Causes of pre-AIDS death
The reported causes of death in the pre-AIDS death group are summarized in Table 2. Acute pneumonia, to which 92 deaths (19.6%) were attributed, was the leading identified cause. Septicaemia was a cause of death in 51 (10.9%) patients. Sixty-two (13.2%) patients died from a liver disease: liver cirrhosis associated with hepatic failure (57), hepatic carcinoma (five). Among these patients, non-A non-B hepatitis or hepatitis C virus infection was recorded for 10 (16%), chronic hepatitis B virus infection for six (10%) and excessive alcohol consumption for seven patients (11%). Of the 62 patients, 25 (40%) were haemophiliacs, 16 (26%) were homosexual men and 15 (24%) were IDU, four (6%) had acquired infection through heterosexual contacts and two (3%) had unreported risk factors for HIV infection.
Forty-nine patients died from malignancies other than hepatic carcinoma. Eight patients died from haematological malignancy: Hodgkin's disease (five), T-cell lymphoma (two), acute lymphocytic leukaemia (one). Eleven patients died from lung cancer: all were male and their mean age (± SD) at death was 59.2 (± 9.1) years. The tumour histology, available for nine patients, was adenocarcinoma (two) and other types of carcinoma (seven). Thirteen patients died from gastrointestinal cancer: carcinoma of the colon (five), carcinoma of the rectum (four), anal cancer (two), cancer of the head of the pancreas (one), and disseminated peritoneal carcinomatosis (one). Thirteen patients died from other specified malignancies: malignant tumour of the testicles (two teratoma and one seminoma), malignant cutaneous tumour (one carcinoma and one melanoma), disseminated sarcoma (two), anaplastic mediastinal cancer (two), kidney histiocytoma (one), prostate carcinoma (one), infiltrating cerebral glioma (one), and nasopharyngeal carcinoma (one). Four patients died from disseminated malignancies for which histology was not reported.
Haemorrhage, including cerebral and gastro-intestinal haemorrhage without specific cause, was reported in 42 (9.0%) further cases mainly in haemophiliac patients.
Among the 46 (9.5%) cardiopulmonary causes of death, 13 were due to myocardial infarction, five to pulmonary embolism, four to primary pulmonary hypertension and three deaths were recorded as due to an asthma attack.
In 99 (21.2%) patients, the cause of pre-AIDS death was not given as disease related: suicide (45), accidental overdose (24), road traffic accident (seven), other accidental causes (17) and homicide (five). Forty-three of the 45 patients who died from suicide were male of whom 28 were probably infected by sex with another man; four of these and 12 further individuals were IDU. Their mean age (± SD) was 31.8 ± 7.8 years. Of the patients who died from accidental overdose (21 males and three females), 13 (54.2%) were IDU, five (20.8%) were homosexual males, three (12.5%) were exposed to HIV infection through both injecting drug use and sex with another male, two (8.3%) were haemophiliac patients and one (4.1%) had undetermined risk exposure to HIV.
Risk factors for pre-AIDS mortality
In the pre-AIDS death group there were 418 (89.3%) males and 50 (10.7%) females (sex ratio, 8:1). These proportions differed slightly but significantly (P = 0.007) from those in the AIDS death group: 7946 (92.7%) males and 628 (7.3%) females (sex ratio, 13:1). Year at death was available for 462 (98.7%) patients in the pre-AIDS death group and for 8367 (97.6%) patients in the AIDS death group. Mean age (± SD) at death in the pre-AIDS group was 40.6 (± 14.6) years (range, < 1–86 years), slightly but significantly greater than the mean age at death in the AIDS group: 38.7 (± 10.4) years (range: < 1–81 years; P <10−5).
Patients who died pre-AIDS were more likely to live outside the Thames region than patients in the AIDS group (49.3% versus 28.8%; P <10−6; OR, 2.40).
Probable routes of HIV infection in individuals in the pre-AIDS death group differed significantly (P <10−6) from those in individuals in the AIDS death group. The proportion in the pre-AIDS group was higher for IDU (15.2% versus 3.7%), haemophiliacs (22.6% versus 5.7%) and individuals who acquired HIV infection through blood transfusion (2.3% versus 0.8%) and lower for homosexual men (41.7% versus 76.4%) and heterosexual individuals (8.8% versus 11.2%). Pre-AIDS death as a proportion of all HIV related deaths reported to CDSC (excluding the 227 deaths for which there was no clinical information) is presented in Fig. 2 by exposure category. Eighteen per cent of HIV-infected IDU, 17.8% of HIV-infected haemophiliacs and 12.9% of HIV-infected blood recipients were identified as having died prior to AIDS during the period study 1982–1996.
The results of multivariate analysis are presented in Table 3 which gives the numbers in the different groups along with adjusted OR and 95% CI. There were no significant interactions (χ2 test of two-way interactions: P = 0.16). The OR in Table 3 represent independent effects, jointly estimated from the multiple regression model. The risk of an HIV-infected individual dying prior to AIDS was related to, as base line, deaths among HIV-positive individuals in the following reference groups: homosexual men, residents of the Thames regions and those < 30 years of age at death. This risk was higher for haemophiliacs, IDU and blood/tissue recipients and also higher for individuals outside the Thames regions, and for individuals who died at an older age.
Surveillance of the HIV epidemic in England, Wales and Northern Ireland during the period 1982–1996 shows that pre-AIDS deaths make a substantial contribution to mortality in HIV-infected individuals. Trends over time in the number of ascertained pre-AIDS and AIDS deaths were not significantly different, the proportion of pre-AIDS deaths being constant at around 5%. Because of possible delay in reporting, pre-AIDS and AIDS deaths identified in 1994, 1995 and 1996 were not included in the time trends analysis, so that the revised AIDS case definition used from 1993  would not have influenced our conclusions.
Pre-AIDS death was more likely to occur in HIV-infected IDU, haemophiliacs and blood transfusion recipients than those infected by other routes, in those reported from outside the Thames regions and in older individuals. HIV transmission by blood transfusion or administration of blood products effectively ceased in developed countries in 1985, but the use of contaminated equipment and needle sharing continue to expose IDU to HIV, as well as to hepatitis B virus (HBV) and hepatitis C virus (HCV). It is therefore likely that IDU will become the leading risk factor of pre-AIDS death in the near future, even in the UK where injecting drug use forms a small part of the total epidemic. In England, Wales and Northern Ireland during the study period, 18.4% HIV-infected IDU died prior to AIDS.
Studies reported from other countries have consistently showed higher pre-AIDS mortality rates in IDU than in other risk groups. For instance, in a cohort in Amsterdam, after 6.5 years of follow-up an estimated 43.7% of the homosexual men were diagnosed with AIDS and 0.7% died without an AIDS diagnosis, whereas 32.7% of the IDU were diagnosed with AIDS and 19.8% died without an AIDS diagnosis . High pre-AIDS mortality rates lead to substantial underestimates of the HIV epidemic in these populations if the estimation relied only on AIDS reporting .
Pneumonia and septicaemia were leading causes of pre-AIDS death identified by the surveillance scheme. Pneumonia due to an unspecified organism became the leading secondary cause of death among individuals dying from HIV infection in the USA during the period 1987–1992 . Male death rates for septicaemia, pneumonia and influenza were higher in high-AIDS-incidence states than in low-AIDS-incidence states in the USA between 1980 and 1987 . In addition, bacterial pneumonia has also been recognized as more frequent in HIV-infected individuals than in the general population, particularly among those with CD4 lymphocyte counts < 0.2 × 106/l and in IDU [25–32].
Liver disease, the second most important cause of pre-AIDS deaths, accounted for 13.2%. Hepatic cirrhosis is commonly reported as the cause of death in IDU and haemophiliacs, two groups which have been recognized as at higher risk of (HBV and HCV) infection through exposure to blood and blood products [33,34]. Although co-infection by HIV and HBV or HCV was mentioned on only a minority of death certificates or death reports, it has been assumed that viral co-infection has affected most of the patients who were reported as having died from hepatic cirrhosis or liver failure. Virological studies have shown that mean HCV RNA levels are much higher in HIV-infected patients than in HIV-negative controls and that the level does not correlate with the CD4 cell count [35,36]. Liver damage, however, correlates with HCV viraemia . These virological data support the likelihood of a more rapidly progressive HCV-related liver disease in HIV-infected patients resulting in cirrhosis or liver failure as likely causes of pre-AIDS death in haemophiliacs and IDU. All previously mentioned cohort studies of HIV-infected haemophiliacs and IDU identified liver failure or hepatic cirrhosis as a major cause of pre-AIDS death [9,11,13,14,17,38,39].
Several malignancies which are not classified as AIDS-defining have been identified. Among them, Hodgkin's disease and anal cancer were expected to be identified as causes of pre-AIDS death because their high incidence and their poorer prognosis has been established in HIV-infected patients [40–46]. Case series of lung cancer in HIV-infected individuals have been reported [47,48] and case-control studies have suggested that in this group the cancer would be more likely to be adenocarcinoma, to occur at a younger age, to be more advanced disease and have a shortened survival compared with all cases of lung cancer [49–50]. Small case series of solid tumours including gastro-intestinal, cutaneous and germ cell carcinoma have been reported in HIV-infected individuals [51–52].
One-fifth of pre-AIDS deaths were due to violent causes including suicides and accidental overdoses. Overdose is widely reported as a leading cause of pre-AIDS death in HIV-infected IDU [10,11,13,14,16,53–55]. We attributed pre-AIDS death to suicide by an overdose when both items were mentioned in the reports. If suicide was not mentioned then pre-AIDS deaths due to overdose were considered to be accidental.
A recent case note audit of HIV-infected individuals in London showed a bimodal distribution of suicidal acts with peaks at or around the diagnosis or infection and again at end-stage illness . Death from suicide/overdose in a cohort of IDU was found to be twice as common in HIV-infected IDU than in HIV-seronegative IDU .
Our surveillance data on pre-AIDS mortality in England, Wales and Northern Ireland has several limitations. There were 8574 cases known to have fulfilled the AIDS case definition by the time of death, but of the 942 reports of deaths in HIV-infected individuals without ascertained AIDS initially included in this review 227 gave no clinical information on the cause of death or the clinical course prior to death. For a further 247 patients we were not certain whether or not they had fulfilled the AIDS case definition by the time of death (131 probable and 116 possible AIDS cases). The majority of these 474 cases will represent unreported AIDS cases but some of them are probably unreported pre-AIDS deaths. The 468 reports included in this study may therefore not represent all cases of pre-AIDS deaths. The estimate for completeness reporting AIDS cases is 87% for England and Wales and if the same under-reporting rate existed for pre-AIDS death, this would add an extra 70 cases.
The quality of our data also depends on how accurately and thoroughly the causes of pre-AIDS death were reported. Causes of death represent what was written on the surveillance form or the death certificate for the national register of deaths held at the ONS. No attempt has been made to verify these data with the clinicians.
Incompleteness of reporting to our surveillance scheme may have biased some of our results. It may explain, for instance, the finding that living outside the Thames regions is an independent risk factor for pre-AIDS mortality among HIV-infected individuals.
A small but constant proportion of patients are known to have died before developing AIDS. This has led to an underestimation of the burden of the HIV epidemic when it has been based on reported AIDS cases. The situation is likely to be less stable in the future as combination therapies and primary prophylaxis for opportunistic infections postpone the occurrence of AIDS-defining events . These are now associated with a lower level of immunosuppression as assessed by the CD4 cell count . Non-AIDS-defining events that could lead to pre-AIDS deaths such as bacterial infections, liver cirrhosis and cancer may be more likely to occur in an ageing group of treated patients. Such factors may compromise the continuing relevance of the clinical AIDS case definition to the objectives of the epidemiological surveillance of HIV infection in industrialized countries . Surveillance of HIV infection should take account of the probably increasing number of pre-AIDS deaths, particularly in IDU, in assessing the development of the HIV epidemic, especially in countries where injecting drug use is an important source of HIV transmission.
We thank the reporters to the National Surveillance Scheme at CDSC and the Oxford Haemophilia Centre who report on behalf of the UK Haemophilia Centre Directors Organisation for patients infected by blood factor treatment. We also acknowledge the help of staff in the PHLS AIDS and STD Centre with data analysis.
1. Centers for Disease Control: Revision of the case definition for acquired immunodeficiency syndrome
2. Centers for Disease Control: 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults
3. Ancelle Park RA: European AIDS definition
4. Quinn TC: Global burden of the HIV pandemic
5. Stoneburner RL, Des Jarlais DC, Benezra D, et al.
: A larger spectrum of severe HIV-1-related disease in intravenous drug users in New York city
6. Perucci CA, Davoli M, Rapiti E, Abeni DD, Forastiere F: Mortality of intravenous drug users in Rome: a cohort study
. Am J Public Health
7. Selwyn PA, Alcabes P, Hartel D, et al.
: Clinical manifestations and predictors of disease progression in drug users with human immunodeficiency virus infection
. N Engl J Med
8. Frisher M, Bloor M, Goldberg D, Clark J, Green S, McKeganey N: Mortality among injecting drug users: a critical reappraisal
. J Epidemiol Community Health
9. Eskild A, Magnus P, Samuelsen SO, Sohlberg C, Kittelsen P: Differences in mortality rates and causes of death between HIV positive and HIV negative intravenous drug users
. Int J Epidemiol
10. Zaccarelli M, Gattari P, Rezza G, et al.
: Impact of HIV infection on non-AIDS mortality among Italian injecting drug users
11. Van Haastrecht HJA, Van den Hoek A(J)AR, Coutinho RA: High mortality among HIV-infected injecting drug users without AIDS diagnosis: implications for HIV infection epidemic modellers? AIDS
12. Van Haastrecht HJA, Van Ameijden EJC, Van den Hoek JAR, Mientjes GHC, Bax JS, Coutinho RA: Predictors of mortality in the Amsterdam cohort of human immunodeficiency virus (HIV)-positive and HIV-negative drug users
. Am J Epidemiol
13. Eyster ME, Schaefer JH, Ragni NV, et al.
: Changing causes of death in Pennsylvania's hemophiliacs 1976 to 1991: impact of liver disease and acquired immunodeficiency syndrome
14. Rosendaal FR, Varekamp I, Smit C: Mortality and causes of death in Dutch haemophiliacs, 1973–86
. Br J Hematol
15. Diamondstone LS, Blakley SA, Rice JC, Clark RA, Goedert JJ: Prognostic factors for all-cause mortality among hemophiliacs infected with human immunodeficiency virus
. Am J Epidemiol
16. Darby SC, Rizza CR, Doll R, Spooner RJD, Stratton IM, Thakrar B: Incidence of AIDS and excess of mortality associated with HIV haemophiliacs in the United Kingdom: report on behalf of the directors of haemophilia centres in the United Kingdom
17. Darby SC, Ewart DW, Giangrande PLF, Spooner JD, Rizza CR, the UK Haemophilia Centre Directors' Organisation: Importance of age at infection with HIV-1 for survival and development of AIDS in UK haemophilia population
18. Madden PB, Lamagni T, Hope V, Bennett D, Golberg D: The HIV epidemic in injecting drug users
. Commun Dis Rep CDR Rev
19. Public Health Laboratory Service AIDS Centre: Clinical reports of deaths in HIV-infected persons without AIDS in the United Kingdom to the end of September 1990
. Commun Dis Rep CDR Wkly
20. Public Health Laboratory Service AIDS Centre: The surveillance of HIV-1 infection and AIDS in England and Wales
. Commun Dis Rep CDR Rev
21. Dean AG, Dean JA, Coulombier D, et al.
: Epi Info Version 6: a Word Processing, Database, and Statistics Program for Epidemiology on Microcomputers.
Atlanta: Centers for Disease Control and Prevention, 1994.
22. Francis B, Green K, Payne C (Eds): The Glim System: Generalised Interactive Modelling.
Oxford: Oxford Science Publications; 1983.
23. Selik RM, Chu SY, Ward JW: Trends in infectious diseases and cancers among persons dying of HIV infection in the United States from 1987 to 1992
. Ann Intern Med
24. Buehler JW, Devine OJ, Berkelman RL, Chevarley FM: Impact of the human immunodeficiency virus epidemic on mortality trends in young men, United States
. Am J Public Health
25. Buehler JW, Hanson DL, Chu SY: The reporting of HIV/AIDS deaths in women
. Am J Public Health
26. Miller R: HIV-associated respiratory diseases
27. Hirschtick RE, Glassroth J, Jordan MC, et al.
: Bacterial pneumonia in persons infected with the human immunodeficiency virus
. N Engl J Med
28. Caiaffa WT, Graham NMH, Vlahov D: Bacterial pneumonia in adult populations with human immunodeficiency virus (HIV) infection
