Luis A. Serrano

Accuracy and Quality of Clinical Decision Rules for Syncope in the Emergency Department: A Systematic Review and Meta-analysis

STUDY OBJECTIVE We assess the methodological quality and prognostic accuracy of clinical decision rules in emergency department (ED) syncope patients. METHODS We searched 6 electronic databases, reviewed reference lists of included studies, and contacted content experts to identify articles for review. Studies that derived or validated clinical decision rules in ED syncope patients were included. Two reviewers independently screened records for relevance, selected studies for inclusion, assessed study quality, and abstracted data. Random-effects meta-analysis was used to pool diagnostic performance estimates across studies that derived or validated the same clinical decision rule. Between-study heterogeneity was assessed with the I(2) statistic, and subgroup hypotheses were tested with a test of interaction. RESULTS We identified 18 eligible studies. Deficiencies in outcome (blinding) and interrater reliability assessment were the most common methodological weaknesses. Meta-analysis of the San Francisco Syncope Rule (sensitivity 86% [95% confidence interval {CI} 83% to 89%]; specificity 49% [95% CI 48% to 51%]) and the Osservatorio Epidemiologico sulla Sincope nel Lazio risk score (sensitivity 95% [95% CI 88% to 98%]; specificity 31% [95% CI 29% to 34%]). Subgroup analysis identified study design (prospective, diagnostic odds ratio 8.82 [95% CI 3.5 to 22] versus retrospective, diagnostic odds ratio 2.45 [95% CI 0.96 to 6.21]) and ECG determination (by evaluating physician, diagnostic odds ratio 25.5 [95% CI 4.41 to 148] versus researcher or cardiologist, diagnostic odds ratio 4 [95% CI 2.15 to 7.55]) as potential explanations for the variability in San Francisco Syncope Rule performance. CONCLUSION The methodological quality and prognostic accuracy of clinical decision rules for syncope are limited. Differences in study design and ECG interpretation may account for the variable prognostic performance of the San Francisco Syncope Rule when validated in different practice settings.

Effect of anticoagulant and antiplatelet therapy in patients with spontaneous intra-cerebral hemorrhage: Does medication use predict worse outcome?

OBJECTIVES To assess the impact of anticoagulants and antiplatelet agents on the severity and outcome of spontaneous non-traumatic intra-cerebral hemorrhage (ICH). To evaluate associations between reversal of anticoagulation and mortality/morbidity in these patients. METHODS Data was collected on a consecutive cohort of adults presenting with ICH to an academic Emergency Department over a 3-year period starting January 2006. RESULTS The final cohort of 245 patients consisted of 125 females (51.1%). The median age of the cohort was 73 years [inter-quartile (IQR) range of 59-82 years]. Antiplatelet (AP) use was seen in 32.6%, 18.4% were using anticoagulant (AC) and 8.9% patients were on both drugs (AC+AP). Patients on AC had significantly higher INR (median 2.3) and aPTT (median 31 s) when compared to patients not on AP/AC (median INR 1.0, median aPTT 24s; p<0.001). Similarly patients on AC+AP also had higher INR (median 1.9) and aPTT (median 30s) when compared to those not on AC/AP (p<0.001). Hemorrhage volumes were significantly higher for patients on AC alone (median 64.7 cm(3)) when compared to those not on either AC/AP (median 27.2 cm(3); p=0.05). The same was not found for patients using AP (median volume 20.5 cm(3); p=0.813), or both AC+AP (median volume 27.7 cm(3); p=0.619). Patients on AC were 1.43 times higher at risk to have intra-ventricular extension of hemorrhage (IVE) as compared to patients not on AC/AP (95% CI 1.04-1.98; p=0.035). There was no relationship between the use of AC/AP/AC+AP and functional outcome of patients. Patients on AC were 1.74 times more likely to die within 7 days (95% CI 1.0-3.03; p=0.05). No relationship was found between use of AP or AC+AP use and mortality. Of the 82 patients with INR>1.0, 52 patients were given reversal (minimum INR 1.4, median 2.3). Therapy was heterogeneous, with fresh frozen plasma (FFP) being the most commonly used agent (86.5% patients, median dose 4U). Vitamin K, activated factor VIIa and platelets were the other agents used. Post reversal, INR normalized within 24h (median 1.2, IQR 1.1-1.3). There was no association between reversal and volume of hemorrhage, IVE, early mortality (death<7 days) or functional outcome. CONCLUSIONS Anticoagulated patients were at 1.7 times higher risk of early mortality after ICH. Reversal of INR to normal did not influence mortality or functional outcome.

Understanding statistical tests in the medical literature: which test should I use?

When writing or reading articles, one should be aware whether the statistical tests performed were appropriate for the type of data collected and used, thereby avoiding misleading conclusions. The goal of all statistical tests is to determine whether two (or more) variables are associated with one another or independent from each other at the population level.

Low Back Pain Emergencies

Low back pain (LBP) is exceedingly common. Most patients with LBP present acutely or subacutely. The clinician is frequently faced with the task of determining whether or not the individual LBP patient has an emergent or soon-to-be-emergent underlying condition. The approach to the patient with acute or subacute LBP includes a search for red flags in the history and careful physical and neurological examinations that can indicate the likelihood of an underlying urgent or emergent condition. In the absence of red flags, patients can be treated conservatively for 1 month or more without diagnostic testing.