By Rebecca H. Allen, MD, MPH, Editor
Synopsis: In this large retrospective cohort study at a single health system, the rate of insufficient endometrial biopsies among 27,456 patients was 12.1%. The largest risk factor for an insufficient sample was postmenopausal status (adjusted odds ratio, 5.27; 95% confidence interval, 4.88-5.70). Among premenopausal women, a history of cesarean delivery (9.6% vs. 7.2%; P < 0.001) and leiomyomas (10.3% vs. 7.2%; P < 0.001) was associated with insufficient samples.
Source: Howe M, Hung YY, Bhatnagar R, et al. Patient and procedural factors associated with insufficient office endometrial biopsy. O&G Open. 2025;2(4):e106.
The authors of this study sought to identify factors that could predict insufficient or scant tissue yield on the office endometrial biopsy. The office endometrial biopsy is a key procedure to evaluate abnormal uterine bleeding and postmenopausal bleeding.
This was a retrospective cohort study from the Kaiser Permanente Northern California healthcare system. Inclusion criteria were all adult patients who underwent office endometrial biopsy between Jan. 1 and Dec. 31, 2023. Exclusion criteria were prior history of an endometrial ablation. The authors collected demographic and clinical data, including age, race, ethnicity, body mass index (BMI), presence of leiomyomas, prior cesarean delivery, prior loop electrosurgical excision procedure (LEEP) or cone biopsy, and chronic pain or mental health diagnoses. Postmenopausal status was defined as age 55 years or older. Insufficient endometrial biopsy samples were identified by pathology reports with the terms, “insufficient,” “not sufficient,” “limited,” “unsatisfactory,” “not satisfactory,” “no endometrial tissue,” “no endometrial epithelium/tissue,” “nondiagnostic specimen,” or “scant.” The primary outcome was the rate of insufficient endometrial biopsies. Analyses were stratified by menopausal status. Both univariable and multivariable logistic regression analysis were performed.
In 2023, there were 27,456 patients who met the study inclusion criteria, with 12.1% insufficient endometrial biopsies. The median age of the study population was 48 years (interquartile range, 42-54 years), with 21.2% aged 55 years or older and considered postmenopausal. The rate of insufficient samples in postmenopausal patients was 29.1% compared with 7.6% of premenopausal patients (P < 0.001). Chronic pain was associated with 14.0% insufficient rates compared to 11.8% among those without chronic pain (P < 0.001). Patients with a BMI of 50 kg/m2 or higher had lower rates of insufficient samples (8.0%) compared with those with a BMI < 50 kg/m2 (12.3%; P < 0.001).
Among premenopausal patients only, significant predictors of insufficient samples were leiomyomas (10.3% vs. 7.2%; P < 0.001) and prior cesarean delivery (9.6% vs. 7.2%; P < 0.001). Among postmenopausal patients, predictors were leiomyomas (35.0% vs. 29.0%; P < 0.001) and anxiety (31.7% vs. 28.5%; P = 0.04). In the multivariable logistic regression model, the strongest predictor of insufficiency was postmenopausal status (adjusted odds ratio [OR], 5.27; 95% confidence interval [CI], 4.88-5.70). Leiomyomas also persisted as a predictor in premenopausal (adjusted OR, 1.47; 95% CI, 1.28-1.70) and postmenopausal patients (adjusted OR, 1.36; 95% CI, 1.07-1.73).
Commentary
Insufficient endometrial biopsies can lead to repeat procedures, often in the operating room with a dilation and curettage hysteroscopy causing patients anxiety and stress. Previous studies examining this issue often were limited by small sample sizes. Therefore, the authors undertook this study in their large health system, collecting a sample size of 27,456 patients in only one year, and had sufficient power to detect differences. However, some limitations of this study include the definition of menopause, which was age-based and not clinically based. Therefore, some patients may have been misclassified as premenopausal who were younger than 55 years of age but were menopausal. Additionally, the authors were unable to identify any instances of failed endometrial biopsies.
Not surprisingly, the investigators found that postmenopausal status increases the risk of insufficient endometrial biopsy samples. Often, these patients have atrophic endometrium that leads to these scant samples. Nevertheless, with postmenopausal bleeding, tissue samples are required to rule out malignancy, especially if the endometrial lining is greater than 4 mm.1 An interesting finding in this study is the fact that patients with a BMI of 50 kg/m2 or more had a lower risk of insufficient sampling. This may be because of the extra adipose tissue causing higher levels of estrogen, which stimulates growth of the endometrial lining, making sampling easier.
The authors suggested that patients with a risk of insufficient sampling be counseled about the potential for inadequate tissue. Although typically we offer patients an attempt at endometrial biopsy in the office, some patients may opt to go directly to the operating room for a dilation and curettage hysteroscopy to ensure an adequate sample. In postmenopausal patients, the investigators suggested possible pre-procedure interventions, such as topical vaginal estrogen for at least two weeks before the procedure or planned use of mechanical cervical dilators.2 Ultrasound guidance also is helpful if mechanical cervical dilation will be performed.
Another option is preprocedural misoprostol, which has been shown to be helpful prior to operative hysteroscopy in both premenopausal and postmenopausal patients in reducing the amount of mechanical dilation required.3 One regimen is 400 mcg of vaginal misoprostol the night before the procedure. The oral route also can be used. Whether this also would apply to endometrial biopsies is not as clear, and the data regarding the efficacy of misoprostol in postmenopausal patients are contradictory.4 Nevertheless, I suspect most of the inadequate sampling in postmenopausal patients has less to do with cervical stenosis and more to do with endometrial atrophy.
Rebecca H. Allen, MD, MPH, is Professor, Department of Obstetrics and Gynecology, Warren Alpert Medical School of Brown University, Women & Infants Hospital, Providence, RI.
References
1. Carugno J. Clinical management of vaginal bleeding in postmenopausal women. Climacteric. 2020;23(4):343-349.
2. Casadei L, Piccolo E, Manicuti C, et al. Role of vaginal estradiol pretreatment combined with vaginal misoprostol for cervical ripening before operative hysteroscopy in postmenopausal women. Obstet Gynecol Sci. 2016;59(3):220-226.
3. Al-Fozan H, Firwana B, Al Kadri H, et al. Preoperative ripening of the cervix before operative hysteroscopy. Cochrane Database Syst Rev. 2015;2015(4):CD005998
4. Wood MA, Kerrigan KL, Burns MK, et al. Overcoming the challenging cervix: Identification and techniques to access the uterine cavity. Obstet Gynecol Surv. 2018;73(11):641-649.
