BACKGROUND: A systematic evaluation focused on sensitivity and specificity of head-up tilt testing (HUT) for diagnosing vasovagal syncope has not been previously performed. We conducted a meta-analysis of studies comparing HUT outcome between patients with syncope of unknown origin and control subjects without previous syncope. METHODS: We searched Pubmed and Embase databases for all English-only articles concerning case-control studies estimating the diagnostic yield of HUT, and selected 55 articles, published before March 2012, including 4361 patients and 1791 controls. The influence of age, test duration, tilt angle, and nitroglycerine or isoproterenol stimulation on tilt testing outcome was analyzed. RESULTS: Head-up tilt testing demonstrated to have a good overall ability to discriminate between symptomatic patients and asymptomatic controls with an area under the summary receiver-operating characteristics curve of 0.84 and an adjusted diagnostic odds ratio of 12.15 (p<0.001). A significant inverse relationship between sensitivity and specificity of tilt testing for each study was observed (p<0.001). At multivariate analysis, advancing age and a 60° tilt angle showed a significant effect in reducing sensitivity and increasing specificity of the test. Nitroglycerine significantly raised tilt testing sensitivity by maintaining a similar specificity in comparison to isoproterenol. CONCLUSIONS: The results from this meta-analysis show the high overall performance of HUT for diagnosing vasovagal syncope. Our findings provide useful information for evaluating clinical and instrumental parameters together with pharmacological stressors influencing HUT accuracy. This could allow the drawing of tilt testing protocols tailored on the diagnostic needs of each patient with unexplained syncope.
BACKGROUND: A systematic evaluation focused on sensitivity and specificity of head-up tilt testing (HUT) for diagnosing vasovagal syncope has not been previously performed. We conducted a meta-analysis of studies comparing HUT outcome between patients with syncope of unknown origin and control subjects without previous syncope. METHODS: We searched Pubmed and Embase databases for all English-only articles concerning case-control studies estimating the diagnostic yield of HUT, and selected 55 articles, published before March 2012, including 4361 patients and 1791 controls. The influence of age, test duration, tilt angle, and nitroglycerine or isoproterenol stimulation on tilt testing outcome was analyzed. RESULTS: Head-up tilt testing demonstrated to have a good overall ability to discriminate between symptomatic patients and asymptomatic controls with an area under the summary receiver-operating characteristics curve of 0.84 and an adjusted diagnostic odds ratio of 12.15 (p<0.001). A significant inverse relationship between sensitivity and specificity of tilt testing for each study was observed (p<0.001). At multivariate analysis, advancing age and a 60° tilt angle showed a significant effect in reducing sensitivity and increasing specificity of the test. Nitroglycerine significantly raised tilt testing sensitivity by maintaining a similar specificity in comparison to isoproterenol. CONCLUSIONS: The results from this meta-analysis show the high overall performance of HUT for diagnosing vasovagal syncope. Our findings provide useful information for evaluating clinical and instrumental parameters together with pharmacological stressors influencing HUT accuracy. This could allow the drawing of tilt testing protocols tailored on the diagnostic needs of each patient with unexplained syncope.
Authors: Roland D Thijs; Michele Brignole; Cristian Falup-Pecurariu; Alessandra Fanciulli; Roy Freeman; Pietro Guaraldi; Jens Jordan; Mario Habek; Max Hilz; Anne Pavy-Le Traon; Iva Stankovic; Walter Struhal; Richard Sutton; Gregor Wenning; J Gert Van Dijk Journal: Clin Auton Res Date: 2021-03-19 Impact factor: 4.435
Authors: Richard Sutton; Artur Fedorowski; Brian Olshansky; J Gert van Dijk; Haruhiko Abe; Michele Brignole; Frederik de Lange; Rose Anne Kenny; Phang Boon Lim; Angel Moya; Stuart D Rosen; Vincenzo Russo; Julian M Stewart; Roland D Thijs; David G Benditt Journal: Eur Heart J Date: 2021-05-01 Impact factor: 35.855