Contact Precautions, Decolonization Best for MRSA Prevention in ICU

October 29, 2010 (Vancouver, British Columbia) — Universal contact precautions (CP) combined with universal decolonization prove to be the best methods for preventing colonization and infection with methicillin-resistant Staphylococcus aureus (MRSA) in the intensive care unit (ICU), according to a group of researchers presenting their findings here at the Infectious Diseases Society of America (IDSA) 48th Annual Meeting.

To estimate the health benefits and cost-effectiveness of different strategies to prevent MRSA transmission among patients in an ICU, the researchers used a Markov microsimulation model, which involved a hypothetical cohort of 100,000 adult patients admitted to an ICU.

Markov modeling is a way to represent a set of changing health states with a known probability of transition from one state to another over time. Such a model uses a specified set of states, the probabilities of a patient moving from one state to another, and the utilities of living in each state. A patient is always in one of a finite set of discrete health states, and an event is a transition from one state to another.

Courtney Gidengil, MD, MPH, from the RAND Corporation and the Division of Infectious Diseases at Children's Hospital Boston, Massachusetts, presented the results of the Markov modeling of various preventive strategies.

"Since it's obviously not very feasible to test all of these strategies and combinations of strategies against each other in a real-life clinical trial, we created a microsimulation model that tests the strategies against each other using the best data that we have available from other studies," she explained.

The strategies tested in the model were:

standard precautions (including careful hand washing by healthcare workers and cleaning and disinfecting of the environment)
universal CP (private room or room with another patient colonized with MRSA; gloves and gowns for anyone entering the room)
universal chlorhexidine gluconate (CHG) baths every day for 5 days
universal decolonization (CHG baths and nasal mupirocin ointment twice a day for 5 days)
universal CP and universal CHG baths
universal CP and universal decolonization.

"Universal" denotes that the strategy is applied to every patient admitted to the ICU.

The Markov model assumed an initial MRSA prevalence of 10% and the efficacy and compliance associated with alternative strategies for prevention. Some other assumptions of the model were an average ICU length of stay of 4 days, a 0.5% to 1.5% risk of acquiring MRSA colonization per day of ICU stay (rate dependent on colonization prevalence), a 0.4% development of MRSA infection per colonization day, efficacy of each intervention strategy, a $2 cost for a CHG bath, a $5 cost for decolonization, and a daily excess cost of $1250 for an MRSA infection.

Based on 100,000 patients, the assumptions of the model, and 1 run of the model, universal CP with universal decolonization prevented the highest proportion of cases of MRSA colonization (54%) and infection (51%), compared with standard precautions alone. Standard precautions alone were calculated to result in 6300 cases of colonization and 470 cases of infection.

Cases of MRSA Colonization and Infection Prevented (Compared With Standard Precautions Alone)

Precautions	Colonization prevented, % (cost per colonization prevented)	Infections prevented, % (cost per infection prevented)
Standard precautions	—	—
Universal CP	29 ($3,250)	26 ($49,000)
Universal CHG baths	12 (cost saving)	23 (cost saving)
Universal decolonization	34 (cost saving)	40 (cost saving)
Universal CP + universal CHG	38 ($1,110)	40 ($14,000)
Universal CP + universal decolonization	54 ($50)	51 ($760)

Dr. Gidengil said that both universal CHG baths and universal decolonization would save costs, compared with standard precautions alone. To implement the various strategies, on the basis of this 100,000-patient model, approximately 375,000 days of CP, 229,000 days of CHG baths, and 227,000 days of decolonization would be incurred.

She said certain limitations apply. One is that the results are preliminary and depend on the quality of the assumptions of the model. Another is that the investigators did not take patient quality of life into account, so that aspect has not been figured into the cost-effectiveness ratios. Finally, they have not yet incorporated the benefit of different strategies in preventing postdischarge MRSA infections or the possible development of resistance to CHG or mupirocin ointment.

"So our findings may be applicable to some hospitals that have similar rates of MRSA, but not necessarily other hospitals where MRSA rates may differ," Dr. Gidengil said. "To provide the best possible guidance to decision makers in different hospital settings, we plan to run the model while varying some key assumptions to see if and how our results might change."

Joel Ernst, MD, vice chair of the IDSA annual meeting and director of the Division of Infectious Disease at New York University School of Medicine in New York City, said that a lot of the Staphylococcus bacteria that infect people in the hospital are brought in by patients. Transmission in the hospital is important, he said, but is not the only way of preventing MRSA infections. "So in elective hospital admissions, it's good to be able to eradicate that carriage state. In emergency hospital admissions . . . that's much more difficult," Dr. Ernst said.

Dr. Gidengil's study focused on eradicating Staphylococcus on patients' skin and mucous membranes but did not consider hospital personnel. "I think that's kind of underemphasized," Dr. Ernst noted. The focus on staff is usually in the context of an outbreak investigation and not done routinely, he said.

Dr. Ernst and Dr. Gidengil have disclosed no relevant financial relationships.

Infectious Diseases Society of America (IDSA) 48th Annual Meeting: Abstract 426. Presented October 22, 2010.