Influency of pentoxifylline treatment for neonatal sepsis: A meta-analysis of randomized controlled studies

Introduction: Pentoxifylline may be an important approach to treat neonatal sepsis. However, its use has not been well established. We conduct a systematic review and meta-analysis to evaluate the efficacy of pentoxifylline treatment for neonatal sepsis. Methods: PubMed, Embase, and the Cochrane Central Register of Controlled Trials are searched. Randomized controlled trials assessing the influence of pentoxifylline treatment on neonatal sepsis are included. Two investigators independently have searched articles, extracted data, and assessed the quality of included studies. This meta-analysis is performed using the random-effect model. Results: Seven randomized controlled trials involving 439 patients are included in the meta-analysis. Compared with control intervention for neonatal sepsis, pentoxifylline treatment is associated with reduced hospital stay (standard mean differences = -0.61; 95% confidence interval = -0.93 to -0.29; p = 0.0002) and metabolic acidosis (risk ratio = 0.38; 95% confidence interval = 0.22 to 0.66; p = 0.0006), but has no remarkable impact on mortality (risk ratio = 0.59; 95% confidence interval = 0.30 to 1.16; p = 0.13), serum tumor necrosis factor-alpha (standard mean differences = -0.38; 95% confidence interval = -1.29 to 0.52; p = 0.41), serum C-reactive protein (standard mean differences = -0.25; 95% confidence interval = -0.92 to 0.42; p = 0.47), plasma interleukin-6 (standard mean differences = -0.13; 95% confidence interval = -0.41 to 0.15; p = 0.37), disseminated intravascular coagulopathy (risk ratio = 0.55; 95% confidence interval = 0.25 to 1.21; p = 0.14), and oliguria/anuria (risk ratio = 0.77; 95% confidence interval = 0.28 to 2.16; p = 0.62). Conclusion: Pentoxifylline treatment may be associated with reduced mortality and hospital stay in neonatal sepsis with caution.