Correlation of Surgical and Computed Tomography Pulmonary Angiographic Level of Disease in Chronic Thromboembolic Pulmonary Hypertension

Abstract

Purpose Although the surgical level of disease in chronic thromboembolic pulmonary hypertension (CTEPH) is routinely described, no study has systematically correlated the preoperative computed tomography pulmonary angiogram (CTPA) with surgical level. Methods We retrospectively reviewed 113 patients undergoing pulmonary endarterectomy for CTEPH confirmed by CTPA between 05/2015 and 12/2017. Thirteen patients were excluded for technically inadequate CTPA. CTPA level of disease was classified by two independent readers blinded to the surgical level as level 1 (main pulmonary artery [PA]), level 2 (lobar PA), level 3 (segmental PA), and level 4 (subsegmental PA). Level 2 was subdivided into 2a (lobar PA) and 2b (lower lobe basal trunk). CTPA level was then compared with surgical level. Results Average patient age was 58 +/- 15 years and 51% were female. The median time between CTPA and PEA was 18 days (range 1 to 615 days). CTPA demonstrated level 1 disease in 20%, 2a in 43%, 2b in 11%, 3 in 23%, and 4 in 3%. Surgical staging was within one level of CTPA staging in 99% of cases. There was a higher proportion of level 2 disease at CTPA compared to surgical staging (54% vs. 31%, p = 0.001). There was a lower proportion of level 4 disease at CTPA compared to surgical staging (3% vs. 16%, p = 0.002). When grouped as proximal (levels 1 and 2) versus distal (levels 3 and 4), the proportion of distal disease was higher at surgery than CTPA (44% vs. 23%, p = 0.008). CTPA level 2b had a higher proportion of surgical distal disease than CTPA level 2a (73% vs. 28%, p = 0.01). The proportion of distal disease was no longer significantly different between surgery and CTPA if level 2b is considered distal (44% vs. 37%, p = 0.3) and concordance in classifying level as proximal or distal on CTPA and at surgery increased from 76% to 84%. If surgical staging is considered gold standard then the sensitivity, specificity, and accuracy of CTPA in identifying proximal disease in a cohort of CTEPH patients is 89.2%, 70.4%, and 81%, respectively. Conclusion CTPA staging of CTEPH closely correlates to surgical level. A subdivision of CTPA level 2 disease into 2a and 2b better distinguished proximal from distal disease. A unified staging system would help to address differences and facilitate CT reporting. Although the surgical level of disease in chronic thromboembolic pulmonary hypertension (CTEPH) is routinely described, no study has systematically correlated the preoperative computed tomography pulmonary angiogram (CTPA) with surgical level. We retrospectively reviewed 113 patients undergoing pulmonary endarterectomy for CTEPH confirmed by CTPA between 05/2015 and 12/2017. Thirteen patients were excluded for technically inadequate CTPA. CTPA level of disease was classified by two independent readers blinded to the surgical level as level 1 (main pulmonary artery [PA]), level 2 (lobar PA), level 3 (segmental PA), and level 4 (subsegmental PA). Level 2 was subdivided into 2a (lobar PA) and 2b (lower lobe basal trunk). CTPA level was then compared with surgical level. Average patient age was 58 +/- 15 years and 51% were female. The median time between CTPA and PEA was 18 days (range 1 to 615 days). CTPA demonstrated level 1 disease in 20%, 2a in 43%, 2b in 11%, 3 in 23%, and 4 in 3%. Surgical staging was within one level of CTPA staging in 99% of cases. There was a higher proportion of level 2 disease at CTPA compared to surgical staging (54% vs. 31%, p = 0.001). There was a lower proportion of level 4 disease at CTPA compared to surgical staging (3% vs. 16%, p = 0.002). When grouped as proximal (levels 1 and 2) versus distal (levels 3 and 4), the proportion of distal disease was higher at surgery than CTPA (44% vs. 23%, p = 0.008). CTPA level 2b had a higher proportion of surgical distal disease than CTPA level 2a (73% vs. 28%, p = 0.01). The proportion of distal disease was no longer significantly different between surgery and CTPA if level 2b is considered distal (44% vs. 37%, p = 0.3) and concordance in classifying level as proximal or distal on CTPA and at surgery increased from 76% to 84%. If surgical staging is considered gold standard then the sensitivity, specificity, and accuracy of CTPA in identifying proximal disease in a cohort of CTEPH patients is 89.2%, 70.4%, and 81%, respectively. CTPA staging of CTEPH closely correlates to surgical level. A subdivision of CTPA level 2 disease into 2a and 2b better distinguished proximal from distal disease. A unified staging system would help to address differences and facilitate CT reporting.