fix

‘ ‘5.
—~%.k‘. in}

311.3%

$31.3: ..
hazita‘fl .3i 3

 

NV:
fixymfir.
fit.§1t%1.

3 i
hays?»
Rainy... er.»
tie!!!
.: t... 13:...» flu
sflfl‘hflfibm. .03... {xxxfi

flair .

. 3.5.!
$3.323“!
. .3 ..

 

 

«was

2008

   

LIBRARY
Michigan State
Universi

 

This is to certify that the
thesis entitled

SCREENING INSTRUMENTS FOR NEUROCOGNITIVE
IMPAIRMENT IN HIV/AIDS PATIENTS: A PILOT STUDY UTILIZING
NON-PHYSICIAN HEALTHCARE PROVIDERS IN KALINGALINGA,

LUSAKA, ZAMBIA

presented by

Michelle Powell Kvalsund

has been accepted towards fulfillment
of the requirements for the

MS. degree in Epidemiology

 

 

W Q
l’ 1

Major Professor’s Signature

c9/ijzcc8

Date

MSU is an affinnative-action, equal-opportunity employer

»- _.-.—.-.—-I-gl_-_Inun-.—.-A-a--o—--.-n-.-—.—.—A-.-—.

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

DATE DUE DATE DUE DATE DUE

OCT 0 8 2011

 

 

SCREENING INSTRUMENTS FOR NEUROCOGNITIVE IMPAIRMENT IN
HIV/AIDS PATIENTS: A PILOT STUDY UTILIZING NON-PHYSICIAN
HEALTHCARE PROVIDERS IN KALINGALINGA, LUSAKA, ZAMBIA

By

Michelle Powell Kvalsund

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Epidemiology

2008

 

ABSTRACT
SCREENING INSTRUMENTS FOR NEUROCOGNITIVE IMPAIRMENT IN

HIV/AIDS PATIENTS: A PILOT STUDY UTILIZING NON-PHYSICIAN
HEALTHCARE PROVIDERS IN KALINGALINGA, LUSAKA, ZAMBIA

By

Michelle Powell Kvalsund

HIV-related neurocognitive impairments (N CI) have important clinical, public health,
and economic implications, particularly in light of the rapid expansion of antiretroviral
access in sub-Saharan Africa. Obstacles to the clinical identification of these disorders in
resource-limited settings are multi-factorial, including limited access to physician-experts
for diagnosis and lack of ecologically-sound screening instruments and normative data.
Screening instruments for HIV-related NCI have been utilized in some investigations in
Africa, but none using the primary non-physician healthcare workers (NPHW) staffing
antiretroviral therapy clinics. This study conducted steps for the formal evaluation of
three locally-adapted screening instruments in Zambia, a Zambian Mini-Mental Status
Examination (ZMMSE), HIV Dementia Scale, and Color Trails tests. Instruments were
administered by NPHW to a group of AIDS patients (n=48) admitted to an urban hospice
and a group of healthy subjects (n=15) from the same community. Hospice patients
performed significantly worse (all p’s <0.001) on all instruments. In addition, 24 of 48
(50%) hospice patients, 55% of patients with WHO HIV Stage IV disease, and 33% of
patients with WHO HIV Stage III disease met the study definition for significant
cognitive impairment. Evaluation of individual test items revealed that the ZMMSE was

more easily administered and had better face and content validin than other instruments.

k

Future steps for formal validation of the ZMMSE in Zambia are outlined.

 

Dedicated to the patients at Our Lady’s Hospice
who graciously donated their time to participate in this study

 

ACKNOWLEDGMENTS

I would like to extend a heartfelt thank you to everyone at Our Lady’s Hospice for
your kindness, assistance, and friendship. Thank you to Sister Maria Crucis, Mr.
Sangayakula Sanga, and Venus Shamoya for the support that made this project possible.
I also thank Naomi Malaya, Agness Mwanza, Annie Sakala, Sonile Mulwira, Mulemi
Lubasi, Bienvenu “Brian” Kongolo Kiolo, Sister Precious Mumeno, Naomi Chulu, Willy
Chilufya, Silvia Kunda, and Martha Mbewe for assistance with interviewing.

Thank you to Febby Phiri, Peter Mwanza, and Father Salimo Hachibamba for the
many hours spent diligently translating study documents. Special thanks to Professor
Alan Haworth, who offered guidance that was instrumental in the success of this project.

There are many to thank at Michigan State University. Thank you to Wenjiang Fu for
his statistical guidance. Thanks to Jayne Goeddeke, Kimberley Rockwell, Natalie
Organek, and Katelyn Esch forassisting me from abroad. Thanks to Dr Ellen Velie for
sharing your methodological expertise. Thank you to Dr Daniel Murman, not only for
facilitating that first trip to Zambia, but for always being willing to share your wisdom.
Finally, thank you to my thesis chair, Dr Gretchen Birbeck for always believing this
project was possible, for always nudging me forward, and for seeing me through to the
end. Gretchen, your dedication is an inspiration to me. Thank you for all the time you
invested in this project and in me.

Thank you to my parents, family, and friends for their endless support. To my best
friend and husband, Pelle, thank you for believing in me, for showing me Africa, and for

joining me to South America. I can’t wait to see where our next adventure takes us.

 

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. viii
LIST OF FIGURES ................................................................................. ix
LIST OF ABBREVIATIONS ....................................................................... x

CHAPTER 1: INTRODUCTION

1.1 Overview ....................................................................... l
1.2 Rationale ........................................................................ 7
1.3 Study Aims ................................................................... 10
1.4 Organization of Paper ....................................................... 12
CHAPTER 2: BACKGROUND
2.1 HIV-related NCI Nomenclature ............................................. 13
2.2 Diagnostic Criteria ............................................................ 13
2.3 Clinical Features .............................................................. 15
2.3.1 Common Symptoms ................................................... 15
2.3.2 Neurological Findings ................................................. 16
2.3.3 Neuropsychological Assessments .................................... 17
2.3.4 Course ................................................................... 18
2.3.5 Risk Factors ............................................................ 18
2.4 Frequency of HIV-related NCI in Low and Middle Income
Countries. .......................................................................... 20
2.4.1 Latin America and the Caribbean .................................... 22
2.4.2 Asia ..................................................................... 27
2.4.3 Africa .................................................................... 30
2.4.4 Summary ............................................................... 34
2.5 Importance of identifying HIV-D in sub-Saharan Africa ............... 35
2.5.1 Initiation of Antiretroviral Therapy ................................. 36
2.5.2 Optimizing Treatment for Patients with HIV-related NCI ....... 37
2.5.3 Communication of Medical Information and Medication and
Adherence ............................................................... 38
2.5.4 Quality of Life .......................................................... 39
2.5.5 Caregiver Burden ...................................................... 40
2.5.6 Impact on Employment ................................................ 41
2.5.7 Understanding HIV -D Pathogenesis ................................. 41

 

CHAPTER 3: METHODS

3.1 Setting ......................................................................... 50
3.1.1 Overview of Zambia .................................................. 50

- 3.1.2 Kalingalinga ............................................................ 51
3.2 Population ..................................................................... 51
3.3 Interview Procedures ......................................................... 52
3.4 Language and Cultural Issues ............................................... 53
3.4.1Translation of Study Instruments .......................................... 53
3.4.2 Neuropsychological Assessments .................................... 52
3.5 HIV/AIDS Diagnosis ........................................................ 54
3.6 Definition of Cognitive Impairment ........................................ 55
3.7 Sample Size Calculation ..................................................... 55
3.8 Statistical Analysis ............................................................ 56
3.8.1 Overview ................................................................ 56
3.8.2 Analysis by Specific Aim ............................................. 56
3.8.2.1 Aim 1 ............................................................. 56

3 8 2.2 Aim 2 ............................................................. 57

3 8 2.3 Aim 3 ............................................................. 58

3 8 2.4 Aim 4 ............................................................. 58
3.8.2.5 Aim 5 ............................................................. 58

3.9 Protection of Human Rights ................................................ 59

CHAPTER 4: RESULTS

4.1 Overview and Descriptive Statistics ....................................... 63
4.2 Results by Specific Aim ...................................................... 64
4.2.1 Aim 1 .................................................................... 64
4.2.1.1 Zambian Mini-Mental State Exam ............................ 64

4.2.1.2 HIV Dementia Scale ............................................ 66

4.2.1.3 Color Trails 1 & 2 ............................................... 66

4.2.2 Aim 2 .................................................................... 66
4.2.2.1 Zambian Mini-Mental State Exam ........................... 66

4.2.2.2 HIV Dementia Scale ............................................ 68

4.2.2.3 Color Trails l .................................................... 68

4.2.3 Aim 3 .................................................................... 69
4.2.4 Aim 4 .................................................................... 69
4.2.5 Aim 5 .................................................................... 69
4.2.6 Aim 6 ..................................................................... 70
CHAPTER 5: CONCLUSIONS ......................................................... 81
5.1 Summary of Findings ......................................................... 81
5.2 Comparison of Findings to Previous Literature ........................... 81
5.2.1 Screening Instruments for HIV -related NCI ........................ 81
5.2.2 Prevalence of Cognitive Impairment .................... , ............ 84

vi

5.3 Strengths and Limitations ................................................... 86

5.4 Suggestions for Future Research ........................................... 87
APPENDIX A .............................................................................. 90
APPENDD( B .............................................................................. 92
REFERENCES ............................................................................. 99 -

 

LIST OF TABLES
Chapter 2:

Table 1: Comparison of American Academy of neurology and HIV
Neurobehavioral Research Center Guidelines for Diagnosis of HIV-related

NCI ........................................................................................... 44
Table 2: Barriers to Knowledge and Comparability of Previous Investigations of
HIV-related NCI In Low and Middle Income Countries ............................. 45
Table 3: Summary of HIV —related NCI Studies in Low and Middle Income
Countries by World Region ............................................................... 46
Chapter 3:

Table 4: Neuropsychological Instruments and Cognitive Domains Assessed. ....60
Table 5: World Health Organization Screening Algorithm for HIV Infection. . ...60
Chapter 4:

Table 6: Baseline Characteristics of Persons Living With AIDS (PLWAS) and
Comparison Group ......................................................................... 73

Table 7: Number of Correct Responses (%) to Zambian Mini—Mental State Exam
Items by Group and Cognitive Domain Assessed ....................................... 73

Table 8: Number of Correct Responses (%) to HIV Dementia Scale Items by
Group and Cognitive Domain Assessed ................................................. 74

Table 9: Neuropsychological Assessments for Persons Living With AIDS
(PLWAS) and Comparison Group ....................................................... 75

viii

 

LIST OF FIGURES

Chapter 3:
Figure 1: Map of Zambia .................................................................. 61
Figure 2: Map of Kalingalinga, Lusaka, Zambra62
Chapter 4:
Figure 3: Flow Chart for Patients and Comparison Group ........................... 76

Figure 4: Distribution of Persons Living With AIDS (PLWAS) and Comparison
Group Zambian Mini-Mental State Exam Total Scores ............................... 77

Figure 5: Distribution of Time (in seconds) to Complete the HIV Dementia Scale
Timed Number Writing Task for Persons Living With AIDS (PLWAS) and
Comparison Group .......................................................................... 78

Figure 6: Distribution of Persons Living With AIDS (PLWAS) and Comparison
Group HIV Dementia Scale Total Scores ............................................. 79

Figure 7: Distribution of Time (in seconds) to Complete Color Trails 1 among
Persons Living With AIDS (PLWAS) and Comparison Group ....................... 80

 

LIST OF ABBREVIATIONS

ADI: AIDS Defining Illness

AAN: American Academy of Neurology

ART: Antiretroviral Therapies

CI: Cognitive Impairment

HAART: Highly Active Antiretroviral Therapy

HDS: HIV Dementia Scale

HIV-D: HIV —associated Dementia

HBC: Home-based care

IHDS: International HIV Dementia Scale

MCMD: Mild cognitive motor disorder

lVflVISE: Mini Mental State Examination

NCI: Neurocognitive impairments

NP: Neuropsychological

NPHW: Non-physician Healthcare Workers

PEPFAR: President’s Emergency Plan for AIDS Relief
PLWAS: Person Living With AIDS

SSA: Sub-Saharan Africa

WHOCCD: World Health Organization Clinical Case Definition

ZMMSE: Zambian Mini-Mental State Examination

 

CHAPTER 1: INTRODUCTION

1.1 Overview:

HIV -associated neurocognitive impairments (NCI) range among a continuum from
subtle, even subclinical deficits in psychomotor speed, memory, and attention to a
fulminant subcortical dementia with marked impairment of cognitive capacity and
functional ability. The evolution of HIV/AIDS care, and most particularly the advent of
Highly Active Antiretroviral Therapy (HAARTY in 1996, has brought shifting patterns of
HIV-associated NCI including changing incidence and prevalence, and variable
progression and clinical features. The pathogenesis and best therapeutic approach to
HIV-related N CI remains ill-defined more than twenty years since these syndromes were
first described. The epidemiology of HIV-related NCI in developing countries, where the
majority of HIV/AIDS patients reside, remains largely unknown.

The course and clinical featUres of HIV—related NCI appear to be changing with
HAART. In the pro-HAART era, the course of HIV dementia (HIV-D), the most severe
form of HIV —related NCI, was generally that of a subacute, progressive dementia with
prominent subcortical features, such as psychomotor slowing. Since the advent of
HAART, longitudinal cohort Studies have reported increased frequency of milder NCI
with phenotypic features of both cortical and subcortical dementias in HAART treated

patients”. HIV-D often now results in a more static or chronic dementia state with

 

' Highly Active Antiretroviral Therapy (HAART) refers to the use of three or more
antiretroviral agents to suppress systemic HIV replication. The use of HAART in clinical
practice became widespread in 1996 when the first protease inhibitor (third class of
antiretroviral agents) was approved for clinical use.

. K

 

varying degrees of reversibility, resulting in some clinicians and researchers to propose
new HIV—D categories to more adequately describe HIV—D subtypes4.

Along with changes in the course and clinical profile of HIV-related NCI, HAART
has reduced the incidence of HIV -related NCI. Before the HAART era, approximately
20 to 30% of AIDS patients in the United States suffered from HIV-D“. An additional
30% of symptomatic HIV patients in the US. suffered from a less severe form of NCI
known as Mild Cognitive Motor Disorder (MCMD).1’3 The incidence of HIV-D has
decreased by roughly 50% in the post-HAART era from 21/1000 person—years to 10/1000
person-yearsHo. However, patients with advanced HIV disease, despite HAART
therapy, are still at significant risk for the development of HIV -related NCI. In the
Northeastern AIDS Dementia (NEAD) cohort (a group of high risk HIV patients with
CD4 counts less than 200 cells/mm3 identified through four university infectious disease
clinics in New York and Maryland), incidence rates remained nearly identical to a similar
pre-HAART cohort; both groups exhibited a 25% cumulative incidence of HIV-D at one
year, and 38% versus 40% cumulative incidence for the NEAD and pre-HAART cohort
respectively after two years3. In addition, a rising incidence of HIV-D was noted
amongst Johns Hopkins HIV clinic attendees in 2003“. This finding may lend support to
previous concerns that poor antiretroviral drug penetration in the Central Nervous System
(CNS) combined with the immunologic peculiarities of the CNS may led to a differential
impact of HAART on the neurological complications of HIV relative to other HIV-
related illnesses (i.e. “CNS escape.”)“.

Overall, prevalence studies in developed countries have suggested Static to increasing

prevalence of all HIV-related NCI with improved survival following HAART3’12'”.

2 L

There is some concern that prevalence rates will continue to rise with aging of the HIV
population, as well as with the development of other co-morbid conditionsls'ls.
Investigators at Johns Hopkins University reported an increasing prevalence of HIV-D
among 1300 HIV clinic attendees from 6.6: 100 person-years in 1994 up to 10.1: 100
person—years in 2000, although the statistical Significance of this trend was not
reportedlg’zo. In Australia, Dore et a1 utilized the National AIDS Registry to follow
survival after a diagnosis of HIV-D as an initial AIDS defining illness (ADI). They
found that HIV -D represented 4.4% of all ADIs in 1992-1995, but this prevalence
increased significantly following the introduction of HAART in 1996 and 1997, with
HIV-D cases representing 6.0% and 6.5% of ADIS, respectively (p=0.02)21. Neuro-
pathological studies conducted in the US. in early HAART treated patients also found a
trend towards increased prevalence of HIV encephalitis (HIV-E), the pathological
correlate of HIV-D, although this trend was not statistically significantzz'za.

Other studies have found static pre- and post-HAART prevalence rates of HIV-related
NCIIZ’”. A cross-sectional study from a tertiary HIV center in Australia found that,
among patients with advanced HIV disease, but without HIV-D, the prevalence of milder
forms of HIV-related NCI remained unchanged between pre- and post-HAART cohorts”.
In another study of 432 HIV patients attending a large HIV referral clinic in Rome, Italy,
the prevalence of both mild HIV—related NCI and HIV-D remained unchanged after seven
years of follow—up with HAART”. These findings suggest that studies conducted early
in the post-HAART era may be impacted by increased survival in patients with fixed

neurological deficits or “bum-out” disease“, while patients identified and treated earlier

with HAART may have more reversible cognitive deficits and/or a higher likelihood of

s a.

halting disease progression to more severe forms of HIV-related NCI. The true reasons
for these differences remain speculative, but may become more clear with increased
duration of follow-up in HAART-treated cohorts with milder HIV disease, higher nadir
CD4 counts (a marker of less severe HIV disease), and increased age. Still, whether the
prevalence of HIV-related NCI remains static or increasing, these neurocognitive deficits
clearly remain important consequences of HIV infection in developed countries despite
the most effective combinations of HIV therapies available today.

Our current understanding of HIV —related NCI epidemiology comes almost
exclusively from well-educated, homosexual Caucasian mens'zs, who may not be
reflective of HIV-related NCI in other groups within the US. or globally. There are
many reasons to suspect that the profile and frequency of HIV -re1ated NCI may differ in
developing countries. Different HIV subtypes found in different world regions may have
variations in degrees of neurotropism. Co-factors such as poor nutrition, low education,
different HIV transmission patterns and patterns of concomitant disease may all result in
different frequency and severity of these inrpairrnents. Unfortunately, the true impact of
these co-factors on HIV-related NCI epiderrriology is largely unknown. Studies from low
and middle income countries assessing HIV-related NCI have varied widely in the
neuropsychological batteries used, guidelines for diagnosis, and the range of
neuropsychological impairment included in their figures. Variations in HIV-related NCI
nomenclature over time have led to the use of a variety of terminology in these studies
and vague categorizations of impairments. The inclusion of patients with mild cognitive
impairment can inflate prevalence figures compared to studies that exclude these patients.

The neuropsychological battery employed for each study may also alter prevalence rates,

 

raising the prevalence with increasing cognitive domains tested. The proportion of
patients with end stage AIDS among a study population will also impact the prevalence
rate, as HIV-D typically occurs with advanced HIV disease”.

The result of these differences in study design has been a wide range of HIV-D
prevalence estimates, both internationally and within geographic regions. In Latin
America and the Caribbean, HIV-related NCI prevalence estimates have ranged from
2.6% to 28.6%26'31. Studies from Asia have reported HIV-D prevalence estimates from
0.1% in Thailand32 to 16.7% in China,33 and within India studies have ranged from 8% to
as high as 35%‘6'34'37. In sub-Saharan Africa, where 22.5 (68%) of the 33.2 million
persons living with HIV reside”, prevalence estimates ranging from 3.2 to 54% have
been reported3944.

Only an estimated 2 million (28%) of the 7.1 million PLWAs globally who require
treatment with antiretroviral therapies (ART) today are able to access these life-saving
treatments”. However, access. to ART is rapidly expanding as a result of efforts by
WHO/UNAIDS and increased financial commitment from funders such as the Global
Fund to fight AIDS, Tuberculosis and Malaria; the President’s Emergency Plan for AIDS
Relief (PEPFAR); the World Bank; and others“. Our inability to anticipate the impact of
widespread distribution of ART on the prevalence of HIV-related NCI may result in
inadequate allocation of resources and significant distress to informal care systems, like

Home-Based Care' (I-IBC)“.

 

‘ Home based care programs offer community-based HIV /AIDS care via volunteer
healthcare workers and community members. By providing care to HIV/AIDS patients at
home, these programs have offered non-institutional care services, lessening the burden
of HIV/AIDS on government hospitals, while shifting the burden of care to families and

communities.
5 ‘

The high demand for antiretroviral treatments, coupled with geographic,
socioeconomic, and infrastructural limitations in sub-Saharan Africa have necessitated
policies that limit ART initiation to patients with very low CD4 counts (<200 cells/mma),
when available, or clinical evidence of advanced HIV disease (i.e. WHO Stage III or
Stage IV). HIV-associated dementia (HIV -D) is a qualifying indication for antiretroviral
therapy (ART) initiation according to World Health Organization (W HO) guidelines for
resource-limited settings“, but is not routinely screened for in ART clinics in sub-
Saharan Africa (SSA) due to a lack of valid instruments with population-based normative
data. HIV-D frequently goes unrecognized by clinicians inexperienced in diagnosing
such disorders and may lead to delays in treatment initiation. A recent editorial
highlighting experience from Uganda has pointed out that 6 of 16 HIV—D patients had
CD4 counts above 200 cells/mm3, making a clinical-based diagnosis essential for ART
commencement in 37.5% of HIV-D patients identified“. Such delays in ART initiation
may result in Significant and irreversible CNS damage that complicates future HIV
management“.

In sub-Saharan Africa, where economic development has nearly halted under the
strain of HIV-associated mortality and morbidity, HIV—related NCI may prevent
desperately needed employees from returning to the workforce”. I-HV-related NCI has
important implications for individual households as well, when significant neurocognitive
impairment strikes the main provider or results in other family members exiting the
workforce to provide care to these individuals”. The success of large-scale, heavily
subsidized antiretroviral treatment initiatives may also be hindered, as HIV-related NCI

has been associated with poor adherence to antiretroviral therapies (ART)51. Poor

6 L

 

adherence is frequently associated with ART failures and development of drug
resistanceSz. Thus, enabling healthcare workers to diagnose HIV-related NCI is of urgent

importance in planning appropriate HIV treatment and prevention initiatives.