. Am J Epidemiol
29. Caiaffa WT, Vlahov D, Graham NMH, et al.
: Drug smoking
, Pneumocystis carinii pneumonia, and immunosuppression increase risk of bacterial pneumonia in human immunodeficiency virus-seropositive injection drug users
. Am J Respir Crit Care Med
30. Selwyn PA, Feingold AR, Hartel D: Increased risk of bacterial pneumonia in HIV-infected intravenous users without AIDS
31. Mientjes GH, Van Ameijden EJ, Van den Hoek AJAR, Coutinho RA: Increasing morbidity without rise in non-AIDS mortality among HIV-infected intravenous users in Amsterdam
32. Centers for Disease Control: Increase in pneumonia mortality among young adults and the HIV epidemic – New York City, United States
33. Majid A, Holmes R, Desselberger U, Simmonds P, McKee T: Molecular epidemiology of hepatitis C virus among intravenous drugs users in rural communities
. J Med Virol
34. Esteban R: Epidemiology of hepatitis C virus infection
. J Hepatol
35. Soriano V, Bravo R, Mas A, Garcia-Samaniego J, the Hepatitis-HIV Spanish study group: Hepatitis C viraemia in HIV-infected patients [letter]
36. Cribier B, Rey D, Schmitt C, Lang JM, Kirn A, Stoll-Keller F: High hepatitis C viraemia and impaired antibody response in patients coinfected with HIV
37. Lau J, Davis G, Kniffen J, et al.
: Significance of serum hepatitis C virus RNA levels in chronic hepatitis
38. Eysler ME, Diadmondstone LS, Lien JM, Ehmann WC, Quan S, Goedert JJ: Natural history of hepatitis C virus infection in multitransfused hemophiliacs: effect of coinfection with human immunodeficiency virus. The Multicenter Haemophilia Cohort Study
. J Acquir Immune Defic Syndr
39. Telfer P, Sabin C, Devereux H, Scott F, Dusheiko G, Lee C: The progression of HCV-associated liver disease in a cohort of haemophilia patients
. Br J Haematolol
40. Schulz TF, Boshoff CH, Weiss RA: HIV infection and neoplasia
41. Reynolds P, Saunders LD, Layefsky ME, Lemp GF: The spectrum of acquired immunodeficiency syndrome (AIDS) - associated malignancies in San Francisco
. Am J Epidemiol
42. Gerolds M, Adler R: Hodgkin's disease as an indicator of AIDS
. Med Hypotheses
43. Kaplan LD: HIV-associated lymphoma
. AIDS Clin Review
44. Melbye M, Cote TR, Kessler L, Gail M, Biggar RJ, AIDS/Cancer Working Group: High incidence of anal cancer among AIDS patients
45. Northfelt DW: Cervical and anal neoplasia and HPV infection in persons with HIV infection
46. Palefsky JM: Anal human papillomavirus infection and anal cancer in HIV-positive individuals: an emerging problem
47. Flores MR, Sridhar KS, Thurer RJ, Saldana M, Raub WA, Klimas NG: Lung cancer in patients with human immunodeficiency virus infection
. Am J Clin Oncology
48. Gunthel CJ, Northfelt DW: Cancers not associated with immunodeficiency in HIV infected persons
49. Karp J, Profeta G, Marantz PR, Karpel JP: Lung cancer in patients with immunodeficiency syndrome
50. Sridhar KS, Flores MR, Raub WA, Saldana M: Lung cancer in patients with human immunodeficiency virus infection compared with historical control subjects
51. Remick SC, Boguniewicz A, Wolf B: Solid tumours in HIV-infected patients other than AIDS defining neoplasms
. Adv Exp Med Biol
52. Remick SC: The spectrum of non AIDS-defining neoplastic disease in HIV infection
. J Invest Med
53. Mezzelani P, Venturini L, Quaglio GL, Lugoboni F, Scientific Intercenters Collaborating Drug Users Group, Des Jarlais DC: A multicenter study on deaths among injecting drug users. AIDS has overtaken overdose as the principal cause of death
. XI International Conference on AIDS
Vancouver, July 1996 [abstract Tu.C.2514].
54. Van Ameijden E, Vlahov D, Van den Hoeck JAR, Flynn C, Coutinho RA: Pre-AIDS mortality and morbidity among injection drug users (IDU) in cohort studies in Amsterdam and Baltimore
. XI International Conference on AIDS
. Vancouver, July 1996 [abstract Th.C.220].
55. Van Haastrecht HJA, Mientjes GHC, Van den Hoek ANJR, Coutinho RA: Death from suicide and overdose among drug injectors after disclosure HIV test result
56. Sherr L: Suicide and AIDS: lessons from a case note audit in London
. AIDS Care
57. Deeks SG, Smith M, Holodnly M, Kahn JO: HIV-1 protease inhibitors. A review for clinicians
58. Rogers PA, Gore SM, Whitmore-Overton SE, Allardice GM, Swan AV, Noone A: United Kingdom AIDS survival in adults
59. Albrecht H: Redefining AIDS: towards a modification of the current AIDS case definition
. Clin Infect Dis
Keywords:© Lippincott-Raven Publishers.
Pre-AIDS mortality; mortality; surveillance