Recent Research

C. diff risk, discharge bundles, transfusion strategies, and more.


Ward-level antibiotic use predicts Clostridium difficile risk

Inpatients' risk of developing Clostridium difficile is related to the rate of antibiotic prescribing on their wards, regardless of whether they themselves took an antibiotic, a recent study found.

The retrospective cohort study looked at 46 months of adult inpatient care (from 2010 to 2014) on 16 wards (including 5 ICUs) at a single teaching hospital. The authors considered patient-level risk factors for C. difficile (including age, comorbidities, hospitalization history, and antibiotic exposure) and ward-level risk factors (including antibiotic use, hand hygiene, and mean patient age). Results were published in the April JAMA Internal Medicine.

Of the more than 34,000 studied patients, 255 developed C. difficile, an incidence rate of 5.95 per 10,000 patient-days. Among the wards, antibiotic use varied from 21.7 days of therapy per 100 patient-days up to 56.4 per 100. The researchers found that each 10% increase in antibiotic use on a ward was associated with a 2.1 per 10,000 patient-day increase in C. difficile (P<0.001). The association was the same regardless of whether an individual patient had recent antibiotic exposure, and it persisted after multilevel, multivariate adjustment for differences in patient risk factors among wards (relative risk, 1.34 per 10% increase in antibiotic use; 95% CI, 1.16 to 1.57).

Ward-level prescribing is associated with a clinically relevant increase in C. difficile risk, the researchers concluded. The finding may be due to increased colonization and shedding of C. difficile in wards with more antibiotic use. Antibiotic use even in patients who never show symptoms of C. difficile could increase overall rates because they become carriers of the infection, a phenomenon the study authors described as “an inverse of herd immunity seen with vaccines.”

The results support antibiotic stewardship efforts aimed at reducing overall antibiotic exposure, not just use of specific high-risk agents, and promoting hand hygiene with soap and water before and after seeing patients on antibiotics, the authors concluded. They noted that the study was limited by its observational and single-center design and called for future research to quantify the mechanisms of the observed association.

Future studies should also investigate the role of C. difficile colonization at hospital admission, which may have confounded the results, an accompanying commentary said. If confirmed, the association between C. difficile and ward-level prescribing might also be relevant to the spread of other multidrug-resistant organisms, including carbapenem-resistant Enterobacteriaceae and vancomycin-resistant enterococci, the commenter added.

Another study published in the Feb. 26 New England Journal of Medicine calculated the overall burden of C. difficile in the U.S. in 2011. Surveillance in 10 geographic areas identified more than 15,000 cases; 65.8% were health care related, but only 24.2% had onset during hospitalization. That extrapolates out to about 453,000 infections, 83,000 recurrent cases, and 29,300 deaths from C. difficile in the U.S. in 2011. These estimates are higher than previous ones (possibly due to the uptake of molecular testing for C. difficile by hospital laboratories during 2011) and show that prevention of C. difficile should extend outside the hospital and focus on appropriate antibiotic use, the authors said.

Bundle of interventions associated with lower mortality, more discharges to home

A stroke bundle that included stroke unit admission, swallow screen, aspirin, and a brain scan improved outcomes in ischemic stroke patients, a Scottish study found.

The national analysis included all 36,055 patients admitted with a stroke to acute hospitals in Scotland between 2005 and 2011. The researchers assessed whether the patients' care met 4 standards: entry into a stroke unit on the day of admission (day 0) or on the following day (day 1), aspirin on day 0 or 1, a brain scan on day 0, and a swallow screen on day 0. The results were adjusted based on the Six Simple Variable model, admission year, and hospital-level random effects. The study was published in the April Stroke.

Receiving care that met more of the standards was associated with lower 30-day all-cause mortality rate, with an adjusted odds ratio (OR) for death of 2.95 (95% CI, 1.91 to 4.55) when no standards were met compared to 4 standards and an OR of 1.21 with 3 standards versus 4 (95% CI, 1.09 to 1.34). The trend continued at 6 months. Three of the standards—stroke unit admission, swallow screen, and aspirin use—were independently associated with lower 30-day mortality, and aspirin and stroke unit admission were associated with lower mortality at 6 months. Patients who received the full bundle were more likely to be living at home at 6 months (OR with 3 vs. 4 standards, 0.91; 95% CI, 0.85 to 0.98), with the chance of home discharge increasing with each component provided.

