Recent Research

Restrictive transfusions, hospice care use, and more.

Restrictive transfusion strategy as safe as liberal strategy after cardiac surgery

A restrictive perioperative transfusion strategy was noninferior for mortality and morbidity to a liberal strategy, in patients undergoing cardiac surgery, a study found.

In a prospective, randomized, controlled clinical noninferiority trial, 502 ICU patients were randomly assigned to a liberal strategy of blood transfusion (to maintain a hematocrit ≥30%) or a restrictive strategy (to maintain a hematocrit ≥24%). Eligible patients underwent elective cardiac surgery with cardiopulmonary bypass between February 2009 and February 2010 at a Brazilian hospital's cardiac surgery referral center. The main outcome measure was a composite end point of 30-day, all-cause mortality and severe morbidity (the latter comprising cardiogenic shock, acute respiratory distress syndrome or acute renal injury) during the hospital stay. A -8% noninferiority margin was predefined, meaning an 8% minimal clinically important increase in occurrence of the composite end point. Results were published in the Oct. 13, 2010 Journal of the American Medical Association.

In the liberal strategy group, hemoglobin concentrations were maintained at a mean of 10.5 g/dL (95% CI, 10.4 to 10.6 g/dL), compared to 9.1 g/dL (95% CI, 9.0 to 9.2 g/dL) in the restrictive strategy group (P<0.001). Seventy-eight percent of patients in the liberal strategy group and 47% in the restrictive strategy group received a blood transfusion (P<0.001). Composite end point occurrence was 10% for the liberal group and 11% for the restrictive group, a non-significant difference. The number of transfused red blood cell (RBC) units was an independent risk factor for clinical complications or death at 30 days, independent of strategy (hazard ratio for each additional unit transfused, 1.2 [95% CI, 1.1 to 1.4]; P=0.002).

These data suggest a restrictive transfusion strategy is as safe as a liberal one, though there was a nonsignificant trend toward higher incidence of cardiogenic shock in patients treated with the restrictive strategy, the authors noted. Strategy aside, clinicians should pay attention to the number of transfused units, as it is an independent risk factor for worse outcomes, they said. “These findings suggest that the primary strategy in patients undergoing cardiac surgery should be to avoid giving RBC solely to correct low hemoglobin levels,” the authors said. “In addition, clinicians caring for patients after surgery should administer only 1 RBC unit at a time, because this may result in less exposure to risks but similar benefits.”

Prior DVT, large catheters raise risk for PICC-associated DVT

Patients who have larger catheters inserted and prior deep vein thrombosis (DVT) are at elevated risk for peripherally inserted central catheter (PICC)-associated DVT, a study found.

In a prospective observational study, researchers at a level I trauma center followed all patients with PICCs inserted during 2008, monitoring them for symptomatic DVT. Venous duplex ultrasonography was ordered based on observation of clinical manifestations of DVT, like swelling in the upper extremity, pain, or leaking at the PICC site. DVT events were identified by analyzing dictated ultrasound reports, with DVT defined as the noncompressibility of the vein using ultrasound probe during direct visualization. Twenty-one risk factors for DVT were examined, including catheter size, length, duration and tip location; surgery duration; BMI; and use of anticoagulants and/or pressors, among others. All patients were followed until five days after PICC removal or hospital discharge; those with DVT were followed for 90 days. Results were published in the October 2010 CHEST.

In 1,728 patients, a total of 2,014 PICCs were inserted. Some patients had multiple PICCs inserted during the same hospitalization; some had several insertions during different hospitalizations. Most PICCs were placed in the right arm (76.9%) and basilic vein (74%), and were double-lumen 5F (75.3%). Sixty PICC insertions in 57 patients developed DVT in the cannulated or adjacent veins. The DVT rate was 0.6% for single-lumen 4F PICCs, 2.9% for double-lumen 5F PICCs and 8.8% for triple-lumen 6F PICCs. The risk of DVT increased with history of previous DVT (odds ratio [OR], 9.92; P<0.001), use of a double-lumen 5F compared with a single lumen 4F PICC (OR, 7.54; P<0.05), and use of a triple-lumen 6F compared with a single lumen 4F PICC (OR, 19.50; P<0.01). The best multivariable predictive model achieved an area under the curve of 0.83, and included previous DVT, PICC size, and surgery duration of more than an hour, though on its own the last variable didn't achieve significance at the P=0.05 level.

The decision to place a PICC should be based on the clinical needs of the patient and catheter size “should be based on compelling clinical indications, not on convenience or department stock,” the authors concluded. There are disadvantages with smaller-size catheters, such as their fragility, which must be taken into account along with the risk of DVT with larger catheters. Still, the results suggest triple-lumen 6F catheters shouldn't be routinely used in the absence of an indication for three lumens, they said. Study limitations include the fact that PICCs were always inserted at the bedside by nurses, thus the results may not apply to those inserted by other personnel or in different patient groups. Also, since only symptomatic DVT was reported in this study, the true incidence of PICC-associated DVT is underreported, the authors noted.

Hospice care use rising, but timing still needs improvement

The use of hospice has increased in recent years, but a significant proportion of patients are still entering the program too early or too late, according to two studies.

A retrospective cohort study of U.S. heart failure patients, which included more than 200,000 Medicare beneficiaries who died in 2000-2007, found that use of hospice increased from 19% to nearly 40% over the course of the study. Yet the average number of days the patients spent in intensive care increased from 3.5 to 4.6 and adjusted Medicare costs increased by 11%. About 80% of patients were hospitalized during their last six months.

The study also assessed the length of patients' stays in hospice and found that the percentage of patients who entered hospice only shortly before their deaths remained steady: about 19% stayed three days or less and about 37% were in hospice for less than a week. It was published Feb. 14 in the Archives of Internal Medicine.

Another study, published in the same issue of Archives, assessed hospice use among men dying of prostate cancer between 1992 and 2005. Of the patients who used hospice (53% of the total), 22% were in it for less than a week. Such short stays don't allow patients to receive the full benefits of enrollment in hospice, the study authors noted.

However, this study did find an increase in the use of hospice over time. By 2003, 62% of the patients were in hospice. Hospice patients were less likely to receive high-intensity care, including imaging, hospitalization, ICU stays, and emergency department visits, the study found. Increasing the appropriate use of hospice care for patients at the end of life could both improve the quality of death and reduce ineffective health care expenditures, the study authors concluded.

Also published in the Archives issue was a study of Canadian heart failure patients at the end of life that found increasing costs between 2000 and 2006, but the Canadian costs were substantially lower (an average cost of 27,983 Canadian dollars in 2006 [equivalent to $24,674 in U.S.] vs. a mean cost of $36,216 in 2007 in the U.S. heart failure study). The percentage of Canadian patients dying in the hospital decreased from 60% to 54% and hospitalization rates during the last six months decreased from 84% to 76% during the study. Study authors noted that further reductions in hospitalization rates and deaths may be limited by the relative unavailability of hospice care in Canada compared to the U.S. Both U.S. and Canadian heart failure studies noted that increasing costs do not appear to be driven by use of costly invasive cardiac procedures, rates of which remained low throughout the study period in both countries.

Dysphagia screening can help predict pneumonia risk after stroke

Screening for dysphagia helped predict pneumonia risk in poststroke patients, a study found.

Using observational data, researchers analyzed outcomes for 18,017 patients with stroke who were discharged from 222 hospitals in six states between March 1 and Dec. 31, 2009. The patients were classified as having been unscreened for dysphagia before oral intake of food, liquid or medicine; screened and passed before oral intake; or screened and failed before oral intake. The main outcome, pneumonia, was defined as having occurred during hospitalization at least 48 hours after admission, and having necessitated antibiotics. The researchers used logistic regression models to determine the association between screening status and pneumonia rate. Results were published in the December 2010 Stroke.

Twenty-five percent (n=4,509) of patients were unscreened; 47% (n=8,406) were screened and passed; and 28% (n=5,099) were screened and failed. Patients who were screened were more likely to have aphasia or weakness compared with those who weren't screened. Patients who failed screening were older (median, 74 years vs. 70 years), and had longer hospital stays (median, 6 days vs. 5 days) compared to the other two groups. Pneumonia rates were lower in patients who were unscreened (4.2%) than those who were screened and failed (6.8%), but higher than those who were screened and passed (2.0%). Unscreened patients were still at a higher risk of pneumonia than screened-and-passed patients after adjustment for clinical and demographic features (odds ratio [OR], 2.2; 95% CI, 1.7 to 2.7). Those who failed screening were also at a higher risk for pneumonia than those who passed (OR, 3.6; 95% CI, 3.0 to 4.3).

The data suggest patients are selectively screened based on their stroke severity, and the higher pneumonia rate in the unscreened versus the screened-and-passed patients shows clinical judgment is imperfect at determining which patients to screen, the authors said. Since patients who failed screening had a higher risk than those who passed, dysphagia screening results appear to add accuracy in predicting pneumonia risk. Heightened vigilance in those who fail screening could help with identifying and treating pneumonia, the authors said. Furthermore, the “lack of National Quality Forum-endorsed DS [dysphagia screening] performance measures in the stroke measure set for The Joint Commission may reduce overall screening rates, which could increase poststroke complication rates,” they wrote. “Given our results, a new stroke measure for DS/pneumonia prevention should be developed for acute stroke care.”

