Clinical Inquiries

What’s the best way to predict the success of a trial of labor after a previous C-section?

Author and Disclosure Information

 

References

EVIDENCE-BASED ANSWER:

While 8 scoring tools predict success rates for a trial of labor after previous cesarean section (TOLAC), it’s unclear which is the best because no trials have compared prediction tools against each other, and each tool has a unique set of variables.

A “close-to-delivery” scoring nomogram predicting the success rate of TOLAC correlates well (90% accuracy) with actual outcomes (strength of recommendation [SOR]: B, prospective and retrospective cohort studies) and has been externally validated with multiple additional cohorts.

All other point-prediction scoring tools are accurate within 10% when predicting the success rate of TOLAC (SOR: B, prospective and retrospective cohort studies).

EVIDENCE SUMMARY

Seven validated prospective scoring systems, and one unvalidated system, predict a successful TOLAC based on a variety of clinical factors (TABLE1-11). The systems use different outcome statistics, so their predictive accuracy can’t be directly compared.12

Grobman: Entry-to-care and close-to-delivery nomograms

Grobman et al created 2 prediction models, an “entry-to-care” model (used at the first prenatal visit), and a “close-to-delivery” model (used on admission to the labor ward).1,2 Both models display a graphic nomogram forecasting the probability of TOLAC success (with 95% confidence intervals [CIs]). The authors compared predicted TOLAC outcomes with actual TOLAC outcomes and found that the model predictions most successfully correlated with high-likelihood outcomes (70% to 90% chance of successful TOLAC, plus or minus approximately 5%). Both models were less accurate with low-likelihood outcomes (40% chance of successful TOLAC, plus or minus approximately 10%).

Many independent authors have validated the close-to-delivery model, comparing predicted with actual TOLAC success rates. In a retrospective cohort study of 490 women, Constantine et al found the correlation between the observed and predicted TOLAC rates to have an r of 0.90, P=.002, with an area under the curve (AUC) of 0.70.3 Yoki et al validated the model in a Japanese cohort of 729 women with an AUC of 0.81, consistent with the AUC of 0.75 reported in the development of the original model.4

Evidence-based answers from the Family Physicians Inquiries Network

Next Article: