(116) Associations between lung T1-MRI, chest CT, and multiple breath washout in young children with mild CF lung disease

Background: Structural and physiologic evidence of lung disease is present in a large proportion of children with CF even in the setting of a normal FEV1. Chest CT can detect changes in lung structure in children with CF, but it exposes them to ionizing radiation. Multiple breath washout (MBW) can detect ventilation inhomogeneity in children with mild CF lung disease. However, it can be time consuming to perform and requires a high degree of operator expertise. Lung T1-MRI can assess pulmonary perfusion and can be quickly performed on clinical MRI scanners using conventional acquisitions and no inhaled or injectable MRI contrast agent. We hypothesized that lung T1-MRI is a sensitive measure of early and mild lung disease in children with CF. To test this hypothesis, we performed T1-MRI, chest CT, and MBW in a cohort of children with CF ages 6-11 years.

Methods: Children with CF ages 6-11 years were recruited from 3 CF Centers (Indiana University, University of Michigan, Case Western Reserve University). MBW was performed according to published guidelines and the lung clearance index (LCI) was measured. Chest CTs were scored using the Brody score. Lung T1-MRI images were acquired using a Modified Look-Locker Imaging acquisition with a short echo time (~10 seconds/imaging slice). Lung T1 maps were threshholded to calculate percent normal lung perfusion (%NLP) for each subject. Descriptive statistics were used to analyze the cohort and Pearson or Spearman correlations were used to study the association between T1-MRI %NLP, LCI, and Brody score, and P≤0.05 was considered significant.

Results: A total of 45 children had acceptable T1-MRI, MBW, and CT data. Their clinical and test data are summarized in the Table. T1-MRI %NLP correlated significantly with CT scores (R=-0.41, P=0.026) but not LCI (R=-0.06, P=0.71). LCI and CT scores were significantly correlated (R=0.43, P=0.017). Analysis of correlations with Brody subscores showed that LCI correlated significantly with air trapping, peribronchial thickening, and central mucus plugging, while %NLP correlated significantly with dilated airways, mucus plugging, and dense opacities.

Table. Selected Clinical Features and Test Results of the Study Cohort. All values are reported as mean ± SD unless otherwise noted.

N	45
Male/Female (%)	51.1/48.9
Race (%) White Black	97.7 2.3
Age at testing	8.9 ± 1.7
FEV1 % predicted	101.6 ± 15.3
BMI Z score	0.4 ± 0.9
LCI	8.6 ± 1.3
CT Brody Score	23.6 ± 26.9
MRI %NLP	87.3 ± 8.9

Conclusions: Both LCI and %NLP are associated with chest CT scores in children with CF. Specifically, %NLP correlated with structural and parenchymal findings, whereas LCI correlated with measures of air trapping and airway disease. Our results are consistent with other studies that suggest lung T1-MRI imaging provides complementary information to that from MBW. The associations observed in our study suggest that T1-MRI in combination with MBW, could serve as a safe, fast method to assess CF lung disease that does not expose children to ionizing radiation and can be performed on clinically available equipment.

Acknowledgements:

Funding: FLASK20A0

References: