Initially inappropriate antibiotic therapy lengthens sepsis patients' stay by two days
Nearly one-third of patients with severe sepsis and septic shock attributable to Gram-negative organisms receive inappropriate antibiotic therapy at first, which prolongs their hospital stay by two days on average, a study found.
Researchers retrospectively reviewed patients with Gram-negative bacteremia and severe sepsis or septic shock at an urban academic hospital over a six-year period. The final cohort included 760 patients, 55% of whom had nosocomial infections. Escherichia coli was the most common pathogen (29.6% of patients); Pseudomonas species were isolated in 17.4% of patients. Most patients were treated in the ICU (78.9%); urine was the most common site of primary infection (51%), followed by the lung (38.8%). Researchers defined initially inappropriate antibiotic therapy as occurring when a patient either was not given an antibiotic within 24 hours of sepsis onset, or was treated with an antibiotic to which the pathogen was resistant. Results were published in the January Critical Care Medicine.
Thirty-one percent of patients received initially inappropriate antibiotic therapy. Those who did were more likely to have a nosocomial infection (69.3% vs. 48.7%; P=0.001); require chronic hemodialysis (14.7% vs. 7.1%; P=0.001); undergo mechanical ventilation (62.6% vs. 51.5%; P=0.005); or have diabetes (27.5% vs. 20.1%; P=0.034). These patients also had an unadjusted mean length of stay (after sepsis onset) of 11 days, compared to nine days in those treated appropriately (P=0.028 by log-rank test). They also had a greater hospital mortality (51.7% vs. 36.4%; P=0.001). Initially inappropriate therapy independently correlated with continued hospitalization in a Cox model that controlled for multiple cofounders like comorbid diseases and illness severity (adjusted hazard ratio, 1.19; 95% CI, 1.01 to 1.40; P=0.044).
Other studies have noted that initially inappropriate antibiotic treatment increases mortality for sepsis patients; this study adds to the analysis by focusing on a different end point—length of stay, the authors noted. By noting the economic costs of a longer length of stay, “physicians can make a strong case to hospital administrators and national policymakers” about the value of addressing initially inappropriate antibiotic therapy, and of treating patients aggressively, they wrote. Ways to minimize inappropriate therapy include creating antibiotic prescribing protocols, routinely discussing antibiotic management strategies in multidisciplinary meetings, and consulting earlier with infectious disease specialists, the authors wrote.
Intervention helps community hospital ICUs adopt evidence-based practices
A multi-pronged quality improvement program helped community hospitals adopt six evidence-based practices in the ICU, a study reported.
The targeted care practices (with process-of-care indicators) were prevention of ventilator-associated pneumonia (indicators: semirecumbent positioning, orotracheal intubation); prophylaxis for deep vein thrombosis (indicators: anticoagulant administration, antiembolic stockings if anticoagulants were contraindicated); daily spontaneous breathing trials (indicators: spontaneous breathing trial or extubation within the last 24 hours); preventing catheter-related bloodstream infections (indicators: completing 7-point checklist for sterile insertion, fulfilling all 7 checklist criteria, anatomical site of catheter insertion); early enteral feeding (indicator: initiation of enteral feeding within 48 hours of ICU admission); and decubitus ulcer prevention (indicator: completion of the Braden index at least twice a day).
In the cluster-randomized trial of 15 community hospital intensive care units (ICUs) in Ontario, Canada, researchers examined 9,269 admissions over 12 months, and 7,141 admissions over a separate decay-monitoring period of nine months. To improve the six care practices, they used a videoconference-based forum with auditing and feedback, expert-led educational sessions, and disbursement of algorithms. ICUs were randomized into two groups. Each group received the intervention, which targeted a new practice every four months, and also acted as a control for another group in which a different practice was targeted in the same time period. The six practices were paired to minimize the potential for quality improvement efforts that targeted one practice to influence process measures related to the other practice.
Adoption of targeted practices was greater in intervention ICUs than in controls (summary ratio of odds ratios [ORs], 2.79; 95% CI, 1.00 to 7.74). In intervention ICUs, care delivery improved the most for semirecumbent positioning to prevent ventilator-associated pneumonia (90.0% of patient-days in last month vs. 50.0% in first month; OR, 6.35; 95% CI, 1.85 to 21.79) and precautions to prevent catheter-related bloodstream infection (70% of patients receiving central lines vs. 10.6%; OR, 30.06; 95% CI, 11.00 to 82.17). Adoption of other practices didn't change much, but several already had high baseline adherence. Results were published Jan. 26 in the Journal of the American Medical Association.
The success of the intervention is noteworthy in that it occurred in community (not academic) ICUs, which admit the majority of critically ill patients, the authors noted. Further, the videoconferencing aspect of the intervention can help health care workers in geographically dispersed hospitals access resources that are usually restricted to academic facilities, they noted. Post hoc analyses indicate the intervention had the largest effect in ICUs with low baseline adherence to certain practices, suggesting similar initiatives should target these types of ICUs and practices, the authors wrote. Generally, large-scale quality improvement initiatives “should choose practices based on measured rather than reported care gaps, consider site-specific (vs. aggregated) needs assessments to determine target care practices, and conduct baseline audits to focus on poorly performing ICUs,” the authors concluded.
Program sustains reduction in MRSA infections after surgery
An intervention to reduce methicillin-resistant Staphylococcus aureus (MRSA) wound infections after cardiothoracic surgery dramatically reduced the infections and sustained the result for at least 30 months, a study found.
Researchers at a community-based medical-surgical hospital examined data from all patients who underwent cardiac surgery and required a median sternotomy incision from Jan. 1, 2004 through Jan. 31, 2010. They compared postoperative wound infection rates for the three years before the intervention and the three years after. The intervention included preoperative screening for MRSA colonization, administration of intravenous vancomycin for MRSA carriers, administration of intranasal mupirocin calcium ointment to all patients for five days starting the day before surgery, application of mupirocin to the chest tube sites of all patients at the time of removal, and—at the program outset—screening of cardiothoracic staff for nasal carriage of MRSA with decolonization if necessary.
Following the intervention, postoperative MRSA infections fell by 93% (32 infections per 2,767 cases during baseline vs. 2 infections per 2,496 cases during intervention; relative risk, 0.069; P<0.001). There was no change in the number of MRSA infections after noncardiac surgery during the intervention period. Overall wound infection rates fell from 2.1% to 0.8% (59 infections per 2,769 cases vs. 20 infections per 2,496 cases; P<0.001). The number needed to treat to prevent one postoperative MRSA wound infection was 93. Results were published in the Jan. 10 Archives of Internal Medicine.
Study limitations include the use of data from a historical control group rather than from a randomized, double-blind, placebo-controlled trial, and the fact that effects from vancomycin vs. mupirocin can't be distinguished since both were introduced at the same time. While it is possible that factors unrelated to the intervention may have affected the observed MRSA rates, hospital-wide rates of nosocomial MRSA surgical site infections didn't change during the intervention period, suggesting other effects weren't operative, the authors said. The intervention program “resulted in a near-complete and sustained elimination of MRSA wound infections after cardiac surgery,” the authors concluded.
Pneumonia guideline adherence increased mortality among intensive care patients
A performance-improvement initiative found that compliance with guidelines may actually have increased mortality from pneumonia among intensive care patients.
Four academic medical centers in the U.S. instituted a program to improve clinicians' compliance with guidelines for management of hospital-acquired, ventilator-associated and health care-associated pneumonias that were issued in 2005 by the American Thoracic Society and the Infectious Diseases Society of America. The study included 303 ICU patients who had risk factors for multidrug-resistant (MDR) pneumonia, most of whom also required mechanical ventilation.
Researchers assessed mortality rates among these patients and compared them with compliance with the ATS/IDSA guidelines, which call for treating patients at risk for MDR with an antipseudomonal cephalosporin, carbapenem, or β-lactam and β-lactamase inhibitor; an aminoglycoside or antipseudomonal fluoroquinolone; and linezolid or vancomycin. The study found that 129 patients had received treatment in compliance with these criteria, but 174 had not. The most common deviation from the guidelines was failure to use a secondary anti-Gram-negative drug (154 patients), followed by failure to use a primary anti-Gram-negative drug (24 patients) and failure to use an anti-MRSA drug (24 patients).
Patients who received guideline-compliant therapy had a 28-day mortality rate of 34% compared to 20% in the non-compliant group. The difference persisted after adjustment for severity of illness, leading study authors to conclude that compliance with the guidelines was associated with increased mortality. They therefore recommended that the guidelines be reassessed, and that a comparison be conducted of single versus dual Gram-negative coverage. One potential explanation for the findings is antibiotic-specific toxic effects, the authors said.
A comment, published with the article in The Lancet Infectious Diseases on Jan. 20, agreed that toxicity was a possible cause, but the author also pointed out several concerns regarding the study. The study authors failed to comment on significant mortality differences among patients depending on which pathogen they had or to measure the timeliness of treatment. The commenter also critiqued the study authors' disregard of treatment de-escalation in classification of compliance. “De-escalation is what matters,” the commenter wrote, although he also acknowledged that the current guidelines may be insensitive to local variations in pathogen prevalence and the need for variations for elderly and severely disabled patients.