Liver transplantation (LT) is the definitive treatment for many end-stage liver diseases. Portopulmonary hypertension (PPH) is a known complication of cirrhosis and is frequently encountered in patients requiring LT. Several definitions for PPH have been proposed, but usually PPH is defined by mean pulmonary artery pressure (mPAP) >25 mm Hg with a normal pulmonary capillary wedge pressure in cirrhotic patients.1–3 The incidence of PPH is difficult to determine in part because of different definitions.4–6 Early in the course of PPH, increased cardiac output and blood flow increase the pulmonary artery (PA) pressure without inducing anatomical changes in the pulmonary vasculature or increasing pulmonary vascular resistance (PVR).7,8 In these patients, right ventricular (RV) ejection fraction (RVEF) is a product of a complex interaction between RV preload, ventricular contractility, and afterload.9
Previous studies have concluded that mild or moderate PPH is not a contraindication for LT.7,10 However, very few studies have reported on the hemodynamic effects of PPH on RV function during LT.11,12 Furthermore, because these studies were not prospective, many confounders (including intraoperative fluid management) were not controlled. The use of healthy, living, related donors in our study helps avoid the confounding effects of variability in quality and ischemia times of cadaveric grafts.13 Early RVEF, especially at the early postreperfusion period was the primary objective for this prospective controlled study. Our secondary objectives were to differentially quantify the impact of mild to moderate PPH on intraoperative RV performance as well as outcome variables in living donor LT (LDLT) recipients.
Twenty adult patients in the Mansoura University Liver Transplantation Program scheduled for LDLT were prospectively enrolled into this controlled, double-blind study from June 2008 to January 2011. The study was approved by the local research ethics committee of the Mansoura University Liver Transplantation Program and written informed consent was secured from all enrolled subjects. Eligibility criteria included age 20 to 60 years, preoperative left ventricular ejection fraction >60%, and no evidence of cardiac ischemia by radionucleotide stress testing. Exclusion criteria included retransplantation, Budd-Chiari syndrome, autoimmune hepatitis, previous upper abdominal surgery, and use of cardiac drugs. A single cardiologist evaluated the patients preoperatively by transthoracic echocardiography and Doppler for initial assessment of systolic PA pressure (sPAP). Although patients with (echocardiographic estimated) sPAP ≥30 mm Hg were considered to have PPH, these patients were enrolled in the PPH group only if mPAP >25 mm Hg was confirmed by PA catheter after induction of anesthesia. All patients with confirmed PPH were included in the study. To nullify the effects of increased experience over time, each PPH case enrolled was followed by random allocation of a patient to the control group, using a computer-generated randomization, read by an anesthesiology resident blinded to the protocol. All patients were unaware of their PA pressures, and a dedicated resident anesthesiologist who was unaware of the nature of the study was responsible for recording intraoperative data.
The transplants were all performed by the same anesthesiology and surgical teams, and the piggyback technique was used. Patients were premedicated with midazolam (3 mg) and pantoprazole sodium (40 mg) IV 15 minutes before anesthesia. Noninvasive arterial blood pressure, electrocardiogram, oxygen saturation, and nasopharyngeal temperature were monitored. Upon admission to the operating theater, lidocaine (60 mg) was injected IV to prepare the vein followed by propofol 1.5 mg · kg−1 and fentanyl 2 μg · kg−1. Tracheal intubation was facilitated by rocuronium bromide 1.5 mg · kg−1. The lungs were ventilated with 40% oxygen in air to keep end-tidal CO2 approximately 33 mm Hg using volume control mode with positive end-expiratory pressure 3 mm Hg. Anesthesia was maintained by continuous IV infusion of fentanyl 2 μg · kg · h−1 and rocuronium bromide 300 μg · kg · h−1 and inhaled sevoflurane to maintain the Bispectral Index around 50. An arterial catheter was introduced into the left radial artery, and Bispectral Index was monitored using an Angstrom AS5 Monitor (Datex-Ohmeda AS5, Microvitec Display LTD, Bradford, UK). A 7.5F modified continuous thermal fiberoptic PA catheter (CCO/SvO2/CEDV; Edwards Lifesciences, Irvine, CA) was introduced into the right internal jugular vein under strict aseptic conditions. Insertion and the final positioning were monitored by fluoroscopy. PA and central venous pressures (CVP) were recorded. The catheter transducers were then attached to a Vigilance monitor (Edwards Lifesciences) for continuous online monitoring of cardiac index, stroke volume index (SVI), RV end-diastolic volume index (RVEDVI), mixed venous oxygen saturation, and blood temperature. Other calculated hemodynamic variables were generated by the Datex-Ohmeda monitor, including systemic vascular resistance index (SVRI), PVR index, the RV stroke work index (RVSWI), and the left ventricular stroke work index.
