The accuracy of endometrial biopsy and saline sonohysterography in the determination of the cause of abnormal uterine bleeding Lillian M. Mihm, MD, Valerie A. Quick, BSN, RN, Jonathan A. Brumfield, RDMS, RT(R), Alfred F. Connors, Jr, MD, and James J. Finnerty, MD Charlottesville, Va OBJECTIVE: The purpose of this study was to determine the accuracy of outpatient endometrial biopsy and saline sonohysterography for the evaluation of abnormal uterine bleeding. STUDY DESIGN: Eligible participants included women aged 25 to 69 years who complained of persistent uterine bleeding, despite medical treatment. One hundred forty-four patients consented and were followed up prospectively: 1 patient did not successfully complete a saline sonohysterography because of discomfort, 143 patients underwent an endometrial biopsy and saline sonohysterography as outpatients, 113 patients underwent a definitive surgical intervention (hysteroscopy/dilatation and curettage or hysterectomy), 20 patients did not complete a gold standard measure, and 10 patients were lost to follow-up. RESULTS: The combination of endometrial biopsy and saline sonohysterography for the 113 patients who completed the study had a sensitivity and specificity for the detection of abnormal pathologic features of 97.0% (95% CI, 88.6-99.5) and 70.2% (95% CI, 55.0-82.2) and a positive and negative predictive value of 82.1% (95% CI, 71.4-89.5) and 94.3% (95% CI, 79.6-99.0) compared with hysteroscopy/curettage or hysterectomy. CONCLUSION: The high sensitivity and high negative predictive value of saline sonohysterography combined with endometrial biopsy make this technique useful for the evaluation of abnormal uterine bleeding. It may allow some patients to avoid more invasive operative procedures; however, it is important to recognize the limitations in the predictive value of this diagnostic modality. (Am J Obstet Gynecol 2002;186:858-60.)
Key words: Abnormal uterine bleeding, ultrasound, saline sonohysterography, hysteroscopy
The high prevalence of abnormal uterine bleeding makes this problem a common complaint in the outpatient clinical setting.1 When conservative measures, such as oral contraceptives and other progestational agents, fail to treat this problem, an anatomic cause for the patient’s abnormal uterine bleeding should be considered.2 The gold standard for the evaluation of abnormal uterine bleeding is hysteroscopy with curettage (D&C).3 An endometrial biopsy in combination with saline sonohysterography, which is performed as an office procedure, is a simple, less-invasive, and less-expensive technique for the evaluation of this problem.4 The accuracy of this procedure has yet to be determined with precision.
From the Departments of Obstetrics/Gynecology and Health Evaluation Sciences, University of Virginia. Supported by a grant from the Research and Development Committee at the Health Sciences Center, University of Virginia. Reprint requests: Lillian M. Mihm, MD, Department of Obstetrics/Gynecology, Jefferson Park Ave, Room 3572, University of Virginia, Charlottesville, VA 22908. E-mail:
[email protected] © 2002, Mosby, Inc. All rights reserved. 0002-9378/2002 $35.00 + 0 6/1/123056 doi:10.1067/mob.2002.123056
858
This study proposes to determine the accuracy of endometrial biopsy/saline sonohysterography compared with the gold standard, hysteroscopy/D&C, or hysterectomy in a prospective, blinded trial. Material and methods From October 1998 through November 2000, 144 women aged 25 to 69 years participated in the study. All patients had persistent abnormal uterine bleeding, despite at least 3 months of medical treatment. Informed consent was obtained with a protocol approved by the University of Virginia Human Investigation Committee. An endometrial biopsy and saline sonohysterography were performed and followed by 1 of 3 surgical interventions: hysteroscopy/D&C, hysteroscopy/endometrial ablation, or hysterectomy, determined by the patient’s primary physician. Each patient underwent an endometrial biopsy approximately 4 to 6 weeks before a saline sonohysterography. Both procedures were performed by resident house staff under clinic attending supervision. Ideally, the ultrasound scan was scheduled soon after menstruation; however, most patterns of bleeding were so irregular that
Mihm et al 859
Volume 186, Number 5 Am J Obstet Gynecol
Table. Diagnosis made by operation* Positive Surgical pathologic features Benign endometrial polyp Malignant endometrial polyp Benign submucosal myoma Malignant submucosal myoma Endometrial hyperplasia Normal intrauterine cavity/surgical pathologic features Total
Negative
No.
%
37 2 22 1 4
32.7 1.8 19.5 0.9 3.5
66
No.
%
47
41.6
47
*Diagnoses were based on findings from hysteroscopy, D&C, or hysterectomy specimens.
