Identifying risk factors for procedural pain in outpatient hysteroscopy: A multivariable analysis.

Office hysteroscopy has become an essential diagnostic and therapeutic modality in contemporary gynecologic practice owing to its minimally invasive nature, high diagnostic accuracy, and feasibility in outpatient settings without general anesthesia. 1 – 3 Compared with transvaginal ultrasonography and saline infusion sonohysterography, hysteroscopy allows direct visualization of the uterine cavity and enables simultaneous diagnosis and treatment of intrauterine pathology, thereby reducing false-negative results and the need for repeated interventions. 4 – 9 Despite these advantages, procedural pain remains a major barrier to the widespread acceptance of office hysteroscopy. 10 , 11 Pain-related intolerance accounts for up to 80% of incomplete procedures, with reported failure rates ranging from 1.3% to 5.2%. 6 , 8 , 12 Pain during hysteroscopy is multifactorial, arising primarily from cervical manipulation, uterine distension, and uterine contractility. Both procedural factors (hysteroscope diameter, distension pressure, and procedure duration) and patient-related characteristics influence pain perception. 5 , 8 , 10 Previous studies have identified parity, menopausal status, anxiety, and anatomical variations as potential predictors of hysteroscopic pain. 13 – 15 However, findings remain inconsistent across populations, and data regarding uterine orientation, dysmenorrhea, and obstetric history are limited. A clearer understanding of patient-specific predictors may enable tailored pain management strategies and improve procedural success. The aim of this study was to evaluate clinical and anatomical factors associated with pain perception during office hysteroscopy, with the goal of identifying predictors that may guide individualized preprocedural counseling and the development of tailored pain mitigation strategies.

This retrospective study reviewed medical records of women who underwent office hysteroscopy at the Division of Gynecologic Endoscopy, Department of Obstetrics and Gynecology, Chang Gung Memorial Hospital at Linkou, Taiwan, between August and November 2024. Patients were included if they underwent hysteroscopy for abnormal uterine bleeding refractory to medical therapy, infertility evaluation, postmenopausal bleeding, or suspected intrauterine pathology. Exclusion criteria included pregnancy, pelvic inflammatory disease, suspected cervical malignancy, intolerable pain leading to procedural discontinuation, or the need for major operative hysteroscopic procedures. All procedures utilized rigid hysteroscopes (2.7-mm diameter; Karl Storz, Tuttlingen, Germany or 4.0-mm diameter; Richard Wolf, Knittlingen, Germany) and were performed without anesthesia. Normal saline served as the distension medium, delivered via an electronic pump system (Endomat, Karl Storz) that maintained an intrauterine pressure between 50 and 100 to ensure optimal visualization. 4 , 6 In routine clinical practice, the 4.0-mm hysteroscope was used as the initial instrument, and the 2.7-mm hysteroscope was used only when the 4.0-mm hysteroscope could not be advanced through the isthmus. Accordingly, hysteroscope diameter was included in the multivariable logistic regression model as a potential confounding factor for pain perception. Preprocedure data included patient demographics such as age, body mass index (BMI), gravidity, sexual experience, history of vaginal delivery, menopausal status, severity of dysmenorrhea, uterine position revealed by ultrasound, and the indications for the procedure. Intraprocedure and postprocedure data were also collected, including cervical width of the uterus and pain intensity. Cervical width was estimated intraoperatively based on the ease of hysteroscope passage through the isthmus. Smooth passage of the 4.0-mm hysteroscope through the isthmus indicated a cervical width ≥4 mm, whereas failure to advance the 4.0-mm hysteroscope through the isthmus indicated cervical narrowing or a cervical width <4 mm. Pain intensity was assessed immediately after the procedure using a 10-cm visual analog scale (VAS). Patients were categorized into no-to-mild pain (VAS <4) and moderate-to-severe pain (VAS ≥4) groups. 16 , 17 Pain intensity was assessed approximately 3–5 minutes after completion of the procedure using a visual analog scale. The VAS score represented the patient’s overall procedural pain experience during office hysteroscopy, rather than pain from individual procedural components. All office hysteroscopy procedures were performed by the first author, who had 3 years of experience in hysteroscopy during the study period. This study was approved by the Institutional Review Board of Chang Gung Memorial Hospital (202501244B0). An Excel database was created to facilitate data entry and retrieval (Excel 2021; Microsoft, Redmond, WA). Group comparisons were conducted using chi-square tests. Univariate and multivariable logistic regression analyses used all relevant variables (age, BMI, gravidity, sexual experience, history of vaginal delivery, menopausal status, severity of dysmenorrhea, uterine position, cervical width of the uterus, and caliber of hysteroscope) to identify factors associated with moderate-to-severe pain. Statistical analyses were performed using SPSS software, Version 30 (IBM Corp., Armonk, NY). A two-sided p < 0.05 was considered statistically significant.

