Research potentials and pitfalls in the use of an HIV clinical database: Chelsea and Westminster Hospital

A causal framework for understanding the effect of losses to follow-up on epidemiologic analyses in clinic-based cohorts: the case of HIV-infected patients on antiretroviral therapy in Africa

Although clinic-based cohorts are most representative of the "real world," they are susceptible to loss to follow-up. Strategies for managing the impact of loss to follow-up are therefore needed to maximize the value of studies conducted in these cohorts. The authors evaluated adult patients starting antiretroviral therapy at an HIV/AIDS clinic in Uganda, where 29% of patients were lost to follow-up after 2 years (January 1, 2004-September 30, 2007). Unweighted, inverse probability of censoring weighted (IPCW), and sampling-based approaches (using supplemental data from a sample of lost patients subsequently tracked in the community) were used to identify the predictive value of sex on mortality. Directed acyclic graphs (DAGs) were used to explore the structural basis for bias in each approach. Among 3,628 patients, unweighted and IPCW analyses found men to have higher mortality than women, whereas the sampling-based approach did not. DAGs encoding knowledge about the data-generating process, including the fact that death is a cause of being classified as lost to follow-up in this setting, revealed "collider" bias in the unweighted and IPCW approaches. In a clinic-based cohort in Africa, unweighted and IPCW approaches-which rely on the "missing at random" assumption-yielded biased estimates. A sampling-based approach can in general strengthen epidemiologic analyses conducted in many clinic-based cohorts, including those examining other diseases.

Demographics of HIV-1 infection in Denmark: Results from the Danish HIV Cohort Study

We used a population-based cohort study design to describe the demographic characteristics of the HIV-infected population in Denmark and their variation over time. HIV treatment in Denmark is restricted to 9 centres, and all 3941 HIV-1 infected patients more than 15 y old seen at these centres in 1995-2003 were included. We found an estimated HIV prevalence of 70 per 100,000, and a mean annual incidence rate of 5.1 per 100,000 persons. The number of newly infected individuals was stable with a median of 231 per y (period 1995-2002), whereas the number of deaths decreased from 166 in 1995 to 50 in 2000 (p=0.000) and remained stable thereafter. Of the enrolled patients, 75% were males, 80% were Caucasian, 13% were black African, and the primary risk behaviour was male-to-male sexual contact (44%), heterosexual contact (36%), and injection drug use (11%). During the y 1995-2003 we found an increase in age at diagnosis (p=0.000), and no major changes in gender, race, mode of infection, or baseline CD4+ cell count and viral load, neither overall not within subgroups of patients. In this period 14.5% had AIDS at the time of HIV diagnosis. Our data do not confirm concerns about unmonitored evolution in the HIV epidemic in Denmark.

The creation of a large UK-based multicentre cohort of HIV-infected individuals: The UK Collaborative HIV Cohort (UK CHIC) Study

OBJECTIVES

This paper describes the development of the UK Collaborative HIV Cohort (CHIC) Study. The aim of the study is to collate routinely collected data on HIV-infected individuals attending one of seven clinical centres in the UK since 1 January 1996, with the objectives of describing changes over time in the frequency of AIDS-defining illnesses, describing the uptake of and response to highly active antiretroviral therapy (HAART), and identifying factors associated with virological and immunological responses to HAART.

METHODS

By December 2002, demographic, clinical and laboratory data had been collected on HIV-positive patients seen at six of the seven HIV centres. Missing and inconsistent data had been investigated and the datasets audited. Records identified as relating to the same patient had been merged, and cross-checks made with UK death registers to improve the accuracy of death reporting.

RESULTS

The cohort currently contains information on 13,833 individuals. Eighty-two per cent of the cohort are male, and the median age was 34 years at first follow-up. The main risk factors for HIV infection have been determined as sex between men (63%) and sex between men and women (24%). Twenty-five per cent of the cohort are known to have developed AIDS, and 8% have died.

CONCLUSIONS

The UK CHIC Study provides important information on the status of individuals infected with HIV in the UK, and provides a means to study the response to HAART and to monitor changes in the clinical event and death rates that have occurred since the introduction of HAART in the UK.

Hepatitis C virus and human immunodeficiency virus coinfection in Spain

OBJECTIVE AND METHODS

In a cross-sectional study, based on a cohort composed of HIV-infected patients of fifteen tertiary level institutions of Spain, the main data of the entire cohort are described, characteristics of patients with or without hepatitis C coinfection are compared, and the possible association of hepatitis C virus coinfection with socioeconomic, HIV-related, and hepatitis B-related variables is assessed.

RESULTS

A total of 4,709 patients are studied. Median of age is 37 years, 78.3% are male. HIV risk behaviours are: parenteral drug use in 63.8% of patients, heterosexual in 22.3%, and homosexual in 10.8%. Serology of hepatitis C is positive in 69.2% of participants. The following variables are associated with increased prevalence of hepatitis C coinfection, both in univariate and in multivariate analysis: HIV risk behaviour, positive anti-HBs, longer time elapsed since HIV infection diagnosis, younger age, lower social status, lower CD4 cell count increase between nadir and last available result, and lower educational level (all P<0.001). Patients with heterosexual behaviour are more frequently coinfected than patients with homosexual behaviour (P<0.001).

CONCLUSION

This study highlights that, in Spain, more than two thirds of patients with HIV infection are coinfected with hepatitis C virus.

Comparison of virologic, immunologic, and clinical response to five different initial protease inhibitor-containing and nevirapine-containing regimens

CONTEXT

The effectiveness of different protease inhibitors (PIs) and non-nucleoside reverse transcriptase inhibitors outside the setting of clinical trials has not been well described.

