Our comparison of the accuracy of aspiration biopsy and D&C in patients diagnosed with preoperative endometrial hyperplasia suggests that D&C is a more accurate endometrial sampling method, particularly for patients with complex hyperplasia, than aspiration biopsy.
In patients diagnosed with endometrial hyperplasia, there are risks of hidden malignancy and the possibility of progression to endometrial carcinoma. The risk of progression is < 5% in women with endometrial hyperplasia without atypia but increases up to 30% in women with atypical endometrial hyperplasia [9]. Therefore, hysterectomy is recommended, particularly in patients with atypical endometrial hyperplasia [2]. However, in those who desire the preservation of their fertility or are unable to tolerate surgery due to coexisting medical conditions, conservative management is inevitable. In these patients, accurate diagnostic evaluation is particularly vital, as their final pathologic results might not be verified. In our study, 210 (84%) of the patients were premenopausal patients for whom special attention is required to exclude coexisting malignancy.
Various methods of endometrial sampling are used in either inpatient or outpatient settings as a basis to diagnose endometrial lesions. Past studies regarding the accuracy of D&C compared to that of endometrial biopsy in diagnosing endometrial hyperplasia show conflicting results. One prospective study investigated the comparable diagnostic value of D&C and biopsy [10]. Of the 70 patients, 55% and 45% of cases were diagnosed with complex atypical hyperplasia using endometrial biopsy and D&C, respectively. No difference in the incidence of coexisting cancer at hysterectomy was observed between the two methods (41% vs 45%; P > 0.05), but this study conclusion was limited by the small sample size. Another study evaluated whether preoperative D&C lowers the risk of unexpected cancer at hysterectomy compared to biopsy alone and reported that D&C lowered the risk of unexpected carcinoma compared to endometrial biopsy in patients with complex atypical hyperplasia (30% vs 45%; P < 0.001) [11]. Dijkhuizen et al. performed a meta-analysis to assess the accuracy of endometrial sampling devices for detecting endometrial carcinoma and endometrial hyperplasia [12]. They concluded that aspiration biopsy with Pipelle is superior to other endometrial instruments with sensitivities of 99.6% and 91% for endometrial carcinoma and endometrial hyperplasia, respectively, and more than 98% specificity. However, this review primarily included studies that examined the final pathology using D&C. Of the 39 studies, only five included studies used hysterectomy as the reference for final pathology [13–17].
Compared to previous studies, our study has several strengths. First, we include a substantial number of patients diagnosed with endometrial hyperplasia, which was one of the limitations of previous studies [10]. Second, we evaluate the final pathology based on hysterectomy specimens to compare two endometrial sampling methods, which provides more accuracy than using D&C as a reference. Third, in our study, the patients with “insufficient tissue for pathological examination” were excluded to further clarify the results. One of the concerns about endometrial sampling is the sufficiency of sample specimen. In some studies, inadequate pathological specimens that may have affected the results were included or the adequacy of the specimens was not considered [8]. Lastly, we also excluded patients who received hormone therapy during the period between endometrial sampling and hysterectomy, as this treatment may significantly affect pathology.
One potential limitation of the current study is that we could not apply the updated endometrial hyperplasia classification. There are two classification systems of endometrial hyperplasia: the WHO94 classification, used in this study, and the endometrial intraepithelial neoplasm (EIN) classification. The current position statement of the American College of Obstetricians and Gynecologists and the Society of Gynecologic Oncology recommends using the EIN classification, which includes three categories: benign (benign endometrial hyperplasia), premalignant (EIN), and malignant (well-differentiated endometrial adenocarcinoma) [18]. The EIN classification is more likely to identify precancerous lesions than the WHO94 classification. However, as our pathology department has only recently converted to the new terminology, the old WHO94 classification was used in this study. The WHO94 classification remains the most commonly used and reported classification in existing literature [4]. Another limitation of the current study is that we did not evaluate the effect of endometrial thickness. Thickened endometrium is known to be associated with endometrial pathologies. However, this may have little impact on the result of our study. A previous study evaluating the effect of endometrial thickness between biopsy methods showed no statistically significant results [6]. The final limitation is that we did not compare various devices used for endometrial sampling. It is possible that differences in sampling devices may affect the accuracy of the results.
In conclusion, as a gold standard method of endometrial sampling, higher accuracy with final pathology is expected from D&C than from aspiration biopsy. For an accurate diagnosis of endometrial hyperplasia, D&C seems superior to aspiration biopsy. When considering management strategies for women with endometrial hyperplasia obtained from aspiration biopsy, physicians should take into account the considerable rate of upgraded diseases.