Use of a learning network to improve variation in interstage weight gain after the Norwood operation.

BACKGROUND: Growth failure is common in infants with single ventricle. This study evaluated the use of a learning network, the National Pediatric Cardiology Quality Improvement Collaborative (NPC-QIC), to spread optimized nutritional practices and improve infant growth.
METHODS: A previously identified Nutritional Bundle was spread among NPC-QIC sites. PRIMARY OUTCOME: interstage weight-for-age z-score change (ΔWAZ) between discharge from stage 1 palliation (S1) and stage 2 surgical palliation (S2). Variation among sites in interstage ΔWAZ was evaluated before (Period 1) and after (Period 2) spread of Nutritional Bundle. We performed an analysis of NPC-QIC registry infants presenting for S2 at sites previously shown to have significant variation in interstage patient growth.
RESULTS: Four hundred seven infants from 15 sites underwent S2 between 2008 and 2013: 158 in Period 1 (December 2008-December 2010) and 249 in Period 2 (December 2010-April 2013). Median age at S2 was 4.9 months (2.6-12.8) with no difference between periods. There was significant variation in interstage ΔWAZ among sites in Period 1 (P = .01) but not in Period 2 (P = .39). More patients had an interstage ΔWAZ <0 in Period 1 (43%) than Period 2 (32%) (P = .03). In Period 1, the median interstage ΔWAZ was <0 in six sites while in Period 2 no site had median interstage ΔWAZ <0. Sites with the worst patient growth in Period 1 had marked improvement in Period 2 (P = .02, .06, and .06, respectively).
CONCLUSIONS: Spread of optimal nutritional practices led to decreased variation in interstage growth with most improvement observed at sites with the worst baseline growth outcomes.