Journal Watch: Recent studies of note

Recent studies about poor blood sugar control, predicting heart failure, and other topics.


Poor blood sugar control associated with higher mortality

A recent large study found that poor glucose control was associated with a fourfold increase in mortality and major complications following cardiac surgery, regardless of whether patients had been diagnosed with diabetes.

In the study of more than 8,000 adults who underwent cardiac surgery between April 1996 and March 2004, researchers found that more than half of patients with moderate (200 to 250 mg/dL) to poor (>250 mg/dL) blood glucose control (BGC) were not previously identified as diabetic. Inadequate BGC was associated with in-hospital mortality (good, 1.8%; moderate, 4.2%; poor, 9.6%; adjusted odds ratio: poor versus good BGC, 3.90; moderate versus good BGC, 1.68) as well as postoperative myocardial infarction and pulmonary and renal complications in patients without known diabetes.

The study, led by researchers at the Bristol Heart Institute and published in the July 8 issue of Circulation, could have major implications for the treatment of all patients admitted for major surgery, said a news release. The study's lead author urged surgeons and intensive care specialists to use strict protocols of active BGC in all patients admitted for major surgery.

Patients' characteristics predict heart failure outcomes

Patients with certain characteristics may be at higher risk for death during heart failure hospitalization, according to a recent study.

Researchers used data from OPTIMIZE-HF to examine factors associated with in-hospital mortality in patients hospitalized for heart failure. Of 45 possible predictor variables, those deemed significant in stepwise logistic regression underwent multivariate analysis. The study appeared in the July 29 Journal of the American College of Cardiology.

A total of 48,612 patients with a mean age of approximately 73 years were enrolled in the OPTIMIZE-HF trial. Of these, slightly over half were women and approximately three-quarters were white. Approximately 4% died in the hospital. Age, heart rate, systolic blood pressure, sodium and creatinine levels, primary hospitalization for heart failure and presence or absence of left ventricular systolic dysfunction were found to predict inhospital death, with the strongest associations seen for serum creatinine at admission, systolic blood pressure at admission, and age. On the basis of these variables, the authors developed a risk-prediction nomogram to evaluate likelihood of death in patients hospitalized for heart failure.

Although the authors found that their nomogram compared favorably with results from previously published trials, they acknowledged that their study did not examine outcomes after hospital discharge and that OPTIMIZE-HF was not a randomized trial. However, they concluded that their scoring method could help to identify patients hospitalized with heart failure who are at high risk for death and might benefit from intensive monitoring.

Beta-blockers improve outcomes in decompensated heart failure

Continuing beta-blocker therapy during hospitalization can improve outcomes in patients with decompensated heart failure, according to another analysis of the OPTIMIZE-HF trial. Researchers used data from study participants who were eligible to receive beta-blockers at discharge after hospitalization for systolic heart failure. The results appear in the July 15 Journal of the American College of Cardiology.

Of the 2,373 patients, 56.9% were taking beta-blockers before admission and continued taking them during their hospital stay, 26.6% began taking beta-blockers while in the hospital, 3.3% had beta-blockers withdrawn, and 12.8% were considered eligible for beta-blockers but did not receive them. Patients whose beta-blocker therapy was continued had a significantly lower risk for propensity-adjusted post-discharge death and death or rehospitalization than those who did not receive beta-blockers. In addition, patients whose beta-blockers were continued had a lower adjusted risk for death than those whose beta-blockers were withdrawn during their hospital stay and those who were eligible for beta-blockers but did not receive them; the latter two groups had similar risks for death.

The authors concluded that continuing beta-blocker therapy during hospitalization for decompensated heart failure decreases mortality risk after hospital discharge and improves rates of treatment. Risk for death and poor outcomes worsens, however, in patients whose beta-blocker therapy is withdrawn.

Staff shortcuts undermine safety aim of bar code technology

Practical problems surrounding the use of bar code medication administration (BCMA) technology in hospitals are prompting staff to develop shortcuts that may be undermining safety, according to a recent study.

In the study, researchers spent several years observing barcode technology in five hospitals and analyzing medication scans. They identified several factors that led nurses to take shortcuts, including some related to technology, such as unreadable or missing bar codes, and other problems related to workflow or efficiency, such as affixing extra copies of patient ID bar codes on desks or clipboards to avoid scanning the patient's actual ID wristband.

The authors identified several categories of causes for these “workarounds”:

  • Technological: Including problems with BCMA software or hardware, such as battery failures of handheld scanners or linked computers, difficult-to-read or -navigate screens, and alert beeps that sound like confirmation beeps.
  • Task-related: For example, nurses may circumvent BCMA protections by carrying several patients' medications on one tray or by not scanning medication IDs or patient IDs.
  • Organizational: For example, patients or medications without bar codes (due to organizational flaws), medication bar codes covered by a label reminding users to scan bar codes, or pharmacies sending only partial doses.
  • Patient-related: For example, staff may not scan patient IDs or document administrations before medications are ingested due to special circumstances, such as patients refusing medications, vomiting medications, sleeping, agitated, receiving central lines, or in contact isolation.
  • Environmental: For example, hospital areas without wireless BCMA connectivity (operating rooms, laboratories) and medications stored far away from the scanner (e.g., those that require refrigeration) necessitating multiple trips to scan, dispense, and return.

The study, published in the July/September 2008 Journal of the American Medical Informatics Association, includes a list of recommendations for hospitals to identify and correct problems associated with using BCMA technology.

Radiofrequency id devices may interfere with equipment

Radiofrequency identification devices (RFIDs) may cause critical care equipment to malfunction and may lead to adverse events, according to a recent study.

RFIDs are increasingly used in medicine for tracking and patient safety purposes, but few if any studies have examined their potential impact on medical equipment. A group of Dutch authors assessed two types of RFIDs, a passive 868-MHz system and an active 125-MHz system, to determine how often they caused incidents of electromagnetic interference (EMI) with critical care equipment.

Incidents were defined as “every unintended change in the function of a medical device” and were classified as hazardous (a direct physical influence on a patient by unintentionally changing equipment function), significant (an influence on monitoring requiring a significant level of attention and substantial distraction from patient care) or light (an influence on monitoring not requiring a significant level of attention). Patients were not connected to the 41 critical care devices examined in the study during the EMI tests. The study was published in the June 25 Journal of the American Medical Association.

The authors conducted 123 EMI tests and found that RFIDs caused 34 reproducible EMI incidents. Of these, 22 were considered hazardous, two were considered significant and 10 were considered light. The passive RFID caused more incidents than the active RFID (26 in 41 tests [63%] vs. 8 in 41 tests [20%]). RFID incidents occurred at a median distance of 30 cm (range, 0.1 to 600 cm).

Although the authors cautioned that their study results apply only to the specific RFIDs tested at maximum power, they concluded that RFIDs can cause harm by interfering with medical equipment. RFIDs should be frequently tested in the health care setting, the authors wrote.

Heparin and enoxaparin may increase bleeding risk after stroke

Bridging with heparin and enoxaparin may increase risk for bleeding after cardioembolic stroke, a recent study reported.

Most patients who have a cardioembolic stroke eventually require anticoagulation, but no consensus exists on how and when to begin such therapy. To compare anticoagulation strategies in this population, researchers retrospectively reviewed data from 204 patients who had been admitted with cardioembolic stroke and had not received tissue plasminogen activator. Patients were grouped according to treatment: none (8 patients), aspirin alone (88 patients), aspirin plus warfarin (35 patients), IV heparin followed by warfarin (heparin bridging, 44 patients), and full-dose enoxaparin plus warfarin (enoxaparin bridging, 29 patients). The study was published online July 14 in Archives of Neurology.

Two patients experienced recurrent stroke. The most common serious adverse event, progressive stroke, occurred in 11 patients (5%), all but one of whom were in the aspirin-only group. Patients who received aspirin alone were 12.5 times as likely to have stroke progression as those who received anticoagulant therapy. Hemorrhagic transformation was seen in 23 patients; of these, three patients, all in the enoxaparin bridging group, had parenchymal hematoma. Two patients developed systemic bleeding, both in the heparin bridging group. Enoxaparin bridging was significantly associated with late symptomatic hemorrhagic transformation, while heparin bridging was significantly associated with systemic bleeding.

The authors acknowledged several limitations of their study, including its retrospective, nonrandomized design and its lack of long-term follow-up. However, they concluded that the data may be useful in guiding decisions about long-term anticoagulation after cardioembolic stroke. It seems safe to start anticoagulation with warfarin at any time during hospitalization, while risk for bleeding may be higher with full-dose enoxaparin or heparin bridging, they wrote.