| Abstrakt: |
Introduction Burn injuries are well known to cause a state of immunosuppression in patients. This can result in wound infections, a common complication in burn injuries, that can lead to sepsis and increased mortality. Human immunodeficiency virus (HIV) is also known to cause immunosuppression in patients. The outcomes of burn patients with pre-existing HIV infections, however, are not yet completely understood. We conducted a systematic review and meta-analysis to compare the outcomes of burn patients with pre-existing HIV against those without this chronic infection. Methods We searched MEDLINE (Pubmed), Google Scholar, Scopus, and Embase for studies that compared outcomes and complications between burn patients with and without HIV. From this search, we screened 445 articles. Through our selection criteria, five articles focusing on HIV patients were selected for systematic review and meta-analysis. Data were analyzed using the Cochrane Review Manager (RevMan) Data Analysis package to produce pooled odds ratios and mean differences from the random effect model. Results Five studies observing a total of 24,419 burn patients, published between 2000 and 2017, were included. Of these, two are prospective studies and three are retrospective chart reviews. The primary outcome of mortality for HIV+ patients compared to HIV- patients had an odds ratio of 2.04 (CI= 0.46–9.14) in the random effects model. Secondary outcomes of sepsis and wound infection odds ratios were 1.47 (CI= 0.44–4.99) and 1.10 (CI= 0.28–4.25), respectively. The length of stay (LOS) between studies showed an overall mean difference of 0.95 (CI=-8.08–9.99). Most studies had a greater proportion of male patients. TBSA between studies ranged from 13.1% and 35%. Conclusions From our results, we concluded that HIV+ had a tendency toward greater mortality (OR=2.04) and sepsis (OR=1.47). However, mortality and sepsis had confidence intervals of [0.46–9.14] and [0.44–4.99], respectively. Therefore, we cannot definitively state that HIV infection is responsible for greater mortality or sepsis in burn patients. Additionally, LOS analysis also showed a wide confidence interval [-8.08–9.99], preventing us from making reliable deductions about this outcome. We believe further research is needed before universal conclusion or recommendations are appropriate. [ABSTRACT FROM AUTHOR] |
