Archives of Gynecology and Obstetrics

, Volume 290, Issue 3, pp 417–423

Methotrexate for ectopic pregnancy: when and how


  • Gustavo Nardini Cecchino
    • Department of ObstetricsFederal University of São Paulo (UNIFESP)
    • Department of ObstetricsFederal University of São Paulo (UNIFESP)
  • Julio Elito Júnior
    • Department of ObstetricsFederal University of São Paulo (UNIFESP)

DOI: 10.1007/s00404-014-3266-9

Cite this article as:
Cecchino, G.N., Araujo Júnior, E. & Elito Júnior, J. Arch Gynecol Obstet (2014) 290: 417. doi:10.1007/s00404-014-3266-9



Ectopic pregnancy is the leading cause of maternal death in the first trimester of pregnancy. The dosage of beta fraction of human chorionic gonadotropin (beta-hCG) and improvement of the transvaginal ultrasound allowed an earlier diagnosis and a conservative management. Currently, the use of systemic methotrexate (MTX) proved to be a great alternative with similar success rates and completely non-invasive.


We searched for the most relevant articles on the use of MTX in ectopic pregnancy published between 2003 and 2013 in high-impact journals. We performed a strategic search at the Centre for Reviews and Dissemination (CRD), Database of Abstracts of Reviews of Effects (DARE), National Institute for Health Research (NHS), International Prospective Register of Systematic Reviews (PROSPERO), The Cochrane Database of Systematic Reviews (CDSR) and Medical Literature Analysis and Retrieval System Online (MEDLINE) according to the descriptors “pregnancy, ectopic” and “methotrexate”, alone or combined.


Thus, we based this review on 32 studies that were classified following the grades of recommendation and levels of evidence proposed by the Oxford Centre for Evidence-Based Medicine. Additionally, selected papers were used. Scientific evidence points to a growing trend in the choice of conservative treatment for ectopic pregnancies, whereas expectant management still lacks studies for definitive conclusions. Indeed, the well-established protocols which exhibit a greater number of studies are still based on the single-dose treatment.


Considering MTX, it proved to be more effective in cases of low titers of beta-hCG and masses with a small diameter, although there is still no uniformity of these parameters. The choice largely depends on the experience of the medical team and ultimately, on the woman’s reproductive desire.


Ectopic pregnancyMethotrexateTherapeuticReproductive health


In spite of new treatment modalities and improvement of diagnostic methods, ectopic pregnancy (EP) is still the leading cause of maternal death in the first trimester of pregnancy, accounting for 6–13 % of all pregnancy-related deaths [15]. In industrialized countries, up to 2 % of all pregnancies are ectopic in location [1, 68]. Therefore, EP remains a major public health issue, considering that there has been a slight upward trend on its incidence over the past years. The contributing aspects are related not only to the increased cases of pelvic inflammatory disease, especially Chlamydia trachomatis infections, but also to the use of assisted reproductive technologies, and to the prevalence of smoking among women of reproductive age [911]. Following the first episode of ectopic pregnancy, the estimated risk of recurrence is up to 15 % and at least 25 % among women with two or more previous episodes [1215].

The most frequent location and the main focus of this article is the tubal ectopic pregnancy [16]. Other localizations of atypical ectopic pregnancies include: the cervix (0.15 %), the ovary (0.15–3 %), the abdomen (1.3 %), the interstice (2.5 %) and the cesarean scar (6 %) [17]. This group represents less than 10 % of all ectopic pregnancies but is associated with greater morbidity [18]. Traditionally, these rare ectopic pregnancies have been lately diagnosed and managed by open surgery. Advances in ultrasound technology have led to an increase in early diagnosis of non-tubal ectopic pregnancies, which allowed them to be managed through minimally invasive access [8, 17, 19].

Currently, therapeutic options for tubal pregnancy include both surgical and nonsurgical approaches, such as medical treatment with systemic methotrexate (MTX) and even expectant management [68, 19, 20]. However, there are few controlled trials and lack of evidence defining the criteria on whether intervention should be conservative or radical [10, 21]. In selected cases, the use of MTX proved to be cost-effective and showed similar results regarding success rates and future fertility [10, 16, 22]. Moreover, drug-based therapy avoids the inherent risks of surgery and anesthesia [1, 20]. The main drawback is the possibility of failure, which may require emergency surgical treatment afterward [20, 23].

MTX is a folic acid antagonist that inactivates dihydrofolate reductase and de novo synthesis of purines and pyrimidines, and therefore, cellular DNA. Thus, methotrexate acts on rapidly dividing trophoblastic cells and prevents their multiplication [1, 2, 8]. Considering the risk of persistent trophoblastic tissue, a rigorous monitoring of serum beta fraction of human chorionic gonadotropin (beta-hCG) is mandatory [68].

