Advertisement

Ultrasonographic features of the endometrium following successful medical termination of early pregnancy

Open AccessPublished:November 25, 2022DOI:https://doi.org/10.1016/j.ejogrb.2022.11.018

      Highlights

      • Endometrial thickness and echogenicity vary considerably following successful early medical abortion.
      • All possible patterns of change in endometrial thickness and echogenicity may be observed in the weeks following induction.
      • Endometrial features should be considered with caution in the decision of surgical intervention of early medical abortion.

      Abstract

      Objective

      Ultrasonographic features of the endometrium are often assessed when deciding the necessity of surgical intervention following early medical abortion. Knowledge is therefore needed on the ultrasonographic appearance of the endometrium following successful medical abortion in order to avoid unnecessary surgical interventions. We aimed to assess endometrial thickness and echogenicity at multiple time points following successful early medical abortion.

      Study design

      We conducted a retrospective study in the largest office-based abortion providing clinic in Denmark. Using archived ultrasonographic images, we assessed endometrial thickness and echogenicity following all early medical abortions that did not need surgical intervention or repeated medication for completion during the years 2014–2017.

      Results

      Ultrasonographic endometrial features were assessed 1854 times following 1074 early medical abortions. Median endometrial thickness in the 1st week from induction was 13 milimeters (mm; lower-upper quartile 11–17 mm). For the 2nd, 3rd, 4th, and >4th week, the median endometrial thickness was found to be 11 mm (9–15 mm), 11 mm (8–14 mm), 12 mm (9–16 mm), and 11 mm (8–14 mm), respectively.
      Of the ultrasonographic examinations performed in the 1st week from medical induction, 24.7 % showed a heterogenous endometrium. For 2nd, 3rd, 4th, >4th week, the frequency of heterogeneity was 23.9 %, 16.3 %, 21.3 %, 18.9 %, respectively.
      A total of 151 abortions (14.1 %) were each examined three times, median time of examination being day 7, 15, and 26 following induction. Among these abortions, the three most common patterns of change in endometrial thickness were “decreasing” (37.7 %), “increasing-decreasing” (23.2 %), and “decreasing-increasing” (21.9 %). Further, 49.7 % of the 151 abortions showed a homogenous endometrium at all three examinations, 17.2 % showed a heterogenous endometrium at first examination and a homogenous endometrium the following two examinations, and 9.9 % showed a heterogenous endometrium at the first two examinations followed by a homogenous endometrium.

      Conclusion

      In early medical abortions completed without secondary intervention, endometrial thickness and echogenicity varied clinically significantly for weeks following the medical induction. Every possible pattern of change in endometrial thickness and echogenicity was observed.

      Keywords

      Introduction

      Medical abortion in early first trimester has proven to be logistically more practical and potentially less harmful than its surgical counterpart, surgical termination of early pregnancy having been linked to increased risk of infection, Asherman’s syndrome, and preterm delivery [
      • Singh S.
      • Remez L.
      • Sedgh G.
      • Kwok L.
      • Onda T.
      Abortion worldwide 2017: uneven progress and unequal access.
      ,
      • Saccone G.
      • Perriera L.
      • Berghella V.
      Prior uterine evacuation of pregnancy as independent risk factor for preterm birth: a systematic review and metaanalysis.
      ,
      • Mentula M.
      • Männistö J.
      • Gissler M.
      • Heikinheimo O.
      • Niinimäki M.
      Intrauterine adhesions following an induced termination of pregnancy: a nationwide cohort study.
      ,
      • Rørbye C.
      • Nørgaard M.
      • Nilas L.
      Medical versus surgical abortion efficacy, complications and leave of absence compared in a partly randomized study.
      ]. Thus, there are important advantages of reducing unintended surgical interventions of early medical abortions.
      Most surgical interventions of early medical abortions are based on clinical impression and thereby potentially avoidable [
      • Allen R.H.
      • Westhoff C.
      • De Nonno L.
      • Fielding S.L.
      • Schaff E.A.
      Curettage after mifepristone-induced abortion: frequency, timing, and indications.
      ]. The ultrasonographic appearance of the endometrium following medical induction often influences the clinical estimation of the necessity of surgical intervention to complete the early medical abortion. Thus, knowledge on the natural appearance of the endometrium following medical termination of pregnancy in early first trimester is important for the prevention of unnecessary surgical interventions. This knowledge, however, is lacking in the literature. Previous studies assessing ultrasonographic features of the endometrium following early medical abortion are sparse, small in sample size, and the majority of studies have only assessed ultrasonographic features of the endometrium at one single time-point following medical induction, thereby not being informative on the natural evolution of the endometrium over time following medical induction [
      • Cowett A.A.
      • Cohen L.S.
      • Lichtenberg E.S.
      • Stika C.S.
      Ultrasound evaluation of the endometrium after medical termination of pregnancy.
      ,
      • Acharya G.
      • Haugen M.
      • Brathen A.
      • Nilsen I.
      • Maltau J.M.
      Role of routine ultrasonography in monitoring the outcome of medical abortion in a clinical setting.
      ,
      • Napolitano R.
      • Ghosh M.
      • Gillott D.J.
      • Ojha K.
      Three-dimensional Doppler sonography in asymptomatic and symptomatic women after medical termination of pregnancy.
      ].
      Aiming to gather descriptive knowledge on the naturally occurring ultrasonographic features of the endometrium following early medical abortion, the objective of this study was to assess endometrial thickness and echogenicity over time following successful medical termination of early pregnancy.

      Materials and methods

