The quality of harvested organs is an extremely important factor for the success of transplantation . Most transplanted organs are recovered from brain-dead organ donors. In these donors, a decline of central regulatory mechanisms leads to a disturbance of homoeostasis, which can cause deterioration of the organs to be harvested . Beside electrolyte imbalances, hypothermia, diabetes insipidus and vasoplegia, endogenous catecholamines might induce myocardial or hepatic necrosis . In a previous study, we were able to show that during organ harvesting, painful stimuli can elicit massive release of endogenous catecholamines . In this study, our hypothesis was that blockade of propagation of painful stimuli with opioids would prevent catecholamine release. Thus, we performed a randomized, controlled and double-blinded trial in brain-dead donors, administering either fentanyl or placebo at the beginning of the retrieval surgery, and compared catecholamine blood concentrations and haemodynamics between the groups.
Our Institutional Ethics Committee approved this study. Seventeen brain-dead organ donors were studied. All were declared brain dead according to a protocol for brain death declaration prescribed by Austrian law, including loss of brain-stem reflexes, negative blood test for narcotics and sedatives and either two zero-line electroencephalogram for 20 min in a 6 h interval, or four vessel angiography. Patient characteristics data are shown in Table 1. As sternotomy was a fixed point of measurement, only donors scheduled for multi-organ donation were accepted into the study. Exclusion criteria were: haemodynamic instability with the need for exogenous catecholamines throughout the observation period, and age below 18 yr. After transfer to the operating room, donors were monitored by invasive measurement of blood pressure, central venous pressure (CVP), electrocardiogram and body temperature. The amount of intravenous (i.v.) fluid given was recorded.
Randomization was first achieved by a computer-generated list (Microsoft Excel® v.4.0). Then the anaesthetist received a syringe, containing either fentanyl 7 μg kg−1 (n = 10) or an identical volume of sodium chloride 0.9% (n = 7), the contents of which were injected i.v. before the start of surgery. The duration of the period between administration of fentanyl and skin incision was 10-12 min. Haemodynamic variables and blood samples were obtained before injection of the bolus, 10 min thereafter, 5 min following skin incision and 5 min following sternotomy.
Arterial blood was sampled into 7 mL ethylenediaminetetraacetic acid (EDTA) tubes, stored immediately on ice and centrifuged at −20°C. Plasma was separated and stored at −20°C until processing. Measurement of serum epinephrine and norepinephrine concentrations was made using a commercially available radioimmunoassay (KatCombi 'high-sensitive'®; IBL Gesellschaft für Immunchemie und Immunbiologie GmbH, Hamburg, Germany). The lower detection level for epinephrine was 3 and 10 pg mL−1 for norepinephrine.
A total of 30 cases were planned with an interim analysis after 15 cases. Comparisons between groups were obtained by a U-test, since multivariate or parametric analysis could not be performed due to violations of the normality assumption. Testing for correlation between catecholamine concentrations and haemodynamic variables was performed by Spearman's rank correlation. P < 0.05 was regarded as significant.
Of the 30 cases planned, only 17 were included as an evaluation of the data after 15 cases (by a statistician not otherwise involved in the protocol) already clearly demonstrated the failure of the treatment.
No differences between the groups regarding patient characteristics, body temperature or i.v. fluids were apparent (Table 1). Averaged (median; 95% CI) haemodynamic variables and catecholamine concentrations are shown in Figure 1. Single values with percent change relative to the preceding value are shown in Table 2 for the fentanyl group and in Table 3 for the placebo group.
Administration of fentanyl did not result in a uniform reaction, but with a great variation of increases and decreases of epinephrine and norepinephrine concentrations. Skin incision led to an increase of epinephrine concentration in all cases, while at sternotomy there was again great variability (Table 2). The increase in epinephrine concentration was significantly higher at sternotomy in the fentanyl group compared to the placebo (P < 0.05). Norepinephrine concentrations increased following skin incision in all but one case, where it decreased.
Major changes in either direction were seen between time points 1 and 2 in one case for epinephrine and four cases for norepinephrine. Skin incision and/or sternotomy led to an increase of epinephrine and norepinephrine concentrations in most cases.
Haemodynamic changes and temperature
Changes in catecholamine concentrations were only very inconsistently mirrored by concomitant changes in mean arterial pressure (MAP) and heart rate (HR). Only a correlation, between the level of epinephrine and MAP was significant (P = 0.04) but with a weak r2 = 0.0784. No other significant correlation between the variables was detectable.
In an attempt to block the release of catecholamines following painful surgical stimulation in brain-dead organ donors, we administered fentanyl 7 μg kg−1. The catecholamine concentrations in our study showed a great variability, as is known to occur in brain-dead donors during organ retrieval surgery, but clearly answers our hypothesis: fentanyl 7 μg kg−1 was not effective in reliably suppressing the catecholamine release elicited by painful stimuli in these brain-dead organ donors.
