Influence of disposable (‘Conchapak’) and reusable humidifying systems on the incidence of ventilation pneumonia

Journal

of Hospital

Influence humidifying

F. Daschner*,

Infection

(1988)

11, 161-168

of disposable (‘Conchapak’) systems on the incidence pneumonia

and reusable of ventilation

I. Kappstein*, F. Schuster*, R. Scholz*, JooPenst and H. Just?

E. Bauery,

D.

of Hospital Epidemiology and TDepartment of Internal *Department Medicine, University Hospital of Freiburg, FRG Accepted for publication

9 June 198 7

Summary: The contamination of disposable (‘Conchapak’) and reusable humidifying systems and their influence on the incidence of pneumonia was studied in 116 patients requiring continuous mechanical ventilation therapy. The water reservoirs of 11 (15.9%) of the 69 disposable systems became colonized, but all reusable systems were found to be sterile. In four of the 11 samples, the organisms isolated corresponded with those cultured from tracheal secretions several days before. Ventilator-associated pneumonia occurred in 36 (31.0%) of the patients, but there was no statistically significant difference in the incidence of pneumonia between the patients treated with the disposable or the reusable humidifying systems. Gramnegative bacteria were the predominant organisms isolated from tracheal aspirates of patients who developed ventilator-associated pneumonia. These results suggest that disposable humidifying systems do not influence the rate of ventilator-associated pneumonia in mechanically ventilated patients. Key words: Disposable ciated pneumonia

humidifiers;

reusable

humidifiers;

ventilator

asso-

Introduction Pneumonia accounts for about 15 % of nosocomial infections and was found to be the most frequent hospital-acquired infection related to death (Gross et al., 1980; Garibaldi et al., 1981; Dixon, 1983; Bryan & Reynolds, 1984). Critically ill patients requiring mechanical ventilation therapy have an increased risk of developing nosocomial infections of the lower respiratory tract (Cross & Roup, 1981; Craig & Connelly, 1984). Bacteria may enter the lung via air inhaled through the lumen of the endotracheal tube; their source can be from the air of the unit, from equipment or the hands of personnel. Another important pathway is aspiration of highly contaminated oropharyngeal or gastric secretions (McCrae & Wallace, 1981; Sanderson, Correspondence to: Prof F. Daschner, Department of Hospital Freiburg, Hugstetter Strasse 55, 7800 Freiburg (FRG). 0195-6701/88/020161+08

Epidemiology,

0

803.00/O

161

University

1988 The Hospital

Hospital

Infection

of

Society

162

F. Daschner

et al.

1983; Daschner, 1985). Previously, the importance of the contaminated reservoirs of nebulizers has been emphasized (Rinarz et al., 1965; Pierce & Sanford, 1973; Cross & Roup, 1981; Craven et al., 1984; Craven, Goularte & Make, 1984; Craven et al., 1986). At present, most mechanical ventilators have cascade humidifiers that do not generate aerosols. However, condensates are formed as the warm humidified air flows through the cooler ventilator tubing and these may be a risk for ventilator-associated pneumonia if the ventilator circuit has been contaminated by extrinsic factors or by retrograde bacterial spread from the patient’s oropharyngeal secretions (Craven et al., 1984). Since the risk of contamination may be decreased by the use of disposable humidifiers, we prospectively compared the influence of the disposable ‘Conchapak’ system and three reusable systems on the development of ventilator-associated pneumonia. Patients

