Clinical predictors of polyps recurring in patients with chronic rhinosinusitis and nasal polyps: a systematic review and meta-analysis.

Identification of perioperative risk factors for recurrent nasal polyps (RNPs) is important for selection of further treatment and determination of appropriate follow-up period. However, the relative prognostic significance of these risk factors has not been investigated.We compared the nasal symptoms, endoscopic polyp and Lund-Mackey computed tomography scores, and the laboratory and pathological findings of RNP and non-RNP patients. The risk of bias was assessed using the Newcastle-Ottawa scale.Patients with poor nasal symptom scores and olfactory dysfunctions and high Lund-Mackey computed tomography scores were at higher risk of postoperative RNPs, as were those with allergic conditions and elevated tissue and serum eosinophil levels. The tissue neutrophil counts/percentages were significantly lower in the RNP than the other group. The tissue eosinophil level was of higher diagnostic utility than the serum eosinophil level. The RNP diagnostic odds ratio afforded by the tissue eosinophil count or percentage was 54.1247. The area under the receiver operating characteristic curve was 0.936. The sensitivity and specificity were 0.8809 and 0.8834, respectively.The tissue eosinophil level reliably predicts RNP after endoscopic sinus surgery.