Normal Values for the Childhood Signal- Averaged ECG ANDREW M. DAVIS, BRIAN W. McCRINDLE, ROBERT M. HAMILTON, PAMELA MOORE-COLEMAN, and ROBERT M. GOW From the Division of Cardiology and Department of Paediatrics, University of Toronto, Hospital for Sick Children. Toronto, Ontario, Canada DAVIS, A., ET AL.: Normal Values for the Childhood Signal-Averaged ECG. To obtain normative data for the childhood signal-averaged ECG (SAECG), we obtained SAECGs in 155 volunteers, aged 5-15 years, using Erank leads and a Eourier transform filter. Unfiltered QRS duration (QRSDU) and filtered QRS du- ration (QRSDF) were significantly longer in males, and the root mean square of the terminal 40 ms {RMS40) was significantly higher in females. There were no gender differences in the duration of high fre- quency low amplitude signals < 40 fiV (DHFLA). All SAEGG variables were significantly related to body surface area (BSA). Regression models were established for SAECG variables. Eor males, the predicted mean + 1.96 standard deviations (SD) for QRSDE (97.5th centile) ranged from 114 ms at BSA 0.70 m^ to 123 ms at 1.90 m^. For females, the predicted mean -h 1.96 SDfor QRSDF ranged from 110 ms at RSA 0.70 m^ to 119 ms at 1.90 m^. For males and females, the predicted mean + 1.96 SD for DHFLA ranged from 34 ms at 0.70 m^ to 38 ms at 1.90 m^. For males, the predicted mean - 1.96 SD (2.5th centile) for RMS40 (based on natural logarithm model) ranged from 30 /xV at 0.70 m^ to 15 fiV at 1.90 m^. For females, the predicted mean - 1.96 SD for RMS40 ranged from 42 ^V at 0.70 m^ to 20 fiVatl.9O nf. In children 5-15 years of age, both gender and BSA need to be taken into account when interpreting the SAEGG. (FACE 1996:19:793-801) signal-averaged electrocardiography, normal values, pediatric Introduction The high resolution signal-averaged electro- cardiogram (SAECG) is a noninvasive investiga- tion that allows demonstration of high frequency low amplitude (HFLA) electrocardiographic sig- nals that correlate with delayed and fragmented activation of pathological myocardium. ^"^ In some situations, such late potentials may represent a potential substrate for ventricular arrhythmias^"^. The technique has been widely applied in adults, especially in patients who have suffered myocar- dial infarction,^• ^"' ^ but also in other patholo- gies.^°~^^ Different filter and lead systems affect Dr. Davis was supported by a Grant from the Hospital for Sick Children Research Institute. Address for reprints: Robert M. Hamilton. M.D.. Division of Cardiology, Hospital for Sick Children. 555 University Av- enue. Toronto, M5G 1X8. Canada. Fax: (416) 813-7547. Received October 31, 1994; revision January 19, 1995; ac- cepted March 29, 1995. SAECG variables and hence it is required to estab- lish normal limits specific to device.^'^• ^• ' ' The fact that adult criteria cannot be applied to children and teenagers has been recognized, but only very limited pediatric data is available.^"^-^^ Several publications have addressed the effects of various pediatric cardiac pathologies on SAECG variables but with little normative data.^*'"^'* This study was performed to accumulate normative data for chil- dren aged 5-15 years and to facilitate the clinical application of the SAEGG to the many and varied pathologies seen by the pediatric cardiologist and electrophysiologist. Material and Methods Study Population Healthy volunteers, aged 5-15 years, from co- operating schools, siblings of patients, and rela- tives of hospital staff were recruited as subjects. The study was approved by the Research Ethics Board at The Hospital for Sick Children, and in- PACE, Vol. 19 May 1996 793