Abstract
BACKGROUND: Under-humidification and associated complications may occur with heated humidifiers. Hygrometric performances of heated wire humidifiers are reduced by high ambient and high outlet ventilator temperatures. Currently, there is no reliable monitoring tool to evaluate humidification performances of heated wire humidifiers in the daily practice. We sought to demonstrate the relation between humidity delivered by heated wire humidifiers and different parameters that could be used to monitor humidity of gas delivered to subjects.
METHODS: On a bench test, we measured heater plate temperature, inlet chamber temperature, and delivered humidity with MR850 system. Temperature displayed on the humidifier was also recorded. The measurements were performed at different ambient temperatures and five minute ventilation levels (5, 7.5, 10, 12.5, 15 L/min). Inlet chamber temperatures varied from 20–40°C. In each condition, hygrometric measurements with the psychrometric method were performed at steady state.
RESULTS: We performed 279 measurements at steady state including all conditions. We found a good correlation between heater plate temperature and absolute humidity delivered (R2 = 0.82). This relationship was hardly affected by ambient temperature, but minute ventilation had more effect. For different minute ventilations, the correlation between heater plate temperature and absolute humidity delivered was very good with coefficient of determination R2 from 0.87–0.98. Heater plate temperature > 62°C was a good predictor of absolute humidity delivered > 30 mg H2O/L (area under the curve = 0.96, sensibility 79%, specificity 94%). No correlation existed between humidity delivered and the outlet chamber temperature (displayed on the humidifier).
CONCLUSIONS: In this bench study, we have shown a good correlation between heater plate temperature and humidity delivered with a heated wire humidifier. This means that a “hidden” hygrometer is built into the heated wire humidifier. Heater plate temperature should be used as a surrogate of humidity to improve the humidification monitoring.
Introduction
Adequate humidification of the gas delivered to the patients during mechanical ventilation is mandatory.1-3 Heated humidifiers (HHs) as well as heat and moisture exchangers (HMEs) may be used for this purpose.1,3 Whereas clinicians frequently consider that HHs provide adequate humidification, it may not be the case in conditions such as high ambient temperature and high outlet ventilator temperature.4 During COVID-19 pandemic, high rates of endotracheal tube occlusions have been described with poor performing HMEs and with HHs when not used in optimal conditions.5-10 Endotracheal tube occlusions were a frequent cause of unexpected cardiac arrest in mechanically ventilated patients with COVID-19, frequently associated with unfavorable outcome.11 Endotracheal tube occlusions are related to under-humidification of the gas delivered to patients.12,13 It is recommended to deliver to mechanically ventilated patients a gas humidity > 30 mg H2O/L with HME (International Organization for Standardization [ISO] 9360) and > 33 mg H2O/L with HH (ISO 8185:1997).3 We showed that this humidity is not attained with many HMEs13 and with heated-wire HHs when ambient temperature is high.4 Measuring and monitoring the level of humidity in inspiratory gas delivered to the intubated patient are difficult, especially in the daily practice.14 It was shown that humidity delivered by HME is somehow related to the condensation in the flex tubing after the Y-piece.15,16 However, condensation on the flex-tubing inner wall or on the HH humidification chamber dome does not occur when ambient temperature is high, which represents the situations during which HH performances may be greatly reduced.4 When ambient temperature is high, heated wire humidifier may deliver humidity as low as 20 mg H2O/L.4 Therefore, a humidity-monitoring tool for these devices is desirable. We recently described a method used during the pandemic to avoid under-humidification in patients with COVID-19.6 As part of a bundle of measures to eliminate the endotracheal tube occlusions experienced at the beginning of the pandemic, we monitored the humidity delivered to the patients by measuring the heater plate temperature. The monitoring of the heater plate temperature (available from the submenus of the HH) by respiratory therapists was useful to optimize the settings of the heated wire humidifiers and ensure proper humidification.6,10 The objective of the present study was to describe the relationship between heater plate temperature and absolute humidity delivered to patients and the parameters that may influence this relation.
QUICK LOOK
Current Knowledge
Heated wire humidifiers are efficient at conditioning gases delivered to patients on invasive mechanical ventilation. When ambient temperature or outlet ventilator temperature is high, the heater plate temperature diminishes as per working principles and humidity delivered to patients may not be sufficient, leading to endotracheal tube occlusion or subocclusion. No monitoring tool exists in daily practice to evaluate humidity delivered by heated wire humidifiers to patients when ambient temperature is high.
