- Email: [email protected]
The effects of circuit and humidifier type on contamination potential during mechanical ventilation: A laboratory study
The effects of circuit and humidifier type on contamination potential during mechanical ventilation: A laboratory study i
lan J. Gilmour, MD a Michael J. Boyle, RRT b Andrew Streifel, MPH d R. Carter McComb, RRT c Minneapolis, Minnesota
Background: This study was undertaken because of concerns that ventilator humidifiers could be exacerbating the problem of nosocomial pneumonia in patients receiving mechanical ventilation. Methods: Four different brands of humidifiers were used in conjunction with a Siemens Servo 900B mechanical ventilator (Siemens Life Support Services, Solna, Sweden). In the first part, the ventilator was operated with humidifiers filled with contaminated water at room temperature. The viability of airborne particles and the effect of flow rates on the number of particles produced were assessed. In the second part, we measured the effect of time and temperature on bacterial survival in humidifier chambers. Because only bubble-through humidifiers were determined to produce infectious particles, the speed with which a contaminated bubble-through humidifier could infect circuit condensate was also determined. Aliquots of chamber water and circuit condensate, as well as air samples and distal circuit swabs, were cultured. Results: Humidifiers other than bubble-through humidifiers did not produce aerosols. Particle production by bubble-through humidifiers varied directly with flow rate (R2 = 0.91). Chamber temperatures did not affect chamber colony counts except in bubble-through humidifiers. Although chamber colony counts in bubble-through humidifiers decreased with time, organisms remained viable throughout the study. When bubble-through humidifiers were heated, both condensate and effluent gas became heavily contaminated within minutes of flow initiation. Conclusions: Bubble-through humidifiers produce aerosols that readily contaminate both circuit condensate and effluent gas. Avoiding bubble-through humidifiers should improve patient safety while allowing changes in practice that can result in significant cost :savings. (AJIC AM J INFECTCONTROL1995;23:65-72)
A l t h o u g h w a r m i n g a n d h u m i d i f i c a t i o n o f inspired gas may be accomplished satisfacl:orily w i t h h e a t - m o i s t u r e e x c h a n g e r s f o r s h o r t From the Department of Anesthesiology, University of Minnesota Medical School,a the Department of Cardiopulmonary Servicesb and Hospital Administration,° University of Minnesota Hospital and Clinics, and the Department of Environmental Health and Safety, University of Minnesota School of Public Health,d Minneapolis. Reprint requests: lan J. Gilmour, MD, Department of Anesthesiology, University of Minnesota Medical School, 420 Delaware St., S.E., Box 294, Minneapolis, MN 55455. Copyright © 1995 by the Association for Professionals in Infection Control and Epidemiology, Inc. 0196-6553/95 $3.00 + 0 17/46/60094
p e r i o d s , a c t i v e h u m i d i f i c a t i o n is still r e c o m mended for long-term mechanical ventilation. 1 Unfortunately, the relationship between added moisture and nosocomial pneumonia has long been acknowledged. Although humidif i e r s a r e t h o u g h t to b e a n u n c o m m o n c a u s e o f t h i s p r o b l e m , 2,3 a r e c e n t s t u d y s u g g e s t s t h a t t h e i r u s e is a s s o c i a t e d w i t h a h i g h e r i n c i d e n c e of pneumonia.* Some time ago, Rhame and coworkers s demonstrated that bubble-through humidifiers produce aerosols, which could transmit bacteria through patient circuits. 5 They theorized that true humidifiers would not do this. 65
AJIC 66
April 1995
Gilmour et al.
Table
1. Characteristics of tested humidifiers
Humidifier F&PMR 630 M SCT3000 RCI Concha lit+ PB Cascade l
Type
Chamber volume (ml)
Passover Passover Wick Bubble-through
40 60 50 850
Humidifier chamber MR 390 541020 Low compliance -
Continuous reservoir feed
Servo controlled
Yes Yes Yes No
Yes Yes Yes No
F&P, Fisher & Paykel; M, Marquest; PB, Puritan-Bennett,
T a b l e 2 . Results of a s s e s s m e n t of viable 3article production Humidifier
Particles/ft s
F&P MR 630 M SCT3000 RCI Concha Ill + PB Cascade I
<10" < 1O* < 1O* > 96*
Swab bacterial growth
Air sample bacterial growth (cfu) m
m
> 21-50
For details of procedure, see Methods. For distribution of particle size,see Fig. 5. Therewas no growth from distal circuit swabswith any of the humidifiers, Bacterial growth from Cassella slit sampler cultures occurred only with the bubble-through humidifier, Abbreviations as in Table 1. *Background count is < 10 particles/ft3,
We investigated this theory because bubblethrough humidifiers are still in c o m m o n use in m a n y places and because there is considerable uncertainty about their role in nosocomial infections. 6 We attempted to determine whether bubble-through humidifiers produce microaerosols that transmit bacteria, whereas true humidifiers do not. We also assessed the effects of time and temperature on bacteria in the chambers of these humidifiers, to ascertain whether safety claims based on these effects were credible. 7 METHODS
Fisher & Paykel MR 630 (Fisher & Paykel Allied Products, Auckland, New Zealand), Marquest SeT 3000 (Marquest Medical Products, Inc., Englewood, Colo.), Hudson RCI Conchatherm III Plus (Hudson Respiratory Care, Inc., Temecula, Calif.), and Puritan-Bennett Cascade I (Puritan-Bennett Corp., Overland Park, Kan.) humidifier bases were used in the study. Comparative information on these humidifiers is provided in Table 1. The humidification chambers selected for use with the appropriate bases were as follows: Fisher & Paykel MR 390, Marquest Extended Run (#541020), and RCI Pediatric Conchatherm. The first part of the study was carried out in a vertical flow high efficiency particulate air-filtered clean room (Enviromedic, Inc., Albuquerque, N.M.). The clean room was operated continuously for at least 15 minutes before study periods, after which control air samples were assessed for
particle counts with an airborne particle counter (eL 8060; Climet Inc., Redlands, Calif.) and for bacteria with a high-volume slit sampler (Cassella and Co., Ltd., Britannia Walk, Northern Ireland). The four humidifier systems were assembled according to manufacturers' instructions and 60 inches of 22 m m internal diameter flexible disposable tubing was attached to each outlet port. A Siemens Servo 900B Ventilator (Siemens Life Support Services, Solna, Sweden) set to deliver a tidal volume of 750 ml at a respiratory rate of 12 breaths/min, with inspiratory time of 33% and medical air (inspired fraction of oxygen 0.21), was used to test each unit. A bacterial filter was placed on the ventilator's outlet port and flexible polyvinyl chloride tubing was used to connect the filter with the inlet port of the humidifier. The dry, unheated systems were purged until fewer than 10 particles/ft3 were being emitted; the humidifier chambers were then filled to m a x i m u m operating level with water containing Pseudomonas cepacia at a concentration of 2.0 x l0 s cells/ml. After inoculation, each system was operated at room temperature for 90 minutes, after which a 2-minute air sample was obtained with the slit sampler and the distal end of the 60-inch tubing was swabbed. The swab samples were streaked on standard methods agar and incubated, along with the air sample cultures, at 35 ° C for 48 hours. Because only the bubble-through humidifiers produced particles, the relationship between flow rate and particle generation by bubble-through hu-
AJIC Volume 23, Number 2
Gilmour et al.
67
Table 3. Bacterial survival in the chamber and the circuit during use of a bubble-through humidifier Time (min)
Chamber
Swab
Cassella
0 5 10 20 30 60 90 120 24 hr
0 1,2 x 105 78 112 31 29 5 10 20
0 21-50 21-50 6-20 1-5 1-5 1-5 1-5 >50
0 140 2 3 2 7 2 3 1
Condensate
>3.0 >3.0 >3,0 >3.0 > 3.0 >3,0 >3,0 >3.0
0 x x x x x x x x
103 103 103 10 a 103 10 s 10 s 103
Samples for assessing bacterial growth were taken from bubble-through humidifier (chamber), dependent toop (condensate), distal circuit (swabs), and humidified gas collected by the Cassella sampler (Cassella). See Fig. 1. The chamber was topped off with contaminated fluid at 9 hours. All data in cfu/ml. Abbreviations as in Table 1.
midifiers was investigated with the same ventilator settings. At each flow (inspiratory time setting), particle counts were measured five times in 1-minute increments. Data were analyzed by regression analysis. In the next part, four each of the selected humidifiers were tested as previously with both standard, unheated circuits and with the heated circuits specified by each humidifier's manufacturer. The bubble-through humidifier's manufacturer does not make a heated circuit. Humidifier chambers were filled to maximum operating level with water contaminated as previously. When applicable (Table 1), reservoirs were also filled with contaminated water; the bubble-through humidifiers was refilled manually when necessary. Humidifier heat controls were set to keep distal ,circuit temperature at 33 ° C. Gas temperature at the ventilator outlet port and both room and chamber water temperatures were monitored. The system was then turned on and initial temperarares were noted. Five-milliliter aliquots of chamber water were obtained by pipette at start-up and at 5, 10, 20, 30, 60, and 90 minutes. Water samples were refrigerated at 4 ° C and plated within 24 hours. Serial dilutions were made to achieve between 20 and 200 colony-forming units (cfu) per plate, s Plates were incubated at 35 ° C for 48 hours before evaluation. Data were analyzed by ANOVA. Because swab cultures of the distal circuit in the first part of the experiment were sterile despite contaminated particle production by the bubblethrough humidifier, we evaluated the effect of chamber contamination on circuit colonization when the bubble-through humidifier was heated. The bubble-through humidifier, circuit, and ventilator were assembled in the clean room as in the second part (Fig. 1) and the system was purged of particles, as in the first part. The chamber was
filled with sterile, distilled water and the system was operated until condensate appeared in the circuit. The chamber was then refilled with water contaminated as previously and the system was operated for 24 hours at a distal circuit temperature of 33 ° C. Air samples, distal circuit swabs, and humidifier chamber aliquots were obtained for bacteriologic assessment as in the first part, along with samples of circuit condensate. Sampling was done before inoculation and at 5, 10, 15, 20, 30, 60, and 90, minutes and 24 hours after inoculation. Samples were treated as in the second part. RESULTS Assessment of viable particle production by different humidifiers
The Fisher & Paykel, Marquest, and RCI humidifiers did not produce particles; there was no growth on culture plates from either airborne samples or distal circuit swabs from these humidifiers (Table 2). With bubble-through humidifiers, particles of various size were produced (Fig. 2); and bacteria were isolated from air samples but not from distal circuit swabs (Table 2). Particle production by bubble-through humidifiers increased with increasing flows, suggesting potential for bacterial transfer at higher flows (Fig. 3). Assessment of effects of time and t e m p e r a t u r e on c o n t a m i n a n t viability
The chamber temperatures necessary to maintain distal circuit temperatures of 33 ° C were much lower for heated circuits than for unheated circuits (p < 0.0001 by ANOVA). This did not have a consistent, predictable effect on chamber water colony counts (Fig. 4). Although colony counts per milliliter in bubble-through humidifier
68
AJIC April 1995
Gilmour et al. c
A
D
\ k '~ J ]
I
I
t
j
Ba.~.eria
F,I,er
a
Cascade Humidifier
0
t~
I.E _III jleooo o.o Oli o,!, u Ventilator
Fig, 1. Schematic depiction of equipment utilized in the final part of the experiment. For details of setup, see Methods. A, Immersion thermometer; B, unheated circuitdependent loop allows sampling of condensate; C, YSI telethermometer; D, Cassella slit sampler.
35% 30%
u
o I'o ¢-
.o n
25% 20% 15% 10% 5% 0% :::L
o
^
:::L
o
It-"
"~
@
=L
"7
,-
"~
r,,.
:::L
u~
d
o"
o
o
Particle Size Fig. 2. Range of particle size produced by an unheated bubble-through humidifier. Particles collected by the Climet particle counter are ranged by aerodynamic particle size as shown. Although the total number of particles was flow dependent (Fig. 2), the percentage of particles within each range produced at different flows was similar.
chamber water decreased with time (Fig. 5), this was not true with the continuous-refill humidifiers (Fig. 4).
during the 24-hour study period. Furthermore, although bacterial counts in air samples diminished with time, they were also never free of P. cepacia.
A s s e s s m e n t of c o n d e n s a t e c o n t a m i n a t i o n p o t e n t i a l of b u b b l e . t h r o u g h humidifiers
DISCUSSION
When bubble-through humidifiers were heated, circuit condensate was contaminated within 5 minutes of infection of the chamber contents and remained heavily contaminated throughout the study (Table 3). Although chamber colony counts decreased as chamber temperatures increased, chamber contents were never free of P. cepacia
We acknowledge some limitations to the clinical applicability of our study. For example, bacteria aerosolized by a bubble-through humidifier would n o t necessarily cause infection in a patient receiving mechanical ventilation, 9 nor would humidifier chambers often be replenished with water as heavily contaminated as that used in our study. We also needed to use unheated humidifiers, because
AJIC
Gilmour et al. 69
Volume 23, Number2
the vapor produced during heated humidification interferes with the optics of the Climet particle counter. Nevertheless, we believe that the results of the study of colony growth in heated bubblethrough humidifiers allow us to draw clinically meaningful conclusions from our data on particle production. In the first part of our study, bacteria grew from air samples but not from distal circuit swabs (Table 2). This probably occurred because everything involved in the experiment was at room temperature so that the relatively few, small particles produced were not deposited in the circuit. When the bubble-through humidifier chamber was heated, the circuit was grossly contaminated within 5 minutes of flow initiation (Table 3). Although other investigators have disputed the significance of bubble-through humidifier nebulization, 7' ~0 the absence of humidity and temperature data make their studies less convincing. Our data suggest not only that viable organisms can be transported directly from the bubblethrough humidifier chamber to the patient, especially at high flows (Fig. 3) and high chamber temperatures, but that one should be particularly concerned about the capability of bubble-through humidifiers infect circuit condensate. In contrast, identically contaminated humidifiers that do not produce aerosols should not directly infect patients. It has been suggested that exposure of bacteria in humidifier chambers to high temperatures for prolonged periods almost guarantees the sterility of chamber contents] This conclusion may not be justified. In our study, the chamber temperature of true humidifiers had a limited impact on P. cepacia survival. Despite the significantly higher chamber temperature seen when unheated circuits were used, there was no difference in bacterial survival. Such design features as method of humidification (wick vs passover), recycling of contents between chamber and reservoir, continuous feed from a contaminated reservoir, and chamber volume may contribute to the better survival rate of bacteria in the chambers of true humidifiers. We must emphasize, however, that the presence of bacteria in the chamber is of much less concern when humidifiers do not produce aerosols. Furthermore, although bacterial survival in bubble-through humidifier chambers diminished with time, we were unable to eliminate P. cepacia completely from the bubble-through humidifier chamber during 24 hours of observation. We cannot account for this contrast with the results of Goularte and colleagues, 7 other than to
Cascade I000
I00
i
[]
10
|
I
I
I
Flow flpmi Fig. 3. Effect of flow rate on particle output from a bubblethrough humidifier. A Siemens Servo 900B was used to generate flows of 18, 27, and 45 L/min. Particles emerging from the distal end of the circuit were counted by the Climet counter at each flow rate, and results are displayed as the log of particles per minute _+ SD. Flow dependence of particle production was confirmed by regression analysis: counts = exp (4.08 + 0.028 L/min) R2 = 0.91.
note that we used different test organisms. We must conclude that, contrary to previous assertions,7" 9 sterilization of bubble-through humidifier chamber contents may fail to occur even after prolonged exposure to high temperatures. Of much more concern are the failure of bacterial counts in contaminated condensate to decrease with time and the emission of contaminated aerosol particles from the patient end of the circuit throughout the last part of the study (Table 3). Because of both the progressive decrease in the level of contamination of air samples and our study design, we believe that these particles originated in the humidifying chamber; we cannot, however, dismiss the possibility that they originated from circuit condensate. These findings are important because previous studies have not addressed the potential for an infected bubblethrough humidifier to contaminate circuit condensate, nor have complete data been available about factors affecting particle production and condensate volume, such as distal circuit tempera-
AJIC 70
April 1995
Gilmour et al.
RCI
Fisher & Paykel
lOO~e6~o
50
1000000
80
10oe¢oo -
-70 " 50
leeeeo2
10000
.
O..
-60
P
10000 -
r,.)
1000-
- 50
7
-,o ~,
%"
r7
,oo
.............
e
1 0 0 -~
- 40 - 30
1o
!
20
I 100
Time (minutes)
A
-- ~-....
-~ ....
--
C1~1 Temp H
H
Time (minutes)
. . . . . . . . . . .:. . . . . . . . . . .
ClqJ/ml U
....
Temp U
..A- . . . .
B
-
-
"£; . . . .
(~U/nd H
. . . . . . . . . -.:. . . . . . . . . .
ClqX/mlU
.....
O ....
Te~p H
....
Temp U
-~r~---
Marquest
ipr ....
~
,~...---b ...............
~ ............
~
,? o - ~ . ........................................ ,oooo-ip'-~ '. ~:
looo
-~o - 50
~
- ~
~
. . . . . . . . . . . .
~
.
.
.
.
.
.
.
~oo- ~...~.......... ~......................._.:f_:...........":"...........~ I0
I
I
25
$0
" 70
~
-
o~ 1
40
-.,.o 10
715
100
Time (minutes)
C
....
-
,.c -
-
"<:" . . . . .
CZqJ/ml H
......... e ......
(:~U/ml U
T~;p
----.-,tL-----
Tmap U
H"
Fig. 4. Effect of time and chamber temperature on bacterial growth within the humidifier chamber: A, RCI; B, Fisher & Paykel; C, Marquest. Temperature represents chamber temperatures; time is from onset of chamber contamination. H, Heated circuit; U, unheated circuit. For all three humidifier brands together, as well as for each brand separately, chamber temperature required to maintain a distal circuit temperature of 33 ° + 3 ° C was greater for an unheated than for a heated circuit (p < 0.001, ANOVA for repeated measures). This chamber temperature difference did not affect bacterial survival in comparisons between either humidifier brands or heated and unheated circuits (Mann-Whitney test, p > 0.5).
ture, delivered humidity, tidal volumes and flow rates.7' lo Eighty-five percent of the particles produced by the bubble-through humidifier in the first part were smaller than 1 ~m (Fig. 5); in other studies, which used heated bubble-through humidifiers, the average particle size was greater than 3 i~m.v' 10 Although this effect of temperature on particle size explains the low incidence of direct patient contamination observed previously, it also accounts for the rapidity with which condensate became infected in the final part of our study. It has been asserted that because aerosol particles produced by bubble-through humidifiers are not of the right size ( < 3.5 ~m) to migrate to terminal bronchioles, the infectious potential of bubblethrough humidifiers is insignificant. 3' 7. 10 Recent data, however, suggest that if aerodynamic par-
ticle size is considered, particles m u c h larger than 3.5/xm can penetrate to the alveolae. 11In addition, bacteria need only be introduced into the airways of patients receiving mechanical ventilation to put them at risk, making the ability of contaminated particles to penetrate to the alveolae of little consequence insofar as development of nosocomial pneumonia is concerned. 12 Although nosocomial pneumonia is most often caused by organisms native to the patient,13 it has been observed repeatedly that patients are at risk w h e n contaminated airway equipment is used during care. 3' 14, 15 Because safe, effective, and reasonably priced alternatives to bubble-through humidifiers are available, we believe that even a slightly increased risk for nosocomial infection can no longer be justified. This study has other clinical implications as
AJIC
V o l u m e 23, N u m b e r
GiImour et al.
2
71
Cascade 1
(
~
60
-
10(X)000
@~ '
I ~ ,
..a . . . . . . . . . .
"h
. . . . . . . . . .
"It
tP
50
lOOOOO "/.. I "'o./ rx,..,.
w 40
/
¢
- 30
1~ 1 / I
I
|
25
$0
75
- 20
100
Time (minutes) . . . . . . . . . -~ . . . . . . . . .
