Predictors of hospital admission after syncope: Relationships with clinical risk scores Francesca Perego a, ⁎ , 1 , Giorgio Costantino b, 1 , Franca Dipaola c, 1 , Emanuela Scannella a, 1 , Marta Borella b, 1 , Andrea Galli d, 1 , Franca Barbic f, 1 , Francesco Casella a, 1 , Monica Solbiati b, 1 , Laura Angaroni b, 1 , Piergiorgio Duca e, 1 , Raffaello Furlan f, g, 1 a Internal Medicine 3, “Luigi Sacco” Hospital, Milan, Italy b Internal Medicine 2, “Luigi Sacco” Hospital, Milan, Italy c Internal Medicine, Sesto S. Giovanni Hospital, Sesto S. Giovanni, Milan, Italy d Emergency Department, Vimercate Hospital, Vimercate, Milan, Italy e Medical Statistics and Biometry Unit, Clinical Science Department, “Luigi Sacco” Hospital, Milan, Italy f Internal Medicine, Istituto Clinico Humanitas, Rozzano, Milano, Italy g University of Milan, Italy article info Article history: Received 13 March 2012 Received in revised form 13 June 2012 Accepted 16 June 2012 Available online 2 July 2012 Keywords: Syncope Decision rules Emergency Department ⁎ Corresponding author at: Department of Internal Medicine, Luigi Sacco Hospital, Milan, Via G. B. Grassi, 74, Milan (MI), Italy. Tel.: + 39 3409050440; fax: + 39 39042573. E-mail address: francescappe@gmail.com (F. Perego). 1 On behalf of the STePS investigators. There is a wide range of hospitalisation rates reported worldwide for individuals referring to the Emergency Department (ED) because of syncope [1–5] in spite of the availability of guidelines and risk scores aimed at standardising syncope management in the emergency setting [5–11]. The present study was prompted by the need to clarify the emergen- cy physician's criteria (Clinical Judgment) leading to hospital admission after syncope. In addition, predictors of hospitalisation obtained from clinical judgment were compared with those identified by two risk scores [8,9]. We enrolled 488 consecutive subjects referred to the ED of two general hospitals for syncope. Demographic and clinical features of the population are summarised in Table 1. Inclusion and exclusion criteria were previously defined [12]. Clinical judgment was defined as the decision-making process used by physicians in the ED to determine hospital admission or discharge [4]. Predictors of hospitalisation obtained from clinical judgment were compared with those identified by two risk scores, i.e. the Osservatorio Epidemiologico Sincope Lazio (OESIL) [8] and the San Francisco Syncope Rule (SFSR) [9]. All patients who were seen in the ED for syncope under- went both the OESIL risk score [8] and the SFSR [9] to stratify their risk. We assumed that low-risk patients would have been discharged where- as high-risk subjects would have been admitted. Patients characterised by a score ≥2 according to the OESIL risk score [8] were assumed to be at high risk. The presence of one or more risk factors of the SFSR [9] characterised high-risk patients (Table 2). ECG was defined as abnormal in the presence of any of the follow- ing conditions: (i) atrial fibrillation or tachycardia; (ii) sinus pause ≥2 s; (iii) sinus bradycardia with heart rate ranging between 35 and 45 beats per minute; (iv) conduction disorders (i.e. bundle branch block, second-degree Mobitz I atrioventricular block); (v) ECG signs of previous myocardial infarction or ventricular hypertrophy; and (vi) multiple premature ventricular beats. Selection of factors associated with hospital admission included patient's clinical history, presenting symptoms, physical examination and laboratory tests. Univariate and multivariate backward stepwise logistic regression identified the variables significantly associated with hospital admission according to clinical judgment. The adjusted odds ratio (OR) and 95% confidence interval (CI) were calculated. K statistic was used to evaluate the concordance between the OESIL risk score, SFSR and clinical judgment [14]. Results were expressed as the mean ± standard deviation (SD). A p value b 0.05 was consid- ered significant. The study was approved by the Ethical Committee of the L. Sacco Hospital. Clinical judgment resulted in the admission of 165 patients (34%). Predictors of hospitalisation were: age > 65 years (OR 3.19; CI 2.0–5.1; p b 0.001), haematocrit b 30% (OR 4.56; CI 1.5–14.3; p = 0.009), abnor- mal electrocardiogram (OR 3.30; CI 2.1–5.2; p b 0.001), absence of pro- dromes (OR 2.83; CI 1.8–4.6; p b 0.001), and trauma associated with syncope (OR 2.28; CI 1.4–3.7; p = 0.001) (Table 2). Agreement values between clinical judgment and OESIL (K statistic = 0.39) and between clinical judgment and SFSR (K statistic = 0.35) were fair. Concordance Table 1 Demographic and clinical features of the population. Population (n) 488 Age ± SD, y 59±22 18–44 y 149 (31) 45–65 y 108 (22) > 65 y 231 (47) Women 270 (55) Past medical history Hypertension 183 (38) Structural heart disease 125 (26) Heart failure 22 (5) Ventricular arrhythmias 8 (2) Cerebrovascular disease 66 (14) Neurological disease 51 (10) Diabetes mellitus 46 (9) COPD 33 (7) Cancer 40 (8) Index syncope history Supine/Sitting 116 (24) Upright posture 364 (74) During exercise 8 (2) First episode 215 (44) Trauma 115 (24) Abnormal ECG at presentation 167 (34) Absence of prodromal symptoms 127 (26) Final diagnosis Neurally mediated 109 (22) Cardiac 16 (3) Orthostatic hypotension 14 (3) Unexplained 295 (61) Other 54 (11) Recurrence within 1 y 64 (13) Values are expressed as mean ± SD or n (%). y indicates years, COPD Chronic Obstruc- tive Pulmonary Disease. 182 Letters to the Editor