INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The spread of infectious agents between subjects who have undergone spirometric testing is extremely uncommon. The one case reported by Hazeleus and colleagues (1) of a patient becoming skin-test-positive for tuberculosis after having had spirometry performed on equipment that had previously been used to test a patient with tuberculosis may have represented a tuberculin booster phenomenon rather than true skin test conversion. Despite this concern, this solitary report is part of the basis for marketing single-patient-use bacterial filters, a practice that substantially increases the cost of spirometric testing. There has been no documented case involving the transmission of pathogenic organisms by spirometry in more than 225,000 tests conducted in our institution during the past 30 yr.

Mucosal contact cross-contamination during routine testing is typically avoided by using disposable mouthpieces. There is adequate documentation that respiratory pathogens are deposited on the walls of testing equipment during forced expiratory maneuvers by culturing the internal surfaces (2, 3). However, it seems unlikely that organisms so deposited would be reaerosolized during routine testing maneuvers. Although there has been ample documentation that aerosol-generating equipment such as humidifiers and nebulizers can spread infectious agents (4), there is no evidence that spirometers generate aerosols. We have found no reports in the literature that unequivocally document the spread of infectious agents between patients by way of spirometers during pulmonary function testing.

We set out to test the following hypothesis: significant transfer of aerosolized bacteria will not occur during a forced inspiratory effort that follows a forced expiration if adequate time is allowed between these two events.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We studied the retrieval of nonpathogenic Escherichia coli that had been aerosolized at various concentrations into standard spirometry tubing. The presence of organisms reaching the distal end of the tubing was verified by culture. After separate contaminated aerosol injections, air from within the tubing was forcibly evacuated at intervals of 0, 1, 5, and 10 min, respectively, through a special petri dish holder. The petri dish was then cultured to detect the recovery of organisms from the air in the tubing.

We utilized a standard nonpathogenic laboratory strain of E. coli ATCC 25922 bacteria. The organisms were inoculated into Mueller Hinton broth and incubated overnight at 37° C. The overnight suspension was adjusted to a McFarland 1 standard dilution of 3 × 10⁸ organisms per milliliter. The organism test concentrations of 1 × 10⁸, 3 × 10⁷, 1 × 10⁷, 3 × 10⁶, and 1 × 10⁶ were obtained by serial dilution of the standardized sample made immediately before use.

Forced breaths were simulated by forcing air into and out of the system using a 3-L calibration syringe (SensorMedics Corp., Yorba Linda, CA) (Figure 1).

View larger version (34K): [in this window] [in a new window]   Figure 1.   1. Remote actuating switch. 2. Calibrated 3-L syringe. 3. Proximal microbial filter. 4. Proximal petri plate assembly. 5. Spirometry tubing. 6. Nebulizer for aerosol generation. 7. Distal petri plate assembly. 8. Distal microbial filter. 9. Rosenthal-French dosimeter. All experiments were done with the hose extended rather than flexed on itself as pictured.

The nebulizer was powered by compressed air that was metered through a Victor two-stage pressure regulator (Victor Medical, Denton, TX) set at 20 psi. This source was connected via standard oxygen tubing to the control solenoid of a Rosenthal-French Nebulization Dosimeter Model D-2A (Applied Immunology Inc., Baltimore, MD). The dosimeter was controlled using a modified manual switch attached to the handle of a 3-L calibration syringe. The control solenoid "on" duration was controlled by an adjustable dial that was calibrated according to the manufacturer's directions. The control solenoid output was attached to a DeVilbiss no. 646 nebulizer (DeVilbiss Medical Division, Somerset, PA) that produced a droplet profile of 0.3 to 5.0 µm (Sunrise Medical Respiratory Products, Somerset, PA, personal communication). The side vent and the thermistor port on the nebulizer were sealed. A 3-ml aliquot of diluted, standardized E. coli solution was placed inside the nebulizer at the beginning of each test run. Care was taken to always maintain the nebulizer assembly in an upright position throughout the procedure. The nebulizer was attached via a custom T-fitting to the proximal end of the pulmonary function tubing, just after the proximal petri plate holder and syringe.

The petri plates were held in place using custom plate holders of our own design, which suspended a standard blood agar petri plate with an orthogonal orientation to the airflow. Prior to each test, the assemblies were soaked in a bactericidal solution (Cidex; Johnson & Johnson Medical Inc., Arlington, TX) for at least 10 min. They were then rinsed in tap water and air-dried. The petri plate holder inlet and outlet orifices were capped after the insertion of a standard 10-cm petri plate into the device utilizing a special jig. The petri plate holders were uncapped immediately prior to insertion into the system for each individual test. After the test was finished, the blood agar plates were removed from the plate holder assemblies and cultured.

