Objective: A subset of intractable allergic rhinitis (iAR) patients experience severe symptoms which cannot be effectively controlled by standard drug therapy and/or antigen specific immunotherapy. In recent decades, endoscopy vidian neurectomy and posterior nasal nerve neurectomy (PNNN) were introduced as treatments of iAR that have shown to be highly successful at symptom management in a number of patients. But some patients experience relapse or suboptimal symptom control postoperation. To improve the effectiveness of PNNN to control iAR, a modified PNNN surgical approach (mPNNN) combined with accessory posterior nasal nerve neurectomy (aPNNN), which called as mPNNN-aPNNN was used. This study aims to compare the effects between mPNNN-aPNNN and PNNN on controlling the symptoms of iAR and evaluate the surgical effectiveness and safety of mPNNN-aPNNN. Methods: The patients with iAR experienced mPNNN-aPNNN or PNNN surgery at the department of Otolaryngology Head and Neck Surgery of the Second Xiangya Hospital, Central South University from January 2018 to December 2019 were analyzed retrospectively. The approach of PNNN, a selective resection of the posterior nasal nerve branches, was modified to the neurectomy of total branches of posterior nasal nerve at the sphenopalatine foramen, and combined the operation of aPNNN in which the accessory posterior nasal nerve at the palatine bone perpendicular plate was resect in our study. Daily Nasal Symptom Scores (DNSS), Total Rhinitis Medication Score (TRMS), and the Rhinoconjunctivitis Qualities of Life Questionnaires Scores (RQLQS) were used to evaluate the complications during the operation and after the operation at the 3rd, 6th, 12th, and 24th month postoperatively. Total Nasal Symptom Scores (TNSS) was used to assess the total effective rate and markedly effective rate of the operations. Results: A total of 140 iAR patients experienced mPNNN-aPNNN or PNNN. Those with concomitant septoplasty and/or inferior turbinate reduction, and were absent during the postoperative follow-up were excluded. The final 62 patients with mPNNN-aPNNN and 34 with PNNN were enrolled. DNSS, TNSS, TRMS, and RQLQS at the postoperation were significantly improved compared with the preoperation in all patients (all P<0.001). Compared with PNNN, the postoperative DNSS, TNSS, and TRMS of mPNNN-aPNNN were obviously improved (all P<0.001). There was a persisted relief of symptoms at the postoperation in all patients with mPNNN-aPNNN. The total effective rate and markedly effective rate at the postoperative 24th month were 100％ and 83.3％, respectively. Furthermore, the postoperative RQLQS decreased significantly (P<0.001). Only 5 sides of all patients (5/192, 2.6％) reported upper palate numbness during the first week after operation, with all recovered spontaneously in 1 month without treatment. No other postoperative complications occurred in mPNNN-aPNNN and PNNN.Conclusion: The surgery of mPNNN-aPNNN improve TNSS more significantly than PNNN. The operation of mPNNN-aPNNN is safe and effective to control iAR symptoms.