临床小儿外科杂志  2024, Vol. 23 Issue (1): 46-50  DOI: 10.3760/cma.j.cn101785-202311024-009

引用本文  

孙祖嵩, 段栩飞, 闫学强. 小儿卵巢扭转坏死的危险因素及其预测价值分析[J]. 临床小儿外科杂志, 2024, 23(1): 46-50.   DOI: 10.3760/cma.j.cn101785-202311024-009
Sun ZS, Duan XF, Yan XQ. Risk factors and predictive values of ovarian torsion necrosis in children[J]. J Clin Ped Sur, 2024, 23(1): 46-50.   DOI: 10.3760/cma.j.cn101785-202311024-009

基金项目

儿童肝胆胰疾病研究室基金(2022FEYJS004)

通信作者

段栩飞, Email: alendxf6@hotmail.com

文章历史

收稿日期:2023-11-16
小儿卵巢扭转坏死的危险因素及其预测价值分析
孙祖嵩 , 段栩飞 , 闫学强     
华中科技大学同济医学院附属武汉儿童医院普外科, 武汉 430015
摘要目的 探讨小儿卵巢扭转坏死的危险因素及其预测价值。方法 回顾性分析华中科技大学同济医学院附属武汉儿童医院2014年5月至2023年5月术中诊断为卵巢扭转的73例患儿临床资料。按照病理结果将患儿分为坏死组(27例)及未坏死组(46例)。比较两组患儿的年龄、侧别、恶心呕吐、发热、腹痛至手术时间、血小板、淋巴细胞、中性粒细胞、中性粒细胞百分比、C-反应蛋白值(C-reaction protein, CRP)、白细胞计数(white blood cell, WBC)、中性粒细胞与淋巴细胞的比值(neutrophil-to-lymphocyte ratio, NLR)、血小板与淋巴细胞的比值(platelet-to-lymphocyte ratio, PLR)、淋巴细胞与CRP的比值(lymphocyte to C-reaction protein, LCR)及附件肿物直径。使用单因素Logistic回归分析小儿卵巢扭转坏死的相关危险因素, 多因素Logistic回归进一步分析其独立危险因素, 绘制受试者工作特征(receiver operating characteristic, ROC)曲线评估各指标对小儿卵巢扭转坏死的预测价值。结果 两组年龄[(7.21±3.65)岁比(8.80±4.17)岁]、侧别(左/右)(9/17比21/26)、血小板[(289.18±94.57)×109/L比(300.06±79.05)×109/L]、淋巴细胞[1.93(1.36, 2.74)×109/L比1.60(1.00, 2.46)×109/L]、中性粒细胞百分比[(72.25±14.23)%比(69.30±18.69)%]、NLR[6.18(3.68, 8.79)比5.02(2.38, 8.49)]、PLR[117.31(101.27, 199.98)比181.99(104.22, 282.81)]、LCR[0.39(0.05, 1.45)比0.10(0.02, 1.73)]、附件肿物直径[5.00(3.97, 6.33)cm比4.67(3.38, 6.17)cm]比较, 差异均无统计学意义(P>0.05)。两组卵巢扭转度数[720.00(720.00, 855.00)°比720.00(360.00, 720.00)°]、恶心及呕吐例数[21例(80.77%)比27例(57.45%)]、发热例数[9例(34.62%)比2例(4.26%)]、腹痛至手术时间[72.00(31.50, 96.00)h比24.00(15.84, 45.60)h]、中性粒细胞[10.29(6.30, 11.61)×109/L比5.98(3.67, 9.04)×109/L]、CRP[9.40(0.78, 36.70)mg/L比0.84(0.78, 2.27)mg/L]、WBC[13.25(7.92, 16.89)×109/L比8.28(6.21, 11.87) ×109/L]比较, 差异均有统计学意义(P<0.05)。ROC曲线结果显示, 腹痛至手术时间对卵巢扭转坏死的预测价值最高, 最佳截断值为26.7 h, ROC曲线下面积(area under the ROC curve, AUC)为0.755, 灵敏度、特异度分别为84.6%、66.0%;其次为卵巢扭转度数, 其最佳截断值为405°, AUC为0.695, 灵敏度、特异度分别为88.5%、42.6%;最后为发热, AUC为0.652, 灵敏度、特异度分别为34.6%、95.7%。将三者联合对卵巢扭转坏死进行预测时, AUC为0.870, 灵敏度、特异度分别为69.2%、93.6%。结论 腹痛至手术时间、卵巢扭转度数及发热可有效预测小儿卵巢扭转坏死。
关键词卵巢扭转    卵巢坏死    外科手术    儿童    
Risk factors and predictive values of ovarian torsion necrosis in children
Sun Zusong , Duan Xufei , Yan Xueqiang     
Department of General Surgery, Wuhan Children's Hospital, Tongji Medical College, Huazhong University of Science &Technology, Wuhan 430015, China
Abstract: Objective To explore the risk factors and predictive values of ovarian torsion (OT) necrosis in children. Methods To summarize the clinical data of 73 children with ovarian torsion diagnosed during operation in Wuhan Children's Hospital, Tongji Medical College, Huazhong University of Science & Technologyfrom May 2014 to May 2014, and to 2023. According to the results of pathological examination, they were assigned into two groups of torsion necrosis (n=27) and non-necrosis (n=46). The age, affected side, malignant behavior and vomiting, fever, time from abdominal pain to operation, platelet, lymphocyte, neutrophil, percentage of Neutrophil, C-reaction protein (CRP), white blood cell (WBC), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), lymphocyte-to-C-reaction protein (LCR) and the diameter of adnexal masses were compared between the two groups. Univariate Logistic regression was used to analyze the risk factors of ovarian torsion necrosis in children. Multivariate Logistic regression was used to analyze the independent risk factors, receiver operating characteristic (ROC) curve was drawn to evaluate the predictive value of each index for torsion necrosis of ovary in children. Results No significant inter-group differences existed in age[(7.21±3.65)year vs. (8.80±4.17)year], The affected side (left/right){[9(34.62%)/17(65.38%)]vs. [21(44.68%)/26(55.32%)]}, platelet[(289.18±94.57)×109/L vs. (300.06±79.05)×109/L], lymphocyte[1.93 (1.36, 2.74)×109/L vs. 1.60 (1.00, 2.46)×109/L], neutrophil[(72.25±14.23)% vs. (69.30±18.69)%], neutrophil-to-lymphocyte ratio (NLR)[6.18 (3.68, 8.79) vs. 5.02 (2.38, 8.49)], platelet-to-lymphocyte ratio (PLR)[117.31 (101.27, 199.98) vs. 181.99 (104.22, 282.81)]or lymphocyte to C-reaction protein (LCR)[0.39 (0.05, 1.45) vs. 0.10 (0.02, 1.73)], the diameter of adnexal masses[5.00 (3.97, 6.33)cm vs. 4.67 (3.38, 6.17)cm](P>0.05). Significant inter-group differences existed in degree of OT[720.00 (720.00, 855.00)° vs. 720.00 (360.00, 720.00)°], nausea and vomiting[21(80.77%) vs. 27(57.45%)], fever[9(34.62%) vs. 2(4.26%)], time from abdominal pain to operation[72.00 (31.50, 96.00)h vs. 24.00 (15.84, 45.60)h], neutrophil[10.29 (6.30, 11.61)×109/L vs. 5.98 (3.67, 9.04)×109/L], C-reaction protein (CRP)[9.40 (0.78, 36.70)mg/L vs. 0.84 (0.78, 2.27)mg/L]and white blood cell (WBC)[13.25 (7.92, 16.89)×109/L vs. 8.28 (6.21, 11.87) ×109/L](P < 0.05). The results of ROC showed that the time from abdominal pain to operation had the highest predictive value for ovarian torsion necrosis, the best cut-off was 26.7 h, the area under the ROC curve (AUC) was 0.755, the sensitivity and specificity were 84.6% and 66.0%. The second is the degree of ovarian torsion, the best cut-off value was 405° with an AUC of 0.695, the sensitivity and specificity were 88.5% and 42.6%. Finally for fever, the AUC was 0.652, the sensitivity and specificity were 34.6% and 95.7%. Combining those three parameters, the AUC was 0.870 with a sensitivity of 69.2% and a specificity of 93.6%. Conclusions Time from abdominal pain to operation, degree of OT and fever may effectively predict the occurrence of OT necrosis in children.
Key words: Ovarian Torsion    Ovarian Necrosis    Surgical Procedures, Operative    Child    

卵巢扭转是指卵巢绕血管蒂旋转,导致部分或完全的血液循环受阻,儿童患病率约4.9/100 000,如未及时干预,可引起卵巢缺血坏死[1]。外科医师选择保留或切除卵巢最重要的依据是术中对卵巢的主观评估,目前尚无明确的客观指标来预测卵巢组织是否发生坏死。本研究收集因卵巢扭转行手术治疗的病例进行回顾性分析,为小儿卵巢扭转坏死的术前及术中预测提供参考依据。

