Recent Research

Hip fractures in nursing homes, cognitive impairment with TIA, beta-blockers before CABG, and more.


Hip fractures among nursing home residents often lead to death, functional decline

The frequently poor outcomes of nursing home residents who have a hip fracture were quantified by a recent study.

The retrospective cohort study included more than 60,000 Medicare beneficiaries who were living in nursing homes when they were hospitalized for hip fracture between 2005 and 2009. Outcomes were evaluated at 180 days after fracture and included death from any cause, dependence in locomotion, and changes in function on activities of daily living (ADL). Results were published in the August JAMA Internal Medicine.

In total, 36.2% of the patients had died by 180 days. Mortality risk was higher among men, as 46% of them had died. Among a healthier subset of patients—those who were not totally dependent in locomotion at baseline—more than half (53.5%) were dead or totally dependent at 180 days, and only about 1 in 5 regained their prefracture level of independence. Overall, function declined substantially after fracture in all ADL domains. The researchers did identify some predictors of death and total dependence, including nonoperative fracture management (relative risk [RR] for combined outcome compared to internal fixation, 1.48; P<0.001) and older age (RR for patients >90 years vs. those ≤75 years, 1.42; P<0.001). The strongest predictor of death and total dependence was very severe cognitive impairment at baseline.

This study confirms and extends prior research showing that survival and functional outcomes are poor in nursing home residents who have a hip fracture, the study authors concluded. The finding of worse outcomes with nonoperative management could result from the sickest patients not undergoing surgery, but it may also suggest that more patients could benefit from surgery, the authors concluded. The other predictors could be used in care planning, and the results should encourage clinicians to counsel patients about their prognoses, goals of care, and options to treat distressing symptoms. Specialized inpatient geriatric fracture programs may also be helpful for nursing home residents.

According to an accompanying commentary, the study highlights the importance of offering palliative care to elderly, vulnerable hip fracture patients. Palliative care should be implemented immediately after the fracture and include “effective communication, care planning and coordination, symptoms management, psychosocial, spiritual and bereavement support, and end-of-life care,” the commentary said. Surgery should also be considered, but its benefits should be balanced with potential risks and patients' life expectancy.

Nearly 40% of TIA patients have cognitive impairment 3 months later

More than a third of patients who had a transient ischemic attack (TIA) had impaired cognitive function in at least 1 area 3 months later, a recent study found.

In a single-center, cross-sectional study, 107 patients aged 45 to 64 years, without prior stroke or dementia, underwent neuropsychological testing within 3 months of having a TIA. The testing assessed executive functioning, information processing speed, attention, working memory, and verbal episodic memory. Researchers also administered a 15-item interview to identify subjective cognitive failures in the previous month. TIA patient assessments were compared with those of a control group of 81 people without incident TIA, except for the test for verbal episodic memory, which was compared to a normative sample in the test manual. For TIA patients, a z score <−1.65 of the control group was used as a cutoff to signify domain-specific impairment. Results were published in the August Stroke.

TIA patients did worse than controls on every individual cognitive test and in all cognitive domains, except episodic memory. The greatest impairment for TIA patients was seen in working memory (24.5% impaired; age- and sex-adjusted odds ratio [OR], 22.5; 95% CI, 2.9 to 174.3); attention (21.7% impaired; OR, 6.8; 95% CI, 1.9 to 24.3); and information processing speed (16.1% impaired; OR, 7.1; 95% CI, 1.5 to 32.5). Thirty-eight percent of TIA patients (OR, 5.9) had impairment of 1 or more cognitive domains (excluding episodic memory). TIA and control groups didn't differ in mean subjective cognitive failure scores. Also, the presence of silent brain infarcts was related to worse executive functioning scores.

Potential study weaknesses include its cross-sectional design; the fact that controls were slightly younger than TIA patients (52.9 years vs. 56.6 years); and the fact that older patients and those with low Mini-Mental State Examination scores were excluded, which limits generalizability of results to the whole TIA population. By administering cognitive testing within 3 months of a TIA, using a younger population, and excluding those with stroke history, the researchers aimed to minimize the effect of possible concomitant causes of cognitive impairment that may have influenced past research, they noted. As such, the 38% prevalence of impairment “suggests a role for TIA itself in cognitive impairment afterward, the mechanism of which remains to be elucidated,” they wrote.

Beta-blockers before coronary artery bypass grafting not associated with better outcomes

Use of beta-blockers in patients who have not had a recent heart attack but are undergoing nonemergency coronary artery bypass grafting (CABG) surgery was not associated with better outcomes, a study found.

Preoperative beta-blocker therapy is a national quality standard, the authors noted. They conducted a retrospective analysis of the Society of Thoracic Surgeons National Adult Cardiac database of U.S. hospitals performing cardiac surgery from 2008 to 2012. The study included 506,110 patients undergoing nonemergency CABG surgery who had not had a heart attack in the previous 21 days or any other high-risk symptoms.

Results appeared in the August JAMA Internal Medicine.

Of the 506,110 patients, 86.2% received preoperative beta-blockers within 24 hours of surgery. A propensity-matched analysis included 138,542 patients. There was no difference between patients who did and did not receive preoperative beta-blockers in rates of:

  • operative mortality (1.12% vs. 1.17%; odds ratio [OR], 0.96; 95% CI, 0.87 to 1.06; P=0.38),
  • permanent stroke (0.97% vs. 0.98%; OR, 0.99; 95% CI, 0.89 to 1.10; P=0.81),
  • prolonged ventilation (7.01% vs. 6.86%; OR, 1.02; 95% CI, 0.98 to 1.07; P=0.26),
  • any reoperation (3.60% vs. 3.69%; OR, 0.97; 95% CI, 0.92 to 1.03; P=0.35),
  • renal failure (2.33% vs. 2.24%; OR, 1.04; 95% CI, 0.97 to 1.11; P=0.30), and
  • deep sternal wound infection (0.29% vs. 0.34%; OR, 0.86; 95% CI, 0.71 to 1.04; P=0.12).

Patients who received preoperative beta-blockers within 24 hours of surgery did have higher rates of new-onset atrial fibrillation than patients who did not (21.50% vs. 20.10%; OR, 1.09; 95% CI, 1.06 to 1.12; P<0.001), the authors wrote.

“[Beta]-blockers are an important and effective tool in the care of patients undergoing cardiac surgery in specific clinical scenarios,” the authors wrote. “However, the empirical use of [beta]-blockers as recommended by the National Quality Forum (without physiologic goals i.e., adequate clinical drug levels) in all patients before CABG may not improve outcomes.”

In an invited commentary, the editorialist states that, because of 5 limitations of the study, physicians should continue to adhere to American College of Cardiology/American Heart Association guidelines for preoperative beta-blockade in CABG surgery. Limitations include the following:

  • Short-term CABG mortality rates, which have decreased to about 1%, may no longer be a sufficiently sensitive outcome to assess the value of perioperative beta-blockade;
  • The study excluded 28% of patients with isolated CABG because of a myocardial infarction within 21 days before surgery, thus eliminating patients with ST-elevation myocardial infarction (STEMI) and non-STEMI;
  • Critical variables necessary to address the efficacy of perioperative beta-blockade are not present in the study, such as the specific beta-blocker used, whether the patient was receiving long-term beta-blocker therapy, the exact timing and protocol for beta-blocker administration, and use of amiodarone hydrochloride;
  • Substantial differences were noted in the baseline characteristics of the beta-blocker and non-beta-blocker groups; and
  • Pharmacogenetic variation may affect the efficacy of preoperative beta-blockade.

“Important considerations include perioperative continuation of [beta]-blockade in patients receiving long-term therapy and administration and titration of [beta]-blockers to optimal heart rate and blood pressure in [beta]-blocker-naïve patients, initiated as long before surgery as possible (preferably weeks before in elective patients),” the editorial concluded.

Varying appearance of generic pills may affect medication adherence after MI

Adherence to medications after myocardial infarction (MI) in patients with cardiovascular disease may be affected by variation in pill appearance, according to a recent study.

