Abstract
Background
For HIV patients admitted with sepsis, ARVs are often stopped or held due to myriad concerns including drug interactions, acute renal failure, gastrointestinal dysfunction, or inability to administer crushed medications down feeding tubes. We seek to examine prescription patterns of HAART for HIV positive patients admitted for sepsis in our healthcare system and the impact of HAART prescription on patient outcomes.
Methods
We identified HIV positive patients from an institutional database of patients admitted for sepsis within our multi-hospital healthcare system and retrospectively extracted further clinical patient and laboratory information as well as information on HAART prescription by chart review. The impact of HAART prescription and immunologic and virologic parameters of HIV infection on mortality was examined.
Results
Inpatient mortality was 35% in HIV patients admitted for sepsis, compared with 17% for all patients with sepsis in our healthcare system. Opportunistic infections were identified in only 25% of patients while 56% had other infections identified. Only 55% of patients had HAART prescribed while inpatient. CD4 count, virologic suppression, APACHE score, presence of an opportunistic infection, admission to a tertiary care hospital, and inpatient prescription of HAART were all predictors of survival.
Table:
Factors impacting mortality in HIV patients with sepsis in univariate analysis.
| Survivors (n = 50) | Deaths (n = 28) | Odds ratio of survival (p value) | |
|---|---|---|---|
| Mean baseline CD4 count | 309 | 64 | (P < 0.01) |
| Virologic suppression (VL<200) | 48% (n = 21 of 44) | 22% (n = 5 of 23) | 3.3 (P < 0.05) |
| Mean APACHE score | 67 (n = 32) | 110 (n = 17) | (P < 0.01) |
| Opportunistic infection | 18% (n = 9) | 39% (n = 11) | 0.34 (P < 0.05) |
| Tertiary hospital admission | 50% (n = 25) | 21% (n = 6) | 3.7 (P < 0.05) |
| Inpatient HAART prescription | 68% (n = 34) | 32% (n = 9) | 4.5 (P < 0.01) |
In a multivariable analysis both CD4 count and inpatient HAART prescription predicted survival in our cohort with an odds ratio of survival of 3.3 for patients prescribed HAART inpatient compared with their untreated peers.
Conclusion
Immunologic and virologic status at time of admission predicted survival in HIV patients admitted for sepsis but prescription of HAART to HIV patients admitted for sepsis may increase survival.
Disclosures
C. Polk, Gilead Sciences: Investigator, Research support; Viiv Healthcare: Investigator, Research support