
Methods: This is a retrospective cohort study including all HIV pts operated between Jan 2010 to Dec 2015 at our hospital. The primary outcome was 30-day mortality and secondary outcome was PO infection. Potential risk factors evaluated included surgical factors, duration of HIV, WHO clinical stage, preoperative (preop) CD4 count and albumin, use of ART, ICU admission and use of antibiotics. For binary variables, chi square test or Fisher exact test and for continuous variables, Mann Whitney test were used. All p values were 2 sided. SPSS was used for statistical analysis.
Results: Among 2844 admissions with HIV, 293 pts underwent 374 surgeries. The median age was 40 (3 - 85 years) with 252 surgeries (67%) on males. Emergency surgeries were 134 (35%), open 292 (78%) and abdomino-pelvic 225 (60%). The commonest indication was debridement or drainage of infection/abscess (101, 27%) followed by obstetric reasons (49, 13%). At the time of surgery, median preop CD4 count was 389 (9 - 2159 ) cells/mm3 and 153 (40%) pts were ART naive. Mortality at day 30 was 2% (n = 8) and PO infection rate was 4% (n = 15). The foci of infection were surgical site in 9 (60%), pneumonia in 5 (33%). Of 38 blood cultures sent, 5 (13%) were positive. The median duration of hospital stay was 5 (1-60) days. Mortality was associated with PO infection (p < 0.001), emergency surgery (p = 0.018), low preop albumin (p = 0.017), ICU stay (p < 0.0001) and empirical antifungal use (p<0.001). PO infection was associated with preop CD4 count < 200 cells/mm3(p = 0.002), low preop albumin (p = 0.007) and ICU admission (p < 0.001).
Conclusion: PO mortality and infections are low in HIV patients and are probably comparable to those without HIV. Identifying preop risk factors may help in stratifying patients at increased risk of PO morbidity and mortality.Addressing these may possibly help in improving outcomes.

D. S,
None
N. Gracelin, None
L. Roy, None
P. Rupali, None