A randomised prospective comparison of three protocols for head-up tilt testing and carotid sinus massage.

INTRODUCTION: Head-up tilt testing is an important tool in the diagnosis of syncope. Several different protocols are in use. This study aimed to compare three different protocols in an unselected population of patients with recurrent unexplained syncope and to assess long-term outcome using conventional tilt-directed management or implantable loop recorder (Reveal Plus)-directed management, allowing evaluation of the sensitivity and specificity of the technique.

METHODS: Patients with recurrent unexplained syncope were randomized to one of three tilt protocols: Drug-free--70 degree tilt, 45 min, CSM at 5 and 45 min. GTN--70 degree tilt, 35 min, CSM at 5 min, 400 microg of glyceryl trinitrate spray administered sublingually at 20 min. Adenosine--70 degree tilt, 5 min, CSM when blood pressure is stable in upright position, adenosine bonus at 150 microg/kg after CSM. Tilts were terminated at the onset of syncope, when systolic BP reached 60 mm Hg, or in the presence of prolonged hypotension (> 3 min systolic BP < 80 mm Hg). Appropriate therapies were commenced according to the result of the tilt test. All patients without a definite indication for immediate cardiac pacing (asystolic tilt) were randomized to conventional management or ILR implantation. Recurrent syncopal events were compared to tilt outcome, allowing estimation of sensitivity and specificity.

RESULTS: Of 214 patients, aged 68+/-18 years, 55% were female, with a median of three previous syncopes. 13 patients received pacemakers due to asystolic syncope during tilt testing. The proportion of VASIS classification diagnoses was similar with each protocol; however more positive diagnoses resulted from the GTN protocol (p=0.0013). 47% of patients achieved a diagnosis with tilt testing. We were able to correlate a subsequent spontaneous syncope to tilt result in 36 patients (18%). Heart rate during a spontaneous event was similar to that obtained during tilt testing (+/- 10%) in 55% of cases. Sensitivities for combined protocols, adenosine, GTN, and drug-free protocols were 50%, 50%, 100%, and 21%, respectively. Specificities were 85%, 100%, 75%, and 71%, respectively.

CONCLUSIONS: A high diagnosis rate for unexplained syncope can be achieved with tilt testing. The GTN protocol resulted in significantly more diagnoses than the other compared protocols with good sensitivity and adequate specificity. Sensitivity of the drug-free tilt test was lower than drug-augmented tilt testing.