SEARCH

SEARCH BY CITATION

Keywords:

  • AIDS-defining illness;
  • cause of death;
  • cohort study;
  • HIV/AIDS;
  • survival;
  • tuberculosis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Appendix

Objectives

To analyse the impact of combined antiretroviral treatment (cART) on survival with AIDS, according to the nature of the first AIDS-defining clinical illness (ADI); to examine trends in AIDS-defining causes (ADC) and non-AIDS-defining causes (non-ADC) of death.

Methods

From the French Hospital Database on HIV, we studied trends in the nature of the first ADI and subsequent survival in France during three calendar periods: the pre-cART period (1993–1995; 8027 patients), the early cART period (1998–2000; 3504 patients) and the late cART period (2001–2003; 2936 patients).

Results

The three most frequent initial ADIs were Pneumocystis carinii (jirovecii) pneumonia (PCP) (15.6%), oesophageal candidiasis (14.3%) and Kaposi's sarcoma (13.9%) in the pre-cART period. In the late cART period, the most frequent ADIs were tuberculosis (22.7%), PCP (19.1%) and oesophageal candidiasis (16.2%). The risk of death after a first ADI fell significantly after the arrival of cART. Lower declines were observed for progressive multifocal leukoencephalopathy, lymphoma and Mycobacterium avium complex infection. After an ADI, the 3-year risk of death from an ADC fell fivefold between the pre-cART and late cART periods (39%vs. 8%), and fell twofold for non-ADCs (17%vs. 9%).

Conclusions

The relative frequencies of initial ADI have changed since the advent of cART. Tuberculosis is now the most frequent initial ADI in France; this is probably the result of the increasing proportion of migrants from sub-Saharan Africa. After a first ADI, cART has a major impact on ADCs and a smaller impact on deaths from other causes. The risk of death from AIDS and from other causes is now similar.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Appendix

Diagnosis of AIDS has been associated with a poor prognosis in HIV-infected patients. This was true in the era before the extensive use of combined antiretroviral treatment (cART), introduced in 1996.

The latest Centers for Disease Control (CDC) definition of AIDS, published in 1993, lists 23 distinct AIDS-defining clinical illnesses (ADIs) [1]. However, each of these illnesses might have different prognoses and might therefore have different clinical significance. For instance, in the pre-cART period, Mocroft et al. [2] showed that the median survival time with AIDS ranged from 2 months when the first ADI was progressive multifocal leukoencephalopathy (PML) to 22 months when it was extrapulmonary tuberculosis.

Many studies have shown a drastic fall in AIDS-related mortality since the advent of potent antiretroviral (ARV) therapy. However, most studies focused on the early cART period (roughly between 1996 and 1998) [2–10]; few have evaluated whether the risk reduction was similar for each specific initial ADI [2,5,9,11,12] in a more recent period [13,14].

It is unknown whether the decline in deaths will persist in more recent times, with long-term exposure to treatment, complications induced by therapies and the ageing of the HIV-infected population. This also explains why the cause of death among HIV-infected patients has recently become a focus of interest [15–20]. Most relevant studies show that the pattern of causes of death has changed over the years [15–17,19,20], with a trend towards lower AIDS-related mortality; however, very few studies focused on survival according to causes of death.

Here we analysed survival with AIDS in a large cohort of patients, comparing the pre-cART period to the early cART period and the late cART period. We also investigated whether the risk reduction was similar according to each initial ADI, and trends in the causes of death in the three periods.

Patients and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Appendix

Patients

Patients were selected from the French Hospital Database on HIV (FHDH), a nationwide hospital-based cohort [21]. This epidemiological network was created in 1989, and 62 participating hospitals across France currently provide data on HIV-infected patients. The only FHDH inclusion criteria are HIV-1 or HIV-2 infection and written informed consent. Trained research assistants prospectively collect clinical, biological and therapeutic data and causes of death from medical records using specialized software (DMI2) developed and owned by the French Ministry of Health. A follow-up form is completed at least every 6 months or at each visit or hospital admission during which a new clinical manifestation is diagnosed, a new treatment is prescribed or a change in a biological marker is noted. In each centre, AIDS diagnoses are validated by an expert in HIV/AIDS. By mid-2005, the FHDH had enrolled 102 777 HIV-1 infected patients. The FHDH covers 53% of AIDS cases diagnosed among adults in France [22].

For this analysis, we considered all HIV-1 infected adults enrolled in the database who had a first diagnosis of AIDS between 1993 and 2003 (n=21 327 patients) (Fig. 1). Patients with diagnosis of AIDS prior to FHDH enrolment were excluded (n=6693 patients). Only initial ADIs corresponding to the clinical definitions of the 1993 CDC revised AIDS case definition [1] were considered. Therefore, a CD4 cell count <200 cells/μL was not considered as an AIDS-defining event.

image

Figure 1.  Patient selection. FHDH, French Hospital Database on HIV; ADI, AIDS-defining clinical illness; cART, combined antiretroviral treatment.

Download figure to PowerPoint

We then selected patients who had had at least one follow-up visit after AIDS onset and had at least 6 months of follow-up in the database (or who died <6 months after enrolment).

The patients were divided into three groups according to treatment availability at the time of AIDS diagnosis: patients diagnosed with AIDS in the pre-cART period (1993–1995), the early cART period (1998–2000) and the late cART period (2001–2003). These periods correspond roughly to the main use of nucleoside reverse transcriptase inhibitor (NRTI) monotherapy, protease inhibitor (PI)-containing triple-drug combinations, and PI- and/or non-nucleoside reverse transcriptase inhibitor (NNRTI)-containing or boosted PI regimens, respectively. We excluded the period 1996–1997 (4360 patients) because it corresponded to the very beginning of PI use in France, and prescriptions were very heterogeneous.

