Analyzing the Risk Factors for Recurrence of Endometrioma and Dysmenorrhea in Patients with Endometriosis

Endometriosis is an estrogen-dependent chronic disease characterized by the presence of endometrial glands and stroma outside the uterine cavity. 1 It affects approximately 6–10% of reproductive-aged women worldwide, 2 with ovarian endometrioma accounting for nearly half of all cases. 3 , 4 Common manifestations include dysmenorrhea, chronic pelvic pain, 5 and infertility, all of which substantially impair quality of life. 6 Surgical removal of ovarian endometriomas remains the mainstay of treatment, particularly for women with pain or infertility. 7 , 8 However, recurrence after surgery is frequent. Reported recurrence rates range from 19–30% for endometrioma and up to 50% for dysmenorrhea. 9 , 10 Repeated surgery further increases the risk of diminished ovarian reserve. In addition, postoperative recurrence may lead to persistent symptoms, repeated medical visits, and further interventions, thereby imposing a substantial long-term burden on patients’ quality of life and reproductive potential. Although adjuvant medical therapy such as oral contraceptives, progestins, and gonadotropin-releasing hormone agonists may reduce recurrence, their long-term efficacy is limited by side effects and poor adherence. 11 , 12 Several studies have examined risk factors for recurrence, including age at surgery, revised American Fertility Society (rAFS) score, 13 and history of prior surgery. 14 However, findings have been inconsistent, partly due to small sample sizes and limited follow-up. 15 , 16 Moreover, recurrence of ovarian endometrioma and recurrence of dysmenorrhea do not always occur simultaneously and may reflect different aspects of disease persistence or progression. Therefore, evaluating these two outcomes separately may provide a more comprehensive understanding of postoperative prognosis. Moreover, evidence from Chinese populations remains scarce, despite potential differences in clinical patterns and management practices. 17 To address these gaps, we conducted a large-scale, long-term follow-up study of 1598 women who underwent surgery for pathologically confirmed endometriosis at a tertiary center in China. The objective was to evaluate recurrence rates of endometrioma and dysmenorrhea and to identify independent risk factors associated with recurrence. This study represents one of the largest cohort analyses of endometriosis recurrence in China and provides valuable evidence for guiding postoperative management.

The baseline characteristics of the 1598 patients included in the analysis are summarized in Table 1 . The median age at surgery was 32 years (IQR, 28–37), and the median BMI was 22.6 kg/m 2 (IQR, 21.5–24.0). A total of 846 patients (52.9%) reported preoperative dysmenorrhea. Table 1 Baseline Characteristics of the 1,598 Patients with Endometriosis Variable All Patients (N = 1,598) Age at surgery, years 32 (28–37) BMI, kg/m 2 22.5 (21.2–24.0) Age at menarche, years 14 (13–15) Parity, n (%) 0 930 (58.2) 1 520 (32.5) ≥2 148 (9.3) Preoperative dysmenorrhea, n (%) 846 (52.9) History of endometriosis, n (%) 71 (4.4) History of infertility, n (%) 365 (22.8) Preoperative VAS (0–10) 1 (0–3) Largest ovarian cyst diameter, cm 4.2 (3.0–5.7) Laterality of ovarian endometrioma, n (%) Left 533 (33.4) Right 490 (30.7) Bilateral 573 (35.9) Menstrual cycle