The pregnancy outcomes of patients with rudimentary uterine horn: A 30-year experience.

Mullerian abnormalities are present in 0.17% of fertile women and 3.5% of infertile women, and unicornuate uterus is observed in 0.4% of women[ 1 – 3 ]. Approximately 84% of unicornuate uteruses have a contralateral rudimentary horn[ 4 ]. An embryo can implant in a uterus with a rudimentary horn or in a unicornuate uterus. Although these conditions are similar, their reproductive outcomes are completely different. Rudimentary horn pregnancy (RHP) is rarer still, with a reported incidence ranging from 1 in 76,000 to 1 in 150,000[ 5 ]. RHP results in the rupture of the horn by the second trimester in 80–90% of all cases. Only 14% of all cases are diagnosed before clinical symptoms occur[ 6 ]. Most cases of RHP provide a diagnostic challenge and are diagnosed after rupture, which leads to emergency surgery, blood transfusions, and increased morbidity[ 7 – 13 ]. Early diagnosis before rupture is essential for the successful management and prevention of maternal morbidity and mortality. The reproductive outcomes of women with unicornuate uteruses are poor; the associated live birth rate is only 29.2% and the prematurity rate is 44%[ 14 , 15 ]. Moreover, women with this anomaly present spontaneous abortion rates of 24.3% in the first trimester and 9.7% in the second trimester[ 1 ]. A lack of published data on unicornuate uterus pregnancy and RHP exists in the medical literature, and most of the available studies are case reports[ 1 , 13 , 15 – 18 ]. A limited number of studies have reported the clinical characteristics and reproductive differences between the subtypes of rudimentary horn according to the American Fertility Society classification (AFSC). The current study aimed to describe the presentations, assessments, treatments, and pregnancy outcomes of 22 women with a rudimentary horn, and we compared the reproductive performance of these groups.

Table 1 compares the characteristics and clinical data from Type A and Type B patients. The mean gestational age of Type A patients (23.5 weeks) was significantly higher ( P = 0.046) than that of Type B patients (10 weeks). Type A patients exhibited higher frequencies of abdominal pain (100% vs 42.9%) than did Type B patients, but the difference was not statistically significant ( P = 0.071). Mean age, gravidity, parity, ipsilateral renal agenesis, intra-abdominal hemorrhaging, and the rate of diagnosis before pregnancy were not significantly different between the two types. Table 2 shows the baseline characteristics of the patients with communicating and noncommunicating rudimentary uterine horns. Of the 22 patients, the mean age of patients in the noncommunicating group (29 years) was significantly higher ( P = 0.036) than the mean age of patients in the communicating group (26 years). Patients in the communicating group exhibited significantly higher frequencies of abdominal pain (100% vs 17.6%) than did the patients in the noncommunicating group, which is consistent with the results shown in Table 1 . Gravidity, parity, gestational age, ipsilateral renal agenesis, abnormal gestational history, surgery history, infertility history, intra-abdominal hemorrhaging, and the rate of RHP diagnosis before pregnancy were not significantly different between the two groups. The reproductive outcomes of the 22 patients are shown in Table 3 . The rudimentary uterine horn carried 4 of 5 (80%) pregnancies in the communicating group. Three pregnancies ruptured before a gestational age of 12 weeks, and one abortion occurred after 12 weeks. In the noncommunicating group, 7 of 17 (41.2%) patients presented with RHPs, and 3 pregnancies ruptured after a gestational age of 12 weeks. The remaining 4 unruptured pregnancies were diagnosed and managed before 12 weeks. However, the number of patients was too small to obtain a significant difference between two groups. Six patients suffered rupture of the pregnant horn with massive intra-abdominal hemorrhaging and shock. The mean intra-abdominal blood loss volume was 2,600±418.33 ml. All of these patients received multiple blood transfusions. In the patient who presented with an abdominal pregnancy, the pregnancy was ultimately confirmed as having developed from a ruptured rudimentary uterine horn. We performed laparotomy at 33 gestational weeks and observed a fragile placenta located in the ruptured rudimentary uterine horn, and a living neonate was retrieved from the peritoneum. The neonate weighed 1,855 g and had 1-minute and 5-minute Apgar scores of 10 and 10, respectively. The neonate had no gross congenital abnormalities. Eleven unicornuate uterus pregnancies were recorded, and 10 reached term. Only one preterm delivery occurred in the communicating group. This neonate weighed 1,150 g and had 1-minute and 5-minute Apgar scores of 10 and 10, respectively. A cesarean section was performed due to the preterm premature rupture of membrane (PPROM), an elevated C-reactive protein level and the patient’s firm request. The noncommunicating group included 9 neonates delivered at term. Certain maternal and fetal abnormalities were observed, including one neonate with macrosomia weighing 4,190 g, one neonate with oligohydramnios, one preterm delivery with premature rupture of membranes (PROM), one case of PROM, and one neonatal malformation (foot inversion). Overall, one-third of these pregnancies involved a breech presentation.

