Recent Research

Short-term NSAIDS and MI risk, VTE after colorectal surgery, and more.


Short-term NSAIDs may still increase myocardial infarction risk after prior heart attacks

Even short-term treatment with most nonsteroidal anti-inflammatory drugs (NSAIDs) was associated with increased risk of death and a subsequent myocardial infarction (MI) in patients with prior heart attacks, a national cohort population study from Denmark has reported.

Because patients may unavoidably receive NSAID treatment for a short period of time despite existing precautions in those with established cardiovascular disease, researchers studied the duration of NSAID treatment and cardiovascular risk. Patients 30 years of age or older admitted for a first MI from 1997 to 2006 and their subsequent NSAID use were identified using ICD codes and pharmacy records. Results appeared in the May 24 Circulation.

Of the 83,677 patients discharged alive after their first heart attack, 42.3% received NSAIDs during follow-up. There were 35,257 deaths or recurrent MIs.

NSAIDs were significantly associated with an increased risk of death/recurrent MI (hazard ratio [HR], 1.45) at the beginning of the treatment, and the risk persisted throughout the treatment course (HR, 1.55). Diclofenac was associated with the highest risk (HR, 3.26 for death/MI at day 1 to 7 of treatment). Rofecoxib was associated with increased risk of death after treatment from 7 to 14 days, while celecoxib was associated with increased risk of death after 14 to 30 days. Ibuprofen showed an increased risk when used for more than one week. Naproxen was not associated with an increased risk of death or MI for the entire treatment duration.

“It is noteworthy that a commonly used nonselective NSAID like diclofenac is associated with an even higher risk of death at the beginning of the course of treatment than the selective COX-2 inhibitor rofecoxib, which was withdrawn from the market in 2004,” the authors wrote. But they later added that while naproxen had the lowest cardiovascular risk, it is associated with more gastrointestinal bleeding events.

“Our data challenge the current recommendations by the American Heart Association regarding NSAID treatment in patients with established cardiovascular disease,” the authors concluded, “because we demonstrate that even short-term NSAID treatment is associated with increased cardiovascular risk in patients with prior MI; i.e., there essentially appears to be no safe therapeutic window for NSAID treatment. Therefore, the current approach of recommending short-duration treatment in patients with established cardiovascular disease who require NSAIDs may need revision.”

VTE less common with laparoscopic versus open colorectal surgery

Venous thromboembolism (VTE) appears to be less common after laparoscopic colorectal surgery than after open colorectal surgery, according to a new study.

Researchers analyzed data from the Nationwide Inpatient Sample to determine how often VTE occurred after colorectal surgery and to identity associated VTE risk factors. Included patients had elective laparoscopic or open colorectal surgery from Jan. 31, 2002 through Dec. 31, 2006. The primary outcome measures were VTE incidence after surgery during the initial hospitalization and VTE according to site of surgery, pathology, and at-risk patient population. The study results were published in the June Archives of Surgery.

A total of 149,304 patients had laparoscopic or open resection during the study period. The mean age was 63.5 years; more than half of the patients were women (53.6%), and most were white (82.8%). VTE incidence was statistically significantly higher in the open group than in the laparoscopic group (2,036 of 141,456 [1.44%] vs. 65 of 7,848 [0.83%]; P<0.001). Patients who had inflammatory bowel disease and those who underwent rectal resection had the highest overall VTE rates, while malignant disease, obesity and congestive heart failure were also found to be statistically significant VTE risk factors regardless of type of surgery. Patients who had laparoscopic resection had significantly shorter hospital stays (6.5 days vs. 9 days) and significantly lower mortality rates (0.8% vs. 3.0%) than those who had open resection (P<0.001 for both comparisons).

The authors noted that they may have missed some cases of laparoscopic surgery in the database and that data on follow-up and thromboprophylaxis were not available, among other limitations. However, they concluded that VTE incidence overall is lower after laparoscopic than open colorectal surgery and that certain patient characteristics and conditions also seemed to increase VTE risk significantly. Their findings may help to guide decisions about appropriate VTE prophylaxis in patients undergoing these procedures, the authors wrote.

