Speed of sepsis care, early feeding for mild pancreatitis, and more

Summaries from ACP Hospitalist Weekly.


Faster antibiotics, full sepsis bundle associated with lower inpatient mortality

Rapid completion of the three-hour sepsis bundle was associated with lower inpatient mortality, according to a recent study of hospitals in New York State.

Researchers used data reported to the New York State Department of Health from April 1, 2014, to June 30, 2016. The study included 49,331 patients treated in the emergency departments of 149 hospitals for sepsis or septic shock. Patients had a three-hour sepsis bundle initiated within six hours of arrival to the ED and completed within 12 hours. Overall, 82.5% got the recommended bundle, which included blood cultures, broad-spectrum antibiotics, and lactate measurement, within three hours. Results were published by the New England Journal of Medicine on May 21 and appeared in the June 8 issue.

The median time to completion of the bundle was 1.30 hours (interquartile range, 0.65 to 2.35 hours). Antibiotics were administered at a mean of 0.95 hour (interquartile range, 0.35 to 1.95 hours) and the median time to completion of the fluid bolus was 2.56 hours (interquartile range, 1.33 to 4.20 hours). Among patients who received the bundle within 12 hours, inpatient mortality risk increased with longer time to completion of the bundle (odds ratio [OR], 1.04 per hour; 95% CI, 1.02 to 1.05) and time to antibiotics (OR, 1.04 per hour; 95% CI, 1.03 to 1.06). Longer time to completion of a fluid bolus was not associated with mortality.

The findings support an association between time to treatment and outcomes among patients with sepsis or septic shock treated in the ED, the study authors said. They noted that time to treatment varied widely across the studied hospitals, with the risk-adjusted and reliability-adjusted rates of meeting the three-hour goal varying from 53% to 97% (median, 83%; interquartile range, 75% to 88%). Overall, however, adherence was better than in comparable quality improvement programs for stroke treatment in New York, they noted. Adherence was highest in smaller nonteaching hospitals, the study found.

The observed lack of association between timely fluid delivery and outcomes “should not be interpreted as evidence in favor of abandoning early fluid resuscitation,” the authors said. This measure would be prone to confounding by indication, as sicker patients receive fluids sooner and are more likely to die. Randomized clinical trials would be needed to determine how rapid fluid delivery affects outcomes, according to the authors. Other limitations of the study include lack of data about appropriateness of antibiotics and inclusion of hospitals from only one geographic region.

An accompanying editorial analyzed the New York regulations about sepsis care that led to the collection of the study data. Recent data shows that compliance with sepsis protocols has increased in the state, but the editorialists still have concerns about the potential negative effects of regulating practice in this way, from antibiotic overuse to failure to apply the latest medical evidence. “It is also essential to have a public debate regarding whether it is best to enforce clinical practice guidelines through regulation rather than allowing the medical community to decide whether to adopt them,” they wrote.

Review finds benefit to early feeding for mild to moderate pancreatitis

Early feeding of patients hospitalized with acute pancreatitis does not appear to increase adverse events and may improve outcomes for some patients, according to a recent systematic review.

The review included 11 randomized trials with 948 patients, all hospitalized with acute pancreatitis. The trials all compared early versus delayed feeding (less or more than 48 hours after hospitalization). A variety of feeding routes were employed, including oral (four studies), nasogastric (two studies), nasojejunal (four studies), and oral or nasoenteric (one study). Most of the outcomes studied were heterogeneous, and this prevented researchers from doing a meta-analysis. Their systematic review was published online May 16 by Annals of Internal Medicine and appeared in the June 20 issue.

Photo by Thinkstock
Photo by Thinkstock

Seven of the reviewed trials (three with low risk of bias) included patients with mild to moderate pancreatitis, and in four of the trials early feeding was associated with reduced length of stay. Three of the trials found lower rates of gastrointestinal symptoms (feeding intolerance, nausea, vomiting, pain) with early feeding. For patients with predicted severe pancreatitis, there were four trials; one had low risk of bias and it showed no difference in length of stay, mortality, symptoms, or necrotizing pancreatitis between early and late feeding. None of the trials showed an increase in adverse events with early feeding.

When to feed acute pancreatitis patients has long been controversial, according to the study authors, who concluded that their review, based on limited data, suggests that early feeding “does not seem to increase adverse events, and, for patients with mild to moderate pancreatitis, may reduce length of hospital stay.” This conclusion supports the consideration of early feeding in such patients, in contrast with 2013 U.S. guidelines that recommended waiting to start feeding until inflammatory markers are improving or abdominal pain has resolved, they noted.