Statistical analysis

All first ADIs that occurred in more than 50 patients in each of the calendar periods were ranked according to their frequency in each period (frequency ranking).

Overall survival after diagnosis of the first ADI was studied using Kaplan–Meier estimates. Survival comparisons between patients who developed AIDS during the three periods were based on the log–rank test. Survival after diagnosis of individual ADIs was estimated after 3 years and ranked in increasing order for each period (survival ranking).

We then distinguished deaths from AIDS (using the AIDS CDC case definition) and deaths from other causes. We took into account the competing risk of deaths from AIDS and deaths from other causes using the methods described by Fine and Gray [23]. This approach avoids overestimating the risk of death from the cause of interest [24,25], and is recommended especially when censoring would have been informative.

The database was last updated on 30 June 2005. We used the following right-censoring strategy to take into account the time lag before deaths were recorded in the database: patients who were known to be alive in the 6 months before the last database update (i.e. between January and July 2005) were censored on 1 July 2005; patients who did not attend a follow-up visit during this 6-month period were considered lost to follow-up and were censored at their last visit.

Multivariate Cox proportional hazards models were adjusted for sex, age at initial AIDS diagnosis (continuous variable), CD4 cell count at AIDS diagnosis (≤50, 50–100, 101–200, >200 cells/μL), region of follow-up (Paris area, Provence–Alpes–Côte d'Azur, rest of metropolitan France, French West Indies and Reunion Islands), HIV exposure category, sub-Saharan African origin, initiation of ART monotherapy, dual nucleoside therapy or cART, and anti-infective prophylaxis. Long stays in sub-Saharan Africa were used as a proxy for sub-Saharan origin. (In comparison to studies in which nationality is recorded, this variable provides a good approximation of migrant status [26,27].)

cART was defined as a regimen comprising at least three ARV drugs, or two boosted PIs, or one PI plus one non-nucleoside analogue, NNRTI.

All tests were two-sided, and P-values below 0.05 were considered to denote statistical significance. Statistical analyses were carried out with the SAS software package (version 9; SAS Institute, Cary, NC, USA). Confidence intervals (CIs) of competing risks were computed with R 2.1.1 software (cmprsk package) (http://cran.r-project.org/).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Appendix

First ADIs were diagnosed in 8027 patients in the pre-cART period, 3504 patients in the early cART period and 2936 patients in the late cART period. The patients' characteristics are described in Table 1. In comparison to the pre-cART period, patients diagnosed with AIDS in the cART periods (early and late cART period pooled together) were more often heterosexual (more than 40%vs. 22%), less often homosexual (29%vs. 44%) and more often migrants from sub-Saharan Africa (14.6%vs. 2.7%); they tended to have higher CD4 cell counts at AIDS diagnosis (median 80 cells/μL vs. 41 cells/μL) and were more likely to be diagnosed with AIDS at enrolment in the FHDH database (38%vs. 26%). Patients' characteristics between the early and late cART periods were similar, except the higher proportion of patients of sub-Saharan origin in the late cART period (11.6%vs. 18.4%, P<0.0001).

Table 1.   Characteristics of patients diagnosed with a first ADI in the pre-cART period (1993–1995), the early cART period (1998–2000) and the late cART period (2001–2003)
 Pre-cART period (1993–1995) (n=8027)Early cART period (1998–2000) (n=3504)Late cART period (2001–2003) (n=2936)P-value
  • ADI, AIDS-defining clinical illness; ART, antiretroviral treatment; cART, combined antiretroviral treatment; FHDH, French Hospital Database on HIV; IQR, interquartile range; IVDU, intravenous drug user.

  • *

    cART was defined as a regimen comprising at least three antiretroviral drugs or two boosted protease inhibitors (PIs) or one PI plus one non-nucleoside reverse transcriptase inhibitor.

  • χ2 test or Kruskal–Wallis test when appropriate.

Men, n (%)6496 (80.9)2607 (74.4)2109 (71.8)< 0.0001
HIV transmission group, n (%)
 Homosexual men3488 (43.5)1078 (30.8)787 (26.8) 
 IVDUs2078 (25.9)636 (18.2)412 (14.0) 
 Heterosexuals1789 (22.3)1431 (40.8)1435 (48.9) 
 Others672 (8.3)359 (10.2)302 (10.3)< 0.0001
Region of follow-up, n (%)
 Paris and suburbs3762 (46.9)1674 (47.8)1529 (52.1) 
 Provence–Alpes–Côte d'Azur1369 (17.1)470 (13.4)304 (10.4) 
 Rest of metropolitan France2496 (31.1)1018 (29.1)750 (25.5) 
 West Indies and Reunion Islands400 (4.9)344 (9.8)353 (12.0)< 0.0001
Age at AIDS onset, median (IQR)35.3 (31.1–42.3)38.2 (33.5–44.7)39.8 (34.5–46.4)< 0.0001
CD4 cell count at AIDS onset, median (IQR)41 (13–123)80 (25–210)79 (26–210)< 0.0001
Patients with CD4<200 cells/μL at AIDS onset, n (%)6154 (76.7)2404 (68.6)2050 (69.8)< 0.0001
Last known ART before AIDS onset, n (%)
 No treatment3961 (49.3)1908 (54.5)1525 (51.9) 
 Nucleoside monotherapy3377 (42.1)48 (1.4)24 (0.8) 
 Dual nucleoside therapy681 (8.5)292 (8.3)131 (4.5) 
 cART*8 (0.1)1256 (35.8)1256 (42.8)< 0.0001
Prophylaxis at AIDS diagnosis, n (%)2682 (33.4)661 (18.9)564 (19.2) 
 Never prophylaxis693 (8.6)547 (15.6)465 (15.8) 
 Prior prophylaxis but none at AIDS onset4652 (58.0)2296 (65.5)1907 (65.0)< 0.0001
AIDS at enrolment in FHDH, n (%)2068 (25.8)1363 (38.9)1105 (37.6)< 0.0001
AIDS or CD4<200 cells/μL at enrolment in FHDH3620 (45.1)1875 (53.5)1568 (53.4)< 0.0001
More than one ADI at AIDS onset, n (%)994 (12.4)440 (12.5)390 (13.3)0.10
Known sub-Saharan origin, n (%)214 (2.7)405 (11.6)539 (18.4)< 0.0001
Duration of follow-up in months, median20.454.726.7 

