This is a cross-sectional study aiming to understand the early characteristics and background of bone health impairment in clinically well children with Fontan circulation.

We enrolled 10 clinically well children with Fontan palliation (operated >5 years before study entrance, Tanner stage ≤3, age 12.1 ± 1.77 years, 7 males) and 11 healthy controls (age 12.0 ± 1.45 years, 9 males) at two children’s hospitals. All patients underwent peripheral quantitative CT. For the Fontan group, we obtained clinical characteristics, NYHA class, cardiac index by MRI, dual x-ray absorptiometry, and biochemical studies. Linear regression was used to compare radius and tibia peripheral quantitative CT measures between Fontan patients and controls.

All Fontan patients were clinically well (NYHA class 1 or 2, cardiac index 4.85 ± 1.51 L/min/m2) and without significant comorbidities. Adjusted trabecular bone mineral density, cortical thickness, and bone strength index at the radius were significantly decreased in Fontan patients compared to controls with mean differences −30.13 mg/cm3 (p = 0.041), −0.31 mm (p = 0.043), and −6.65 mg2/mm4 (p = 0.036), respectively. No differences were found for tibial measures. In Fontan patients, the mean height-adjusted lumbar bone mineral density and total body less head z scores were −0.46 ± 1.1 and −0.63 ± 1.1, respectively, which are below the average, but within normal range for age and sex.

In a clinically well Fontan cohort, we found significant bone deficits by peripheral quantitative CT in the radius but not the tibia, suggesting non-weight-bearing bones may be more vulnerable to the unique haemodynamics of the Fontan circulation.

Fontan survivors have depressed cardiac index that worsens over time. Serum biomarker measurement is minimally invasive, rapid, widely available, and may be useful for serial monitoring. The purpose of this study was to identify biomarkers that correlate with lower cardiac index in Fontan patients.

This study was a multi-centre case series assessing the correlations between biomarkers and cardiac magnetic resonance-derived cardiac index in Fontan patients ⩾6 years of age with biochemical and haematopoietic biomarkers obtained ±12 months from cardiac magnetic resonance. Medical history and biomarker values were obtained by chart review. Spearman’s Rank correlation assessed associations between biomarker z-scores and cardiac index. Biomarkers with significant correlations had receiver operating characteristic curves and area under the curve estimated. In total, 97 cardiac magnetic resonances in 87 patients met inclusion criteria: median age at cardiac magnetic resonance was 15 (6–33) years. Significant correlations were found between cardiac index and total alkaline phosphatase (−0.26, p=0.04), estimated creatinine clearance (0.26, p=0.02), and mean corpuscular volume (−0.32, p<0.01). Area under the curve for the three individual biomarkers was 0.63–0.69. Area under the curve for the three-biomarker panel was 0.75. Comparison of cardiac index above and below the receiver operating characteristic curve-identified cut-off points revealed significant differences for each biomarker (p<0.01) and for the composite panel [median cardiac index for higher-risk group=2.17 L/minute/m2 versus lower-risk group=2.96 L/minute/m2, (p<0.01)].

Higher total alkaline phosphatase and mean corpuscular volume as well as lower estimated creatinine clearance identify Fontan patients with lower cardiac index. Using biomarkers to monitor haemodynamics and organ-specific effects warrants prospective investigation.