Study Covariates
We included baseline demographic characteristics, past medical history,
drug history, family history, and syncopal history using the
abovementioned databases. Other than demographic characteristics and
past medical history, we carefully selected the study covariates that
may interfere with serum levels of vitamin B12 or folate to minimize
their confounding effects. For this purpose, we specifically recorded
the history of hypo-/hyperthyroidism, and taking proton pump inhibitors,
H2 receptor blockers, metformin, and oral contraceptive
pills, which may interfere with levels of vitamin B12 or folate. We also
investigated the family history of syncope, seizure, or sudden death.
The evidence of possibly lower serum levels of thyroid-stimulating
hormone (TSH) in pediatric patients with VVS and positive HUTT compared
to negative HUTT23 encouraged us to measure serum TSH
as a possible confounder in this study. We did so using the Abbott
Diagnostic chemiluminescence kit and the abovementioned protocol. In the
case group, we inquired about the last year and the lifetime number of
syncopal episodes. According to the number of lifetime syncopal
episodes, we categorized patients into frequent VVS with ≥3 episodes
versus infrequent VVS with <3 episodes in the lifetime. All
patients in the case group underwent HUTT according to the Italian
protocol.28 According to the modified VASIS
classification,29 we categorized the hemodynamic
response in a positive HUTT into 1) Type 1, mixed; 2) Type 2A,
cardioinhibition without asystole; 3) Type 2B, cardioinhibition with
asystole; and 4) Type 3, vasodepressor.