Decontaminate all incoming ICU patients to reduce infections, says RCT

"Decolonizing" New ICU Patients Reduces Bloodstream Infections: NEJM

This article had an erratum posted in NEJM: read more here.

In humankind's battle against bacteria, the ICU is the front line. And with MRSA infection rates doubling in the past 5 years, and the more recent and scary spread of lethal pan-resistant Enterobacteriae, lately the bugs have won a few rounds.

The problem isn't just catching a deadly bacteria from the patient in the next room by way of your doctor's dirty stethoscope. ICU patients are at greatest risk from the bacteria they carry in on their own bodies (their microbiota). As a critically ill patient's immune system becomes weakened by stress, illness and invasive devices, these bacteria previously living harmlessly in the body (MRSA in the nose, gram-negative rods in the colon) jump the fences and wreak havoc throughout the body -- usually in the blood or lungs. Add selective pressure from antibiotics and the presence of antibiotic-resistant bacteria able to share their secret DNA codes, and you get today's superbug-saturated ICU, a "meta-threat" to patients already fighting serious illnesses or injuries.

It's enough of a clear and present danger that 9 states now legally mandate MRSA screening for ICU patients. But MRSA screening and barrier precautions alone are of debatable benefit: a Swiss study in JAMA and the large STAR*ICU trial in NEJM suggested they're almost useless. An Ann Int Med study argued MRSA screening works beautifully, but only when there is a rapid test turnaround, and the VA reported big success, as well. In the real world, meanwhile, MRSA and other resistant bacteria have continued to spread throughout U.S. ICUs despite the widespread adoption of screening and barrier precautions.

The next escalation in the campaign against noscomial infections was to "decolonize" the noses and skin of patients who were discovered to be carrying MRSA; this reduced colonization but not necessarily infections. Undaunted, Susan S. Huang et al (working along with the Agency for Healthcare Research and Quality, and the Centers for Disease Control and Prevention) asked, what if we decolonize every patient who comes into the ICU with a thorough antiseptic scrubdown? This proactive tactic has been shown to reduce infections in the ICU before. Their impressive results of the "REDUCE MRSA" trial are in the June 13, 2013 New England Journal of Medicine.

What They Did

74  ICUs in 45 HCA hospitals in 16 U.S. states (caring for 74,256 patients) were randomized using cluster randomization to one of three arms:

• MRSA screening in the nares on admission to the ICU; contact precautions if positive (the current standard of care; the control group);

• Screening for MRSA of all incoming ICU patients, and if positive, "targeted decolonization" with 5 days of nasal mupirocin and baths with chlorhexidine-soaked cloths like these;

• No screening for MRSA on incoming ICU patients, with "universal decolonization" with mupirocin and chlorhexidine skin baths.

Because all the hospitals were HCA, they had shared policies and procedures, and could use internal company resources (posters, etc) to try to boost compliance with protocols. More than 90% of patients in the screening arms underwent screening.

The primary outcome was any new MRSA-positive culture attributable to being in the ICU. Other measured outcomes were bloodstream infections in the ICU from MRSA or by any pathogen. Cultures were obtained at doctors' discretion for clinical reasons (i.e., there were no surveillance cultures). Randomization appeared to be adequate and interventions performed appropriately and evenly across sites.

What They Found

Comparing the 12-month baseline period to the 18-month intervention period,

• The control group saw roughly an 8% drop in confirmed MRSA infections (from 3.4 to 3.2 per 1,000 patient-days).

• Targeted decolonization of patients found to be MRSA-positive at ICU admission reduced MRSA infections by 25% (4.3 --> 3.2 per 1,000 patient-days).

• ICUs implementing universal decolonization on all patients (not bothering to screen) saw reductions of MRSA infections of 37% (3.4 -- > 2.1 per 1,000 patient-days), statistically signfiicant.

Universal decolonization also resulted in a 44% reduction in ICU bloodstream infections from any pathogens; routine screening and isolation (control group) saw no reduction in overall ICU bloodstream infections, while the targeted decolonization group had a 22% reduction in ICU bacteremia (statistically significant).

MRSA bloodstream infections dropped by 28% after implementation of universal decolonization, while MRSA bacteremia rose in the other two groups; however, these events occurred less frequently and the finding wasn't statistically significant.

A possible caveat: The universal decolonization group was randomly allocated 3 out of the 4 hospitals performing bone marrow and solid organ transplants -- which happened to be where the highest rates of ICU infections occurred. This could conceivably have caused universal decolonization to seem to perform better. After trying to control for this, universal decolonization seemed to retain the proof of its benefit.

What It Means

Authors estimate their intervention would prevent one ICU bloodstream infection for every 54 patients cleaned, and one MRSA infection for every 181 patients treated. If this is a real effect, a lot of lives could be saved: up to 24,000 people might die of ICU bloodstream infections each year. Is it feasible and cost effective? Authors say their intervention put little extra burden on nursing staff, and only increased costs by $40 per patient. That's only about $2,160 per ICU bacteremia prevented, and $7,240 per MRSA infection.

Sage Products has mainly been selling its chlorhexidine-impregnated cloths as prevention against surgical site infections. If they and their competitors can convince hospital administrators of the benefits of antiseptic bathing of all ICU patients, they'll have a massive new market. With each catheter-related bloodstream infection estimated to cost $18,000 (and an extra 12 hospital days), and with Medicare and private insurers refusing to pay for infections acquired in the hospital, this should be an easy sell.

Until now, HCA has been best known for paying the largest settlement in history for Medicare fraud (almost $2 billion), and for more recent allegations that hundreds of unnecessary cardiac procedures were performed at HCA hospitals. If more hospitals do implement universal cleansing of ICU patients and see similar benefits to those in this trial, HCA will get to polish up its tarnished brand ... a little.

Susan S. Huang et al. Targeted versus Universal Decolonization to Prevent ICU Infection. N Engl J Med 2013; 368:2255-2265