“Despite these encouraging results, questions persist about feeding in patients with pancreatitis,” the authors added. Issues mitigating the certainty of their findings include lack of data on patients with severe pancreatitis, the variation in the route and timing of early feeding, and uncertainty about treatment of patients who do not tolerate an initial oral diet.

An accompanying editorial urged even more caution when applying the review's results to practice. The editorialists' concerns included that three of the studied trials were abstract-only, that three permitted use of parenteral nutrition, and that patients were insufficiently stratified by disease severity and not stratified by feeding method. The editorialists noted that the 2016 nutrition support guidelines from the Society of Critical Care Medicine and American Society for Parenteral and Enteral Nutrition “set the stage for nutrition support clinicians to improve specialized nutritional support.”

Program for advanced recovery after surgery associated with shorter stays, better outcomes

A program focused on enhanced recovery after surgery (ERAS) was associated with decreased length of stay and postoperative complication rates, according to a recent study.

Researchers conducted a pre-post difference-in-differences study before and after implementation of an ERAS program at 20 medical centers in the Kaiser Permanente Northern California health care system. Target populations of patients undergoing elective colorectal resection and patients undergoing emergency hip fracture repair were compared with contemporaneous surgical comparator groups undergoing elective gastrointestinal surgery and emergency orthopedic surgery. ERAS implementation started in February and March 2014 and was completed by the end of 2014. The main outcome measure was hospital length of stay, with hospital mortality, home discharge, 30-day readmission rates, and complication rates as secondary outcome measures.

The ERAS program focused on optimizing nutrition, mobility, pain management, and patient engagement. For nutrition, prolonged postsurgical fasting was reduced by providing patients with a high-carbohydrate beverage two to four hours before surgery, solids eight to 12 hours before surgery, or both; after surgery, nutrition was provided within 12 hours. To improve mobility, patients who were ambulatory were encouraged to start walking within 12 hours of surgery completion and to walk at least 21 feet in the first three days after surgery. Opioid-sparing multimodal analgesia was used for pain management, including IV acetaminophen, NSAIDs, and perioperative IV lidocaine for colorectal surgery or peripheral nerve blocks for hip fracture repair. For patient engagement, patients received a calendar with infographics to detail what they could expect from the night before surgery through discharge; patient education was also addressed through an informational video series. Study results were published online May 10 by JAMA Surgery and appeared in the July issue.

Overall, 3,768 patients having elective colorectal resection and 5,002 patients having emergency hip repair were included in the study. Mean age was 62.7 years and 79.5 years, respectively, and 48.1% and 31.7% of patients were men. The comparator surgical group included 5,556 patients having elective gastrointestinal surgery and 1,523 patients having emergency orthopedic surgery. Significant changes in most process metrics were seen in the ERAS target populations, and while process metrics changes were also seen in the comparator groups between the pre- and postintervention phases, these differences were more modest.

ERAS implementation was associated with increased rates of early ambulation in patients undergoing colorectal resection and hip fracture repair (rate ratios, 1.99 [95% CI, 1.80 to 2.21] and 4.44 [95% CI, 3.19 to 6.21], respectively; P<0.001 for both comparisons), as well as with decreased opioid use (rate ratios, 0.79 [95% CI, 0.71 to 0.89] and 0.73 [95% CI, 0.63 to 0.85]; P<0.001 for both comparisons). Lower hospital mortality rates (rate ratio, 0.17; 95% CI, 0.03 to 0.86; P=0.03) and lower rates of major complications (rate ratio, 0.28; 95% CI, 0.12 to 0.68; P=0.005) were associated with ERAS implementation in patients undergoing colorectal resection. In patients undergoing hip fracture repair, ERAS implementation was associated with higher rates of home discharge (rate ratio, 1.24; 95% CI, 1.06 to 1.44; P=0.007). ERAS implementation was not associated with differences in the rate of 30-day readmissions in patients receiving either type of surgery.

The study authors noted that their evaluation was not a randomized controlled trial, that the comparator groups and target populations had different surgical procedures, and that they looked only at short-term outcomes, among other limitations. However, they concluded that the ERAS program tested in their study resulted in significant changes in process-of-care metrics and was associated with significant absolute and relative improvements in both length of stay and rates of surgical complications.

The authors of an accompanying invited commentary praised the study's design and implementation and said its findings are relevant to clinical practice, research, and policy and make an important contribution to population health. “The investigators have robustly taken implementation science to the next level, thus showing that thoughtfully planned quality-improvement endeavors that are integrated with robust research evaluation measures can positively affect our surgical patients,” the commentary authors wrote.

Clinical decision tool may predict benefit of intra-arterial treatment after acute ischemic stroke

A clinical decision tool may help determine which acute ischemic stroke patients may benefit from intra-arterial treatment.