Follow-up lowers readmissions, return ED visits for COPD patients

Chronic obstructive pulmonary disease (COPD) patients who had a follow-up visit shortly after hospital discharge were less likely to visit the ED or be readmitted within 30 days than those without follow up, a study found.

In a retrospective cohort study, researchers used enrollment and claims data from 62,746 Medicare patients who were hospitalized for COPD at least once between 1996 and 2006. For patients with more than one admission, the researchers randomly selected one admission per patient per year; all patients were discharged to home. Also, all patients had an identifiable primary care provider (PCP), meaning they had visited that PCP three or more times in the year before hospitalization. The researchers used Cox proportional hazard regression with time-dependent covariates to determine the risk of 30-day ED visit and readmission for patients with or without a follow-up visit to their PCP or pulmonologist within 30 days of discharge. About 74% of patients were exclusively treated by their PCP, while 26% were cotreated with a pulmonologist in the year before hospitalization. Results were published in the October 11, 2010 Archives of Internal Medicine.

Sixty-seven percent of patients with COPD had an outpatient visit with their PCP or pulmonologist within 30 days of discharge (that percentage increased from 65% in 1996 to 71% in 2006, P<0.001). Among those who had follow-up, 86.7% occurred with a PCP, 6.7% with a pulmonologist, and 6.6% with both. Patients who had no follow-up visit were more likely to have a longer length of hospital stay, be older than 85 years, have multiple comorbidities, be black and be admitted through the ED. Receiving care at a large hospital or teaching hospital, and/or in a city, was also associated with less follow up. The 30-day rate of postdischarge ED visits in patients with follow up was 21.7% compared to 26.3% for non-follow up patients (P<0.001). Thirty-day readmissions were 18.9% for the follow-up patients and 21.4% patients without follow up (P<0.001). Of those readmitted, the top three reasons were COPD, pneumonia or respiratory infection, and heart failure. In a multivariate, time-dependent analysis, follow-up visits were associated with 14% fewer ED visits (hazard ratio [HR], 0.86; 95% CI, 0.83 to 0.90) and 9% fewer readmissions (HR, 0.91; 95% CI, 0.87 to 0.96).

The results suggest discharge planning that emphasizes early outpatient follow up with particular emphasis on vulnerable populations (older and/or black adults, and those with multiple comorbidities) may reduce readmissions for COPD patients, they said. Study limitations include that some patients may have had follow-up visits with non-PCP/pulmonologist clinicians; that information on quality of care during hospitalization wasn't taken into account; and that the observational design means associations could come from confounding variables, the authors noted.

Model finds multimodal CT cost-effective to identify intra-arterial therapy candidates

Multimodal CT appears to save costs compared to noncontrast CT (NCCT) when identifying acute stroke patients who need intra-arterial (IA) therapy, a modeling study found.

Multimodal CT, which includes NCCT, CT with contrast, CT angiography and perfusion CT, is more costly than NCCT alone, but is an alternative to conventional cerebral angiography for identifying clots. Researchers designed a Markov model to calculate future costs and quality of life for a hypothetical cohort of stroke patients who presented within three hours of symptoms, and who would be considered for intravenous (IV) tissue plasminogen activator (tPA) or IA procedures when needed. They compared two strategies: NCCT followed by cerebral angiography to screen candidates for IA procedures vs multimodal CT followed by IA procedures in patients with identified intraluminal clot. Candidates for IA therapy (any emergent endovascular thrombolysis, thrombectomy, or combination) included patients in whom IV tPA was contraindicated, or who didn't improve after IV tPA. The researchers considered a three-month time horizon, since data for IV tPA and stroke outcome are well-established at three-month follow up, as well as a time horizon of the cohort lifetime. Outcomes, probabilities and direct costs in 2008 U.S. dollars were estimated from the literature. Results were published Nov. 9, 2010 in Neurology.

For patients with symptoms of stroke severe enough to warrant tPA, the model predicted 0.171 quality-adjusted life-years (QALYs) and costs of $20,165 for multimodal CT; QALYs for NCCT were 0.167 and costs were $21,881. With lower incremental costs and greater QALYs, multimodal CT was the dominant imaging strategy and the cost-effective choice 100% of the time for a $100,000/QALY willingness-to-pay. The number needed to screen with multimodal CT to avoid one diagnostic angiogram was two. In the lifetime horizon, multimodal CT was associated with 6.671 QALY and $223,229, while NCCT was associated with 6.663 QALY and $225,287. An alternative model, which evaluated a cohort with more severe symptoms and a higher prevalence of thrombus, showed that multimodal CT was cost-effective at three months, but NCCT became the dominant strategy over the lifetime of the cohort.

This model's limitations include that only one acute event was followed, and risks of exacerbating renal disease or a contrast reaction weren't included. Still, the results suggest that, in the short term, multimodal CT is a cost-effective screening tool for patients with acute stroke who might be considered for IV tPA or IA procedures. More data are needed to truly determine the long-term cost-effectiveness, however, the authors concluded.

Drug company information might influence prescribing habits

Information from drug companies does not improve prescribing, according to a meta-analysis.

Researchers reviewed studies of prescribing physicians who were exposed to sales visits, journal advertisements, attendance at pharmaceutical company-sponsored meetings, mailed information, prescribing software and participation in sponsored clinical trials. The outcomes measured were quality, quantity and cost of physicians' prescribing.

The meta-analysis included mostly cross-sectional studies, but also included cluster randomized controlled trials, time series analyses, before-after studies, cohort studies, and case-control studies. Studies were included if they both measured exposure to any type of information directly provided by pharmaceutical companies and physicians' prescribing. Studies were excluded if they looked at the indirect information, such as continuing medical education courses funded by unrestricted grants, or if they were case series, case reports, abstracts, news items and short reports. Results were published Oct. 19, 2010 in PLoS Medicine.

Of the studies, 38 found associations between exposure to drug company information and more frequent prescribing, while 13 did not. Among the many analyses of potentially influential factors:

  • Of studies of pharmaceutical sales representative visits, 17 found an association with increased prescribing of the promoted drug. None found less frequent prescribing. Of the remaining 11, six had mixed results.
  • Of the four studies that measured journal advertisements and included statistical tests, one found that journal advertisements had a more pronounced effect on market share for the advertised drug than did positive scientific information published in medical journals.
  • Of eight studies of pharmaceutical company-sponsored meetings, five found positive associations with prescribing frequency and three did not.
  • Of three studies of mailed promotional material, one found an association with increased prescribing and two did not.
  • A single study that examined the effect of advertising in clinical practice software found no association with prescribing frequency for six medications and less prescribing of one medication.
  • Several studies combined the outcome measures for various exposures to pharmaceutical company information or measured overall promotional investment, a proxy for the amount of exposure to information from pharmaceutical companies. Three studies found that total promotional investment was positively associated with prescribing frequency. Two studies found both positive results and no association. One study did not detect an association.

While the limitations of the original studies and of the nature of meta-analysis itself means that researchers cannot prove causality between drug company information and prescribing habits, they found some evidence of increased costs and decreased quality of prescribing.

Researchers did not find evidence of net improvements in the quality of prescribing associated with exposure to information from pharmaceutical companies. In the absence of such evidence, researchers wrote, “We recommend that practitioners follow the precautionary principle and thus avoid exposure to information from pharmaceutical companies unless evidence of net benefit emerges.”

Nonrespiratory causes of death becoming more common in oxygen-dependent COPD

Mortality rates are decreasing for respiratory disease and increasing for nonrespiratory causes in patients receiving long-term oxygen therapy for chronic obstructive pulmonary disease (COPD), according to a recent study.

Swedish researchers performed a prospective study of 7,628 patients with COPD (4,027 women, 3,601 men) who started long-term oxygen therapy between Jan. 1, 1987 and Dec. 31, 2004. Patients were followed until the end of the study, until oxygen therapy was stopped, or until death. The goal of the study was to determine whether there have been any time trends in overall and cause-specific mortality, especially as more older people and more women with severe COPD have begun receiving long-term oxygen therapy. The study's primary end point was cause of death according to the Swedish National Causes of Death Register. The study results were published online Jan. 7 by the American Journal of Respiratory and Critical Care Medicine and appeared in the April 15 print issue.

Median follow-up was 1.7 years (range, 0 to 18 years). Overall, 5,457 patients (71.5%) died during the study, with a 1.6% (95% CI, 0.9% to 2.2%; P<0.001) annual increase in crude overall mortality rate. Seventy-one percent of deaths were due to respiratory causes, 16% were due to circulatory causes, 7.6% were due to cancer and 1.1% were due to digestive organ disease. The absolute risk for death increased by 2.8% annually (95% CI, 1.3% to 4.3%; P<0.001) for circulatory disease and by 7.8% (95% CI, 1.9% to 14.0%; P=0.009) for digestive organ disease but decreased by 2.7% annually (95% CI, 2.0% to 3.3%; P<0.001) for respiratory disease and by 3.4% (95% CI, 1.1% to 5.7%; P=0.004) for lung cancer.

The authors cautioned that their mortality estimates are based on death certificate data and that estimates of cause-specific mortality could have been affected by changes in such factors as diagnostics and coding. However, they concluded that both overall mortality rates and mortality rates due to nonrespiratory causes have increased over time in patients with oxygen-dependent COPD. “The present study supports the important prognostic role of co-morbidity in oxygen-dependent COPD and shows that mortality has increased for non-respiratory causes, such as cardiovascular disease,” the authors wrote.