Atelectasis and fever; heparin dosing frequency for VTE prophylaxis; perioperative cardiac risk calculator; diagnosing subarachnoid hemorrhage without an LP; model to predict risk of bleeding on warfarin; risk of death with tiotropium use in COPD; BNP to predict perioperative mortality; beta-blockers and COPD.
by Aaron Hamilton, MD, MBA, Vijaiganesh Nagarajan, MD, MRCP, Anbazhagan Prabhakaran, MD, MRCP, Thadeo Catacutan, MD, Chadi Alraies, MD, Rubin Bahuva, MD
A guide to this month’s studies
Clinical question: Is atelectasis really a major cause of early (up to 48 hours) postoperative fever (EPF)?
Background: Both fever and atelectasis are common findings in the postoperative period. EPF is believed to be noninfectious, and many textbooks consider atelectasis to be the most common cause. However, this association is controversial with no clear evidence.
Study design: Systematic review of prospective studies evaluating atelectasis and postoperative fever using PubMed and Scopus databases.
Setting: Postoperative patients (predominantly cardiac, maxillofacial, and abdominal surgeries). Lung surgery patients were excluded.
Synopsis: Eight prospective studies (four interventional and four observational) with 998 patients were included for review. All studies diagnosed atelectasis based on chest imaging but only three studies used the conventional definition of ≥38°C for fever. Seven studies individually reported no association between atelectasis and EPF.
Only five studies had eligible data for pooling and analysis. EPF was found to be a very weak indicator (diagnostic OR 1.4; 95% CI 0.92-2.12) of atelectasis. EPF also fared poorly for ruling out (sensitivity 13% to 47%) or ruling in (specificity 41% to 87%) the diagnosis of atelectasis with similarly poor positive and negative predictive values.
The results of this study, however, should be interpreted with caution. It was not a formal meta-analysis, due to the heterogeneity of the studies included with regard to the definition of fever, time points of imaging, and the variation of end points.
Bottom line: Since there is no clinical evidence to prove an association between atelectasis and fever, it is presumed that atelectasis may not be a cause of EPF.
Citation: Mavros MN, Velmahos GC, Falagas ME. Atelectasis as a cause of postoperative fever: where is the clinical evidence? Chest. 2011;140:418-424.
Clinical question: Which is the best dosing frequency of unfractionated heparin (UFH) in preventing venous thromboembolism?
Background: Low-dose UFH is commonly used in hospitals for pharmacologic prophylaxis against venous thromboembolism. However, the risks and benefits of BID vs. TID dosing are not clear.
Study design: Mixed-treatment comparison (MTC) meta-analysis of RCTs.
Setting: RCTs on thromboprophylaxis regimens, selected from two previous systematic reviews and an updated literature search.
Synopsis: Included in the analysis were 27,667 patients from 16 RCTs comparing three prophylactic regimens (UFH BID, UFH TID, or low-molecular-weight heparin) with each other or with controls. Stroke and some myocardial infarction patients were excluded. The outcomes measured were DVT, pulmonary embolism (PE), major bleeding, and death. As compared with controls, all three regimens significantly reduced DVT (ranging from 58% to 72%), showed a nonsignificant trend toward reduction in PE (by 46% to 67%), and had no difference in risk of major bleeding or death.
UFH BID vs. TID were compared indirectly by using data from their trials against control patients or low-molecular-weight heparin. There was no significant difference between UFH TID and BID in reducing DVT (RR 1.56, CI 0.64-4.33), PE (RR 1.67, CI 0.49-208.9), mortality (RR 1.17, CI 0.72-1.95), or causing major bleeding (RR 0.89, CI 0.08-7.05). Additionally, both UFH dosing frequencies were similar to low-molecular-weight heparin in all four measured outcomes. This evidence is of moderate quality due to the lack of a direct comparison between UFH BID vs. TID.
Bottom line: Both BID and TID dosing of UFH are acceptable thromboprophylaxis regimens in hospitalized medical patients with no difference in effect on DVT, PE, major bleeding, or death.
Citation: Phung OJ, Kahn SR, Cook DJ, et al. Dosing frequency of unfractionated heparin thromboprophylaxis: a meta-analysis. Chest. 2011;140: 374-381.
Clinical question: Can a more accurate risk calculator than the Revised Cardiac Risk Index (RCRI) be developed and validated to predict postoperative cardiac events?
Background: The majority of perioperative deaths are secondary to cardiac-related events. The RCRI is the most commonly used preoperative risk stratification tool, but it has limitations and low discriminatory ability.
Study design: Multicenter prospective National Surgical Quality Improvement Program database study.
Setting: More than 250 academic and community U.S. hospitals.
