Radionuclide evaluation of lung perfusion after transcatheter closure of patent ductus arteriosus: analysis of frequently used two different type of coils

Abstract

Impaired left lung perfusion (LLP) has been described after transcatheter closure of the patent ductus arteriosus (PDA). In this study, we aimed to evaluate lung perfusion scans (LPSs) following occlusion of PDA with two frequently used coils: Cook detachable coil and Gianturco coil. A prospective study of 89 patients who underwent PDA occlusion using the Cook coils in 49 and the Gianturco coils in 40, was conducted. LPSs were performed after a median period of 18 months. The relationships between the LPSs and continuous variables were assessed by Pearson correlation analysis and the cut-off value of the best correlated parameters to predict abnormal LPSs obtained by the Receiver Operating Characteristic (ROC) analysis. Decreased left lung perfusion (LLP) was found in 13 patients including 10 after using Cook coils and 3 after using Gianturco coils (p=0.077). LLP values were significantly correlated with the loops deployed at the pulmonary side, coil/ductal diameter ratio and number of coils deployed (p=0.002, p=0.006 and p=0.009, respectively). Number of loops deployed at the pulmonary side >1.4 (specificity 77%, sensitivity 85%, area under the ROC curve 0.804, 95%CI 0.661-0.947, p<0.001) and first coil/ductal diameter ratio >1.85 (specificity 70%, sensitivity 77%, area under the ROC curve 0.747, 95%CI 0.423-0.790, p<0.005) were the best discriminative cut-off values of decreased LLP. Impaired LLP may appear following transcatheter closure of PDA with either Cook detachable coils or Gianturco coils. This situation may be complicated with loops deployed at the pulmonary side, the use of relatively large size of coil with respect to the ductal diameter and the use of multiple coils.