The amplitude of circadian FEV1 variation suitable for phase determination by cosinor analysis in stable asthmatic children.

To elucidate the criterion for amplitude of circadian FEV1 variation suitable for phase analysis, six FEV1 measurements within a day were made on 346 days in 125 stable asthmatic children aged 8 to 12 years. Since coefficient of variation (CV) of FEV1 within a day in nonasthmatic children in our laboratory is less than 5 percent, circadian FEV1 variations were divided into CV greater than or equal to 5 percent and CV less than 5 percent groups. Forty-two (12 percent) variations with significant fit of cosinor rhythm were CV greater than or equal to 5 percent. Troughs in 32 (76 percent) variations of the 42 were distributed between 12 PM and 6 AM, and the deviation was highly significant against a uniform distribution (p less than 0.001). This was consistent with nocturnal asthma. Thirty-five (10 percent) variations with significant fit of cosinor rhythm were CV less than 5 percent. Troughs in 12 (34 percent) variations of the 35 were distributed between 12 PM and 6 AM, and the deviation from a uniform distribution was nonsignificant. These suggest that 5 percent or more of CV of FEV1 may be clinically suitable for analysis of circadian FEV1 variation.