Congenital uterine anomalies and congenital heart disease: An underappreciated association?

Harrison J. VanDolah: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Kathy Mostajeran: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Luis Goncalves: Writing – review & editing, Validation, Supervision, Investigation, Formal analysis. Wayne J. Franklin: Writing – review & editing, Validation, Supervision, Project administration, Formal analysis. Melissa Parks: Writing – review & editing, Writing – original draft, Validation, Supervision, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Tabitha G. Moe: Writing – review & editing, Writing – original draft, Validation, Supervision, Methodology, Investigation, Formal analysis, Conceptualization.

No funding was used or received for this work.

We performed a single center (Phoenix Children's Hospital, Phoenix, AZ) retrospective cohort study of all patients assigned female at birth with ICD-10 codes corresponding to congenital heart disease, and a completed pelvic MRI or pelvic US that included adequate imaging of the uterus to assess for CUA. IRB approval was obtained from the Phoenix Children's Hospital IRB Board (IRB-23-175, reference number 041055). Since an electronic health record was established at our institution in 2011, only imaging studies that were uploaded since that time were able to be included. Baseline demographics and comorbidities were documented as well as the specific congenital heart disease and congenital uterine anomaly. CHD diagnoses were confirmed by authors' review and stratified into “simple”, “moderate”, or “complex” in accordance with the 2018 ACHD guidelines, with the exception of certain pathologies the authors feel are better categorized as “simple”, for example bicuspid aortic valve [ 11 ]. CUA were classified into uterine agenesis, hypoplastic uterus, septate uterus, bicornuate uterus, unicornuate uterus, uterus didelphys, or arcuate uterus according to the 2021 ASRM guidelines [ 12 ]. Statistical analysis was performed on all variables, with normally distributed continuous variables described as mean±standard deviation and test for difference with Student's t-test. Categorical variables were compared using Chi-squared tests for expected values more than 5, and Fisher's Exact test for expected values less than 5. P-values <0.05 were considered statistically significant.

197 patients met the inclusion criteria, 140 of whom had pelvic US and 57 had pelvic MRI. Initially, 26 patients were found to have had CUA, however after review of all studies by our radiologist co-investigator (LG), an additional 7 patients were found to have had CUA. This led to a total prevalence of 33 of 197 total patients, or 16.7 %. Basic demographics and comorbidities of all patients are summarized in Table 1 . Notable statistically significant associations with the presence of a CUA include the presence of renal malformations, anal malformations, whether a patient had pelvic ultrasound or MRI performed, Trisomy 21 (which was negatively associated), and our category for “other chromosomal abnormality”. The included diagnoses are specifically summarized in Table 2 . VACTERL association was found in 7/33 (21 %) patients with CUA, and 7/12 (58 %) of all patients with VACTERL in the cohort had CUA, which was statistically significant (p = 0.001). Of the patients with VACTERL who had CUA, two had Tetralogy of Fallot, one had partial anomalous pulmonary venous return, and the remaining four had simple CHD. The only specific CHD diagnosis that was statistically significantly associated was partial anomalous pulmonary venous return (PAPVR). Complexity of CHD was not statistically significantly associated with CUA. CHD diagnoses are summarized in the Central Illustration. Table 1 Demographics, comorbidities, and reason for imaging. Table 1 No CUA (#) % Positive CUA (#) % P-value Total 164 83.20 % 33 16.80 % Age at image (avg, yrs) 13.29 ±7.97 11.85 ±6.67 0.333 ∗ Image Obtained 0.002 † Ultrasound 124 MRI 40 Race/Ethnicity   White 91 55.49 % 17 51.52 % 0.676 †   Black 10 6.10 % 1 3.03 % 0.695 ¥   Hispanic 56 34.15 % 11 33.33 % 0.928 †   Native American 4 2.44 % 3 9.09 % 0.093 ¥   Asian 2 1.22 % 1 3.03 % 0.425   Other 1 0.61 % 0 0.00 % 1.000 Simple CHD 99 60.37 % 16 48.48 % 0.206 † Moderate CHD 44 26.83 % 12 36.36 % 0.268 † Complex CHD 21 12.80 % 5 15.15 % 0.778 ¥ Renal Malformation 15 9.15 % 10 30.30 % 0.003 ¥ Anal Malformation 8 4.88 % 8 24.24 % 0.001 ¥ Systemic Genetic Syndrome 73 44.51 % 18 54.55 % 0.292 † Reason for Imaging 0.921 †   Gynecologic (eg dysmenorrhea, evaluation for ovarian torsion) 114 69.51 % 22 66.67 %   Vascular anatomy 9 5.49 % 2 6.06 %   vOrthopedic 9 5.49 % 2 6.06 %   Associated congenital anomalies 12 7.32 % 4 12.12 %   Other 18 10.98 % 3 9.09 % Footnote: CUA = congenital uterine anomaly, MRI = magnetic resonance imaging, CHD = congenital heart disease. P-values were obtained with Student's t-test (∗), Chi test (†), or Fischer's exact test (¥). Bolded if statistically significant (p < 0.05). Table 2 Genetic syndromes and associations. Table 2 Syndrome/Association No CUA (#) Positive CUA (#) P-value Trisomy 21 15 0 0.008 DiGeorge syndrome 4 1 1.000 Turner syndrome 12 1 0.700 Heterotaxy syndrome 5 2 0.333 CHARGE syndrome 4 1 1.000 VACTERL association 5 7 0.001 Marfan syndrome 4 1 0.524 Noonan syndrome 2 0 1.000 Jacobsen syndrome 1 0 1.000 Alagille syndrome 2 0 1.000 Kabuki syndrome 1 1 0.309 Cornelia de Lange syndrome 1 0 1.000 Baraitser-Winter 1 0 1.000 Trisomy 18 1 0 1.000 Neurofibromatosis 1 1 0 1.000 Other chromosomal abnormality 7 5 0.032 Costello syndrome 2 0 1.000 Mucolipidosis type 3 1 0 1.000 Pitt-Hopkins syndrome 0 1 0.168 Cardio-facio-cutaneous 0 1 0.168 Pierre-Robin sequence 1 1 0.308 Trisomy 8 0 1 0.168 Scimitar syndrome 0 1 0.168 Loeys-Dietz syndrome 5 2 0.333 Wolf-Hirschhorn syndrome 0 1 0.168 Joubert syndrome 0 1 0.168 Desbusquois syndrome 1 0 1.000 CLOVE syndrome 1 0 1.000 Smith-Magenis syndrome 1 0 1.000 Footnote: CHARGE =  c oloboma, h eart defects, a tresia of the nasal choanae, r estricted growth and/or development, g enital and/or urinary abnormalities, e ar abnormalities, VACTERL =  v ertebral defects, a nal atresia, c ardiac defects, t racheo- e sophageal fistula, r enal anomalies, l imb abnormalities. All statistical tests performed for this table were done with Fischer's exact test. Demographics, comorbidities, and reason for imaging. Footnote: CUA = congenital uterine anomaly, MRI = magnetic resonance imaging, CHD = congenital heart disease. P-values were obtained with Student's t-test (∗), Chi test (†), or Fischer's exact test (¥). Bolded if statistically significant (p < 0.05). Genetic syndromes and associations. Footnote: CHARGE =  c oloboma, h eart defects, a tresia of the nasal choanae, r estricted growth and/or development, g enital and/or urinary abnormalities, e ar abnormalities, VACTERL =  v ertebral defects, a nal atresia, c ardiac defects, t racheo- e sophageal fistula, r enal anomalies, l imb abnormalities. All statistical tests performed for this table were done with Fischer's exact test. While the most common indication for pelvic imaging within the cohort was a gynecological complaint (66 %), the percentage was similar in both groups, regardless of whether they were ultimately diagnosed with a CUA or not (66.67 % versus 69.51 % respectively) with nonsignifcant p-value (p = 0.921). Notably, the subtypes of CUA appeared to show some association with the complexity of CHD. Arcuate uterus, the most simple CUA, was almost exclusively seen in patients with simple CHD (6/7). The other CUA subtypes were not statistically associated with CHD subtypes, however the numbers are extremely small to be able to be confident with this analysis. Interestingly, the overall prevalence of subtype of CUA favored uterine didelphys (27.3 %) which is the rarest CUA. The most common CUA within the general population, septate uterus, was less well represented in our population (12.1 %). Fig. 1 presents the CUA cases by subtype as well as by CHD stratification. Fig. 1 Congenital uterine anomalies by congenital heart disease complexity in our cohort. The only statistically significant association was for arcuate uterus and simple CHD. Fig. 1 Congenital uterine anomalies by congenital heart disease complexity in our cohort. The only statistically significant association was for arcuate uterus and simple CHD.

