Demographic Data Collection

We collected all patients with diagnoses of left ventricular noncompaction from our reference database and reviewed all patients using the Siemens KinetDx workstation (Siemens Medical Solutions USA, Inc, Mountain View, CA). We reviewed all data from January 1, 2000, to June 30, 2008. Demographic data included age, body surface area, sex, and age at presentation. The outcomes data collected included death, heart transplantation, and alive without heart transplantation. If a patient was alive without heart transplantation, he or she was assigned to group 1. If a patient either died or required heart transplantation, he or she was assigned to group 2. This study was approved through our institutional review board (protocol 14881), and all data were made anonymous in accordance with the Health Insurance Portability and Accountability Act.

Echocardiographic Data Collection

Ultrasound equipment included the Siemens Acuson C512, revision 12.0 (Siemens Medical Solutions USA, Inc), the Philips iE33 (Philips Medical Systems, Bothell, WA), and the Hewlett-Packard 5500 (Hewlett-Packard, Andover, MA). We carefully reviewed all patients in our database who had diagnoses of left ventricular noncompaction. Strict criteria were used to ensure the appropriate diagnosis in that all patients had the following left ventricular findings as described by Oechslin et al and Jenni et al :

• •
  

  multiple trabeculations and recesses ( Figures 1 and 2 ),

  
  

  
  

  
  

  

  
  

  
  

  Midsystolic 4-chamber view also used to establish deep recesses and trabeculations as a part of the diagnosis of left ventricular noncompaction. This example illustrates the deep recesses and trabeculations within the left ventricular apex, incidentally discovered prior to chemotherapy for osteogenic sarcoma (patient 28 in Table 1 ).

  
  
  
  
• •
  

  distinct compacted and noncompacted layers,

  
  
• •
  

  low–Nyquist limit color mapping delineating continuity of intertrabecular recesses with ventricular cavity, and

  
  
• •
  

  a noncompaction/compaction ratio ≥2.0 during systole.

The initial echocardiogram at presentation was used to make all measurements. The severity of noncompaction was determined using a noncompaction/compaction ratio ≥ 2.0, which distinguishes this lesion from other diseases with hypertrophy. In addition, we evaluated the severity of noncompaction by calculating the average noncompaction/compaction ratio for each segment. The number of segments with noncompaction were counted and analyzed in the parasternal short-axis views of the left ventricle at the level of the mitral valve, papillary muscle, and apex using the 16-segment model described by the American Society of Echocardiography and the American Heart Association ( Figure 1 ). Other echocardiographic data that were obtained included the shortening and ejection fractions. The shortening fraction was measured using the standard M-mode measurement of the left ventricle just below the mitral valve, and the ejection fraction was calculated offline using the modified Simpson’s rule, defined by the border of the left ventricle using the apical 2-chamber and 4-chamber views.

Statistical Analysis

A Kaplan-Meier curve was used to depict the percentage of freedom from death or transplantation in all of the patients included in the study. Groups 1 and 2 were directly compared to detect statistically significant differences. Fisher’s exact test was used to compare the proportions of patients in groups 1 and 2 who had congenital heart disease and associated cardiomyopathy. Data with normal distributions, such as follow-up time period, shortening fraction, ejection fraction, and noncompaction/compaction ratio, were compared using Student’s 2-tailed t test. The number of segments with noncompaction/compaction ratios > 2.0 and median age were compared using Mann-Whitney U tests, because they were nonparametric data. Among patients in each group, an analysis of variance was conducted to determine if there were differences in the average noncompaction/compaction ratio between the basal, midpapillary, and apical regions. The post hoc Bonferroni test was used to establish which region was more involved than the others. Functional data such as ejection and shortening fractions were also compared with the number of segments involved using Spearman’s correlation. Statistical analysis was conducted using Microsoft Excel (Microsoft Corporation, Redmond, WA) and Stata release 7.0 (StataCorp LP, College Station, TX).