icon-folder.gif   Conference Reports for NATAP  
  ICAAC 2015 55th Interscience Conference
on Antimicrobial Agents and Chemotherapy
September 5-9, 2015, San Diego, CA
Back grey_arrow_rt.gif
HIV Hospital Mortality Flat From 1993 to 2013, But Non-AIDS Death Rate Surges...... Over that period the proportion of in-hospital deaths attributable to non-AIDS causes rose from 43% to 70.5% (P < 0.0001)
  ICAAC 2015, September 17-21, 2015, San Diego
Mark Mascolini
Hospital mortality among HIV-positive people barely changed from 1993-2002 to 2003-2013 in a 1937-person study in Spain [1]. Non-AIDS illness accounted for just over one third of hospital deaths in the earlier period and for two thirds in the later period.
To get a better understanding of how hospital admission, readmission, and mortality rates have changed since the early days of combination antiretroviral therapy, clinician-researchers at the 1422-bed NIBIC-University Hospital in La Coruna, Spain, conducted this retrospective analysis. La Coruna is a provincial capital of a quarter-million people on the Bay of Biscay in northwest Spain.
For this analysis, the researchers divided follow-up time into two periods, 1993-2002 and 2003-2013. During 22,901 patient-years 1937 HIV-positive people get admitted to the hospital 6917 times. Three quarters of those 1937 were men, and 37% had HCV infection. Their age averaged 36 years (+/- 11).
The overall hospital admission rate fell from 30.7 per 100 patients in 1993 to 19.9 in 2013, a highly significant decline (P < 0.001). The admission rate fell 4.9% yearly after 1996, but the rate dropped only 1.7% yearly among people coinfected with HIV and HCV. Coinfected people had more hospital admissions than HIV-monoinfected people over the study period (3 versus 2 per patient, P < 0.001), and they had a higher 30-day readmission rate (21.5% versus 19.1%, P < 0.001).
For all patients, median hospital stay dwindled from 9.0 days in 1993-2002 to 8.0 days in 2003-2013, a nonsignificant change (P = 0.2). From the earlier period to the later period, the 30-day readmission rate varied little, from 19.8% to 20.4% (P = 0.6).
From 1993-2002 to 2003-2013, two reasons for admission dropped significantly--infectious disease (49.1% to 35.3%) and psychiatric illness (9.1% to 5.2%). But admission rates rose significantly for malignancies (3.6% to 7.8%), cardiovascular disease (1.2% to 3.7%), gastrointestinal disease (9.2% to 16.1%), and chronic respiratory disease (5.6% to 10.9%).
In-hospital mortality did not change substantially from 1993-2002 (6.8%) to 2003-2013 (6.3%). Non-AIDS illnesses accounted for just more than one third of in-hospital deaths in 1993-2002 (37.9%); that proportion climbed to two thirds in 2003-2013 (68.3%).
From the earlier period to the later period, AIDS encephalitis, mycobacterial infection, and Pneumocystis pneumonia accounted for falling proportions of deaths. Rising proportions of deaths could be attributed to AIDS cancers and non-AIDS cancers, bacterial pneumonia, sepsis, liver disease, chronic respiratory disease, and cardiovascular disease.
A recently published study from Yale-New Haven Hospital found that the proportion of HIV-positive hospitalized patients who died dropped from 6.2% in 1995 to 1.5% in 2011, a highly significant decline (P< 0.0001) [2]. Over that period the proportion of in-hospital deaths attributable to non-AIDS causes rose from 43% to 70.5% (P < 0.0001), usually non-AIDS infection (20.3%), cardiovascular disease (11.3%), liver disease (8.5%), and non-AIDS malignancy (7.8%). Multivariate logistic regression linked non-AIDS death to 4.5 times higher odds of underlying liver disease and 4.2 times higher odds of underlying cardiovascular disease.
1. Meijide Míguez H, Mena de Cea A, Rodríguez Osorio I, et al. Trends in hospitalizations, readmissions and in-hospital mortality in HIV-infected patients between 1993-2013: impact of hepatitis C coinfection. ICAAC 2015, September 17-21, 2015, San Diego.
2. Cowell A, Shenoi SV, Kyriakides TC, Friedland G, Barakat LA. Trends in hospital deaths among human immunodeficiency virus-infected patients during the antiretroviral therapy era, 1995 to 2011. J Hosp Med. 2015;10:608-614. http://onlinelibrary.wiley.com/doi/10.1002/jhm.2409/abstract