An experimental set-up to test heat-moisture exchangers

Humidification performance of 48 passive airway humidifiers: comparison with manufacturer data

INTRODUCTION

Heat and moisture exchangers (HMEs) are increasingly used in the ICU for gas conditioning during mechanical ventilation. Independent assessments of the humidification performance of HMEs are scarce. The aim of the present study was thus to assess the humidification performance of a large number of adult HMEs.

METHOD

We assessed 48 devices using a bench test apparatus that simulated real-life physiologic ventilation conditions. Thirty-two devices were described by the manufacturers as HMEs, and 16 were described as antibacterial filters. The test apparatus provided expiratory gases with an absolute humidity (AH) of 35 mg H(2)O/L. The AH of inspired gases was measured after steady state using the psychrometric method. We performed three hygrometric measurements for each device, measured their resistance, and compared our results with the manufacturer data.

RESULTS

Of the 32 HMEs tested, only 37.5% performed well (>or= 30 mg H(2)O/L), while 25% performed poorly (< 25 mg H(2)O/L). The mean difference (+/- SD) between our measurements and the manufacturer data was 3.0 +/- 2.7 mg H(2)O/L for devices described as HMEs (maximum, 8.9 mg H(2)O/L) [p = 0.0001], while the mean difference for 36% of the HMEs was > 4 mg H(2)O/L. The mean difference for the antibacterial filters was 0.2 +/- 1.4 mg H(2)O/L. The mean resistance of all the tested devices was 2.17 +/- 0.70 cm H(2)O/L/s.

CONCLUSIONS

Several HMEs performed poorly and should not be used as HMEs. The values determined by independent assessments may be lower than the manufacturer data. Describing a device as an HME does not guarantee that it provides adequate humidification. The performance of HMEs must be verified by independent assessment.

A novel method of evaluation of three heat-moisture exchangers in six different ventilator settings

OBJECTIVE

The purpose of this study was to assess and compare the humidification, heating, and resistance properties of three commercially available heat-moisture exchangers (HMEs). To mimic clinical conditions, a previously validated, new, realistic experimental set-up and measurement protocol was used.

DESIGN

Prospective, comparative experimental study.

SETTING

Surgical Intensive Care Unit, University Hospital of Rotterdam.

MATERIALS

An experimental set-up consisting of a patient model, measurement systems, and ventilator and three different HME types.

INTERVENTIONS

The air flow, pressure in the ventilation circuit, pressure difference over the HME, and partial water vapour pressure and temperature at each side of the HMEs were measured. Measurements were repeated every 30 min during the first 2 h and every hour up to 24 h for each HME at six different ventilator settings. The mean inspiratory and maximum expiratory resistance, flow-weighted mean absolute humidity and temperature outputs, and humidification and heating efficiencies of HMEs were calculated.

MEASUREMENTS AND RESULTS

The Dar Hygroster had the highest humidity output, temperature output, humidification efficiency, and heating efficiency values throughout the study (32.8 +/- 21. mg/l, 32.2 +/- 0.8 degrees C, 86.3 +/- 2.3%, and 0.9 +/- 0.01%, respectively) in comparison to the Humid-Vent Filter (25.3 +/- 3.2 mg/l, 31.9 +/- 0.8 degrees C, 72.2 +/- 5.3%, 0.9 +/- 0.02%, respectively) and the Pall Ultipor BB100 breathing circuit filter (23.4 +/- 3 mg/l, 28.3 +/- 0.7 degrees C, 68.8 +/- 5.9%, 0.8 +/- 0.02%, respectively). The inspiratory and expiratory resistance of the HMEs remained below clinically acceptable maximum values (2.60 +/- 0.04 and 2.45 +/- 0.05 cmH2O/l per s, respectively).

CONCLUSION

The Dar Hygroster filter was found to have the highest humidity and temperature output of all three HMEs, the Humid-Vent filter had a satisfactory humidity output only at low tidal volume flow rate and minute volume settings, whereas the Pall Ultipore BB 100 never achieved a sufficient humidity and temperature output.