 |
|
|
| |
Changing Patterns of Causes of Death in the Swiss HIV Cohort Study (SHCS) 2005-2009
|
| |
| |
Reported by Jules Levin
CROI 2011 March 2 Boston
From Jules: they reported high rates of deaths due to non-AIDS malignancies (19%) & liver disease/HCC (18%), and non-AIDS causes of death is higher than AIDS (15%) as a cause of death, and particularly stunning note is 32% of deaths in HCV+ was due to HCV & HCC with the reported average age at 44 in HCV+.
"Malignancies were the most important underlying cause of death (total of non-AIDS + AIDS-malignancies = 25.5%), whereas AIDS-related deaths decreased to 15%. HCV co-infection substantially influenced the distribution of causes of death. The median CD4 count at time of death was increasing over calendar years to a value of modest cellular immunodeficiency (321 cells/μL)...... median age at death, 47 years (IQR 42 to 56)..... Underlying causes of death were: non-AIDS-defining malignancies (n = 87, 19.0%), including hepatocellular carcinoma, n = 13 (2.8%); AIDS (n = 74, 16.1%); liver diseases (excluding hepatocellular carcinoma) (n = 68, 14.8%); non-AIDS-defining infections (n = 42, 9.2%); heart disease and stroke (n = 35, 7.6%); alcohol and substance use (n = 33, 7.2%); suicide (n = 29, 6.3%)"
Martha Ruppik 1, Bruno Ledergerber 1, Martin Rickenbach 2, Hansjakob Furrer 3, Manuel Battegay 4, Matthias Cavassini 2, Bernard Hirschel 6, Enos Bernasconi 7, Pietro Vernazza 8, Markus Flepp 9, Rainer Weber 1, and the Swiss HIV Cohort Study
1 University Hospital and University of Zurich, Zurich, Switzerland
2 University Hospital Lausanne, Lausanne, Switzerland
3 University Hospital Bern, Bern, Switzerland
4 University Hospital Basel, Basel, Switzerland
6 University Hospital Geneva, Geneva, Switzerland
7 Cantonal Hospital Lugano, Lugano, Switzerland
8 Cantonal Hospital St. Gallen, St. Gallen, Switzerland
9 Zentrum fur Infektionskrankheiten, Klinik Im Park, Zurich, Switzerland
ABSTRACT
Background: Data on causes of death influence priorities in medical care and preventive measures. We aimed to describe characteristics of participants of the Swiss HIV Cohort Study (SHCS) who died between 2005 and 2009, and to study causes of death.
Methods: The SHCS is a prospective observational database established in 1988. Causes of death were previously documented, but the SHCS started to more comprehensively collect data associated with death in 2005 using the structured CoDe questionnaire (www.cphiv.dk/CoDe/tabid/55/Default.aspx). Because autopsy rates are decreasing, and many patients die outside the institution providing HIV care, the CoDe documentation aims to collect all necessary clinical data to determine immediate, contributing, and underlying causes of death.
Results: A total of 459 (5.1%) of 9053 SHCS participants under follow-up between 2005 and 2009 died. Characteristics of persons who died were: 340 male, 119 female; median age at death, 47 years (IQR 42 to 56); median duration since diagnosis of HIV infection, 13.9 years (range 0 to 27); ever on ART, 93%; median duration of ART, 9.5 years (IQR 5.5 to 12.2); median last CD4 count before death, 251 (114 to 430) cells/μL; HIV transmission risk, IDU (44%), heterosexual (28%), MSM (24%); hepatitis C virus (HCV) infection, 44%; active hepatitis B virus (HBV) infection, 11.1%; smoking, 65%; and former smoking, 17%. Autopsies were done in 89 (19%) deaths. Underlying causes of death were: non-AIDS-defining malignancies (n = 87, 19.0%), including hepatocellular carcinoma, n = 13 (2.8%); AIDS (n = 74, 16.1%); liver diseases (excluding hepatocellular carcinoma) (n = 68, 14.8%); non-AIDS-defining infections (n = 42, 9.2%); heart disease and stroke (n = 35, 7.6%); alcohol and substance use (n = 33, 7.2%); suicide (n = 29, 6.3%); other (n = 77, 16.7%); and unknown (n = 14, 3.1%). Proportions of causes of death are shown in the figure, stratified in patients with (n = 202) and without (n = 244) HCV co-infection (HCV status missing, n = 13). Median age at death increased from 45 years (IQR 41 to 52) in 2005 to 49 (44 to 56) in 2009; median CD4 counts increased from 257 (118 to 429) in 2005 to 321 (157 to 519) in 2009.