- Medical Knowledge: congenital heart disease (CHD), particularly of moderate to complex classification, are associated with many extracardiac manifestations, and our data demonstrates a previously undefined association with congenital uterine anomalies (CUAs) - Patient Care: our data suggests there may be a role for screening patients with moderate to complex CHD with pelvic US or MRI for congenital uterine anomaly, particularly in those with genetic syndromes, gynecologic symptoms, or considering pregnancy or long-acting reversible contraception. Medical Knowledge: congenital heart disease (CHD), particularly of moderate to complex classification, are associated with many extracardiac manifestations, and our data demonstrates a previously undefined association with congenital uterine anomalies (CUAs) Patient Care: our data suggests there may be a role for screening patients with moderate to complex CHD with pelvic US or MRI for congenital uterine anomaly, particularly in those with genetic syndromes, gynecologic symptoms, or considering pregnancy or long-acting reversible contraception. - Our data is chiefly hypothesis-generating for many forms of CHD and CUAs that will need to be explored further with larger-scale prospective studies to precisely identify the prevalence of CUAs amongst patients with CHD Our data is chiefly hypothesis-generating for many forms of CHD and CUAs that will need to be explored further with larger-scale prospective studies to precisely identify the prevalence of CUAs amongst patients with CHD

This is the first large, published study examining the relationship between congenital heart disease (CHD) and congenital uterine anomalies (CUAs). Within the notable limitations of a retrospective study at a single tertiary referral center, patients with CHD appear to have a high prevalence of CUA. The prevalence is potentially higher than the general population, which is estimated to range from 0.5 % to 5 %, a finding that carries a number of important clinical implications as outlined below [ 1 , 2 ]. However, given the small sample size of patients able to get proper imaging assessment, our study is unable to accurately assess the prevalence of CUAs within the CHD population as well as associations with specific CHD diagnoses. The very large percentage of patients with genetic syndromes (almost 50 % of all patients) included in our cohort suggests our study is also not representative of a general CHD clinic. This is likely from referral bias as our cohort received care at a tertiary referral center and therefore more likely to have complex CHD patients with genetic syndromes. Importantly, our data suggests that the association between CUA and CHD may lie moreso in the prevalence of genetic syndromes in patients with CHD. So while our data suggests that there may be a high prevalence of CUA in patients with CHD, our data does not suggest that CHD alone is independently associated with uterine abnormalities. Additionally, all patients had some indication for uterine imaging, which may overestimate prevalence compared to an asymptomatic population. Though gynecologic indication did not differ between the comparison groups, our sample size may be too small to detect a difference. As such, we recommend future prospective multicenter studies to determine true prevalence should also undertake careful comparison between patients with and without genetic syndromes. The specific gynecological implications of CUAs warrant further discussion. The clinical presentation can range from an asymptomatic coincidental finding to severe primary dysmenorrhea or primary amenorrhea with cyclical pelvic pain due to menses in the setting of an outlet obstruction, such as a due to a transverse vaginal septum or an obstructive non-communicating horn. The impact of delayed diagnosis can be lifelong, as retrograde menstruation is hypothesized to be responsible for the high proportion of concurrent diagnosis of endometriosis in 40 % of patients diagnosed with obstructive CUAs [ 13 ]. Other long-term consequences include an increased risk for developing pelvic adhesions due to ascending infections and the associated lifelong gynecological sequelae, such as chronic pelvic pain or infertility. With