Preop Hemodialysis or Intraop Ultrafiltration for Patients With Severe Renal Dysfunction Undergoing Open Heart Surgery

1. BACKGROUND

1.1. RENAL DYSFUNCTION AND OPEN HEART SURGERY:

The incidences of both cardiovascular disease (CVD) and chronic renal dysfunction (RD) are increasing with the aging population in the western world (1). The intense relationship between the pathogenesis of CVD and chronic RD has recently been reviewed by Schiffrin et al, in detail (2). They both have common risk factors such as diabetes, hypertension, activation of renin-angiotensin system, endothelial dysfunction, oxydative stress, etc. Besides, each has an impact on the other's outcome. On the one hand, CVD is the most frequent cause of death in chronic RD patients (3). On the other hand, even mild chronic RD is one of the major risk factors of postoperative mortality and morbidity after cardiac operations (4, 5). The mechanism is not clear yet, however, volume overload, electrolyte imbalance and inflammatory state created by cardiopulmonary bypass (CPB) may have an impact. Zakeri et al showed that in-hospital mortality after isolated primary coronary artery bypass grafting (CABG) increases exponentially with increasing levels of renal dysfunction (6). They reported an in-hospital mortality of 2.2%, 4.3%, 9.3% and 14.8% in patients who have a preoperative serum creatinine level (SCr) of <130 µmol/L, 130-149 µmol/L, 150-179 µmol/L and 180-199 µmol/L, respectively. These results were similar to the study published previously by Weerasinghe et al with the same cut-off levels of SCr (7). Using the Glomerular Filtration Rate (GFR) instead of SCr, Cooper et al. came to the same conclusion after analysing 483,914 patients receiving isolated CABG in the Society of Thoracic Surgeons (STS) National Adult Cardiac Database (5). They reported that operative mortality rose inversely with declining renal function, from 1.3% for those with normal renal function to 1.8%, 4.3% and 9.3% for patients with mild, moderate and severe RD, respectively. Another study regarding the effect of preoperative RD on mortality after valve surgery was also published with a relatively smaller patient population (8). Although the RD group had significantly worse outcomes with regard to postoperative ventilation time, re-operation, blood transfusion and length of hospital stay, operative mortality was not statistically different between the two groups (3.4% for RD group vs. 2.3% for the control group), probably because of small sample size. However, Filsoufi et al. reported an increased mortality for patients having SCr of >2.5 mg/dL after single valve replacement (25.0% vs. 2.4%),multiple valve replacement (26.7% vs. 3.4%), and combined valve replacement with CABG (28.0% vs. 4.6%) in a large, single-center cohort (9). Regarding long-term survival, Devbhandari reported 1-, 3- and 5-year survival rates following on-pump coronary bypass surgery as 90.3%, 83.2% and 71.4% for non-dialysis dependent renal dysfunction (NDDRD) patients, and 97.4%, 94.6% and 91.0% for patients with no history of RD, respectively (10). Chronic RD affects not only the operative mortality, but also the morbidity after open heart surgery. It has been shown that preoperative RD is an independent predictor of postoperative acute RD and hemodialysis (HD) (5, 7, 9-12) as well as gastrointestinal (GI) (4, 9), respiratory (5, 9), infectious (5) and neurological (5) complications.

1.2. HEMODIALYSIS:

HD is the most common renal replacement therapy for decades, for those who have end-stage RD and have not received renal transplantation. Intermittent HD is a very efficient method to decrease blood urea and creatinine as well as to treat volume overload. Intermittent HD can be performed temporarily in the setting of acute RD or permanently in the setting of chronic RD. In chronic RD, 3 sessions of 4 hours are usually prescribed to adequately substitute the renal function. A good vascular access is essential to perform HD. A temporary dual- or tri-lumen dialysis catheter has to be inserted into a central vein such as the internal jugular, the subclavian or the femoral vein.

1.3. ULTRAFILTRATION:

Intraoperative ultrafiltration has been used widely in pediatric open heart surgery for decades, reducing total body water, increasing hematocrit (Htc) levels, removing inflammatory mediators, thus improving the operative outcome (13). In the 90's, Naik et al. modified the technique (14), and reported better outcomes with modified ultrafiltration (MUF) in pediatric population (15). However, use of MUF has been limited to end-stage RD patients with volume overload undergoing open heart surgery, as an adjunct to pre- and postoperative HD in the adult population. The Verona group reported fewer respiratory, neurological, GI complications, and less blood product transfusion in the group of patients who received MUF after CPB, however mortality, overall morbidity, length of Intensive Care Unit (ICU) stay and length of hospital stay were comparable between MUF and control groups including 573 consecutive patients (16). A meta-analysis evaluating the effects of ultrafiltration on postoperative blood product use and perioperative bleeding in adult patients revealed fewer bleeding complications and reduced blood product use after intraoperative ultrafiltration (17). Boga et al reported improved cardiac performance after CABG surgery with MUF. However, they could not find any difference in Interleukin-6, Interleukin-8 and Neopterin levels. They attributed this effect to prevention of hemodilution and hypervolemia (18). In summary, no clear evidence is available at the present regarding the impact of intraoperative MUF on the operative outcome of NDDRD patients undergoing open heart surgery. Capuano et al. recently (19) reported successful results in a NDDRD patient who required urgent coronary revascularisation. Nevertheless, the impact of intraoperative MUF on the outcome of NDDRD patients undergoing open heart surgery remains unclear, and is worth investigation.

1.4. PREVIOUS STUDIES:

The quest to improve the outcome of NDDRD patients undergoing open heart surgery has been in the agenda of some groups to date. Two pioneering studies were recently published from Turkey (20, 21). The target patient population was NDDRD patients undergoing elective isolated primary CABG surgery. Patients were randomized into two groups prospectively, one group received 2 doses of prophylactic HD just before surgery whereas the other did not, and served as control. Both studies reported reduced operative mortality rates, reduced postoperative need for HD, and shorter length of stay in the prophylactic HD groups. However, these two studies had very limited number of patients with a short period of follow-up, excluded valve surgery, and did not analyse cost-effectiveness. Furthermore, intraoperative ultrafiltration was not studied.

1.5. ASSESSMENT OF RENAL FUNCTION:

GFR is the best measure of overall kidney function (22). The Cockroft-Gault formula is a commonly used way to predict GFR (23). GFR <30 mL/min/1.73 m2 is accepted as "severe RD" (22). SCr is a simple and practical universal biologic marker used for estimating glomerular filtration. Although SCr does not have a linear association with GFR, it has also been reported to be a powerful predictor of operative mortality (6). Thus, SCr and GFR were both accepted as preoperative indicators of RD with the cut-off levels of 180 µmol/L (or 2.0 mg/dL) and 30 mL/min/1.73 m2, respectively.

1.6. CONCLUSION:

In summary, this data mandates us a well defined strategy for patients with NDDRD in order to obtain better operative outcome. Under the guidance of the current literature, a randomized controlled trial (RCT) with a larger number of patients undergoing open heart surgery will provide precise answers for these questions. Comparison of hospital costs may add an extra value for the assessment of cost-effectiveness as well.