Predictive value of the penilo-cavernosus reflex

To determine the predictive values and specificity of clinical and neurophysiologic testing of the penilo-cavernosus reflex.

A group of 22 control men and a group of 53 men with chronic cauda equina lesions (patients) were studied. The penilo-cavernosus reflex was elicited clinically and measured neurophysiologically using single and double electrical stimulation. The sensory threshold, reflex threshold and stimulation strength in all controls, and latency of the response in all subjects, were compared to reference intervals. Sensitivity, specificity, positive and negative predictive values, and likelihood ratios of the penilo-cavernosus reflex testing were calculated.

Sensitivity of the penilo-cavernosus reflex was 72-83% and 81-83%, specificity 52-57% and 90-91%, positive predictive value 83-97% and 95-96%, negative predictive value 52-57% and 66-68%, and likelihood ratios 1.9-2.8 and 8.3-9.0 for clinical (circumcised subjects: included/excluded; reflex pathologic when diminished/non-elicitable) and neurophysiologic testing (using three stimulation modalities), respectively. Specificities of the sensory threshold, reflex threshold and stimulation strength were approximately 90% or higher.

The present study demonstrated higher specificity, predictive values and likelihood ratios of neurophysiologic testing of the penilo-cavernosus reflex compared to clinical testing. The results support the use of clinical penilo-cavernosus reflex testing as an addition to bladder emptying difficulties and saddle sensory loss as criteria for patient referral for uro-neurophysiologic examination, and measurement of the reflex as a valuable part of such testing. Neurourol. Urodynam. 28:390-394, 2009. (c) 2008 Wiley-Liss, Inc.