Dexamethasone with antibiotics shortens length of hospital stay for CAP patients

Adding dexamethasone to antibiotic treatment reduced the length of stay by one day for inpatients with community-acquired pneumonia (CAP), a study found.

In a double-blind trial, Dutch researchers randomly assigned 304 adults with confirmed CAP at two hospitals to receive a daily intravenous bolus of 5 mg (1 mL) of dexamethasone or 1 mL of sterile water (placebo) for four days from admission. Patients were excluded if they were immunocompromised, needed immediate transfer to the ICU, or were already receiving corticosteroids or other immunosuppressive drugs. All patients received antibiotics before study treatment was given. Selection, duration and administration of the antibiotics were decided by the medical team and were based on national guidelines. The primary endpoint was length of stay until hospital discharge or death. Results were published in the June 11 The Lancet.

Median length of stay in the dexamethasone group was 6.5 days compared to 7.5 days in the placebo group (13% reduction; P=0.0480). Hospital mortality and rates of admission to ICUs didn't differ between groups. Hyperglycemia was more common in the dexamethasone group compared to the control group (44% vs. 23%, P<0.0001), but only 5% of patients in the former group and 3% in the latter needed additional glucose-lowering treatment during their stay. Thirty-eight patients didn't complete the treatment course because they were admitted to the ICU, had protocol violations or died. Antibiotic treatment was similar in both groups. In a secondary analysis, patients in the dexamethasone group had better quality of life than those in the control group in terms of social functioning by day 30 after hospital admission. Serious adverse events were rare, though one patient in the dexamethasone group developed a gastric perforation on day three that the authors said could be attributed to use of the drug.

Pneumonia severity index risk classes 4 and 5 were more common in the treatment group, an imbalance that may have led to underestimating the effect of dexamethasone, since a higher risk class usually leads to a longer length of stay, the authors noted. The study is limited in that results can't be generalized to all CAP patients; for example, chronic obstructive pulmonary disease (COPD) patients were underrepresented since they usually need treatment with systemic corticosteroids (which was an exclusion for study participation). Eleven percent of enrolled patients had COPD compared to 21% in the screened population. Also, guidelines for antibiotic treatment in the Netherlands differ from the U.S. In the former, amoxicillin is standard therapy for CAP of severity risk 1 and 2, and is combined with a fluoroquinolone or macrolide antibiotics for higher severity patients, the authors noted. The apparent benefit of dexamethasone seen in this study should be weighed against potential disadvantages of this drug class, such as gastric disturbances and superinfections, they concluded.

Telemedicine reduced mortality, length of stay in ICUs

Implementation of a tele-intensive care unit (ICU) reduced mortality and length of stay at one academic medical center.

The prospective study included more than 6,000 adults admitted to seven ICUs at the University of Massachusetts from April 2005 through September 2007. Outcomes before and after the implementation of the tele-ICU were compared. The hospital mortality rate decreased from 13.6% to 11.8% and length of stay decreased from 13.3 days to 9.8. Study authors attributed some of the success on these measures to improved adherence to best practices for preventing certain conditions: deep vein thrombosis (prophylaxis increased from 85% to 99%), stress ulcers (83% to 96%), cardiovascular problems (80% to 99%), and ventilator-associated pneumonia (33% to 52%). The study found significant decreases in the rate of ventilator-associated pneumonia (from 13% to 1.6%) and catheter-related bloodstream infections (from 1% to 0.6%).

Researchers also observed that the tele-ICU care sped up response to alerts of patients' physiological instability and allowed patients admitted during off-hours to have assistance with their care plan and monitoring by a rested, on-duty intensivist. More patients were also put on noninvasive ventilation instead of mechanical ventilation, which the study authors speculated could be due to emergency physicians being more comfortable using noninvasive ventilation when they knew patients would be monitored by the tele-ICU.