First ADIs

Table 2 shows the distribution of the initial ADI in the three periods, and the corresponding median CD4 cell counts. The most frequent ADIs were Pneumocystis carinii (jirovecii) pneumonia (PCP) (15.6%), oesophageal candidiasis (14.3%) and Kaposi's sarcoma (KS) (13.9%) in the pre-cART period; PCP (19.9%), tuberculosis (18.2%) and oesophageal candidiasis (14.5%) in the early cART period; and tuberculosis (22.7%), PCP (19.1%) and oesophageal candidiasis (16.2%) in the late cART period. The distribution of ADIs differed significantly over the three periods of interest (P<0.0001) and between the early and late cART periods (P=0.001). The median CD4 cell count at AIDS onset increased significantly across the three periods for all individual ADIs, except PCP (36 and 33 cells/μL in the pre-cART and late cART periods, respectively) and cryptococcosis.

Table 2.   First ADI in the pre-cART (1993–1995), early cART (1998–2000) and late cART periods (2001–2003), ordered by the increasing frequency rank in the late cART period
First ADIAIDS in 1993–1995 (n=8027)AIDS in 1998–2000 (n=3504)AIDS in 2001–2003 (n=2936)CD4 median comparison P-value
nMedian CD4 count at AIDS (cells/μL)Pre-cART rank*Relative frequency (%)nMedian CD4 count at AIDS (cells/μL)Early cART rank*Relative frequency (%)nMedian CD4 count at AIDS (cells/μL)Late cART rank*Relative frequency (%)
  • ADI, AIDS-defining clinical illness; cART, combined antiretroviral treatment; CMV, cytomegalovirus infection; MAC, Mycobacterium avium complex; PCP, Pneumocystis carinii pneumonia; PML, progressive multifocal leukoencephalopathy.

  • *

    Frequency rank in the period.

  • Relative frequency of the ADI in the period.

  • Kruskal–Wallis test.

Tuberculosis846121410.5638150218.2667135122.70.0096
PCP125236115.669930119.956033219.10.1719
Candidiasis (oesophageal or respiratory)114738214.350767314.547666316.2< 0.0001
Toxoplasmosis7143468.937941410.829542410.00.0033
Kaposi's sarcoma111365313.9374137510.729215759.9< 0.0001
Lymphoma31191113.924519767.016818865.7< 0.0001
CMV7382059.21723974.91574175.3< 0.0001
HIV encephalopathy26065133.214411684.11188884.00.0004
Cryptococcosis17330142.29625122.7862792.90.2842
Herpes31753103.913014393.778159102.70.0001
Recurrent bacterial pneumonia30397123.8111228103.275222112.6< 0.0001
PML13558151.766107151.969108122.40.0005
MAC4461385.69835112.86827132.30.0001
Wasting syndrome4773675.98741142.56564142.20.0037
Cryptosporidiosis3243794.09053132.66146152.10.0467
Deaths4618   760   452    

Survival with AIDS

The patients diagnosed with a first ADI in the pre-cART period had a total of 29 986 person–years of follow-up (PYFU). Among those patients, 4918 died, of whom two thirds (3301; 67%) died from an AIDS-defining cause (ADC). In comparison, about half the patients diagnosed in the two cART periods died of AIDS. Among the patients diagnosed with a first ADI in the early cART period (14 095 PYFU), 760 died, of whom 365 (48%) died from an ADC. Among the patients diagnosed in the late cART period (6601 PYFU), 452 died, of whom 213 (47%) died from an ADC. The crude incidence rate of death after AIDS onset fell from 16.4% PYFU in the pre-cART period to 5.4% PYFU and 6.8% PYFU in the early and late cART periods, respectively.

After adjustment for gender, age at AIDS diagnosis, CD4 cell count at AIDS diagnosis, region of follow-up, HIV exposure category, ART monotherapy, dual-agent therapy or cART and anti-infective prophylaxis, the risk of death was significantly lower among those who were diagnosed in the cART periods compared to those diagnosed in the pre-cART period [hazard ratio (HR)=0.31, 95% CI 0.28–0.33 in the early cART period; HR=0.27, 95% CI 0.24–0.30 in the late cART period].

The 5-year overall survival rate after AIDS diagnosis was 40.0% (95% CI 38.8–41.1) in the pre-cART period and 77.4% (95% CI 75.9–78.9) in the early cART period. The 5-year survival estimates were not available for the late cART period, and we therefore report the 3-year survival probabilities for the three periods (Table 3).

