By John C. Hobbins, MD

Professor, Department of Obstetrics and Gynecology, University of Colorado School of Medicine, Aurora

Dr. Hobbins reports no financial relationships relevant to this field of study.

SYNOPSIS: A recent randomized study suggested that vaginal progesterone is at least as good as, and may be superior to, intramuscular 17 alpha-hydroxy progesterone caproate in preventing recurrent preterm birth, but shortcomings in the study indicate that more investigation is needed.

SOURCE: Saccone G, Khalifeh A, Elimian A, et al. Vaginal progesterone vs intramuscular 17 alpha-hydroxy progesterone caproate for prevention of recurrent spontaneous preterm birth in singleton gestations: Systematic review and meta-analysis of randomized controlled trials. Ultrasound Obstet Gynecol 2017;49:315-321.

Preterm birth (PTB) is the leading cause of perinatal morbidity and mortality, and until 2003, when Meis et al1 demonstrated benefit in preventing PTB with 17 alpha-hydroxy progesterone caproate (17 P), no real preventive options had been available. Vaginal progesterone came on the scene later.2 Although the investigative focus of the sentinel 17 P study was to prevent recurrent PTB, vaginal progesterone studies mostly have been directed toward patients with short cervices, regardless of history.

Thus far, the efficacy of the two methods has been based on comparisons with placebos. However, Saccone et al recently conducted a meta-analysis in which the two methods of progesterone delivery were pitted against each other in patients with a history of PTB. The authors found three studies from hospitals in the United States, Saudi Arabia, and Iran that met their inclusion criteria. The meta-analysis contained a total of 680 patients who had prior PTB. These patients were given weekly injections of 250 mg of 17 P or either 90 mg, 100 mg, or 200 mg delivered vaginally daily. The medications were started between 14 and 20 weeks and continued until 36 weeks. The authors were interested in the incidence of spontaneous delivery at varying gestational ages, newborn special care unit (NBSCU) admissions, and side effects.

The vaginal progesterone had significantly lower rates of PTB < 34 weeks (17.5% vs. 25%; relative risk [RR], 0.71; 95% confidence interval [CI], 0.53-0.95) and < 32 weeks (8.9% vs. 14.5%; RR, 0.62; 95% CI, 0.40-0.94). Interestingly, if using < 37 weeks and < 28 weeks, there were no significant differences. Vaginal progesterone had lower rates of NBSCU admissions (18.7% vs. 23.5%; RR, 0.63; 95% CI, 0.47-0.83). Also, vaginal progesterone had a lower rate of “adverse drug reactions” compared with 17 P (7.1% vs. 13.2%; RR, 0.53; 95% CI, 0.31-0.91).


I have chosen this study for two reasons: 1) to use it as a springboard for discussion of the benefits and shortcomings of meta-analyses, and 2) to discuss an option to the standard, but expensive and sometimes uncomfortable, use of 17 P in patients with a history of previous PTB.

The advantage of a meta-analysis is that by increasing the number of collected patients, one can enhance the ability to attain statistical significance when individual studies are underpowered to do so. The shortcoming is that incorporating the warts of one or more of the weaker studies taints the whole study. The way to assure the quality of any meta-analysis is to carefully and objectively select the studies at the front end and to subject the findings to a rigorous critique once the analysis is complete.

The authors found only five studies in the literature that even dealt with the comparison between 17P and vaginal progesterone and immediately excluded two for obvious reasons. Then, at the completion of the study, the authors applied a Cochrane-fashioned tool to assess the quality of the evidence based on the number of limitations and biases incorporated into the studies (the GRADE approach).

In this study, the findings that attained statistical significance were a lower rate of PTB at < 34 weeks and at < 32 weeks, as well as a lower risk of NBSCU admissions and adverse reactions with vaginal progesterone. On the surface this may seem compelling, but the above findings were GRADE-rated as being either “low quality of evidence” or “very low quality of evidence” — meaning that further investigation would be needed before solid conclusions could be made regarding the superiority of vaginal progesterone in this setting. In fact, “it is possible, or even likely,” that the conclusions could be reversed. This is not surprising because of the heterogeneity of the study populations, the three different vaginal progesterone doses used, the low number of patients enrolled in one study, and only one of the three studies attained statistical significance for any of the PTB variables.

So, what can we take away from this imperfect study? I think we can say that vaginal progesterone could be better, but certainly seems no worse, than 17 P in preventing recurrent PTB while causing fewer side effects. Therefore, while waiting for more evidence to surface, vaginal progesterone may represent a viable alternative to 17 P — a sometimes uncomfortable and expensive method that attained favor simply by being the first one on the block to show benefit.


  1. Meis PJ, Klebanoff M, Thom E, et al. Prevention of recurrent preterm delivery by 17 alpha-hydroxy progesterone caproate. N Engl J Med 2003;348:2379-2385.
  2. Hassan SS, Romero R, Vidyadhari, et al. Vaginal progesterone reduces the rate of preterm birth in women with a sonographic short cervix: A multicenter, randomized, double-blind, placebo-controlled trial. Ultrasound Obstet Gynecol 2011;38:18-31.