Researchers performed cohort and nested case-control studies using claims from a U.S. commercial health insurance database to examine whether inconsistent appearance of generic medications was associated with inconsistent use in patients with cardiovascular disease after MI. Patients who were hospitalized for an MI and discharged between 2006 and 2011 and started treatment with a generic beta-blocker, angiotensin-converting enzyme (ACE) inhibitor, angiotensin II-receptor blocker (ARB), or statin were included.

Case patients were those who stopped their index medication for at least a month, and control patients were those who continued treatment. The researchers matched control patients to case patients by therapeutic medication class, sex, age, and number of times the drug was dispensed before treatment nonpersistence. Rates of changes in pill color and shape in the year after MI were determined, and the 2 refills preceding nonpersistence were examined to see whether pill color or shape had changed. The study results appeared in the July 15 Annals of Internal Medicine.

A total of 11,513 patients began treatment with a generic prescription drug in the class of interest within 90 days after hospital discharge for MI. Most of the patients were men with commercial insurance; the average age was 57.7 years. A generic beta-blocker was the most common drug prescribed (75.4% of patients), followed by a generic ACE inhibitor or ARB blocker (52.8%) and a generic statin (40.0%). Change in pill shape or color affected 29% of patients (3,286 of 11,513) during the study. Statins changed appearance most often, and beta-blockers changed appearance least often. There were 4,573 episodes of treatment nonpersistence matched to 19,881 control episodes. Odds of nonpersistence in case patients increased 34% after a pill color change (adjusted odds ratio, 1.34; 95% CI, 1.12 to 1.59) and 66% after a pill shape change (adjusted odds ratio, 1.66; 95% CI, 1.43 to 1.94).

The authors acknowledged that they examined only 3 categories of drugs prescribed after MI, that they did not evaluate effects on clinical outcomes, and that they had no data on socioeconomic status or enrollment in automatic refill programs. However, they concluded that variation in appearance of generic pills is associated with nonpersistent use after MI in patients with heart disease. “Until the FDA or manufacturers of generic drugs take the initiative to make consistent pill shape or color an industry standard, it is incumbent on prescribers and pharmacists to take steps to warn patients about the diversity of the shapes and colors of the pills containing their generic cardiovascular drugs to reduce the burden of these changes on the public health,” the authors wrote.

Linking inpatient and outpatient buprenorphine treatment reduced opioid abuse

A program that started hospitalized, opioid-dependent patients on buprenorphine and linked them to outpatient care effectively reduced their opioid use, a recent study found.

The trial randomly assigned 139 opioid-dependent patients who were hospitalized between Aug. 1, 2009, and Oct. 31, 2012, to either 5-day buprenorphine detoxification with treatment referral information at discharge or an alternative intervention. The intervention consisted of buprenorphine induction, inpatient dose stabilization, and postdischarge transition to maintenance buprenorphine opioid agonist treatment (OAT) provided by the hospital's primary care clinic. Results were published in the August JAMA Internal Medicine.

Patients in the intervention group were significantly more likely to enter maintenance OAT than those in the detox group (72.2% vs. 11.9%; P<0.001). At 6 months, which was the end of the study's follow-up, 12 (16.7%) of the intervention group and only 2 (3.0%) of the detox patients were receiving buprenorphine OAT (P=0.007). At the same time, the researchers queried patients about their illicit opioid use and found that the intervention group reported much less use in the prior 30 days (incidence rate ratio, 0.60; P<0.01). Available urine drug tests were used to confirm that the participants' self-reports were not dramatically underreporting use.

The intervention was an effective means for getting hospitalized patients who were not seeking addiction treatment connected with a program and using fewer opioids, the researchers concluded. However, the decline in OAT participation during the 6-month follow-up is a problem, they acknowledged. The same treatment program has a retention rate of 51% at 12 months for patients who initiate outpatient therapy, although those patients are likely to be more motivated and less medically ill. Another limitation of the study is its generalizability: Hospitals without affiliated outpatient programs might have more difficulty linking patients to postdischarge treatment.

If hospitals wanted to implement such a program, they would need to identify drug users systematically, develop an active referral network of buprenorphine prescribers, and have a dedicated inpatient substance use consulting team to initiate and bridge treatment, the study authors said. According to an accompanying editorial, the study is “groundbreaking,” but one of the major obstacles to widespread adoption of the model is a shortage of outpatient physicians and inpatient clinical staff (physicians or nonphysicians who could advise them) trained in buprenorphine treatment.

Some other recent publications assessed the use, and probable overuse, of opioids in the U.S. A survey of active military (more than 2,000 recently deployed members of an infantry brigade) found that 44% reported chronic pain and 15.1% reported opioid use in the past month. Almost half of those who had used opioids had no or mild pain in the past month, which should be cause for concern about overuse, concluded the authors of the research letter, also published in the August JAMA Internal Medicine.

Rates of opioid prescribing vary widely from state to state, according to a report in the CDC's July 1 Morbidity and Mortality Weekly Report. Overall, in 2012, prescribers wrote 82.5 opioid prescriptions per 100 persons, but rates varied 2.7-fold and were highest in the South. Prescriptions for long- acting/extended-release and high-dose opioids, which may lead to more abuse and overdoses, were highest in the Northeast. The results indicate the need to identify problematic prescribing practices, the authors concluded. Another report in the same issue described how the state of Florida has successfully accomplished this and reduced deaths from prescription opioid use.

Additional heart rhythm monitoring may help detect afib after cryptogenic stroke

Additional heart rhythm monitoring after cryptogenic stroke helped detect atrial fibrillation in 2 recent studies published in the June 26 New England Journal of Medicine.

In the EMBRACE trial, patients 55 years of age and older without known atrial fibrillation who had had a cryptogenic ischemic stroke or transient ischemic attack (TIA) in the previous 6 months were randomly assigned to additional monitoring with noninvasive ambulatory electrocardiography (ECG). The intervention group was monitored via a 30-day event-triggered recorder, while the control group received a conventional 24-hour monitor. The study's primary outcome was newly detected atrial fibrillation of 30 seconds or longer in the 90 days after randomization, while secondary outcomes were atrial fibrillation of 2.5 minutes or longer and anticoagulation status at 90 days.

Of 280 patients in the intervention group, 45 (16.1%) had atrial fibrillation lasting 30 seconds or longer compared with 9 of 277 patients (3.2%) in the control group (absolute difference, 12.9 percentage points; P<0.001; number needed to screen, 8). Twenty-eight of 284 patients in the intervention group (9.9%) had atrial fibrillation of 2.5 minutes or longer compared with 7 of 277 (2.5%) in the control group (absolute difference, 7.4 percentage points; P<0.001). Fifty-two of 280 patients (18.6%) in the intervention group had been prescribed anticoagulant therapy by 90 days versus 31 of 279 patients (11.1%) in the control group (absolute difference, 7.5 percentage points; P=0.01).

The authors noted that the total burden of atrial fibrillation for each patient could not be determined and that the prevalence of atrial fibrillation in the intervention group is probably a conservative estimate. However, they concluded that noninvasive ECG monitoring on an outpatient basis for 30 days after cryptogenic stroke or TIA outperforms standard short-term ECG monitoring in detecting atrial fibrillation.

In the CRYSTAL AF trial, patients at least 40 years of age who had no evidence of atrial fibrillation on 24 hours or more of ECG monitoring after ischemic stroke were randomly assigned within 90 days of the index event to receive long-term monitoring with an insertable cardiac monitor (ICM) or conventional follow-up (defined as assessment at scheduled and unscheduled visits and ECG monitoring at the site investigator's discretion). The primary end point was time to first detection of atrial fibrillation lasting at least 30 seconds in a 6-month period, while time to first detection of atrial fibrillation lasting at least 30 seconds in a 12-month period was one of the secondary end points. The study was funded by Medtronic.

