Diagnostic value and concordance of endometrial thickness, smear, and biopsy results in women with abnormal uterine bleeding

S.B. Torumtay Alic, H.E. Malatyalioglu

Department of Obstetrics and Gynecology, The Private Corum Hospital, Corum, Türkiye. sbtorumtay@gmail.com

---

• Obstetrics

OBJECTIVE: This study was conducted to investigate whether an effective screening method for endometrial hyperplasia and endometrial cancer can be identified by comparing the results of transvaginal ultrasonography (TV-USG), liquid-based endometrial cytology, and endometrial biopsy.

MATERIALS AND METHODS: The study included a total of 200 patients with a mean age of 50.04 ± 10.39 years (range: 33-85), who presented with abnormal uterine bleeding during the premenopausal and postmenopausal periods. Following a detailed clinical examination, endometrial thickness was measured using TV-USG in all cases. Subsequently, liquid-based endometrial cytology was obtained using an endometrial brush, followed by an endometrial biopsy using a Karman cannula. All patients were evaluated by the same clinician, and both the cytological and biopsy specimens were assessed by the same pathologist.

RESULTS: No endometrial pathology was detected in cases with an endometrial thickness of ≤ 5 mm as measured by TV-USG. The mean endometrial thickness was found to be 12.66 mm for simple hyperplasia without atypia, 18 mm for complex atypical hyperplasia, and 17.00 mm for endometrial adenocarcinoma. A statistically significant difference was observed among these three groups with endometrial thicknesses (p < 0.05). The endometrial thickness in cases with complex atypical hyperplasia and endometrial cancer was significantly higher compared to the others (p < 0.05). While 58% of endometrial cytology samples and 34% of biopsy samples were found to be insufficient for evaluation, cytology still appeared to be a useful method for detecting malignant lesions. The high insufficiency rate observed in our series likely reflects the sampling technique employed, as an endocervical-type brush was used rather than a dedicated endometrial sampler, which may have reduced cellular yield and affected adequacy rates. For the diagnosis of malignant lesions, cytology demonstrated a sensitivity of 100%, a specificity of 96.9%, a positive predictive value of 71.43%, and a negative predictive value of 100%.

CONCLUSIONS: Although TV-USG is not a method that provides a definitive diagnosis, it is highly effective in identifying cases that require further investigation. Moreover, due to its non-invasive nature, it should be considered the first-line diagnostic tool in patients suspected of having endometrial pathology. Despite the high diagnostic accuracy of endometrial cytology in detecting malignant lesions, the relatively high rate of insufficient sample both in premenopausal and postmenopausal patients compared to biopsy represents a notable disadvantage. Therefore, we do not consider endometrial cytology to be a standalone method suitable for screening endometrial cancer and its precursors. However, since malignant lesions were reliably detected when adequate cytology material was obtained, dedicated endometrial samplers and emerging molecular adjuncts should be evaluated in future prospective studies to determine whether cytology can contribute to a multimodal screening strategy.

Free PDF Download