Excluded versus included patients in a randomized controlled trial of infections caused by carbapenem-resistant Gram-negative bacteria: relevance to external validity.

BACKGROUND: Population external validity is the extent to which an experimental study results can be generalized from a specific sample to a defined population. In order to apply the results of a study, we should be able to assess its population external validity. We performed an investigator-initiated randomized controlled trial (RCT) (AIDA study), which compared colistin-meropenem combination therapy to colistin monotherapy in the treatment of patients infected with carbapenem-resistant Gram-negative bacteria. In order to examine the study's population external validity and to substantiate the use of AIDA study results in clinical practice, we performed a concomitant observational trial. METHODS: The study was conducted between October 1st, 2013 and January 31st, 2017 (during the RCTs recruitment period) in Greece, Israel and Italy. Patients included in the observational arm of the study have fulfilled clinical and microbiological inclusion criteria but were excluded from the RCT due to receipt of colistin for > 96 h, refusal to participate, or prior inclusion in the RCT. Non-randomized cases were compared to randomized patients. The primary outcome was clinical failure at 14 days of infection onset. RESULTS: Analysis included 701 patients. Patients were infected mainly with Acinetobacter baumannii [78.2% (548/701)]. The most common reason for exclusion was refusal to participate [62% (183/295)]. Non-randomized and randomized patients were similar in most of the demographic and background parameters, though randomized patients showed minor differences towards a more severe infection. Combination therapy was less common in non-randomized patients [31.9% (53/166) vs. 51.2% (208/406), p = 0.000]. Randomized patients received longer treatment of colistin [13 days (IQR 10-16) vs. 8.5 days (IQR 0-15), p = 0.000]. Univariate analysis showed that non-randomized patients were more inclined to clinical failure on day 14 from infection onset [82% (242/295) vs. 75.5% (307/406), p = 0.042]. After adjusting for other variables, non-inclusion was not an independent risk factor for clinical failure at day 14. CONCLUSION: The similarity between the observational arm and RCT patients has strengthened our confidence in the population external validity of the AIDA trial. Adding an observational arm to intervention studies can help increase the population external validity and improve implementation of study results in clinical practice. TRIAL REGISTRATION: The trial was registered with ClinicalTrials.gov, number NCT01732250 on November 22, 2012.

Extended Infusion of ß-Lactams for Bloodstream Infection in Patients With Liver Cirrhosis: An Observational Multicenter Study.

BACKGROUND: We analyzed the impact of continuous/extended infusion (C/EI) vs intermittent infusion of piperacillin-tazobactam (TZP) and carbapenems on 30-day mortality of patients with liver cirrhosis and bloodstream infection (BSI). METHODS: The BICRHOME study was a prospective, multicenter study that enrolled 312 cirrhotic patients with BSI. In this secondary analysis, we selected patients receiving TZP or carbapenems as adequate empirical treatment. The 30-day mortality of patients receiving C/EI or intermittent infusion of TZP or carbapenems was assessed with Kaplan-Meier curves, Cox-regression model, and estimation of the average treatment effect (ATE) using propensity score matching. RESULTS: Overall, 119 patients received TZP or carbapenems as empirical treatment. Patients who received C/EI had a significantly lower mortality rate (16% vs 36%, P = .047). In a Cox-regression model, the administration of C/EI was associated with a significantly lower mortality (hazard ratio [HR], 0.41; 95% confidence interval [CI], 0.11-0.936; P = .04) when adjusted for severity of illness and an ATE of 25.6% reduction in 30-day mortality risk (95% CI, 18.9-32.3; P < .0001) estimated with propensity score matching. A significant reduction in 30-day mortality was also observed in the subgroups of patients with sepsis (HR, 0.21; 95% CI, 0.06-0.74), acute-on-chronic liver failure (HR, 0.29; 95% CI, 0.03-0.99), and a model for end-stage liver disease score ≥25 (HR, 0.26; 95% CI, 0.08-0.92). At competing risk analysis, C/EI of beta-lactams was associated with significantly higher rates of hospital discharge (subdistribution hazard [95% CI], 1.62 [1.06-2.47]). CONCLUSIONS: C/EI of beta-lactams in cirrhotic patients with BSI may improve outcomes and facilitate earlier discharge.

