Birth after caesarean section.

Abstract

Birth after caesarean section (CS) is an issue of growing importance. As a consequence of increasing CS rates, more women having experienced a prior CS will need counselling about preferred second delivery mode. There are two choices: elective repeat caesarean section (ERCS) or trial of labour (TOL). It is well known, that a TOL ending in an emergency CS carries the largest risks for mother and child. We wanted to investigate for which women and infants it would be safest to recommend an ERCS or a TOL. We assumed, that underlying conditions/indications for the first CS performed would often recur in the second pregnancy and be important for the second delivery outcome. Therefore, a hierarchical system was developed, in which efforts were made to classify according to underlying conditions instead of focusing on conditions appearing during labour, in an attempt to diminish the subjective impact of diagnoses recorded after delivery. The hierarchical system was used through the four papers. We investigated women in the Swedish medical birth register with their first two deliveries 1987-2007 (Paper I-III), or giving birth at least twice, including one CS and at least one delivery after the CS 1992-2011 (Paper IV). In Papers I-III, we have shown that all first CS indications had a statistically significant risk to recur in the second pregnancy/ delivery. Women with a first CS were older, shorter, and had a higher body mass index than women with a first vaginal delivery. The risk for unplanned CS in TOL increased by the women’s age, body mass index, and smoking, while increasing height lowered the risk. Women with a prior CS had an all-over increased risk for unplanned CS in TOL, compared with primiparous women. Infants born to mothers with one prior CS had an almost doubled risk for low Apgar score and perinatal death compared with infants of women with one prior vaginal birth. The risk was lower but still statistically significant after adjustment for possible maternal and fetal/infant confounders. For infants of women with one previous CS, the risk for low Apgar score was higher after a TOL than after an ERCS. In all studies, the risk for adverse outcomes differed substantially between hierarchical indications for the first CS performed. When the first CS was performed without medical indication, no inreased risk for low Apgar score or perinatal death could be detected. The results suggest that underlying conditions, not the previous CS per se, contributed to the risk increase. In Paper IV, we validated a widely used prediction model for chance of successful TOL after CS, developed by Grobman et al. (2007) for US conditions. As the original model was not directly applicable for Swedish settings, we modified it stepwise. The final, new model included maternal age, body mass index, prior vaginal birth, prior vaginal birth after CS, maternal height, first CS hierarchical indication, and the rates of ERCS and unplanned CS in the respective delivery wards. We reached an excellent predictability for vaginal birth in TOL after CS. Counselling about the safest delivery mode after one CS is a challenge. Our study results, combined with previous findings, add important scientific knowledge. However, non-medical factors are vital in the decision-making after one CS, and a trust between the woman, her partner, the obstetrician and the midwife is fundamental. Considering the new information would possibly make counselling easier and, hopefully, lower the rate of unplanned CS in TOL after CS and decrease the rate of low Apgar score and perinatal death in the birth after a caesarean delivery. (Less)