Cyclosporine's effect on canine pancreatic endocrine function.

This study was designed to investigate the mechanism, and reversibility, of glucose intolerance following the acute administration of cyclosporine (CsA) in a canine model. Three groups underwent a baseline intravenous glucose tolerance test (IVGTT; 0.5 g/kg of glucose), with simultaneous insulin determinations. In groups A and B, repeat IVGTTs were performed, at three-day intervals, after 2, 4, and 6 mg/kg of intravenous CsA + solvent (group A; n = 8), or the solvent alone (group B; n = 5). Repeat IVGTTs were performed in group A, 24 and 72 hr after the last CsA infusion. In group C (n = 5), IVGTTs were performed, 4, 24, 48, and 72 hr after oral CsA (12.5 mg/kg). In each group, the rate of glucose clearance (k value,--per cent min), and basal-to-peak insulin difference (uU/ml), for each IVGTT were compared with the baseline results. In group A, the basal-to-peak insulin difference was significantly lower than baseline (81.9 +/- 13.6) after 2 mg/kg (27.3 +/- 3.1; P less than 0.005), 4 mg/kg (22.7 +/- 3.7; P less than 0.001); and 6 mg/kg (16.8 +/- 3.2; P less than 0.001) of CsA, and returned to baseline within 24 hr (81.4 +/- 3.7). Corresponding K values were also significantly different in group A. In group B, there were no significant differences in these parameters from controls, at the equivalent doses of the solvent alone. At 4 hr after oral CsA (group C), there was a reduction in the basal-to-peak insulin difference (37.2 +/- 9.1 vs. 22.5 +/- 4.1) and K values (-3.20 +/- 0.4 vs. -1.96 +/- 0.3), with the change in K values being statistically significant (P less than 0.05). A return to baseline levels was present at 24 h. This study demonstrates that, in the canine model, therapeutic doses of intravenous and oral CsA acutely impair glucose regulation. This acute effect is secondary to decreased peripheral insulin levels, is reversible at 24 h, and is not evident with CsA solvent alone. The mechanism of decreased insulin secretion following CsA administration requires further elucidation.

The effect of terbutaline on exocrine function in the denervated canine pancreas.

This study was designed to evaluate the effects of the beta adrenergic agonist, terbutaline, on pancreatic exocrine secretion in the denervated canine pancreas. In vitro assessment was performed by evaluating the effect of terbutaline on 10(-8) OP-CCK stimulated amylase release of pancreatic tissue slices incubated at 37 degrees C in Krebs-bicarbonate media. In vivo assessment was accomplished in animals with pancreatic autografts and functioning pancreaticocystostomies, by evaluating the effect of intravenous terbutaline (0.075 mg/kg over 15 min) on the basal, and OP-CCK (125 ng/kg/hr)--stimulated, rate of secretion of urinary (autograft) amylase and bicarbonate. Incubation of tissue slices with terbutaline had no significant effect on OP-CCK-stimulated amylase release. The intravenous terbutaline infusion resulted in a decrease in the basal rate of amylase (U/min) and bicarbonate (mmol/min) secretion, with the bicarbonate inhibition being significantly decreased, when compared with controls (0.073 +/- .04 vs. 000 +/- .00; P less than 0.05). Following the terbutaline infusion, there was also a significant decrease in OP-CCK-stimulated amylase (140.3 +/- 23.3 vs. 24.6 +/- 11.9; P less than 0.005) and bicarbonate release (.069 +/- .03 vs. .003 +/- .001; P less than 0.05). This inhibition persisted until the study was terminated 3 hr after the terbutaline infusion. These studies demonstrate that terbutaline causes a significant and prolonged decrease in autograft exocrine secretion--and, as a result, may have a therapeutic role in reducing the exocrine complications associated with pancreatic transplantation. The mechanism of action of this agent in the denervated pancreas requires further elucidation.

Enteral pancreatic enzyme feedback inhibition of the exocrine secretion of the human transplanted pancreas.

The mechanism of regulation of negative feedback inhibition of the exocrine pancreas and its possible role in decreasing the exocrine secretion of the grafted human pancreas is unknown. To evaluate this we studied the effect of oral pancreatic enzymes on the stimulated transplanted pancreatic exocrine secretion in eight patients with allograft pancreaticocystostomies. After an 8-hr fast, all graft exocrine secretions via graft stent, fistula, and urinary anastomosis were collected for a 1-hr basal period. A standard 300-ml Lundh test meal was then ingested, and all exocrine secretions were collected in 30-min intervals for 3 hr. This test was repeated with 6 capsules of pancrelipase (24,000 units of lipase, 120,000 units of amylase, and 150,000 units of protease) given with the Lundh test meal. Stent, urine and fistula volume, amylase, and pH were measured for each collection period. The total 3-hr amylase secreted after the test meal and the test meal plus pancrelipase were compared. The period of peak amylase secretion after the test meal alone was compared with the same period after the test meal plus pancrelipase and the premeal basal period. The total amylase decreased 34% from 5550 +/- 1000 to 3680 +/- 740 IU/3 hr (P < .03) with pancrelipase. The peak amylase secretion decreased 63% from 1520 +/- 271 to 567 +/- 185 IU/30 min (P < .02) with the addition of pancrelipase to the test meal. Pancrelipase eliminated all meal-stimulated amylase secretion with the mean secretion 16% below the basal secretion of 674 +/- 117 IU/30 min. We conclude that pancreatic negative feedback inhibition significantly decreases meal-stimulated and basal exocrine secretion in the transplanted human pancreas.

Allograft viability determined by enzyme analysis.

Enzymatic analysis of the venous effluent of ischemically injured kidney failed to predict accurately the ability of an isochemically injured kidney to support life. Postoperative serum assay of lactic dehydrogenase (LDH) is of value in the assessment of the functional status of the kidney and correlates with response of the rejection episode to immunosuppression. However, by itself it cannot be a sole guide to withholding of therapy. Successful treatment is associated with a decline in LDH level, and failure to return to base line serves as a guide to irreversibility of the rejection reaction.

Cavoportal hemitransposition in liver transplantation.

Over the last decade a large number of patients with portal vein thrombosis have undergone successful liver transplantation. In most of these patients, simple modifications in vascular reconstruction techniques are adequate. However, anastomosis of the donor portal vein may not be possible in the presence of extensive portal and superior mesenteric venous thrombosis and in the absence of any other large tributary of the portal venous system. Cavoportal hemitransposition has been described as a salvage technique under these circumstances. We report a 43-year-old patient who underwent such a procedure and remains well 1 year later. We review the literature and discuss the implications of cavoportal hemitransposition.

Evaluation of the thromboxane A2 synthetase inhibitor OKY-046 in a warm ischemia-reperfusion rat model.

The pathophysiology of ischemia-reperfusion renal injury is mediated, in part, by the generation of the vasoconstricting prostanoid thromboxane A2 (TXA2). This study was undertaken to evaluate the renoprotective effects, as well as the optimal timing and dosage, of a selective thromboxane synthetase inhibitor, OKY-046, in a unilateral nephrectomized, 60 min ischemia, 72 hr reperfusion, rodent model. Forty-one rats were subjected to right nephrectomy only (group A), or right nephrectomy with 60 min of left renal ischemia and treatment with inactive vehicle only (group B), or 2 mg/kg or 4 mg/kg of OKY-046 administered intravenously before (groups C and D) or after (groups E and F) pedicle clamping. Outcome variables included animal survival; change in kidney weight; 0, 24, and 72 hr plasma creatinine (CR); urea nitrogen (BUN); thromboxane B2 (TXB2) and 6-keto prostaglandin F(1alpha) (6 kPGF(2alpha)) levels; creatinine clearance (CRCL); and histologic evidence of renal injury. Animal survival and postperfusion kidney weight were not significantly different among the groups. However, renal functional parameters were significantly improved with the 2 mg/kg dose of OKY-046 administered after renal ischemia. (group B 72 hr Cr= 8.01 +/- 1.1 mg% vs. group E=3.99 +/- 1.5 mg%, and group B 72 hr BUN=241.3 +/- 32.8 mg% vs. group E=52.6 +/- 22.5 mg%). The CRCL was also improved in group E vs. group B, although these results did not reach statistical significance (group B=0.069 ml/min vs. group E=0.194 ml/ min). The 24 hr TXB2 levels were significantly increased in group B (0 hr=754.1 +/- 219.4 pg/ml vs. 24 hr=2055.9 +/- 550.0 pg/ml), and pre- or posttreatment with OKY-046 abrogated this increase (group C 0 hr=517.1 +/- 80.9 pg/ml vs. 24 hr=384.7 +/- 251.5 pg/ml, and group E 0 hr=781.6 +/- 390.4 pg/ml vs. 24 hr=183.0 +/- 81.4 pg/ml). The 24 hr 6 kPGF(1alpha) levels decreased in all groups, whereas 72 hr 6 kPGF(1alpha) levels increased above baseline in groups A, C, and E, but not in group B. These data demonstrate the beneficial effects of thromboxane A2 synthesis inhibition in the setting of ischemia-reperfusion injury and suggest that this renoprotection correlates with late vasodilatory prostanoid synthesis.

The effect of octreotide acetate on meal-stimulated exocrine secretion in canine pancreatic autografts.

Octreotide acetate (Sandostatin), a long-acting somatostatin analogue, has been demonstrated to have an inhibitory effect on exocrine secretion in the neurally intact pancreas. This study was designed to evaluate the effect of this agent on exocrine secretion in the denervated canine pancreas, utilizing animals with pancreatic autografts and functioning pancreaticocystostomies. The rates of secretion of urinary (autograft) amylase (units/min) and bicarbonate (mM/min), over a five-hr interval, were determined in the basal state (group A, n = 10), after a bolus injection of 400 micrograms of Sandostatin (group B, n = 5), after a standard meal (group C, n = 5), or a meal preceded by 400 micrograms of Sandostatin (group D, n = 5). Basal secretion of amylase was decreased for 4 hr following Sandostatin, although this decrease was not significant. Conversely, basal bicarbonate secretion was not inhibited by Sandostatin. When compared with group C (22.4 +/- 3.2), a significant inhibition of meal-stimulated amylase release was demonstrated in group D (5.4 +/- 0.21, P = 0.0006) during the first hour after Sandostatin was given. This inhibition remained significant at 2 hr (group C = 38.5 +/- 5.2 versus group D = 9.4 +/- 0.8; P = 0.0006) and 3 hr (group C = 38.6 +/- 6.3 versus group D = 17.5 +/- 0.9; P = 0.0108) after Sandostatin was given. In addition, meal-stimulated bicarbonate secretion was significantly inhibited for 2 hr following Sandostatin (group C = 0.19 +/- 0.03 versus group D = 0.07 +/- 0.02, P = 0.0096; and group C = 0.23 +/- 0.03 versus group D = 0.10 +/- 0.01, P = 0.0018, respectively). These studies demonstrate that Sandostatin has a profound inhibitory effect on meal-stimulated enzyme and bicarbonate release in a denervated canine autograft model. Although the site of action of this agent remains to be defined, Sandostatin may have therapeutic potential in clinical pancreas transplantation.

Renoprotective effects of the 21-aminosteroid U74389G in ischemia-reperfusion injury and cold storage preservation.

Free radical mediated lipid peroxidation (LPO) has been implicated in the pathogenesis of ischemic-reperfusion injury (IRI). To address the renoprotective effect(s) of LPO inhibition, the efficacy of the 21 aminosteroid U74389G was evaluated in three IRI models. In Model 1 51 unilateral nephrectomized rats that underwent 60 min of warm ischemia followed by a 72-hr reperfusion interval were treated with the test vehicle only, or 3, 6, or 12 mg/kg of U74389G intravenously, 5 min pre- or postischemia. In Model 2 Sprague-Dawley rats underwent sham operation (n=9), or 45 min of warm ischemia and 10 min of reperfusion with U74389G (6 mg/kg; n=10) or test vehicle only (n=10) administered intravenously over 10 min beginning 5 min prior to clamp release. After reperfusion, LPO was determined by assay of snap frozen tissue for thiobarbituric acid (TBA) concentrations (nmol/g tissue weight). In Model 3 domestic lean maid pigs (14-18 kg) underwent left nephrectomy with 30 min of warm ischemia, Collins C-4 flush, and 24 hr of cold storage preservation. Heterotopic autotransplantation and immediate contralateral nephrectomy was then performed in Group A-nonischemic controls (n=4), Group B-ischemic controls (n=5), and Group C-U74389G (6 mg/kg) administered preischemia and at autotransplantation (n=5). In Model 1 maximal renoprotection was demonstrated with the 6 mg/kg dose of U74389G administered after ischemia (ischemic control 72-hr serum creatinine (Cr) = 8.01+/-1.1 mg% vs. 3.32+/-0.96 mg%; ischemic control creatinine clearance = 0.069+/-0.03 ml/min vs. 0.206+/-0.04 ml/min; P<0.05). In Model 2 TBA levels were significantly lower in U74389G treated animals (88.5+/-10.0 vs. ischemic controls = 296.8+/-81.4; P=0.02). In Model 3 graft survivals were 100%, 0%, and 60% respectively. Peak Cr and BUN (mg%) were significantly greater in Group C vs. Group A, (Group A Cr = 8.59+/-0.63 vs. Group C = 12.8+/-1.01; Group A BUN = 64.1+/-2.73 vs. Group C = 104.9+/-12.21)--however, by day 10, thee were no significant differences in renal function: (Group A Cr = 2.15+/-0.3 vs. Group C = 2.10+/-0.06; Group A BUN = 27.0+/-6.0 vs. Group C = 31.1+/-6.4). These results support the beneficial effects of LPO inhibitors in models of ischemia-reperfusion, as well as preservation/transplantation, and suggest that this renoprotection correlates with decreased membrane lipid peroxidation.

Spontaneous rupture of the liver upon revascularization during transplantation.

Spontaneous rupture of the liver has been described in association with many benign and malignant conditions. We report, to our knowledge, the first case of spontaneous rupture of the liver upon revascularization, requiring total hepatectomy and portocaval shunt, followed by successful retransplantation. Routine pathological examination of the explanted liver failed to reveal the etiology of the rupture. However, electron microscopy demonstrated abnormal collagen in the hepatic arterial wall compatible with a collagen disorder such as Ehlers-Danlos type IV disease. We conclude that the donor liver had a previously undiagnosed collagen disorder. Review of the literature does not preclude the use of livers from donors with a history of connective tissue disorders. Based on our experience one should exercise caution when using livers from such donors. With a history of connective tissue disorder in an immediate family member, further tests should be performed in the donor to rule out a subclinical connective tissue disorder. In addition, a review of all patients reported thus far to have undergone total hepatectomy and portocaval shunt, followed by liver transplantation as a two-stage procedure is presented.

Prognostic value of intravenous dipyridamole thallium imaging in patients with diabetes mellitus considered for renal transplantation.

Patients with diabetes and end-stage renal failure are known to have a high risk for cardiac morbidity and mortality associated with renal transplantation. The most efficient method to determine preoperative cardiac risk has not been established. To determine the effectiveness of intravenous dipyridamole thallium imaging in predicting cardiac events, 40 diabetic renal transplant candidates were studied preoperatively in a prospective trial. The study group consisted of 40 patients whose average age was 42 years (range 27 to 64); 34 (85%) were hypertensive and 21 (53%) were cigarette smokers. Cardiac history included chest pain in 6 patients and prior myocardial infarction in 3 patients. Dipyridamole thallium imaging showed reversible defects in 9 patients, fixed defects in 8 patients and normal scans in 23 patients. Dipyridamole thallium imaging was performed using 0.56 mg/kg of dipyridamole infused intravenously over 4 minutes. Cardiac events occurred only in patients with reversible thallium defects, of which there were 6. Of these 6 patients, 3 had cardiac events before transplantation and 3 had them in the early postoperative phase (within 6 weeks of surgery). Of 21 patients who underwent renal transplantation, 3 had cardiac events within 6 weeks of transplantation. The average duration of follow-up was 11 months (range 1 to 21). Thus, dipyridamole thallium imaging is an effective method of identifying renal transplant candidates likely to develop cardiac complications. Routine coronary angiography may not be necessary to screen all renal transplant candidates for coronary artery disease before surgery.

Significance of the in-training examination in a surgical residency program.

The exact role that the American Board of Surgery In-Training Examination plays in resident evaluation remains poorly defined. We have required that all residents take the In-Training Examination annually. An analysis was performed of the results of the In-Training Examination and the Qualifying Examination of the American Board of Surgery for 16 residents who completed their residencies between July 1976 and July 1981. Twelve graduates passed their initial Qualifying Examinations with a mean +/- 1 SEM score of 81 +/- 2. Four graduates failed with a score of 70 +/- 0.3. Corresponding scores on their final In-Training Examinations were 42 +/- 9 and 11 +/- 6 (p less than 0.001). Each year the In-Training Examination Scores obtained by the residents who passed the Qualifying Examination were significantly higher than were those scores obtained by the graduates who failed. Linear regression analysis identified a significant correlation between the graduates' initial (r = 0.676) and final (r = 0.760) In-Training Examination scores and the Qualifying Examination score. In our resident training program, In-Training Examination results correlated well with Qualifying Examination results and may be used as an objective determinant for remedial measures and resident retentions. These data should be developed on a national level. While Board certification was or likely will be accomplished by all our residents, our goal is to strengthen the academic characteristics of the training program to produce uniform success on the initial Qualifying Examination.

Purulent pericarditis in children.

Acute purulent pericarditis was treated successfully in five children between the ages of 27 months and 11 1/2 years during the past 5 years. The responsible organism was Hemophilus influenzae, type b, in two cases and Meningococcus, Pneumococcus, and coagulase-positive Staphylococcus aureus in one case each. No primary source of infection could be identified in two patients. A high index of suspicion, combined with immediate echocardiograms and pericardiocentesis, led to the diagnosis. Immediate antibiotic therapy was instituted on the basis of the gram stain of the pericardial fluid. All five patients had a pericardial window established--four through subxyphoid approach and the fifth, because of a left pleural effusion, through a left thoracotomy. When the subxyphoid approach was used, sump drains were left for postoperative suction and irrigation. All five patients survived without sequalae during follow-up periods of from 18 months to 5 years. We advocate an aggressive approach to the diagnosis and treatment of this problem. This report documents the safety, ease, and effectiveness of the subxyphoid approach as a means of drainage.

Simultaneous cadaver renal and pancreas transplantation in type I diabetes.

Between February 1985 and April 1986, we performed 11 simultaneous cadaver kidney and segmental pancreatic transplants in patients with type I diabetes. There were nine men and two women ranging in age from 25 to 47 years (mean, 38.5 years). All pancreatic grafts were extraperitoneal, and the pancreatic duct was managed by pancreaticocystostomy utilizing an internal stent. Three patients died from two to six weeks postoperatively of septic complications. Four pancreatic grafts were functioning at 2, 5, 11, and 14 months after operation, and eight patients had had functioning renal allografts from two to 14 months (mean, 6.8 months) with a mean serum creatinine level of 2.4 mg/dL (210 mumol/L). Graft failure occurred in the other four patients from vascular thrombosis (three patients) or hemorrhagic pancreatitis (one patient). Significant morbidity included an infected arterial anastomosis (two patients), pancreatic fistulas (four patients), and bladder leak (four patients). In conclusion, this procedure is an effective option for selective diabetics with end-stage renal disease. Although technical complications were frequent, no adverse effect on renal allograft function was evident. With technical refinements, this procedure should be applicable to most type I diabetics with renal failure.

A five-year experience with the bovine heterograft for vascular access.

We retrospectively analyzed 385 bovine heterografts inserted in 331 patients receiving dialysis to determine patency and complication rates. Eighty-two grafts were inserted in diabetics, 303 in nondiabetics; 129 were primary grafts, 256 secondary. There were 251 "distal" grafts using the distal radial artery and 134 "proximal" grafts using the distal brachial artery (127 grafts) or the superficial femoral artery (seven grafts). During this five-year interval, 209 complications were identified with the most prevalent being thrombosis (160), aneurysms (18), and puncture site (18) or wound (eight) infections. An aggressive surgical approach to these complications prolonged graft patency, as evidenced by only 82 irreversible graft failures. The overall patency rate at one through four years was 79%, 69%, 63%, and 51% respectively. One- and two-year patency for diabetics (81%, 74%) vs nondiabetics (78%, 70%), primary (85%, 77%) vs secondary grafts (73%, 64%), and proximal (80%, 66%) vs distal grafts (75%, 64%) were not significantly different. Long-term patency was unaffected by diabetes, previous access procedures, or graft location.

Aseptic hip necrosis after renal transplantation.

To help determine the etiology of posttransplant aseptic hip necrosis, 11 stable renal allograft recipients (group A) who developed aseptic hip necrosis were compared with 89 patients (group B) without this complication. A comparison of mean age, duration of dialysis, mean daily prednisone dose, and incidence of rejection in the first year following transplant, sex, donor source, incidence of posttransplant parathyroidectomy, and mean serum calcium and alkaline phosphatase levels identified no significant differences between groups A and B. The mean serum creatinine value at three (2.2 +/- 0.31 mg/dL [190 +/- 30 mumol/L] vs 1.9 +/- 0.10 mg/dL [170 +/- 10 mumol/L]) and 12 (2.3 +/- 0.35 mg/dL [200 +/- 30 mumol/L] vs 1.9 +/- 0.10 mg/dL [170 +/- 10 mumol/L]) months and the serum phosphate value at three (3.0 +/- 0.19 mg/dL [0.97 +/- 0.06 mmol/L] vs 2.8 +/- 0.08 mg/dL [0.90 +/- 0.03 mmol/L]) and six (3.2 +/- 0.25 mg/dL [1.03 +/- 0.08 mmol/L] vs 2.9 +/- 0.25 mg/dL [0.94 +/- 0.08 mmol/L]) months were significantly greater in group A. Eight patients in group A underwent 13 total hip replacement an average of 16.5 +/- 3.1 months following transplant without significant complications. In conclusion, posttransplant aseptic hip necrosis occurs frequently, and renal allograft dysfunction may contribute significantly to its pathogenesis. When indicated, total hip replacement is both safe and effective.

Selective posttransplantation bilateral native nephrectomy. Indications and results.

Twenty-five patients underwent bilateral native nephrectomy one to 68 months (mean, 15.6 months) following renal transplantation. The indications were erythrocytosis in two patients, recurrent urinary tract infection in three, medically uncontrolled hypertension in 18, and hypertension and urinary tract infection in two. One patient died two months after the nephrectomy, and one allograft was lost because of acute tubular necrosis. Both patients with erythrocytosis had prompt return of the hematocrit level and RBC mass to normal. Native nephrectomy eradicated the infection in each of the five patients with recurrent urinary tract infections. Results of nephrectomy for hypertension were classified as excellent in six patients, good in nine, and poor in four. Native renal-vein renin ratios of patients with excellent or good responses were not statistically different when compared with those of poor responders.

Recipient race as a risk factor in renal transplantation.

We analyzed 118 renal transplants performed from November 1977 through October 1981 to determine the effect of recipient race on graft and patient survival. Fifty-one cadaver and three living related transplants were performed in the black recipients and 41 cadaver and 23 living related transplants in the white recipients. No significant differences existed between the groups in regard to age, incidence of pretransplant nephrectomy or splenectomy, warm ischemia time, perfusion time, panel-reactive alloantibody (PRA) status, or number of pretransplant blood transfusions. The mean HLA-A and -B locus match was significantly less in black cadaver recipients and the incidence of malignant hypertension was significantly greater in black recipients. The one-year graft survival rate was 54.9% for black cadaver recipients and 48.7% for white cadaver recipients with a corresponding patient survival rate of 86.2% and 85.3%. The one-year graft survival rate for living-related recipients was 100% for blacks and 73.9% for whites with a corresponding patient survival rate of 100% and 86.9%. These results were not significantly different. When the recipients were matched for age, pretransplant transfusions, HLA-A and -B locus matching, and PRA status, no difference in transplant outcome was identified. We concluded that recipient race is not of prognostic significance in determining the outcome of either cadaver or living related donor transplantation.

In situ cold perfusion of kidneys for transplantation. An experimental and clinical evaluation.

Cadaver kidneys from donors who have sustained cardiac standstill are often unsuitable for transplantation due to prolonged warm ischemic time. In an attempt to increase the salvage rate of these kidneys, the efficacy of in situ intra-aortic cold perfusion in producing immediate function of allografted kidneys was assessed in the nonheart beating canine model. In the first experiment, kidneys harvested after cardiac standstill and immediate intra-aortic cold perfusion were demonstrated to function equally as well as kidneys harvested "optimally." In the second experiment, evaluation of intra-aortic cold perfusion on renal core temperature demonstrated rapid cooling of these kidneys to a protective range. Early clinical results in donors whose kidneys were removed after cardiac arrest demonstrate that most of the kidneys allotransplanted after in situ intra-aortic cold perfusion functioned immediately. These preliminary experimental and clinical results demonstrate the effectiveness of in situ intra-aortic cold perfusion. Wider application of this technique to donors with cardiac standstill should increase the available organs for transplantation.

Etiology and management of bovine graft aneurysms.

From April 1975 to September 1980, 16 symptomatic bovine graft aneurysms were identified in 200 grafts used for long-term hemodialysis. Aneurysms occurred in five (3.6%) straight forearm grafts, two (33%) reversed forearm grafts, two (33%) straight thigh grafts, and seven (18.9%) upper arm grafts. Mean blood pressures were similar in the aneurysm and non-aneurysm groups (142/81 vs 153/83 mm Hg). At operation, the aneurysms were determined to be true in three patients and false in 13. Five patients were treated by graft ligation and aneurysm excision and 11 by aneurysm excision and graft revision, with additional graft survivals of one to 25 months. In conclusion, bovine graft aneurysms are usually false and are more frequent in proximal grafts. In the absence of infection, resection is indicated and will result in considerable prolongation of graft survival.

Renal cortical levels of adenosine triphosphate: restoration after prolonged ischemia by in situ perfusion of ATP-MgCl2.

The metabolic consequences of low-temperature kidney preservation by pulsatile perfusion and cold storage were evaluated in canine kidneys by serial adenosine triphosphate (ATP) determinations. Serial cortical ATP levels were determined (1) after optimal harvest, (2) after 60 minutes of warm ischemia, and (3) after 60 minutes of warm ischemia followed by an intra-aortic infusion of ATP-MgCl2. After 24 hours of preservation, cortical ATP levels were significantly greater when pulsatile perfusion was used-both after optimal harvest and after 60 minutes of warm ischemia. Cortical ATP levels were significantly greater at 24 hours in both perfused and cold-storage kidneys when an intra-aortic infusion of ATP-MgCl2 was used following ischemia. These results may explain the 24-hour limitation of cold storage as well as its inferior results following ischemia injury. Furthermore, since ATP is critical in cellular metabolism, the technique of intra-aortic ATP-MgCl2 infusion warrants further investigation in the field of organ preservation.