Examining the Evidence
Does US measurement of cervical length to determine the need for cerclage reduce preterm delivery?
This methodologically rigorous trial explored the utility of US imaging versus history in establishing the “need” for cervical cerclage. I have deliberately placed the word “need” in quotations because, as yet, the efficacy of cerclage under any circumstances has yet to be definitively established.
In their discussion of the findings, Simcox and colleagues assert that a trial to settle the questions frequently raised in the cerclage debate would require thousands of patients and may not be feasible.
Rigorous trial is impressive,
but doesn’t settle key questions
The authors performed an excellent randomized, controlled trial, and their intention-to-treat analysis is laudable. They are also to be congratulated for remaining focused on the primary outcome of delivery before 34 weeks’ gestation. It is notable that the primary outcome was essentially the same in each group, regardless of the treatment, be it 1) US screening and cerclage for cervical length <20 mm or 2) no screening and cerclage for historical indications. I recall a conference on prematurity from the mid-1980s that included, as one of its conclusions, the observation that as many as 70% of patients who have historical “indications” for cerclage will deliver at term in their next pregnancy if left untreated.
Unresolved questions in regard to cervical cerclage include:
- What is the best way to determine who is a candidate?
- What is the best type of cerclage?
- What is the most appropriate outcome to be measured?
- Is there a place in practice for “universal” screening of cervical length?
- What is the true cost (in terms of both dollars and morbidity) of intervention versus no intervention?
- What are the medicolegal implications of each approach?
This study by Simcox and colleagues settles none of these questions.
High-risk women may benefit from US imaging, but the data from this study do not support that conclusion. Nor is the best type of cerclage defined, though there is ample opinion on this topic.
Is 34 weeks’ gestation the appropriate primary outcome? More and more, we read about late preterm or so-called near-term outcomes being less optimal than they once were thought to be—though delivery at 34 to 37 weeks would seem to be preferable to delivery at less than 34 weeks.
The cost of each approach is unclear. How many “unnecessary” cerclages would be needed to prevent one very-low-birth-weight delivery? And how “risky” is elective cerclage placement in skilled hands?
Finally, not many patients or physicians are likely to want to embrace a wait-and-see approach if they have already had one or more adverse outcomes, and the risk of doing nothing may be considerably greater in medicolegal terms than the risk of proceeding with what may be an unnecessary intervention that ends in a term or near-term delivery.
WHAT THIS EVIDENCE MEANS FOR PRACTICE
On the basis of these results, I think the practitioner should rely on history to make a clinical judgment about the need for cerclage. Ultrasonographic imaging may not only be of little help, but it may lead to greater intervention than would otherwise be needed. Perhaps a return to clinical basics, such as detailed history taking and physical examination, is a good message for these economic times.—JOHN T. REPKE, MD