A case-control study of airborne dissemination of bacteria during CT colonography

Abstract

Objectives:

The COVID-19 pandemic has highlighted the importance of aerosol spread of infection. We have conducted a study to detect bacterial contamination of room surfaces and room air during CT colonography and confirm the efficacy of room disinfection procedures carried out between examinations.

Methods:

Systematic sampling of the CT examination couch and horizontal surfaces 1 m and 3 m from the couch was performed before and after patient examinations. A 1 m³ sample of room air was obtained during patient examinations. Samples were processed using routine laboratory methods. A case–control study design was used (30 CT colonography and 30 routine body CT scans).

Results:

Evidence of airborne dissemination of bacteria was detected in >30% of CT colonography examinations and <10% of control examinations (p = 0.01). No pathogenic bacteria were detected in surface samples taken before patient examinations.

Conclusion:

The room disinfection policy in use in our CT department is effective in eliminating pathogenic bacteria from surfaces in the patient environment. CT colonography causes contamination of room air with enteric bacteria in a significant proportion of cases.

Advances in knowledge:

CT colonography may possibly be an aerosol-generating procedure. Larger-scale investigation is needed to fully evaluate this potential infection risk.

Introduction

The current pandemic caused by the coronavirus SARS-CoV-2 has required implementation of stringent infection control measures to reduce the risks of cross-infection, both between patients and to health-care staff.^1,2 SARS-CoV-2 is transmitted primarily either directly by droplets from the respiratory tract of infected persons, or on the hands following contact with contaminated fomites. In addition, virus-contaminated aerosol particles (of the order of 10 microns diameter or less) are considered particularly hazardous, as they may remain airborne for an extended period and can penetrate deep into the respiratory tract, circumventing the body’s first-line defences against infection.³

While respiratory droplets are the principal vector for transmission, faecal–oral transmission of SARS-CoV-2 has also been suspected, viable virus particles have been isolated from the stools of infected patients and prolonged faecal excretion of virus antigens following clinical recovery from respiratory infection is well described.⁴

Control measures implemented to date have focussed on areas of clinical practice where environmental contamination by respiratory droplets is intense. Published guidance to the National Health Service in England defines a graded approach to patient care and personal protection, with the highest-risk treatments and procedures classified as “aerosol-generating procedures” (AGP).¹ The impact of this guidance extends far beyond immediate care of infected patients, profoundly impacting the practice of surgery under anaesthesia, dentistry and certain invasive diagnostic procedures. The measures required for management of AGP have adversely affected productivity, resulting in delays to patient care. Concern about the harms resulting from delayed investigation and treatment is particularly acute with respect to cancer services.⁵ However, safe restoration of services must recognise health-care employees’ right to adequate protection in the workplace.

Diagnosis of cancer of the digestive tract relies primarily on endoscopic biopsy techniques (oesophagogastroduodenoscopy and colonoscopy), of which only the former is currently classified as an AGP.¹ In recent years, computed tomography colonography (CTC) has become a validated radiological alternative to colonoscopy.⁶ CTC requires purgative bowel preparation and controlled insufflation of the large bowel via a rectal catheter before an abdominopelvic CT scan is performed. Up to 4 litres of carbon dioxide is insufflated, much of which is subsequently passed as flatus. Faecal soiling of the examination couch is also frequent. Despite this, CTC is not currently classified as an AGP.¹

We therefore designed a case–control observational study to detect coliform bacterial contamination of the patient environment during routine body CT examinations and CTC as a surrogate marker of faecal contamination. We quantified the number of bacteria on surfaces to indicate direct or droplet transmission, and sampled room air to detect airborne dissemination. The study was conducted as a service review with the support of our institution’s (The University Hospitals of Morecambe Bay NHS Foundation Trust) department of research and development. The requirement for research ethics committee approval was waived.

Methods

CTC was performed according to our standard operating procedure; patients were changed into a hospital gown and a silicon balloon-tipped rectal catheter inserted with the patient in left decubitus position. The gown and a cotton sheet were used to cover the patients’ buttocks during the examination. Carbon dioxide was insufflated up to a maximum of 4 litres using a proprietary colon insufflator (ProtoCO2l^TM, Bracco Inc. Milan, Italy).

Contact samples were taken from the CT examination couch and horizontal surfaces at 1 m and 3 m from the couch (workbenches, electrical plant cabinets). Samples were collected immediately prior to patients entering the room and immediately after patients left the room, using a proprietary double-sided culture medium (plate count agar and violet red bile glucose agar) (Oxoid^TM dip-slide Oxoid Ltd. Basingstoke UK ). A total of six swabs was taken from each CT examination. Swabs were incubated in an aerobic atmosphere at 36^o C for 24 h and processed using routine microbiological techniques.

Air samples were taken using a proprietary microbial air sampler (Merck^TM MAS-100, Merck KGaA, Darmstadt, Germany) developed to monitor air quality in operating theatres and validated to United Kingdom Accreditation Services (UKAS, Staines-upon-Thames, UK) standards to ensure accuracy of measurement. The device was placed 1 m from the examination couch. A 1 m³ sample of room air was aspirated over a 10 min cycle, inoculating a 90 mm diameter Petridish containing cystine–lactose–electrolyte-deficient agar and replicating our standard air sampling protocol for operating theatres. Plates were incubated and processed in the same way as the contact samples.

To ensure consistency, all samples were collected by the same investigator. Samples were collected from 30 CT colonography examinations and 30 control CT examinations (out-patient and emergency body scans). Samples were collected from three different CT rooms at two different hospitals over a period of 8 weeks between August and October, 2020. Radiology department staff had no prior knowledge of when samples would be collected, ensuring that room preparation was representative of routine practice. The investigator responsible for sample collection was blinded to the results until the study had been completed.

Results

Contact samples

Of 180 contact samples collected from control CT examinations (6 swabs from each of 30 examinations), none yielded positive cultures for coliform bacteria. Of the surface swabs collected from CTC examinations, none of 90 pre-procedure swabs yielded positive cultures, while 4 swabs taken after the procedure yielded positive cultures for coliform bacteria (two from the patient examination couch, one from a surface 1m from the couch and one from a surface 3 m from the couch). (Results summarised in Table 1).

Table 1.

Results of post-examination contact sampling of room surfaces

	No coliforms detected	Coliforms detected	Total
Controls	30	0	30
CTC	26	4 (NS)	30

CTC, computed tomography colonography.

Air samples

All of the air samples had detectable total viable counts (TVC) of non-pathogenic bacteria, as expected given the presence of staff and patients in the room during sampling, with a mean colony count of 129 in the control group and 96 in the CTC group. Of 30 samples in the control group, 2 yielded positive cultures for coliform bacteria, which we were using as a marker of faecal contamination. Of 30 samples in the CTC group, 11 yielded positive cultures for coliform bacteria. The mean colony count for coliforms was two colony-forming units (range 1–8). There did not appear to be a difference in the coliform colony counts between the control and CTC groups. The two positive cultures in the control group were obtained from routine out-patient body CT scans at two different hospital sites on different dates. Using a Fisher’s exact test, the probability that the difference in numbers of positive coliform cultures between the control and CTC groups had occurred by chance was calculated to be p = 0.01 (Results summarised in Table 2).

Table 2.

Results of room air sampling during examinations

	No coliforms detected	Coliforms detected	Total
Controls	28	2	30
CTC	19	11 (p = 0.01)	30

CTC, computed tomography colonography.

Discussion

This study was prompted by concern that if CTC is indeed an AGP, this could pose a threat of faecal–oral transmission of SARS-CoV-2, the virus responsible for COVID-19. Isolation of viable SARS-CoV-2 virus particles from stool remains technically challenging, while tests for viral antigens can be falsely positive due to prolonged excretion of non-viable virus components. We therefore searched instead for evidence of contamination by enteric bacteria, using laboratory methods commonly used for monitoring operating theatres and catering facilities. This enabled us to detect viable organisms considerably larger than intact virus particles, allowing the inference that dissemination of virus particles could also have occurred.

The results of surface sampling confirm the efficacy of current room hygiene practice in controlling the risk of nosocomial infection as a result of faecal contamination of the patient environment. Detection of contamination of room surfaces exclusively among the post-examination samples in the CTC group and not the control group suggests that contamination occurred as a result of the examination, but this observation did not reach statistical significance.

The frequency with which coliforms were detected in room air among the CTC group supports the study hypothesis that CTC causes airborne dissemination of enteric microorganisms. The statistical analysis of our data suggest that the findings are likely to be externally valid.

The extent to which our findings can be generalizsd is limited by the small sample size used. Furthermore, we were limited to taking a single air sample from each patient examination (due to the requirement to resterilise the air sampling head after each use), and chose a fixed distance of 1 m from the examination couch. Particles detected at this range could in theory have been of a size conventionally considered to be droplets (>5 micron diameter) rather than aerosols (<5 micron diameter). However, a recent study⁷ has suggested that airborne droplets may spread much further from source than previously appreciated under cool, humid atmospheric conditions, and the infection hazard presented by airborne particles is increasingly viewed as a continuum for the purposes of infection control practice. The only case in which we detected surface contamination 3 m from the examination couch also had a positive air sample.

The density of coliform bacteria detected (1–8 colony-forming units per 1 m³ air sample) is below the estimated threshold for human infection by most bacterial pathogens,⁸ suggesting that the absolute risk of infection by this route is extremely small. Recent studies addressing the risk of COVID-19 transmission following resumption of out-patient colonoscopy and CTC services^9,10 have both indicated that cross-infection is unlikely to occur if appropriate infection-control procedures are implemented.

The influence of room ventilation systems on our findings was not addressed by our study. Current guidance in the United Kingdom¹¹ recommends that CT scanner rooms should be supplied with a minimum of 10 air exchanges per hour, but we have been unable to locate data indicating how frequently healthcare providers are granted exemptions from this requirement, or the extent to which installation of air filtration systems has been shown to be effective in improving air quality at sites where the recommended frequency of air exchanges cannot be achieved. We observed an inverse linear relationship between room volume and the proportion of positive cultures obtained across the 3 CT rooms included in our study, but this did not reach statistical significance.

We suggest that the findings from this study warrant larger-scale investigation to fully evaluate the potential for nosocomial transmission of infection during CTC and establish appropriate procedures to mitigate risk to staff and other service users. Staff at our institution currently follow the guidance published jointly by the British Society for Gastrointestinal and Abdominal Radiology and Society and College of Radiographers.¹² It seems prudent to perform CTC on dedicated lists whenever practicable, with a 30-min interval between the last CTC and room cleaning, to allow any contaminated airborne particles to settle. Service managers should also review the room ventilation specification of their CT scanners, as suboptimal ventilation may result in an increased risk of environmental contamination during CTC.