Ann Thorac Surg 2004;77:61-65
© 2004 The Society of Thoracic Surgeons
Original article: cardiovascular
Predictors of mortality at initiation of peritoneal dialysis in children after cardiac surgery
Harald Boigner, MDa*,
Werner Brannath, PhDb,
Michael Hermon, MDa,
Elisabeth Stoll, MDa,
Gudrun Burda, MDa,
Gerhard Trittenwein, MDa,
Johann Golej, MDa
a Department of Neonatology and Pediatric Intensive Care, University Children's Hospital of Vienna, Vienna, Austria
b Institute of Medical Statistics, University of Vienna, Vienna, Austria
Accepted for publication July 29, 2003.
* Address reprint requests to Dr Boigner, Department of Neonatology and Pediatric Intensive Care, University Children's Hospital of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria.
e-mail: harald.boigner{at}akh-wien.ac.at
 |
Abstract
|
|---|
BACKGROUND: The development of renal dysfunction in the postoperative course of cardiac surgery is still associated with high mortality in pediatric patients. In particular for small infants peritoneal dialysis offers a secure and useful treatment option. The aim of the present study was to investigate if routinely used laboratory and clinical variables could help predict mortality at initiation of peritoneal dialysis.
METHODS: We performed a retrospective chart analysis of pediatric intensive care unit patients with renal dysfunction who were treated with peritoneal dialysis after cardiac surgery between 1993 and 2001 and analyzed variables obtained 3 hours or less before starting peritoneal dialysis.
RESULTS: Results are documented as means and standard errors. A total of 1,141 children underwent a cardiac operation on cardiopulmonary bypass. Sixty-two children (5.4%) were treated with peritoneal dialysis. Mortality was 40.3% (37 survivors, 25 nonsurvivors). The pH in survivors was 7.35 (0.01); in nonsurvivors it was 7.23 (0.03; p = 0.0037). Base excess in survivors was -1.37 mmol/L (0.61); in nonsurvivors it was -7.17 mmol/L (1.49; p = 0.0026). Lactate in survivors was 4.5 mmol/L (0.60); in nonsurvivors it was 10.5 mmol/L (1.78; p = 0.0089). Positive inspiratory pressure in survivors was 24.6 cmH2O (0.78); in nonsurvivors it was 28.9 cmH2O (1.08; p = 0.0274). Tidal volume per kilogram bodyweight in survivors was 11.0 mL/kg (0.48); in nonsurvivors it was 8.7 mL/kg (0.50; p = 0.0493).
CONCLUSIONS: We conclude from our data that the consideration of pH, base excess, lactate, positive inspiratory pressure, and tidal volume per kilogram bodyweight help predict mortality at initiation of peritoneal dialysis. We were able to observe significant differences between survivors and nonsurvivors using these variables.
|
Introduction
|
|---|
Treatment of acute renal failure after cardiac surgery remains to be a challenge in pediatric critical care. Despite progress in cardiac operation and intensive care techniques the mortality rate is still high in this group of patients [1, 2]. The modalities of renal replacement therapy are influenced by the size of the child and the personal experience of the medical staff. In neonates and small infants acute renal replacement therapy is still preferably carried out as peritoneal dialysis because the procedure is simple and safe; furthermore cannulation of the great vessels, systemic anticoagulation, and the associated risks of ischemic and embolic complications, as described during extracorporeal techniques, are avoided [3, 4]. Suggesting significant differences of laboratory and clinical variables in survivors and nonsurvivors at the time of peritoneal dialysis initiation we reviewed the medical records of 62 pediatric intensive care unit patients treated by peritoneal dialysis for renal dysfunction after cardiac surgery, thus identifying children at high risk of mortality.
|
Material and methods
|
|---|
Study population
The records of all children with peritoneal dialysis after cardiac surgery by means of cardiopulmonary bypass who were treated in our pediatric intensive care unit between June 1993 and December 2001 were reviewed retrospectively. Exclusion criteria were a postoperative course of extracorporeal membrane oxygenation, peritoneal dialysis for less than 1 day, or incomplete data. Hemodynamic data were continuously monitored by means of a Hewlett-Packard monitoring system (Model 68 S; Hewlett-Packard, Palo Alto, CA). The patients were ventilated with Baby-log 8000 or with EVITA conventional ventilators (Dräger, Lübeck, Germany).
Measurements
Age, weight, congenital heart defect, cause of death, indication for peritoneal dialysis, time interval between operation and peritoneal dialysis, and duration of peritoneal dialysis were obtained from the medical records. Heart frequency before and at initiation of peritoneal dialysis, pH, bicarbonate, base excess, central venous oxygen saturation, serum-lactate, serum-potassium, serum-sodium, serum-calcium, serum-phosphate, blood urea nitrogen, serum-creatinine, positive inspiratory airway pressure (PIP), positive endexpiratory airway pressure, tidal volume, mean arterial blood pressure, central venous blood pressure, and urinary output were extracted from the medical records at the following point of time: last blood sample taken before initiation of peritoneal dialysis (
3 hours before initiation of peritoneal dialysis). Fluid balance and fluid supply per kilogram bodyweight during the last 3 hours before initiation of peritoneal dialysis, as well as tidal volume per kilogram bodyweight (TV/kg) were calculated.
Indication for peritoneal dialysis
The indication for peritoneal dialysis after cardiac surgery included hypervolemia (namely physical evidence of total body fluid overload, central venous pressure > 10 cm H2O, and positive fluid balance that were not treatable otherwise); oliguria (urinary output < 1 mL/kg per hour); anuria for more than 4 hours unresponsive to adjustment in fluid therapy, diuretics, inotropic support, or any combination thereof; and hyperkalemia (serum potassium concentration 
6.5 mmol/L).
Performance of peritoneal dialysis
An original curled and double-cuffed peritoneal Tenckhoff catheter (Quinton; Sherwood Medical Company, St. Louis, MO) was inserted either in the intensive care unit or in the operating room. The tubing system was a Safe Lock PD System (Fresenius Medical Care AG, Bad Homburg, Germany), and the heat exchanger was an Astotherm (Stikler Electronic GmbH, Stuttgart, Germany). In all cases the applied dialysis solution was a potassium-free bicarbonate solution containing variable glucose concentrations (1.5, 2.5, and 4.25 g/dL). Dialysis was initiated with a 1.5 g/dL solution. After two quick initial cycles to remove air and enable unrestricted flow through the catheter, 10 mL per kg body weight warmed dialysis solution was infused during each cycle and was exchanged every hour. The cycle length and dialyses solution were adapted to the patient's hemodynamic and metabolic condition.
In case of visible clots in the tubing system or suspected infection, heparin (200 IE/L dialysis solution) and cefamandol (250 mg/L dialysis solution) were added to the dialysis solution. The cycle length was extended up to 6 hours if hypervolemia resolved and hemodynamic state improved.
Data analysis
Data are summarized by means and standard errors. For univariate two group comparisons Westfall and Young's nonparametric step down permutation test algorithm was used [5]. Multivariate influences on mortality were evaluated by a stepwise logistic regression analysis restricted to the variables that showed significant influence in the univariate analysis. Bivariate associations were analyzed using Spearman's rank correlation coefficient. A p value less than 0.05 is considered to indicate statistical significance.
|
Results
|
|---|
Between June 1993 and December 2001, 1,141 children underwent a cardiac operation on cardiopulmonary bypass at the University hospital of Vienna. A total of 62 of the 1,141 children (5.4%) were treated with peritoneal dialysis and were eligible for this retrospective study. Table 1
shows the different congenital heart defects and the operations performed. Table 2
shows the number of patients of each indication for peritoneal dialysis. The cause of death was low cardiac output in 21 cases (acute ventricular dilation [9 of 21], myocardial infarction [3 of 21], arrhythmia [4 of 21], and ventricular pump failure [5 of 21]), sepsis in 3 cases, and cerebral hemorrhage in 1 case. Data of 62 children, 37 in the survivor group and 25 in the nonsurvivor group, were analyzed. All measured variables are reported separately for survivors and nonsurvivors in Table 3.
View this table:
[in this window]
[in a new window]
|
Table 3. Comparison of Survivors and Nonsurvivors for All Measured Variables (3 Hours Before Initiation of Peritoneal Dialysisa
|
|
The Westfall and Young multiple test procedure shows significant differences between survivors and nonsurvivors for the variables pH, base excess, lactate, PIP, and TV/kg (see p values in Table 3). The correlation analysis indicates that pH correlates positively with base excess (r = 0.81). Lactate correlates negatively with pH (r = -0.44) and with base excess (r = -0.47). The stepwise logistic regression analysis indicates a multivariate influence of pH (p = 0.017), PIP (p = 0.003), and TV/kg (p = 0.0068) on mortality, namely each of these variables shows a significant influence even when accounting for the influence of the others. The mortality rate is found to increase with increasing PIP and with decreasing pH and TV/kg. Owing to their univariate significance and the correlation with pH, the variable base excess and lactate can be expected to influence mortality in a similar way as pH, with a reversed sign for lactate however.
|
Comment
|
|---|
In this retrospective study we reviewed the medical records of pediatric intensive care unit patients treated for renal dysfunction after cardiac surgery. When we compared survivors and nonsurvivors we found that pH, base excess, lactate, PIP, and TV/kg could significantly identify children at high risk of mortality at time of peritoneal dialysis initiation after cardiac surgery whereas mean arterial pressure, central venous blood pressure, central venous oxygen saturation, and urinary output could not.
Renal dysfunction after cardiac surgery has been reported to have an important impact on postoperative mortality [1]. A postoperative mortality rate as high as 79% has been noticed in this group of patients [6, 7]. The mortality rate in our group was 40.3%, which is comparable with others, but variations in patient demographics, dialysis modality, and indication for initiation of renal support makes a comparison of the outcome difficult [2, 8]. There is concern about variables that improve surveillance of patients after cardiac surgery and additionally have a predictive value for outcome. Therefore we have analyzed laboratory and clinical variables routinely used at our pediatric intensive care unit in the care of postoperative pediatric cardiac patients.
Base excess has been used as an indicator of the severity of shock, of the efficacy of resuscitation in animal models [9, 10], and as a predictor of mortality when obtained at admission in a retrospective study of 515 pediatric trauma patients [11]. In postoperative pediatric patients base excess has failed to predict mortality as yet but titration of bicarbonate during therapy may have influenced the predictive value of this variable in these studies [12, 13]. Contrasting these findings base excess levels taken before initiation of peritoneal dialysis in our study did not only discriminate significantly survivors from nonsurvivors but also correlated strongly with lactate, another valuable indicator, which is generated by anaerobic metabolism and known to reliably detect inadequate tissue oxygen delivery. Unlike base excess, lactate is not affected directly by the use of buffer and is an often reported predictor of mortality. Particularly in pediatric patients undergoing cardiac surgery for congenital heart disease elevated serum lactate levels have been found to be a serious indicator of potential mortality especially if serum lactate is greater than 4.2 mmol/L [14, 12]. These findings are in accordance with our results as we found a lactate level of 4.5 mmol/L (0.60) in survivors and 10.5 mmol/L (1.78) in nonsurvivors, which demonstrated that serum lactate levels taken before initiation of peritoneal dialysis after cardiac surgery differed significantly between survivors and nonsurvivors. It remains speculative if the predictive value of serum lactate in our study could have been increased by serial measurements as demonstrated in a study by Charpie and coworkers [13].
Although postoperatively increased lactate levels most likely result from low cardiac output, direct measurement of cardiac output could only be accomplished by means of a pulmonary artery catheter. The value of direct measurement of cardiac output in the pediatric population still remains under discussion. Although direct measurement of cardiac output by pulmonary artery catheters has been described as a safe technique in some studies it may be complicated by technical difficulties especially in the smallest infants [11, 16]. In addition data about improved outcome by means of direct measurement of cardiac output in the pediatric population are still missing. Alternatively central venous oxygen saturation might be used as an estimate of mixed venous oxygen saturation obtained by pulmonary artery catheter and reflects the balance of oxygen supply and demand. In a study of 51 adult patients central venous oxygen saturation was found to be a good estimate of mixed venous oxygen saturation [17]. In children after open heart surgery, even in cyanotic patients central venous oxygen saturation proved to be a significant cofactor for the evaluation of postoperative mortality [18]. On the contrary the reliability of central venous oxygen saturation as a substitute of mixed venous oxygen saturation has been doubted by Martin and colleagues [19] and by Vincent [20]. Although central venous oxygen saturation was not significant in predicting mortality in our study, which may be explained by patient selection, we consider central venous oxygen saturation to be a helpful tool for therapeutic decision making in pediatric postoperative care.
Deterioration of pulmonary function after cardiac surgery has been due to intraoperative mechanical lung damage, cardiopulmonary bypass induced systemic inflammation, and in particular to impairment of fluid management. Although we cannot exclude a certain influence of the first and second factor, impairment of fluid management seems to be of considerable importance in patients suffering from renal dysfunction. It has been shown most recently that peritoneal dialysis, especially carried out as low-volume peritoneal dialysis, can achieve adequate fluid balance owing to ultrafiltration even in hemodynamically instable pediatric intensive care unit patients [21]. Werner and associates [2] postulated a reduction of hypervolemia and extravascular lung water by means of peritoneal dialysis as reason for improved pulmonary function in postoperative pediatric patients treated with peritoneal dialysis. However, the need for high PIP and resulting low TV/kg showed a significantly higher mortality in our study. Because of the combination of impaired pulmonary function, low base excess, and high lactate in nonsurvivors one could suggest that these patients were in protracted shock. In this situation peritoneal dialysis might be irrelevant for a better outcome. Maybe more invasive treatment procedures could have improved outcome. However peritoneal dialysis seemed justified as priority treatment for postoperative renal failure because of obvious problems in fluid management and an expected additional beneficial effect on low cardiac output.
In summary we have reviewed laboratory and clinical data of pediatric patients suffering from renal dysfunction after cardiac surgery before initiation of peritoneal dialysis. When we compared survivors with nonsurvivors we found significant differences between the groups with regard to pH, base excess, lactate, PIP, and TV/kg. These variables revealed children at high risk of mortality before initiation of peritoneal dialysis treatment in our study. We therefore suggest that in those patients an early initiation of alternative circulatory support by means of extracorporeal devices should be considered. [15]
|
Acknowledgments
|
|---|
The authors would like to thank Thomas Lang, MS, for his expertise in biostatistics.
|
References
|
|---|
- Fleming F., Bohn D., Edwards H., et al. Renal replacement therapy after repair of congenital heart disease in children. A comparison of hemofiltration and peritoneal dialysis. J Thorac Cardiovasc Surg 1995;109:322-331.[Abstract/Free Full Text]
- Werner H.A., Wensley D.F., Lirenman D.S., LeBlanc J.G. Peritoneal dialysis in children after cardiopulmonary bypass. J Thorac Cardiovasc Surg 1997;113:64-70.[Abstract/Free Full Text]
- Flynn J.T. Choice of dialysis modality for management of pediatric acute renal failure. Pediatr Nephrol 2002;17:61-69.[Medline]
- Stromberg D., Fraser C.D., Jr, Sorof J.M., Drescher K., Feltes T.F. Peritoneal dialysis. An adjunct to pediatric postcardiotomy fluid management. Tex Heart Inst J 1997;24:269-277.[Medline]
- Westfall P.H., Young S.S. Resampling-based multiple testing. . New York: Wiley, 1993.
