比较不同策略治疗剖宫产瘢痕妊娠(CSP)的临床疗效及并发症发生情况。

采用回顾性分析方法，选择2014年9月1日至2017年5月31日，于武汉大学人民医院妇产科、生殖医学中心就诊的103例CSP患者为研究对象。按照CSP治疗方案将其分为3组：A组[n＝26，采取甲氨蝶呤(MTX)＋手术治疗]，B组[n＝35，采取子宫动脉栓塞术(UAE)＋手术治疗]，以及C组(n＝42，采取单纯手术治疗)。根据MTX治疗方式的不同，将A组患者进一步分为A1亚组(n＝12，采取孕囊局部注射MTX＋手术治疗)，A2亚组(n＝14，采取肌内注射MTX＋手术治疗)。收集A、B及C组CSP患者的临床病例资料，对于治疗前血清β-人绒毛膜促性腺激素(hCG)水平、住院时间、出院时血清β-hCG水平下降比例等呈非正态分布的计量资料比较，采用Kruskal-Wallis H秩和检验，进一步两两比较，采用Mann-Whitney U检验，并采用Bonferroni法调整检验水准。A、B及C组患者术中大出血、术后宫腔黏连、妊娠组织残留发生率等计数资料比较，采用χ²检验，进一步两两比较，采用Fisher确切概率法，并调整检验水准。本研究遵循的程序符合2013年修订的《世界医学协会赫尔辛基宣言》要求。

①A、B及C组CSP患者的年龄、孕次、既往剖宫产术次数、孕龄、阴道流血及下腹疼痛发生率、孕囊直径、残存宫腔壁厚度、宫腔壁厚度≤3 mm所占比例、胎心搏动发生率及浅表型、部分型、完全型CSP所占比例分别比较，差异均无统计学意义(P>0.05)。②A、B及C组患者治疗前血清β-hCG水平分别为120 004 IU/L(16 720～181 727 IU/L)、38 219 IU/L(23 194～100 029 IU/L)、22 557 IU/L(9 113～49 573 IU/L)，3组总体比较，差异有统计学意义(χ²＝9.987，P＝0.007)。进一步进行两两比较的结果显示，A组患者治疗前血清β-hCG水平明显高于B、C组，差异均有统计学意义(U＝266.000、262.000，P＝0.013、0.009)。③A、B及C组患者住院时间、术中大出血发生率、出院时血清β-hCG水平下降比例分别总体比较，差异均有统计学意义(χ²＝21.010、9.786、37.590，P<0.001、＝0.044、<0.001)。进一步进行两两比较的结果显示，A组患者住院时间显著长于B、C组，C组患者术中大出血发生率显著高于B组，A、B组患者出院时血清β-hCG水平下降比例均显著高于C组，并且差异均有统计学意义(住院时间：U＝170.000、176.000，均为P<0.001；术中大出血：P＝0.014；出院时血清β-hCG水平下降比例：U＝171.000、412.000，P<0.001、＝0.001)。A1亚组中，无一例患者发生术中大出血。A1亚组患者住院时间显著长于B、C组，出院时血清β-hCG水平下降比例显著高于A2亚组及B、C组，并且差异均有统计学意义(U＝53.000、43.000、37.000、100.000、36.000，P<0.001、<0.001、＝0.016、＝0.018、<0.001)。④A、B及C组患者术后1个月妊娠组织残留发生率分别为0(0/26)、2.9%(1/35)、21.4%(9/42)，术后6个月宫腔黏连发生率分别为0(0/26)、20.0%(7/35)、0(0/35)，A、B及C组术后1个月妊娠组织残留、术后6个月宫腔黏连发生率分别总体比较，差异均有统计学意义(χ²＝11.250、14.590，P<0.001、＝0.004)。进一步进行两两比较的结果显示，C组患者术后1个月妊娠组织残留发生率明显高于A组，B组患者术后6个月宫腔黏连发生率明显高于A、C组，差异均有统计学意义(P＝0.010、0.016、0.002)。

局部或全身注射MTX联合手术治疗CSP，疗效可靠、并发症少。由于UAE联合手术治疗CSP可对患者生育功能造成严重影响，临床应谨慎选择。因为本研究仅为回顾性分析，采取局部或全身注射MTX、UAE等不同方案治疗CSP的疗效，仍然有待多中心、大样本、随机对照研究进一步证实。

To compare the treatment effects and complications of different treatment methods for cesarean scar pregnancy (CSP) and explore the optimal treatment method for CSP.

