You are here
Editor’s comment: Prof. Yves Ville: An unsuspected enemy is doubly dangerous (L. Frank Baum)
(Editor’s comment relates to the 3 AJOG articles below: Aubry et al, Khalifeh et al & Saade et al)
Over a third of infant deaths arise from complications related to preterm births, making prematurity the most frequent cause of infant mortality. Health complications are also a lifelong burden of survivors including mental retardation, cerebral palsy, learning and behavioral problems, respiratory problems, vision and hearing loss, but also diabetes, high blood pressure, and heart disease. This inventory has become a classic lament of perinatologists facing the implacable consequences of prematurity.
Trends in prematurity have been either on the rise (1990-2006) or flattening (2006-2013) with medically indicated « late » prematurity as the main adjustment variable. However around 1/25 pregnancies still face unexpected delivery before 33 weeks and roughly half of them are primigravidas without anticipated risk factors.
Prevention measures are mainly directed at strengthening cervical tolerance and uterine quiescence in pregnancy, including vaginal progesterone administration, cerclage and pessary placement that have shown definite but partial positive effect at delaying delivery. This is likely to reflect the incomplete adequacy between the causative factor and the limited prevention measures at hand.
Although the underlying causes of preterm birth are incompletely understood, a previous premature birth, a multiple pregnancy, uterine abnormalities, and a number of medical conditions involving inflammation or impacting placental implantation are at increased risk of preterm birth. Genetic profiling is till a long shot, and social, lifestyle and environmental factors also play a role. Overall about half of the cases of severe prematurity involve primigravidas with no (as yet) identified risk factor.
There is therefore a need for both early identification and wide dissemination of potent screening tools. Outside previous history of PTB, short cervical length (CL) detected by transvaginal ultrasound (TVU) in the second trimester is the best current predictor of subsequent PTB. However 20 years down the line, following on J. Iams’ seminal paper in the New England Journal of Medicine, there have been only few calls for universal TVU CL screening, and mainly limited to singleton gestations.
In the May issue of AJOG, Berghella et al make a strong case for universal TVU CL screening. The natural history of CL shortening leading to PTB is well understood. The technique is standardized and reproducible and it is an acceptable test for pregnant women. Resources may well be adequate in industrialized countries where second trimester ultrasound screening is well established. Identification of a short cervix through universal CL screening is a continuous process that prevents a prevalent condition. Facilities for diagnosis and treatment are available with consensual indications for vaginal progesterone, cerclage and pessary in the short-term.
The number needed to screen to prevent one PTB before 33 weeks of gestation is approximately 357 (compared with approximately 1140 to screen to prevent one death from cervical cancer). Prevention of PTB with progesterone would have been expected to have shown a greater effect in the years after its introduction, yet the decline in PTB has remained fairly steady in the USA since 2006. It may be due to a gradual uptake in this practice or a combination of both demographic and interventional changes. A strong call for universal TVU CL screening could only help prevention to show its effect however limited that is.
However, algorithms based solely on clinical, demographic and biophysical data will remain limited. Mass screening would be facilitated by the existence of biomarkers for preterm birth, that are still lacking although multiple molecular pathways are known to be aberrant in women who ultimately deliver preterm. In the May issue of AJOG, Saade et al have demonstrated the value of serum screening using a combination of proteins whose expression was shown to be disregulated in early second trimester in women destined to deliver prematurely. The elegance of study is to be recommended. Indeed the authors have prospectively applied proteomics in maternal blood in a cohort of pregnant women and identified a pannel of proteins that appeared to be significantly either upregulated (Insulin-like growth factor binding protein 4 (IBP4)) or down-regulated (sex hormone binding globulin (SHBG). They then built-up a serum screening test using the log ratio of the measures of IBP4 and SHBG (IBP4/SHBG) in a clinical validation study to classify spontaneous preterm delivery cases (<370/7 weeks gestational age) in a nested case-control cohort different from the discovery study. The predictor had an area under the receiver operating characteristic curve value of 0.75. At a sensitivity and specificity of 0.75 and 0.74, respectively, a positive test showed an odds ratio of 5.04 for spontaneous preterm delivery. The test performed better for earlier gestational ages at delivery.
These two papers may well mark a turning point in offering the possibility of building- up a combined screening test in early second trimester. Using the successful model of combined aneuploidy screening in the first trimester, bringing up two strong predictors in addition to personal history and physical examination in early second trimester, should hopefully allow to apply preventive measures and give more room to personal medicine taking into account medical factors, genetic predisposition as well as social, lifestyle and environmental factors.
This editorial is about bringing attention to stepwise progress and emerging pathways in obstetrics. However I would like to also encourage you to also read what Aubry and Nesbitt wrote nearly 50 years ago in AJOG in the early days of the continuous battle for a better perinatal outcome.
(Aubry RH, Nesbitt RE Jr . High risk obstetrics. I. Perinatal outcome in relation to a broadened approach to obstetric care for patients at special risk. Am J Obstet Gynecol. 1969 Sep 15;105(2):241-7.)