phase at surgery, n (%) Proliferative 719 (45.0) Secretory 879 (55.0) rAFS score 45 (28–60) EFI score (0–10) 6 (5–7) LF score 6 (5–7) Notes : Data are presented as median (interquartile range, IQR) for continuous variables and as number (percentage) for categorical variables. Baseline Characteristics of the 1,598 Patients with Endometriosis Notes : Data are presented as median (interquartile range, IQR) for continuous variables and as number (percentage) for categorical variables. During a median follow-up of 58.0 months (IQR, 44.0–72.0 months), 348 patients (21.8%) developed recurrence of endometrioma. As shown in Table 2 , patients with recurrence had significantly higher preoperative VAS scores, higher rAFS scores, greater intraoperative bleeding, longer operative time, and longer hospital stay compared with those without recurrence (all P < 0.05). Conversely, lower LF and EFI scores were observed in the recurrent group. Table 2 Comparisons of Patients with and without Recurrence of Endometrioma Variable Nonrecurrent (n = 1,250) Recurrent (n = 348) P value Age at surgery, years 32 (28–37) 31 (27–36) 0.028 BMI, kg/m 2 22.6 (21.2–24.0) 22.3 (20.9–23.7) 0.051 Age at menarche, years 14 (13–15) 14 (13–15) 0.099 Parity, n (%) 0.058  0 725 (58.0) 205 (58.9)  ≥1 525 (42.0) 143 (41.1) Preoperative infertility, n (%) 271 (21.7) 94 (27.0) 0.023 History of endometriosis, n (%) 11 (0.9) 60 (17.2) <0.001 Preoperative dysmenorrhea, n (%) 628 (50.2) 218 (62.6) <0.001 Preoperative VAS score (0–10) 1 (0–3) 2 (0–4) <0.001 Largest ovarian cyst diameter, cm 4.1 (3.0–5.7) 4.3 (3.0–6.0) 0.015 Laterality of ovarian endometrioma, n (%) <0.001  Left 455 (36.4) 78 (22.4)  Right 401 (32.1) 89 (25.6)  Bilateral 394 (31.5) 179 (51.4) Cyst size, n (%) 0.566  <5 cm 191 (15.3) 50 (14.4)  5–10 cm 1015 (81.3) 282 (81.0)  ≥10 cm 43 (3.4) 16 (4.6) rAFS score 36 (20–52) 69 (28–100) <0.001 LF score 6 (5–7) 5 (4–6) <0.001 EFI score 6 (5–7) 5 (4–7) <0.001 Intraoperative bleeding, mL 50 (30–100) 50 (50–100) 0.021 Operative time, min 65 (50–90) 75 (55–100) <0.001 Intraoperative adhesion, n (%) 853 (68.2) 260 (74.7) 0.011 Operation mode, n (%) 0.031  Laparoscopy 1155 (92.4) 331 (95.1)  Laparotomy 95 (7.6) 16 (4.6) Days of hospitalization 6 (5–7) 7 (5–7) 0.001 Preoperative medical therapy, n (%) 36 (2.9) 146 (42.0) <0.001 Postoperative medical therapy, n (%) 705 (56.4) 155 (44.5) <0.001 Drug use time after operation, months 3 (0–3) 0 (0–3) <0.001 Notes : Data are presented as median (IQR) for continuous variables and as number (percentage) for categorical variables. P values were calculated using the Mann–Whitney U -test for continuous variables and the χ 2 -test or Fisher’s exact test for categorical variables. Comparisons of Patients with and without Recurrence of Endometrioma Notes : Data are presented as median (IQR) for continuous variables and as number (percentage) for categorical variables. P values were calculated using the Mann–Whitney U -test for continuous variables and the χ 2 -test or Fisher’s exact test for categorical variables. Bilateral ovarian involvement was more common among recurrent cases, and postoperative medical therapy was less frequently used (P = 0.011). Kaplan–Meier analysis showed that the cumulative recurrence rates of endometrioma at 1, 3, 5, and 7 years after surgery were 12%, 38%, 55%, and 65%, respectively ( Figure 2A ). The cumulative hazard function demonstrated a