What This Paper Contributes to Our Knowledge
This study shows that heater plate temperature is well correlated to the absolute humidity delivered at the Y-piece. This correlation is influenced by ambient temperature and minute ventilation. A minimal value of heater plate temperature should be targeted to ensure sufficient humidity delivered to patients and avoid complications related to under-humidification.
Methods
We conducted a bench study to evaluate the relationship between the heater plate temperature and the inspiratory humidity delivered to the patients with the MR850 HH (Fisher & Paykel Healthcare, Auckland, New Zealand) with settings recommended by the manufacturers (37°C at the humidification chamber and 40°C at the Y-piece).
We measured relative and absolute humidity of the inspiratory gases with the psychrometric method, as previously described.4,13 The bench test is described in Figure 1. An ICU ventilator, Puritan Bennett 840 (Medtronic, Dublin, Ireland), was set to provide different minute ventilations (5, 7.5, 10, 12.5, and 15 L/min). The tidal volume was maintained at 500 mL, and the breathing frequency was modified to obtain the targeted minute ventilation. We varied the ambient temperatures in the laboratory from 20–30°C and the inlet humidification chamber temperature of the HH from 20–40°C. Inlet chamber temperature variations were obtained by covering the dry line (between the ventilator outlet and the inlet humidification chamber) with some pieces of wet paper, as previously described.4
Diagram of the temperature regulation system of a heated humidifier with a heated circuit and of parameters recorded during the bench study. The relationship between the heater plate temperature and the humidity delivered at the Y-piece was specifically evaluated. Modified from reference 4.
The ambient temperatures, outlet ventilator, and inlet chamber were collected with high-precision thermometers (Duotemp, Fisher & Paykel). We recorded the temperature displayed by the HH main screen, which indicates the outlet chamber temperature, usually set at 37°C (as in the current protocol), and the heater plate temperature in the submenus of the MR850 HH. The different temperatures were recorded at the same time as the hygrometric measurements. Hygrometric measurements were performed after 2 h of steady state using the psychrometric method4,13,17,18 (Fig. 1).
Statistical Analysis
Data were expressed using mean ± SD to summarize the data. Relationship between absolute humidity (mg H2O/L) delivered at the Y-piece and heater plate temperature as well as by the minute ventilation was analyzed using linear regression models. Univariable and multivariable analyses were performed. Plots of externally studentized residuals and Cook D statistic revealed that no observation had influenced the fit model. Influence and fit diagnostics were examined, and the conclusion was data support the assumptions underlie the linear regression. Receiver operating characteristic (ROC) analysis was performed to determine the minimum heater plate temperature that produced an inspiratory absolute humidity ≥ 30 mg H2O/L. Optimal threshold values for ROC curves were determined according to the Youden index and the sensitivity/specificity balance. Relationships between tests were measured using the Pearson correlation coefficient. Statistical significance was present with a 2-tailed P < .05. Analyses were performed using SAS version 9.4 (SAS Institute, Cary, North Carolina).
Results
In total, we performed 279 hygrometric measurements in the tested conditions and collected all the different parameters and temperatures for all these measurements.
Overall, we found a good correlation between the heater plate temperature and absolute humidity delivered to subjects (R2 = 0.82, P < .001) (Fig. 2). The change in absolute humidity delivered for an increase of 1°C of the heater plate temperature was 0.74 mg H2O/L with a 95% CI 0.70–0.78. The correlation was barely influenced by the ambient temperature, whereas the minute ventilation had a significant impact (Fig. 3). After adjusting for ambient temperature only, the estimate remained unchanged. Adding the minute ventilation, the estimate in absolute humidity delivered for an increase of 1°C was inflated to 0.86 mg H2O/L (0.83–0.89). Here, there is clear evidence of a confounding effect due to the minute ventilation. Overall, the ROC curve indicates that a heater plate temperature > 62°C indicates an absolute humidity > 30 mg H2O/L with a specificity and sensibility of 94% and 79%, respectively.
Relationship between absolute humidity (mg H2O/L) delivered at the Y-piece and heater plate temperature (°C) of the MR850 for all conditions. The blue dotted line represents the regression line; the equation and the coefficient of determination are represented. When heater plate temperature is above 62°C, absolute humidity is most frequently above 30 mg H2O/L or near this value.
Relationship between absolute humidity (mg H2O/L) delivered at the Y-piece and heater plate temperature (°C) for different minute ventilation (from 5–15 L/min). The blue dotted lines represent the regression lines; the equations and the coefficient of determination are shown.
With a minute ventilation of 10 L/min, a heater plate temperature > 61°C predicted a humidity delivered to subjects > 30 mg H2O/L with a specificity of 98.7% and a sensibility of 100% (Fig. 4). The minimum heater plate temperatures to ensure adequate humidity went from 56 to 67°C for minute ventilations, going from 5 to 15 L/min (Figs. 3 and 4).
Receiver operating characteristic curves at the different minute ventilations tested (from 5–15 L/min) to determine the minimum temperatures of the heater plate to ensure absolute humidity at the Y-piece above 30 mgH2O/L. The Youden index is indicated for each minute ventilation. V̇E = minute ventilation.
With heater plate temperature > 62°C, adequate humidity was ensured with minute ventilations up to 12.5 L/min. For higher minute ventilations, heater plate temperature should be kept > 65°C (Figures S1 and S2, see related supplementary materials at http://www.rcjournal.com). We propose a simple algorithm to improve humidification of the gases when heater plate temperature is too low (Figure S3, see related supplementary materials at http://www.rcjournal.com).
There was no relationship between the temperature indicated on the MR850 screen (in most instances, the outlet humidification chamber temperature) and the humidification performance of the device (Figure S4, see related supplementary materials at http://www.rcjournal.com).
The high temperature of the heater plate did not translate into a high temperature of the gas delivered to the patient side. Indeed, according to the operating principles of the heated wire humidifier, the temperature of the gas in Y-piece remains at 40°C (Figure S5, see related supplementary materials at http://www.rcjournal.com). A similar relationship between heater plate temperature and humidity delivered likely exists with other heated wire humidifiers (Figure S6, see related supplementary materials at http://www.rcjournal.com).
Discussion
In this study, we tested the hypothesis that heater plate temperature was related to the humidity delivered. Heater plate temperature was well correlated with the inspiratory gas humidity measured at the Y-piece with the psychrometric method. This makes it possible to consider a new method of monitoring the humidity delivered by the MR850 heated wire humidifier with the heater plate temperature. Ambient temperature had little impact, but minute ventilation may influence this relationship.
The heater plate temperature variations are related to the working principles of the heated wire humidifiers as previously described, leading to humidity variations.17 The humidifier algorithm maintains the outlet chamber temperature around 37°C with recommended settings. When inlet chamber temperature is too high (due to high ambient temperature or high outlet ventilator temperature), the heater plate temperature is reduced to maintain the outlet chamber temperature at 37°C for usual settings. When the water bath temperature in the humidification chamber is too low, evaporation (change from liquid to gaseous state of water when there is enough energy) is reduced. Consequently, the water content of the gas (or absolute humidity) is reduced such as humidity delivered to the patients. The evaporation rate is influenced by several parameters, among which are the temperature of the water, the air temperature above the water, and the flow of the gas passing through the humidification chamber. The basic heated wire humidifier algorithm does not take into account several of these parameters but only tries to maintain constant the outlet chamber temperature. Some advanced algorithms to avoid under-humidification in case of high ambient temperature have been developed (compensation algorithm). However, these algorithms partially improve the humidification performances in these situations.4
To date, no monitoring tool exists to verify that the heated wire humidifier is providing proper humidification to avoid the complications of under-humidification.5,6,8,19 Monitoring of the humidity through the condensation at the flex tube has been proposed for HMEs.15,16 Similarly, the condensation at the dome of the HH humidification chamber may be used when ambient temperature is normal.4 However, this method is ineffective when the ambient temperature is high, a condition that prevents condensation from occurring on the dome wall. Recently, it has been proposed to monitor the temperature inside the endotracheal tube.20 However, this variable does not provide sufficient monitoring as only temperature associated with relative humidity may provide insight on water content of the gas. Some HHs monitor relative humidity, but the accuracy of this measurement is unknown.
The new method of monitoring that we propose, based on heater plate temperature monitoring, was used during the pandemic to detect under-humidification.6 At the beginning of the pandemic, several tube occlusions occurred in a short period of time in mechanically ventilated patients whose humidification system was an HH. This was likely related to very high ambient temperatures after the installation of negative-pressure devices to manage patients with COVID-19.6,10 After the implementation of several measures including monitoring the heater plate temperature, there were no more tube occlusions.6 A low heater plate temperature (< 62°C) triggered a change in the HH settings to increase the outlet chamber temperature and, therefore, raise the heater plate temperature. This value of 62°C was derived from a set of experiments obtained on a bench with similar evaluations compared to those conducted on the present study.
