About a month ago I did a little research from a slightly different angle. Since ENSO conditions are only available from 1950, I decided to check El Nino Region 3.4 anomalies against all of the very active seasons since 1950. I defined an active season as 14 named storms or more (there have been 10 of these) and ENSO neutral conditions as anything from +0.5C to -0.5C. I used the average sea surface temperature anomaly for the three month period of May, June and July (figuring that a three month lag was reasonable between Pacific ENSO conditions and Atlantic activity). The results coincide nicely with the findings of Ostro & Lyons.

1953...14 storms...+0.4 anomaly
1969...17 storms...+0.4 anomaly
1990...14 storms...+0.3 anomaly
1995...19 storms...+0.1 anomaly
1998...14 storms...+0.1 anomaly
2000...14 storms...-0.6 anomaly (weakening La Nina)
2001...15 storms...+0.1 anomaly
2003...15 storms.....0.0 anomaly
2004...15 storms...+0.4 anomaly
2005...28 storms...+0.3 anomaly

Conclusion: Although the dataset is limited, it appears that very active seasons generally occur under initially 'neutral' ENSO conditions. A strong La Nina doesn't really enhance activity, but a strong El Nino (1992, 1997, 2006) will limit activity.
Cheers,
ED