Magnetic resonance imaging presentation of deep infiltrating endometriosis nodules before and after pregnancy: A case series

Anne Elodie Millischer, A. Millischer, Millischer AE, Louis Marcellin, Santulli P, Pietro Santulli, Piétro Santulli, Chloé Maignien, Maignien C, Chloe Maignien, Mathilde Bourdon, Bourdon M, Bruno Borghese, François Goffinet, Goffinet F, Chapron C, Charles Chapron
PloS one · 2019 · vol. 14(10) , pp. e0223330 · doi:10.1371/journal.pone.0223330 · PMID:31584969 · PMC6777797 · W2978023201
article OA: gold CC0 ⤵ 6 in-corpus citations
📄 Open PDF Full text JSON View on OpenAlex View on PubMed View at publisher
⚙ AI-generated summary by claude@2026-06, 2026-06-07 ⓘ

This case series compared MRI features of deep infiltrating endometriosis nodules in 21 women before and after pregnancy, finding significantly smaller lesion volumes after pregnancy.

One-sentence paraphrase of the abstract; not a substitute for reading it. No clinical advice. How this works

⚙ AI-generated deep summary by claude@2026-06, 2026-06-07 · read from full text ⓘ

This retrospective case series studied 21 women with MRI-diagnosed deep infiltrating endometriosis (DIE) who had monofetal pregnancies and no prior endometriosis surgery, comparing pelvic MRI lesion volumes obtained before pregnancy and again after delivery. Using the same 1.5T MRI protocol interpreted by a single experienced radiologist, the authors found that overall DIE lesion volume was significantly higher before pregnancy than after pregnancy (2,552 ± 3,315 mm3 vs. 1,708 ± 3,266 mm3, p < 0.01), and the largest lesion volume per patient was also significantly larger before pregnancy (4,728 ± 4,776 mm3 vs. 3,165 ± 5,299 mm3, p < 0.01). Limitations explicitly include the case-series design and the fact that all participants were on hormonal treatment at the time of the pre-pregnancy MRI, potentially confounding lesion changes. This paper is centrally about endometriosis — it directly assesses MRI volume changes of deep infiltrating endometriosis nodules before and after pregnancy.

Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works

Abstract

OBJECTIVE: To compare the magnetic resonance imaging (MRI) features of deep infiltrating endometriosis (DIE) lesions before and after pregnancy. DESIGN: Retrospective study. SETTING: A single French university tertiary referral hospital. PATIENTS: Twenty-one women without a prior history of surgery for endometriosis with a radiological diagnosis by MRI with two sets of examinations performed before and after pregnancy. INTERVENTIONS: The volumes of the lesions were compared using the same protocol before and after pregnancy based on MRI (1.5 T) examinations by a single experienced radiologist who is a referring practitioner for image-based diagnosis of endometriosis. MAIN OUTCOME MEASURE(S): The DIE lesion volume. MEASUREMENTS AND MAIN RESULTS: Between October 2012 and December 2016, a total of 21 patients (67 lesions) were included and compared before and after pregnancy. The mean time interval between the MRI before pregnancy and delivery was 19.6 ± 8.5 months (median: 17.6, IQR 13.5-25.2 months). The mean time interval between delivery and the MRI after pregnancy was 11.0 ± 6.4 months (median: 8.3, IQR 6-15.2 months). The mean overall DIE lesion volume by MRI was significantly higher before pregnancy compared to after pregnancy (2,552 ± 3,315 mm3 vs. 1,708 ± 3,266 mm3, respectively, p < 0.01). The mean volume by MRI of the largest lesion of each patient was significantly higher before pregnancy compared to after pregnancy (4,728 ± 4,776 mm3 vs. 3165 ± 5299 mm3; p < 0.01). CONCLUSION: Our data indicate a favorable impact of pregnancy on DIE lesion volumes as measured by MRI.
Full text 16,543 characters · extracted from pmc-nxml · 5 sections · click to expand

Intro

Endometriosis is a common chronic benign hormone-dependent gynecological disorder associated with persistent pelvic pain and/or infertility [ 1 , 2 ] that is characterized by the development of glandular and stromal endometrium-like tissues at ectopic locations.[ 3 ] At present, imaging procedures such as ultrasonography (US),[ 4 ] and magnetic resonance imaging (MRI) are suitable for diagnosing and for providing follow-up of endometriotic lesions (endometrioma—OMA and deep infiltrating endometriosis—DIE), and allow for a non-surgical diagnosis of endometriosis,[ 5 , 6 ] without a need for histological confirmation.[ 5 , 7 , 8 ] The data available regarding the interrelationships between pregnancy and endometriosis are limited or of poor-quality.[ 9 – 11 ] Historically, pregnancy is typically considered to have a positive -albeit temporary- effect on endometriosis, which is most likely due to anovulation, amenorrhea, and pregnancy-related hormonal changes that prevent bleeding in the ectopic endometriotic tissue.[ 12 ] Varying degrees of growth changes of ectopic endometriotic lesions have been observed.[ 11 ] In addition, severe but rare and unpredictable complications during pregnancy have been reported with puerperal changes in ectopic lesions.[ 13 ] Preliminary data indicate that pregnancies in conjunction with endometriosis are associated with increased adverse pregnancy outcomes such as spontaneous miscarriage,[ 14 ] preterm birth, preeclampsia, small-for-gestational-age offspring, and obstetric hemorrhages.[ 15 – 22 ] One of the suspected etiologies stems from the potential increased risk of defective deep placentation.[ 23 , 24 ] Moreover, endometriosis-like glands have previously been reported to be in close contact with the fetal membranes in pregnant women who have endometriosis.[ 25 ] During pregnancy, the endometrium undergoes decidualization (initiated during the secretory phase of the menstrual cycle and persisting when implantation of the blastocyst occurs) that is mainly induced by progesterone.[ 26 ] This decidualization results in conversion of the endometrium into a specialized uterine tissue,[ 27 ] thereby controlling trophoblast invasion and guaranteeing optimal accommodation of the gestation.[ 28 ] This transformation of the endometrium involves different metabolic, hormonal, angiogenic, and immune system changes related to steroid exposure.[ 29 ] Decidualization also occurs in ectopic endometriotic lesions.[ 30 , 31 ] It is not clear whether these changes in ectopic lesions persist after delivery. A previous study evaluated the clinical features and the change in size based on US during pregnancy of 25 OMA,[ 32 ] and a 3-case series of deep infiltrating endometriosis (DIE) lesions,[ 30 ] both of which exhibited a change in size. However, no data are available regarding the progression of DIE lesions before and after pregnancy based on MRI. Therefore, the aim of this retrospective study was to compare the DIE lesion volume by MRI before and after pregnancy.

Results

Between October 2012 and December 2016, 21 DIE patients without a prior history of surgery for endometriosis underwent MRI before and after pregnancy. The characteristics of the women and the pregnancies are presented in Table 1 . No endometriosis-related complications were reported during the pregnancies. All of these 21 women were undergoing hormonal treatment at the time of the MRI prior to the pregnancy. The mean time interval between the MRI before pregnancy and delivery was 19.6 ± 8.5 months (median: 17.6, IQR 13.5–25.2 months). The mean time interval between delivery and the MRI after pregnancy was 11.0 ± 6.4 months (median: 8.3, IQR 6–15.2 months). The mean number of DIE lesions was 3.2 ± 1.1 (median: 4, IQR 3–4) ( Table 1 ). The distribution of the DIE location and the OMA based on MRI are presented in Table 2 : 19/21 patients exhibited vaginal posterior fornix involvement (90.5%), 20/21 uterosacral involvement (95.2%), 14/21 bowel involvement (66.7%), 3/21 bladder involvement (14.3%), and 15/21 OMAs (71.4%) ( Table 2 ). ART: assisted reproductive technology; WG: weeks of gestation, SGA: small for gestational age offspring. OMA: Ovarian endometrioma * 1: unilateral lesion 2: bilateral lesion Fig 1 depicts the DIE lesions of a woman before ( Fig 1A, 1C and 1E ) and after ( Fig 1B, 1D and 1F ) pregnancy based on MRI, with the bowel ( Fig 1A and 1B ), uterosacral ligaments and the posterior vaginal fornix ( Fig 1C and 1D ), and OMAs ( Fig 1E and 1F ) exhibiting a qualitative thinning and reduction in the size of the lesions. (A,B) sagittal plane, deep infiltrative endometriosis of the bowel (white arrow); (C,D) axial plane, deep infiltrative endometriosis of the uterosacral ligaments and posterior vaginal fornix (white circle); (E,F) axial plane, ovarian endometriosis, axial plane with fat saturation, illustrating a left endometrioma in T1-hypersignal (white arrow). A total of 67 DIE lesions were compared before and after pregnancy. For these DIE lesions, the volume decreased for n = 49/67 lesions (73.1%), was stable for n = 5/67 lesions (7.5%), and increased for n = 13/67 lesions (19.4%). The mean overall DIE lesion volume by MRI was significantly higher before pregnancy compared to after pregnancy (2,552 ± 315 mm 3 vs. 1,708 ± 3,266 mm 3 , respectively, p < 0.01) ( Table 3 ). In case of multiple DIE lesions, the mean volume of the largest lesion of each patient was significantly higher before pregnancy compared to after pregnancy (4,728 ± 4,776 mm 3 vs. 3,165 ± 5,299 mm 3 ; p < 0.01) ( Table 3 ). The OMA sizes were significantly higher before pregnancy ( Table 3 ). DIE: deep infiltrating endometriosis, OMA: endometrioma, MRI: magnetic resonance imaging, USL uterosacral ligament. All of the measurements are presented as means ± the standard derivation.

Conclusions

This study demonstrates, for the first time, a favorable effect of pregnancy on the volume of DIE lesions, which regressed significantly after pregnancy based on MRI evaluation. Longer follow-up is required to more extensively evaluate the change in DIE lesions during and after pregnancy.