Despite being widely used, there is no established consensus concerning the optimal candidates for MTX treatment, the preferential route of administration of the drug or the post-treatment follow-up. There are two commonly used protocols: single dose or multiple doses of MTX [3, 9]. Recently, new studies have been conducted as an attempt to establish a more secure regimen, with similar efficacy and minimal side effects. For this purpose, some researchers developed clinical trials with a double-dose protocol or combining MTX with other drugs [3, 24, 25].

When it comes to ectopic pregnancy of atypical location, other methods have been described besides the use of isolated systemic MTX, including the ultrasound-guided application of local MTX in the gestational sac, conventional embolization of the uterine artery with or without the use of transcatheter intra-arterial MTX (chemoembolization), dilation and curettage, or even hysterectomy in the failure of conservative treatments. Such modalities have been performed especially in cases of cesarean scar pregnancy and cervical pregnancy [2634].

Given the diverging information as to the applicability of MTX, the aim of this review was to gather the most recent and qualified evidence regarding the multiple aspects on the use of this drug in ectopic pregnancy to guide daily challenges on the subject and also to assess state of the art.

Methods and results

We searched for the most relevant articles on the use of MTX in EP published between 2003 and 2013 in high-impact journals. Only articles published in English and Portuguese were selected for this review, according to the descriptors “pregnancy, ectopic” and “methotrexate”, alone or combined. Using these keywords separately at the Centre for Reviews and Dissemination (CRD), a total of 111 revisions were found within the Database of Abstracts of Reviews of Effects (DARE) and the National Institute for Health Research (NHS). After reading the title or abstract, and considering the most appropriate articles on the subject, 5 reviews of interest were gathered. The same strategy was implemented on the basis of the International Prospective Register of Systematic Reviews (PROSPERO) that listed a total of 12 revisions but none was selected after reading the titles and abstracts. In contrast, while searching The Cochrane Database of Systematic Reviews (CDSR), 35 revisions and 27 trials were obtained. A careful analysis of the titles and abstracts allowed the selection of a single systematic review and 7 trials of effect. Finally, we also performed a strategic search in the Medical Literature Analysis and Retrieval System Online (MEDLINE) by combining the two descriptors and applying filters, which allowed the selection of articles with the highest level of evidence. The filters used (inclusion criteria) were studies conducted in humans, systematic reviews, trials, meta-analysis, and multicentre trials. A total of 36 articles were obtained and 29 of these were used after thorough reading of the title and summary. Of the above-mentioned articles, 10 were already obtained from the other searched databases. Thus, we based this review on 32 studies that were classified following the grades of recommendation and levels of evidence proposed by the Oxford Centre for Evidence-Based Medicine [35]. Additionally, selected papers were used. In Fig. 1, we present a flowchart of search strategy.
Fig. 1

Flowchart showing the search strategy


Recently, Mol et al. did a systematic review and meta-analysis demonstrating the current evidence regarding the surgical treatment, medical or expectant management in cases of tubal pregnancy. Laparoscopic surgical approach was the most cost-effective method, but in carefully selected cases, the use of systemic MTX proved to be a great alternative with similar success rates, and it is completely non-invasive [6].

Concerning the expectant management, there are very few published articles to date. Hajenius et al. performed a randomized study comparing expectant management with MTX treatment and observed that 60 % of women had an uneventful clinical course after receiving expectant management. They also ended up indicating surgery on 1/41 (2 %) patient in the MTX group and 4/32 (13 %) in the expectant management group. The single indication for surgical intervention was the complaint of persistent pain. No tubal rupture was observed on those women. These findings might indicate that MTX could be withheld in selected cases [36]. Upcoming results from ongoing trials may clarify actual aspects related to the expectant management [37, 38].

The inclusion and exclusion criteria considered in the choice of drug treatment are still controversial [8]. Most authors agree on the use of MTX preferentially in cases of hemodynamic stability, absence of fetal heartbeat, and no signs of rupture [5, 8, 11, 39]. They also emphasize the importance of excluding evidence of allergy to MTX, as well as kidney, liver and hematologic diseases, in addition to the need for informed consent and rigorous follow-up after the drug use [5, 11, 39].