      We conducted a descriptive study, retrospectively assessing the thickness and echogenicity of the endometrium following medical termination of singleton pregnancies with a gestational age of <63 days in women aged 15–49 years during 2014–2017 at the Gynecology Clinic Hvidovre, the largest office-based abortion providing site in Denmark during the study period.
      We included all abortions completed with the standard drug regimen of 200 milligrams (mg) mifepristone followed 48 h later by 0.8 mg home-administrated vaginal misoprostol. Thus, we excluded medical abortions that needed subsequent surgical intervention (uterine vacuum aspiration, dilatation and curettage, hysteroscopic excision of tissue) or repetition of medical induction. This information was extracted from patient records and from The Danish National Patient Register to ensure information on surgical interventions done outside the opening hours of the clinic at Danish hospitals (The Nordic Medico-Statistical Committee Classification of Surgical Procedures codes KMBA00, KMBA03, KLCH00, KLCH03, KLCH13, KLCB98, KLCB25, KULC02 given within eight weeks from mifepristone administration) [
      • Lynge E.
      • Sandegaard J.L.
      • Rebolj M.
      The Danish national patient register.
      ]. This national register holds information on all surgical procedures executed at Danish hospitals since the year 1977 [
      • Lynge E.
      • Sandegaard J.L.
      • Rebolj M.
      The Danish national patient register.
      ].
      Basic characteristics of the study population, including gestational age, maternal age, and reproductive history, were extracted from patient records and national health registers (Supplementary Table S1) [
      • Lynge E.
      • Sandegaard J.L.
      • Rebolj M.
      The Danish national patient register.
      ,
      • Tølbøll Blenstrup L.
      • Knudsen L.B.
      Danish registers on aspects of reproduction.
      ,
      • Knudsen L.B.
      • Olsen J.
      The Danish medical birth registry.
      ].
      For all included abortions, endometrial thickness and echogenicity were assessed from archived ultrasonographic images taken during transvaginal ultrasonographic examinations, which were routinely scheduled following medical induction. In general, consultations were scheduled with one-week-intervals to ensure completeness of the abortion, the physical and mental well-being of the woman, and to discuss future use of contraception. Normally, three consultations were scheduled. If the gestational sac was expelled, future contraception method found, and the woman in well-being with no personal need for further consultations, the abortion provider could finalize the course before the third consultation. If the abortion provider was not sure of the completeness of the abortion, or if the woman wished for further consultations, the woman was offered more than the usual three consultations post induction.
      Endometrial thickness was measured as the maximum measurement in the sagittal plane, including both endometrial layers. In accordance with previous publications on the topic, if present, intracavitary lesions such as product of conception, fluid, blood clots, were included in the assessment of the endometrial thickness.
      Endometrial echogenicity was categorized as either homogenous or heterogenous. If the endometrial echogenicity was non-uniform, or if intracavitary lesions were present, the endometrium was considered heterogenous.
      The study was approved by the Danish Data Protection Agency, the Danish Health Data Board, and the Danish Patient Safety Authority.

      Results

      During the four-year-long study period, 1316 singleton-pregnancies were medically terminated before a gestational age of 63 days among women aged 15–49 years at study site. Of these, 1209 (92 %) were successfully terminated with a single mifepristone-misoprostol dose and without surgical intervention within eight weeks from medical induction. However, only 1074 of the 1209 successful early medical abortions (89 %) were included in the study due to missing ultrasonographic images of the uterus post induction of 135 abortions.
      Characteristics of the included abortions are shown in Table 1. The majority of abortions were induced in the 5th-6th gestational week in 20–29-year-old nulliparous women (Table 1). Gestational and maternal age of the excluded abortions did not differ from those of included (p = 0.3 and p = 0.2, respectively).
      Table 1Characteristics of the 1074 included medical abortions.
      Abortions N (column %)
      Gestational age (week)
      5 + 6689 (64.2)
      7214 (19.9)
      8147 (13.7)
      924 (2.2)
      Maternal age (year)
      15–1999 (9.2)
      20–24272 (25.3)
      25–29264 (24.7)
      30–34204 (19)
      35–39157 (14.6)
      40–4472 (6.7)
      45–496 (0.6)
      Previous deliveries
      Nulliparous542 (50.5)
      Only vaginal deliveries421 (39.2)
      ≥1 cesarean section111 (10.3)
      Previous medical abortion331 (30.8)
      Previous surgical abortion251 (23.3)
      The 1074 early medical abortions were induced in 1009 women. A total of 1854 transvaginal ultrasonographic examinations were performed post induction of the 1074 abortions with 473 (44.0 %), 436 (40.6 %), 151 (14.1 %), and 14 (1.3 %) abortions undergoing sonographic examination once, twice, three, and four times following medical induction, respectively.
      Fig. 1 illustrates endometrial thickness according to time from mifepristone administration. A total of 372 (34.6 %) abortions were examined in the first week following mifepristone administration, and the median endometrial thickness (first quantile-third quantile) among these abortions was 13 mm (11–17 mm; Fig. 1). Similarly, among the 660 (61.5 %), 496 (46.2 %), 178 (16.6 %), and 148 (13.7 %) abortions examined in the second, third, fourth, and >fourth week following mifepristone administration, respectively, the median endometrial thickness was 11 mm (9–15 mm), 11 mm (8–14 mm), 12 mm (9–16), and 11 mm (8–14 mm; Fig. 1), respectively. Similar characteristics of the endometrial thickness were observed in the subpopulation of only first-time abortions (women with no history of induced abortions) and when only considering a single image per abortion (from the first examination following induction).
      Figure thumbnail gr1
      Fig. 1Endometrial thickness in millimeters according to time since mifepristone administration. Endometrial thickness illustrated with boxplots (horizontal lines: 25th, 50th, and 75th percentiles; vertical lines: minimum value and maximum value; bullets: outliers; red number: median endometrial thickness; black number: number of observations). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
      Median endometrial thickness increased 1–2 mm for one week increase in gestational age in the first, second, and third week following mifepristone administration.
      Fig. 2 illustrates the different patterns of change in endometrial thickness over time observed per abortion among the 151 abortions with three measurements of endometrial thickness post induction. The median time of the first, second, and third measurement was 7, 15, 26 days following mifepristone administration, respectively. All possible patterns of change in endometrial thickness were observed, with a decrease in endometrial thickness over the three time points being the most frequent pattern observed (37.7 %), followed by the patterns “increasing-decreasing” (23.2 %) and “decreasing-increasing” (21.9 %). These three patterns were the most common in all gestational age groups.
      Figure thumbnail gr2
      Fig. 2Change in endometrial thickness following medical induction per abortion in 151 medical abortions examined three times post induction. Pattern name: First part of the name illustrating the pattern between first and second measurement; Last part of the name illustrating the pattern between second and third measurement. Percentages: Percentages of the 151 abortions with the particular pattern. Day from mifepristone administration: Median day of first, second, and third measurement of the abortions with the particular pattern.
      Of the 1854 ultrasonographic examinations, 397 (21.4 %) visualized a heterogenous endometrium. The proportion of abortions presenting with a heterogenous endometrium was 24.7 %, 23.9 %, 16.3 %, 21.3 %, and 18.9 % in the first, second, third, fourth, and >fourth week following induction. These proportions did not materially change across gestational ages and in a sub-assessment only including the first ultrasonographic examination per abortion.
      Changes in echogenicity over time per abortion is shown in Fig. 3 for the 151 abortions with three observations following medical induction. All types of changes were observed. A total of 49.7 % of the 151 abortions showed a homogenous endometrium at all three examinations done at median day 7, 15, and 26 following mifepristone administration. Second most observed pattern of change in endometrial echogenicity (17.2 %) was a heterogenous endometrium at the first examination followed by a homogenous endometrium at the following two scans. The third most observed pattern of change (9.9 %) was a heterogenous endometrium at the first and second scan followed by a homogenous endometrium at the third scan following induction. These findings persisted when stratified according to gestational age at time of induction, with the exception that the third most common pattern in abortions induced in the 7th or 8th gestational week was a heterogenous endometrium at all three examinations post induction.
      Figure thumbnail gr3
      Fig. 3Change in endometrial echogenicity following medical induction per abortion in 151 medical abortions examined three times post induction. Red triangle: Heterogenous endometrium (non-uniform and/or intracavitary lesion) Green triangle: Homogenous endometrium (uniform and no intracavitary lesion) Day from mifepristone administration: Median day of first, second, and third measurement. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

      Discussion

      This study shows that endometrial thickness and echogenicity may vary considerably following successful medical termination of pregnancy in early first trimester. A thick and heterogenous endometrium was observable for weeks following induction, and every possible pattern of change in endometrial thickness and echogenicity following induction was observed.
      We found no study assessing the pattern of change in endometrial thickness and echogenicity per abortion within the following month from induction.
      Pooling data from two multicenter randomized trials, Reeves et al. (2009) found the median endometrial thickness to be 10 mm (25th–75th percentile: 8–14 mm) five to eleven days following misoprostol administration [
      • Reeves M.F.
      • Fox M.C.
      • Lohr P.A.
      • Creinin M.D.
      Endometrial thickness following medical abortion is not predictive of subsequent surgical intervention.
      ]. We found a similar variation in endometrial thickness at the same time point post induction (median: 13 mm; 25th–75th percentile: 10–16 mm). In a prospective cohort study, Rørbye et al. (2004) measured endometrial thickness 14–16 days after mifepristone administration and found the median to be 10 mm (25th-75th percentile: 7–13) [
      • Rørbye C.
      • Nørgaard M.
      • Nilas L.
      Prediction of late failure after medical abortion from serial β-hCG measurements and ultrasonography.
      ]. Accordingly, we observed a median endometrial thickness of 11 mm (25th–75th percentile: 8–14) 14–16 days following the administration of mifepristone.
      In a prospective cohort study of 36 early medical abortions, Markovitch et al. (2006) observed an intrauterine echogenic mass 14 days post medical induction in 50 % of the successful abortions included [
      • Markovitch O.
      • Tepper R.
      • Klein Z.
      • Fishman A.
      • Aviram R.
      Sonographic appearance of the uterine cavity following administration of mifepristone and misoprostol for termination of pregnancy.
      ]. The smaller proportion of heterogeneity observed in our study could be explained by the higher dose of misoprostol used for induction in the current study [
      • Markovitch O.
      • Tepper R.
      • Klein Z.
      • Fishman A.
      • Aviram R.
      Sonographic appearance of the uterine cavity following administration of mifepristone and misoprostol for termination of pregnancy.
      ]. G. Acharya et al. (2004) reported intracavitary content in 9.4 % of 585 successful early medical abortions two-three weeks following medical induction [
      • Acharya G.
      • Haugen M.
      • Brathen A.
      • Nilsen I.
      • Maltau J.M.
      Role of routine ultrasonography in monitoring the outcome of medical abortion in a clinical setting.
      ]. We found the endometrium to be heterogeneous (non-uniformed echogenicity or existence of intracavitary content) in 24 % and 16 % of abortions examined in the second and third week post induction, respectively.
      Obvious strengths of this study include the size of the study population and the eight weeks of full follow-up of all included abortions, ensuring the inclusion of only abortions completed without surgical intervention or repetition of medical induction. Furthermore, to our knowledge, this is the first study assessing the ultrasonographic features of the endometrium at multiple time points following medical termination of early pregnancy, allowing us to report findings on changes in endometrial thickness and echogenicity over time. This is important, since clinicians are recommended to practice watchful waiting if in doubt of the medical necessity of a surgical intervention. This promotes additional examinations and thereby increases the risk of clinicians assessing changes in endometrial thickness and echogenicity and letting this assessment influence the decision to surgically intervene.
      Furthermore, to our knowledge, this is the first study reporting on endometrial features >3 weeks post induction (as many as 16.6 % and 13.7 % of the included successful medical abortions were examined four and >four weeks following induction). This is highly relevant, since most surgical interventions on early medical abortions occur 3–5 weeks following induction [
      • Allen R.H.
      • Westhoff C.
      • De Nonno L.
      • Fielding S.L.
      • Schaff E.A.
      Curettage after mifepristone-induced abortion: frequency, timing, and indications.
      ].
      The exclusion of 11 % of all abortions eligible for inclusion could be viewed as a limitation of the study. However, we were able to characterize the excluded abortions and compare them with the included, and we found no considerable distinction between the two.
      Considering that >3 examinations following medical induction were not routine, but only scheduled when found necessary by the provider or the woman, observations from these examinations may be non-representative (selection bias). However, only 14 out of the 1074 abortions (1 %) were examined >3 times following induction. Furthermore, our main findings persisted when only considering the first ultrasonography per abortion, eliminating the potential selection bias occurring due to previous scans/consultations.
      This study was not designed to compare the endometrial thickness and echogenicity post induction between medical abortions completed with and without surgical intervention/repetition of medication. It is acknowledged that endometrial thickness following medical induction is a poor predictor of the necessity of surgical intervention [
      • Reeves M.F.
      • Fox M.C.
      • Lohr P.A.
      • Creinin M.D.
      Endometrial thickness following medical abortion is not predictive of subsequent surgical intervention.
      ,
      • Rørbye C.
      • Nørgaard M.
      • Nilas L.
      Prediction of late failure after medical abortion from serial β-hCG measurements and ultrasonography.
      ]. Nonetheless, ultrasonography is an essential part of the examinations leading to the decision of secondary intervention of early medical abortions. Thus, knowledge on the changes in endometrial presentation over time following induction in successful medical abortions is crucial for the avoidance of over-treatment with surgical intervention.

      Conclusion

      Our findings contribute to the understanding of the naturally occurring ultrasonographic features of the endometrium post medical induction, emphasizing the considerable variation in endometrial thickness, echogenicity, and the pattern change of these features over time. Despite the descriptive nature of the study and the lack of comparison between successful medical abortions and medical abortions needing further intervention, the observed variation in endometrial appearance among successful medical abortions strongly suggest the importance of being careful when including endometrial ultrasonographic features in the clinical assessment of the necessity of surgical intervention.

      Funding

      The authors received no financial support.

      Declaration of Competing Interest

      The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

      Appendix A. Supplementary data

      The following are the Supplementary data to this article:

      References

        • Singh S.
        • Remez L.
        • Sedgh G.
        • Kwok L.
        • Onda T.
        Abortion worldwide 2017: uneven progress and unequal access.
        Guttmacher Institute, New York2018
        • Saccone G.
        • Perriera L.
        • Berghella V.
        Prior uterine evacuation of pregnancy as independent risk factor for preterm birth: a systematic review and metaanalysis.
        Am J Obstet Gynecol. 2016; 214: 572-591
        • Mentula M.
        • Männistö J.
        • Gissler M.
        • Heikinheimo O.
        • Niinimäki M.
        Intrauterine adhesions following an induced termination of pregnancy: a nationwide cohort study.
        BJOG. 2018; 125: 1424-1431
        • Rørbye C.
        • Nørgaard M.
        • Nilas L.
        Medical versus surgical abortion efficacy, complications and leave of absence compared in a partly randomized study.
        Contraception. 2004; 70: 393-399
        • Allen R.H.
        • Westhoff C.
        • De Nonno L.
        • Fielding S.L.
        • Schaff E.A.
        Curettage after mifepristone-induced abortion: frequency, timing, and indications.
        Obstet Gynecol. 2001; 98: 1149-1150
        • Cowett A.A.
        • Cohen L.S.
        • Lichtenberg E.S.
        • Stika C.S.
        Ultrasound evaluation of the endometrium after medical termination of pregnancy.
        Obstet Gynecol. 2004; 103: 871-875
        • Acharya G.
        • Haugen M.
        • Brathen A.
        • Nilsen I.
        • Maltau J.M.
        Role of routine ultrasonography in monitoring the outcome of medical abortion in a clinical setting.
        Acta Obstet Gynecol Scand. 2004; 83: 390-394
        • Napolitano R.
        • Ghosh M.
        • Gillott D.J.
        • Ojha K.
        Three-dimensional Doppler sonography in asymptomatic and symptomatic women after medical termination of pregnancy.
        J Ultrasound Med. 2014; 33: 847-852
        • Lynge E.
        • Sandegaard J.L.
        • Rebolj M.
        The Danish national patient register.
        Scand J Public Health. 2011; 39: 30-33
        • Tølbøll Blenstrup L.
        • Knudsen L.B.
        Danish registers on aspects of reproduction.
        Scand J Public Health. 2011; 39: 79-82
        • Knudsen L.B.
        • Olsen J.
        The Danish medical birth registry.
        Dan Med Bull. 1998; 45: 320-323
        • Reeves M.F.
        • Fox M.C.
        • Lohr P.A.
        • Creinin M.D.
        Endometrial thickness following medical abortion is not predictive of subsequent surgical intervention.
        Ultrasound Obstet Gynecol. 2009; 34: 104-109
        • Rørbye C.
        • Nørgaard M.
        • Nilas L.
        Prediction of late failure after medical abortion from serial β-hCG measurements and ultrasonography.
        Hum Reprod. 2004; 19: 85-89
        • Markovitch O.
        • Tepper R.
        • Klein Z.
        • Fishman A.
        • Aviram R.
        Sonographic appearance of the uterine cavity following administration of mifepristone and misoprostol for termination of pregnancy.
        J Clin Ultrasound. 2006; 34: 278-282