Synopsis: Data were obtained from patients over a two-year period (2007 and 2008). From the 2007 data set (n=211,410), perioperative myocardial infarction or cardiac arrest (MICA) was seen in 1,371 patients (0.65%). After multivariate analysis on the 2007 data set, five risk predictors were obtained (increasing age, American anesthesiology class, dependent functional status, abnormal serum creatinine of >1.5 mg/dL, and type of surgery). This was validated utilizing the 2008 data set (n=257,385), where MICA was seen in 1,401 patients (0.54%).
The risk-predictive model showed excellent discrimination (distinguishing between events and nonevents) after application of C statistics to the dataset. The discriminatory ability was better when compared with the RCRI model. Limitations included nonavailability of information on preoperative stress test, arrhythmia, and aortic valve disease.
Bottom line: The new risk calculator model would help predict MICA more accurately, which in turn would help in preoperative optimization and patient counseling.
Citation: Gupta PK, Gupta H, Sundaram A, et al. Development and validation of a risk calculator for prediction of cardiac risk after surgery. Circ. 2011;124:381-387.
Clinical question: Are modern third-generation CT scans good enough to exclude subarachnoid hemorrhage (SAH) without a lumbar puncture (LP)?
Background: SAH is a neurosurgical emergency identified in about 1% of patients with headache in the emergency department. As the standard of care, all patients with suspected SAH have to undergo LP if a CT scan of the brain is normal. However, LP causes pain and delays discharge from the emergency department.
Study design: Prospective multicenter cohort study.
Setting: Eleven tertiary-care Canadian emergency departments.
Synopsis: From November 2000 to December 2009, data on all alert patients (n=3,132) who presented with acute headache and underwent emergent head CT were collected. Of these, 240 had SAH (7.7%). The sensitivity of CT overall for detecting SAH was 92.9% and the specificity was 100%. For the 953 patients scanned within six hours of headache onset, all 121 patients with SAH were identified by CT, yielding a sensitivity of 100% and specificity of 100%.
The study was limited largely by the lack of a consensus definition on the diagnosis of SAH and by some patient enrollment issues in the emergency department. Overall, these findings should give clinicians more confidence in forgoing an LP in patients with a negative head CT if done within six hours of the onset of their headache.
Bottom line: Modern third-generation CT scans are extremely sensitive for SAH if performed within six hours of the headache onset and interpreted by a qualified radiologist, thus possibly excluding the need for an LP.
Citation: Perry JJ, Stiell IG, Sivilotti ML, et al. Sensitivity of computed tomography performed within six hours of onset of headache for diagnosis of subarachnoid haemorrhage: prospective cohort study. Br Med J. 2011;343:d4277.
Clinical question: Can a simple scoring model accurately assess the risk of warfarin-associated bleeding in a cohort of patients with atrial fibrillation?
Background: It is well known that anticoagulants, such as warfarin, dramatically reduce the risk of thromboembolic events in patients with atrial fibrillation. Despite this, clinicians often find themselves weighing the risks and benefits of anticoagulation in this cohort of patients, and improved models to assess those risks are needed.
Study design: Retrospective cohort study.
Setting: Kaiser Permanente of Northern California.
Synopsis: From a cohort of 13,559 adult patients with atrial fibrillation, the investigators used chart review to determine hemorrhagic events in this population and developed a model using Cox regression to assess hemorrhagic risk in certain patient populations. Final input variables for the model included anemia, severe renal disease, age ≥75, prior hemorrhage, and hypertension. When collapsed into three risk tiers (low, intermediate, and high), the scoring model nicely differentiated low (<1% annual) from high (5.8% annual) bleeding risk.
This study is limited by the lack of information on concomitant use of NSAIDs or aspirin in these patients and the lack of external validation of the model. Despite those limitations, it may serve as a valuable tool for clinicians. As the number of alternatives to warfarin rise and as those agents become more familiar, it will become increasingly important to accurately assess hemorrhage risk with various anticoagulants.
Bottom line: The Anticoagulation and Risk Factors in Atrial Fibrillation (ATRIA) risk scoring system is a reliable and easy way for clinicians to estimate the degree of bleeding risk in patients anticoagulated with warfarin for atrial fibrillation.
Citation: Fang MC, Go AS, Chang Y, et al. A new risk scheme to predict warfarin-associated hemorrhage: the ATRIA (anticoagulation and risk factors in atrial fibrillation) study. J Am Coll Cardiol. 2011;58:395-401.
Clinical question: Does the mist-inhaler formulation of tiotropium increase mortality in patients with chronic obstructive pulmonary disease (COPD) when compared with placebo?
Background: Tiotropium is used in patients with COPD to reduce both symptoms of dyspnea and exacerbations of COPD. Tiotropium comes in two formulations: a powder (approved in the U.S.) and the mist inhaler (not approved in the U.S. but approved in 55 other countries). There are concerns based on recent studies that tiotropium may increase cardiovascular events and death.
Study design: Meta-analysis of five randomized controlled trials (RCTs) comparing tiotropium mist inhaler with a placebo.
