| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2009;9:476-479. doi:10.1510/icvts.2009.202085 © 2009 European Association of Cardio-Thoracic Surgery
Donor pharmacological hemodynamic support is associated with primary graft failure in human heart transplantation
a Department of CT Surgery, ISMETT (Mediterranean Institute for Transplantation and Advanced Specialized Therapies), UPMC (University of Pittsburgh Medical Center), Via Tricomi 1, 90127 Palermo, Italy Received 5 January 2009; received in revised form 18 May 2009; accepted 12 June 2009
*Corresponding author. Tel.: +39-091-2192111; fax: +39-091-2192354.
The aim of this study was to test the impact of donor and recipient characteristics on the development of primary graft failure (PGF) after heart transplantation (HT) by focusing on the donor's inotropic support. Heart donors and matched recipients data were prospectively collected. Univariate and multivariate analyses were used to determine independent predictors for PGF and peri-operative mortality. The donor's high inotrope requirement was defined as sustained need for dopamine exceeding 10 µg/kg/min and/or alpha agonists exceeding 0.06 µg/kg/min. PGF instead was defined as need for immediate post-HT mechanical circulatory support. Since 2006, we have performed 37 HTs. PGF occurred in six patients (16.2%). Although four patients (66.6%) were weaned off circulatory support, two of them (33.3%) died on mechanical assistance. Total in-hospital mortality was 10.8% (4/37). Upon multivariate analysis, pre-harvesting donor high inotrope dosage was the major determinant for PGF (P=0.03, OR=10.8). Given the organ shortage, many centers accepted marginal hearts assuming the donor's pre-harvest hemodynamic managing has a reduced impact on PGF development. As PGF remains the most lethal postoperative complication, the hazards should be carefully considered when using pre-harvesting high inotrope infusion rates.
Key Words: Primary graft failure; Heart transplant
Heart transplantation (HT) represents the best option for the growing number of patients suffering end-stage heart failure. Due to the chronic donor shortage and increasing number of patients on the waiting lists, several transplant centers have started accepting organs with various imperfections (the so-called marginal organs). As Wittwer and Wahlers summarized [1], the concept of marginal donor has continuously varied from an elderly to an HCV-positive or coronary disease donor. As a consequence, several strategies have been suggested to ‘resuscitate' marginal hearts and reduce the impact of brain death over the donor's heart. In a similar context, there is still controversy as to which pre-harvesting donor characteristics independently impact on primary graft failure (PGF) occurrence, the most common cause of early mortality after HT. The aim of this study is to identify, by means of multivariable analyses, independent determinants for PGF after HT focusing particularly on the impact of donor pre-harvest drug management.
We retrospectively analyzed the population of isolated HT performed at our institution from January 2006 to October 2008. Donors and recipients variables were recorded. Donor high inotrope requirement was defined as sustained need for dopamine >10 µg/kg/min and/or alpha agonists (adrenaline and noradrenaline) >0.06 µg/kg/min. None of the donors included in this study had vasopressin infusion. PGF was defined by requirement for mechanical circulatory support (other than IABP) to wean off cardiopulmonary bypass immediately post HT. Hearts were retrieved with standard technique and preserved with infusion of 2 l of CelsiorTM solution at 4 °C.Bicaval surgical technique was adopted for all transplant procedures. In the event of PGF, mechanical support with extracorporeal membrane oxygenator (ECMO) (PLS®, Maquet, Rastatt, Germany) or CentriMag centrifugal pump (Levitronix® LLC, Waltham, MA) was enforced based on the failing side of the heart and the expected recovery time. Patients with and without PGF were compared (No-PGF; Yes-PGF). Wilk–Shapiro test was used to assess normality. Differences between the two groups were identified using the unpaired Student t-test, the Mann–Whitney test, and the Fisher exact test, as appropriate.Multivariable analysis by means of logistic regression was performed to identify independent determinants for PGF and peri-operative mortality occurrence. All statistical analyses were performed using SPSS (SPSS Inc, Chicago, IL, US).
