Acta Anaesthesiol &and 1990: 34: 592-595
Peroperative fluid management of the brain-dead multiorgan donor T. RANDELL, R. ORKOand K. HOGKERSTEDT Departments of Anaesthesia and Surgery IV, Helsinki University Central Hospital, Helsinki, Finland
Brain-dead organ donors are often dehydrated and have serum electrolyte disorders. This study was designed to analyse the haemodynamic condition and serum electrolyte balance of liver donors. Two different fluid management plans for the harvesting operation were studied. Sixteen consecutive organ donors were included. They were randomly infused either with a combination of colloid (hydroxy ethyl starch) and electrolyte solution (group COL) or with crystalloid fluid alone (group CR). Arterial pressures, heart rate, central venous pressure and oesophageal temperature were monitored and serum electrolytes were analysed before the beginning of the operation and during harvesting. The amount of fluid needed in the COL group was significantly less ( P < O . O l ) than in the CR group. There were no statistical differences between the groups in the haemodynamic parameters during the study period. The oesophageal temperature was maintained in both groups. All donors were initially hypernatraemic, but the serum sodium values returned towards normal during surgery in both groups. Immediate function was seen in all livers. In conclusion, the haemodynamic stability is maintained with a smaller infused volume if hydroxyethyl starch is combined with crystalloid fluids. The formation of interstitial oedema will be less when colloids are used, but its significance in organ donation needs further evaluation. Received 29 October 1989, accepted for publication 25 April 1990
Key words: Brain death; colloids; hydroxyethyl starch; organ donation; transplantation.
Total brain infarction (brain death) is associated with certain pathophysiological sequelae. The vasomotor and temperature regulations are lost, and diabetes insipidus is a common finding (1). Unless closely monitored and treated, a donor will become dehydrated, which causes rapid haemodynamic deterioration. Anaerobic metabolism is increased. Depletion of myocardial high energy compounds (adenosine triphosphate and creatine phosphate) occurs, leading to impaired cardiac function (2). As 12% of liver grafts function poorly and 3% fail to function at all after transplantation (3), concentration on donor care seems most important. In our centre, we predominantly harvest organs from hospitals in Finland but also from several hospitals in other parts of Scandinavia. Therefore, the fluid balance of the donors is primarily managed by other than the transplantation unit physicians and the fluid therapy is based on local hospital policy. The donors referred to us have often been hypernatraemic and hypovolaemic. This study was prompted by a case in which cardiac arrest occurred in a liver donor at the beginning of the operation. We wanted to evaluate the haemodynamic condition of liver donors and to compare two different fluid management plans during multiorgan harvesting. At present, there is no general agreement on the treatment of the multiple organ donor.
PATIENTS AND METHODS All liver donors operated on by the transplantation team from January 1988 to March 1989 were studied. This included 16 multiorgan donors. The study was approved by the Ethical Committee of Helsinki University Central Hospital. When the transplantation centre surgeon (K.H.) had accepted a liver donor, the transplantation team anaesthesiologist immediately contacted the donor hospital physicians and guidelines for fluid management were given. The transplantation team took over management on arrival in the donor hospital, or, on three occasions, when the donor arrived in our own hospital. The charts of the donors were reviewed for preoperative fluid therapy and electrolyte balance. During surgery, ECG and oesophageal temperature were continuously monitored. An artery of either upper extremity was cannulated for continuous arterial pressure monitoring and for collection of blood samples. Central venous pressure (CVP) was measured at 10-min intervals. Urine output was recorded at 1-h intervals. The i.v. fluid infusion rate was also adjusted according to the surgeons' evaluation of the condition of the operation field and the liver. Laboratory data. Blood gases and serum sodium and potassium levels were measured at the beginning of surgery and thereafter at 1-h intervals. Haemoglobin and blood glucose were analysed before the start of the fluid management plan. Prior to surgery the donor was randomly allocated to either a colloid (COL) or a crystalloid (CR) group. Heart rate, arterial pressures, CVP and temperature were recorded and blood samples were obtained. In the COL group the donor was infused with 500 ml 6% hydroxyethyl starch (HES)(Plasmafusin", Orion, Finland). If CVP was less than 5 mmHg (0.67 kPa), an additional 50@1000 ml of hydroxyethyl starch was infused to increase the CVP to the
MANAGEMENT O F MULTIORGAN DONOR desired level. Thereafter the donor was infused with an electrolyte solution. In the CR group the donor was given only Ringer's acetate (Ringersteril", Medipolar, Finland) or 0.45% NaCl solution, depending on the serum sodium value. Warmed fluids were used. The electrolyte contents and pH of the fluids are shown in Table 1. Fluid was infused to maintain the CVP above 5 mmHg (0.67 kPa). If the arterial pressure was below 90 mmHg ( 1 2 kPa) for more than 1 min, a dopamine infusion was started unless the hypotension was caused by surgical manipulation. The infusion rate was adjusted to maintain the systolic arterial pressure above 90 mmHg (12 kPa). The donor procedure included hepatectomy and bilateral nephrectomy in all cases and removal of the heart in three cases. At laparotomy, both kidneys were first prepared free but left in place. The liver hilus was also dissected and in order to maintain the donor body temperature, thoracotomy was usually done at this late stage when the suprahepatic vena cava was entered. Perfusion cannulae were placed, both kidneys were removed and perfused on a back table. In vivo perfusion of the liver through the portal vein was begun and the liver was removed and further perfused with Eurocollins" preservation fluid (4) on the back table. In case of heart removal, whole body perfusion through the aorta was used.
Statistics The parametric data were analysed by ANOVA, and for nonparametric data contingency table analysis was used (5). The difference between the study groups was considered statistically significant if P< 0.05.
RESULTS The groups were comparable with respect to donor age, sex, time from the diagnosis of brain death to the start of surgery and length of the operation (Table 2). In the COL group, the brain death was caused by trauma to the head in 7 cases and by intracranial bleeding in 2 cases. In the CR group, the respective figures were 5 and 1. In one case in the CR group
Table 1 Electrolyte contents and pH of the fluids. Plasmafusin"
NaCl 0.45%
RingersteriP
154
77
130 4 109 5-7
Na' (mmol/l) K' (mmol/l) Cl- (mmol/l) PH
-
154 4.5-7.5
77
5-6.5
Table 2 Characteristics of the study groups (mean 2 s.d.). Col= colloid group; CR = crystalloid group.
COL (n=9) CR (n=7)
Age (years)
Sex (M/F)
Time from diagnosis of brain death to start of surgery (h)
35k 14
7/2
4.4k 1.8
171k 13
35+9
3/4
7.1 k3.9
169 2 37
Length of operation (min)
593
the brain death was caused by cerebral herniation following a diagnostic lumbar puncture. The heart was removed in 2 cases in the COL group and in 1 case in the CR group.
Preoperative management The objectives of fluid management of a potential organ donor in different hospitals were in accordance with the guidelines given by the transplantation centre in Finland. The patient charts did not allow detailed analysis of primary fluid therapy. The fluids given before or after brain death could not be distinguished. O n 9 occasions, fluid management included volume expanders in combination with electrolyte solutions. Two donors received 4% albumin, 4 donors were infused with gelatin solution and 3 donors with HES. In one case, no electrolyte solutions were given. In 11 cases, the donors were also infused with 5% glucose solution. Two donors were given parenteral nutrition prior to organ harvesting. The electrolyte balance was analysed according to the local hospital policy. The serum sodium values during the preoperative phase were high, the mean being 156 mmol/l (range 148-174 mmol/l). The potassium levels tended to be low, 3.2 mmol/l on average. The lowest value was 1.6 mmol/l. Potassium substitution was given when required. Peroperative management The haemodynamic changes in the different study groups during surgery are summarised in Table 3. Table 3 Haemodynamic data of the study groups. The means and the standard deviations are shown. COL = the colloid group, CR = the crystalloid group. I =initial recordings, 30 = 30 min after start of operation, 60 = 60 min from start of operation, TO = thoracotomy, T30 = 30 min after thoracotomy, T60 = 60 min after thoracotomy. Arterial pressures mmHg (kPa) Systolic COL
CR
Heart rate b.p.m.
Diastolic COL
CR
COL
CR
I
135+38 135245 (1825) (18k6)
90+30 (12f4)
83k38 (11k5)
90+31
83+17
30
120223 113523 ( l 6 k 3 ) (15k3)
75k15 (1022)
68k38 (9k5)
103+30
99k23
60
113k23 105k15 ( 1 5 k 3 ) (1452)
68223 (9+3)
68223 (9+2)
102520 101k23
TO
113k23 113k30 (15+3) (1554)
68+23 (923)
68523 (9+3)
103519 102+17
T30 105+30 9 8 k 8 (14+4) (13+1)
53k23 (752)
60k23 (8k2)
105k13 106k16
T60 9 8 k 2 3 (13k2)
5358 (7k1)
68k8 (9k1)
9 9 k 1 8 120k4
98+23 (13k2)
594
T. RANDELL E T AL.
There are no statistical differences between the groups. A dopamine infusion was required in 819 donors of the COL group and in all of the CR group. The mean infusion rate was 4.3 pg/kg/min in both groups. No blood was transfused to the donors. The urine output was satisfactory in all donors. The serum sodium and potassium levels are shown in Table 4. The blood glucose level varied greatly (range 3.4-36.6 mmol/l). None of the donors were given insulin, however. The oesophageal temperature remained stable in both groups. The initial oesophageal temperature (mean k s.d.) was 33.6 k 2.0"C in the COL group and 33.6k 1.9"C in the CR group. O n starting the perfusion of the liver, the temperatures were 32.9 & 1.2"C and 33.5 & 1.6"C, respectively. The amount of HES infused averaged (mean f s.e.mean) 1080 f 300 ml. The total amount of fluid given was 5310 740 ml in the COL group and 9960 f 1360 ml in the C R group. The difference between the groups is statistically significant (P< 0.0 1). There were no cases of primary nonfunction of the liver. DISCUSSION The organ donor is often polyuric, hypotensive and hypernatraemic. Intensive monitoring and fluid management, even if initiated as late as at the start of the operation, were sufficient to maintain the haemodynamic stability during the operation. Adequate treatment of brain-dead organ donors has frequently been neglected, possibly due to the emotional stress involved in the situation. Also, for centres not actively participating in organ transplantation, the treatment of a brain-dead donor may be considered secondary to other activities in the unit. Therefore, the monitoring of the donor is sometimes inadequate, resulting in failure to correct the ensuing fluid and electrolyte imbalance. In the present study, we were not able to analyse the preoperative treatment in detail. The fluid balance sheets showed only total amounts of fluids given during Table 4 The electrolyte balance of the donors (mean & s.d.). COL= colloid group. CR = crystalloid group. 0 =start of operation. 1 = 1 h after start of operation. 2 = 2 h after start of operation. S-Na (mmol/l) 0 1 2
S-K (mmol/l)
COL
CR
COL
CR
156+10 151 f 7 14829
154f7 151f6 146k7
3.8 f 0.8 4.0 f 0.7
3.7 k 0.4 3.8 0.4 3.8 0.5
3.9 2 0.6
a certain 24-hour period. Therefore, comparisons could not be made between those patients infused only with crystalloids and those also given colloids during the preoperative period. Furthermore, the choice of the solution infused was based on the practice of the local hospital, and great variation was found. Brain death is followed by pituitary dysfunction causing diabetes insipidus. Polyuria, hypernatraemia, hypokalaemia, hyperosmolality and dehydration are characteristic findings of the untreated condition. The specific treatment of diabetes insipidus is antidiuretic hormone (ADH) (6, 7). However, during organ harvesting, diuresis is considered a valuable sign of adequate kidney perfusion (8). Also, ADH is known to cause vasoconstriction and even hypoxia in the splanchnic bed (9). The significance of it is obscure in the liver donor. We suggest that dehydration and associated hypernatraemia are managed with vigorous fluid infusion. Both electrolyte solutions and colloid fluids are used for volume expansion in surgical patients. The amount needed to correct hypovolaemia with crystalloids is 3 to 4 times greater than when using colloid solutions. Electrolyte solutions escape into the interstitial space, causing oedema of the tissues (10, 1 1) . The importance of an excessive water content of the organs to be transplanted needs further evaluation. I n experimental studies we have found that excessive use of crystalloids in treating hypovolaemic shock leads to impaired oxygenation of the liver ( 12). In addition to the adverse effects of crystalloid solutions mentioned above, fluid management with colloids seems less time- and effort-consuming. Also, if not adequately warmed, the great amounts of fluid needed to maintain haemodynamic stability may contribute to cooling of the donor. I n this study, there was no decrease in core temperature in either group. This was probably due to our vigorous attempt to warm the donor with heated gases and fluids, and warm blankets. Donor age and length of intensive care treatment have been associated with poor liver graft function (3). The nonfunction or delayed function of kidney grafts has been attributed to hypotension or the requirement of high-dose dopamine (13, 14). As also suggested by Yanaga and others (3), maintenance of haemodynamic stability in the multiple organ donor seems important to preserve optimal liver perfusion. In this study, primary function of all livers was noted, which compares favourably with the results of other studies (3, 15). In conclusion, if multiple organ donors are intensively monitored and treated, their haemodynamic stability can be maintained and normal electrolyte bal-
MANAGEMENT O F MULTIORGAN DONOR
ance restored. Colloid solutions may be beneficial in the treatment of a brain-dead organ donor.
REFERENCES I . Powner D J, Lagler R G, Jastremski M. Medical management of the brain-dead patient in preparation for organ donation. Indiana Med 1986: 79: 966-968. 2. Novitzky D, Wicomb W N, Cooper D K, Tjaalgard M A. Improved cardiac function following hormonal therapy in braindead pigs: relevance to organ donation. Cryobiology 1987: 24: 1-10, 3. Yanaga K, Tzakis A G, Starzl T E. Personal experience with the procurement of 132 liver allografts. Transplant Int 1989: 2: 137-142. 4. Dreikorn K, Horsch R, Rohl L. 48- to 96-hour preservation of canine kidneys by initial perfusion and hypothermic storage using Euro-Collins solution. Eur Urol 1980: 6: 22 1. 5. Glanz S A. Primer of biostatistics. New York: McGraw-Hill Book Company, 1987. 6. Blaine E M, Tallrnan R D, Frolicher D, Jordan M A, Bluth L L, Howie M B. Vasopressin supplementation in a porcine model of brain-dead potential organ donors. Transplantation 1984: 38: 459-464. 7. Streeten D H P, Moses A M, Miller M. Disorders ofthe neurohypophysis. In: Thorn G W, Adams R D, Braunwald E, Isselbacher K J, PetersdorfR G, eds. Harrison’s principles ofinternal medicine. Tokyo: McGraw Hill Kogakusha Ltd., 1977: 490-501. 8. Luksza A R. Brain-dead kidney donor: selection, care, and administration. Br Med J 1979: 1: 1316-1319.
595
9. Korsback C, Hockerstedt K. Small bowel and liver gas tensions during intravenous vasopressin infusion and 60% oxygen ventilation. Res Exp Med 1984: 184: 243-249. 10. Hauser C J, Shoemaker W C, Turpin I, Golberg S J. Oxygen transport responses to colloids and crystalloids in critically ill surgical patients. Surg Gynecol Obstet 1980: 150: 81 1-816. 11. Linko K , Makelainen A. Hydroxyethyl starch 120, dextran 70 and acetated Ringer’s solution: hemodilution, albumin, colloid osmotic pressure and fluid balance following replacement of blood loss in pigs. Act0 Anaesthesia1 Scand 1988: 32: 228-233. 12. Makisalo H , Soini H, Heino A, Korsback C, Hockerstedt K. Correction of hemorrhagic shock-induced liver hypoxia with whole blood, Ringer’s solution or with hetastarch. Res Exp Med. (In press). 13. Halloran P, Aprile M, Farewell V. Factors influencing early renal function in cadaver kidney transplants. Transplantation 1988: 45: 12 2- 12 7. 14. Meurisse M, Albert A, Defraigne J 0 et al. Multiple risk factor analysis of non-immunological delayed graft function after kidney transplantation. Clin Transplant 1988: 2: 31 2-318. 15. Bismuth H, Ericzon B G, Rolles K , Castaing D, Otte J B, Ringe B. Hepatic transplantation in Europe. First report of the European Liver Transplant Registry. Lancet 1987: ii: 674-676. Address: I: Randell, M.D. Department of Anaesthesia Surgical Hospital Kasarmikatu 11-13 SF-00 130 Helsinki Finland
Peroperative fluid management of the brain-dead multiorgan donor T. RANDELL, R. ORKOand K. HOGKERSTEDT Departments of Anaesthesia and Surgery IV, Helsinki University Central Hospital, Helsinki, Finland
Brain-dead organ donors are often dehydrated and have serum electrolyte disorders. This study was designed to analyse the haemodynamic condition and serum electrolyte balance of liver donors. Two different fluid management plans for the harvesting operation were studied. Sixteen consecutive organ donors were included. They were randomly infused either with a combination of colloid (hydroxy ethyl starch) and electrolyte solution (group COL) or with crystalloid fluid alone (group CR). Arterial pressures, heart rate, central venous pressure and oesophageal temperature were monitored and serum electrolytes were analysed before the beginning of the operation and during harvesting. The amount of fluid needed in the COL group was significantly less ( P < O . O l ) than in the CR group. There were no statistical differences between the groups in the haemodynamic parameters during the study period. The oesophageal temperature was maintained in both groups. All donors were initially hypernatraemic, but the serum sodium values returned towards normal during surgery in both groups. Immediate function was seen in all livers. In conclusion, the haemodynamic stability is maintained with a smaller infused volume if hydroxyethyl starch is combined with crystalloid fluids. The formation of interstitial oedema will be less when colloids are used, but its significance in organ donation needs further evaluation. Received 29 October 1989, accepted for publication 25 April 1990
Key words: Brain death; colloids; hydroxyethyl starch; organ donation; transplantation.
Total brain infarction (brain death) is associated with certain pathophysiological sequelae. The vasomotor and temperature regulations are lost, and diabetes insipidus is a common finding (1). Unless closely monitored and treated, a donor will become dehydrated, which causes rapid haemodynamic deterioration. Anaerobic metabolism is increased. Depletion of myocardial high energy compounds (adenosine triphosphate and creatine phosphate) occurs, leading to impaired cardiac function (2). As 12% of liver grafts function poorly and 3% fail to function at all after transplantation (3), concentration on donor care seems most important. In our centre, we predominantly harvest organs from hospitals in Finland but also from several hospitals in other parts of Scandinavia. Therefore, the fluid balance of the donors is primarily managed by other than the transplantation unit physicians and the fluid therapy is based on local hospital policy. The donors referred to us have often been hypernatraemic and hypovolaemic. This study was prompted by a case in which cardiac arrest occurred in a liver donor at the beginning of the operation. We wanted to evaluate the haemodynamic condition of liver donors and to compare two different fluid management plans during multiorgan harvesting. At present, there is no general agreement on the treatment of the multiple organ donor.
PATIENTS AND METHODS All liver donors operated on by the transplantation team from January 1988 to March 1989 were studied. This included 16 multiorgan donors. The study was approved by the Ethical Committee of Helsinki University Central Hospital. When the transplantation centre surgeon (K.H.) had accepted a liver donor, the transplantation team anaesthesiologist immediately contacted the donor hospital physicians and guidelines for fluid management were given. The transplantation team took over management on arrival in the donor hospital, or, on three occasions, when the donor arrived in our own hospital. The charts of the donors were reviewed for preoperative fluid therapy and electrolyte balance. During surgery, ECG and oesophageal temperature were continuously monitored. An artery of either upper extremity was cannulated for continuous arterial pressure monitoring and for collection of blood samples. Central venous pressure (CVP) was measured at 10-min intervals. Urine output was recorded at 1-h intervals. The i.v. fluid infusion rate was also adjusted according to the surgeons' evaluation of the condition of the operation field and the liver. Laboratory data. Blood gases and serum sodium and potassium levels were measured at the beginning of surgery and thereafter at 1-h intervals. Haemoglobin and blood glucose were analysed before the start of the fluid management plan. Prior to surgery the donor was randomly allocated to either a colloid (COL) or a crystalloid (CR) group. Heart rate, arterial pressures, CVP and temperature were recorded and blood samples were obtained. In the COL group the donor was infused with 500 ml 6% hydroxyethyl starch (HES)(Plasmafusin", Orion, Finland). If CVP was less than 5 mmHg (0.67 kPa), an additional 50@1000 ml of hydroxyethyl starch was infused to increase the CVP to the
MANAGEMENT O F MULTIORGAN DONOR desired level. Thereafter the donor was infused with an electrolyte solution. In the CR group the donor was given only Ringer's acetate (Ringersteril", Medipolar, Finland) or 0.45% NaCl solution, depending on the serum sodium value. Warmed fluids were used. The electrolyte contents and pH of the fluids are shown in Table 1. Fluid was infused to maintain the CVP above 5 mmHg (0.67 kPa). If the arterial pressure was below 90 mmHg ( 1 2 kPa) for more than 1 min, a dopamine infusion was started unless the hypotension was caused by surgical manipulation. The infusion rate was adjusted to maintain the systolic arterial pressure above 90 mmHg (12 kPa). The donor procedure included hepatectomy and bilateral nephrectomy in all cases and removal of the heart in three cases. At laparotomy, both kidneys were first prepared free but left in place. The liver hilus was also dissected and in order to maintain the donor body temperature, thoracotomy was usually done at this late stage when the suprahepatic vena cava was entered. Perfusion cannulae were placed, both kidneys were removed and perfused on a back table. In vivo perfusion of the liver through the portal vein was begun and the liver was removed and further perfused with Eurocollins" preservation fluid (4) on the back table. In case of heart removal, whole body perfusion through the aorta was used.
Statistics The parametric data were analysed by ANOVA, and for nonparametric data contingency table analysis was used (5). The difference between the study groups was considered statistically significant if P< 0.05.
RESULTS The groups were comparable with respect to donor age, sex, time from the diagnosis of brain death to the start of surgery and length of the operation (Table 2). In the COL group, the brain death was caused by trauma to the head in 7 cases and by intracranial bleeding in 2 cases. In the CR group, the respective figures were 5 and 1. In one case in the CR group
Table 1 Electrolyte contents and pH of the fluids. Plasmafusin"
NaCl 0.45%
RingersteriP
154
77
130 4 109 5-7
Na' (mmol/l) K' (mmol/l) Cl- (mmol/l) PH
-
154 4.5-7.5
77
5-6.5
Table 2 Characteristics of the study groups (mean 2 s.d.). Col= colloid group; CR = crystalloid group.
COL (n=9) CR (n=7)
Age (years)
Sex (M/F)
Time from diagnosis of brain death to start of surgery (h)
35k 14
7/2
4.4k 1.8
171k 13
35+9
3/4
7.1 k3.9
169 2 37
Length of operation (min)
593
the brain death was caused by cerebral herniation following a diagnostic lumbar puncture. The heart was removed in 2 cases in the COL group and in 1 case in the CR group.
Preoperative management The objectives of fluid management of a potential organ donor in different hospitals were in accordance with the guidelines given by the transplantation centre in Finland. The patient charts did not allow detailed analysis of primary fluid therapy. The fluids given before or after brain death could not be distinguished. O n 9 occasions, fluid management included volume expanders in combination with electrolyte solutions. Two donors received 4% albumin, 4 donors were infused with gelatin solution and 3 donors with HES. In one case, no electrolyte solutions were given. In 11 cases, the donors were also infused with 5% glucose solution. Two donors were given parenteral nutrition prior to organ harvesting. The electrolyte balance was analysed according to the local hospital policy. The serum sodium values during the preoperative phase were high, the mean being 156 mmol/l (range 148-174 mmol/l). The potassium levels tended to be low, 3.2 mmol/l on average. The lowest value was 1.6 mmol/l. Potassium substitution was given when required. Peroperative management The haemodynamic changes in the different study groups during surgery are summarised in Table 3. Table 3 Haemodynamic data of the study groups. The means and the standard deviations are shown. COL = the colloid group, CR = the crystalloid group. I =initial recordings, 30 = 30 min after start of operation, 60 = 60 min from start of operation, TO = thoracotomy, T30 = 30 min after thoracotomy, T60 = 60 min after thoracotomy. Arterial pressures mmHg (kPa) Systolic COL
CR
Heart rate b.p.m.
Diastolic COL
CR
COL
CR
I
135+38 135245 (1825) (18k6)
90+30 (12f4)
83k38 (11k5)
90+31
83+17
30
120223 113523 ( l 6 k 3 ) (15k3)
75k15 (1022)
68k38 (9k5)
103+30
99k23
60
113k23 105k15 ( 1 5 k 3 ) (1452)
68223 (9+3)
68223 (9+2)
102520 101k23
TO
113k23 113k30 (15+3) (1554)
68+23 (923)
68523 (9+3)
103519 102+17
T30 105+30 9 8 k 8 (14+4) (13+1)
53k23 (752)
60k23 (8k2)
105k13 106k16
T60 9 8 k 2 3 (13k2)
5358 (7k1)
68k8 (9k1)
9 9 k 1 8 120k4
98+23 (13k2)
594
T. RANDELL E T AL.
There are no statistical differences between the groups. A dopamine infusion was required in 819 donors of the COL group and in all of the CR group. The mean infusion rate was 4.3 pg/kg/min in both groups. No blood was transfused to the donors. The urine output was satisfactory in all donors. The serum sodium and potassium levels are shown in Table 4. The blood glucose level varied greatly (range 3.4-36.6 mmol/l). None of the donors were given insulin, however. The oesophageal temperature remained stable in both groups. The initial oesophageal temperature (mean k s.d.) was 33.6 k 2.0"C in the COL group and 33.6k 1.9"C in the CR group. O n starting the perfusion of the liver, the temperatures were 32.9 & 1.2"C and 33.5 & 1.6"C, respectively. The amount of HES infused averaged (mean f s.e.mean) 1080 f 300 ml. The total amount of fluid given was 5310 740 ml in the COL group and 9960 f 1360 ml in the C R group. The difference between the groups is statistically significant (P< 0.0 1). There were no cases of primary nonfunction of the liver. DISCUSSION The organ donor is often polyuric, hypotensive and hypernatraemic. Intensive monitoring and fluid management, even if initiated as late as at the start of the operation, were sufficient to maintain the haemodynamic stability during the operation. Adequate treatment of brain-dead organ donors has frequently been neglected, possibly due to the emotional stress involved in the situation. Also, for centres not actively participating in organ transplantation, the treatment of a brain-dead donor may be considered secondary to other activities in the unit. Therefore, the monitoring of the donor is sometimes inadequate, resulting in failure to correct the ensuing fluid and electrolyte imbalance. In the present study, we were not able to analyse the preoperative treatment in detail. The fluid balance sheets showed only total amounts of fluids given during Table 4 The electrolyte balance of the donors (mean & s.d.). COL= colloid group. CR = crystalloid group. 0 =start of operation. 1 = 1 h after start of operation. 2 = 2 h after start of operation. S-Na (mmol/l) 0 1 2
S-K (mmol/l)
COL
CR
COL
CR
156+10 151 f 7 14829
154f7 151f6 146k7
3.8 f 0.8 4.0 f 0.7
3.7 k 0.4 3.8 0.4 3.8 0.5
3.9 2 0.6
a certain 24-hour period. Therefore, comparisons could not be made between those patients infused only with crystalloids and those also given colloids during the preoperative period. Furthermore, the choice of the solution infused was based on the practice of the local hospital, and great variation was found. Brain death is followed by pituitary dysfunction causing diabetes insipidus. Polyuria, hypernatraemia, hypokalaemia, hyperosmolality and dehydration are characteristic findings of the untreated condition. The specific treatment of diabetes insipidus is antidiuretic hormone (ADH) (6, 7). However, during organ harvesting, diuresis is considered a valuable sign of adequate kidney perfusion (8). Also, ADH is known to cause vasoconstriction and even hypoxia in the splanchnic bed (9). The significance of it is obscure in the liver donor. We suggest that dehydration and associated hypernatraemia are managed with vigorous fluid infusion. Both electrolyte solutions and colloid fluids are used for volume expansion in surgical patients. The amount needed to correct hypovolaemia with crystalloids is 3 to 4 times greater than when using colloid solutions. Electrolyte solutions escape into the interstitial space, causing oedema of the tissues (10, 1 1) . The importance of an excessive water content of the organs to be transplanted needs further evaluation. I n experimental studies we have found that excessive use of crystalloids in treating hypovolaemic shock leads to impaired oxygenation of the liver ( 12). In addition to the adverse effects of crystalloid solutions mentioned above, fluid management with colloids seems less time- and effort-consuming. Also, if not adequately warmed, the great amounts of fluid needed to maintain haemodynamic stability may contribute to cooling of the donor. I n this study, there was no decrease in core temperature in either group. This was probably due to our vigorous attempt to warm the donor with heated gases and fluids, and warm blankets. Donor age and length of intensive care treatment have been associated with poor liver graft function (3). The nonfunction or delayed function of kidney grafts has been attributed to hypotension or the requirement of high-dose dopamine (13, 14). As also suggested by Yanaga and others (3), maintenance of haemodynamic stability in the multiple organ donor seems important to preserve optimal liver perfusion. In this study, primary function of all livers was noted, which compares favourably with the results of other studies (3, 15). In conclusion, if multiple organ donors are intensively monitored and treated, their haemodynamic stability can be maintained and normal electrolyte bal-
MANAGEMENT O F MULTIORGAN DONOR
ance restored. Colloid solutions may be beneficial in the treatment of a brain-dead organ donor.
REFERENCES I . Powner D J, Lagler R G, Jastremski M. Medical management of the brain-dead patient in preparation for organ donation. Indiana Med 1986: 79: 966-968. 2. Novitzky D, Wicomb W N, Cooper D K, Tjaalgard M A. Improved cardiac function following hormonal therapy in braindead pigs: relevance to organ donation. Cryobiology 1987: 24: 1-10, 3. Yanaga K, Tzakis A G, Starzl T E. Personal experience with the procurement of 132 liver allografts. Transplant Int 1989: 2: 137-142. 4. Dreikorn K, Horsch R, Rohl L. 48- to 96-hour preservation of canine kidneys by initial perfusion and hypothermic storage using Euro-Collins solution. Eur Urol 1980: 6: 22 1. 5. Glanz S A. Primer of biostatistics. New York: McGraw-Hill Book Company, 1987. 6. Blaine E M, Tallrnan R D, Frolicher D, Jordan M A, Bluth L L, Howie M B. Vasopressin supplementation in a porcine model of brain-dead potential organ donors. Transplantation 1984: 38: 459-464. 7. Streeten D H P, Moses A M, Miller M. Disorders ofthe neurohypophysis. In: Thorn G W, Adams R D, Braunwald E, Isselbacher K J, PetersdorfR G, eds. Harrison’s principles ofinternal medicine. Tokyo: McGraw Hill Kogakusha Ltd., 1977: 490-501. 8. Luksza A R. Brain-dead kidney donor: selection, care, and administration. Br Med J 1979: 1: 1316-1319.
595
9. Korsback C, Hockerstedt K. Small bowel and liver gas tensions during intravenous vasopressin infusion and 60% oxygen ventilation. Res Exp Med 1984: 184: 243-249. 10. Hauser C J, Shoemaker W C, Turpin I, Golberg S J. Oxygen transport responses to colloids and crystalloids in critically ill surgical patients. Surg Gynecol Obstet 1980: 150: 81 1-816. 11. Linko K , Makelainen A. Hydroxyethyl starch 120, dextran 70 and acetated Ringer’s solution: hemodilution, albumin, colloid osmotic pressure and fluid balance following replacement of blood loss in pigs. Act0 Anaesthesia1 Scand 1988: 32: 228-233. 12. Makisalo H , Soini H, Heino A, Korsback C, Hockerstedt K. Correction of hemorrhagic shock-induced liver hypoxia with whole blood, Ringer’s solution or with hetastarch. Res Exp Med. (In press). 13. Halloran P, Aprile M, Farewell V. Factors influencing early renal function in cadaver kidney transplants. Transplantation 1988: 45: 12 2- 12 7. 14. Meurisse M, Albert A, Defraigne J 0 et al. Multiple risk factor analysis of non-immunological delayed graft function after kidney transplantation. Clin Transplant 1988: 2: 31 2-318. 15. Bismuth H, Ericzon B G, Rolles K , Castaing D, Otte J B, Ringe B. Hepatic transplantation in Europe. First report of the European Liver Transplant Registry. Lancet 1987: ii: 674-676. Address: I: Randell, M.D. Department of Anaesthesia Surgical Hospital Kasarmikatu 11-13 SF-00 130 Helsinki Finland
