Vaginal Birth After Cesarean 

Abstract & Commentary

In the July 5, 2001, issue of the New England Journal of Medicine, the lead article was entitled, "The Risk of Uterine Rupture during Labor among women with a Prior Cesarean Delivery." Lydon-Rochelle and colleagues retrospectively analyzed data from the Washington state birth-events database from 1987 to 1996. Information was available on 20,095 patients who had had cesarean delivery and then delivered another child vaginally during this study period.

Uterine rupture occurred in 1.6 per 1000 patients who had repeat cesarean delivery without labor and in 5.2 per 1000 of those who had spontaneous onset of labor. If labor was induced with oxytocin, the rupture rate was 7.7 per 1000 and 24.5 per 1000 when a prostaglandin (PG) preparation was used. Since misoprostol (PGE1) was introduced to Washington state the year the study was completed, it is likely that few of the PG-induced patients actually got this agent.

The relative risk (RR) of uterine rupture for repeat cesarean section without labor was 3.3 (confidence interval [CI] 1.8-6.0), induction of labor without PGE was 4.9 (CI, 2.4-9.7), and labor with prostaglandins was 15.6 (CI, 8.1-30.0). (Lydon-Rochelle M, et al. N Engl J Med. 2001;345:3-8).

Comment by John C. Hobbins, MD

The national cesarean delivery rate peaked in 1989 to 25%, and fell to 21% in 1996 and has since remained the same. Although there were many contributors to this downward trend, such as monitoring through peer review, liberalization of second-stage progress requirements, and perhaps, a more active management of labor, the emphasis on the option of vaginal birth after cesarean delivery (VBAC) had to have played a major role.

The popularity of VBAC was predicated on the excellent safety record generated during many early trials. Now that there is evidence emerging that VBAC may not be a good idea for everyone. Studies have surfaced, like the New England Journal of Medicine report above, that have shown a greater risk of uterine rupture when PGs are used for induction. In the New England Journal of Medicine article, it was unclear which type of PG was used (intravaginal PGE2 suppositories or intracervical PGE2 gel), but the timing of the study precludes PGE1 as being a contributor to uterine rupture in this study. However, there are a few recent reports that suggest that misoprostol has an even higher rate of uterine rupture in VBACs, and this has caused the American College of Obstetrics and Gynecology to advise against its use in VBACs.

The original rationale for the use of various PG derivatives was not necessarily to induce labor, but to ripen the cervix either through generating mini-contractions or to "soften" the cervix through an effect on collagen and elastin. What seems to have been overlooked was the possibility that while softening the cervix, one might also be softening the uterine scar, and PGE1 might be the best scar softener around.

On the other hand, PG may not be the culprit here. By implication, PG derivatives are used predominantly when the cervix is unripe, and the increased risk of rupture may well be more related to the process of trying to initiate an induction in a woman who has a scar in her uterus and a cervix that is unready for labor. This could also explain the difference in uterine rupture in the above study between patients in spontaneous labor and those induced with oxytocin.

It is important to make a clear distinction between induction and augmentation. By definition, the latter is applied to those already in spontaneous labor—those who have passed through some natural early physiologic processes but who need a little help with contractions. Naiden and Deshpande recently published 10 years of experience in which an active management of labor protocol (with liberal use of oxytocin augmentation) was used successfully to diminish the overall cesarean section rate from 16.6% to 10.9%, and the repeat cesarean section rate from 7.3% to 3.8%. During this time, there were 2 uterine ruptures in the 1200 successful VBACs. Neonatal mortality did not change during the study period. So, undoubtedly, judicious use of oxytocin, especially for augmentation, should not be interdicted in VBACs.

On a pragmatic note, whether we like it or not, almost everything we do today is scrutinized according to its cost-effectiveness. In an ambitious study, Chung and colleagues subjected data from 22 years worth of VBAC studies in the literature to mathematical modeling in order to determine its cost-effectiveness compared with outright repeat cesarean section. When taking into account the cost of the cesarean delivery vs. vaginal delivery, and folding in the expense of the maternal and neonatal complications, Chung et al found that the bottom line was very dependent upon the success of the endeavor. VBAC was clearly cost-effective only when the success rate exceeded 75%.

Is there a way to identify patients with uterine scars who are at the greatest risk of rupture? In a study from Japan, Gotoh and associates attempted to predict uterine rupture by evaluating the thickness of the scar with transabdominal and transvaginal ultrasound. They found that the magic number below which there was a substantial risk of rupture was 2 mm. If the thinnest portion of the uterus in the neighborhood of the scar exceeded 2.9 mm, the risk of rupture was negligible.

Our initial excitement about the study has been dampened somewhat by our difficulty in obtaining adequate transvaginal images of the uterine scar when enough urine is in the patient’s bladder to precisely measure wall thickness. Also, in 1 case of a uterine "window" and another case of frank rupture, the wall thickness exceeded 2.9 mm just prior to delivery.

There are many disincentives for the obstetrician to embark upon VBAC, such as risk of uterine rupture, the necessity of onsite supervision, and the questionable cost-effectiveness of the venture. However, perhaps the greatest deterrent is the potential for a bad outcome to wind up in the courts. So, should VBAC be abandoned? No! There is solid evidence that while VBAC may not be for everyone, it is safe and effective for some, especially since many patients are highly motivated to choose this option. The ideal patient would be one who has an excellent chance of succeeding and in whom a minimum of medication and intervention would be needed. For example, this would be a patient who has entered into spontaneous labor or whose cervix is easily inducible with oxytocin (Bishop score greater than 4). The estimated fetal weight should be reasonable and the patient should make excellent progress through labor.

Most importantly, before embarking on VBAC, the patient must be carefully counseled about its risks, benefits, and expectations of success. The stark wording of some consent forms, such as those drafted by malpractice insurers, can make some patients wonder if pregnancy is a good idea. Therefore, verbal embellishment is essential in the consent process.

Suggested Reading

1. Wing DA, et al. Obstet Gynecol. 1998;91:828-830.

2. Induction of labor with misoprostol. ACOG Committee Opinion Number 228. Presented at meeting of the American College of Obstetrics and Gynecology; November 1999; Washington, DC.

3. Naiden J, Deshpande P. Am J Obstet Gynecol. 2001; 184:1535-1543.

4. Chung A, et al. Obstet Gynecol. 2001;97:932-941.

5. Gotoh H, et al. Obstet Gynecol. 2000;95:596-600.