The most important parameter for drug therapy with MTX in EP is the beta-hCG. It is known that success rates decline with increasing beta-hCG levels. In a systematic review of 503 patients treated with MTX, Menon et al. found that the rates of treatment failure are substantially and statistically greater if the initial values of beta-hCG exceed 5,000 mIU/mL. In this study, the results indicate that for every 10 treatments of patients with such exceeding beta-hCG levels, there will be at least one more failure [40]. More recently, Sagiv et al. published the results of a long non-randomized clinical trial in which 238 patients were treated with MTX. A logistic regression analysis showed that levels of beta-hCG greater than 2,000 mIU/mL increased the odds of failure by about 4.5 times [41]. As the differences between the criteria used in determining the cutoff value are not always explicit in different studies, it is very difficult to compare the results.

The maximum size of the mass evaluated on transvaginal ultrasound that can be treated with MTX is also a diverging point among experts. While some authors only admit drug treatment in masses with a diameter up to 3.5 cm [11], others extend the use for masses larger than 3.5 cm [8, 39]. In 2003, Cobellis et al. [5] reported a success rate of 91 % using MTX for masses of up to 5 cm and beta-hCG levels of up to 10,000 mIU/mL. In contrast, another prospective observational study revealed a worse prognosis and higher rates of surgical conversion for masses greater than 3 cm [19].

Several predictors of failure of drug therapy have been identified over the years, and the most common are embryonic heart activity, size and volume of the mass greater than 4 cm, initial concentration of beta-hCG greater than 5,000 mIU/mL, presence of blood in the peritoneal cavity, increasing rate of beta-hCG above 50 % within 48 h prior to MTX, besides rapid and continuous increase of the beta-hCG level during MTX therapy (Table 1) [11, 39]. Through a prospective observational study, da Costa Soares et al. [42] demonstrated that an increase in beta-hCG concentrations lower than 11.1 % in 48 h resulted in a success rate of 86 % using MTX. In the same article, it became apparent that a significant and rapid increase in the beta-hCG level in 48 h imposes an additional risk for therapeutic failure.
Table 1

Predictors of methotrexate treatment failure

Embryonic cardiac activity

Size and volume of the gestational mass (>4 cm)

High initial beta-hCG level (>5,000 mUI/mL)

Presence of free peritoneal blood

Rapidly increasing beta-hCG titers before methotrexate (>50 %/48 h)

Continued rapid rise in beta-hCG titers during methotrexate

More recent studies correlate endometrial thickness with the indices of success or failure of MTX therapy. In the first published studies, da Costa Soares et al. found mean values of 6.4 mm for endometrial thickness and 1,936.2 mIU/mL of beta-hCG in cases of success with drug therapy. In cases of failure, the mean values of endometrial thickness and beta-hCG were 11.7 mm and 6,831.3 mIU/mL, respectively [20]. Subsequently, Takacs et al. [43] in a cohort of 73 patients showed that the group with endometrial thickness above 12 mm had significantly higher failure rates. These findings suggest that endometrial thickness reflects the hormonal levels and that the higher the level of beta-hCG, the higher endometrial thickness and the worse the prognosis of MTX therapy becomes [20, 43].

Despite new perspectives and recent studies with different suggestions on the use of MTX, most of the systematic reviews and of the randomized trials available used the protocol of a single dose or multiple doses, as indicated in Table 2 [1, 4, 9]. The largest meta-analysis comparing these two protocols dates from 2003, when Barnhart et al. analyzed data from 26 articles totaling 1,327 patients treated with MTX. The overall success rate reported was 89 %, and when the two groups were evaluated separately, the success rates of the group of women treated with single-dose protocol was 88.1 %, while patients treated with multiple doses succeeded 92.7 % of the time. However, it is important to note that none of the studies included in this analysis were controlled or blind [1].
Table 2

Comparison of the single-dose and multi-dose protocol of methotrexate


Single dose

Multiple doses



50 mg/m2

1 mg/kg



0.1 mg/kg

Route of administration




One dose: if the difference of beta-hCG on days 4 and 7 <15 % repeat weekly until undetectable (maximum of 3 doses)

Up to four doses of each (methotrexate and leucovorin) in alternate daily doses until serum beta-hCG declines by 15 %

Monitor beta-hCG

Day 0, day 4, and day 7

Day 0 and before each injection of methotrexate until beta-hCG declines 15 % from the previous value

Surveillance beta-hCG (after initial treatment)

Weekly until beta-hCG undetectable

Weekly until beta-hCG undetectable

The efficacy and safety of MTX in cases of tubal pregnancy have been widely demonstrated [44]. In the literature, the success rate of MTX ranges from 75 to 96 %, regardless of the mode of administration [9]. Two prospective randomized trials have failed to identify statistically significant differences between groups receiving MTX as a single dose or in multiple doses [4, 9]. Even though there have been no dose-finding studies on the administration of MTX, several groups have reported good success rates of between 85.4 and 98.7 % using low-dose MTX therapy [16]. The single-dose protocol is the most commonly used because it requires fewer days of hospitalization and has few side effects [9]. In tubal pregnancy with titers of beta-hCG lower than 5,000 mIU/mL, the treatment with a single dose of MTX is preferred. In cases of atypical location (cervical, cesarean scar, interstitial portion of the tube and ovary) without embryonic cardiac activity and elevated beta-hCG levels, the protocol with multiple doses must be employed [11, 39].

