Abstract:

Objective To investigate the risk factors of postoperative recurrence in patients with primary brain glioma and to construct a prediction model. Methods A total of 98 patients with primary brain glioma treated by radical surgery in Nanchong Central Hospital from January 2018 to January 2021 were retrospectively included， and were divided into recurrent group （40 cases） and non-recurrent group （58 cases） according to whether there was recurrence or not during the follow-up period. The independent influencing factors for postoperative recurrence in patients with primary brain glioma were evaluated by multivariate logistic regression. Logistic prediction model of postoperative recurrence risk of patients with primary brain glioma was established， and the predictive efficacy of each index was calculated by receiver operator characteristic （ROC） curve. Results There were statistically significant differences between recurrent group and non-recurrent group in glioma World Health Organization （WHO） grade（χ²=12.48，P<0.001）， isocitrate dehydrogenase （IDH）1/2 mutation （χ²=13.24， P<0.001）， mean platelet volume （MPV）（t=5.34， P<0.001）， and MPV/platelet count （PLT）（t=9.73，P<0.001）. Multivariate analysis showed that WHO grade Ⅲ-Ⅳ （OR=8.54， 95%CI：1.62-44.99， P=0.011）， IDH1/2 wild type （OR=9.08， 95%CI：1.68-49.19， P=0.010）， low MPV （OR=0.46， 95%CI：0.21-0.99， P=0.048） and low MPV/PLT （OR=0.02， 95%CI：0.01-0.03， P<0.001） were independent risk factors for postoperative recurrence in patients with primary brain glioma. The logistic prediction model based on the above indicators was logit（P）=11.78+2.15×WHO grade+2.21×IDH1/2 mutation situation-0.78×MPV-200.70×MPV/PLT （R²=0.785）.The ROC curve analysis results showed that WHO grade， IDH1/2 mutation， MPV， MPV/PLT， and logistic prediction model P-value could all be used to predict the risk of postoperative recurrence in patients with primary brain glioma； The areas under the curve were 0.681， 0.684， 0.783， 0.920 and 0.964， respectively. In the area under ROC curve comparison of each indicator， the predictive performance of the logistic regression model P-value was significantly higher than that of other indicators （all P<0.05）. Conclusion Postoperative recurrence in patients with primary brain glioma may be related to glioma WHO grade， IDH1/2 mutation and platelet related laboratory indexes. The model constructed based on the above indicators can be used to predict the recurrence risk in patients with primary brain glioma after surgery.

Key words: Glioma, Recurrence, Prognosis, Blood platelets