Endometriosis is a chronic inflammatory gynecological disease characterized by the presence of endometrial-like tissue outside the uterine cavity. It predominantly affects women during the reproductive years and is associated with dysmenorrhea, chronic pelvic pain, dyspareunia, infertility, and impaired quality of life (1). The clinical presentation is heterogeneous, and symptom severity does not always correspond directly to the anatomical extent of disease (2).

Management generally involves medical therapy, surgery, or a combination of both. Treatment is individualized according to symptoms, disease extent, age, reproductive plans, previous treatment, and patient preferences (3,4). Conservative surgery is commonly undertaken in women wishing to preserve fertility, particularly when ovarian endometriomas or pelvic endometriotic lesions are present. The aim is to treat visible disease while preserving ovarian and reproductive function (5).

Despite symptomatic improvement after surgery, recurrence remains an important clinical problem. Recurrent disease may manifest as renewed pelvic pain, recurrent endometrioma, infertility, or radiological evidence of disease. Published recurrence estimates vary according to the definition used, duration of surveillance, disease severity, surgical technique, and postoperative medical treatment (6–8).

Ovarian endometrioma deserves particular attention because surgical removal can improve disease-related symptoms while potentially affecting ovarian reserve. Consequently, clinicians must balance adequate treatment of disease against preservation of healthy ovarian tissue, particularly in patients with fertility goals (9,10).

Disease severity may also influence postoperative recurrence. Advanced endometriosis can involve extensive pelvic disease, adhesions, bilateral ovarian involvement, and greater overall disease burden (11). Postoperative hormonal suppression is frequently considered in patients who are not immediately attempting pregnancy and may reduce symptom recurrence and delay disease recurrence in selected patients (3,4,12,13).

The identification of clinical and disease-related characteristics associated with recurrence can assist postoperative counseling and surveillance planning. The present study therefore evaluated recurrence following conservative surgery and examined its relationship with demographic characteristics, symptoms, infertility, lesion laterality, rAFS stage and score, cyst diameter, previous surgery, postoperative hormonal therapy, and follow-up duration.

This retrospective cohort study was based on a dataset containing 148 records of patients undergoing conservative surgery for endometriosis. The available variables included patient identification, hospital, city, age, BMI, parity, preoperative dysmenorrhea, infertility, lesion side, rAFS stage, rAFS score, cyst diameter, prior surgery, postoperative hormonal therapy, recurrence, and follow-up duration. The principal postoperative outcome was documented recurrence of endometriosis. Recurrence was coded as yes or no. Continuous variables were summarized as mean ± standard deviation (SD), while categorical variables were summarized as frequencies and percentages. Continuous variables were compared between recurrence groups using Welch's independent-samples t-test. Categorical associations were assessed using Pearson's chi-square test or Fisher's exact test when sparse or zero cell frequencies were present. A two-sided p-value <0.05 was considered statistically significant. Because several predictors demonstrated complete or quasi-complete separation of recurrent and non-recurrent cases, univariable and multivariable Firth penalized logistic regression were used to reduce small-sample bias and address separation. Odds ratios (ORs), adjusted odds ratios (aORs), and 95% confidence intervals (CIs) were reported. A reduced multivariable model was used because of strong correlations among age, BMI, parity, rAFS score, cyst diameter, and follow-up duration. Age, rAFS score, and postoperative hormonal therapy were retained as representative clinically relevant predictors. The supplied dataset is explicitly labeled as simulated. Accordingly, the present analysis represents an analysis of the supplied records and should not be interpreted as evidence from actual patients or named institutions unless independently validated against source clinical records.

A total of 148 records were analyzed. Recurrence was documented in 44 patients (29.7%), while 104 (70.3%) had no documented recurrence. The overall mean age was 32.49 ± 5.02 years, mean BMI was 22.54 ± 2.16 kg/m², mean parity was 1.39 ± 1.12, mean rAFS score was 29.74 ± 11.17, mean cyst diameter was 6.74 ± 1.50 cm, and mean follow-up duration was 28.76 ± 11.68 months.

Table 1. Demographic and continuous clinical characteristics according to recurrence.

Patients who developed recurrence were significantly younger and had lower BMI and parity. They also had substantially higher rAFS scores and larger cysts. The mean rAFS score was 43.45 ± 3.14 among recurrent cases compared with 23.93 ± 7.71 among non-recurrent cases (p<0.001). Mean cyst diameter was 8.63 ± 0.47 cm versus 5.94 ± 0.99 cm, respectively (p<0.001).