Table 3.   Survival at 3 years according to the first ADI, and risk of death between the cART period and the pre-cART period, ordered by the increasing survival rank* in the late cART period
First ADIAIDS in 1993–1995 (n=8027) Survival at 3 years (95% CI)AIDS in 1998–2000 (n=3504) Survival at 3 years (95% CI)AIDS in 2001–2003 (n=2936) Survival at 3 years (95% CI)HR of death (2001–2003 vs. 1993–1995) Adjusted HR (95% CI)P-value
  • ADI, AIDS-defining clinical illness; cART, combined antiretroviral treatment; CI, confidence interval; CMV, cytomegalovirus infection; HR, hazard ratio; MAC, Mycobacterium avium complex; PCP, Pneumocystis carinii pneumonia; PML, progressive multifocal leukoencephalopathy.

  • *

    Survival rank: Kaplan–Meier survival estimates at 3 years.

  • HR of death in the cART vs. pre-cART period adjusted for sex, age at AIDS onset, CD4 cell count, region of follow-up, region of origin, HIV exposure category, ART monotherapy, dual therapy or cART, and prophylaxis.

MAC30.5 (26.1–34.9)71.6 (61.0–79.7)63.0 (48.3–74.5)0.41 (0.25–0.67)0.0004
Lymphoma32.3 (27.1–37.7)62.1 (55.6–68.0)63.9 (55.1–71.3)0.42 (0.30–0.58)< 0.0001
PML25.6 (18.3–33.6)63.0 (49.5–73.8)68.9 (56.3–78.6)0.46 (0.27–0.78)0.004
HIV encephalopathy42.7 (36.5–48.8)73.8 (65.5–80.4)74.3 (63.6–82.2)0.34 (0.22–0.54)< 0.0001
Toxoplasmosis37.8 (34.1–41.5)82.3 (77.9–85.9)80.3 (74.8–84.8)0.24 (0.18–0.33)< 0.0001
Candidiasis (oesophageal or respiratory)46.3 (43.3–49.2)81.0 (77.1–84.2)81.5 (77.1–85.1)0.35 (0.27–0.45)< 0.0001
Herpes52.5 (46.7–58.0)88.0 (80.5–92.7)82.5 (69.2–90.4)0.25 (0.13–0.49)< 0.0001
Cryptococcosis42.6 (34.8–50.1)79.3 (69.5–86.3)83.9 (73.9–90.4)0.25 (0.13–0.48)< 0.0001
CMV29.7 (26.3–33.2)80.0 (73.0–85.3)84.2 (76.6–89.4)0.17 (0.11–0.26)< 0.0001
Recurrent bacterial pneumonia70.9 (65.2–75.8)90.7 (82.8–95.1)85.7 (72.9–92.7)0.40 (0.18–0.89)0.02
Cryptosporidiosis40.2 (34.7–45.7)83.9 (74.3–90.1)85.9 (71.5–93.3)0.11 (0.05–0.25)< 0.0001
Wasting syndrome34.7 (30.3–39.2)73.0 (62.2–81.2)86.0 (74.9–92.5)0.21 (0.11–0.39)< 0.0001
PCP50.6 (47.6–53.4)88.0 (85.2–90.2)87.3 (83.8–90.0)0.19 (0.15–0.26)< 0.0001
Tuberculosis (pulmonary or extrapulmonary)67.1 (63.6–70.2)89.9 (87.1–92.0)88.8 (85.5–91.4)0.32 (0.23–0.43)< 0.0001
Kaposi's sarcoma42.3 (39.3–45.3)89.7 (86.0–92.4)89.7 (85.2–92.9)0.15 (0.10–0.22)< 0.0001
All44.5 (43.3–45.6)82.0 (80.7–83.4)82.9 (81.3–84.5)0.27 (0.24–0.30)< 0.0001

Survival according to initial ADI

Survival after the first ADI increased considerably after the advent of cART, whatever the first ADI (Table 3). The worst prognosis was associated with Mycobacterium avium complex (MAC), lymphoma and PML (3-year survival rate <70%) in the late cART period, and with PML, cytomegalovirus disease (CMV) and MAC in the pre-cART period (3-year survival rate <30%). At the time of death, CD4 cell counts were higher in the cART period than in the pre-cART period (median 91 vs. 22 cells/μL, respectively; P=0.0001).

Compared to the pre-cART period, the largest decline in risk of death was observed when cryptosporidiosis (89%), KS (85%), CMV disease (83%), PCP (81%) or toxoplasmosis (76%) was the first ADI, whereas the smallest declines were for PML (54%), lymphoma (58%), MAC (59%) and recurrent bacterial infection (60%) (Fig. 2). The overall risk of death continued to decline in the late cART period relative to the early cART period (HR=0.87, 95% CI 0.77–0.98), although statistical significance was not reached for any specific ADI.

image

Figure 2.  Adjusted hazards of death according to the initial AIDS-defining illness between the combined antiretroviral treatment (cART) periods (2001–2003 and 1998–2000) and the pre-cART period (1993–1995). Bars indicate the 95% CI. CI, confidence interval; HR, hazard ratio; PCP, Pneumocystis carinii pneumonia; PML, progressive multifocal leukoencephalopathy.