the current study demonstrating an overall prevalence of almost 18 % for CUAs in this subpopulation, screening and diagnosis prior to menarche could serve a role in decreasing gynecological complications in patients with CHD. The use of intrauterine contraceptive devices can be particularly favorable in adolescents with CHD who desire an effective, safe, long-acting reversible contraceptive method. Furthermore, the levonorgestrel intrauterine device (IUD) has the additional benefit of menstrual suppression, which can be particularly favorable in the setting of patients requiring chronic anticoagulation therapy. Imaging prior to or after insertion of the device is not currently standard of care. Unfortunately, uterine abnormalities such as septate uterus, bicornuate uterus, and even uterine didelphys can be missed on routine clinical exam [ 3 ]. The safety and efficacy of intrauterine contraceptive devices are unknown in the population of patients with CUAs, although one systematic review of published cases and case series demonstrated that there was an increased risk for device expulsion or a mispositioned device with subsequent unplanned pregnancy in patients with CUAs [ 14 ]. In a patient population with CHD proactively attempting to avoid an unplanned pregnancy due to the increased risk of adverse outcomes, image based screening prior to placement should at least be considered to ensure efficacious placement of the contraceptive device. Obstetrical complications occur in approximately 25 % of patients with a known CUA [ 15 ]. Research investigating perinatal outcomes in patients with CUAs has demonstrated a multitude of adverse effects, ranging from infertility and recurrent pregnancy loss, premature delivery, fetal growth restriction, placental abruption, and an increased risk of malpresentation and cesarean delivery [ 2 , 4 , 5 ]. Many of these same complications are well-documented in the congenital heart disease and pregnancy literature, specifically infertility, recurrent pregnancy loss, premature delivery, and intra-uterine growth restriction [ 10 , 16 , 17 ]. Knowledge of CUA in CHD patients by way of screening, ideally prior to menarche for the reasons listed above, would allow for appropriate preconception counseling and delivery planning. This seems critical particularly given the probable increased prevalence of these disorders within this subpopulation. As the obstetrical risk assessment is multifaceted for patients with CHD, particularly those with complex CHD, these findings add another variable that should be considered in patients with CHD considering pregnancy [ [16] , [17] , [18] ]. The abstract data presented by Farahzad et al. is the only other large retrospective cohort of patients with CHD and CUA, however it was limited to patients at least 12 years old and included CT imaging as well as US and MRI, which makes their data not exactly comparable to ours [ 8 ]. Limitations aside, they did find eight patients with CUA of the 44 patients that met inclusion criteria, which is a similar prevalence to our data of about 18 %. Of their 8 cases, six were complex CHD and nearly all had complex CUA, in keeping with our findings. Four of their patients had obstructive anomalies necessitating hormonal suppression, surgical intervention, or both. Their cases are summarized alongside our cases in Table 3 . Table 3 Summary of all CHD and CUA cases. Table 3 CHD Diagnosis (es) Comorbidities CUA Subtype ∗∗SIMPLE∗∗ Patient 1 ASD, PDA Loeys Dietz, pectus excavatum Bicornuate Patient 2 Simple aortic valve abnormality Chromosomal deletion (4q2), duplication 16q, VUR, malrotation Didelphys Patient 3 ASD VACTERL, TEF, partial vertebral body fusion, urogenital sinus Didelphys Patient 4 Aortic root dilation Marfan syndrome Didelphys Patient 5 Bicuspid aortic valve, left-sided SVC Turner syndrome, gonadal dysgenesis Hypoplastic Patient 6 Simple aortic valve abnormality Joubert syndrome Septate Patient 7 VSD Chromosome 8 abnormality, absent corpus callosum Arcuate Patient 8 ASD VACTERL, imperforate anus, VUR, horseshoe kidney Arcuate Patient 9 Simple aortic valve abnormality Loeys Dietz Arcuate Patient 10 VSD Chromosome 9/16 translocation, imperforate anus, cystic kidneys Uterine agenesis Patient 11 Dysplastic mitral valve Cardio-facio-cutaneous syndrome Bicornuate Patient 12 VSD Pierre Robin, right renal agenesis Didelphys Patient 13 ASD, aberrant right subclavian Wolf-Hirschhorn syndrome Arcuate Patient 14 VSD None Arcuate Patient 15 ASD, VSD VACTERL, left renal agenesis, imperforate anus Didelphys Patient 16 ASD Partial trisomy terminal 1q and partial monosomy of terminal 7q Arcuate Farahzad et al. 2023 VSD VACTERL with cloaca, urinary tract anomalies, skeletal anomalies Didelphys Farahzad et al. 2023 ASD None Septate ∗∗MODERATE∗∗ Patient 17 PAPVR with sinus venosus ASD VACTERL, imperforate anus, horseshoe kidney Uterine agenesis Patient 18 PAPVR with sinus venosus ASD MECP2 duplication, 3p deletion syndrome Bicornuate Patient 19 Severe pulmonary stenosis, ASD Kabuki-like syndrome (HNRNPK mutation, Au Kline syndrome) Bicornuate Patient 20 VSD, pulmonary stenosis Pitt-Hopkins syndrome, VACTERL, imperforate anus Hypoplastic Patient 21 bicuspid aortic valve, parachute mitral valve trisomy 8 mosaic Didelphys Patient 22 PAPVR (right lung to IVC), left SVC Scimitar syndrome, right lung hypoplasia Didelphys Patient 23 left pulmonary artery stenosis left lung hypoplasia, left ectopic kidney, MRKH Uterine agenesis Patient 24 VSD, cleft mitral valve MRKH Uterine agenesis Patient 25 Aortic arch hypoplasia, bicuspid aortic valve, VSD left lung hypoplasia Septate Patient 26 Tetralogy of Fallot q22 deletion, VACTERL, double hemivertebrae, ambiguous genitalia Uterine agenesis Patient 27 Ebstein anomaly CHARGE, blindness, scoliosis, WPW Hypoplastic Patient 28 Tetralogy of Fallot VACTERL, cloacal anomaly, tethered cords Didelphys ∗∗COMPLEX∗∗ Patient 29 Dextrocardia, left AV valve atresia, pulmonary atresia, TGA Heterotaxy with asplenia Septate Patient 30 PA/IVS None Bicornuate Patient 31 Shone complex None Didelphys Patient 32 Azygous continuation of IVC, situs inversus Heterotaxy with polysplenia, anterior anus Arcuate Patient 33 Shone complex None Septate VanDolah and Moe 2024 ccTGA T1DM, atrial fibrillation Bicornuate VanDolah and Moe 2024 Tricuspid atresia None Bicornuate VanDolah and Moe 2024 HLHS, DORV, interrupted IVC Heterotaxy with asplenia Bicornuate Farahzad et al. 2023 Tricuspid atresia None Uterine agenesis Farahzad et al. 2023 HLHS, DORV, malposed great arteries Chromosome 7 trisomy mosaic, skeletal abnormalities Uterine agenesis Farahzad et al. 2023 TGA, pulm atresia, VSD, right aortic arch Urinary tract abnormalities and skeletal abnormalities Bicornuate Farahzad et al. 2023 Tricuspid atresia Urinary tract abnormalities and skeletal abnormalities Didelphys Farahzad et al. 2023 HLHS, unbalanced AVSD Bardet-Biedel, urinary and skeletal malformations Unicornuate Farahzad et al. 2023 TGA, DILV, interrupted aortic arch Skeletal abnormalities Bicornuate Calaluce et al. 1995 [ 20 ] HLHS, infradiaphragmatic TAPVR heterotaxy Bicornuate Maaswinkel-Mooij and Stovkis-Brantsma 1992 [ 21 ] HLHS heterotaxy with polysplenia, horseshoe kidney Bicornuate vs unicornuate Kramer et al. 2022 [ 22 ] Sinus of valsalva aneurysm mosaic trisomy 13 Didelphys Scheurle 2008 [ 23 ] 5 patients with DiGeorge syndrome DiGeorge syndrome Didelphys, uterine agenesis Soekersi and Hernowo 2023 [ 24 ] Situs inversus MRKH Uterine agenesis Suri et al. 2007 [ 25 ] HLHS Meacham syndrome Bicornuate uterus Footnote: Abbreviations in order presented: ASD = atrial septal defect, PDA = patent ductus arteriosus, VUR = vesicoureteral reflux, VACTERL =  v ertebral defects, a nal atresia, c ardiac defects, t racheo- e sophageal fistula, r enal anomalies, l imb abnormalities, TEF = tracheo-esophageal fistula, SVC = superior vena cava, VSD = ventricular septal defect, PAPVR = partial anomalous pulmonary venous return, MRKH = Mayer-Rokitansky-Küster-Hauser syndrome, CHARGE =  c oloboma, h eart defects, a tresia of the nasal choanae, r estricted growth and/or development, g enital and/or urinary abnormalities, e ar abnormalities, WPW=Wolff-Parkinson White syndrome, AV = atrioventricular, TGA = transposition of the great arteries, PA/IVS = pulmonary atresia with intact ventricular septum, IVC = inferior vena cava, ccTGA = congenitally corrected transposition of the great arteries, HLHS = hypoplastic left heart syndrome, DORV = double