Patient misreporting may lead to underestimation of cough events.

It was with great interest that we read the article by Tutuian et al1 in a recent issue of CHEST (August 2006). The authors investigated temporal relationships between non-acid reflux and cough events in patients with persistent cough, utilizing impedance/pH monitoring and subjective recording of cough events by means of a diary and a self-operated digital data logger. Examining temporal relationships is important since both cough and reflux are frequent events; hence, coexistence alone does not imply causation. Establishing a reflux-cough association within a given time period may help to determine cause and effect, and may allow better patient selection for surgical intervention. However, there are issues we would like to highlight regarding the identification of cough. First, the authors reported a total median number of coughs of 8 (interquartile range, 3 to 21 coughs) over the entire 24-h monitoring period. This rate is orders of magnitude lower than those reported by Birring et al2 (mean [ SD] cough rate, 43 8 coughs per hour) and those determined in our own department (median rate, 11.06 cough-seconds per hour; range, 1.06 to 46 cough-seconds per hour).3 Second, the use of the subjective reporting of cough correlates poorly with objectively monitored cough sounds,4 particularly at night.5 The symptom index is a ratio of the number of cough events preceded by reflux to the number of cough events during the monitoring period, expressed as a percentage, and is taken to be positive if 50%. It seems likely that subjective reporting misses cough events, thus altering this ratio and increasing the chance of an apparently positive relationship. A validated ambulatory monitor, recording individual cough sounds, could be time-synchronized to the impedance/pH trace. This would enable a more accurate determination of the temporal relationship between cough and reflux.