In the News

Rehospitalization after MI, hospitalists and outcomes, and more.


Majority of rehospitalizations after MI are for unrelated causes

More than 40% of 30-day rehospitalizations after a myocardial infarction (MI) were related to the MI, while the rest were due to other causes or for unclear reasons, a study reported.

In addition, comorbid conditions, longer length of stay, and complications of angiography and revascularization or reperfusion were associated with increased 30-day rehospitalization risk, according to results published in the July 3 Annals of Internal Medicine. The retrospective cohort study used a population-based registry in Olmsted County, Minn., of 3,010 patients who were hospitalized with first-ever MI from 1987 to 2010.

A total of 643 rehospitalizations occurred among 561 (18.6%) patients within 30 days of discharge. The most common reasons were ischemic heart disease, respiratory or chest symptoms, and heart failure. Overall, 42.6% of the rehospitalizations were related to the first MI or its treatment, whereas 30.2% were unrelated and 27.2% had an unclear relationship (most often cited as atypical chest pain). Unrelated rehospitalizations were more common in women and patients with non-ST-elevated MI.

Among the conditions independently associated with increased risk for rehospitalization after MI were:

  • diabetes mellitus (hazard ratio [HR], 1.34; 95% CI, 1.10 to 1.63),
  • chronic obstructive pulmonary disease (HR, 1.43; 95% CI, 1.15 to 1.79),
  • anemia (HR, 1.25; 95% CI, 1.03 to 1.50),
  • Killip class 2 to 4 at presentation (HR, 1.22; 95% CI, 1.01 to 1.46),
  • four- to seven-day length of stay during index MI hospitalization (HR, 1.34; 95% CI, 1.05 to 1.70),
  • more than seven-day length of stay during index hospitalization (HR, 1.65; 95% CI, 1.27 to 2.14),
  • complication of angiography during index hospitalization (HR, 2.40; 95% CI, 1.43 to 4.01) and
  • complication of reperfusion or revascularization during index hospitalization (HR, 2.12; 95% CI, 1.61 to 2.80).

“[P]atients affected by a vascular or bleeding complication, stroke, or acute kidney injury after angiography or revascularization or reperfusion represent a high-risk population for rehospitalization,” the authors concluded. “Prevention of complications and close follow-up for patients who have had a complication may be of particular importance for preventing rehospitalizations… .[P]atients with MI have many comorbid conditions, which may affect rehospitalization. The management of patients with multiple comorbid conditions and competing risks is of increasing importance as the population ages.”

Hospitalist presence didn't impact death, readmission rates for common conditions

Hospitals with hospitalists fared the same on mortality and readmission measures for three specific conditions as those without hospitalists, with one exception, a recent study found.

Researchers identified hospitals with 25 or more acute myocardial infarction (AMI) discharges in 2008 using American Hospital Association data, and called them up to six times until they reached a target sample size of 600 participants. They administered surveys to hospital leaders via phone and fax between February 2010 and January 2011 to obtain data on the presence or absence of hospitalists and characteristics of the hospitalist services. For each admission, researchers calculated 30-day risk-standardized mortality and readmission rates for AMI, heart failure and pneumonia for Medicare patients age 65 and up. Presence of hospitalists was correlated with hospital- and patient-level characteristics, and with performance on each outcome measure. Researchers also assessed the relationship between the percentage of patients admitted by hospitalists and each outcome measure.

Of the 598 survey respondents, 72% (n=429) reported use of hospitalist services. There was no statistically significant difference between facilities that used hospitalists or didn't on any outcome measure except risk-stratified readmissions for heart failure, in multivariable models or bivariate analysis. Sites with hospitalists had a lower heart failure readmission rate than those without (24.7% vs. 25. 4%; P<0.0001 on bivariate analysis); the researchers noted the effect size of this finding was small. In the hospitals that used hospitalists, there was no change in any outcome measure associated with increasing percentage of patients admitted by hospitalists (versus admission by another type of clinician). Results were published in the July/August Journal of Hospital Medicine.

Despite concerns that hospitalists negatively impact continuity of care, these data didn't find an association between higher readmission rates and use of hospitalists, the researchers noted. Study limitations include the lack of a standardized industry definition of “hospitalist,” which could have affected survey response validity; reliance on self-reported data; and the fact that hospitalists may be less likely to provide care for AMI and heart failure than other conditions, so outcome data may not fully capture their influence. “It is likely that multiple factors contribute to performance on outcome measures, including type and mix of hospital personnel, patient care processes and workflow, and system level attributes,” the authors concluded. “Interventions leading to improvement on core outcome measures are more complex than simply having a hospital medicine program.”

IDSA diabetic foot infection guidelines cover diagnosis and treatment

A clinical practice guideline from the Infectious Diseases Society of America (IDSA) provides detailed instruction on diagnosis and treatment of diabetic foot infections.

The guideline notes that not all foot wounds in diabetic patients are infected. Infections should be diagnosed by the presence of at least two classic symptoms or signs of inflammation or purulent secretions, the IDSA said. Infections can then be classified into mild, moderate or severe. These classifications, along with vascular assessment, will help determine the need for hospitalization, surgery or amputation. Imaging is helpful in most cases, the guideline said, beginning with plain radiographs and progressing to magnetic resonance imaging if more information is needed. The guidelines were published June 15 by Clinical Infectious Diseases.

Hospital admission is recommended for all patients with a severe infection, selected patients with a moderate infection with complicating features, patients unable to comply with outpatient treatment regimens, or patients who fail to improve on outpatient therapy. Before discharge, patients should be clinically stable; have any urgently needed surgery; achieve glycemic control; be able to manage outpatient self-care; and have a well-defined plan, an off-loading scheme, specific wound care instructions and planned follow-up.

For either in- or outpatient treatment, the guideline also provides instructions on culturing diabetic foot infections and prescribing antibiotics based on the results of cultures. Wounds with no evidence of infection do not require antibiotic therapy, and aerobic gram-positive cocci are the most common causative organisms of infections, the IDSA said.

Recommendations on how to diagnose and treat osteomyelitis are also included in the guideline. Multidisciplinary team care is recommended whenever possible, and the guideline includes advice on when to refer care to specialists. The appropriate use of surgical interventions (which are required for most diabetic foot infections) and best methods of wound care are also described in the guideline, which is presented as a series of questions and answers.

Risk model may help assess which heart failure patients to admit, discharge

A mortality risk model may be useful in assessing which heart failure patients in the emergency department need to be hospitalized and which can be treated and sent home.

Researchers abstracted clinical data from 12,591 patients in Ontario, Canada who presented in an emergency department (ED) for heart failure and were discharged or admitted from April 1, 2004 through March 2007. They developed a multivariate risk index for death in the seven days after ED presentation using initial vital signs, clinical and presentation features, and easily available laboratory tests. Of the initial patients, 7,433 comprised the derivation cohort and 5,158 comprised the validation cohort. Results were published online June 4 by Annals of Internal Medicine.

Predictors of seven-day mortality, which were used to create the risk score, were older age (P<0.001), transport by emergency medical services (P<0.001), lower systolic blood pressure (P<0.001), heart rate outside the range of 80 to 120 beats/min (P=0.017), lower oxygen saturation (P=0.033), higher creatinine (P<0.001), elevated troponin level (P<0.001), active cancer (P=0.002), use of metolazone at home (P=0.036) and potassium ≥4.6 mmol/L (P=0.012) or ≤3.9 mmol/L (P=0.70). Mortality rates in the highest two deciles were 3.5% and 8.5%, respectively—more than 10-fold and 25-fold higher than the 0.3% event rate in the two lowest risk quintiles.

One of the model's strengths is that it was created by including all patients who presented to the ED with heart failure, which makes it more useful to guide decisions on whether to admit or discharge patients than a model derived solely from hospitalized patients, the authors noted. The model also addresses a need for methods of acute prognostication that don't depend on knowing ventricular function status and the cause of heart failure, since these may not be known in the ED setting, they said. The model should be used in conjunction with clinical judgment of other factors, such as symptom improvement, in deciding whether to admit or discharge patients, the authors said.

Observation care rate rises, inpatient rate drops, for Medicare patients

The prevalence of observation care increased between 2007 and 2009, while it fell for inpatient care, a recent analysis of Medicare patients found.

Researchers examined Medicare enrollment and claims data from 2007 through 2009. The study population was constructed on a monthly basis; each month, about 29 million beneficiaries met all inclusion criteria for the analysis. Researchers used revenue center codes and the Healthcare Common Procedure Coding System to identify observation stays, and to count the total hours for which observation services were provided. Results were published in the June Health Affairs.

The prevalence of observation stays rose from an average of 2.3 per 1,000 beneficiaries per month in 2007 to 2.9 in 2009. Meanwhile, the prevalence of inpatient stays declined from 23.9 per 1,000 beneficiaries per month to 22.5 in 2009. This means the ratio of observation stays to inpatient admissions rose 34%, from an average of 86.9 observation stays per 1,000 inpatient admissions monthly in 2007 to 116.6 in 2009. The number of hours a patient was held for observation per episode also increased by more than 7%, from an average of 26.2 hours in 2007 to 28.2 hours in 2009. In 2007, 23,841 beneficiaries were held in observation for at least 72 hours, compared to 44,843 in 2009—an 88% increase.

The rising prevalence of hospital observation care is in line with a general shift of Medicare-covered services from inpatient to outpatient settings—which is due in part to policies that affect reimbursement, the authors noted. “Facing more stringent criteria for inpatient admissions and uncertainties over the prospects of retroactive payment denial, physicians may choose to place their patients under observation more often than they would otherwise,” they wrote. Readmission penalties may also encourage hospitals to classify patients as under observation rather than inpatient, they noted. Doing so can expose patients to greater out-of-pocket expenses if they are eventually admitted to nursing homes, because they haven't hit the three-day inpatient stay requirement for Medicare coverage of nursing homes, the authors noted. More clarity is needed in clinical practice and Medicare guidelines regarding observation care, they concluded.

Postdischarge med errors common in heart patients despite pharmacist intervention

Postdischarge medication errors were common in patients with acute coronary syndromes (ACS) or acute decompensated heart failure even with a pharmacist-led intervention, a recent study has found.

Researchers performed a randomized, controlled trial at two tertiary care academic hospitals to determine whether a tailored, pharmacist-led intervention would affect clinically important medication errors after hospital discharge among patients with ACS or acute decompensated heart failure. The intervention consisted of pharmacist-assisted medication reconciliation, inpatient counseling by a pharmacist, low-literacy adherence aids, and individualized postdischarge follow-up by telephone.

The number of clinically important medication errors per patient in the first 30 days after discharge, including preventable or ameliorable adverse drug events (ADEs) and potential ADEs caused by discrepancies or lack of adherence, was the study's primary outcome. Preventable or ameliorable ADEs, potential ADEs caused by discrepancies or lack of adherence, and preventable or ameliorable ADEs considered to be serious, life-threatening or fatal were the secondary outcomes. The study appeared in the July 3 Annals of Internal Medicine.

Four hundred thirty patients were assigned to the intervention group, and 432 were assigned to usual care, defined as medication reconciliation and discharge counseling by the treating physicians and nurses. Seven patients in the intervention group and four in the usual care group died in the hospital or withdrew their consent, meaning 851 patients were included in the intention-to-treat analysis. The patients' mean age was 60 years, and 41.4% were women. Health literacy was adequate in approximately 10% and marginal in 8.7%; in addition, 11.5% had some cognitive impairment. Sixty-one percent had only ACS, 31% had only acute heart failure, and 7% had both.

Overall, 432 patients (50.8%) had at least one clinically important medication error, 22.9% of which were considered serious and 1.8% of which were considered life-threatening. Two hundred fifty-eight patients (30.3%) had ADEs and 253 (29.7%) had potential ADEs. The per-patient numbers of clinically important medication errors and ADEs were not significantly affected by the intervention (unadjusted incidence rate ratios, 0.92 [95% CI, 0.77 to 1.10] and 1.09 [95% CI, 0.86 to 1.39], respectively). Potential ADEs tended to be less common in the intervention group (unadjusted incidence rate ratio, 0.80 [95% CI, 0.61 to 1.04]).

The authors acknowledged that their study involved patients from only two hospitals and that the results therefore may not be generalizable, among other limitations. However, they concluded that clinically important medication errors are common within 30 days after hospitalization for a cardiac condition, and that the pharmacist-led intervention they tested did not improve overall medication safety. “Reducing ADEs and potential ADEs in the postdischarge period is becoming more critical as hospitals have increasing financial penalties tied to rehospitalization rates,” they wrote. “Further work is needed to develop and test interventions in this setting, including strategies for higher-risk populations, as well as additional methods, such as postdischarge medication reconciliation…or closer postdischarge surveillance.”

Prepare for carbapenem-resistant Enterobacteriaceae

Two U.S. cases of a particularly problematic strain of carbapenem-resistant Enterobacteriaceae (CRE) were recently described by the Centers for Disease Control and Prevention (CDC).

The two patients with CRE containing New Delhi metallo-beta-lactamase (NDM) were treated at a Rhode Island hospital in March. The first patient had recently been hospitalized in Vietnam and was found to have a Klebsiella pneumoniae isolate containing NDM susceptible only to tigecycline, colistin, and polymyxin B. The second patient was being treated on the same hematology/oncology unit. These two patients bring to 13 the number of cases of NDM reported in the U.S., according to the CDC's Morbidity and Mortality Weekly Report.