¢l~/ml
~--~--
T~mp
Fig. 5. Effects of time and chamber temperature (°C) on bacterial growth within the humidifier chamber for a bubble humidifier, unheated circuit only. There was a significant inverse correlation between time and colony count (Regression analysis R2 = 0.86, p < 0.0001).
well. For example, true humidifiers should not directly cause nosocomial infection even when their chambers and reservoirs are grossly contaminated. Accordingly, our data support the growing practice of using other than sterile, distilled water in these systems. Switching to distilled water from sterile water in true humidifier systems could cut water costs by one third. In addition, if use of bubble-through humidifiers is discontinued, the only mechanism by which ventilator circuits can become contaminated from external sources would be through opening of the ,circuit. Although less frequent circuit changes do :not appear to increase the risk of nosocomial infections,2,7. ~0, 16 the expected decrease has not been observed. Possible explanations for this include the use of bubble-through humidifiers in many of these studies, inadequate humidification, and failure to allow for the effect of confounding variables such as mechanical ventilator flow rates, method of dealing with circuit condensate, method and frequency of suctioning, and circuit type (disposable vs nondisposable). 17 Nevertheless, if true humidifiers are used, increasing the interval between circuit changes appears to be both prudent and cost-effective. 16, ~8, 19 In response to concerns about the infectious hazard posed by contaminated condensate, only the patient portion of the circuit may be changed, leaving t h e true humidifier and associated connections in place. Because bacteria are not eliminated from humidifier chambers during normal use, caregivers must be extremely careful when refilling noncontinu-
ous-feed humidifiers, both to avoid contaminating the humidifier and to prevent transfer of chamber contaminants to other patients. In short, aseptic precautions such as those suggested by the Centers for Disease Control and Prevention are mandatory when handling humidifiers and circuits. Finally, our findings suggest that the conclusions and recommendations of researchers who used bubble-through humidifiers in their studies on humidity 2°22 and nosocomial infection 17 must be reevaluated, with particular attention to the handling of circuit contents, the water content of inspired gas, distal circuit temperature, type of humidifier, and other variables. For example, considerable importance has been ascribed to the role of circuit condensate in the etiology of nosocomial pneumonia, 2, 12,23-25 but no data are available on the confounding effects of such condensate controls as water traps or heated circuits, both of which are less than ideal. 26,27 We believe that additional work needs to be done in the area of condensate control and on the effect of such control on nosocomial infections. i~eferences 1. Cohen IL, Weinberg PF, Fein A, Rowinski GS. Endotracheal tube occlusion associated with the use of heat and moisture exchangers in the intensive care unit. Crit Care Med 1988;16:277-9. 2. CravenDE, Kunches LM, KilinskyV, etal. Risk factors for pneumonia and fatality in patients receiving continuous mechanical ventilation. Am Rev Respir Dis 1986;133: 792-6.
72
AJIC April 1995
G i l m o u r et al.
3. Reinarz JA, Pierce AK, Mays BB, et al. The potential role of inhalation therapy equipment in nosocomial pulmonary infection. J Clin Invest 1965;44:831-9. 4. Tortes A, Aznar R, Gatell JM, et al. Incidence, risk and prognosis factors of nosocomial pneumonia in mechanically ventilated patients. Am Rev Respir Dis 1990;142: 523-8. 5. Rhame FS, Streifel A, McComb RC, et al. Bubbling humidifiers produce micro-aerosols which can carry bacteria. Infect Control 1986;7:403-6. 6. Centers for Disease Control and Prevention. Draft guidelines for prevention of nosocomial pneumonia: notice of comments. Federal Register 1994;59:4980-5022. 7. Goularte TA, Manning M, Craven DE. Bacterial colonization in humidifying Cascade reservoirs after 24 and 48 hours of continuous mechanical ventilation. Infect Control 1987;8:200-3. 8. American Public Health Association. Heterotrophic plate count, sec 907. Standard methods for the evaluation of water and waste water. 17th ed. New York: APHA, 1989:860-70. 9. Williams TA, Elder L. Bacterial migration through infant ventilator circuits. Respir Care 1990;35:1089-90. 10. Goularte TA, Craven DE. Results of a survey of infection control practices for respiratory therapy equipment. Infect Control 1986;7:327-30. 11. 7th ed. American Conference of Governmental Industrial Hygienists, Section J. Air sampling instruments for evaluation of atmospheric contaminents. Cincinnati: ACGHI, 1989:205. 12. Craven DE, Steger KA. Nosocomial pneumonia in the intubated patient. Infect Dis Clin North Am 1989;3:84366. 13. Meduri GU. Ventilator associated pneumonia in patients with respiratory failure. Chest 1990;97:1208-19. 14. Hartstein AI, Rashad AL, Liebler JM, et al. Multiple intensive care unit outbreak of Acinetobacter calcoaceticus subspecies amitratus respiratory infection and colonization associated with contaminated, reusable ventilator circuits and resuscitation bags. Am J Med 1988;85: 624-31. 15. Vandenbroucke-Granls CMJE, Kerver AGH, Rommes JH, et al. Endemic Acinetobacter anitratus in a surgical intensive care unit: mechanical ventilators as a reservoir. Eur J Microbiol Infect Dis 1988;7:485-9.
16. Dreyfuss D, Djedaini K, Weber P, et al, Prospective study of nosocomial pneumonia and of patient and circuit colonization during mechanical ventilation with circuit changes every 48 hours versus no change. Am Rev Respir Dis 1991;143:738-43. 17. Malecka-Griggs B, Kennedy C, Ross B. Microbial burdens in disposable and nondisposable ventilator circuits used for 24 and 48 hours in intensive care units. J Clin Microbiol 1989;27:495-503. 18. Fink J, Mahlmeister M, York M, Cohen N. A comparison of organism growth in ventilator circuits at 48 hours versus seven days [Abstract]. AJIC AM J INFECTCONTROL 1992;20:103. 19. Boher M, Lohse S, Glasby C, et al. Impact of seven day ventilator tubing changes on nosocomial lower respiratory tract infections [Abstract]. AJIC AMJ INFECTCONTROL 1992;20:103. 20. Bengston JP, Sonander H, Stenquist O. Preservation of humidity and heat of respiratory gases during anaesthesia: a laboratory investigation. Acta Anaesthesiol Scand 1987;31:127-31. 21. Kuo CD, Lin SE, Wang JH. Aerosol, humidity and oxygenation. Chest 1991;99:1352-6. 22. Misset B, Escudier B, Rivara D, et al. Heat and moisture exchanger vs heated humidifier during long-term mechanical ventilation. Chest 1991; 100:160-3. 23, Craven DE, Goularte TA, Make BJ. Contaminated condensate in mechanical ventilatory tubing: a risk factor for nosocomial pneumonia? Am Rev Respir Dis 1984;129: 625-8. 24, Craven DE, Steger KA. Pathogenesis and prevention of nosocomial pneumonia in the mechanically ventilated patient. Respir Care 1989;34:85-97. 25. Cardinal P, Jessamine P, Carter-Snell C, et al. Contribution of water condensation in endotracheal tubes to contamination of the lungs. Chest 1993; 104:127-9. 26. Gilmour IJ, Boyle MJ, Rozenberg A, Palahniuk RJ. The effect of heated wire circuits on humidification of inspired gases. Anesth Analg 1994;79:160-4. 27. Miyao H, Hirokawa T, Miyasaka K, et al. Relative humidity not absolute humidity is of great importance when using a humidifier with a heating wire. Crit Care Med 1992;20:674-9.
Availability of JOURNALback issues As a service to our subscribers, copies of back issues of AJIC: AMERICANJOURNALOFINFECTION CONTROLfor the preceding 5 years are maintained and are available for purchase from the publisher, Mosby, at a cost of $9.00 per issue. The following quantity discounts are available: 25% off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby-Year Book, Inc., Subscription Services, 11830 Westline Industrial Dr,, St. Louis, MO 63146-3318, or call (800) 453-4351 or (314) 453-4351 for information on availability of particular issues. If unavailable from the publisher, photocopies of complete issues are available from University Microforms International, 300 N. Zeeb Rd., Ann Arbor, MI 48106 (313) 761-4700,
lan J. Gilmour, MD a Michael J. Boyle, RRT b Andrew Streifel, MPH d R. Carter McComb, RRT c Minneapolis, Minnesota
Background: This study was undertaken because of concerns that ventilator humidifiers could be exacerbating the problem of nosocomial pneumonia in patients receiving mechanical ventilation. Methods: Four different brands of humidifiers were used in conjunction with a Siemens Servo 900B mechanical ventilator (Siemens Life Support Services, Solna, Sweden). In the first part, the ventilator was operated with humidifiers filled with contaminated water at room temperature. The viability of airborne particles and the effect of flow rates on the number of particles produced were assessed. In the second part, we measured the effect of time and temperature on bacterial survival in humidifier chambers. Because only bubble-through humidifiers were determined to produce infectious particles, the speed with which a contaminated bubble-through humidifier could infect circuit condensate was also determined. Aliquots of chamber water and circuit condensate, as well as air samples and distal circuit swabs, were cultured. Results: Humidifiers other than bubble-through humidifiers did not produce aerosols. Particle production by bubble-through humidifiers varied directly with flow rate (R2 = 0.91). Chamber temperatures did not affect chamber colony counts except in bubble-through humidifiers. Although chamber colony counts in bubble-through humidifiers decreased with time, organisms remained viable throughout the study. When bubble-through humidifiers were heated, both condensate and effluent gas became heavily contaminated within minutes of flow initiation. Conclusions: Bubble-through humidifiers produce aerosols that readily contaminate both circuit condensate and effluent gas. Avoiding bubble-through humidifiers should improve patient safety while allowing changes in practice that can result in significant cost :savings. (AJIC AM J INFECTCONTROL1995;23:65-72)
A l t h o u g h w a r m i n g a n d h u m i d i f i c a t i o n o f inspired gas may be accomplished satisfacl:orily w i t h h e a t - m o i s t u r e e x c h a n g e r s f o r s h o r t From the Department of Anesthesiology, University of Minnesota Medical School,a the Department of Cardiopulmonary Servicesb and Hospital Administration,° University of Minnesota Hospital and Clinics, and the Department of Environmental Health and Safety, University of Minnesota School of Public Health,d Minneapolis. Reprint requests: lan J. Gilmour, MD, Department of Anesthesiology, University of Minnesota Medical School, 420 Delaware St., S.E., Box 294, Minneapolis, MN 55455. Copyright © 1995 by the Association for Professionals in Infection Control and Epidemiology, Inc. 0196-6553/95 $3.00 + 0 17/46/60094
p e r i o d s , a c t i v e h u m i d i f i c a t i o n is still r e c o m mended for long-term mechanical ventilation. 1 Unfortunately, the relationship between added moisture and nosocomial pneumonia has long been acknowledged. Although humidif i e r s a r e t h o u g h t to b e a n u n c o m m o n c a u s e o f t h i s p r o b l e m , 2,3 a r e c e n t s t u d y s u g g e s t s t h a t t h e i r u s e is a s s o c i a t e d w i t h a h i g h e r i n c i d e n c e of pneumonia.* Some time ago, Rhame and coworkers s demonstrated that bubble-through humidifiers produce aerosols, which could transmit bacteria through patient circuits. 5 They theorized that true humidifiers would not do this. 65
AJIC 66
April 1995
Gilmour et al.
Table
1. Characteristics of tested humidifiers
Humidifier F&PMR 630 M SCT3000 RCI Concha lit+ PB Cascade l
Type
Chamber volume (ml)
Passover Passover Wick Bubble-through
40 60 50 850
Humidifier chamber MR 390 541020 Low compliance -
Continuous reservoir feed
Servo controlled
Yes Yes Yes No
Yes Yes Yes No
F&P, Fisher & Paykel; M, Marquest; PB, Puritan-Bennett,
T a b l e 2 . Results of a s s e s s m e n t of viable 3article production Humidifier
Particles/ft s
F&P MR 630 M SCT3000 RCI Concha Ill + PB Cascade I
<10" < 1O* < 1O* > 96*
Swab bacterial growth
Air sample bacterial growth (cfu) m
m
> 21-50
For details of procedure, see Methods. For distribution of particle size,see Fig. 5. Therewas no growth from distal circuit swabswith any of the humidifiers, Bacterial growth from Cassella slit sampler cultures occurred only with the bubble-through humidifier, Abbreviations as in Table 1. *Background count is < 10 particles/ft3,
We investigated this theory because bubblethrough humidifiers are still in c o m m o n use in m a n y places and because there is considerable uncertainty about their role in nosocomial infections. 6 We attempted to determine whether bubble-through humidifiers produce microaerosols that transmit bacteria, whereas true humidifiers do not. We also assessed the effects of time and temperature on bacteria in the chambers of these humidifiers, to ascertain whether safety claims based on these effects were credible. 7 METHODS
Fisher & Paykel MR 630 (Fisher & Paykel Allied Products, Auckland, New Zealand), Marquest SeT 3000 (Marquest Medical Products, Inc., Englewood, Colo.), Hudson RCI Conchatherm III Plus (Hudson Respiratory Care, Inc., Temecula, Calif.), and Puritan-Bennett Cascade I (Puritan-Bennett Corp., Overland Park, Kan.) humidifier bases were used in the study. Comparative information on these humidifiers is provided in Table 1. The humidification chambers selected for use with the appropriate bases were as follows: Fisher & Paykel MR 390, Marquest Extended Run (#541020), and RCI Pediatric Conchatherm. The first part of the study was carried out in a vertical flow high efficiency particulate air-filtered clean room (Enviromedic, Inc., Albuquerque, N.M.). The clean room was operated continuously for at least 15 minutes before study periods, after which control air samples were assessed for
particle counts with an airborne particle counter (eL 8060; Climet Inc., Redlands, Calif.) and for bacteria with a high-volume slit sampler (Cassella and Co., Ltd., Britannia Walk, Northern Ireland). The four humidifier systems were assembled according to manufacturers' instructions and 60 inches of 22 m m internal diameter flexible disposable tubing was attached to each outlet port. A Siemens Servo 900B Ventilator (Siemens Life Support Services, Solna, Sweden) set to deliver a tidal volume of 750 ml at a respiratory rate of 12 breaths/min, with inspiratory time of 33% and medical air (inspired fraction of oxygen 0.21), was used to test each unit. A bacterial filter was placed on the ventilator's outlet port and flexible polyvinyl chloride tubing was used to connect the filter with the inlet port of the humidifier. The dry, unheated systems were purged until fewer than 10 particles/ft3 were being emitted; the humidifier chambers were then filled to m a x i m u m operating level with water containing Pseudomonas cepacia at a concentration of 2.0 x l0 s cells/ml. After inoculation, each system was operated at room temperature for 90 minutes, after which a 2-minute air sample was obtained with the slit sampler and the distal end of the 60-inch tubing was swabbed. The swab samples were streaked on standard methods agar and incubated, along with the air sample cultures, at 35 ° C for 48 hours. Because only the bubble-through humidifiers produced particles, the relationship between flow rate and particle generation by bubble-through hu-
AJIC Volume 23, Number 2
Gilmour et al.
