Eur Radiol (2014) 24:2385–2393 DOI 10.1007/s00330-014-3289-4
CHEST
Portopulmonary hypertension: Improved detection using CT and echocardiography in combination Anand Devaraj & Robert Loveridge & Diana Bosanac & Konstantinos Stefanidis & William Bernal & Christopher Willars & Julia A. Wendon & Georg Auzinger & Sujal R. Desai
Received: 21 January 2014 / Revised: 27 May 2014 / Accepted: 24 June 2014 / Published online: 24 July 2014 # European Society of Radiology 2014
Abstract Objectives To establish the relationship between CT signs of pulmonary hypertension and mean pulmonary artery pressure (mPAP) in patients with liver disease, and to determine the additive value of CT in the detection of portopulmonary hypertension in combination with transthoracic echocardiography. Methods Forty-nine patients referred for liver transplantation were retrospectively reviewed. Measured CT signs included the main pulmonary artery/ascending aorta diameter ratio (PA/ AAmeas) and the mean left and right main PA diameter (RLPAmeas). Enlargement of the pulmonary artery compared to the ascending aorta was also assessed visually (PA/AAvis). CT measurements were correlated with right-sided heart catheter-derived mPAP. The ability of PA/AAvis combined with echocardiogram-derived right ventricular systolic pressure (RVSP) to detect portopulmonary hypertension was tested with ROC analysis. Results There were moderate correlations between mPAP and both PA/AAmeas and RLPAmeas (rs =0.41 and rs =0.42, respectively; p 240 dynes.s.cm -5 and pulmonary artery occlusion pressure (PAOP)≤15 mmHg [7, 19]; ii) patients with established or incipient portopulmonary hypertension. This group included not only those patients who met strict ERS criteria for portopulmonary hypertension, but also those that had elevated pulmonary artery pressure with suspected incipient vascular remodelling based on an elevated transpulmonary pressure gradient (TPG; [where TPG =mPAP – PAOP]). Patients Table 1 Demographic and clinical data in 49 patients with chronic liver disease and portal hypertension Patient Demographics Gender (M/F) Age (median; range) Cirrhotic/Non-Cirrhotic Aetiology of Liver Disease Alcoholic liver disease Hepatitis B/C Cryptogenic cirrhosis Primary sclerosing cholangitis Hemachromatosis Non-alcoholic steatohepatitis Primary biliary cirrhosis Secondary biliary cirrhosis Wilson’s disease Sarcoidosis Portal vein thrombosis Nodular regenerative hyperplasia MELD Score (median [range])
32 / 17 54 years; 23-72 48 / 1 18 9 8 3 2 2 2 1 1 1 1 1 18 (8-40)
Eur Radiol (2014) 24:2385–2393
2387
included in this group had an mPAP≥25 mmHg and TPG> 12 mmHg [5, 6, 20] (Table 2). This group included patients in whom either the PVR did not reach the defining threshold level because of a preponderance of a high cardiac output state, or strict ERS were not fulfilled because the PAOP was above the 15 mmHg threshold. Patients without ERS-defined portopulmonary hypertension or incipient portopulmonary hypertension included all patients with a TPG ≤12 mmHg. Right-sided Heart catheterization Right-sided heart catheterisation was performed using an Edwards Vigilance CCombo™ continuous cardiac output pulmonary artery flotation catheter (Edwards Lifesciences Corporation, CA, USA). The mean pulmonary artery pressure (mPAP) and pulmonary artery occlusion pressure (PAOP) were recorded. The transpulmonary gradient (TPG) was derived by subtracting PAOP from mPAP. Thermodilution cardiac output was calculated continuously with energy emitted by a thermal filament. Data on mPAP measurements were available in all patients. Cardiac output data and, by implication, PVR measures, were available in 48/49 (98 %) patients.
CT acquisition Contrast-enhanced thoracic CT examinations were available in 49 patients. Thirteen out of 49 (26.5 %) patients had been referred for CT pulmonary angiography (CTPA). Examinations were performed on one of two CT machines with the following imaging parameters: i) Siemens Sensation 16 (Siemens Healthcare, Forchheim, Germany); 120 kV, 100 mAs (with a maximum of 600 mAs with CareDose 4D dose reduction), pitch 1.15, 0.5 s tube rotation, 16×0.75 mm collimation; and ii) GE 64 VCT Lightspeed (GE Healthcare, Waukesha, WI, USA); 100 kV, Auto mA/Smart mA (maximum mA= 700 mA); pitch 1.375, 0.5 s tube rotation, collimation 64× 0.65 mm. The entire thorax was examined in a single breath hold, after injection of 100 ml of contrast medium was injected intravenously (Iohexol 300 mg/ml, GE Healthcare AS, Nycoveien 1-2, Nydalen, Norway) at 5 ml per second. Thoracic CTs were acquired at 30 s after contrast medium injection and bolus tracking was used for CTPAs. Images were reconstructed contiguously at either 1 mm, 1.25 mm, or 2 mm slice thickness, using soft and sharp algorithms.
Two-dimensional transthoracic echocardiography
CT image analysis – electronic measurements
Transthoracic echocardiography was performed according to international standards [21]. Right ventricular systolic pressure (RVSP) was calculated by measuring the peak tricuspid regurgitant (TR) velocity on continuous wave Doppler, using the modified Bernoulli equation: RVSP (mmHg)=4 x TR2 + estimated RAP; the RAP was derived from inferior vena cava diameter and collapsibility data [22].
To test the relationship between CT signs of pulmonary hypertension and right-sided heart catheterisation-derived mPAP, arterial dimensions on CT were measured by one observer (KS; 7 years of experience) blinded to right-sided heart catheterisation and transthoracic echocardiographic data in accordance with previous studies [17, 23]. Using electronic callipers, the following markers of pulmonary hypertension
Table 2 Comparison of physiologic data and CT measurements in 49 patients with and without portopulmonary hypertension Group
mPAP TPG PVR PA/AAmeas RLPAmeas† sABRmeas RV †
ERS-PoPH mPAP≥25 PVR>240* PAOP12
Patients without PoPH TPG≤12
n=17
n=32
44.2 (38-50) 32.8 (25-38) 357.8 (243-464) 1.09 (0.84-1.43) 27.4 (25.4-29.5) 1.84 (1.41-2.45) 42.7 (33.1-53.2)
42 (28-57) 24.6 (14-48) 224.4 (61-464) 1.04 (0.79-1.49) 26.0 (19.7-30.4) 1.49 (1.01-2.45) 43.9 (33.1-55.3)
23.2 (8-45) 6.9 (2-12) 54.9 (15-184) 0.91 (0.72-1.40) 22.4 (18.0-26.4) 1.23 (0.97-1.81) 40.8 (28.0-53.5)
p value*
CHEST
Portopulmonary hypertension: Improved detection using CT and echocardiography in combination Anand Devaraj & Robert Loveridge & Diana Bosanac & Konstantinos Stefanidis & William Bernal & Christopher Willars & Julia A. Wendon & Georg Auzinger & Sujal R. Desai
Received: 21 January 2014 / Revised: 27 May 2014 / Accepted: 24 June 2014 / Published online: 24 July 2014 # European Society of Radiology 2014
Abstract Objectives To establish the relationship between CT signs of pulmonary hypertension and mean pulmonary artery pressure (mPAP) in patients with liver disease, and to determine the additive value of CT in the detection of portopulmonary hypertension in combination with transthoracic echocardiography. Methods Forty-nine patients referred for liver transplantation were retrospectively reviewed. Measured CT signs included the main pulmonary artery/ascending aorta diameter ratio (PA/ AAmeas) and the mean left and right main PA diameter (RLPAmeas). Enlargement of the pulmonary artery compared to the ascending aorta was also assessed visually (PA/AAvis). CT measurements were correlated with right-sided heart catheter-derived mPAP. The ability of PA/AAvis combined with echocardiogram-derived right ventricular systolic pressure (RVSP) to detect portopulmonary hypertension was tested with ROC analysis. Results There were moderate correlations between mPAP and both PA/AAmeas and RLPAmeas (rs =0.41 and rs =0.42, respectively; p 240 dynes.s.cm -5 and pulmonary artery occlusion pressure (PAOP)≤15 mmHg [7, 19]; ii) patients with established or incipient portopulmonary hypertension. This group included not only those patients who met strict ERS criteria for portopulmonary hypertension, but also those that had elevated pulmonary artery pressure with suspected incipient vascular remodelling based on an elevated transpulmonary pressure gradient (TPG; [where TPG =mPAP – PAOP]). Patients Table 1 Demographic and clinical data in 49 patients with chronic liver disease and portal hypertension Patient Demographics Gender (M/F) Age (median; range) Cirrhotic/Non-Cirrhotic Aetiology of Liver Disease Alcoholic liver disease Hepatitis B/C Cryptogenic cirrhosis Primary sclerosing cholangitis Hemachromatosis Non-alcoholic steatohepatitis Primary biliary cirrhosis Secondary biliary cirrhosis Wilson’s disease Sarcoidosis Portal vein thrombosis Nodular regenerative hyperplasia MELD Score (median [range])
32 / 17 54 years; 23-72 48 / 1 18 9 8 3 2 2 2 1 1 1 1 1 18 (8-40)
Eur Radiol (2014) 24:2385–2393
2387
included in this group had an mPAP≥25 mmHg and TPG> 12 mmHg [5, 6, 20] (Table 2). This group included patients in whom either the PVR did not reach the defining threshold level because of a preponderance of a high cardiac output state, or strict ERS were not fulfilled because the PAOP was above the 15 mmHg threshold. Patients without ERS-defined portopulmonary hypertension or incipient portopulmonary hypertension included all patients with a TPG ≤12 mmHg. Right-sided Heart catheterization Right-sided heart catheterisation was performed using an Edwards Vigilance CCombo™ continuous cardiac output pulmonary artery flotation catheter (Edwards Lifesciences Corporation, CA, USA). The mean pulmonary artery pressure (mPAP) and pulmonary artery occlusion pressure (PAOP) were recorded. The transpulmonary gradient (TPG) was derived by subtracting PAOP from mPAP. Thermodilution cardiac output was calculated continuously with energy emitted by a thermal filament. Data on mPAP measurements were available in all patients. Cardiac output data and, by implication, PVR measures, were available in 48/49 (98 %) patients.
CT acquisition Contrast-enhanced thoracic CT examinations were available in 49 patients. Thirteen out of 49 (26.5 %) patients had been referred for CT pulmonary angiography (CTPA). Examinations were performed on one of two CT machines with the following imaging parameters: i) Siemens Sensation 16 (Siemens Healthcare, Forchheim, Germany); 120 kV, 100 mAs (with a maximum of 600 mAs with CareDose 4D dose reduction), pitch 1.15, 0.5 s tube rotation, 16×0.75 mm collimation; and ii) GE 64 VCT Lightspeed (GE Healthcare, Waukesha, WI, USA); 100 kV, Auto mA/Smart mA (maximum mA= 700 mA); pitch 1.375, 0.5 s tube rotation, collimation 64× 0.65 mm. The entire thorax was examined in a single breath hold, after injection of 100 ml of contrast medium was injected intravenously (Iohexol 300 mg/ml, GE Healthcare AS, Nycoveien 1-2, Nydalen, Norway) at 5 ml per second. Thoracic CTs were acquired at 30 s after contrast medium injection and bolus tracking was used for CTPAs. Images were reconstructed contiguously at either 1 mm, 1.25 mm, or 2 mm slice thickness, using soft and sharp algorithms.
Two-dimensional transthoracic echocardiography
CT image analysis – electronic measurements
Transthoracic echocardiography was performed according to international standards [21]. Right ventricular systolic pressure (RVSP) was calculated by measuring the peak tricuspid regurgitant (TR) velocity on continuous wave Doppler, using the modified Bernoulli equation: RVSP (mmHg)=4 x TR2 + estimated RAP; the RAP was derived from inferior vena cava diameter and collapsibility data [22].
To test the relationship between CT signs of pulmonary hypertension and right-sided heart catheterisation-derived mPAP, arterial dimensions on CT were measured by one observer (KS; 7 years of experience) blinded to right-sided heart catheterisation and transthoracic echocardiographic data in accordance with previous studies [17, 23]. Using electronic callipers, the following markers of pulmonary hypertension
Table 2 Comparison of physiologic data and CT measurements in 49 patients with and without portopulmonary hypertension Group
mPAP TPG PVR PA/AAmeas RLPAmeas† sABRmeas RV †
ERS-PoPH mPAP≥25 PVR>240* PAOP12
Patients without PoPH TPG≤12
n=17
n=32
44.2 (38-50) 32.8 (25-38) 357.8 (243-464) 1.09 (0.84-1.43) 27.4 (25.4-29.5) 1.84 (1.41-2.45) 42.7 (33.1-53.2)
42 (28-57) 24.6 (14-48) 224.4 (61-464) 1.04 (0.79-1.49) 26.0 (19.7-30.4) 1.49 (1.01-2.45) 43.9 (33.1-55.3)
23.2 (8-45) 6.9 (2-12) 54.9 (15-184) 0.91 (0.72-1.40) 22.4 (18.0-26.4) 1.23 (0.97-1.81) 40.8 (28.0-53.5)
p value*
