The authors report no financial relationships relevant to this article.
CASE 1: Pelvic pain and few signs of intrauterine pregnancy
A 24-year-old woman, para 1-0-0-1, visits the hospital emergency department complaining of pelvic pain. She says the pain arose suddenly and reports that she had a positive urine pregnancy test earlier in the week. When asked about her obstetric history, she reports vaginal delivery of an 8 lb, 8 oz infant at 38 weeks’ gestation 2 years earlier. Her human chorionic gonadotropin (hCG) level is 3,000 mIU/mL, but ultrasonography (US) reveals no evidence of pregnancy. She is discharged with instructions to follow up with her physician in 2 days.
When her abdominal pain worsens, she returns to the emergency department. Physical examination reveals significant tenderness of the abdomen and moderate to severe tenderness of the cervix upon motion. Transvaginal US shows a uterus of normal size with a 5-mm endometrial lining and no gestational sac. The patient’s abdomen is full of fluid, with large, hypodense areas adjacent to the uterus bilaterally but larger on the right. The preoperative diagnosis: ruptured ectopic pregnancy.
During the diagnostic laparoscopy that follows, approximately 500 mL of blood is discovered in the abdomen and pelvis, and a gestational sac is found to be densely adherent to the right pelvic sidewall, where the ureter nears the uterine vessels. The sac, which has partially separated from the sidewall, is bleeding.
The surgeon peels the sac off the sidewall and controls bleeding with electrocautery and liquid thrombin. The final pathology report describes the tissue as an organizing blood clot with trophoblasts, consistent with ectopic pregnancy.
At a follow-up visit 3 weeks later, the patient reports persistent symptoms of pregnancy. Repeat US reveals a twin intrauterine pregnancy with two sacs, only one of which has a heartbeat. One week later, US shows confluence of the sacs, with a single viable fetus at 8 weeks and 2 days of gestation.
Could heterotopic pregnancy have been diagnosed earlier?
This case illustrates challenges inherent in the diagnosis of heterotopic pregnancy, which is much more common today than it was when it was first described 300 years ago. Incidence has increased from approximately 1 in 30,000 pregnancies to 1 in 2,600 pregnancies annually. When assisted reproductive technologies (ART) are used, the incidence may be as high as 1 in 100 pregnancies.1
The rising incidence suggests that the diagnosis of intrauterine pregnancy can no longer be used to exclude the presence of ectopic pregnancy, and vice versa. Instead, steps must be taken to rule out both when a woman exhibits pain and signs of pregnancy.
In this article, we discuss the causes, diagnosis, and treatment of heterotopic pregnancy, including the necessity of a high index of suspicion, the unreliability of US imaging in 50% of cases, and the need to avoid curettage in the treatment of ectopic pregnancy until an empty uterus can be confirmed.
The literature on heterotopic pregnancy is slight
Duverney was the first to report heterotopic pregnancy, in 1708, after finding an intrauterine pregnancy during the autopsy of a woman who had died from a ruptured ectopic pregnancy.1 It was 165 years, however, before the first review of the phenomenon was written.2 By 1970, only 479 such cases had been reported.20
Determining incidence remains a challenge—except that it is rising
In 1948, DeVoe and Pratt calculated the incidence of heterotopic pregnancy by multiplying the incidence of two-ovum pregnancy by the incidence of ectopic pregnancy, reaching an estimate of 1 in 30,000 pregnancies annually.2
In 1965, Rothman and Shapiro found that only about 500 cases of heterotopic pregnancy had been reported cumulatively.7 They reasoned that, if the incidence of fraternal twins is 1 in 110 and the incidence of ectopic pregnancy is 1 in 250, heterotopic pregnancy should occur at a rate of 1 in 27,500 gestations.7 They also speculated that many heterotopic pregnancies go undiagnosed because of early pregnancy loss.7
In 1971, Payne and colleagues hypothesized that ovulation-induction agents increase the incidence of heterotopic pregnancy, and McLain and associates reached a similar conclusion in 1987.9 The incidence of multiple pregnancy after oral ovulation induction is 8% to 10%, and it is 20% to 50% with injectable agents.9 In 1994, Crabtree and associates reported that both abdominal and heterotopic pregnancies appear to be increasing in incidence.17
Today, the incidence of heterotopic pregnancy is thought to be about 1 in 2,600 pregnancies annually—primarily because of assisted reproduction.2 The calculated risk of heterotopic pregnancy is 1 in 119, and it rises to 1 in 45 with embryo transfer.8