Can preoperative neutrophil-to-lymphocyte ratio predict in-hospital mortality in postoperative patients with Stanford type A aortic dissection? Evidence-based appraisal by meta-analysis and GRADE

Abstract

Objectives: In-hospital mortality in postoperative patients with type A aortic dissection (AAD) is high. Neutrophil-to-lymphocyte ratio (NLR) is a novel predictor of adverse outcomes in many cardiovascular diseases. We examine NLR as a predictive tool in AAD in this meta-analysis. Materials and Methods: We systematically searched in four databanks. Risk of bias was appraised using the Quality Assessment of Diagnostic Accuracy Studies 2 tool. The midas and metandi commands in Stata 15 were used for the meta-analysis. The certainty of evidence was assessed by the Grading of Recommendations, Assessment, Development, and Evaluation methodology (GRADE). Results: Four studies with 502 cases in total were included. Pooled sensitivity and specificity were 0.71 (95% confidence interval [CI] = 0.52–0.79) and 0.64 (95% CI = 0.55–0.71), respectively. Area under the hierarchical summary receiver operating characteristic curve yielded 0.73 (95% CI = 0.68–0.76). The diagnostic odds ratio was 4.42 (95% CI = 2.56–7.62). Pooled positive and negative likelihood ratios yielded 1.98 (95% CI = 1.53–2.55) and 0.45 (95% CI = 0.32–0.62), respectively. When the pretest probabilities were 25%, 50%, and 75%, the positive posttest probabilities were 40%, 66%, and 86%, and the negative posttest probabilities were 13%, 31%, and 57%, respectively, according to the Fagan's nomogram plot. The overall certainty of evidence in GRADE was low and very low in sensitivity and specificity, respectively. Conclusion: The pooled diagnostic values of preoperative NLR, an inexpensive and routine laboratory examination, provide a practicable help for predicting in-hospital mortality for patients with postoperative AAD in our meta-analysis.