Download figure to PowerPoint

Causes of death

Figure 3 shows the cumulative incidence of death in the three periods, after taking competing risks into account. Among patients diagnosed with AIDS in the pre-cART period, the cumulative 3-year incidence of death was 39% for ADC (95% CI 38–40) and 17% for other causes (95% CI 16–17). The corresponding values among patients diagnosed in the early cART period were 9% (95% CI 8–10) and 9% (95% CI 8–10), respectively; in the late cART period, values were 8% (95% CI 7–9) and 9% (95% CI 8–10), respectively.

image

Figure 3.  Cumulative incidence of death after the first AIDS-defining illness, according to the cause of death, in the pre-combined antiretroviral treatment (cART) period (1993–1995) and in the cART periods (1998–2000 and 2001–2003): a competing risk analysis.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Appendix

We analysed trends in survival after the first ADI and in causes of death over a 10-year period (1993–2003), based on a very large French prospective hospital cohort. We found that the relative frequencies of initial ADIs changed markedly over this period. Tuberculosis is now the most frequent first ADI in France. Survival with AIDS increased markedly, regardless of the first ADI. The smallest increases were observed when PML, lymphoma, MAC infection or recurrent bacterial infection was the first ADI. Interestingly, cART also had a positive impact on the incidence of deaths from causes unrelated to AIDS, albeit to a lesser extent.

Limits and strengths

Two important strengths of this study are the fact that it is based on a large prospective population of patients belonging to all HIV transmission groups (who were recruited both before and after the advent of potent ART) and the large number of PYFU. Another strength is the use of competing risk analysis in order to avoid overestimating the risk of death from causes related or unrelated to AIDS.

Here, causes of deaths were split into ADC of death and non-AIDS-defining cause (non-ADC) of death. A more detailed analysis of causes of death would have required the use of a more specific questionnaire during the three study periods. Such studies are ongoing in France [15]. Under-reporting of deaths and imperfect documentation of causes of death might also be of concern in our hospital cohort. The quality of reporting may depend on the cause of death and on the participating centre. Indeed, deaths from AIDS might be more likely to be reported. Moreover, some centres might be more likely than others to collect information on the vital status of patients lost to follow-up. However, despite some misclassification of AIDS-related deaths and deaths from other causes, there is no obvious reason why the quality of this information should have varied during the study period. In addition, because most of the deaths (6047; 99%) were recorded in the year following the last hospital visit, biases caused by inappropriate censoring are unlikely here. Loss of follow-up is always a concern in survival analyses – one reason for being lost to follow-up is death. In our study, 9% of patients were censored in the 3 years following the first ADI and could therefore be considered as lost to follow-up. The censoring of these patients might have slightly overestimated the survival estimates [27] provided in Table 3.

Changing pattern of initial ADIs

The relative frequencies of the different initial ADIs evolved with time. Tuberculosis is now the most frequent ADI in France, while it ranked in fourth place in the pre-cART period. This is no doubt related to the changing epidemiology of AIDS. AIDS is now diagnosed more frequently in patients from sub-Saharan Africa (18% in the late cART period vs. 3% in the pre-cART period; P<0.0001) and also in heterosexual patients (49%vs. 22%). Diagnoses of tuberculosis involved patients from sub-Saharan Africa in 10.2%, 29.3% and 35.7% of cases (P<0.0001) in the pre-cART, early cART and late cART eras, respectively (data not shown). When the first ADIs were considered separately according to the origins (data not shown), tuberculosis was the most frequent ADI in sub-Saharan African patients over the three periods (40%, 46% and 44%, respectively). In contrast, PCP was the most frequent ADI over the three periods in patients without known sub-Saharan origin (16%, 22% and 21%, respectively); tuberculosis (10%, 15% and 18%, respectively) increased from the fourth to the second most frequent ADI in the late cART period. Similar findings were found in the Swiss HIV Cohort [28]. Moreover, almost half the cases of tuberculosis (44% in the late cART period) had been diagnosed at inclusion in the database. This confirms that tuberculosis is associated with late access to care and underlines the need for programmes aimed at encouraging this population to seek HIV screening and care.

Lymphoma was the 11th most frequent first ADI in the pre-cART period and the 6th in the late cART era. It has been reported elsewhere that the fall in the incidence of lymphoma that followed the advent of cART was accompanied by a changing profile of AIDS-related lymphoma (less frequent systemic and brain involvement) and also by prolonged survival [29].

Like other authors [12,13], we found that the CD4 cell count at initial diagnosis of AIDS tends to be higher than in the pre-cART period. Nowadays, thanks to cART, more patients have higher CD4 cell counts than in the pre-cART period. This change in CD4 cell count distribution among patients at risk of AIDS may explain the higher median CD4 cell counts among patients diagnosed with AIDS – even if the risk of ADI in a given CD4 stratum did not necessarily change. The presence of an impaired qualitative immunological response may also explain the fact that ADI occurs despite the higher CD4 cell count. However, two arguments do not support this explanation. Firstly, one study suggested that the risk of onset of an ADI in a given CD4 cell count stratum has changed little since the pre-cART era [29]. Secondly, the fact that anti-infective primary or secondary prophylaxis can be interrupted safely when the CD4 cell count reaches 200 cells/μL on cART [30–32] indicates that the immune response is at least partially functional.

Survival according to first ADI

Survival with AIDS improved markedly after the advent of cART, irrespective of the nature of the first ADI. It is encouraging that, like others [33], we found that this improvement continued between the early and late cART periods, despite concerns over viral mutants, treatment complications and the ageing of the HIV-infected population.

The gap between the best and worst survival rates after diagnosis of the different first ADIs tended to become smaller with time. Indeed, the 3-year survival estimates ranged from 26% to 71% (for PML and recurrent bacterial pneumonia, respectively) in the pre-cART period, and from 63% to 90% (for MAC and KS, respectively) in the late cART period. Improvement in prognoses might be related either to the change in CD4 level at diagnosis or to a better immune response induced by ARV drugs for a given disease. The fact that ADIs do not have the same prognostic importance might have implications for the design of clinical trials for inclusion criteria as well as endpoint definitions.