The diagnosis and management of the rudimentary uterine horn remains challenging. Many women with a rudimentary uterine horn present with acute uterine rupture during pregnancy. Early diagnosis is the key to successful management. Medical and radiological personnel must maintain a high degree of alertness to prevent the morbidity associated with this condition. In particular, patients with RHP (type A), who have a higher chance being misdiagnosed before 12 gestational weeks, have higher risks of potential complications. If an RHP is diagnosed, excision of the pregnant horn is recommended because the risk of rupture of an RHP in the second trimester is very high, regardless of the type of unicornuate uterus.

We reviewed the hospital data regarding the pregnancy outcomes of patients with a rudimentary horn who were managed at our institute, the Peking Union Medical College Hospital, over the last 30 years. Twenty-two pregnant patients with a rudimentary horn who presented at our institute from January 1, 1986, to December 31, 2016, were enrolled in the present study. We conducted this study on April 30, 2017. Patient age, gravidity, parity, diagnosis before pregnancy, abnormal gestational history, surgery history, estimated blood loss, gestational weeks, abdominal pain and vaginal bleeding at presentation, and abdominal bleeding were recorded. Abdominal pain defined as severe,acute pain of abdomin, and it is one of the main complaints of patients. Uterine anatomy was evaluated using a routine vaginal or abdominal ultrasound before or after the patient became pregnant. Some patients were underwent hysterosalpingography, laparoscopy or were diagnosed by macroscopic detection during cesarean section. Patients with RHP were managed by excision of the rudimentary horn combined with ipsilateral salpingectomy. Patients with a unicornuate uterus pregnancy were diagnosed with unicornuate uterus pregnancy featuring a contralateral rudimentary horn upon cesarean section. Patients with RHP were divided into the following two groups according to the AFSC classification: A rudimentary horn with a cavity that communicates with the uterus. A rudimentary horn with a cavity that does not communicate with the uterus. The other 11 patients with a unicornuate uterus pregnancy were definitively diagnosed with a rudimentary uterine horn during cesarean section. If the horn did not communicate with the main cavity of the contralateral hemiuterus, then we were unable to confirm that it contained a functional endometrium. Thus, we classified all 22 patients into communicating group or noncommunicating group according to the anatomical connection of the rudimentary horn to the contralateral hemiuterus (see S1 Table ). The local ethical committees of Peking Union Medical College Hospital granted a waiver of approval for this retrospective study, and patients provided informed consent for the use of their medical records in retrospective studies prior to surgery. Nonnormally distributed continuous variables were compared using the Mann-Whitney U test and the Kruskal-Wallis test. Two-tailed P -values are reported, and the alpha for all tests was set to 0.05. All statistical analyses were performed using SPSS 12.0 (SPSS, Inc., Chicago, IL, USA).

(XLSX) Click here for additional data file.