Evaluation of tidal volume measurements of the anaesthesia device Tafonius® in vitro and in vivo

Abstract

The anaesthesiologic risk is higher in horses than in other domesticated animal species and thereby ventilation is a vital problem. The aim of this study was to evaluate the accuracy of the anaesthesia device Tafonius (R) tidal volume measurements in vitro as well as in vivo. In an in vitro study the whole system was flushed by air to reduce the air moisture to a minimum. The preparation of the anaesthesia device was exactly the same, as if there was a surgery on a patient. But instead of a horse, a seven liter calibration syringe was connected to the Y-piece of the anaesthesia device and the whole system was flooded by pure oxygen. By touch screen on the monitor of the anaesthesia device, a CPAP (Continuous Positive Airway Pressure), a tidal volume of 20 liters and an airway pressure of 20 cm H2O were adjusted. The leakage test was repeated and the connections tightened until the leak was 168 mL gas per minute. All the measurements were made randomly, without breaks or changes of the adjustments. For tidal volume measurements a seven liter calibration syringe was used. A connecting piece was specially manufactured, to attach the syringe to the Tafonius (R) without gas leakage.14 different tidal volumes were evaluated in a random order, covering the range from 500 mL to 7000 mL using 500 mL steps. To imitate respiration during anaesthesia, three respiratory rates were simulated: 30, 15 and 6 times per minute. All the measurements were performed in duplicate, once with NICO (R) interconnected and once without it. The respiratory rates and tidal volumes were chosen in random order, as well as whether the flow meter was interconnected or not. In a in vivo study twenty-one patients with a mean weight of 516 kilograms submitted for elective surgery were included. All of them received a balanced anaesthesia regime (inhalation anaesthesia with isoflurane and constant rate infusion with medetomidine (3.5 mcg/kg/h) and controlled mechanical ventilation. 16 horses were positioned in dorsal and five in lateral recumbency. The flow meter NICO (R) was connected ones a steady state in the plane of anaesthesia was reached. The flow meter was left during 38 to 190 breaths. A total of 2368 breaths were measured. Mean tidal volumes reached from 4107 mL to 8964 mL. In vitro data were analysed by descriptive statistics to determine measurement accuracy. Bland-Altman plots with limits of agreement (loa) were used to analyze in vivo data. The mean difference between the volume given by the calibration syringe and the volume shown on the Tafonius (R) screen was -19.45% (standard deviation 11%). The data stored on the computer showed an average difference of -20.5% (standard deviation 15.57%). Whether NICO (R) was interconnected or not resulted in a change in accuracy of 0.36% in the volume measurement of the Tafonius (R). The mean difference of the volume produced with the calibration syringe and the measured volume of the NICO (R) was 0.73% (standard deviation 4.3%). The Bland-Altman analysis from NICO (R) to Tafonius (R) gave a lower limit of agreement of -2427 mL and an upper limit of agreement of 1436 mL and a bias of -496 mL. From 500 mL to seven liter the Tafonius (R) measures tidal volumes 20% lower than a reference volume provided by a calibration syringe. Inadequate ventilation of the lung might be a consequence of this in horses under controlled mechanical ventilation. In vivo gained tidal volume data automatically stored by Tafonius (R) were very variable and probably a result of outliers. In vivo analyses of accuracy of tidal volume measurement should be repeated at the bedside, using data displayed on the screen of the device instead of automatically stored data.