67
Table 3. Bacterial survival in the chamber and the circuit during use of a bubble-through humidifier Time (min)
Chamber
Swab
Cassella
0 5 10 20 30 60 90 120 24 hr
0 1,2 x 105 78 112 31 29 5 10 20
0 21-50 21-50 6-20 1-5 1-5 1-5 1-5 >50
0 140 2 3 2 7 2 3 1
Condensate
>3.0 >3.0 >3,0 >3.0 > 3.0 >3,0 >3,0 >3.0
0 x x x x x x x x
103 103 103 10 a 103 10 s 10 s 103
Samples for assessing bacterial growth were taken from bubble-through humidifier (chamber), dependent toop (condensate), distal circuit (swabs), and humidified gas collected by the Cassella sampler (Cassella). See Fig. 1. The chamber was topped off with contaminated fluid at 9 hours. All data in cfu/ml. Abbreviations as in Table 1.
midifiers was investigated with the same ventilator settings. At each flow (inspiratory time setting), particle counts were measured five times in 1-minute increments. Data were analyzed by regression analysis. In the next part, four each of the selected humidifiers were tested as previously with both standard, unheated circuits and with the heated circuits specified by each humidifier's manufacturer. The bubble-through humidifier's manufacturer does not make a heated circuit. Humidifier chambers were filled to maximum operating level with water contaminated as previously. When applicable (Table 1), reservoirs were also filled with contaminated water; the bubble-through humidifiers was refilled manually when necessary. Humidifier heat controls were set to keep distal ,circuit temperature at 33 ° C. Gas temperature at the ventilator outlet port and both room and chamber water temperatures were monitored. The system was then turned on and initial temperarares were noted. Five-milliliter aliquots of chamber water were obtained by pipette at start-up and at 5, 10, 20, 30, 60, and 90 minutes. Water samples were refrigerated at 4 ° C and plated within 24 hours. Serial dilutions were made to achieve between 20 and 200 colony-forming units (cfu) per plate, s Plates were incubated at 35 ° C for 48 hours before evaluation. Data were analyzed by ANOVA. Because swab cultures of the distal circuit in the first part of the experiment were sterile despite contaminated particle production by the bubblethrough humidifier, we evaluated the effect of chamber contamination on circuit colonization when the bubble-through humidifier was heated. The bubble-through humidifier, circuit, and ventilator were assembled in the clean room as in the second part (Fig. 1) and the system was purged of particles, as in the first part. The chamber was
filled with sterile, distilled water and the system was operated until condensate appeared in the circuit. The chamber was then refilled with water contaminated as previously and the system was operated for 24 hours at a distal circuit temperature of 33 ° C. Air samples, distal circuit swabs, and humidifier chamber aliquots were obtained for bacteriologic assessment as in the first part, along with samples of circuit condensate. Sampling was done before inoculation and at 5, 10, 15, 20, 30, 60, and 90, minutes and 24 hours after inoculation. Samples were treated as in the second part. RESULTS Assessment of viable particle production by different humidifiers
The Fisher & Paykel, Marquest, and RCI humidifiers did not produce particles; there was no growth on culture plates from either airborne samples or distal circuit swabs from these humidifiers (Table 2). With bubble-through humidifiers, particles of various size were produced (Fig. 2); and bacteria were isolated from air samples but not from distal circuit swabs (Table 2). Particle production by bubble-through humidifiers increased with increasing flows, suggesting potential for bacterial transfer at higher flows (Fig. 3). Assessment of effects of time and t e m p e r a t u r e on c o n t a m i n a n t viability
The chamber temperatures necessary to maintain distal circuit temperatures of 33 ° C were much lower for heated circuits than for unheated circuits (p < 0.0001 by ANOVA). This did not have a consistent, predictable effect on chamber water colony counts (Fig. 4). Although colony counts per milliliter in bubble-through humidifier
68
AJIC April 1995
Gilmour et al. c
A
D
\ k '~ J ]
I
I
t
j
Ba.~.eria
F,I,er
a
Cascade Humidifier
0
t~
I.E _III jleooo o.o Oli o,!, u Ventilator
Fig, 1. Schematic depiction of equipment utilized in the final part of the experiment. For details of setup, see Methods. A, Immersion thermometer; B, unheated circuitdependent loop allows sampling of condensate; C, YSI telethermometer; D, Cassella slit sampler.
35% 30%
u
o I'o ¢-
.o n
25% 20% 15% 10% 5% 0% :::L
o
^
:::L
o
It-"
"~
@
=L
"7
,-
"~
r,,.
:::L
u~
d
o"
o
o
Particle Size Fig. 2. Range of particle size produced by an unheated bubble-through humidifier. Particles collected by the Climet particle counter are ranged by aerodynamic particle size as shown. Although the total number of particles was flow dependent (Fig. 2), the percentage of particles within each range produced at different flows was similar.
chamber water decreased with time (Fig. 5), this was not true with the continuous-refill humidifiers (Fig. 4).
during the 24-hour study period. Furthermore, although bacterial counts in air samples diminished with time, they were also never free of P. cepacia.
A s s e s s m e n t of c o n d e n s a t e c o n t a m i n a t i o n p o t e n t i a l of b u b b l e . t h r o u g h humidifiers
DISCUSSION
When bubble-through humidifiers were heated, circuit condensate was contaminated within 5 minutes of infection of the chamber contents and remained heavily contaminated throughout the study (Table 3). Although chamber colony counts decreased as chamber temperatures increased, chamber contents were never free of P. cepacia
We acknowledge some limitations to the clinical applicability of our study. For example, bacteria aerosolized by a bubble-through humidifier would n o t necessarily cause infection in a patient receiving mechanical ventilation, 9 nor would humidifier chambers often be replenished with water as heavily contaminated as that used in our study. We also needed to use unheated humidifiers, because
AJIC
Gilmour et al. 69
Volume 23, Number2
the vapor produced during heated humidification interferes with the optics of the Climet particle counter. Nevertheless, we believe that the results of the study of colony growth in heated bubblethrough humidifiers allow us to draw clinically meaningful conclusions from our data on particle production. In the first part of our study, bacteria grew from air samples but not from distal circuit swabs (Table 2). This probably occurred because everything involved in the experiment was at room temperature so that the relatively few, small particles produced were not deposited in the circuit. When the bubble-through humidifier chamber was heated, the circuit was grossly contaminated within 5 minutes of flow initiation (Table 3). Although other investigators have disputed the significance of bubble-through humidifier nebulization, 7' ~0 the absence of humidity and temperature data make their studies less convincing. Our data suggest not only that viable organisms can be transported directly from the bubblethrough humidifier chamber to the patient, especially at high flows (Fig. 3) and high chamber temperatures, but that one should be particularly concerned about the capability of bubble-through humidifiers infect circuit condensate. In contrast, identically contaminated humidifiers that do not produce aerosols should not directly infect patients. It has been suggested that exposure of bacteria in humidifier chambers to high temperatures for prolonged periods almost guarantees the sterility of chamber contents] This conclusion may not be justified. In our study, the chamber temperature of true humidifiers had a limited impact on P. cepacia survival. Despite the significantly higher chamber temperature seen when unheated circuits were used, there was no difference in bacterial survival. Such design features as method of humidification (wick vs passover), recycling of contents between chamber and reservoir, continuous feed from a contaminated reservoir, and chamber volume may contribute to the better survival rate of bacteria in the chambers of true humidifiers. We must emphasize, however, that the presence of bacteria in the chamber is of much less concern when humidifiers do not produce aerosols. Furthermore, although bacterial survival in bubble-through humidifier chambers diminished with time, we were unable to eliminate P. cepacia completely from the bubble-through humidifier chamber during 24 hours of observation. We cannot account for this contrast with the results of Goularte and colleagues, 7 other than to
Cascade I000
I00
i
[]
10
|
I
I
I
Flow flpmi Fig. 3. Effect of flow rate on particle output from a bubblethrough humidifier. A Siemens Servo 900B was used to generate flows of 18, 27, and 45 L/min. Particles emerging from the distal end of the circuit were counted by the Climet counter at each flow rate, and results are displayed as the log of particles per minute _+ SD. Flow dependence of particle production was confirmed by regression analysis: counts = exp (4.08 + 0.028 L/min) R2 = 0.91.