The authors noted that their medical center had previously tried to improve best practice adherence and preventable complication rates through educational outreach and checklist-based reminders, without achieving the results shown by the tele-ICU. They concluded that tele-ICUs can provide benefits even to hospitals that already have daytime intensivist staffing and active quality improvement efforts. Results were published in the June 1 Journal of the American Medical Association.

An accompanying editorial noted that another recent trial of a tele-ICU found no benefit to the intervention. A key difference was that on-site physicians in the prior study could decline the assistance of the remote intensivist, while this study had no opt-out provision. The tele-intensivists in the current study were also more active participants, reviewing patients' care to make sure it conformed to best practices and stated care plans, in addition to remote monitoring. The differing results show that implementation of a tele-ICU is not a universal solution to care deficiencies, but a potential tool for quality improvement, the editorialist concluded.

Reduced kidney function and albuminuria strongly associated with atrial fibrillation

Patients with kidney damage—manifested as microalbuminuria or macroalbuminuria—or decreased kidney function had a higher atrial fibrillation (AF) risk, researchers concluded.

In the Atherosclerosis Risk in Communities (ARIC) Study, 10,328 men and women aged 45 to 64 years who were free of AF were recruited from four geographically disparate U.S. regions from 1996 to 1998. Participants had four exams—a baseline, and three that were three years apart. Results appeared in the June 28 Circulation.

Atrial fibrillation was ascertained through the end of 2007 by hospital discharge codes and death certificates. During a median follow-up of 10.1 years, 788 atrial fibrillation cases occurred. Lower levels of eGFRcys were associated with a higher risk of AF, even after adjustment for potential confounders. Compared to patients with eGFRcys ≥90 mL per min per 1.73 m2, patients with National Kidney Foundation classification eGFRcys levels of 60 to 89, 30 to 59, and 15 to 29 mL per min per 1.73 m2 had multivariable hazard ratios for AF of 1.3, 1.6 and 3.2 (P for trend <0.0001), respectively.

Microalbuminuria and macroalbuminuria were associated with a higher risk, as well. Compared with patients with an albumin-to-creatine ratio (ACR) <30 mg/g, those with an ACR of 30 to 299 mg/g had twice the incidence of AF and for an ACR of ≥300 mg/g, the incidence was increased 3.2 times.

Risk of AF was particularly elevated in those with both low eGFRcys and macroalbuminuria (hazard ratio, 13.1, comparing individuals with ACR ≥300 mg/g and eGFRcys of 15 to 29 mL per min per 1.73 m2 and those with ACR <30 mg/g and eGFRcys ≥90 mL per min per 1.73 m2). An elevated risk of AF was observed even among individuals with mildly decreased kidney function measured by cystatin C-based glomerular filtration rate (eGFRcys), independent of lifestyle factors, cardiovascular disease, sex, race or hypertension.

“Given the growing burden of CKD in the general population and the potential for its prevention, future studies should focus on understanding the specific mechanisms underlying this association,” the authors wrote. “Furthermore, strategies for the prevention of AF will have to consider CKD as a preventable risk factor for AF in addition to other well-established risk factors.”

Cardiac rehabilitation after PCI associated with reduced mortality

Patients who participated in cardiac rehabilitation (CR) after percutaneous coronary intervention (PCI) were less likely to die than those who didn't, a study found.

In a retrospective analysis of data from 2,395 consecutive PCI patients, researchers examined the association of CR with all-cause mortality (primary outcome) and cardiac mortality, myocardial infarction and revascularization (secondary outcomes). They used three statistical techniques to examine the data, which came from patients in Olmsted County, Minnesota between 1994 and 2008. CR participation was defined as having attended at least one outpatient session within three months of the index PCI. The mean number of CR sessions per participant was 13.5. Patients were followed for a median of 6.3 years. Results were published in the May 31 Circulation.