Two hundred twenty-one patients were assigned to the ICM group (of these, 208 received the device), and 221 patients were assigned to the control group. During the trial, 12 patients in the ICM group crossed over to the control group and 6 patients in the control group crossed over to the ICM group, while 12 patients and 13 patients in each group, respectively, exited the study. A total of 8.9% of patients in the ICM group had atrial fibrillation detected by 6 months, compared with 1.4% of patients in the control group (hazard ratio, 6.4; P<0.001). At 12 months, these percentages were 12.4% and 2.0%, respectively (hazard ratio, 7.3; P<0.001).

The authors noted that they could not be sure whether atrial fibrillation detected via monitoring had caused the index stroke and that ICMs may not have detected all episodes of atrial fibrillation, among other limitations. However, they concluded that ICMs detected atrial fibrillation more frequently after cryptogenic stroke than did conventional follow-up.

The author of an accompanying editorial noted that more information is needed on other sources of embolism and improved markers of stroke mechanisms, in addition to appropriate therapy for subclinical atrial fibrillation. However, he wrote that “the weight of current evidence suggests that subclinical atrial fibrillation is a modifiable risk factor for stroke recurrence, and its presence should be thoroughly ruled out in this high-risk population.”

Most patients with cryptogenic stroke or TIA should have rhythm monitoring for at least several weeks, he wrote, and those in whom subclinical atrial fibrillation is detected should usually be switched from antiplatelet therapy to anticoagulants. “At the least, patients should be followed closely in order to detect progression to clinically apparent atrial fibrillation, in which case the evidence unambiguously supports anticoagulant therapy for the secondary prevention of stroke,” the editorialist concluded.

Non-opioid pain management regimen may reduce costs, length of stay, overall opioid use

Surgical patients treated with a non-opioid pain management regimen had less overall opioid use and shorter length of stay and incurred fewer hospital costs than those treated with an opioid-based regimen, a recent analysis found.

The pooled analysis of several prospective, open-label, phase 4 studies compared a multimodal analgesia regimen to an opioid-based patient-controlled analgesia (PCA) regimen in 191 adults undergoing open colectomy, laparoscopic colectomy, or ileostomy reversal. The 86 patients in the multimodal analgesia group got a single intraoperative administration of 266 mg of liposome bupivacaine, with the dose expanding to a total volume of 40 mL in studies involving colectomy and 30 mL in studies involving ileostomy reversal. Liposome bupivacaine is made by Pacira Pharma-ceuticals, Inc., for which the study's lead author is a consultant and speaker, and from which 2 of the other 3 authors have received research funding.

The multimodal analgesia regimen patients also received a single administration of 30 mg of IV ketorolac at the end of surgery, then 1,000 mg of oral or IV acetaminophen plus 600 mg of oral ibuprofen every 6 hours for 72 hours after surgery. The 105 patients in the opioid-based group received IV morphine or hydromorphone after surgery on a PCA basis.

The multimodal analgesia patients had lower postsurgical opioid consumption than the IV opioid-based PCA group (38 mg vs. 96 mg; P<0.0001); lower length of stay after surgery (2.9 days vs. 4.3 days; P<0.0001); lower total hospitalization costs ($8,271 vs. $10,726; P=0.01); and a lower proportion of opioid-related adverse events (9% vs. 27%; P=0.003). There were no significant between-group differences in patient satisfaction scores or hospital readmissions at 30 days. Results were published online June 24 by the Journal of Pain Research.

The researchers noted the study was limited by its dependence on a post-hoc analysis of data from open-label studies. Still, they noted several apparent advantages of the multimodal analgesia regimen, which is useful given that postsurgical pain is “often undermanaged and undertreated, partially as a result of concern over the prospect of opioid-related [adverse events].”

These summaries come from ACP Hospitalist Weekly, an e-newsletter provided every Wednesday by ACP Hospitalist. To receive it, call Customer Service at 800-543-1546, ext. 2600, or direct at 215-351-2600 (M-F, 9 a.m. to 5 p.m. EST); send an email; or subscribe online.