Patients with liver cirrhosis and concomitant coronary

Transcription

1 Outcomes of Simultaneous Liver Transplantation and Elective Cardiac Surgical Procedures Brian Lima, MD, Edward R. Nowicki, MD, Charles M. Miller, MD, Koji Hashimoto, MD, Nicholas G. Smedira, MD, and Gonzalo V. Gonzalez-Stawinski, MD Departments of Thoracic and Cardiovascular Surgery, Heart and Vascular Institute, and General Surgery, Cleveland Clinic, Cleveland, Ohio Background. Many centers are reticent to list patients for liver transplantation until coexistent cardiac disease is surgically corrected. Previous studies have documented considerable morbidity and mortality in liver failure patients undergoing cardiac operations. This study examined whether elective cardiac operations at the time of hepatic transplantation would yield enhanced outcomes. Methods. Between July 1999 and June 2010, 10 patients underwent simultaneous liver transplantation and elective cardiac operations at a single institution. Postoperative outcomes were analyzed using a prospectively maintained database. Results. The 10 patients were men (mean age, years): 7 were in Child-Pugh class B and 3 were in class C. Mean Model for End-Stage Liver Disease score was Cardiac operations included coronary artery bypass grafting in 1, aortic valve replacement in 4, coronary artery bypass grafting and aortic valve replacement in 3, coronary artery bypass grafting and mitral valve repair in 1, and tricuspid valve repair in 1. In-hospital mortality was 20%. Mean postoperative length of stay was 23 8 days. Actuarial survival at 3 years was 70%. Conclusions. Survival was modestly improved relative to that observed in previous studies of advanced liver failure patients undergoing heart operations without concomitant hepatic replacement. Moreover, the medium-term survival outcomes approach those documented with liver transplant alone. Further studies are warranted with this combined surgical strategy to determine if such an approach would be routinely preferable to staged repair of cardiac pathology and liver transplant. (Ann Thorac Surg 2011;92:1580 5) 2011 by The Society of Thoracic Surgeons Patients with liver cirrhosis and concomitant coronary or valvular heart disease, or both, present a challenging clinical dilemma. Liver transplantation centers often mandate that such cardiac lesions must be surgically treated before listing for transplantation can be entertained. However, the published experience with cardiac surgical procedures in cirrhotic patients has been fraught with high rates of morbidity and mortality, notably in patients with more advanced disease states (Child- Pugh class B or C) [1 5]. In light of these poor outcomes, many cardiac surgeons have become reluctant to offer elective operations to this high-risk patient population [6], thereby diminishing the likelihood of eventual liver transplant and opportunity for meaningful survival. A theoretically plausible treatment strategy for these liver transplant candidates could entail simultaneous cardiac operation and hepatic transplantation. Little data exist on the viability of this approach, with only isolated case reports having been published to date [7 11]. The purpose of the present study was to delineate outcomes of combined elective cardiac operations and liver transplantation in a series of 10 patients. We hypothesized that this combined approach would yield readily apparent Accepted for publication June 14, Address correspondence to Dr Gonzalez-Stawinski, Cleveland Clinic, 9500 Euclid Ave, Desk J4-1, Cleveland, OH 44195; gonzalg@ ccf.org. reductions in morbidity and mortality compared with that commonly encountered with cardiac operations in patients with cirrhosis. Patients and Methods A retrospective analysis of 10 consecutive patients undergoing simultaneous orthotopic liver transplantation and elective cardiac surgical procedures between July 1999 and June 2010 at the Cleveland Clinic was performed. Data are prospectively collected into the Cardiovascular Information Registry, which inclusive with the current study, is approved by the Institutional Review Board of the Cleveland Clinic, and the need for informed consent was waived for this study. Requisite criteria for eligibility for this combined surgical strategy included preserved left ventricular ejection fraction and relatively straightforward cardiac pathology not requiring an extensive duration of cardiopulmonary bypass (CPB), such as for redo sternotomy for cardiac reoperation, or any level of hypothermic circulatory arrest as with atheromatous ascending aorta or other aortic arch pathology. The Child-Pugh classification was defined by the score calculated from the degree of hepatic encephalopathy (grade 1 to 3), ascites (absent, mild, moderate), total bilirubin (mg/dl) and serum albumin (g/dl) levels, and prothrombin time (seconds). The resultant score yielded a classification of A (5 to 6), B (7 to 2011 by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg LIMA ET AL 2011;92: CONCOMITANT CARDIAC OPERATIONS AND LTX 9), or C (10 to 15). The preoperative Model for End-Stage Liver Disease score [12] was also calculated for each patient. The donors were a mean age of 42 years (range, 18 to 58 years). The mean cold ischemia time was hours (range, 6.1 to 10.6 hours). Three recipients had portal vein thrombosis at the time of transplantation. All recipients received blood type identical whole-liver allografts. Two liver allografts were preserved with University of Wisconsin solution, and 8 were preserved with histidinetryptophan-ketoglutarate solution. In April 2004 we started using induction therapy with basiliximab, an interleukin-2 receptor blocker, on postoperative days 0 and 4, after administration of 1,000 mg of methylprednisolone during the anhepatic phase. Steroids were rapidly tapered and discontinued on postoperative day 22, except for patients with primary sclerosing cholangitis. Tacrolimus was administered to all recipients as the calcineurin inhibitor of choice. Tacrolimus trough levels were maintained between 8 and 12 ng/ml for the first 3 months and then between 6 and 8 ng/ml thereafter. Cyclosporine was the second choice when recipients had serious side effects from tacrolimus. Mycophenolate mofetil was added to delay tacrolimus administration when recipients had renal dysfunction and used as an adjunct to lessen the incidence and severity of calcineurin inhibitor toxicities. Sirolimus was used to prevent hepatocellular carcinoma recurrence or when patients had mycophenolate mofetil toxicities. Surgical Procedures For each patient, the elective cardiac surgical procedure preceded the liver transplantation. All cardiac surgical procedures were performed through a full median sternotomy on CPB with the heart arrested. Full systemic heparinization was instituted in each case, with target activated clotting time exceeding 450 seconds. Protamine was used to for heparin reversal when the procedure was finished. The chest was left open until the liver transplant 1581 was concluded. Mediastinal and pleural drains were placed, and sternal reapproximation was achieved with stainless steel wires. Liver transplantation was performed in the standard fashion [13] with conventional caval interposition technique in 4 patients and with venovenous bypass or the piggyback technique in 6, without bypass. No recipient required an interposition graft for portal vein reconstruction. Biliary anastomosis was performed in 9 patients using an end-to-end choledochocholedochostomy and in 1 patient with an end-to-side choledochojejunostomy with Roux-en-Y. Patient Data and Analysis The present report includes all data collected through the most recent follow-up visit. Deaths from all causes were included in the analysis. In addition, the Social Security Death Index was queried ( to confirm all patient deaths. Major postoperative (inhospital) complications were catalogued and included stroke, renal failure requiring renal replacement therapy, sepsis, and liver allograft rejection. Patient characteristics are reported as percentages for discrete variables. Means and standard deviations are provided for continuous variables. For comparisons, the Wilcoxon rank sum test was used for continuous variables, and the 2 was used for categoric variables, with an alternative hypothesis that the rates between groups were not different. Long-term survival was computed using Kaplan-Meier survival analysis. All statistical analyses were computed using SAS software (SAS Inc, Cary, NC). Results Patient Population All 10 patients were men (Table 1). Mean age was 59 9 years (range, 44 to 72 years). The mean preoperative left ventricular ejection fraction was Five patients ADULT CARDIAC Table 1. Preoperative Patient Characteristics Pt a Age (yrs) Cardiac Diagnosis LVEF Cause of Liver Cirrhosis Child-Pugh Class MELD Score 1 57 AS 0.55 EtOH B AS, CAD, PE 0.75 EtOH C AI 0.55 Viral C CAD 0.60 NASH C CAD, AS 0.55 EtOH B CAD, MR 0.55 Viral, HCC B AS 0.65 Viral, HCC B CAD, AS 0.60 HCC B AS 0.65 Viral B TR 0.60 PSC B 13 a All patients were men. AI aortic insufficiency; AS aortic stenosis; CAD coronary artery disease; EtOH alcoholic cirrhosis; HCC hepatocellular carcinoma; LVEF left ventricular ejection fraction; MELD Model for End-Stage Liver Disease; MR mitral regurgitation; NASH nonalcoholic steatohepatitis; PE pulmonary embolism; PSC primary sclerosing cholangitis; Pt patient; TR tricuspid regurgitation.

3 1582 LIMA ET AL Ann Thorac Surg CONCOMITANT CARDIAC OPERATIONS AND LTX 2011;92: (50%) had severe coronary artery disease and met the operative indications for coronary artery bypass grafting (CABG). Six patients (60%) presented with severe aortic stenosis. Other cardiac diagnoses necessitating surgical intervention included pulmonary embolism (noted intraoperatively) in 1 patient, aortic valve insufficiency, and mitral and tricuspid valve regurgitation. Etiologies of liver failure included alcoholic cirrhosis (30%), hepatitis C (40%), hepatocellular carcinoma (30%), nonalcoholic steatohepatitis (10%), and primary sclerosing cholangitis (10%). Of the 10 patients 7 were Child-Pugh class B, and the remaining 3 were class C. The mean Model for End-Stage Liver Disease score was Cardiac Procedures Coronary artery bypass grafting was done in 5 patients (50%) with 1 (n 2),2(n 2), or 3 (n 1) bypass grafts. For each of the 7 aortic valve replacements, a bioprosthetic was implanted to avoid the anticoagulation necessary for a mechanical prosthesis. Mitral valve repair was performed in patient 6, who required concomitant CABG. Patient 10 required tricuspid valve repair for severe regurgitation. Patient 2 experienced intraoperative hemodynamic instability during the liver transplantation, prompting transesophageal echocardiographic evaluation and subsequent discovery of large pulmonary embolus. CPB was reinstituted and a formal pulmonary embolectomy was performed. Postoperative Complications No patients experienced postoperative in-hospital episodes of liver allograft rejection (Table 2). Two patients (1 Child-Pugh class B, 1 class C) required surgical reexploration for hemorrhage in the mediastinum (n 1) or abdomen (n 1). Stroke, as confirmed by formal neurologic assessment and radiographic diagnostics, occurred in 1 patient with pretransplant Child-Pugh class C cirrhosis. Renal dysfunction necessitating hemodialysis was observed in 3 patients (1 Child-Pugh class B, 2 class C). Two in-hospital deaths occurred, both in Child-Pugh class C patients. The first death occurred in patient 2, who required an emergency embolectomy for an intraoperative pulmonary embolus. He subsequently required exploratory laparotomy with small-bowel resection secondary to smallbowel infarction. His postoperative course was further complicated by stroke and renal failure. The second postoperative death occurred in patient 3, who was at Child-Pugh class C and underwent aortic valve replacement. He also required reexploration for hemorrhage and repair of patent foramen ovale. Acuteon-chronic pulmonary hypertension developed that was refractory to inhaled nitric oxide and sildenafil. Central catheter-associated sepsis also developed, along with renal failure that required dialysis. Cardiac arrest occurred on postoperative day 12, and he could not be resuscitated. Follow-up was 100% complete for all 10 patients. Actuarial survival was 70% at 3 years (Fig. 1). Two deaths occurred after hospital discharge. Patient 4 died secondary to complications of septic shock at 97 days. Patient 1 died of complications after a traumatic fall down a flight of stairs at his home nearly 4 years after the operation. Comment This study reports a large descriptive analysis of outcomes after combined elective cardiac operations with liver transplantation. All procedures were performed with CPB. The 30-day and in-hospital mortality was 20%, with both deaths occurring in 2 of the 3 Child-Pugh class C patients (67%) and no deaths occurring in the remaining 7 class B patients (0%). Major postoperative complications, such as stroke, renal failure, hemorrhage, and sepsis, occurred in 40% patients overall (25% Child-Pugh class B and 67% of C). The addition of a cardiac operation to liver transplantation also did not appear to have an appreciable negative effect on long-term survival, with a 3-year actuarial survival rate (70%) closely approximating that of modern series of liver transplant alone [14]. These outcomes also compare very favorably in relation to the reported experience of cardiac operations in patients with cirrhosis not undergoing concomitant transplantation. A study from our institution examined outcomes