Patient-reported menstrual and obstetric outcomes following hysteroscopic adhesiolysis for Asherman syndrome.

Patients who underwent hysteroscopy with lysis of intrauterine adhesions at the Center for Minimally Invasive Gynecologic Surgery at Newton Wellesley Hospital from January 1, 2015 to March 1, 2019 by one of the three minimally invasive gynecologic surgical specialists were identified. Patients were identified through our institution’s electronic medical records via the research patient data registry (RPDR) using the diagnosis code for Asherman syndrome, N85.6 Intrauterine synechiae (2018 ICD-10-CM Diagnosis Code), and the procedure codes for hysteroscopy, CPT Code 58555 (Hysteroscopy, diagnostic), and/or CPT Code 58559 (Hysteroscopy, with lysis of intrauterine adhesions). For validation, all outpatient records from the Department of Minimally Invasive Gynecologic Surgery also were reviewed and checked with the list produced by the RPDR search of electronic medical records to ensure no patients were missed for possible inclusion during this defined timeframe. Evaluation of distance traveled per patient was calculated to understand the efforts patients went through to seek specialized gynecologic care by using the web mapping service Google Maps (Google, Menlo Park, CA) calculating the distance from the patients documented hometown to our institution’s location in Newton, Massachusetts. Patients underwent a transvaginal pelvic ultrasound and a 5-mm (outer diameter) continuous flow therapeutic office hysteroscopy during their initial patient encounter. Intrauterine adhesions were lysed under direct visualization utilizing 5-French hysteroscopic scissors alone until normal uterine cavity anatomy was restored by 1 of 3 gynecologic surgical providers. The vast majority of these procedures (89.3% of hysteroscopic procedures) were performed in the outpatient office setting, with the operating room reserved only for patients unable to tolerate an office procedure (10.7% of hysteroscopic procedures) during the initial attempt. Hysteroscopic entry into the uterine cavity was performed using the vaginoscopic technique without any parenteral or local anesthesia. All findings and a March classification of the disease severity were documented. Patients were then started on oral estradiol 2 mg twice daily for 30 days, followed by medroxyprogesterone acetate 10 mg daily for the last 5 days of this regimen to induce a withdrawal bleed for only one cycle. Patients were then seen between 2 and 3 weeks postoperatively for repeat adhesiolysis, and then again at 6 weeks postoperatively for repeat adhesiolysis if warranted, with the goal of complete restoration of the normal uterine cavity anatomy. A retrospective review of the electronic health records identified patient’s perioperative characteristics. Patients were classified via the March classification system ( Supplemental Fig. 1 , available online) as has been the traditional practice of our clinic. Given the low incidence of any chronic medical conditions among our patient cohort, we created a dichotomous variable called a “chronic medical condition.” Patients were categorized as having a “chronic medical condition” if they had ≥1 of any of the following chronic medical conditions: obesity, chronic hypertension, diabetes mellitus, thyroid disease, polycystic ovarian syndrome, asthma, smoker, chronic obstructive pulmonary disease, or inflammatory bowel disease. Patients were contacted via telephone and invited to complete a scripted telephone survey. Verbal consent for participation was obtained, and answers were recorded in a secure electronic database, REDCap (research electronic data capture). This study was approved by the Institutional Review Board (IRB) at Newton Wellesley Hospital via the Partners Human Research Committee, the IRB of Partners HealthCare Protocol 2018P002095, obtained on February 19, 2019. To assess menstrual outcomes following hysteroscopic lysis of adhesions for Asherman syndrome, presenting menstrual pattern variables were compared with follow-up menstrual patterns from the telephone survey. To evaluate more accurately the return of menses, we specifically looked at patients that presented with amenorrhea and their follow-up menstrual patterns. The three different patient-reported obstetric outcomes of the study were number of pregnancies, number of miscarriages, and number of live births, using data from the telephone survey. Each of the variables was used to construct dichotomous outcomes for pregnancy rate, miscarriage rate, and live birth rate. Patients were either classified as having 0 or ≥1 pregnancies, 0 or ≥1 miscarriages, and 0 or ≥1 live births. For patient-reported obstetric outcomes, pregnancy was defined as any positive urine or serum evidence of pregnancy via human chorionic gonadotropin or any evidence of intrauterine pregnancy via abdominal or pelvic ultrasound. Miscarriage was defined as any pregnancy loss at <24 weeks’ gestational age, excluding termination of pregnancy or ectopic pregnancy. Live birth was defined as any birth at ≥24 weeks’ gestational age. ART utilization was defined as either utilization of intrauterine insemination or in vitro fertilization. Descriptive statistics were used to summarize the characteristics of the total clinic population and patients that completed the telephone survey. Bivariate analyses were performed to examine sample differences across March classification and telephone survey completion using the ANOVA F -test and unpaired t tests for continuous variables and the χ 2 test for categorical variables. Lastly, multivariable logistic regression analyses were performed to examine whether the severity of Asherman syndrome was an independent indicator of obstetric outcomes after controlling for patient characteristics and medical and obstetric or gynecological history. Due to the small sample size, we only focused on indicators of obstetric outcomes. In addition, we included all patient characteristics regardless of their statistical significance to ensure that observed associations were not confounded by these variables. All analyses were performed using the statistical software package SAS, version 9.4 (Cary, NC).

A total of 355 patients were evaluated and treated for Asherman syndrome within the clinic during the study period. In total, 150 (42.3%) were successfully contacted and completed the telephone survey ( Supplemental Fig. 2 , available online). These patients presented from a total of 41 different states within the United States and from 6 different countries; 233 patients (65.6%) presented from outside our institution’s home state of Massachusetts, and 6 patients (1.7%) presented from outside of the United States ( Supplemental Fig. 3 , available online). The mean distance traveled per patient was 571.3 ± 849.1 miles (median, 205.0 miles). Of the 150 patients contacted, 61 patients (40.7%) had mild disease, 79 (52.7%) had moderate disease, and 10 (6.6%) had severe disease ( Table 1 ). The mean patient age was 35.2 years, and the mean gravidity was 2.1 pregnancies. Patients with moderate disease patients had the lowest mean patient age at 34.2 years ( P <.05). The most common indications for the evaluation and treatment of Asherman syndrome were infertility and menstrual irregularities at 68.7% and 19.3%, respectively, for all classification groups. There were no significant differences in the presenting menstrual pattern when evaluating patients by their classification group, with a total of 38 (25.3%) of the 150 patients initially presenting with amenorrhea. “D&C/D&E–Early Pregnancy Loss or Elective Termination” was identified as being the most common presumed etiology for both mild and moderate Asherman syndrome at 52.5% and 45.6%, respectively, whereas “D&C/D&E–Postpartum” accounted for the most common presumed etiology for severe Asherman syndrome at 30.0% ( P <.001). There was no difference among the classification groups in the follow-up period calculated as the time from initial patient encounter to the date of the telephone survey, with a mean follow-up period of 807.6 days, or 2.21 years. Of the patients contacted, 127 (84.7%) were attempting conception, with a significantly lower proportion of patients with severe disease attempting conception compared with patients with mild and moderate disease ( P = .045). Among those patients attempting conception, there was no statistical difference among utilization of ART when stratified by classification group, with 51.9% utilizing ART while attempting conception. Table 1 Patient characteristics (completed telephone survey) by March classification. March Classification Total (n = 150) P value Mild (n = 61) Moderate (n = 79) Severe (n = 10) Age a 36.2 (5.3) 34.2 (4.2) 36.0 (4.3) 35.2 (4.8) <.05 Gravidity a 1.8 (1.4) 2.2 (1.9) 2.8 (2.0) 2.1 (1.8) .14 Parity a 0.6 (0.9) 1.0 (1.2) 1.9 (2.4) 0.9 (1.2) <.05 Chief Complaint b .632  Infertility 43 (70.5%) 52 (65.8%) 8 (80.0%) 103 (68.7%)  Recurrent Pregnancy Loss 3 (4.9%) 5 (6.3%) 0 8 (5.3%)  Menstrual Irregularity 12 (19.7%) 16 (20.3%) 1 (10.0%) 29 (19.3%)  Dysmenorrhea 1 (1.6%) 1 (1.3%) 1 (10.0%) 3 (2.0%)  Noncyclic pelvic pain 2(3.3%) 5 (6.3%) 0 7 (4.7%) Presenting Menstrual Pattern b .29  Normal 23 (37.7%) 22 (27.9%) 1 (10.0%) 46 (30.7%)  Light 19 (31.2%) 35 (44.3%) 7 (70.0%) 61 (40.7%)  Absent 16 (26.2%) 20 (25.3%) 2 (20.0%) 38 (25.3%)  Heavy 3 (4.9%) 2 (2.5%) 0 5 (3.3%) Presumed Etiology b <.001  D&C/D&E: Early Pregnancy Loss or Elective Termination 32 (52.5%) 36 (45.6%) 2 (20.0%) 70 (46.7%)  D&C/D&E: Postpartum 15 (24.6%) 24 (30.4%) 3 (30.0%) 42 (28.0%)  Hysteroscopic Polypectomy 3 (4.9%) 1 (1.3%) 0 4 (2.7%)  Hysteroscopic Metroplasty 1 (1.6%) 0 0 1 (0.7%)  Hysteroscopic Myomectomy 1 (1.6%) 1 (1.3%) 1 (10.0%) 3 (2.0%)  Laparoscopic Myomectomy 1 (1.6%) 1 (1.3%) 0 2 (1.3%)  Abdominal Myomectomy 3 (4.9%) 3 (3.8%) 1 (10.0%) 7 (4.7%)  Cesarean Section 1 (1.6%) 10 (12.7%) 1 (10.0%) 12 (8.0%)  Endometrial Ablation 0 1 (1.3%) 2 (20.0%) 3 (2.0%)  Unclear 4 (6.6%) 2 (2.5%) 0 6 (4.0%) Medical History b .55  Yes 9 (14.8%) 8 (10.1%) 2 (20.0%) 19 (12.7%)  No 52 (85.3%) 71 (89.9%) 8 (80.0%) 131 (87.3%) Previous Miscarriages a 1.0 (1.0) 1.1 (1.3) 0.6 (0.7) 1.0 (1.1) .39 D&C/D&E–Early Pregnancy Loss or Elective Termination a 0.8 (0.8) 0.8 (1.0) 0.6 (1.0) 0.8 (0.9) .75 D&C/D&E–Postpartum a 0.3 (0.6) 0.3 (0.5) 0.3 (0.5) 0.3 (0.5) .96 Follow-Up Period (Days) a 769.7 (439.0) 815.9 (428.3) 974.2 (666.0) 807.6 (455.0) .888 Attempting Conception a 55 (90.2%) 66 (83.5%) 6 (60.0%) 127 (84.7%) .045 ART Utilization a 29 (52.7%) 35 (53.0%) 2 (33.3%) 66 (51.9%) .269 Note : Percentages may not add to 100% because of rounding. ART = assisted reproductive technology; D&C = dilation and curettage; D&E = dilation and evacuation. a F -test (ANOVA), all groups based on March classification system were compared (mild, moderate, and severe). b χ 2 test, all groups based on March classification system were compared (mild, moderate, and severe). Patient characteristics (completed telephone survey) by March classification. Note : Percentages may not add to 100% because of rounding. ART = assisted reproductive technology; D&C = dilation and curettage; D&E = dilation and evacuation. F -test (ANOVA), all groups based on March classification system were compared (mild, moderate, and severe). χ 2 test, all groups based on March classification system were compared (mild, moderate, and severe). The only significant difference in patient characteristics between those who completed the telephone survey and those who did not complete it was a higher percentage of patients presenting with infertility (68.7% vs. 62.9%) as their primary chief complaint among those that completed the telephone survey ( Supplemental Table 1 , available online). Patients who completed the telephone survey were thus overall representative of the entire clinic population. The most common presenting menstrual pattern for the mild disease was normal flow at 37.7%; for moderate disease, light flow at 44.3%; and for severe disease, light flow at 70.0%. There was no difference in the rate of amenorrhea when stratified by classification group ( Table 1 ). Upon follow-up