Nutrition, glycemic control, and more

Summaries from ACP Hospitalist Weekly.

Individualized nutritional support in the hospital associated with slightly better clinical outcomes

Individualized nutritional support appeared to improve survival and other clinical outcomes in at-risk medical inpatients versus standard hospital food, according to a recent study.

Image by Getty Images
Image by Getty Images

Researchers in Switzerland performed an open-label multicenter study to determine whether medical inpatients at nutritional risk benefited from individualized nutritional support. Medical inpatients from eight hospitals who were considered to be at nutritional risk and were expected to be hospitalized for more than four days were randomly assigned to receive individualized nutritional support, or standard hospital food. Nutritional risk was defined as a score of three or more points on nutritional risk screening.

Patients in the intervention group had protein and caloric goals determined by dietitians and began receiving nutritional support no more than 48 hours after hospital admission, and nutritional intake was reassessed by a dietitian every 24 to 48 hours throughout the hospital stay. Patients in the control group received standard hospital food and did not receive nutritional support. The study's primary composite end point was any adverse clinical reaction, defined as all-cause mortality, ICU admission, nonelective hospital readmission, major complications, and decreased functional status at 30 days. The study results were published April 25 by The Lancet and appeared in the June 8 issue.

A total of 2,088 patients were recruited for the study and were monitored between April 1, 2014, and Feb. 28, 2018. The intervention group included 1,050 patients, and the control group included 1,038 patients. Thirty-five patients in the intervention group and 25 in the control group withdrew their consent during the trial, leaving 1,015 patients and 1,013 patients, respectively, for analysis. Patients' mean age was 72.6 years, and mean body mass index was 24.8 kg/m2.

Caloric and protein goals were reached during the hospital stay in 800 (79%) and 770 (76%) patients, respectively, in the intervention group and 547 (54%) and 557 (55%) patients, respectively, in the control group. Adverse clinical outcomes occurred at 30 days in 232 patients in the intervention group and 272 patients in the control group (23% vs. 27%; adjusted odds ratio, 0.79; 95% CI, 0.64 to 0.97; P=0.023). Seventy-three patients in the intervention group versus 100 patients in the control group died by day 30 (7% vs. 10%; adjusted odds ratio, 0.65; 95% CI, 0.47 to 0.91; P=0.011). The activities of daily living score was better at 30 days in the intervention group versus the control group. Length of hospital stay between groups did not differ.

The researchers noted that their trial was not blinded and that a significant proportion of patients in the intervention group did not meet their caloric (21%) or protein (24%) goals, among other limitations. They concluded, however, that early use of individualized nutritional support to help patients reach protein and caloric goals during a hospital stay appears to improve outcomes in medical patients who are considered to be at nutritional risk. “Our findings strongly support the concept of systematically screening medical inpatients on admission to hospital for nutritional risk, irrespective of any underlying conditions, followed by a nutritional assessment and introduction of individualised nutritional support in at-risk patients,” the authors wrote.

Azithromycin may reduce treatment failure for acute COPD exacerbations

Patients hospitalized for an acute exacerbation of chronic obstructive pulmonary disease (COPD) may be less likely to experience treatment failure if they also receive azithromycin, a recent study found.

Researchers in Belgium performed a multicenter, double-blind, placebo-controlled trial in patients hospitalized for an acute COPD exacerbation to determine whether a three-month course of low-dose azithromycin initiated at admission could decrease rates of COPD treatment failure. Patients who had a smoking history of 10 or more pack-years and had had at least one COPD exacerbation within the previous year were randomly assigned within 48 hours of admission for an exacerbation to receive azithromycin, 500 mg/d for three days, or placebo. The study drug was given along with systemic corticosteroids and antibiotics and was continued for three months at a dose of 250 mg every two days.

Patients were followed for nine months, including six months after the study drug was withdrawn. The composite end point was treatment failure over three months in the intention-to-treat population, where treatment failure was defined as a composite of treatment intensification with systemic corticosteroids, antibiotics, or both; stepped-up hospital care or a respiratory-related readmission; or all-cause mortality. The results of the study were published May 3 by the American Journal of Respiratory and Critical Care Medicine and appeared in the Oct. 1 issue.

Three hundred one patients were included in the study, 147 assigned to azithromycin and 154 assigned to placebo. One hundred eighteen patients in the treatment group and 115 in the placebo group completed the study (80% vs. 75%). Rates of treatment failure by three months were 49% in the azithromycin group and 60% in the placebo group (hazard ratio [HR], 0.73; 95% CI, 0.53 to 1.01; P=0.0526). Rates of treatment intensification, stepped-up hospital care or respiratory-related readmission, and all-cause mortality were 47% versus 60% (HR, 0.70; 95% CI, 0.51 to 0.97; P=0.0272), 13% versus 28% (HR, 0.43; 95% CI, 0.25 to 0.75; P=0.0024), and 2% versus 4% (HR, 0.62; 95% CI, 0.15 to 2.59; P=0.5075), respectively. Differences between treatment groups could no longer be detected six months after azithromycin was withdrawn.

The authors noted that their trial did not meet its target enrollment and that their findings lack external validity and should not be widely generalized to other populations of COPD patients, among other limitations. They concluded that their results indicate a potential benefit of low-dose azithromycin taken during the three-month period after hospitalization for an acute COPD exacerbation, when patients are at high risk for relapse and death. “Prolonged treatment, however, appears to be needed to sustain clinical benefits,” the authors wrote. “A careful and individualized approach to the selection of patients, with regards to pro-arrhythmic effects and the development of antibiotic resistance, would be recommended.”

Lower glucose levels before hospital discharge associated with readmission, death in diabetes patients

Lower glucose levels during the 24 hours prior to discharge were associated with higher risk of readmission and mortality in a recent study of patients with diabetes.

The retrospective cohort study included 843,978 patients with diabetes admitted to Veterans Affairs hospitals in 2000 through 2014. The authors collected patients' minimum point-of-care glucose values during the last 24 hours of hospitalization and compared them to adjusted rates of readmission and mortality after discharge. Results were published by the Journal of Clinical Endocrinology and Metabolism on May 1.

Overall, 17.3% of the patients were readmitted within 30 days and 30-, 90-, and 180-day crude mortality rates were 2.3%, 6.0%, and 10%, respectively. The study used glucose values of 100 to 109 mg/dL (5.6 to 6.1 mmol/L) as the reference range. Patients with minimum glucose readings of 90 to 99 mg/dL (5.0 to 5.5 mmol/L) on the last day of hospitalization had similar rates of readmission and postdischarge mortality to the reference population. However, levels below that range were associated with progressively increasing risk of readmission and mortality. For example, compared to the reference group, the event rate ratio for 30-day readmission was 1.04 with a minimum glucose level of 80 to 89 mg/dL (4.4 to 4.9 mmol/L) and 1.46 for a glucose level of 10 to 19 mg/dL (0.6 to 1.1 mmol/L). Similarly, the event rate ratio for mortality within 180 days increased from 1.07 with a minimum glucose level of 80 to 89 mg/dL (4.4 to 4.9 mmol/L) to 1.96 for a glucose of 10 to 19 mg/dL (0.6 to 1.1 mmol/L).

The authors identified thresholds at which risk for these adverse outcomes increased: 92.9 mg/dL (5.2 mmol/L) for 30-day readmission, 45.2 mg/dL (2.5 mmol/L) for 30-day mortality, 65.8 mg/dL (3.7 mmol/L) for 90-day mortality, 67.3 mg/dL (3.7 mmol/L) for 180-day mortality, and 87.2 mg/dL (4.8 mmol/L) for the combined outcome of readmission or mortality within 30 days. They concluded that patients with diabetes who had hypoglycemia or near-normal glucose values during the last day of hospitalization had higher risk of readmission and mortality, but noted that the reasons for this association are unknown. “The transition of care from the inpatient to the outpatient setting is often challenging, leading to adverse events, poor glycemic control, increased emergency room visits, higher hospital readmission rates and costs,” the study said.

