Head-impulse and caloric tests in patients with dizziness.

OBJECTIVE: To test the performance of the head-impulse and caloric tests in terms of sensitivity, specificity, and predictive efficiency.

STUDY DESIGN: This was an open and prospective study conducted at a tertiary care center in which 265 patients were subjected to the head-impulse test and caloric test on the same day. The results of the head-impulse test were considered as normal or pathologic. In a similar way, the caloric test was rated as normal when the difference in canal paresis was less than 22 percent and directional preponderance less than 28 percent, and abnormal if canal paresis was more than 22 percent and/or directional preponderance was more than 28 percent.

MAIN OUTCOME MEASURES: The results of each test were compared with obtain the specificity, sensitivity, and positive and negative predictive values. A receiver operating characteristics (ROC) curve was obtained from the false-alarm rate and the hit rate value of the head impulse test.

RESULTS: The specificity of the head impulse test was 0.91 and the sensitivity was 0.45. The positive predictive value was 0.92, the negative predictive value was 0.41, and the area under the ROC curve was 0.866. A canal paresis value of 42.5 percent was considered to be the limit of the normal response, as seen when the head impulse test was used to predict a normal or abnormal result in a given patient.

CONCLUSION: The head impulse test, when used as a bedside test, and the caloric test are by no means redundant methods. The information obtained form both can be used in combination to obtain a better insight into the degree of vestibular dysfunction of patients.