15 December 2008

File under Unintended Consequences, 1

My friend and colleague Helen Branswell of the Canadian Press reports (via the Toronto Sun) on the cruel and accidental irony behind an outbreak of healthcare-associated infections at Toronto General Hospital between Dec. 2004 and Mar. 2006. Based on a new paper in Infection Control and Hospital Epidemiology, it's a fascinating and bizarre tale of the unpredictable hurdles that a hospital can face in attempting to eradicate HAIs.

It seems the hospital, in an attempt to reduce HAIs, installed hand hygiene stations in each room in its medical-surgical intensive care unit, in between the patient's bed and a countertop that held patient-care materials. This would seem like good design: The sink was right in the middle of the "zone of action" in the room, so health care workers would be reminded to use it (unlike, for instance, retrofitted rooms I have seen where the sink is away from the bed or out of the path between the bed and the door, and where health care workers have to consciously think about using it rather than having it be automatic). And the sinks were of a particular design meant to reduce accidental contamination of health care workers' hands: When the water was turned on, it flowed from a high gooseneck faucet straight down into the sink drain, without washing around the sink's side.

But it turns out that design and location both had unanticipated flaws. Water flowing straight into the drain was more likely to splash from the drain back out of the bowl; when investigators marked the sinks with fluorescent dye, they found splashes up to a yard away. Because the sinks were so close to the patient beds, the water was able to contaminate the patients, and the countertops on the other side as well. And because the water was falling directly into the sink drains, without the reduction in velocity caused by allowing it to wash around the sides of the sink, it was able to dislodge biofilm colonies of drug-resistant Pseudomonas aeruginosa, a moisture-loving organism that was growing in the sink pipes — which then splashed out of the sinks in the water bouncing back from the drain.

When the investigators found that, they had an explanation for why 36 transplant patients in that ICU had become colonized with MDR pseudomonas over 18 months. Twenty-four of the patients developed invasive infections, and 17 died; 12 of those deaths were either caused or closely related to pseudomonas infection.

The investigators tried multiple times to decontaminate the sink drains; in a few cases, they were successful, but the drains became recolonized and grew fresh biofilms. It was not practical to relocate the sinks. Nevertheless, they shut down the outbreak: They swapped out the faucets, decreased the water pressure, put a splash barrier on the sides of the sink, and moved patient care materials on the counter next to the sink elsewhere in the ICU rooms. Once those rearrangements were complete, the outbreak stopped.

This outbreak obviously was not MRSA, and in the strictest sense it is not relevant to MRSA, which is not an organism that lives in sink drains. But in a broader sense — as an illustration of the completely unpredictable hurdles that can stand in the way of excellent infection control — it is a useful and tragic cautionary tale.

The abstract is here. The cite is: Susy Hota, MD; Zahir Hirji, MHSc; Karen Stockton, MHSc; et al. Outbreak of Multidrug-Resistant Pseudomonas aeruginosa Colonization and Infection Secondary to Imperfect Intensive Care Unit Room Design. Infection Control and Hospital Epidemiology 2009 30:1, 25-3.

No comments: