- Email: [email protected]
Lethal complications of bilateral nephrectomy and splenectomy in hemodialyzed patients
Lethal Complications of Bilateral Nephrectomy and Splenectomy in Hemodialyzed Patients Arthur J. Matas, MD, Minneapolis, Minnesota Richard L. Slmmons, MD, Minneapolis, Minnesota Thecdor J. Buselmeier, MD, Minneapolis, Minnesota John S. Najarign, MD, Minneapolis, Minnesota Carl M. Kjellstrand, MD, Minneapolis, Minnesota
The routine removal of both kidneys and the spleen in preparation for kidney transplantation has become a controversial issue. The question of whether the benefits to post-transplant graft function outweigh the hazards of the operation (mortality) has yet to be answered. This paper reviews twenty pretransplantation deaths in 421 patients
undergoing bilateral nephrectomy
and splenecto-
my (BNS). We have analyzed the causes of death and the preoperatively definable risk factors in an effort to minimize the risk of this procedure in the future. Methods and Material Bilateral nephrectomy and splenectomy are performed routinely in all patients accepted to the University of Minnesota Transplantation Program. Before BNS, each patient receives intensive medical care, including at least one week of hemodialysis to stabilize his condition. Hemodialysis is always performed forty-eight and twenty-four hours before nephrectomy so that dialysis, with its inherent need for heparin anticoagulation, will not be necessary in the immediate postoperative period [I]. The evening before the operation, a long arm intravenous line is placed to monitor central venous pressure. At the same time, hematocrit, serum electrolyte levels, and blood sugar values are determined to ensure that the patient has a satisfactory fluid and electrolyte balance. BNS is performed through a transabdominal midline incision [2]. Postoperatively, the patient reFrom the Departmentsof Surgery
and Medicine, University of Minnesota, Minneapolis. Minnesota. This work was supported by USPHS Grant AM 13063. Reprint requests should be addressed to C. M. Kjelistrand, MD. Box 123-Mayo Memorial Bullding. University of Minnesota, Minneapolis. Minnesota 55455.
816
ceives intensive nursing care for twenty-four hours, with frequent monitoring of vital signs and central venous pressure. Electrolytes, hemoglobin, hematocrit, and glucose levels (in diabetic patients) are monitored at four to six hour intervals for the first forty-eight hours or until they are stable. Ambulation is begun the evening of operation, and the patient is started on solid foods on the third or fourth postoperative day. The patient usually receives his first dialysis under regional heparinization on the second or third postoperative day [3]. From January 1, 1968 to July 30, 1974, 421 pretransplant bilateral nephrectomies and splenectomies were performed; twenty deaths (4.8 per cent) occurred prior to renal transplantation. Seven deaths occurred within the first thirty postoperative days, and five additional deaths were attributed to the operative procedure itself, giving an overall surgical mortality of 2.8 per cent. (Tables I, II, and III.) Results Mortality in Diabetic Patients. There were eight deaths in fifty-two diabetic patients (15.4 per cent) 3 to 120 days after operation. (Table I.) Five of these deaths (9.6 per cent) must be considered operative deaths since they occurred within the first thirty postoperative days, although one of these occurred after the patient was discharged. Three other diabetic patients died 44 to 120 days after operation. Their deaths were not directly related to the procedure. Four of the eight deaths in the diabetic patients were secondary to myocardial infarction, and three of these patients died after they were discharged from the hospital and while they were awaiting transplantation. All four of these patients had severe hypertension before BNS and two had had
The Amarlcan Journal of Surgery
BNS in Hemodialyzed
TABLE I __-.
Mortality
&a (yr) and Sex
Patient
in Diabetic
Patients
____
Dialysis Time before BNS ____-
~_
Patients
_~ _ _~ Post-
Risk Factors
operative Day of Death
Cause of Death
--
Operative Mortality 1
36,M
4 mo
2
53,M
4mo
3 4
35,M 32, F
1 yr 1 mo
5
45,F
4mo
Severe hypertension, old myocardial infarction Hypertension, old myocardial infarction, old cerebrovascular accident, congestive heart failure Moderate hypertension Moderate hypertension, hypothyroidism Hypertension, hypothyroidism
Myocardial
infarction
Myocardial
infarction
3
embolus
30 7 11
Septicemia Septicemia Pulmonary
after discharge
Nonoperative Mortality 6
36, F
2mo
Severe
7
28,M 32.M
1 yr 4mo
Severe hypertension
8
TABLE II
Patient
Mortality
&a (yr) and Sex
hypertension
-____ Myocardial infarction, pericardial effusion after discharge Myocardial infarction after discharge Fluid overload
in Nondiabetic
21
on dialysis
60 44 120
Patients
Underlying Disease
Dialysis Time before BNS
Risk Factors
Cause of Death
Postoperative Day of Death
Operative Mortality 1
32,M
Chronic
pyelonephritis
2mo
2
32,F
Chronic
pyelonephritis
2wk
3 4 5 6 7
25,M 35,M 54,M 48,M 47,M
Radiation nephritis Chronic glomerulonephritis Polycystic kidneys Chronic pyelonephritis Hemolytic uremic syndrome
2wk 2wk 2 yr 3mo 2 mo
lleal loop, infected kidneys Infected kidneys, perinephric abscess Morbid obesity Severe hypertension Polycystic kidneys Infected kidneys Severe hypertension
Drug hypersensitivity
33
Sepis
31
Sepsis Myocardial infarction Sep&s Sepsis Technical complications (retroperitoneal hemorrhage)
38 3 40 45 7
Cerebra! bleeding Infection 1 yr post BNS Myocardial infarction after discharge Pericardial effusion
35 360 31
Nonoperative Mortality 8 9 10
45,F 50, F 30,M
Malignant nephrosclerosis Malignant nephrosclerosis Chronic glomerulonephritis
3mo 2 yr 4mo
Severe hypertension Severe hypertension Severe hypertension
11
29, F
Chronic
2mo
Systemic
6 mo
erythematous Hypoproteinemia age of 18 mo
12
1.5,M
glomerulonephritis
Congenital
nephrotic
syndrome
myocardial infarcts. Two other deaths in the diabetic patients were secondary to infection and one was due to a pulmonary embolus. The eighth death was caused by fluid overload while the patient was being maintained on hemodialysis and awaiting a cadaver donor kidney.
Vohmo
129, June 1975
lupus at
Cerebral bleeding, bleeding diathesis
180 36
It is difficult to distinguish any specific risk factors in the diabetic patient. These patients are uniformly hypertensive and most of them have pre-existing cardiovascular disease. In addition, most of these patients had reported previous episodes of urinary tract infection.
617
Matas et
TABLE III
al
Cause of Death in Diabetic Patients
and Nondiabetic
Patients Cause of Death Cardiovascular Myocardial infarction Cerebral bleeding Infection Complication of dialysis Other Pulmonary embolus Drug hypersensitivity Technical problem Total deaths Number of patients operated on Pretransplantation mortality (%) Operative mortality (%)
Diabetic
Nondiabetic
4
2 2 5 1
2 1 1
8 52 15.4 9.6
1 1 12 369 3.25 1.9
Mortality in Nondiabetic Patients. Of the nondiabetic patients, 12 of the 369 patients (3.25 per cent) who underwent BNS died before transplantation (compared with 15.4 per cent of the diabetic patients). (Tables II and III.) Two of these deaths (0.6 per cent) occurred within the thirty postoperative days traditionally related to operative mortality statistics; however, seven of the twelve deaths (1.9 per cent) were actually related to intraor postoperative complications. Four of the twelve deaths (33 per cent) were attributed to sepsis within sixty days after BNS, and a fifth death was caused by infection one year postoperatively. Two patients died after cerebral hemorrhage. Complications of dialysis, retroperitoneal hemorrhage, and a drug hypersensitivity reaction each resulted in one death. Myocardial infarction was responsible for the death of two of twelve nondiabetic patients (17 per cent). Both patients (numbers 4 and 10, Table II) were young men, thirty and thirty-five years old, with severe, uncontrollable hypertension. Eleven nondiabetic patients who died (91.6 per cent) had preoperatively definable factors that placed them in a high risk group. Five patients had uncontrollable hypertension, three had recurrent pyelonephritis, one patient had systemic lupus erythematosus, another, an eighteen month old child, had hypoproteinemia, and one patient had morbid obesity. Contrary to our findings in the diabetic patients, there is an evident cause and effect relationship between risk factors and death in a majority of these nondiabetic patients. Vascular catastrophes were responsible for three deaths; all
616
three of these patients had severe hypertension. This can be compared with an incidence of uncontrollable hypertension of approximately 10 per cent in our dialyzed population. Death was secondary to sepsis in five patients, three of whom had infected kidneys (one was operated on because of a perinephric abscess) and one of whom had morbid obesity. One death was due to a cerebral hemorrhage in a child with a bleeding diathesis secondary to a congenital nephrotic syndrome. There was no direct correlation between a risk factor and the cause of death in four patients. One of these deaths resulted from a technical complication. In one patient on dialysis while awaiting transplantation, pericardial effusion and tamponade developed after six months, and one patient died of drug hypersensitivity. The fourth patient died from infection after another operative procedure one year after BNS.
Comments Our data indicate that death after BNS and before kidney transplantation occurs in two preoperatively definable groups of patients. The first group, those with juvenile onset diabetes, have a 15.4 per cent pretransplantation mortality (9.6 per cent operative and 5.8 per cent nonoperative). Fifty per cent of these deaths were caused by myocardial infarction that occurred either in the immediate postoperative period or while the patient was being maintained on hemodialysis and awaiting transplantation. This high incidence of fatal cardiovascular complications is consistent with the fate of this group of patients on maintenance hemodialysis [4,5]. The second group that can be defined preoperatively are patients whose kidney disease or associated diseases (that is, severe hypertension, infected kidneys, or both) place them in a high risk group. The cause of death in these patients is closely related to their preoperative high risk factors. The severely hypertensive patients died from myocardial infarction or intracerebral hemorrhage, whereas patients with chronic pyelonephritis and perinephric abscess died from sepsis. Paradoxically, it is these patients in whom pretransplant nephrectomy is indicated at institutions that do not perform routine pretransplant nephrectomy and splenectomy [6]. The question of whether to perform routine bilateral nephrectomy and splenectomy before transplantation remains unanswered. Splenectomy is advocated as a supportive immunosuppressant
The American Journal of Surgery
BNS in Hemodialyzed
and, in addition, to allow administration of larger amounts of immunosuppressive drugs. However, several researchers diversely suggest that splenectomy (1) influences graft survival [7], (2) has no effect on first graft survival but is important in second graft survival [8], or (3) has no effect at all on graft survival 19,I 01. Whether splenectomy does affect tolerance to azathioprine is also still controversial; some investigators report that the patient who has undergone splenectomy can better tolerate the drugs [II], whereas other authors have found no difference [12]. In our experience, patients who have not undergone splenectomy do not tolerate our routine post-transplant immunosuppressive regimen (antilymphocyte globulin, azathioprine, and prednisone [2]) as well as do patients who have undergone splenectomy (unpublished observations). Unquestionably, splenectomy benefits patients with hypersplenism [13,14], but they comprise only 5 to 10 per cent of pretransplant patients; in the patient who has not had splenectomy, a 27 per cent incidence of post-transplant hypersplenism has been reported [15]. There have been no randomized studies on the benefits of pretransplant nephrectomy; however, the latest report of the Human Renal Transplant Registry shows that post-transplant two year survival in all disease categories is better in patients who have undergone nephrectomy than in those who have not [16]. This finding may reflect the fact that the largest and most experienced transplantation centers perform routine pretransplantation nephrectomy or that patients who could not tolerate the transplantation procedure itself are weeded out by pretransplantation bilateral nephrectomy and splenectomy. Patients who have undergone nephrectomy require increased numbers of transfusions, which increases the risk of having hepatitis develop and of being sensitized by foreign histocompatibility antigens [l’i]. The strongest arguments against routine nephrectomy and splenectomy, however, are the reported morbidity and mortality associated with the procedure [6,28]. Because of this association, a number of institutions now perform nephrectomy only in selected cases. Although nephrectomy has been advocated for certain dialysisrelated problems (polyneuropathy [19,20] and uncontrollable thirst [20]), all authorities agree that nephrectomy is indicated for hypertension uncontrolled by dialysis and medications, renal infections, ureteral reflux, polycystic kidneys, protein-
Volume
129.June
1975
Patients
uria with hypoproteinemia, and antiglomerular basement membrane antibody nephritis [4,18,2124]. Paradoxically, in our series the deaths occurred in just this group of patients who would be selected for nephrectomy at transplantation centers where routine pretransplantation nephrectomy is not performed. Ten of the twelve deaths (83 per cent) among the nondiabetic patients occurred in those who required nephrectomy by the foregoing criteria. Surgeons who perform selective nephrectomy believe that, if necessary, the patient’s own kidneys can be removed successfully post transplantation. Indeed, this can usually be carried out safely [18,25], although Husberg, Nilsson, and Fritz [26] have reported three cases in which post-transplantation removal of the patient’s own kidneys was associated with temporary elevation of the serum creatinine level. Review of our data has altered our approach to nephrectomy and splenectomy. Diabetic patients no longer undergo routine bilateral nephrectomy and splenectomy. The surgical mortality of 9.6 per cent is excessive, as is the mortality post.nephrectomy while the patient is being maintained on chronic hemodialysis (5.8 per cent). At present, bilateral nephrectomy and splenectomy are only carried out in diabetic patients who have one or more of the risk factors discussed previously. The remainder undergo transplantation with their kidneys in place, and transplantation is usually carried out while the serum creatinine levels are in the range of 5 to 8 mg/lOO ml. Splenectomy is performed at the time of transplantation [27]. Ail other patients continue to undergo routine pretransplantation BNS. Because we now recognize certain factors as high risk, we can take certain steps to reduce mortality. For example, in addition to monitoring the fluid and electrolyte balance of hypertensive patients before and after BNS, we also monitor them during chronic dialysis while they are awaiting transplantation. Preoperatively the severely hypertensive patient receives a slight fluid overload to avoid large pressure drops with removal of the kidneys. In addition, transplantation is carried out with minimal delay. The patient with a history of recurrent pyelonephritis or reflux is not operated on until cultures are sterile, and is treated with prophylactic antibiotics pre-, intra-, and postoperatively. With this approach our mortality in the last eighteen months has decreased from 4.7 per cent to 1.8 per cent before transplantation.
619
Matas et
al
Summary
Routine bilateral nephrectomy and splenectomy (BNS) in uremic patients before transplantation are relatively safe procedures except when there is pre-existing sepsis, diabetes, or severe hypertension. A review of 421 patients undergoing routine pretransplantation BNS reveals that death before transplantation occurs in two definable groups of patients. In our series, the first group, those with juvenile onset diabetes, have a 15.4 per cent pretransplantation mortality (9.6 per cent operative and 5.8 per cent nonoperative) while being maintained on hemodialysis and awaiting transplantation. The second group, nondiabetic patients with other preoperatively definable risk factors such as severe hypertension and infected kidneys, had a 3.25 per cent pretransplantation mortality (1.9 per cent operative and 1.25 per cent nonoperative) while on hemodialysis. Paradoxically, these same factors are used as absolute criteria for pretransplantation nephrectomy at institutions where this operation is not a routine part of the pretransplantation regimen.
References 1. Kjellstrand CM, Simmons RL. Buselmeier TJ, Najarian JS: Recipient selection, medical management, and dialysis (in renal transplantation), p 418. Transplantation (Najarian JS, Simmons RL, ed). Philadelphia, Lea & Febiger, 1972. 2. Simmons RL, Kjellstrand CM, Najarian JS: Technique, complications, and results (in renal transplantation), p 445. Transplantation (Najarian JS, Simmons RL, ed). Philadelphia, Lea & Febiger, 1972. 3. Kjellstrand CM, Buselmeier TJ: A simple method for anticoagulation during pre and postoperative hemodialysis, avoiding rebound phenomenon. Surgery 72: 630, 1972. 4. White N, Snowden SA, Parsons V, et al: The management of terminal renal failure in diabetic patients by regular dialysis therapy. Nephron 11: 26 1, 1973. 5. Leonard A, Comay C, Raij L, et al: The natural history of regularly dialyzed diabetics. Trans Am Sot Artif lntem Organs 19: 282, 1973. 6. Rosenberg JC, Azcarate J, Fleischmann LE, et al: Indications for pretransplant nephrectomy. Arch Surg 107: 233, 1973. 7. Kaufman HM, Swanson MK, McGregor WR, et al: Splenecto-
620
8.
9. 10. 11.
12.
13.
14.
15.
16.
17.
18. 19.
20.
21. 22.
23.
24. 25.
26.
27.
my in renal transplantation. Surg Gynecol Obstet 139: 33, 1974. Pierce JC, Hume DH: The effect of splenectomy on the survival of first and second renal homotransplants in man. Surg Gynecol Obstet 127: 1300, 1968. Bennett WM. Boileau M, Barry JM, et al: Role of splenectomy in renal transplantation. Urology 11: 542, 1973. Opelz G, Terasaki PI: Effect of splenectomy on human renal transplants. Transplanfation 15: 605, 1973. Woods JE, DeWeerd JH, Johnson WJ, Anderson CF: Splenectomy in renal transplantation. Influence on azathioprine sensitivity. JAMA 218: 1430, 1971. Veith FJ, Luck RJ, Murray JE: The effects of splenectomy on immunosuppressive regimes in dog and man. Surg Gyneco/Obstet 121: 299, 1965. Berne TV, Bischel MD, Payne JE, Barbour BH: Selective splenectomy in chronic renal failure. Am J Surg 126: 27 1, 1973. Hartley LCI, Morgan TO, lnnis MD, Clunie GIA: Splenectomy for anemia in patients on regular hemodialysis. Lancef 2: 1343,1971. Kauffman HN, Fox PS, Swanson MK, et al: Post transplant hypertension (abstr). Eighth Annual Meeting Assoc Acad Surg, 1974, p 61. Advisory Committee of the Renal Transplant Registry: The twelfth report of the Human Renal Transplant Registry. JAMA in press. Van Ypersele De Strihou C, Stragier A: Effect of bilateral nephrectomy on transfusion requirements of patients undergoing chronic dialysis. Lancet 2: 705. 1969. Cohen SL: Hypertension in renal transplant recipients: role of bilateral nephrectomy. Br Med J 3: 78, 1973. Popovtzer MM, Rosenbaum BJ. Gordon A, Maxwell MH: Relief of uremic polyneuropathy after bilateral nephrectomy. N Engl J Med 281: 949, 1969. Rogers PW, Kurtzman NA: Renal failure, uncontrollable thirst, and hyper-reninemia. Cessation of thirst with bilateral nephrectomy. JAMA 225: 1236, 1973. Mahoney JF, Gibson GR, Sheil AGR, et al: Bilateral nephrectomy for malignant hypertension. Lancet 1: 1036, 1972. Donahue JP, Bohnert WW, Shires DL, Bradley KP: Bilateral nephrectomy: its role in management of the malignant hypertension of end-stage renal disease. J Ural 106: 488, 1971. Siegel RR: The basis of pulmonary disease resolution after nephrectomy in Goodpasture’s syndrome. Am J Med Sci 259: 201,197o. Vertes V, Cangiano JI, Berman LB, Gould A: Hypertension in end-stage renal disease. N fngl J Med 280: 978, 1969. Mitchell TS, Halasz NA, Gittes RF: Renal transplantation: selective preliminary bilateral nephrectomy. J Ural 109: 796, 1973. Husberg B, Nilsson T, Fritz H: Acute rejection of allogenic transplanted kidneys following nephrectomy of autogenic pyelonephritic kidneys. Transplantation 11: 502, 1971, Kjellstrand CM, Shideman JR, Simmons RL, et al: Renal transplantation in insulin-dependent diabetic patients. KidneyInt6: 515, 1974.
