CO2激光初治外阴尖锐湿疣的评价与单次疗效预测

Outcome prediction of single-period CO2 laser therapy for vulvar condylomata acuminate

目的评价CO2激光初治外阴尖锐湿疣的疗效，探讨单次疗效影响因素并建立预测模型。方法前瞻性观察2009年3月至2010年12月，在北京协和医院、浙江大学医学院附属妇产科医院、华中科技大学同济医学院附属同济医院3家医院初诊为外阴尖锐湿疣并接受CO2激光治疗的患者，收集人口学和临床资料。Logistic回归多因素分析并建立预测模型。结果160例患者完成6个月的随访，失访16例，失访率9．1％。CO2激光单次治愈131例，单次治愈率82％；总体治愈150例，治愈率94％；50例出现不良反应，发生率31％，无严重不良反应发生。病变面积〉8cm2、阴道受累、失业是单次治疗失败的危险因素，术前疼痛是保护因素。建立单次疗效预测模型：Logit（Po）=-1．511＋1．573X1＋1．679X2＋3．254X3—1．685X4（X1～X4分别代表病变面积〉8cm2、阴道受累、失业和疼痛），模型ROC曲线下面积为0．816（P〈0．01），Po的最佳分界值为0．35。以此预测的灵敏度、特异度、阳性预测值和阴性预测值分别为58．6％、91．6％、60．7％和90．9％。结论CO2激光是治疗外阴尖锐湿疣安全有效的方法，单次疗效预测模型对临床决策有指导意义。

Objective To evaluate the outcome of CO2 laser treatment as primary therapy for vulvar condylomata acuminate and examine the risk factors and prediction model of single-period CO2 laser treatment. Methods Between March 2009 and December 2010, a muhicenter prospective study was conducted at three 3A hospitals of China （Peking Union Medical College Hospital, Zhejiang Women＇s Health Hospital ＆ Tongji Hospital）. All enrolled patients of vulvar condylomata acuminata received CO2 laser vaporization as the primary therapy and had return visits at 1, 3 and 6 months individually after treatment. Therapeutic recurrence and side effects were recorded. Logistic regression was used to analyze the associations between demographic or clinical characteristics and the outcome of single-period CO2 laser treatment and a prediction model was established subsequently. The optimal cutoff value of model was evaluated by area under the receiver operating characteristic curve （AUC ROC）. Results A total of 160 patients completed a 6-month follow-up with a loss rate of 9. 1% （16/176）. And 131 patients （82%） were cured after the single-period CO2 laser therapy with a total recovery rate of 94% （ 150/160 ）. Side effects occurred in 50 （31% ） patients with a complete self-recovery within 6 months. The most common side effects were local ulceration, pain and edema. No severe side effect was present. Large area of lesion （ 〉 8 cm2 ）, vagina involved and unemployment were associated with the failure of single-period treatment while pain symptom was a protective factor of effectiveness. Age, marital status, symptom-free and vaginal involvement were not related with outcome. A prediction model was established as follows： Logit （P0） = -1. 511 ＋ 1.573X1＋ 1.679X2 ＋ 3.254X3 - 1.685X4 （X1 - X4 representing area of lesion 〉 8 cm2, vaginal involvement, unemployment and pain symptom respectively）. The optimal cutoff value of P0 was 0. 35 with AUC ROC of 0. 816 （ P 〈 0. 01 ） . The sensitivity, specificity, positive predictive value and negative predictive value of model were 58.6%, 91.6%, 60.7% and 90. 9% respectively. Conclusion CO2 laser is effective and safe therapy for vulvar condylomata acuminata. A prediction model has been proposed to predict the outcome of single-period CO2 laser therapy in initially diagnosed patients. It may guide clinical decision-making.