To nullify a confounding effect of fluids on the RV preload state and subsequently RVEF, a standard goal-directed approach for fluid management was implemented in all patients. Fluids were infused to maintain SVI within 20% of baseline value and to keep mean arterial blood pressure >65 mm Hg by using 250-mL boluses of either Voluven® (Fresenius Kabi AG, Bad Homburg, Germany) or albumin 4%. Initially, Voluven was used until a maximal volume of 30 mL · kg−1 was reached, then boluses of albumin 4% were used thereafter. However, albumin was used instead of Voluven whenever serum albumin was <3.5 mg · dL−1. Blood component transfusion followed the American Society of Anesthesiologists' guidelines for both cell saver and allogeneic blood products.14 On reperfusion, norepinephrine was administered only if systolic blood pressure was <90 mm Hg for >3 minutes, with CVP >4 mm Hg and >30% reduction of SVRI from its baseline value. All monitored and calculated variables were recorded at 6 time points: 10 minutes after PA catheter insertion (baseline), at end hepatectomy, immediately before portal unclamping (anhepatic), 5 and 30 minutes postreperfusion (5PR and 30PR), and at skin closure.
Previous work in our center on 23 LT recipients with normal PA pressures established a 20% decrease in RVEF at 5 minutes PR (primary objective) as a clinically significant effect with a standard deviation of 8.6% (unpublished data). Sample size calculation aiming for 90% power at 0.05 significance predicted that a minimum sample size of 8 cases in each group would detect clinically significant differences. The data are presented as mean ± SD unless otherwise indicated. The quantitative data distribution was tested for normality using the Shapiro-Wilk test. All data exhibited normal distribution (all P > 0.485). Two-tailed independent sample t test “equal variance assumed” was used to test for significant differences between groups. Data were analyzed by using STATISTICA® version 8 (StatSoft, Inc., Tulsa, OK). Statistical significance was set at P < 0.05.
Seventeen patients were diagnosed with PPH by transthoracic Doppler (sPAP >30 mm Hg). Of these, 10 patients were enrolled in the PPH group after confirming a baseline PA catheter reading of ≥25 mm Hg. One patient in the PPH group did not complete the study and was excluded because of a monitor malfunction, leaving 9 patients for analysis in the PPH group. In the PPH group, 6 patients (66.7%) had mild PPH (mPAP 25–34 mm Hg) and 3 (33.3%) had moderate PPH (mPAP 35–44 mm Hg). As shown in Table 1, there were no statistically significant differences in terms of demographic, perioperative, and outcome data, except Child-Pugh and Model of End-Stage Liver Disease (MELD) scores, which were significantly higher in the PPH group compared with controls (Child-Pugh 12.2 ± 4.1 vs 9.2 ± 2.0, P = 0.04; MELD 17 ± 5 vs 14 ± 2, P = 0.02). One patient (11.1%) in the PPH group was transplanted for hepatocellular carcinoma versus 30.0% of controls. None of the subjects in either group required norepinephrine intraoperatively or in the immediate postoperative period. Six-month patient and graft survival did not differ significantly between groups. One PPH patient (baseline mPAP 32 mm Hg) died from intracerebral hemorrhage 39 days postoperatively.
Hemodynamic data are presented in Table 2. The primary objective in this trial was the RVEF 5 minutes PR and it was significantly lower in the PPH group compared with the control group. RVEF was also lower in the PPH group at 30 minutes PR and at closure (Fig. 1). The PPH group had a lower SVI than the control group after portal unclamping, at 5 minutes PR, 30 minutes PR, and closure. As expected, mPAP and PVR index were significantly higher whereas SVRI was significantly lower in the PPH group versus controls (Table 2, Fig. 2). RVSWI was similar in the 2 groups (Fig. 3). Compared with controls, the PPH group had significantly higher CVP and RVEDVI (RV preload) values from early reperfusion until closure (Table 2, Fig. 3).
We investigated the impact of PPH on the RV dynamics in LDLT recipients. Despite differences in the RV performance between PPH and control groups, mild to moderate PPH did not lead to clinically significant intraoperative outcome differences including early graft failure, persistent RV failure, or death.
The RV is challenged during LT. Increased RV preload due to volume overload and unfavorable changes in the pulmonary vascular system in patients with a cirrhotic liver constitute a constant strain on the RV. Hyperdynamic circulation with low systemic vascular resistance is typical in cirrhotics because of increased levels of nitric oxide, proinflammatory mediators, and other endogenous vasodilators.2,15 Poor cardiac contractility is another frequent complication of liver disease.16
Using a definition of RV dysfunction of RVEF <30%,17 patients with PPH in our study had RV dysfunction with significantly lower RVEF than the control group throughout the transplant with further exacerbation after portal unclamping.
RVEF is dependent on RV preload, afterload, and contractility.9 In this study, analysis of these determinants revealed the degree of contribution of each to poor RV performance in the PPH group. RVEDVI reflects diastolic function of the RV and is a sensitive preload indicator.18 Della Rocca et al.16 point out that volumetric indices are more sensitive than pressure indices for assessing volume status. Siniscalchi et al.19 found that RVEDVI is a sensitive predictor of preload, even in patients with low RVEF during LT. In the current study, RVEDVI was similar in both groups until portal unclamping but thereafter was significantly increased in the PPH group. Likely blood loss during hepatectomy and reduced venous return with partial cross-clamping of the vena cava during the anhepatic phase resulted in reduced central filling pressures, keeping the RV unstressed, and explaining the similar RV performance in the 2 groups in the early phases of LT.10,16 It is also worth noting that preload was well maintained in our patients, such that an underloaded RV was not likely a contributor to the reduced RVEF in the PPH group.
With a low muscle mass, the RV is sensitive to pressure rather than volume loads.20 Although PVR reflects the static portion of RV afterload, PVR has been used as a measure of RV afterload.9 In patients with mild to moderate PPH, increased flow across vasoconstricted, but anatomically normal, pulmonary vasculature increases the mPAP with only modest increases in PVR.1,3 Furthermore, the PR preload surge did not lead to a significant increase in the PVR in the PPH (or control) group.
Based on Frank-Starling relationships, increasing RVEDVI should induce proportional increases in RVEF through increasing the RV contractile forces. In the PPH group, the increased RVEDVI observed after unclamping was associated with a significant reduction of the RVEF as well as the SVI compared with baseline values, whereas the RV afterload did not differ from pre-reperfusion readings. However, the relationship between RVEDVI and RVEF in controls did not significantly change after reperfusion. Overall, these data point to failure of the RV contractile reserve in the PPH patients to maintain RVEF at baseline levels with the stress of reperfusion. Acosta et al.12 studied RV contractile reserve during LT in 9 patients versus 20 controls using RVSWI as an indicator of RV contractility. They concluded that cirrhotic patients with mild PPH retain the capacity to increase RV contractility, and also attributed this capacity to Frank-Starling forces based on finding that RVEDVI increased at the same time as RVSWI. Our finding that RVSWI did not change significantly during LT in either group may seem inconsistent with the conclusion of Acosta et al.; however, that group used dobutamine as a stress to study RV contractility and pulmonary vasodilatation reserve. The use of dobutamine with its inotropic effect may explain why the PPH patients in the study by Acosta et al. maintained RVEF whereas patients in our study did not. A report by De Wolf et al.21 on RV function during LT suggested that increased RV oxygen consumption could generate RV ischemia, in turn contributing to impaired RV contractility at reperfusion.
Hoeper et al.22 found that measuring RVEF in patients with pulmonary hypertension by using PA catheters underestimates RVEF relative to more accurate magnetic resonance imaging techniques. Therefore, the low RVEF values in PPH patients in our study may underestimate the actual RVEF because of an inherent defect in the measuring technique. If so, the RVEF values after reperfusion in PPH patients could actually be higher than recorded, and therefore, in healthier clinical ranges. Another limitation of our study is the unavailability of intraoperative transesophageal echocardiography. Nonetheless, even without the added information and confirmation of findings by transesophageal echocardiography, we believe our study shows that patients with mild to moderate PPH, who receive a healthy graft, are likely to be stable hemodynamically during LT.
In conclusion, although RVEF was lower in patients with PPH than controls, and RV hemodynamics deteriorated after reperfusion in the PPH group, these patients experienced no significant morbidity or mortality with LT. The lack of clinical impact of PPH in this study is likely attributable to the mild to moderate degree of PPH, with some contribution from healthy grafts with short ischemia times. Our results and conclusions cannot be extrapolated to patients with severe PPH who are at high risk of major perioperative LT morbidity and mortality.
Name: Amr M. Yassen, MD.
Contribution: This author helped design and conduct the study, analyze data, and write the manuscript.
Attestation: Amr M. Yassen reviewed the original study data and its analysis, approved the final manuscript, and is the author responsible for archiving the study files.
Name: Waleed R. Elsarraf, MD.
Contribution: This author helped conduct the study.
Attestation: Waleed R. Elsarraf has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Mohamed Elsadany, MD.
Contribution: This author helped design the study.
Attestation: Mohamed Elsadany has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Mohamed M. Elshobari, MD.
Contribution: This author helped analyze the data.
Attestation: Mohamed M. Elshobari has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Tarek Salah, MD.
Contribution: This author helped write the manuscript.
Attestation: Tarek Salah has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Ahmed M. Sultan, MD.
Contribution: This author helped analyze the data and write the manuscript.
Attestation: Ahmed M. Sultan has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
This manuscript was handled by: Edward C. Nemergut, MD.
1. Hoeper MM. The new definition of pulmonary hypertension. Eur Respir J 2009;34:790–1
2. Herve P, Lebrec D, Brenot FS, Humbert M, Sitbon O, Duroux P. Pulmonary vascular disorders in portal hypertension. Eur Respir J 1998;11:1153–66
3. Hoeper MM, Krowka MJ, Strassburg CP. Portopulmonary hypertension and hepatopulmonary syndrome. Lancet 2004;363:1461–8
4. Herve P, Le Pavec J, Sztrymf B, Decante B, Savale L, Sitbon O. Pulmonary vascular abnormalities in cirrhosis. Best Pract Res Clin Gastroenterol 2007;21:141–59
5. Starkle P, Vera A, Gunson B, Mutimer D. Outcome of liver transplantation for patients with pulmonary hypertension. Liver Transpl 2002;4:382–8
6. Krowka MJ. Pulmonary hypertension, high risk of orthotopic liver transplantation, and home lessons from “primary” pulmonary hypertension. Liver Transpl 2002;8:389–90
7. Michael AR, Simpson BR, Nguyen A, Ramasy KJ, Cara E, Klintmalm BG. Severe pulmonary hypertension in liver transplantation candidates. Liver Transpl Surg 1997;3:494–500
8. Krowka MJ, Plevak DJ, Findlay JY, Rosen CB, Wiesener RH, Krom AR. Pulmonary hemodynamics and perioperative cardiopulmonary-related mortality in patients with portopulmonary hypertension undergoing liver transplantation. Liver Transpl 2000;6:443–50
9. Haddad H, Hunt SA, Rosenthal DN, Murphy DJ. Right ventricular function in cardiovascular disease. Part I. Anatomy, physiology, aging, and functional assessment of the right ventricle. Circulation 2008;117:1436–48
10. Taura P, Garica-Valdecasas JC, Beltran J, Izquierdo E, Navasa M, Sala-Blanch J, Mas A, Balust J, Grande L, Visa J. Moderate primary pulmonary hypertension in patients undergoing liver transplantation. Anesth Analg 1996;83:675–80
11. Acosta F, Sansano T, Palenciano CG, Domenech P, Falcon L, Robles R, Bueno FS, Ramirez P, Parrilla P. Portopulmonary hypertension and liver transplantation: hemodynamic consequences at reperfusion. Transplant Proc 2005;37:3865–6
12. Acosta F, Sansano T, Palenciano CG, Falcon L, Domenech P, Robles R, Bueno FS, Ramirez P, Parrilla P. Effects of dobutamine on right ventricular function and pulmonary circulation in pulmonary hypertension during liver transplantation. Transplant Proc 2005;37:3869–70
13. Olivari MT, Fiorentini C, Polese A, Guazzi MD. Pulmonary hemodynamic and right ventricular function in hypertension. Circulation 1978;57:1185–90
14. American Society of Anesthesiologists Task Force on Perioperative Blood Transfusion And Adjuvant Therapy. Practice guidelines for perioperative blood transfusion and adjuvant therapies. Anesthesiology 2006;105:198–208
15. Atucha NM, Nadal FJ, Iyu D, Alcaraz A, Rodriguez-Barbero A, Ortiz MC, Lopez-Novoa JM, Garcia-Estan J. Role of vascular nitric oxide in experimental liver cirrhosis. Curr Vasc Pharmacol 2005;3:81–5
16. Della Rocca G, Pompei L, Costa MG, Coccia C, Rossi M, Berloco PM, Pietropaoli P, Cortesini R. Hemodynamic-volumetric versus pulmonary artery catheter monitoring during anesthesia for liver transplantation. Transplant Proc 2001;33:1394–6
17. Kim YK, Shin WJ, Song JG, Jun IG, Kim HY, Seeong SH, Huh IY, Hwang GS. Effect of right ventricular dysfunction on dynamic preload indices to predict a decrease in cardiac output after inferior vena cava clamping during liver transplantation. Transplant Proc 2010;42:2585–9
18. Della Rocca G, Costa MG, Feltracco P, Biancofiore G, Begliomini B, Taddei S, Coccia C, Pompei L, Di Marco P, Pietropaoli P. Continuous right ventricular end diastolic volume and right ventricular ejection fraction during liver transplantation: a multicenter study. Liver Transpl 2008;14:327–32
19. Siniscalchi A, Pavesi M, Piraccini E, De Pietri L, Braglia V, Di Benedetto F, Lauro A, Spedicato S, Dante A, Pinna AD, Faenza S. Right ventricular end-diastolic volume index as a predictor of preload status in patients with low right ventricular ejection fraction during orthotopic liver transplantation. Transplant Proc 2005;37:2541–3
20. Chin K, Kim NH, Rubin LJ. The right ventricle in pulmonary hypertension. Coron Artery Dis 2005;16:13–8
21. De Wolf AM, Begliomini B, Gasior TA, Kang Y, Pinsky MR. Right ventricular function during orthotopic liver transplantation. Anesth Analg 1993;67:562–8
22. Hoeper MM, Tongers J, Leppert A, Baus S, Maier R, Lotz J. Evaluation of right ventricular performance with a right ventricular ejection fraction thermo dilution catheter and MRI in patients with pulmonary hypertension. Chest 2001;120:502–7