Conclusions: Malignancies were the most important underlying cause of death (total of non-AIDS + AIDS-malignancies = 25.5%), whereas AIDS-related deaths decreased to 15%. HCV co-infection substantially influenced the distribution of causes of death. The median CD4 count at time of death was increasing over calendar years to a value of modest cellular immunodeficiency (321 cells/μL).
Background
Data on causes of death inform priorities in medical care and preventive measures.
Patterns of morbidity and mortality in HIV-positive individuals taking antiretroviral therapy (ART) are changing as a result of immune reconstitution and improved survival (⇒ Figure 1).
Study Aims
We aimed to describe characteristics of participants of the Swiss HIV
Cohort Study (SHCS) who died between 2005 and 2009, and to
study causes of death.
Methods
Study population: The Swiss HIV Cohort Study (SHCS) is a
prospective observational cohort, established in 1988, with
continuous enrolment (www.shcs.ch).
Data collection: The SHCS
- initially described causes of death using broad categories (HIV-related, overdose of narcotics, suicide, accident, homicide, other,
unknown),
- introduced ICD-10 coding in 1999, and
- started to collected more comprehensive data associated with
death in 2005 using the structured CoDe form (http://www.cphiv.dk).
Autopsy rates are decreasing, and many patients die outside of an institution providing HIV care. Therefore, the CoDe documentation
aims to collect all necessary and available clinical and laboratory
data to determine
- immediate,
- contributing, and
- underlying causes of death.
Patient characteristics: ⇒ Table 1 summarizes patient characteristics: A total of 459 (5.1%) of 9053 SHCS participants
under follow-up died between 2005 and 2009. (Status of HCV-coinfection was available for 446 participants.)
Characteristics of persons who died were: 340 male, 119 female;
median age at death, 47 (IQR 42-56) yrs; median duration since
diagnosis of HIV infection, 14 (range 0-27) yrs; ever on ART, 93%;
median duration of ART, 9.5 (IQR 5.5-12.2) yrs; median last CD4
count before death, 251 (114-428) cells/μL; HCV and HBV infection, 45.3% and 11.4%; smoking (current plus ever), 82.3%.
Autopsies: were done in 89 (19%) deaths.
Underlying causes of death: ⇒ Table 1
- Non-AIDS-defining malignancies (n=85, 19.1%) [incl. hepatocellular carcinoma (HCC), n = 13 (2.8%)];
- AIDS (n = 73, 16.4%);
- liver diseases [excl. HCC] (n = 67, 15.0%);
- non-AIDS-defining infections (n = 42, 9.4%);
- heart diseases/stroke (n = 30, 6.7%);
- alcohol/substance use (n = 31, 7.0%);
- suicide (n = 28, 6.3%);
- unknown (n = 11, 2.5%).
Proportions of causes of death are shown in ⇒ Figure 2, stratified
according to CD4 cell count, age, and status of HCV infection (202
participants with, and 244 without HCV co-infection, HCV status
missing, n=13).
The most notable causes of death are LIVER DISEASE and HCC which appears to be the highest percent of deaths (15%+2.8%=18%) and non-AIDS malignancies which is 19%, which are the 2 leaders in cause of death, AIDS is 15%. If you look at the demographics in Table 1, the average age of HCV+ is only 44 (41-49) and of those HCV-neg (52 (44-64). Note the high suicide rate in HIV+. 32% of deaths in HCV+ is due to HCC or HCV. A point of the authors is that non-AIDS causes of death have surpassed AIDS causes by percent of deaths.
Note the high smoking rate of 80% which for sure contributes to non-AIDS malignancies and other causes of death and high psych & depression rates, and high alcohol & injection drug use rates.
|
| |
|
|
|
|
|