Being admitted to a stroke unit significantly increased the chances that the other standards would be met, the study also found. Only 58.7% of the patients were admitted to a unit on day 0 or 1. Aspirin was the most commonly achieved standard (84.4%), followed by the swallow screen (77.4%) and brain scan (77.1%). Compliance with the standards increased over the course of the study.

The researchers concluded that achieving more components of this simple bundle at admission was associated with improved outcomes in stroke patients for up to 6 months. However, the retrospective nature of the study means that other factors, such as perceived futility of some interventions for sicker patients, could have confounded the results, despite adjustment. Conversely, the lack of independent benefit from a brain scan may relate to patients with more severe strokes being imaged more rapidly. The results are worthy of further exploration, the authors said.

Restrictive vs. liberal transfusion strategies may not matter after cardiac surgery

Restrictive transfusion thresholds after cardiac surgery didn't improve outcomes compared to liberal ones, a study found.

The Transfusion Indication Threshold Reduction (TITRe2), a multicenter, parallel-group trial, recruited patients age 16 and older undergoing nonemergency cardiac surgery at 17 British centers. Patients with a postoperative hemoglobin level of less than 9 g/dL were randomly assigned to a restrictive transfusion threshold of hemoglobin level <7.5 g/dL (n=1,000) or a liberal transfusion threshold of <9 g/dL (n=1,003). In the liberal-threshold group, patients received 1 unit of red cells immediately after randomization and received another unit if the hemoglobin level remained or dropped below 9 g/dL again. In the restrictive-threshold group, patients received 1 unit if the hemoglobin level dropped below 7.5 g/dL and another unit if the level remained or dropped below 7.5 g/dL again.

The primary outcome was a composite of a serious infection (sepsis or a wound infection) or an ischemic event such as a permanent stroke, myocardial infarction, infarction of the gut, or acute kidney injury within 3 months. Health care costs, excluding the index surgery, were estimated from the day of surgery to 3 months afterward. Results appeared in the March 12 New England Journal of Medicine.

Transfusion rates after randomization were 53.4% in the restrictive strategy group and 92.2% in the liberal strategy group. Primary outcomes occurred in 35.1% of the patients in the restrictive-threshold group and 33.0% of the patients in the liberal-threshold group (odds ratio, 1.11; 95% CI, 0.91 to 1.34; P=0.30). There were more deaths in the restrictive-threshold group than in the liberal-threshold group (4.2% vs. 2.6%; hazard ratio=1.64; 95% CI, 1.00 to 2.67; P=0.045). The researchers noted that serious postoperative complications, excluding primary-outcome events, occurred in 35.7% of participants in the restrictive-threshold group and 34.2% of participants in the liberal-threshold group. Total costs did not differ significantly between the groups.

The study creates new uncertainty about using a restrictive threshold for transfusion after cardiac surgery, the researchers wrote. The TITRe2 trial differs from previous observational studies of transfusion thresholds in that all the participants had cardiovascular disease and a substantial proportion of participants were likely to have been dependent on oxygen supplementation immediately after surgery. This group may have been at the limits of their cardiovascular reserve and may benefit from higher hemoglobin levels, the authors speculated. “Patients with cardiovascular disease may represent a specific high-risk group for which more liberal transfusion thresholds are to be recommended,” they wrote.

The results are surprising, said an accompanying editorial. “Cardiac surgery departments should review the findings of the TITRe2 trial and decide which threshold they deem to be most appropriate for transfusion. Protocols should be developed to minimize deviation from the agreed-upon approach, and feedback should be provided to hold operators accountable to the institution's standard of care,” the editorialist wrote.

Continuous PPI use associated with C. difficile recurrences

Recurrences of Clostridium difficile were significantly more common among patients who continuously took proton-pump inhibitors (PPIs) than among those who didn't, a recent study found.

The retrospective cohort study included 754 patients at 2 Canadian hospitals who developed health care-associated C. difficile infection in 2010 to 2013 and survived for a minimum of 15 days after their initial episode. Continuous PPI use was common (60.7% of patients), and a record review showed that only 47.1% of these prescriptions were evidence-based. The study was published by JAMA Internal Medicine on March 2.