and

methods

Study population One hundred and sixteen patients in the medical and surgical intensive care units of the University Hospital of Freiburg, FRG, receiving mechanical ventilation were studied. Patients with short-term (below 24 h, 32 patients) and long-term ventilation therapy (above 24 h, 84 patients) were included. Each patient was ventilated via an endotracheal or treacheostomy tube connected to one of the following mechanical ventilators: Bennett MA-l B, Bennett 7200, Servo-Vent, Drager UV 1, Bird Mark 14, Engstrijm Erica and Jetronic High Frequency. The airways of 69 patients were humidified with the disposable ‘Conchapak’ system (Kendall, FRG) which consists of a heater, a humidifying cylinder and a 1500 ml reservoir for sterile water (Figure 1). In 47 patients, one of the following reusable humidifying systems were used and were changed daily: Bennett-Cascade Humidifyer I, Drager Humidifyer 19 and Bird 500 m3 Inline Long-Term Nebulizer. Data collection laboratory results, underlying diseases and Daily patient temperatures, drug therapy were recorded prospectively on standardized forms by infection control nurses. The diagnosis of ventilator-associated pneumonia required four of the following five criteria developing not less than 48 h after the start of mechanical ventilation: (1) a new and persistent infiltrate in a chest X-ray, (2) purulent tracheal secretion, (3) a significant pathogen isolated from cultures of tracheal aspirate, (4) body temperature above 38°C and (5) a peripheral leukocyte count of more than 10,000 mm3 -’ and physical signs of pneumonia. Of these criteria, a positive chest X-ray must always be present.

Disposable Resplrotory

v reusable

163

humidifiers

gas Water

reservoir

Sterile water Site of sampling

Figure

1. Conchapak

disposable

humidifiers.

Microbiological sampling 10 (1) Reusable systems Reservoir samples were obtained by removing to 20 ml of water with a sterile pipette each day prior to changing. Each sample was deposited in a sterile screw-capped tube and promptly cultured for bacteria and fungi. Samples were filtered by vacuum filtration (Millipore filters, pore size 0.22 cl) and the filters cultured on nutrient agar containing disinfectant neutralizers (0.5% sodium thiosulfate, 0.1% histidine, 0.5% ‘Tween’ 80, 0.07% lecithin). All plates were incubated at 37°C for 48 h and then for 4 additional days at room temperature for slow-growing organisms. (2) Disposable systems Reservoir samples were obtained and cultured as described above. Humidifying cylinders were changed after 3 days, at the latest, and sampled by rinsing with 20 ml of tryptone soya broth containing disinfectant neutralizers. A 10 ml sample of the rinsing fluid was cultured as above; a second 10 ml was incubated at 37°C and after 24 h plated on blood and nutrient agar and subsequently incubated at 37°C for 48 h. Cultures of tracheal secretion were (3) Trachea-bronchial secretions collected by suction directly from the lower respiratory tract via the tracheostomy or endotracheal tube. Organisms were identified according to routine microbiological methods. No cultures were made for anaerobic bacteria, legionella, mycoplasma or viruses.

164

F. Daschner

et al.

Culturally identical organisms isolated from tracheal secretions and water reservoirs were typed by phages (Staphylococcus aweus) or by agglutination (Pseudomonas aeruginosa) . Results

Admission characteristics in the 116 study All patients required mechanical ventilation tracheostomy tube.

patients therapy

are shown in Table I. via an endotracheal or

Humidifying reservoir colonization Reservoir samples from 69 disposable systems and 47 reusable systems were investigated for bacterial contamination. The reusable systems showed no growth of bacteria, whereas 11. (1 S-9’/,) of the 69 disposable water reservoirs became colonized (Table II). Ten of these 11 patients acquired ventilator-associated pneumonia. In four patients, the bacteria isolated from the reservoir samples (one S. aureus, three P. aeruginosa) were found to be the same organisms by typing as cultured from tracheal secretions several days before. Therefore, the patient’s tracheal secretions were considered to be the source of reservoir colonization. In six patients with ventilator-associated pneumonia, the bacteria isolated from the reservoir (one P. aeruginosa, four spore-forming organisms, one samples micrococcus) did not correspond with those recovered from tracheal Table I. Characteristics of the 116 study patients humidified either with disposable (‘Conchapak’)

whose airways were or reusable systems

Number Disposable systems (n= 69) Male sex Severe head injury Thoracic injury Upper abdominal surgery Lower abdominal surgery Tracheostomy Prior aspiration Prior antacids Alcoholism History of smoking Prior hospital stay 1 day 2-7 days 7 days Lung oedema Underlying diseases (chronic lung disease, diabetes mellitus, neoplasia/cytostatic therapy, heart insufficiency)

52

of patients

with

Reusable systems (n = 47)

:f 16 6 5 1.5 4

32 14 8 12 6 10 10 1

1:

1:

8 9 6 18 26

3 7 5

Disposable Table

II. Microbial

Humidifying systems used Disposable Reusable

v reusable

humidifiers

165

contamination of disposable (‘Conchapak’) reusable humidifying systems Number of humidifiers sampled

Number of contaminated systems (%)

69 47

11 (15.9) 0

and

and were cultured between 7 and 39 days after development of In the one patient without pneumonia, ventilator-associated pneumonia. the water reservoir was colonized with S. epidermidis. No contamination was found in the humidifying cylinders of the disposable system.

aspirates

Ventilator-associated pneumonia Pneumonia occurred in 36 (3 1 .O%) of the 116 study patients. The rate of pneumonia in the 69 patients whose airways were humidified with the was 26.1% (18 patients) and in the 47 disposable system ‘Conchapak’ patients whose airways were humidified with reusable systems was 38.3% (18 patients) (Table III). The difference between the groups was not statistically significant (PaO.OS), but there was a trend toward a lower rate of ventilator-associated pneumonia in the disposable system group. In the 84 patients with long-term ventilation therapy, the rate of ventilator-associated pneumonia was 42.8%. Gram-negative bacteria were the predominant organisms, being isolated from 59.4% of tracheal aspirates of patients who developed ventilator-associated pneumonia. The most common species isolated were S. aweus (20.3%), P. aeruginosa (14.1%) and Proteus mirabilis (9.4%) (Table IV). Pneumonia developed in 87.5% of the 36 patients who acquired ventilator-associated pneumonia within 7-8 days of ventilation therapy. On average, the duration of ventilation therapy until the onset of

Table III. Frequency of ventilator-associated pneumonia in mechanically ventilated patients treated with either the disposable (‘Conchapak’) or the reusable humidifying systems Number patients

of

69 47 Total 116

Humidifying systems used Disposable Reusable

Frequency of ventilation pneumonia (%) 18 (26.1) 18 (38.3)

number 36 (31.0)

166

F. Daschner Table

IV.

Organisms

et al.

isolated from tracheal secretions ventilation pneumonia

of patients

with

Number of isolates (%) (n = 64)

Organism Staphylococcus aureus Pseudomonas aeruginosa Proteus mirabilis Haemophilus influenzae Escherichia coli Klebsiella pneumoniae Candida spp. Other fungi Enterococci Enterobacter cloacae E. aerogenes Streptococcus pneumoniae Citrobacter diversus C. freundii Group C streptococci Neisseria meningitidis Coagulase-negative staphylococci Acinetobacter calcoaceticus, var. anitratus

13 (20.3) 9 (14.1) 6 (9.4) 5 (7.8) 4 (6.3) 4 (6.3) 4 (6.3) i 2

g::j (3.1)

; ::::i 1 (1.6) 1 (1.6) 1 (1.6) 1 (1.6) 1 (1.6) 1 (1.6)

ventilator-associated pneumonia was 6.3 f 0.5 days in the disposable group and 5.9 f O-5 days in the reusable group. The mean age of the patients with ventilator-associated pneumonia was 48.2 f 2.9 yrs and that of the patients who did not acquire pneumonia was 41.8 f 2.1 yrs (P
Ventilators with cascade humidifiers do not generate contaminated aerosols, unlike those with mainstream nebulizers (Reinarz et al., 1965), but the presence of contaminated tubing condensates may be a risk factor for ventilator-associated pneumonia (Craven et al., 1984). Such highly contaminated condensates may be a risk factor for ventilator-associated pneumonia in that they may enter the patient’s respiratory system, for example after turning the patient or other simple nursing techniques, possibly not considered to be dangerous. In our study, we compared different humidifying systems, one disposable and three reusable, with regard to contamination of the water reservoirs and the frequency of ventilator-associated pneumonia. We did not find a significant difference in the frequency of ventilator-associated pneumonia between the two groups, but we observed a marked collection of tubing condensate in the disposable system requiring frequent tubing

Disposable

v reusable

humidifiers

167

disconnections, which increases the risk of exogenous contamination and cross-infection by the hands of nursing staff. In addition, in four patients who developed ventilator-associated pneumonia, we found the water reservoir of the disposable system to be colonized with the same organisms as cultured from tracheal secretions several days before. In the three reusable humidifying systems tested, collection of tubing condensates was less frequent and could possibly explain the absence of contamination of water reservoirs in these systems. The overall rate of ventilator-associated pneumonia in our study population was 31 .O%. In the group of 84 patients requiring long-term ventilation therapy 36 (42.8%) acquired pneumonia. Other investigators have found pneumonia rates ranging from 21% to 66% (Cross & Roup, 1981; Craven et al., 1986). Comparing patients whose airways were humidified with the disposable system or with the reusable systems we found only a trend toward a lower pneumonia rate in the disposable group. However, tracheostomy was significantly more frequent in the reusable group. In the study of Cross & Roup (1981), ventilator-associated pneumonia occurred in 66% of tracheostomized patients. Intubation and tracheostomy impair the natural host defences and enable pharyngeal and gastric secretions highly contaminated with Gram-negative bacteria to enter the lower respiratory system (Sanderson, 1983). About 60% of the bacteria isolated from tracheal secretions of patients with ventilator-associated pneumonia were Gram-negative species, a finding which correlates with the results of other authors (LaForce, 1981; Craven et al., 1986). The present study indicates that use of a disposable humidifying system has no influence on the frequency of ventilator-associated pneumonia in mechanically ventilated patients. We would

like to thank

Prof

G. A. J. Ayliffe

for reviewing

our manuscript.

References Bryan, Ch. S. & Reynolds, K. L. (1984). Bacteremic nosocomial pneumonia. American Reviews of Respiratory Diseases 129, 668-671. Craig, Ch. P. & Connelly, S. (1984). Effect of intensive care unit nosocomial pneumonia on duration of stay and mortality. American Journal of Infection Control 12, 233-238. Craven, D. E., Lichtenberg, D. A., Goularte, Th. A., Make, M. J. & McCabe, W. R. (1984). Contaminated medication nebulizers in mechanical ventilator circuits. The American rournal of Medicine 77, 834838. Craven, D. E., Goularte, Th. A. & Make, B. J. (1984). Contaminated condensate in mechanical ventilator circuits. American Reviews of ResPivatory Diseases 129. 625-628. Craven, D. E., Kunches, L. M., Kilinsky, V., Lichtenberg,-D. A.,-Make, B. J. & McCabe, W. R. (1986). Risk factors for nneumonia and fatalitv in natients receiving continuous mechanical ventilation. Ame&an Reviews of Respiratory biseases 133, 792:796. Cross, A. S. & Roup, B. (1981). Role of respiratory assistance devices in endemic nosocomial pneumonia. The American Journal of Medicine 70, 681-685. Daschner, F. D. (1985). Useful and useless hygienic techniques in intensive care units. Intensive Care Medicine 11, 28+283.

168

F. Daschner

et al.

Dixon, R. E. (1983). Nosocomial respiratory infections. Infection Control 4, 376-381. Garibaldi, R. A., Britt, M. R., Coleman, M. L., Reading, J. C. & Pace, N. L. (1981). Risk factors for postoperative pneumonia. The American Journal of Medicine 70, 677-680. Gross, P. A., Neu, H. C., Aswapokee, P., Van Antwerpen, C. & Aswapokee, N. (1980). Deaths from nosocomial infection: experience in a university hospital and a community hospital. The AmericanJournal of Medicine 68, 219-223. LaForce, F. M. (1981). Hospital-acquired Gram-negative rod pneumonias: an overview. American Yournal of Medicine 70, 664-669. McCrae, W. I& Wallace,, P. (19Si). Aspiration around high volume, low pressure endotracheal cuff. Brztzsh Medical Yournal2. 1220-1221. Pierce, A. K. & Sanford, J. P. (1973). Bacterial contamination of aerosols. Archives of Internal Medicine 131, 156-l 59. Reinarz, J. A., Pierce, A. K., Mays, B. B. & Sanford, J. P. (1965). The potential role of inhalation therapy equipment in nosocomial pulmonary infection. Journal of Clinical Investigation 44, 831-839. Sanderson, P. J. (1983). Colonization of the trachea in ventilated patients: What is the bacterial pathway? Journal of Hospital Infection 4, 15-l 8.