HH manufacturers usually claim a humidity delivered of 44 mg H2O/L at the Y-piece, but much lower humidity is commonly delivered to patients.4,5 In optimal conditions, absolute humidity around 35–40 mg H2O/L may be delivered, whereas in nonoptimal conditions (high ambient air, high outlet ventilator temperature, or when the sun hits the humidifier directly) absolute humidity may be < 20 mg H2O/L,4 which greatly increases the risk of occlusion or subocclusions.3,6,13 In these situations, the monitoring of the heater plate may be useful to reduce the risk related to under-humidification. As a rule of thumb, with heater plate temperature > 62°C, the inspiratory humidity should be acceptable. Interestingly, as this study shows, the outlet chamber temperature displayed on the HH screen only indicates that the HH is switched on and operating according to its working principle. It provides no indication whatsoever on the humidity delivered.
The objective of clinicians should not be to reach the maximum temperature of the heater plate but to maintain this temperature above a minimum value to ensure good humidity of the gas delivered to the patients. If the heater plate temperature is too high, there is a potential risk of over-humidification and excessive condensation in the circuits. Under normal conditions, very high humidity or temperature of the gas is not expected to occur with the tested heated wire humidifier. In the present study, the maximum absolute humidity reached was 40.7 mg H2O/L, which is not considered as over-humidification21,22; and heated wire temperature was 77°C, which is also the maximum value that was reached in this study. Consequently, heater plate temperatures up to 75°C are safe for the patients, without risk of over-humidification. In addition, it must be emphasized that even with a high heater plate temperature the temperature of the Y-piece does not increase beyond the limit imposed by the humidifier (generally 40°C). Thus, the temperature in the airways is not modified by high heater plate temperatures (Figure S5, see related supplementary materials at http://www.rcjournal.com).
Monitoring the heater plate provides additional safety information for clinicians when HHs are used in patients. Indeed, contrary to usual belief, latest-generation HHs (heated wire humidifiers) do not always provide adequate humidity and may be responsible for endotracheal tube occlusions.6,8 This HH monitoring may be integrated in the regular monitoring conducted by the respiratory therapists or in situations at risk of under-humidification (ambient temperature > 25°C, turbine ventilators, south-facing rooms, inefficient air conditioning). If heater plate temperature is below a minimum value (ie, below 61°C with minute ventilation around 10 L/min), the humidifiers settings should be modified. Compensation algorithm may be activated, or gradient temperature between the humidification chamber and the Y-piece should be reduced.4 Further monitoring of the heater plate should be done 15–30 min after the modification of the HH settings. Ideally, the heater plate temperature should be easily accessible to clinicians, whereas it actually requires the exploration of submenus.The newer generation of heated wire humidifiers take into account all these parameters in the algorithm and seem to provide more stable humidity over a large range of ambient temperatures and minute ventilation.23
This paper has limitations. This was a bench study, and all critical scenarios may not have been covered. However, we think that the usual range of minute ventilation and ambient temperature was evaluated. With minute ventilations lower than those evaluated in this study (the lowest minute ventilation was 5 L/min) as used in infants and neonates, the relationship with heater plate temperature described in this study may differ. The clinical application at the bedside of data collected on a bench study may be questioned. However, the conditions tested on bench reproduced the temperature and ventilator settings frequently encountered in the clinical settings. In addition, this method of monitoring has already been used and was helpful to control a clinical situation leading to severe under-humidification due to high temperatures.6
A relationship between HH and absolute humidity likely exists for all heated wire humidifiers, but each relationship is unique and may be modified by many factors (eg, area of exchange for the water and of the dome of the humidification chamber, properties of the materials used to make the humidifiers). Consequently, the targeted minimum heater plate temperatures proposed in this study should be used only with the MR850.
Conclusions
We demonstrated that heater plate temperature may be considered as a hygrometer integrated to the humidifier and may be used to monitor the humidity delivered by the MR850 HHs. In general, a heater plate temperature > 62°C will ensure a sufficient humidification of the gas delivered to the patient. When minute ventilation is high (> 12.5 L/min), slightly higher heater plate temperature may be targeted.
Footnotes
- Correspondence: François Lellouche MD PhD, Centre de recherche de l’Institut universitaire de cardiologie et de pneumologie de Québec. 2725, Chemin Sainte-Foy. G1V4G5, Québec, QC, Canada. E-mail: francois.lellouche{at}criucpq.ulaval.ca
Dr Lellouche discloses a relationship with Fisher & Paykel Healthcare. The other authors have disclosed no conflicts of interest.
Devices and consumables were provided by Fisher & Paykel Healthcare, who had no other involvement in the study.
Supplementary material related to this paper is available at http://www.rcjournal.com.
- Copyright © 2022 by Daedalus Enterprises