S-100 After Correction of Congenital Heart Defects in Neonates: Is It a Reliable Marker for Cerebral Damage?

Size: px

Start display at page:

Download "S-100 After Correction of Congenital Heart Defects in Neonates: Is It a Reliable Marker for Cerebral Damage?"

Carol Wells
5 years ago
Views:

1 S-100 After Correction of Congenital Heart Defects in Neonates: Is It a Reliable Marker for Cerebral Damage? Michael A. Erb, MD, Markus K. Heinemann, MD, Hans P. Wendel, PhD, Leo Häberle, MD, Ludger Sieverding, MD, Christian P. Speer, MD, and Gerhard Ziemer, MD, PhD Division of Thoracic and Cardiovascular Surgery, Department of Surgery, Department of Anesthesiology, Divisions of Cardiology and Neonatology, Department of Pediatrics, Tübingen University Hospital, Tübingen, Germany Background. Newborns undergoing cardiac operation may acquire some extent of neuronal damage. An early diagnosis is especially hard regarding neonates. In the past years, S-100 has been widely discussed as a marker revealing perioperative damage to the brain. Methods. Sequential blood samples from 33 neonates undergoing repair of congenital heart disease were taken perioperatively. Samples of 12 healthy neonates were taken at birth as a control group. The newborns were divided into four groups: cyanotic and acyanotic disease operated on in deep hypothermic circulatory arrest, operation without deep hypothermic cardiac arrest, and operation without extracorporeal circulation. Results. Even in healthy neonates, serum S-100 levels were at 10-fold values compared with adults. On admission, S-100 values in the operative groups were similar. During extracorporeal circulation, levels rose to a certain degree. Cyanotic newborns operated on in deep hypothermic cardiac arrest had significantly higher S-100 levels compared with acyanotic newborns also operated on in deep hypothermic cardiac arrest (p < 0.001). Two newborns who experienced seizures postoperatively had the highest absolute S-100 levels. One child with a poor neurologic outcome but no seizures did not have different values when compared with her group. Conclusions. In this study, S-100 seemed to be a possible marker for a certain degree of neurologic deficit after cardiac operation in neonates, especially regarding postoperative seizures. The missing peaks of this protein in one newborn with poor neurologic outcome show that it is not possible to exclude damage to the brain with normal postoperative values. These results suggest that the mechanism of cerebral damage and S-100 release into the blood in neonates with a developing central nervous system and blood brain barrier is not fully understood. (Ann Thorac Surg 2000;69:1515 9) 2000 by The Society of Thoracic Surgeons Cyanotic and acyanotic newborns undergoing cardiac operation under extracorporeal circulation (ECC) with or without deep hypothermic circulatory arrest (DHCA) are at risk of acquiring cerebral complications. Specifically, the presentation of seizures is found to correlate with poor long-term neuropsychological and developmental outcomes [1]. Clinical expression of early cerebral complications is often blunted by sedation and relaxation. Common methods of diagnosis, such as transcranial Doppler ultrasonography, continuous electroencephalography, near-infrared spectroscopy, and magnetic resonance imaging, are used but are timeconsuming and often expensive. A specific biochemical marker for early detection of cerebral complications, especially after cardiac operations in newborns, would be of great value for clinical practice [2]. S-100 protein is a small, dimeric, cytosolic protein with Accepted for publication Nov 26, Address reprint requests to Dr Erb, Division of Thoracic and Cardiovascular Surgery, Department of Surgery, Tübingen University Hospital, Hoppe Seyler Str 3, D Tübingen, Germany, mike.erb@uni-tuebingen.de. a molecular weight of 22 kd. It exists in various forms depending on its chain structure ( or ). The -form predominantly occurs in astroglial and Schwann cells. S-100 is metabolized in the kidney and excreted in urine. The biologic half-life in adults is approximately 113 minutes [3, 4]. There is relatively little literature regarding expression of the S-100 protein after pediatric cardiac operation [5, 6] and its relevance regarding cerebral damage; moreover, there is basically no investigation looking only at neonates and differentiating between cyanotic and acyanotic cardiac malformations. It has been demonstrated that a marked rise of S-100 protein after cardiac operation in adults, and especially after DHCA, correlates with cerebral damage and neurobehavioral outcome [4, 7 10]. We conducted the following study to evaluate the perioperative course of S-100 protein in neonates undergoing repair of congenital heart disease. Points of special interest were whether there would be a difference between cyanotic and acyanotic children undergoing operation with or without DHCA and whether a rise of S-100 would correlate with neurologic outcome by The Society of Thoracic Surgeons /00/$20.00 Published by Elsevier Science Inc PII S (00)