During the study follow-up, 503 patients died (199 cardiac deaths), 394 had myocardial infarction, and 755 had revascularization procedures. Forty percent of the cohort (n=964) participated in CR, which was associated with a significant decrease in all-cause mortality by all three statistical techniques (hazard ratio, 0.53 to 0.55; P<0.001). The association was similar for men and women, older and younger patients, and for patients undergoing elective or nonelective PCI. A nonsignificant trend toward decreased cardiac mortality was also seen in CR participants, though there was no observed effect for myocardial infarction or revascularization. The number of PCI patients needed to treat with CR to prevent one death was 34 at one year after PCI and 22 at five years after PCI.

It's well known that CR helps lower mortality rates after myocardial infarction, but a similar association has been scarcely explored with PCI, the authors wrote. The current study finds CR participation is associated with lower mortality rates, which bolsters support for existing clinical practice guidelines, performance measures, and insurance coverage protocols that recommend CR after PCI, they said. The association of lower mortality with CR might occur due to the beneficial physical effects of exercise, improved medication adherence, risk factor control, reduced inflammation, increased identification and treatment for depression, and increased psychosocial support, the authors wrote.

Multidetector computed tomography should be standard care for suspected appendicitis

In patients with suspected appendicitis, multidetector computed tomography (MDCT) is a sensitive and specific test for identifying whether the condition exists, and helps identify probable alternative causes for those without appendicitis, an analysis found.

Researchers reviewed hospital records for 2,871 adults at an academic medical center who were referred to radiology for suspected appendicitis between 2000 and 2009. Original MDCT findings were evaluated at presentation by board-certified radiologists, with the findings used to determine diagnostic performance for appendicitis. The researchers used the final surgical pathology report as the reference standard for acute appendicitis, along with intraoperative findings and/or clinical follow-up, as appropriate. The surgical procedure, length of stay, and presence or absence of perforation to the appendix were also recorded. Results were published in the June 21 Annals of Internal Medicine.

Almost twenty-four percent of patients had confirmed acute appendicitis. The sensitivity of MDCT was 98.5%; specificity was 98%; and negative and positive predictive values were 99.5% and 93.9%, respectively. The overall rate of negative findings during appendectomy was 7.5%, but would have decreased to 4.1% if surgery had been avoided in 26 cases with true-negative findings on MDCT. The perforation rate declined from 28.9% in 2000 to 11.5% in 2009. MDCT either provided or suggested an alternative diagnosis in 42% of patients who didn't have appendicitis or appendectomy. Categories of suspected disease for these non-appendicitis patients included gastrointestinal conditions (55.5%), gynecologic conditions (21.7%) and genitourinary conditions (16.5%).

Study limitations included a possible referral bias, as some patients whose appendicitis was difficult to diagnose clinically may not have been referred for MDCT to evaluate suspected appendicitis. The diagnostic performance of MDCT didn't differ by sex. Although women had a higher rate of negative findings at appendectomy, the authors noted that this difference was driven by patient selection and the underlying prevalence of appendicitis rather than MDCT performance. The results “support routine use of preoperative MDCT as the standard of care for suspected appendicitis in adults,” as it can identify those who need urgent surgery, and offer a probable cause for those without appendicitis, the authors concluded.

STOPP criteria work better than Beers to halt adverse drug events

Adverse drug events in elderly patients could potentially be reduced by use of the STOPP (Screening Tool of Older Persons' potentially inappropriate Prescriptions) criteria, a new study suggested.

The prospective study included 600 patients 65 years or older admitted to an Irish teaching hospital. Researchers observed 329 adverse drug events in the records of 158 of the patients. About 66% of those adverse events had contributed to or caused the hospital admission, according to the researchers' evaluation. Of those causal/contributory events, 68.9% were avoidable or potentially avoidable. The study was published in the June 13 Archives of Internal Medicine.