Causes of death

Competing risk analysis showed that the 3-year risk of death from AIDS fell fivefold between the pre-cART period and the late cART period (39%vs. 8%), and that the risk of death from other causes fell twofold (17%vs. 9%). These results are consistent with recent reports describing causes of death in various populations of HIV-infected patients and in various settings. For instance, a competing risk analysis of Spanish HIV-infected haemophiliacs showed that the marked improvement in survival was largely attributable to the reduction in AIDS-related deaths, and that the rate of deaths from liver disease had risen [34]. Similar increases in the proportion of deaths from causes unrelated to AIDS have been found in North America, in the HIV Outpatient Study (HOPS) cohort [19], the Women's Interagency HIV Study (WIHS) [18], in New York City [20], the Cascade collaboration [35] and the French Mortality 2000 study [15]. Interestingly, in the last study, about half of the AIDS-related deaths were caused by non-Hodgkin lymphoma [15,36], confirming the poor prognosis of lymphoma even in the cART period. Here, we found that cART was associated with far fewer AIDS-related deaths and also with a smaller reduction in deaths by other causes. This probably indicates that – from an individual and public health perspective – it might be interesting to identify and follow other clinical conditions than those listed in the CDC AIDS-defining conditions [37].

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Appendix

The relative frequencies of initial ADIs have changed over time. Tuberculosis is now the most frequent first ADI in France. Survival with AIDS has improved markedly, regardless of the initial clinical event, although some AIDS-inaugurating diseases are associated with better outcome than others. There has been a marked reduction in deaths from AIDS, and also a smaller reduction in deaths from other causes. The absolute risks of death from ADCs and non-ADCs are now similar.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Appendix

FHDH is supported by Agence Nationale de Recherches sur le SIDA et les hépatites virales (ANRS), Fondation pour la Recherche Médicale, Institut National de la Santé et de la Recherche Médicale (INSERM) and the French Ministry of Health.

The authors are grateful to all FHDH participants and research assistants, without whom this work would not have been possible.

Potential conflicts of interest

D.C. has received: honoraria from Abbott, GlaxoSmithKline, Bristol-Myers Squibb, Gilead, Janssen, Roche and Boehringer Ingelheim; research grants from Abbott, GlaxoSmithKline, Bristol-Myers-Squibb, Gilead, Janssen, Roche and Boehringer Ingelheim; and travel grants from Abbott, GlaxoSmithKline and Roche.

C.A. has received honoraria from Roche, GlaxoSmithKline and Tibotec.

S.G., E.L., M.M.-K., M.B., P.F., A.M. and C.R. have no conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Appendix
  • 1
    Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. Morb Mortal Wkly Rep 1992; 41: 119.
  • 2
    Mocroft AJ, Lundgren JD, D'Armino Monforte A et al. Survival of AIDS patients according to type of AIDS-defining event. The AIDS in Europe Study Group. Int J Epidemiol 1997; 26: 400407.
  • 3
    Egger M, Hirschel B, Francioli P et al. Impact of new antiretroviral combination therapies in HIV infected patients in Switzerland: prospective multicentre study. BMJ 1997; 315: 11941199.
  • 4
    Mocroft A, Vella S, Benfield TL et al. Changing patterns of mortality across Europe in patients infected with HIV-1. EuroSIDA Study Group. Lancet 1998; 352: 17251730.
  • 5
    Fordyce EJ, Singh TP, Nash D, Gallagher B, Forlenza S. Survival rates in NYC in the era of combination ART. J Acquir Immune Defic Syndr 2002; 30: 111118.
  • 6
    Palella FJ Jr, Delaney KM, Moorman AC et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study investigators. N Engl J Med 1998; 338: 853860.
  • 7
    Hogg RS, Heath KV, Yip B et al. Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA 1998; 279: 450454.
  • 8
    Detels R, Muñoz A, McFarlane G et al. Effectiveness of potent antiretroviral therapy on time to AIDS and death in men with known HIV infection duration. Multicenter AIDS Cohort Study investigators. JAMA 1998; 280: 14971503.
  • 9
    Ives NJ, Gazzard BG, Easterbrook PJ. The changing pattern of AIDS-defining illnesses with the introduction of highly active antiretroviral therapy (HAART) in a London clinic. J Infect 2001; 42: 134139.
  • 10
    Lee LM, Karon JM, Selik R, Neal JJ, Fleming PL. Survival after AIDS diagnosis in adolescents and adults during the treatment era, United States, 1984–1997. JAMA 2001; 285: 13081315.
  • 11
    Conti S, Masocco M, Pezzotti P et al. Differential impact of combined antiretroviral therapy on the survival of Italian patients with specific AIDS-defining illnesses. J Acquir Immune Defic Syndr 2000; 25: 451458.
  • 12
    Dore GJ, Li Y, McDonald A, Ree H, Kaldor JM, Kaldo JM. Impact of highly active antiretroviral therapy on individual AIDS-defining illness incidence and survival in Australia. J Acquir Immune Defic Syndr 2002; 29: 388395.
  • 13
    Schneider MF, Gange SJ, Williams CM et al. Patterns of the hazard of death after AIDS through the evolution of antiretroviral therapy: 1984–2004. AIDS 2005; 19: 20092018.
  • 14
    Mocroft A, Ledergerber B, Katlama C et al. Decline in the AIDS and death rates in the EuroSIDA Study: an observational study. Lancet 2003; 362: 2229.
  • 15
    Lewden C, Salmon D, Morlat P et al. Causes of death among human immunodeficiency virus (HIV)-infected adults in the era of potent antiretroviral therapy: emerging role of hepatitis and cancers, persistent role of AIDS. Int J Epidemiol 2005; 34: 121130.
  • 16
    Crum NF, Riffenburgh RH, Wegner S et al. Comparisons of causes of death and mortality rates among HIV-infected persons: analysis of the pre-, early, and late HAART (highly active antiretroviral therapy) eras. J Acquir Immune Defic Syndr 2006; 41: 194200.
  • 17
    Krentz HB, Kliewer G, Gill MJ. Changing mortality rates and causes of death for HIV-infected individuals living in Southern Alberta, Canada from 1984 to 2003. HIV Med 2005; 6: 99106.
  • 18
    Cohen MH, French AL, Benning L et al. Causes of death among women with human immunodeficiency virus infection in the era of combination antiretroviral therapy. Am J Med 2002; 113: 9198.
  • 19
    Palella FJ Jr, Baker RK, Moorman AC et al. Mortality in the highly active antiretroviral therapy era: changing causes of death and disease in the HIV outpatient study. J Acquir Immune Defic Syndr 2006; 43: 2734.
  • 20
    Sackoff JE, Hanna DB, Pfeiffer MR, Torian LV. Causes of death among persons with AIDS in the era of highly active antiretroviral therapy: New York City. Ann Intern Med 2006; 145: 397406.
  • 21
    Grabar S, Le Moing V, Goujard C et al. Clinical outcome of patients with HIV-1 infection according to immunologic and virologic response after 6 months of highly active antiretroviral therapy. Ann Intern Med 2000; 133: 401410.
  • 22
    Bernillon P, Lievre L, Pillonel J, Laporte A, Costagliola D. Record-linkage between two anonymous databases for a capture–recapture estimation of underreporting of AIDS cases: France 1990–1993. The Clinical Epidemiology Group from Centres d'Information et de Soins de l'Immunodeficience Humaine. Int J Epidemiol 2000; 29: 168174.
  • 23
    Fine J, Gray R. A proportional hazards model for subdistribution of a competing risk. J Am Stat Assoc 1999; 94: 496509.
  • 24
    Satagopan JM, Ben-Porat L, Berwick M, Robson M, Kutler D, Auerbach AD. A note on competing risks in survival data analysis. Br J Cancer 2004; 91: 12291235.
  • 25
    Alberti C, Metivier F, Landais P, Thervet E, Legendre C, Chevret S. Improving estimates of event incidence over time in populations exposed to other events: application to three large databases. J Clin Epidemiol 2003; 56: 536545.
  • 26
    Lert F, Obadia Y, Equipe de l' étude Vespa. Comment vit-on en France avec le VIH/sida? [How do we live in France with HIV/AIDS?] Pop Soc 2004; 406: 14.
  • 27
    Lanoy E, Lewden C, Lievre L et al. What happens to patients from the French Hospital Database on HIV lost to follow-up and consequences on mortality estimates. European AIDS Conference. Madrid, Spain, October 2007 [Abstract P184/10].
  • 28
    Staehelin C, Rickenbach M, Low N et al. Migrants from Sub-Saharan Africa in the Swiss HIV Cohort Study: access to antiretroviral therapy, disease progression and survival. AIDS 2003; 17: 22372244.
  • 29
    Besson C, Goubar A, Gabarre J et al. Changes in AIDS-related lymphoma since the era of highly active antiretroviral therapy. Blood 2001; 98: 23392344.
  • 30
    Abgrall S, Rabaud C, Costagliola D. Incidence and risk factors for toxoplasmic encephalitis in human immunodeficiency virus-infected patients before and during the highly active antiretroviral therapy era. Clin Infect Dis 2001; 33: 17471755.
  • 31
    Kirk O, Reiss P, Uberti-Foppa C et al. Safe interruption of maintenance therapy against previous infection with four common HIV-associated opportunistic pathogens during potent antiretroviral therapy. Ann Intern Med 2002; 137: 239250.
  • 32
    Ledergerber B, Mocroft A, Reiss P et al. Discontinuation of secondary prophylaxis against Pneumocystis carinii pneumonia in patients with HIV infection who have a response to antiretroviral therapy. Eight European Study Groups. N Engl J Med 2001; 344: 168174.
  • 33
    Lohse N, Hansen AB, Pedersen G et al. Survival of persons with and without HIV infection in Denmark, 1995–2005. Ann Intern Med 2007; 146: 8795.
  • 34
    Del Amo J, Perez-Hoyos S, Moreno A et al. Trends in AIDS and mortality in HIV-infected subjects with hemophilia from 1985 to 2003: the competing risks for death between AIDS and liver disease. J Acquir Immune Defic Syndr 2006; 41: 624631.
  • 35
    Smit C, Geskus R, Walker S et al. Effective therapy has altered the spectrum of cause-specific mortality following HIV seroconversion. AIDS 2006; 20: 741749.
  • 36
    May T, Lewden C, Bonnet F et al. Causes and characteristics of death among HIV-1 infected patients with immunovirologic response to antiretroviral treatment. Presse Med 2004; 33: 14871492.
  • 37
    Mocroft A. Clinical endpoints for randomized clinical trials: all AIDS-defining conditions are not created equal. 14th Conference on Retroviruses and Opportunistic Infections. Los Angeles, CA, February 2007 [Abstract 80].

