Anesth Analg 2006;102:2-7
© 2006 International Anesthesia Research Society
doi: 10.1213/01.ane.0000189555.70938.e2
CARDIOVASCULAR ANESTHESIA
Kenneth J. Tuman
Duration of Dialysis Is a Significant Predictor of Prolonged Postoperative Mechanical Ventilation in Dialysis-Dependent Patients Undergoing Cardiac Surgery
Masato Nakasuji, MD,
Shinichi Nishi, MD, PhD,
Kae Nakasuji, MD,
Naoya Hamaoka, MD, PhD,
Kazutoshi Ikeshita, MD, and
Akira Asada, MD, PhD
Department of Anesthesiology and Intensive Care Medicine, Osaka City University Medical School, Osaka, Japan
Address correspondence and reprint requests to Masato Nakasuji, MD, Department of Anesthesiology and Intensive Care Medicine, Osaka City University Medical School, 1-5-7 Asahimachi, Abeno-ku, Osaka 545-8586, Japan. Address e-mail to nksuji{at}msic.med.osaka-cu.ac.jp.
 |
Abstract
|
|---|
Prolonged mechanical ventilation is reported to correlate with increased risk of mortality after cardiac surgery. We designed the present study to determine the preoperative and intraoperative risk factors that could predict postoperative prolonged mechanical ventilation in dialysis-dependent patients undergoing cardiac surgery with cardiopulmonary bypass. Forty-four dialysis-dependent patients were divided into two groups; patients of group E were tracheally extubated within 24 h after admission to the intensive care unit postoperatively (n = 19) and patients of group L (n = 25) required more than 24 h of mechanical ventilation. All patients received hemofiltration during cardiopulmonary bypass and continuous veno-venous hemodialysis postoperatively. A multiple logistic regression analysis showed that duration of dialysis (>10 yr) and duration of surgery (>8 h) were independent risk factors of prolonged mechanical ventilation (>24 h). On admission to the intensive care unit, Pao2/Fio2 of group L was significantly lower than that of group E (294 ± 135 versus 415 ± 99 mm Hg) and the circulatory status of group L was worse than that of group E. The median (interquartile range) duration of intensive care unit stay in group E was 3 (3.00) days, which was significantly shorter than that of group L (5 [2.75] days). It is possible that longer surgery increases the likelihood of cardiac dysfunction and poor oxygenation in patients with a long history of dialysis.
|
Introduction
|
|---|
Cardiac disease is a leading cause of death in dialysis-dependent patients. Surgical treatment of cardiac patients, including surgeons’ skill, technique for providing cardiopulmonary bypass (CPB), myocardial protection during CPB, anesthetic management, and postoperative management in the intensive care unit (ICU), have improved recently. However, the mortality rate for dialysis-dependent patients undergoing cardiac surgery with CPB in the early postoperative period remains frequent and varies from 4.4% to 16.9% (1–8) compared with patients with normal renal function, even in institutions that manage many surgical cases annually. Therefore, careful postoperative management in accordance with duration of CPB and circulatory and respiratory status just after surgery (i.e., goal for cardiac output to be achieved, timing of tracheal extubation, and initiation of fluid removal in the early postoperative period) is important to enhance a good prognosis. The use of prolonged mechanical ventilation (PMV) postoperatively is especially reported to correlate with frequent mortality after cardiac surgery (9). In addition, PMV increases overall hospital costs as well as ICU costs and reduces the availability of ICU beds for other critically ill patients. Our hypothesis is that PMV can be predicted before admission to the ICU. The present study was designed to determine the preoperative and intraoperative risk factors that could predict the need for postoperative PMV in dialysis-dependent patients.
|
Methods
|
|---|
After receiving approval from the Ethics Review Committee of Osaka City University and informed consent from the participants, we retrospectively investigated the medical records of 44 dialysis-dependent patients undergoing cardiac surgery with CPB from January 1997 to December 2004 at Osaka City University Hospital. All surgeries were conducted under the supervision of the same cardiac surgeon and intensivist during the study period. Similarly, the protocols followed for anesthetic management during the same period were the same. Patients receiving peritoneal dialysis were excluded from this study.
All patients received conventional hemodialysis the day before surgery. The goal of preoperative hemodialysis was to remove fluid to maintain ideal dry body weight. Monitoring consisted of electrocardiography, pulse oximetry, capnography, catheterization of peripheral artery and pulmonary artery, and monitoring of peripheral and bladder temperature. Anesthesia was induced with 5–10 µg/kg fentanyl and 0.1 mg/kg midazolam and maintained with 20–30 µg/kg fentanyl and continuous propofol or intermittent midazolam. Vecuronium bromide was administered initially at 0.1 mg/kg IV for tracheal intubation followed by intermittent injection during surgery. Nitrous oxide was not used throughout surgery. Potassium-free crystalloid was infused in all patients before CPB. For myocardial protection, all patients received cold blood cardioplegia and CPB was performed at moderate systemic hypothermia (30°C to 32°C). All patients underwent hemofiltration during CPB. A polysulfone hemofilter (PS-1.0UW; Kawasumi Kagaku CO, Tokyo, Japan) was connected in parallel to the CPB circuit and replacement fluid was infused into the blood reservoir. Dopamine was IV infused continuously at a rate of 3–5 µg · kg–1 · min–1 in all patients after CPB as inotropic support. An intraaortic balloon pump was applied or other inotropic drugs (adrenaline or noradrenalin) were infused when the cardiac index decreased to <2.5 L/min/m2 and the systemic arterial blood pressure could not be maintained above 80 mm Hg despite the optimal filling pressure and dopamine infusion at 10 µg · kg–1 · min–1. All patients were transferred to the ICU before emergence from anesthesia and they received mechanical ventilation overnight under continuous IV infusion of midazolam (5–8 mg/h) from 1997 to 1999 and propofol (2–3 mg · kg–1 · h–1) from 2000 to 2004.
Irrespective of the circulatory status, continuous veno-venous hemodialysis (CVVHD) was planned as soon as possible when hemostasis was confirmed as appropriate (<50 mL/h from chest tube) and activated clotting time (ACT) was <150 s. The ACH-07S (Asahi Medical CO, Tokyo, Japan) was used as the CVVHD apparatus. Blood access was achieved principally by an indwelling flexible double-lumen catheter inserted in the femoral vein. A polyacrylonitril dialyzer (APF 10S, Asahi Medical CO) was used. The dialysate (containing in mmol/L, Na+ 138, K+ 2.0, Ca2+ 3.8, Mg2+ 1.5, Cl– 107, and L-lactate 38) was circulated in a countercurrent direction toward the blood flow from the dialyzer inlet and the diafiltrate was drained from the dialyzer outlet. The blood flow rate was maintained at 100 mL/min. The flow rates of the dialysate and diafiltrate were preset at 900 mL/h and 1000 mL/h, respectively. The flow rate of plasma fluid for removal was approximately 100 mL/h according to this preset value. Nafamostat mesilate was administered at 20 mg/h as an anticoagulant through the predialyzer compartment of the circuit. ACT was maintained between 140 and 160 s and monitored at least every 8 h. Potassium-free crystalloid (500–1000 mL/day) was infused and transfusion was performed on the request of intensivists to maintain a stable hemodynamic status. When arterial blood gas values matched our criteria (Pao2/Fio2 >350 mm Hg) the first morning after surgery, in addition to stable hemodynamic status, we discontinued propofol/midazolam infusion and patients were weaned from mechanical ventilation. Patients were tracheally extubated when they satisfied the following criteria: 1) Pao2/Fio2 >300 and Paco2 <45 mm Hg receiving 5 cm H2O pressure support ventilation, 2) respiratory rate <20 breaths/min, 3) stable circulatory status without arrhythmia and blood loss <50 mL/h, and 4) clear consciousness.
We used CVVHD in all patients until the first postoperative morning and the CVVHD was discontinued before tracheal extubation. When patients were not extubated within 24 h after admission to the ICU, continuation of CVVHD or conversion to conventional hemodialysis after the second postoperative day depended solely on circulatory status. Our therapeutic goal for dialysis-dependent patients undergoing cardiac surgery was for patients to be tracheally extubated within 24 h after admission to the ICU and discharged from the ICU on the second postoperative day.
According to our therapeutic protocol, patients were discharged from the ICU on the second postoperative day if they were tracheally extubated within 24 h after admission to the ICU. If not, they stayed in the ICU for more than 3 days. Therefore, 44 patients were divided into 2 groups according to the duration of mechanical ventilation. Group E consisted of patients who were tracheally extubated within 24 h after admission to the ICU. Group L consisted of patients who required mechanical ventilation in the ICU for more than 24 h. We compared the following variables in the two groups: 1) preoperative demographic data, 2) intraoperative factors, 3) ventilatory, circulatory, and biochemical status on admission to the ICU, 4) postoperative status and mortality. Fluid balance included the sum of water balance (filtrated fluid volume subtracted from crystalloid infusion volume excluding the priming volume for CPB) and blood balance (bleeding volume subtracted from transfusion volume). We also evaluated preoperative and intraoperative risk factors of PMV (>24 h).
All statistical analyses were performed using SigmaStat 3.0®software. First, the normality test (Kolmogorov-Smirnov) and equal variance test were applied to all data. When these tests confirmed normal distribution of the data, the Student’s t-test or Welch’s t-test was used for comparison of baseline data of the two groups and the results were expressed as mean ± sd. When the data were not normally distributed, the Mann-Whitney U-test was used and the results were expressed as median (interquartile range). 
2 and Fisher’s exact tests were used for comparison of categorical data between groups. When a significant difference was seen between the two groups, power analysis was performed and the sample size was evaluated. Preoperative and intraoperative variables considered to be associated with PMV, and variables considered clinically significant, e.g., duration of dialysis, left ventricular ejection fraction (LVEF), and aortic and mitral valve calcification as preoperative risk factors and duration of surgery as an intraoperative risk factor, were entered into a multiple logistic regression model for multivariate analysis to identify independent risk factors of PMV (>24 h). The odds ratio, 95% confidence intervals (CI) and P values were calculated for each risk factor in the final model. The difference was considered statistically significant when the P value was <0.05.
|
Results
|
|---|
Tracheal extubation within 24 h after admission to the ICU was accomplished in 19 patients (group E) whereas 25 patients required PMV (group L). There were significant differences between groups E and L with regard to preoperative (duration of dialysis, LVEF, and calcification of both aortic and mitral valves detected by echocardiogram) (Table 1) and intraoperative (duration of surgery, CPB, and aortic cross-clamp) factors (Table 2). The power, defined as the ability of a test to detect an effect given that the effect actually exists, of duration of dialysis, LVEF, and aortic and mitral valve calcification were 0.820, 0.600, and 0.538, respectively. The power of duration of surgery, CPB and aortic cross-clamp were 0.985, 0.696, and 0.538, respectively. There were no significant differences between the two groups with regard to preoperative respiratory factors (nor of intraoperative fentanyl and vecuronium bromide dosing) that were considered to influence to postoperative respiratory management. The rates of patients requiring PMV in the earlier period (from 1997 to 1999, n = 13) and in the latest period (from 2000, n = 31) were 60% and 55%, respectively.
We considered LVEF of <45%, duration of dialysis of >10 yr, and existence of both aortic and mitral valve calcification as preoperative risk factors and duration of surgery of >8 h as an intraoperative risk factor. These cut-off values represented the mean values recorded in group L. These 4 factors were subsequently entered into multiple logistic regression analysis to identify independent risk factors for PMV (>24 h). The analysis showed that duration of dialysis and duration of surgery were independent risk factors (Table 3).
On admission to the ICU, the respiratory and circulatory status of patients of group L was worse than that of group E, as evidenced by the significantly lower mean arterial blood pressure, significantly more rapid heart rate, higher mean pulmonary artery pressure and central venous pressure, and a lower Pao2/Fio2 ratio (Table 4). Furthermore, serum lactate was also significantly higher in group L than in group E. Postoperative total fluid balance during 12 h after initiation of CVVHD of group E was significantly less than that of group L (–1.0 ± 0.8 L versus –0.4 ± 0.8 L).
By definition, we achieved almost the optimal therapeutic goal in group E, i.e., tracheal extubation within 24 h after admission to the ICU and discharge from the ICU on the second postoperative day. On the other hand, the duration of mechanical ventilation and ICU stay in group L were much longer (Table 5), but early mortality (<30 days) was similar in both groups. Only one patient of group L died within 30 days after surgery. This patient had undergone double valve replacement and died as a result of rupture of the left ventricle on the second postoperative day.
|
Discussion
|
|---|
The number of registered dialysis-dependent patients in Japan was 206,134 at the end of the year 2000 (10) and this number has been steadily increasing in recent years. Cardiac disease is a leading cause of death in dialysis-dependent patients. Surgical treatment for cardiac patients, including surgeon skill, technique for providing CPB, myocardial protection during CPB, anesthetic management, and postoperative management in the ICU, has improved recently. However, the early mortality (<30 days) rate of dialysis-dependent patients who underwent cardiac surgery reported in a review of previous studies was frequent and varied from 4.4% to 16.9% (1–8). The reason for this wide range of early mortality rate was the inhomogeneous background of pooled patients resulting from the small numbers of participants (from 17 to 124). Unfortunately, even in "experienced" cardiac surgical institutions, the mortality rate after cardiac surgery is still more frequent than in patients with normal renal function. The hospital mortality rate after cardiac surgery is closely related to PMV (9). In addition, PMV increases overall hospital costs as well as ICU costs and reduces the availability of ICU beds for other critically ill patients. In dialysis-dependent patients, oxygenation often deteriorates as a result of lung congestion while fluid removal by hemodialysis is withheld and PMV complicates postoperative management. In this study, therefore, we investigated the risk factors of PMV in dialysis-dependent patients after undergoing cardiac surgery.
Univariate analysis showed that cardiac calcification, preoperative LVEF, duration of dialysis and duration of surgery were important predictors of PMV. Calcification of the coronary artery and cardiac valve are common in patients on dialysis for more than 10 years (11–12). Cardiac valve calcification was also found to be a significant predictor for mortality in long-term dialysis patients (13). Unfortunately, we did not use electron-beam computed tomography preoperatively to screen for calcification of the coronary artery. However, the proportion of patients of group L with calcification of both aortic and mitral valve was significantly larger than that of group E. Cardiac calcification may correlate with poor preoperative LVEF and difficult protection of the myocardium during CPB in addition to being associated with left ventricular hypertrophy and subendocardial ischemia (14). Furthermore, it also complicates the surgical procedure and prolongs the duration of surgery. Based on the poor preoperative LVEF and the long duration of the surgical procedure, the circulatory status on admission to the ICU was worse in group L than in group E, and the lower Pao2/Fio2 ratio suggested possible lung congestion resulting from cardiac dysfunction. In addition, long duration of CPB itself induced postoperative hypoxemia resulting from atelectasis (15). However, cardiac calcification and preoperative LVEF were not independent risk factors of PMV by multivariate analysis. This means "confounding" exists between LVEF and duration of dialysis and between calcification of the cardiac valves and duration of dialysis. We considered LVEF and calcification of the cardiac valves were closely related to duration of dialysis. Hence LVEF and calcification of the both cardiac valves were not considered to be "independent" risk factors, and we concluded duration of dialysis was the most important predictor of PMV.
The benefit of continuous renal replacement therapy (CRRT) is controversial, even for acute renal failure. Some authors showed that early and intensive CRRT for renal insufficiency after cardiac surgery was associated with a better survival rate (16–17). However Bouman et al. (18) concluded that early initiation of CRRT for critically ill patients did not improve survival at 28 days and recovery of renal function. Kellum et al. (19,20) indicated that current evidence was insufficient to draw strong conclusions regarding the mode of renal replacement therapy although they agreed with the life-saving potential. In chronic renal failure, CRRT after cardiac surgery has been used in a few institutions (6–7,16) but the time and conditions for the initiation were not described in detail. We consider that CRRT is a better mode of renal replacement therapy than conventional intermittent hemodialysis in the presence of unstable circulatory status just after cardiac surgery and pulmonary congestion together with base deficit. Therefore, we routinely start CVVHD just after cardiac surgery irrespective of circulatory status. The overall early mortality rate after cardiac surgery in the present study was 2.5%, which was less than that reported in other studies, especially in cardiac valve surgery (1–8). The better rate might be attributable to the removal of excessive fluid during the immediate postoperative period, as CVVHD decreases cardiac edema and induces a rapid recovery of cardiac function earlier than for patients on conventional intermittent hemodialysis, which is usually applied on the first postoperative day. Of course, cardiac function usually recovers with postoperative time lapse. However, excessive fluid is never removed without restarting dialysis and that time lag until restarting dialysis is critical and may influence prognosis. In this regard, keeping fluid balance negative during the immediate postoperative period may be the key to early tracheal extubation because postoperative total fluid balance for 12 hours after initiation of CVVHD of group E was significantly less than that of group L.
The limitation and shortcomings of our study were the small sample size with a long study period. We considered that the two risk factors of PMV, i.e., duration of dialysis and duration of surgery, were reliable, but we cannot exclude the possibility that other factors might become significant if the sample size is increased. Also, the change of our postoperative sedative protocol from midazolam to propofol might have influenced PMV. In addition, the latest management might be influenced by our experiences, as our study was not blinded. For example, CVVHD might have been started earlier in patients with unstable circulatory status. Unfortunately, <10 dialysis-dependent patients underwent cardiac surgery annually in our hospital and this number is quite similar to previous reports (1–8). Despite the small sample size, we consider it important to identify the risk factors of PMV because perioperative management of dialysis-dependent patients undergoing cardiac surgery with CPB is challenging for anesthesiologists and intensivists. A prospective multicenter study should be performed in the future.
In summary, we investigated the risk factors of PMV in 44 dialysis-dependent patients after cardiac surgery. The duration of dialysis and operative time were significant and independent predictors of PMV. Patients who required more than 24 hours of mechanical ventilation spent a longer time in the ICU. Therefore, efficient removal of excessive fluid and inotropic support for deteriorated cardiac function in the early postoperative period is important to avoid PMV especially in patients with a long history of dialysis who undergo prolonged surgical procedures.
|
Footnotes
|
|---|
Accepted for publication September 6, 2005.
|
References
|
|---|
- Christiansen S, Claus M, Philipp T, Reidemeister JC. Cardiac surgery in patients with end-stage renal failure. Clin Nephrol 1997;48:246–52.[Web of Science][Medline]
- Frenken M, Krian A. Cardiovascular operations in patients with dialysis-dependent renal failure. Ann Thorac Surg 1999;68:887–93.[Abstract/Free Full Text]
- Horst M, Mehlhorn U, Hoerstrup SP, et al. Cardiac surgery in patients with end-stage renal disease: 10-year experience. Ann Thorac Surg 2000;69:96–101.[Abstract/Free Full Text]
- Khaitan L, Sutter FP, Goldman SM. Coronary artery bypass grafting in patients who require long-term dialysis. Ann Thorac Surg 2000;69:1135–9.[Abstract/Free Full Text]
- Gelsomino S, Morocutti G, Gianluca M, et al. Open heart surgery in patients with dialysis d-dependent renal insufficiency. J Card Surg 2001;16:400–7.[Web of Science][Medline]
- Okamura Y, Mochizuki Y, Iida H, et al. Coronary artery bypass in dialysis patients. Artif Organs 2001;25:256–9.[Web of Science][Medline]
- Nishida H, Uchikawa S, Chikazawa G, et al. Coronary artery bypass grafting in 105 patients with hemodialysis-dependent renal failure. Artif Organs 2001;25:268–72.[Web of Science][Medline]
- Jault F, Rama A, Bonnet N, et al. Cardiac surgery in patients receiving long term hemodialysis: short and long term results. J Cardiovasc Surg (Torino) 2003;44:725–30.[Medline]
- Pappalardo F, Franco A, Landoni G, et al. Long-term outcome and quality of life of patients requiring prolonged mechanical ventilation after cardiac surgery. Eur J Cardiothorac Surg 2004;24:548–52.
- The current state of chronic dialysis treatment in Japan (as of December 31, 2000). Ther Apher Dial 2003;7:3–35.[Web of Science][Medline]
- Goodman WG, Goldin J, Kuizon BD, et al. Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis. N Engl J Med 2000;342:1478–83.[Abstract/Free Full Text]
- Raggi P, Boulay A, Chasan-Taber S, et al. Cardiac calcification in adult hemodialysis patients: a link between end-stage renal disease and cardiovascular disease. J Am Coll Cardiol 2002;39:695–701.[Abstract/Free Full Text]
- Wang AYM, Wang M, Woo J, et al. Cardiac valve calcification as an important predictor for all-cause mortality and cardiovascular mortality in long-term peritoneal dialysis patients: a prospective study. J Am Soc Nephrol 2003;13:159–68.
- Parfrey PS, Harnett JD, Barre PE. The natural history of myocardial disease in dialysis patients. J Am Soc Nephrol 1991;2:2–12.[Abstract]
- Weiss YG, Merin G, Koganov E, et al. Postcardiopulmonary bypass hypoxia: a prospective study on incidence, risk factors, and clinical significance. J Cardiothorac Vasc Anesth 2000;14:506–13.[Web of Science][Medline]
- Bent P, Tan HK, Bellomo R, et al. Early and intensive continuous hemofiltration for severe renal failure after cardiac surgery. Ann Thorac Surg 2001;71:832–7.[Abstract/Free Full Text]
- Demirkilic U, Kuralay E, Yenicesu M, et al. Timing of replacement therapy for acute renal failure after cardiac surgery. J Card Surg 2004;19:17–20.[Web of Science][Medline]
- Bouman CSC, Oudemans-van Straaten HM, Tijssen JG, et al. Effect of early high-volume continuous veno-venous hemofiltration on survival and recovery of renal function in intensive care patients with acute renal failure: a prospective, randomized trial. Crit Care Med 2002;30:2205–11.[Web of Science][Medline]
- Kellum JA, Angus DC, Johnson JP, et al. Continuous versus intermittent renal replacement therapy: a meta-analysis. Intensive Care Med 2002;28:29–37.[Web of Science][Medline]
- Kellum JA, Meath RL, Angus DC, et al. The first international consensus conference on continuous renal replacement therapy. Kidney Int 2002;62:1855–63.[Web of Science][Medline]
Top
Abstract
Introduction