outlet right ventricle, DILV = double inlet left ventricle, TAPVR = total anomalous pulmonary venous return. Summary of all CHD and CUA cases. Footnote: Abbreviations in order presented: ASD = atrial septal defect, PDA = patent ductus arteriosus, VUR = vesicoureteral reflux, VACTERL =  v ertebral defects, a nal atresia, c ardiac defects, t racheo- e sophageal fistula, r enal anomalies, l imb abnormalities, TEF = tracheo-esophageal fistula, SVC = superior vena cava, VSD = ventricular septal defect, PAPVR = partial anomalous pulmonary venous return, MRKH = Mayer-Rokitansky-Küster-Hauser syndrome, CHARGE =  c oloboma, h eart defects, a tresia of the nasal choanae, r estricted growth and/or development, g enital and/or urinary abnormalities, e ar abnormalities, WPW=Wolff-Parkinson White syndrome, AV = atrioventricular, TGA = transposition of the great arteries, PA/IVS = pulmonary atresia with intact ventricular septum, IVC = inferior vena cava, ccTGA = congenitally corrected transposition of the great arteries, HLHS = hypoplastic left heart syndrome, DORV = double outlet right ventricle, DILV = double inlet left ventricle, TAPVR = total anomalous pulmonary venous return. Other reported cases include the case series by VanDolah and Moe which presented three patients with complex CHD who were found to have CUA during and after pregnancy and delivery complications [ 9 ]. One patient had a history of heterotaxy, hypoplastic left heart syndrome (HLHS), and didelphys uterus, one patient with tricuspid atresia and bicornuate uterus, and one patient with congenitally corrected transposition of the great arteries (ccTGA) and bicornuate uterus. There are also many single case reports of patients with complex CHD and CUA, which are also summarized in Table 3 . These cases are again hypothesis-generating for an association between more complex CUA with complex CHD. Though the total number of patients with a CUA in our study is small, there are a few notable associations within the study population that warrant future investigation. Three of the four patients with partial anomalous pulmonary venous return (PAPVR) were noted to have CUAs. Fig. 2 highlights one of these cases of a patient with Scimitar syndrome and a didelphys uterus. When paired with the other two cases of heterotaxy syndrome and CUAs, there is a suggested association with disorders of laterality that may share an underlying embryologic pathophysiology. Fig. 2 14 year old female with exertional dyspnea undergoes cardiac CT angiography, demonstrating partial anomalous pulmonary venous return with the right lower pulmonary vein connecting to the IVC-RA junction, shown from posterior view on 3-D reconstruction (Panel A). She also is found to have right lung hypoplasia consistent with Scimitar syndrome. Later in adolescence, she has significant dysmenorrhea and menorrhagia despite oral contraception, prompting a pelvic ultrasound which suggests a congenital uterine anomaly. Pelvic MRI is obtained which confirms a uterus didelphys with dominant left hemiuterus on coronal view (Panel B). Fig. 2 14 year old female with exertional dyspnea undergoes cardiac CT angiography, demonstrating partial anomalous pulmonary venous return with the right lower pulmonary vein connecting to the IVC-RA junction, shown from posterior view on 3-D reconstruction (Panel A). She also is found to have right lung hypoplasia consistent with Scimitar syndrome. Later in adolescence, she has significant dysmenorrhea and menorrhagia despite oral contraception, prompting a pelvic ultrasound which suggests a congenital uterine anomaly. Pelvic MRI is obtained which confirms a uterus didelphys with dominant left hemiuterus on coronal view (Panel B). Also notable is the association of patients with VACTERL and CUAs, particularly when more complex CHD is present. The association of VACTERL with CUA is well-described. Fig. 3 highlights one of these patients with Tetralogy of Fallot with pulmonary atresia who was found to have uterine didelphys. This patient as well as the patient in Fig. 4 both highlight the number of CUAs that were not noted on initial imaging reads but were found on our re-evaluation for the purpose of this study. The patient highlighted in Fig. 4 is a particularly interesting example of this as she underwent heart transplant early in childhood, emphasizing the importance of recalling a transplant patient's index diagnosis and how that may impact extracardiac findings and symptoms later in life. Given the increasing interest in assisted reproductive technology and usage in patients after heart transplant, the possible presence of CUA carries implications for success of these therapies [ 19 ]. Fig. 3 1 month old female with VACTERL association and Tetralogy of Fallot with pulmonary atresia undergoes pre-surgical planning CT angiography with 3-D reconstruction showing absent main pulmonary artery with hypoplastic left and right pulmonary arteries in blue, and a large patent ductus arteriosus in green, which is being kept open by prostaglandin E infusion to provide pulmonary blood flow (Panel A). Later in life, she undergoes pelvic MRI for tethered cord pre-surgical planning, which initially is read as a normal uterus. However, on review by our radiologist for this study, she is found to have a uterus didelphys (Panel B). Fig. 3 Fig. 4 15 year old female with a history of hypoplastic left heart syndrome, valvular and subvalvular aortic stenosis, and mitral stenosis consistent with Shone complex (Panel A). She undergoes orthotopic heart transplant at age 5. In early adolescence, she develops dysmenorrhea and menorrhagia. Pelvic MRI obtained for orthopedic purposes is read as normal uterus, however on review for this study shows a septate uterus (Panel B). Fig. 4 1 month old female with VACTERL association and Tetralogy of Fallot with pulmonary atresia undergoes pre-surgical planning CT angiography with 3-D reconstruction showing absent main pulmonary artery with hypoplastic left and right pulmonary arteries in blue, and a large patent ductus arteriosus in green, which is being kept open by prostaglandin E infusion to provide pulmonary blood flow (Panel A). Later in life, she undergoes pelvic MRI for tethered cord pre-surgical planning, which initially is read as a normal uterus. However, on review by our radiologist for this study, she is found to have a uterus didelphys (Panel B). 15 year old female with a history of hypoplastic left heart syndrome, valvular and subvalvular aortic stenosis, and mitral stenosis consistent with Shone complex (Panel A). She undergoes orthotopic heart transplant at age 5. In early adolescence, she develops dysmenorrhea and menorrhagia. Pelvic MRI obtained for orthopedic purposes is read as normal uterus, however on review for this study shows a septate uterus (Panel B). Finally, we noted statistically significant associations with Trisomy 21 (which was negative) and with our created category of “other chromosomal abnormality”. It isn't suprising we saw Trisomy 21 was not associated with CUA, since Trisomy 21 is a well-described genetic syndrome and CUA is rarely concomitantly seen. The association with “other chromosomal abnormality” is interesting but difficult to interpret, given this category represented patients with genetic chromosomal abnormalities that are not well understood or classically categorized into more common genetic syndromes. Larger studies may be able to identify more specific associations that could guide screening for patients found to have the same abnormalities. As discussed above, future studies should be prospective to eliminate the inherent biases of our retrospective study and are needed to obtain the most accurate estimation of prevalence of CUA in the CHD population to accurately guide screening recommendations. We recommend that future study be done with pelvic ultrasound obtained specifically for CUA, to eliminate any referral bias and allow for close attention from the reading physician. Additionally, given the extremely heterogeneous nature of CHD diagnoses, it is difficult to make specific associations with diagnoses given the low overall numbers, making our findings chiefly hypothesis-generating and multicenter collaboration paramount. Finally, we recommend that future similar studies should compare all born females patients who would've qualified had they gotten adequate pelvic imaging, since this will help with understanding just how selected of a patient population is being examined.