- Picca S., Principato F., Mazzera E., et al. Risks of acute renal failure after cardiopulmonary bypass surgery in children: a retrospective 10-year case-control study. Nephrol Dial Transplant 1995;10:630-636.[Abstract/Free Full Text]
- Reznik V.M., Griswold W.R., Peterson B.M., Rodarte A., Ferris M.E., Mendoza S.A. Peritoneal dialysis for acute renal failure in children. Pediatr Nephrol 1991;5:715-717.[Medline]
- Rigden S.P., Barratt T.M., Dillon M.J., De Leval M., Stark J. Acute renal failure complicating cardiopulmonary bypass surgery. Arch Dis Child 1982;57:425-430.[Abstract/Free Full Text]
- Davis J.W., Shackford S.R., Holbrook T.L. Base deficit as a sensitive indicator of compensated shock and tissue oxygen utilization. Surg Gynecol Obstet 1991;173:473-476.[Medline]
- Davis J.W. The relationship of base deficit to lactate in porcine hemorrhagic shock and resuscitation. J Trauma 1994;36:168-172.[Medline]
- Kincaid E.H., Chang M.C., Letton R.W., Chen J.G., Meredith J.W. Admission base deficit in pediatric trauma: a study using the National Trauma Data Bank. J Trauma 2001;51:332-335.[Medline]
- Siegel L.B., Dalton H.J., Hertzog J.H., Hopkins R.A., Hannan R.L., Hauser G.J. Initial postoperative serum lactate levels predict survival in children after open heart surgery. Intens Care Med 1996;22:1418-1423.[Medline]
- Charpie J.R., Dekeon M.K., Goldberg C.S., Mosca R.S., Bove E.L., Kulik T.J. Serial blood lactate measurements predict early outcome after neonatal repair or palliation for complex congenital heart disease. J Thorac Cardiovasc Surg 2000;120:73-80.[Abstract/Free Full Text]
- Cheifetz I.M., Kern F.H., Schulman S.R., Greeley W.J., Ungerleider R.M., Meliones J.N. Serum lactates correlate with mortality after operations for complex congenital heart disease. Ann Thorac Surg 1997;64:735-738.[Abstract/Free Full Text]
- Flori H.R., Johnson L.D., Hanley F.L., Fineman J.R. Transthoracic intracardiac catheters in pediatric patients recovering from congenital heart defect surgery: associated complications and outcomes. Crit Care Med 2000;28:2997-3001.[Medline]
- Fuenfer M.M., Georgeson K.E., Cain W.S., et al. Etiology and retrieval of retained central venous catheter fragments within the heart and great vessels of infants and children. J Pediatr Surg 1998;33:454-456.[Medline]
- Berridge J.C. Influence of cardiac output on the correlation between mixed venous and central venous oxygen saturation. Br J Anaesth 1992;69:409-410.[Abstract/Free Full Text]
- Trittenwein G., Pansi H., Graf B., et al. Proposed entry criteria for postoperative cardiac extracorporeal membrane oxygenation after pediatric open heart surgery. Artif Organs 1999;23:1010-1014.[Medline]
- Martin C., Auffray J.P., Badetti C., Perrin G., Papazian L., Gouin F. Monitoring of central venous oxygen saturation versus mixed venous oxygen saturation in critically ill patients. Intens Care Med 1992;18:101-104.[Medline]
- Vincent J.L. Does central venous oxygen saturation accurately reflect mixed venous oxygen saturation? Nothing is simple, unfortunately. Inten Care Med 1992;18:386-387.
- Golej J., Kitzmueller E., Hermon M., Boigner H., Burda G., Trittenwein G. Low-volume peritoneal dialysis in 116 neonatal and paediatric critical care patients. Eur J Pediatr 2002;161:385-389.[Medline]
Top
Abstract
Introduction