Materials|Methods

This study was approved by the National Data Protection Authority (Commission Nationale de l’Informatique et des Libertés, CNIL n° 1755849). French regulations stipulate that this study is exempt from Institutional Review Board (IRB) review because it is an observational study using anonymized data from medical records. The study's exempt status was confirmed by the IRB Ile-de-France. During the study period, women were routinely informed that their records could be used for evaluation of medical practices and that they had the right to opt-out of these studies. This was a retrospective analysis of a series of severe-stage DIE patients without a prior history of surgery for endometriosis and who had a monofetal pregnancy followed in a tertiary care center. Women with a diagnosis of DIE based on MRI criteria [ 5 , 33 ] obtained before pregnancy were retained for the study. After delivery, an MRI was carried out of women with DIE due to painful symptoms. All of the pregnant women received prenatal care at our tertiary university hospital before 22+0 weeks and they gave birth there after 24+0 weeks in accordance with the protocols of the department.[ 34 ] The two sets of pelvic MRI examinations were performed by a single experienced radiologist, who is referring practitioner for image-based diagnosis of endometriosis, using the same protocol before and after pregnancy on a 1.5 T MRI machine (Sonata, Siemens; Erlangen, Germany) based on stringent previously published criteria.[ 5 , 7 , 8 , 35 , 36 ] The MRI obtained before and after pregnancy were reviewed independently, one month apart, by the same experienced radiologist to avoid memory contamination and inter-operator variability.[ 37 ] In keeping with the literature, OMA exhibited pathognomic features of signal loss on T2-weighted imaging, referred to as “T2 shading”, and high signal intensity on T1-weighted images with and without fat saturation. These elements reflect the hemorrhagic nature of OMA and they help differentiate them from other T1-hyperintense lesions, such as dermoid and hemorrhagic cysts.[ 38 , 39 ] DIE nodules are composed of fibromuscular hyperplasia surrounding endometrial glands, displaying T2-hypointense signal, T1 intermediate signal intensity, with masses that had irregular edges and/or thickening of soft tissues leading to distortion of the normal pelvic anatomy and adhesion formation.[ 38 , 40 ] Deep endometriosis involving the uterosacral ligaments appears as T2-hypointense thickening or nodularity. Rectosigmoid involvement exhibits a specific morphological pattern, in the anterior wall, featuring a semi-lunar hypointense T2 signal nodule, recently described as a “mushroom cap sign”.[ 41 ] The DIE lesions were classified based on five locations (i.e., uterosacral ligament(s), the vagina, bladder, intestine, and ureter),[ 42 ] and they were diagnosed according to stringent imaging criteria published elsewhere.[ 36 ] In cases with multiple DIE lesions, the patients were classified according to the worst location (least to most severe: uterosacral ligament(s), the vagina, bladder, intestine, and ureter).[ 42 ] The volumes of the DIE lesions and the sizes of the OMA were assessed by MRI before and after the pregnancy by the same radiologist (AEM). The nodule volume of the uterosacral ligament(s), the bladder, and the intestine were calculated with the volume formula for a general ellipsoid solid (V = π/6 [(a + b)/2]). Posterior vaginal fornix nodules were considered to be spheroids, and the volumes were calculated with the volume formula for a general sphere (V = 4/3 π r 3 ).[ 43 ] For each patient, their personal history data and pregnancy-related information were collected including the maternal age, weight, size, assisted reproductive technology (ART), gestity, parity, gestational hypertension, mode of delivery, gestational age at delivery, preterm birth < 37 weeks of gestation, birth weight, small for gestational age offspring, the time interval between the MRI before pregnancy and delivery, and the time interval between delivery and the MRI after pregnancy. For each patient, the distribution of DIE lesions and the presence of OMA were recorded. The continuous data are presented as means with the standard deviation and as medians and the interquartile range (IQR) 25–75%. Paired analyses were used when comparing pre- and post-pregnancy lesion volumes (mm 3 ) by MRI, and comparisons were performed according to the DIE location. In case of multiple DIE lesions, paired comparisons of lesion volumes before and after pregnancy were also performed according to the largest lesion of each patient. Paired OMA sizes (mm) were compared before and after pregnancy. The quantitative data were compared using the Wilcoxon matched-pairs test. A p-value < 0.05 was considered statistically significant. The data were analyzed using STATA software for Macintosh (Stata/IC 11.0 for Mac, StataCorp College Station, TX, USA).

Supplementary Material

(XLSX) Click here for additional data file.

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

⚙ Ask this paper AI returns verbatim quotes from the full text · source: pmc-nxml ⓘ

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Condition tags

endometriosisdie_deep_infiltrating

MeSH descriptors

Biopsy Biopsy Endometriosis Endometriosis Magnetic Resonance Imaging Magnetic Resonance Imaging Pregnancy Complications Adult Endometriosis Female Humans Organ Size Pregnancy Retrospective Studies

Citation neighborhood

Papers in the corpus that this work cites (lower rings, blue) and that cite this one (upper rings, green). Dot size scales with the paper's in-corpus citation count — bigger dot = more influential within the endo/adeno field. Click a dot to open that paper. [ expand to 2 hops ] — adds papers reached through this work's immediate citers/citees. Heavier; up to 60 extra dots.

References (100)

Cited by (7)

Source provenance

crossref
last seen: 2026-06-25T06:33:43.642476+00:00
europepmc
last seen: 2026-07-26T06:08:39.051465+00:00
openalex
last seen: 2026-06-10T17:14:06.276822+00:00
pubmed
last seen: 2026-05-13T22:22:29.487098+00:00
License: CC0 · commercial use OK