This is the largest published study on the relationship between congenital heart disease (CHD) and congenital uterine anomalies (CUAs). While this study is too limited to determine true prevalence of CUAs in the CHD population, the results are hypothesis-generating for an association between CUAs and CHD that necessitates further larger multicenter prospective study, given the important clinical implications for reproductive health among CHD patients. Based on the results of this study, we believe it is reasonable to consider non-invasive screening of patients with congenital heart disease around menarche with pelvic US for CUA, particularly for patients with moderate or complex CHD and especially so for those with renal or anal anatomical malformations or VACTERL. Other important clinical takeaways include consideration of imaging prior to IUD placement in similar patients, the importance of careful imaging attunement towards CUA diagnosis, and recalling the index diagnosis for patients who have undergone childhood cardiac transplant given important later implications for extracardiac manifestations such as CUA.

Harrison VanDolah declares no industry relationships. Kathy Mostajeran declares no industry relationships. Luis Goncalves declares no industry relationships. Wayne Franklin declares no industry relationships. Melissa Parks declares no industry relationships. Tabitha Moe declares no industry relationships.

Congenital uterine anomalies (CUAs), also known as Müllerian duct anomalies, are the result of an abnormality of at least one of three processes during embryological development of the uterus: formation, fusion, or resorption. While the true prevalence of CUAs is difficult to ascertain, population estimates range from 0.5 % to 5.5 % [ 1 , 2 ]. CUAs present unique considerations for both obstetrical and gynecological care, with primary dysmenorrhea due to an obstructed uterine horn or transverse septum and intrauterine device mal-positioning and contraception failure as recognized complications [ 3 ]. Additionally, the obstetrical implications of CUAs often require specialized care, from episodes of recurrent pregnancy loss, premature delivery, fetal growth restriction, placental abruption, and an increased risk of cesarean delivery due to malpresentation of the fetus at term [ 4 , 5 ]. CUAs are best assessed with MRI or US given the ability to define both the internal and external contours of the uterus [ 6 ]. Patients with congenital heart disease (CHD), particularly of higher complexity, often have extracardiac manifestations [ 7 ]. However, the relationship between CUAs and CHD is not well-described, with only one small retrospective study in abstract form and one small case series that have been published looking at this relationship specifically [ 8 , 9 ]. Given the already high-risk nature of many of these patients from the obstetrical perspective as well as the known increased rates of gynecologic conditions, more precise data on the prevalence of CUAs in patients with CHD is needed to better guide pediatric and adolescent gynecology practices and preconception counseling [ 10 ].

Congenital heart disease Computed tomography Congenital uterine anomaly International classification for diseases, tenth edition Magnetic resonance imaging Vertebral anomalies, Anorectal anomalies (anal atresia), Cardiac anomalies, TracheoEsophageal fistula or atresia, Renal anomalies, Limb anomalies

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.