Comparison of and After Pediatric Cardiovascular Operation Franz F. Immer, MD, Franco P. Stocker, MD, Andrea M. Seiler, MD, Jean-Pierre Pfammatter, MD, Gerhardt Printzen, MD, and Thierry P. Carrel, MD Division of Pediatric Cardiology, Department of Clinical Chemistry, and Department of Cardiovascular Surgery, University Hospital, Berne, Switzerland Background. Although the diagnostic value of troponin-t in childhood is documented, little is known about the significance of troponin-i. It was the aim of this study to compare the diagnostic value of troponin-i and troponin-t in children and newborns to assess the perioperative potential myocardial damage. Methods. Forty-eight children, mean, 51 54 months (mean value 1 standard deviation) (range, 1 day to 204 months) undergoing cardiac operation were prospectively enrolled in the present study., troponin-t, creatine kinase (CK), and the MB isoenzyme were measured before operation and postoperatively within 2 days. Results. Postoperative values of troponin-i for children undergoing extracardiac operation were in the normal range. In children with interventions through the right atrium (n 10) the mean value increase to 6.5 6.1 g/l (range, 1.8 to 24.3 g/l) and even to a mean of 29.9 21.1 g/l (range, 7.5 to 90 g/l) (p < 0.01) in children with atrial and additional ventricular surgical approach (n 23). was of equal specificity and sensitivity compared to troponin-t, excepted in patients with postoperative renal failure in whom troponin-t raised to false pathological results. Conclusions. For detection of perioperative myocardial damage troponin-i shows a higher specificity than CK-MB activity and CK-MB mass. The diagnostic value of troponin-i is similar to troponin-t, but compared with troponin-t, it has the advantage of not being influenced by renal failure. (Ann Thorac Surg 1998;66:2073 7) 1998 by The Society of Thoracic Surgeons (ctni) and troponin-t (ctnt) are wellestablished markers of myocardial damage in adults [1 3]. Both are reported to be specific for myocardial injury, but in adults renal failure has been demonstrated to increase ctnt levels significantly [4 7]. It was the aim of this study to evaluate the diagnostic impact of ctni in children with myocardial damage and to compare it with that of ctnt, creatine kinase (CK), and the MB isoenzyme, which are actually the standard biochemical parameters in the assessment of myocardial damage [8]. As a model of myocardial damage children undergoing cardiac operation were studied, with a defined extent of myocardial lesion induced by the surgical manipulation, which leads to a systemic release of the troponin regulatory complex, as an indirect sign of myocardial damage. Patients and Methods Patients Forty-eight neonates and children suffering mainly from congenital heart disease and scheduled for cardiac operation met the entry criteria: there were no signs of myocardial damage preoperatively and repetitive simultaneous blood samples for serial measurement of ctni, ctnt, as well as CK activity and its isoenzyme CK-MB Accepted for publication May 19, 1998. Address reprint requests to Dr Stocker, Division of Pediatric Cardiology, University Hospital, 3010 Berne, Switzerland. were obtained. These children were studied prospectively between March 1995 and June 1996. The study protocol was approved by the University Hospital ethical committee. The average age of the patients was 51 54 months (range, 1 day to 204 months), 29 were boys. Forty-six children were operated for various congenital heart lesions, one presented with left atrial myxoma and one suffered from rheumatic valve disease. The patients were divided into three groups according to the surgical procedure. Group A consisted of 10 children (mean age, 49 35 months; range, 1 day to 92 months) undergoing surgical repair for aortic coarctation. Resection was followed by end-to-end anastomosis without any additional intracardiac operation in all patients. In 3 newborns the operation was performed as an emergency procedure. Group B consisted of 15 children (mean age, 92 65 months; range, 3 days to 204 months) operated on using moderate hypothermia in cardiopulmonary bypass including atrial surgical manipulation. In 8 patients a secundum atrial septal defect was closed, in 3 valve grafting was performed, 3 patients underwent arterial switch operation for simple transposition of the great arteries, and excision of a left atrial myxoma was performed in 1 child. Group C consisted of 23 patients (mean age, 25 32 months; range, 2 months to 133 months) operated with moderate to deep hypothermia and cardiopulmonary bypass with surgical approach through 1998 by The Society of Thoracic Surgeons 0003-4975/98/$19.00 Published by Elsevier Science Inc PII S0003-4975(98)00795-4
