Background—Long-term outcomes for childhood left ventricular non-compaction (LVNC) are uncertain. We examined late outcomes for children with LVNC enrolled in a national population-based study.Methods—The National Australian Childhood Cardiomyopathy Study includes all children in Australia with primary cardiomyopathy diagnosed <10 years of age between 1987 and 1996. Outcomes for LVNC subjects with a dilated phenotype (LVNC-D) were compared to those with dilated cardiomyopathy (DCM). Propensity-score analysis was used for risk factor adjustment.Results—There were 29 subjects with LVNC (9.2% of all cardiomyopathy subjects) with a mean annual incidence of newly diagnosed cases of 0.11 per 100,000 at-risk persons. Congestive heart failure was the initial symptom in 24 (83%) of 29 subjects, and 27 (93%) had a dilated phenotype (LVNC-D). The median age at diagnosis was 0.3 (interquartile interval 0.08 - 1.3) years of age. The median (interquartile interval) duration of follow-up was 6.8 (0.7-14.1) years for all subjects and 24.7 (23.3 - 27.7) years for surviving subjects. Freedom from death or transplantation was 48% (95% CI 30 - 65%) at 10 years after diagnosis and 45% (95% CI 27-63%) at 15 years. By competing risk analysis, 21% of LVNC subjects were alive with normal LV systolic function and 31% were alive with abnormal function at 15 years. Propensity-score matching between LVNC-D and DCM subjects suggested a lower freedom from death/transplantation at 15 years after diagnosis in the LVNC-D subjects (LVNC-D: 46% (95% CI 26-66%) vs. DCM: 70% (95% CI 42-97%), p=0.08). Using propensity-score inverse probability of treatment weighted Cox regression, we found evidence that LVNC-D was associated with a greater risk of death or transplantation (HR 2.3, 95% CI 1.4-3.8, p=0.0012).Conclusions—Symptomatic children with LVNC usually present in early infancy with a predominant dilated phenotype. Long-term outcomes are worse than for matched children with DCM.

背景-儿童左心室非紧致症（LVNC）的长期结果尚不确定。我们检查了参加全国人口研究的LVNC儿童的晚期结局。方法—澳大利亚国家儿童心脏病研究包括1987年至1996年在澳大利亚诊断为＜10岁的所有原发性心肌病的儿童。比较了具有扩张表型（LVNC-D）的LVNC受试者与扩张型心肌病（DCM）的结果。 。倾向得分分析用于风险因素调整。结果-有29例LVNC受试者（占所有心肌病受试者的9.2％），新诊断病例的年平均发病率为每100,000名高危人群中0.11例。充血性心力衰竭是29位受试者中的24位（83％）的初始症状，其中27位（93％）具有扩张的表型（LVNC-D）。诊断时的中位年龄为0.3岁（四分位间距为0.08-1.3）。所有受试者的中位随访时间（四分位数间隔）为6.8（0.7-14.1）年，存活受试者为24.7（23.3-27.7）年。在诊断后10年，死亡或移植的自由度为48％（95％CI 30-65％），在15年时为45％（95％CI 27-63％）。通过竞争风险分析，在15岁时，有21％的LVNC受试者活着，其左室收缩功能正常，而有31％的存活者，其功能异常。LVNC-D和DCM受试者之间的倾向得分匹配表明，LVNC-D受试者诊断后15年的死亡/移植自由较低（LVNC-D：46％（95％CI 26-66％），而DCM： 70％（95％CI 42-97％），p = 0.08）。使用治疗性加权Cox回归的倾向得分逆向概率，我们发现有证据表明LVNC-D与更大的死亡或移植风险相关（HR 2.3，95％CI 1.4-3.8，p = 0.0012）。结论—有症状的LVNC儿童通常在婴儿早期出现，主要表现为扩张型。长期结局要比匹配的DCM儿童差。