steeper increase after 72 months, indicating an acceleration in recurrence risk ( Figure 2B ). Figure 2 Kaplan–Meier curves of endometrioma recurrence. ( A ) The cumulative recurrence rates of endometrioma were 12%, 38%, 55%, and 65% at 1, 3, 5, and 7 years after surgery, respectively. ( B ) The estimated cumulative hazard function for endometrioma recurrence increased gradually over time, with a more pronounced rise after approximately 72 months postoperatively. Kaplan–Meier curves of endometrioma recurrence. ( A ) The cumulative recurrence rates of endometrioma were 12%, 38%, 55%, and 65% at 1, 3, 5, and 7 years after surgery, respectively. ( B ) The estimated cumulative hazard function for endometrioma recurrence increased gradually over time, with a more pronounced rise after approximately 72 months postoperatively. Among the 846 patients with preoperative dysmenorrhea, 187 (22.1%) experienced recurrence after surgery ( Figure 1 ). As shown in Table 3 , patients with recurrence had significantly higher preoperative VAS scores and rAFS scores, longer hospital stay, and were more likely to have a history of endometriosis (all P < 0.05). Postoperative medical therapy was less common among recurrent cases (P = 0.011). Table 3 Baseline Characteristics of Patients with and without Recurrence of Dysmenorrhea Variable Nonrecurrent (n = 659) Recurrent (n = 187) P value Age at surgery, years 32 (28–37) 31 (27–36) 0.210 BMI, kg/m 2 22.7 (21.5–24.0) 22.5 (21.2–23.8) 0.267 Age at menarche, years 14 (13–15) 14 (13–15) 0.574 Parity, n (%) 0.182  0 402 (61.0) 121 (64.7)  ≥1 257 (39.0) 66 (35.3) Preoperative infertility, n (%) 157 (23.8) 45 (24.1) 0.508 History of endometriosis, n (%) 17 (2.6) 19 (10.2) <0.001 Preoperative dysmenorrhea, n (%) 27 (4.1) 10 (5.3) 0.289 Preoperative VAS score 3.4 (3.0–3.6) 4.1 (3.7–4.4) <0.001 Largest ovarian cyst diameter, cm 3.8 (3.5–4.5) 4.0 (3.7–4.5) 0.097 Laterality of ovarian endometrioma, n (%) 0.066  Left 202 (30.7) 53 (28.3)  Right 201 (30.5) 43 (23.0)  Bilateral 255 (38.7) 90 (48.1) Cyst size, n (%) 0.705  <5 cm 81 (12.3) 26 (13.9)  5–10 cm 549 (83.3) 151 (80.7)  ≥10 cm 29 (4.4) 10 (5.3) rAFS score 45 (28–52) 55 (50–60) <0.001 LF score 5.5 (5.3–5.6) 5.6 (5.3–5.8) 0.653 EFI score 5.8 (5.6–5.9) 5.5 (5.3–5.8) 0.129 Intraoperative bleeding, mL 50 (30–100) 55 (40–110) 0.606 Operative time, min 65 (50–90) 70 (55–95) 0.289 Intraoperative adhesion, n (%) 549 (83.3) 151 (80.7) 0.705 Operation mode, n (%) 0.522  Laparoscopy 179 (27.2) 51 (27.3)  Laparotomy 480 (72.8) 136 (72.7) Days of hospitalization 6 (5–7) 7 (6–7) 0.045 Preoperative medical therapy, n (%) 37 (5.6) 31 (16.6) <0.001 Postoperative medical therapy, n (%) 399 (60.5) 95 (50.8) 0.011 Duration of postoperative therapy, months 3 (0–3) 2 (0–3) 0.468 Notes : Data are presented as median (IQR) for continuous variables and as number (percentage) for categorical variables. P values were calculated using the Mann–Whitney U -test for continuous variables and the χ 2 -test or Fisher’s exact test for categorical variables. Baseline Characteristics of Patients with and without Recurrence of Dysmenorrhea Notes : Data are presented as median (IQR) for continuous variables and as number (percentage) for categorical variables. P values were calculated using the Mann–Whitney U -test for continuous variables and the χ 2 -test or Fisher’s exact test for categorical variables. The recurrence rates of dysmenorrhea at 1, 3, 5, and 7 years after surgery were 28%, 52%, 68%, and 75%, respectively ( Figure 3A ). The cumulative hazard of dysmenorrhea recurrence increased approximately linearly over time ( Figure 3B ). Figure 3 Kaplan–Meier curves of dysmenorrhea recurrence. ( A ) The cumulative recurrence rates of dysmenorrhea were 28%, 52%, 68%, and 75% at 1, 3, 5, and 7 years after surgery, respectively. ( B ) The cumulative hazard function for dysmenorrhea recurrence increased almost linearly over time, suggesting a relatively constant hazard rate throughout the follow-up. Kaplan–Meier curves of dysmenorrhea recurrence. ( A ) The cumulative recurrence rates of dysmenorrhea were 28%, 52%, 68%, and 75% at 1, 3, 5, and 7 years after surgery, respectively. ( B ) The cumulative hazard function for dysmenorrhea recurrence increased almost linearly over time, suggesting a relatively constant hazard rate throughout the follow-up. Variables with P < 0.10 in univariable comparisons were entered into Cox proportional hazards models ( Tables 4 and 5 ). Table 4 Multivariable Cox Regression for Risk of Endometrioma Recurrence (Primary Model) Variable Hazard Ratio (95% CI) P value Bilateral cyst (yes vs unilateral) 1.357 (1.060–1.737) 0.015 Age at surgery (per year) 1.040 (1.004–1.077) 0.029 Postoperative medical therapy (No vs Any) 2.103 (1.279–3.457) 0.003 Postoperative VAS (per 1-point increase) 1.100 (1.053–1.151) < 0.001 Notes : Hazard ratios (HRs) with 95% confidence intervals (CIs) are shown. Variables included in the model were selected based on clinical relevance and univariable analysis (P < 0.10). Table 5 Multivariable Cox Regression for Risk of Dysmenorrhea Recurrence (Primary Model) Variable Hazard Ratio (95% CI) P value Endometriosis history (Yes vs No) 2.06 (1.19–3.57) 0.010 Preoperative VAS (per 1-point increase) 1.14 (1.02–1.28) 0.026 Postoperative VAS (per 1-point increase) 1.27 (1.18–1.39) < 0.001 Notes : Hazard ratios (HRs) with 95% CIs are shown. Variables included in the model were selected based on clinical relevance and univariable analysis (P < 0.10). Multivariable Cox Regression for Risk of Endometrioma Recurrence (Primary Model) Notes : Hazard ratios (HRs) with 95% confidence intervals (CIs) are shown. Variables included in the model were selected based on clinical relevance and univariable analysis (P < 0.10). Multivariable Cox Regression for Risk of Dysmenorrhea Recurrence (Primary Model) Notes : Hazard ratios (HRs) with 95% CIs are shown. Variables included in the model were selected based on clinical relevance and univariable analysis (P < 0.10). For endometrioma recurrence ( Table 4 ), bilateral cysts (HR 1.36, 95% CI 1.06–1.74, P = 0.015), older age at surgery (HR 1.04, 95% CI 1.00–1.08, P = 0.029), lack of postoperative medical therapy (HR 2.10, 95% CI 1.28–3.46, P = 0.003), and higher postoperative VAS scores (HR 1.10, 95% CI 1.05–1.15, P < 0.001) were independent predictors of recurrence. For dysmenorrhea recurrence ( Table 5 ), a history of endometriosis (HR 2.06, 95% CI 1.19–3.57, P = 0.010), higher preoperative VAS scores (HR 1.14, 95% CI 1.02–1.28, P = 0.026), and higher postoperative VAS scores (HR 1.27, 95% CI 1.18–1.39, P < 0.001) were independent risk factors, suggesting that both baseline pain severity and persistent postoperative pain are associated with recurrence and may serve as clinical markers of persistent disease activity.

This study retrospectively screened women of childbearing age who underwent surgery for endometriosis at the Obstetrics and Gynecology Hospital of Fudan University in Shanghai from January 2009 to December 2010. A total of 1740 patients were identified, and the diagnosis of endometriosis was confirmed by two independent pathologists. Inclusion criteria were pathologically confirmed endometriosis and completion of at least one postoperative follow-up. Exclusion criteria included incomplete clinical data, concomitant malignancies, or loss to follow-up within 12 months. Of the 1740 patients, 1598 (91.8%) were successfully followed and included in the final analysis. During follow-up, 348 patients (21.8%) developed recurrence of endometrioma, and 187 of 846 patients (22.1%) with baseline dysmenorrhea experienced recurrence of dysmenorrhea. Follow-up was conducted via outpatient visits and standardized telephone interviews every three months during the first postoperative year and every six months thereafter. Routine transvaginal ultrasonography was performed to confirm endometrioma recurrence, and pain symptoms were systematically assessed at each visit ( Figure 1 ). Figure 1 Study flowchart of patient selection and follow-up. A total of 1,598 patients with surgically confirmed endometriosis were enrolled. Among them, 348 developed recurrence of endometrioma during follow-up, and 846 patients had baseline dysmenorrhea, of whom 187 experienced recurrence. Study flowchart of patient selection and follow-up. A total of 1,598 patients with surgically confirmed endometriosis were enrolled. Among them, 348 developed recurrence of endometrioma during follow-up, and 846 patients had baseline dysmenorrhea, of whom 187 experienced recurrence. The study was approved by the Ethics Committee of Obstetrics & Gynecology Hospital of Fudan University (No.: 2025–283). Written informed consent was obtained from all patients at the time of surgery. Patients were followed for a median of 58.0 months (IQR, 44.0–72.0 months). Data collected from medical records included demographic characteristics, clinical symptoms, pelvic examination findings, surgical details, postoperative medical therapy, and clinical scores, including the revised American Fertility Society (rAFS) score, Endometriosis Fertility Index (EFI), and Least Function (LF) score. Outcomes: Endometrioma recurrence was defined as the detection of an ovarian cyst ≥3 cm with typical ultrasonographic features on at least two consecutive menstrual cycles, irrespective of pain symptoms. The recurrence interval was defined as the time from surgery to the first confirmed recurrence. 15 , 16 The recurrence of dysmenorrhea was defined as the case when the patient with dysmenorrhea reported at least 3 months of alleviation of pain immediately after surgery yet the pain recurred later with the severity score equal to or higher than that before the surgery. Pain intensity was assessed using a 10-point visual analog scale (VAS), with higher scores indicating more severe pain. Both preoperative and postoperative VAS were analyzed as continuous variables in Cox models. These two recurrence outcomes were predefined separately because imaging-confirmed endometrioma recurrence and pain recurrence represent different clinical dimensions of postoperative disease persistence. Scoring systems: AFS classification was used to stage lesions. EFI score (0–10) was used to assess fertility potential. LF score was calculated as the sum of the lowest function score of each adnexal structure (fallopian tube, fimbria, ovary). Statistical analyses were performed using SPSS version 16.0 (SPSS Inc., New York, NY, USA). Continuous variables were expressed as median (interquartile range [IQR]) and categorical variables as number (percentage). Comparisons of continuous variables between groups were performed using the Mann–Whitney U -test, and categorical variables were compared using the χ 2 -test or Fisher’s exact test, as appropriate. Time-to-event outcomes (endometrioma and dysmenorrhea recurrence) were analyzed using Kaplan–Meier curves with Log rank tests. Multivariable Cox proportional hazards models were applied to estimate hazard ratios (HRs) with 95% confidence intervals (CIs). Clinically relevant variables and those with P < 0.10 in univariable analyses were considered for inclusion. The proportional hazards assumption was verified using Schoenfeld residuals. Postoperative medical therapy was coded as a binary variable (none vs. any therapy). Preoperative and postoperative VAS scores were analyzed as continuous variables, reflecting baseline pain severity and residual pain, respectively. ΔVAS (preoperative − postoperative) was additionally explored in sensitivity analyses but not included in the primary models. A two-sided P < 0.05 was considered statistically significant.

In summary, recurrence of endometrioma and dysmenorrhea after surgery remains high in Chinese women. Independent risk factors include bilateral ovarian involvement, older age at surgery, lack of postoperative medical therapy, and both higher preoperative and postoperative pain scores. These findings emphasize the need for long-term follow-up, optimized surgical strategies, and sustained adjuvant pharmacological therapy to reduce recurrence and improve patients’ quality of life.

