| ObjectiveTo investigate the factors affecting olfactory recovery after endoscopic sinus surgery(ESS)in patients with chronic rhinosinusitis with nasal polyposis(CRSw NP).To evaluate the effect of glucocorticoid(GC)and olfaction training(OT)on olfactory dysfunction(OD)in patients with eosinophilic chronic rhinosinusitis with nasal polyps(ECRSw NP)and non Eosinophilic chronic rhinosinusitis with nasal polyps(Non ECRSw NP).Methods(1)Patients with bilateral CRSw NP and olfactory dysfunction(OD)(n = 213)underwent ESS and were defined as the olfactory recovery group(n = 99)according to TDIS ≥ 30.75 at 2 weeks after surgery,otherwise as the olfactory dysfunction group(n = 114).Logistic regression analysis and support vector machine(SVM)were used to construct clinical prediction models based on preoperative basic clinical data including Patient Health Questionnaire-2(PHQ-2),the 22-item sino-nasal outcome tests(SNOT-22),Questionnaire of olfactory disorders-negative statements(QOD-NS),Lund-Kennedy endoscopic scores(LKES),Lund-Mackay scores(LMS)and Olfactory cleft scores(OCS).(2)Patients in the olfactory dysfunction group were randomized to receive GC(n =38),OT(n = 36),and GC combined with OT(GCw OT)(n = 34)to complete treatment follow-up.And the phenotype of patients(ECRSw NP and Non ECRSWNP)was analyzed retrospectively.(3)Patients were followed up at 1 month,3 months and 6 months after treatment,including SNOT-22,olfactory fluctuation questionnaire,LKES and TDIS.Results(1)Age,sex,smoking and OCS were the factors affecting olfactory recovery in CRSw NP patients after ESS,of which the most important were older age and higher preoperative OCS values.SVM model(AUC = 0.918)classification accuracy was higher than logistic regression model(AUC = 0.775).(2)OCS values were positively correlated with LMS values in patients in the ECRSw NP group(r = 0.731,p < 0.001),and OCS values were also positively correlated with LMS values in patients in the Non ECRSw NP group(r = 0.473,p <0.001).(3)QOD-NS values were negatively correlated with LMS values(r =-0.247,p =0.010),but not significantly correlated with LKES,OCS and TDIS(p > 0.05).(4)The change value of SNOT-22 score,LKES values and TDIS values from baseline to each follow-up visit after treatment was recorded as △SNOT-22,△LKES and △TDIS,respectively,using the preoperative evaluation as the baseline.There was no statistical difference in △SNOT-22 between the groups during the treatment observation period.In the ECRSw NP group,△LKES was significantly greater in the GCw OT group than in the OT group at 1 month of treatment(p < 0.05),and △TDIS was significantly greater in the GCw OT group than in the GC group at 3 and 6 months of treatment(p < 0.05).In the Non ECRSw NP group,△LKES was significantly greater in the GCw OT group than in the OT group at 1 and 3 months of treatment(p < 0.05),and △TDIS was significantly greater in the GCw OT group than in the OT group at 1 month of treatment(p < 0.05).(5)According to the change value of TDIS ≥ 6,olfactory therapy is defined as effective,otherwise,it is ineffective.No statistical difference was observed in the comparison of olfactory treatment response rates between the groups during the treatment observation period.Fluctuations in olfaction were considered to occur at frequencies greater than once a week,otherwise they were considered non-fluctuating.The incidence of olfactory fluctuations in the GCw OT group was significantly lower than in the GC group during the treatment observation period in the ECRSw NP group(p < 0.0125).Conclusion(1)Age and preoperative OCS were the most significant factors affecting olfactory recovery in CRSw NP patients after surgery.(2)The clinical prediction model constructed by SVM method is more suitable for predicting olfactory recovery in CRSw NP patients after surgery than logistic regression model.(3)In olfactory therapy,the GCw OT group showed a lower incidence of olfactory fluctuations in the ECRSw NP group compared with the GC group. |
PDF Full Text Request