资料与方法 一、研究对象

本研究为回顾性研究,选取2014年5月至2023年5月在华中科技大学同济医学院附属武汉儿童医院接受手术治疗,且术中诊断为卵巢扭转的73例患儿作为研究对象,中位年龄8.2岁。纳入标准:①年龄≤18岁;②病例资料完整、规范。排除标准:①双侧卵巢扭转;②出生前已明确存在腹腔肿物或卵巢扭转。本研究通过华中科技大学同济医学院附属武汉儿童医院伦理委员会审查(2022R098-E01),患儿家属均知情同意。

二、观察指标及分组方式

将所有纳入病例根据病理检查结果分为坏死组及未坏死组。将病理提示伴有卵巢组织部分坏死的病例纳入坏死组;将病理提示未见卵巢坏死的病例及未行病理检查的病例纳入未坏死组。本组73例患儿中,坏死组26例,未坏死组47例。比较两组患儿年龄、临床症状(发热、腹痛持续时间、恶心及呕吐症状)、血液检查结果[C-反应蛋白值(C-reaction protein, CRP)、白细胞计数(white blood cell, WBC)、中性粒细胞百分比值、中性粒细胞与淋巴细胞的比值(neutrophil-to-lymphocyte ratio, NLR)、血小板与淋巴细胞的比值(platelet-to-lymphocyte ratio, PLR)、淋巴细胞与CRP的比值(lymphocyte to C-reaction protein, LCR)]、超声检查结果(肿块直径)、术中所见(卵巢扭转的患侧、卵巢扭转度数及病因)以及肿块病理类型。

三、统计学处理

运用SPSS 27.0进行数据分析。采用Shapiro-Wilk法对计量资料进行正态性检验。分类资料以例(%)表示,无序分类资料应用χ2检验比较组间差异;服从正态分布的计量资料用x±s表示,两组间比较采用独立样本t检验;不服从正态分布的计量资料用M(Q1Q3)表示,两组间比较采用独立样本Wilcoxon秩和检验。应用Logistic回归模型进行单因素及多因素分析,将单因素分析中P<0.1的自变量纳入多因素分析,将多因素分析结果中有统计学意义的指标进行受试者操作特性(receiver operating characteristic, ROC)曲线分析,得出其ROC曲线下面积(area under the ROC curve, AUC)、最佳截断值、灵敏度、特异度。P<0.05为差异有统计学意义。

结果

73例卵巢扭转均为单侧,其中右侧43例、左侧30例。原发性卵巢扭转11例,继发于腹股沟斜疝的卵巢扭转4例,继发于输卵管积水肿胀的卵巢扭转1例,继发于输卵管索带形成的卵巢扭转1例,继发于附件肿块的卵巢扭转56例(其中卵巢成熟性畸胎瘤27例,卵巢单胚层畸胎瘤1例,黄体囊肿4例,卵巢滤泡囊肿3例,浆液性囊腺瘤1例,卵巢单纯性囊肿13例,卵巢苗勒氏源性囊肿2例,输卵管副中肾管源性囊肿2例,输卵管系膜囊肿3例)。坏死组附件肿块直径为5.00 (3.97, 6.33)cm,未坏死组为4.67 (3.38, 6.17)cm,差异无统计学意义(P>0.05)。

在单因素分析中,两组卵巢扭转度数、恶心及呕吐、发热、腹痛至手术时间、中性粒细胞、CRP、WBC差异有统计学意义(P<0.05),将上述因素纳入二元Logistic回归模型后发现, 腹痛至手术时间是小儿卵巢扭转坏死的危险因素(P < 0.05)。见表 1表 2

表 1 卵巢扭转坏死危险因素的单因素分析结果 Table 1 Results of univariate analysis of risk factors for ovarian torsion necrosis

表 2 卵巢扭转坏死危险因素的二元Logistic回归分析结果 Table 2 Binary Logistic regression analysis of risk factors for ovarian torsion necrosis

单项指标中,诊断卵巢扭转伴坏死效能最高的是腹痛至手术时间(其AUC为0.755);其次为卵巢扭转度数(其AUC为0.695)。联合腹痛至手术时间、发热、卵巢扭转度数绘制ROC曲线,AUC=0.870。见表 3图 1

表 3 单个指标及联合指标对小儿卵巢扭转坏死的预测效能 Table 3 Predictive potency of single and combined parameters for ovarian torsion necrosis in children


图 1 卵巢扭转坏死危险因素的ROC曲线分析 Fig.1 ROC curve of risk factors for ovarian torsion necrosis    预测因素包括腹痛至手术时间、发热、卵巢扭转度数,三者联合诊断时曲线下面积=0.870
讨论

