Article, Respiratory Medicine

Flow resistance, work of breathing of humidifiers, and endotracheal tubes in the hyperbaric chamber

Original Contribution

Flow resistance, work of breathing of humidifiers, and endotracheal tubes in the hyperbaric chamber

Ran Arieli PhD a, Yohanan Daskalovic PhD a, Ofir Ertracht MSc a, Yehuda Arieli PhD a,

Yohai Adir MD a, Amir Abramovich MD a, Pinchas Halpern MD b,?

aIsrael Naval Medical Institute, Haifa, Israel

bEmergency Department, Tel Aviv Medical Center, Tel Aviv, Israel affiliated to the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv 64239, Israel

Received 25 January 2010; accepted 10 February 2010

Abstract Humidification of inspired gas is critical in ventilated patients, usually achieved by heat and moisture exchange devices (HMEs). HME and the endotracheal tube add airflow resistance. Ventilated patients are sometimes treated in hyperbaric chambers. Increased gas density may increase total airway resistance, peak pressures (PPs), and mechanical work of breathing (WOB). We tested the added WOB imposed by HMEs and various sizes of ETT under hyperbaric conditions. We mechanically ventilated 4 types of HMEs and 3 ETTs at 6 minute Ventilation volumes (7-19.5 L/min) in a hyperbaric chamber at pressures of 1 to 6 atmospheres absolute (ATA). Peak pressure increased with increasing chamber pressure with an HME alone, from 2 cm H2O at 1 ATA to 6 cm H2O at 6 ATA. Work of breathing was low at 1 ATA (0.2 J/L) and increased to 1.2 J/L at 6 ATA at minute ventilation = 19.5 L/min. Connecting the HME to an ETT increased PP as a function of peak flow and chamber pressure. Reduction of the ETT diameter (9 N 8 N 7.5 mm) and increase in chamber pressure increased the PP up to 27.7 cm H2O, resistance to 33.2 cmH2O*s/L, and WOB to 3.76 J/L at 6 ATA with a 7.5-mm EET. These are much greater than the usually accepted critical peak pressures of 25 cm H2O and WOB of 1.5 to 2.0 J/L. Endotracheal tubes less than 8 mm produce significant added WOB and airway pressure swings under hyperbaric conditions. The hyperbaric critical care clinician is advised to use the largest possible ETT. The tested HMEs add negligible resistance and WOB in the chamber.

(C) 2011


Humidification of inspired respiratory gas is critical in the care of the mechanically ventilated patient. Humidification minimizes airway mucosal damage and Endotracheal tube occlusion by accreted secretions [1]. There are a number of methods to achieve respiratory gas humidifica-

* Corresponding author. Tel.: +972 3 6973829; fax: +972 3 6974670.

E-mail address: [email protected] (P. Halpern).

tion, but because of cost and ease of use considerations, disposable heat and moisture exchange devices (HMEs) have become increasingly popular [2]. Heat and moisture exchange devices, however, create an added respiratory dead space and may affect airflow resistance and airway pressures [3]. Hyperbaric oxygen therapy is commonly used in critically ill, Mechanically ventilated patients, for condi- tions such as carbon monoxide intoxication, anaerobic soft tissue infections, Crush injury, and others. The increased density of hyperbaric respiratory gas may adversely affect respiratory mechanics [4,5] and further exacerbate the effects

0735-6757/$ – see front matter (C) 2011 doi:10.1016/j.ajem.2010.02.003

of an HME on total airway resistance and peak pressures as well as on the patient’s work of breathing (WOB). In addition, ETTs, an unavoidable component of the respiratory circuit in ventilated patients, may also impose significant added resistance and WOB in inverse proportion to their diameter [6,7]. We tested a number of commercially available HMEs and various sizes of ETTs on imposed resistance and WOB under hyperbaric conditions to establish their effect on airway pressures and added WOB in the hyperbaric chamber.