Despite advances in surgical approaches such as laparoscopy and laparotomy, recurrence of endometriosis-related symptoms remains a major clinical challenge. 18 , 19 Identifying risk factors for recurrence is crucial to guide individualized management and long-term follow-up strategies. 15 , 20 Previous studies outside of China have reported heterogeneous recurrence rates, ranging from 8–30% within 3 years after surgery, with risk factors including disease stage, age at surgery and prior surgical history. 21 , 22 However, most of these studies were limited by relatively small sample sizes and shorter follow-up durations. Our study, based on a large cohort of 1,598 patients with a median follow-up of nearly 5 years, provides one of the most comprehensive datasets in China. 16 , 23 , 24 We observed cumulative recurrence rates of endometrioma at 1, 3, 5, and 7 years of 12%, 38%, 55%, and 65%, respectively, which are comparable to those reported in previous studies. Similarly, recurrence rates of dysmenorrhea reached 28%, 52%, 68%, and 75% at 1, 3, 5, and 7 years, respectively. These findings highlight the need for continued long-term follow-up to better understand recurrence patterns in different populations. In multivariable analysis, bilateral ovarian involvement, older age at surgery, lack of postoperative medical therapy, and higher postoperative VAS scores were independent predictors of endometrioma recurrence. The association between age and recurrence has been reported previously; women aged >30 years were found to have an increased risk compared with younger women. 19 Our findings are consistent, suggesting that age at surgery may influence recurrence risk. It should be noted that while univariable analysis suggested younger patients tended to recur more frequently, multivariable Cox regression identified older age as an independent predictor. This discrepancy highlights the importance of adjusting for confounding factors and warrants cautious interpretation. This reversal in direction may reflect the influence of confounding factors correlated with age, such as reproductive history, disease severity, or postoperative management, which were accounted for in the multivariable model. The role of postoperative medical therapy remains debated. Several randomized controlled trials have demonstrated that long-term use of oral contraceptives or progestins after surgery significantly reduces recurrence, 25 , 26 while other studies found no clear benefit. 2 In our cohort, lack of postoperative therapy doubled the risk of recurrence, supporting the integration of adjuvant pharmacotherapy into long-term management strategies. Notably, postoperative pain scores (VAS) were positively associated with recurrence risk. This finding suggests that persistent or worsening pain after surgery may indicate residual disease or ongoing pathological activity, which predisposes patients to recurrence. Clinically, this highlights the importance of adequate postoperative pain control not only for improving quality of life but also as a potential marker for recurrence surveillance. Sensitivity analysis of ΔVAS (preoperative – postoperative change) showed no independent association with recurrence, further indicating that absolute postoperative pain levels, rather than pain improvement alone, may better reflect recurrence risk. For dysmenorrhea recurrence, independent predictors included a history of endometriosis, higher preoperative VAS scores, and higher postoperative VAS scores. These findings suggest that both baseline pain severity and persistent postoperative pain are strong prognostic indicators. Previous reports also support the role of severe preoperative pain in predicting recurrence risk. 19 , 27 Taken together, our data emphasize that pain trajectories before and after surgery deserve greater attention in clinical follow-up. This study has several strengths, including its large sample size, long-term follow-up, and comprehensive evaluation of both clinical and surgical variables. To our knowledge, this is the first large-scale study in China to analyze recurrence of both endometrioma and dysmenorrhea with up to 7 years of follow-up. Nevertheless, several limitations should be noted. First, the retrospective design introduces potential selection bias and limits causal inference. Second, heterogeneity in postoperative medical regimens (types, duration, adherence) could not be fully standardized. Third, pain assessment relied on VAS scores, which are subjective and may vary with cultural or personal perception. Moreover, postoperative VAS was assessed at fixed follow-up intervals, which may not capture dynamic changes in pain perception over time. Finally, this study was conducted at a single tertiary center, and findings may not be generalizable to broader populations. In summary, our findings highlight the importance of identifying high-risk patients—those with bilateral ovarian involvement, older age at surgery, prior endometriosis history, and persistent pain after surgery—for closer surveillance and individualized treatment. Long-term follow-up and appropriate postoperative medical therapy should be considered essential components of management to reduce recurrence and improve patient outcomes.