卵巢扭转可能引起卵巢出血、坏死,继而影响患儿生育功能的保留[2]。其常见表现为腹痛、恶心及呕吐、发热及白细胞增多[3]。不全性卵巢扭转(扭转度数<360°)或间歇性卵巢扭转可表现为持续性或间歇性疼痛,有自行好转的可能,临床易被误诊、漏诊[4]。本研究中有43例(58.9%)患儿右侧卵巢发生扭转,这可能与乙状结肠位于左下腹,限制了左侧卵巢活动有关。

卵巢扭转可分为原发性卵巢扭转及继发性卵巢扭转。本组共11例诊断为原发性卵巢扭转,考虑与儿童子宫相对较小,输卵管、输卵管系膜、卵巢固有韧带相对较长,卵巢活动度较大有关[4];62例诊断为继发性卵巢扭转,其中附件肿物合并卵巢扭转56例。值得注意的是,本组无一例病理检查提示卵巢恶性肿瘤,这可能与恶性肿瘤呈浸润性生长、外形不规则、与周围组织粘连较重、较难引起卵巢扭转有关[5-6]

妇科超声检查是诊断卵巢扭转的首选检查方式,其影像学特点包括单侧卵巢增大、卵泡外周移位、附件区混合回声肿块、卵巢移位、子宫远离中线、盆腔游离液体、whirlpool征或“漩涡征”等[4, 6]。超声对发现卵巢肿物有较高价值,有研究显示卵巢直径增大可诱发卵巢扭转[2],也有研究显示坏死的卵巢组织可萎缩变小[7],本研究发现卵巢大小与卵巢扭转坏死的发生无统计学层面的关联性。

本研究发现恶心及呕吐为预测卵巢扭转坏死的危险因素,这可能是由于卵巢长时间扭转引起剧烈疼痛或腹膜刺激引起迷走神经反射,从而引起恶心、呕吐等消化道症状。恶心及呕吐表现是预测卵巢扭转的危险因素[4, 6-8]。女性患儿一旦出现原因不明的急腹症,伴有恶心、呕吐等消化道症状,应警惕卵巢扭转的可能。另外中性粒细胞、CRP、WBC水平升高也是预测卵巢扭转坏死的危险因素。Hajime等[9]认为WBC水平升高可能提示卵巢扭转发生不可逆缺血改变。Hong等[10]认为CRP升高预示卵巢坏死。若在中性粒细胞、CRP、WBC水平升高前对疑似卵巢扭转的患儿进行手术治疗,将有利于保留卵巢功能。Nissen等[11]认为术前LCR水平降低可以作为1岁以上小儿卵巢坏死的评估指标,但本研究中坏死组与未坏死之间LCR差异无统计学意义,这可能与本研究样本量较少有关。

本研究结果显示,腹痛至手术的时间预测卵巢扭转坏死的AUC最大,达0.755,最佳截断值为26.7 h,灵敏度、特异度分别为84.6%、66.0%。本研究中,坏死组中有22例(81.5%)患儿腹痛至手术的时间大于24 h,且均不少于10 h;而未坏死组有17例(36.2%)在腹痛超过24 h后进行手术。因此本研究认为术前一旦有可靠临床证据怀疑卵巢发生扭转,应尽早行手术治疗[3]。其次为卵巢扭转度数,AUC为0.695,其最佳截断值为405°,灵敏度、特异度分别为88.5%、42.6%。由于卵巢拥有双重血供的特性,当卵巢发生扭转时首先累及静脉血流,其次才是动脉血流,卵巢扭转的严重程度取决于血管蒂的旋转程度,当仅发生卵巢短暂扭转或卵巢不完全性扭转时,卵巢仍可能存在动脉血流灌注,故本研究认为完全性卵巢扭转并不能有效预测卵巢扭转坏死,但随着卵巢扭转度数的增加,未解除扭转的动脉闭塞更容易引起出血性梗死[12-13]。再次为发热,AUC为0.652,灵敏度、特异度分别为34.6%,95.7%。本研究中发热11例,其中坏死组9例,未坏死组2例,有研究显示儿童卵巢扭转后发热的发生率为4% ~22%,当发生卵巢坏死时,可能出现发热,与本研究相符[3]。将腹痛至手术时间、发热、卵巢扭转度数联合绘制ROC曲线时,AUC=0.870,其灵敏度及特异度分别为69.2%、93.6%。

