Over the past decade, there has been an evolution in the screening and management of cervical dysplasia toward a more conservative approach. Accumulating data from large studies demonstrating a low risk of cervical cancer among adolescents and women 21-24 years of age supported these changes. However, a major stimulus for the new guidelines was concern regarding the potential adverse effects of excisional procedures on subsequent pregnancy outcomes, namely, preterm birth. The published evidence overall reflects an increased risk of preterm birth in women with a history of a cervical excisional procedure. However, recent studies have found a similar increased risk for preterm birth in women with cervical dysplasia without a history of cervical excision. This suggests that perhaps cervical excisional procedures do not cause subsequent preterm birth, but rather the risk factors for cervical dysplasia are similar to those for preterm birth. As a result of these recent findings, there has been a renewed interest in clarifying the relationship among cervical dysplasia, excisional procedures, and preterm delivery.

Published in this issue of the journal, the study by Miller et al aims to further investigate whether cervical dysplasia without a cervical excisional procedure is associated with an increased risk of preterm birth, and to determine if the possible increased risk can be attributed to the presence of a short cervix. Miller et al conducted a retrospective cohort study including 18,528 pregnant women who underwent routine cervical length screening over the 4-year study period. These women were stratified into 3 groups based on review of medical records: no prior dysplasia, history of dysplasia without cervical excisional procedure, and prior excisional procedure. Using multivariable logistic regression equations, they found that after adjusting for the presence of a short cervix and other potential confounders, a prior excisional procedure (adjusted odds ratio [aOR], 1.31; 95% confidence interval [CI], 1.04–1.64), but not prior dysplasia alone (aOR, 1.02; 95% CI, 0.85–1.21) was associated with preterm birth.

One of the key strengths of this study was the incorporation of cervical length data in the analysis. This is fundamental to understanding possible mechanisms by which cervical excisional procedures increase the risk of preterm delivery. The authors report that women with a prior cervical excisional procedure had the highest incidence of short cervix (2.2%) within the cohort. Despite this, prior excisional procedure was independently associated with an increased risk of preterm birth regardless of cervical length. The implication is that the increased risk of preterm birth following a cervical excisional procedure is not solely the result of the loss of cervical tissue leading to a short cervix. This is further supported by the finding that an interaction term for short cervical length and prior excisional procedure in relation to preterm birth was not significant.

Ideally, the question of whether or not a cervical excisional procedure actually causes subsequent preterm birth would be best answered in a randomized controlled trial. However, this is not possible for obvious ethical reasons, thus we rely on large, well-conducted observational studies like this study performed by Miller et al to guide clinical care. However, there are some limitations to this study that should be considered. First, dysplasia history and history of an excisional procedure were determined by patient report on the prenatal record, thereby increasing the likelihood of recall bias and misclassification bias. Women are more likely to recall and report cervical excisional procedures than untreated dysplasia, which could have biased the results away from the null. In addition, the authors do not report the amount of missing data with regards to dysplasia history. It is important to note that the primary outcome in this study was preterm birth <37 weeks, which includes both spontaneous and indicated preterm births, and may not accurately reflect the mechanism of preterm birth as a result of cervical excision. Despite including all preterm births, the overall preterm birth rate in this study (6.6%) is low, making the results less generalizable to high-risk populations. Finally, as the authors discuss, there is the potential for residual confounding, which may account for their results. In this study, they did not control for severity of dysplasia, depth of cervical excision, or socioeconomic factors. Due to these limitations, especially in a study where the CI of the positive finding so closely approaches 1, the results should be interpreted with caution.

When a new study is published, it is wrong to immediately incorporate the results into clinical care, and disregard prior studies on the subject. We must instead integrate these new findings with previous literature. Ideally, this assessment can be aided by patient level–based quality metaanalyses.

To properly assess the effect of cervical dysplasia on the incidence of preterm birth, women with cervical dysplasia but no excisional procedures should be compared to similar women with neither cervical dysplasia nor prior excisional procedures. No metaanalysis of such studies exists, and data are very limited. One study including 75 women with precancerous lesions and no treatment found no increased risk of preterm birth. Another study reported instead increased risk of premature delivery in women with untreated cervical dysplasia.

To better delineate the effect of excisional procedures in women with precancerous changes of the cervix, the most appropriate control group is women with similar precancerous changes of the cervix, but no excisional procedure. One metaanalysis of such studies shows a significant 27% increased incidence of preterm birth associated with excisional treatment, after adjusting for confounders. In contrast, a metaanalysis limited to only loop electrosurgical excision procedures showed no increased risk of preterm birth when women with prior loop electrosurgical excision procedure were compared to women with history of cervical dysplasia but no cervical excision. Despite the major contribution to assessing the association between precancerous cervical lesions (dysplasia alone) and preterm birth made by Miller et al in this edition of our journal, clearly more data are needed to make a final conclusion.

Although the study from Miller et al shows an independent association between history of cervical excisional procedure and the risk of preterm delivery, the magnitude of the risk should be acknowledged when considering the clinical implications of this finding. In this study, a history of cervical excision was associated with only a 30% increased risk (aOR, 1.3) of preterm delivery, consistent with the metaanalysis data. In contrast, use of in vitro fertilization to conceive was associated with a 130% increased risk (aOR, 2.3) of prematurity in the study. Furthermore, the risk of delivering prematurely was >5 times higher if a woman had a cervical length of ≤2.5 cm, regardless of her history of cervical excisional procedure. In light of this, it seems reasonable to counsel women about a potential increased risk of preterm delivery if they have had a prior cervical excisional procedure. However, without data from studies that specifically evaluate potential interventions to prevent prematurity in women with a prior excisional procedure, these women should receive routine prenatal care, which may include cervical length measurement as part of universal cervical length screening. Cerclage placement in the absence of other indications is not currently supported by the literature.

Clearly there is a relationship between prior cervical excisional procedure and preterm birth, however the nature and etiology of this association remains undefined. The study by Miller et al shows that cervical shortening may not be the only specific process that leads to an increased risk for preterm birth in these women. Further research is needed focusing on clarifying the biological mechanism by which prior cervical excisional procedure increases the risk of preterm delivery. Until this relationship is further delineated, given the conflicting data, the current less interventional approach to the management of cervical dysplasia among young women seems prudent.