Perioperative Beta-blocker and Statin use in Patients with Coronary Artery Disease Undergoing Orthotopic Liver Transplantation

Perioperative Beta-blocker and Statin use in Patients with Coronary Artery Disease Undergoing Orthotopic Liver Transplantation Stephen J Bagley1, Meredith Grzybowski2, Andrew J Litwack3 and Scott M. Lilly4* 1Department of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA 2Penn Transplant Institute, University of Pennsylvania, Philadelphia, PA 19104, USA 3Department of Medicine, Cardiovascular Division, University of Pennsylvania, Philadelphia, PA 19104, USA 4Division of Cardiovascular Medicine, Heart and Vascular Center, Ohio State University, Columbus, OH 43210, USA Journal of Clinical Trials in Cardiology Open Access Research Article


Introduction
Patients with End-Stage Liver Disease (ESLD) have a prevalence of Coronary Artery Disease (CAD) up to 27% , exceeding that of the general population [1][2][3][4].Moreover, patients with CAD undergoing Orthotopic Liver Transplantation (OLT) experience worse short and long-term outcomes compared to those without CAD [5][6][7].While these prognostic implications may limit transplant eligibility, OLT is still frequently offered to patients with CAD [8].
There are few guidelines that address the perioperative management of cardiovascular risk in this specific population [9,10].Established guidelines do exist for the perioperative management of patients with known CAD undergoing intermediate or high-risk noncardiac surgery.These include the continuation of BB and statin therapy throughout the perioperative period in an effort to reduce cardiovascular events [11].However, the adherence and utility of this practice is unknown in the OLT population, as BB therapy may be initiated primarily due to concern for variceal bleeding and statin therapy may be withheld due to concern for hepatotoxicity.
The present study was undertaken to describe patterns of beta-blocker and statin use among patients with CAD undergoing OLT at a tertiary care hospital.In addition, we sought to evaluate

Methods
After institutional review board approval and the granting of a waiver of informed consent, we searched the institutional transplant database for patients who underwent OLT at the Hospital of the University of Pennsylvania between 2002 and 2010.This search produced a list of 1,075 patients, which was then cross-referenced to a list of all inpatients and outpatients in the University of Pennsylvania Health System who were billed for an ICD-9 diagnostic code related to CAD between 1998 and 2010.These codes included: "acute myocardial infarction", "other acute and subacute forms of ischemic heart disease", "old myocardial infarction", "angina pectoris", "other forms of chronic ischemic heart disease", "aortocoronary bypass status", and "percutaneous transluminal coronary angioplasty status".This cross-reference yielded 251 patients who had undergone OLT between 2002 and 2010 and had been billed for a diagnosis or related diagnosis of CAD (either before or after OLT).Each of these 251 charts was then manually evaluated to corroborate the existence of known CAD, as defined in Table 1, at the time of the patient's OLT.A total of 47 patients met the criteria for analysis.Two of these patients were excluded because they underwent simultaneous liver and heart transplants, leaving 45 patients included in the study.
These 45 patients additionally had their medical records reviewed for the use of a beta blocker or statin 1) during the 30 days prior to hospitalization for OLT; 2) at the time of admission for OLT; 3) during hospitalization for OLT; 4) at discharge from OLT hospitalization; and 5) following discharge up to 30 days posttransplant.Ordinal variables were created for BB use, including "uninterrupted therapy", "interrupted therapy" (defined in Table 1), "no therapy", and " at hospital discharge", and for statin use, including "preoperative" and "at hospital discharge".Chisquare, Fisher's exact test, and analysis of variance (ANOVA) were employed to compare BB and statin use with the following outcomes: 1) myocardial infarction (MI, as defined in Table 1) at 30 days and 12 months post-OLT, 2) mortality at 30 days and 12 months, 3) composite death and myocardial infarction at 30 days and 12 months, and 4) length of hospital stay (LOS).

Results
Baseline characteristics are listed in Table 2.The cohort predominantly comprised of Caucasian males (91%), with a mean age of 61 years.The most common etiologies for ESLD were viral hepatitis (33%) and non alcoholic steatohepatitis (33%).Thirtysix percent of the patients had a history of coronary artery bypass grafting (CABG), 33% had a history of percutaneous coronary intervention (PCI), 24% had been managed medically with an obstructive lesion, ≥50% on coronary angiography, and the remaining 7% had been managed medically with a documented history of myocardial infarction.No patients were allergic to beta-blockers.One patient had charted allergies to atorvastatin and rosuvastatin.
The rates of BB and statin use are listed in Table 3. Thirtythree patients in our cohort (73%) were on a BB at admission,

Coronary artery disease
Any one or more of the following: 1) history of ≥ 1 documented MI, 2) history of coronary artery bypass grafting (CABG), 3) history of percutaneous coronary intervention (coronary artery angioplasty with or without stenting), 4) coronary angiography demonstrating ≥ 1 significant lesion causing ≥ 50% stenosis

Myocardial infarction
Detection of rise in serum cardiac biomarkers (troponin T) with any of the following evidence of ischemia: 1) chest pain, 2) ECG ST-T wave changes indicative of ischemia, 3) development of pathologic Q waves on ECG, 4) imaging evidence of new loss of viable myocardium or new regional wall motion abnormality 23 and out of these 20 (61%) had BB therapy withheld for > 48 hours during their hospitalization.Twenty-seven (60%) patients were discharged on a BB.Among the patients who were administered a BB postoperatively, the mean time between OLT and start of beta-blockade was 2.4 ± 4.9 days.Eight patients (18%) were on statin therapy chronically and had the drug continued to the time of OLT.Following transplantation, 1 patient (2%) was discharged on a statin.
Tables 4 and 5 display the outcomes of myocardial infarction, death, and hospital length of stay following OLT.In our cohort, 5 patients (11%) experienced an in-hospital MI.Of these, 4 occurred in patients that had BB therapy withdrawn (Figure 1), although this was not a statistically significant difference as compared to patients in whom BB therapy was uninterrupted (p = 0.118).Of the 5 patients with in-hospital MI, one was on statin therapy.In total, there were 5 deaths (11%) at 12 months posttransplant, 3 on BB at discharge and none on statin therapy at discharge.Composite 1-year mortality and MI was 22% and did not differ with regard to BB (provided 22%; not provided 22%; p = 0.637) or statin prescription (provided 0%; not provided 23%; p = 0.601) at the time of discharge (Figure 2).
Lastly, univariate analysis revealed that mean LOS differed with respect to BB use and was longer in those that had interrupted (20.3 ± 13.9 days) compared with uninterrupted BB therapy (11.5 ± 7.1; p = 0.020) (Figure 3).LOS did not vary between those provided preoperative statins (13.6 days) compared to those that were not (17.0days; p = 0.422).

Discussion
The significant prevalence of CAD in the ESLD population has recently been recognized, and the presence of CAD has been demonstrated to predict poor outcomes in patients undergoing OLT.Three-year mortality rates in this population have been reported as high as 50%, with half of the deaths occurring in the first 35 days after transplant [5].In addition, cohort studies by Deidrich et al. [6] and Yong et al. [7]

LOS (days) BB therapy
Values expressed as mean ± standard deviation (13%), and composite mortality/MI (22%) rates, similar to those identified in the aforementioned cohorts.

LOS vs. perioperative BB use
Management of perioperative cardiovascular risk in this population might include the provision of BB and statin drugs -the continuation of which is recommended in an effort to reduce cardiovascular events and mortality in those with CAD undergoing high risk surgeries [11][12][13][14][15][16].We identified a high rate of perioperative BB withdraw-for which there are several potential explanations.At baseline, patients with cirrhosis are peripherally vasodilated and have low systemic vascular resistance, [17] and hypotension is often more pronounced around the time of transplant due to complicating occurrences (e.g.hemorrhage, sepsis), or the sudden hemodynamic changes that accompany liver transplantation [18].An alternative explanation for frequent perioperative BB withdrawal is that many OLT candidates might have been initiated on chronic nonselective BB therapy for variceal bleeding prophylaxis, rather than for cardiac indications.Since acquisition of a healthy liver mitigates the risks of variceal bleeding, patients on nonselective beta-blockers prior to transplant may have these drugs discontinued perioperatively despite having CAD as an indication for continuation.Importantly, we observed a trend toward increased 30-day MI and a statistically significant increase in hospital LOS among those with BB withdraw.In a retrospective study of 403 patients (with and without CAD) undergoing OLT, BB use was associated with reduced 30-day mortality and reduced 30-day death and myocardial infarction [8].Whether or not the LOS or trend toward increased MI in the present investigation are related to BB withdraw, or representative of interceding co-morbidities is unclear but warrants further investigation in a prospective fashion.
Regarding statin therapy, few patients were on a statin at the time of OLT, most were not administered a statin postoperatively, and there was no relationship between statin therapy and outcomes.We speculate that common concerns regarding statin use in the OLT community may include 1) statin-associated transaminitis following OLT complicating the detection of acute rejection, and 2) fear of potential hepatotoxicity associated with statin use.Results from studies in patients with elevated liver enzyme levels, non alcoholic fatty liver disease, hepatitis C, and cirrhosis have shown benefit of statin therapy without increased risk of adverse effects [19].In addition, between 16% and 43% of liver transplant recipients have increased plasma cholesterol levels (in part due to their immunosuppressive regimens), [20,21] and achievement of therapeutic goals among OLT patients is attainable with a low incidence of adverse events [22].Thus, while the theoretical risk of confusing a statin-associated rise in liver enzymes with acute rejection remains a concern in patients with CAD undergoing OLT, prospective trials are needed to weigh the risk/benefit of statin administration in this setting.
To our knowledge, the present study represents the largest reported cohort of patients with CAD who underwent OLT, and the only one that describes patterns of BB and statin use in this population, along with associated outcomes.However, the present study has limitations: it is retrospective and single center.Moreover, although significant with respect to patterns of statin use, the number of patients on statin therapy in this population was low, precluding meaningful conclusions regarding outcomes.
Whether perioperative beta-blockers and statins improve outcomes in patients with CAD undergoing liver transplantation is unknown.Our study suggests a protective effect of continuing BB therapy perioperatively in those already taking a BB, but our results may be confounded by the medical complications of transplant that might necessitate interruption of BB therapy.Nonetheless, our observations call for identifying reasons for BB withdrawal and provide support for increased vigilance with respect to BB provision in this population.Prospective trials are needed to elucidate the impact of both beta-blockers and statins on CAD patients undergoing OLT.If these therapies are shown to be advantageous, then patients who have historically been excluded from consideration may eventually benefit from this lifesaving surgery.

Figure 2 :
Figure 2: Rate of 12-month MI/death vs. BB prescription at discharge.

Table 2 :
Baseline Demographic and Clinical Features.

Table 3 :
have established that this population fares reliably worse than those without known CAD.The present study identified high 1 year-mortality (11%), MI Peri-operative Beta-blocker and Statin use.

Table 4 :
30 days Outcomes and Peri-operative BB and Statin Therapy.

Table 5 :
12 months Outcomes and Peri-operative BB and Statin Therapy.All values are expressed as number (%).

therapy Rate of 30-day MI vs. perioperative BB therapy Figure 1:
Rate of 30-day MI vs. perioperative BB therapy.