The ability of WHO clinical staging to predict CD4 cell counts of 200 cells/μl or less was evaluated among 1221 patients screened for antiretroviral therapy (ART). Sensitivity was 51% and specificity was 88%. The positive predictive value was 64% and the negative predictive value was 81%. Clinical criteria missed half the patients with CD4 cell counts of 200 cells/μl or less, highlighting the importance of CD4 cell measurements for the scale-up of ART provision in resource-limited settings.
aRakai Health Sciences Programme, Rakai, Uganda
bJohns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
cInstitute of Public Health, Makerere University, Kampala, Uganda
dthe National Institute of Allergy and Infectious Diseases, Bethesda, Maryland, USA
eJohns Hopkins School of Medicine, Baltimore, Maryland, USA.
Received 29 January, 2007
Accepted 14 February, 2007
The World Health Organization (WHO) guidelines recommend that antiretroviral therapy (ART) be started for all patients with CD4 cell counts of 200 cells/μl or less irrespective of clinical stage or total lymphocyte counts less than 1200 cells/μl in symptomatic patients. When CD4 cell counts or total lymphocyte counts cannot be performed, patients with WHO clinical stage III or IV disease may be offered ART . In sub-Saharan Africa, when access to CD4 cell counts or total lymphocyte counts is limited, clinical criteria are frequently used to initiate therapy. It is not known, however, how well the clinical criteria predict a CD4 cell count of 200 cells/μl or less. Studies have documented a higher mortality risk or development of AIDS among patients who started ART with very low CD4 cell counts [2,3].
If WHO clinical criteria lag behind immunological criteria, this could potentially result in treatment delays and poorer outcomes among patients started on ART based on clinical criteria alone.
There are limited data to evaluate the utility of the WHO clinical criteria to predict CD4 cell counts of 200 cells/μl or less. One hospital-based study in Cambodia compared the use of clinical criteria alone versus a combination of the 2003 WHO clinical and immunological criteria, and showed high sensitivity (96%) and accuracy (89%) . Findings from that study only apply to very sick patient populations, because approximately 86% of the study population had clinical stage III or IV disease. Another study conducted in primary healthcare facilities in south Africa showed that 24% of patients classified as WHO stage 1 and 46% classified as stage II had CD4 cell counts of 200 cells/μl or less . A substantial proportion of stage I and II cases would thus not have initiated ART if therapy was based on clinical criteria alone. Clinical assessment in that study was primarily done by nurses and there was a time lag of up to 90 days between clinical assessment and CD4 cell counts, potentially leading to misclassification.
We analysed data on patients seen in the Rakai Health Sciences Programme (RHSP) community-based ART programme to assess whether WHO clinical stages III or IV could be used to identify individuals with CD4 cell counts of 200 cells/μl or less. The majority of patients were participants in the population-based Rakai Community Cohort Study (RCCS) and thus were likely to be representative of the HIV-infected population in Rakai.
In 2004, RHSP started the provision of ART in rural Rakai district in southwestern Uganda, with funding from the Presidential Emergency Plan for AIDS Relief through the Centers for Disease Control and Prevention, Uganda. Patients for this evaluation participated in RCCS and the ART-related clinical study evaluating the epidemiological, demographic and behavioural effects of antiretroviral drugs in Rakai. Through the RCCS, serosurveys are conducted in 48 communities every 12–16 months among 13 000 participants aged 15–49 years. Approximately 94% of participants provided blood for HIV testing and approximately 80% agreed to receive their HIV results and post-test counseling. HIV-positive participants were referred to 16 community-based clinics, to screen for ART eligibility. Patients enrolled in the clinics were invited to participate in the ART-related clinical study.
A fluorescence-activated cell sorter count (Becton Dickson, Franklin Lakes, New Jersey, USA) located at the RHSP facility in Rakai was used for CD4 cell counts. Clinical staging was conducted by clinicians who were trained in HIV care at the Infectious Diseases Institute in Kampala, Uganda; additional weekly technical support was provided on-site in Rakai by an expert infectious disease physician from the Infectious Diseases Institute. Clinical assessment was carried out 2 weeks after the CD4 cell counts.
The RCCS and ART-related clinical studies were approved by the Ugandan Science and Ethics committee and the US Western internal review board. Patients provided consent to have their clinical data used for research.
Sensitivity and specificity were calculated by computing 2 × 2 tables, comparing clinical stage III or IV with CD4 cell counts of 200 cells/μl or less (as the reference). The Wilson score method without continuity correction was used to calculate 95% confidence intervals (CI) around the estimates. The analysis used STATA software (release 8.0; Stata Corp., College Station, Texas, USA).
As shown in Table 1, a total of 1221 patients were evaluated. The median age was 34 years (interquartile range 29–40), and 65% of patients were women. Overall, 24% of patients had clinical stage III or IV, 16% had CD4 cell counts of 100 cells/μl or less, 14% had CD4 cell counts between 100 and 201 cells/μl, and 70% had CD4 cell counts above 200 cells/μl.
The sensitivity of clinical stage III or IV to detect individuals with CD4 cell counts of 200 cells/μl or less was 51% (95% CI 46–56%), specificity was 88% (95% CI 85–90%), the positive predictive value was 64% (186/292, 95% CI 58–69%), and the negative predictive value was 81% (751/929, 95% CI 78–83%). Sensitivity was higher (64%, 95% CI 57–70%) for the subgroup of patients with CD4 cell counts of 100 cells/μl or less compared with patients with CD4 cell counts of 100–200 cells/μl (37%, 95% CI 30–44%; P < 0.001).
The WHO clinical criteria missed approximately half the patients with CD4 cell counts of 200 cells/μl or less. The proportion of patients missed was highest (63%) for patients with CD4 cell counts between 100 and 201 cells/μl and lowest (36%) for patients with CD4 cell counts of 100 cells/μl or less.
Our findings support previous evidence that WHO clinical criteria provide an unsatisfactory screening tool for ART eligibility in settings in which CD4 cell technologies are unavailable. The exclusive use of clinical criteria would better identify patients with CD4 cell counts of 100 cells/μl or less, but this would translate into the initiation of ART in patients with advanced disease and potentially led to poor treatment outcomes.
As the patient population with advanced HIV declines, as a result of mortality and the availability of antiretroviral drugs, the clinical criteria will become less sensitive and thus less useful as a screening tool.
Our findings have important implications for the scale-up of ART in resource-limited settings in which the exclusive use of the clinical criteria could deny or delay ART to approximately half the patients with CD4 cell counts of 200 cells/μl or less, and patients who eventually start ART would potentially be at an increased risk of poor treatment outcomes.
In conclusion, CD4 cell screening facilities are urgently needed as an adjunct to clinical criteria, to facilitate screening for ART in resource-limited settings.
Sponsorship: This evaluation was supported by grants from the Presidential Emergency Plan for AIDS Relief, through the Centers for Disease Control and Prevention, Uganda, and the National Institute of Allergy and Infectious Diseases/National Institutes of Health.
1. World Health Organization. Antiretroviral therapy for HIV infection in adults and adolescents in resource-limited settings: towards universal access: recommendations for a public health approach
(2006 revision). Geneva, Switzerland: WHO; 2006.
2. Egger M, May M, Chene G, Phillips AN, Ledergerber B, Dabis F, et al
. Prognosis of HIV-1-infected patients starting HAART: a collaborative analysis of prospective studies. Lancet 2002; 360:119–129.
3. Kaplan JE, Hanson DL, Cohn DL, Karon J, Buskin S, Thompson M, et al
. When to begin highly active antiretroviral therapy? Evidence supporting initiation of therapy at CD4+ lymphocyte counts < 350 cells/microl. Clin Infect Dis 2003; 37:951–958.
4. Lynen L, Thai S, De Munter P, Leang B, Sokkab A, Schrooten W, et al
. The added value of a CD4 count to identify patients eligible for highly active antiretroviral therapy among HIV-positive adults in Cambodia. J Acquir Immune Defic Syndr 2006; 42:322–324.
5. Martinson N, Heyer A, Steyn J, Struthers H, McIntyre J, Gray G, Pronyk P. Does WHO clinical stage reliably predict who should receive ARV treatment?
In: 3rd IAS Conference on HIV Pathogenesis and Treatment
. Rio de Janeiro, Brazil, 24–27 July 2005 [Abstract WeFo0304].