timing was difficult. The saline sonohysterography was performed, and sterile saline solution was used to dilate the uterine cavity. An open-sided vaginal speculum was used to visualize the cervix, which was prepared with povidone-iodine (Betadine) before placement of a hysterosalpingogram (H/S) inflatable balloon catheter (Ackrad Laboratories, Inc, Cranford, NJ). Approximately 2 mL of sterile water was used to inflate the balloon catheter to confirm endocervical placement. The speculum was removed, and the transvaginal ultrasound probe was introduced into the vagina and followed by the distention of the uterine cavity with 10 to 20 mL sterile saline solution. Subsequently, an appropriate operative procedure was chosen and planned with the surgeon, who was blinded only to the results of saline sonohysterography. Afterward, operative reports were reviewed to determine the intraoperative findings on hysteroscopy, and pathologic reports were obtained to determine the results from the D&C or hysterectomy specimens. The operative findings and pathologic results were then compared with the findings of the endometrial biopsy and saline sonohysterography. We defined normal surgical pathologic findings as secretory, proliferative, or atrophic endometrium. Abnormal surgical pathologic findings included simple or complex hyperplasia, atypia, or malignancy. We described normal saline sonohysterography findings as a smooth, sharply contoured cavity and uniform-appearing endometrium. Abnormal sonohysterography findings included polyps, submucosal fibroids, or irregularly, thickened endometrium.4 Data from 113 patients who completed surgical interventions were analyzed for sensitivity, specificity, and positive and negative predictive values of the endometrial biopsy and saline sonohysterography compared with the actual surgical findings or pathologic results. A 95% CI was estimated.5 Results Of the 144 patients who were enrolled, 113 patients completed the study; 10 patients were lost to follow-up
Figure. Results from surgery (n = 113). EM Bx, Endometrial biopsy; PPV, positive predicted value; NPV, negative predicted value.
after initial evaluation with biopsy and saline sonohysterography, and 1 patient did not complete the saline sonohysterography because of discomfort. The remaining 20 patients did not proceed with a gold standard measure. The actual intraoperative findings are described in the Table, with associated statistical analysis in the Figure. There were no procedural complications. Comment Our study is 1 of the largest prospective, blinded trials to address the accuracy of saline sonohysterography and endometrial biopsy in the diagnosis of endometrial pathologic features. This combination of testing has excellent sensitivity (few patients with disease were missed) and excellent predictive value (a negative test indicated a very low likelihood of disease). Williams and Marshburn2 determined a sensitivity of 100%, a specificity of 85%, a positive predictive value of 75%, and a negative predictive value of 100% when they compared the diagnostic accuracy of transvaginal ultrasound scanning and hydrosonography with office hysteroscopy. These results were based on a total of 39 patients. Widrich et al6 compared saline sonohysterography with flexible office hysteroscopy in 113 patients and determined a sensitivity of 96% and specificity of 88%, concluding the accuracy of this technique in the evaluation of abnormal uterine bleeding. Only 28% of gynecologists in the United States perform office hysteroscopy. We found a lower specificity and positive predictive value in our study compared with the findings of Williams and Marshburn2 and Widrich et al.6 The reason for this is unclear but may be explained, in part, by the difficulty in differentiating anatomic abnormalities from other intrauterine irregularities. Our false-positive findings included intrauterine debris, blood clots, thickened endometrial folds, and fragments of endometrium that
860 Mihm et al
were mistakenly identified as abnormal anatomic structures on sonohysterography. In some instances, patients may undergo unnecessary surgical procedures in the face of an abnormal saline sonohysterography. The high sensitivity and negative predictive value of saline sonohysterography and endometrial biopsy make it a good predictor of the necessity and type of further surgical intervention. In the face of a negative saline sonohysterography and a normal endometrial biopsy, one may opt to continue with conservative medical treatment, knowing that the yield of an operative procedure is likely to be low. The preoperative evaluation with saline sonohysterography may assist the surgeon in triaging patients for the best surgical treatment (such as D&C, endometrial ablation, or a more extensive resectoscope procedure). For example, if a surgeon suspects a larger submucosal myoma that extends into the endometrial cavity, the surgeon may plan for hysteroscopic resection with a longer operative time or for an inpatient procedure, such as hysterectomy. Furthermore, saline sonohysterography has been found to be a useful diagnostic adjunct for infertility patients in the evaluation of intrauterine cavities for assisted reproductive technology.7 We recognize certain limitations of our investigation: patients were not stratified by menopausal status, and treatment decisions may differ based on that fact. In addition, patients were not randomized to the type of surgical intervention used. Future studies would assist clinicians in determining the least invasive and most cost efficient procedure for individual patients.
May 2002 Am J Obstet Gynecol
On the basis of this investigation, we recommend the use of saline sonohysterography before operation to determine the best surgical means of treating anatomic causes for abnormal uterine bleeding. Furthermore, we believe that, after malignancy is excluded, the excellent sensitivity of saline sonohysterography allows for continued conservative treatment and the possible avoidance of unnecessary surgical procedures. We thank Amir Jazaeri, MD, for his efforts in initiating this study and Joann Pinkerton, MD, and Kathie Hullfish, MD, for their dedication in editing this manuscript. REFERENCES
1. Goldstein SR. Use of ultrasonohysterography for triage of perimenopausal patients with unexplained uterine bleeding. Am J Obstet Gynecol 1994;170:565-70. 2. Williams CD, Marshburn PB. A prospective study of transvaginal hydrosonography in the evaluation of abnormal uterine bleeding. Am J Obstet Gynecol 1998;179:292-8. 3. Gimpelson R, Roppold HA. Comparative study between panoramic hysteroscopy with directed biopsies and dilatation and curettage. Am J Obstet Gynecol 1988;158:489-94. 4. Parsons AK, Lense JJ. Sonohysterography for endometrial abnormalities: preliminary results. J Clin Ultrasound 1993;21: 87-95. 5. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. New York: John Wiley; 1981. 6. Widrich, T, Bradley LD, Mitchinson AR, Collins RL. Comparison of saline infusion sonography with office hysteroscopy for the evaluation of the endometrium. Am J Obstet Gynecol 1996; 174:1327-34. 7. Syrop CH, Sahakian V. Transvaginal sonographic detection of endometrial polyps with fluid contrast augmentation. Obstet Gynecol 1992;79:1041-3.