A total of 37 isolated HTs were performed in our institution; 31 patients were male, and the mean age was 54.2± 9.2 years. All pre-harvest donor echocardiographies showed normal left ventricular contractility and anatomy. PGF occurred in six patients (16.2%). A significantly higher rate of recipients developing PGF had their hearts retrieved from donors undergoing high inotropes dosage [7 (22.6%) No-PGF patients vs. 4 (66.7%) PGF patients; P=0.03] (Table 1). The donor's cause of death was also heterogeneously distributed in the two groups. The majority of recipients that failed to develop PGF had donors deceased for head trauma [18 (58.1%) No-PGF patients vs. 0 PGF patients; P=0.009] (Table 1). Furthermore, the mean age of the donors dying from spontaneous intracranial bleeding was significantly higher than the age of the donors dying for other reasons (44.3±14.8 vs. 32.3±14.5, P=0.02).
No additional statistically-significant differences in donor and recipient variables were reported between the PGF and non-PGF groups (Tables 1 and 2).
Four of the six patients experiencing PGF were promptly assisted with artero–venous ECMO. One patient also required a 6-day support on biventricular Levitronix after being placed on ECMO. One patient underwent 17 days on right Levitronix. Four (66%) of the six PGF patients were weaned off the support and three (50%) survived to be discharged. The patient undergoing ECMO and Levitronix assistance was weaned off the support to die after one month from systemic aspergillosis. The two remaining PGF patients who were critically ill at the time of the transplant, died after three and seven days of mechanical support. Total in-hospital mortality was 10.8% (4/37). At logistic regression analysis, the donor pre-harvesting high inotropes dosage was the sole independent determinant for PGF (P=0.03, OR=10.8). Death from head trauma had a slight protective effect toward development of PGF (P=0.04; OR=–0.5). Furthermore, PGF was an independent determinant for mortality (P=0.02, OR=50) (Table 3).
Donor age failed to impact on PGF or mortality even when it was analyzed separately for each cause of death.
PGF after HT is a treacherous occurrence accounting for up to 40% of peri-operative deaths [2]. Although donor graft imperfections can theoretically enhance PGF occurrence, there are limited specific studies in literature to identify independent determinants for PGF. Homogeneous definition of PGF is also lacking, which results in a wide range of incidence and mortality. Lima et al. [3] have recently focused on the incidence and predictors of PGF. In their report, the authors investigate whether marginal cardiac allografts allocated to an alternate list of recipients, and not meeting standard criteria for HT, have an impact on the development of PGF. Interestingly, the incidence of PGF does not differ between alternate list and standard list of HT recipients. Pre-transplant ventricular assisting device and prolonged total ischemic time (>4.5 h) were the only independent predictors of PGF [3]. In the light of their findings, the authors support aggressive usage of marginal donor hearts in alternate list HT. Segovia et al. have recently proposed a risk scoring for PGF. Their logistic model identified four variables from the recipient (right atrial pressure >10 mmHg, age >60 years, inotrope dependence, and diabetes), one from the donor (age >30), and one from the procedure (ischemic time >240 min). The relative risks for all six factors were similar [4]. Differently from any other published data, our analysis also assessed the impact of donors' drug management on PGF. In our experience, pre-harvesting hemodynamic management can play a key role in PGF occurrence. Indications for inotropes' use in donors are quite controversial. In this regard, detailed cardiac recommendations were obtained from the ‘Maximizing Use of Organs Recovered from the Cadaver Donor’ conference [5]. As suggested, aggressive hemodynamic management should be achieved in the pre-harvest phases mainly by hormonal resuscitation, fluid adjustment, alpha agonists minimization, and administering dopamine and dobutamine at doses <10 µg/kg/min. In contrast, other authors suggest a more liberal use of inotropes in the pre-harvest phase is not associated with an increased risk of PGF [6]. In Silva's experience, a 4% incidence of PGF was reported among 27 recipients of HTs from donors managed with high doses of dopamine and/or noradrenaline [6]. Differently, Stoica and colleagues showed an impaired, load-independent, right ventricular function in donors treated with empirical noradrenaline infusions [7]. Although the policy of our institution is overall in line with international recommendations, as a result of donor shortage we have been recently prone to also retrieve hearts undergoing high inotrope rates. This more aggressive strategy resulted in a 16.2% PGF rate higher than the one reported in the International Society for Heart and Lung Transplantation (ISHLT) annual report [2]. Our multivariable analysis identified pre-harvesting high inotropes' dosage (dopamine/dobutamine >10 µg/kg/min and/or adrenaline/noradrenaline >6 µg/kg/min) as the strongest independent determinant for PGF with an odds ratio reaching 10 [8]. Although in the light of our findings we suggest caution in selecting hearts from donors undergoing high inotropes' dosages, we are, however, aware the response threshold to cathecolamines is subject to a wide variability mainly based on differences in inter-individual pharmacokinetics. For this reason, we believe further investigation on this topic is necessary. The role of donor's cause of death in determining transplant outcome is still controversial. ISHLT data show no connection between the donor's cause of death and the recipient's outcome [2]. In a recent multicenter study, Ganesh et al. showed an apparent relationship between donor's cause of death and post-transplant survival. This association was not maintained after adjustment for confounding variables [9]. In more specific analyses, the donor's cause of death was identified as a potential risk factor for early mortality after HT [8, 10, 11]. In particular, spontaneous intracranial bleeding could hinder a history of untreated hypertension in the donor [11]. In a similar scenario, the impact of pre-harvesting inotropes' use may be enhanced with detrimental effects on cardiac physiology due to catecholamine surge and intracranial pressure rise [12–14]. On the other hand, early experimental studies and clinical observations suggest that after the ‘catecholamine storm’ a reduction in circulating catecholamines takes place reaching levels lower than those recorded prior to brain death [15]. This condition may lead to reducing systemic vascular resistance and inotropism. In our analysis, hearts of donors deceased for head trauma had a reduced risk for PGF. In fact, of the 18 hearts harvested from patients with cranial trauma, none developed PGF. At multivariable analysis, death for head trauma also had a ‘protective’ effect on PGF (OR=–0.5). This concept could be rephrased by saying that death for intracranial bleeding is a risk factor for PGF, although in our specific analysis the donor intracranial bleeding was not an independent determinant for PGF. Also, our organ acceptance/rejection pattern shows how, in the same time frame of the present study, 45% of the 268 rejected hearts belonged to donors who died from spontaneous intracranial bleeding and 36% to donors who died from trauma. In the 37 accepted organs, the ratio is inverted with 38% of the hearts retrieved from donors affected by intracranial bleeding and 49% by lethal trauma. Donor age apparently has an independent impact upon recipient survival and PGF occurrence [2, 4]. In the light of previous findings, we specifically analyzed the impact of donor age considering the overall population of donors and the age in sub-groups divided by cause of death. Although, upon univariate analysis, the donors dying from spontaneous intracranial bleeding were significantly older, we were unable to define an age cut-off that could independently impact upon PGF and/or mortality. Adequate knowledge of PGF determinants should trigger development of appropriate treatment strategies. Once PGF occurs, we ensure prompt assistance with ECMO PLS or CentriMag. In spite of the advanced biocompatibility and excellent hemodynamic performance of these devices, PGF occurrence carries a heavy burden. In this regard, we experienced a mortality risk increased by 50 times. The deleterious complications associated with mechanical hemodynamic assistance are obviously emphasized by the patients' preoperative systemic conditions, as shown in our findings where the only two patients deceased while on support had an extremely complex preoperative status. Although, in our experience, post-PGF heart recovery occurred in 66% of the cases, we suggest cautiously allocating marginal organs considering the recipients' preoperative conditions and their chances to withstand the complication burden associated with hemodynamic mechanical assistance.
Although the limited number of patients included in the study could be interpreted as a major limitation, we were able to build a multivariable analysis model with good statistical significance. We were, however, unable to analyze the impact of the pre-op mechanical assisting device on PGF as only two of our patients were on long-term VAD. Finally, donors' echocardiographic parameters were not considered in the analysis as all pre-harvesting investigations had shown normal left ventricular contractility and anatomy.
In an era of donor shortage, correct selection of organs should be sustained by scientific evidence. Although modern and prompt circulatory mechanical support could lead us to accept the risk of marginal donors, specific distinctions should be made. A careful consideration of the hazards should be ensured when high inotrope infusion rates are used, especially in donors dying from causes other than head trauma. Furthermore, a thorough staging of the recipient preoperative condition should be performed to reasonably justify prolonged circulatory support should PGF ensue.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction