Regional Control of a large KPC outbreak: The Israeli Experience

The oubreak DID NOT occur here.
However, it IS St. Patrick's Day.

Hello everybody.  I hope you're all having a great St. Patrick's Day and a great Match Day.  Not sure those two days should be combined, at least for the safety of the future of the medical profession, but for some reason, that decision is not left up to me...

There is a report out electronically in CID (scheduled for April 1) by Mitch Schwaber et al. that describes the containment of a country-wide, carbepenem-resistant Klebsiella pneumoniae outbreak in Israel.  The outbreak of a highly-resistant strain, typically susceptible only to gentamicin and colistin, began in 2006 in multiple Israeli hospitals. The resistance was mediated by KPC-3 and local efforts to control the outbreak were largely unsuccessful. By March 31, 2007 there had been 1275 patients in 27 hospitals affected (13,040 beds).

In March 2007, the Israel Ministry of Health implemented a 3 component intervention: 1) Mandatory reporting of every patient with a carbepenem-resistant Enterobacteriaceae (CRE); 2) mandatory contact isolation of all known CRE carriers within self-contained nursing units in single rooms or cohorts with dedicated equipment AND dedicated nursing; and 3) creation of a nationwide task-force with statutory authority to intervene as necessary to control the outbreak.

Did it all work?  Well, they probably wouldn't have published this if it didn't work. They'd still be working too hard trying to stop the problem!  From the peak of 185 cases (56 cases/100,000) in March 2007 (92% of CREs were Klebsiella) infections fell to a low of 45 (12 cases/100,000) in May 2008, a 79% decline. I have pasted the incidence curve below. So it worked, but there are still too many CREs. If they let their guard down, the outbreak could easily reoccur.

As far as the study design, the usual caveats apply.  Despite great statistical control, they did not include a non-equivalent control group, etc, so perhaps these findings could be partially explained by regression to the mean or other biases, such as non-recorded interventions. Do I think that is what is going on here?  No.  I think the nationwide effort probably worked and their analysis and interpretation are correct.  This overwhelming response might be needed more often in the future given the lack of new antimicrobials, poor overall support for infection control (everywhere, not just in Israel) and continued overuse of the antimicrobials we do have.

Reference: Schwaber MJ et al. CID April 2011

Note: Dan posted on the CDC Guidance for Carbapenem-Resistant Enterobacteriaceae a couple years ago.