aaBreathing airflow is a physiological variable commonly monitored for assessing respiratory patterns and nocturnal events in studies aimed at diagnosing sleep disorders [1][2][3]. Given that in sleep studies the use of a pneumotachograph, which is the reference transducer for measuring flow, is rather cumbersome, airflow is routinely assessed by means of thermally sensitive devices such as thermistors and thermocouples. The main advantage of these devices is that they are small and light and, consequently, their use causes a minimum amount of disturbance to the patient during sleep. However, thermistors and thermocouples are semiquantitative devices since the flow signal they provide is not a direct measure of actual flow [4][5][6]. This drawback, which does not pose major problems for detecting apnoeas, is however relevant to scoring hypopnoeas [7,8], given that the definition of these events is made in terms of quantitative reduction in airflow.Precise characterization of the performance of thermistor/thermocouples when used for measuring airflow in sleep studies requires a laboratory study under controlled conditions. Indeed, contrary to what happens when testing linear transducers, the analysis of a slow-response nonlinear system such as a thermistor/thermocouple [6] requires subjecting it to a variety of input airflows covering the ranges of amplitudes, frequencies and waveforms found in the particular application. Such a systematic study cannot be carried out in patients during sleep since it is not possible to modify the pattern of the breathing airflow in a selective and controlled way. Moreover, pneumo-tachographs, which may modify the thermal conditions in the thermistor/thermocouple, or thoracoabdominal bands, which may not be adequate during padoxical breathing, are not perfect reference flow transducers in this application. The aim of this study was to characterize the accuracy of thermistor/thermocouples as flow-measuring devices for detecting hypopnoeas. To this end, we set up a respiratory model to reproduce the measuring conditions of nasal airflow by means of thermally sensitive devices.
Materials and methodsThe response of thermistors and thermocouples was studied with the respiratory system model shown in figure 1. The model was based on a 60 L methacrylate chamber which was used as a buffer of heated air at 37°C. The air temperature in the chamber was measured with a mercury-bulb thermometer and was maintained (±0.2°C) by means of a 120 W heating resistance supplied with the required voltage. Homogeneous distribution of air temperature in the chamber was achieved by means of mixing airflow generated by an internal fan. A flow generator based on a servocontrolled linear motor attached to rubber bellows [9,10] We concluded that thermistor/thermocouples are inaccurate flow-measuring devices when used at the airflow conditions typical of sleep studies. Their use for quantifying hypopnoeas may lead to considerable underdetection of these respiratory events.