1.2 Rationale

Improved access to antiretroviral therapies in sub-Saharan Africa and other regions
with resource limitations are likely to cause significant shifts in the epidemiology of HIV
and HIV-associated illnesses. As more prominent systemic infections decline and HIV
survival increases, HIV-related NCI are likely to become increasingly prevalent though
the healthcare sector may be slow or unable to recognize this readily. The ability of the
primary non-physician healthcare workers who staff ART clinics in sub-Saharan Africa
to diagnose HIV-related NCI is limited by the lack of valid instruments to identify these
impairments. The use of extensive neuropsychological batteries administered by
neurologists, psychiatrists, and neuropsychologists in previous investigations of HIV -D
make these methods impractical for routine surveillance of HIV-D in understaffed,
overcrowded ART Clinics in Africa. For example, the WHO Neuropsychiatric AIDS
study reported a mean time required for experienced neuropsychologists to conduct the
neuropsychological battery of 2 hours”. The development of a simple instrument
reasonably sensitive and specific to identifying the cognitive deficits seen in HIV -related
NCI that could be administered by non-physicians would prove more feasible for
screening and surveillance in countries with a lack of general or specialist physicians,

limited financial means, and already overburdened medical resources.

In the US, there are several instruments used in routine clinical settings to assess
cognitive status. The Mini Mental Status Examination (MMSE) is a brief screening
instrument typically used to identify patients with cortical dementia syndromes, such as
Alzheimer’s disease“. Cognitive domains assessed include orientation, attention,
language, constructional praxis, and memory and recall. The MMSE is generally
considered to be insensitive to the cognitive deficits seen in HIV -D, although the clinical
utility may perform better in post-HAART cohorts with more cortical features of HIV-
related N C14855 . MMSE performance is known to be impacted by factors like respondent
education and many items may be culturally dependent“. However, the MMSE has been
modified and validated in several resource limited-settings and illiterate populationsSMI.
The MMSE has the advantage of widespread familiarity and frequent informal use. In
Zambia, neuropsychiatric officers, clinical officers with specialized training in
neuropsychiatry, often use questions adapted from the MMSE for cognitive assessment in
their patients (personal reference). Unfortunately, there has never been a formal
translation or validation of these questions for NCI assessment in the Zambian
population.

Color Trails l & 2 are designed specifically for cross-cultural use and assess attention,
psychomotor speed, and executive function”. These are all cognitive domains that are
frequently compromised in HIV-D. Color Trail tests are variations of the better-known
Trail Making A & B tests, but replace the use of the English alphabet in Trail Making B.
In Color Trails 1, patients are required to draw lines connecting numbers consecutively
from one to 25 during a timed trial. Color Trails 2 contains two sets of numbers, each of

different colors, and patients must alternate each advancing number with a different color

 

(i.e. yellow 1, pink 2, yellow 3). This test (also timed) requires higher executive
functioning skills to sort and retrieve correct responses and inhibit selection of incorrect
items. A disadvantage of Color Trails 1 & 2 is the need for instrumentation, most notably
color copies of the test document.

The HIV Dementia Scale (HDS) has been widely used as a sensitive screening tool for
HIV-D in the U.S.62, but its utility in other populations has not been evaluated. The HDS
includes tests of motor speed (timed written alphabet), memory (recall four words at 5
minutes), constructional praxis (timed drawing of 3D cube), and its original form also
included a test of executive function (antisaccadic error test). The antisaccadic error test,
however, was noted to be poorly administered by non-neurologists and was subsequently
removed from the HDS63.

Another scale for HIV Dementia, the International HIV Dementia scale (II-IDS)“, has
recently been developed specifically for cross-cultural use. The IHDS contains tests of
motor speed (timed finger tapping), psychomotor speed (timed alternating hand sequence
test), and memory (recall of four words). The IDHS was shown to have 64% sensitivity
and 71% specificity for identifying HIV-D in a group of ambulatory HIV patients (n=81)
in Kampala, Uganda“. Although the neuropsychological (NP) instruments utilized in
this study for HIV-D consensus diagnosis and for comparative performance on the IHDS
have never been validated in Uganda, the authors were able to collect data on a group of
100 seronegative controls from the same clinic to provide some normative data for
diagnosis and comparison purposes“. The IHDS was also able to identify significant
differences in performance between HIV patients and a group of normative controls in an

Indian population based on H-IDS cut-off scores derived from studies in the US. and

 

Uganda”. In a recently published study from Ethiopia, the IHDS failed to identify
significant differences in performance between patients with untreated, moderately
advanced HIV disease (i.e. median CD4 <300) and a group of seronegative controls“.
This finding suggests a decreased prevalence of HIV-related NCI or insensitivity of the
IHDS to milder NCI in this population.

It is notable that in all studies utilizing the IHDS the test has been administered by
physicians, although physician-level care represents only a small minority of primary
HIV care in many resource-limited settings. In sub-Saharan Africa, clinical officers, who
receive three years of post-secondary clinical training, provide the bulk of medical and
HIV care38'65. If an instrument is to be feasibly incorporated into the routine care setting
in Africa, then studies of this instrument should be conducted utilizing clinical officers
for administration and scoring procedures. There may also be a need to expand the IHDS
to include assessment of the mixed phenotypic presentation of HIV—D noted in US.
populations following widespread introduction of HAART (i.e. with both subcortical and

cortical deficits seen).

1.3 Specific Aims

This study aims to investigate the feasibility of using screening instruments including
a locally-adapted Zambian MMSE (ZIVIMSE), the HIV Dementia Scale (HDS), and
Color Trails 1 & 2 in identifying HIV-related NCI among AIDS patients admitted to Our
Lady’s Hospice, in Kalingalinga, Lusaka, Zambia. The IHDS was not available at the
time of this study and was therefore not assessed. This study will also describe the

prevalence of HIV-related NCI among these patients. If successful, these instruments

 

may be used by paramedical healthcare workers in clinical settings, as a method for

surveillance of HIV-related NCI, and may be adaptable to other African populations as

well.

The specific aims of this study are:

1) To assess cultural relevance and adaptability of screening instruments,
including the ZMMSE, HDS and Color Trails l & 2, to identify cognitive
impairment among Persons Living with AIDS (PLWAS) in Lusaka, Zambia
through focus group discussions with local medical professionals, community

piloting, and in study participants.

2) To evaluate the distribution of individual neuropsychological test items

between PLWAS and healthy community members in Lusaka, Zambia.

3) To evaluate differences in mean ZMMSE, HDS, and Color Trails l & 2
between PLWAS admitted to Our Lady’s Hospice in Lusaka, Zambia and healthy

persons from the same community without evidence of HIV/AIDS
4) To evaluate WHO HIV Stage as a potential effect modifier of performance on
the ZMMSE, HDS, and Color Trails l & 2 among PLWAS admitted to Our

Lady’s Hospice in Lusaka, Zambia.

5) T 0 determine the prevalence of HIV-related NCI among PLWAS admitted to

Our Lady’s Hospice in Lusaka, Zambia, as defined by scoring 2 2 standard

. x

deviations (SD) below the comparison group on the ZMMSE alone, or Z 2 SD

below the mean on both the HDS and Color Trails 1.

6) To provide an overview of potential methods for formal cross-cultural
validation of locally-adapted screening instruments for HIV-related NCI in

Lusaka, Zambia based on pilot study findings.

1.4 Organization of Paper

This thesis includes an introduction, background and literature review, methods,
results, and conclusions. In addition, Appendix 1 includes a copy of the ZMMSE.
Appendix 2 includes a manuscript for submission to a peer—reviewed journal, and thus

contains information redundant to the thesis manuscript.

l2

CHAPTER 2: BACKGROUND

2.1 HIV-related NCI Nomenclature

HIV-D has been and continues to be described in the literature under various
terminology. HIV encephalopathy, AIDS Dementia Complex (ADC), and HIV -D are all
synonymous terms to describe the clinical triad of cognitive, motor, and behavioral
abnormalities that result in functional impairment in HIV patients. Minor cognitive
motor disorder (MCMD) is used to describe HIV patients who have significant changes
in cognition, motor dysfunction, and behavioral abnormalities when these changes are not
severe enough to interfere with activities of daily living“. HIV Encephalitis (HIV-E) is a
pathologic diagnosis reserved to describe the hallmark neuropathologic features of
clinically evident HIV -D, including white matter pallor, microglial nodules,

multinucleated giant cells, and gliosis67.

2.2 Diagnostic Criteria

The gold standard for diagnosing HIV-D is the American Academy of Neurology
(AAN) clinical criteria“. According to the AAN guidelines, a diagnosis of HIV-D can be
reached in patients with at least a one month history of cognitive decline resulting in
impairment of activities of daily living in at least two of the following cognitive domains:
attention/concentration, speed of processing of information, abstraction/reasoning,
visuospatial skills, memory/learning, and speech/language. Impairment is usually
identified through neuropsychiatric testing batteries. In addition, a diagnosis requires

finding either objective motor dysfunction or behavioral abnormalities. Objective motor

13

dysfunction may be identified on neurological exam as slowed rapid movements,
abnormal gait, limb incoordination, hyperreflexia, hypertonia, or weakness. Behavioral
abnormalities include change in personality with apathy, inertia, irritability, emotional
lability, impaired judgment, or disinhibitionés. The diagnosis of HIV—D also requires
exclusion of other potential causes of cognitive impairment through CSF evaluation and
neuroimaging. The diagnosis of MCMD would require all of the above evidence in the
absence of functional impairment.

The AAN guidelines, as proposed in 1991, have been criticized for a number of
limitations that restrict, complicate, and lead to overlap between the clinical diagnoses of
HIV-D and MCMD. For example, motor and behavioral abnormalities must be present
to lead to a diagnosis of HIV-D. Thus, patients with severe NCI, but without motor or
behavioral symptoms cannot be diagnosed with HIV —D. Although previous research had
indicated that NCI without behavioral or motor symptoms was rare in the natural history
of HIV-D, this may be changing with HAARTSS. Conversely, patients without NCI can
receive a diagnosis of MCMD if there is evidence of motor or behavioral abnormalities.
While a diagnosis of HIV—D requires impairment in two cognitive domains, the number
of domains impaired in MCMD is not delineated. In addition, the AAN criteria fail to
specify the degree of neuropsychiatric (NP) impairment required to differentiate HIV-D
from MCMD. This allows the same NP presentation to qualify for both HIV-D and
MCMD diagnoses. Quantification of degrees of functional impairment are also lacking.
Further, patients with reliably documented cognitive difficulties that do not interfere with
activities of daily living may not meet criteria for either diagnosis. Recently, the HIV

Neurobehavioral Research Center (HNRC), an independent research institution in San

14

 

k

Diego, California, has proposed new categories to reduce overlap of diagnoses of mild
HIV-D and MCMD, to reflect changes in the clinical profile of HIV-D cases since
widespread HAART use, and to improve identification of asymptomatic NCI that
correlates with HIV-E findings at autopsy and yields important prognostic value for
clinicansss. The HNRC proposes three categories including 1) HIV-D; 2) Mild
Neurocognitive Disorder; and 3) Asymptomatic neurocognitive impairment. In contrast
to the AAN criteria, the HNRC criteria place added emphasis on N CI manifestations in
HAART treated patients. HNRC criteria include categorization of NCI as normal, mild
to moderate, or moderate to severe global NP impairment based on a standard NP battery
with demographically adjusted normative means. There is also specification about the
degree of functional impairment required for each diagnosis. Differences between the
AAN and HNRC criteria are summarized in Table 1. When comparing AAN and HNRC
categories to 39 autopsy cases with evidence of HIV -E, the HNRC categories had
improved sensitivity (67% versus 56%), improved specificity (92% versus 83%), and
improved positive predictive power (95% versus 88%), though the authors did not
compare these differences statistically. Since these guidelines have only recently been
published (2007), all studies to date have used the AAN criteria as the gold standard for

HIV-related NCI. However, future research may benefit from using the HNRC criteria.

2.3 Clinical Features.
2.3.1 Common Symptoms
Early in HIV-D, patients often complain of lack of concentration, decreased

attentiveness, and a general slowing of thought processes”. Motor complaints are also a

. x

prominent feature of the disease. In one case series of 111 HIV—D patients, 75% of HIV -
D patients complained of gait Iunsteadiness and 50% noted clunrsiness and changes in
handwriting”. Among behavioral symptoms, apathy and social withdrawal are the most
common early manifestations, with agitation, mania, and hallucinations occurring less
frequently”. Depression has been reported to be present in up to 30% of HIV -D cases”.
Amongst the very limited studies that present clinical features of HIV-related NCI
outside the US, Europe, and Australia, the pattern (i.e. early and late symptoms) of
clinical complaints seems to be consistent. In an Indian series of ADC cases,
forgetfulness associated with slowed mental abilities, motor Signs including loss of
balance and leg weakness, and apathy and social withdrawal were frequently reported”.
A case series of 6 ADC cases in Shanghai, China reported forgetfulness and inattention
aslcommon early symptoms”. Other symptoms encountered included inability to
calculate, disorientation to time and place, difficulty recognizing family members, and
reduced external responsiveneSs. As ADC advanced, personality change, apathy, and
social withdrawal were common. Two patients reported incontinence, one had visual
hallucinations, and one was unable to write.
2.3.2 Neurological Findings

Neurological signs are non-focal and may include gait disturbance, impaired rapid eye
movements, hyperreflexia, and frontal lobe release signsz. Additional findings may
include slowing of fine finger movements, mild leg weakness, and inability to perform
tandem gait. An estimated one third of patients will also have a peripheral neuropathy
associated with depressed or absent ankle jerks“. In the pre-HAART era, these

disturbances invariably progressed over a period of weeks to months to involve

. x

significant global dysfunction including mutism, abulia, and ataxia, often accompanied
by hypertonia, paraplegia, and bladder/bowel incontinence.
' 2.3.3 Neuropsychological Assessments

Neuropsychological assessments are important in HIV-D because the results provide
important information about the types of deficits patients encounter, and provide
temporal information regarding progression of deficits and response to treatment when
testing is repeated over time. A recent meta-analysis of Studies characterizing the
neuropsychological profile of HIV-D in the pre-HAART era has reported that domains of
learning, motor coordination, language fluency, and memory were the most severely
impaired, followed by complex attention and simple processing speed which were
moderately to severely impaired". AIDS patients without HIV-D also had significant
impairments compared to seronegative controls on neuropsychological test performance,
especially in the domains of complex attention and psychomotor speed. Similar, but
milder impairments were found in patients with symptomatic HIV infection. The WHO
Neuropyschiatric AIDS study, conducted in the pre—HAART era, reported a consistent
pattern of deficits in fine motor control, sustained and selective attention, cognitive
flexibility, and verbal learning across five geographically diverse countries including,
Germany, Brazil, former Zaire, Kenya, and Thailand”. In Kampala, Uganda similar
patterns of neuropsychological testing abnormalities in verbal learning and memory,
speed of processing, attention, and executive functioning were seen amongst HIV
outpatients with 28% receiving HAART therapy”.

There is some indication, although indirect, that the pattern of deficits seen in HIV-

related NCI may be changing with HAART. For example, Cysique et al found that

 

HAART treated patients had an increased frequency of verbal memory impairment
compared to pre—HAART patients”. Brew et al found that characteristic
hypermetabolism of the basal ganglia found on positron emission tomography (PET) of
HIV-D patients in the pre-HAART era is now much less prominent“. Instead, there
appears to be more mesial temporal lobe hypometabolism in HAART treated patients on
PET scan. Further research will need to be conducted to identify if these preliminary
findings represent true variations in the neuropsychological profiles of HIV-related NCI
related to ART.
2.3.4 Course

In the pre-HAART era, the onset of HIV-D occurred insidiously with progressively
worsening cognitive function and death over a period of months. Mean survival
following a diagnosis of HIV-D was 3 to 6 months25'74. Since HAART, the progression
of deficits seen in HIV-related NCI is much more variable leading some to propose new
HIV-D subtype categories, such as subacute aggressive, chronic active, chronic inactive,
and reversible dementiaz.
2.3.5 Risk Factors

In the pre-HAART era, risk factors for HIV—D were highly associated with markers of
advanced immunosuppression, including high viral load, low CD4 count, constitutional
symptomss, clinical features like depression69 and sustained declines in psychomotor
slowing”. Some of these risk factors have changed with HAART treatment, namely
HIV-D cases are more equally distributed over a range of CD4 cell countslo'21’76.

Screening for HIV -related NCI is still recommended among HIV/AIDS care providers.

 

Age has been investigated as a risk factor for development of HIV -D. A study of
Michigan HIV/AIDS registry patients found that 5.6% of patients aged >50 at HIV
diagnosis compared to only 3.3% of persons between 13-49 years of age at HIV
diagnosis received a diagnosis of HIV—D (p=0.001)15. Failure to control for duration and
stage of HIV infection may play a role in study, as older HIV patients were significantly
more likely to present with CD4 counts below 200 than younger HIV patients at
diagnosis. However, the Hawaii Aging with HIV-1 Cohort found that even after
adjusting for differences in education, race, substance abuse, HAART status, viral load,
and CD4 count, older patients were 3.3 (1.32-8.07) times more likely to meet criteria for
HIV—D.17 A study of HIV patients in Kampala, Uganda has also reported that age was a
significant risk factor for HIV-D, with each additional 10 years of age conferring a 2.06
increased odds of HIV-D (p<0.05).44 These findings have led to concerns that aging of
the HIV population may lead to additional increases in HIV-D prevalence.

Female gender has also been investigated as a risk factor for HIV-D, but many studies
have been limited by failure to control for care-seeking patterns or patterns of care
received by HIV positive women that may be associated with increased risk for HIV -
related NCI. A recent study of Puerto Rican women, almost all of who where receiving
HAART, reported an extremely high prevalence of HIV—D at 28.6%“. The authors
found that many women in this study had detectable plasma (82.3%) and CSF (37.5%)
HIV viral loads, suggesting poor HIV control and treatment failure. In a large
multicenter retrospective study of European patients, HIV-D was present in 8.9% of
women compared to 4.1% of men (p=0.0002)77. Historically recognized patterns of

homosexual transmission may have resulted in more advanced HIV disease at diagnosis

 

among women in this study of HIV cases between 1979 and 1989. Indeed, Robertson et
al reported no gender differences in neuropsychological test performance or progression
of neurological disease in a prospective cohort of HIV patients followed for three years
and matched for disease duration and markers of immunosuppression at a university
clinic in North Carolina”. Given the high prevalence of HIV infection of women in many
developing countries, these findings warrant further investigation of gender differences in

HIV-related NCI.

2.4 Frequency of HIV -related NCI in Low and Middle Income Countries

In the US, studies of the pre—HAART prevalence of HIV-D revealed that
approximately 20—30% of HIV patients suffered from HIV-D3, with an annual incidence
of 7% among AIDS patientss. HAART has resulted in significant decreases in the
incidence of HIV-D, although prevalence estimates have remained unchanged and in
some reports increased with improved survival3‘8’l3. If similar trends are seen in
resource-limited settings, then HIV-D will likely become the leading cause of dementia
in adults under 40 worldwide”.

Unfortunately, very little is known about the epidemiology of HIV -D in low and
middle income countries. Obstacles to identification and understanding of these
disorders include universal non-resource related factors, limitations of infrastructure and
human resources, as well as factors related to healthcare worker perceptions (Table 2).
Variations in terminology, ambiguities in the current AAN criteria, and variable severity
of HIV illness among the population being studied can all impact research findings and

limit comparability. In addition, there has been a tradition of dependence on physician-

20 x

specialists for diagnosis of these disorders. In the US, Europe, and Australia,
investigations of HIV-related'NCI often employ a series of physicians and other
specialists, including neurologists, psychiatrists, and neuropsychologists for diagnosis
and interpretation of findings. Specialist services are far out of reach in many regions of
the world, where non-physician healthcare workers are responsible for HIV primary care
and management. These investigations also employ neuropsychiatric tests batteries that
are lengthy and time-consuming, and not likely to be employed for routine clinical
purposes in resource-limited settings. Even more importantly, NP test instruments are
known to be age, education, and culturally dependent. Thus, NP tool development and
validation in the population of interest are necessary first tasks in HIV -related NCI
studies, which are expensive and ambitious research agendas in themselves. Healthcare
worker perceptions may also play a role, particularly in retrospective studies. Mental
health disorders are highly stigmatized conditions and frequently blamed upon witchcraft
in sub-Saharan Africa and perhaps other regions. Thus, healthcare workers may fear that
labeling patients with such disorders will result in withdrawal of care and/or pursuit of
traditional medicine. Anecdotal beliefs among healthcare workers that short survivorship
among HIV/AIDS patients in Africa and other regions make HIV-related NCI rare, if not
unimportant, also likely contribute to under-identification of the disorder. Another factor
may be that, in contrast to other HIV-associated opportunistic infections, there is no
treatment for HIV-related N CI beyond the use of HAART. Thus, the impetus among
healthcare workers to identify these conditions, at least in the past, has been low.

Still, review of the literature from low and middle income countries suggests that

HIV-related NCI are important manifestations of HIV in these settings (Table 3). The

21

 

following section, organized by world region, summarizes findings, strengths, and
limitations of these studies. The studies reported were identified through all PubMed
entries submitted through August 2007, utilizing the term HIV plus one of the following
terms: cognition; neuropsychiatric, or dementia. Additional articles were identified
through review of citations from the full articles retrieved via PubMed.
2.4.1 Latin America and the Caribbean

Studies of HIV-D from Latin America and the Caribbean have been conducted in
Mexico, Puerto Rico, and Brazil (see Table 3). Studies within Mexico have revealed
very consistent prevalence estimates, ranging from 8.5-10.8%. Still, there are many
limitations and factors that are not accounted for in these studies. A study of 120 HIV
patients (all of whom met Center for Disease Control criteria for AIDS) referred for
neurological evaluation to two main neurological institutes in Mexico City, Mexico
reported 13 cases (10.8%) of ADC between 1986 to 198830. A strength of this study was
that other causes of cognitive impairment in HIV patients were able to be excluded via
CSF analysis and computed tomography (CT) scan in all ADC cases. In addition, one
third of the ADC cases underwent Magnetic Resonance Imaging (MRI), and four cases
had postmortem histopathologic examinations. However, it is unlikely that all HIV
patients have equitable access to neurological services in Mexico, so this study
population may represent only a subset of Mexican HIV patients that can access
specialist services. Thus, the external validity of this study is difficult to assess. Another
concern is that this study was conducted before any standardized guidelines were
published to diagnose HIV-D. The authors use the terms HIV encephalopathy and ADC

interchangeably, and it is unclear if this estimate includes patients with MCMD. It is also

22

 

unclear if neuropsychological tests were employed for evaluation of cognitive function.
Another study conducted in Mexico City, Mexico by Gongora—Rivera et a1,
retrospectively evaluated all HIV patients evaluated at the National Institute of
Neurology and Neurosurgery between 1990 and 1998.28 This study identified 13 cases
(8.7%) of ADC amongst 149 HIV/AIDS patients. The ability of the investigators to
exclude other causes of altered cognition was again excellent, as all patients were
evaluated by a neurologist and had access to a wide variety of diagnostic studies
including CSF analysis, CT, and MRI. However, abnormal mental status was an initial
complaint in 21 (14.1%), including 15 patients with subacute behavioral abnormalities,
four cases with acute confusional state, and two patients with chronic cognitive
impairment. It is possible that some of these cases may have represented early
manifestations of ADC or milder forms such as MCMD that did not fulfill the study
definition for ADC. Conversely, it is possible that the additional cases were found to
have other causes for altered mental status, such as CNS-opportunistic infections, but the
final diagnoses of these cases were not reported or unknown due to loss of follow-up or
insufficient documentation. Sirrrilar to the previously mentioned study, the
generalizability of the study findings are difficult to assess because information about
systems of referral and access to this institute are not reported. In addition, if any
objective measures were used to identify cognitive impairment, the authors do not report
NP assessments. There are several additional factors that may impact HIV-D prevalence
that are not addressed in this study. Firstly, this study evaluated HIV cases from 1990
and 1998 and therefore some patients may have been receiving monotherapy, dual

therapy, or even HAART, but ART usage is not reported. Another concern is that there

23

is also no indication of stage of HIV infection of study participants, a factor known to
impact prevalence rates. In another study from Mexico, HIV encephalopathy was found
as an initial ADI in 101 (8.5%) of 1,175 AIDS cases reported to the Yucatan State
Department of Health between 1983 to 200127. This estimate is slightly higher than
reports in US. populations in which ADC represents only 3% of all initial ADIsmj.
Although stage of HIV infection is not reported, patients with neurological manifestations
had lower mean CD4 cell counts and higher viral load compared to patients without
neurological manifestations, suggesting advanced immunosuppression of these cases.
Neurological manifestations between the time period 1983-1992 were much higher
compared to 1993-2001. The authors interpret this finding as related to earlier diagnosis
of HIV during the latter time period, but this may also reflect increases in ART usage that
may have occurred. However, ART usage is not reported. HIV encephalopathy
diagnosis was “based on clinical data of patients with deterioration of motor and
behavioral cognitive functions, having ruled out opportunistic infections or neoplasms.”
How cognitive function was assessed or the criteria used to define cognitive impairment
are not reported in this article.

Neurological manifestations of HIV in Brazilian HIV/AIDS patients have also been
reported in several studies. In Rio de Janeiro, Brazil a retrospective review of 653
patients admitted to a university hospital between 1985 and 1989 reported 17 (2.6%)
cases of “subacute encephalitis”29. Diagnosis was based on clinical history, neurological
exam findings, and exclusion of other causes of HIV —related NCI through CT scan or
CSF analysis. Neuropsychiatric tests, if utilized, are not reported. Emilio Hibas Hospital

in S50 Paulo, Brazil participated in the WHO Neuropsychiatric AIDS study, the only

24

 

international study of neuropsychiatric manifestations of HIV to date72’80. Between
October 1990 and August 1991, every third patient attending medical or infectious
disease clinics were invited to undergo a comprehensive neuropsychological battery
covering 8 cognitive domains, as well as a neurological evaluation and a psychiatric
evaluation. HIV positive subjects were categorized as asymptomatic and symptomatic
and matched by gender, age, education, and HIV at—risk group to HIV negative subjects
from the same clinic. Global NP impairment, defined as scoring 22 SD below the mean
of the HIV negative group on 3 or more tests, was identified in 9.1% of asymptomatic
and 13% of symptomatic HIV patients. Although the AAN clinical criteria were not
available at the time of this study, Maj et al reported a 6.5% prevalence of dementia
applying both DSM-HI-R and ICD-10 criteria for dementia. All cases were identified
among symptomatic HIV patients. Among symptomatic HIV patients with a mean CD4
count of 350 (370) cells/mm3, 40.1% meet CDC criteria for AIDS, and 4.3% were
receiving Zidovudine. This study was Strengthened by the use of a comprehensive
neuropsychiatric battery piloted before study commencement and access to a wide range
of diagnostic capabilities as a result of access to neurologists, psychiatrists, experienced
neuropsychologists, access to a microbiological unit and neuroimaging (CT scan).
However, the study population may not have been reflective of the HIV population
generally as a result of Site selection and exclusion criteria, which included number
literacy. In 1995, Camara et al reported clinico-pathologic correlations of 154 AIDS
patients with evidence of CNS pathology at autopsy from a university hospital in Niteroi,
Brazil“. Among ten cases (6.5%) with HIV encephalitis identified on histopathologic

examination, retrospective review of clinic and in-patient records revealed dementia

25 ;

 

syndromes in nine patients (5.8%). Unfortunately, definitions for dementia are not
provided in this study. In the post-HAART era, a cross—sectional study of all adult HIV
patients admitted to Hospital Eduarado de Menezes, the main tertiary hospital for
HIV/AIDS patients in the large metropolitan area of Belo Horizonte, found HIV
dementia in 9 (2.2%) of 417 HIV cases“. This study utilized the AAN criteria for HIV-D
diagnosis, though like previous studies from this region, neuropsychological assessment
descriptions are lacking. This study focused on the prevalence of all neurological
manifestations of HIV. Therefore, descriptive data about HIV-D specifically are not
reported. However, amongst the patients with neurological conditions, the mean CD4
count was 93.9 (SD=110.6) suggesting these patients had advanced HIV disease.
Additionally, 76.1% had been on ART in the past 6 months including 79.3% on HAART,
10.8% on dual therapy, and in 9.9% there was no information regarding antiretroviral use
available.

A recent study investigated, the prevalence of a range of HIV-related NCI amongst
women attending two HIV clinics in Puerto Rico”. This study clearly delineates both the
neuropsychological assessment instruments used, and cut-off criteria for HIV —related
NCI. Another strength of this study was the use of normative data from 34 seronegative
Puerto Rican women of similar age and education to assess HIV—related NCI
performance. HIV dementia was diagnosed in 14 (28.6%), MCMD was diagnosed in 8
(16.3%), and asymptomatic NCI was apparent in 16 (32.7%) of 49 HIV positive women.
No significant differences were found between HIV-related NCI and CD4 cell count or

ART treatment.

26

 

2.4.2 Asia

The frequency of HIV-related NCI in Indian HIV/AIDS patients has varied in the
literature over time. Initial reports suggested a very low prevalence of the disorder. For
example, two studies of HIV patients attending neurological institutes reported HIV—D
prevalence estimates of less than or equal to 2%35'37. These studies suffered from lack of
normative data for HIV dementia diagnosis and possibly from referral bias. For example,
patients with HIV dementia have reduced survival and might not survive inpatient
admissions leading to underestimated prevalence rates in outpatient referral clinics.
Indeed, recent publications suggest that HIV-D prevalence in India approaches numbers
closer to those seen the US. and other developed countries. A retrospective study
investigating the neurological manifestations of disease amongst 1,606 HIV inpatients
admitted to an Indian hospital between 1993 and 2003, reported an HIV-D prevalence of
8.03%”. In 2006, Yepthomi et al compared 30 treatment-naive HIV patients with
median CD4 cell count 97 cells/mm3 (range 14-209) and 30 healthy controls matched for
age, education, and gender on neuropsychological test performance”. Significant
differences were identified between the groups on tests of verbal learning, motor speed,
attention, and executive function. Using a cut-off of 1.5 standard deviations below the
control group on at least two cognitive domains to define impairment, impairment was
present in 56% of the HIV patients. Although HIV-D prevalence was not able to be
measured in this study directly because of lack of other information (i.e. activities of
daily living), this study suggests that HIV-related NCI are also extremely prevalent in

treatment na'l've Indian patients with advanced HIV disease. Similarly, significant

27

 

differences in performance on the IHDS were identified between HIV patients with CD4
counts less than 200 cells/mm3 and seronegative controls in Pune, India“. The overall
prevalence of impairment on the IHDS was 35% amongst HIV patients, suggesting a high
prevalence of HIV -related NCI in this area of west India. In a study from South India
conducted between October 2003 and December 2004, 119 treatment na'l've HIV
outpatients attending two clinics in Bangalore were administered a comprehensive
neuropsychological battery83. These patients had a mean CD4 count of 396.8 cells/mm3
(SD 212.5), suggesting mild to moderate immunosuppression. The NP test battery
included tests of motor Speed, verbal fluency, working memory, planning, and verbal
learning and memory. This study was strengthened by the availability of age, gender,
and education adjusted normative data on 540 healthy subjects recruited from the same
community. A deficit on NP test items was defined as falling 1 standard deviation below
the mean score of the normative group, and further categorized as mild to moderate
(deficits on 4-6 of 12 items), moderate to severe (deficits on 7-9 of 12 items), and severe
(deficits on >9 of 12 items). Despite specifically excluding patients with neurological
manifestations (n=18), including patients with HIV-D (n=2), this study found mild to
moderate cognitive deficits in 57.2% of patients, and moderate to severe cognitive
deficits in 3.3%. No cases of severe cognitive deficits were identified, likely owing to the
study’s exclusion criteria.

Only three studies of HIV -D frequency in Asian countries outside of India were
identified. One of these studies identified 36 AIDS patients managed at two hospitals in
Shanghai, China between 1999 and 200333. Amongst this group, 6 (16.7%) cases of

“ADC” were diagnosed according to AAN criteria, although this nomenclature is not

28

recommended under the current AAN diagnostic guidelines. Therefore, it is possible that
AIDS patients with mild NCI may have not been included. Indeed, staging of cases
revealed only patients with moderate and severe ADC. MMSE and HDS evaluations
were used for neuropsychological assessment, but normative data on these scales in this
population are not provided. Nevertheless, MMSE scores ranged from 22 to 29 and.
suggested mild deficits in attention and short term memory in 50% of patients. HDS
scores ranged from 3-9, with a median HDS score of 6.17 out of 16 maximum points. All
ADC cases had CD4 counts below 200 cells/mm3. Only one case out of six had received
HAART prior to hospital admission.

A study from Bangkok, Thailand has reported the lowest frequency of HIV-D of all
studies from around the globe at less than 0.1%”. This study identified 2,261 HIV cases
admitted for the first time to medical wards, infectious disease wards, private wards, a
gynecology ward and a diarrhea] disease ward at a public hospital in Bangkok between
1993 and 1996. On-going chart reviews were conducted during each admission as part of
an increased internal HIV surveillance system. Patients in this study had evidence of
advanced HIV illness, with 1,553 (68.7%) cases presenting with an ADI. Of 509 patients
in which CD4 counts were available, 88% had CD4 counts <200 cells/mm3. Only two
cases of HIV encephalopathy were definitively diagnosed, although 19 additional cases
were suspected by clinicians but could not be confirmed because of lack of diagnostic
capabilities. No information is provided as to what criteria or tools are utilized for HIV-
related NCI diagnosis at this hospital. The very low prevalence in this study is likely
related to study design, which utilized chart reviews to identify the frequencies of all

HIV-related illnesses. Such a design, versus an active screening protocol for HIV-D, is

29

likely to result in under-identification of HIV-related NCI. This may be especially
pertinent given the severity of HIV disease in this population and high likelihood of
treatment for more conspicuous systemic illnesses overshadowing HIV-related N CI
identification.

The previously mentioned WHO Neuropsychiatric AIDS study was also conducted at
Chulalongkom University outpatient clinics in Bangkok, Thailand72’80. Global
neuropsychiatric impairments were common, with 5.7% and 22.1% impaired amongst
asymptomatic and asymptomatic HIV patients, respectively. However, no cases of
dementia were identified. This may be related to the mild HIV disease seen in this
population, as reflected by mean CD4 counts of 640 cells/mm3 for asymptomatic and 460
cells/mm3 for symptomatic HIV patients. Only 25.3% of patients meet criteria for AIDS
at this site.

2.4.3 Africa

An estimated 68% of all persons living with HIV/AIDS reside in SSA, yet only a
handful of studies have exarrrined HIV-related N CI in these populations, and populations
studied are likely highly selected and not necessarily representative of most people with
HIV in the region. Bélec et a1 conducted an investigation of the prevalence of
neurological complications among seropositive inpatients meeting the World Health
Organization Clinical Case Definition (W HOCCD) for AIDS in Bangui, Central African
Republic and reported a prevalence of severe dementia of 3.2%, but it is unclear what
clinical assessments were used for dementia diagnosis“. In Tanzania, 54% of hospital
patients were reported to have ADC“. However, this estimate included patients with

psychomotor retardation and confusion, as well as frank dementia. Both of these studies

30

were conducted before any standardized guidelines were developed for diagnosing HIV -
D and thus are difficult, if not erroneous, to compare. CSF analysis was conducted in the
study from Central African Republic, but not in the Tanzanian study, and neither study
had access to neuroimaging facilities for exclusion of other causes of cognitive
impairment in AIDS patients. In 1992, Perriéns et al found an 8.7% (95% CI 4-16%)
prevalence of possible HIV —D among hospital patients in Kinshasa, former Zaire using
the AAN criteria for diagnosis”. The WHO Neuropsychiatric AIDS study found
significant differences on numerous neuropsychological tests between symptomatic HIV
positive outpatients and HIV negative controls in Kinshasa, former Zaire and Nairobi,
Kenya53’84. The overall prevalence of dementia was estimated at between 4.4 and 6.9%
in both countries". A retrospective review of HIV/AIDS patients with neurological
disease attending an HIV clinic and medical in—patient ward at a Nigerian hospital
reported a 10.5% prevalence of AIDS Dementia Complex”. A clinical definition of
ADC was provided, but a description of how clinicians at this hospital reach this
diagnosis was not reported. Another study, conducted in 2004 at a university hospital in
Benin City, Nigeria, reported significant differences in mean verbal and nonverbal
memory NP scores between symptomatic HIV patients (p<0.001), but not asymptomatic
patients (p>0.05), and a group of seronegative controls”. Tests of psychomotor speed
and attention were significantly impaired in both groups compared to controls (both p-
values <0.001). All patients were ART-naive in this study. This study used a
computerized neuropsychological test battery, which consisted of tests of reaction time,
memory recognition, visual scanning, and abstraction tests. The authors are quick to

point out that computer literacy is not necessary for administration of this test, however

31

clearly accessibility and need for maintenance of testing equipment make computerized
NP testing a poor choice for routine use in resource-limited settings. A group of controls
were identified through HIV -negative hospital staff members for comparison purposes
that were “randomly” selected, although the randomization procedures for healthy
controls, and patients for that matter, are not reported. This study did not attempt to
exclude opportunistic CNS infections as causes for impairment. The prevalence rates of
neuropsychiatric testing abnormalities, MCMD, or HIV-D are not reported.

Pierotti et al conducted a study at a hospital in northern Uganda, the only study based
in an rural African setting to date“. Amongst 44 antiretroviral naive HIV patients with
neurological impairments admitted to the internal medicine ward over a seven month
period, seven ( 15.9%) cases “consistent with ADC” were identified. Cases were
identified through neurological exam, but neuropsychiatric measures are not reported.
Also, this study did not have access to neuroimaging or CD4 counts, but CSF analysis
was performed for all patients. Another Ugandan study of 78 ambulatory HIV patients in
Kampala, Uganda has reported a very high prevalence of HIV-related N CI, 31% for mild
HIV dementia or worse and 47% for MCMD. In this study, HIV-D and MCMD
diagnoses were based on comparison of mean scores on NP tests between HIV patients
(28% of whom were receiving HAART) and a seronegative group from the same clinic
base, followed by assessment by two primary physician examiners, a HIV neurologist,
and an HIV neuropsychologist.

In 2004, a study of a well-characterized cohort of 1400 Ethiopian textile and
sugarcane factory workers attending free clinics in two communities outside Addis

Ababa failed to find differences between 73 ART-naive HIV positive and 87 HIV

32

 

negative clinic attendees on total or sub-item IHDS scores, timed gait, grooved pegboard,
or verbal fluency tests. In this study, only timed dominant-hand finger tapping
performance was significantly different between HIV positive and HIV negative groups
(p = 0.01). The authors point to the potential differential impact of clade C and C’ virus
on HIV-related neurological manifestations as a potential explanation for the very low
prevalence of NP deficits in this Ethiopian cohort. However, 61 (86%) patients reported
full time working status, suggesting relatively mild HIV illness. Indeed, markers of
immunosuppression indicated only mild to moderate HIV illness with CD4 counts
ranging from 168 to 451 (median=260) cells/mm3 and mean HIV viral load at 4.3 log
copies/ml. Functional abilities of the cohort were also extremely good with a mean
Karnofsky score of 95.7 out of a maximum score of 100 among HIV patients. Another
important concern is that receiving free healthcare based on factory employment may
lead to patients with more severe systemic illness or NP manifestations seeking care
elsewhere for fear of job insecurity, although this study captured greater than 70% of all
HIV cases in these communities.

In a cross-sectional study of black South African patients admitted to inpatient
medical wards at Helen Joseph Hospital in Johannesburg, HIV -D was the most common
neurological disorder with 108 (38%) cases of HIV-D diagnosed in 506 consecutive HIV
positive patients“. Decreasing CD4 count correlated with HIV-D diagnoses, with 108
cases having CD4 below 50 cells/mm3, 38 between 51 and 100 cells/mm, 29 between 101
and 200 cells/mm3, 12 between 201 and 500 cells/mm3, and in 5 cases CD4 count was
>500 cells/mm3 (p for trend not reported). None of the patients had received ART. Other

causes of cognitive impairment were excluded through CSF analysis and CT scans. The

33

MMSE was utilized for neuropsychological assessment, though it is unclear how
cognitive impairment was defined or if normative data was available for this screening
instrument in South Africa. In addition, the AAN criteria were utilized for diagnosis, but
no cases of MCMD were reported. It is therefore unclear if mild cognitive impairments
were excluded from this analysis or simply included in the HIV-D diagnoses.
2.4.4 Summary

When excluding studies conducted before any standardized guidelines for HIV -D
diagnosis had been published, the prevalence estimates of most studies appears within the

27.28.30.31. 3. 4,36, . .
d 3 3 87, and are srrrnlar to

range from 8.5 to 38% in most regions of the worl
prevalence estimates from the US. in the pre-HAART era (i.e. 20-30%)3. Notable
exceptions are Brazil and Thailand, which have reported significantly lower
estimate326’29'32’8'. It is unclear if these are true differences related to the
neuropathogenesis of HIV in these countries or reflect differences in diagnosis, sampling,
or other methodological differences. Although comparisons of these studies are difficult
at best, HIV-related NCI appear to be common amongst HIV patients in low and middle
income countries. In addition, the only cross-cultural investigation, the WHO
neuropsychiatric AIDS study, found similar prevalences of global neuropsychiatric
impairments in asymptomatic and symptomatic HIV patients in Brazil, former Zaire,
Kenya, and Thailand72'80. Trends similar to those seen in developed countries following

availability of HAART may also be seen, highlighting the need for improved

identification and surveillance of HIV -D in resource poor countries.

34

2.5 Importance of Identifying HIV-D in sub-Saharan Africa

Most of the burden of HIV/AIDS lies outside the developed world. An estimated 22.5
million (68%) of the 33.2 million persons infected with HIV globally are found in SSA38.
It is estimated that 2.8 million new HIV infections occurred in SSA in 2006, and
accounted for more than 2 million deaths that same year.

The burden of HIV/AIDS in SSA has lead to a dramatic increase in funding for ART
and infrastructures to support ART distribution. In 2003, it was estimated that less than
100,000 Persons Living With AIDS (PLWAS) in Africa had access to these life saving
treatments”. By 2005, approximately 810,000 PLWAS were receiving ART thanks to
efforts by the WHO ‘3 by 5’ initiative. Improving access to ART is likely to significantly
improve mortality of HIV-related infections. However, morbidity from HIV infection
may increase as patients gain systemic control, but live to bear the more chronic Stigmata
of HIV infection, such as HIV-related NCI. Our ability to understand and respond to
these changes is essential for planning and program development, and for allocation of
preciously limited resources. Sadly, the impact ART may have on the prevalence of
neurological disease and other chronic health conditions related to HIV infection were
more or less omitted from the long, political discourses that preceded the widespread
entry of ART into African healthcare systems. The lack of background knowledge we
have about these conditions in sub-Saharan Africa and other resource-limited settings
makes it difficult to anticipate the impact ART will have. However, there are several
areas of concern that speak to the need for increased clinical identification and improved

surveillance of HIV -related NCI in such countries.

35

2.5.1 Initiation of Antiretroviral Therapy

The limited supply and high demand for ART in sub-Saharan Africa and other
resource-limited settings has resulted in clinical guidelines that support initiation in
PLWAS who are in the most need of ART treatment. The WHO recommends ART
initiation in patients who 1) have a WHO stage IV disease, regardless of CD4 count, 2)
have WHO Stage III disease and CD4 count <350 cells/mm3, or 3) patients at any WHO
stage with CD4 count <200 cells/mm388. HIV-D is considered a stage Iv WHO illness
and is an indication for ART inflation”. It is estimated that HIV-D occurs as the initial

1’25, although estimates as high

manifestation of HIV/AIDS in approximately 3% of cases
as 8.5% were seen in a Mexican population”. However, clinical officers, who provide
most of the care in ART clinics in SSA38’65, have little training in identifying HIV-D and
it is likely that the condition generally goes unrecognized. Changes in cognition,
behavior, and personality may be wrongly attributed to depression, other psychiatric
disease or even witchcraft by healthcare workers and family members. Thus, patients
may receive treatment for psychiatric disorders or seek help from traditional healers,
leading to delays in ART initiation. Such delays may allow progression of CNS injury
and result in reduced efficacy of ART to reverse neurocognitive deficits. A prospective
cohort of HAART—treated patients with HIV -related NCI identified through the National
Institute of Infectious Diseases Lazzaro Spallanzani in Rome, Italy found that the best
predictor of reversibility of NP deficits was baseline neuropsychiatric test peformance48.
After controlling for gender, education, Hepatitis C-positive serology, and CD4 count,

poor performance on the baseline composite NP score was associated with a 3.07 (1.54-

6.08, p=0.001) increased odds of having persistent NP deficits after HAART initiation.

36

This finding suggests that early HAART treatment in patients with HIV-related
N CI may be essential to avoid irreversible neurological damage.
2.5.2 Optimizing Treatment for Patients with HIV -related N CI

HAART appears to be at least partially effective in improving HIV -related NCI, but
there remains little empirical evidence or consensus among experts on how to optimize
treatment. The main goal of treatment is the same for all HIV patients (i.e. reduce the
systemic burden of HIV as measured by HIV viral load and CD4 count). The importance
of CNS-penetrant drugs, defined as drugs with cerebrospinal fluid (CSF) concentrations
that exceed the level necessary to inhibit replication of HIV4, is largely unknown. In a
study among Johns Hopkins University HIV neurology clinic attendees, no difference
was found between patients receiving HAART regimens with single versus multiple
CNS-penetrant drugs on tests of motor and psychomotor function”. In contrast, another
study found that HAART regimens containing two CNS-penetrant drugs were associated
with transient improvements in motor and psychomotor speed in individuals with
moderately advanced HIV illness9'. Cysique et al found that amongst a group of
impaired patients (i.e. scoring 2 SD below the mean on 2 NP measures), those receiving
HAART with 3 or more CNS-penetrant drugs performed significantly better on tests of
memory learning (p = 0.04), short-term recall (p = 0.03), and long-term recall (p = 0.02)
than patients with other HAART regimens”.

It is generally agreed that patients with HIV-related NCI should receive HAART
therapies with good CNS penetration“. In the US, this has become an important
consideration among HIV/AIDS healthcare providers in the planning and selection of

HAART regimens“. This individualized care is in stark contrast to the situation in SSA,

37

where cost-saving, simplified regimens considered safe and effective in most patients are
emphasized. Triomune, containing the antiretroviral agents stavudine, larnivudine, and
nevirapine, is the only regimen available in many public ART clinics in sub-Saharan
Africa”. Stavudine and nevirapine are considered CNS-active drugs, but the efficacy of
a two CNS-penetrant drug cocktail in controlling HIV replication in the CNS has not
been established. Thus, limited access to alternate treatments may differentially impact
HIV -related NCI in SSA.
2.5.3 Communication of Medical Information and Medication Adherence

AS mentioned earlier, symptoms of HIV-D may be subtle and go unrecognized or
confused with depression by healthcare workers. This Speaks to the need for healthcare
workers to adequately assess cognition in their patients in order to ensure that proper
medical information is accurately conveyed to the patient, or a caregiver if needed. This
is of great importance for HIV/AIDS patients, who may be on complicated medication
regimens such as ART, anti-tuberculosis medications, other antimicrobials, psychotropic
drugs, and others. Medication adherence is known to be compromised in cognitively
impaired HIV patients in the US$134. Hinkin et al found that cognitive impairment in
HIV patients increased the odds of poor adherence to HAART regimens by 2.5 (95% CI
1.19-5.35) compared to HIV patients without cognitive impairments 1. Complexity of
treatment regimen was also found to interact with cognitive impairment to further
increase the likelihood of poor adherence”. Therefore, HIV -D patients may be at risk for
development of drug resistance, ART regimen failure, and death if not properly assessed.
Adherence to ART must exceed 95% to prevent development of drug-resistant virus and

subsequent regimen failure”. Improved ability to diagnose HIV-D patients would allow

38

healthcare workers in SSA to identify appropriate caregivers to administer ART, or at the
very least Simplify regimen complexity. The development of drug-resistant HIV has far
reaching public health implications, and could impact the success of large-scale ART
intervention efforts.
2.5.4 Quality of Life

As ART access becomes more equitable in developing countries, quality of life
measures will become increasingly important. Several studies have found an
association with decreased quality of life in patients with HIV-D, even after adjustment
for other confounding factors such as depression and other comorbid medical
condition595'97. In an Italian HIV cohort, Tozzi et al. found that cognitively impaired
patients reported significantly lower quality of life measures (p<0.05) in all domains
tested, including social functioning, health distress, and general health perceptions,
among othersg6. Identification of HIV -related NCI facilitates the ability of clinicians to
plan and implement interventions that can improve quality of life and functional abilities,

including HAART initiation. HAART has been shown to reverse or stabilize NCI in

98-100
D 101

some patients with HIV- and can improve patient quality of life . Patient quality
of life may also be impacted by stigma surrounding HIV /AIDS. HIV-related NCI
patients may be more likely to experience abuse, neglect, and abandonment. In Zambia,
patients with epilepsy, a highly stigmatized chronic health condition, are more likely to
experience physical violence and rape than persons with non-stigmatized chronic health
conditions, and were found to be significantly disadvantaged both socially and

102

economically . Although formal investigations of stigma and HIV -D have not been

39

formally carried out, HIV -related chronic health conditions are likely to be highly
stigmatized conditions as well.
2.5.5 Caregiver Burden

AS cognitive functioning progressively worsens, HIV —D patients often need 24 hour
care to meet their daily needs and maintain safety. In a case series of 11 caregivers of
patients with HIV -D, caregivers reported similar stresses and concerns of caregivers of
patients with other types of dementia. Some common themes reported were feelings of
social isolation, difficulty coping with problematic behaviors, and practical problems of

‘03. However, there are also important differences

care (i.e. eating, bathing, toileting)
between caregiving for HIV-D patients and patients with other dementias. For example,
the onset of dementia often occurs during young or middle adulthood. HIV-D patients
are more likely to have financial or family responsibilities (i.e. dependent children at
home). In this circumstance, HIV-D caregivers must remedy the financial or child care
consequences of losing a spouse or other member of their household, while at the same
time attending to the needs of an increasingly dependent loved-onem. Caregivers of
HIV-D patients must also cope not only with dementia, but also with societal attitudes
and stigma surrounding HIV/AIDS. This may strain social networks for caregivers,
leading to increased feelings of despair and isolation. Further, caregivers of HIV-D
patients may be living with HIV themselves, increasing their own anxieties about
dementia and dying'os. Other important and unique cultural factors may be important in
SSA and other resource-limited countries, but the impact of HIV-relate NCI on caregiver

burden has never been formally assessed in these environments. Still, many of these

factors are likely to be important in such countries, and the clinical identification of HIV-

40

D would allow clinicians the opportunity to provide advice and prognostic information
about the disorder, including how to handle difficult behaviors such as combativeness or
agitation. In addition, HBC resources could be prioritized to these patients to allow
caregiver respite.
2.5.6 Impact on Employment

At a macroeconomic level, the HIV /AIDS epidemic has led to a dramatic decrease in
economic growth and development in African countries, reducing national growth rates
to 2-4% across African”. While access to ART is likely to reduce mortality and
morbidity of HIV illness, increasing HIV -D prevalence may prevent employee (re)entry
or reduce worker productivity. Patients with HIV -related NCI are more likely to be
unemployed than cognitively normal HIV patients with similar HIV stage in the U.S50"07.
When comparing employment rates in HIV patients with NCI to unimpaired HIV
patients, Heaton et al found that HIV patients with NCI were three times more likely to
be unemployed (p: 0.0001)50"07. Unemployment rates remained higher even when
considering patients with only mild impairment (p<0.05). In addition, HIV patients with
NCI who were employed were significantly more likely to complain of difficulty with
work tasks (p<0.001). Since ART may improve neuropsychological function or stabilize
HIV—related NCI, identification of HIV -D may allow some patients to return to work and
improve productivity amongst those who are employed.
2.5.7 Understanding HIV-D Pathogenesis

The opportunity to identify HIV-D patients in regions with non-B HIV clades (or
subtypes), the most common clade found in the US, Europe, and Australia would

provide valuable information about HIV-D pathogenesis and may lead to further

41

elucidation of clinical treatments. The infidelity of the HIV reverse transcriptase enzyme
and lack of a proof-reading mechanism has led to diversity of HIV in different world
regionslog. There are three phylogenetic groups, including M (major), O (outlier), and N
(new). Groups 0 and N are found only in central Africa, especially Cameroon. Group
M accounts for roughly 90% of HIV infections globally and is subdivided further into 9
subtypes based on phylogenetic classifications of variations in HIV envelope nucleotide
sequences. Group M subtypes (or clades) include A-D, F-H, J and K. Some of these
subtypes have been shown to preferentially influence disease progression and death and
thus may impact HIV-related disease epidemiology, including HIV -re1ated neurological
complications. In Rakai, Uganda, a retrospective study found that 59% of patients with
clade D progressed or died compared with only 29% of patients with clade A who
progressed (and none died) during a 5 .1 year mean follow-up period (p<0.01)'09. Similar
findings were seen among a group of pregnant mothers in Dar es Salaam, Tanzania, with
HIV positive women with clade D experiencing more rapid disease progression to death
(HR 2.27, 95% CI 1.46-3.52) compared to HIV positive women with clade A, even after
adjusting for viral load and CD4 cell count1 '0.

Animal retrovirus models have also shown that different HIV strains can preferentially
induce brain pathologyl I I" '6. Further support for the importance of clade diversity on
neurological outcomes comes from ex vivo findings indicating decreased viral fitness of
Clade C virus in macrophages relative to Clade B isolatesm. Since macrophages are a
predominant HIV -infected cell within the central nervous system, some authors have
extended this finding to explain the regional differences in HIV -related NCI prevalence

between the previously mentioned Ethiopian cohort (where clade C predominates) and

42

Ugandan cohort (where clades A and D predominate)108. However, recent studies from
South India, also a Clade C predominant region, have indicated that HIV -related NCI are
extremely common manifestations among Indian HIV/AIDS patients, with estimates
ranging from 35% to 60.5%34‘83. These conflicting findings are likely to result from
differences in population sampling, selection of control subjects for normative data, and
neuropsychological assessment batteries utilized. No studies have formally investigated
the impact of HIV clade on HIV-related neurological manifestations. Further research
would provide valuable information about HIV-D pathogenesis and may lead to further

elucidation of clinical treatments.

43

BERKWFD an ”magma“. :8 @2352 8: NZ $6336 :8 cohavom um
.93” ”255m 6:30:52 a @5986 o>E=moooSe=
cosioommmaag Ao:u_ocoeo::EE_ ENE—E :8 «recto ESEEE we :oanficcgo Ewe—2:39:02 ._n Ho 550:0 Soc @833:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m m m m m 2:0an 32820 :8 awe—owe 550 02
m m M m m 55:: _ 3:2 3 55:5
m x m m m mEfiB :8 305.5638 5305.6
Z m Z m z 2e>em 9 ounce—cog 3:35:35 15¢.
Z m m Z mD 2:2 ”EQEEEQEH AQ<
M Z Z Z Z 0:02 ”EeEEEEZ 40¢.
Z Z Z Z M bsanE: 3.5333 :0 :92: nocte>
m m mD m m: 555m AZ Ego—m 298 9 282.02
a m m3 Z mD bto>om m2 .320 2980:: 9 22
+N +N mD +N +~ RE“ HE .8888 me u
m m m m m Essences 63%
mv +m mD +m mD Siam—:8 o>E:m8 automortom Mo 63:52
5:33 52581: .owSEo $35209 G ”:oEEEOOQE Am
wEoEoBE
@0303 Cd ”boEoE REEF: Am ”mEBBm 3:2: 3
”:ocmbcoocoflgufig notafif
no N “$2 2
Z Z M Z m beam; 3 5:22:55: EcogsonhoaoEBEEon
~Z< DZE QED—2 0-2: 0&2: Q55 Z<<
DMZ: DMZ: Z<<
82:85 CZ cos—ox->_: 550 0-23 I
LUZ @832

-2: w:_mo:mw5 :8 85—0220 :850 ganged 383239302 2: t5 awe—9:52 we iguana 585:5. no :OmEmEoU A 035.

 

803—006 530: 3808 *0 :owfiaagum

0:32:20

.52 8053, HUZ 003—0:->~I :0“— 080 :0 808308 02
005 8:80:33. 90 £23m

80:3:
053000800103
030530:

 

80800 :38 3 80:2:mom 00 800030 30:20:05

30

.mmu FD .22 0.5 52 :0 000000 :050 :0 :03388 .50 0002
5:02.? 0:: 8080203: _000 058:0»30802

:58: 5.20.. 2200:8802:

80:00:84 00883—

 

 

gaming

>080 :_ 3805808 :0 $0808-30: Hg: 00 080355
080358 52 003—0:

-23 8088_ @0680 0:23:88 8 $05: a: 90 bt0>0m
8:08:58 3000 0335030080: 00 008038058800
808.308 0>E=m00 00 835.00 30:00:00 02

0:23:85 808:0 Z<<

08: :0>0 32088.80 52 00:20:55 8 0:08:83

 

0:080:85 38983

 

 

008800 08005 0622 0:0 33 8 ~02 003.0%-22 .«0 82380002: 000305 00 35.339800 0:: 030—30; 9 83.83 .N 0509

45

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

08.0 535:5 33.5
53673085 :25 8:08.. 8: o. 82:60 8:08: +33 :85 389
03d Q55 53.3. 525.: Cw": 88:5wa E05888 .0808 Z<< mo 33.: .0:o:8m-mmo:u 6.55:2: 20m 0:035
m 8:: :0 2050
93.0 m 0-2: 8:62: 020 5? a: 82 38:
m2: 0::0E0Q vfln: 00:20:00 8: £852: 825.0 :0 03:00 03008850”: :85 ._o._0._z «8:80
8000 o. :0 mN :0 £888 .21 320: 020:: :8
00m: 0::0E0Q 0:G:>oEN :o 3: Om NM wctonvm 8 85.8: 52 2320 >H: 08899:?
0? .mm 2: 0:089:53 850:5: 0>E:w00 w 3688: 0.8. m2 :8 ouanaan:
CZ .820 :0 03.: ST: 8:008 0:850: 9.09 )3?sz mo 3:; 35:80:85 :85 .220: 0mm $3: .82
:0: 3:8: 322:: 28:
3:280:00 8:608 o. 83:60 9: Z: :85 820m
058:5 0:0: mmcn: 3:20:00 5: .2883: 325.0 .6 :38 0>:00%o:0m .8253. 0: 2m 4:283:
g
000% 9-2: :2 g cm NM 5 £2 mm
0002 QED—2 :o :00E 05 320: Gm _N u .:0E.:0:E_ 025.0 >5
0:th a 050:5: 0>E:w8 n 8:80 3:25me 95 :0 00:00:00 8:83 383
ozafianmmxx «03 av”: AZ w .52 SEE: :8 Z<<-E 5:; Z<< mo 330 35580085 8E 28:: 05.03
8;. 88am ”NHEEU
a»: .23: $88
wt; 2 8:80: £D< REE 858%
§m.w WE: 8:08: .o: n: 858% 5: £853: 3255 Mo :28 0>so0awohom 08% 5:00.; -0500
22:05 :02 28:02 888
.0 :08 an mn=<\>E 8:82 08>;—
§fiw UQ< 895:0: .o: o3 u: 858% .0: £7,885 .0255 .«o :28 03:88:53: 55 02x02 -8850
20: mo: 50E
08?: 858% 03. 9 80:68 9: mag 00:82 $33
UQ<E->E 0:02 on _ n: 83:88 8: .8286 306:0 mo 03:; 3:283:85 .55 8:82 EFF—H
g
380:. 0N6 :ocszmom .5 .00:_>o:m\0§m C003
0:58m >M< w:_>_08: 00 0.:Emw OZ :8 0:050 38:85 :8 :mioa 33m 55 55=<

 

 

 

 

 

 

 

 

 

 

00:32 ”ngoo

88:80 0% mot0E< E81— :2 0M 2.53

 

 

:2m0m 2:03 03 8:880 0:505 0:022 :50 3Q.— E 863m CZ 88.2-2: ho meEzm .m 030,—.

 

46

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

088:0 80:08:: 0: 8 88m:
:0 _ .ov m->=.: 00880: :0: ENN M: 005000: :02 +>E :0 80:00 08000880”: :0:w::m _3=0:B:m::0=m::.b
0000:00: :80: o. :0 mN :0 0:00:00
:0:0E00 02 -2: 30.0: Om NM w:::00m m: 00::00 2:00.388 Em 00:88:80?
03.3 :02 0320 .m:_:::00 w :0 E88 AZ 0:: 00:59:83:
:02 050.0 0:0: 8 _ u: 08080 :0:0E00 0: -Q0: 2-202%: :0 >008 _::0000m-mm80 xoxwgm $00: .32
0::=::.:. u :=00
a:
::0Emm0mm: :08: 0:3 :: 8:003 mo?
00:0— 0D< aid emu: :Z :0: mo: 0:: wwEE .:_:0::0 Z<< :0 >02: _::0000m-mm80 _::w:::m A808 :3
::_:0 ”#08000
92:
00me E0: 80. .3: :0 £8300 0>0::::0: :0 ::0E:_::E: _:0_w0_0::0:
86:00 :0 5:08: 320: Om _A m: 005:00 8.0::3 3:00:98 20::
0>0_:w00 0:0: 2 _n: 0:080:90: ESQ n:Z 0>_m:0:0:0E00 Z: :0 3:8 030008-880 :Som .0:0_:w::m $88 8000
802800 0>0Ew00 82:00 _:::0:0:
00% 03: :0 000:» comramEoo 30.0: Om >=.: >580. :: 02 30.0:
0:0E:_:0E: m: m: 00:000 20:80:08: $05000: 85:00 :00 :30 :03 m8: :_0::
0>_::w00 0:0: emu: 855:8 00:: 8:88 :2 00805055 >5 :0 x008 _::0000m-mm80 :.=0m ._:::0:0 68$ E559;
0858:: 5380:
00mm Q->=.: .: 08v 3000 30 :03 mi
03:30: 008000: :0: wvn: 0.:0m :0:0E0O >5 _::00:::0::: >5 :0 3:8 _:_:0000m-mm80 20:: 803 6:0: seem: 3003:
8:08: a
008.: D->=: 008000: :0: coo—n: 00:80:00 :0: .m_m0:m:_0 _:0_:=0 Z: :0 80:00 03808003: 30:: ::=0m 38$ 20.:
80:00:88
::0m :0 :_> 33:0 :0::0 :0 005330 _:0:w0_8=0: :0: 088:0: ma :_0:_
00m.— Q->=.: 008000: :0: 82": 0:: mmEE :0 08:: w:m0:w::0 _:0_::0 >5 :0 >008 _::0000m-wm80 :50m .0:0_:m::m Coca: 30:3
0am] 0_::0 0:800: 50:08: :8: 20:: 8084
0% :0:0::00 008000: :0: ON M: 088050 :0: .m:m0:w::0 _:0_:=0 +2: :0 80:00 05000853— :Som .0:0_:w::m ::0:::0:m0:w
3.30m 380:: 08m :02 :0: :8080 0:80: :5 :00:_: 0: 0% : E09 03m gm A;
>m< 0EE:m :0 .00::>8n<0§m
m:_>_000: 0:: 5:0
a
3?: ”:0: 0M 0803

 

A0088: m 2:5.

47

 

v.33 328:5

 

 

 

 

 

 

 

 

 

 

 

 

3062: :5 3::o a:
$65.: a 832: $89
$2: 0% cow—one: 8: Sn u: 333% 3.3.850 Ho: .mtouto 32:20 mo :28 egoamobom atoflz €992?
338: .83
8 8:08:38. 323—23:
858% as, 35:8 Z: 883
$3.2 UQ< 0:0: Sun: 83.850 8: 6235830 33%—8:02 he saw 3538-380 .6ng 505.52 9205
§o.o-m.n
«5:0an 8:52 w? u:
52 Boo—O ”933— ”umcox 98. 3 ho mm :0 $828 $5 322. 8:23:50
9:56-»; mace—5Q cg": Om «N groom ma vac—ho: H02 1320 a: 26889:? i=3—
”ofiuwm ~02 ”BEN .2350: 33:30 m 9:383 33: AZ :5 ogfioEEzma .5832 :5 BEN €33
.320 ”BEN Beam 0:0: hog—om .atoEo «35:8: EACH 2-5-2ma .8 33m 35:03-380 Bar—om dwggg “:2
$2 :2: 3:8; 35
mad 9202 :35 8 35:35 a: Egon .3235— Ammo:
$3 52 0:0: vo _n: 358% .o: 38. m Z .atoEu Z<< mo 33m 3:288:85 \SASM .5882 got—om
gum—=23
owvm 553532 :5 £2.52: .boEoE £28338 .838:
5889.0me :28: :5 3:82 8: was AZ 533 8 95:35 mm: «ESE; Aawo:
§vm UQ< 05: com": «cs—:0: n :8. AZ _A E EoEEKE— mo 35% 35:03-380 .o:oN E0532 3230:
5&8: :35
3 ma? :8 DOUG—$5
mEBoE 35:85 23:3“ 58$? Ema:
§N.m «3:0an 0:0: no H: @2383 8_Eofio 8: .35th Rafi—U .8 33m 38:08-380 35:00 ._:w§m 8:5
382:
>m< oflm :omwom 8 C35
3:.QO @368: 05 295% 62 5.. «12:0 usagwfio :ocflsmom a. :wfiofl 33m .oo:_>o:m\38m $30 55:2

 

 

2v aoE< ”:2me 2.83

 

 

3:8; _

.5253 m 2%...

