Fetal scalp sampling in labour.

In the linked randomised controlled trial, Wiberg-Itzel and colleagues compare the effectiveness of two methods of diagnosing hypoxia in the fetus during labour— scalp pH analysis and scalp lactate analysis.1 Fetal wellbeing during labour depends on several factors—an adequate supply of oxygenated maternal blood reaching the placental intervillous space, gas exchange across an undamaged placenta, supply of oxygenated blood to the fetus through an open umbilical vein, and sufficient metabolic reserve in the fetus to withstand the hypoxic effects of uterine contractions. The fetus may be compromised by maternal hypotension, prolonged uterine contractions, placental abruption, umbilical cord occlusion, and fetal growth restriction. Distress may also occur for no obvious reason in apparently normal labours. The traditional mainstay of fetal assessment during labour is monitoring of the fetal heart rate—either by intermittent auscultation (using a Pinard stethoscope or hand held Doppler device) or by continuous electronic monitoring. Compared with intermittent auscultation, continuous electronic monitoring has the disadvantage that it restricts the woman’s movement. It also increases the need for caesarean section (relative risk 1.66, 95% confidence interval 1.30 to 2.13) and instrumental vaginal delivery (1.16, 1.01 to 1.32), but is less likely to result in the neonate having seizures because of hypoxic brain insults (0.50, 0.31 to 0.80).2 A trade-off therefore exists between the benefits and harms of the two methods. If 628 women had continuous electronic monitoring there would be one less neonatal seizure and 11 more caesarean sections compared with intermittent auscultation.2 Both approaches to monitoring fetal heart rate lack specificity. Intermittent slowing of the heart rate does not necessarily indicate serious compromise. In particular, “variable decelerations”—caused by cord compression— may be associated with marked and audible drops in the fetal heart rate but are often benign. Adjunctive methods of fetal assessment are therefore needed in the presence of worrying features that are not severe enough to warrant immediate delivery. Fetal scalp sampling to measure capillary blood pH has provided that function for decades.3 The baby’s scalp is visualised through a tapered cylindrical amnioscope that is inserted into the vagina, and capillary blood is obtained via a stab wound. However, the procedure is awkward, intrusive, and uncomfortable. A recent study found that it takes longer to perform than most obstetricians realise. Of 100 consecutive attempts at scalp sampling, 11 failed completely. The mean time between making the decision to sample and obtaining a pH result was 18 minutes. The median time to take a sample after preparations were made and the woman positioned was 14 minutes. The median time to abandoning the procedure in cases of failure was 26 minutes.4 The value of estimating capillary blood pH is also uncertain. A recent review on fetal heart monitoring found no evidence that access to scalp sampling reduced the risk of neonatal seizures or of caesarean section.2 The intrapartum care guideline issued by the National Institute for Health and Clinical Excellence in 2007 acknowledged the lack of a satisfactory evidence base, but it concluded that clinical experience and indirect research comparisons justified a recommendation that fetal scalp sampling should be used when the fetal heart trace was pathological.5 Consequently, and because this is a procedure that features prominently in an expensive medicolegal arena, anything that improves the ease of scalp sampling or provides a less invasive alternative is welcome. Wiberg-Itzel and colleagues report a multicentre Swedish randomised controlled trial of almost 3000 women, which compared scalp pH analysis with scalp lactate analysis.1 Lactate concentrations reflect tissue hypoxia and consequent anaerobic metabolism. Measurement of fetal capillary lactate concentrations during labour was described in the early 1980s,6 but the technique failed to catch on in clinical practice. With technological advances, lactate can be reliably measured using small amounts of fetal blood (5 μl).7 In contrast, pH analysis requires 30-50 μl of blood. Wiberg-Itzel and colleagues found no significant difference between rates of metabolic acidaemia at birth after use of lactate analysis or pH analysis of fetal scalp blood samples to determine hypoxia during labour (0.91, 0.61 to 1.36 v 0.84, 0.47 to 1.50). However, significantly more protocol violations occurred for pH estimation (10.4%) than for lactate estimation (1.2%), mainly because of failed sampling. The authors did not record the time taken to perform sampling, but it was thought to be shorter for lactate estimation because of the smaller samples needed. Economic data were not reported. This study, together with a smaller earlier trial,8 supports the use of lactate measurement rather than pH estimation at fetal scalp sampling. Do less invasive alternatives exist? The most promising is analysis of the ST segment of the electrocardiographic signal obtained from an electrode attached to the fetal scalp. The adjunctive use of ST segment analysis along with cardiotocography decreases the need for dk im ag es