The variability seen in our results requires some comment. This finding correlates well with the results reported in previous studies of catecholamine concentrations and other stress-related hormones and has been attributed to the breakdown of homoeostasis in brain death [4,5]. Several possible mechanisms might contribute to this variability. Deficits at the receptor level have been described, which could inhibit the function of the catecholamines released . Hypothermia and acidosis might also impair the effect of catecholamines released, or directly cause myocardial depression . Vasomotor tone dysregulation and microcirculatory maldistribution lead to an oxygen extraction deficiency , which, like the failure of thyroid function, might cause a depletion of energy-rich phosphates . The data obtained in our preceding studies also varied considerably and only showed an inconstant correlation to haemodynamic alterations, such as hypertension or tachycardia [4,5]. In the current investigation, we also found that the increases in epinephrine and norepinephrine concentrations did not arise simultaneously and the only correlation found between the catecholamines and hormones was a very weak correlation between epinephrine and MAP.
The most striking finding in this study was that the most pronounced catecholamine releases were in the group receiving fentanyl - leading to a significant difference between the groups in epinephrine concentrations following sternotomy. However, as it was never the purpose of this study to investigate the mechanisms of fentanyl action, the groups are far too small and their numbers lack the power to allow detailed interpretation. Pharmacodynamics and pharmacokinetics of opioids in brain death are unknown, and the variability of the serum concentrations might be exaggerated by the unknown time course and the shape of the response between catecholamines and painful stimulation. However, we emphasize that the epinephrine and norepinephrine increases seen in the donors receiving fentanyl were sufficient to permit us to reject our hypothesis that opioids could suppress the pain-related catecholamine release.
The variability in the response following the painful stimulation (on skin incision and sternotomy) gave us the impression that, despite their excellent condition, these brain-dead donors had to meet the study entry criteria (no exogenous catecholamines) - thus we were observing a disintegration of regulatory mechanisms. Hence, the reaction of the organism in this state seems to display a variety of quantitatively and qualitatively different pictures depending on the resting function of the regulatory systems. Deterioration of cardiac function in brain death was found to be unattributable to sympathetic or parasympathetic mechanisms, nitric oxide or endogenous opioid peptides . A marked uncoupling of sympathetic signal transduction was described to occur in acutely failing donor hearts . Thus, the situation observed in our cases may reflect the failure of the many biochemical processes reported for brain-dead patients .
Possibly our results reflect the fact that the dosage of the opioid administered was too small. Although premedication, nitrous oxide and other drugs potentiating the opioid effect during standard anaesthetic procedures were missing in our donors this dosage usually ensures sufficient analgesia in surgical patients. Even a dose of only 0.2 mg 70 kg−1 has been described to blunt the catecholamine response to the much weaker stimulus of endotracheal intubation . Also, we were concerned that higher doses might lead to haemodynamic instability. However, we cannot exclude the fact that a different opioid, or a different dosage, would be more efficient than fentanyl 7 μg kg−1 of used in this protocol.
We conclude that fentanyl at the dose used was ineffective in suppressing the catecholamine release in brain-dead organ donors following painful stimuli. There was great variability in the reaction patterns displayed in our brain-dead organ donors and this seems to be related to the disintegration of regulatory mechanisms, during the course of brain death.
1. Wahlers T, Cremer J, Fieguth, et al.
Donor heart-related variables and early mortality after heart transplantation. J Heart Lung Transplant
2. Mackersie RC, Bronsther OL, Shackford SR. Organ procurement in patients with fatal head injuries. Ann Surg
3. Mertes PM, Carteaux JP, Jaboin Y, et al.
Estimation of myocardial interstitial norepinephrine release after brain death using cardiac microdialysis. Transplantation
4. Fitzgerald RD, Dechtyar I, Templ E, Fridrich P, Lackner FX. Cardiovascular and catecholamine response during surgery in brain-dead organ donors. Anaesthesia
5. Fitzgerald RD, Dechtyar I, Templ E, Pernerstorfer T, Hackl W, Lackner FX. Endocrine stress reaction in brain dead organ donors. Transpt Intern
6. Depret J, Teboul JL, Benoit G, Mercat A, Richard C. Global energetic failure in brain-dead patients. Transplantation
7. Ghosh S, Bethune DW, Hardy I, Kneeshaw J, Latimer RD, Oduro A. Management of donors for heart and heart-lung transplantation. Anaesthesia
8. Wicomb WN, Cooper DK, Lanza RP, Novitzky D, Isaacs S. The effects of brain-death and 24 hours storage by hypothermic perfusion on donor heart function in the pig. J Cardiovasc Surg
9. Galinanes M, Hearse DJ. Brain-death-induced cardiac contractile dysfunction: studies of possible neurohormonal and blood-borne mediators. J Mol Cell Cardiol
10. White M, Wiechmann RJ, Roden RL, et al.
Cardiac beta-adrenergic neuroeffector systems in acute myocardial dysfunction related to brain injury. Evidence for catecholamine-mediated damage. Circulation
11. Tolksdorf W, Schafer E, Pfeiffer J, Mittelstaedt G. Adrenalin-, Noradrenalin-, Blutdruck und Herzfrequenzverhalten während der Intubation in Abhängigkeit unterschiedlicher Fentanyl-Dosen. Anästh Intensivther Notfallmed