If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Infectious and Tropical Diseases Unit, Department of Medicine, Bayero University Kano, PMB 3011, Aminu Kano Teaching Hospital, Kano, Kano State, Nigeria
Infectious and Tropical Diseases Unit, Department of Medicine, Bayero University Kano, PMB 3011, Aminu Kano Teaching Hospital, Kano, Kano State, Nigeria
Infectious and Tropical Diseases Unit, Department of Medicine, Bayero University Kano, PMB 3011, Aminu Kano Teaching Hospital, Kano, Kano State, Nigeria
To estimate the burden of HIV neurocognitive impairment (NCI) among adult patients on and off antiretroviral therapy (ART) in Sub-Saharan Africa.
Methods
Estimates were derived from a random effects meta-analysis of prospective studies reporting HIV status, utilization of ART, and the presence of NCI determined using the International HIV Dementia Scale.
Results
Sixteen studies with quality data from seven countries in Sub-Saharan Africa up to June 2012 were included. Among HIV patients, the frequency of NCI pre-ART was 42.37% (95% confidence interval (CI) 32.18–52.56%), and among those on ART for ≥6 months was 30.39% (95% CI 13.17–47.61%). Respective NCI estimates in studies from Uganda were 46.49% (95% CI 30.62–62.37%) and 28.50% (95% CI −1.31–58.30%). NCI was more common among patients with a concomitant psychiatric ailment. HIV-positive patients compared to HIV-negative controls were predisposed to NCI (odds ratio (OR) 6.49, 95% CI 1.68–25.08); the estimated unadjusted attributable risk of HIV infection leading to NCI was 85%. Meta-regression showed no associations between age, gender, CD4 cell counts, or years of education with NCI. Patients on ART were less likely to have NCI compared to HIV-infected pre-ART patients, with OR 0.36 (95% CI 0.19–0.69). In longitudinal studies with the same patients followed before and at ≥6 months after ART, the OR of NCI after ART compared to pre-ART was 0.23 (95% CI 0.14–0.37). The combined burden of NCI among pre-ART and on-ART patients in Sub-Saharan Africa was estimated at 8 121 910 (95% CI 5 772 140–10 471 680). No publication bias was observed, although residual confounding from differing environmental factors, stages of HIV infection, and viral clades might be a limitation.
Conclusions
HIV strongly predisposes to NCI leading to a huge burden in Sub-Saharan Africa, and scale-up of ART can substantially reduce it.
Early in the epidemic, several studies assessed the presence of NCI among infected adults, with variable results arising from differences regarding the dominant viral clade, methods used for NCI detection, stage of infection, and whether patients were on or off antiretroviral therapy (ART).
Successive attempts have been made to harmonize and unify the definition and nosology of NCI, eventually culminating in a definition formulated by the American Academy of Neurology in 2007 (AAN2007) referred to as HIV Associated Neurocognitive Deficits (HAND).
This categorizes subjects into ‘normal asymptomatics’, ‘asymptomatic neurocognitive impairment’ (ANI), ‘mild neurocognitive deficit’ (MND), and the severest form referred to as ‘HIV-associated dementia’ (HAD),
previously referred to as ‘AIDS dementia complex’ (ADC). As the neuropathological correlate of NCI is thought to be primarily of sub-cortical origin, assessment methods used in Sub-Saharan Africa for cortical diseases may be insensitive for evaluating it. Thus, there have been attempts to adapt and standardize more sensitive assessment tools and test batteries, mindful of local cultures, ease of administration, interpretation, and comparability between populations. This has led to the use of the extensive multi-domain test batteries largely in industrialized countries (e.g., the full or modified versions of the HIV Neurobehavioral Research Center, University of California San-Diego (HNRC) battery) and the less extensive but more easily administered screening tools such as the International HIV Dementia Scale (IHDS) mostly used in developing countries, especially in Sub-Saharan Africa.
The IHDS examines the three main domains of motor speed, psychomotor speed, and memory registration/recall. It consists of three subtests: timed finger tapping, timed alternating sequence test, and registration/recall of four items at 2 min.
A previous meta-analysis and a systematic review addressed detailed neuropsychological profiles and NCI among HIV-infected patients before and after the highly active antiretroviral therapy (HAART) era.
In these reviews, data presented were extracted from studies conducted worldwide with varied tools, and no point-estimates could be derived thereby limiting their epidemiological significance, especially in Sub-Saharan Africa. In this study, a systematic review and meta-analysis of studies that used IHDS was conducted to obtain more precise prevalence estimates of NCI, approximate the burden of NCI, and approximate the potential impact of ART on NCI among HIV-infected adults in Sub-Saharan Africa.
2. Materials and methods
This study was conducted and reported in accordance with the following available guidelines: Meta-analysis of Observational Studies in Epidemiology (MOOSE) and Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) statements.
Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.
To build up the meta-analysis and systematic review, we searched for relevant English-language papers in MEDLINE, the Cochrane Database, African Journals On-Line (AJOL) database, Global Health, PsycINFO, existing systematic reviews, specialty journals, other search engines, e.g. Google, and several websites. Medical subject heading (MeSH) terms were used in the search. ‘Africa’, ‘AIDS’, ‘adults’, cognition’, ‘cognitive deficit’, ‘cognitive impairment’, ‘dementia’, ‘depression’, ‘HIV’, ‘impairment’, ‘memory’, ‘mental’, ‘neuro-cognitive impairment’, ‘neuro-psychological impairment’, ‘neuro-psychiatric’, ‘psychological impairment’, ‘Sub-Saharan Africa’, and ‘studies’ were used in various combinations as primary search keywords, with no period restriction until the end of June 2012. The review was restricted to studies that defined NCI using IHDS among adults aged ≥18 years, with HIV-1 (excluding HIV-2) infections, and to Sub-Sahara Africa. We also used our knowledge, experiences, and discussions with colleagues and recognized experts, as well as previous publications on the subject. When required, we contacted authors and also manually searched the reference lists of all identified publications and recent systematic reviews. Books, chapters, and review articles on the subject were also consulted.
2.1 Selection criteria
Two reviewers independently assessed each potentially relevant study for eligibility. Disagreements were resolved by consultation with a third reviewer. The full texts of all accessible English-language studies evaluating or reporting on NCI among HIV-infected adults in Sub-Saharan Africa were selected, applying the following inclusion criteria: (1) prospective observational studies that used IHDS for detecting NCI; and (2) when HIV status and ART use was stated or could be discerned. Studies were excluded if they did not fulfill the inclusion criteria (e.g., if they did not use IHDS or defined NCI clinically), or lacked or had insufficient data or desired information.
2.2 Data extraction
Data were dredged, ascertained, and obtained on a standardized form used to record relevant items and entered into a database; the following were recorded: study characteristics, authors, year of publication, country, study design, setting, test battery, cut-off scores for defining NCI, age of participants, gender, HIV status and stage of infection, formal education, CD4 cell counts, proportion of participants with NCI, duration of ART, viral clades, potential confounders, sources of bias, quality measures, and comments. Another reviewer checked the extracted data for accuracy.
2.3 Quality assessment
The study quality was assessed using a modified version of the Downs and Black checklist,
The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions.
and assessment included other items also known to be associated with study quality.
2.4 Data analysis
The prevalence of NCI as a percentage with the respective 95% confidence interval (95% CI) was calculated for each study, among HIV-infected adults on or off ART and among HIV-uninfected controls. We also determined the odds of NCI as the number of patients who developed it divided by the number of HIV-infected patients who did not develop it for HIV-infected adults on/off ART and for HIV-negative controls for relevant studies. When no or zero NCI cases were found in the numerator, we added 0.5 to the numerator and denominator of both groups when estimating the odds.
We did not assume zero NCI cases when this was not reported among controls, only when it was stated. The odds ratio (OR) and respective 95% CI for NCI in the HIV-positive compared to the HIV-negative control group and for those on ART compared to those off ART were calculated. The log OR and the standard error of log OR were computed for the respective studies. Meta-analyses were then conducted for prevalence estimates and for the OR from comparative studies.
Because of the differences expected between studies, we combined results and obtained meta-analysis estimates using a random effects model (REM). We evaluated for statistical heterogeneity by conducting tests of between-study heterogeneity and I2 statistics with I2 > 50% denoting substantial heterogeneity. Sensitivity analyses primarily consisted of investigating the influence of a single study on the overall summary estimate.
Restricted scenario analyses were then performed on data derived from studies with distinct characteristics. Publication bias and small study effect were assessed using adjusted rank correlation tests and the regression asymmetry test, termed Begg's and Egger's tests.
publication bias was considered to exist only if detected in both tests. The crude attributable risk and population attributable risk percent were computed using the REM-derived OR with Levin's formula.
We also performed univariate, weighted, least-squares meta-regressions to identify study-level characteristics (mean or median age of participants, years of formal education, CD4 cell counts, and proportion of female participants) associated with the risk of NCI.
The estimates derived for the prevalence of NCI among the HIV-infected on and off ART were applied to the total adult HIV-infected populations on and off ART in Sub-Saharan Africa, respectively.
The two were then summed to give an estimated burden of NCI in Sub-Saharan Africa. All analyses were carried out using Stata version 11.0 (Stata Corp., College Station, TX, USA).
3. Results
A flow diagram of the process of article selection for the meta-analysis and review is shown in Figure 1.
Figure 1Flow diagram of the process of article selection for the meta-analysis and review.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
Association between apolipoprotein E4 genotype and human immunodeficiency virus-associated dementia in younger adults starting antiretroviral therapy in South Africa.
Reliability of the lay adherence counsellor administered Substance Abuse and Mental Illness Symptoms Screener (SAMISS) and the International HIV Dementia Scale (IHDS) in a primary care HIV clinic in Cape Town, South Africa.
The International HIV Dementia Scale is a useful screening tool for HIV-associated dementia/cognitive impairment in HIV-infected adults in Yaoundé, Cameroon.
Association between apolipoprotein E4 genotype and human immunodeficiency virus-associated dementia in younger adults starting antiretroviral therapy in South Africa.
The International HIV Dementia Scale is a useful screening tool for HIV-associated dementia/cognitive impairment in HIV-infected adults in Yaoundé, Cameroon.
The International HIV Dementia Scale is a useful screening tool for HIV-associated dementia/cognitive impairment in HIV-infected adults in Yaoundé, Cameroon.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
Reliability of the lay adherence counsellor administered Substance Abuse and Mental Illness Symptoms Screener (SAMISS) and the International HIV Dementia Scale (IHDS) in a primary care HIV clinic in Cape Town, South Africa.
Characteristics of the studies and quality indices assessed are presented in Table 1, Table 2. The studies were from seven countries in Sub-Saharan Africa, comprising South Africa (n = 4), Uganda (n = 6), Zambia (n = 2), Botswana (n = 1), Cameroon (n = 1), Malawi (n = 1), and Nigeria (n = 1). These 16 studies had point-prevalence for NCI in HIV-infected adult patients pre-ART (n = 7),
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
Reliability of the lay adherence counsellor administered Substance Abuse and Mental Illness Symptoms Screener (SAMISS) and the International HIV Dementia Scale (IHDS) in a primary care HIV clinic in Cape Town, South Africa.
three of the latter studies were longitudinal and had NCI prevalence computed for the same subjects at both pre-ART and ≥6 months after commencement of ART
(Table 1). All five studies showed lower prevalence of NCI after ≥6 months on ART compared to off- or pre-ART (Table 1). All the studies used an IHDS cut-off of ≤10 for the definition of NCI except two studies that used ≤9.5 and <10.5, respectively,
Reliability of the lay adherence counsellor administered Substance Abuse and Mental Illness Symptoms Screener (SAMISS) and the International HIV Dementia Scale (IHDS) in a primary care HIV clinic in Cape Town, South Africa.
Further to using IHDS with a cut-off of ≤10, four studies used additional tests, and their combined assessments were used to assign a Memorial Sloane Kettering (MSK) dementia stage.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
Six of the 16 studies had a healthy/HIV-negative control group; NCI was detected among controls in three of the studies, while the remaining three studies only generated normative data and cut-offs for defining NCI among the controls (Table 1). Most of the studies were cross-sectional (n = 12) and the remaining studies were either cohort (n = 1) or longitudinal (n = 3).
Table 1Characteristics of included studies on neurocognitive impairment in HIV-1-infected adults in Sub-Saharan Africa
Study No.
Author, year, Ref.
Country
Study design
Symptom status
CD4 cell count
Demographics
Prevalence of NCI between groups (HIV-pos/neg or ART+/−) (n/N)
Reliability of the lay adherence counsellor administered Substance Abuse and Mental Illness Symptoms Screener (SAMISS) and the International HIV Dementia Scale (IHDS) in a primary care HIV clinic in Cape Town, South Africa.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
ART, antiretroviral therapy; ART+, on ART for at least 6 months; ART−, not on ART; Asymp, asymptomatic; C-S, cross-sectional; IHDS, International HIV Dementia Scale; NR, not reported; OIs, opportunistic infections; Symp, symptomatic.
Data are mean (SD) or median (IQR). Cofounders include ethanol, depression, psychiatric diseases, opportunistic infections, low Karnofsky score, and comorbid conditions. Some studies did not exclude all of them.
Reliability of the lay adherence counsellor administered Substance Abuse and Mental Illness Symptoms Screener (SAMISS) and the International HIV Dementia Scale (IHDS) in a primary care HIV clinic in Cape Town, South Africa.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
The proportion of female participants reported in 15 studies ranged from 50% to 79.2%. The mean or median ages of participants in 15 studies ranged from 29.75 to 40 years. The mean or median duration of formal education among participants in nine studies ranged from 7.2 to 10.5 years. Mean or median CD4 cell counts were reported in 10 studies and ranged from 35 to 360.4 cells/ml. Participants recruited in the studies comprised asymptomatic HIV patients (n = 1), a mixed group of asymptomatic and symptomatic infections (n = 3), and symptomatic stable patients (n = 12). Four studies recruited participants with CD4 cell counts <200 cells/ml.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
Several studies evaluated the functional activity and performance status of participants using the Karnofsky scale and excluded those with Karnofsky scale <50%. HIV viral load was reported in studies from Botswana and Nigeria.
Studies addressed other potential confounders and some excluded participants with psychiatric morbidity, ethanol/substance use, and possible central nervous system (CNS) opportunistic infections (OIs) (Table 1, Table 2).
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
The REM estimates of NCI among HIV-infected adults pre-ART from 12 studies was 42.37% (95% CI 32.18–52.56%) (Figure 2). There was no publication bias (Begg's test, p = 0.945; Egger's test, p = 0.827) (Figure 7a). Sensitivity analysis showed that no study unduly influenced the findings (Figure 3).
Figure 2Prevalence (%) of neurocognitive impairment in HIV-infected adults in Sub-Saharan Africa (pre-ART or off ART).
Figure 3Sensitivity analyses showing effect of dropping individual studies on pooled REM prevalence (%) estimates of remaining 11 studies (combined estimates from the 12 comparisons are depicted by the 3 vertical lines).
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
the REM pooled estimates of NCI among infected patients pre-ART was 46.49% (95% CI = 30.62–62.37%). While the corresponding estimates was 28.50% (95% CI = 1.31–58.30%) for patients on ART from 4 studies. The prevalence of NCI among ART-naïve patients reported by two studies each from South Africa and Zambia is given in Table 1.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
where those with active psychiatric disease, use of anti-psychiatric medication, alcohol or substance use were excluded, the prevalence estimate of NCI was 35.66% (95% CI 24.44–46.87%). The pooled estimate from three studies
that included participants with psychiatric diseases or substance use was 62.56% (95% CI 39.28–85.84%).
Meta-regression was conducted only on studies reporting NCI pre-ART or off-ART. In the analysis, study-specific NCI estimates showed no significant associations with four study-level characteristics: mean/median age of patients in years ranging from 29.5 to 38.1 years had a slope coefficient of −4.95 (95% CI −10.07 to 0.16, p = 0.056) in 12 studies; the slope of the line in the relationship with the proportion of women across a wide range from 50% to 79.2% had a coefficient of 1.06 (95% CI −4.49 to 6.60, p = 0.680) in 12 studies; the slope of the line in the relationship with mean/median formal years spent in education across a wide range from 7.2 to 10.5 years had a coefficient of 2.49 (95% CI −14.95 to 19.94; p = 0.739) in eight studies; and the slope of the line in the relationship with mean/median CD4 cell counts across a wide range from 35 to 355 cells per ml had a coefficient of −0.13 (95% CI −0.29 to 0.035; p = 0.103) in eight studies (Figure 4).
Figure 4Meta-regression of NCI in HIV infected adults (pre-ART) by study-level parameters.
When compared to HIV-negative adult controls, the OR for developing NCI in HIV-infected adults was 6.49 (95% CI 1.68–25.08) (Figure 5). There was no publication bias (Begg's test, p = 0.602; Egger's test, p = 0.469.). Sensitivity analysis showed none of the studies unduly weighed on the estimates. The estimated unadjusted attributable risk of HIV infection leading to NCI was 85% and the population attributable risk percent was estimated as 9.89% assuming a conservative 2.0% adult HIV seroprevalence in Sub-Saharan Africa.
Reliability of the lay adherence counsellor administered Substance Abuse and Mental Illness Symptoms Screener (SAMISS) and the International HIV Dementia Scale (IHDS) in a primary care HIV clinic in Cape Town, South Africa.
was 30.39% (95% CI 13.17–47.61%) (Figure 6). There was no publication bias (Begg's test, p = 0.175; Egger's test, p = 0.287). The sensitivity analysis showed that no study unduly influenced the findings.
Figure 6Prevalence (%) of Neuro-Cognitive Impairment (NCI) in HIV infected adults on ART in 9 studies in SSA.
When five studies reporting on NCI among those on ART compared to pre-ART were analyzed, the pooled REM OR was 0.36 (95% CI 0.19–0.69) with no publication bias (Begg's test, p = 0.462; Egger's test, p = 0.914) (Figure 7, Figure 8). Three of the five were longitudinal studies with the same patients followed before and at ≥6 months after ART yielding an OR of NCI after ART compared to pre-ART of 0.23 (95% CI 0.14–0.37).
Figure 7Publication bias on studies of NCI: (a) among patients pre- or off- ART by Egger's plot; (b) comparing [Odds Ratio(95% CI)] of those on-ART to those off-ART by Begg's funnel plot.
The burden of NCI was obtained as sum of product of 42.37% (95% CI 32.18–52.56%) of HIV-infected adults not on ART (assumed 16 300 000; UNAIDS 2009) and the product of 30.39% (95% CI 13.17–47.61) of HIV-infected on ART (assuming 80% of the 5 million patients on ART are adults, i.e., 4 million adults (ONE International)).
Thus, the combined burden of NCI among pre-ART and on-ART patients totaled 8 121 910 (95% CI 5 772 140–10 471 680) in Sub-Saharan Africa.
4. Discussion
This study confirmed that HIV infection is a major predisposing factor for NCI, leading to a huge burden of over 8.1 million adult patients in Sub-Saharan Africa. It reaffirmed that ART confers substantial benefit against NCI, lowering it by over 60%. Indeed, the burden may be an underestimation, as a higher prevalence of NCI has been reported in some studies,
and the exclusion of children, adolescents, and HIV-2 infections, yielded lower burden estimates. The overall estimates are similar to the 8 million derived from studies conducted in Uganda.
In Sub-Saharan Africa, patients generally present with a more advanced symptomatic infection, often with anemia, poor performance status, wasting and severe immunosuppression with low CD4 cell counts, conditions known to predispose to NCI.
Perhaps this accounts for the higher NCI prevalence recorded compared to the cumulative lifetime risk of developing NCI from well-conducted cohort studies in developed countries, as in the USA where 15–20% was observed in the Multicenter AIDS Cohort Study in the pre-HAART era.
A previous meta-analysis reported on the indices traditionally used in psychological studies, such as quantification of the effect size and assigning of impairment ratings from comparisons with normative data among controls.
This was not adopted in this review. The strengths of our approach include an ability to derive an accurate point prevalence and therefore burden of NCI. This would not have been feasible with the former approach. Several studies and previous systematic reviews have suggested that ART has beneficial effects on NCI, but to our knowledge no previous report has quantified the extent of the benefit.
Our analysis suggests that ART improves NCI by 63% when compared to similar subjects not on ART and by 77% when the same patients are tested before and at least 6 months after commencement of ART. Such longitudinal studies have the advantage of allowing for within-subject assessments. However, it is likely the actual benefit could be slightly less, as serially conducted testing in the same patient population introduces a practice effect and apparent improvements in performance. Furthermore, as HIV-infected persons age, residual long-term effects can still be seen in the brain structure which likely stem from damage done prior to ART. As antiretroviral (ARV) drugs have different CNS penetrating effectiveness (CPE), it is likely the effect of ART regimens may also be variable. However, our results strongly advocate for scale-up of HAART as an intervention for reducing the massive burden of NCI in the region.
To improve population homogeneity, analysis was conducted on adults only and NCI was defined with the same test, the IHDS. However, there are several sources of heterogeneity including differences resulting from differing circulating HIV-1 viral clades, and differing study designs and data quality. However attempts were made at deriving estimates from more homogeneous populations, e.g. from individual countries (Uganda), thereby reducing differences from viral clades and cultural/environmental effects. Secondly, the prevalence of NCI for those with and without psychiatric diseases and alcohol/substance use was obtained separately to reduce their confounding effects. NCI was higher in studies with participants with psychiatric diseases or substance use.
It has been suggested that sole psychiatric presentation may be the early phase of NCI. Indeed initial symptoms may be subtle and are therefore easily overlooked and can be misdiagnosed as depression or mental retardation.
Meta-regression of study-specific NCI prevalence conducted to explore associations with study-level parameters failed to confirm significant relationships. However, there was a non-significant decreasing trend in NCI with increasing age and CD4 cell counts. Education increases the so-called cognitive ‘cerebral reserve’,
but duration of education had no relationship to NCI. Similarly, no association between gender and NCI was shown, although studies have found it to be higher among females
Consistent with previous reports, estimates derived from studies in Uganda with more homogeneous viral populations and a predominance of clades D/A yielded a higher prevalence of NCI. It has been suggested that clade D viruses have more neurotropic activity and generally have a more aggressive natural history.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
with circulating recombinant form (CRF) 02AG (CRF_02AG) and clade A respectively yielded estimates of 29% and 21%, compared to 89% reported in clade D-infected patients in Uganda. Indeed, the preliminary analysis from Nigeria, involving sequencing of partial pol amplicons from viral isolates, revealed that two of three patients with NCI were infected with subtype G virus and one with the CRF02_AG, while all four individuals without NCI were infected with CRF_02AG.
The findings should be interpreted in the light of several caveats. Firstly, IHDS is a screening test for NCI. Indeed, in Guinea-Bissau it has been suggested that IHDS may be a less effective screening tool for HAND/NCI, especially in settings of lower educational attainment.
Furthermore it does not assess depression and other important domains like executive functions and may underestimate the burden of NCI. Investigators in Cameroon using an extensive battery of 19 tests detected significantly lower performance on executive functions as evaluated using three tests (Category Test, Color Trails II, and Wisconsin Card Sorting Test-64), and South African patients with HAD, who screened negative on the IHDS, also performed poorly on some tests of executive functions.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
but the addition of assessments missing in IHDS (e.g., executive functions) in combination will definitely improve its accuracy. Combinations of assessments and tests that will yield accurate results while retaining the balance of brevity and simplicity should be explored. Secondly, only about half of the studies included in the review excluded (or declared) participants with active OIs or substance use. Few of the studies excluded those with neurological comorbidities and none utilized neuroimaging for excluding alternative diagnoses to NCI. This could have overestimated NCI, but provides some advantages from the point of view of generalizability of findings to diverse routine patient populations such as exist in HIV clinics in Sub-Saharan Africa.
Publication bias did not affect the results as our analyses using two concurrent approaches excluded it. Bias from non-inclusion of literature in other languages is unclear but likely to be minimal.
In conclusion, HIV strongly predisposes to NCI, leading to a huge burden in Sub-Saharan Africa. Early commencement of ART can improve NCI substantially, and its scale-up as an intervention for reducing the huge burden in the region is strongly advocated.
Conflict of interest: We declare that we have no conflicts of interest.
Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.
The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions.
HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda.
Association between apolipoprotein E4 genotype and human immunodeficiency virus-associated dementia in younger adults starting antiretroviral therapy in South Africa.
Reliability of the lay adherence counsellor administered Substance Abuse and Mental Illness Symptoms Screener (SAMISS) and the International HIV Dementia Scale (IHDS) in a primary care HIV clinic in Cape Town, South Africa.
The International HIV Dementia Scale is a useful screening tool for HIV-associated dementia/cognitive impairment in HIV-infected adults in Yaoundé, Cameroon.
00223-3&id=gr1.jpg){kind=link}
00223-3&id=gr2.jpg){kind=link}
00223-3&id=gr3.jpg){kind=link}
00223-3&id=gr4.jpg){kind=link}
00223-3&id=gr5.jpg){kind=link}
00223-3&id=gr6.jpg){kind=link}
00223-3&id=gr7.jpg){kind=link}
00223-3&id=gr8.jpg){kind=link}