The study authors then compared the drugs that had caused the adverse events with two established criteria for avoiding potentially inappropriate medications in elderly patients—the STOPP and the Beers criteria. They found that the drugs involved in the adverse events were 2.54 times more likely to meet STOPP criteria than Beers criteria. Looking only at the events that were avoidable and contributory to the hospitalization, the researchers found that they were 2.8 times more likely to be on the STOPP list (P<0.001).

After adjustment for a number of factors, the researchers found that patients taking a medication on the STOPP list were 85% more likely to have an adverse drug event than those who weren't (odds ratio, 1.847; CI, 1.506-2.264; P<0.001). In contrast, prescription of potentially inappropriate medications from the Beers list did not significantly increase the likelihood of an adverse drug event (odds ratio, 1.276; CI, 0.945-1.722; P<0.11). The results strengthen the argument for using STOPP criteria as a routine screening tool in everyday clinical practice, the authors concluded. Further research is needed to determine definitively whether use of the STOPP criteria reduces adverse events, medication costs and health care utilization, the authors said.

Tools such as STOPP cannot capture all potentially inappropriate prescribing, the authors cautioned, and should be used to enhance, rather than replace, clinical judgment in prescribing for older patients. An accompanying commentary also noted that tools have to be implemented effectively to provide benefit. Greater incorporation of electronic prescribing into clinicians' workflows should help with this, the commentary author wrote. He also suggested that future research assess any unintended consequences of efforts to avoid potentially dangerous medications, for example, pain in a patient taken off opiate therapy to reduce the risk of falls.

43% of ischemic stroke patients discharged with high blood pressure

Forty-three percent of patients hospitalized with acute ischemic stroke were discharged with elevated blood pressure, and 33% had uncontrolled blood pressure six months later, a new analysis found.

Researchers examined a sample of patients (n=3,987) who had were admitted to a Veterans Affairs Medical Center and hospitalized for ischemic stroke in 2007. They analyzed blood pressure control (defined as <140/90 mm Hg) at discharge, and excluded 347 who had died, enrolled in hospice, or had unknown discharge disposition. Another 258 patients were excluded for missing race data, leaving 3,382 patients from 129 facilities for the first analysis. Researchers also examined all antihypertensive medications prescribed at admission and discharge, and compared to see if patients received a new prescription for a drug class at discharge. In a second analysis, they looked at blood pressure control within six months after stroke, excluding those who had died, were readmitted within 30 days, were lost to follow-up, or didn't have blood pressure or race recorded, leaving 1,915 from 125 facilities. Results were published in the July Circulation: Cardiovascular Quality and Outcomes.

About sixty-three percent of the study population was white, and 98% were men. Forty-seven percent were younger than age 65, 29% had a history of cerebrovascular disease, and 37% had a history of cardiovascular disease. Among the stroke patients in the first analysis, 43% had their last documented blood pressure before discharge as ≥140/90 mm Hg. Black race, diabetes and hypertension history were associated with lower odds for controlled blood pressure at discharge. Of the stroke patients seen within six months of their index event, 32.8% still had uncontrolled blood pressure. By six months after the event, neither race nor diabetes was associated with blood pressure control, while history of hypertension continued to predict lower odds of control. For each 10-point increase in systolic blood pressure at discharge over 140 mm Hg, the odds of control within six months after discharge decreased by 12%. Receipt of a new blood pressure medication at discharge was associated with decreased odds of blood pressure control at six months, possibly because sicker patients are more likely to get new prescriptions and have poor control, the authors wrote.

The study data suggest that heightened efforts to improve management of hypertension at discharge and follow-up may benefit certain subgroups of patients, given that hypertension is causally involved in nearly 70% of all stroke cases and puts patients at risk for cardiovascular events, the authors said. Secondary prevention should include efforts to start risk factor control and antihypertensive medication before discharge. Future interventions could target those at highest risk for poorly controlled blood pressure, including those with a prior diagnosis of hypertension and multiple comorbidities, including diabetes, the authors said.