Does the absence of comorbidities really identify low-risk syncope patients?