Secondary Logo

Journal Logo


Coronary Bypass Grafting for Patients Dependent on Dialysis: Modified Ultrafiltration for Perioperative Management

Otaki, Masaki*; Enmoto, Takeshi*; Oku, Hidetaka

Author Information
doi: 10.1097/01.MAT.0000094632.66217.EE
  • Free


With increasing numbers of patients requiring chronic dialysis for end-stage renal failure, the number of patients in this population who require coronary bypass grafting has increased. 1 Previous reports revealed that coronary bypass grafting for patients with dialysis dependent renal failure carries an increased risk of operative mortality in long-term survival. 2 Special management for coronary bypass grafting in these patients is required to solve their problems in association with the balance of fluid and electrolytes, the maintenance of hemodynamics by the control of perioperative bleeding and infectious complications, and the timing of postoperative dialysis. 3

Here, we report perioperative managements for renal support, operative results, and long-term survival in patients dependent on chronic dialysis undergoing coronary bypass grafting. To our knowledge, there have been very few reports regarding the clinical use of modified ultrafiltration (UF) for renal support after the cessation of cardiopulmonary bypass for patients dependent on dialysis. Therefore, the effect of modified UF was emphasized in this analysis.

Patients and Methods

An analysis was undertaken with 33 patients (4.5% of total coronary bypass graft patients) on chronic dialysis who required coronary artery bypass grafting at Kinki University Hospital and Saiseikai Kyoto Hospital since January 1993.

The mean age of the patients was 60.6 ± 8.4 years, and the mean duration of maintenance dialysis was 4.7 ± 1.8 years. The pathologic lesions contributing to the development of renal failure were diabetic nephropathy (19 patients, 58%), nephrosclerosis (8 patients, 24%), and chronic glomerulonephritis (6 patients, 18%). The coronary risk factors were systemic hypertension (20 patients, 61%), diabetes mellitus (18 patients, 55%), smoking (17 patients, 52%), family history of coronary artery disease (18 patients, 55%), and hyperlipidemia (15 patients, 45%).

Twenty-six (79%) patients had triple-vessel coronary disease, and 12 (36%) patients had significant left main coronary disease (75% or more). The mean left ventricular ejection fraction (EF) was 0.50 ± 0.12. Twenty-four (73%) patients were in the New York Heart Association (NYHA) class III to IV congestive heart failure category, and 23 (70%) patients were in the NYHA anginal class III to IV. Twenty-three (70%) patients had a history of myocardial infarction, and 11 (33%) patients underwent coronary bypass grafting on an emergent basis. Antiplatelet agents were administrated in 30 (90%) patients preoperatively, and antifibrinolytic therapy was performed in no patients before coronary bypass grafting.

Two methods for UF have been applied to clinical use. With conventional UF during cardiopulmonary bypass, a filter is connected in parallel with a cardiopulmonary bypass circuit, and a patient’s circuit blood can be filtered during cardiopulmonary bypass. Modified UF is connected in a series with the cardiopulmonary bypass circuit and is started immediately after the patient is weaned from cardiopulmonary bypass. Modified UF can allow for both greater efficiency of filtration and greater concentration of red blood cells in the cardiopulmonary bypass circuit. 4

Strategy for Perioperative Management

Our strategy for perioperative management has changed since April 1998 (Table 1). The previous approach with 12 patients dependent on dialysis (group A) was with the performance of conventional continuous hemofiltration (CHF) during and after cardiopulmonary bypass in the operating room, CHF in the intensive care unit, with patients returning to regular dialysis in the dialysis unit when hemodynamically stable. In contrast, the new approach (since April 1998) with 21 patients (group B) was a modified UF approach. Modified UF was conducted immediately after cardiopulmonary bypass, and further artificial renal support was not performed during surgery. In the intensive care unit, continuous hemodialysis (CHD) and early reinstitution of regular dialysis in the dialysis unit were scheduled whenever the patient was stable hemodynamically.

Table 1
Table 1:
Strategy for Perioperative Artificial Renal Support for Patients on Chronic Dialysis

Perfusion and Ultrafiltration Techniques during Surgery

The conventional continuous UF in group A was performed during and after cardiopulmonary bypass. Just after the initiation of cardiopulmonary bypass, a patient’s venous blood from the reservoir was conducted to the hemofiltration circuit and hemoconcentrated using a filter. The amount of fluid removal was 100 ml/hour, and concentrated blood was returned to the reservoir. When the reservoir level was low, Ringer’s lactate was added to maintain the level. Modified UF in group B was performed using UF immediately after the cessation of cardiopulmonary bypass, as described by Eliott. 4 The patient’s arterial blood was drawn into the modified UF circuit through an arterial perfusion line of the cardiopulmonary bypass circuit. The arterial blood in the modified UF circuit was ultrafiltrated using a filter, and the ultrafiltrated arterial blood was returned to the patient’s right atrium through the venous line of the cardiopulmonary bypass circuit. Ringer’s lactate was added to the reservoir when the level was low. The total amount of fluid removal by modified UF was 1800 ml during approximately 15 minutes of filtration.

Coronary artery grafting was performed using cardiopulmonary bypass and cold blood cardioplegia. The left internal thoracic artery was placed on the left anterior descending artery, and saphenous veins were used for other coronary targets. The following were considered indications for left ventricular volume reduction: (1) EF less than 30%, (2) left ventricular end-diastolic volume index more than 145 ml/m2, and (3) left ventricular end-systolic volume index more than 100 ml/m2. Volume reduction was performed by linear ventriculotomy between the left anterior descending coronary artery and posterolateral branch. Left ventricular volume reduction surgery was concomitantly performed with coronary bypass grafting in three patients. The blood transfusion was indicated when the hematocrit value was less than 15% during cardiopulmonary bypass or 25% in the operating room or intensive care unit. Shed mediastinal blood was used in all patients after cardiopulmonary bypass, and coagulation factors were not indicated for all patients of this series.

Operative mortality and morbidity were defined as occurring within 30 days of surgery. Fisher’s exact test was used for nonparametric variables, and a comparison of continuous variables was performed using Student’s t-test. Cardiac event-free survival was calculated by the Kaplan-Meier method. All values are expressed as the mean ± standard error. p < 0.05 was considered statistically significant. Mean follow-up period was 5.3 ± 1.8 years in group A and 2.2 ± 0.8 years in group B.


There were no statistically significant differences between group A and group B in regard to age distribution, male/female ratio, preoperative left ventricular EF, history of myocardial infarction, and NYHA III to IV congestive heart failure class and anginal class (Table 2). The number of coronary bypass grafts, cardiopulmonary bypass time, aortic cross clamp time, and total operating time were not statistically significant between the two groups (Table 3). The operative mortality in group A was 17% (2 deaths) and 0% (no deaths) in group B (p < 0.05). One patient in group A died of arrhythmogenic complications on day 7 postoperatively, and another patient in group A had a massive stroke and died after coronary surgery.

Table 2
Table 2:
Preoperative Patient Profile
Table 3
Table 3:
Operative Information

Postoperative complications are shown in Table 4. Three patients in group A and one in group B had bleeding complications requiring reoperation (25%vs. 5%, p < 0.05). Three patients in group A and one in group B developed congestive heart failure necessitating intraaortic balloon pump (IABP) support postoperatively (25%vs. 5%, p < 0.05). Four patients in group A and one in group B had respiratory failure requiring long-term ventilation of more than 3 days (33%vs. 5%, p < 0.05). Two patients in group A and two in group B had wound infection (16%vs. 10%, p = ns). There were five patients in group A and two in group B requiring long-term management of more than 4 days in the intensive care unit (41%vs. 10%, p < 0.05). Postoperative blood loss and transfusion requirements are shown in Table 5. Postoperative bleeding within 24 hours was 1310 ± 536 ml in group A and 623 ± 312 ml in group B (p < 0.05). Transfusion requirements were 9.3 ± 3.0 units in group A and 3.0 ± 1.7 units in group B (p < 0.05). Eleven (92%) patients in group A and 16 (76%) patients in group B needed blood transfusion because of chronic anemia caused by renal failure. The blood loss between 24 and 48 hours after surgery was not statistically different in either group.

Table 4
Table 4:
Postoperative Complications
Table 5
Table 5:
Postoperative Blood Loss and Transfusion Requirements

Nineteen patients underwent postoperative angiographic examinations at the time of discharge. All of 19 inferior temporal artery grafts to the left anterior descending artery were patent. Two saphenous vein grafts were occluded, one of two occluded grafts to the right coronary artery in group A and the other to the obtuse marginal branch in group B.

During long-term follow-up, there were no significant differences in the long-term survival, NYHA functional class, or incidence of recurrent angina between patients in groups A and B. One patient in group A died of hepatoma 18 months after surgery, and one death occurred because of stroke 10 months after surgery in group B. No patients had myocardial infarction. Two patients (1 in group A and 1 in group B) had recurrent angina because of the development of coronary artery disease after 10 months and 13 months, respectively, with the overall incidence of recurrent angina of 2.0% per patient year in all patients, although, angiographically, all grafts were patent. These two patients were successfully treated pharmacologically. One patient in group B had ventricular tachycardia diagnosed by Holter monitoring, but this was well controlled pharmacologically. Overall cardiac event-free rates after coronary bypass grafting at 1, 3, and 5 years were 88%, 73%, and 67%, respectively. The mean NYHA congestive heart failure class in all operative survivors fell from 3.2/IV preoperatively to 1.6/IV, and the mean NYHA anginal class fell from 3.3/IV preoperatively to 1.5/IV during long-term follow-up. Twenty-six long-term survivors were all able to return to daily life with regular dialysis maintenance three times a week.


Ischemic heart diseases remain the major cause of death in patients with end-stage renal failure, accounting for 30% to 53% of deaths among patients on chronic dialysis. 5 It is generally understood that patients on dialysis have a greater risk of developing coronary artery disease, 6,7 and coronary bypass grafting for patients on chronic dialysis carries increased risk for operative mortality and long-term survival. 2

Percutaneous transluminal coronary angioplasty (PTCA) in patients on chronic dialysis is technically feasible and provides relief for angina. However, the restenosis rate after PTCA has been high, and aggressive restenosis limits the long-term clinical benefit. 8 Because of these disadvantages of PTCA in patients dependent on dialysis, coronary bypass grafting is considered a more preferable therapy for these patients. Patients with chronic renal failure on dialysis often have multiple comorbid disorders, including hypertension and diabetes, fluid balance and electrolyte management, 9 and serious complications after coronary artery bypass graft. 10–15

The primary question is whether protection from cardiac related events is improved by surgical revascularization in patients on dialysis. Although relief of cardiac symptoms has been reported, overall functional status after coronary bypass in patients on dialysis and long-term survival have not been adequately studied. The current study 2 demonstrated that overall functional status was stable at discharge in operative survivors who were very functionally limited preoperatively.

Concerning long-term survival, few studies have provided follow-up beyond 2 years after coronary bypass in patients on dialysis. 2 Jahangiri et al.1 reported that 1, 2, and 3 year survival rates were 87%, 78%, and 58%, respectively. These survival rates are notably lower than those for the normal coronary population, which have been reported to be 95%, 88%, and 75% at 1, 5, and 10 years, respectively. In contrast, one recent study revealed favorable clinical results. The actuarial survival rates at 1, 3, and 5 years were 89%, 84%, and 71%, respectively, whereas estimates for cardiac deaths were 93%, 93%, and 82%, respectively. 16

The overall operative mortality in our series was 6.6% in patients dependent on dialysis, which is comparatively higher than the general population of patients receiving coronary bypass but reasonably acceptable compared with patients that are medically treated. During long-term follow-up in our series, there were two deaths, with an incidence of 2.0% per patient year, notably better than previous reports in the literature. There were two patients with recurrence of chest pain caused by developing coronary artery disease, with an incidence of 2.0% per patient year, and one arrhythmogenic complication, with an incidence of 1.0% per patient year.

As far as the special management of patients on chronic dialysis in the perioperative period is concerned, dialysis 24 hours before revascularization and dialysis as close to bypass surgery as possible are reported as effective in the literature. 1,17 Previously, the basic management of patients on chronic dialysis undergoing coronary bypass grafting in our series was dialysis 24 hours before surgery, CHF during and after cardiopulmonary bypass, CHF in the intensive care unit, and then routine hemodialysis in the dialysis unit.

Recently, our basic approach has changed, with dialysis performed 24 hours before revascularization, modified UF immediately after cardiopulmonary bypass, and then CHD or routine dialysis in the intensive care unit. With the clinical application of modified UF immediately after cardiopulmonary bypass, perioperative fluid balance and electrolyte management can be controlled more easily and effectively than CHF alone during and after cardiopulmonary bypass.

After modified UF was performed, further hemodialysis was not usually performed until the patient was in the intensive care unit. Patients treated with modified UF had several advantages, including more stable hemodynamics after surgery, easier weaning from mechanical respiratory support, and lower postoperative blood loss and transfusion requirements than with CHF alone during cardiopulmonary bypass. A few studies have been published regarding the benefit of modified UF, although the clinical efficacy of modified UF remains to be studied adequately. To our knowledge, there are no reports describing the clinical use of modified UF with patients dependent on dialysis as a perioperative renal support after coronary bypass grafting.

Modified UF was generally reported to significantly diminish the total body water increase and donor blood requirements and to return the contents of the cardiopulmonary bypass circuit to the patients in a concentrated form. 4,18–21 As a result, modified UF can improve respiratory function and hemodynamics after surgery and improve contraction of the left ventricle. Elliott 4 reported that the use of modified UF after cessation of cardiopulmonary bypass increased hemoglobin and hematocrit levels and reduced postoperative chest drain loss and blood transfusion requirements. Inflammatory mediators such as interleukins, tumor necrosis factor, and activated complement components are reported to be removed by modified UF. 21,22 Elliott 4 also demonstrated that a reduction in the myocardial water content and inflammatory mediators may result in the improvement of myocardial performance and accelerate myocardial recovery. Kiziltepe and associates 23 reported a similar result in which modified UF improved hemodynamic, hemostatic, and pulmonary function after open heart surgery in adult patients at high risk. In addition, Onoe et al.20 emphasized the clinical use of modified UF more as a means of blood purification with blood apheresis by adding a crystalloid solution more than as a means to concentrate blood. Our findings also demonstrate that modified UF improves cardiac, pulmonary, and hemostatic functions after coronary bypass grafting even in patients with end-stage renal failure.

In conclusion, the operative mortality and morbidity for coronary bypass grafting in patients dependent on dialysis were reasonably acceptable compared with previous reports in the literature. As a perioperative management, modified UF can play a more important role in reducing bleeding complications and improving cardiac and pulmonary functions after coronary bypass grafting than can CHF during cardiopulmonary bypass. Further clinical application of modified UF may be warranted for coronary bypass grafting in patients at high risk, especially those on chronic dialysis. During follow-up, long-term survival, the NYHA functional class, and incidence of recurrent angina or myocardial infarction were considered favorable in patients dependent on dialysis. However, the number of patients included in each group is small, and this study has the limitation of being a single center study. Therefore, a multicenter study in a prospective manner would be an appropriate method of investigation in the future.


1. Jahangiri M, Wright J, Edmondson S, et al: Coronary artery bypass surgery in dialysis patients. Heart 78: 343–345, 1997.
2. Owen CH, Cummings RG, Sell TL, et al: Coronary artery bypass grafting in patients with dialysis-dependent renal failure. Ann Thorac Surg 58: 1729–1733, 1994.
3. Ko W, Kreiger KH, Isom OW: Cardiopulmonary bypass procedures in dialysis patients. Ann Thorac Surg 55: 677–684, 1993.
4. Elliott MJ: Ultrafiltration and modified ultrafiltration in pediatric open heart operations. Ann Thorac Surg 56: 1518–1522, 1993.
5. Batiuk TD, Kurtz SB, Oh JK, Orszulak TA: Coronary artery bypass operation in dialysis patients. Mayo Clin Proc 66: 45–53, 1991.
6. Kaul JK, Fields BL, Reddy MA, et al: Cardiac operation in patients with end-stage renal disease. Ann Thorac Surg 57: 691–696, 1994.
7. Port FK: The end-stage renal disease prognosis trends over the past 18 years. Am J Kidney Dis 20 (Suppl): 3–7, 1992.
8. Kahn JK, Rutherford BD, McConahay DR, et al: Short and long-term outcome of percutaneous transluminal coronary angioplasty in chronic dialysis patients. Am Heart J 119: 484–489, 1990.
9. Cruz D, Bia MJ: Coronary revascularization in patients on dialysis. What treatment option should we choose? ASAIO J 42: 139–141, 1996.
10. Mujais S: Coronary revascularization in dialysis patients: the need for vigilance. Int J Artif Organs 14: 7–9, 1991.
11. Khaitan L, Sutter FP, Goldman SM: Coronary artery bypass grafting in patients who require long-term dialysis. Ann Thorac Surg 69: 1135–1139, 2000.
12. Rostand SG, Kirklin KA, Rutsky EA, et al: Results of coronary artery bypass grafting in end-stage renal disease. Am J Kidney Dis 12: 266–270, 1988.
13. Monson BK, Wickstrom PH, Haglin JJ, et al: Cardiac operation and end-stage renal disease. Ann Thorac Surg 30: 267–272, 1980.
14. Opsahl JA, Husebye DG, Helseth HK, et al: Coronary artery bypass-surgery in patients on maintenance dialysis: long-term survival. Am J Kidney Dis 12: 271–274, 1988.
15. Franga DL, Kratz JM, Crumbley AJ, et al: Early and long-term results of coronary artery bypass grafting in dialysis patients. Ann Thorac Surg 70: 813–819, 2000.
16. Nakayama Y, Sakata R, Ura M, et al: Coronary artery bypass grafting in dialysis patients. Ann Thorac Surg 68: 1257–1261, 1999.
17. Laws KH, Merrill WH, Hammon JW, et al: Cardiac surgery in patients with end-stage renal disease. Ann Thorac Surg 42: 152–157, 1986.
18. Daggett CW, Lodge AJ, Scarborough JE, et al: Modified ultrafiltration versus conventional ultrafiltration: a randomized prospective study in neonatal piglets. J Thorac Cardiovasc Surg 115: 336–342, 1998.
19. Draaisma AM, Hazekamp MG, Frank M, et al: Modified ultrafiltration after cardiopulmonary bypass in pediatric cardiac surgery. Ann Thorac Surg 64: 521–525, 1997.
20. Onoe M, Magara T, Yamamoto Y, et al: Modified ultrafiltration removes serum interleukin-8 in adult cardiac surgery. Perfusion 16: 37–42, 2000.
21. Journois D, Pouard P, Greeley WJ, et al: Hemofiltration during cardiopulmonary bypass in pediatric cardiac surgery. Anesthesiology 81: 1181–1189, 1994.
22. Wang MJ, Chiu IS, Hsu CM, et al: Efficacy of ultrafiltration in removing inflammatory mediators during pediatric cardiac operations. Ann Thorac Surg 61: 651–656, 1996.
23. Kiziltepe U, Uysalel A, Corapcioglu T, et al: Effects of combined conventional and modified ultrafiltration in adult patient. Ann Thorac Surg 71: 684–693, 2001.
Copyright © 2003 by the American Society for Artificial Internal Organs