There are many different approaches to the management of rare ectopic pregnancies. As of now, no universal guideline has been published. Series of cases of cesarean scar and cervical pregnancies demonstrated outstanding results combining transvaginal ultrasound-guided sac aspiration under general anesthesia with a local 16- to 22-gage needle injection of MTX (varying doses of 1 mg/kg or standard single dose of 50 mg), an advantage comparing to 70 % of failure using uterine curettage [29, 33, 45, 46]. Adjunctive techniques for controlling hemorrhage such as cervical cerclage and uterine artery embolization should also be considered as options for urgent assistance [34]. Surgery remains as the preferred treatment option for ovarian, cornual and abdominal ectopic pregnancies, including cases of rupture, hemodynamic instability, gestational sac diameter greater than 4 cm (late diagnose) or persistent pain [17, 18].

Until recently, beta-hCG was the only parameter capable of predicting the success of the single-dose protocol [47]. However, new parameters are being proposed in different experimental studies, as demonstrated in a prospective observational study conducted in France in 2011, which revealed a new predictor of success for the single-dose protocol: the dosage of creatine phosphokinase (CPK) on the day of the first injection. According to the analysis of the receiver operating characteristic (ROC) curve, authors identified that CPK levels above 109 could predict success rates with 100 % specificity and 100 % positive predictive value. Thus, all patients with CPK levels above the cutoff mentioned responded to treatment with only one injection of the drug [23].

In this context, it is important to remember the contraindications and side effects of MTX. Among the absolute contraindications, the most important include intrauterine pregnancy or breastfeeding; evidence of immunodeficiency; hepatic, renal, or hematologic dysfunction; pulmonary disease or active peptic ulcer disease, as well as sensitivity to MTX. Relative contraindications include those already mentioned: detectable heartbeat, inability to follow-up, and initial concentrations of beta-hCG above 5,000 mIU/mL. The main side effects reported on the use of MTX are related to the gastrointestinal tract, such as nausea and vomiting. Nevertheless, there are reports of pneumonia, alopecia, elevation of liver enzymes, skin lesions, and others [11, 39].

Finally, aspects related to future fertility of women treated with MTX, which can be determined directly by a subsequent spontaneous pregnancy, and indirectly through hysterosalpingography, should be pointed out [48, 49]. Some studies have shown that treatment with MTX does not affect the tubal patency or even the ovarian reserve [49, 50]. Elito et al. [49] demonstrated a rate of up to 84 % of patency for the ipsilateral tube and 97 % of the contralateral tube. However, there is a proportional relationship between high levels of beta-hCG (>5,000 mIU/mL) treated medically and higher risk for tubal obstruction [49]. The cumulative ectopic pregnancy rate after MTX treatment on Gervaise’s study was 15.4 % after 1 year and 23.7 % after 2 years [13], diverging from other studies with rates as low as 2.7 % [14]. Despite the treatment modality there is no severe impact on women’s subsequent reproductive performance and long-term fertility outcome [49, 50].


The evidence points to a growing trend in the choice of conservative treatment for ectopic pregnancies. Expectant management still lacks studies for definitive conclusions. Indeed, the well-established protocols that exhibit a greater number of studies are still based on the single-dose treatment. Considering MTX in tubal pregnancy, it proved to be more effective in cases of low titers of beta-hCG and masses with a small diameter, although there is still no uniformity of these parameters. Additionally, non-tubal ectopic pregnancies are associated with greater morbidity and may require medical treatment with direct injection or systemic MTX, which have been used effectively. Clinical treatment avoids surgeries that would have otherwise been essential in cases of EP of unusual localization. Above all, the choice largely depends on the experience of the medical team and ultimately, on the woman’s reproductive desire. Counseling about the chances and possible consequences of the next pregnancy is important as well as clarifying that either treatment modality will not have a direct impact on her subsequent reproductive performance.

Conflict of interest


Copyright information

© Springer-Verlag Berlin Heidelberg 2014