Journal watch: Recent studies of note

Recent studies of note.

Bar codes may improve medication administration safety

Using an electronic medication administration system that incorporates bar code technology appears to help reduce medication errors, according to a study.

Researchers at Brigham and Women's Hospital in Boston performed a before-after study at an academic medical center that was in the process of implementing a bar code-equipped medication administration system. Outcome measures were error rates for order transcription and medication administration on hospital units before and after the system was implemented. The authors classified errors involving early or late administration of medications as errors of timing; all other errors were classified as nontiming errors. Each error was reviewed by two clinicians, who determined whether the errors could have harmed patients. Those that could have were considered potential adverse drug events. Clinician disagreements were resolved by consensus. The study results were in the May 6 New England Journal of Medicine.

The study involved data from 14,041 administrations of medication and 3,082 order transcriptions. Most medication administrations were observed on a weekday from 7 a.m. to 3 p.m. Seven hundred seventy-six nontiming errors occurred on units that did not use the bar code-equipped system, while 495 occurred on units that did (error rates, 11.5% vs. 6.8%). This translated to a 41.4% relative reduction in medication errors with bar code technology (P<0.001). The rate of potential adverse drug events not associated with timing errors was 3.1% before the system was implemented and 1.6% afterward, a relative reduction of 50.8% (P<0.001). The rate of timing errors decreased by 27.3% (P<0.001), while the rate of potential adverse drug events associated with these errors did not see a significant change. Units that didn't use the bar code system had a transcription error rate of 6.1%, while those that did had a transcription error rate of 0%.

The authors acknowledged that their study involved only one hospital and looked only at potential, not actual, adverse drug events, among other limitations. However, they concluded that bar code technology within a medication administration system reduces errors and improves patient safety. Although they called for further studies to examine optimal implementation of such systems, their findings “provid[e] support for the inclusion of this technology as a 2013 criterion for achieving meaningful use under the American Recovery and Reinvestment Act,” they wrote.

Magnetic resonance angiography shouldn't be standard practice for diagnosing PE

Physicians should use magnetic resonance pulmonary angiography to diagnose pulmonary embolism (PE) only in patients for whom standard tests are contraindicated, and only at centers that routinely perform it well, a study found.

In a prospective, multicenter study between April 2006 and September 2008, researchers examined 371 adults at seven hospitals who had PE either diagnosed or excluded. They measured sensitivity, specificity and likelihood ratios for PE diagnosis by comparing gadolinium-enhanced magnetic resonance angiography with a reference standard that included computed tomographic angiography and venography, ventilation-perfusion lung scan, venous ultrasonography, D-dimer assay and clinical assessment.

Criteria for diagnosing acute PE with magnetic resonance imaging–which was performed within 72 hours of the reference test–were a partially occlusive intraluminal filling defect or complete arterial occlusion with termination of the column of contrast material in a meniscus that outlined the trailing edge of the embolus. Both the angiogram and venogram were required to be technically adequate to exclude PE; the combined result was considered positive if either result was positive. Study results were published in the April 6 Annals of Internal Medicine.

The reference test was positive in 104 of 371 patients. Eighty-four percent of patients received gadobenate dimeglumine; 14% received gadopentetate dimeglumine. Magnetic resonance angiography identified 57% of patients with PE, including patients with technically inadequate images; the angiography was technically inadequate due to poor quality in 25% of patients, averaged across centers. The adequacy of images varied widely, ranging from 11% to 51% of images being technically inadequate at various centers. Among patients who didn't have PE according to the reference test, angiography excluded PE in 75%, including those with technically inadequate images. When images were technically adequate, magnetic resonance angiography was 78% sensitive and 99% specific. Combined venography and angiography was 92% sensitive and 96% specific, though 52% of patients had technically inadequate results as images were harder to obtain in the lungs and lower extremities.

Study limitations included that many of the patients who had suspected PE declined to participate in the study, or were ineligible to do so, mostly due to concerns about nephrogenic systemic fibrosis, the authors said. Also, imaging results may not apply to pregnant women, patients with renal failure, or patients who are critically ill, on ventilator support, or in shock. The results suggest that magnetic resonance pulmonary angiography is often technically inadequate, and should therefore be reserved for use at centers that consistently perform it well, and with patients who have contraindications to standard tests, the authors concluded.

ABCD2 score can predict long-term risk of stroke recurrence

In addition to predicting short-term risk, the ABCD2 score can predict the risk of having another stroke up to four years after an initial minor ischemic stroke or transient ischemic attack, research has found.

Researchers in Hong Kong followed 490 consecutive patients, for an average of 40.5 months, who had been hospitalized between January 2004 and December 2005 for transient ischemic attack (TIA) or minor stroke. Patients with atrial fibrillation were excluded, because the risk of further stroke strongly depended on warfarin use, and this drug may be underused in the Chinese population, the researchers noted. The ABCD2 score was calculated retrospectively from hospital records and from a prospective stroke registry. Results were published June 1 in Stroke.

Fifteen-and-a-half percent of patients had stroke during follow-up, and 12.7% of patients died. Causes of death were ischemic stroke (n=4), hemorrhagic stroke (n=2), pneumonia (n=16), cancer (n=11), ischemic heart disease (n=4), other causes (n=21) and unknown causes (n=4). Cox multivariate regression analysis found that an ABCD2 score of 4 was an independent risk factor for future stroke (hazard ratio [HR], 2.27; 95% CI, 1.36 to 3.80; P<0.001). Previous cerebral vascular disease, ischemic heart disease and smoking were independent predictors of death. The cumulative risk of later stroke was 7% at 12 months, 11% at 24 months, 15% at 36 months, 17% at 48 months, and 17% at 54 months.

While previous studies have showed the ABCD2 score can predict immediate stroke risk, the current study predicted long-term stroke or death risk up to 54 months after a TIA or minor stroke, the authors noted. Study limitations include the use of retrospective calculation of the ABCD2 score, which might be less predictive than a prospective score. For patients with elevated long-term stroke risk as identified by ABCD2 score, urgent intervention and aggressive evaluation may be warranted, the authors concluded.

No difference in VAP rates with early vs. late tracheotomy

Early tracheotomy doesn't significantly reduce the incidence of ventilator-associated pneumonia compared with late tracheotomy, nor does it lower mortality risk, a study found.

In a randomized, controlled trial, Italian researchers recruited 600 patients from 12 ICUs from June 2004 to June 2008. Patients were age 18 or older, didn't have lung infection, had been ventilated for at least 24 hours, had a Simplified Acute Physiology Score II between 35 and 65, and had a sequential organ failure assessment (SOFA) score of at least 5. They were randomized 48 hours after enrollment to early tracheotomy (after 6 to 8 days of laryngeal intubation) or late tracheotomy (after 13 to 15 days of intubation) if their respiratory conditions worsened, if their sequential organ failure assessment score remained unchanged or worsened, and if they didn't have pneumonia. Of the 600, 419 patients met the tracheotomy criteria. The study was published in the April 21 Journal of the American Medical Association.

Of the 419 patients randomized to bedside tracheotomy, 69% in the early tracheotomy group actually received the procedure, compared to 57% in the late tracheotomy group. There was no significant difference in ventilator-associated pneumonia (VAP) rates between the early tracheotomy and late tracheotomy patients (14% vs. 21% incidence, P=0.07) 28 days after randomization. There was also no difference in length of stay, mortality at one year, and need for care at a long-term care facility. The number of ICU-free and ventilator-free days was higher in the early tracheotomy group; the adverse event rate was 39% in both groups.

The study data suggest tracheotomy shouldn't be performed earlier than after 13 to 15 days of endotracheal intubation, the authors concluded, since planning an earlier tracheotomy increased the number of patients who received a tracheotomy, and more than one-third of patients experienced an adverse event related to the tracheotomy.

Non-conventional approach succeeds in extubating unweanable patients

Unweanable patients with neuromuscular disease (NMD) can be successfully extubated using a non-conventional approach, a study found.

Researchers collected data on 157 consecutive patients who were unweanable—i.e., who couldn't pass spontaneous breathing trials or conventional ventilator weaning parameters. Before hospitalization, 61% of patients had no experience with noninvasive mechanical ventilation support (NIV), 26% used it less than 24 hours a day, and 13% were continuously NIV-dependent. As treatment, patients were extubated to full NIV and aggressive mechanically assisted coughing (MAC) once pulse oxyhemoglobin saturation (SpO2) was maintained at ≥95% in ambient air. NIV and MAC were used to maintain or return SpO2 to ≥95%. Extubation was considered successful if patients didn't require reintubation while hospitalized, and success was considered as a function of diagnosis, patient group, vital capacity, preintubation NIV experience, and assisted cough peak flow (CPF) at extubation.

On the first attempt at the protocol, 95% of patients were successfully extubated. On patients with assisted CPF ≥160 L/m, all 98 extubation attempts were successful. Dependence on continuous NIV and the duration of dependence before intubation correlated with extubation success (P <0.005). Six of eight patients who initially failed extubation succeeded on later attempts; the remaining two patients, who had no measurable assisted CPF, underwent tracheotomy. Results were published in the May issue of Chest.

Benefits of extubating unweanable patients included no mortality, fewer days intubated, no tracheostomies, and return home, the authors said. Avoiding tracheostomy can improve quality of life and diminish long-term pneumonia rates, they noted. The idea that early tracheotomy after intubation facilitates ventilator weaning should be reassessed for patients with NMD, the authors added. “Patients with measurable assisted cough flows should no longer be advised to refuse intubation for fear of extubation failure and tracheotomy,” they wrote.

Overuse of PPIs may raise risk of C. diff infections, fractures

Widespread overuse of proton-pump inhibitors appears to increase patients' risk of Clostridium difficile infections and fractures, several studies concluded.

One study, which analyzed data on more than 100,000 patients discharged from a tertiary care hospital over five years, found that the risk of nosocomial C. difficile infection increased among patients on acid suppressive therapy. Patients not on acid suppressive therapy had a 0.3% risk of nosocomial C. difficile infection compared with 0.6% in those on an H2 antagonist, 0.9% in those taking daily proton-pump inhibitors (PPIs), and 1.4% in those taking more frequent PPI therapy. Another retrospective study found that PPI use during C. difficile treatment was associated with a 42% increase in risk of recurrence of infection. Both studies appeared in the May 10 Archives of Internal Medicine.

Three other studies in the same issue of Archives brought up other potential risks associated with overuse of PPIs. One study, based on Women's Health Initiative data, looked at the impact of PPI use on fractures in postmenopausal women and concluded that PPIs were associated with an increased risk of spine, lower arm and total fractures (but not hip fracture). In addition, a meta-analysis demonstrated that higher doses of PPIs (equivalent to an 80-mg bolus followed by 8 mg/h for 72 hours or similar dose) were not more effective than lower doses in decreasing the rates of rebleeding, surgical intervention or mortality among patients with bleeding peptic ulcers.

A fifth study found that a guideline on appropriate PPI use decreased inpatient and discharge PPI therapy, but only among patients who were not taking the medication at admission. The overall rate of use was driven up by the high rate of outpatient PPI use at admission.

Overuse of PPIs can be tied to the prevalence of dyspepsia and the propensity of physicians to prescribe a pill rather than consider other possible treatments, said an accompanying editorial. However, for many patients the risks of PPIs may outweigh the benefits, the editorial continued, adding that PPIs are also known to increase the risk of hospital and community-acquired pneumonia.

While PPIs relieve symptoms of dyspepsia, they are overprescribed, especially in the absence of ulcer disease, esophagitis or severe gastroesophageal reflux disease, the editorialist said. Clinicians should offer alternative treatments for functional dyspepsia, prescribe short courses of treatment, and consider discontinuing therapy in asymptomatic patients. In addition, patients who know the risks may want to consider non-drug options such as waiting for symptoms to resolve over time and making behavioral changes.

Noncardiac surgery within six weeks of stent implantation raises complications risk

Patients who have noncardiac surgery within six weeks of receiving a coronary stent are more likely to die or have heart complications.

In a country-wide, retrospective cohort study, researchers linked hospital admission data to the Scottish Coronary Revascularisation Register to examine cardiac outcomes in 1,953 patients who received stents between April 2003 and March 2007, and had subsequent non-cardiac surgery. Twenty-nine percent of the patients were treated with at least one drug-eluting stent, and 71% with bare metal stents only. Results were published June 1 in Circulation: Cardiovascular Interventions.

For at least two years after implantation, there was no difference by stent type in occurrence of death, ischemic cardiac events or acute myocardial infarction (MI); there was also no temporal difference in outcomes between these groups. Perioperative death and ischemic cardiac events occurred more frequently when noncardiac surgery was performed within 42 days of stent implantation versus after 42 days (42.4% vs. 12.8%, P<0.001), particularly in patients revascularized after acute coronary syndrome compared with those who underwent stent implantation for stable coronary artery disease (65% vs. 32%, P=0.037).