examination of all patients who initially presented with amenorrhea (38 patients), those with mild disease had a significantly higher rate of returning menstruation (93.8%; P <.05) ( Table 2 ). Table 2 Follow-up menstrual pattern for patients who presented with amenorrhea by March classification. Absent Presenting Menstrual Pattern March Classification P value N = 38 patients Follow-Up Menstrual Pattern Mild Moderate Severe  Normal 8 (50.0) 12 (60.0) 0 (0.0) <.05  Light 7 (43.8) 1 (5.0) 1 (50.0)  Absent 1 (6.3) 3 (15.0) 1 (50.0)  Heavy 0 (0.0) 4 (20.0) 0 (0.0) Note : Values are expressed as the number (percentage) of patients. Follow-up menstrual pattern for patients who presented with amenorrhea by March classification. Note : Values are expressed as the number (percentage) of patients. Among the 127 patients who attempted conception, 43.3% had mild, 52.0% had moderate, and 4.7% had severe Asherman syndrome ( Table 3 ). Of the 127 patients, 104 reported ≥1 pregnancy following hysteroscopic treatment at our institution, for a 81.9% cumulative pregnancy rate. Although not statistically significant, there was a decreasing trend in pregnancy rate with increasing severity ( P = .47). The miscarriage rate was highest in the severe Asherman syndrome patient group ( P <.05). Sixty-five patients reported ≥1 live births, for a 51.2% cumulative live birth rate. There was no statistically significant difference in live birth rate when stratified by classification group ( P = .21). Additionally, 22 patients were actively pregnant at the time of the telephone survey. Of these patients, 6 had already reported ≥1 previous live birth since hysteroscopic treatment for their Asherman syndrome at our institution, whereas 16 patients had reported no previous live births since hysteroscopic treatment for their Asherman syndrome at our institution. Table 3 Patient-reported pregnancy outcomes. Patient Outcomes Mild Moderate Severe Total P value All Patients Attempting Conception (N = 127) (n = 55) (n = 66) (n = 6) (n = 127)  ≥1 Pregnancy 47 (85.5) 53 (80.3) 4 (66.7) 104 (81.9) .47  ≥1 Miscarriage 20 (36.4) 12 (18.2) 3 (50.0%) 35 (27.6) <.05  ≥1 Live Birth 28 (50.9) 36 (54.6) 1 (16.7%) 65 (51.2) .21 Pregnancy Outcomes Mild Moderate Severe Total All Pregnancies (N = 149) (n = 70) (n = 72) (n = 7) (n = 149)  Preterm & Full-Term Births 29 (41.4) 39 (54.2) 1 (14.3%) 69 (46.3)  SAB/TAB/Ectopic 32 (45.7) 21 (29.2) 5 (71.4%) 58 (38.9)  Active Pregnancies 9 (12.9) 12 (16.7) 1 (14.3%) 22 (14.8) Trimester of Active Pregnancies (N = 22) Mild (n = 9) Moderate (n = 12) Severe (n = 1) Total (n = 22)  First 5 (55.6) 3 (25.0) 0 (0.0%) 8 (36.4)  Second 4 (44.4) 4 (33.3) 1 (100.0%) 9 (40.9)  Third 0 (0.0) 5 (41.7) 0 (0.0%) 5 (22.7) Note : Values are expressed as the number (percentage) of patients. SAB = spontaneous abortion; TAB = therapeutic abortion. Patient-reported pregnancy outcomes. Note : Values are expressed as the number (percentage) of patients. SAB = spontaneous abortion; TAB = therapeutic abortion. A total of 149 pregnancies were reported among the 104 patients who had reported ≥1 pregnancy after hysteroscopic treatment at our institution ( Table 3 ). Sixty-nine (46.3%) pregnancies were “Preterm & Full-Term Births,” 58 (38.9%) were “SAB/TAB/Ectopic,” and 22 (14.8%) were “Active Pregnancies.” “Preterm & Full-Term Births” referred to any pregnancy resulting in delivery at ≥24 weeks’ gestational age. “SAB/TAB/Ectopic” referred to any pregnancy resulting in a pregnancy loss at <24 weeks’ gestational age, termination of pregnancy at any gestational age, or an ectopic pregnancy at any gestational age. Complete survey results are available in Supplemental Table 2 (available online). We assessed Asherman syndrome disease severity as an independent risk factor for pregnancy outcomes using multivariable analysis to control for several potential confounding variables, such as patient age, gravidity, parity, presenting menstrual pattern, chronic medical condition, previous miscarriages, previous uterine instrumentation, and use of in vitro fertilization ( Table 4 ). Asherman syndrome disease severity was not a predictor for ≥1 pregnancy or ≥1 live births when adjusted for potential confounders. It was a predictor of ≥1 miscarriage, and patients with moderate Asherman syndrome demonstrated the lowest rate of miscarriages ( P <.05). Table 4 Multivariable analysis for patient-reported pregnancy outcomes. Characteristic Patient-Reported Pregnancy Outcomes Having ≥1 pregnancies Having ≥1 miscarriages Having ≥1 live births Adjusted OR (95% CI) P value Adjusted OR (95% CI) P value Adjusted OR (95% CI) P value March Classification  Mild Ref. - Ref. - Ref. -  Moderate 0.3 (0.1, 1.1) .06 0.3 (0.1, 0.8) <.05 1.0 (0.4, 2.1) .95  Severe 0.3 (0.02, 3.4) .31 1.8 (0.3, 11.2) .51 0.1 (0.01,1.2) .06 Age 0.8 (0.6, 0.9) <.001 1.0 (0.9, 1.1) .69 0.9 (0.8, 0.9) <.05 Gravidity 1.4 (0.3, 7.7) .71 1.5 (0.6, 3.6) .38 1.1 (0.4, 3.0) .71 Parity 0.6 (0.1, 3.7) .59 0.6 (0.2, 1.6) .33 1.2 (0.5, 3.2) .84 Presenting Menstrual Pattern  Normal Ref. - Ref. - Ref. -  Light 2.3 (0.6, 8.6) .21 1.7 (0.6, 4.7) .31 0.6 (0.2, 1.5) .19  Absent 1.2 (0.2, 6.3) .8 1.4 (0.4, 4.8) .6 0.7 (0.2, 1.9) .28  Heavy 3.2 (0.2, 65.6) .45 2.2 (0.3, 17.0) .47 0.5 (0.1, 4.3) .58 Chronic Medical Condition 0.6 (0.1, 2.6) .48 0.8 (0.2, 3.1) .79 0.7 (0.2, 2.4) .63 Previous Miscarriages (Reported at time of clinic) 0.7 (0.1, 3.9) .67 1.0 (0.4, 2.4) .98 0.6 (0.3, 1.6) .4 D&C/D&E–Early Pregnancy Loss or Elective Termination 3.2 (1.0, 10.8) .06 0.7 (0.4, 1.5) .4 1.8 (0.9, 3.6) .12 D&C/D&E–Postpartum 1.4 (0.3, 6.3) .67 2.2 (0.7, 6.7) .15 0.7 (0.3, 1.9) .41 ART Utilization 0.7 (0.2, 2.4) .6 1.3 (0.5, 3.4) .54 0.6 (0.2, 1.3) .19 Note : ART = assisted reproductive technology; D&C = dilation and curettage; D&E = dilation and evacuation. Multivariable analysis for patient-reported pregnancy outcomes. Note : ART = assisted reproductive technology; D&C = dilation and curettage; D&E = dilation and evacuation.