A total of 103 women who were diagnosed as CSP in Department of Obstetrics and Gynecology and Centre for Reproductive Medicine, Renmin Hospital of Wuhan University from September 1, 2014, to May 31, 2017 were selected as research subjects. According to different treatment methods of CSP, they were divided into group A (n=26) who received the combination of local or systematic methotrexate (MTX) injection and surgery, group B (n=35) who received the combination of uterine arterial embolization (UAE) and surgery, and group C (n=42) who only received surgery. And the patients of group A were further divided into subgroup A1 (n=12, with treatment of gestational sac local injection of MTX and surgery), and subgroup A2 (n=14, with treatment of whole body intramuscular injection MTX and surgery) according to different treatment methods of MTX. The clinical data of group A, B and C were collected by retrospective method. The measurement data with abnormal distribution, such as the initial serum β-human chorionic gonadotropin (hCG) level before treatment, length of hospital stay, the decrease ratio of serum β-hCG level at the time of discharge among group A, B and C were analyzed with Kruskal-Wallis H rank sum test, and Mann-Whitney U test with Bonferroni adjustment was applied for further comparision between each two different groups. The categorical data, such as the incidences of intraoperative hemorrhage, intrauterine adhesion and embryo residue were analyzed with chi-square test among the three groups (group A, B and C) or four groups (subgroup A1 and A2, and group B, C). Fisher′s exact test with adjustment of P value was applied for further comparision between each two groups. The study was in accordance with World Medical Association Declaration of Helsinki revised in 2013.

①There were no statistical differences in age, gravidity, times of previous cesarean section, gestational age, the incidences of vaginal bleeding and lower abdomen pain, mean diameter of gestation sac, thickness of remnant myometrial wall, proportion of myometrial wall thickness ≤3 mm, incidence of fetal heartbeat, and proportions of superficial, partial, and full CSP among group A, B and C (P>0.05). ②The initial serum β-hCG level before treatment of group A, B and C was 120 004 IU/L (16 720-181 727 IU/L), 38 219 IU/L (23 194-100 029 IU/L) and 22 557 IU/L (9 113-49 573 IU/L), respectively, and the difference was statistically significant (χ²=9.987, P=0.007). The results of further comparison showed that the initial serum β-hCG level before treatment of group A was significantly higher than that of group B (U=266.000, P=0.013) and group C (U=262.000, P=0.009). ③There were significant differences in the initial serum β-hCG level before treatment (χ²=21.010, P<0.001), length of hospital stay (χ²=9.786, P=0.001) and decrease ratio of serum β-hCG level at the time of discharge (χ²=37.590, P<0.001) among group A, B and C. For further comparision, the length of hospital stay was significantly longer in group A compared with group B and C (U=170.000, 176.000, both P<0.001), the incidence of intraoperative hemorrhage of group C was significantly higher compared with group B (P=0.014), and the decrease ratio of serum β-hCG level at the time of discharge was significantly higher in group A and B compared with group C (U=171.000, P<0.001; U=412.000, P=0.001). None intraoperative hemorrhage in occurred subgroup A1. The length of hospital stay was significantly longer in subgroup A1 compared with group B and C (U=53.000, 43.000, both P<0.001), and the decrease ratio of serum β-hCG level at the time of discharge was significantly higher in subgroup A1 compared with subgroup A2 and group B, C (U=37.000, P=0.016; U=100.000, P=0.018; U=36.000, P<0.001). ④The incidence of embryo residue in one month after surgery was 0 (0/26), 2.9% (1/35) and 21.4% (9/42) in group A, B and C, respectively. The incidence of intrauterine adhesion in six months after surgery was 0 (0/26), 20.0% (7/35) and 0 (0/35) in group A, B and C, respectively. There were significant differences in the incidence of embryo residue in one month after surgery (χ²=14.590, P<0.001) and the incidence of intrauterine adhesion in six months after surgery (χ²=11.250, P=0.004) among group A, B and C. For further comparision, the incidence of embryo residue in one month after surgery was higher in group C compared with group A (P=0.010), the incidence of intrauterine adhesion in six months after surgery was higher in group B compared with group A and C (P=0.016, 0.002).

Ultrasound-guided local or systematic injection of MTX combined with surgery for CSP is reliable and leads to few complications. The combination of UAE with surgery should be chosen carefully because of its potential fertility complication. Because this study is just a retrospective research, the effects of local or systemic injection of MTX, UAE and other different treatment methods for CSP, still need to be confirmed by multi-center, large-sample, randomized controlled studies.