Patient selection criteria
This study was reviewed and approved by the Human Ethics Committee of the University of Teikyo (trial registration number: 18-233). The medical records of 151 female patients with unilateral endometrioma, including 58 right-sided and 93 left-sided cases, from June 1, 2014, to December 31, 2019, were reviewed retrospectively. We defined the former cases as the right-sided endometrioma group (REG, n = 58) and the latter cases as the left-sided endometrioma group (LEG, n = 93). In this study, we did not include patients with recurrent endometriosis. As shown in Figure 1, 151 of the 302 patients who underwent laparoscopic surgery for new onset endometrioma detected during outpatient examinations were excluded for the following reasons: 95 patients were excluded because they were diagnosed with bilateral endometrioma and 56 patients were excluded for other reasons, including a lack of cystic lesions detected during surgery (n = 23); a diagnosis of other main ovarian cystic diseases, such as mature cystic teratoma (n = 8); ruptured endometrioma (n = 4); and prior abdominal surgery (n = 21). Since patients meeting the last criterion (prior abdominal surgery) were excluded after patients meeting other criteria were excluded, this number may be an underestimation (n = 21). Among the 151 remaining patients, we performed 123 laparoscopic cystectomies, including 9 that were performed simultaneously with laparoscopic myomectomies; 9 laparoscopic hysterectomies; 23 laparoscopic salpingo-oophorectomies, including 3 surgeries performed simultaneously with laparoscopic hysterectomies; and 5 laparoscopic-assisted cystectomies, including 3 surgeries performed simultaneously with laparoscopic-assisted myomectomies.
Collection of patient characteristics
After extracting the medical records of these 151 patients, the following patient characteristics were collected: 1) size of the ovarian endometrioma before and during surgery; 2) ASRM score (points) [5]; 3) patient’s age at the time of the operation; 4) presence of adhesions predicted before the operation; 5) history of use of assisted reproductive technology (ART); 6) presence of complicated adenomyomas, uterine fibroids or other ovarian tumours; and 7) serum carbohydrate antigen 125 (CA125) level (U/ml). Magnetic resonance imaging (MRI) was performed on almost all patients (148/151 cases) to assess the size of the endometrioma. MRI was also used to predict adhesions, particularly in the Douglas pouch. In this study, ART included both in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI). Twenty-eight of these 151 patients used hormone drugs, including oral contraceptives, gonadotropin-releasing hormone agonists and other drugs. However, the indications for the use of hormone drugs varied, and thus this factor was not considered in the present study.
Classification of endometrial lesions
We divided the endometrial lesions based on the location of the endometrioma to compare the tendency of endometrial lesion spread between patients with right or left ovarian endometrioma. Briefly, the so-called ASRM score chart, in which the severity of endometriosis is standardized by roughly classifying endometrial lesions [5], was divided into four parts. First, we calculated the total ASRM score for the uterine adnexa on the same side as the ovarian endometrioma and excluded the score for the endometrioma itself, as shown in Figure 2-A. Second, a similar calculation was performed for the uterine adnexa on the contralateral side of the ovarian endometrioma (Figure 2-B). The remaining endometrial lesions were divided into peritoneal (Figure 2-C) or Douglas pouch endometrial lesions (Figure 2-C). In summary, by referencing the sheet of “Revised ASRM classification of endometriosis, 1996.” [5], we classified the ASRM scores into the following four categories: 1) “same endometrial lesions (SELs)”, including Ovary-ENDOMETRIOSIS-Superficial, Ovary-ADHESIONS and Tube-ADHESIONS on the same side, as the main ovarian endometrioma; 2) “contralateral endometrial lesions (CELs)”, including Ovary-ENDOMETRIOSIS-Superficial, Ovary-ADHESIONS and Tube-ADHESIONS on the contralateral side of the main ovarian endometrioma; 3) “peritoneal endometrial lesions (PELs)”, including Peritoneum-ENDOMETRIOSIS; and 4) “Douglas pouch endometrial lesions (DELs)”, including POSTERIOR CULDESAC OBLITERATION. As an example, this classification is shown in Figure 2 for REG. The SEL (Figure 2-A), CEL (Figure 2-B), PEL (Figure 2-C) and DEL (Figure 2-C) scores are the total scores in each cell, as indicated by an italicized letter and thick-bordered box. In the patients with left ovarian endometrioma, the SEL and CEL scores were reversed. In this analysis, the theoretical maximum values of SEL, CEL, PEL and DEL were 36, 36, 10 and 40 points, respectively. We not only compared each average value of SEL, CEL, PEL and DEL between patients in the REG and LEG but also counted each number of patients with these four lesions in both the REG and LEG.
Statistical analysis
The primary outcomes included the difference in the ASRM scores detected during laparoscopic surgery between the patients with right- and left-sided endometriomas. We assessed the influences of the following 8 factors on the ASRM score: 1) “right-sided endometrioma”; 2) “predicted adhesion”, which was defined as the detection of the possible presence of an adhesion in a patient during a pelvic examination or MRI; 3) “ART history”, which was defined as patients who underwent ART; 4) “adenomyoma”, which was defined as the current presence of adenomyoma based on clinical images; 5) “uterine fibroid”, which was defined as the current presence of a uterine fibroid based on clinical images; 6) “other ovarian tumour”, which was defined as the current presence of an ovarian tumour other than endometrioma based on clinical images; 7) “positive marker”, which was defined as serum CA125 levels ≥ 35 U/ml; and 8) “large endometrioma”, which was defined as a tumour size of at least 50 mm (described in Table 2). All 151 patients were divided into two groups based on the presence (= 1) or absence (= 0) of these 8 factors. The difference in the ASRM score was compared using Welch’s t-test. We also performed a multivariate regression analysis using the least squares method to reduce confounding factors and confirm this difference. These statistical analyses were performed using JMP version 12 for Windows (SAS Institute, Inc., Tokyo, Japan). The total values of SEL, CEL, PEL and DEL between patients in the REG and LEG were also compared with Welch’s t-test. The data are presented as means ± standard deviations. A p-value less than 0.05 was considered statistically significant.