Limitations of the study include that fewer patients in the drug-eluting stent group than the bare metal stent group underwent percutaneous coronary intervention after presenting with acute coronary syndrome. Also, national data on the use of perioperative antiplatelets and on bleeding complications weren't available for analysis, the authors noted. Possible reasons for the apparent excess risk with noncardiac surgery after stents include that physiological stressors (e.g., anemia) may lead to myocardial ischemia when fixed coronary obstruction is present; withdrawal of antiplatelet therapy may lead to an increase in atherothrombotic events; and the pro-inflammatory state of noncardiac surgery may contribute to developing vulnerable or inflamed plaques in the coronary vascular bed, the authors said.

New drug-eluting stents outperform old ones except in diabetics

Second-generation drug-eluting stents that released everolimus lowered rates of stent thrombosis and restenosis compared to paclitaxel-eluting stents, a trial found.

The trial randomized 3,687 patients to receive one or the other drug-eluting stents without routine follow-up angiography. Over the next year, rates of cardiac death, target-vessel myocardial infarction and ischemia-driven target-lesion revascularization (combined as target-lesion failure) were compared between the two groups. The study was published in the May 6 New England Journal of Medicine.

Everolimus-eluting stents performed significantly better than paclitaxel-eluting ones on the composite endpoint of target-lesion failure (4.2% vs. 6.8% of patients; relative risk, 0.62, 95% CI, 0.46 to 0.82; P=0.001). The newer stents also significantly reduced the risk of ischemia-driven target-lesion revascularization (P=0.001), myocardial infarction (1.9% vs. 3.1%, P=0.02) and stent thrombosis (0.17% vs. 0.85%, P=0.004). Study authors noted that the rate of stent thrombosis was among the lowest ever reported for a drug-eluting stent. The everolimus stents were non-inferior to the paclitaxel stents in the rate of cardiac death and target-vessel myocardial infarction.

There was no significant difference between the two stents in patients with diabetes, however, similar to what the recent COMPARE trial found. That finding suggests that the mechanism of restenosis and/or the response to antiproliferative agents may differ in diabetic patients, said an accompanying editorial. Therefore, the paclitaxel-eluting stent may be more appropriate for these patients, the editorialist concluded.

In patients without diabetes, more data about cost-effectiveness is needed to determine whether the higher cost of the everolimus-eluting stents (about $300 extra) is justified by the reduction in adverse events, the editorialist concluded. He also called for additional research on the stents' effect on stent thrombosis and MI when prasugrel instead of clopidogrel is used as anti-platelet therapy.

Antiplatelet, anticoagulation meds raise bleeding risk with cardiac device implantation

Periprocedural (“bridging”) heparin and dual antiplatelet therapy raise the risk of bleeding complications at the time of cardiac device implantation, a study found.

Researchers reviewed 1,512 charts, and included 1,388 in the study, of patients who had pacemakers or implantable cardioverter-defibrillator (ICD) devices implanted between August 2004 and August 2007. Forty percent got a pacemaker; 60% received an ICD. If aspirin or clopidogrel was taken, it was taken within five days of implantation. A significant bleeding complication was defined as a need for pocket exploration or blood transfusion, hematoma requiring pressure dressing or change in anticoagulation therapy, or prolonged hospitalization. The study was in the May 25 Journal of the American College of Cardiology.

Seventy-one patients (5.1%) had bleeding complications. Patients who took combined aspirin and clopidogrel (n=139) had significantly greater bleeding risk than control patients (n=255) not taking antiplatelet therapy (7.2% vs. 1.6%; P=0.004). Bleeding risk wasn't significantly higher for those taking aspirin alone (n=536) vs. control. Patients who used periprocedural heparin (n=154) had significantly higher bleeding risk compared to those who had warfarin held until their international normalized ratio (INR) was normal (n=258; 14.3% vs. 4.3%; P<0.001) and those who got no anticoagulation therapy (14.3% vs. 1.6%; P<0.0001). There was no statistical difference in bleeding risk between those who had warfarin held until INR was normal and those who continued on warfarin with an INR ≥1.5 (n=46).

Study limitations include its retrospective, observational nature and a small sample size for each medication subgroup that may have been underpowered to examine certain associations. Also, there was no uniform protocol for antiplatelet and anticoagulant management. Still, the results suggest it would be wise to hold antiplatelet or anticoagulation medications for three to five days before device implantation for patients at low risk for thromboembolic events, the authors said. Patients at high risk for thromboembolic events should continue warfarin throughout the periprocedural phase, thus avoiding the need for heparin “bridging,” they said.