2074 IMMER ET AL Ann Thorac Surg TROPONIN-I AND TROPONIN-T IN CHILDREN 1998;66:2073 7 the atria and the ventricles. Ventricular manipulation was either done by a transatrial approach or through a ventriculotomy. Of these children closure of a ventricular septal defect was performed in 10, correction of a complete atrioventricular defect in 7, repair of tetralogy of Fallot or double-outlet right ventricle in 4, and an isolated procedure for relief of right ventricular outflow tract stenosis in 2 patients. Methods Blood samples were taken preoperatively and at 8, 18 to 28, and 42 to 55 hours postoperatively. As the ethical committee allowed only blood samples for all markers at the time of routine blood tests, it was not possible to have a complete follow-up in all patients between 42 and 55 hours postoperatively. The following tests were used: ctni (Sandwich-Immunoassay-Test, Dade, Miami, FL), ctnt (Immunoassay-Test, Boehringer Mannheim, Mannheim, Germany), CK activity (CK-NAC; Boehringer Mannheim), CK-MB activity (Boehringer Mannheim), and CK-MB mass (Sandwich-Immunoassay-Test). All analyses were performed at 37 C, except ctnt, which was analyzed at 25 C. In 8 children the measurements of ctni were performed in frozen serum, not older than 4 months. For CK-MB activity results in the range between 6% and 20% of total CK activity were considered a sign of myocardial damage in accordance with Stein and colleagues [9]. Values more than 20% of total CK activity were excluded as they can be attributable to macro-ck [10]. For ctni the upper limit of normal is set at 0.6 g/l [11] and for ctnt at 0.1 g/l [12]. For CK activity the upper normal limit is defined at 136 U/L during the first year of life and at 94 U/L later on [13]. The upper limit of normal for CK-MB mass is defined at 4 g/l without age differences and CK-MB activity was considered normal up to 6% of total CK activity [9, 13]. Patients were assigned to one of the three groups according to the extent of reported surgical manipulation. Duration of cardiopulmonary bypass, aortic crossclamping time, and reperfusion time, as well as intraoperative technical details were collected. Serum creatinine values were obtained at different times during the early postoperative period. Group A patients underwent operation without extracorporeal circulation. Myocardial protection in groups B and C was realized by intermittent antegrade cold blood cardioplegia and hot-shot before reperfusion. Short periods of circulatory arrest in deep hypothermia were reported in some patients of group C. Statistical Analysis Statistical tests were performed by using a two-tailed z-test for average differences of a standard repartition. A p value less than 0.05 was considered significant. Results were expressed as mean values with indication of the ranges and the standard deviation ( 1 standard deviation). Fig 1. (A) and troponin-t (B) in the early postoperative period. Mean values and standard deviation ( 1 standard deviation). Dotted line shows upper limit of norm ( 1.0). Preoperative and maximum values up to 28 hours postoperatively for each group (A, B, C). Total number of patients and total number of pathological results are indicated below the graph. Results All 48 children had normal preoperative ctni values, with a mean value of 0.06 0.3 g/l (range, 0 to 0.5 g/l) (Fig 1A). In group A, children undergoing extracardiac operation, no postoperative myocardial damage was expected and all children effectively demonstrated normal values except one newborn in whom ctni increased to 3.6 g/l. All patients of group B and C had, during the same early postoperative period, maximum values for ctni clearly exceeding the upper limit of normal. There was a significant difference (p 0.01) between these two groups with a mean maximum value of 5.5 6.1 g/l (range, 1.8 to 24.3 g/l) in group B and a mean value of 29.9 21.1 g/l (range, 7.5 to 90.0 g/l) in group C, indicating that the surgical trauma was effectively more important in group C. These results lead to a positive predictive value for ctni of 100% and a negative predictive value of 98%. Creatine Kinase and Its Isoenzyme MB Preoperatively, 21 children (44%) had already pathological CK activity values. In the first 28 postoperative hours
Ann Thorac Surg IMMER ET AL 1998;66:2073 7 TROPONIN-I AND TROPONIN-T IN CHILDREN 2075 Fig 2. Creatine kinase (A), CK-MB activity in percentage of total creatine kinase (B), and CK-MB mass (C) in the early postoperative period. Mean values and standard deviation ( 1 standard deviation). Dotted line shows upper limit of norm ( 1.0). Preoperative and maximum values up to 28 hours postoperatively for each group (A, B, C). Total number of patients and total number of pathologic results are indicated below the graph. all group A patients (extracardiac operation) demonstrated pathological results as did group B and C children (Fig 2A). Between these three groups no significant difference of the maximum values could be observed. Therefore, the positive predictive value was 100% and the negative predictive value was only 47%. Preoperative CK-MB activity levels were found to be elevated in 11 children, indicating myocardial damage (mean, 13.8% 4.1% of total CK activity; range, 3.3% to 20%), whereas in only 1 child preoperatively a normal value with 3.3% was found (Fig 2B). The remaining 36 children had to be excluded, as their values exceeded 20%, which could be related to macro-ck. Postoperatively in group A patients, all values were normal, except in 2 newborns in whom CK-MB activity was slightly elevated (mean, 4.4% 1.6%; range, 2.1% to 8.6%). In group B and C patients pathological values were found but there is a great overlap with the normal range. In group B patients 10 values (67%) and in group C patients 3 values (13%) were within the normal range. On the basis of these results a positive predictive value of 66% and a negative predictive value of 68% was calculated. CK-MB Mass The preoperative mean value was in the normal range with some overlapping into the pathological range in 5 children (10%). Postoperatively elevated CK-MB mass was observed in all group A children, in whom normal results after extracardiac operation were expected. CK-MB mass in group B and C patients showed a similar behavior to that of ctni with all values in the pathological range and with a significant difference between the values of groups B and C. Positive predictive value was calculated to be 100% and the negative predictive value was found to be 74%. Comparison of and In contrast to ctni, preoperative ctnt was slightly elevated in 2 children, both suffering from impaired renal function with elevated serum creatinine (Fig 1B). Among group A patients 3 newborns, who underwent emergency operation for aortic coarctation postoperatively, had pathological maximum values for ctnt during the first 28 hours, as compared to only 1 patient with an elevated value of ctni. In group B and C patients ctni and ctnt demonstrated very similar results (p not significant) during the first 28 postoperative hours (Fig 1A,B). The positive predictive value for ctnt was as high as 100% and the negative predictive value was 97%. This observation cannot be extrapolated for the late period between 28 and 55 hours after operation (Table 1) in 17 children of group C with a complete follow-up up to this late period (Fig 3). The course of ctni in all of these children showed progressive regression, whereas ctnt values showed a decrease only in some children. However 7 children (41%) had values remaining on a plateau or even increasing at this time. Studying the relation between late postoperative course of ctni and ctnt in relation to serum creatinine we found a different behavior corresponding to the serum creatinine level (Fig 3). The value of R 2 for ctnt of 0.0003 is statistically significant (p 0.01) compared with the value of 0.2218 for ctni.
2076 IMMER ET AL Ann Thorac Surg TROPONIN-I AND TROPONIN-T IN CHILDREN 1998;66:2073 7 Table 1. in 17 Children of Group C With Complete Follow-up up to 55 h Postoperatively (t 8) (t 18 28) (t 12 55) Group C (n 23) No. of patients 17 17 17 17 17 17 Mean value 5.35 28.89 6.14 25.62 6.68 8.89 Min/max 0.43/13.87 6.7/76.8 0.95/13.49 6.5/62.7 2.3/25.4 0.63/23.25 No. of pathologic values 17 17 17 17 17 17 Comment The advantage of ctni and ctnt over creatine kinase and its MB isoenzyme in the assessment of myocardial injury in adult patients with myocardial infarction, in patients undergoing percutaneous transluminar coronary angiography, or cardiac operation have been demonstrated in numerous publications [1 3]. Several studies on the value of ctnt confirmed its diagnostic accuracy in children [8, 12, 14 17], but little is known about the value of ctni [8, 11, 18] in neonates and children. Our aim was to investigate the relation between ctni and CK and its MB isoenzyme and ctnt. Because myocardial infarction is extremely rare in children, we studied three groups of children who underwent either extracardiac or intracardiac operation that caused a welldefined surgical trauma to the myocardium. The group of children with surgical repair for aortic coarctation served as the control group. Compared with creatine kinase and its MB isoenzyme, ctni determination showed a clear advantage with regard to specificity and sensitivity in the assessment of myocardial damage in the children studied. This is underlined by a preoperative elevation of CK, which cannot be explained, whereas the postoperative increase in group A patients could be attributable to skeletal muscle damage after thoracotomy. This observation is also confirmed by the postoperative increase of CK-MB mass values in all group A children and by the normal CK-MB activity of a substantial number of children in groups B and C, in addition to a nonexplained preoperative increase in all 3 groups, which probably is attributable to macro-ck [10]. Comparing ctni to ctnt two different conclusions can be drawn. In the early postoperative course up to 28 hours postoperatively, ctni and ctnt behaved in a similar way. The significant difference in the values for ctni and ctnt between groups B and C indicates the correlation to the extent of myocardial damage, in accordance with reports in adults [19, 20] and in children [8, 12, 18]. On the other hand, looking at the late postoperative course from 28 to 55 hours, it was found that ctni and ctnt do not behave similarly. This is shown by the results of 17 children of group C who had a complete follow-up. Although a steady decrease of ctni levels was observed in all children, the course of ctnt did not show the expected regression. In some patients, instead of a decrease a plateau or even an increase was found. As there was no evidence for a new myocardial damage during the postoperative course, an influence of impaired renal function on ctnt was suspected, as it has been reported in adults [4 7]. Indeed we were able to show a correlation between the course of ctnt and serum creatinine levels in this late postoperative period (Fig 3). Despite the identical behavior in the early postoperative phase, this late postoperative difference indicates an important advantage of ctni over ctnt. Fig 3. Relation of troponin-i and troponin-t with serum creatinine. Changes of values between 18 to 28 hours compared with values between 42 to 55 hours. Results are given in percentage of increase or decrease.
Ann Thorac Surg IMMER ET AL 1998;66:2073 7 TROPONIN-I AND TROPONIN-T IN CHILDREN 2077 For clinical use ctni will improve the assessment of postoperative myocardial damage, for example, in children after arterial switch operation. Concerning the influence of the duration of the cardiopulmonary bypass and of aortic cross-clamping time, as well as younger age, a trend toward higher ctni and ctnt values with longer duration and with younger age was observed. But this may be attributable to the age of the patients (younger children, in the first year of life) and to the length and more complex surgical procedures; a conclusive interpretation cannot be drawn in the present study. This tendency was already described by Taggart and colleagues [18] in a recently published study. We conclude that ctni as well as ctnt are more specific and more sensitive markers of the extent of myocardial injury in children (including newborns) compared with CK and MB isoenzyme. Furthermore ctni has the advantage of not being influenced significantly by impaired renal function; as a result we prefer to use ctni in our clinical routine. References 1. Apple FS, Sharkey SW, Henry TD. Early serum cardiac troponin I and T concentrations after successful thrombolysis for acute myocardial infarction. Clin Chem 1995;41; 1197 8. 