Failure rates for vacuum extraction of between one in 16 and one in 600 have been reported. Most studies report that unexpected failure carries a greater risk to both mother and fetus. The aim of this study was to determine factors that were likely to predict success or failure in trials of vacuum extraction.

At the Port Moresby General Hospital, 59 trials of vacuum extraction were performed between 1 December 1997 and 30 November 1999. These cases were analysed according to whether vacuum extraction was achieved with more than or less than three pulls, or an alternative method of delivery was required to effect delivery.

Factors that were predictive of failure were: (i) Highlands origin of the mother; (ii) longer duration of the second stage of labour; (iii) severe moulding of the fetal head; (iv) cup detachments and deflexing cup applications; and (v) operator persisting with the procedure after three pulls.

The cervix being less than fully dilated when the trial was commenced was not associated with a higher risk of failure, nor was it associated with a significantly higher risk of cervical trauma.

Perinatal death and serious fetal scalp trauma were associated with deflexing cup applications, making more than three pulls and failed vacuum extractions.