A Meta-analysis of Diagnostic Test Agreement Between Eucapnic Voluntary Hyperventilation and Cardiopulmonary Exercise Tests for Exercise-Induced Bronchoconstriction.

INTRODUCTION

Exercise-induced bronchoconstriction (EIB) is very common in athletes. Cardiopulmonary exercise tests (CPET) have traditionally been used for the diagnosis of EIB. However, alternative indirect bronchoprovocation tests have recently been used as surrogate tests. One of these is the eucapnic voluntary hyperventilation (EVH). This meta-analysis studied the agreement between the two tests.

METHODS

An extensive search in PubMed and Medline was conducted for studies where participants underwent both CPET and EVH with measurement of forced expiratory volume in 1-second (FEV1). After extracting data using two-by-two contingency tables, pooled positive and negative agreements were first calculated between the two tests, with EVH benchmarked against CPET, and then, pooled positive and negative agreements were calculated with CPET benchmarked against EVH.

RESULTS

The pooled positive and negative agreements between EVH and CPET (with CPET as the reference) were 0.62 [(95% confidence interval 0.54-0.70), I2 77%] and 0.61 [(0.56-0.65)), I2 81%]. The pooled positive and negative agreements between CPET and EVH (with EVH as the reference) were 0.36 [(0.30-0.42), I2 93%] and 0.82 [(0.77-0.86), I2 78%]. The average of positive test results with EVH across all studies was greater than that of CPETs (58.84% vs. 39.51%).

CONCLUSIONS

Results of this meta-analysis show poor positive agreement between the two tests but high negative agreement (specifically using EVH as reference), suggesting that either test can be used for correctly identifying those without EIB. Results also suggest that the chances of a test resulting positive are higher with EVH than with CPET.