2. De Winter RJ, Koster RW, Sturk A, Sanders GT. Value of myoglobin, troponin T, and CK MB mass in ruling out an acute myocardial infarction in the emergency room. Circulation 1995;92:3401 7. 3. Lee TH, Thomas EJ, Ludwig LE, et al. Troponin T as a marker for myocardial ischemia in patients undergoing major noncardiac surgery. Am J Cardiol 1996;77;1031 6. 4. Bhayana V, Gououlias T, Cohoe S, Henderson AR. Discordance between results for serum troponin T and troponin I in renal disease. Clin Chem 1995;41:312 7. 5. Braun SL, Baum H, Neumeier D, Vogt W. Troponin T and troponin I after coronary artery bypass grafting: discordant results in patients with renal failure. Clin Chem 1996;42: 781 3. 6. Croitoru M, Taegtmeyer H. Spurious rises in troponin T in end stage renal disease. Lancet 1995;346:974. 7. Li D, Jialal I, Keffer J. Greater frequency of increased cardiac troponin T than increased troponin I in patients with chronic renal failure [Letter]. Clin Chem 1996;42:114 5. 8. Taggart DP, Hadjinikolas L, Wong K, et al. Vulnerability of paediatric myocardium to cardiac surgery. Heart 1996;76: 214 7. 9. Stein W, Bohner J, Renn W, Maulbetsch R. Macro creatine kinase type 2: results of prospective study in hospitalized patients. Clin Chem 1985;31:1959 64. 10. Collinson PO, Rosalki SB, Flather M, Wolman R, Evans T. Early diagnosis of myocardial infarction by timed sequential enzyme measurements. Ann Clin Biochem 1988;25:376 82. 11. Grant JW, Canter CE, Spray TL, Landt Y. Elevated donor cardiac troponin I: a marker of acute graft failure in infant heart recipients. Circulation 1994;90:2618. 12. Immer FF, Stocker F, Seiler AM, Pfammatter J P, Printzen G, Peheim E. Cardiac troponin T: improved diagnostic assessment of myocardial damage in childhood. Acta Paediatrica 1997;86:1321 7. 13. Thomas L. Creatine kinase (Gesamt CK), creatine kinase MB (CK MB). In: Thomas L, ed. Labor und Diagnose. (3. Auflage). Marburg: Die Medizinische Verlagsgesellschaft, 1988:91 105. 14. Agnoletti G, Panthegini M, Borghi A. Troponin T as a diagnostic tool in ischemic myocardial dysfunction of the newborn [Abstract]. Cardiology in the Young 1996;6(Suppl 1):30. 15. Emmel M, Bauer I, Mennicken U, Borowski A, de Vivie ER, Geisel J. Kardiales Troponin T als Marker für perioperative Myokardzellschädigung im Kindesalter [Abstract]. Zeitschrift für Kardiologie 1994;83:692. 16. Fink FM, Genser N, Fink C, et al. Cardiac troponin T and creatine kinase MB mass concentrations in children receiving anthracycline chemotherapy. Med Pediatr Oncol 1995; 25:185 9. 17. Kaku S, Mirand AF, Pachica F, Pinto F, Trigo C. Is cardiac troponin T a good marker for assesement of cardiac injury in children? [Abstract]. Cardiol Young 1996;6(Suppl 1):33. 18. Taggart DP, Hadjinikolas L, Hooper J, et al. Effects of age and ischemic times on biochemical evidence of myocardial injury after pediatric cardiac operations. J Thorac Cardiovasc Surg 1997;113:728 35. 19. Omura T, Teragaski M, Takagi M, et al. Myocardial infarction size by serum troponin T and myosin light chain 1 concentration. Jpn Circ J 1995;59:154 9. 20. Wagner I, Mair J, Fridrich L, et al. Cardiac Troponin T release in acute myocardial infarction is associated with scintigraphic estimates of myocardial scar. Coron Artery Dis 1993;4:537 44. New Senior Production Editor It is with regret that we announce the resignation of our Senior Production Editor, Mr Roger D. Morrison. Roger worked in the offices of our publisher, Elsevier Science Inc, New York, and was responsible for all aspects of journal production, including design layout and quality control. Our office and his were in constant communication, often as many as a dozen times a day. Roger has worked in this capacity for The Annals for 11 years. We shall certainly miss him, and wish him well in his new venture. At the same time, we look forward to working with our new Senior Production Editor, Mr Kenneth J. Che. Ken is new to The Annals, but not to Elsevier, having worked with the company for 2 years. But for the future, Ken is our man in New York. Welcome, Ken. Tom Ferguson Editor, The Annals of Thoracic Surgery 1998 by The Society of Thoracic Surgeons Ann Thorac Surg 1998;66:2077 0003-4975/98/$19.00 Published by Elsevier Science Inc