Main Findings
In this study, a clinical multivariate risk model of failed mid-cavity VAD was developed and validated. The multivariate analysis identified seven statistically significant and independent risk factors for failed mid-cavity VAD: a head-perineum distance greater than or equal to 51 mm; the presence of a bulky caput succedaneum, a maternal height less than or equal to 1.54 m; a duration of expulsive efforts of less than 21 minutes or greater than 30 minutes before beginning the VAD; a duration of greater than or equal to three hours from a cervical dilation of five centimetres to complete cervical dilation; a term of delivery greater than or equal to 42 weeks of gestation; and an occiput-posterior or occiput-transverse fetal head position. The statistical method used in this study calculated the independent OR associated with each of these variables and allowed a combination of these elements in a quantitative manner to establish a reliable (Hosmer-Lemeshow test = 8.5; p = 0.39) and accurate (concordance index = 0.74) score. With the use of a simple seven-variables scoring system, the risk of failed mid-cavity VAD can be evaluated quickly to guide the obstetrician in choosing the most appropriate room for delivery. The threshold to predict a failed mid-cavity VAD was set at 0.163. Below this value, the VAD attempt could be conducted in a classical delivery room. Conversely, for a score value greater than 0.163, an attempt in an operating theatre with anticipated extension of epidural analgesia and advanced preparation of C-section equipment should be considered.
The most important risk factor for failed VAD was a head-perineum distance greater than or equal to 61 mm, corresponding to high-cavity VAD. High-cavity VADs are difficult and rarely attempted, but their success rate was close to 50 % in our study. The implementation of the abovementioned precautions could make it possible to attempt this type of OVD in the operating theatre with experienced operators and reduce the number of C-sections.
The results concerning the secondary objective confirmed the interest of this score. In the case of failed mid-cavity VAD, advanced preparation of equipment for C-section and anticipated extension of epidural analgesia were significantly associated with a decrease in the time interval between mid-cavity VAD and birth. The transfer to an operating theatre and the anticipated extension of epidural analgesia were associated with a significant decrease in the rate of general anaesthesia for the mother. No improvement of the other maternal or neonatal outcomes appeared with these precautions in our study. The low number of women included for this secondary objective (194) may explain this result. When the women were transferred to an operating theatre for an ultimately successful VAD, the neonatal outcomes were globally poorer than those of successful VADs conducted in a standard delivery room. Two factors could explain these results: the transfer to the operating theatre was probably decided in the most critical cases, and the delay associated with the transfer to the operating theatre may worsen neonatal outcomes. These results confirmed the utility of our predictive score: to better assess the risk of failed mid-cavity VADs to ensure that only women with a major risk of failed VAD are transferred to an operating theatre.