{"paper_id":"0a3c0604-3ba8-427b-8b41-e65a21bccaba","body_text":"UCSF\nUC San Francisco Previously Published Works\nTitle\nUterine Artery Pseudoaneurysm in the Setting of Deep Endometriosis: An Uncommon Cause \nof Hemoperitoneum in Pregnancy\nPermalink\nhttps://escholarship.org/uc/item/75f6m9m9\nAuthors\nFeld, Zoe M\nRowen, Tami\nCallen, Andrew\net al.\nPublication Date\n2017-09-27\n \nPeer reviewed\neScholarship.org Powered by the California Digital Library\nUniversity of California\n\nEmergency Radiology  \nUterine Artery Pseudoaneurysm in the Setting of Deep Endometriosis: An Uncommon\nCause of Hemoperitoneum in Pregnancy\n--Manuscript Draft--\n \nManuscript Number: EMRA-D-17-00155\nFull Title: Uterine Artery Pseudoaneurysm in the Setting of Deep Endometriosis: An Uncommon\nCause of Hemoperitoneum in Pregnancy\nArticle Type: Case Report\nKeywords: uterine artery;  pseudoaneurysm;  deep endometriosis;  hemoperitoneum in pregnancy\nCorresponding Author: Zoe McGovern Feld, B.A.\nFlorida International University Herbert Wertheim College of Medicine\nMiami, UNITED STATES\nCorresponding Author Secondary\nInformation:\nCorresponding Author's Institution: Florida International University Herbert Wertheim College of Medicine\nCorresponding Author's Secondary\nInstitution:\nFirst Author: Zoe Feld, B.A.\nFirst Author Secondary Information:\nOrder of Authors: Zoe Feld, B.A.\nTami Rowen, M.D.\nAndrew Callen, M.D.\nRuth Goldstein, M.D.\nLiina Poder, M.D.\nOrder of Authors Secondary Information:\nFunding Information:\nAbstract: Uterine, ovarian, and placental pathologies are among the differential considerations\nfor a pregnant woman presenting with abdominal and pelvic pain. Imaging plays a key\nrole in the initial workup of these patients. Sonography is often the first line test,\nhowever evaluation of pelvic pathology can be limited in the gravid state, especially in\nmid- or late- term pregnancy. We present a case of a pregnant woman who came to\nthe emergency room at 25 weeks with acute abdominal and pelvic pain. Both\nultrasound and MR imaging findings revealed intraperitoneal hemorrhage, initially of\nunknown origin, as well as endometriomas and deep endometriosis. Only postpartum\nimaging confirmed a uterine artery pseudoaneurysm (PSA) presumably due to\ndecidual reaction in deep endometriosis. We speculate the intraperitoneal hemorrhage\nwas subsequently due to the PSA. This case demonstrates that if hemorrhage is not\nrecognized promptly, it can lead to hemodynamic instability, as well as premature labor\nand delivery.\nPowered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation\n\nTitle Page\nClick here to access/download\nTitle Page\nTitle page with author information - Emergency\nRadiology.docx\n\nAbstract:  \nUterine, ovarian, and placental pathologies are among the differential considerations for a \npregnant woman presenting with abdominal and pelvic pain. Imaging plays a key role in the \ninitial workup of these patients. Sonography is often the first line test, however evaluation of \npelvic pathology can be limited in the gravid state, especially in mid- or late- term pregnancy. \nWe present a case of a pregnant woman who came to the emergency room at 25 weeks with \nacute abdominal and pelvic pain. Both ultrasound and MR imaging findings revealed \nintraperitoneal hemorrhage, initially of unknown origin, as well as endometriomas and deep \nendometriosis. Only postpartum imaging confirmed a uterine artery pseudoaneurysm (PSA) \npresumably due to decidual reaction in deep endometriosis. We speculate the intraperitoneal \nhemorrhage was subsequently due to the PSA. This case demonstrates that if hemorrhage is not \nrecognized promptly, it can lead to hemodynamic instability, as well as premature labor and \ndelivery. \nBackground: \nUterine, ovarian, and placental pathologies are among the differential considerations for a \npregnant woman presenting with abdominal and pelvic pain. Imaging plays a key role in the \ninitial workup of these patients. Sonography is often the first line test, however evaluation of \npelvic pathology can be limited in the gravid state, especially in mid- or late- term pregnancy. In \nthese cases, magnetic resonance imaging (MRI) can be helpful but may also be difficult to \ninterpret in advanced stages of pregnancy. Endometriosis-associated spontaneous \nhemoperitoneum in pregnancy (SHiP) is a very rare condition and is difficult to diagnose \npreoperatively. Patients may be initially hemodynamically stable with a nonspecific physical \nexam.1 Complications of endometriosis, specifically vascular compromise, are not common \nenough to be routinely considered in the patient’s initial workup but the following case \nhighlights the importance of this diagnosis. We present a case of a pregnant woman who came to \nthe emergency room at 25 weeks with acute abdominal and pelvic pain. Both ultrasound and MR \nimaging findings revealed intraperitoneal hemorrhage, initially of unknown origin, as well as \nendometriomas and deep endometriosis. Only postpartum imaging confirmed a uterine artery \npseudoaneurysm (PSA) presumably due to decidual reaction in deep endometriosis. We \nspeculate the intraperitoneal hemorrhage was subsequently due to the PSA. We found no \npreviously reported cases in the literature of spontaneous PSA developing in decidualized \nendometriosis. Intraperitoneal hemorrhage is a serious complication during pregnancy. This case \ndemonstrates that if hemorrhage is not recognized promptly, it can lead to hemodynamic \ninstability, as well as premature labor and delivery. \nCase presentation: \nA 37-year-old G1P0000 woman at 23w2d was admitted for significant abdominal pain. She had \na past medical history significant for chronic microcytic anemia and granulomatosis with \npolyangiitis (GPA), placental mosaicism and intrauterine growth restriction as well as a bladder \nmass noted on routine obstetric ultrasound, felt to be benign on cystoscopy (and turned out to be \nanother site of deep endometriosis). On labor and delivery, she was found to have frequent \ncontractions on tocometer. Her exam was unremarkable but her laboratory workup showed a \nBLIND Manuscript--SHOULD NOT CONTAIN AUTHOR\nINFORMATION\nClick here to view linked References\n\nhemoglobin was noted to drop from 11.5 10 days prior to 8.9 on the day of admission. A limited \nobstetric ultrasound showed no evidence of placental abruption. She had no cervical change and \nwas presumed to have pain from preterm contractions, was placed on disability due to the pain \nand followed up as an outpatient. She was seen for a formal obstetric ultrasound 11 days after \ndischarge, now 25w1d and was continuing to report severe abdominal pain, reporting 5 days of \nacute on chronic sharp right lower quadrant abdominal pain. There was associated anorexia \nwithout nausea, vomiting, vaginal bleeding or discharge. On exam she was afebrile and without \nperitoneal signs. Due to concern for ovarian torsion, pelvic ultrasound was performed, in which \nthe ovaries were not adequately visualized, and multiple dilated veins were seen in the left \nadnexal region (Figure 1). MRI without contrast was recommended to exclude ovarian torsion \nand look for other potential causes of pelvic pain. MRI was limited due to the patient’s \ndiscomfort during the exam but revealed extrauterine locules of T1 hyperintense material mostly \nsurrounding left adnexa and posterior cul-de-sac in keeping with intraperitoneal hemorrhage. \n(Figure 2) The left ovary had no signs of torsion but contained a complex cyst with internal \nblood products, thought to represent hemorrhagic cyst versus an endometrioma. Incidentally \nnoted was left uterosacral ligament asymmetric thickening with heterogeneous T2 signal with \ninternal hyperintense foci on T1 (Figure 3). Asymmetric hydronephrosis was present on the left \nside with narrowing of the distal ureter likely due to endometriosis.  \nThe patient was admitted for pain control and diagnostic workup. Her pain worsened and during \nthe observation, she reported a sudden and worsening pain, and an increase in uterine \ncontractions with cervical change. Repeat labs showed her hemoglobin dropped from 8.7 to 6.3, \nrequiring transfusion of two units of packed red blood cells. Bedside ultrasound revealed large \nvolume free fluid in her abdomen. The patient was taken for diagnostic laparoscopy, during \nwhich 1600 cc of organized blood was evacuated from the pelvis. The left ovarian lesion was \nidentified that was moderately bleeding, intraoperatively thought to represent a hemorrhagic cyst \nor endometrioma; peritoneal biopsies were obtained and sent for pathologic analysis, which \nsubsequently demonstrated decidual reaction. On post-operative day 1, the patient experienced \npreterm premature rupture of membranes and delivered via spontaneous vaginal delivery at \n25wk6d. The neonate expired at four days of life. The patient’s clinical status stabilized post-\npartum and she was discharged from the hospital with a presumed diagnosis of endometriosis. \nFollow up MRI with contrast three weeks after discharge revealed deep endometriosis involving \nthe left pelvic sidewall and left uterosacral ligament, decreased in extent compared to her prior \nintra-partum exam. Within the region of deep endometriosis was a focus of intense enhancement \nmeasuring 11 mm, suggestive of pseudoaneurysm. CT angiogram of the abdomen and pelvis \ndelineated the pseudoaneurysm deriving from a branch of the left uterine artery, measuring 14 x \n9 mm and with no evidence of active extravasation. Subsequent embolization attempt by \ninterventional radiology confirmed the pseudoaneurysm but embolization was unsuccessful due \nto uterine artery spasm.  (Figure 4) Two weeks later, repeat MRI showed resolution of the \npseudoaneurysm. Patient is currently being followed for treatment of her newly diagnosed deep \nendometriosis with the goal of conceiving in the near future and carrying to term. \nDiscussion: \n\nWhen a pregnant patient presents with lower abdominal or pelvic pain, transabdominal \nsonography should be performed as the initial imaging modality.  As was the case with this \npatient, peritoneal implants in endometriosis can often be difficult to visualize with ultrasound.2 \nIf the pathologic process is incompletely characterized, as in this case, it is reasonable to proceed \nto noncontrast MRI, which is more specific for the diagnosis of endometriosis.3  \nDeep endometriosis is defined as subperitoneal invasion by endometriotic lesions causing \nreactive fibrous changes which typically present as T2 hypointense masses or nodules with \nirregular, indistinct or stellate margins. Intermingled T2 hyperintense foci may be observed, \nreflecting ectopic endometrial glands. There may or may not be T1 hyperintense internal \nendometriotic foci. The fibrous and reactive changes may demonstrate enhancement on post \ncontrast imaging and may raise concern for a possible neoplastic process. Typical anatomic \nlocations include the rectovaginal septum, uterosacral ligaments, vesicovaginal septum, \nvesicouterine pouch, prevesical space, alimentary tract, and urinary tract. Involvement of the \nurinary tract or uterosacral ligaments can lead to ureteral strictures and hydronephrosis. The \npresence of these characteristic signal characteristics in the typical anatomic locations are \ndiagnostic of deep endometriosis.4 \nThe natural history of endometriosis involves a gradual decrease in symptoms and imaging \nfindings throughout pregnancy, with the notable exception of the phenomenon of \ndecidualization. With decidual reaction induced from the gravid state, neovascularity can \ndevelop.  In this case, we speculate decidualization and subsequent neovascularity led to the \neventual development of pseudoeneurysm.5 \nUterine artery pseudoaneurysm is a rare complication of endometriosis.  Ferrero et al. described \nuterine artery pseudoaneurysm after excision of pelvic endometriosis, but no case has been \nreported in the literature of a pseudoaneurysm arising from decidiualized endometriosis without \nprevious surgery or an inciting event.6 Although no cases have been reported of primary uterine \nartery aneurysm secondary to GPA, it is possible that this patient’s condition predisposed her to \nvascular injury. \nIn this case, secondary arterial pseudoaneurysm was not detected until post partum contrast \nenhanced MRI. Gadolinium is generally avoided in pregnancy due to its ability to cross the \nplacenta and potentially harm the fetus. The American College of Obstetrics and Gynecology \nrecommends the use of MRI contrast should be limited to situations where the diagnostic \nbenefits clearly outweigh the possible risks.7 Although difficult to determine prospectively, in \nthis case the risk from complications of intraperitoneal hemorrhage outweighs the potential risk \nfrom gadolinium administration. We postulate that if contrast imaging was obtained earlier, the \nPSA would have been detected and treated sooner, and premature delivery potentially could have \nbeen avoided.   \nIn summary, understanding and utilizing ultrasound and MRI as complementary modalities in \nthe pregnant patient with abdominal pain can expedite the diagnosis and lead to expedited \ntreatment and preservation of pregnancy.  This case highlights a rare but important complication \nof endometriosis in pregnancy. \n\n \nConflict of Interest: The authors declare that they have no conflict of interest. \n \nInformed Consent: For this type of study, formal consent is not required.  \n \nReferences \n1. Leone Roberti Maggiore U, Ferrero S, Mangili G et al. A systematic review on \nendometriosis during pregnancy: diagnosis, misdiagnosis, complications and outcomes. \nHuman Reproduction Update. 2015;22(1):70-103. \n2. Olive DL, Schwartz LB. Endometriosis. N. Engl. J. Med. 1993;328 (24): 1759-69 \n3. Woodward PJ, Sohaey R, Mezzetti TP. Endometriosis: radiologic-pathologic correlation. \nRadiographics. 21 (1): 193-216 \n4. Coutinho et al . MR Imaging in Deep Pelvic Endometriosis: A pictorial Essay. \nRadioGraphics 2011; 31:549–567 \n5. Poder L, Coakley FV, Rabban JT, Goldstein RB, Aziz S, Chen LM. Decidualized \nendometrioma during pregnancy: recognizing an imaging mimic of ovarian malignancy. J \nComput Assist Tomogr. 2008;32(4):555–8. \n6. Ferrero et al. Unusual complication of excision of pelvic endometriosis: pseudoaneurysm \nof the left uterine artery. Fertil Steril. 2010 Jan;93(1):264-6 \n7. Guidelines for diagnostic imaging during pregnancy and lactation. Committee Opinion \nNo. 656. American College of Obstetricians and Gynecologists. Obstet Gynecol \n2016;127:e75–8 \nFigures \n \n \nFigure 1: Initial gray scale (A)  duplex US (B)  and B-mode (C) did not clearly visualize adnexa. \nDilated pelvic veins were described which were thought to represent physiologic changes. Only \nretrospectively was a focal areal of increased vascularity identified correlating to the PSA on \npostpartum MRI and CTA. (arrows)  \n\n\n \nFigure 2: A. T1 axial image demonstrating hyper intense material surrounding left adnexa and \ncut de sac (arrow), representing extra uterine blood products consistent with intraperitoneal \nhemorrhage. \n \n \n \n \nFigure 3: A. T2 axial image demonstrated possible hemorrhagic cyst versus endometrioma \n(arrow). Incidentally noted is asymmetric thickening of left uterosacral ligament (asterisk) and \nsecondarily dilated left ureter (arrow head). B. T1 hyperintense foci in the thickened left \nuterosacral ligament (arrow). The constellation of findings on T1 and T2 are strongly suggestive \nof deep endometriosis involving the left uterosacral ligament. \n\n\n \nFigure 4: A. T1 post gadolinium image demonstrates mildly enhancing thickened left \nuterosacral ligament (arrowheads) with focal rounded area of brisk enhancement (arrow). B. \nBranch uterine artery PSA was confirmed on subsequent CT. C. Pre-embolization angiogram \nconcurred with the cross-sectional imaging findings (arrow)","source_license":"CC0","license_restricted":false}