Recent Research

Lower-extremity revascularization in nursing home residents, guidance on contact precautions for hospital visitors, treatment in-hospital strokes versus community-onset strokes, and more.


Lower-extremity revascularization appears to yield poor functional outcomes in nursing home residents

Most nursing home residents who undergo lower-extremity revascularization are unlikely to benefit from it, according to a recent study.

Researchers used Medicare claims data from 2005 to 2008 with follow-up to 2009 and linked them to the Minimum Data Set for Nursing Homes (MDS) to examine the effect of lower-extremity revascularization in long-term nursing home residents. MDS assessments are mandatory for all U.S. nursing home facilities that participate in Medicare or Medicaid and are done at nursing home admission, readmission, quarterly, and at change in clinical status. Patients who had had a lower-extremity revascularization procedure in the 2 years before the study period were excluded. The MDS for activities of daily living score was used to determine changes in the 3 main outcome measures: functional status, ambulatory status, and death. The study results were published in the June JAMA Internal Medicine.

Data from 10,784 long-term nursing home residents who had lower-extremity revascularization were included in the study. Patients' mean age was 82.1 years. Sixty-three percent were women, and 80% were white. Sixty percent had cognitive impairment. Seventy-five percent of patients were not ambulatory before the revascularization procedure was performed, and 40% had had functional decline in the 6 months before surgery. At 1 year after the surgery, overall mortality rates were 51% among ambulatory residents and 53% among nonambulatory residents, while overall 82% of patients had died or were nonambulatory.

Of 1,672 patients who were ambulatory before having surgery, 63% had died or were nonambulatory 1 year afterward. Of 7,188 patients who were nonambulatory before surgery, 89% had died or were nonambulatory at 1 year. In the first 3 months after surgery, 40% of the entire cohort and 56% of surviving patients experienced functional decline compared with before surgery. By 1 year after surgery, 36% of surviving patients had improved or maintained their function and in 64% function had declined. The researchers performed multivariate adjustment and found that age 80 years or older, cognitive impairment, congestive heart failure, renal failure, emergent surgery, nonambulatory status before surgery, and decline in activities of daily living before surgery were all factors independently associated with death or with nonambulatory status after surgery.

The authors noted that detailed information about surgery indications was not available and that they did not know which patients with critical limb ischemia did not have revascularization and instead were treated with palliative amputation or pain control and wound care. However, they concluded that although many nursing home residents in the U.S. have lower-extremity revascularization, few appear to gain much benefit from it and the procedure helps a nonambulatory patient to become ambulatory only rarely. They also noted that while ambulatory function may be a goal of care, it might not be attainable in many cases.

“Treatment decisions for elderly, frail patients with critical limb ischemia are complex; shared decision-making is needed,” the authors wrote. “Our findings can inform conversations between physicians, patients, and families about the risks and expected outcomes of surgery and whether the surgery is likely to be worthwhile.”

The author of an accompanying invited commentary called the findings “balanced and valuable” and pointed out that since many of the patients in the study were not walking at surgery, the procedures in these cases were probably performed to relieve symptoms of ischemic leg pain, wounds that were not healing, or worsening gangrene. “In this context, lower extremity revascularization should be viewed as a palliative measure rather than as a definitive therapeutic procedure to extend life or ambulatory function,” the commentary author wrote.

The commentary author also noted that this and other studies could help inform a different approach to caring for frail elderly patients who are experiencing chronic pain and discomfort but whose lifespans are limited. “The best care will be patient and family centered, interdisciplinary, and characterized by communication and determining the goals of care,” the commentary author wrote. “Attention to pain control and other symptoms can reasonably include selective surgical intervention.”

SHEA issues guidance on contact precautions for hospital visitors

The Society for Healthcare Epidemiology of America (SHEA) recently released expert guidance on isolation precautions for hospital visitors.

The recommendations are based on a synthesis of limited evidence, theoretical rationales, practical conditions, expert opinions (including a survey of SHEA members' practices), and consideration of potential harm. The authors attempted to balance visitor and patient safety, potential pathogen spread, psychosocial effects of isolation, and feasibility in their recommendations, which were published by Infection Control & Hospital Epidemiology on April 10.

In general, use of isolation precautions should be based on the specific pathogen, the underlying infectious condition, and the endemicity of the organism in the hospital and community, the guidance said. In addition, all visitors should perform hand hygiene on entering and leaving a patient room, with either soap and water or an alcohol-based rub.

More specifically, the guidance said that:

  • Contact isolations precautions are not recommended routinely for visitors in endemic situations of methicillin-resistant Staphylococcus aureus or vancomycin-resistant enterococci. However, if such visitors will be interacting with multiple patients, they may be at greater risk of transmission.
  • Contact precautions should be considered for visitors to patients colonized with extensively drug-resistant gram-negative organisms (e.g., Klebsiella pneumoniae carbapenemase).
  • Contact precautions are suggested for visitors to patients infected with enteric pathogens (e.g., Clostridium difficile, norovirus).
  • Isolation precautions may not be practical for visitors making an extended patient visit, such as overnight, and benefit from wearing gowns and gloves in these scenarios is unclear. If such visitors are assisting health care workers with care, and contact with body fluids is anticipated, standard precautions should be utilized.
  • Surgical masks are suggested for visitors of patients on droplet precautions, and recommended for visitors of patients on airborne precautions, although visitors with extensive prior exposure to the patient may be excluded.
  • In situations of potential outbreaks or novel, virulent pathogens (e.g., Ebola, Middle East respiratory syndrome coronavirus), enforcing isolation precautions should be prioritized, and hospitals should consider restricting nonessential visitors.

“These recommendations should be adapted to hospitals based on their specific situation, annual risk assessment, regional epidemiology, and individual patient needs, particularly among the pediatric population,” the authors concluded, also calling for additional research in this area.

Treatment slower for in-hospital strokes than community-onset ones

Patients who had strokes while already in the hospital waited longer for treatment and had worse outcomes than those whose strokes occurred outside the hospital, a recent study found.

The prospective cohort study included all adult acute stroke patients treated at regional stroke centers in Ontario, Canada, between July 2003 and March 2012. More than 28,000 patients with community-onset stroke were compared to 973 patients with in-hospital stroke. Results were published by JAMA Neurology on May 4.

The time from symptom recognition to neuroimaging was significantly longer in the in-hospital stroke group (median, 4.5 hours vs. 1.2 hour; P<0.001), and only 32% of the inpatients had imaging within 2 hours, compared to 63% of community-onset stroke patients. The inpatients also had a longer time from stroke recognition to administration of thrombolysis (median 2 hours vs. 1.2 hours; P<0.001).

After adjustment for a number of factors, including age and stroke severity, the 2 groups had similar mortality rates at 30 days and 1 year. The in-hospital stroke patients had a higher prevalence of comorbid illness and ineligibility for thrombolysis, the study authors noted. They also had a longer median length of stay after stroke (17 days vs. 8 days) and were more likely to be dead or disabled at discharge (77% vs. 65% with a modified Rankin Scale score of 3 to 6; P<0.001).

The study's results are consistent with previous research and indicate that there are delays in investigation and treatment of patients with in-hospital stroke, the authors concluded. Although these patients and their situations may be significantly different from community-onset patients, a standardized approach to recognition and management of in-hospital stroke could improve care, the authors wrote. They suggested development of “code gray” stroke protocols for inpatient populations similar to those used in the ED.

The delay between stroke recognition to neuroimaging in in-hospital strokes is one of the study's most striking findings, according to an accompanying editorial. The evidence from this and other studies “reveal a paradox in which a critical therapy is limited or delayed in a group of patients whose need for it may be greater,” wrote the editorialist.

Risk factors predict progression from sepsis to septic shock

Progression from sepsis to septic shock often occurred more than 4 hours after ED presentation and was more likely if patients displayed certain clinical characteristics, a recent study found.

The observational study included a retrospective chart review of 18,100 patients admitted from the EDs of 2 large academic medical centers between September 2010 and February 2011. Patients were excluded if they presented with ST-elevation myocardial infarction, acute stroke, trauma, or cardiac arrest or if they were pregnant, admitted from a psychiatric unit, or transferred from another hospital. Results were published in the May Critical Care Medicine.

Of the 18,100 patients, 3,960 had 2 or more systemic inflammatory response criteria, and 1,316 were found to have sepsis within 4 hours of ED arrival. Fifty of those patients (3.6%) developed septic shock within 4 hours of arrival, but 111 (8.1%) progressed to shock between 4 and 48 hours after arrival. Of the 111 patients, 95 (85.6%) developed septic shock between 4 and 24 hours. The most common source of infection was pneumonia, followed by gastrointestinal infection.

Patients were more likely to develop septic shock between 4 and 48 hours if they were female (odds ratio [OR], 1.59; 95% CI, 1.02 to 2.47) or had nonpersistent hypotension (OR, 6.24; 95% CI, 3.58 to 10.86), bandemia at least 10% (OR, 2.60; 95% CI, 1.50 to 4.51), lactate level at least 4.0 mmol/L (OR, 5.30; 95% CI, 2.59 to 10.84), or history of coronary artery disease (OR, 2.01; 95% CI, 1.26 to 3.44).

The study also looked at how patients were treated in the ED and found that 83.3% of the overall sepsis group received antibiotics in the ED, as did 87.4% of the patients who progressed to septic shock between 4 and 48 hours. Only 59.5% of that latter group was directly admitted to the ICU from the ED, the researchers noted. The study couldn't determine why some patients did not receive antibiotics in the ED, but it may be because the ED physician didn't recognize sepsis as a diagnosis, they speculated. These findings demonstrate the importance of aggressive treatment and monitoring of sepsis patients during the first 24 hours, the authors said.

The study shows that it is difficult to determine which few patients will progress to septic shock, the authors also said. They called for additional studies to similarly search for markers or risk factors of progression and assist with clinical decisions about which patients to admit to the ICU. Development of specific sepsis ED/hospital units could also improve monitoring of sepsis patients for progression, the authors suggested.

Hospital-acquired conditions have a July effect, analysis finds

Inpatients were significantly more likely to develop a hospital-acquired condition if they were admitted during the month of July, a recent study found.

Using discharge data from the Nationwide Inpatient Sample database from 2008 to 2011, researchers looked at rates of hospital-acquired conditions (HACs) as defined by CMS. The study evaluated HAC incidence as a function of admission month and adjusted for demographic and hospital factors. Results were published by the Journal of Hospital Medicine on March 31.

In the overall analysis of more than 143 million admissions, 4.7% of patients had a HAC. Of patients admitted during the month of July, 4.9% developed a HAC, which worked out to a 6% increase in HAC risk (P<0.001). Patients who developed a HAC had almost double the risk of prolonged length of stay and had significantly higher inpatient costs than those who didn't (P<0.001 for both comparisons).

July admissions are associated with a higher risk of HAC occurrence, the study authors concluded. Previous research has shown a July effect on other outcomes, such as mortality, but this is the first to assess the association with HACs, which may be a better, less confounded measure of the inefficiency and inexperience of new hospital staff. The July effect is not just a problem affecting teaching hospitals, since hiring of recently graduated physicians and nurses is common during the summer at nonteaching hospitals, too, the authors noted.

They also reported that the July effect spills over into August, with a similar increased risk of HACs found when August admissions were combined with July admissions and compared to the rest of the year. The results may represent “breakdowns in organization structure distinct from traditional quality measures,” the study authors said, recommending that protocols or guidelines be developed in response to this problem.

PE recurrence same with vena cava filter plus anticoagulation as anticoagulation alone

Adding an inferior vena cava filter did not reduce the risk of recurrent pulmonary embolism (PE) compared to anticoagulation alone, according to a recent study of French patients hospitalized with PE.

The open-label trial included patients who were hospitalized with acute, symptomatic PE associated with lower-limb venous thrombosis and who met at least 1 criterion for severity between 2006 and 2012. The patients were randomized to retrievable inferior vena cava filter implantation plus anticoagulation (n=200) or anticoagulation alone (n=199) and were followed for 6 months. Patients with a contraindication to anticoagulation therapy and those with recurrence despite adequate anticoagulation were excluded. Results were published in the April 28 Journal of the American Medical Association.

Filters were successfully inserted in 193 patients, and they were retrieved, at 3 months after placement, in 153 of the 164 patients in whom retrieval was attempted. By 3 months, 6 patients in the filter group had a recurrent PE (3.0% of the group, all were fatal), compared to 3 patients in the anticoagulation-only group (1.5% of the group; 2 fatal; P=0.50 for PE recurrence). The groups also had similar rates of recurrence within 6 months, symptomatic deep venous thrombosis, major bleeding, and mortality within 3 or 6 months. Filter thrombosis occurred in 3 patients. One patient died of cardiac arrest during filter insertion.

Implantation of the filters did not reduce risk of symptomatic PE recurrence, and thus the study's results do not support use of inferior vena cava filters in PE patients who can be treated with anticoagulation, the study authors concluded. The rate of PE recurrence in the study was less than expected, but this shows that “modern management with full-dose anticoagulation is likely very effective even in patients usually considered to be at high risk for recurrence,” the authors wrote.

The study was limited by its open-label design and its small size, and the results do not exclude the possibility of benefit from filters in certain subgroups of patients with venous thromboembolism, such as those with PE and hemodynamic instability. The study used the ALN filter, but given existing evidence of similarities among retrievable filter models, the results are probably applicable to other retrievable filters, the authors said.

Guideline offers recommendations to prevent acute COPD exacerbations

An evidence-based guideline from the American College of Chest Physicians and Canadian Thoracic Society offers recommendations on preventing exacerbations in patients with chronic obstructive pulmonary disease (COPD).

“This guideline is unique because a group of interdisciplinary clinicians who have special expertise in COPD clinical research and care led the development of the guideline process with the assistance of methodologists,” the guideline states.

The guideline was published in the April CHEST.

Among the full set of 33 recommendations are the following:

  • In patients with COPD, administer the 23-valent pneumococcal vaccine as part of overall medical management, although there isn't sufficient evidence that it prevents acute exacerbations (weak recommendation, low- or very low-quality evidence).
  • In patients with COPD, administer the influenza vaccine annually to prevent acute exacerbations of COPD (strong recommendation, moderate-quality evidence).
  • In patients with COPD, include smoking cessation counseling and treatment as part of a comprehensive clinical strategy to prevent acute exacerbations of COPD (weak recommendation, low- or very low-quality evidence).
  • In patients with moderate, severe, or very severe COPD who have had an exacerbation within the previous 4 weeks, offer pulmonary rehabilitation to prevent acute exacerbations of COPD (strong recommendation, low- or very low-quality evidence).
  • In patients with moderate, severe, or very severe COPD who have had an exacerbation prior to the past 4 weeks, do not offer pulmonary rehabilitation to prevent acute exacerbations of COPD (weak recommendation, moderate-quality evidence).

The guidelines also address the use of inhaled corticosteroids and long- and short-acting muscarinic antagonists and beta2-agonists based on patients' severity of disease.

The guideline states, “It is clear that large gaps in knowledge currently exist about exacerbation prevention that limit our ability to prioritize one type of therapy over another or make recommendations about combinations of therapy to prevent exacerbations. … Newer therapies that are soon to be released for clinical use or that are currently under investigation that focus on the prevention of COPD exacerbations also promise to rapidly improve the future armamentarium for the treatment of the patient with COPD.”

