We thank Campion et al. for their interest in our study examining the accuracy of malignant hyperthermia (MH) diagnoses in hospital discharge records1 and appreciate the opportunity to answer their questions. First, they ask why we chose the expert panel approach over the Delphi method for the medical record review. As described in our article, we used two expert panelists to independently review each medical record with a standard data abstraction form and the MH Clinical Grading Scale (CGS). When there was discrepancy between the two panelists, a third expert on site was consulted to reconcile the difference. Consensus on final CGS scores was reached in all cases without difficulty. The Delphi method is a consensus-building technique involving multiple rounds of polling a panel of experts, who usually remain anonymous to each other. The Delphi method was neither appropriate nor practical for use in our study, which was performed in six medical centers in compliance with privacy rules and the requirements of institutional review boards.
Second, Campion et al. wondered why we did not provide CGS scores. As a qualitative measure, the CGS is designed for ranking the likelihood that an adverse anesthetic event represents an incident of MH.2 The expert panelists’ assessment of the MH diagnoses (shown in table 1 of our article1 ) was based on the individual CGS scores along with other clinical characteristics. We did not present CGS scores because the number of MH incident cases (n = 11) included in our study was too small; providing individual characteristics for such a small sample would be problematic according to the conventional statistical reporting standards and personal privacy protection laws.
Third, Campion et al. question our decision to include MH susceptibility based on positive medical history in assessing the accuracy of MH diagnoses. Although we agree that the documentation of MH susceptibility in the medical record may not be entirely accurate, we do not agree that “the data of the study itself suggest that this is not valid.” Our study provides valuable data for assessing and interpreting MH diagnoses ascertained from hospital discharge records. It is noteworthy that diagnoses in hospital discharge records include both incident and prevalent cases. Therefore, it is necessary for us to consider both MH episodes occurring during the hospitalization and preexisting MH susceptibility in assessing the accuracy of MH diagnoses in hospital discharge records. The results of our study indicate that MH diagnosis listed in hospital discharge records has a positive predictive value of approximately 70%, which is similar to the findings regarding diagnoses of cerebrovascular disease and congestive heart failure.3
Finally, Campion et al. suggest that we should have used the caffeine–halothane contracture test, instead of medical record review, as the reference standard. Although we agree that the caffeine–halothane contracture test is widely regarded as the “accepted standard” for determining MH susceptibility, its specificity is less than 80%,4 and performing this test to confirm the diagnosis of MH susceptibility would not have been feasible for our study. We reviewed medical records for six institutions to retrospectively estimate the accuracy of coding for MH in hospital discharge records. The study proposed by Campion et al. could only be performed prospectively. Given the rarity of MH cases, such a study is unlikely to ever be performed.
This study was supported in part by the Malignant Hyperthermia Association of the United States (Sherburne, New York).
Dr. Rosenberg received a one-time speaking fee from Eagle Pharmaceuticals (Woodcliff Lake, New Jersey), a company that manufactures Ryanodex, a concentrated formulation of dantrolene approved for the treatment of malignant hyperthermia. The other authors declare no competing interests.