Journal Watch

Recent studies about perioperative consultation, increased U.S. staph infections, and other topics.

Perioperative consultation has inconsistent effects on outcome, study finds

Perioperative medical consultation with an internist doesn't necessarily improve patients' outcomes, according to a study.

Researchers performed an observational study of 1,282 patients who had surgery at the University of California, San Francisco, Medical Center between May 1, 2004, and May 31, 2006. The authors examined costs, length of hospital stay, preventive therapy and outcomes in those who received a medical consultation and those who did not. The study results were published in the Nov. 26 Archives of Internal Medicine.

One hundred seventeen patients (9.1%) had a medical consultation the day before, the day after, or the day of their surgery. Patients who had a consultation were, in general, sicker than those who did not. After adjustment for illness severity and consultation likelihood, the authors found that patients in both groups were equally likely to have well-controlled glucose levels, to receive beta-blockers and to receive prophylaxis for venous thromboembolism. In addition, patients who received a consultation had longer adjusted hospital stays.

The authors noted a high likelihood that their results were affected by methodologic considerations. In addition, consultants' recommendations may have had too narrow a focus and in some cases may not have been followed. However, the authors concluded that perioperative medical consultation has “inconsistent effects” on care efficiency and quality and that the current consultation system could be improved. They called for further study of different models of surgical comanagement.

New estimates support trend of increased U.S. staph infections

Infections with Staphylococcus aureus, including methicillin-resistant strains, have increased dramatically in U.S. hospitals, a study has found. Researchers used data from the National Hospital Discharge Survey and data on S. aureus resistance to determine hospitalizations and deaths associated with S. aureus in U.S. hospitals from 1999 to 2005. The results appeared in the December Emerging Infectious Diseases.

Over the seven-year study period, estimated S. aureus-associated hospitalizations increased by 62% (from 294,570 to 477,927), while estimated hospitalizations related to methicillin-resistant S. aureus (MRSA) more than doubled (from 127,036 to 278,203). In addition, S. aureus resistance to cephalothin, erythromycin and ampicillin/sulbactam increased 35%, 27% and 21%, respectively, from 1999 to 2005, while resistance to trimethoprim-sulfamethoxazole and gentamicin decreased 64% and 76%, respectively. Marked increases were also observed in skin and soft-tissue infections, which are usually community-associated.

These results indicate that MRSA and S. aureus should be priorities for disease control in the U.S., the authors wrote. Possible next steps could include enhanced surveillance or reporting programs and additional efforts to control infections in hospitals, they suggested.

Positive observational findings cited despite negative RCT results

Some observational research findings—such as the cardiovascular benefits of vitamin E—continue to be cited long after they have been contradicted by randomized, controlled trials, according to an analysis.

For the analysis, researchers collected journal articles from 1997, 2001 and 2005 which referenced one or both of two 1993 epidemiological studies proposing cardiovascular benefits of vitamin E. A major randomized trial published in January 2000 found no benefit from vitamin E and a 2004 meta-analysis found that high doses increased the risk of death, the study authors noted. Despite the contradiction of the original studies' claims, 50% of the articles published in 2005 that cited the first studies described them favorably.

When the researchers then looked at all the 2005 publications which referenced the contradictory evidence (the randomized trial that refuted the benefits of vitamin E), they found that 41.4% of the references were unfavorable to the randomized trial and cited a variety of counterarguments. The analysis was published in the Dec. 5 Journal of the American Medical Association.

The study authors also looked at two other observational findings which were later contradicted by randomized trials—the protective effects of beta-carotene on cancer and estrogen on Alzheimer's disease. The findings on beta-carotene were published in 1981 and contradicted in 1994-1996, and the estrogen/Alzheimer's research was published in 1996 and contradicted in 2004. Yet in 2006, more than 60% of the articles that cited these observational studies referred to them favorably, the new analysis found.

The study authors concluded that researchers, especially those publishing in specialty journals, are slow to give up some positive ideas proposed by observational studies even when refuted by controlled trials. They concluded that wish bias may be affecting researchers' interpretations of study results, and suggested that better communication between research specialists and evidence-based clinical science could lead to more rational use of research.

Many in-hospital defibrillations are delayed, study finds

Almost a third of patients who have in-hospital cardiac arrests are not defibrillated within the recommended two minutes.

The observational research included 6,789 patients who had cardiac arrest due to ventricular fibrillation or pulseless ventricular tachycardia at 369 hospitals. Although the overall median time to defibrillation among the patients was one minute, 30% of the studied patients received delayed defibrillation. The researchers found several factors that increased the likelihood of delay: patients who were black, had a noncardiac admitting diagnosis, were at a hospital with fewer than 250 beds, or were on an unmonitored hospital unit. Delays were also more likely after hours (between 5 p.m. and 8 a.m. and on weekends).

The study also found that patients whose defibrillation was delayed were 48% less likely to survive to hospital discharge (22% in the group who were defibrillated within two minutes died vs. 39% of the delayed). A graded association was found between increasing time to defibrillation and poorer survival, both before and after two minutes. Patients who received delayed defibrillation also were more likely to have major disabilities in neurologic or functional status, according to the study.

The study authors noted that although several of the delaying factors were clearly hospital-related and resulted from limited availability of personnel and failure to quickly recognize ventricular arrhythmia, other associated factors, such as black race, raise concerns of disparity and merit further study. The study was published in the Jan. 3 New England Journal of Medicine.

An accompanying editorial suggested that automatic detection algorithms could help remedy the problems highlighted by the study. Electrodes on the patient's skin could wirelessly transmit electrocardiogram data to a centralized alarm and computer station, which would alert nurses if the algorithm indicated ventricular tachycardia or ventricular fibrillation. Such a system could alleviate staffing-related factors as well as any disparities due to bias, the editorialist said.

ED prescriptions for pain vary by ethnicity

Nonwhites presenting to the emergency department are less likely than whites to be prescribed opioids for pain, according to a new study.

Over the past decade, The Joint Commission and the Veterans Health Administration have conducted campaigns to increase the awareness of appropriate analgesia and improve treatment for pain. Researchers used data from the National Hospital Ambulatory Medical Care Survey to examine whether these national initiatives affected prescription of opioids for pain in U.S. emergency departments. The results appear in the Jan. 2 Journal of the American Medical Association.

From 1993 to 2005, 42% of ED visits (156,729 of 374,891) were pain-related. Opioids were prescribed for 23% of pain-related ED visits in 1993 and 37% in 2005, with a more pronounced trend toward an increase in such prescribing from 2001 to 2005 (P = 0.02).

However, although opioid prescribing increased overall, racial disparities in opioid prescribing did not change. For all years examined, white patients were more likely than black, Hispanic or Asian/other patients to receive opioids for pain in the ED (31% vs. 23%, 24% and 28%, respectively). In 2005, opioid prescribing rates were 40% for white patients and 32% for patients of all other ethnicities. Racial disparities were observed for all types of pain-related visits and persisted after adjustment for pain severity and other factors.

The authors concluded that national initiatives to improve appropriate analgesia resulted in increased opioid prescribing for pain-related ED visits overall but had no effect on racial disparities in such prescribing. While these results may be attributed to overprescribing of opioids for pain among whites, the authors believe they are more likely to indicate undertreatment of pain among racial and ethnic minorities. The authors noted that such disparities are complex and that ethnic and racial bias alone is unlikely to completely explain their findings. System-level changes and future initiatives involving patients and nurses as well as physicians could help decrease these disparities, the authors wrote.

Rural patients less likely to get organ transplants

Americans who live in rural areas are less likely to receive organ transplants, or even to be put on waiting lists for hearts, livers and kidneys, according to a study in the Jan. 9/16 Journal of the American Medical Association.

The findings came from a cohort study of more than 174,000 patients who were wait-listed and underwent heart, liver or kidney transplantation between 1999 and 2004. Compared with urban residents, patients who lived in small towns and isolated rural areas were 9% less likely to be placed on the list for a heart, 14% less likely to make the liver list and 8% less likely to be on the list for kidney transplants. In addition, rural residents were 12% less likely to get a heart, 20% less likely to get a liver, and 10% less likely to receive a kidney transplant than urbanites. After receiving a transplant, the rural and urban patients did not have any significant difference in outcomes.

The researchers could not definitively determine whether the observed disparities were due to underlying disease burden or specific barriers faced by rural residents. However, they did suggest the disparity may reflect the difficulties faced by rural patients in completing the complex referral and evaluation process required for transplantation. Delayed referrals to specialists and greater distances to transplant centers could also contribute to the problem. With the increasing concentration of transplant centers in urban areas, it is possible that this disparity will only grow, and further assessment of the burdens facing rural residents requiring transplants is needed, the authors concluded.