Scholarlink

Objective <p>For individuals diagnosed with obstructive sleep apnea (OSA), uvulopalatopharyngoplasty (UPPP) represents a frequently utilized surgical intervention, yet its postoperative efficacy varies significantly among individuals. Identifying key preoperative predictors of surgical success is crucial for optimizing patient selection and enabling personalized treatment strategies.</p> Methods <p>This retrospective study included 191 OSA patients who underwent UPPP. Postoperative outcomes were assessed at a three-month follow-up, and a multivariate logistic regression model combining body mass index (BMI), apnea–hypopnea index (AHI), peripheral blood oxygen saturation (SpO₂), Epworth Sleepiness Scale (ESS) score, visual analogue scale (VAS) score, and Friedman staging was constructed to predict surgical efficacy. Surgical success was defined as a postoperative AHI reduction &gt; 50% or a final AHI &lt; 20 events/hour. Multivariate logistic regression and ROC curve analysis were used to identify predictors of surgical efficacy.</p> Results <p>UPPP significantly reduced AHI, ESS, and VAS scores, while elevating SpO₂, with 64.4% of patients meeting efficacy criteria. Preoperative SpO₂ (<i>OR</i> = 1.297), ESS (<i>OR</i> = 1.351), and VAS (<i>OR</i> = 3.635) were independent positive predictors, while BMI (<i>OR</i> = 0.644) was a negative predictor. Friedman stage I (<i>OR</i> = 0.054) and II (<i>OR</i> = 0.002) predicted better outcomes than stage III. The combined prediction model achieved an area under the curve (AUC) of 0.972. Patients with fewer residual symptoms postoperatively had higher preoperative SpO₂, lower BMI and VAS scores, and significantly higher satisfaction (95.8% vs. 62.2%). Complication rates were lower in both the surgical success group and the low-residual symptom groups.</p> Conclusion <p>The multivariate model integrating key preoperative predictors demonstrated excellent discriminative performance (AUC = 0.972) and offers substantial clinical utility for improving preoperative risk assessment, optimizing surgical decision-making, and enhancing postoperative patient satisfaction.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predictive factors for surgical outcomes and symptom persistence after uvulopalatopharyngoplasty in patients with obstructive sleep apnea

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月18日

查看全文
  • Xiajun Ling,
  • Jun Zheng,
  • Qiurong Liu,
  • Chunlin Shi,
  • Chi Fan

摘要

Objective

For individuals diagnosed with obstructive sleep apnea (OSA), uvulopalatopharyngoplasty (UPPP) represents a frequently utilized surgical intervention, yet its postoperative efficacy varies significantly among individuals. Identifying key preoperative predictors of surgical success is crucial for optimizing patient selection and enabling personalized treatment strategies.

Methods

This retrospective study included 191 OSA patients who underwent UPPP. Postoperative outcomes were assessed at a three-month follow-up, and a multivariate logistic regression model combining body mass index (BMI), apnea–hypopnea index (AHI), peripheral blood oxygen saturation (SpO₂), Epworth Sleepiness Scale (ESS) score, visual analogue scale (VAS) score, and Friedman staging was constructed to predict surgical efficacy. Surgical success was defined as a postoperative AHI reduction > 50% or a final AHI < 20 events/hour. Multivariate logistic regression and ROC curve analysis were used to identify predictors of surgical efficacy.

Results

UPPP significantly reduced AHI, ESS, and VAS scores, while elevating SpO₂, with 64.4% of patients meeting efficacy criteria. Preoperative SpO₂ (OR = 1.297), ESS (OR = 1.351), and VAS (OR = 3.635) were independent positive predictors, while BMI (OR = 0.644) was a negative predictor. Friedman stage I (OR = 0.054) and II (OR = 0.002) predicted better outcomes than stage III. The combined prediction model achieved an area under the curve (AUC) of 0.972. Patients with fewer residual symptoms postoperatively had higher preoperative SpO₂, lower BMI and VAS scores, and significantly higher satisfaction (95.8% vs. 62.2%). Complication rates were lower in both the surgical success group and the low-residual symptom groups.

Conclusion

The multivariate model integrating key preoperative predictors demonstrated excellent discriminative performance (AUC = 0.972) and offers substantial clinical utility for improving preoperative risk assessment, optimizing surgical decision-making, and enhancing postoperative patient satisfaction.