Antimicrobial stewardship and infection control: October 2018

“Decontaminating” the oropharynx and/or the GI tract does not reduce the incidence of MDR bloodstream infection (MDR-BSI) in patients on mechanical ventilation. This month JAMA reports an RTC conducted in 13 European ICUs with an ESBL prevalence of >5% among bloodstream infections in ventilated patients. The interventions were daily chlorhexidine body washes and a hand hygiene campaign for staff (standard care), oropharyngeal decontamination (cleaning with a colistin-tobramycin-nystatin mouthpaste), or GI tract decontamination (the mouthpaste plus an ingested suspension of the same antibiotics). The trial used an 18-month crossover design, with each ICU using each intervention for 6 months in a randomly assigned order. The primary outcome was the incidence of MDR-BSI; the secondary outcome was 28-day mortality.

The 8665 patients enrolled in the study acquired a total 154 MDR-BSIs. Compared to the baseline 6-14 month periods recorded before each ICU’s enrollment, the incidence of MDR-BSI was not meaningfully reduced with standard care (0.3%; 95% CI -0.6% to 1.1%), oropharyngeal decontamination (0.6%; 95% CI -0.2% to 1.4%), or GI tract decontamination (0.8%; 95% CI 0.1% to 1.6%). Crude rates of mortality were 32%, 33%, and 32% respectively, and adjusted odds ratios for mortality were nonsignificant for each intervention.

Take home: antibiotic decontamination of the mouth and gut to reduce MDR-BSI doesn’t work. In fact, not only can the oropharyngeal part be quite painful for patients, the antibiotic exposure probably selects for gastrointestinal MDR colonization in the long run. So we should stop doing it. 30347072

Antimicrobial lock therapy decreases the incidence of CLABSI in patients with long-term dialysis catheters. This multinational European RTC randomized 270 patients with ESRD on hemodialysis via a long-term central venous catheter to have CVC locks with either heparin (control group) or an antimicrobial solution containing trimethoprim, ethanol, and Ca2+-EDTA. The patients were followed for a mean 5.3 months, and the primary outcome of interest was a composite of CLABSI or need for intra-catheter thrombolytic therapy, adjudicated by a blinded committee. Compared to heparin locks, antimicrobial locks were associated with a 78% reduction in the incidence of CLABSI (0.09 versus 0.41 infections per 1000 catheter days; p = 0.03) but a 330% increase in the need for thrombolytic therapy (12% versus 40%; p < 0.001). Rates of catheter removal for any reason were similar between the heparin and antibiotic lock groups (13% versus 11%).

More dialysis line thromboses for fewer CLABSIs sounds like a good trade to me, especially when the line removal rate is unchanged. But hey, I’m an ID doc, not a nephrologist. 30281074

Restricting quinolone use improves quinolone susceptibility rates in the community. The authors report their 10-year experience with restricting the prescription of quinolones at a large academic medical center. They had noticed large reductions in quinolone susceptibility beginning in 1998, and in 2005 began a policy of requiring preauthorization for the use of quinolones. The authors compared the rate of change in quinolone susceptibilities between the preintervention period (1998-2005) and the postintervention period (2006-2016). First, they report that their intervention was successful, achieving a nearly two-thirds reduction in quinolone use, measured as quinolone-days prescribed per 1,000 patient days. Second, they report small but statistically significant improvements in quinolone susceptibility for Acinetobacter spp. (RR 1.04), E. cloacae (RR 1.03), and P. aeruginosa (RR 1.01). Susceptibility rates remained stable for K.pneumoniae, and in E. coli they continued to decline (RR 0.98).

The authors conclude that a prescription restriction program was able to reverse or stop the progression of quinolone resistance for most gram-negative organisms, with the notable exception of E.coli. The benefits seen here may seem small, but given that 2006-2016 was a time for dramatic decreases in quinolone susceptibility at my local hospitals, I thought the results were impressive. 30296959

Urine cultures predict blood culture susceptibilities in bacteremic UTI and can be used for early targeting of antibiotic therapy. The authors reviewed the cases of all adults admitted to their academic medical center over a 5-year period with a diagnosis of bacteremic UTI and the same organism in urine and blood. They compared the susceptibility profiles of the urine and blood isolates and calculated the diagnostic accuracy of the urine culture sensitivities for the blood culture sensitivities.

Among the 428 patients included in the study, the most common infecting organisms were E.coli (69%), Klebsiella (14%), and Enterobacter (5%). In total, 98.5% of the individual antimicrobial susceptibility results were concordant between paired blood and urine cultures, and 89% of the pairs had identical susceptibility patterns. When the susceptibilities varied between the blood and urine cultures, the urine culture was the more resistant isolate about two thirds of the time. As a predictor of blood culture isolate’s antimicrobial susceptibilities, the urine culture was 99% sensitive and 98% specific, with a PPV of 99% and a NPV of 98%. Urine culture susceptibilities reported about a day earlier than blood culture susceptibilities.

In conclusion, narrowing antibiotic coverage in bacteremic UTI based on urine culture sensitivities when the blood culture isolate is still in process is safe and should save patients about a day of empiric broad-spectrum therapy. 30323048