Abstract
Objective. Based on a comparison of the clinical indications for cesarean section (CS) in two Danish counties and a review of the literature regarding this issue the aim of this study was to discuss possible explanations for variations in CS rates in twin pregnancies. The comparison of indications for CS in twin pregnancies was made between two Danish counties, one with a high and one with a low overall CS rate in twin deliveries, taking into account the distribution of parity, mother's age, gestational age at birth, and birth weight.
Design. A population based, historic follow-up study based on antecedent data.
Setting. Two Danish counties, with a CS rate in twin pregnancies of 57% and 28%, respectively.
Subjects. All women with twin pregnancies who delivered in 1989 in the two counties.
Main outcome measures. Comparison of the CS rates in the two counties according to indications and fetal presentation.
Secondary measures. Perinatal and maternal outcome.
Results. The difference in CS rates between the two counties could not be explained by different distributions of background characteristics. Different attitudes were found towards CS in cases with previous CS, with twin A in breech presentation and in cases with vertex-breech deliveries. These differences could explain less than two thirds of the overall 29% (CI: 12-46%) difference in risk of CS between the two counties, indicating more subtle reasons for the discrepancy. No difference between the two counties in perinatal morbidity and mortality was seen.
Conclusion. In order to understand and discuss regional variations in the use of CSs in twin deliveries the subjects must be addressed in different ways: the unequivocal indications related to fetal presentations and previous CS can be subjected to randomised controlled trials or large scaled follow-up studies regarding maternal and perinatal morbidity and mortality. Other more subtle determinants of the physicians’ and the pregnant women's attitude towards CS, however, seem quantitatively important, and these can only be evaluated in observational studies and through discussions.