48

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3: mo; 53:. E 2:23 5 SE< 2.83
fiwm Q43: 25: can": MmEE wENES .atofio Z<< +>=.~ mo 33m 3553-380 Enom $538523 :52
QEUE 809.33 £53 .353 ASONV

.0 Q55 02 0:0: 00 _ n: 858% 52 Dose mo 3:; .mcouoomémouo «£25m Eot=U

mEanv
«Sm Q->=.~ 03::on w 953. E2535 25.0 :35 36:28 3:239 €88
9A.: QED—2 §w~ mun: EuEmmommm AZ 0 .utoEo Z<< +2: ..o 355 Ecozoomémohu «9.me .EanEaM mac?
BE >m< am a 93
3.30% 9.568.. Q» 035% 52 he «FEB osmocwma coca—anon can cwfiofl 32m .oo:_>o.&\o§m :90 553x

 

 

5 8:2 a2 oz 253

 

 

 

.9558 m 2an

49

CHAPTER 3: METHODS

3.1 Setting
3.1.1 Overview of Zambia

The Republic of Zambia is a landlocked country in southern Africa bordered by seven
countries including Democratic Republic of Congo in the north; Tanzania, Malawi and
Mozambique to the east; Zimbabwe, Botswana, and Namibia to the south; and Angola to
the west (Figure 1). The country is slightly larger in size than the state of Texas. Zambia
has a population of 11.5 million, of which approximately 44% reside in urban centers and
1.2 million (10%) reside in the capital city, Lusaka] ‘3. English is the official language of
Zambia, but there are over 70 local languages and dialects. Chibemba is the most widely
spoken language in Zambia (30.1%), but Chinyanja was identified as the predominant
language spoken at home in Lusaka (52.8%) during the 2000 national census and is
considered the lingua franca by most1 '9.

Zambia ranks among the poorest countries in the world. According to the United
Nations Development Programme’s Human Development Report 2006, 75.8% of the
population live on less than $1.00 a day120 and Zambia ranks 87th out of 102 developing
countries on the Human Poverty Index.‘

Many factors have contributed to Zambia’s economic struggles including high
unemployment, falling copper prices, foreign debt, droughts and food insecurity, and the

HIV/AIDS epidemic. The HIV/AIDS epidemic continues to ravage Zambia, with an

 

I. The Human Poverty Index is calculated based on the following indicators: probability at birth of not
surviving to age 40, adult literacy rate, population without sustainable access to an improved water source,
children underweight for age, population below income poverty line

50

estimated 1.1-1.2 million HIV infected individuals and an HIV prevalence between 13.3
and 20% in 2004”".
3.1.2 Kalingalinga

This study took place in Kalingalinga, a shantytown settlement wedged between the
University of Zambia and the Government Airport on the west-side of Lusaka, on the
south side of the main Great East Road (Figure 2). It is a densely populated area with
extreme poverty, poor sanitation, and limited access to healthcare. The official population
of Kalingalinga is unknown, but was widely quoted by hospice administration staff to
number around 100,000 during the study period. The HIV seroprevalence in
Kalingalinga, as estimated from antenatal clinic attendees, was reported at 26.5% in

2002‘”.

3.2 Population

At the time of this study (2004), Our Lady’s Hospice was a newly opened facility with
a blossoming community outreach program in Kalingalinga and surrounding areas.
Patients were identified through key community members employed by the hospice, who
also resided within the communities. Once identified, patients were evaluated and
followed at home by a team of hospice healthcare workers. Depending on need and
availability of beds, patients could then be referred to the hospice for in-patient
admission. The hospice administration placed emphasis on home-based care and
generally restricted in-patient admissions to five days or less. However, longer stays
were common. During the study period, the hospice recruited over 500 hundred patients

within Kalingalinga and three surrounding communities, and had a 24 inpatient

51

admission capacity. The hospice was dependent upon donor funding and governmental
support for medications and supplies, and at the time of this study had no available

laboratory or radiological equipment on site.

3.3 Interview Procedures

Each morning, a list of admissions occurring the evening before was reviewed for
eligibility. All patients aged 18 or older, and admitted for the first time to the in-patient
ward between January 21, 2004 and March 22, 2004, were invited to participate.
Readmissions were not evaluated because of the high turnover of patients and the heavy
clinical responsibilities of study interviewers. Other exclusion criteria included a past
medical history of a chronic neurologic disorder, psychiatric disorder, alcohol abuse, or a
physical deficit or severe medical illness that would interfere with study performance.
After informed consent procedures, patients underwent a structured interview to assess
demographic information, medical history, history of changes in memory and thinking,
and cognitive and neurological status. A medical chart review was also undertaken and
all information regarding medical history, medications, and available laboratory testing
was recorded. No laboratory testing was done as part of this study. In cases in which a
patient was unable to provide informed consent, a family member or friend provided
informed consent, and answered questions on behalf of the patient regarding
demographic characteristics and medical history. The interview was conducted in the
patient’s or advocate’s language of choice between Chinyanja, the most prevalent
vernacular language in Lusaka, and English, the official language and the language of

instruction in schools. Several hospice volunteers fluent in these languages were trained

52

to conduct the interview procedures. An additional group of healthy individuals also
underwent the neuropsychiatric assessment to provide normative data. These individuals
were family members or friends of patients attending the outpatient clinics at the hospice
and other interested members of the community. All interviews were conducted by
clinical officers and hospice volunteers who underwent training in administration of the
standardized study instrument. The author of this thesis was present during all interview

procedures.

3.4 Language and Cultural Considerations
3.4.1 Translation of Study Instruments

A group of translators with extensive experience in translating clinical research
instruments were employed to translate consent documents and interview instruments
into Chinyanj a. This group has been previously employed by the WHO for translation of
research instruments. A team of local professionals including a psychiatrist, internists,
and medical students, among others, were also asked for their input. A linguist then
translated the documents back into English. Following back-translation, discrepancies
were assessed and corrected and documents were then carried into the community to fine-
tune the documents. This procedure was repeated until content was preserved and
vocabulary was easily understandable to all parties.
3.4.2 Neuropsychological Assessment

No instruments were formally available for the assessment of cognition in Zambian
HIV patients. Therefore, several neuropsychological assessment instruments were

chosen, including the MMSE54, the HD862, and Color Trails 1 and 2122. These

53

instruments were chosen because of their familiarity to healthcare workers (MMSE) and
sensitivity to the cognitive deficits seen in HIV-related NCI (HDS and Color Trails 1). In
addition, these instruments had the advantage of simplicity of application and short time
for administration, which was considered of fundamental importance for this population
of terminally ill patients, who would be unlikely to tolerate a lengthy and complex
examination. Local physicians, including a psychiatrist and two internists, medical
students, and clinical officers were consulted regarding the cross-cultural applicability
and appropriateness of the instruments. Items of concern were identified through group
meetings, piloting in the community, and finally during data collection for the study.
Table 4 summarizes the neuropsychological instruments and the respective cognitive

domains assessed following modifications to the instruments for cultural appropriateness.

3.5 HIV/AIDS Diagnosis

Patients were interviewed regarding knowledge of their HIV /AIDS status. It was
anticipated that very few patients would be aware of their HIV status, and so patients
were also assessed using the World Health Organization Clinical Case Definition
(W HOCCD) for AIDS123 using a previously described al gorithm.124 The sensitivity and

124. Patients

specificity of this algorithm has been reported at 71% and 80% respectively
who did not meet the WHOCCD for AIDS were excluded from the analysis. All patients
who meet WHOCCD for AIDS and consented to participation were included in the study
analysis and also assessed for stage of HIV infection accordingly. Stage 2 (mild), Stage 3

(moderate), and Stage 4 (severe) were the possible staging categories. Stage 1

(asymptomatic) was not assessed because of the reliance of WHOCCD on symptomatic

54

HIV infection. WHOCCD for AIDS was preferred over conducting HIV testing because
of the expense of such tests, limited access to HIV serological testing and appropriate
counselors for consenting procedures, lack of access to antiretroviral treatment, and the

overall sensitivity of the issue during the time this study was conducted.

3.6 Definition of Cognitive Impairment

Cognitive impairment (CI) was defined as scoring two or more standard deviations
below the mean of the normative group on the ZMMSE, or scoring greater than two
standard deviations below the mean of the comparison group on the HDS and Color
Trails l. Impairment on neuropsychological test performance has previously been
defined as greater than two standard deviations below healthy subjects in other cross-
cultural investagions72‘84. The term cognitive impairment, as opposed to neurocognitive
impairment, is utilized because of the lack of diagnostic capabilities available for
differentiating the cause of cognitive impairment in hospice patients at the time of this

study.

3.7 Sample Size Calculation

In a US. study on the validity of the HDS comparing AIDS patient with and without
dementia, asymptomatic HIV patients, and normal controls, an effect size of >1 was
identified between groups on scores of both the HDS and MMSE62. Extrapolating this
effect size to this population, an estimated 17 patients per group are needed to identify
significant differences in mean NP test scores with 0L=.05 and [3:20 for Specific Aim 3.

For specific aim 5, a sample size of 88 is anticipated to be necessary to establish 95%

55

confidence that between 5% and 20% of AIDS patients have suspected HIV-D in this
population, assuming that Zambian AIDS patients have a similar incidence of HIV -D as

US. AIDS patients in the pre-HAART era (i.e. 15%)5.

3.73.8 Statistical Analysis
3.8.1 Overview

Data was coded and entered into an Excel spread sheet. All statistical analyses were
carried out using SPSS version 13.0 or SAS version 9.1. Demographic variables were
compared using Student’s t-tests for continuous variables including age and education.
Chi-square tests were performed to identify differences between groups for categorical
variables such as gender and language of interview. Patients and comparison controls
with missing data points for neuropsychological tests were evaluated and cases removed
from data analysis when necessary. Color Trails 1 was the only test in which missing
data values were problematic, with one (6.7%) comparison group participant in which a
poor copy of the testing material was administered with missing numeric values; and in
three PLWAS Color Trails l was not assessed because of patient refusal (n=2) or
inadequate test documents (n=1). On occasions in which participants were unable to
complete the task at the five-minute mark, a test time of 300 seconds was imputed
(PLWAS, n=16; Comparison group, n=2).
3.8.2 Data Analysis by Specific Aim
3.8.2.1 Aim 1

To assess cultural relevance and adaptability of screening instruments, including the

ZMMSE, HDS and Color Trails 1 & 2, to identify cognitive impairment among PLWAS in

56

Lusaka, Zambia through focus group discussions with local medical professionals,

community piloting, and in study participants.

Evaluations of study instruments for cultural relevance and adaptability were

conducted through:

3.8.2.2 Aim 2

Focus group discussions with local medical professionals, including a
psychiatrist, and two internists, as well as medical students to assess
individual neuropsychological test items for cultural, contextual, and
linguistic appropriateness

Consensus agreement amongst local healthcare professionals regarding
appropriate adaptations for items identified as culturally problematic in
focus group discussions

Forward and back translations of study instruments from English into
Chinyanja, the most prevalent vernacular language in Lusaka, Zambia,
utilizing emic translation procedures

Interview-based field piloting of the adapted and translated study
documents and neuropsychiatric instruments to volunteer community
members prior to study commencement, and finally, with study participant

administration.

To evaluate the distribution of individual neuropsychological test items between

PLWAS and healthy community members in Lusaka, Zambia.

57

The frequency (%) of correct responses on sub-items of the neuropsychiatric tests
were evaluated in both PLWAS and the healthy comparison group. In addition, the
distribution of each individual neuropsychiatric test score was evaluated graphically.
3.8.2.3 Aim 3

To evaluate differences in mean ZMMSE, HDS, and Color Trails I & 2 between
PLWAS admitted to Our Lady’s Hospice in Lusaka, Zambia and healthy persons from the
some community without evidence of HIV/AIDS.

Mann-Whitney U Tests and Student’s t-tests were performed to compare ZMMSE,
HDS, and Color Trails 1 scores between PLWAS and the comparison group using p g
0.05 as the level of statistical significance.
3.8.2.4 Aim 4

To evaluate WHO HIV Stage as a potential effect modifier of performance on the
ZMMSE, HDS, and Color Trails I & 2 among PLWAS admitted to Our Lady’s Hospice in
Lusaka, Zambia.

To assess the impact of WHO Stage on neuropsychological test performance, Kruskal-
Wallis and one-way AN OVA tests were performed to evaluate between group and within
group differences. In addition, a chi-square test for trend in binomial proportions was
used to assess the relationship between CI and WHO staging of HIV infection. A
significance level of 0.05 was considered statistically significant.
3.8.2.5 Aim 5

To determine the prevalence of HIV-related NCI among PLWAS admitted to Our

Lady ’s Hospice in Lusaka, Zambia, as defined by scoring _>_ 2 standard deviations (SD)

58

below the comparison group on the ZMMSE alone, or Z 2 SD below the mean on both
the HDS and Color Trails I.
The prevalence of HIV-related NCI is reported as the proportion (%) of hospice

patients identified meeting the study definition for significant cognitive impairment.

3.9 Protection of Human Rights
This project was approved by both the Michigan State University Committee on
Research Involving Human Subjects and the University of Zambia Research Ethics

Committee.

59

TABLES

Table 4. Neuropsychological Instruments and Cognitive Domains Assessed

 

 

 

 

 

 

 

Instrument Domains Assessed
ZMMSE Orientation, memory registration, language, attention, memory
recall
HDS Psychomotor speed, and memory recall
Color Trails 1 & 2 Attention and concentration, motor speed/fine motor control,
executive function

 

Table 5. World Health Organization Screening Algorithm for HIV Infection*

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Definition of Suspected HIV infection:

1) 1 cardinal OR

2) 2 characteristic OR

3) 1 characteristic and 2 associated

4) OR 3 associated

Symptoms, Signs, & Diagnosis Caggories
Cardinal Characteristic Associated

Kaposi’s sarcoma Oral Candidiasis Weight loss >10%

Pneumocystis carinii Hairy leukoplakia Fever >1 month

pneumonia

Toxoplasmic Encephalitis Tuberculosis Diarrhea > 1 month

Esophageal candidiasis Herpes Zoster Genital Ulcers >1
month

Cytomegalovirus retinitis Pruritic dermatitis Cough >1 month

Cryptococcal Meningitis B cell lymphoma Drug reactions
Skin infections
Seborrheic dermatitis
Neurologic symptoms

 

 

 

*Adapted from Miller et al. Diagnosis and screening of HIV/AIDS using clinical criteria
in Tanzanian adults. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 9(4):410.124

60

FIGURES

Figure 1. Map of Zambia

l

 

   

   

K.) 3.21mi}.

1'. {.5 '.vi gas
1' 8:38

  
 
 
 

 

Cinema

 

Kaplra Mpiib'fl.

    
  
   
 

'Ka bwe

.MOSQU LUSAKA*

.w...al,ivingsétor?j
. T‘s—IIJEFT’NA

  

 

 

 

61

Figure 2. Map of Lusaka, Zambia and Location of Kalingalinga Compound*

      

KA nAsouM: /

4 .
%YTN‘“ ./

*Arrow indicates location of Kalingalinga

62

CHAPTER 4: RESULTS

4.1 Overview and Descriptive Statistics

Patient flow is summarized in Figure 3. The final study group was comprised of 48
PLWAS. Only one patient was currently receiving antiretroviral therapy, and none of
the study group reported previous use of antiretroviral therapy. A group of 18 healthy
volunteer individuals from the Kalingalinga community underwent the same
evaluation. Three volunteers reported a past medical history of possible
signs/symptoms of HIV including herpes zoster, cryptococcal meningitis, and a
history of chronic persistent diarrhea. These volunteers were excluded from the
analysis and the remaining 15 formed the comparison group. As suspected, very few
participants were aware of their HIV status, with only 5 (10.4%) PLWAS reporting
having had an HIV test in the past (4 positive, 1 negative). Only one comparison
group participant (6%) reported knowledge of HIV status, having had a negative HIV
test in the past. Baseline characteristics for the PLWAS and comparison groups are
summarized in Table 5. No significant differences were found between or within
PLWAS and comparison subjects with regard to age, education, or gender. Among all
PLWAS, 25 (52.1%) reported noticing changes in memory and 16 (33.3%) reported
noticing changes in thinking. No one in the comparison group reported problems
with memory or thinking. Three PLWAS (6%) were identified as having WHO Stage

H, 14 (29%) WHO Stage 1H, and 31 (65%) WHO Stage IV HIV disease.

63

4.2 Results by Specific Aim
4.2.1 Aim 1

To assess cultural relevance and adaptability of screening instruments, including the
ZMMSE, HDS and Color Trails 1 & 2, to identify cognitive impairment among PLWAS in
Lusaka, Zambia through focus group discussions with local medical professionals,
community piloting, and in study participants.
4.2.1.1 Zambian Mini Mental State Examination

Several modifications were made for cultural appropriateness on the locally adapted
ZMMSE. Amongst items assessing orientation to time, season ascertainment was noted
to be an item of difficulty because the Chinyanja translation for “what season is it,”
literally translates as “what time is it?” The result was that patients were often confused
if the item referred to time of day or time of year. To circumvent this confusion and
provide context to the question, patients were provided with cues. Cues included all
possible correct answers to the question, including hot, cold, rainy, or dry. During the
study period, the only correct answer to this question was “rainy.” No difficulties were
encountered with other time orientation assessments including year, date, day, and
month. The MMSE as designed by Folstein et a1, includes five items assessing
orientation to place, such as state, county, city or town, place (i.e. hospital or clinic), and
floor of building. These items were altered on the ZMMSE to include province, city, area
of city, name of hospice, and floor of building. No changes were made regarding item
naming, including identifying a wristwatch and pencil by the appropriate names.
However, while assessment of language repetition and fluency with the statement, “no ifs

ands or buts,” was maintained in interviews conducted in English, another statement

64

needed to be identified for Chinyanja interviews. “Kulibe kuchita, kulibe kusachita,”
which translates to “there is no doing, there is no not doing,” was identified as a potential
phrase through another group in Lusaka conducting a concurrent HIV -D study, and was
utilized because of the lexical and syntactical similarities to the English phrase.

Education and rate of literacy were assumed to be relatively low in Kalingalinga and
items requiring reading and writing were considered to be inappropriate in such a
population. Attention tasks such as spelling world backwards or serial 73 were therefore
replaced with a different task, which required study participants to name the days of the
week backwards, starting with Sunday. Another item of concern because of education
and literacy rates was a visual command task in which patients are asked to read the
visual command, “close your eyes.” Instead, patients were asked by the examiner to
“look at me and do exactly as I do.” The examiner then closes his/her eyes for three
seconds, and observes and records the patient response. This procedure has been used in
other studies of the MMSE in illiterate populations“).

Similarly, the sentence—writing task was identified as another problematic item in this
population. In a Hindi adaptation of the MMSE in an illiterate population, this item was
replaced with “Tell me something about your house”60. However, in our community
piloting of the tests, this item often lead to confusion and required additional explanation.
Another appropriate substitution was never identified, so we decided to include the
sentence-writing task for completeness. All other items were maintained as originally

described. The final ZMMSE is shown in Appendix 1.

65

4.2.1.2 HIV Dementia Scale

Previous investigations have indicated that most adults are able to count numbers from
one through twenty-five, regardless of literacy levelm. Therefore, the HDS timed-
alphabet task was changed to a timed-number writing task. The 3D cube drawing was
dropped from the HDS because very few participants, healthy or otherwise, could
complete the drawing accurately. Thus, the maximum HDS score was six points versus
twelve on the original HDS scale.
4.2.1.3 Color Trails 1 & 2

No modifications were made to Color Trails 1 & 2. While field piloting and testing in
the comparison group revealed no administrative difficulties, Color Trails 2 was revealed
to be poorly tolerated by hospice patients. A significant number declined to complete the
task, were unable to complete in the allotted seven minutes or made five or more errors
resulting in discontinuation of the test. Thus, only Color Trails 1 was included in the data
analysis.
4.2.2 Aim 2

To evaluate the distribution of individual neuropsychological test items between

PLWAS and healthy community members in Lusaka, Zambia.
4.2.2.1 Zambian Mini-Mental State Exam

The frequency of correct response to ZMMSE test items among comparison group
participants and PLWAS are reported in Table 7. The ZMMSE appeared to have good
face validity among the healthy comparison group. Items including assessment of year,
day of week, name of place, item naming, pentagon drawing, and word registration were

answered correctly by 15 (100%) of the comparison group. Other items including

66

season, month, floor, town, compound, phrase repetition, naming days of the week
backwards, and follows a 3-step command also performed reasonable well, with at least
13 (86.7%) of comparison participants providing the correct response. Other items
proved less reliable. As suspected, sentence writing proved to be one of the most
difficult tasks with only 11 (73.3%) able to complete the task correctly. Day of month,
province, and word recall items were also less reliable with 11 (73.3%), 12 (80%), and 12
(80%) of comparison group participants answering these items appropriately,
respectively. These items might be considered for revision in future cross-cultural
validation studies.

PLWAS performed worse than the comparison group on all ZMMSE items. There
were no items on which all PLWAS were able to answer correctly. PLWAs performed
reasonably well on orientation questions, with the exception of day of month and day of
week in which only 17 (35.4%) answered correctly. Language naming items, including
naming “chair” and “pen,” appeared to be intact among PLWAS with 46 (95.8%) able to
generate correct responses. Other language tasks such as phrase repetition, following a
three-step command, and sentence—writing proved more difficult for PLWAS with 33
(68.6%), 31 (65.6%), and 20 (41.7%) providing correct responses, respectively. PLWAS
performed very poorly overall on items of attention, constructional praxis, and memory
recall (table 7).

The distribution of ZMMSE total scores for both the comparison group and PLWAS

showed left-skewing, as seen in Figure 4.

67

4.2.2.2 HIV Dementia Scale

Performance on the HDS was markedly difficult for both PLWAS and the comparison
group. The distribution of time in seconds to complete the HDS timed number writing
task are provided in Figure 5 for both groups. Ten PLWAS (20.8%) and one (6.7%)
comparison group participant were unable to complete the timed number writing task
correctly. Among those completing the task, timed number writing was significantly
slower among the comparison group participants compared to data from US. normative
controls on timed alphabet number writing, and point distributions based on U.S.-derived
scores required reassessment. As seen in Figure 5, no comparison group participants were
able to complete timed number writing in under 25 seconds, resulting in maximum task
score of 1 based on US. derived cut-off scores. Thus, the distribution of time in seconds
for normal controls was reassessed, and a new scoring system was derived, as outlined in
table 8. Memory recall task also proved difficult, with only 3 of 15 (20%) comparison
subjects and 7 of 48 (14.3%) PLWAS able to correctly recall all four items. HDS
summary score distributions for the comparison group and PLWAS are summarized
graphically in Figure 6.
4.2.2.3 Color Trails 1

The distribution of time in seconds required to complete Color Trails l are shown in
Figure 7 for both study groups. Time to complete Color Trails 1 among comparison
group participants, ranged from 42 seconds to a maximum of 300 seconds. In two
instances comparison group participants were unable to complete the task in the allotted 5
minute time period, resulting in imputed scores of 300 seconds. The remaining 12

(85.7%) performed the task correctly. Among PLWAS, time in seconds ranged from 32

68

seconds to 300 seconds. However, 12 (25%) PLWAS were unable to complete the task in
5 minutes resulting in a 300 second imputed value.
4.2.3 Aim 3

To evaluate difierences in mean ZMMSE, HDS, and Color Trails 1 & 2 between
PLWAS admitted to Our Lady ’s Hospice in Lusaka, Zambia and healthy persons from the
some community without evidence of HIV/AIDS.

Mean and median scores of PLWAS and comparison group participants on the
screening instruments are summarized in Table 9. PLWAS scored significantly worse
than the comparison group on all neuropsychological tests as a group (all p’s <0.001).
4.2.4 Aim 4

To evaluate WHO HIV Stage as a potential effect modifier of performance on the
ZJMMSE, HDS, and Color Trails I & 2 among PLWAs admitted to Our .Lady’s Hospice in
Lusaka, Zambia.

After stratifying by WHO stage, highly significant differences were identified between
the comparison group and WHO Stage IV PLWAs on all tests (all p’s < 0.001), as seen in
Table 9. No statistically significant differences were identified between the comparison
group and WHO Stage Hill] on the ZMMSE (p =.081), HDS (p =.086), or Color Trails 1
(p =.093). However, WHO Stage IV patients differed significantly from the comparison
group on the ZMMSE (p<0.001), the HDS (p<0.001), and Color Trails l (p<0.003).
WHO Stage IV differed from WHO Stage II/III on the HDS only (p =.011).

4.2.5 Aim 5
To determine the prevalence of H1 V-related N CI among PLWAS admitted to Our

Lady 's Hospice in Lusaka, Zambia, as defined by scoring 2 2 standard deviations (SD)

69

below the comparison group on the ZMMSE alone, or 2 2 SD below the mean on both
the HDS and Color Trails I.

Twenty-four PLWAS (50%) scored >2 SD below the comparison mean on the
ZMMSE, and 8 of those 24 also scored >2 SD below the comparison mean on both the
HDS and Color Trails 1. No additional CI cases were identified using the latter criteria
alone. Twenty-four PLWAS (50%) scored >2 SD below the comparison mean on the
HDS. Finally, 13 PLWAs (27%) scored >2 SD below the comparison mean on Color
Trails 1. Overall, 24 (50%) PLWAS in this Lusaka-based hospice met the study definition
for significant NCI. Among WHO staging categories, there was one stage H (33%), 6
stage III (42%), and 17 stage IV (55%) patients with significant NCI, but this trend was
not statistically significant (p = 0.24).

4.2.6 Aim 6

To provide an overview of potential methods for formal cross-cultural validation of
locally-adapted screening instruments for HIV-related NCI in Lusaka, Zambia based on
pilot study findings.

This pilot study represented a first step in cross-cultural validation of a
neuropsychiatric instrument useful for identification of CI in HIV/AIDS patients by non-
physician healthcare workers. Based on our initial findings of high face and content
validity of the ZMMSE, the following steps, as outlined by other researchers adapting
Western instruments for cross-cultural use125 , will need to be undertaken for formal

cross-cultural validation:

70

1) Standardization of the instrument among a normal HIV seronegative population
sample to determine normal age, gender, and education-specific performance
distributions on ZMMSE performance.

a. Using logistic regression analysis with age, sex, and education included as
explanatory variables, the predicted probability of answering each item
correctly should be assessed. Identification of items with significant age,
gender, and education effects should be further modified to ensure that this
item can be used on all HIV patients irrespective of these factors.

b. Create graphical representations of observed and predictive probabilities
of answering a given test item correct for visual inspection and
identification of discrepancies.

c. Calculation of a “goodness of fit” statistic to evaluate statistical
significance of the fitted curve to representative data.

2) Reliability of Each Item

a. Clinical officer interobserver and intraobserver reliability should be
assessed by calculating Kappa statistics for each ZMMSE test item. Items
with poor interobserver and intraobserver reliability among clinical
officers should be further modified to ensure the instrument can be used in
routine care settings.

3) Consensus Meeting

a. Further revisions and modifications as determined by standardization and

reliability findings.

71

4) Further evaluation in the HIV population, including asymptomatic and
symptomatic patients, to evaluate the sensitivity, specificity, and positive predictive
value of the instrument for identifying HIV -D, MCMD, and other HIV-related NCI

compared to the American Academy of Neurology Clinical Criteria.

72

Tables

Table 6. Baseline Characteristics of Persons Living With AIDS (PLWAS) and

Comparison Group

 

 

 

 

 

 

 

 

 

Characteristic PLWAS Comparison
11 = 48 n = 15 p-value*

Age, yrs; mean (SD) 34.7 (9.53) 32.4 (7.39) 0.32
Education, yrs; mean (SD) 7.2 (4.35) 7.5 (3.11) 0.71
Sex, female (%) 30 (62.5) 9 (60) 0.87
Language of Interview 32 (67) 10 (67) 0.30
(Chinyanja)

(%)

 

*p-value based on Students T-tests for age and education, Chi square tests for gender and
language of interview

Table 7. Number of Correct Responses (%) to Zambian Mini-Mental State Exam Items
by Group and Cognitive Domain Assessed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PLWAS Comparison PLWAs Comparison
(n= 48) (n=15) (n=48) (n=15)

Orientation Year Langu_age

Year (1) 34 (70.8%) 15 (100%) Names 46 (95.8%) 15 (100%)
“chair” (1)

Season (1) 35 (72.9%) 13 (86.7%) Names 46 (95.8%) 15(100%)
“Pen” (1)

Month (1) 36 (75%) 13 (86.7%) Repeats 33 (68.8%) 14 (93.3%)
phrase (1)

Day of Month (1) 17 (35.4%) 11 (73.3%) Sentence 20 (41.7%) 11 (73.3%)
Writing (1)

Day of week (1) 29 (60.4%) 15 (100%) Follows 3- 31 (65.6%) 13 (86.7%)
step
command
(3)

Name of placgl) 39 (81.3%) 15 (100%) Constructional Praxis

Floor (1) 30 (62.5%) 14 (93.3%) Pentagon 16 (33.3%) 15 (100%)
Drawing
(1)

Town (1) 4fl89.S%) 14 (93.3%) Memory and Recall

Compound (1) 33 (68.8%) 14 (93.3%) Word 39 (81.3%) 15 (100%)
Registratio
n (3)

Province (1) 37 (77.1%) 12 (80%) Word 20 (41.7%) 12 (80%)
Recall (3)

Attention

Days of week 32 (66.7%) 13 (86.7%)

backward (5)

 

PLWAS: Persons Living with AIDS; Comparison: Comparison Group

73

 

 

Table 8. Number of Correct Responses (%) to HIV Dementia Scale Items by Group and
Cognitive Domain Assessed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PLWAS Comparison Group
Motor Speed
Timed Number Writing
<30 seconds (2 point) 9 (18.8%) 8 (53.3%)
31-40 seconds (1 points) 5 (10.4) 4 (26.7%)
>40 seconds (0 points) 34 (70.8%) 3 (20%)
Verbal Learning
Memory Recall (4 points) 7 ( 14.3%) 3 (20%)
Memory Recall (3 points) 13 (27%) 9 (60%)
Memory Recall (2 points) 10 (20.8%) 3 (20%)
Memory Recall (1 point) 5 (10.4%) 0
Memory Recall (0 points) 13 (27%) 0

 

PLWAS: Persons Living with AIDS

74

 

Amvncv 3.34m .Awmucv Z owfim 03>? AEME anew acmtmanU ”mos—g wean);
<>OZ< E :56 v a .9; 8% OPE 23,5 3. is 88858...

<>oz< 3 85 v a .Z 8% 0:3 a, BB away. on}
<>oz< E n an; 880 no mood vn ”we: one 52% no 8an .Z ownnm 0:3 .3 88868...

 

 

Emma: .8622
83.8 v2 88-va meow Aoom-mmv 3: 88-3; 5
Emv 532 1 £8...
H G69 ENE *Glmv finfl €63 mdf GAD Veno— 880
Emmet 5622
6-9 m 6-8 N 8-8 m 6-9 v
amv :32
to: am in: on 9: Nam 2.: 3. ma:
Amman: 3622
82: mm amév om 8m-mC mm Smémv mm
Emv :82
“8.8 v.3 *Aodv _dm 2.0 w.mm 3.: Nam mmEEN
$.qu =< R": in: m E":
Z owSm EB owSm comtquoU
0:3 0:3

 

@220 cemggou can Am<>>qmv me< 5;? w5>5 mcomnom n8 mucoEmmommxx _mflwo—onozmmoSoZ .0 2an

75

 

76 hospice patients
admitted

 

 

 

 

 

 

FIGURES

Figure 3. Flow Chart for Patients and Comparison Group.

 

l8 volunteer
community members

 

 

 

Expired or
Transferred
n=17

 

 

 

 

Screened
n=59
(5 aware of HIV
serostatus)

 

 

 

 

Did not met
WHOCCD
for AIDS
n=10

 

 

 

 

 

 

Screened
n=59
(5 aware of HIV
serostatus)

 

 

 

 

Did not met
WHOCCD
for AIDS
n=10

 

 

 

 

 

Completed Interview

 

 

Screened
n=59
(5 aware of HIV
serostatus)

 

 

 

 

 

 

HIV

signs/symptoms

n=3

 

 

 

Completed Interview

 

 

 

n=48

PLWAS
31 Stage IV
14 Stage III
3 Stage II

 

 

 

n=15

Comparison Group

 

WHOCCD: World Health Organization Clinical Case Definition
Stage defined by WHO classification criteria20

76

 

23m .3: mmEEN
nmmmmwnm mmmmnummmmom 22329 2 m m n

3unoo

comcmanos

m<>>._n__U
320

 

noncom 30,—. 83m 28m ESEAEE :mEEaN @380 833800 28 $3543 me< 5;? wag; 288m mo .8335me .v 23$

77

ouoo.m cmA ”_uomgmmA ”unoom mmémA
mlo8 OWRA ”38w 5-va ”muoom VNANA nonoom HNV ”monoom manta,» “Bang? 38: n8 £0-30 wo>touum5 8865 m30h<*

mucouom E 08:.

 

 

 

 

 

a 1 m m a _n a: 3%
:2; Z: Z: Z: Z: Z:
In
3
o
um
comcmanoN
w<>>._n:U
320 Z L:

 

 

n85 nonnannnoo nnn 334% men. 55,
mag...— mcomnon— “8 fish wast? noon—=2 gait 28m «:5an an 05 803800 on Amwaooom no 085. mo nozsngwa m oSME

78

23m .23 8:
m m w m n _. o

o
mm
m
0
3 m
m.
E.
nomnmaEnoH
35.58 mg

%So

8.00m :38. 28m «:5an Km 95.5 838800 98 9.4.345 me< £5» wEZA 285m .8 cousnwawa .c 05me

 

79

mucouum E 0E2.

 

 

 

 

 

 

1m

1.0—.”

n

, W

1.9.
comtmanoN
m<>>|_&_U

macaw . , . . . Ton

 

@305 flaw—«9:00

nnn @555 mm? an,» 35 nnonnn wnonnn _ £5. 860 92995 2 @288 nc on? no nonnnnnnfi .n nnnmE

80

CHAPTER 5: CONCLUSIONS

5.1 Summary of Findings

In this pilot study, non-physician healthcare workers utilized brief screening
instruments to identify significant differences in neuropsychological test performance
between PLWAS and a healthy comparison group. We found that 50% of all PLWAS,
and 55% of patients with WHO Stage IV HIV disease met our study definition of CI.
Among the study instruments tested, the ZMMSE was able to identify more cases of CI
compared to the use of the m-HDS and Color Trails 1. Comparison group participants
also performed more reliably on the ZMMSE, indicating improved cultural relevance in

this population.

5.2 Comparison of Findings to Prior Literature
5.2.1 Screening Instruments for HIV -related NCI

The pattern of deficits on the ZMMSE, including impaired attention and verbal
memory recall among PLWAS is consistent with previous investigations of HIV -related
NCI69'71’73. Verbal fluency tests are frequently impaired in patients with HIV-related
NCI, and other HIV -D investigations have employed timed semantic and phonemic
fluency tasks, which are likely more specific to the psychomotor slowing deficits seen in
HIV -D and might be considered as an appendix to the ZMMSE in future investigations of
HIV -D.

The patterns of deficits seen in HIV-related NCI in regions outside of the US,
Europe, and Australia have not been well documented and tests of cortical impairment

were also included for evaluation on the ZMMSE. Orientation and constructional praxis

81

were preserved amongst PLWAS, but these items may still have clinical utility in
identifying patients with other co—morbidities that result in cognitive impairment in
HIV/AIDS patients with or without ART treatment.

N europsychiatry officers in Zambia commonly use questions adapted from the MMSE
for cognitive status assessment in their patients. Thus, the MMSE has the benefit of
widespread familiarity in Zambia, and perhaps in other African settings as well. An
MMSE was utilized for neuropsychiatric assessment in a study of HIV inpatients in
Johannesburg, South Africa”. This study does not report if the instrument was adapted
for local use and normative data was not provided. In fact, no previous studies have
examined the reliability or validity of the MMSE in Africa. An addition to the ZMMSE
of more culturally appropriate tests of psychomotor retardation may improve
identification of HIV-related CI in such populations and should be considered in future
investigations.

Although the HDS is considered more specific to the cognitive deficits seen among
HIV/AIDS patients in the US, few study participants, healthy or otherwise, were able to
complete timed 3-D cube drawing in this study, and timed number writing also appeared
to be a difficult task. It was observed that PLWAS frequently omitted numbers and/or
wrote numbers backwards, suggesting that number writing requirements and possibly the
paper and pencil format are unfamiliar tasks amongst patients in this community.

Similarly, our normative comparison group performed much more slowly on Color

Trails 1 than other cross-cultural studies of healthy subjectsm'127

, emphasizing the need
for population-specific normative data. In a Chinese study of mandarin-speaking

volunteers with mean age of 35.77 and mean years of education of 9.06, the mean time in

82

seconds to complete Color Trails 1 was 44.7 seconds, compared to our healthy

127. Although designed to minimize

comparison group with a mean of 104.4 seconds
cultural bias by eliminating the use of the English alphabet, the Color Trail tests were
originally validated amongst populations with high educational ascertainment, with all
participants having at least 10 years of education and a mean and standard deviations of
16.4 years in Bangkok, Thailand; 15.7 years in Kinshasa, former Zaire; and 17.2 years in
Naples, Italy, whilst our normative controls and PLWAS had a mean education of 7.5 and
7.2,respectivc1y‘22. In addition, 13 study participants, including 12 (25%) PLWAs and 1
(6.7%) normative control, were unable to complete the task in the recommended allotted
time of 5 minutes. Thus, the paper and pencil format of the Color Trails test may produce
a significant educational bias and reduce the sensitivity of the test for identifying HIV-D
in areas with low education and literacy rates, reducing the cross-cultural relevance for
use. In addition, the cost of purchasing the colored documents is prohibitive to routine
use in resource-limited settings. This study did not formally evaluate the correlation
between education, HDS score, and Color Trails 1 time, and thus the causes of the
administrative difficulties observed remain speculative, but suggest that both are poor
choices for use in resource-limited settings.

The International HIV Dementia Scale (IHDS) may prove to be a useful screening
instrument, but was not available at the time of this study“. The IHDS contains tests of
motor speed (timed finger tapping), psychomotor speed (timed alternating hand sequence
test), and memory (recall of four words). The IDHS was shown to have 64% sensitivity
and 71% specificity for identifying HIV-D in a group of 81 ambulatory HIV patients

(mean CD4 219 cells/mm3) in Kampala, Uganda“. However, the IHDS failed to identify

83

any significant differences in performance between factory and sugarcane workers in
Ethiopia with untreated and moderately advanced HIV disease (i.e. median CD4 <300
cells/mm3) and a group of seronegative controls“. This may be related to selection of
control patients for normative data, who were slightly younger in the Ugandan study, or
suggests that the IHDS may be insensitive to milder HIV-related NCI. Application of the
ZMMSE and the IHDS in another larger study in a rural Zambian population may help
elucidate further strengths and limitations of these two instruments.
5.2.2 Prevalence of Cognitive Impairment

This study found a very high prevalence of cognitive impairment (50%) among a
group of high-risk, antiretroviral na'i've AIDS patients in Lusaka, Zambia. Statistical
power was not sufficient to ensure that this finding was not due to chance alone.
However, other African investigations have found equally high prevalence estimates in
hospitalized patients40’42’44’84. In Tanzania, 54% of PLWAS had neuropsychiatric
abnormalities including psychomotor slowing, confusion, and dementia utilizing
unspecified neuropsychological tests of orientation, memory, attention, and calculation”.
Although the WHO staging scheme was not utilized in the Tanzanian study, 62% of
patients with greater than three months duration since AlDS diagnosis and 65% of
PLWAS that died in the hospital (suggesting advanced HIV disease) were identified as
having CI. A study from former Zaire identified neuropsychiatric abnormalities in 43 of
104 (41%) HIV inpatients”. However, the neuropsychiatric instruments utilized are not
reported in this study. Among the HIV seropositive patients, nine cases of HIV -D and
three cases of MCMD were identified. All of theses cases were identified among WHO

stage H1 (41%) and stage IV (58%) PLWAS. In Johannesburg, South Africa, HIV-D was

84

the most common neurological disorder with 108 (38%) cases of HIV -D diagnosed in
506 consecutive antiretroviral-naive HIV positive inpatients as assessed by MMSE87.
Decreasing CD4 count correlated with HIV-D diagnoses, with 108 cases having CD4
below 50, 38 between 51 and 100, 29 between 101 and 200, 12 between 201 and 500, and
in 5 cases CD4 count was >500 (p for trend not reported). Belec et al found a very low
occurrence of dementia at 3.2% amongst 93 hospitalized patients in Bangui, Central
Africa Republic“). The reasons for this disparate finding are unclear, owing to lack of
specification of clinical criteria for diagnosis in this study. It is reasonable that HIV-D
patients may be overrepresented among hospitalized patients, owing to the increased
caregiving and dependency needs of patients with cognitive impairment.

Studies of non-hospitalized HIV/AIDS patients also suggest that HIV-related NCI are

39’44’72, with the exception of an

common in ambulatory HIV patients in SSA(i.e. 4.5-31%)
Ethiopian study which failed to find any significant HIV-related NCI“. It is important to
note, however, that this study utilized employer-based free clinics for patient
identification. Thus, patients with more disabling neurological manifestations may have
sought care elsewhere for fear of employment insecurity (although >70% of the HIV
population is characterized in this study). HIV-D patients also have reduced survival
without access to HAART and may also have contributed to the absence of significant
impairments in this cohort.

In the US, several studies have shown that subtle cognitive deficits among
asymptomatic HIV patients (equivalent to WHO stage I) increase in magnitude with

progression to symptomatic HIV (equivalent to WHO stage H and HI) and full-blown

AIDS (equivalent to WHO stage IV) 128'130. In this study, there appeared to be important

85

differences between the Stage II/III and Stage IV PLWAS on all neuropsychological tests,
but this trend did not reach statistical significance. This latter finding may reflect the
limited number of patients among the study population with Stage II or Stage III HIV
disease. Studies with larger sample size will be needed to further examine the

relationship between stage of HIV disease and CI among African PLWAS.

5.3 Strengths and Limitations

This study was limited by small sample size. Another limitation of our study was that
causes of CI likely included HIV-D, opportunistic infections of the CNS, metabolic
encephalopathy, as well as mixed cases. Our ability to define the cause of CI was
hindered by the lack of available diagnostic testing and the expense of conducting such
tests, but these conditions reflect the reality of HIV care provisions in SSA. Lack of
access to diagnostic and physician level healthcare are challenges faced daily by African
patients and their healthcare providers. In addition, the screening instruments utilized in
this study only assessed cognition, but behavioral abnormalities and functional abilities
are also important components of the clinical triad of HIV-related NCI. Still,
identification of patients with Cl may prompt additional referral and/or lead to
identification of other potentially treatable causes, such as neurosphyillis or depression.

In comparison to previous investigations of CI among African PLWAS that utilized
extensive neuropsychiatric tests administered and interpreted by neuropsychologists and
physician-specialists, a strength of this study is that the methodology is reproducible in
other resource-limited setting. These preliminary findings indicate that the ZMMSE is

easily administered and interpreted by non-physician healthcare workers and that further

86

modifications to the instrument and evaluations may allow routine screening for
cognitive impairment in HIV patients in SSA.
5.4 Suggestions for Future Research

As ART becomes increasingly available in SSA, HIV-related NCI have important
pubic health implications and may become increasingly prevalent. Enabling clinical
identification of the disorder is crucial to monitoring shifts in the epidemiology of these
disorders and planning appropriate resource allocations.

This pilot study conducted preliminary steps for formal validation of the ZMMSE in
Zambia. Additional studies will need to be conducted to standardize the instrument, and
will likely result in further modifications and improvements. Further, reliability of test
items within and between observers should be assessed if the instrument is to be used for
routine surveillance of HIV-related NCI by clinical officers or other paramedical
healthcare workers. Screening instruments for identifying HIV-related NCI provide a
viable opportunity to improve diagnosis in resource-limited settings where physician-
experts are most often unavailable.

Many aspects of HIV-related NCI remain unknown, including optimal treatment
schemes and diagnostic importance of CSF markers of HIV suppression and
inflammation (i.e. CSF viral load and cytokines) in guiding treatment decisions. The
pathogenesis of the disorder may be further elucidated by studies of frequency, clinical
features, and course of HIV-related NCI in patients with different patterns of disease
transmission and HIV-related disease, as well as diverse HIV clades. Enabling early

clinical identification of HIV-related NCI by clinical officers is a first step in improving

87

understanding of the epidemiology of these disorders in SSA, as well as providing

maximal clinical benefit to patients through ART initiation.

88

APPENDICES

89

APPENDIX A

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

... co o
a a 3 a
l: 8. 0 8. . . . . .
8 3 g 3 Zambian Mini-Mental State Examination
'5 ‘3‘ U 9‘ Kuyambisa (Orientation)
O Ndi caka bwanji (what is the year)?
0 Ndi ntawu bwanji (ntawi ya mpepo, maiza, chi la la)
(what is the season)?
0 Ndi mwezi bwanji (what is the month)?
0 Ndi siku la bwanji mu mwezi (what day of the month
is it)?
0 1 Ndi siku la bwanji mu week (what day of the week is
it)?
0 l Tili muziko bwanji (what is the name of this place)?
0 1 Kodi tili pa floor bwanji? (what floor are we on?)
0 1 Kodi tili mumuzinda bwanji (what city are we in)?
0 l Kodi tili muyumba btani (what are of that city are we
in)?
0 1 Tili muprovince bwanji (what province are we in)?
Cilankulidwe (Langlage)
Kodi nicani ici?/Kambani zina la cisulo ici
(What is the name of this (Show item ) )?
0 l Nkoloko (wrist watch)
0 1 Pensulu (pen)
0 Mubwezepo pambuyo panga: ‘kulibe kuchita kulibe
kusachita’
Ask the patient to repeat this phrase, only one trial: ‘no
ifs, ands, or buts’
Khalani cete (Attention)
Can you name the days of the week backward? For
example, before Sunday comes Saturday; what comes
before that?
0 1 Friday
0 1 Thursday
0 1 Wednesday
0 1 Tuesday
0 1 MondaL

 

9O

 

 

Incorrect

Response

Correct

Response

Zambian Mini-Mental State Examination
(Page 2)

 

Konkani njila zitatu (3-Stage Command)

 

Tenga iyi pepala mukwanja yamanja, upeteke pakati
ndipo uike pansi.

(Take this paper in your right hand, fold it in half and
put it on the floor. Do NOT prompt or repeat
command when patient is performing task)

 

Right hand

 

Folds

 

COO

Places on floor

 

Lamulo lolembedwa (Written Command)

 

Welengani mau aya ndi kucita zimene inena
(MVALAN I MASO AN U)

(say look at me and do exactly what I do. DO NOT
TELL PATIENT TO CLOSE EYES).

 

Lembani mau (Write a Sentence)

 

Lembani mau Lembani mau aliyense (Have the patient
write a sentence)

 

Cigwilizano (Coordination)

 

 

Lembani ici cintuzi bwinobwino mwamene
mungalembele

(All ten angles must be present and the two pentagons
must intersect)

 

 

 

 

| TOTAL

 

91

 

APPENDIX B

Introduction

HIV-associated dementia (HIV -D) is a qualifying indication for antiretroviral therapy
(ART) initiation according to World Health Organization (WHO) guidelines for resource-
limited settings“, but is not routinely screened for in ART clinics in sub-Saharan Africa
(SSA) due to a lack of valid instruments with population-based normative data. Delays
in treatment initiation may lead to irreversible neurologic damage that complicates future
HIV management“, but the non-physician healthcare workers (NPHW) providing the
bulk of HIV care in SSA38'65‘l3 1, have little experience in diagnosing these disorders.

Previous investigations for HIV-D in SSA have utilized physician specialists for
evaluation4l‘64'87. No previous studies have utilized NPHW for cognitive assessments.
We trained NPHW in administration of locally-adapted screening instruments. We report
cross-sectional results of performance and frequency of cognitive impairment (CD in
Persons Living With AIDS (PLWAS) admitted to an urban hospice facility in Lusaka,

Zambia.

Methods
Population

PLWAS were recruited from a hospice in Kalingalinga, Lusaka, Zambia and controls
were recruited from the same community. HIV seroprevalence in Kalingalinga was
reported at 26% in 2002121. The hospice had a 24—bed capacity, but no laboratory or
radiological equipment available on site.

Participants

92

All PLWAS, aged 18 or older, admitted for the first time to the hospice between
January 21, 2004 and March 22, 2004 were interviewed. Exclusion criteria included
failure to meet World Health Organization Clinical Case Definition for AIDS
(W HOCCD)89, past medical history of a chronic neurologic disorder, psychiatric
disorder, substance abuse, or a physical deficit or severe medical illness that would
interfere with study performance. Family members of patients attending outpatient
clinics also underwent the evaluation to provide normative data.

Interview Procedures

This project was approved by the Michigan State University Committee on Research
Involving Human Subjects and the University of Zambia Research Ethics Committee.
Twelve NPHW were trained in administration of study instruments. After informed
consent procedures, participants underwent a structured interview assessing demographic
information and past medical history. A chart review was also undertaken. Interviews
were conducted in the patient’s language of choice between Chinyanja (67%) and English
(33%).

HIV/AIDS Diagnosis

HIV status was assessed by the WHOCCDgg'm. Comparison group participants with
signs/symptoms of HIV were excluded from analysis (n=3). WHOCCD for AIDS was
preferred over conducting HIV testing because of limited access to HIV testing. Patients
were also assessed by WHO HIV staging criteria”. Subsidized ART were generally not
available in Zambia during the study period.

Neuropsychological Screening Instruments

93

The Mini Mental State Exam54 (MMSE) was selected because of frequent informal
use among clinical psychiatric officers in Lusaka. The HIV Dementia Scale62 (HDS) and

1122 (CT 1) were chosen because of the specificity of each in assessing HIV -

Color Trails
D deficits. To adapt these instruments, focus group discussions were conducted with a
local psychiatrist (A.H.), internists, and medical students to identify test items likely to be
culturally or otherwise problematic. Once identified, the cognitive domain tested by each
item and relationship to HIV-D was discussed and adaptation suggestions were provided.
Adapted items involved tasks requiring reading and writing and were replaced with more
appropriate items deemed to have content validity by the focus group. However, an
adaptation for 3-D cube drawing was not identified and this item was omitted from the
summary score. Following a careful emic translation procedure by experienced medical
translators, instruments were back—translated into English by a linguist and medical
professionals and pre-piloted in the community prior to study commencement.
Statistical Analysis

Data was coded and entered into an Excel spread sheet. All statistical analyses were
carried out using SPSS version 13.0 or SAS version 9.1. Demographic variables were
compared using Student’s T-tests or Chi-square tests. Mann-Whitney U tests and
Student’s t-tests were performed to compare MMSE, HDS, and CTl scores between
PLWAS and the comparison group. Very few patients had evidence of only mild HIV
disease and patients with WHO Stage II Illness and WHO Stage III together for
comparisons. Kruskal-Wallis tests and one-way AN OVA tests were performed to

evaluate between group differences on WHO Stage II/III, WHO Stage IV, and the

comparison group. Since all nonparametric tests supported the findings of the Student’s

94

 

t-tests and AN OVA, only the latter are reported to maintain the raw data in original scale.
Significant CI was defined as scoring >2 SD below the mean of the comparison group on
the MMSE, or >2 SD below the mean of the comparison group on both the HDS and
CTl, similar to previous investigations72’84. The prevalence of CI is reported as the
proportion (%) of PLWAS meeting the study criteria for significant CI. A chi-square test
for trend in binomial proportions was used to assess the relationship between CI and
WHO HIV stage. A significance level of 0.05 was considered statistically significant for
all tests.
Results

The final study group was comprised of 48 PLWAS (Figure 3). One patient was
currently receiving ART, and none reported previous use of ART. Only one comparison
group volunteer (6%) reported having a negative HIV test. N 0 significant differences
were found between PLWAS and healthy subjects with regard to age, education, or
gender (see Table 6). Among all PLWAS, 25 (52.1%) reported noticing changes in
memory and 16 (33.3%) reported changes in thinking. No one in the comparison group
had these complaints.
Neuropsychological Comparisons

PLWAS scored significantly worse than the comparison group on all tests as a group,
as summarized in Table 9. Overall, three PLWAS (6%) were identified as having WHO
Stage H, 14 (29%) WHO Stage III, and 31 (65%) WHO Stage IV disease. After
stratifying by WHO stage, highly significant differences were identified between the
comparison group and WHO Stage IV PLWAS on the MMSE (p<0.001), HDS (p<0.001),

and CT] (p<0.003). No statistically significant differences were identified between the

95

comparison group and WHO Stage II/III on the MMSE (p =.081), HDS (p =.086), or CTl
(p =.093). WHO Stage IV differed from WHO Stage II/III on the HDS only (p =.011).
Twenty-four PLWAS (50%) scored >2 SDs below the comparison mean on the MMSE,
and 8 of those 24 also scored >2 SD below the comparison mean on both the HDS and
CT]. No additional cases were identified using the latter criteria alone. Twenty-four
PLWAS (50%) scored >2 SD below the comparison mean on the HDS. Finally, 13
PLWAS (27%) scored >2 SD below the comparison mean on CT]. Overall, 24 (50%)
PLWAS met the study definition for significant CI. Among WHO staging categories,
there was one stage H (33%), 6 stage III (42%), and 17 stage IV (55%) patients with
significant CI, but this trend was not statistically significant (p = 0.24).
Discussion

In our pilot study, NPHW utilized locally adapted screening instruments to identify
significant differences in neuropsychological performance between PLWAs and a healthy
comparison group. The MMSE was noted to be most easily administered and have better
face validity among control subjects. An addition to the MMSE of more culturally
appropriate tests of psychomotor retardation and visuo-spatial abilities may improve
identification of HIV-related CI and should be considered in future investigations.
Clinical psychiatry officers in Zambia commonly use MMSE items for cognitive
assessments, and therefore the MMSE has the benefit of widespread familiarity. The
International HIV Dementia Scale (IHDS) may also prove to be a useful screening
instrument, but was not available at the time of our study“. We are applying the adapted

MMSE and the IHDS in another larger study in a rural Zambian population utilizing

96

NPHW, which may help elucidate further strengths and limitations of these two
instruments for routine clinical assessments.

We found that 50% of all PLWAS, and 55% of patients with WHO Stage IV HIV
disease met our study definition of CI. In addition, the identification of one stage II
(33%) and 6 stage III (42%) patients with C1 suggests that screening is important for all
symptomatic HIV patients. Other African investigations have found equally high
prevalence estimates in hospitalized patients40’42‘44’84. HIV patients with Cl may be
overrepresented among hospitalized patients owing to the increased dependency and
caregiving needs, but our findings suggest that CI in PLWAS are disorders that consume
a great deal of hospital resources in SSA. Given that delayed treatment may result in
irreversible CI, resources should be allocated to allow increased training in identifying
and managing such cases, as well as allowing for caregiver respite.

Screening instruments alone cannot be used alone for HIV-D identification, but may
help to narrow the ART access gap for patients that would not otherwise qualify for ART
initiation. In Kampala, Uganda, 6 of 16 HIV-D patients had CD4 counts greater than
200, making a clinical-based diagnosis essential for ART commencement in 37.5% of
HIV-D patients identified“. Use of such screening instruments may facilitate triage to
specialists or diagnosis of other causes of CI in PLWAS. The time to enable clinical
identification is now, so that these important disorders can be considered in public health
policy planning and resource allocation. Our preliminary findings indicate that simple
screening examinations performed by NPHW can identify HIV patients with significant

CI.

97

REFERENCES

98

REFERENCES

J anssen RS, Nwanyanwu OC, Selik RM, Stehr-Green J K. Epidemiology of
human immunodeficiency virus encephalopathy in the United States. Neurology
1992;42(8):1472-6.

McArthur JC, Brew BJ, Nath A. Neurological complications of HIV infection.
Lancet Neurol 2005;4(9):543-55.

Sacktor N, McDerrnott MP, Marder K, Schifitto G, Selnes OA, McArthur J C,
Stem Y, Albert S, Palumbo D, Kieburtz K, De Marcaida J A, Cohen B, Epstein L.
HIV-associated cognitive impairment before and after the advent of combination
therapy. J Neurovirol 2002;8(2): 136-42.

Wojna V, Nath A. Challenges to the diagnosis and management of HIV dementia.
AIDS Read 2006;16(11):615-6, 621-4, 626, 629-32.

McArthur J C, Hoover DR, Bacellar H, Miller EN, Cohen BA, Becker JT, Graham
NM, McArthur J H, Selnes OA, Jacobson LP, et al. Dementia in AIDS patients:
incidence and risk factors. Multicenter AIDS Cohort Study. Neurology
1993;43(11):2245-52.

Moore R, Keruly J C, Gallant J, Chaisson RE. Decline in Mortality rates and
opportunistic disease with combination antiretroviral therapy. 12th World AIDS
Conference. Geneva, 1999.

Brodt HR, Karnps BS, Gute P, Knupp B, Staszewski S, Helm EB. Changing
incidence of AIDS-defining illnesses in the era of antiretroviral combination
therapy. Aids 1997;1 1(14):173l-8.

d'Arminio Monforte A, Cinque P, Mocroft A, Goebel FD, Antunes F, Katlama C,
Justesen US, Vella 8, Kirk O, Lundgren J. Changing incidence of central nervous
system diseases in the EuroSIDA cohort. Ann Neurol 2004;55(3):320-8.

Mocroft A, Katlama C, Johnson AM, Pradier C, Antunes F, Mulcahy F, Chiesi A,
Phillips AN, Kirk O, Lundgren JD. AIDS across Europe, 1994-98: the EuroSIDA
study. Lancet 2000;356(9226):291-6.

99

10.

11.

12.

13.

14.

15.

l6.

17.

18.

Sacktor N, Lyles RH, Skolasky R, Kleeberger C, Selnes OA, Miller EN, Becker
JT, Cohen B, McArthur J C. HIV-associated neurologic disease incidence
changeszz Multicenter AIDS Cohort Study, 1990-1998. Neurology

2001 ;56(2):257-60.

McArthur JC. HIV dementia: an evolving disease. J Neuroimmunol 2004;157(1-
2):3-10.

Cysique LA, Maruff P, Brew BJ. Prevalence and pattern of neuropsychological
impairment in human immunodeficiency virus-infected/acquired
immunodeficiency syndrome (HIV/AIDS) patients across pre- and post-highly
active antiretroviral therapy eras: a combined study of two cohorts. J Neurovirol
2004;10(6):350-7.

Dore GJ, McDonald A, Li Y, Kaldor J M, Brew BJ. Marked improvement in
survival following AIDS dementia complex in the era of highly active
antiretroviral therapy. Aids 2003; 17(10): 1539-45.

Tozzi V, Balestra P, Lorenzini P, Bellagamba R, Galgani S, Corpolongo A, Vlassi
C, Larussa D, Zaccarelli M, Noto P, Visco-Comandini U, Giulianelli M, Ippolito
G, Antinori A, Narciso P. Prevalence and risk factors for human
immunodeficiency virus-associated neurocognitive impairment, 1996 to 2002:
results from an urban observational cohort. J Neurovirol 2005;11(3):265-73.

Inungu JN, Mokotoff ED, Kent JB. Characteristics of HIV infection in patients
fifty years or older in Michigan. AIDS Patient Care STDS 2001;15(11):567-73.

Mehta NA, Ranjith PV, Keshav BS, Chandak NC. Neurologic disorders
complicating HIV Infection. A study of 52 cases. Neurology India
1998;44(Suppl):21.

Valcour V, Shikuma C, Shiramizu B, Watters M, Poff P, Selnes O, Holck P,
Grove J, Sacktor N. Higher frequency of dementia in older HIV-1 individuals: the
Hawaii Aging with HIV-1 Cohort. Neurology 2004;63(5):822-7.

Valcour VG, Shikuma CM, Watters MR, Sacktor NC. Cognitive impairment in
older HIV-l-seropositive individuals: prevalence and potential mechanisms. Aids
2004;18 Suppl 1:S79-86.

100

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

Moore RD, Chaisson RE. Natural history of HIV infection in the era of
combination antiretroviral therapy. Aids 1999;13( 14): 1933-42.

McArthur JC, Haughey N, Gartner S, Conant K, Pardo C, Nath A, Sacktor N.
Human immunodeficiency virus-associated dementia: an evolving disease. J
Neurovirol 2003;9(2):205-21.

Dore GJ, Correll PK, Li Y, Kaldor JM, Cooper DA, Brew BJ. Changes to AIDS
dementia complex in the era of highly active antiretroviral therapy. Aids
1999;13(10): 1249-53.

Masliah E, DeTeresa RM, Mallory ME, Hansen LA. Changes in pathological
findings at autopsy in AIDS cases for the last 15 years. Aids 2000;14(1):69-74.

Neuenburg J K, Brodt HR, Hemdier BG, Bickel M, Bacchetti P, Price RW, Grant
RM, Schlote W. HIV -related neuropathology, 1985 to 1999: rising prevalence of
HIV encephalopathy in the era of highly active antiretroviral therapy. J Acquir
Immune Defic Syndr 2002;31(2): 171-7.

Brew BJ. Evidence for a change in AIDS Dementia‘Complex in the era of highly
active antiretroviral therapy and the possibility of new forms of AIDS Dementia
Complex. AIDS 2004;18(Suppl 1):S75-78.

Navia BA, Jordan BD, Price RW. The AIDS dementia complex: I. Clinical
features. Ann Neurol 1986;19(6):517-24.

Camara VD, Tavares W, da Rocha MP, Chimelli LC, Dumas-Hahn M. [The
knowledge of neurological changes in patients with AIDS]. Arq Neuropsiquiatr
1995;53(1):53-9.

Castro-Sansores CJ, Gongora-Biachi RA, Gonzalez-Martinez P. HIV -
encephalopathy as initial manifestation of acquired immunodeficiency syndrome
in Yucatan State, Mexico. Arch Med Res 2004;35(1):49-53.

Gongora-Rivera F, Santos-Zambrano J, Moreno-Andrade T, Calzada-Lopez P,
Soto-Hemandez J L. The clinical spectrum of neurological manifestations in AIDS
patients in Mexico. Arch Med Res 2000;31(4):393-8.

101

29.

30.

31.

32.

33.

34.

35.

36.

37.

Puccioni-Sohler M, Correa RB, Perez MA, Schechter M, Ramos Filho C, Novis
SA. [Neurological complications in acquired immunodeficiency syndrome:
experience at the Hospital Universitario Clementino Fraga Filho-Universidade
Federal do Rio de J aneiro]. Arq Neuropsiquiatr 1991;49(2):159-63.

Trujillo JR, Garcia-Ramos G, Novak IS, Rivera VM, Huerta E, Essex M.
Neurologic manifestations of AIDS: a comparative study of two populations from
Mexico and the United States. J Acquir Immune Defic Syndr Hum Retrovirol
1995;8(1):23-9.

Wojna V, Skolasky RL, Hechavania R, Mayo R, Selnes O, McArthur JC,
Melendez LM, Maldonado E, Zorrilla CD, Garcia H, Kraiselburd E, Nath A.
Prevalence of human immunodeficiency virus-associated cognitive impairment in
a group of Hispanic women at risk for neurological impairment. J Neurovirol

2006; 12(5):356-64.

Tansuphasawadikul S, Amomkul PN, Tanchanpong C, Limpakamjanarat K,
Kaewkungwal J, Likanonsakul S, Eampokalap B, N aiwatanakul T, Kitayapom D,
Young NL, Hu DJ, Mastro TD. Clinical presentation of hospitalized adult patients
with HIV infection and AIDS in Bangkok, Thailand. J Acquir Immune Defic
Syndr 1999;21(4):326-32.

Wu YC, Zhao YB, Tang MG, Zhang-Nunes SX, McArthur J C. AIDS dementia
complex in China. J Clin Neurosci 2007; 14(1):8-1 1.

Riedel D, Ghate M, Nene M, Paranjape R, Mehendale S, Bollinger R, Sacktor N,
McArthur J, Nath A. Screening for human immunodeficiency virus (HIV)
dementia in an HIV clade C-infected population in India. J Neurovirol
2006;12(1):34-8.

Satishchandra P, Nalini A, Gourie-Devi M, Khanna N, Santosh V, Ravi V, Desai
A, Chandramuki A, Jayakumar PN, Shankar SK. Profile of neurologic disorders
associated with HIV/AIDS from Bangalore, south India (1989-96). Indian J Med
Res 2000;111:14-23.

Teja VD, Talasila SR, Vemu L. Neurologic manifestations of HIV infection: an
Indian hospital-based study. AIDS Read 2005;15(3):139-43, C3.

Wadia RS, Pujari SN, Kothari S, Udhar M, Kulkarni S, Bhagat S, Nanivadekar A.
Neurological manifestations of HIV disease. J Assoc Physicians India
2001;49:343-8.

102

38.

39.

40.

41.

42.

43.

45.

46.

47.

Muula AS, Chipeta J, Siziya S, Rudatsikira E, Mataya RH, Kataika E. Human
resources requirements for highly active antiretroviral therapy scale-up in Malawi.
BMC Health Serv Res 2007;7z208.

Abayomi O, Okubadejo N, Danesi M, Kolapo K, Osalusi B, Boyle B. Pattern and
Outcome of Neurological Manifestations of HIV /AIDS- A Review if 154 Cases in
a Nigerian University Teaching Hospital-A Preliminary Report African J of
Neurological Sciences 2005;24(1):29—36.

Belec PL, Testa J, Vohito MD, Gresenguet G, Martin MI, Tabo A, Di Costanzo
B, Georges AJ. [Neurologic and psychiatric manifestations of AIDS in Central
African Republic]. Bull Soc Pathol Exot Filiales 1989;82(3):297-307.

Clifford DB, Mitike MT, Mekonnen Y, Zhang J, Zenebe G, Melaku Z, Zewde A,
Gessesse N, Wolday D, Messele T, Teshome M, Evans S. Neurological
evaluation of untreated human immunodeficiency virus infected adults in
Ethiopia. J Neurovirol 2007; 13(1):67-72.

Howlett WP, Nkya WM, Mmuni KA, Missalek WR. Neurological disorders in
AIDS and HIV disease in the northern zone of Tanzania. Aids l989;3(5):289-96.

Perriens J H, Mussa M, Luabeya MK, Kayembe K, Kapita B, Brown C, Piot P,
Janssen R. Neurological complications of HIV-l-seropositive internal medicine
inpatients in Kinshasa, Zaire. J Acquir Immune Defic Syndr 1992;5(4):333-40.

Wong MH, Robertson K, Nakasujja N, Skolasky R, Musisi S, Katabira E,
McArthur J C, Ronald A, Sacktor N. Frequency of and risk factors for HIV
dementia in an HIV clinic in sub—Saharan Africa. Neurology 2007;68(5):350-5.

World Health Organization. Towards universal access by 2010: How WHO is
working with countries to scale-up HIV prevention, treatment, care and support.
Available online at: http://www.who.int/hiv/toront02006/towardsuniversalaccess.
Last accessed April 15, 2007.

Birbeck GL. Human immunodeficiency virus dementia patients in Africa: How
many? Who cares? And where to from here? J Neurovirol 2005;11 Suppl 3:30-3.

McGrath NM, Cooke GS. Frequency of and risk factors for Hiv dementia in an
Hiv clinic in sub-Saharan Africa. Neurology 2007;69(4):411-2; author reply 412-
3.

103

48.

49.

50.

51.

52.

53.

54.

55.

56.

Tozzi V, Balestra P, Bellagamba R, Corpolongo A, Salvatori MF, Visco-
Comandini U, Vlassi C, Giulianelli M, Galgani S, Antinori A, N arciso P.
Persistence of neuropsychologic deficits despite long-term highly active
antiretroviral therapy in patients with HIV-related neurocognitive impairment:
prevalence and risk factors. I Acquir Immune Defic Syndr 2007;45(2):174—82.

Albert SM, Marder K, Dooneief G, Bell K, Sano M, Todak G, Stern Y.
Neuropsychologic impairment in early HIV infection. A risk factor for work
disability. Arch Neurol 1995;52(5):525-30.

Benedict RH, Mezhir JJ, Walsh K, Hewitt RG. Impact of human
immunodeficiency virus type-l-associated cognitive dysfunction on activities of
daily living and quality of life. Arch Clin Neuropsychol 2000;15(6):535-44.

Hinkin CH, Hardy DJ, Mason KI, Castellon SA, Durvasula RS, Lam MN,
Stefaniak M. Medication adherence in HIV-infected adults: effect of patient age,
cognitive status, and substance abuse. Aids 2004;18 Suppl 1:Sl9-25.

Chesney M. Adherence to HAART regimens. AIDS Patient Care STDS
2003; 17(4): 169-77.

Maj M, J anssen R, Satz P, Zaudig M, Starace F, Boor D, Sughondhabirom B,
Bing EG, Luabeya MK, Ndetei D, et al. The World Health Organization's cross-
cultural study on neuropsychiatric aspects of infection with the human
immunodeficiency virus 1 (HIV-1). Preparation and pilot phase. Br J Psychiatry
1991;159:3516

Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method
for grading the cognitive state of patients for the clinician. J Psychiatr Res
1975;12(3): 189-98.

Chemer M, Cysique L, Heaton RK, Marcotte TD, Ellis RJ, Masliah E, Grant I.
Neuropathologic confirmation of definitional criteria for human
immunodeficiency virus-associated neurocognitive disorders. J Neurovirol
2007;13(1):23-8.

Ramirez M, Teresi J A, Holmes D, Gurland B, Lantigua R. Differential item
functioning (DIF) and the Mini-Mental State Examination (MMSE). Overview,
sample, and issues of translation. Med Care 2006;44(11 Suppl 3):S95-S 106.

104

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

Brito-Marques PR, Cabral—Filho JE. The role of education in mini-mental state
examination: a study in Northeast Brazil. Arq Neuropsiquiatr 2004;62(2A):206-
l 1.

Brito-Marques PR, Cabral-Filho JE. Influence of age and schooling on the
performance in a modified Mini-Mental State Examination version: a study in
Brazil northeast. Arq Neuropsiquiatr 2005;63(3A):583-7.

Fuzikawa C, Lima-Costa MF, Uchoa E, Shulman K. Correlation and agreement
between the Mini-mental State Examination and the Clock Drawing Test in older
adults with low levels of schooling: the Barnbui Health Aging Study (BHAS). Int
Psychogeriatr 2007: 1-1 1.

Ganguli M, Ratclif G. A Hindi Version of the MMSE: The development of a
cognitive screening instrument for a largel illiterate rural elderly population in
India. Int J Geriatr Psychiatry 1995;10:36-7.

Shaji S, Promodu K, Abraham T, Roy KJ, Verghese A. An epidemiological study
of dementia in a rural community in Kerala, India. Br J Psychiatry
1996;168(6):745-9.

Power C, Selnes OA, Grim J A, McArthur JC. HIV Dementia Scale: a rapid
screening test. J Acquir Immune Defic Syndr Hum Retrovirol l995;8(3):273-8.

Davis HF, Skolasky RL, Jr., Selnes OA, Burgess DM, McArthur JC. Assessing
HIV-associated dementia: modified HIV dementia scale versus the Grooved
Pegboard. AIDS Read 2002;12(l):29-3l, 38.

Sacktor NC, Wong M, Nakasujja N, Skolasky RL, Selnes OA, Musisi S,
Robertson K, McArthur J C, Ronald A, Katabira E. The International HIV
Dementia Scale: a new rapid screening test for HIV dementia. Aids

2005; 19( 13): 1367-74.

Stringer J S, Zulu I, Levy J, Stringer EM, Mwango A, Chi BH, Mtonga V, Reid S,
Cantrell RA, Bulterys M, Saag MS, Marlink RG, Mwinga A, Ellerbrock TV,
Sinkala M. Rapid scale-up of antiretroviral therapy at primary care sites in
Zambia: feasibility and early outcomes. J AMA 2006;296(7):782-93.

Ances BM, Ellis RJ. Dementia and neurocognitive disorders due to HIV-1
infection. Semin Neurol 2007;27(1):86-92.

105

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

Budka H, Wiley CA, Kleihues P, Artigas J, Asbury AK, Cho ES, Comblath DR,
Dal Canto MC, DeGirolami U, Dickson D, et al. HIV-associated disease of the
nervous system: review of nomenclature and proposal for neuropathology-based
terminology. Brain Pathol 1991;1(3): 143-52.

Nomenclature and research case definitions for neurologic manifestations of
human immunodeficiency virus-type 1 (HIV-1) infection. Report of a Working
Group of the American Academy of Neurology AIDS Task Force. Neurology
1991;41(6):778-85.

Brew BJ. HIV Neurology. Contemporary Neurology Series Oxford University
Press, Inc., 2001.

Catalan J. Psychiatric manifestations of HIV disease. Baillieres Clin Gastroenterol
1990;4(2):547-62.

Cysique LA, Maruff P, Brew BJ. The neuropsychological profile of symptomatic
AIDS and ADC patients in the pre-HAART era: a meta-analysis. J Int
Neuropsychol Soc 2006;12(3):368-82.

Maj M, Satz P, J anssen R, Zaudig M, Starace F, D'Elia L, Sughondhabirom B,
Mussa M, Naber D, Ndetei D, et al. WHO Neuropsychiatric AIDS study, cross-
sectional phase II. Neuropsychological and neurological findings. Arch Gen
Psychiatry 1994;51(1):51-6l. ‘

Robertson KR, Nakasujja N, Wong M, Musisi S, Katabira E, Parsons TD, Ronald
A, Sacktor N. Pattern of neuropsychological performance among HIV positive
patients in Uganda. BMC Neurol 2007;728.

Portegies P, Enting RH, de Gans J, Algra PR, Derix MM, Lange JM, Goudsmit J.
Presentation and course of AIDS dementia complex: 10 years of follow-up in
Amsterdam, The Netherlands. Aids 1993;7(5):669-75.

Sacktor NC, Bacellar H, Hoover DR, Nance-Sproson TE, Selnes OA, Miller EN,
Dal Pan GJ, Kleeberger C, Brown A, Saah A, McArthur J C. Psychomotor
slowing in HIV infection: a predictor of dementia, AIDS and death. J Neurovirol
1996;2(6):404—10.

Maschke M, Kastrup O, Esser S, Ross B, Hengge U, Hufnagel A. Incidence and
prevalence of neurological disorders associated with HIV since the introduction of

106

77.

78.

79.

80.

81.

82.

83.

84.

85.

highly active antiretroviral therapy (HAART). J Neurol Neurosurg Psychiatry
2000;69(3):376-80.

Chiesi A, Vella S, Dally LG, Pedersen C, Danner S, Johnson AM, Schwander S,
Goebel FD, Glauser M, Antunes F, et al. Epidemiology of AIDS dementia
complex in Europe. AIDS in Europe Study Group. J Acquir Immune Defic Syndr
Hum Retrovirol 1996;11(1):39-44.

Robertson KR, Kapoor C, Robertson WT, Fiscus S, Ford S, Hall CD. No gender
differences in the progression of nervous system disease in HIV infection. J
Acquir Immune Defic Syndr 2004;36(3):817-22.

Sacktor N, Nakasujja N, Skolasky R, Robertson K, Wong M, Musisi S, Ronald A,
Katabira E. Antiretroviral therapy improves cognitive impairment in HIV+
individuals in sub-Saharan Africa. Neurology 2006;67(2):311-4.

Maj M, Janssen R, Starace F, Zaudig M, Satz P, Sughondhabirom B, Luabeya
MA, Riedel R, Ndetei D, Calil HM, et al. WHO Neuropsychiatric AIDS study,
cross-sectional phase I. Study design and psychiatric findings. Arch Gen
Psychiatry l994;51(1):39—49.

Oliveira J F, Greco DB, Oliveira GC, Christo PP, Guimaraes MD, Oliveira RC.
Neurological disease in HIV-infected patients in the era of highly active
antiretroviral treatment: a Brazilian experience. Rev Soc Bras Med Trop
2006;39(2): 146-51.

Yepthomi T, Paul R, Vallabhaneni S, Kumarasamy N, Tate DF, Solomon S,
Flanigan T. Neurocognitive consequences of HIV in southern India: a preliminary
study of clade C virus. J Int Neuropsychol Soc 2006; 12(3):424-30.

Gupta JD, Satishchandra P, Gopukumar K, Wilkie F, Waldrop-Valverde D, Ellis
R, Ownby R, Subbakrishna DK, Desai A, Kamat A, Ravi V, Rao BS, Satish KS,
Kumar M. Neuropsychological deficits in human immunodeficiency virus type 1
clade C-seropositive adults from South India. J Neurovirol 2007;13(3):195-202.

Sebit MB. Neuropsychiatric HIV-1 infection study: in Kenya and Zaire cross-
sectional phase I and 11. Cent Afr J Med 1995;4l(10):315-22.

Odiase FE, Ogunrin OA, Ogunniyi AA. Memory performance in HIV/AIDS--a
prospective case control study. Can J Neurol Sci 2007;34(2): 154-9.

107

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

Pierotti C, Zabulon Y, Ayo H, Opira C, Drileba P, Bestetti A, Lazzarin A, Cinque
P. A prospective study on neuroAIDS with emphasis on cerebrospinal fluid
analysis in an Ugandan rural hospital. Journal of Neurovirology 2002;82SP224.

Modi G, Hari K, Modi M, Mochan A. The frequency and profile of neurology in
black South African HIV infected (clade C) patients - a hospital-based
prospective audit. J Neurol Sci 2007;254(1-2):60-4.

World Health Organization. Antiretroviral Therapy for HIV Infection in Adults
and Adolescents: Recommendations for a Public Health Approach. Available at:
http://www.who.int/hiv/pub/guidelines/adult/enl. Last accessed April 16, 2007.

World Health Organization. Interim proposal for a WHO Staging System for HIV
infection and Disease. Wkly Epidemiol Rec 1990;65(29):221—4.

Sacktor N, Tarwater PM, Skolasky RL, McArthur J C, Selnes OA, Becker J,
Cohen B, Miller EN. CSF antiretroviral drug penetrance and the treatment of
HIV -associated psychomotor slowing. Neurology 2001;57(3):542—4.

Marra CM, Lockhart D, Zunt JR, Perrin M, Coombs RW, Collier AC. Changes in
CSF and plasma HIV-1 RNA and cognition after starting potent antiretroviral
therapy. Neurology 2003;60(8): 1388-90.

Cysique LA, Maruff P, Brew BJ. Antiretroviral therapy in HIV infection: are
neurologically active drugs important? Arch Neurol 2004;6l(1 1): 1699-704.

Birbeck G, Chomba E, Ddumba E, Kauye F, Mielke J. Lack of appropriate
treatment for people with comorbid HIV /AIDS and epilepsy in sub-Saharan
Africa. Epilepsia 2007;48(7): 1424-5.

Hinkin CH, Castellon SA, Durvasula RS, Hardy DJ, Lam MN, Mason KI,
Thrasher D, Goetz MB, Stefaniak M. Medication adherence among HIV+ adults:

effects of cognitive dysfunction and regimen complexity. Neurology
2002;59( 12): 1944-50.

Osowiecki DM, Cohen RA, Morrow KM, Paul RH, Carpenter CC, Flanigan T,
Boland RJ. Neurocognitive and psychological contributions to quality of life in
HIV-l-infected women. Aids 2000;14(10):1327—32.

108

96.

97.

98.

99.

100.

101.

102.

103.

104.

Tozzi V, Balestra P, Galgani S, Murri R, Bellagamba R, Narciso P, Antinori A,
Giulianelli M, Tosi G, Costa M, Sampaolesi A, Fantoni M, Noto P, Ippolito G,
Wu AW. Neurocognitive performance and quality of life in patients with HIV
infection. AIDS Res Hum Retroviruses 2003;19(8):643-52.

Tozzi V, Balestra P, Murri R, Galgani S, Bellagamba R, Narciso P, Antinori A,
Giulianelli M, Tosi G, Fantoni M, Sampaolesi A, Noto P, Ippolito G, Wu AW.
Neurocognitive impairment influences quality of life in HIV-infected patients
receiving HAART. Int J STD AIDS 2004;15(4):254-9.

Ferrando S, van Gorp W, McElhiney M, Goggin K, Sewell M, Rabkin J. Highly
active antiretroviral treatment in HIV infection: benefits for neuropsychological
function. Aids 1998;12(8):F65-70.

Price RW, Yiannoutsos CT, Clifford DB, Zaborski L, Tselis A, Sidtis J], Cohen
B, Hall CD, Erice A, Henry K. Neurological outcomes in late HIV infection:
adverse impact of neurological impairment on survival and protective effect of
antiviral therapy. AIDS Clinical Trial Group and Neurological AIDS Research
Consortium study team. Aids 1999; 13(13): 1677-85.

Tozzi V, Balestra P, Galgani S, Narciso P, Ferri F, Sebastiani G, D'Amato C,
Affricano C, Pigorini F, Pau FM, De Felici A, Benedetto A. Positive and
sustained effects of highly active antiretroviral therapy on HIV-l-associated
neurocognitive impairment. Aids 1999;13(14): 1889-97.

Parsons TD, Braaten A], Hall CD, Robertson KR. Better quality of life with
neuropsychological improvement on HAART. Health Qual Life Outcomes
2006;4:l l.

Birbeck G, Chomba E, Atadzhanov M, Mbewe E, Haworth A. The social and
economic impact of epilepsy in Zambia: a cross-sectional study. Lancet N eurol
2007;6(1):39-44.

Meadows J, Le Marechal K, Catalan J. The burden of care: the impact of HIV -
associated dementia on caregivers. AIDS Patient Care STDS l999;13(1):47-56.

LoGiudice D, Hassett A. Uncommon dementia and the carer's perspective. Int
Psychogeriatr 2005;17 Suppl 1:8223-31.

109

105.

106.

107.

108.

109.

110.

111.

112.

113.

Harvath TA, Patsdaughter CA, Bumbalo JA, McCann MK. Dementia-related
behaviors in Alzheimer's disease and AIDS. J Psychosoc Nurs Ment Health Serv
1995;33(1):35-9.

Guinness L, Walker D, Ndubani P, J ama J, Kelly P. Surviving the impact of HIV -
related illness in the Zambian business sector. AIDS Patient Care STDS
2003;17(7):353-63.

Heaton RK, Velin RA, McCutchan J A, Gulevich SJ, Atkinson JH, Wallace MR,
Godfrey HP, Kirson DA, Grant I. Neuropsychological impairment in human
immunodeficiency virus-infection: implications for employment. HNRC Group.
HIV Neurobehavioral Research Center. Psychosom Med 1994;56(1):8-17.

Sacktor N, Nakasujja N, Robertson K, Clifford DB. HIV -associated cognitive
impairment in sub-Saharan Africa--the potential effect of clade diversity. Nat Clin
Pract Neurol 2007;3(8):436-43.

Laeyendecker 0, Li X, Arroyo M, McCuthan F, Gray R, Wawer M, Serwadda D,
Nalugoda F, Kigozi G, Quinn T. The Effect of HIV Subtype on Rapid Disease
Progression in Rakai, Uganda. 13th Conference on Retroviruses and
Opportunistic Infections. Denver, CO., 2006.

Vasan A, Renjifo B, Hertzmark E, Chaplin B, Msamanga G, Essex M, Fawzi W,
Hunter D. Different rates of disease progression of HIV type 1 infection in
Tanzania based on infecting subtype. Clin Infect Dis 2006;42(6):843-52.

Georgsson G, Houwers DJ, Palsson PA, Petursson G. Expression of viral antigens
in the central nervous system of visna-infected sheep: an immunohistochemical
study on experimental visna induced by virus strains of increased neurovirulence.
Acta Neuropathol 1989;77(3):299-306.

Masuda M, Remington MP, Hoffman PM, Ruscetti SK. Molecular
characterization of a neuropathogenic and nonerythroleukemogenic variant of
Friend murine leukemia virus PVC-211. J Virol 1992;66(5):2798—806.

Portis JL, Czub S, Garon CF, McAtee F]. N eurodegenerative disease induced by
the wild mouse ecotropic retrovirus is markedly accelerated by long terminal
repeat and gag-pol sequences from nondefective Friend murine leukemia virus. J
Virol 1990;64(4):l648-56.

110

114.

115.

116.

117.

118.

119.

120.

121.

122.

Rassart E, Nelbach L, J olicoeur P. Cas-Br—E murine leukemia virus: sequencing
of the paralytogenic region of its genome and derivation of specific probes to
study its origin and the structure of its recombinant genomes in leukemic tissues. J
Virol 1986;60(3):910-9.

Sharma DP, Zink MC, Anderson M, Adams R, Clements J E, J oag SV, Narayan
O. Derivation of neurotropic simian immunodeficiency virus from exclusively
lymphocytetropic parental virus: pathogenesis of infection in macaques. J Virol
1992;66(6):3550—6.

Szurek PF, Floyd E, Yuen PH, Wong PK. Site-directed mutagenesis of the codon

for Ile—25 in gPr80env alters the neurovirulence of tsl, a mutant of Moloney
murine leukemia virus TB. J Virol 1990;64(1 1):5241-9.

Ball SC, Abraha A, Collins KR, Marozsan AJ, Baird H, Quinones-Mateu ME,
Penn-Nicholson A, Murray M, Richard N, Lobritz M, Zimmerman PA,
Kawamura T, Blauvelt A, Arts EJ. Comparing the ex vivo fitness of CCRS-tropic
human immunodeficiency virus type 1 isolates of subtypes B and C. J Virol
2003;77(2): 1021-38.

The Central Intelligence Agency. The World Factbook: Zambia. 2007. Available
online at: https://www.cia.gov/library/publications/the-world-
factbook/geos/za.html. Last Accessed April 17, 2008.

Central Statistical Office, Zambia. National Statistics from the Government of
the Republic of Zambia. Census data from Zambia. Available at:
http://www.zamstats.gov.zm/census.php. Last accessed April 17, 2008.

United Nations Development Programme Human Development Report. Beyond
scarcity: Power, poverty and the global water crisis. 2006. Available at:
http://hdr.undp.org/en/reports/global/hdr2006l. Last accessed: April 17, 2008.

WHO/UNAIDS. Epidemiological Fact Sheet on HIV/AIDS and Sexually
Transmitted Infections: Zambia. 2004. Available online at:
ttp://www.who.int/3by5/support/en/EFS2004_zmb.pd. Last accessed: May 14,
2007.

Maj M, D'Elia L, Satz P, J anssen R, Zaudig M, Uchiyarna C, Starace F, Galderisi
S, Chervinsky A. Evaluation of two new neuropsychological tests designed to
minimize cultural bias in the assessment of HIV-1 seropositive persons: a WHO
study. Arch Clin Neuropsychol 1993;8(2): 123-35.

111

123.

124.

125.

126.

127.

128.

129.

130.

131.

CDC/WHO. Acquired Immunodeficiency Syndrome (AIDS). 1987 Revision of
CDC/WHO Case Definition for AIDS. Wkly Epidem Rec 1988;1:2-8.

Miller WC, Thielman NM, Swai N, Cegielski JP, Shao J, Manyenga D, Mlalasi J,
Lallinger GJ. Diagnosis and screening of HIV/AIDS using clinical criteria in
Tanzanian adults. J Acquir Immune Defic Syndr Hum Retrovirol 1995;9(4):408-
14.

Gladstone M, Lancaster G, Jones A, Maleta K, Mtitimila E, Ashom P, Smyth R.
Can Western developmental screening tools be modified for use in a rural
Malawian setting? Arch Dis Child 2007;22:22.

Dugbartey AT, Townes BD, Mahurin RK. Equivalence of the Color Trails Test
and Trail Making Test in nonnative English-speakers. Arch Clin Neuropsychol
2000;15(5):425-31.

Hsieh SLJ, Tori CD. Normative data on cross-cultural neuropsychological tests
obtained from Mandarin-speaking adults across the life span. Arch Clin
Neuropsychol 2007;22(3):283-296.

Baldewicz 'I'T, Leserrnan J, Silva SG, Petitto JM, Golden RN, Perkins DO,
Barroso J, Evans DL. Changes in neuropsychological functioning with
progression of HIV -1 infection: results of an 8-year longitudinal investigation.
AIDS Behav 2004;8(3):345-55.

Heaton RK, Grant I, Butters N, White DA, Kirson D, Atkinson J H, McCutchan
J A, Taylor M], Kelly MD, Ellis RJ, et al. The HNRC 500--neuropsychology of
HIV infection at different disease stages. HIV Neurobehavioral Research Center.
J Int Neuropsychol Soc 1995; 1(3):23 1-51.

Van Gorp WG, Miller EN, Satz P, Visscher B. Neuropsychological performance
in HIV -1 immunocompromised patients: a preliminary report. J Clin Exp
Neuropsychol 1989;1 1(5):763-73.

Kober K, Van Damme W. Scaling up access to antiretroviral treatment in
southern Africa: who will do the job? Lancet 2004;364(9428): 103-7.

112

 

 

 

Il‘

1L lrlll .r’EH‘

.‘3 I It

N
.nu

11

1