Guidelines for health care-associated pneumonia not always followed
Physicians are aware of and even agree with guidelines for antibiotic treatment of health care-associated pneumonia but don't always follow them, a recent survey found.
Guidelines for treatment of health care-associated pneumonia (HCAP) stress appropriate use of antibiotics to avoid development of multidrug resistance. Researchers conducted an online survey at four academic medical centers to determine whether physicians would select guideline-approved therapy in nine hypothetical cases, seven involving HCAP and two involving community-acquired pneumonia (CAP). Physicians were also asked about their knowledge of and agreement with current guideline recommendations for HCAP. The study results were published in the Dec. 15, 2009 Clinical Infectious Diseases.
Overall, 855 of 1,313 physicians completed the survey (response rate, 65%). Sixty percent of respondents were hospitalists or internists, 25% were emergency medicine physicians, and 13% specialized in critical care. Respondents were much more likely to select guideline-concordant therapy for the CAP cases than for the HCAP cases (78% vs. 9%). Hospitalists and internists performed worse on HCAP questions than emergency medicine and critical care physicians, but the scores overall were so low that differences were too small to be meaningful, the study authors wrote. Seventy-one percent of physicians reported being familiar with the HCAP guidelines, and 79% reported agreeing with and following them in clinical practice.
Study limitations include uncertainty about how closely physicians' responses to hypothetical cases mirrored their clinical practice and potential lack of generalizability. However, the findings could have “potentially serious implications” since inappropriate antibiotic treatment is known to increase mortality rates in patients with HCAP, the authors said.
PCI use on STEMI patients varies widely among hospitals
Hospitals that are capable of performing percutaneous coronary intervention (PCI) on ST-segment elevation myocardial infarction (STEMI) patients don't always do so, and certain clinical factors make PCI use less likely, a study found.
Researchers analyzed STEMI patients in the National Registry of Myocardial Infarction who were at PCI-capable hospitals between July 1, 2000, and December 31, 2006. At the 444 hospitals, 25,579 patients received primary PCI and 14,332 received fibrinolytic therapy. While overall PCI use increased over the study period, reperfusion strategies varied widely among hospitals. The study was published in the December 2009 Circulation.
The strongest association was for patients who presented on weekends and in the evening; they had an approximately 70% lower likelihood of undergoing primary PCI (P<0.0001; adjusted odds ratio, 0.27; 95% CI, 0.25 to 0.29). Cardiogenic shock was associated with greater use of primary PCI (P<0.0001; OR, 2.14; 95% CI, 1.72 to 2.66), as was delayed presentation (P<0.0001; OR, 1.18; 95% CI, 1.09 to 1.27)—although nearly 25% of patients with the former received fibrinolytic therapy. A Thrombolysis in Myocardial Infarction risk score ≥5 wasn't associated with greater PCI use.
The finding that off-hours presentation was associated with a lower likelihood of undergoing PCI may partially reflect the expectation of long time-to-treatment delays, thus “leading to the appropriate selection of fibrinolytic therapy,” the authors said. Because there was no interaction between key clinical factors in the PCI-preferred group and the time of hospital arrival, it appears the use of PCI during off-hours wasn't being reserved for patients most likely to benefit, they added.
Simple tool effective in predicting mortality for dialysis patients
A simple five-point prognostic tool was effective in predicting six-month survival of patients with end-stage renal disease, a study reported.
Researchers monitored survival for up to 24 months of 512 patients at five dialysis clinics and then tested the prognostic model with a validation cohort of 514 patients at eight clinics. Five variables were independently associated with early mortality: older age, dementia, peripheral vascular disease, decreased albumin and a response of “no” to a “surprise question” (nephrologists were asked “Would I be surprised if this patient died within the next six months?”). The results were published Jan. 1 in the Clinical Journal of the American Society of Nephrology.
Nephrologists are often hesitant to give a prognosis for dialysis patients due to the questionable accuracy of existing prediction tools, the authors noted. The new predictive model, which showed a relatively high level of accuracy, represents a potentially valuable tool for physicians to identify patients with a poor prognosis who could benefit from palliative care and support services, the authors said.
The surprise question is an innovative aspect of the new model and played a key role in the accurate risk stratification of patients, researchers said. At the end of the study, almost 55% of patients nephrologists classified into the “No” group had died compared with 17% in the “Yes” group.
The new prognostic model is more specific and sensitive than any one of its components and is a significant improvement over existing methods at predicting survival of patients on dialysis, researchers concluded. They noted that future research should focus on other instruments that combine actuarial and clinical estimates of survival.