Table 2. Categorical clinical characteristics according to recurrence.

Preoperative dysmenorrhea, infertility, rAFS stage, prior surgery, and postoperative hormonal therapy were significantly associated with recurrence. All 44 recurrent records were stage IV, whereas none of the 75 stage III records had documented recurrence. Lesion laterality was not significantly associated with recurrence (p=0.491).

Table 3. Univariable Firth logistic regression for recurrence.

Greater rAFS score and larger cyst diameter were strongly associated with increased odds of recurrence. Stage IV disease showed a strong association with recurrence. The very large inverse association for postoperative hormonal therapy reflects complete separation because all patients receiving therapy were classified as non-recurrent.

Table 4. Reduced multivariable Firth logistic regression for recurrence.

After adjustment, age and rAFS score were no longer independently associated with recurrence in the reduced model. Postoperative hormonal therapy remained strongly inversely associated with documented recurrence; however, the magnitude is driven by complete separation and should not be interpreted as a conventional causal treatment-effect estimate.

The present analysis found documented recurrence in 44 of 148 records (29.7%) following conservative surgery. Recurrence is a recognized challenge in endometriosis management because surgery removes visible disease but does not necessarily eliminate the biological processes responsible for persistence and recurrence (14–16). Advanced disease was strongly associated with recurrence in the supplied records. All recurrent records were classified as stage IV, while no stage III record had documented recurrence. Advanced disease generally reflects greater anatomical disease burden and may include extensive adhesions, ovarian involvement, and other manifestations of severe endometriosis (19). The complete separation observed here is unusually strong and should therefore be interpreted cautiously. rAFS score was substantially higher among recurrent records. The mean score was approximately 43.5 compared with 23.9 among non-recurrent records. Cyst diameter was also greater among recurrent records (8.63 vs. 5.94 cm). These findings support a descriptive association between greater disease burden and recurrence, while the close relationship between stage, score, and cyst size limits claims about independent effects. Postoperative hormonal therapy showed a marked inverse association with recurrence. Hormonal suppression is commonly used after conservative surgery in patients who are not immediately attempting pregnancy and may reduce symptom recurrence in selected patients (3,4,12,13). However, every patient receiving hormonal therapy was classified as non-recurrent and every recurrent record lacked hormonal therapy. This complete separation prevents interpretation of the estimated magnitude as a realistic causal treatment effect. Preoperative dysmenorrhea and infertility were also associated with recurrence. These are common clinical manifestations of endometriosis and may identify patients with a more symptomatic or reproductive phenotype (1,2,23). Nevertheless, complete separation limits causal interpretation. Previous surgery showed an inverse association with recurrence in the supplied records, with no recurrent case having previous surgery documented. This should not be interpreted as evidence that prior surgery protects against recurrence because the pattern may reflect the structure of the supplied dataset. Follow-up duration differed markedly between groups. Patients without documented recurrence had substantially longer follow-up than recurrent records. This should not be interpreted as evidence that longer follow-up prevents recurrence. Time-to-event analysis would be preferable when exact dates of surgery and recurrence are available. The major limitation is that the supplied dataset is explicitly labeled simulated. Additional limitations include complete separation between recurrence and several predictors, strong correlations among age, BMI, parity, disease severity, cyst diameter and follow-up, and absence of pain scores, quality-of-life measures, ovarian reserve, fertility outcomes, complications, and reoperation. The actual recurrence rate in the supplied records is 29.7%, differing from the dataset description's target of approximately 20%. Despite these limitations, the analysis identifies clinically relevant patterns within the supplied records, particularly the association of recurrence with advanced disease and larger cyst size. Studies using verified clinical records, standardized recurrence definitions, uniform surveillance, and time-to-event methods would provide stronger evidence.

In the supplied 148 records, recurrence following conservative surgery for endometriosis was documented in 44 patients (29.7%). Recurrence was associated with younger age, lower BMI and parity, infertility, preoperative dysmenorrhea, advanced rAFS stage, higher rAFS score, larger cyst diameter, absence of postoperative hormonal therapy, and shorter recorded follow-up. Disease severity and cyst burden showed particularly strong associations. Because the dataset is simulated and several variables demonstrate complete separation, the findings should be interpreted as descriptive associations rather than causal effects. Validation using real clinical data with standardized follow-up is required before clinical conclusions can be drawn.