Cesarean Delivery for Fetal Macrosomia
ABSTRACT & COMMENTARY
Synopsis: A policy of elective cesarean delivery for ultrasonographically diagnosed fetal macrosomia is medically and economically unsound in normal pregnant women.
Source: Rouse DJ, et al. JAMA 1996;276:1480-1486.
To determine the potential benefits, risks, and financial impact of a policy of elective cesarean delivery for fetal macrosomia identified by ultrasound, Rouse et al performed a decision analysis comparing three strategies: first, patient care without ultrasound; second, ultrasound followed by elective cesarean delivery if the estimated fetal weight was 4000 g or more (4000 g policy), and, third, ultrasound followed by elective cesarean delivery if the estimated fetal weight was thought to be 4500 g or more (4500 g policy). The analysis was performed for patients with and without diabetes mellitus. This study focused on the rates of shoulder dystocia with resulting permanent brachial plexus injury compared to the number of additional cesarean deliveries and the cost that would result from strategies designed to avoid permanent brachial plexus injury. A birth weight of 4000-4499 g was assumed to occur in 8.2% of births in nondiabetic women and a weight of 4500 g or more in 1.5% of these births. The figures for infants of diabetic mothers were 17.1% and 6.1%, respectively. In the standard management group, cesarean delivery rates of 27% for infants weighing 4000-4499 g and 45% for infants over 4500 g were applied. The likelihood of shoulder dystocia was thought to be 6.7% and 14.5% in the two study groups in nondiabetic women, increasing to 13.9% and 52% when the pregnancy was complicated by diabetes. The likelihood that brachial plexus injury would result from shoulder dystocia was thought to be the same whether or not the pregnancy was complicated by diabetes18% for a birth weight of 4000-4499 g and 26% for those above 4500 g. The authors assumed that 6.7% of brachial plexus injuries would be permanent. Ultrasound was assumed to have a 60% sensitivity for the identification of macrosomia and 90% for its exclusion.
Applying these data to one million pregnancies in non-diabetic women demonstrated that, for each permanent brachial plexus injury prevented by the 4000 g policy, 2345 cesarean deliveries would be performed at a cost of $4.9 million per injury avoided. For the 4500 g policy, 3695 cesarean deliveries would be performed at a cost of $8.7 million per permanent brachial plexus injury prevented. For one million women whose pregnancies were complicated by diabetes, the results were more favorable. For the 4500 g policy, 443 cesarean deliveries at a cost of $930,000 would be required to avoid a permanent brachial plexus injury, while with the 4000 g policy, 489 cesarean deliveries would be performed at a cost of $880,000 to prevent one permanent traumatic injury.
The authors conclude that a policy of elective cesarean delivery for ultrasonographically diagnosed fetal macrosomia is medically and economically unsound in normal pregnant women. For pregnancies complicated by diabetes, this strategy appears to be more reasonable.
COMMENT BY STEVEN G. GABBE, MD
Shoulder dystocia complicates one in 200 vaginal deliveries and occurs most often in fetuses weighing more than 4000 g, particularly in infants of diabetic mothers because excessive fat is deposited around the fetal shoulders. While the rate of permanent brachial plexus injury resulting from shoulder dystocia is low, the long-term consequences for the infant, ranging from impaired to total loss of the use of an arm, are so devastating obstetricians are likely to consider a cesarean delivery when a macrosomic fetus is anticipated. Determining how large the fetus will be is just one part of this challenging clinical problem. For example, in obese women, a group at increased likelihood to deliver a macrosomic fetus, clinical and ultrasound estimates of weight are extremely difficult. Based on the available data, Rouse et al have challenged the use of any weight cutoffs in reaching a decision to perform a cesarean delivery to avoid shoulder dystocia, although they note that pregnancies complicated by diabetes may benefit from this approach.
Like all decision analysis, the conclusions reached are only as good as the data on which the analysis is performed. To account for this, Rouse et al performed sensitivity analyses varying the rates of key parameters, using a rate of permanent brachial plexus injury as high as 32% and increasing the sensitivity and specificity of ultrasound, for example. The cost associated with a policy of performing elective cesarean deliveries for macrosomia in women without diabetes remained excessively high. Unfortunately, this analysis failed to examine fully some of the other costs. For example, what are the costs of medicolegal settlements and litigation of such cases? Nor could the analysis examine important factors in the decision to allow a trial of labor or perform a cesarean delivery, such as the patient’s prior obstetric history, clinical pelvimetry, or progress during labor.
In my own practice, I continue to consider cesarean delivery in pregnancies complicated by diabetes with an estimated fetal weight of 4250 g or greater. In normal pregnancies, while taking my clinical and ultrasonographic estimate of fetal weight into consideration as well as other important clinical parameters, I will allow a carefully monitored trial of labor.