The experimental setup was constructed within a multi- place hyperbaric chamber. A breathing machine (type 1, Reimers Eng Inc, Alexandria, Va) drove air back and forth, either through a humidification box or bypassing it, to an HME that was either connected or disconnected from an ETT, which was open to chamber air (Fig. 1). The frequency and tidal volume of the breathing machine were regulated by controls located outside the hyperbaric chamber. Inspired and expirED volumes were measured using an in-line turbine spirometer (Spirometer, K L Engineering, Northridge, Calif). The pressure difference over the HME and ETT was measured using a pressure transducer (Valydine DP-45, Northridge, Calif). Humidity and temperature were measured inside the inspiratory tube leading from the simulator to the HME (EE20FT, EE Electronics, Linz, Austria) proximal to it. The output from the volume meters and the pressure transducer were recorded on a strip-chart recorder (Gould Inc, Cleveland, Ohio) and sampled using a data logger (CODAS, DATAQ Instruments Inc, Akron, Ohio) for further analysis.

Four types of HMEs were evaluated:

  1. Adult (DAR, Mallincrodt Medical, Hazelwood, Mo)
  2. Adult (Inter-Therm HMEF, Intersurgical, Woking- ham, Berkshire, UK)

Fig. 1 Experimental setup. The experimental setup was mounted inside the hyperbaric chamber. The breathing machine drove air back and forth, measuring volumes via volume meters. The “inspired gas” was either bubbled through humidification box or bypassed it. The tidal ventilation passed through the humidity and moisture exchanger (HME) and the ETT.

  1. Pediatric model (Clear-Therm Mini, Intersurgical, Wokingham, Berkshire, UK)
  2. Adult Hygrovent-S (Medisize, Hillegom, the Netherlands)

Three internal diameters (IDs) of ETTs were tested: 7.5, 8.0, and 9.0 mm.

Experimental procedure

The HME was connected to the experimental setup with or without the ETT. At an ambient pressure of 1 atmosphere absolute [ATA]), the breathing machine was operated in a sequence of all 6 combinations of the 2 frequencies 10 and 15 breaths per minute (bpm) and 3 tidal volumes of 0.7, 1.0, and 1.3 L in a sinusoidal cycle. The various combinations yielded MV rates of 7 to 19.5 L/min. A steady state in each combination was reached within a few simulated breaths and the relevant variables were sampled. The hyperbaric chamber pressure was raised sequentially to 2, 3, 4, 5, and 6 ATA. The 6 combinations of frequency and tidal volume were tested at each pressure level. The temperature within the hyperbaric chamber was maintained at a steady 27?C to 28?C.

Experimental protocol

The combinations of tested HMEs, ETTs, and humidity are listed below. First, the various HMEs were compared without the ETTs, then the effect of the added ETT was assessed, and then the effect of added air humidification was tested. The respiratory pattern was sinusoidal with a 1:1 inspiratory to expiratory ratio (except for protocol 7).

Protocol 1: HME A without humidification Protocol 2: HME B without humidification Protocol 3: HME C without humidification Protocol 4: HME D without humidification

Protocol 5: HME A + ETT 7.5 mm ID without humidification

Protocol 6: HME A + ETT 8 mm ID without humidification, (I/E ratio 1/1 and I/E ratio 2/3)

Protocol 7: HME A + ETT 9 mm ID without humidification

Protocol 8: HME A + ETT 8 mm ID with humidification.


The data logger program (CODAS) was used to find peak inspired and expired pressures (cm H2O) and peak flows (L/s) at times of peak pressures, which coincided with the steepest part of the volume record. Work of breathing in joules per liter was calculated from the area of the pressure volume loop through the respiratory cycle. The mechanical system was repeatable with negligible differences between breaths for the same conditions, and so it was assumed that there is no variation in the same condition.


Heat and moisture exchange device alone

      1. Peak pressures

Peak inspiratory and expiratory pressures showed a linear relationship with the peak flow both during inspiration and expiration. Peak pressures and peak flows during expiration were similar to those during inspiration; therefore, only the values for inspiration are presented (Fig. 2). Peak pressures at the same peak flow increased with increasing chamber pressures. Thus, peak pressure ranged up to 2 cm H2O at 1 ATA in all 4 HMEs and rose to 6 cm H2O at 6 ATA. Flow resistance at peak flow during inspiration and expiration changed linearly with peak flow.

      1. Work of breathing

Work of breathing imposed by the HME alone was low at

1 ATA (up to 0.2 J/L) and increased gradually to approximately 1.2 J/L at 6 ATA at minute ventilation (MV) = 19.5 L/min (Fig. 3). The increase in WOB as a function of MV was fairly linear and moderate at low chamber pressures. Work of breathing increased significant- ly faster at higher chamber pressures, and there were no significant differences between the 4 tested HMEs.

Heat and moisture exchange device with ETT attached

      1. Peak pressure

Peak pressure with the combination of HME and ETT also increased as a function of peak flow and chamber

Fig. 2 Linear relationship of peak inspiratory pressure plotted as a function of peak flow in the 4 tested humidifiers A to D. Lines in each panel represent relationships at each of the 6 chamber pressures tested. Chamber pressure values in ATA are shown near the lines.

Fig. 3 Work of breathing in the 4 tested humidifiers A to D plotted as a function of minute ventilation. The different symbols in each panel represent the results at the 6 pressures tested.

pressure (Fig. 4). As the diameter of the ETT declined from 9 mm (lower right panel) to 8 mm (upper left panel) to 7.5 mm (lower left panel), the peak pressure increased. Table 1

Fig. 4 Linear relationship of peak inspiratory pressure is plotted as a function of the peak flow with HME A (no humidification) connected to an ETT. Endotracheal tubes with inner diameters of 7.5 and 9.0 mm are shown in the lower 2 panels and those with an 8.0-mm inner diameter with or without humidification are shown in the upper 2 panels. The 6 lines in each panel represent the results for the 6 chamber pressures tested. Pressure values in ATA are shown near the lines.

7.5-mm ETT

9.0-mm ETT











Peak pressure





(cm H2O)






shows the extremes of generated pressures, ranging from 3.2 cm H2O at 1 ATA with a 9-mm-ID ETT to 27.7 cm H2O at6 ATA with a 7.5-mm-ID ETT.

Table 1 Resistance, peak inspiratory pressure, and WOB for HME A (no humidification) with the narrower (7.5 mm) and wider (9.0 mm) ETT at the highest inspiratory flow at 1 and 6 ATA


      1. Resistance

The effect of ETT size on flow resistance was similar to the effect on peak pressure. The 9-mm-ID ETT had a minor effect on resistance at 1 ATA [2.8 cmH2O*s/L, Table 1), but resistance increased with increasing ambient pressure [9.9 cmH2O*s/L at 6 ATA, Table 1]. The 7.5-mm-ID ETT introduced much larger resistances, that is, 6.7 and 33.2 cm H2O/(s . L) at 1 and 6 ATA, respectively. Thus, increased ambient pressure had a prominent effect on flow resistance with small-diameter ETTs.

      1. Work of breathing

Mechanical (imposed) WOB at the extreme combinations of narrow ETTs and high ambient pressure (6 ATA and 7.5 mm ETT) was very high (3.76 J/L; Fig. 5 left lower panel and Table 1). It was much lower (0.34 J/L) at 1 ATA with a 9- mm-ID ETT.

      1. Humidity

There was no discernible effect of added humidity on resistance at peak flow. The addition of humidification to 99% relative humidity produced no additional effect on peak pressures (Fig. 4, upper panels).

      1. I/E ratio

The 2 cycles tested in protocol 6, that is, a sinusoidal cycle (I/E ratio = 1/1) and a 2/3 I/E ratio cycle, yielded the same peak pressure as a function of peak flow.

      1. Laminar vs turbulent flow

Flow in all test systems was laminar or nearly laminar at all minute volume, HME, ETT, and chamber pressure combinations. This was evidenced by the linear relationship between peak pressure and peak flow. We also calculated the Reynolds number at each condition (the Reynolds number denotes the transition between laminar and turbulent flow; it is calculated as flow velocity x fluid density x tube diameter/viscosity. Air flow becomes turbulent when the

Fig. 5 Work of breathing of the HME A (no humidification) with ETTs of 7.5 to 9.0 mm inner diameter at 1 to 6 ATA ambient pressures as a function of MV. The 6 different symbols in each panel represent the results for the 6 ambient pressures.

Reynolds number exceeds 2000). With an HME alone (protocols 1 to 4), the Reynolds number never exceeded 2000. With the addition of an ETT, however, the Reynolds numbers ranged between 2000 and 2500 in 21/144 measurements at peak flow, and only when transitional times were short (i.e., 20-60 milliseconds).


Because HMEs and ETTs are placed distal to the Y connector of the ventilation circuit, they are not just part of the circuit but also part of the artificial airway [2]. They are expected to increase inspiratory and expiratory resistance and therefore add mechanical WOB and produce high negative inspiratory and positive expiratory pressures. Work of breathing added by mechanical ventilators and ETTs was shown to vary widely between 0.3 and 2.5 J/L [8]. Those authors considered an added WOB of 1.5 J/L as “critical.” Warkander et al [9] also suggests a permissible maximal external WOB of 1.5 to 2.0 J/L, whereas the US Navy diving authorities permit an added WOB of 1.37 J/L (Table 3.3 in reference [10]). Girault et al [3] showed that use of a humidification device in difficult to wean patients increased all inspiratory effort variables.

Heat and moisture exchange device alone

In the present study, ventilation with an HME alone of all types and at all chamber pressures resulted in an added WOB

of less than 1.2 J/L, that is, less than “critical.” The added resistance and consequent mechanical WOB imposed by the HME alone are within acceptable physiological limits.

Added ETT

The addition of an ETT resulted in an added WOB N1.5 J/ L at ETT diameters of 8 mm or less at MV volumes of more than 15 L/min (Fig. 5).

Airway pressure

In the present study, we also recorded negative inspiratory pressure values that varied from as low as 0.7 cm H2O at 6 ATA at low MV volumes with an adult HME without an ETT to extremely high values of up to 27.7 cm H2O with an HME and a 7.5-mm ETT at an MV volume of 19.5 L/min at 6 ATA chamber pressure. Acceptable negative inspiratory pressures are variously quoted as -11 cm H2O [9] and -25 cm H2O [11]. Thus, the maximal pressure values of 6 ATA that were measured in the present study are more than twice the maximal recommended US values and well above the European limits. Negative pressure breathing would be detrimental in spontaneously breathing patients, potentially resulting in pulmonary edema and rapid respiratory muscle fatigue [8]. The large differences between ETT sizes must be emphasized, with a 9-mm-ID ETT producing just 5 cm H2O and an 8-mm-ID ETT producing 18 cm H2O negative inspiratory pressure at 6 ATA.

Most hyperbaric therapy for non-diving-related injuries

is administered at pressures of up to 3 ATA, yet even at these pressures, with a small-diameter ETT and high minute volume or flow requirements, pressures and WOB become excessive. Certainly, if ambient pressures greater than 3 ATA are used or contemplated, such as occurs with treatment of arterial air embolism or diving accidents due to very deep diving, the issue of added WOB and resistance becomes major and should be taken into account.

Many patients are mechanically ventilated while main- taining spontaneous ventilation as well. They are, by necessity, required to breathe via the ventilator circuit, and the added WOB as well as the negative inspiratory pressures generated by the circuit tubing, HME, and ETT is already considerable and often requires compensation by the addition of airway pressure support, various tube resistance compensation systems, and other means. In the hyperbaric chamber, the elevated gas density contributes to the production of very high WOB and negative inspiratory pressures when there are combinations of small-caliber ETTs, high inspiratory flows, and high ambient pressures. The HME itself, however, seems to add little clinically relevant resistance or WOB. If the hyperbaric critical care physician has to make technical adjustments to minimize hyperbaric airway airflow resistance and added mechanical WOB, choosing the largest acceptable ETT and minimizing

minute volume seem to be the most effective options. It must also be noted that most hyperbaric chamber mechanical ventilators have only limited intrinsic capacity to compensate for resistance and to maintain constant inspiratory volumes in the face of large increases in resistance [12]. Decreased tidal volumes are thus the norm rather than the exception in the chamber, and they are largely explained by the added mechanical resistance of the artificial airway. Expiratory resistance may result in prolongation of expiration, dynamic hyperinflation, air trapping, and autopositive end-expiratory pressure [2].

Limitations of the study

The present study was performed using a mechanical breathing simulator producing a sinusoidal curve. Actual patients may produce flow patterns that are different and may thus engender different responses by the respiratory tubing, both for better and for worse. Any Clinical decisions need to be monitored closely for actual clinical impact.


The ETT diameter is critical in its effect on airway pressures and added WOB in the hyperbaric chamber, especially at pressures of 3 ATA and above. Endotracheal tubes with an internal diameter smaller than 8 mm may produce unacceptable resistance and added mechanical WOB at high MV volumes and high chamber pressures, and should probably be avoided in adults. A HME adds little resistance and WOB even at high chamber pressures and can be safely used.


  1. Branson RD. Secretion management in the mechanically ventilated patient. Respir Care 2007;52:1328-42 [discussion 1342-1347].
  2. Iotti GA, Olivei MC, Braschi A. Mechanical effects of heat-moisture exchangers in ventilated patients. Crit Care 1999;3:R77-R82.
  3. Girault C, Breton L, Richard JC, et al. Mechanical effects of airway humidification devices in difficult to wean patients. Crit Care Med 2003;31:1306-11.
  4. Calzia E, Stahl W, Radermacher P. Work of breathing imposed by different ventilators under hyperbaric conditions (Abstract). Crit Care 1998;2(Suppl 1):106.
  5. Kindwall EP, Whelan HT, editors. Hyperbaric medicine practice, 2nd ed. Flagstaff (Ariz): Best Publishing Company; 1999. p. 271-5.
  6. Bock KR, Silver P, Rom M, Sagy M. Reduction in tracheal lumen due to endotracheal intubation and its calculated clinical significance. Chest 2000;118:468-72.
  7. Haberthur C, Elsasser S, Eberhard L, et al. Total versus tube-related additional work of breathing in ventilator-dependent patients. Acta Anaesthesiol Scand 2000;44:749-57.
  8. Bersten AD, Rutten AJ, Vedig AE, Skowronski GA. Additional work of breathing imposed by endotracheal tubes, breathing circuits, and intensive care ventilators. Crit Care Med 1989;17:671-7.
  9. Warkander DE, Norfleet WT, Nagasawa GK, Lundgren CEG. Physiologically and subjectively acceptable breathing resistance in divers’ breathing gear. Undersea Biomed Res 1992;19: 427-45.
  10. U.S. Navy unmanned test methods and performance goals for underwater breathing apparatus. Panama City (Fla): Navy Experimen- tal Diving Unit; 1994. Technical manual no. 01-94.
  11. European Standard EN 250:2000/A1:2006. Respiratory equipment – Open-circuit self-contained Compressed air diving apparatus – Requirements, testing, marking. Brussels, Belgium, European Com- mittee for Standardization, 1993, p. 7.
  12. Stahl W, Radermacher P, Calzia E. Functioning of ICU ventilators under hyperbaric conditions – comparison of volume- and pressure- controlled modes. Intensive Care Med 2000;26:442-8.