The startle test revisited.

OBJECTIVE

Our purpose was to the compare fetal heart reaction to external physical stimulation with the nonstress test (NST).

STUDY DESIGN

This prospective study evaluates documentation of fetal heart rate accelerations by two methods. The standard NST was performed prior to the ultrasound evaluation. The NST results were unavailable to the ultrasonographer. M-mode ultrasound was used initially to detect a stable heart rate. Then the ultrasound transducer was used to stimulate fetal movement by indenting the uterus over the fetal small parts. A second fetal heart rate was determined within 15 seconds after stimulation.

RESULTS

A total of 194 patients underwent 235 studies. The patient population included four sets of twins. When the NST was reactive, 151/216 had an ultrasound startle response of > or =15 beats per minute; however, all nonreactive NSTs were associated with an ultrasound response of 14 beats per minute or less (p<0.001). A receiver-operating characteristic curve comparing the ultrasound fetal response to the startle with the NST identified the area under the curve to be 0.948, consistent with high specificity and sensitivity.

CONCLUSION

The fetal heart rate response to external stimulation identifies 67% of patients with a reactive NST.