Appendix

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Appendix
Clinical epidemiology group of the French Hospital Database on HIV
  • Scientific committee: S Abgrall, F Barin, M Bentata, E Billaud, F Boué, C Burty, A Cabié, D Costagliola, L Cotte, P De Truchis, X Duval, C Duvivier, P Enel, L Fredouille-Heripret, J Gasnault, C Gaud, J Gilquin, S Grabar, C Katlama, MA Khuong, JM Lang, AS Lascaux, O Launay, A Mahamat, M Mary-Krause, S Matheron, JL Meynard, J Pavie, G Pialoux, F Pilorgé, I Poizot-Martin, C Pradier, J Reynes, E Rouveix, A Simon, P Tattevin, H Tissot-Dupont, JP Viard, N Viget.
  • DMI2 coordinating centre: French Ministry of Health (A Pariente-Khayat, V Salomon), Technical Hospitalization Information Agency, ATIH (N Jacquemet, A Rivet).
  • Statistical analysis centre: INSERM U720 (S Abgrall, D Costagliola, S Grabar, M Guiguet, I Kousignian, E Lanoy, L Lièvre, M Mary-Krause, V Potard, H Selinger-Leneman).
  • CISIH:
    Paris area: CISIH de Bichat-Claude Bernard (Hôpital Bichat-Claude Bernard: E Bouvet, B Crickx, JL Ecobichon, C Leport, S Matheron, C Picard-Dahan, P Yeni), CISIH de Paris-Centre Ouest (Hôpital Européen Georges Pompidou: D Tisne-Dessus, L Weiss; GH Tarnier-Cochin: D Salmon, D Sicard; Hôpital Saint-Joseph: I Auperin, J Gilquin; Hôpital Necker Adultes: L Roudière, JP Viard), CISIH de Paris-Sud (Hôpital Antoine Béclère: F Boué, R Fior; Hôpital de Bicêtre: JF Delfraissy, C Goujard; Hôpital Henri Mondor: C Jung, P Lesprit; Hôpital Paul Brousse), CISIH de Paris-Est (Hôpital Saint-Antoine: N Desplanque, JL Meynard, MC Meyohas, O Picard; Hôpital Tenon: J Cadranel, C Mayaud, G Pialoux), CISIH de Pitié-Salpétrière (GH Pitié–Salpétrière: F Bricaire, S Herson, C Katlama, A Simon), CISIH de Saint-Louis (Hôpital Saint-Louis: JP Clauvel, JM Decazes, L Gerard, JM Molina; GH Lariboisière-Fernand Widal: M Diemer, P Sellier), CISIH 92 (Hôpital Ambroise Paré: H Berthé, C Dupont; Hôpital Louis Mourier: C Chandemerle, E Mortier; Hôpital Raymond Poincaré: P de Truchis), CISIH 93 (Hôpital Avicenne: M Bentata, P Honoré; Hôpital Jean Verdier: V Jeantils, S Tassi; Hôpital Delafontaine: D Mechali, B Taverne).
    Outside Paris area: CISIH Auvergne-Loire (CHU de Clermont-Ferrand: F Gourdon, H Laurichesse; CHRU de Saint-Etienne: A Fresard, F Lucht); CISIH de Bourgogne-Franche Comté (CHRU de Besançon; CHRU de Dijon; CH de Belfort: P Eglinger, JP Faller; CHRU de Reims); CISIH de Caen (CHRU de Caen: C Bazin, R Verdon), CISIH de Grenoble (CHU de Grenoble), CISIH de Lyon (Hôpital de la Croix-Rousse: A Boibieux, D Peyramond; Hôpital Edouard Herriot: JM Livrozet, JL Touraine; Hôtel-Dieu: L Cotte, C Trepo), CISIH de Marseille (Hôpital de la Conception: I Ravaux, H Tissot-Dupont; Hôpital Nord: JP Delmont, J Moreau; Institut Paoli Calmettes: JA Gastaut; Hôpital Sainte-Marguerite: I Poizot-Martin, F Retornaz, J Soubeyrand; CHG d'Aix-En-Provence: T Allegre, PA Blanc; Centre Pénitentiaire des Baumettes: A Galinier, JM Ruiz; CH d'Arles; CH d'Avignon: G Lepeu; CH de Digne les Bains: P Granet-Brunello; CH de Gap: JP Esterni, L Pelissier; CH de Martigues: R Cohen-Valensi, M Nezri; CHI de Toulon: S Chadapaud, A Laffeuillade), CISIH de Montpellier (CHU de Montpellier: J Reynes; CHG de Nîmes), CISIH de Nancy (Hôpital de Brabois: T May, C Rabaud), CISIH de Nantes (CHRU de Nantes: E Billaud, F Raffi), CISIH de Nice (Hôpital Archet 1: P Pugliese, C Pradier; CHG Antibes Juan les Pins), CISIH de Rennes (CHU de Rennes: C Arvieux, C Michelet), CISIH de Rouen (CHRU de Rouen: F Borsa-Lebas, F Caron), CISIH de Strasbourg (CHRU de Strasbourg: P Fraisse, JM Lang, D Rey; CH de Mulhouse), CISIH de Toulouse (CHU Purpan: E Arlet-Suau, L Cuzin, P Massip, MF Thiercelin Legrand; Hôpital la Grave; CHU Rangueil), CISIH de Tourcoing-Lille (CH Gustave Dron; CH de Tourcoing: Y Yasdanpanah), CISIH de Tours (CHRU de Tours; CHU Trousseau).
    Overseas: CISIH de Guadeloupe (CHRU de Pointe-à-Pitre), CISIH de Guyane (CHG de Cayenne: R Pradinaud, M Sobesky), CISIH de Martinique (CHRU de Fort-de-France), CISIH de La Réunion (CHD Félix Guyon: C Gaud, M Contant).