Diagnosis, management of pancreatic cysts covered in new guideline

The American Gastroenterological Association (AGA) released a guideline recently on the diagnosis and management of asymptomatic neoplastic pancreatic cysts.

Incidental identification of pancreatic cysts is growing due to increased use of more sophisticated imaging techniques, and a reliable strategy is needed to determine which of these cysts are related to cancer or high-grade dysplasia, the guideline said. While the guideline is based on a technical review of the literature, the available evidence on management of pancreatic cysts is very low quality and is mostly from case series. However, the AGA Clinical Practice Guideline Committee stated that it was still important to develop a guideline based on this evidence because the clinical problem is so complex. The committee rated each recommendation as strong, meaning that it should apply to most patients most of the time, or conditional, meaning that the decision is more nuanced and there could be different approaches for a significant number of patients.

Regarding surveillance, the AGA suggested the following conditional recommendations:

  • Patients who have pancreatic cysts smaller than 3 cm with no solid component or dilated pancreatic duct should have an MRI in 1 year, then every 2 years for a total of 5 years if the cyst does not change in size or characteristics.
  • Pancreatic cysts with 2 or more high-risk features, such as a size of 3 cm or greater, a dilated main pancreatic duct, or an associated solid component should be examined with endoscopic ultrasonography (EUS)-fine-needle aspiration (FNA).
  • Patients who have no concerning findings on EUS-FNA should have MRI after 1 year and then every 2 years to make sure that risk of malignancy hasn't changed.
  • EUS-FNA should be performed in patients whose cysts develop significant changes, such as development of a solid component, increased size of the pancreatic duct, and/or a diameter of 3 cm or more.
  • Continued surveillance is not necessary if the characteristics of the cyst have not changed significantly after 5 years of surveillance or if the patient is no longer a candidate for surgery.

The AGA also suggested surgery in patients whose cysts have a solid component and a dilated pancreatic duct and/or concerning features on EUS-FNA (conditional recommendation) and recommended referral to a center that has demonstrated expertise in pancreatic surgery (strong recommendation).

After surgical resection of a cyst with invasive cancer or dysplasia, the AGA suggests MRI surveillance of the remaining pancreas every 2 years. Patients whose cysts do not show high-grade dysplasia or malignancy at surgical resection do not need routine surveillance, the AGA suggests. Both of these recommendations are conditional.

The complete guideline was published online March 25 by Gastroenterology.

A related commentary published online March 30 by Annals of Internal Medicine put the guideline into perspective for physicians in general internal medicine. The commentary author said because the available evidence is limited, the technical review on which the guideline is based cannot definitively answer the questions that would be most helpful to patients and physicians, namely how often an incidental cyst harbors unsuspected cancer or precancer that will probably progress to a life-threatening condition; how well imaging or clinical characteristics help select patients most likely to benefit from surgery; what further diagnostic procedures and tests add; whether surgery at an asymptomatic stage and postsurgical surveillance would reduce mortality from pancreatic cancer; and whether the benefits of surgery would outweigh the harms and costs of work-up and treatment.

The commentary author said that the AGA position offers a middle ground between pursuing an aggressive strategy and deferring action until there is more evidence is available. He noted that the AGA takes into account the concern that aggressive strategies are often of low value and that any benefit may not justify harms and costs, while also recognizing “the problems of complete inaction in this situation.” The commentary author also congratulated the AGA for its “bold recommendations” and noted that recommending less aggressive care is often difficult for any group to do.

“Until better evidence is available, we will not know if the AGA guidelines provide greater value than previous ones, but they thoughtfully adopted a value framework,” he wrote. “To get value from new technology, we have to learn to use it with the restraint necessary to minimize harms and costs while preserving benefit.”