The American Journal of Surgery
The routine removal of both kidneys and the spleen in preparation for kidney transplantation has become a controversial issue. The question of whether the benefits to post-transplant graft function outweigh the hazards of the operation (mortality) has yet to be answered. This paper reviews twenty pretransplantation deaths in 421 patients
undergoing bilateral nephrectomy
and splenecto-
my (BNS). We have analyzed the causes of death and the preoperatively definable risk factors in an effort to minimize the risk of this procedure in the future. Methods and Material Bilateral nephrectomy and splenectomy are performed routinely in all patients accepted to the University of Minnesota Transplantation Program. Before BNS, each patient receives intensive medical care, including at least one week of hemodialysis to stabilize his condition. Hemodialysis is always performed forty-eight and twenty-four hours before nephrectomy so that dialysis, with its inherent need for heparin anticoagulation, will not be necessary in the immediate postoperative period [I]. The evening before the operation, a long arm intravenous line is placed to monitor central venous pressure. At the same time, hematocrit, serum electrolyte levels, and blood sugar values are determined to ensure that the patient has a satisfactory fluid and electrolyte balance. BNS is performed through a transabdominal midline incision [2]. Postoperatively, the patient reFrom the Departmentsof Surgery
and Medicine, University of Minnesota, Minneapolis. Minnesota. This work was supported by USPHS Grant AM 13063. Reprint requests should be addressed to C. M. Kjelistrand, MD. Box 123-Mayo Memorial Bullding. University of Minnesota, Minneapolis. Minnesota 55455.
816
ceives intensive nursing care for twenty-four hours, with frequent monitoring of vital signs and central venous pressure. Electrolytes, hemoglobin, hematocrit, and glucose levels (in diabetic patients) are monitored at four to six hour intervals for the first forty-eight hours or until they are stable. Ambulation is begun the evening of operation, and the patient is started on solid foods on the third or fourth postoperative day. The patient usually receives his first dialysis under regional heparinization on the second or third postoperative day [3]. From January 1, 1968 to July 30, 1974, 421 pretransplant bilateral nephrectomies and splenectomies were performed; twenty deaths (4.8 per cent) occurred prior to renal transplantation. Seven deaths occurred within the first thirty postoperative days, and five additional deaths were attributed to the operative procedure itself, giving an overall surgical mortality of 2.8 per cent. (Tables I, II, and III.) Results Mortality in Diabetic Patients. There were eight deaths in fifty-two diabetic patients (15.4 per cent) 3 to 120 days after operation. (Table I.) Five of these deaths (9.6 per cent) must be considered operative deaths since they occurred within the first thirty postoperative days, although one of these occurred after the patient was discharged. Three other diabetic patients died 44 to 120 days after operation. Their deaths were not directly related to the procedure. Four of the eight deaths in the diabetic patients were secondary to myocardial infarction, and three of these patients died after they were discharged from the hospital and while they were awaiting transplantation. All four of these patients had severe hypertension before BNS and two had had
The Amarlcan Journal of Surgery
BNS in Hemodialyzed
TABLE I __-.
Mortality
&a (yr) and Sex
Patient
in Diabetic
Patients
____
Dialysis Time before BNS ____-
~_
Patients
_~ _ _~ Post-
Risk Factors
operative Day of Death
Cause of Death
--
Operative Mortality 1
36,M
4 mo
2
53,M
4mo
3 4
35,M 32, F
1 yr 1 mo
5
45,F
4mo
Severe hypertension, old myocardial infarction Hypertension, old myocardial infarction, old cerebrovascular accident, congestive heart failure Moderate hypertension Moderate hypertension, hypothyroidism Hypertension, hypothyroidism
Myocardial
infarction
Myocardial
infarction
3
embolus
30 7 11
Septicemia Septicemia Pulmonary
after discharge
Nonoperative Mortality 6
36, F
2mo
Severe
7
28,M 32.M
1 yr 4mo
Severe hypertension
8
TABLE II
Patient
Mortality
&a (yr) and Sex
hypertension
-____ Myocardial infarction, pericardial effusion after discharge Myocardial infarction after discharge Fluid overload
in Nondiabetic
21
on dialysis
60 44 120
Patients
Underlying Disease
Dialysis Time before BNS
Risk Factors
Cause of Death
Postoperative Day of Death
Operative Mortality 1
32,M
Chronic
pyelonephritis
2mo
2
32,F
Chronic
pyelonephritis
2wk
3 4 5 6 7
25,M 35,M 54,M 48,M 47,M
Radiation nephritis Chronic glomerulonephritis Polycystic kidneys Chronic pyelonephritis Hemolytic uremic syndrome
2wk 2wk 2 yr 3mo 2 mo
lleal loop, infected kidneys Infected kidneys, perinephric abscess Morbid obesity Severe hypertension Polycystic kidneys Infected kidneys Severe hypertension
Drug hypersensitivity
33
Sepis
31
Sepsis Myocardial infarction Sep&s Sepsis Technical complications (retroperitoneal hemorrhage)
38 3 40 45 7
Cerebra! bleeding Infection 1 yr post BNS Myocardial infarction after discharge Pericardial effusion
35 360 31
Nonoperative Mortality 8 9 10
45,F 50, F 30,M
Malignant nephrosclerosis Malignant nephrosclerosis Chronic glomerulonephritis
3mo 2 yr 4mo
Severe hypertension Severe hypertension Severe hypertension
11
29, F
Chronic
2mo
Systemic
6 mo
erythematous Hypoproteinemia age of 18 mo
12
1.5,M
glomerulonephritis
Congenital
nephrotic
syndrome
myocardial infarcts. Two other deaths in the diabetic patients were secondary to infection and one was due to a pulmonary embolus. The eighth death was caused by fluid overload while the patient was being maintained on hemodialysis and awaiting a cadaver donor kidney.
Vohmo
129, June 1975
lupus at
Cerebral bleeding, bleeding diathesis
180 36
It is difficult to distinguish any specific risk factors in the diabetic patient. These patients are uniformly hypertensive and most of them have pre-existing cardiovascular disease. In addition, most of these patients had reported previous episodes of urinary tract infection.
617
Matas et
TABLE III
al
Cause of Death in Diabetic Patients
and Nondiabetic
Patients Cause of Death Cardiovascular Myocardial infarction Cerebral bleeding Infection Complication of dialysis Other Pulmonary embolus Drug hypersensitivity Technical problem Total deaths Number of patients operated on Pretransplantation mortality (%) Operative mortality (%)
Diabetic
Nondiabetic
4
2 2 5 1
2 1 1
8 52 15.4 9.6
1 1 12 369 3.25 1.9
Mortality in Nondiabetic Patients. Of the nondiabetic patients, 12 of the 369 patients (3.25 per cent) who underwent BNS died before transplantation (compared with 15.4 per cent of the diabetic patients). (Tables II and III.) Two of these deaths (0.6 per cent) occurred within the thirty postoperative days traditionally related to operative mortality statistics; however, seven of the twelve deaths (1.9 per cent) were actually related to intraor postoperative complications. Four of the twelve deaths (33 per cent) were attributed to sepsis within sixty days after BNS, and a fifth death was caused by infection one year postoperatively. Two patients died after cerebral hemorrhage. Complications of dialysis, retroperitoneal hemorrhage, and a drug hypersensitivity reaction each resulted in one death. Myocardial infarction was responsible for the death of two of twelve nondiabetic patients (17 per cent). Both patients (numbers 4 and 10, Table II) were young men, thirty and thirty-five years old, with severe, uncontrollable hypertension. Eleven nondiabetic patients who died (91.6 per cent) had preoperatively definable factors that placed them in a high risk group. Five patients had uncontrollable hypertension, three had recurrent pyelonephritis, one patient had systemic lupus erythematosus, another, an eighteen month old child, had hypoproteinemia, and one patient had morbid obesity. Contrary to our findings in the diabetic patients, there is an evident cause and effect relationship between risk factors and death in a majority of these nondiabetic patients. Vascular catastrophes were responsible for three deaths; all
616
three of these patients had severe hypertension. This can be compared with an incidence of uncontrollable hypertension of approximately 10 per cent in our dialyzed population. Death was secondary to sepsis in five patients, three of whom had infected kidneys (one was operated on because of a perinephric abscess) and one of whom had morbid obesity. One death was due to a cerebral hemorrhage in a child with a bleeding diathesis secondary to a congenital nephrotic syndrome. There was no direct correlation between a risk factor and the cause of death in four patients. One of these deaths resulted from a technical complication. In one patient on dialysis while awaiting transplantation, pericardial effusion and tamponade developed after six months, and one patient died of drug hypersensitivity. The fourth patient died from infection after another operative procedure one year after BNS.
Comments Our data indicate that death after BNS and before kidney transplantation occurs in two preoperatively definable groups of patients. The first group, those with juvenile onset diabetes, have a 15.4 per cent pretransplantation mortality (9.6 per cent operative and 5.8 per cent nonoperative). Fifty per cent of these deaths were caused by myocardial infarction that occurred either in the immediate postoperative period or while the patient was being maintained on hemodialysis and awaiting transplantation. This high incidence of fatal cardiovascular complications is consistent with the fate of this group of patients on maintenance hemodialysis [4,5]. The second group that can be defined preoperatively are patients whose kidney disease or associated diseases (that is, severe hypertension, infected kidneys, or both) place them in a high risk group. The cause of death in these patients is closely related to their preoperative high risk factors. The severely hypertensive patients died from myocardial infarction or intracerebral hemorrhage, whereas patients with chronic pyelonephritis and perinephric abscess died from sepsis. Paradoxically, it is these patients in whom pretransplant nephrectomy is indicated at institutions that do not perform routine pretransplant nephrectomy and splenectomy [6]. The question of whether to perform routine bilateral nephrectomy and splenectomy before transplantation remains unanswered. Splenectomy is advocated as a supportive immunosuppressant
The American Journal of Surgery
BNS in Hemodialyzed
and, in addition, to allow administration of larger amounts of immunosuppressive drugs. However, several researchers diversely suggest that splenectomy (1) influences graft survival [7], (2) has no effect on first graft survival but is important in second graft survival [8], or (3) has no effect at all on graft survival 19,I 01. Whether splenectomy does affect tolerance to azathioprine is also still controversial; some investigators report that the patient who has undergone splenectomy can better tolerate the drugs [II], whereas other authors have found no difference [12]. In our experience, patients who have not undergone splenectomy do not tolerate our routine post-transplant immunosuppressive regimen (antilymphocyte globulin, azathioprine, and prednisone [2]) as well as do patients who have undergone splenectomy (unpublished observations). Unquestionably, splenectomy benefits patients with hypersplenism [13,14], but they comprise only 5 to 10 per cent of pretransplant patients; in the patient who has not had splenectomy, a 27 per cent incidence of post-transplant hypersplenism has been reported [15]. There have been no randomized studies on the benefits of pretransplant nephrectomy; however, the latest report of the Human Renal Transplant Registry shows that post-transplant two year survival in all disease categories is better in patients who have undergone nephrectomy than in those who have not [16]. This finding may reflect the fact that the largest and most experienced transplantation centers perform routine pretransplantation nephrectomy or that patients who could not tolerate the transplantation procedure itself are weeded out by pretransplantation bilateral nephrectomy and splenectomy. Patients who have undergone nephrectomy require increased numbers of transfusions, which increases the risk of having hepatitis develop and of being sensitized by foreign histocompatibility antigens [l’i]. The strongest arguments against routine nephrectomy and splenectomy, however, are the reported morbidity and mortality associated with the procedure [6,28]. Because of this association, a number of institutions now perform nephrectomy only in selected cases. Although nephrectomy has been advocated for certain dialysisrelated problems (polyneuropathy [19,20] and uncontrollable thirst [20]), all authorities agree that nephrectomy is indicated for hypertension uncontrolled by dialysis and medications, renal infections, ureteral reflux, polycystic kidneys, protein-
Volume
129.June
1975
Patients
uria with hypoproteinemia, and antiglomerular basement membrane antibody nephritis [4,18,2124]. Paradoxically, in our series the deaths occurred in just this group of patients who would be selected for nephrectomy at transplantation centers where routine pretransplantation nephrectomy is not performed. Ten of the twelve deaths (83 per cent) among the nondiabetic patients occurred in those who required nephrectomy by the foregoing criteria. Surgeons who perform selective nephrectomy believe that, if necessary, the patient’s own kidneys can be removed successfully post transplantation. Indeed, this can usually be carried out safely [18,25], although Husberg, Nilsson, and Fritz [26] have reported three cases in which post-transplantation removal of the patient’s own kidneys was associated with temporary elevation of the serum creatinine level. Review of our data has altered our approach to nephrectomy and splenectomy. Diabetic patients no longer undergo routine bilateral nephrectomy and splenectomy. The surgical mortality of 9.6 per cent is excessive, as is the mortality post.nephrectomy while the patient is being maintained on chronic hemodialysis (5.8 per cent). At present, bilateral nephrectomy and splenectomy are only carried out in diabetic patients who have one or more of the risk factors discussed previously. The remainder undergo transplantation with their kidneys in place, and transplantation is usually carried out while the serum creatinine levels are in the range of 5 to 8 mg/lOO ml. Splenectomy is performed at the time of transplantation [27]. Ail other patients continue to undergo routine pretransplantation BNS. Because we now recognize certain factors as high risk, we can take certain steps to reduce mortality. For example, in addition to monitoring the fluid and electrolyte balance of hypertensive patients before and after BNS, we also monitor them during chronic dialysis while they are awaiting transplantation. Preoperatively the severely hypertensive patient receives a slight fluid overload to avoid large pressure drops with removal of the kidneys. In addition, transplantation is carried out with minimal delay. The patient with a history of recurrent pyelonephritis or reflux is not operated on until cultures are sterile, and is treated with prophylactic antibiotics pre-, intra-, and postoperatively. With this approach our mortality in the last eighteen months has decreased from 4.7 per cent to 1.8 per cent before transplantation.
619
Matas et
al
Summary
Routine bilateral nephrectomy and splenectomy (BNS) in uremic patients before transplantation are relatively safe procedures except when there is pre-existing sepsis, diabetes, or severe hypertension. A review of 421 patients undergoing routine pretransplantation BNS reveals that death before transplantation occurs in two definable groups of patients. In our series, the first group, those with juvenile onset diabetes, have a 15.4 per cent pretransplantation mortality (9.6 per cent operative and 5.8 per cent nonoperative) while being maintained on hemodialysis and awaiting transplantation. The second group, nondiabetic patients with other preoperatively definable risk factors such as severe hypertension and infected kidneys, had a 3.25 per cent pretransplantation mortality (1.9 per cent operative and 1.25 per cent nonoperative) while on hemodialysis. Paradoxically, these same factors are used as absolute criteria for pretransplantation nephrectomy at institutions where this operation is not a routine part of the pretransplantation regimen.
References 1. Kjellstrand CM, Simmons RL. Buselmeier TJ, Najarian JS: Recipient selection, medical management, and dialysis (in renal transplantation), p 418. Transplantation (Najarian JS, Simmons RL, ed). Philadelphia, Lea & Febiger, 1972. 2. Simmons RL, Kjellstrand CM, Najarian JS: Technique, complications, and results (in renal transplantation), p 445. Transplantation (Najarian JS, Simmons RL, ed). Philadelphia, Lea & Febiger, 1972. 3. Kjellstrand CM, Buselmeier TJ: A simple method for anticoagulation during pre and postoperative hemodialysis, avoiding rebound phenomenon. Surgery 72: 630, 1972. 4. White N, Snowden SA, Parsons V, et al: The management of terminal renal failure in diabetic patients by regular dialysis therapy. Nephron 11: 26 1, 1973. 5. Leonard A, Comay C, Raij L, et al: The natural history of regularly dialyzed diabetics. Trans Am Sot Artif lntem Organs 19: 282, 1973. 6. Rosenberg JC, Azcarate J, Fleischmann LE, et al: Indications for pretransplant nephrectomy. Arch Surg 107: 233, 1973. 7. Kaufman HM, Swanson MK, McGregor WR, et al: Splenecto-
620
8.
9. 10. 11.
12.
13.
14.
15.
16.
17.
18. 19.
20.
21. 22.
23.
24. 25.
26.
27.
my in renal transplantation. Surg Gynecol Obstet 139: 33, 1974. Pierce JC, Hume DH: The effect of splenectomy on the survival of first and second renal homotransplants in man. Surg Gynecol Obstet 127: 1300, 1968. Bennett WM. Boileau M, Barry JM, et al: Role of splenectomy in renal transplantation. Urology 11: 542, 1973. Opelz G, Terasaki PI: Effect of splenectomy on human renal transplants. Transplanfation 15: 605, 1973. Woods JE, DeWeerd JH, Johnson WJ, Anderson CF: Splenectomy in renal transplantation. Influence on azathioprine sensitivity. JAMA 218: 1430, 1971. Veith FJ, Luck RJ, Murray JE: The effects of splenectomy on immunosuppressive regimes in dog and man. Surg Gyneco/Obstet 121: 299, 1965. Berne TV, Bischel MD, Payne JE, Barbour BH: Selective splenectomy in chronic renal failure. Am J Surg 126: 27 1, 1973. Hartley LCI, Morgan TO, lnnis MD, Clunie GIA: Splenectomy for anemia in patients on regular hemodialysis. Lancef 2: 1343,1971. Kauffman HN, Fox PS, Swanson MK, et al: Post transplant hypertension (abstr). Eighth Annual Meeting Assoc Acad Surg, 1974, p 61. Advisory Committee of the Renal Transplant Registry: The twelfth report of the Human Renal Transplant Registry. JAMA in press. Van Ypersele De Strihou C, Stragier A: Effect of bilateral nephrectomy on transfusion requirements of patients undergoing chronic dialysis. Lancet 2: 705. 1969. Cohen SL: Hypertension in renal transplant recipients: role of bilateral nephrectomy. Br Med J 3: 78, 1973. Popovtzer MM, Rosenbaum BJ. Gordon A, Maxwell MH: Relief of uremic polyneuropathy after bilateral nephrectomy. N Engl J Med 281: 949, 1969. Rogers PW, Kurtzman NA: Renal failure, uncontrollable thirst, and hyper-reninemia. Cessation of thirst with bilateral nephrectomy. JAMA 225: 1236, 1973. Mahoney JF, Gibson GR, Sheil AGR, et al: Bilateral nephrectomy for malignant hypertension. Lancet 1: 1036, 1972. Donahue JP, Bohnert WW, Shires DL, Bradley KP: Bilateral nephrectomy: its role in management of the malignant hypertension of end-stage renal disease. J Ural 106: 488, 1971. Siegel RR: The basis of pulmonary disease resolution after nephrectomy in Goodpasture’s syndrome. Am J Med Sci 259: 201,197o. Vertes V, Cangiano JI, Berman LB, Gould A: Hypertension in end-stage renal disease. N fngl J Med 280: 978, 1969. Mitchell TS, Halasz NA, Gittes RF: Renal transplantation: selective preliminary bilateral nephrectomy. J Ural 109: 796, 1973. Husberg B, Nilsson T, Fritz H: Acute rejection of allogenic transplanted kidneys following nephrectomy of autogenic pyelonephritic kidneys. Transplantation 11: 502, 1971, Kjellstrand CM, Shideman JR, Simmons RL, et al: Renal transplantation in insulin-dependent diabetic patients. KidneyInt6: 515, 1974.
The American Journal of Surgery