Pneumonia therapies, hospital flu spread, and more

Research summaries from ACP Hospitalist Weekly.


Antibiotics, corticosteroids most effective for COPD exacerbations, review finds

Antibiotics and systemic corticosteroids reduce treatment failure in adults with chronic obstructive pulmonary disease (COPD) exacerbations, but there is little evidence to support other pharmacologic interventions, a systematic review and meta-analysis found.

Researchers looked at 68 randomized controlled trials that enrolled 10,758 adults with COPD exacerbations treated in non-ICU inpatient and outpatient settings and compared pharmacologic therapies, placebo, or usual care. Nearly two-thirds of the trials involved patients treated in the hospital or ED, and follow-up ranged from one to 12 days. The main outcomes were clinical resolution, treatment failure, repeated exacerbations, death, quality of life, hospital admission/readmission, ICU admission, and need for intubation. The researchers also looked at other outcomes, including functional capacity and adverse events. Results were published online on Feb. 24 by Annals of Internal Medicine and appeared in the March 17 issue.

Compared with placebo or management without antibiotics, antibiotics given for three to 14 days were more likely to resolve the exacerbation (odds ratio [OR], 2.03; 95% CI, 1.47 to 2.80; moderate strength of evidence) and less likely to have treatment failure (OR, 0.54; 95% CI, 0.34 to 0.86; moderate strength of evidence), independent of severity of exacerbations in both inpatients and outpatients. Compared with placebo in inpatients and outpatients, systemic corticosteroids given for nine to 56 days were associated with less treatment failure at the end of the intervention (OR, 0.01; 95% CI, 0.00 to 0.13; low strength of evidence) but also with more total and endocrine-related adverse events. Compared with placebo or usual care in inpatients, other pharmacologic interventions (aminophyllines, magnesium sulfate, anti-inflammatory agents, inhaled corticosteroids, and short-acting bronchodilators) had insufficient evidence support, were not effective (with the exception of the mucolytic erdosteine), or were associated with improvement only in lung function.

Among other limitations, the analysis did not include trials published in languages other than English, and only 65% of included trials reported adverse events, the study authors said. In addition, for most interventions, only one or two trials reported the main outcomes of interest, they said.

“The results of this systematic review support use of antibiotics and systemic corticosteroids in COPD exacerbations, independent of severity,” they wrote, adding that they do not recommend using other pharmacologic interventions to treat exacerbations.

Study finds no mortality benefit of anti-MRSA therapy for inpatients with pneumonia

There may be no mortality benefit of empirical methicillin-resistant Staphylococcus aureus (MRSA) therapy over standard antibiotics for patients hospitalized with pneumonia, including those with MRSA risk factors, a retrospective cohort study found.

Researchers used treatment-weighted propensity score matching to compare mortality among patients hospitalized for pneumonia who received empirical anti-MRSA therapy (with or without standard antibiotics) within the first day of hospitalization versus standard therapy alone. The study included all hospitalizations in which patients received either therapy for community-onset pneumonia in the Veterans Health Administration health care system from 2008 through 2013. The main outcome was 30-day all-cause mortality after adjustment for patient comorbidities, vital signs, and laboratory results. Subgroup analyses included patients with initial ICU admission, MRSA risk factors, positive results of a MRSA surveillance test, and positive results of a MRSA admission culture. Results were published on Feb. 17 by JAMA Internal Medicine.