本研究缺乏对卵巢扭转的超声特征的完整记录,因为不同超声医师的书写报告存在差异,缺乏对部分超声特征的描述,加强超声医师对卵巢扭转及卵巢扭转坏死的认识,可能有助于报告形式的统一。综上所述,腹痛至手术时间、卵巢扭转度数及发热表现可有效预测小儿卵巢扭转坏死。当术中证实为卵巢扭转时,结合患儿术前出现发热、腹痛至手术时间超过26.7 h、扭转度数超过405°应提高对卵巢扭转坏死的警惕。

利益冲突  所有作者声明不存在利益冲突

作者贡献声明  研究设计与酝酿为段栩飞、闫学强,文献检索为孙祖嵩,论文调查设计为孙祖嵩,数据收集与分析为孙祖嵩,论文结果撰写为孙祖嵩,论文讨论分析为孙祖嵩,对文章知识性内容进行审阅为段栩飞、闫学强

参考文献
[1]
Epstein KN, Trout AT, Debnath P, et al. Rapid, free-breathing non-contrast MRI for first-line imaging evaluation of ovarian torsion in the emergency department[J/OL]. Pediatr Radiol, 2023. https://doi.org/10.1007/s00247-023-05827-y. DOI: 10.1007/s00247-023-05827-y.
[2]
Prete S, Abbomerato M, Simon EL. Ovarian torsion masquerading as a ureteral stone[J/OL]. Am J Emerg Med, 2023. https://doi.org/10.1016/j.ajem.2023.11.046. DOI: 10.1016/j.ajem.2023.11.046.
[3]
Novoa M, Friedman J, Mayrink M. Ovarian torsion: can we save the ovary?[J]. Arch Gynecol Obstet, 2021, 304(1): 191-195. DOI:10.1007/s00404-021-06008-8
[4]
Chang-Patel EJ, Palacios-Helgeson LK, Gould CH. Adnexal torsion: a review of diagnosis and management strategies[J]. Curr Opin Obstet Gynecol, 2022, 34(4): 196-203. DOI:10.1097/GCO.0000000000000787
[5]
Hao J, Sun ZH, Song GX. Laparoscopy for ovary-sparing tumorectomy in children with ovarian tumors: a clinical retrospective analysis[J]. J Laparoendosc Adv Surg Tech A, 2023, 33(3): 296-302. DOI:10.1089/lap.2022.0372
[6]
熊婧, 路丽, 吕志宝. 儿童卵巢扭转相关危险因素分析[J]. 中华小儿外科杂志, 2020, 41(5): 390-394.
Xiong J, Lu L, Lyu ZB. Analysis of risk factors corelated with ovarian torsion in children[J]. Chin J Pediatr Surg, 2020, 41(5): 390-394. DOI:10.3760/cma.j.cn421158-20200106-00008
[7]
Tielli A, Scala A, Alison M, et al. Ovarian torsion: diagnosis, surgery, and fertility preservation in the pediatric population[J]. Eur J Pediatr, 2022, 181(4): 1405-1411. DOI:10.1007/s00431-021-04352-0
[8]
Aiob A, Lowenstein L, Borik I, et al. The value of clinical symptoms, the neutrophil-to-lymphocyte ratio, and ultrasonographic features in predicting adnexal torsion: a case-control study[J]. J Obstet Gynaecol Res, 2023, 49(1): 289-295. DOI:10.1111/jog.15474
[9]
Takayasu H, Masumoto K, Tanaka N, et al. A clinical review of ovarian tumors in children and adolescents[J]. Pediatr Surg Int, 2020, 36(6): 701-709. DOI:10.1007/s00383-020-04660-w
[10]
Zhang H, Bai JX, Zhang B, et al. Characteristics of ovarian necrosis in the neonatal ovarian tumor: a single-center retrospective study and review of literature[J]. Pediatr Surg Int, 2022, 39(1): 42. DOI:10.1007/s00383-022-05277-x
[11]
Nissen M, Sander V, Rogge P, et al. Neutrophil to lymphocyte ratio and platelet to lymphocyte ratio might predict pediatric ovarian torsion: a single-institution experience and review of the literature[J]. J Pediatr Adolesc Gynecol, 2021, 34(3): 334-340. DOI:10.1016/j.jpag.2020.12.003
[12]
Duan N, Chen X, Rao M, et al. CT predictive model for torsion angle as a marker for risk of necrosis in patients with adnexal torsion[J]. Clin Radiol, 2021, 76(7): 540-546. DOI:10.1016/j.crad.2021.02.027
[13]
Renganathan R, Subramaniam P, Deebika S, et al. Scoring system for predicting ovarian necrosis in adnexal torsion using an ultra-short optimized MRI protocol[J]. Abdom Radiol (NY), 2023, 48(6): 2122-2130. DOI:10.1007/s00261-023-03886-1