The spirometry hose used was a standard pulmonary function test tubing 1 m in length and 3.34 cm in diameter (part no. 894787 and 894788; SensorMedics). Before each run, the tubing was sterilized in a bactericidal solution (Cidex) and thoroughly rinsed. It was subsequently cleaned after the testing of each concentration of E. coli suspension was completed. Cultures were periodically obtained by swabbing the inside of the tubing to verify that viable E. coli could be recovered from the inside wall.

Microbial filters were placed between the air injection syringe and the rest of the assembly, as well as the distal end of the test fixture. These were used to minimize E. coli contamination of the surrounding environment and syringe interior.

At the beginning and end of each test run, control plates were obtained utilizing a standardized concentration of E. coli suspension. For these runs, the equipment was configured so that the distal petri plate holder was placed immediately downstream from the contaminated aerosol injection port, with no tubing in between. Three liters of air were then forcibly injected through the system. E. coli colonies were counted on the petri plates after incubation at 37° C for 24 h. The spirometry tubing was then inserted between the aerosol injection port and the distal petri plate holder. The forcible injection of 3 L of air past the aerosol nebulizer port into the tubing was repeated. The dosimeter was set so that nebulization of organisms would start after airflow had started and would cease before airflow stopped. This insured that no retrograde movement of aerosol could occur to contaminate the proximal petri plate. The presence of contaminated aerosol at the distal end of the tubing was verified by culturing the distal tube petri plate. After a time delay of either 0, 1, 5, or 10 min, air was then forcibly withdrawn from the proximal tube so that the 3-L syringe was filled in 0.5 s, resulting in an average flow rate of 6 L/s. The timing was confirmed with a stopwatch. This air was drawn past the proximal tube petri plate holder. Both distal and proximal petri plates were incubated for 24 h at 37° C, and colonies were manually counted.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The data generated from 50 separate trials, including 18 controls, are shown in Table 1. Each numbered pair represents an average of the colony counts recovered on the petri plates from the air that passed the proximal (numerator) and distal (denominator) petri holders. The control values represent a mean colony count of the trials without intervening spirometry tubing between the aerosol port and the distal petri plate assembly for each concentration of organisms.

All trials with no time between ejection and withdrawal of the aerosol had organisms identified on both proximal and distal plates. When 1 min was allowed between injection of organisms and withdrawal of air from the tubing, there was minimal contamination of the proximal plate noted only at the higher concentration of injected organisms. No E. coli were found on the proximal plate if an interval of 5 or 10 min was allowed, no matter what concentration of E. coli was used.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The subject of interpatient transmission of infections is a frequent topic of concern among directors of pulmonary function laboratories. Part of the reason for this concern is the absence of published data to provide direction. This has allowed the manufacture of single-use bacterial filters to be promoted to prevent a problem for which, for all intents and purposes, may have no basis in fact.

Procedures to minimize mucosal cross-contamination are well documented (5). The most recent recommendations are outlined in the Official Statement of the American Thoracic Society Standardization of Spirometry 1994 Update. These procedures include prevention of infection transmission to technicians exposed to contaminated spirometer surfaces through proper hand-washing between patients or the use of latex gloves if there are open cuts or sores on the technician's hands. Mouthpieces, noseclips, and any other equipment that is in direct mucosal contact should be changed between each subject. Nondisposable equipment should be properly sterilized. If the hose to the spirometer has begun to collect visible condensation, and certainly if there is any pooling of secretions in the tubing, it must also be changed. There are recommendations in addition to those listed that are too detailed to be reviewed in detail at this time. The reader is referred to the statement for the complete recommendations. It should be noted that several of the recommendations are empiric rather than being based upon scientific data.

Our data indicates that the recovery of E. coli organisms appears to be a function of both the concentration of the organisms and the elapsed time between the forced injection of organisms into the equipment and the withdrawal of air from the system. Even when high concentrations of aerosolized E. coli were aerosolized into the spirometry tubing, no bacteria could be recovered from the air in the tubing as long as it was not withdrawn sooner than 5 min after the injection. Five minutes appears to be an adequate interval to allow organisms to be removed by gravitational sedimentation or brownian movement from the air in the tubing even though organisms could be recovered by swabbing the walls of the tubing. The absence of aerosolized E. coli at 5 and 10 min after injection into spirometry tubing supports the hypothesis that there was no transfer of aerosolized organisms from the pulmonary function tubing as long as there was at least a 5-min delay between injection of the aerosolized organisms into the tubing and forcible withdrawal of the air from the tubing.

Although we anticipate that this would be the case for any spirometer that requires a long hose to interface between the patient and the spirometer (such as with volume-based spirometers), it may be different with spirometer systems that are flow-based since a large tube is not required with many of these systems. We, therefore, feel that the hypothesis needs further testing using intact spirometric equipment of various types in use on real patients.