Of 88,605 hospitalized patients (98% men; median age, 70 years), 33,632 (38%) received empirical anti-MRSA therapy, and 8,929 (10%) died within 30 days. Vancomycin accounted for 98% of anti-MRSA therapy administered, and standard antibiotics were typically a fluoroquinolone or the combination of a beta-lactam plus a macrolide or doxycycline. Compared with standard therapy alone, empirical anti-MRSA therapy plus standard therapy was associated with increases in mortality (adjusted risk ratio [RR], 1.4; 95% CI, 1.3 to 1.5), development of kidney injury (adjusted RR, 1.4; 95% CI, 1.3 to 1.5), and secondary infection with Clostridioides difficile (adjusted RR, 1.6; 95% CI, 1.3 to 1.9), vancomycin-resistant Enterococcus species (adjusted RR, 1.6; 95% CI, 1.0 to 2.3), or gram-negative bacilli (adjusted RR, 1.5; 95% CI, 1.2 to 1.8). Similar associations between anti-MRSA therapy and 30-day mortality were found in an instrumental variable analysis (adjusted RR, 1.6; 95% CI, 1.4 to 1.9) and among patients admitted to the ICU (adjusted RR, 1.3; 95% CI, 1.2 to 1.5), at high risk for MRSA (adjusted RR, 1.2; 95% CI, 1.1 to 1.4), or with MRSA detected on surveillance testing (adjusted RR, 1.6; 95% CI, 1.3 to 1.9). No significant association was found in patients with MRSA detected on culture (adjusted RR, 1.1; 95% CI, 0.8 to 1.4).

Limitations of the study include the possibility of residual confounding and the reliance on diagnosis codes assigned at the end of the hospitalization, among other limitations, the study authors noted. An accompanying editorial said the findings must be interpreted with caution, given that patients treated with anti-MRSA regimens were sicker than those treated with standard regimens. Nonetheless, the findings echo the results of other studies, and MRSA is rare in community-onset pneumonia (only 2% of the study population had MRSA detected on culture), the editorial said.

“The messages are clear: empirical coverage with broad-spectrum agents is not indicated in most patients being treated for pneumonia, and if the diagnosis of pneumonia is uncertain, there is no harm and likely some benefit in taking some time to acquire more diagnostic clarity before treating (such as further imaging, laboratory testing, treating concurrent conditions, and/or observation), so long as the patient is clinically stable,” the editorialist wrote.

In-ward transmission of influenza frequent, according to a single-center study

Influenza may be frequently transmitted between hospitalized adults, a retrospective study found.

Image by Getty Images
Image by Getty Images

Researchers identified hospitalized adults with laboratory-verified influenza A virus infection during the 2016-2017 influenza season at a large hospital in Sweden. They characterized cases according to age, sex, comorbidity, receipt of antiviral therapy, viral load expressed as cycle threshold values, length of stay, 30-day mortality, and whether the infection met criteria for a health care-associated influenza (influenza-like illness or acute respiratory infection with onset of symptoms 48 hours or more after hospital admission or 48 hours or less after discharge).

To detect clustering, respiratory samples positive for influenza A virus collected on the same wards within seven days were chosen for whole genome sequencing, and phylogenetic analysis was performed. Concurrent influenza A virus strains from patients with community-acquired infection were included for reference. Results were published online on Feb. 3 by Clinical Infectious Diseases.

Overall, 435 cases of influenza A virus infection were identified, and 114 (26%) met the criteria for health care-associated infection. Seventy-four (65%) of the health care-associated infections possibly involved in-ward transmission. The overall 30-day mortality rate was higher in patients with health care-associated infection compared to those with infection that wasn't associated (9.6% vs. 4.6%), although the difference was not significant in multivariable analysis, where age was the only independent risk factor for death (P<0.05).

Whole genome sequencing was successful in 124 hospital cases, and 60 (48%) sequenced samples belonged to an in-ward cluster or pair. Analysis identified eight closely related clusters (involving three or more cases) and another 10 pairs of strains supporting in-ward transmission. Clusters were distributed between December and May, and a total of 20 different wards were affected, 14 (70%) of which were internal medicine or cardiology wards. In three cases, there was a close relationship between an in-ward cluster and a single strain from another ward (recent transfer from an affected ward in one case and adjacent localization of wards in the other two cases). A few reference cases from the community corresponded well in time to an in-ward cluster.

The study authors noted that all data were collected retrospectively, and no information was available regarding adherence to infection control measures. They added that they did not assess potential links between patients and health care workers.

“In conclusion, we found that in-ward transmission of [influenza A virus] occurs frequently and that health-care-associated influenza may have a severe outcome,” the authors wrote. “[Whole genome sequencing] may be used for outbreak investigations as well as for evaluation of the effect of preventive measures.”

Insulin order set reduced risk of hypoglycemia from hyperkalemia treatment

An order set for insulin administration reduced hypoglycemia among inpatients being treated for hyperkalemia, a recent study found.

The observational, prospective study included adult nonobstetric patients treated for acute hyperkalemia (serum potassium ≥5.1 mEq/L) on medical, ICU, and surgical units at the University of California, San Francisco, Medical Center between January 2016 and September 2017. Results were published in the February Journal of Hospital Medicine.

In January 2016, the hospital implemented an initial hyperkalemia treatment order set, which added blood glucose checks before insulin injection and at one, two, four, and six hours after insulin. In March 2017, it was replaced with a new order set, which added weight-based dosing of insulin, alerts to identify patients at higher risk of hypoglycemia, and tools to guide decision making based on blood glucose levels. The study compared 225 insulin orders made under the old order set with 145 issued under the new one.

Under the first order set, the iatrogenic hypoglycemia rate (defined as a glucose level <70 mg/dL [3.9 mmol/L]) was 21%, with 92% of episodes occurring within three hours of first insulin treatment. The study identified some risk factors for hypoglycemia: decreased renal function (serum creatinine >2.5 mg/dL), a high dose of insulin (>0.14 units/kg), and retreatment when blood glucose was below 140 mg/dL (7.8 mmol/L). Under the updated order set, the rate of iatrogenic hypoglycemia decreased to 10%. Rates of severe hypoglycemia (glucose level <40 mg/dL [2.2 mmol/L]) were 5% and 2% under the respective order sets.

The study also showed a trend toward improved hyperglycemia rates and significantly decreased serum potassium levels under the revised order set. The latter might have been due to weight-based dosing of insulin, the study authors speculated.

“Prior studies have likely underestimated the incidence of hyperkalemia treatment-associated hypoglycemia as glucose levels are rarely checked within three hours of insulin administration,” they wrote. The authors noted that greater adherence to the order set would probably have provided even more improvement in glycemic control. In all five of the 14 hypoglycemic events under the updated order set, a recommended bolus of dextrose was not provided at the appropriate time. The order set generally was only used for about 75% of hyperkalemia patients, and adherence to the glucose checks was suboptimal, the authors noted.

Such frequent glucose monitoring during hyperkalemia treatment (two checks in a six-hour window) may be difficult in some clinical environments, but checking blood sugar before insulin administration and within four hours after should be a priority, they said. The authors also noted that they plan to update their order set with a best practice alert about the additional recommended dextrose bolus.

Physical activity level in the hospital linked to outcomes at discharge in elderly patients

Elderly adults with low levels of in-hospital activity were at higher risk for bedsores and were less likely to return home after discharge, a recent study found.

Researchers in Switzerland performed a cohort study in hospitalized patients ages 65 years and older to assess levels of physical activity in the hospital and their association with functional decline and other outcomes at discharge. Physical activity during the daytime and over 24-hour periods was assessed by wrist accelerometers. The study's main outcome measures were functional decline (defined as a decrease of at least 5 points on the modified Barthel Index), risk for bedsores, length of stay, and inability to return home. The results were published Jan. 31 by JAMA Network Open.

Image by Getty Images
Image by Getty Images

One hundred seventy-seven patients who were admitted from Feb. 1 through Nov. 30, 2018, were included in the study. One hundred six (59.9%) were men, and the median age was 83 years. Patients were more likely to have lower physical activity levels during hospitalization if they had used walking aids before admission, if they were admitted for a reason related to functional decline, and if they were prescribed physiotherapy. At hospital discharge, 63 patients (35.6%; 95% CI, 25.6% to 43.1%) had functional decline, 78 (44.1%; 95% CI, 36.6% to 51.7%) had bedsore risk, and 82 (46.3%; 95% CI, 38.8% to 54.0%) were not able to return home. No association was found between daytime and 24-hour physical activity levels and functional decline (P=0.69 and P=0.45, respectively) or length of stay (P=0.93 and P=0.52, respectively) in multivariate analysis. Patients who were at risk for bedsores had significantly lower daytime and 24-hour physical activity levels compared with those who were not at risk (P=0.008 and P=0.01, respectively), and patients who could return home had significantly higher daytime and 24-hour physical activity levels than those who could not (P=0.04 and P=0.009, respectively).

The authors noted that the study involved only one hospital ward at one university hospital, that risk of bedsores rather than presence of bedsores was assessed, and that activities of daily living and physical activity can be interdependent, complicating interpretation of the results. However, they concluded that among the elderly, less physical activity during a hospital stay is associated with increased risk for bedsores and inability to return home at discharge but not with hospital-acquired functional decline or length of stay. “Future studies should try to estimate the minimum amount of [physical activity] needed to prevent increased in-hospital morbidity or [length of stay] in elderly hospitalized patients,” the authors wrote.

Frailty assessment index based on lab tests may help predict outcomes in hospitalized older adults

A frailty assessment tool based on laboratory tests at hospital admission may help predict adverse outcomes in older adults when combined with chronic frailty assessment, according to a recent study.

Researchers in the United Kingdom created a frailty index using data from 27 laboratory values routinely assessed within 72 hours of hospital admission. This frailty index (FI-Laboratory) was developed in a prospective cohort of older adults consecutively admitted to the acute geriatric medicine service at a large tertiary care hospital between April 2015 and January 2017. Hospital admissions shorter than 24 hours were excluded.

Acute health status was assessed with FI-Laboratory, while chronic health status was assessed with the Clinical Frailty Scale (CFS). To obtain a score on FI-Laboratory, the proportion of abnormal test results was calculated from the total number of tests ordered. The CFS is scored from 1 to 9, with 1 indicating very fit and 9 indicating terminally ill. The study's main outcomes were total hospital days, need for a higher level of care, readmission during the enrollment period, and mortality from the date of hospital admission to Dec. 11, 2018. Results were published by CMAJ on Jan. 6.

Overall, 2,552 admissions for 1,750 patients were included in the analysis, and laboratory values for the index were available for 2,254 admissions (88.3%, 1,580 patients). Patients' average age was 84.8 years, and 55.3% were women. The FI-Laboratory correlated weakly with the CFS. An increase in CFS score and three additional abnormal values on FI-Laboratory were each associated with more hospital days (rate ratios, 1.43 [95% CI, 1.35 to 1.52] and 1.47 [95% CI, 1.41 to 1.54], respectively), more frequent discharge to a higher level of care (odds ratios, 1.39 [95% CI, 1.27 to 1.52] and 1.30 [95% CI, 1.16 to 1.47]), higher readmission rates (hazard ratios, 1.26 [95% CI, 1.17 to 1.37] and 1.18 [95% CI, 1.11 to 1.26]), and higher mortality rates (hazard ratios, 1.39 [95% CI, 1.28 to 1.51] and 1.45 [95% CI, 1.37 to 1.54]).

The researchers noted that their study was done at only one center and that the FI-Laboratory score was missing in 11.6% of cases, among other limitations. However, they concluded that this frailty index based on commonly obtained laboratory values can help in early identification of acutely ill adults who are admitted to the hospital and are at risk for adverse outcomes. Future research should focus on whether the FI-Laboratory can improve clinical outcomes and whether changes in the index during hospital admission provide prognostic information, the authors wrote.

GI bleeding prophylaxis recommended in high-risk critically ill patients

Critically ill patients considered to be at high risk for gastrointestinal (GI) bleeding may benefit from prophylaxis, according to a Rapid Recommendation statement from the BMJ.

The statement was developed by a guideline panel and is based on a systematic review and meta-analysis that evaluated the efficacy and safety of GI bleeding prophylaxis with proton-pump inhibitors (PPIs), histamine-2 receptor antagonists (H2RAs), or sucralfate, as well as no prophylaxis, in critically ill patients in the intensive care unit. Both the statement and the review and meta-analysis were published by the BMJ on Jan. 6.

The review and meta-analysis included 72 trials involving 12,660 patients and found that both PPIs and H2RAs decreased risk for clinically important GI bleeding, especially in those at higher risk. The review also found that both types of treatment may increase risk for pneumonia versus no prophylaxis but probably do not affect mortality or other important outcomes. PPIs appeared to reduce bleeding risk more than H2RAs, the review noted.

The guideline panel made a weak recommendation for use of GI bleeding prophylaxis in critically patients at high risk for clinically important GI bleeding, defined as greater than 4%, and against prophylaxis in those at lower risk. The panel suggested that a PPI be used rather than an H2RA (weak recommendation) and recommended against using sucralfate (strong recommendation). The panel noted that clinicians should ensure acid suppression medications are stopped after patients recover from their critical illness or are no longer considered at risk for GI bleeding, unless another indication for prophylaxis is present.

“In most critically ill patients, the reduction in clinically important gastrointestinal bleeding from gastric acid suppressants is closely balanced with the possibility of pneumonia,” the statement authors wrote. “Clinicians should consider individual patient values, risk of bleeding, and other factors such as medication availability when deciding whether to use gastrointestinal bleeding prophylaxis.”

tPA after DOAC not associated with increased risk for intracerebral hemorrhage

Patients with acute ischemic stroke who have received a direct oral anticoagulant (DOAC) before tissue plasminogen activator (tPA) bolus and in the absence of reversal agents may not have an increased risk of symptomatic intracerebral hemorrhage after IV thrombolysis compared to those who have received warfarin or no prior anticoagulation, a meta-analysis found.

After a systematic review of the literature, researchers conducted a meta-analysis to compare the rate of symptomatic intracerebral hemorrhage in patients pretreated with a DOAC versus stroke patients without prior anticoagulation and patients on warfarin with an international normalized ratio below 1.7. The review was limited to patients who received IV thrombolysis within 48 hours of last DOAC intake.

Additional meta-analyses were conducted for the following subgroups: 1) administration of a DOAC within 48 hours versus an unknown interval before IV thrombolysis and 2) consideration of symptomatic intracerebral hemorrhage outcome according to the National Institute of Neurological Disorders and Stroke (NINDS) versus the European Cooperative Acute Stroke Study II (ECASS-II) criteria. Results were published online on Dec. 30, 2019, by Stroke and appeared in the February issue.

The meta-analysis included six studies with a total of 52,823 patients with acute ischemic stroke. Studies included 366 patients on DOACs, 2,133 on warfarin, and 50,324 without prior anticoagulation. There was no additional risk of symptomatic intracerebral hemorrhage following IV thrombolysis among patients taking DOACs within 48 hours of tPA bolus. For DOACs versus warfarin, the odds ratio (OR) was 0.55 (95% CI, 0.19 to 1.59) with the NINDS criteria and 0.77 (95% CI, 0.28 to 2.16) with the ECASS-II criteria. For DOACs versus no anticoagulation, the OR was 1.23 (95% CI, 0.46 to 3.31) with the NINDS criteria and 0.87 (95% CI, 0.32 to 2.41) with the ECASS-II criteria. Similarly, there was no additional risk detected with no time limit between last DOAC intake (ORs for DOACs vs. warfarin, 0.85 [95% CI, 0.49 to 1.45] with NINDS and 1.11 [95% CI, 0.67 to 1.85] with ECASS-II; ORs for DOACs vs. no anticoagulation, 1.17 [95% CI, 0.43 to 3.15] with NINDS and 0.87 [95% CI, 0.33 to 2.41] with ECASS-II). There was no evidence of heterogeneity across included studies.

Among other limitations, there were inadequate data regarding the exact time of last DOAC intake, and the sample size of patients who received a DOAC was considerably smaller than that of the control groups, which may have affected the power of the study, the authors noted. In addition, because the safety of IV thrombolysis was evaluated in selected study participants with acute ischemic stroke who were pretreated with DOACs, the findings are not easily generalizable to other groups, they said.

A 2018 American Heart Association/American Stroke Association guideline allowed IV tPA after 48 hours following last DOAC intake, the study authors noted. “The results of our study indicate that prior intake of DOACs within 48 hours of tPA bolus is not associated with an increased risk of symptomatic hemorrhagic transformation in selected [acute ischemic stroke] patients,” they concluded, adding that additional studies are required before generalizing these results.