Setting: Multinational studies.
Synopsis: This study of 6,522 patients with COPD showed a 52% increased risk of all-cause mortality with the use of the tiotropium mist inhaler when compared with placebo. It is important to note that there are data showing higher plasma concentrations with the approved mist-inhaler doses when compared with the powder formulation doses. Further, a possible dose effect was seen in this study (though not statistically significant), with higher tiotropium doses associated with a high-risk ratio for the mortality endpoint.
Limitations of this study include the fact that the dosage of the tiotropium varied, as did the length of follow-up for patients. Given that death was a relatively rare event (<1%), estimates are imprecise. Even given these limitations, this study sheds light on the debate over the safety of tiotropium, specifically the mist-inhaler formulation. Caution should be used when prescribing the mist-inhaler formulation of tiotropium, and an understanding of the potential cardiovascular risks should be communicated to patients prior to initiating therapy.
Bottom line: This study shows that the mist-inhaler formulation of tiotropium is associated with an increased risk of cardiovascular mortality.
Citation: Singh S, Loke YK, Enright PL, Furnberg CD. Mortality associated with tiotropium mist inhaler in patients with chronic obstructive pulmonary disease: systematic review and meta-analysis of randomised controlled trials. Br Med J. 2011;342: d3215.
Clinical question: Can preoperative natriuretic peptide levels be used to independently predict perioperative cardiovascular events in patients undergoing vascular surgery?
Background: Currently we use the type of surgery, exercise tolerance, and clinical risk factors to predict perioperative cardiovascular risk. Clinical risk factors, based on the Revised Cardiac Risk Index, or RCRI, include history of ischemic heart disease, heart failure, cerebrovascular events, diabetes mellitus, and renal insufficiency. Recent studies have shown that the pre-operative natriuretic peptides can independently predict perioperative cardiovascular events.
Study design: Individual patient meta-analysis.
Setting: Data sets obtained from six multinational studies.
Synopsis: This meta-analysis included datasets from five studies that used BNP (632 patients) and one study that used NT-proBNP (218 patients) to assess the postoperative cardiovascular events in patients undergoing vascular surgery. Patients with elevated BNP level are at a higher risk of cardiac death (OR 4.3, 95% CI: 1.7-11.3) and all-cause mortality (OR 3.1, 95% CI: 1.4-6.7) within 30 days of vascular surgery. When the RCRI-based groups were reclassified using natriuretic peptide level, the improvement in discrimination was statistically significant. Limitations of this study include: 1) Individual patient data was not obtained for all studies that met the search criteria; and 2) Different types of BNP assays were used in different studies included.
Bottom line: Preoperative BNP level is an independent predictor of cardiovascular events at 30 days after vascular surgery. The addition of preoperative BNP level improves the predictive performance of the RCRI score.
Citation: Rodseth RN, Lurati Buse GA, Bolliger D, et al. The predictive ability of pre-operative B-type natriuretic peptide in vascular patients for major adverse cardiac events: an individual patient data meta-analysis. J Am Coll Cardiol. 2011;58:522-529.
Clinical question: Is it beneficial to use beta-blockers in patients with chronic obstructive pulmonary disease (COPD) if there is an indication?
Background: Patients with COPD may have concomitant cardiovascular disease, which may warrant use of beta-blockers. Many physicians are concerned about using beta-blockers in COPD patients due to the risk of bronchospasm. Evidence suggests that cardio-selective beta-blockers do not cause deterioration of pulmonary status in COPD patients. There is also growing evidence that beta-blockers may be beneficial in patients with COPD.
Study design: Retrospective cohort study.
Setting: Data obtained from a disease-specific (COPD) database in Scotland.
Synopsis: This study included 5,977 patients who were older than 50 and excluded patients with history of malignancy. Beta-blockers were associated with a 22% reduction in all-cause mortality. There was no significant difference between cardio-selective and nonselective beta-blockers. The benefits of beta-blockers in COPD patients were independent of history of cardiovascular disease. There was no significant decline in pulmonary function (FEV1) over time. Beta-blocker usage also reduced the number of hospital admissions for COPD exacerbation. These benefits were shown in patients using different type of inhalers.
Bottom line: In patients older than 50, beta-blockers may not only reduce COPD exacerbations and hospital admissions, but also reduce all-cause mortality without adversely affecting pulmonary function.
Citation: Short PM, Lipworth SI, Elder DH, Schembri S, Lipworth BJ. Effect of beta blockers in treatment of chronic obstructive pulmonary disease: a retrospective cohort study. Br Med J. 2011;342:d2549.
The Hospitalist newsmagazine reports on issues and trends in hospital medicine. The Hospitalist reaches more than 25,000 hospitalists, physician assistants, nurse practitioners, residents, and medical administrators interested in the practice and business of hospital medicine.