Spontaneous miscarriage in the first trimester

Miscarriage in the first trimester affects about 15% of all pregnancies. It can be classified into early fetal loss (previously called missed abortion or blighted ovum) and retained products of conception (previously called incomplete abortion). Traditionally, both types of first trimester miscarriage are treated by surgical uterine evacuation. Increasingly, however, women are offered medical management (using misoprostol) or expectant management (avoiding treatment and letting the miscarriage take its natural course).

There have been numerous studies of miscarriage management, but they have generally been underpowered to assess anything other than the need for unplanned uterine curettage.¹² The miscarriage treatment (MIST) study, a large, multicentre, randomised trial reported by Trinder and colleagues on p 1235, is therefore welcome.³

In the MIST study 1200 women with early fetal loss or retained products of conception were randomly allocated to receive expectant, surgical, or medical management. The primary outcome was infection, and there was no significant difference between the three groups when judged according to strict criteria. Women in the surgical group were, however, given significantly more prescriptions of antibiotics for clinically diagnosed gynaecological infections than the expectantly managed group.

In the surgical group most women had a curettage, compared with just under half in the expectant group. Most of the operations in the expectant group were for persistent retained products of conception diagnosed on ultrasound scan or were done at the request of the patient. In the misoprostol group around a third underwent curettage, mostly for failing to pass the products of conception eight hours after treatment, although some did not have surgery until two weeks later, when ultrasound showed persistent retained products of conception. The need for curettage after two weeks in the expectant management arm of this trial (50% with early fetal loss and 25% for retained products of conception) is similar to that reported in a large observational series.⁴ The need for curettage in the misoprostol arm, however—38% for early fetal loss and 29% for retained products of conception—is very high compared with other series, which report surgical evacuation rates of around 15% and 5% respectively.^5-7

The reason for the very high rates lies with the medical management protocol, which many would consider to be outdated, although it reflects the time when the trial was started, nine years ago. The dose and route of administration of misoprostol are consistent with current best practice,⁸ but the outcome of treatment was assessed too early. Participants were assessed only eight hours after taking misoprostol and offered surgical evacuation at that point if the products of conception had not been passed. In most studies women go home after completing the misoprostol protocol even if products have not been passed and have been reassessed at 7-14 days.⁶ This assessment strategy increases the success rate of medical management considerably. Indeed there is a crude correlation between success rate in various trials and time of assessment.⁹ The lesson for doctors offering medical management of first trimester miscarriage is that it is important to delay the clinical assessment of “failure” for at least seven days to prevent unnecessary surgical intervention.

Moreover, asymptomatic women who have retained products of conception or a thickened endometrium on ultrasound scan at follow-up are no longer offered surgical evacuation because in the absence of clinical symptoms these findings have no bearing on finally assessed success rates in spontaneous or induced miscarriage.¹⁰,¹¹ This may have also falsely increased the need for surgical evacuation in the medical and expectant management arms of the MIST study.

What about the harms associated with each type of management of miscarriage? Concerns that expectant management might reduce subsequent fertility have already been allayed.¹² In the MIST study, vaginal bleeding stopped significantly earlier in the surgical group than in the medical group and expectant management group, but this did not affect haemoglobin concentration and packed cell volume at 10-14 days. Other outcomes such as return to normal activities and rates of anxiety and depression were the same in all arms of the trial. Of concern is the rate of surgical complications (mainly uterine perforation), which was 2% in the curettage arm.

The primary outcome of the MIST study was infection, and the results are reassuring. Previous evidence indicates that surgical evacuation may increase infection rates after miscarriage. Medical management would usually be expected to have the lowest infection rates because it combines relatively rapid evacuation of the uterus with low surgical intervention rates. But rates of infection in all arms of the MIST study were low even without prophylactic antibiotics or bacteriological screening. The same may not be true in parts of the world with high rates of latent sexually transmitted infection or where many of those who present with incomplete miscarriage have attempted illegal abortion. In these settings many practitioners will continue to avoid expectant management and use prophylactic antibiotics for both surgical and medical evacuation in order to keep infection rates low.¹³

In areas without these problems a woman with spontaneous miscarriage in the first trimester should be offered a choice of management. Does she wish to have surgical evacuation that will provide a rapid resolution to the problem but has a 1:50 complication rate, or expectant management which—although unpredictable—will probably allow her to avoid admission to hospital? Medical management should speed up her miscarriage and reduce the likelihood of her having surgical intervention to around 5-15% depending on the type of miscarriage. In choosing, she can be reassured that the risks of infection and bleeding are low, whatever her decision.