We characterized the menstrual pattern and obstetric outcomes following hysteroscopic adhesiolysis among Asherman syndrome patients when stratified by disease severity. We discovered that disease severity predicted returning menstruation but did not accurately predict pregnancy rate or live birth rate following treatment at a mean follow-up period of 2.21 years. The rate of amenorrhea reported among all patients with Asherman syndrome presenting initially for care has been described from 0.0%–100.0% ( 24 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 ), with only one paper reporting amenorrhea rate by disease severity via the American Fertility Society classification at 0%, 2.6%, and 32.1% for mild, moderate, and severe disease, respectively ( 30 ). Resolution of amenorrhea following treatment has been reported from 29.0%–100.0% ( 7 , 23 , 24 , 28 , 29 , 30 , 32 , 33 , 34 , 35 , 36 , 39 , 40 , 41 , 42 , 43 ). In our study, there was no significant difference in patients presenting with amenorrhea when stratified by March classification. This is potentially because the menstrual history is not incorporated within the March classification system, unlike the American Fertility Society classification system, which does give significant weight to patient-reported amenorrhea when calculating the disease severity. Thus, with the March classification, it is possible for two patients both to report amenorrhea at the initial consultation, even if one patient has mild disease with only focal adhesions forming an outlet obstruction and the other patient has severe disease with a significant percentage of adhesions with little functioning endometrium. Our study validates this concept, as we identified a much higher rate of return of menses after hysteroscopic lysis of adhesions among patients with mild disease compared with those with severe disease. The majority of our patients present with a chief complaint of infertility, making it essential to understand obstetric outcomes after hysteroscopic adhesiolysis. The pregnancy rate among all Asherman syndrome patients attempting conception after hysteroscopic treatment varies in the reported literature from 32.1%–85.0%, with a cumulative pregnancy rate of 56.2% (1467/2609 patients) ( Supplemental Table 3 , available online), although the definition of pregnancy is absent in a many of these studies ( 7 , 23 , 24 , 28 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 39 , 40 , 41 , 42 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 ). Additionally, the definition of live birth rate varies among the published literature, but when defined as the total number of patients with ≥1 live birth divided by the total number of patients attempting conception following treatment of Asherman syndrome, the live birth rate ranges from 14.3% to 78.0% with a cumulative rate of 36.8% (960/2609 patients) ( Supplemental Table 3 , available online). In our practice, both cumulative pregnancy rate and live birth rate were higher than the cumulative averages calculated from our literature review. We identified no significant difference in the pregnancy rate with increasing disease severity. We hypothesize this may be due to two factors: the similar pregnancy rates among both the mild and moderate Asherman syndrome patients (85.5% vs. 80.3%, respectively) and the low number of patients in the cohort completing the telephone survey with severe disease. The multivariable analysis performed demonstrated that the disease severity was not able to predict pregnancy rate effectively, even after accounting for confounding factors. We identified a significant decrease in miscarriages among patients with moderate Asherman syndrome compared with those with mild Asherman syndrome after hysteroscopic adhesiolysis. This may be due to an increase in underlying conditions (i.e., old age, chronic medical conditions, endometriosis, oocyte quality) that may exist with a greater incidence among patients with mild Asherman syndrome experiencing these first trimester miscarriages. Among completed telephone survey patients, there was both a higher mean age (36.2 vs. 34.2 years old) and a higher unadjusted rate of chronic medical conditions (14.8% vs. 10.1%) among patients with mild Asherman syndrome compared with those with moderate Asherman syndrome; however, specific chronic medical conditions and their impact on miscarriages were not subanalyzed in this study design. We also demonstrated no statistically significant difference in live birth rate when stratified by disease severity; again, we believe this is because of the similar live birth rates between patients with mild and moderate Asherman syndrome (50.9% vs. 54.6%, respectively) and the low number of patients with severe Asherman attempting conception. Although not statistically significant, patients with severe Asherman syndrome did have the lowest pregnancy rate and live birth rates at 66.7% and 16.7%, respectively. We encourage further research into identifying additional patient characteristics that may impact the menstrual and obstetric outcomes among patients with Asherman syndrome, so we may better guide prognostic counseling and follow-up plans after hysteroscopic adhesiolysis. The strengths of our study include the large cohort of patients with Asherman syndrome treated within the United States. We are the first group to investigate the impact of disease severity via the March classification system on menstrual and obstetric outcomes while utilizing multivariable analysis to investigate the impact of confounding variables. Some limitations of this study include its retrospective nature, with only 42.3% of all clinic patients completing the telephone survey leaving room for possible sampling error with our results to be interpreted with this caution. Additionally, as we serve largely as a referral center for specialized gynecologic surgical intervention, with patients returning to their original providers (often remote from our location, as described in the Results) for complete fertility management, we were unable to assess in our analysis other factors that may impact fertility such as tubal patency and semen analysis. Lastly, we assessed menstrual outcomes using subjective questions and without quantitative measurements of menstrual bleeding. In conclusion, our study demonstrated that overall menstrual and obstetric outcomes are very promising after hysteroscopic management of Asherman syndrome. However, the development of a more comprehensive prognostic tool is necessary to counsel patients better on pregnancy and live birth rates following hysteroscopic adhesiolysis, as disease severity alone is not a strong predictor of pregnancy outcomes.