We thank Dr Storm for the comments regarding our article.1We agree with Dr Storm that in awake patients, stressors such as nausea, vomiting, and anxiety influence the number of fluctuations in skin conductance (NFSC) and, therefore, limit the specificity of the Medstorm device as a measure of postoperative pain. These variables would inevitability be present in the clinical context of postoperative pain in children.
This is not the first study to demonstrate that the Medstorm device has poor sensitivity and specificity for pain in the postoperative period. A study of 100 adults by Ledowski et al .2indicated an optimized (by receiver operating characteristic curve analysis) NFSC cutoff of 0.1, which resulted in a sensitivity of 58% and a specificity of 61%, for a numeric pain rating score of more than 5.
We suggest that a test that is sensitive, but not specific, is not clinically useful. Therefore, a cutoff of 0.0 NFSCs would yield a sensitivity of 100% but a specificity of 0% and would clearly not be useful. We suspect that few clinicians would benefit from a device that “gives an indication on when to ask a patient about their pain” when it is relatively simple to routinely ask all patients about their pain level. We believe that the averaging interval should be a magnitude greater than the NFSC. In our clinical experience, postoperative pain does not last for only 15 s nor would it require a pharmaceutical intervention if it did occur for this short period.
The Medstorm device may have utility for detecting intraoperative pain; the variables of movement and anxiety can be appropriately controlled. However, in the complex setting of postoperative pain, the accuracy of NFSC measurements is severely compromised by numerous nonnoxious confounders of sympathetic activity.
*BC Children's Hospital, Vancouver, British Columbia, Canada. email@example.com.