of 44 cirrhotic patients undergoing elective cardiac operations within the same interval as the present analysis [15]. Of these, 31 had Child-Pugh class A cirrhosis, with only 12 class B patients (27%) and 1 class C patient (2%). Mortality was reported as 3.2%, 42%, and 100% for the Child-Pugh class A, B, and C patients, respectively. Hepatic decompensation was also an important prognosticator and occurred in 27% patients. In a Cornell study of cardiac operations in 13 Child- Pugh class B cirrhotic patients, perioperative morbidity was 100% and mortality was 80% [4]. Hayashida and colleagues [3], in their series of 18 patients, also reported a 100% mortality rate among Child-Pugh class B patients Table 2. Early (In-Hospital) Postoperative Complications Complication a Child-Pugh Stroke Renal Failure b Hemorrhage c Sepsis Death Class No. (%) No. (%) No. (%) No. (%) No. (%) B 0 (0) 1 (14) 1 (14) 0 (0) 0 (0) C 1 (33) 2 (67) 1 (33) 1 (33) 2 (67) Total 1 (10) 3 (30) 2 (20) 1 (10) 2 (20) a Rejection was not among the documented complications. b Renal failure requiring dialysis. c Hemorrhage requiring operative reexploration.

4 Ann Thorac Surg LIMA ET AL 2011;92: CONCOMITANT CARDIAC OPERATIONS AND LTX Fig 1. Survival is shown after cardiac operation combined with hepatic transplantation. when CPB was used (50% mortality without CPB). In that study, the preponderance of perioperative morbidity and death coincided with the implementation of CPB. As summarized in a subsequent review by the same authors, this trend is recapitulated in other published series [2], compelling them to deem Child-Pugh B or C cirrhosis as contraindications to cardiac interventions using CPB. Interestingly, that review also highlights that the incidence of postoperative morbidity remains considerable even among patients with less severe hepatic dysfunction (Child-Pugh A), ranging from 25% to 100%. One of the most recent and comprehensive experience was published by Filsoufi and colleagues [1] from Mount Sinai and included 27 patients (10 Child-Pugh class A, 11 class B, and 6 class C) undergoing cardiac operations with and without CPB. Overall hospital mortality was 26% (10% class A, 18% class B, 67% class C). Major postoperative complications occurred in 20%, 56%, and 100% of class A, B, and C patients, respectively. Blood product requirement for patients with or without CPB did not reach statistical significance. Compared with patients in a contemporaneous control group undergoing cardiac operations alone, long-term survival was significantly compromised in these patients, with 1-year survival rates of 80% for Child-Pugh class A, 45% for class B, and 16% for class C. This latter observation underscores the ominous prognostic implications of persistent hepatic dysfunction on long-term survival after a successful cardiac operation. The decision to offer cardiac surgical procedures with liver transplant to a patient with cirrhosis is largely determined by the degree of hepatic dysfunction and the severity of the cardiac lesion. Patients with most forms of cardiac pathology who are at Child-Pugh class A are considered for cardiac operation because most patients in this class have satisfactory short-term and long-term outcomes after cardiac operations [15]. For patients with 1583 Child-Pugh class B, a cardiac operation alone is plausible in those with early stages of hepatic dysfunction who need simple cardiac intervention (eg, CABG, simple aortic valve replacement). Child-Pugh class C patients are universally offered medical management, given the prohibitive risk of cardiac operations in this population. Generally, part of the evaluation process for all patients considered for concomitant cardiac operation and liver transplant aims to determine two critical elements: (1) the level of complexity of the cardiac surgical procedure, and (2) the cardiovascular risk, short-term and long-term, of the cardiac pathology itself. Work-up is tailored to identify stress-induced myocardial dysfunction or stress-induced pulmonary congestion or hypertension and reversibility of the pulmonary hypertension and identification of portopulmonary hypertension. The simplest scenario is severe left main or triple-vessel disease, or coronary artery disease with large ischemic areas and myocardium at risk. Here, a combined surgical approach of CABG and hepatic transplantation is used. Patients with degenerative mitral valve disease, normal pulmonary artery pressures, and normal exercise or dobutamine stress test result can safely undergo isolated liver transplant. When pulmonary hypertension is present, whether is it is primary to secondary to valve pathology is determined by the hemodynamic assessment. If the pulmonary hypertension is fixed, then the patient has portopulmonary hypertension. We acknowledge the limitations of the present study, conceding that the observed results are promising but merit further investigation in larger, prospective series. The feasibility of such investigations would be predicated on, as in this study, a concerted, multidisciplinary, and cooperative approach among cardiac surgeons, cardiologists, hepatologists, and hepatobiliary surgeons. These studies would also potentially enable statistical stratification of outcomes by the Model for End-Stage Liver Disease score, which was not feasible in this study. Critics may also cite that these patients were a highly select group with normal ventricular function and relatively straightforward cardiac pathology. Although this is true, the high-risk nature of such combined procedures may only be foreseeably achievable in such select patients. What the present study does provide is proof of principle that by eliminating hepatic dysfunction from the outcome equation, patients with liver failure, namely those with Child-Pugh class B, may derive the greatest benefit, even in cases where CPB must be instituted. On the basis of these results, we would also contend that liver failure patients with cardiac disease should not be automatically disqualified from consideration for liver transplantation. Unfortunately, this combined surgical strategy did little to affect the uniformly dismal outcomes of cardiac operations in Child-Pugh class C cirrhotic patients. Perhaps clinical efforts in these specific patients should center on medical optimization to approach class B status. In conclusion, appropriate management of patients with cirrhosis and cardiac disease requires thoughtful consideration of the hepatic disease burden, candidacy for liver transplantation, and cardiac procedural risk ADULT CARDIAC

5 1584 LIMA ET AL Ann Thorac Surg CONCOMITANT CARDIAC OPERATIONS AND LTX 2011;92: assessment determined by ventricular function, surgical complexity, and other patient comorbidities. Simultaneous cardiac operation and liver transplantation can be safely performed in select Child-Pugh class B patients with normal ventricular function, no history of cardiac operations, and with a straightforward cardiac pathology. References 1. Filsoufi F, Salzberg SP, Rahmanian PB, et al. Early and late outcome of cardiac surgery in patients with liver cirrhosis. Liver Transpl 2007;13: Hayashida N, Aoyagi S. Cardiac operations in cirrhotic patients. Ann Thorac Cardiovasc Surg 2004;10: Hayashida N, Shoujima T, Teshima H, et al. Clinical outcome after cardiac operations in patients with cirrhosis. Ann Thorac Surg 2004;77: Klemperer JD, Ko W, Krieger KH, et al. Cardiac operations in patients with cirrhosis. Ann Thorac Surg 1998;65: Lin CH, Lin FY, Wang SS, Yu HY, Hsu RB. Cardiac surgery in patients with liver cirrhosis. Ann Thorac Surg 2005;79: Modi A, Vohra HA, Barlow CW. Do patients with liver cirrhosis undergoing cardiac surgery have acceptable outcomes? Interact Cardiovasc Thorac Surg 2010;11: Benedetti E, Massad MG, Chami Y, Wiley T, Layden TJ. Is the presence of surgically treatable coronary artery disease a contraindication to liver transplantation? Clin Transplant 1999;13: Eckhoff DE, Frenette L, Sellers MT, et al. Combined cardiac surgery and liver transplantation. Liver Transpl 2001;7: Massad MG, Benedetti E, Pollak R, et al. Combined coronary bypass and liver transplantation: technical considerations. Ann Thorac Surg 1998;65: Morris JJ, Hellman CL, Gawey BJ, et al. Case Three patients requiring both coronary artery bypass surgery and orthotopic liver transplantation. J Cardiothorac Vasc Anesth 1995;9: Parker BM, Mayes JT, Henderson JM, Savage RM. Combined aortic valve replacement and orthotopic liver transplantation. J Cardiothorac Vasc Anesth 2001;15: Kamath PS, Wiesner RH, Malinchoc M, et al. A model to predict survival in patients with end-stage liver disease. Hepatology 2001;33: Llado L, Figueras J. Techniques of orthotopic liver transplantation. HPB (Oxford) 2004;6: Lipshutz GS, Hiatt J, Ghobrial RM, et al. Outcome of liver transplantation in septuagenarians: a single-center experience. Arch Surg 2007;142:775 81; discussion Suman A, Barnes DS, Zein NN, Levinthal GN, Connor JT, Carey WD. Predicting outcome after cardiac surgery in patients with cirrhosis: a comparison of Child-Pugh and MELD scores. Clin Gastroenterol Hepatol 2004;2: INVITED COMMENTARY The morbidity and mortality associated with liver transplantation is significantly increased in patients with cardiac disease, and has especially been the case with coronary occlusive disease. As a result, the presence of untreated cardiac pathology has traditionally been considered a contraindication to liver transplantation. Unfortunately, previous studies have reported that in the presence of cirrhosis, attempts at correction of cardiovascular disease with cardiac surgery similarly carry a high risk of complications or death. Lima and colleagues [1] set out to explore whether the combination of elective cardiac surgical operations at the time of liver transplantation would result in improvement of outcomes. Although several previous reports of simultaneous cardiac operations and liver transplantation have been published, this represents the largest series to date. There are several promising findings in this study. When compared with the literature, the outcomes for morbidity and mortality compare favorably with those of patients with cirrhosis undergoing cardiac procedures without concomitant hepatic replacement. This seems to provide, as the authors state, the proof of concept that with the elimination of hepatic dysfunction as a postoperative complication, the surgical outcomes would be improved. Also, the addition of the cardiac surgical procedure did not appear to negatively affect acute or long-term hepatic allograft function. In addition, the presence of cardiac disease did not seem to alter the medium-term survival with respect to the occurrence of adverse cardiac events. As a result, it would seem plausible that many liver transplant programs might reconsider their stance and therapeutic strategies with regard to the presence of cardiac disease in potential recipients. Unfortunately, questions remain in this difficult patient population. As the authors freely admit, this was a highly selected patient group whose requisite variables for eligibility were stringent. Patients with reduced ventricular function and more advanced cardiac pathology were excluded and certainly might have altered the results. The reality is that contemporary liver transplant recipients are presenting with escalating risk profiles. In addition to valvular and coronary artery disease, patients with end-stage liver disease may have an associated cirrhotic cardiomyopathy and portopulmonary hypertension. Neither is a surgical entity, and both may respond only partially to medical management. When severe, they can be associated with high mortality rates postoperatively. Currently, there are no specific recommendations for the cardiovascular assessment of a potential liver transplant recipient. It seems that a thoughtful preoperative evaluation, as the authors describe, is paramount. Also highlighted in the manuscript, and that which is consistent with current data, is that the Childs class C patients continue to have dismal results regardless of the management strategy. The best predictive scale among the many available, including mean end-stage liver disease, criteria, posthepatectomy liver failure, nutritional index, and Child-Turcotte-Pugh, remains unclear. What is clear is that advanced end-stage liver disease is a systemic problem, and even with eliminating liver failure as a postoperative complication, the morbidity and mortality remain high by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc doi: /j.athoracsur