Clinical Application Of Lung Ventilation/Perfusion Imaging In Pulmonary Hypertension

Part 1 Comparison of Planar and Single Photon Emission Computed Tomography Ventilation/Perfusion Imaging for the Diagnosis of Chronic Thromboembolic Pulmonary HypertensionObjective:To compare the diagnostic efficacy of ventilation/perfusion planar(V/Q-Planar),V/Q single photon emission computed tomography(V/Q-SPECT)and SPECT perfusion scanning combined with low dose CT(Q-LDCT)in the diagnosis of chronic thromboembolic pulmonary hypertension(CTEPH).Methods:298 consecutive participants suspected of pulmonary hypertension(PH)from Fuwai hospital were prospectively enrolled in the study,and PH patients due to left heart disease and lung diseases were excluded.All participants underwent V/Q-Planar,V/Q-SPECT and chest LDCT,and were confirmed by pulmonary angiography(PA)and right cardiac catheterization(RHC)within one week.V/Q images were interpreted by the 2009 EANM guidelines criteria.Using PA as reference standard,diagnostic performance of each method was evaluated.Statistical significance of the difference in diagnostic performance of three imaging methods was assessed for individual participants and for individual lung segments by the McNemar test and generalized estimating equations analysis,respectively.Receiver operating characteristic(ROC)curve analysis was used to calculate area under curve(AUC),and their differences among three imaging methods were tested by DeLong method.Kappa test was used to analyze the consistency among the methods.ResultsOf the 208 participants included in this study,69 were confirmed as CTEPH,and 139 was excluded.V/Q-Planar,V/Q-SPECT,and Q-LDCT evaluations were positive for CTEPH in 76,79,and 79 participants,respectively,without of uncertain diagnosis.The sensitivity,specificity,accuracy of V/Q-Planar were 96%(66/69),93%(129/139),94%(195/208);for V/Q-SPECT were 97%(67/69),91%(127/139),93%(194/208),for Q-LDCT were 96%(66/69),91%(126/139),92%(192/208),respectively.AUC values for CTPA,V/Q-Planar and Q-LDCT were 0.94,0.94 and 0.93,respectively.There was no differences between techniques for any of the three calculations(P>0.05).V/Q-SPECT was more sensitive(V/Q-SPECT,79%,P=0.012 vs.V/Q-Planar,76%;V/Q-SPECT,79%,P<0.01 vs.Q-LDCT,75%),and V/Q-Planar was more specific(V/Q-Planar,54%P<0.01 vs.V/O-SPECT,46%;V/Q-Planar,54%,P=0.001 vs.Q-LDCT)as well as both V/Q-Planar and V/Q-SPECT were more accurate than Q-LDCT(V/Q-SPECT,67%,P=0.003 vs.Q-LDCT,64%;V/Q-Planar,68%,P=0.005 vs.Q-LDCT,64%)for segmental analysis.ConclusionsBoth V/Q-Planar and V/Q-SPECT were highly effective for diagnosing CTEPH,and Q-LDCT may be a reliable alternative method for patients who are unsuitable for ventilation imaging.Part 2 Comparison of Lung Ventilation/Perfusion Imaging and CT Pulmonary Angiography for the Diagnosis of Chronic Thromboembolic Pulmonary HypertensionObjective:To compare the diagnostic performance of ventilation/perfusion(V/Q)scanning and CT pulmonary angiography(CTPA)in chronic thromboembolic pulmonary hypertension(CTEPH).Methods:A total of 229 participants with suspected CTEPH from Fuwai hospital were evaluated via V/Q single photon emission computed tomography(V/Q-SPECT),V/Q planar scintigraphy(V/Q-Planar),CTPA,and PA in this prospective study.Participants underwent all four procedures within one week.V/Q images were interpreted by the 2009 EANM guidelines criteria.Using PA as reference standard,diagnostic performance of each method was evaluated for individual participants and lung segments.Statistical significance of the difference in sensitivity,specificity and accuracy of three imaging methods was assessed for individual participants and for individual lung segments by the McNemar test and generalized estimating equations analysis,respectively.Receiver operating characteristic(ROC)curve analysis was used to calculate area under curve(AUC),and their differences among three imaging methods were tested by DeLong method.Kappa test was used to analyze the consistency among the methods.ResultsOf the 150 participants included in this study,69 were confirmed as CTEPH,and 99 was excluded.In individual participants,V/Q-SPECT,V/Q-Planar,and CTPA evaluations were positive for CTEPH in 61,59,and 52 participants,and negative for CTEPH in 89,91,and 98 participants,respectively.All evaluations were diagnostic.When compared to PA,the sensitivity,specificity,accuracy of V/Q-Planar were 98%(50/51),91%(90/99),93%(140/150);for V/Q-SPECT were 98%(50/51),89%(88/99),92%(138/150),for CTPA were 94%(48/51),96%(95/99),95%(143/150),respectively.There was no statistical difference in the sensitivity,specificity,accuracy of the three techniques(P>0.05).However,both V/Q scanning techniques were more sensitive(V/Q-SPECT,85%,P<0.001 vs.CTPA,67%;V/Q-Planar,83%;P<0.001 vs.CTPA,67%),and less specific(V/Q-Planar,51%,P=0.03 vs.CTPA,60%;V/Q-SPECT,42%,P<0.01 vs.CTPA,60%)than CTPA for segmental analysis.AUC values for CTPA,V/Q-Planar and V/Q-SPECT were 0.95,0.95 and 0.94,respectively(all P>0.05),for individual analysis,and 0.64,0.67 and 0.64 by segment,respectively(V/Q-Planar vs.V/Q-SPECT,P=0.02;V/Q-Planar vs.CTPA,P=0.08;V/Q-SPECT vs.CTPA,P=0.94).ConclusionsBoth V/Q scanning(planar and SPECT)and CTPA showed good efficacy for diagnosing CTEPH at the patient level.V/Q scanning was more sensitive and less specific than CTPA for detecting vascular obstructions in segmental pulmonary arteries.Part 3 Clinical study of characteristic perfusion pattern by lung perfusion scintigraphy in idiopathic pulmonary hypertensionObjectiveTo characterized the lung perfusion pattern assessed by lung perfusion scintigraphy in idiopathic pulmonary arterial hypertension(IPAH)and evaluate the potential prognostic significance of the patchy pattern perfusion defect.MethodsA total of 318 IPAH patients from Fuwai hospital who performed lung perfusion scintigraphy were collected retrospectively in the current study.Lung perfusion images are categorized as normal or patchy pattern.All patients were regularly contacted every 6 months by telephone interview,and medical records were reviewed over the follow-up period.The primary endpoint was defined as all-cause mortality.Differences between the two groups were analyzed by the unpaired Student t test for continuous variables and the Chi-square test for categorical variables.The univariate and multivariable Cox proportional hazards model was used to test independent predictors of endpoint events,with results presented as hazard ratio(HR)with 95%confidence interval(C.I).Survival curves of endpoint were estimated by the Kaplan-Meier method and compared by the log-rank test.ResultsOf 318 IPAH patients,184 patients(57.9%)showed patchy perfusion defects and other 134 patients(42.1%)had normal lung perfusion.In comparison to patients with normal lung perfusion,patients with patchy perfusion defects experienced significantly higher mean pulmonary arterial pressure(58.0±15.4 mmHg vs 54.1±16.2mmHg,P=0.027)and total pulmonary resistance(1192.6±533.7dyn·s·cm-5 vs 1067.2±549.3dyn·s·cm-5,P=0.042).Other hemodynamics parameters between two groups were not statistically different.During a median follow-up period of 884.0 days,53 patients reached the primary end-point of all-cause mortality.On univariate Cox analysis,the patchy pattern of perfusion defect was significantly associated with the all-cause mortality(Hazard ratio[HR]2.47,95%confidence interval[C.I]1.32-4.63,P=0.005).Patients with patchy perfusion defects had a worse outcome(log-rank=8.605,P=0.003).On multivariate analysis,the patchy pattern remained as a significant independent predictor of the endpoint(HR 2.30,95%C.I 1.22-4.31,P=0.010).ConclusionsIPAH patients presented with heterogeneity in lung perfusion and the patchy pattern of lung perfusion defect commonly existed.Patients with patchy pattern identifi ed by lung perfusion scintigraphy were associated with more severe disease and worse outcome.