Hysteroscopy was successfully completed in all but one patient (0.3%), a postmenopausal woman who presented with bleeding but had no abnormal findings on sonography. The procedure failed due to severe intrauterine/cervical synechiae rather than intolerable pain; therefore, this case was excluded from the final analysis. Ultimately, a total of 338 women undergoing outpatient hysteroscopy were included in the analysis (Fig. 1 ). The mean age was 43.3 ± 11.2 years, and the mean BMI was 23.7 ± 4.4 kg/m 2 . Most patients were premenopausal (85.8%) and sexually active (95.6%). Nulligravid women accounted for 39.1% of the cohort, while 64.2% had a history of vaginal delivery. Moderate-to-severe dysmenorrhea (VAS ≥4) was reported by 25.5% of premenopausal women. An anteflexed uterus was observed in 77.2% of patients, and cervical width ≥4 mm was present in 83.7%. The most common indication for hysteroscopy was suspected intrauterine mass (42.9%). Baseline characteristics are summarized in Table 1 . Baseline characteristics of the study population (n = 338) Values are presented as mean ± SD (range) or n (%). VAS = 10-cm visual analog scale. Postmenopausal women were excluded from the dysmenorrhea analysis. Some patients had more than one indication, but the dominant indication was selected for analysis. Study flowchart and factors associated with pain during office hysteroscopy. aOR = adjusted odds ratio; VAS = 10-cm visual analog scale. Of the study population, 200 women (59.2%) experienced no-to-mild pain (VAS <4), while 138 women (40.8%) reported moderate-to-severe pain (VAS ≥4) during the procedure. There were no significant differences between the two groups with respect to age, BMI, or menopausal status. However, several patient-related and anatomical factors were significantly associated with higher pain intensity. Women in the moderate-to-severe pain group were more likely to be nulligravid (50.0% vs 31.5%; p < 0.01) and less likely to have a history of vaginal delivery (23.9% vs 44.0%; p < 0.01). Dysmenorrhea (VAS ≥4) was markedly more prevalent among women reporting higher pain levels (48.7% vs 9.4%; p < 0.01). In addition, retroflexed uterine position (29.0% vs 18.5%; p = 0.024) and cervical width <4 mm (26.1% vs 9.5%; p < 0.01) were significantly more frequent in the moderate-to-severe pain group. Because the indication for office hysteroscopy may influence procedural complexity and patient pain perception, each indication was analyzed individually. However, no specific indication was significantly associated with higher pain scores. Detailed comparisons are presented in Table 2 . Comparison of clinical characteristics according to pain intensity during office hysteroscopy Values are presented as mean ± SD or n (%). p Values were calculated using the Student t test, the chi-square test, or Fisher exact test, as appropriate. VAS = 10-cm visual analog scale. Postmenopausal women were excluded from the dysmenorrhea analysis. In univariate logistic regression analysis, nulligravidity, absence of prior vaginal delivery, dysmenorrhea (VAS ≥4), cervical width <4 mm, and a retroflexed uterus were significantly associated with moderate-to-severe pain during office hysteroscopy, whereas age, BMI, menopausal status, and sexual activity were not significantly associated with pain severity. After adjustment for potential confounders in the multivariable logistic regression model, dysmenorrhea (VAS ≥4) emerged as the strongest independent predictor of moderate-to-severe pain (adjusted odds ratio [aOR], 9.42; 95% CI, 4.87-18.22; p < 0.001). Cervical width <4 mm also remained independently associated with increased pain risk (aOR, 6.51; 95% CI, 2.79-15.16; p < 0.001). In addition, use of the 2.7-mm hysteroscope was significantly associated with higher odds of moderate-to-severe pain compared with the 4.0-mm hysteroscope (aOR, 3.61; 95% CI, 1.24-10.51; p = 0.019). However, this association should be interpreted cautiously, as the 2.7-mm hysteroscope was used only when the 4.0-mm hysteroscope could not be advanced through the isthmus, suggesting that its use may reflect technically difficult cases rather than a direct causal effect of instrument diameter. The absence of a history of vaginal delivery and a retroflexed uterus showed borderline associations with moderate-to-severe pain (aOR, 1.86; 95% CI, 0.96-3.90; p = 0.087 and aOR, 1.74; 95% CI, 0.95-3.19; p = 0.075, respectively). None of the remaining variables retained statistical significance in the multivariable model. These results are summarized in Table 3 . Univariate and multivariable logistic regression analyses of factors associated with moderate-to-severe pain during office hysteroscopy Outcome variable: moderate-to-severe pain (VAS >4). OR = odds ratio; VAS = 10-cm visual analog scale.

This retrospective study identified several patient-related factors independently associated with increased pain during office hysteroscopy, including nulligravidity, absence of prior vaginal delivery, dysmenorrhea, and narrow cervical width. Nulligravid women and those without a history of vaginal delivery may experience greater discomfort due to reduced cervical compliance and smaller cervical dimensions, which increase resistance during hysteroscope insertion. 15 Cervical width <4 mm was a strong predictor of pain, consistent with previous reports emphasizing the importance of cervical anatomy in hysteroscopic tolerance. 5 , 10 Although uterine axis deviation did not reach statistical significance in the multivariate analysis, marked deviations (pronounced anteversion or retroversion) were still clinically associated with increased pain levels during office hysteroscopy. This observation remains relevant in clinical practice, as altered uterine angulation may complicate hysteroscope insertion and contribute to procedural discomfort. Dysmenorrhea emerged as the strongest predictor of pain, potentially reflecting heightened uterine sensitivity or underlying conditions such as adenomyosis or endometriosis. In contrast, age, menopausal status, and cesarean delivery history were not independently associated with pain, suggesting that mechanical rather than hormonal factors play a predominant role. The association between dysmenorrhea and significant pain during office hysteroscopy may be explained by both peripheral and central mechanisms. Dysmenorrhea is closely associated with increased uterine prostaglandin production, particularly prostaglandin F2α and prostaglandin E2. As described in previous reviews, these prostaglandins can induce myometrial hypercontractility, vasoconstriction, uterine ischemia, and pelvic pain. 18 , 19 This prostaglandin-mediated response may contribute to peripheral sensitization and lower the pain threshold during uterine instrumentation, isthmic passage, or uterine distension. In addition, recurrent menstrual pain may contribute to altered pain processing and central sensitization. Previous studies have suggested that women with dysmenorrhea may have increased somatic or visceral pain sensitivity. 19 Therefore, patients with dysmenorrhea may experience greater procedural pain during office hysteroscopy. However, because this was a retrospective study, we could not directly assess prostaglandin levels, central sensitization, or pain sensitivity. Further prospective studies are needed to clarify these mechanisms. These results emphasize the importance of implementing individualized pain management strategies for office hysteroscopy. Identifying high-risk patients preoperatively may facilitate the use of tailored interventions, including anticipatory analgesia, smaller-diameter hysteroscopes, cervical priming, or scheduling the procedure in a more controlled setting such as an operating room under anesthesia for select patients. Additionally, these findings highlight the importance of thorough preprocedural counseling to set appropriate expectations and improve the overall patient experience. With respect to analgesia, nonsteroidal anti-inflammatory drugs (NSAIDs) have been shown to effectively reduce pain both during and after office hysteroscopic procedures, without a corresponding increase in adverse effects. 3 Several studies have demonstrated that preprocedural administration of oral NSAIDs (such as diclofenac, mefenamic acid, and naproxen) can significantly enhance patient comfort by attenuating uterine cramping and minimizing the inflammatory response. The use of NSAIDs, in combination with the risk factors identified in this study, should be carefully considered in clinical practice; furthermore, this represents an important area that warrants deeper investigation in future studies. The challenge of effective preoperative communication may be mitigated through multidimensional strategies shown to improve patient understanding and satisfaction. 5 , 20 As patients remain conscious during office hysteroscopy, it is equally critical to maintain real-time intraprocedural communication, informing them of each procedural step. This approach can help reduce anxiety, promote patient cooperation, and improve overall procedural tolerance. 15 This study has several limitations. Operative time was not included in the analysis, although longer procedures may contribute to increased patient discomfort. In addition, detailed, step-specific procedural information was unavailable; therefore, we could not determine whether pain originated predominantly from cervical passage, isthmic entry, uterine distension, or intrauterine manipulation. While all procedures were performed by a single operator to minimize interoperator variability, operator-specific technique and experience might have influenced procedural efficiency and patient discomfort, potentially limiting the generalizability of our findings. Furthermore, anxiety levels were not formally assessed using a validated scale. Lastly, prior hysteroscopy experience and the use of preprocedural analgesics or NSAIDs were not recorded, though these factors may significantly influence pain perception. Future prospective studies incorporating operative time, procedural details, operator-related variables, and validated anxiety measures are needed to provide a more comprehensive understanding of pain perception during office hysteroscopy. In conclusion, dysmenorrhea and narrow cervical width were identified as strong independent predictors of moderate-to-severe pain during office hysteroscopy. In addition, the absence of prior vaginal delivery and a retroflexed uterus showed borderline associations with increased pain risk. Based on these findings, patients may be clinically stratified into risk categories according to the presence of the two independent predictors: low risk if neither factor is present, intermediate risk if one factor is present, and high risk if both factors are present. Recognition of these patient-specific risk factors may facilitate individualized procedural planning, targeted pain management strategies, and improved patient tolerance during office hysteroscopy.

This study was supported by grants from the Chang Gung Medical Foundation (CMRPG3N0691), but this study was not sponsored by any surgical device company.