| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
© 2004 European Association of Cardio-Thoracic Surgery
Early and intensive continuous veno-venous hemofiltration for acute renal failure after cardiac surgeryDepartment of Cardiothoracic Surgery, St Thomas' Hospital, 4 Deauville Court, Elanor Close, London SE 16 6PY, UK
* Corresponding author. Tel.: +44-207-922-8005; fax: +44-207-955-4858 Received January 31, 2004; received in revised form March 4, 2004; accepted March 5, 2004
Various forms of renal replacement therapies are available to treat acute renal failure (ARF) after cardiac surgery. The objective of this study was to assess the incidence of ARF developing postoperatively necessitating continuous veno-venous hemofiltration (CVVH) in adult patients requiring cardiopulmonary bypass (CPB), to determine the factors which influence the outcome in these patients and to assess the outcome following the use of early and intensive CVVH. During the study period, i.e. August 2000 to July 2002, 2355 adult patients underwent surgery under CPB, of whom 159 (6.7%) developed renal failure (creatinine >200µmol/l) and 116 (5%) needed CVVH. Patients excluded were those who died within 24 h and those who underwent coronary artery bypass grafting without utilising CPB, thoracoabdominal aneurysm operations and pericardial surgery. Average age, Parsonnet score and Euroscore in the study population were 69.9 years, 21 and 7.70, respectively. Of the 116, 45 died in the intensive care unit (38.8% mortality). Relatively more non-survivors suffered from diabetes and preoperative renal dysfunction  Adverse outcome was also more likely if patient suffered from postoperative cardiac failure or had gastrointestinal complications or had more than two organ systems failing Mortality was 100% if hepatic failure ensued.
Key Words: Kidney; Cardiopulmonary bypass; Complications; Morbidity; Postoperative care; Risk analysis
Acute renal failure (ARF) is a well-known complication of cardiac surgery. The incidence varies between 1 and 15% [1,2]. Although the mortality a decade ago was reported to be 70–90% [2], recently better outcome has been reported following the treatment of ARF [3,4]. This improvement in outcome has been multifactorial and has been attributed to, increase in awareness by the clinicians of the pathophysiology of organ dysfunction, advances in the field of cardiac surgery to reduce the side effects of cardiopulmonary bypass (CPB), and advances in the renal replacement therapy itself. The contribution of renal replacement therapy to clinical outcome remains a subject of intensive investigation and controversy. This is due in part of underlying illness severity, which remains an overwhelming determinant of the outcome, especially now that ARF is commonly seen in the setting of multiorgan failure [5]. Most common techniques used in clinical practice to treat ARF are intermittent hemodialysis (IHD) and continuous veno-venous hemofiltration (CVVH). IHD is largely being replaced by CVVH with the advantage of its ability to control patient's volume status [6]. Early and intensive application of CVVH as a strategy has also helped to improve the outcome [3,4]. CVVH however has not been without criticism because of the perception that it is costly, resource intensive and in reality may not improve the outcome [5,7]. There is also need to predict which patients would benefit from such an intensive application of CVVH so as to use it more rationally [8]. In our hospital over the last five years there has been gradual but a total shift from utilising intermittent dialysis for treating postcardiac surgery ARF to use of CVVH. This study was aimed to assess (a) the incidence of postoperative ARF necessitating CVVH, (b) the factors influencing mortality and (c) outcome following use of early and intensive CVVH as the strategy for renal replacement therapy.
2.1. Patients This was a prospective, observational case series. During a 24-month period from August 2000 to July 2002, patients who underwent cardiac surgery using CPB and developed ARF necessitating CVVH were identified prospectively and included in the study. Exclusion criteria were patient who underwent coronary artery bypass grafting without the use of CPB, thoracoabdominal aneurysm surgery, patients who were discharged and were readmitted, and patients who died within 24 h of surgery. ARF was defined as urine output <480 ml in the preceding 24 h, or urea  35 mmol/l, or creatinine 300µmol/l. CVVH was initiated when there was no response to fluid and/or hemodynamic resuscitation and frusemide administration within 24 h. 2.2. Renal support Renal support was provided in all patients by CVVH using polyacrylonitrile filters (AN69, Hospal, Lyon, France). Vascular access was established by insertion of double lumen catheter into a femoral, internal jugular or subclavian vein. The blood pump was set to deliver 200–250 ml/min aiming for an ultrafiltration rate of 25–35 ml/kg/h. Anticoagulation of the extracorporeal circuit was maintained by a heparin infusion (25–1000 U/h) through the inflow side of the circuit. In patients with thrombocytopenia or excessive bleeding due to any other cause, heparin was replaced with continuous prostacyclin infusion.2.3. Data collection and analysis Using medical notes, intensive care unit (ICU) database and cardiac surgical database details of the patients with regards to demographics, preoperative morbidities, surgical characteristics, postoperative complications, and outcome data were recorded. Gastrointestinal complications were defined as gastrointestinal bleeding or need for laparotomy for any indication. Details of CVVH with reference to surgery to CVVH duration, indication for CVVH and renal function at start, during and after CVVH were recorded. Severity of illness on the day of initiation of CVVH was recorded using organ-based scoring system, reported by Knauss et al. and later modified for use in postcardiac surgical patients by Ostermann et al. (Table 1). Patients were divided into two groups, i.e. survivors and non-survivors for analysis. Data was analysed using SPSS version 10.0. Data are expressed throughout as mean±SDs. When appropriate, statistical comparisons were made using Mann Whitney U test for non-parametric data. Incidence and mortality rates were compared using 2 test.
During the study period 2355 patients underwent cardiac surgery using CPB and met the inclusion criteria. Hundred and fifty-nine patients (6.7%) developed ARF and 116 (5%) were treated with an early intensive CVVH. Their demographic and clinical characteristics are presented in Table 2. Type of surgery and associated incidence of renal failure and mortality are also presented in Table 2. Observed mortality was 38.8% (45/116).
3.1. Preoperative risk factors Average age, Parsonnet and Euroscore were 69.9 years, 21 and 7.7, respectively, with no statistically significant difference between the two groups. There was no difference in sex distribution between survivors and non-survivors (Table 2). Relatively more non-survivors suffered from diabetes and preoperative renal dysfunction (creatinine 200µmol/l). Three patients were known to have cirrhosis preoperatively, all died postoperatively with multiorgan failure. 3.2. Postoperative events Postoperatively, 66 patients who developed ARF were on inotropic support (Table 2), of which 34 also required intra aortic balloon pump (IABP). Sixteen patients were reoperated on either for bleeding or for tamponade. Twenty-two patients suffered from septicaemia, of which 13 were amongst the non-survivors. On logistic regression, predictors of mortality were redo-surgery postoperative cardiac failure  respiratory failure  and gastrointestinal complications  3.3. Organ system failure Survival correlated inversely with the number of organ systems that have failed (Table 3). Adverse outcome was more likely with more than two organ systems failing On the other hand there were better chances of survival if patient suffered from only renal failure (all 13 patients survived) or from two organ system failures other than liver failure as the mortality was 100% (5/5) if hepatic failure ensued.
3.4. Hemofiltration 3.4.1. Indications Oliguia (urine output  480 ml/24 h) was the primary indication for CVVH, followed by azotaemia (urea 35 mmol/l, creatinine 300µmol/l; Table 4).
3.4.2. Time course The time from surgery to initiation of CVVH varied from immediate postoperative period to 23 days in survivors; and from immediate postoperative period to 19 days in non-survivors. There was significant difference in duration of haemofiltration between the two groups, with non-survivors undergoing CVVH for a longer duration (Table 4).
3.4.3. Efficiency 3.5. Renal outcome Forty-five patients died during their stay in ICU (38.8%). Of the 71 patients who survived, 69 patients recovered enough renal function to discontinue CVVH. The remaining two continued to need IHD, but recovered renal function within 6 months.
Despite major advances in cardiac surgical techniques, anaesthesia and CPB, serious complications such as ARF still happen [1,2]. ARF is independently associated with early mortality following cardiac surgery, even after adjustment from co-morbidity and postoperative complications [9]. The reported incidence of ARF following CPB varies between 1 and 15%, and is due to different definition in each study [1,2]. The mortality in the group of patients requiring renal replacement therapy is substantial in contrast to those patients with mild to moderate ARF who will respond to medical therapy such as fluid resuscitation and frusemide administration [8,9]. Patients necessitating renal replacement therapy usually deteriorate further, need ITU admission and may develop multi-system failure necessitating measures such as ventilation, ionotropic support, IABP, etc. [1–5]. Mortality in these patients despite intermittent dialysis has remained high even in 1990s [5,9]. This may be due to suboptimal removal of uraemic toxins in critically ill patients and aggravation of inflammatory response by bioincompatible membranes [5]. Furthermore, IHD causes serious hemodynamic instability. Continuous hemofiltration is a highly effective system for the replacement of renal function in patients with ARF [10]. The control of biochemical characteristics is constant, and the patient need never have any unwanted alterations of extracellular fluid volume, even if large quantities of fluid are administered. CVVH offers continuous and steady replacement of fluid and removal of uraemic toxins. Its intensity can be titrated to avoid rapid fluid shifts. Myocardial depressant factors may be removed and myocardial performance improved [11].
One of the major reasons for differences in the success with CVVH has been attributed to the differences in the ultrafiltration rate during CVVH [5]. A prospective randomised trial by Ronco and colleagues was aimed at identifying impact of different doses of ultrafiltration during CVVH on survival [12]. Patients were randomly assigned to ultrafiltration at 20 ml/kg/h (group 1, Another observation with regards to better outcome with CVVH is in relation to its early start [5]. Baudouin and colleagues reported a mortality of 80% following CVVH when the treatment was initiated much later (>8 days postoperatively) and ultrafiltration rate was also approximately 50% of that recommended by Ronco and colleagues [14]. This observation is further supported by Bent and colleagues who have recently published their experience with use of early and aggressive CVVH strategy [4]. Sixty-five patients were treated with early and intensive CVVH with a mean operation to CVVH time of 2.38 days and pump controlled ultrafiltration rate of 2 l/h. Patient population was older, mean age of 70.5±6.8 years, 20% patients were operated on as emergencies and in 33% patients IABP was required postoperatively the observed mortality was only 40%. The predicted mortality using a previously validated ARF specific model, Liano score was 66%. Our results are in keeping with those of Bent and colleagues with a mortality of 38.8%. CVVH has several disadvantages, including intensive nursing requirements, continuous anticoagulation, patient immobility, and expense [5,7]. The cost difference between CVVH and IHD in one study was $261 per treatment day [15]. Hence, it is important to analyse the results periodically so as to find out predictors of mortality, which may help to predict outcome and help proper allocation of resources. Tsang and colleagues found no survivors when the cardiac index was less than 1.7 l/m2 and adrenaline requirements was more than 30 µg/min before starting continuous hemofiltration, although only 60% of patients had cardiac output monitored there [7]. Bent and colleagues on multivariate analysis found hypotension, reoperation, acidosis and neurological dysfunction to be predictors of adverse outcome [4]. Using neural networks, the authors went ahead to describe a predictive model for adverse outcome. Such models, which are developed and applied in each unit, can assist clinicians in making difficult therapeutic decisions and wide use of available resources. Dialysis has always been perceived as a bad prognostic indicator by cardiac surgeons, but advances in diagnostic and treatment modalities have helped to reduce the mortality associated with postoperative ARF. Ostermann and colleagues compared their results from 1989–1990 and 1997–1998 in patients with ARF following cardiac surgery. [3]. Despite operating on older patients in 1997–1998 (65.3 vs 56 years) and sicker as evidenced by higher percentage of redo cardiac operations (30 vs 8.6%) and emergency operations (50 vs 25.7%), there was a reduction in mortality from 83 to 54%. They also studied association between number of failed organ systems and mortality. They confirmed inverse relationship between number of systems failed and the outcome. Specially, poor cardiac function was associated with a particularly poor outcome. We have observed similarly poor outcome in patients with more than two organ systems failing. All five patients with acute liver failure died. Association between cardiac failure and poor outcome was also observed (Table 3). In conclusion, the strategy of early and aggressive CVVH appears to achieve an acceptable survival in patients suffering from ARF after cardiac surgery, despite older and higher risk patients undergoing surgery. Strategies focused prevention and management of multi-organ failure may have significant impact on the outcome. doi:10.1016/j.icvts.2004.03.002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
), 35 ml/kg/h (group 2, 
), or 45 ml/kg/h (group 3, 
). The primary end point was survival at 15 days after stopping haemofiltration. All patients reached values of ultrafiltration of at least 85% of prescribed dose with 91%. Seventy-seven percent of their patients were surgical. They found that the survival in group 1 was significantly lower than in the other two groups. The frequency of complications and incidence of full recovery of renal function was similar in all groups. They concluded that the optimal rate of ultrafiltration during CVVH should be prescribed as per patients body weight and should at least be 35 ml/kg/h. This explains the high mortality observed by Levy and colleagues who reported a mortality of 87.5% in 16 patients in whom CVVH therapy was used 
