C. Attoyan et al., PREDICTIVE FACTORS OF VENTRICULAR-FIBRILL ATION IN THE WOLFF-PARKINSON-WHITE SYNDROME, Archives des maladies du coeur et des vaisseaux, 87(7), 1994, pp. 889-897
The incidence of sudden death in the Wolff-Parkinson-White (WPW) syndr
ome is not well documented and probably underestimated. This retrospec
tive study concerned 28 consecutive patients presenting with ventricul
ar fibrillation either spontaneously (20) or during electrophysiologic
al investigation (8) but whose characteristics allowed them to be assi
milated into a single group. Their clinical and electrophysiological c
haracteristics were compared with those of 60 consecutive patients wit
h the WPW syndrome who had documented atrial fibrillation (and even re
ciprocating tachycardia) but never ventricular fibrillation. There wer
e no significant differences between the two groups with respect to th
e following clinical parameters: sex, duration of symptoms, the type o
f tachycardia previously recorded, history of syncope and presence of
underlying cardiac disease.With respect to the electrophysiological da
ta, there were no differences in the point of anterograde block, the e
ffective anterograde refractory period of the accessory pathway, the e
ffective and functional refractory periods of the right atrium and atr
ial vulnerability. On the other hand, a significant difference was obs
erved in the age of patients with ventricular fibrillation (29 +/- 13
years vs 36 +/- 12 years; p < 0.02), the prevalence of multiple access
ory pathways (25 % vs 7 %; p < 0.04) with a dominant localisation in t
he postero-septal region (75 % vs 47 %, p < 0.026), preexcitation duri
ng exercise stress testing and under antiarrhythmic therapy (95 % vs 6
8 %, p < 0.037). The most discriminating parameter was the shorter RR
interval during atrial fibrillation (172 +/- 23 ms vs 230 +/- 50 ms, p
< 0.008). Multivariate analysis only showed one independant predictiv
e factor: the minimum preexcited RR interval. Therefore, a postero-sep
tal bundle of Kent and the multiplicity of these pathways were only pr
edictive in univariate analysis because they are associated with the m
inimum RR intervals during preexcitation. The ROC curve (Receiver Oper
ating Characteristics) shows that a minimal RR interval during atrial
fibrillation less-than-or-equal-to 190 ms is the optimal indicator of
the risk of sudden death: sensitivity = 82 %, specificity = 98 %, posi
tive predictive value = 87 %, negative predictive value = 83 %.