RESUMEN
BACKGROUND: The assessment of lung ventilation by radionuclide imaging has proved useful for the optimization of aerosol therapy in children with cystic fibrosis. Further analysis of lung perfusion may provide additional information. METHODS: Quantitative analysis of regional lung aerosol distribution (Tc phytates) and perfusion (Tc macroaggregates) homogeneity was performed in 18 children with cystic fibrosis, using the third and fourth spatial moments (skew and Kurtosis) of count distribution. Patients were chosen from a prospective study whose goal was to compare the efficacy of two nebulization methods of a radiolabelled aerosol: one session involved a nebulizer activated by patient inspiratory flow (control session), whereas the other involved a nebulizer powered by a pressure support device (PS session). RESULTS: Quantitative regional distribution of perfusion was similar to aerosol distribution, although skew and Kurtosis were lower, indicating better homogeneity. Perfusion skew was inversely correlated with pulmonary volumes and Shwachman score, even more significantly than ventilation skew. Using receiver operating characteristic curve analysis, a perfusion skew threshold of 0.67 was predictive of disease severity (FEV1 > or =60% or FEV1 <60%) with 86% sensitivity and 91% specificity. Furthermore, same skew threshold allowed the identification of patients who were 'PS responders' (greater amount of radioactivity deposited after the PS session) or 'PS non-responders' with 80% sensitivity and 77% specificity. CONCLUSION: Quantification of regional lung perfusion is easy to perform and heterogeneity of the distribution is closely correlated to disease severity. Moreover, perfusion skew can identify patients who are likely to benefit from pressure support (to optimize aerosol therapy) and may be helpful for orienting potential non-responders towards alternative therapies.