ABSTRACT

Background: Acb has been increasingly identified as a cause of nosocomial infections in ICU in the last few years. Because of a perceived increase in the number of respiratory isolates occurring in the ICUs during the first few months of 1993., intensive surveillance of Acb isolation and infection was initiated. Following the documentation of a dramatic increase in Acb respiratory isolates an investigation was initiated.

Purpose: To define the epidemiology of the outbreak, to conduct a microbiologic environmental survey in an effort to identify a source and to determine risk factors for colonization and infection with Acb in the ICU.

Method: Data was collected regarding the isolation of Acb from all patients within the hospital during the months of 1993 and compared to baseline rates from the 5 preceding years. Extensive environmental cultures were done in the ICUs, respiratory therapy department, and sterile processing and distribution (SPD) departments. Molecular subtyping using pulsed-field gel electrophoresis (PFGE) was conducted on 100 selected isolates and a case control study was conducted to identify risk factors for colonization and infection.

Results: A total of 98 Acb isolates were identified from respiratory sites within the 5 ICUs representing 92% of total hospital isolates during the first 7 months of 1993. Defining a positive BAL as a marker for pneumonia there were 43 cases of nosocomial pneumonia. A single strain of Acb identified by PFGE was found in 29/33 (88%) environmental isolates tested and was concordant with 93.5% if 31 respiratory tract isolates tested. Quantitative cultures of broth in which the temperature probe had been inserted revealed Acb at a concentration of >10⁴ CFU/ml. Risk factors significantly associated with colonization or infection identified to date include tracheostomy (p=0.007), length of ventilation (p<0.0001), concomitant isolation of Xanthomonas maltophilia (p<0.02) and use of enteral nutrition (p=0.0003). Respiratory isolation of Acb was significantly associated with mortality (p=0.02). Only after sterilization of the temperature probes was introduced did the isolate of Acb decrease to endemic rates in the ICU.

Conclusions: We conclude that contaminated temperature probes may have played a significant role in this outbreak of Acb in multiple ICUs.

BACKGROUND:

Acinetobacter baumannii (Acb) is a recognized nosocomial organism which has been seen with increased frequency in the hospital setting in recent years. Several outbreaks [1-3] involving Acb have been reported in the Intensive Care Unit (ICU). Most Acb infections are opportunistic in nature and occur in severely compromised patients. Environmental reservoirs which have been implicated in the transmission of Acb in the ICU setting most often include mechanical ventilators and respiratory equipment [1-5]. In addition, transmission may occur on the hands of health-care workers [2-6]. Recently, the ventilator circuit temperature probe has been implicated as a potential source for the transmission of Pseudomonas cepacia in the ICU setting [7]. To our knowledge, the temperature probe has not been previously implicated as a potential source of contamination responsible for respiratory tract colonization and infection due to Acb. With a documented increase in the number of respiratory isolated of Acb noted from several ICUs at the Toronto General Division during the first few months of 1993, an outbreak investigation was launched with the following objectives.

PURPOSE:

The objectives of our investigation were to define the epidemiology of the outbreak, to conduct a microbiologic environmental survey in an effort to identify a source and to determine the risk factors for colonization and infection with Acb in the ICU.

METHODS:

Patient Population and Facility Description

The Toronto Hospital is a 1,220 bed, medical school affiliated 2 site acute care tertiary referral centre. There are 5 ICUs at the General Division (GD) and 2 at the Western Division (WD). The GD has a Surgical, Medical, 2 Cardiovascular, and 1 Coronary ICU and the WD has a Neurosurigcal and combined Medical-Surgical ICU. The ICUs have mainly single and double rooms with some 3 and 4 bed rooms in two units. Each room has a sink at each bedside and a sink in a shared bathroom. Respiratory equipment is replaced every 48 hours and between each patient. A common respiratory therapy room serviced all the units at the General Division. Temperature probes, unable to be steam or gas sterilized, were wiped with a 4% chlorhexidine preparation. Ventilator circuits were pasteurized. Body substance precautions were used for all patients.

The Outbreak

Data was collected regarding the isolation of Acb from all patients from all sites and from respiratory sites within the General Division during the first months of 1993 and compared to baseline rates from the 4 preceding years. Recognizing the distinction between respiratory tract colonization and pneumonia is difficult; data was also collected on the number of positive bronchoaveolar lavages and/or brushings for Acb in one ICU, using this data as a surrogate marker for invasive infection.

Environmental and Patient Cultures

Multiple environmental cultures including quantitative cultures when appropriate were obtained from the Respiratory Therapy (RT) department, the ICUs, and the Supply, Processing, and Distribution (SPD) department. Sterile cotton swabs, moistened with Brain-Heart Infusion broth (BHI), were used to collect cultures from faucets, water, sink, drains, suction equipment, ventilator tubing, in use "water-traps", and humidifiers from the 5 ICUs. Similarly, cultures were obtained from 0, tanks, humidifiers, filters, respiratory circuits, water traps, Laerdal bags, bronchoscopes, and temperature probes from the RT department. Specimens from the pasteurizer surfaces, water, filters, and the dryers from the SPD department were also obtained. In addition handwashing cultures were done in sterile plastic bags containing 50 ml of BHI. Cultures were plated on the blood and MacConkey agar and Acb isolates were presumptively identified by morphology, gram stain, and glucose fermentation. Species level identification was done using standard techniques.

Strain Characterization

Molecular subtyping was performed using low-frequency cleaving restriction endonucleases, Sma I or Sfi I with subsequent pulsed-field gel electrophoresis (PFGE) to determine patterns of separation. Subtyping was performed on 100 non-duplicate isolates obtained over a 4 month period.

Case-Control Study

For the case-control study, all patients colonized with Acb during interval was matched by admission APACHE 11 Score with control patients who were in the ICU at the same time but who did not become colonized with Acb.

RESULTS:

The Outbreak

The number of isolates of Acb from the entire hospital, the total number of respiratory isolates from the hospital and the number of ICU respiratory isolates are shown in Figure 1. The majority of the isolates were from the 5 ICUs (Figure 2). The number of BALs from which Acb was isolated increased dramatically in 1993 (Figure 3). A total of 98 Acb isolates were identified from respiratory sites within the 5 ICUs during the first 7 months of 1993, representing 92% of total hospital isolates.

Environmental and Patient Cultures

All cultures of the bronchoscopes and related equipment were negative for Acb. Similarly, all environmental cultures in the SPD were negative. An initial survey in the 5 ICUs of all patients and their respiratory circuits, and water traps revealed Acb in 7/12 (58%) inhalation line water traps including one patient on whom Acb had never been cultured previously. A more extensive survey in the 5 ICUs revealed Acb in 7/8 inhalation line water traps in concentrations exceeding 10⁶ CFU/ml (Figure 4). Similarly cultures of the temperature probes revealed Acb in 7/7 (100%) cases. Of these 7 cases, Acb had never been cultured previously in the majority of cases (6/7). Cultures from the RT department of clean ventilators revealed a temperature probe, a water trap, and an expiratory circuit filter to be positive for Acb. Acb was isolated from 17% of staff hands in the ICU.

Strain Characterization

Of the 100 isolates examined 36/40 (90%) isolates from the GD ICU (5) were of an identical pattern, including 93.5% of respiratory isolates and 77.7% of blood isolates. Of 33 environmental and handwashing isolates tested 88% were of the same concordant pattern as the ICU respiratory isolates. In contrast, only 1/15 (6.7%) Acb "ward" isolates had the same pattern as the ICU isolates.

Case-Control Study

In the case-control analysis, risk factors for acquisition of Acb were tracheostomy (p=0.006), ventilation > 3 days (p<0.0001), enteric feeding (p<0.0001), dialysis (p=0.03), and colonization with Xanthomonas maltophilia (p=0.02).

Outbreak Control

Control measures were implemented in the ICUs, RT, and SPD departments relating to the care of the respiratory equipment, including circuit changes in the ICUs, disassembly of circuits, assembly of ventilators, and storage. In addition, patients in the ICUs were cohorted and strict handwashing was enforced. Temperature probes were subjected to gas sterilization. Following full implementation of this last recommendation, there was a dramatic decline in the number of isolates of Acb (Figure 5).

CONCLUSIONS:

Although it may be found as a commensal. Acb may cause respiratory tract colonization and invasive infection in ICU patients.

Rapid colonization of ventilator circuits with proliferation to high levels may occur with Acb.

Prolonged ventilation, enteric feeds, tracheostomy, dialysis and colonization with Xanthomonas maltophilia are associated risk factors for colonization with Acb.

The homogeneity of the isolates on macrorestriction analysis provides compelling evidence that this outbreak was due to a clonal "epidemic" strain of Acb.

Although many factors may have contributed to the outbreak, contamination of the temperature probes likely was a major factor.

REFERENCES:

Castle M, Tenney JH, Weinstein MP, Eickboff TC. Outbreak of a multiple resistant Acinetobacter in a surgical intensive care unit: epidemiology and control. Heart Lung 1978:641-644.

Vandenbroucke-Grauls CMJE, Kerver AJH, Rommes JH, Jansen R, den Dekker C, Verhoef J. Endemic Acinetobacter anitratus in a surgical intensive care unit: mechanical ventilators as reservoir. Eur J Clin Microbial Infect Dis 1988:485-489.

Cefai C, Richards J, Gould FK, McPeake P. An outbreak of Acinetobacter respiratory tract infection resulting from incomplete disinfection of ventilatory equipment. J Hosp Infect 1990:15:177-182.

Hartstein AI, Rashad AL, Liebler JM et al. Multiple intensive care unit outbreak of Acinetobacter calcoaceticus subspecies anitratus respiratory infection and colonization associated with contaminated, reusable ventilator circuit and resuscitation bags. Am J Med 1988:85:625-631.

Stone JW, Das BC. Investigation of an outbreak of infection with Acinetobacter calcoaceticus in a special care baby unit. J Hosp Infect 1985:6:42-48.

Beck-Sague CM, Jarvis WR, Brook JH et al. Epidemic bacteraemia due to Acinetobacter baumannii in five intensive care units. Am J Epidemiol 1990:132:723-733.

Berthelot P, Grattard F, Mahul P, Jospe R, Pozzetto B, Ros A, Gaudin OG, Auboyer C. Ventilator temperature sensors: and unusual source of Pseudomonas cepacia in nosocomial infection. J Hosp Infect 1993:25:33-43.

Back to Posters and Abstracts