OBJECTIVES

To compare five different PI-and nevirapine (NVP)-containing regimens on virologic, immunologic, and clinical outcomes and treatment discontinuation.

DESIGN AND SETTING

Observational cohort study based on an HIV clinic in London.

PATIENTS

A total of 690 patients who received either saquinavir hard gel (SQV HG) (n = 183), indinavir (IDV) (n = 189), nelfinavir (NFV) (n = 109), ritonavir (RTV) (n = 42), ritonavir with saquinavir hard gel (RTV/SQV HG) (n = 45), or NVP (n = 122) as part of an initial PI-or NVP-containing treatment regimen between November 1994 and December 1998. A total of 351 (51%) patients had prior exposure to nucleoside reverse transcriptase inhibitors (NRTIs).

MAIN OUTCOME MEASURES

The main outcome measures were virologic undetectability, subsequent virologic rebound, CD4 cell count rise, development of AIDS, and treatment discontinuation. All analyses were stratified for year of initiation of the PI-or NVP-containing regimen.

RESULTS

Overall, 63% of patients attained an undetectable viral load (VL) within 6 months of starting their PI or NVP regimen. The adjusted relative hazard (95% confidence interval [CI]) for an undetectable VL relative to SQV HG was (in rank order): 2.77 (CI: 1.84-4.17) for NFV, 2.54 (CI: 1.81-3.57) for IDV, 2.43 (CI: 1.52-3.87) for RTV, 2.08 (CI: 1.28-3.37) for RTV/SQV HG, and 1.96 (CI: 1.35-2.85) for NVP. Forty-nine percent of patients experienced VL rebound within 12 months of initial attainment of undetectability, but relative to SQV HG, this did not differ significantly across the different PI and NVP regimens. The CD4 cell count response and rate of AIDS events were also similar across the different regimens. No independent predictors of VL undetectability were identified, but prior NRTI exposure was associated with VL rebound, and a lower baseline VL and CD4 cell count were associated with a reduced CD4 count response. The frequency (95% CI) of treatment discontinuation differed across the regimens; at 6 months, it was lowest for NFV (18% [CI: 13%-24%]), IDV (25% [CI: 22%-29%]), and NVP (28% [CI: 22%-34%]) and highest for RTV (41% [CI: 31%-52%]) and SQV HG (52% [CI: 48%-57%]).

CONCLUSIONS

Although PI- and NVP-containing regimens were similar in their CD4 cell count response and rates of subsequent VL rebound, differences were observed in time to VL undetectability and discontinuation rates relative to SQV HG. SQV HG was consistently inferior to the other PIs and NVP. The use of NFV and IDV was associated with the highest rates of undetectability, and together with NVP, the lowest rates of discontinuation.

Factors associated with clinical and virological failure in patients receiving a triple therapy including a protease inhibitor

OBJECTIVE

To determine the predictors of virological and clinical failure in patients receiving a protease inhibitor as part of triple therapy.

METHODS

From the French Hospital Database on HIV, 1402 protease inhibitor-naive patients starting a highly active antiretroviral therapy regimen with ritonavir, saquinavir-hard gel capsule (hgc) or indinavir between July 1996 and March 1997, and with measured HIV RNA at baseline and at 12 months, were studied for progression to a new AIDS-defining event (ADE) or death. Virological failure was defined as plasma HIV RNA > 1000 copies/ml at 12 months. Multivariate analyses were performed using Cox models for clinical outcomes and logistic regression for virological outcomes.

RESULTS

During a median follow-up of 14.1 months, 94 (6.7%) patients experienced an ADE or died. At 12 months, 700 patients (49.9%) had virological failure. In the multivariate analysis, baseline CD4 cell count and viral load were found to be independent predictors of both virological and clinical failure. Neither the type of the first protease inhibitor taken nor previous antiretroviral therapy experience was associated with risk of clinical progression. For virological failure, the use of saquinavir-hgc was associated with a significant 1.96-fold increase in risk compared with indinavir; pre-treated patients were at higher risk than antiretroviral therapy-naive patients.

CONCLUSION

In this study with large samples of patients, the use of saquinavir-hgc was associated with higher risk of virological failure at 12 months than were ritonavir and indinavir; no differences between protease inhibitors were found for clinical progression. As biases cannot be excluded, a longer duration of follow-up will be necessary to answer the question of whether the results are really discrepant or simply reflect the delay between virological failure and clinical manifestations.

Sequence variations in human immunodeficiency virus type 1 Nef are associated with different stages of disease

nef alleles derived from a large number of individuals infected with human immunodeficiency virus type 1 (HIV-1) were analyzed to investigate the frequency of disrupted nef genes and to elucidate whether specific amino acid substitutions in Nef are associated with different stages of disease. We confirm that deletions or gross abnormalities in nef are rarely present. However, a comparison of Nef consensus sequences derived from 41 long-term nonprogressors and from 50 individuals with progressive HIV-1 infection revealed that specific variations are associated with different stages of infection. Five amino acid variations in Nef (T15, N51, H102, L170, and E182) were more frequently observed among nonprogressors, while nine features (an additional N-terminal PxxP motif, A15, R39, T51, T157, C163, N169, Q170, and M182) were more frequently found in progressors. Strong correlations between the frequency of these variations in Nef and both the CD4(+)-cell count and the viral load were observed. Moreover, analysis of sequential samples obtained from two progressors revealed that several variations in Nef, which were more commonly observed in patients with low CD4(+)-T-cell counts, were detected only during or after progression to immunodeficiency. Our results indicate that sequence variations in Nef are associated with different stages of HIV-1 infection and suggest a link between nef gene function and the immune status of the infected individual.