Researchers used data from two randomized controlled trials to determine whether a clinical decision tool based on several characteristics could improve selection of patients for intra-arterial treatment after acute ischemic stroke by predicting individual potential benefit. The Multicenter Randomised Clinical Trial of Endovascular Treatment for Acute Ischaemic Stroke (MR CLEAN), which involved 500 patients at 16 hospitals in the Netherlands, served as the derivation cohort, and data from a subgroup of 260 patients who had an occlusion on CT angiography in the Interventional Management of Stroke III (IMS III) trial, which involved 58 hospitals in the U.S., Canada, Australia, and Europe, served as the validation cohort.

Modified Rankin Scale (mRS) score at 90 days after stroke was the primary outcome measure. The researchers used an ordinal logistic regression model to examine the effect on benefit of treatment of the following variables: age, baseline National Institutes of Health Stroke Scale (NIHSS) score, systolic blood pressure, treatment with IV tissue plasminogen activator, history of ischemic stroke, atrial fibrillation, diabetes mellitus, prestroke mRS score, Alberta Stroke Program Early Computed Tomography Score (ASPECTS), occlusion location, collateral score, time to treatment, and corresponding interactions with treatment. The researchers defined treatment benefit as the difference between the predicted probability of good functional outcome with and without intra-arterial treatment. The study results were published online May 3 by BMJ.

The variables included in the final multivariable model were age, baseline NIHSS score, systolic blood pressure, treatment with IV tissue plasminogen activator, history of ischemic stroke, diabetes mellitus, prestroke mRS score, ASPECTS, occlusion location, collateral score, and time from stroke onset to groin puncture. An externally validated C statistic of 0.69 (95% CI, 0.64 to 0.73) was found for the ordinal model, and an externally validated C statistic of 0.73 (95% CI, 0.67 to 0.79) was found for the prediction of good functional outcome. The probability of an mRS score of 0 to 2 was 11.8% higher with intra-arterial treatment than without, and this varied from −2.3% to 24.3% among individual patients in the combined derivation and validation cohort. Individual predictions of benefit varied substantially in patients who had poor collateral scores (defined as a low score on a four-point scale grading the collateral flow of the occluded territory on imaging) and low ASPECTS (indicating early ischemic changes on noncontrast CT), subgroups in which no treatment effect was found overall.

The authors noted that the model had only modest discriminative ability in the external validation cohort and that the sample size was relatively small, among other limitations. However, they concluded that the model may provide clinicians with guidance when weighing intra-arterial treatment decisions for patients presenting with acute ischemic stroke. They stressed that their study indicates that intra-arterial treatment should not be withheld based on one characteristic only, since individual patients who are in subgroups often considered to derive no benefit from treatment may in fact benefit substantially if they have other favorable characteristics.

“This emphasises the importance of making personalised treatment decisions, instead of using average treatment effects, and shows the need for combining multiple clinical and radiological baseline characteristics instead of withholding treatment based on one characteristic,” the researchers wrote. They called for additional analyses with larger sample sizes to help refine their results and improve the model's validity. Their model is available online.

An editorial that accompanied the study emphasized the difficulty of making treatment decisions in patients presenting with acute ischemic stroke. “Even when patients are first seen at a comprehensive stroke centre, proceeding with endovascular thrombectomy incurs substantial opportunity costs because resources are potentially diverted from other patients,” the editorialist wrote. The editorialist agreed that additional future studies are necessary and reiterated the limitations of the current study but wrote that it “will help clinicians make decisions about endovascular thrombectomy in individual patients and, by providing a quantitative estimate of the potential benefit (or harm), help patients and their families make informed decisions at a critical time.”

Ventilator bundle reduced adverse events, including pneumonia

A bundle of interventions reduced ventilator-associated events in ICUs in Maryland and Pennsylvania, a recent study found.

The study was conducted in 56 ICUs at 38 hospitals in the two states from October 2012 to March 2015. The ICUs established quality improvement teams and implemented a multifaceted intervention that included head-of-bed elevation, use of subglottic secretion drainage endotracheal tubes, oral care, chlorhexidine mouth care, and daily spontaneous awakening and breathing trials—all interventions recommended by the Society for Healthcare Epidemiology of America. Results were published online by Critical Care Medicine on April 26 and appeared in the July issue.

Photo by Thinkstock
Photo by Thinkstock

Using 69,417 patient-days and 1,022 unit-months of data, researchers found that compliance with all of the interventions improved over the study period. According to data reported to the CDC National Healthcare Safety Network, the ventilator-associated event rate significantly decreased from 7.34 per 1,000 ventilator-days to 4.58 per 1,000 ventilator-days 24 months after implementation (P=0.007). Infection-related ventilator complications also decreased from 3.15 to 1.56 per 1,000 ventilator-days (P=0.018) while possible and probable ventilator-associated pneumonia rates decreased from 1.41 to 0.31 per 1,000 ventilator-days (P=0.012).

“Our study is the largest to date affirming that best practices can prevent ventilator-associated events,” the study authors concluded. They noted that the preventability of ventilator-associated events has previously been questioned. Because the interventions were bundled, the study wasn't able to determine which had the most impact, but the authors said that compliance with head-of-bed elevation and subglottic secretion drainage endotracheal tubes did not change substantially during the intervention (with the former being consistently high and the latter low), so it's more likely other interventions were responsible for the results.

The authors said that analyses are underway to evaluate the relative importance of each intervention in the bundle. The researchers also continue to modify the bundle based on emerging evidence and have identified early mobility and low tidal volume interventions as promising candidates to add to the bundle. The study was limited by its nonrandomized design and lack of a control group.

Benefits of endovascular therapy continue out to two years after stroke

Patients who received endovascular treatment for acute ischemic stroke had better functional outcomes than control patients at two years after treatment, just as they had at 90 days, according to a follow-up analysis of a randomized trial.

Researchers assessed clinical outcomes two years after 500 patients were randomized to either endovascular or conventional treatment as part of the MR CLEAN trial. The study had information on functional status for 391 patients and information on death for 459 patients. The primary outcome was score on the modified Rankin scale, and secondary outcomes included mortality and quality of life. Results were published in the April 6 New England Journal of Medicine.

The modified Rankin scale scores at two years were significantly better in the endovascular therapy group (adjusted common odds ratio, 1.68; 95% CI, 1.15 to 2.45; P=0.007). The difference between groups was very similar to that at 90 days, when the odds ratio was 1.67. At two years, the mean quality of life score was higher in the endovascular patients as well (0.48 vs. 0.38; mean difference, 0.10; 95% CI, 0.03 to 0.16; P=0.006). The endovascular group also had a slightly lower mortality rate, although the difference was not statistically significant (26.0% vs. 31.0%; adjusted hazard ratio, 0.9; 95% CI, 0.6 to 1.2; P=0.46). The mortality difference was a change from the 90-day results, at which time the two groups had similar death rates, the study authors noted.

Overall, they concluded that the beneficial effect of endovascular treatment on functional outcomes in patients with acute ischemic stroke was similar at two years to what it was at 90 days according to the original trial. The results are consistent with other smaller studies which looked at outcomes a year after a stroke, the authors said. One limitation of this study was that patients with missing outcome data had worse clinical characteristics and 90-day outcomes and were more likely to be in the control group, which could bias the result. However, a sensitivity analysis which imputed missing outcomes found similar results to the main analysis, the authors said.

Cefazolin associated with lower mortality than other beta-lactams in MSSA infections, study finds

Patients with methicillin-susceptible Staphylococcus aureus (MSSA) infections and bacteremia who received cefazolin had lower risk of mortality than those who took nafcillin or oxacillin, a study found.

The retrospective study included 3,167 patients admitted to 119 Veterans Administration hospitals between 2003 and 2010. They all had a blood culture positive for MSSA and received definitive therapy with cefazolin, nafcillin, or oxacillin. Outcomes were 30- and 90-day all-cause mortality and recurrent MSSA bacteremia within a year. Results were published online by Clinical Infectious Diseases on March 31 and appeared in the July 1 issue.

Thirty-seven percent of the studied patients were treated with cefazolin, and they had a significantly lower mortality rate than the other patients: 30-day mortality hazard ratio, 0.63 (95% CI, 0.51 to 0.78) and 90-day mortality hazard ratio, 0.77 (95% CI, 0.66 to 0.90), both after adjustment. The patients had similar rates of recurrent infection, regardless of which of the drugs they received.

The results support current guidelines listing cefazolin as an alternative therapy to treat severe, invasive MSSA infections, according to the authors. They did warn that, despite efforts to adjust for a number of factors, the study's results could have been confounded by indication. They concluded that clinicians “may want to consider prescribing cefazolin for patients with MSSA infections complicated by bacteremia not involving the central nervous system” but noted that randomized trials are needed.

An accompanying editorial expressed even greater concern that confounding may have been responsible for “the striking and biologically somewhat implausible greater efficacy of cefazolin.” The editorialist expressed particular concern about using cefazolin in MSSA with the type A Bla-cefazoln inoculum effect phenotype and suggested that cefazolin is still most appropriate for use in uncomplicated MSSA bacteremia or the consolidation/completion phase of treatment for complicated infections.