The CDC called for a robust infection control effort to limit or slow the spread of all CRE, including NDM, at the local, national and international levels. Acute- and chronic-care facilities should have a written plan that clearly describes how they will detect CRE and limit transmission before it becomes endemic, the agency advised.

MRSA rates in the U.S. appear to be decreasing, study indicates

Rates of methicillin-resistant Staphylococcus aureus (MRSA) appear to be decreasing in the U.S., according to a recent study.

Researchers used data from U.S. Department of Defense beneficiaries to examine incidence and trends of community- and hospital-onset S. aureus bacteremia and skin and soft-tissue infections (SSTIs), including the proportion due to MRSA. Beneficiaries included active duty members, retirees, guard and reservists, and their immediate family. S. aureus blood, wound or abscess cultures were classified as community- or hospital-onset infections and as methicillin-susceptible S. aureus or MRSA. Main outcome measures were unadjusted incidence rates per 100,000 person-years, proportion of infections due to MRSA, and annual trends. The study results appeared in the July 4 Journal of the American Medical Association.

The Department of Defense databases included 62,326 positive blood cultures and 181,317 positive wound or abscess cultures from 2005 through 2010. Among these, 12% of blood cultures and 62% of wound or abscess cultures yielded S. aureus isolates. Over 56 million person-years (47 million on nonactive duty and 9 million on active duty), 2,643 blood cultures and 80,281 wound or abscess cultures tested positive for S. aureus. Annual incidence rates were 3.6 to 6.0 per 100,000 person-years for S. aureus bacteremia and 122.7 to 168.9 per 100,000 person-years for SSTIs due to S. aureus. From 2005 to 2010, a decrease was seen in annual incidence rates of community-onset MRSA bacteremia (1.7 per 100,000 person-years vs. 1.2 per 100,000 person-years, respectively; P=0.005 for trend) and hospital-onset MRSA bacteremia (0.7 per 100,000 person-years vs. 0.4 per 100,000 person-years, respectively; P=0.005 for trend). Community-onset SSTIs due to MRSA reached a peak of 62% in 2006 but decreased each year thereafter to 52% in 2010 (P<0.001 for trend).

The authors noted that no pre-2005 data were available and that they were therefore unable to determine trends in S. aureus bacteremia and SSTIs before community-acquired MRSA emerged. Data on race, ethnicity and clinical outcomes were also unavailable, among other limitations. However, the authors concluded that while S. aureus bacteremia and SSTIs continue to place a substantial burden on the U.S. military health system, rates of community-onset MRSA and methicillin-susceptible S. aureus bacteremia and hospital-onset MRSA bacteremia decreased from 2005 to 2010, along with the proportion of community-onset SSTIs due to MRSA.

“These observations, taken together with results from others showing decreases in the rates of health care-associated infections from MRSA, suggest that broad shifts in the epidemiology of S. aureus infections may be occurring,” the authors wrote. “Additional studies are needed to assess whether these trends will continue, which prevention methods are most effective, and to what degree other factors may be contributing.”

Recommendations address cardiac evaluation in liver, kidney transplant candidates

The American Heart Association (AHA) and the American College of Cardiology Foundation released a scientific statement recently on cardiac disease evaluation and management in candidates for liver and kidney transplants.

The statement was based on a comprehensive literature review conducted by the AHA Writing Committee on Cardiac Disease Evaluation and Management Among Kidney and Liver Transplantation Candidates, covering English-language studies conducted from 1990 through March 2010. For all solid organ transplant candidates, a thorough history and physical to identify active cardiac conditions is recommended, the statement said.

Additional guidelines for kidney transplant recipients included the following areas:

  • noninvasive stress testing for at-risk patients without active cardiac conditions, regardless of functional status;
  • cardiac surveillance after listing for transplantation;
  • supplemental testing, including resting echocardiography (ECG), 12-lead ECG, biomarkers, and cardiac computed tomography;
  • referral to a cardiologist;
  • coronary revascularization and related care pre-transplant;
  • lipid management; and
  • perioperative medical management of cardiovascular risk.

Additional guidelines for liver transplant recipients included the following areas:

  • evaluation for coronary artery disease,
  • management of flow-limiting coronary artery disease,
  • evaluation for pulmonary hypertension, and
  • medical management of cardiovascular risk.

The authors noted that the recommendations for kidney transplant candidates are more extensive because the target population is four times larger than that of liver transplant candidates, and the available literature reflected that difference.

The full text of the statement was published in the July 31 Circulation.