Liver International ISSN 1478-3223
CIRRHOSIS AND LIVER FAILURE
Characteristic haemodynamic changes of cirrhosis may influence the diagnosis of portopulmonary hypertension n1,3, Judith Go mez-Camarero1, Teresa Chiva1,2,3,*, Cristina Ripoll1,3,*, Fernando Sarnago3,4, Diego Rinco 2,3 1,3 3,4 1,2,3,5 ~ ares Evelyn Galindo , Marıa-Vega Catalina , Jaime Elizaga and Rafael Ban 1 2 3 4 5
~o n, Madrid, Spain Liver Unit Digestive Disease Department, Hospital General Universitario Gregorio Maran n Biom Centro de Investigacio edica de Enfermedades Hep aticas y Digestivas (CIBEREHD), Madrid, Spain n Sanitaria Gregorio Maran ~o n (IISGM), Madrid, Spain Instituto de Investigacio ~o n, Madrid, Spain Cardiac haemodynamic Lab, Cardiology Department, Hospital General Universitario Gregorio Maran Facultad de Medicina, Universidad Complutense, Madrid, Spain
Keywords liver transplantation – portopulmonary hypertension – pulmonary arterial hypertension – pulmonary vascular resistance – transpulmonary gradient Abbreviations CO, cardiac output; FHVP, free hepatic venous pressure; HVPG, hepatic venous pressure gradient; MELD, model of end-stage liver disease; MPAP, mean pulmonary arterial pressure; PAH, pulmonary arterial hypertension; PCWP, pulmonary capillary wedge pressure; POPH, portopulmonary hypertension; PVR, pulmonary vascular resistance; RAP, right atrial pressure; SVR, systemic vascular resistance; TPG, transpulmonary gradient; WHVP, wedge hepatic venous pressure. Correspondence ~ares, Liver Unit Digestive Disease Rafael Ban Department, Hospital General Universitario ~o n, CIBEREHD, IISGM, Dr Gregorio Maran Esquerdo 46; Madrid (28007), Spain Tel: (+34) 915868699 Fax: (+34) 915868018 e-mail: [email protected]
Abstract Background & Aims: Diagnosis of portopulmonary hypertension (POPH) is based on the presence of portal hypertension and the same haemodynamic criteria as pulmonary arterial hypertension (PAH). However, the typical hyperdynamic circulation of cirrhosis may have some impact on the diagnosis of POPH. The aim was to compare the haemodynamic pattern of the pulmonary circulation between cirrhotics and non-cirrhotics, including patients with PAH. Patients and Methods: 600 patients with cirrhosis [male 77.5%, age 54 (47–60) years, Child A: 14.7%, B: 54.3%, C: 31%] received right heart catheterization. For comparison, 118 non-cirrhotic patients [male 60%, age 64 (53–65) years] with right heart catheterization and PCWP 25 mmHg, PVR 120–240 dyn s cm 5 and PCWP 15 mmHg with TPG ≥12 mmHg); C: pulmonary arterial hypertension (same criteria as B except PVR ≥240 dyn s cm 5). Results: Distribution of patients with cirrhosis was A 583, B 7 and C 10. Prevalence of POPH was 1.7%. Cirrhotics had lower SVR and greater CO than non-cirrhotics (P < 0.05). Interestingly, patients with cirrhosis without PAH (groups A and B) had lower PVR (P < 0.05) when comparing with non-cirrhotics, while no differences in PVR were observed in group C. However, mean TPG was greater in group C of cirrhotics [36.6 mmHg (12.2) vs. 27.1 mmHg (10.1); P = 0.034]. Conclusions: Patients with cirrhosis have lower PVR. TPG is greater in POPH than PAH. Characteristic haemodynamic changes of cirrhosis may influence the diagnosis of POPH.
Received 28 October 2013 Accepted 1 April 2014 DOI:10.1111/liv.12562 Liver Int. 2015; 35: 353–361
Portopulmonary hypertension (POPH) is a rare pulmonary complication of cirrhosis defined by the presence of portal hypertension and pulmonary artery hypertension (PAH) with a reported 2–16% prevalence depending on the study population (1–9). POPH belongs to the Group I of diseases causing pulmonary arterial hypertension according to the 4th World Symposium on pulmonary *Both authors have equally contributed to this study and share the first authorship.
Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
hypertension classification (Dana point 2008) (10), previously known as primary pulmonary hypertension (11). The haemodynamic criteria that define PAH have evolved along the years (12) to the present definition established in 2004: an increased mean pulmonary arterial pressure (MPAP) >25 mmHg, pulmonary capillary wedge pressure (PCWP) 5 mmHg) is also required. Recently, some
353
Chiva et al.
Pulmonary pressure in cirrhosis
reports have also proposed inclusion of another haemodynamic criteria as defined by a transpulmonary gradient (TPG = MPAP PCWP) value ≥12 mmHg (8, 12). This criteria allows the diagnosis of pulmonary arterial hypertension in those cases in which a simultaneous venous pulmonary hypertension (PCWP >15 mmHg) is observed. Use of TPG in the diagnosis of pulmonary arterial hypertension (which also includes POPH), contemplates the realistic possibility that there can be more than one simultaneous mechanism that lead to pulmonary arterial hypertension. However, the haemodynamic abnormalities of end-stage liver disease could have some impact on the diagnostic criteria of POPH. Indeed, patients with cirrhosis have characteristically low systemic vascular resistance caused by the intense splanchnic vasodilatation that leads to central hypovolaemia and increased cardiac output (14–16). Although the haemodynamic characteristics of the pulmonary circulation in cirrhosis are not well defined, one could hypothesize that there would also be a decrease in calculated pulmonary vascular resistance, as defined by Ohm’s law [PVR = (MPAP PCWP) 9 80/CO], given the characteristic increase in cardiac output. However, there are no studies that have specifically evaluated this issue. If this hypothesis were true, and in order to achieve the threshold pulmonary vascular resistance required for diagnosis of pulmonary artery hypertension, patients with cirrhosis could require a greater increase of the gradient of pressure in the pulmonary circulation. Therefore, a greater lesion in the distal pulmonary arteries would be required. Interestingly, recent studies could support this hypothesis as they suggest that POPH patients have worse prognosis with greater mortality rate than idiopathic pulmonary arterial hypertension patients and perhaps even worse response to treatment, despite having a better haemodynamic profile as suggested by lower PVR, lower MPAP and higher cardiac output (17, 18). To our knowledge, no study has been specifically aimed to evaluate the haemodynamic pattern of pulmonary circulation in patients with cirrhosis. Therefore, the main purpose of this study was to compare the haemodynamic pattern of the pulmonary vascular circulation between patients with and without cirrhosis including patients with pulmonary arterial hypertension.
than 12-month lapse between transthoracic echocardiography and haemodynamic evaluation [median time lapse 2 (IQR: 2–11) days], Figure 1. This latter criterion was chosen taking into account the fact that progressive pulmonary haemodynamic changes have been described, so that repeat echocardiography in liver transplant candidates should be performed on a yearly basis (5, 19, 20). Most patients (83%) had less than 3-month time lapse between the two procedures. Splanchnic and right heart catheterization and echocardiography formed part of the routine work-up in all liver transplant candidates in our centre during the complete study period. We collected demographical and clinical variables at the moment of the haemodynamic study: age, sex, aetiology of the liver disease, Child-Pugh class, presence of transjugular intrahepatic portosystemic shunt, hepatocellular carcinoma, and treatment with betablockers, as well as blood test parameters in order to calculate MELD score. Haemodynamic parameters collected from the right heart catheterization included mean pulmonary artery pressure (MPAP), pulmonary capillary wedge pressure (PCWP), right atrium pressure (RAP), cardiac output (CO) and systemic and pulmonary vascular resistance (SVR and PVR). The hepatic venous pressure gradient (HVPG) from the splanchnic catheterization was registered. On the other hand, to compare the pulmonary haemodynamic pattern, 118 consecutive patients without volume overload (PCWP 25 