note that we used different test organisms. We must conclude that, contrary to previous assertions,7" 9 sterilization of bubble-through humidifier chamber contents may fail to occur even after prolonged exposure to high temperatures. Of much more concern are the failure of bacterial counts in contaminated condensate to decrease with time and the emission of contaminated aerosol particles from the patient end of the circuit throughout the last part of the study (Table 3). Because of both the progressive decrease in the level of contamination of air samples and our study design, we believe that these particles originated in the humidifying chamber; we cannot, however, dismiss the possibility that they originated from circuit condensate. These findings are important because previous studies have not addressed the potential for an infected bubblethrough humidifier to contaminate circuit condensate, nor have complete data been available about factors affecting particle production and condensate volume, such as distal circuit tempera-
AJIC 70
April 1995
Gilmour et al.
RCI
Fisher & Paykel
lOO~e6~o
50
1000000
80
10oe¢oo -
-70 " 50
leeeeo2
10000
.
O..
-60
P
10000 -
r,.)
1000-
- 50
7
-,o ~,
%"
r7
,oo
.............
e
1 0 0 -~
- 40 - 30
1o
!
20
I 100
Time (minutes)
A
-- ~-....
-~ ....
--
C1~1 Temp H
H
Time (minutes)
. . . . . . . . . . .:. . . . . . . . . . .
ClqJ/ml U
....
Temp U
..A- . . . .
B
-
-
"£; . . . .
(~U/nd H
. . . . . . . . . -.:. . . . . . . . . .
ClqX/mlU
.....
O ....
Te~p H
....
Temp U
-~r~---
Marquest
ipr ....
~
,~...---b ...............
~ ............
~
,? o - ~ . ........................................ ,oooo-ip'-~ '. ~:
looo
-~o - 50
~
- ~
~
. . . . . . . . . . . .
~
.
.
.
.
.
.
.
~oo- ~...~.......... ~......................._.:f_:...........":"...........~ I0
I
I
25
$0
" 70
~
-
o~ 1
40
-.,.o 10
715
100
Time (minutes)
C
....
-
,.c -
-
"<:" . . . . .
CZqJ/ml H
......... e ......
(:~U/ml U
T~;p
----.-,tL-----
Tmap U
H"
Fig. 4. Effect of time and chamber temperature on bacterial growth within the humidifier chamber: A, RCI; B, Fisher & Paykel; C, Marquest. Temperature represents chamber temperatures; time is from onset of chamber contamination. H, Heated circuit; U, unheated circuit. For all three humidifier brands together, as well as for each brand separately, chamber temperature required to maintain a distal circuit temperature of 33 ° + 3 ° C was greater for an unheated than for a heated circuit (p < 0.001, ANOVA for repeated measures). This chamber temperature difference did not affect bacterial survival in comparisons between either humidifier brands or heated and unheated circuits (Mann-Whitney test, p > 0.5).
ture, delivered humidity, tidal volumes and flow rates.7' lo Eighty-five percent of the particles produced by the bubble-through humidifier in the first part were smaller than 1 ~m (Fig. 5); in other studies, which used heated bubble-through humidifiers, the average particle size was greater than 3 i~m.v' 10 Although this effect of temperature on particle size explains the low incidence of direct patient contamination observed previously, it also accounts for the rapidity with which condensate became infected in the final part of our study. It has been asserted that because aerosol particles produced by bubble-through humidifiers are not of the right size ( < 3.5 ~m) to migrate to terminal bronchioles, the infectious potential of bubblethrough humidifiers is insignificant. 3' 7. 10 Recent data, however, suggest that if aerodynamic par-
ticle size is considered, particles m u c h larger than 3.5/xm can penetrate to the alveolae. 11In addition, bacteria need only be introduced into the airways of patients receiving mechanical ventilation to put them at risk, making the ability of contaminated particles to penetrate to the alveolae of little consequence insofar as development of nosocomial pneumonia is concerned. 12 Although nosocomial pneumonia is most often caused by organisms native to the patient,13 it has been observed repeatedly that patients are at risk w h e n contaminated airway equipment is used during care. 3' 14, 15 Because safe, effective, and reasonably priced alternatives to bubble-through humidifiers are available, we believe that even a slightly increased risk for nosocomial infection can no longer be justified. This study has other clinical implications as
AJIC
V o l u m e 23, N u m b e r
GiImour et al.
2
71
Cascade 1
(
~
60
-
10(X)000
@~ '
I ~ ,
..a . . . . . . . . . .
"h
. . . . . . . . . .
"It
tP
50
lOOOOO "/.. I "'o./ rx,..,.
w 40
/
¢
- 30
1~ 1 / I
I
|
25
$0
75
- 20
100
Time (minutes) . . . . . . . . . -~ . . . . . . . . .
¢l~/ml
~--~--
T~mp
Fig. 5. Effects of time and chamber temperature (°C) on bacterial growth within the humidifier chamber for a bubble humidifier, unheated circuit only. There was a significant inverse correlation between time and colony count (Regression analysis R2 = 0.86, p < 0.0001).
well. For example, true humidifiers should not directly cause nosocomial infection even when their chambers and reservoirs are grossly contaminated. Accordingly, our data support the growing practice of using other than sterile, distilled water in these systems. Switching to distilled water from sterile water in true humidifier systems could cut water costs by one third. In addition, if use of bubble-through humidifiers is discontinued, the only mechanism by which ventilator circuits can become contaminated from external sources would be through opening of the ,circuit. Although less frequent circuit changes do :not appear to increase the risk of nosocomial infections,2,7. ~0, 16 the expected decrease has not been observed. Possible explanations for this include the use of bubble-through humidifiers in many of these studies, inadequate humidification, and failure to allow for the effect of confounding variables such as mechanical ventilator flow rates, method of dealing with circuit condensate, method and frequency of suctioning, and circuit type (disposable vs nondisposable). 17 Nevertheless, if true humidifiers are used, increasing the interval between circuit changes appears to be both prudent and cost-effective. 16, ~8, 19 In response to concerns about the infectious hazard posed by contaminated condensate, only the patient portion of the circuit may be changed, leaving t h e true humidifier and associated connections in place. Because bacteria are not eliminated from humidifier chambers during normal use, caregivers must be extremely careful when refilling noncontinu-
ous-feed humidifiers, both to avoid contaminating the humidifier and to prevent transfer of chamber contaminants to other patients. In short, aseptic precautions such as those suggested by the Centers for Disease Control and Prevention are mandatory when handling humidifiers and circuits. Finally, our findings suggest that the conclusions and recommendations of researchers who used bubble-through humidifiers in their studies on humidity 2°22 and nosocomial infection 17 must be reevaluated, with particular attention to the handling of circuit contents, the water content of inspired gas, distal circuit temperature, type of humidifier, and other variables. For example, considerable importance has been ascribed to the role of circuit condensate in the etiology of nosocomial pneumonia, 2, 12,23-25 but no data are available on the confounding effects of such condensate controls as water traps or heated circuits, both of which are less than ideal. 26,27 We believe that additional work needs to be done in the area of condensate control and on the effect of such control on nosocomial infections. i~eferences 1. Cohen IL, Weinberg PF, Fein A, Rowinski GS. Endotracheal tube occlusion associated with the use of heat and moisture exchangers in the intensive care unit. Crit Care Med 1988;16:277-9. 2. CravenDE, Kunches LM, KilinskyV, etal. Risk factors for pneumonia and fatality in patients receiving continuous mechanical ventilation. Am Rev Respir Dis 1986;133: 792-6.
72
AJIC April 1995
G i l m o u r et al.
3. Reinarz JA, Pierce AK, Mays BB, et al. The potential role of inhalation therapy equipment in nosocomial pulmonary infection. J Clin Invest 1965;44:831-9. 4. Tortes A, Aznar R, Gatell JM, et al. Incidence, risk and prognosis factors of nosocomial pneumonia in mechanically ventilated patients. Am Rev Respir Dis 1990;142: 523-8. 5. Rhame FS, Streifel A, McComb RC, et al. Bubbling humidifiers produce micro-aerosols which can carry bacteria. Infect Control 1986;7:403-6. 6. Centers for Disease Control and Prevention. Draft guidelines for prevention of nosocomial pneumonia: notice of comments. Federal Register 1994;59:4980-5022. 7. Goularte TA, Manning M, Craven DE. Bacterial colonization in humidifying Cascade reservoirs after 24 and 48 hours of continuous mechanical ventilation. Infect Control 1987;8:200-3. 8. American Public Health Association. Heterotrophic plate count, sec 907. Standard methods for the evaluation of water and waste water. 17th ed. New York: APHA, 1989:860-70. 9. Williams TA, Elder L. Bacterial migration through infant ventilator circuits. Respir Care 1990;35:1089-90. 10. Goularte TA, Craven DE. Results of a survey of infection control practices for respiratory therapy equipment. Infect Control 1986;7:327-30. 11. 7th ed. American Conference of Governmental Industrial Hygienists, Section J. Air sampling instruments for evaluation of atmospheric contaminents. Cincinnati: ACGHI, 1989:205. 12. Craven DE, Steger KA. Nosocomial pneumonia in the intubated patient. Infect Dis Clin North Am 1989;3:84366. 13. Meduri GU. Ventilator associated pneumonia in patients with respiratory failure. Chest 1990;97:1208-19. 14. Hartstein AI, Rashad AL, Liebler JM, et al. Multiple intensive care unit outbreak of Acinetobacter calcoaceticus subspecies amitratus respiratory infection and colonization associated with contaminated, reusable ventilator circuits and resuscitation bags. Am J Med 1988;85: 624-31. 15. Vandenbroucke-Granls CMJE, Kerver AGH, Rommes JH, et al. Endemic Acinetobacter anitratus in a surgical intensive care unit: mechanical ventilators as a reservoir. Eur J Microbiol Infect Dis 1988;7:485-9.
16. Dreyfuss D, Djedaini K, Weber P, et al, Prospective study of nosocomial pneumonia and of patient and circuit colonization during mechanical ventilation with circuit changes every 48 hours versus no change. Am Rev Respir Dis 1991;143:738-43. 17. Malecka-Griggs B, Kennedy C, Ross B. Microbial burdens in disposable and nondisposable ventilator circuits used for 24 and 48 hours in intensive care units. J Clin Microbiol 1989;27:495-503. 18. Fink J, Mahlmeister M, York M, Cohen N. A comparison of organism growth in ventilator circuits at 48 hours versus seven days [Abstract]. AJIC AM J INFECTCONTROL 1992;20:103. 19. Boher M, Lohse S, Glasby C, et al. Impact of seven day ventilator tubing changes on nosocomial lower respiratory tract infections [Abstract]. AJIC AMJ INFECTCONTROL 1992;20:103. 20. Bengston JP, Sonander H, Stenquist O. Preservation of humidity and heat of respiratory gases during anaesthesia: a laboratory investigation. Acta Anaesthesiol Scand 1987;31:127-31. 21. Kuo CD, Lin SE, Wang JH. Aerosol, humidity and oxygenation. Chest 1991;99:1352-6. 22. Misset B, Escudier B, Rivara D, et al. Heat and moisture exchanger vs heated humidifier during long-term mechanical ventilation. Chest 1991; 100:160-3. 23, Craven DE, Goularte TA, Make BJ. Contaminated condensate in mechanical ventilatory tubing: a risk factor for nosocomial pneumonia? Am Rev Respir Dis 1984;129: 625-8. 24, Craven DE, Steger KA. Pathogenesis and prevention of nosocomial pneumonia in the mechanically ventilated patient. Respir Care 1989;34:85-97. 25. Cardinal P, Jessamine P, Carter-Snell C, et al. Contribution of water condensation in endotracheal tubes to contamination of the lungs. Chest 1993; 104:127-9. 26. Gilmour IJ, Boyle MJ, Rozenberg A, Palahniuk RJ. The effect of heated wire circuits on humidification of inspired gases. Anesth Analg 1994;79:160-4. 27. Miyao H, Hirokawa T, Miyasaka K, et al. Relative humidity not absolute humidity is of great importance when using a humidifier with a heating wire. Crit Care Med 1992;20:674-9.
Availability of JOURNALback issues As a service to our subscribers, copies of back issues of AJIC: AMERICANJOURNALOFINFECTION CONTROLfor the preceding 5 years are maintained and are available for purchase from the publisher, Mosby, at a cost of $9.00 per issue. The following quantity discounts are available: 25% off on quantities of 12 to 23, and one third off on quantities of 24 or more. Please write to Mosby-Year Book, Inc., Subscription Services, 11830 Westline Industrial Dr,, St. Louis, MO 63146-3318, or call (800) 453-4351 or (314) 453-4351 for information on availability of particular issues. If unavailable from the publisher, photocopies of complete issues are available from University Microforms International, 300 N. Zeeb Rd., Ann Arbor, MI 48106 (313) 761-4700,