The Association Between Empirical Antibiotic Treatment and Mortality in Severe Infections Caused by Carbapenem-resistant Gram-negative Bacteria: A Prospective Study.

Background: Empirical colistin should be avoided. We aimed to evaluate the association between covering empirical antibiotics (EAT) and mortality for infections caused by carbapenem-resistant gram-negative bacteria (CRGNB). Methods: This was a secondary analysis of a randomized controlled trial, including adults with bloodstream infections, pneumonia, or urosepsis caused by CRGNB. All patients received EAT followed by covering targeted therapy. The exposure variable was covering EAT in the first 48 hours. The outcome was 28-day mortality. We adjusted the analyses by multivariable regression analysis and propensity score matching. Results: The study included 406 inpatients with severe CRGNB infections, mostly Acinetobacter baumannii (312/406 [77%]). Covering EAT was given to 209 (51.5%) patients, mostly colistin (n = 200). Patients receiving noncovering EAT were older, more frequently unconscious and dependent, carrying catheters, and mechanically ventilated with pneumonia. Mortality was 84 of 197 (42.6%) with noncovering vs 96 of 209 (45.9%) with covering EAT (P = .504). Covering EAT was not associated with survival in the adjusted analysis; rather, there was a weak association with mortality (odds ratio [OR], 1.37; 95% confidence interval [CI], 1.02-1.84). Results were similar for colistin monotherapy and colistin-carbapenem combination EAT. In the propensity score-matched cohort (n = 338) covering antibiotics were not significantly associated with mortality (OR, 1.42; 95% CI, .91-2.22). Similar results were obtained in an analysis of 14-day mortality. Conclusions: Empirical use of colistin before pathogen identification, with or without a carbapenem, was not associated with survival following severe infections caused by CRGNBs, mainly A. baumannii.

The Effectiveness and Safety of High-Dose Colistin: Prospective Cohort Study.

BACKGROUND: Optimizing colistin dosing should translate to improved patient outcomes. METHODS: We used data from 2 prospective cohort studies performed between 2006 and 2009 and between 2012 and 2015. In the latter period, a new policy of high-dose colistin (9 million international units [MIU] loading dose followed by 9 MIU daily for normal renal function) was introduced in 2 participating hospitals. We included adult inpatients with invasive infections caused by carbapenem-resistant gram-negative bacteria treated with colistin. Our primary exposure variable was colistin dose, dichotomized to high-dose vs other regimens. The primary outcome was 28-day mortality. We generated a propensity score for high-dose colistin and conducted propensity-adjusted multivariable and matched-cohort analyses for mortality. RESULTS: Of 529 consecutive patients fulfilling inclusion criteria, 144 were treated with high-dose colistin and 385 with lower-dose colistin regimens. The median daily dose in the high-dose group was 9 MIU (interquartile range [IQR], 9-9) vs 4 MIU (IQR, 3-6) with other regimens. There were 50 of 144 (34.7%) deaths with high-dose colistin vs 165 of 385 (42.9%) with low-dose colistin (P = .1). The propensity-adjusted odds ratio (OR) for mortality was 1.07 (95% confidence interval [CI], .63-1.83) for high-dose colistin. Similar results were obtained when using the study period as the exposure variable, in the subgroup of bacteremic patients (n = 207) and in the propensity-matched cohort (OR, 1.11 [95% CI, .67-1.82]). Nephrotoxicity (RIFLE injury or higher; OR, 2.12 [95% CI, 1.29-3.48]; n = 396) and seizures were significantly more common with high-dose colistin. CONCLUSIONS: In a large cohort, we found no association between high colistin dosing and all-cause mortality. High dosing was associated with more nephrotoxicity.