The authors recommended several potential approaches to this problem that should be tested in future trials, including delaying patients' release from the hospital until normoglycemia is achieved, modifying outpatient diabetes medication regimens, and advising patients to perform frequent glucose monitoring or use continuous glucose monitoring devices. They noted that the study was limited because it included only veterans, who are more likely to be male, elderly, and chronically ill than the general population. The results also do not apply to patients without a diagnosis of diabetes.

Infective endocarditis increases ischemic stroke risk for up to two years

There is an increased risk of ischemic stroke for up to two years after a hospitalization for infective endocarditis, a recent study found.

The study used the Danish national registry to identify 9,312 patients who had a first case of left-sided infective endocarditis from 1977 to 2015. They were compared to 91,996 matched controls who did not have infective endocarditis, and all were followed for up to 38.5 years. Results were published by Clinical Infectious Diseases on April 27.

Compared to the matched controls, patients with infective endocarditis had a significantly increased risk of ischemic stroke. The risk was highest during the first four weeks after endocarditis diagnosis (hazard ratio [HR], 57.20; 95% CI, 45.58 to 71.78; P<0.0001). It then gradually decreased, but it persisted for two years (HR for 4 weeks to 3 months, 5.40 [95% CI, 4.11 to 7.19]; HR for 3 months to 2 years, 1.73 [95% CI, 1.48 to 2.01]). The risk of stroke was not explained by the presence of atrial fibrillation or mechanical valves. Patients who received anticoagulants in the period from four weeks to three months after infective endocarditis had a lower risk of stroke compared to those who did not (HR, 0.30; 95% CI, 0.10 to 0.96; P=0.04).

The cause of this persistent increase in stroke risk is unknown, the study authors said, although they offered potential explanations, including that paroxysmal atrial fibrillation is difficult to diagnose and that systemic bacterial infections generally have been previously associated with embolic events. The results might call for longer follow-up after infective endocarditis to determine whether anticoagulation is indicated, the authors said. However, the risk of cerebral bleeding in this patient population also needs to be considered, and there is a lack of randomized trials in this area.

The study was limited by its use of ICD discharge codes and lack of information about which microorganisms and heart valves were involved in the infective endocarditis cases. However, strengths include that the data came from a registry covering an entire country with free access to health care, the authors said.

1 in 4 PCI patients have unplanned readmission within six months, study finds

Approximately one in four patients who underwent percutaneous coronary interventions (PCI) had an unplanned readmission within six months, a study found.

Researchers used the U.S. Nationwide Readmission Database to identify patients undergoing PCI between 2010 and 2014 and analyze the rate, causes, predictors, and cost of unplanned readmissions up to 180 days after discharge. Results were published by JACC: Cardiovascular Interventions on March 27 and appeared in the April issue.

Among more than 2.4 million studied patients, 2.5% were readmitted between 0 and 7 days, 7.6% were readmitted between 8 and 30 days, 8.9% were readmitted between 31 and 90 days, and 8.0% were readmitted between 91 and 180 days (cumulative rates, 2.5%, 9.9%, 18.0%, and 24.8%, respectively). There was a decline in readmissions in all time ranges between 2011 and 2014. The median time to readmission was 35 days (interquartile range, 14 to 79 days), and peak readmission rate occurred at seven days. Daily readmission rates were 0.35% for 0 to 7 days, 0.33% for 8 to 30 days, 0.15% for 31 to 90 days, and 0.09% for 91 to 180 days.

Most readmissions during each time period were due to noncardiac causes (53.1% to 59.6%). Nonspecific chest pain was the most common noncardiac cause of readmission within each time period (14.2% to 22.7% of noncardiac readmissions). The most common cardiac cause for readmission was coronary artery disease, including angina (37.4% to 39.3% of cardiac readmissions). Other common cardiac causes of readmission were acute myocardial infarction between 0 and 7 days (27.6% of cardiac readmissions) and heart failure during all subsequent time periods (22.2% to 23.7% of cardiac readmissions).

Causes of readmission depend on when they are assessed, and noncardiovascular causes are more common farther out from the initial admission, the authors noted. Any interventions developed to reduce unplanned readmissions should consider these factors, they advised.

An editorial cautioned against legislation that financially penalizes institutions when they fail to meet an “arbitrary benchmark” for readmissions, which may cause unintended consequences, citing the example of the Hospital Readmissions Reduction Program, where a reduction of heart failure admissions has been associated with an increase in mortality.

“In conclusion, understanding how often and why patients ‘bounce back’ after PCI should be added to our quality efforts. However, it is unclear how much effort we should expend on preventing readmissions until we know better what to target and how to do it effectively,” the editorial stated.

Clinicians, patients/surrogates often disagreed about ICU treatment, study finds

Patients and their surrogates often disagree with clinicians about appropriateness of ICU treatment, leading to lower satisfaction, according to a recent study.

Researchers performed a multicenter, prospective, observational study of adult patients at six ICUs in the U.S. and Hungary to determine how often patients or their surrogates perceived treatment to be inappropriate. They also examined the effect of this perception on adverse outcomes, as well as discordance with the opinions of clinicians. Patients, surrogates, physicians, and nurses were surveyed daily during the ICU stay about perceived inappropriate treatment, reasons for perceived inappropriate treatment, and associated distress. In addition, patient and surrogate satisfaction was assessed daily and at ICU discharge, patient and surrogate trust was assessed daily, and patient and surrogate anxiety and depressive symptoms were measured at ICU discharge. The study results were published March 25 by CHEST and appeared in the June issue.

The study included 151 patients for whom 1,332 surveys of patients, surrogates, nurses, and physicians were collected. In 8% of patients, patients and surrogates reported too much treatment, and in 6%, they reported too little. Too much treatment was defined as treatment that would not benefit the patient, that caused unnecessary burden, or was contrary to patient wishes, while too little treatment was defined as any additional treatment that the patient should be receiving. Twenty-six percent of patients and surrogates disagreed with clinicians about whether too much treatment was administered, while 10% disagreed about whether too little treatment was administered.

An association was seen between disagreement about perceived inappropriate treatment and both prognostic discordance and lower patient or surrogate satisfaction (P=0.02 for both comparisons). In addition, perceived inappropriate treatment was associated with moderate or high distress in 55% of patient and surrogate respondents and 35% of physician and nurse respondents. Patients and surrogates who perceived inappropriate treatment were more likely to also report lower satisfaction and less trust in the clinical team. However, no statistically significant difference was seen between perception of inappropriate treatment and patient or surrogate depression or anxiety.

The researchers noted that the patients in this study may have been less ill than other populations and that the sample size was small, among other limitations. They concluded that ICU patients and surrogates may often disagree with clinicians regarding appropriateness of treatment and that such disagreements may be linked to disagreements about prognosis and to lower patient and surrogate satisfaction. “Understanding differences in perception of the appropriateness of treatment among patients, family members, and clinicians may play an important role in developing effective interventions to improve communication, satisfaction, and goal concordant treatment,” the authors wrote.

Hospitalizations for heart failure, but not MI, more common during influenza season

Higher influenza activity in a given month was associated with an increase in hospitalizations for heart failure, but not myocardial infarction, within the same month, a recent study found.

Image by Getty Images
Image by Getty Images

Researchers used surveillance data from the Atherosclerosis Risk in Communities study, which included 451,588 people ages 35 to 84 years living in Jackson, Miss., some city suburbs of Minneapolis, and two rural counties in North Carolina and Maryland. From October 2010 to September 2014, researchers compared the monthly frequency of hospitalizations for heart failure and myocardial infarction to influenza activity (defined as the percentage of visits to sentinel clinicians for influenza-like illness by state), as reported by the CDC and Prevention Surveillance Network. Results were published March 27 by JAMA Cardiology.

During the study period, there were 4,321 hospitalizations due to heart failure (47.3% women; 53.3% white) and 3,541 due to myocardial infarction (45.1% women, 57.4% white). On average, a 5% monthly absolute increase in influenza activity was associated with a 24% increase in heart failure hospitalization rates within the same month after multivariable adjustment (incidence rate ratio, 1.24; P<0.001). These results suggest that in a month with high influenza activity, about 19% (95% CI, 10% to 28%) of heart failure hospitalizations could be attributable to influenza.

On the other hand, overall influenza activity was not associated with myocardial infarction hospitalizations (incidence rate ratio, 1.02; P=0.72). During the two most severe influenza seasons (2010-2011 and 2012-2013), however, there was a 22% and 20% temporally associated increased risk of heart failure (incidence rate ratios, 1.22 [P=0.04] and 1.20 [P=0.02], respectively) as well as a positive but not significant association with myocardial infarction in the 2012-2013 season (incidence rate ratio, 1.12; P=0.25). Influenza activity in the months before hospitalization was not associated with either outcome.

The study authors noted limitations, such as the fact that individual influenza illness and vaccination status of hospitalized patients were not known. They also noted the possibility that other infectious causes may parallel influenza activity and account for some of the association.

“These data suggest that while hospitalizations for influenza and pneumonia, and deaths associated with these, are greatest during severe influenza seasons, acute [cardiovascular] events are also likely increased during periods of peak influenza activity. . . . Addressing influenza activity may be valuable in efforts to prevent [heart failure] hospitalizations,” the authors concluded.

Score predicts risk of stroke within a year of sepsis hospitalization

Researchers developed a score to help predict patients' risk of stroke in the year after a sepsis hospitalization.

They used ICD-9 codes from the 2007–2009 California State Inpatient Database from the Health Care Utilization Project to identify 121,947 adult patients who were hospitalized with sepsis or bloodstream infection and survived until discharge. The primary outcome was a primary diagnosis of ischemic or hemorrhagic stroke on a subsequent hospitalization within a year. Results were published by Stroke on March 21 and appeared in the May issue.

Overall, 0.5% (n=613) of the patients were diagnosed with stroke within a year of their hospitalization for sepsis or bloodstream infection. The authors identified significant predictors of these strokes and developed a risk score based on them: valvular heart diseases (1 point), congestive heart failure (1 point), renal failure (1 point), lymphoma (2 points), peripheral vascular diseases (2 points), pulmonary circulation disorders (2 points), and coagulopathy (3 points). Every one-point increase in a patient's score was associated with a 43% increase in the risk of stroke (odds ratio, 1.43; 95% CI, 1.37 to 1.48), with greater effect among younger patients. The C statistic for the receiver-operating characteristic curve was 0.68.

Coagulopathy was the most significant predictor of stroke, carrying a threefold increase in ischemic stroke risk and a sevenfold increase in hemorrhagic stroke risk. The observed linking of sepsis and stroke by coagulopathy is supported by current biological knowledge, said the authors, who noted that previous research has found more than 80% of sepsis patients to have either clinical or subclinical coagulopathy.

The study's results could have clinical implications if validated in a larger population, the authors added. “If risk factors for stroke among patients with sepsis or bloodstream infection can be reliably identified, then clinicians may consider treating those sepsis patients at highest risk with therapies, such as aspirin or statins, that have been proven to prevent cerebrovascular disease in other settings. An episode of sepsis may provide a particularly appropriate time to calculate a patient's cardiovascular risk profile,” they wrote.

The study had a number of limitations, including lack of data on the timing of sepsis and studied comorbidities and past medication usage or infectious conditions. It was also subject to the risks of covariate misclassification and collider bias. Finally, the overall risk of stroke among these sepsis patients was low, suggesting that the link may be of limited clinical significance, the authors said.