Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Methotrexate for ectopic pregnancy


What can you do when you want to compare two treatments, but there are no randomised trials, no comparative trials, and only observational data? A meta-analysis of two treatment regimens for methotrexate in ectopic pregnancy [1] reminds us of some of the problems. These are likely to be in the type of patient treated, and what the treatment actually is.


The focus for the review was two main medical treatments that use methotrexate for ectopic pregnancy. One involves the use of a single dose of methotrexate, usually based on an injected dose of about 50 mg/sq metre. The alternative is to use 0.1 mg/kg intramuscularly on alternate days for up to four doses.

Searching used a single database for English-only studies. Included were those reporting the number of patients treated with either of the two standard protocols. Nonstandard dosing protocols, tiny studies with fewer than 10 cases, and studies examining interstitial, cervical or ovarian pregnancy were excluded. Studies were rated according to quality and completeness of data reporting, design, diagnosis reporting, and inclusion and exclusion criteria.

Information was available on individual patients. The outcome was failure of treatment defined as abandonment of medical management in favour of surgical management. Information on failure or success was collected together with human chorionic gonadotrophin (hCG) level, the presence of embryonic cardiac activity, the protocol, the number of doses, adverse events and hospital admission. Logistic regression was used to determine associations between outcome and factors.


Information was available on 1,327 cases of women diagnosed with ectopic pregnancy from 26 studies, treated with methotrexate, mostly (80%) with a single dose regimen. Individual studies reported on 12 to over 300 women. Five of the studies included duplicated information on at least some women, and this duplicate information was not used.

The actual number of doses used was not always in accord with the protocol (Figure 1). More than a single dose was used in 15% of women in whom a single dose was planned, and a single dose only was used in 10% of women in whom it was planned to use multiple doses.

Figure 1: Number of doses of methotrexate used in the two regimens

Failure of medical management occurred in 11% of women treated with a single dose and 7% of women treated with multiple doses. Women treated with multiple doses had a significantly higher hCG level, and increasing hCG was significantly associated with failed treatment. The presence of embryonic cardiac activity was significantly associated with the failure of medical treatment. A statistically significantly higher failure rate was seen for single dose compared to multiple dose regimen, particularly when the analysis was controlled for the hCG value and the presence of embryonic cardiac activity.

There was no difference in hospital admission (12%), abdominal pain (22%) or other adverse event rates (33%) when the analysis was adjusted for the hCG level.


Overall, the success rate was 89%, with adverse events were minor and self-limiting. Only 1 woman in 10 treated with methotrexate needed surgery.

Women treated with single dose methotrexate were more likely to fail medical management of ectopic pregnancy than those treated with multiple doses. But women treated with single doses had lower hCG levels, and lower hCG was independently associated with lower failure rates. Clinicians may have been using the single dose regimen for women with good prognosis and using the multiple dose regimen preferentially for women with poorer prognosis, or who were more advanced in their pregnancy. They may have been using the right treatment for the right patient.

Determining whether there is a true difference in efficacy or harm proved to be extremely difficult, even with a good analysis of a reasonable number of individual patients. Perhaps what we learn most from this interesting meta-analysis, apart from confirming the utility of randomised trials from their absence, is that this treatment is effective and that clinicians seem to be making the best choice for their patients. But the optimum dosing may be more than one, and not more than four doses.


  1. KT Barnhart et al. The medical management of ectopic pregnancy: a meta-analysis comparing single dose and multidose regimens. Obstetrics & Gynecology 2003 101: 778-784.

previous or next story