mmHg, PVR 120– 240 dyn s cm 5 and PCWP 25 mmHg, PVR >240 dyn s cm 5 and PCWP 120 dyn s cm 5 (1, 6, 23). TPG ≥12 mmHg has been added in the definitions to consider the possibility of associated pulmonary venous hypertension. An initial analysis comparing the three groups (A, B and C) in the patients with cirrhosis was done, and then, the main analysis comparing patients with and without cirrhosis was performed. The statistical analysis was carried out using the SPSS program (version 15.0) and MedCalc program (version 11.1.1.0). Categorical variables were described with proportions, and continuous variables were described with medians and interquartile range. Normal distribution was tested with the Kolmogorov–Smirnov test. The association between categorical variables was evaluated with Chi-square and Fisher’s exact test and U-Mann Whitney or Kruskall–Wallis were used for continuous variables. In order to compare the differences of the haemodynamic variables between groups (A, B and C) among cirrhotic and non-cirrhotic patients, Student t-test was used. A P value of 0.05 was considered significant. Results
The study population finally consisted of 600 patients with cirrhosis (Fig. 1) and 118 patients without cirrhosis. Patients with cirrhosis
Demographical and biochemical data are shown on Table 1. Patients were mostly Child-Pugh class B (A: 14.7%, B: 54.3%, C: 31%) and had a median MELD score of 15.0 (12.0–19.0). More than half of the patients [327 (54.5%)] patients received liver transplantation in a median of 6 month time period after the haemodynamic study (IQR 3.0–10.5 months). The prevalence of POPH was 1.7% (95% CI: 0.82–3.08) in this group of patients, which had not been previously selected with heart ultrasound to make a right heart catheterization. The prevalence of POPH in the preselected population with PSAP >30 mmHg on heart ultrasound would be 4.7% (95% CI 2.2–8.7). Most patients belonged to group A (without criteria of pulmonary arterial hypertension), while very few patients belonged to group B (intermediate pulmonary haemodynamics) and C (PAH), see Table 2. When analysing the differences among the three groups in patients with cirrhosis, no differences were found regarding demographics and blood test values. Interestingly, a trend to lower HVPG in patients with greater PVR was observed (A: 19.0 (15.0–22.5) mmHg; B: 17.5 (12.5– 23.4) mmHg; C: 12.0 (5.6–20.4) mmHg; P = 0.095) (see Fig. 1). Individual pulmonary haemodynamics data of patients included in group C is shown on Table 3. According to the severity of POPH, there were 2 mild
355
Chiva et al.
Pulmonary pressure in cirrhosis
Table 1. Demographical, biochemical and haemodynamic baseline characteristics of patients with or without cirrhosis Variable Sex Male, n (%) Age (years) Aetiology Alcohol, n (%) HCV, n (%) HBV, n (%) Cholestatic, n (%) Other causes, n (%) Child-Pugh A, n (%) B, n (%) C, n (%) MELD Bilirubin (mg/dl) INR Creatinine (mg/dl) HCC, n (%) Use of betablockers, n (%) TIPS, n (%) LT, n (%) Tricuspid regurgitation, n (%) HVPG (mmHg) MPAP (mmHg) PCWP (mmHg) TPG (mmHg) RAP (mmHg) PVR (dyn s cm 5) SVR (dyn s cm 5) CO (L/min) HR (bpm) MAP (mmHg)
Cirrhosis
Non-cirrhosis
465 (77.5) 54 (47–60) 245 (41.0) 229 (38.4) 60 (10.1) 35 (5.9)
71 (60.0) 64 (53–65) –
28 (4.7)
–
85 (14.7) 315 (54.3) 180 (31.0) 15.0 (12.0–19.0) 2.5 (1.6–4.6) 1.4 (1.2–1.7) 0.9 (0.7–1.1)
– – – –
140 (23.5) 227 (40.5)
– –
19 (3.2) 327 (54.5) 234 (39.4)
– – –
18.5 (15.0–22.5) 14.5 (11.5–18.0) 8.5 (6.0–11.5) 6.0 (4.0–7.5) 5.0 (3.0–7.0) 62.7 (45.4–87.7) 848.0 (653.0–1102.0) 7.1 (5.7–8.8) 75 (65–88) 81.0 (72.0–90.0)
– 22.0 (18.0–28.5) 14.0 (10.7–16.0) 8.5 (6.0–11.0) 8 (5.6–12.2) 127.3 (85.5–192.8) 1185.7 (939.2–1561.1) 5.2 (4.5–6.2) 75 (65–88) 92.5 (80.0–104.0)
HCV, hepatic C virus; HBV, hepatic B virus; HCC, hepatocellular carcinoma; MELD, model of end- stage liver disease; INR, international normalizated ratio; TIPS, transjugular intrahepatic portosystemic shunt; LT, liver transplantation; HVPG, hepatic venous pressure gradient; MPAP, mean pulmonary arterial pressure; PCWP, pulmonary capillary wedge pressure; TPG, transpulmonary gradient; RAP, right atrial pressure; PVR, pulmonary vascular resistance; SVR, systemic vascular resistance; CO, cardiac output; HR, heart rate; MAP, mean arterial pressure.
(MPAP ≥25–35 mmHg), 3 moderate (MPAP ≥35– 45 mmHg) and 5 severe (MPAP ≥45 mmHg) cases. Interestingly, among patients from group A and B (without portopulmonary hypertension), no difference was observed in pulmonary vascular resistance according to use of betablockers (62 dyn s cm 5 vs. 61 dyn s cm 5, P = 0.528). Patients in group C with betablockers (n = 4) had a trend towards a higher pulmonary vascular resistance than those without betablockers
356
(n = 6) (PVR 602 dyn s cm 5 vs. 314 dyn s cm 5, P = 0.088). None of the patients with portopulmonary hypertension were finally transplanted. Although beyond the aim of this study, the comparison between the pre-transplant evaluation and liver transplantation haemodynamic data could be done in 62 of the 327 patients who were transplanted, with a median time span of 6.2 (3.0– 8.3) months between both measurements. The comparison was done with the baseline haemodynamic measurements at liver transplantation before laparotomy. As could be expected given the difference in the measurement conditions (intubation, anaesthesia), there were differences between both measurements. Proportionally there was an increase in all measurements at the time of liver transplantation [PMAP 30% (12–42%), PCP 29% (6–56%) and TPG 19 ( 32–50%)], although no patient was diagnosed of portopulmonary hypertension at the time of liver transplantation. Patients without cirrhosis
Among the patients without cirrhosis (n = 118), 71 were males (60%) and had a median age of 64 years (IQR 53–65). Haemodynamic parameters are described in Table 1. The indication for right heart catheterization was: follow-up after heart transplantation in 30 patients, 44 patients were suspected to have pulmonary hypertension on cardiac ultrasound and 44 patients had right heart catheterization in the context of evaluation previous to cardiac surgery because of left heart diseases. As could be expected taking into account the selection criteria of these patients, the prevalence of PAH was fairly high, 15.0% IC: 9.1–22.7 (18 patients) using the current criteria. The causes of PAH of these patients (group C) were: five idiopathic pulmonary arterial hypertension, two congenital heart disease, four patients with associated left heart dysfunction, five underlying lung diseases and two patients with chronic thromboembolic disease. Comparison of haemodynamic characteristics in patients with and without cirrhosis
Pulmonary and systemic haemodynamic parameters were compared between patients with and without cirrhosis (Table 4). As expected because of the hyperdynamic circulation, patients with cirrhosis had lower SVR and higher cardiac output than patients without cirrhosis. In the pulmonary circulation, PVR was similar in patients with and without cirrhosis in the group with pulmonary artery hypertension (group C), while cirrhotic patients in group A (absence of PAH) and group B (intermediate) had significantly lower PVR [67.3 (34.3) dyn s cm 5 vs. 111.0 (57.3) dyn s cm 5 (P < 0.050); 156.1 (27.5) vs. 192.6 (31.4) dyn s cm 5 (P = 0.029) respectively] than patients without cirrhosis. Lastly, TPG was significantly greater in patients with cirrhosis and POPH than in non-cirrhotic patients with Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Chiva et al.
Pulmonary pressure in cirrhosis
Table 2. Definition of the study groups and distribution of the patients Definition of groups Group A Absence of pulmonary arterial hypertension Group B Intermediate MPAP ≥25 mmHg, PVR 120–240 dyn s cm 5 and PCWP
CIRRHOSIS AND LIVER FAILURE
Characteristic haemodynamic changes of cirrhosis may influence the diagnosis of portopulmonary hypertension n1,3, Judith Go mez-Camarero1, Teresa Chiva1,2,3,*, Cristina Ripoll1,3,*, Fernando Sarnago3,4, Diego Rinco 2,3 1,3 3,4 1,2,3,5 ~ ares Evelyn Galindo , Marıa-Vega Catalina , Jaime Elizaga and Rafael Ban 1 2 3 4 5
~o n, Madrid, Spain Liver Unit Digestive Disease Department, Hospital General Universitario Gregorio Maran n Biom Centro de Investigacio edica de Enfermedades Hep aticas y Digestivas (CIBEREHD), Madrid, Spain n Sanitaria Gregorio Maran ~o n (IISGM), Madrid, Spain Instituto de Investigacio ~o n, Madrid, Spain Cardiac haemodynamic Lab, Cardiology Department, Hospital General Universitario Gregorio Maran Facultad de Medicina, Universidad Complutense, Madrid, Spain
Keywords liver transplantation – portopulmonary hypertension – pulmonary arterial hypertension – pulmonary vascular resistance – transpulmonary gradient Abbreviations CO, cardiac output; FHVP, free hepatic venous pressure; HVPG, hepatic venous pressure gradient; MELD, model of end-stage liver disease; MPAP, mean pulmonary arterial pressure; PAH, pulmonary arterial hypertension; PCWP, pulmonary capillary wedge pressure; POPH, portopulmonary hypertension; PVR, pulmonary vascular resistance; RAP, right atrial pressure; SVR, systemic vascular resistance; TPG, transpulmonary gradient; WHVP, wedge hepatic venous pressure. Correspondence ~ares, Liver Unit Digestive Disease Rafael Ban Department, Hospital General Universitario ~o n, CIBEREHD, IISGM, Dr Gregorio Maran Esquerdo 46; Madrid (28007), Spain Tel: (+34) 915868699 Fax: (+34) 915868018 e-mail: [email protected]
Abstract Background & Aims: Diagnosis of portopulmonary hypertension (POPH) is based on the presence of portal hypertension and the same haemodynamic criteria as pulmonary arterial hypertension (PAH). However, the typical hyperdynamic circulation of cirrhosis may have some impact on the diagnosis of POPH. The aim was to compare the haemodynamic pattern of the pulmonary circulation between cirrhotics and non-cirrhotics, including patients with PAH. Patients and Methods: 600 patients with cirrhosis [male 77.5%, age 54 (47–60) years, Child A: 14.7%, B: 54.3%, C: 31%] received right heart catheterization. For comparison, 118 non-cirrhotic patients [male 60%, age 64 (53–65) years] with right heart catheterization and PCWP 25 mmHg, PVR 120–240 dyn s cm 5 and PCWP 15 mmHg with TPG ≥12 mmHg); C: pulmonary arterial hypertension (same criteria as B except PVR ≥240 dyn s cm 5). Results: Distribution of patients with cirrhosis was A 583, B 7 and C 10. Prevalence of POPH was 1.7%. Cirrhotics had lower SVR and greater CO than non-cirrhotics (P < 0.05). Interestingly, patients with cirrhosis without PAH (groups A and B) had lower PVR (P < 0.05) when comparing with non-cirrhotics, while no differences in PVR were observed in group C. However, mean TPG was greater in group C of cirrhotics [36.6 mmHg (12.2) vs. 27.1 mmHg (10.1); P = 0.034]. Conclusions: Patients with cirrhosis have lower PVR. TPG is greater in POPH than PAH. Characteristic haemodynamic changes of cirrhosis may influence the diagnosis of POPH.
Received 28 October 2013 Accepted 1 April 2014 DOI:10.1111/liv.12562 Liver Int. 2015; 35: 353–361
Portopulmonary hypertension (POPH) is a rare pulmonary complication of cirrhosis defined by the presence of portal hypertension and pulmonary artery hypertension (PAH) with a reported 2–16% prevalence depending on the study population (1–9). POPH belongs to the Group I of diseases causing pulmonary arterial hypertension according to the 4th World Symposium on pulmonary *Both authors have equally contributed to this study and share the first authorship.
Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
hypertension classification (Dana point 2008) (10), previously known as primary pulmonary hypertension (11). The haemodynamic criteria that define PAH have evolved along the years (12) to the present definition established in 2004: an increased mean pulmonary arterial pressure (MPAP) >25 mmHg, pulmonary capillary wedge pressure (PCWP) 5 mmHg) is also required. Recently, some
353
Chiva et al.
Pulmonary pressure in cirrhosis
reports have also proposed inclusion of another haemodynamic criteria as defined by a transpulmonary gradient (TPG = MPAP PCWP) value ≥12 mmHg (8, 12). This criteria allows the diagnosis of pulmonary arterial hypertension in those cases in which a simultaneous venous pulmonary hypertension (PCWP >15 mmHg) is observed. Use of TPG in the diagnosis of pulmonary arterial hypertension (which also includes POPH), contemplates the realistic possibility that there can be more than one simultaneous mechanism that lead to pulmonary arterial hypertension. However, the haemodynamic abnormalities of end-stage liver disease could have some impact on the diagnostic criteria of POPH. Indeed, patients with cirrhosis have characteristically low systemic vascular resistance caused by the intense splanchnic vasodilatation that leads to central hypovolaemia and increased cardiac output (14–16). Although the haemodynamic characteristics of the pulmonary circulation in cirrhosis are not well defined, one could hypothesize that there would also be a decrease in calculated pulmonary vascular resistance, as defined by Ohm’s law [PVR = (MPAP PCWP) 9 80/CO], given the characteristic increase in cardiac output. However, there are no studies that have specifically evaluated this issue. If this hypothesis were true, and in order to achieve the threshold pulmonary vascular resistance required for diagnosis of pulmonary artery hypertension, patients with cirrhosis could require a greater increase of the gradient of pressure in the pulmonary circulation. Therefore, a greater lesion in the distal pulmonary arteries would be required. Interestingly, recent studies could support this hypothesis as they suggest that POPH patients have worse prognosis with greater mortality rate than idiopathic pulmonary arterial hypertension patients and perhaps even worse response to treatment, despite having a better haemodynamic profile as suggested by lower PVR, lower MPAP and higher cardiac output (17, 18). To our knowledge, no study has been specifically aimed to evaluate the haemodynamic pattern of pulmonary circulation in patients with cirrhosis. Therefore, the main purpose of this study was to compare the haemodynamic pattern of the pulmonary vascular circulation between patients with and without cirrhosis including patients with pulmonary arterial hypertension.
than 12-month lapse between transthoracic echocardiography and haemodynamic evaluation [median time lapse 2 (IQR: 2–11) days], Figure 1. This latter criterion was chosen taking into account the fact that progressive pulmonary haemodynamic changes have been described, so that repeat echocardiography in liver transplant candidates should be performed on a yearly basis (5, 19, 20). Most patients (83%) had less than 3-month time lapse between the two procedures. Splanchnic and right heart catheterization and echocardiography formed part of the routine work-up in all liver transplant candidates in our centre during the complete study period. We collected demographical and clinical variables at the moment of the haemodynamic study: age, sex, aetiology of the liver disease, Child-Pugh class, presence of transjugular intrahepatic portosystemic shunt, hepatocellular carcinoma, and treatment with betablockers, as well as blood test parameters in order to calculate MELD score. Haemodynamic parameters collected from the right heart catheterization included mean pulmonary artery pressure (MPAP), pulmonary capillary wedge pressure (PCWP), right atrium pressure (RAP), cardiac output (CO) and systemic and pulmonary vascular resistance (SVR and PVR). The hepatic venous pressure gradient (HVPG) from the splanchnic catheterization was registered. On the other hand, to compare the pulmonary haemodynamic pattern, 118 consecutive patients without volume overload (PCWP 25 mmHg, PVR 120– 240 dyn s cm 5 and PCWP 25 mmHg, PVR >240 dyn s cm 5 and PCWP 120 dyn s cm 5 (1, 6, 23). TPG ≥12 mmHg has been added in the definitions to consider the possibility of associated pulmonary venous hypertension. An initial analysis comparing the three groups (A, B and C) in the patients with cirrhosis was done, and then, the main analysis comparing patients with and without cirrhosis was performed. The statistical analysis was carried out using the SPSS program (version 15.0) and MedCalc program (version 11.1.1.0). Categorical variables were described with proportions, and continuous variables were described with medians and interquartile range. Normal distribution was tested with the Kolmogorov–Smirnov test. The association between categorical variables was evaluated with Chi-square and Fisher’s exact test and U-Mann Whitney or Kruskall–Wallis were used for continuous variables. In order to compare the differences of the haemodynamic variables between groups (A, B and C) among cirrhotic and non-cirrhotic patients, Student t-test was used. A P value of 0.05 was considered significant. Results
The study population finally consisted of 600 patients with cirrhosis (Fig. 1) and 118 patients without cirrhosis. Patients with cirrhosis
Demographical and biochemical data are shown on Table 1. Patients were mostly Child-Pugh class B (A: 14.7%, B: 54.3%, C: 31%) and had a median MELD score of 15.0 (12.0–19.0). More than half of the patients [327 (54.5%)] patients received liver transplantation in a median of 6 month time period after the haemodynamic study (IQR 3.0–10.5 months). The prevalence of POPH was 1.7% (95% CI: 0.82–3.08) in this group of patients, which had not been previously selected with heart ultrasound to make a right heart catheterization. The prevalence of POPH in the preselected population with PSAP >30 mmHg on heart ultrasound would be 4.7% (95% CI 2.2–8.7). Most patients belonged to group A (without criteria of pulmonary arterial hypertension), while very few patients belonged to group B (intermediate pulmonary haemodynamics) and C (PAH), see Table 2. When analysing the differences among the three groups in patients with cirrhosis, no differences were found regarding demographics and blood test values. Interestingly, a trend to lower HVPG in patients with greater PVR was observed (A: 19.0 (15.0–22.5) mmHg; B: 17.5 (12.5– 23.4) mmHg; C: 12.0 (5.6–20.4) mmHg; P = 0.095) (see Fig. 1). Individual pulmonary haemodynamics data of patients included in group C is shown on Table 3. According to the severity of POPH, there were 2 mild
355
Chiva et al.
Pulmonary pressure in cirrhosis
Table 1. Demographical, biochemical and haemodynamic baseline characteristics of patients with or without cirrhosis Variable Sex Male, n (%) Age (years) Aetiology Alcohol, n (%) HCV, n (%) HBV, n (%) Cholestatic, n (%) Other causes, n (%) Child-Pugh A, n (%) B, n (%) C, n (%) MELD Bilirubin (mg/dl) INR Creatinine (mg/dl) HCC, n (%) Use of betablockers, n (%) TIPS, n (%) LT, n (%) Tricuspid regurgitation, n (%) HVPG (mmHg) MPAP (mmHg) PCWP (mmHg) TPG (mmHg) RAP (mmHg) PVR (dyn s cm 5) SVR (dyn s cm 5) CO (L/min) HR (bpm) MAP (mmHg)
Cirrhosis
Non-cirrhosis
465 (77.5) 54 (47–60) 245 (41.0) 229 (38.4) 60 (10.1) 35 (5.9)
71 (60.0) 64 (53–65) –
28 (4.7)
–
85 (14.7) 315 (54.3) 180 (31.0) 15.0 (12.0–19.0) 2.5 (1.6–4.6) 1.4 (1.2–1.7) 0.9 (0.7–1.1)
– – – –
140 (23.5) 227 (40.5)
– –
19 (3.2) 327 (54.5) 234 (39.4)
– – –
18.5 (15.0–22.5) 14.5 (11.5–18.0) 8.5 (6.0–11.5) 6.0 (4.0–7.5) 5.0 (3.0–7.0) 62.7 (45.4–87.7) 848.0 (653.0–1102.0) 7.1 (5.7–8.8) 75 (65–88) 81.0 (72.0–90.0)
– 22.0 (18.0–28.5) 14.0 (10.7–16.0) 8.5 (6.0–11.0) 8 (5.6–12.2) 127.3 (85.5–192.8) 1185.7 (939.2–1561.1) 5.2 (4.5–6.2) 75 (65–88) 92.5 (80.0–104.0)
HCV, hepatic C virus; HBV, hepatic B virus; HCC, hepatocellular carcinoma; MELD, model of end- stage liver disease; INR, international normalizated ratio; TIPS, transjugular intrahepatic portosystemic shunt; LT, liver transplantation; HVPG, hepatic venous pressure gradient; MPAP, mean pulmonary arterial pressure; PCWP, pulmonary capillary wedge pressure; TPG, transpulmonary gradient; RAP, right atrial pressure; PVR, pulmonary vascular resistance; SVR, systemic vascular resistance; CO, cardiac output; HR, heart rate; MAP, mean arterial pressure.
(MPAP ≥25–35 mmHg), 3 moderate (MPAP ≥35– 45 mmHg) and 5 severe (MPAP ≥45 mmHg) cases. Interestingly, among patients from group A and B (without portopulmonary hypertension), no difference was observed in pulmonary vascular resistance according to use of betablockers (62 dyn s cm 5 vs. 61 dyn s cm 5, P = 0.528). Patients in group C with betablockers (n = 4) had a trend towards a higher pulmonary vascular resistance than those without betablockers
356
(n = 6) (PVR 602 dyn s cm 5 vs. 314 dyn s cm 5, P = 0.088). None of the patients with portopulmonary hypertension were finally transplanted. Although beyond the aim of this study, the comparison between the pre-transplant evaluation and liver transplantation haemodynamic data could be done in 62 of the 327 patients who were transplanted, with a median time span of 6.2 (3.0– 8.3) months between both measurements. The comparison was done with the baseline haemodynamic measurements at liver transplantation before laparotomy. As could be expected given the difference in the measurement conditions (intubation, anaesthesia), there were differences between both measurements. Proportionally there was an increase in all measurements at the time of liver transplantation [PMAP 30% (12–42%), PCP 29% (6–56%) and TPG 19 ( 32–50%)], although no patient was diagnosed of portopulmonary hypertension at the time of liver transplantation. Patients without cirrhosis
Among the patients without cirrhosis (n = 118), 71 were males (60%) and had a median age of 64 years (IQR 53–65). Haemodynamic parameters are described in Table 1. The indication for right heart catheterization was: follow-up after heart transplantation in 30 patients, 44 patients were suspected to have pulmonary hypertension on cardiac ultrasound and 44 patients had right heart catheterization in the context of evaluation previous to cardiac surgery because of left heart diseases. As could be expected taking into account the selection criteria of these patients, the prevalence of PAH was fairly high, 15.0% IC: 9.1–22.7 (18 patients) using the current criteria. The causes of PAH of these patients (group C) were: five idiopathic pulmonary arterial hypertension, two congenital heart disease, four patients with associated left heart dysfunction, five underlying lung diseases and two patients with chronic thromboembolic disease. Comparison of haemodynamic characteristics in patients with and without cirrhosis
Pulmonary and systemic haemodynamic parameters were compared between patients with and without cirrhosis (Table 4). As expected because of the hyperdynamic circulation, patients with cirrhosis had lower SVR and higher cardiac output than patients without cirrhosis. In the pulmonary circulation, PVR was similar in patients with and without cirrhosis in the group with pulmonary artery hypertension (group C), while cirrhotic patients in group A (absence of PAH) and group B (intermediate) had significantly lower PVR [67.3 (34.3) dyn s cm 5 vs. 111.0 (57.3) dyn s cm 5 (P < 0.050); 156.1 (27.5) vs. 192.6 (31.4) dyn s cm 5 (P = 0.029) respectively] than patients without cirrhosis. Lastly, TPG was significantly greater in patients with cirrhosis and POPH than in non-cirrhotic patients with Liver International (2015) © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Chiva et al.
Pulmonary pressure in cirrhosis
Table 2. Definition of the study groups and distribution of the patients Definition of groups Group A Absence of pulmonary arterial hypertension Group B Intermediate MPAP ≥25 mmHg, PVR 120–240 dyn s cm 5 and PCWP
