The prevention paradox, as described in 1981, is the “seemingly contradictory situation where the majority of cases of a disease come from a population at low or moderate risk of that disease, and only a minority of cases come from the high risk population (of the same disease). This is because the number of people at high risk is small”, see. In our world this reflects the question how to prevent transmission of ESBL-producing E. coli (ESBL-EC) or K. pneumoniae (ESBL-KP), or both. A new study may help to decide.

The problem: we want to prevent nosocomial transmission of ESBL-producing strains, yet these measures come at a certain price (work, money and motivation) and we strive for the best valued healthcare. Carriage with ESBL-EC is more prevalent than carriage with ESBL-KP and the same holds for the incidence of infections. On the other hand, ESBL-KP might be more susceptible to become CPE, whereas ESB-EC might be more susceptible to be upgraded into “mcr1-4 ESBL-EC knighthood”. ESBL-EC is now endemic in non-hospitalized subjects, which implies that screening at the gate would send about 3-8% of admitted patients straight into isolation (or other control measures).

The latter only helps if those measures prevent transmission, or better said if (enough) transmission occurs without such measures. That has been questioned in some settings, but may well occur in absence of proper infection control. For the pro’s and con’s see this.

There is a widely sensed feeling that ESBL-KP does a better job in spreading than ESBL-EC. The new study confirms this gut feeling. Some years ago we performed a large international study in 13 ICUs during 2 years, in which all patients (about 14,000) were screened for carriage with these bacteria on admission and twice weekly. After a baseline period (6 months), the WHO hand hygiene program and universal chlorhexidine body washing were implemented (6 months) and only in the last period (12 months) screening results for carriage were communicated to the floor (and carriers were placed in isolation if possible). The bad news: none of these interventions reduced the acquisition rate for carriage with ESBL-EC and ESBL-KP. But in the memory of a great Dutch philosopher: Every disadvantage has an advantage. Without effective intervention we had a huge observational study to quantify transmission of ESBL-EC and ESBL-KP in a setting where control measures were exactly the same for both species. We found that – in the setting tested – ESBL-KP was three times more transmissible than ESBL-EC.

If ESBL-EC enters the hospital on a daily basis, and if basic infection control measures effectively prevent uncontrolled spread, one might consider to focus on the (much) less prevalent introduction of ESBL-KP. They have the larger (estimated 3 times) potential to spread and may well have the larger potential to become CPE. Just a thought.