The patients taking PPIs had a 28.8% risk of recurrent C. difficile, compared to 20.6% in nonusers (P=0.007), and they were more likely to die within 15 to 90 days of the initial episode (10.3% vs. 4.7%; P=0.007). According to a multivariable Cox proportional hazards model, the hazard ratio of recurrence associated with PPI use was 1.5 (95% CI, 1.1 to 2.0), which was equal to that for age older than 75 years (1.5 [95% CI, 1.1 to 2.0]) and greater than the hazard ratios associated with antibiotic reexposure or longer length of stay (1.3 [95% CI, 0.9 to 1.7] and 1.003 per day [95% CI, 1.002 to 1.004], respectively).

The authors noted that PPIs were discontinued in only 3 of the patients with C. difficile. They suggested that clinicians should almost certainly stop non-evidence-based PPI prescriptions at the time of C. difficile diagnosis. “The cost of PPI discontinuation is negligible, and tapering the PPI dosage before discontinuation may help to curb any complications such as symptoms of rebound hyperacidity,” they wrote.

Although previous research has looked at the association between PPI use and C. difficile, few studies have analyzed the drugs' relationship with disease recurrences. Preventing C. difficile recurrences by avoiding PPIs would reduce costs for the health care system, even as new treatments for C. difficile are being developed, the authors said.

Meta-analysis looks at heparin safety, efficacy in sepsis

Heparin may be associated with decreased mortality in sepsis and septic shock, according to a recent study.

Researchers performed a systematic review and meta-analysis of randomized, controlled trials to examine the efficacy and safety of heparin in patients who had sepsis, severe sepsis, septic shock, or disseminated intravascular coagulation associated with infection. Included were randomized trials of unfractionated or low-molecular-weight heparin in patients with any of these conditions. The study's primary outcome was mortality; safety outcomes included hemorrhage, transfusion, and thrombocytopenia. Results were published in the March Critical Care Medicine.

Nine trials involving 2,637 patients were included in the study. The risk ratio for heparin-associated death was 0.88 (95% CI, 0.77 to 1.00) compared with placebo or usual care (6 trials, 2,477 patients, moderate strength of evidence) and 1.30 (95% CI, 0.78 to 2.18) compared with other anticoagulants (3 trials, 160 patients, low strength of evidence). For major hemorrhage, the risk ratio was 0.79 (95% CI, 0.53 to 1.17) for heparin compared with placebo or usual care (3 trials, 2,392 patients) and 2.14 (95% CI, 1.07 to 4.30) for heparin compared with other anticoagulants (1 trial, 48 patients). Too few events were reported for other secondary outcomes to allow comparison.

The researchers noted that their study included notable between-trial variability in patient population, heparin dosing, duration of drug exposure, and follow-up duration and that many adverse events were not completely reported, among other limitations. They concluded that heparin may be associated with lower mortality rates in sepsis, septic shock, or disseminated intravascular coagulation associated with infection, although they said that its overall impact remains uncertain and that the possibility of increased bleeding with heparin cannot be excluded. “Large rigorous randomized trials are needed to more carefully evaluate the efficacy and safety of heparin in patients with sepsis, severe sepsis, and septic shock,” they wrote.

ACP issues pressure ulcer prevention, treatment guidelines

Guidelines on risk assessment, prevention, and treatment of pressure ulcers were recently issued by ACP.

The guidelines, published in the March 3 Annals of Internal Medicine, are based on evidence reviews, which were published as a supplement. The guideline on risk assessment and prevention of pressure ulcers includes 3 recommendations:

  • ACP recommends that clinicians should perform an assessment to identify patients who are at risk of developing pressure ulcers (weak recommendation, low-quality evidence).
  • ACP recommends that clinicians should choose an advanced static mattress or advanced static overlay in patients who are at an increased risk of developing pressure ulcers (strong recommendation, moderate-quality evidence).
  • ACP recommends against using alternating-air mattresses or alternating-air overlays in patients who are at increased risk of developing pressure ulcers (weak recommendation, moderate-quality evidence).

The authors of the guideline noted that although risk assessment tools can be helpful in some cases, “the current evidence does not conclusively show a difference between clinical judgment and risk assessment scales in reducing pressure ulcer incidence.” As for preventive interventions, the evidence supported the benefits of advanced static mattresses or overlays (although no specific brand) over standard mattresses but found no clear additional benefit from alternating-air beds and overlays, which are significantly higher cost.

The guideline on treatment of pressure ulcers also provided 3 recommendations:

  • ACP recommends that clinicians use protein or amino acid supplementation in patients with pressure ulcers to reduce wound size (weak recommendation, low-quality evidence).
  • ACP recommends that clinicians use hydrocolloid or foam dressings in patients with pressure ulcers to reduce wound size (weak recommendation, low-quality evidence).
  • ACP recommends that clinicians use electrical stimulation as adjunctive therapy in patients with pressure ulcers to accelerate wound healing (weak recommendation, moderate-quality evidence).

The evidence was insufficient to recommend an optimal dose or form of protein, the guideline authors noted. There was also insufficient evidence on the effectiveness of alternating-air chair cushions, 3-dimensional polyester overlays versus gel overlays, zinc supplementation, L-carnosine supplementation, other wound dressings, topical phenytoin, maggot therapy, other biological agents, surgical techniques, or hydrotherapy.

Practicing physicians will have to make treatment decisions in these areas despite the lack of evidence and thus might have to also consult other, opinion-based guidelines, according to an accompanying editorial. There is a dire need for additional research in this field, the editorialist concluded.

Shorter hospital stay after hip fracture associated with increased mortality risk, study finds

Hospitalization for 10 days or less after a hip fracture was associated with increased mortality risk after discharge, according to a recent study.

Researchers in Sweden used data from a closed nationwide cohort to examine whether inpatient length of stay after hip fracture was associated with post-discharge risk for death. Patients living in Sweden as of Dec. 31, 2005, who were at least 50 years of age and had a first hip fracture in 2006-2012 were included in the study. The main outcome measure was death within 30 days of discharge as related to length of stay and after adjustment for covariates. The study results were published online Feb. 20 by BMJ.

Overall, 116,111 patients had a first hip fracture during the study period, at a mean age of 82.2 years. Of these, 5,863 patients died during their hospital stay, 6,377 died within 30 days of discharge, and 30,052 died within 1 year after admission. The mean inpatient stay after hip fracture was 14.2 days in 2006 versus 11.6 days in 2012 (P<0.001). The researchers found a nonlinear association between length of stay and mortality risk after discharge with a threshold of approximately 10 days. A total of 59,154 patients had a length of stay of 10 days or less, and among this group, each 1-day reduction in length of stay increased risk of death within 30 days of discharge in both 2006 (odds ratio, 1.08; 95% CI, 1.04 to 1.12) and 2012 (odds ratio, 1.16; 95% CI, 1.12 to 1.20). This association was not seen for the 56,957 patients with a stay of 11 days or more.

The authors noted that they could not be certain of patients' causes of death and that they had little information on whether patients were discharged to home or to other facilities, among other limitations. However, they concluded that a hospital stay of 10 days or less after hip fracture was associated with increased mortality risk after hospital discharge. “In addition to evaluation of other diagnoses than hip fractures, further research should seek to gain a better understanding of the underlying cause of the increased risk of death after discharge in surgical patients, and evaluate whether early discharge to rehabilitation centers or nursing homes is associated with a worse outcome,” the authors wrote. They noted that early discharge may mean less care by nurses who have at least a bachelor's degree and less time for comprehensive geriatric assessment, factors which may lead to worse outcomes.

The authors of an accompanying editorial discussed the trend in modern health care toward shorter lengths of stay and said that “it is far from clear that reductions in length of stay do not cause harm.” However, they also pointed out that the current study excluded patients who died during the index hospital admission and that patients with very short stays might have had worse prognoses at baseline than those with longer stays.

“As clinicians we should be aware that, although hospitals are scarce resources, rapid discharge of patients should be considered carefully,” the editorialists wrote. “For some patients, early discharge is likely beneficial and for other perhaps not. As policy makers, we should be aware that, while we have the tools to reduce length of stay, the potential for unintended consequences should not be forgotten.”

Fondaparinux associated with better outcomes than LMWH in NSTEMI study

Fondaparinux was associated with lower risk of major bleeding and death than low-molecular-weight heparin (LMWH) for patients with non-ST-segment elevation myocardial infarction (NSTEMI), a recent prospective study found.

Researchers used data from a Swedish registry that included 40,616 consecutive patients who had an NSTEMI and were treated with either fondaparinux or LMWH between 2006 and 2010 to conduct their prospective multicenter cohort study, in which patients were followed for up to 180 days. Results were published in the Feb. 17 Journal of the American Medical Association.

Overall, 36.4% of the patients were treated with fondaparinux and 63.6% received LMWH. The fondaparinux group had a lower rate of in-hospital bleeding events: 1.1% versus 1.8% (adjusted odds ratio [OR], 0.54; 95% CI, 0.42 to 0.70). They were also less likely to die while in the hospital: 2.7% versus 4.0% (OR, 0.75; 95% CI, 0.63 to 0.89). The differences between groups on these outcomes extended out to 30 and 180 days, and there were no significant differences in rates of recurrent myocardial infarction (MI) or stroke during the study follow-up. There were lower rates of in-hospital death for fondaparinux versus LMWH in all renal function categories, except for the small group with estimated glomerular filtration rate ≤15 mL/minute/1.73 m2, which showed no difference. There was no statistically significant difference in the association between treatment and outcome in patients who received in-hospital percutaneous coronary intervention (PCI) and those who didn't.

This registry study confirms previous randomized, controlled trial evidence that fondaparinux is associated with lower bleeding and mortality risk in NSTEMI patients than LMWH, the study authors concluded. This real-world investigation included more patients with a history of MI, stroke, or renal dysfunction than the randomized OASIS-5 trial, they noted. However, the present study was also limited by the likelihood of residual confounding, possible underestimations of bleeding events, and lack of information about the dose and duration of fondaparinux or LMWH.

Under European Society of Cardiology guidelines, fondaparinux is recommended as the first-choice anticoagulant for NSTEMI patients whether or not they undergo PCI. However, the American College of Cardiology/American Heart Association guidelines on the topic don't choose a specific anticoagulant for PCI patients and recommend either fondaparinux or enoxaparin for other NSTEMI patients, the study authors noted.

Comprehensive geriatric unit improved outcomes in hip fracture patients

Treating hip fracture patients on a geriatric unit instead of an orthopedic one resulted in slightly improved mobility, quality of life, and activities of daily living 4 months after hospitalization, a study found.

The Norwegian trial randomized about 400 previously community-dwelling patients, age 70 or over, to either usual orthopedic care for a hip fracture or comprehensive geriatric care. The geriatric unit had a higher clinician-to-patient ratio, as well as a system of structured interdisciplinary care focused on early mobilization and rehab, somatic health, mental health, function, social determinants, and early discharge planning. Results were published in The Lancet on Feb. 4.

At 4 months, 174 patients remained in the geriatric group and 170 in the orthopedic group, with most lost due to death. The geriatrics group had a higher mean score on the Short Physical Performance Battery (5.12 vs. 4.38; difference 0.74; 95% CI, 0.18 to 1.30; P=0.01). They also scored better on activities of daily living, fear of falling, and quality of life at both 4 and 12 months. The geriatric patients had a longer length of stay than orthopedic patients: 12.6 days versus 11.0, which may be due to geriatric care being time-consuming, the authors noted. After hospitalization, geriatric patients were more likely to be discharged to home (25% vs. 11%; P=0.001). Inpatient costs were higher in the geriatric group, but analysis of the overall results showed comprehensive geriatric care to be a cost-effective alternative to orthopedic care, the study authors said.

The study suggests that hip fractures in older patients should be treated with orthogeriatric care, which is in accordance with guidelines on the treatment of fragility fractures, the authors said. This was a single-center trial, which raises questions about generalizability, but since the intervention required many clinicians, the results are likely not due to any single person's skills or competence, the authors suggested.

Additional research is needed to confirm the findings, according to an accompanying comment. “However,” the authors wrote, “the availability of comprehensive geriatric care facilities should be increased to respond better to the growing needs of a frail, older, population.”