Resolution of ectopic pregnancy

• Very low quality evidence from four randomised controlled trials (n=236) did not demonstrate any clinically important difference in the resolution of ectopic pregnancy between those who received expectant or medical management. Subgroup analyses (by hCG levels or embryo size at presentation) provided moderate to very low quality evidence which did not detect a clinically significant difference between treatment arms.

Tubal rupture

• Low quality evidence from two randomised controlled trials (n=96) did not demonstrate any clinically important difference in tubal rupture rate between those who received expectant or medical management (no events in either group).

Need for additional treatment

• Very low quality evidence from four randomised controlled trials (n=236) did not demonstrate any clinically important difference in the need for additional treatment between those who received expectant or medical management. Subgroup analyses (by hCG levels or embryo size at presentation) provided moderate to very low quality evidence which did not detect a clinically significant difference between treatment arms.

Health-related quality of life

• Moderate to low quality evidence from a single randomised controlled trial (n=57) did not demonstrate a clinically important difference in health status (as measured by the short-form 36 [SF-36] and Rotterdam symptom checklist), depression or anxiety (as measured by the Hospital Anxiety and Depression Scale) between those who received expectant or medical management.

The outcomes that matter most

The committee identified 3 outcomes of critical importance: maternal mortality, resolution of tubal ectopic pregnancy, and rupture rate. These 3 outcomes were selected as critical since they provide direct evidence about the effectiveness of the interventions in resolving an ectopic pregnancy without leading to adverse events. Additionally, the committee identified the need for additional treatments, future ectopic pregnancy rates, future fertility, and patient satisfaction as important outcomes.

The quality of the evidence

Four randomised controlled trials were included in this review. The quality of the evidence was assessed according to GRADE criteria and ranged from very low to moderate quality evidence. The main reason for downgrading was imprecision – the trials had few participants, and therefore the confidence intervals for the estimates were wide. Some of the trials were also downgraded because of high to very high risk of bias. This was assessed with The Cochrane Risk of Bias Tool. The main sources of potential bias were: lack of information regarding how the randomisation was performed or concealed; or because women, clinicians and/or outcome assessors were aware of treatment allocation. Two of the trials had not registered their protocol, therefore were downgraded for high risk of reporting bias. There was no evidence available for the outcomes on maternal mortality, future ectopic rates or future fertility rates.

Benefits and harms

The evidence did not show any significant difference between expectant or medical management for ectopic pregnancy resolution, tubal rupture prevention, additional treatment requirements, or health-related quality of life. The committee therefore agreed that expectant management could be offered based on the clinical suitability of a woman, an assessment of the risks and benefits, and the preferences of the woman. The committee agreed that women who were suitable for expectant management were similar to the inclusion criteria in the clinical studies – for example those women who were clinically stable without pain, who had a small ectopic pregnancy, and who had low serum hCG levels (1500 IU/L or lower). Although the inclusion criteria of the four studies permitted women with a range of hCG levels to enter the trials, the committee noted that the majority of participants had relatively low levels of hCG (typically <1000 IU/l). The committee therefore considered that the strongest evidence for the comparison of expectant and medical management was in women with hCG levels <1000 IU/L. Expectant or medical management were both thought to be entirely suitable options to offer these women. The committee considered that expectant management may also be suitable for women with higher hCG levels (up to 1500 IU/L), but there was less evidence to support this. Therefore they made a recommendation that expectant management could be considered for women with hCG levels above 1000 IU/L but below 1500 IU/L.. Similarly, the two studies which reported on the size of the adnexal mass showed that most participants had an adnexal mass of <35mm, therefore this was considered a reasonable threshold to recommend expectant management. All studies in this review excluded women with an ectopic pregnancy with fetal heart activity, therefore the use of expectant or medical management in these women has not been assessed. The committee noted that their clinical experience also supported these thresholds as reasonable for the use of medical or expectant management, and reflected these in the recommendations.

In terms of follow-up care, the committee agreed that serum hCG levels should be carefully monitored regardless of the treatment choice, to ensure they were falling. The committee were aware of a number of studies that had defined a meaningful drop in hCG to be 15% and so they adopted this value. Based on their experience the committee were aware that hCG levels should be checked after 2 days, 4 days and then again at 7 days. If hCG levels plateau or rise, the women should be reviewed by a senior gynaecologist, and a discussion with the woman about other treatment options may be needed.

Based on their clinical expertise, the committee outlined some risks and benefits that should be considered when discussing expectant management with women. The committee outlined that the main benefits of expectant management included a similar rate of resolution of ectopic pregnancy compared to medical management with methotrexate, while avoiding the side effects of methotrexate, such as nausea, anaemia, vomiting or diarrhoea, potentially mild abnormalities in liver and renal function tests, and the need to avoid pregnancy for 3 months. A disadvantage of expectant management is that women may need to be urgently admitted into hospital if their clinical condition worsens, although this may also be the case for women who have received methotrexate.

As there was no evidence available from this review regarding future fertility/pregnancy rates, the committee based the recommendations relating to this on their clinical knowledge and expertise. In addition, there was no evidence relating to the time for resolution of an ectopic pregnancy following medical or expectant management, but the committee were aware that the time was similar in clinical practice and so included this in their recommendations.

The committee noted that healthcare professionals counselling women with an ectopic pregnancy should be sensitive to the woman’s emotions, but did not make a separate recommendation about this as it is already covered in the support and information giving section of the guideline. An ectopic pregnancy can be devastating news and some women experience the same grief as when losing a family member. The committee were also aware that some women consider medical management with methotrexate as a type of abortion and express feelings of guilt. While of course equating treatment of ectopic pregnancy to terminating a pregnancy is not accurate, offering an alternative treatment route of expectant management if clinically appropriate can help such women from an emotional perspective.

Resolution of ectopic pregnancy

Note: The subgroup analysis for hCG 501-1000IU/L was noted to have I²= 55%, therefore a random effects model was considered. However, the same subgroup analysis for the ‘reversed’ outcome (need for additional intervention) was identified as 0%. Therefore the moderate heterogeneity was noted, but a random effects model was not used for this analysis. (more...)

Additional treatment needed