2 1516 ERB ET AL Ann Thorac Surg S-100 PROTEIN AFTER CHD REPAIR 2000;69: Table 1. Study Population Group n Age (days) BSA (kg/m 2 ) Op ECC X-clamp DHCA Healthy TGA b c b DHCA a b c b ECC CoA d a p 0.01 for DHCA group versus all. b p for TGA and DHCA groups versus ECC and CoA groups. c p for TGA and DHCA groups versus ECC group. d p 0.05 for CoA group versus all. Mean values standard deviation. Age age in days at operation. BSA body surface area; CoA coarctation of the aorta; DHCA duration of deep hypothermic cardiac arrest; ECC duration of extracorporeal circulation; Op operation time; TGA transposition of the great arteries; X-clamp aortic clamping time. Material and Methods This study was approved by written consent from the ethics committee of the University of Tübingen. In an ongoing study that started in 1997, we prospectively studied the perioperative course of the S-100 protein in neonates undergoing early repair of congenital heart malformations. In four operative groups, S-100 was measured prospectively in 33 neonates. As a baseline control, serum drawn on the day of birth from 12 healthy full-term newborns was also analyzed for S-100 expression. Blood collection was performed on the day of admission in our pediatric cardiology department, after induction of anesthesia, aortic cross-clamping, and discontinuation of ECC. In addition samples were drawn on reaching the intensive care unit and on the first and third postoperative day. The newborns were divided into four different groups. The first group (transposition of the great arteries; TGA) consisted of 11 cyanotic newborns with simple d-tga with intact ventricular septum and patent ductus arteriosus, operated on in DHCA. Children with TGA presenting with other accompanying malformations, ie, ventricular septal defect, were excluded. In the second group (DHCA), 7 acyanotic newborns operated on in DHCA were examined. This group contained 2 children with truncus arteriosus communis, 2 with an interrupted aortic arch, 1 with malalignment ventricular septal defect and hypoplastic aortic arch, and 2 with total anomalous pulmonary venous drainage. The third group (ECC) consisted of 4 patients who underwent a cardiac operation with ECC but without DHCA. Three of these children had critical aortic stenosis and 1 had pulmonary stenosis and a large patent ductus arteriosus. The fourth group (coarctation of the aorta; CoA) was intended as an operative control with 11 newborns operated on for critical aortic coarctation but without ECC. Children with Down syndrome were excluded. It is known that S-100 has different kinetics in patients with Down syndrome and absolute serum levels are elevated [5, 11]. This is probably because of the third chromosome 21 present in these patients and encoding of S-100 on this chromosome. Anesthesia was induced with midazolam (0.2 to 0.5 g/kg) and alfentanil (40 g/kg); before intubation vecuronium bromide (0.1 g/kg) was applied for relaxation. Continuous infusion of alfentanil (4 to 8 g kg 1 min 1 ) and bolus injections of midazolam (0.2 g/kg) were given to maintain anesthesia. Cardiopulmonary bypass was established in the usual technique, and nonpulsatile perfusion was performed using a membrane oxygenator and arterial filtration. Pump flow and mean arterial blood pressure were adjusted to the body surface and body temperature. Deep hypothermic circulatory arrest was achieved with rectal or bladder temperature at 18 C or esophageal temperature at 14 C. We used ph-stat management and near normal blood glucose levels. The hematocrit for neonates with acyanotic defects was kept more than 30%, for cyanotic children more than 40%. Postoperative neurologic assessment was routinely performed with neurologic examination and cephalic ultrasound; when applicable, electroencephalography (EEG) and computed tomographic scans were performed. Blood samples were collected at the times stated above, then centrifuged for separation of serum. The serum was then frozen at 20 C. S-100 concentration was analyzed using a commercially available monoclonal two-site immunoradioactive assay (Sangtec Medical, Bromma, Sweden). This assay detects the and dimers of the S-100 protein [7]. Lower sensitivity of the assay was 0.2 g/l. Patient-related data are shown in Table 1. We did not observe renal dysfunction or kidney failure in our patients. Operating, ECC, and aortic cross-clamping times were naturally lower in the ECC and CoA groups compared with TGA and DHCA groups ( p 0.001). Duration of DHCA between the TGA and DHCA groups was comparable. The CoA group was older ( p 0.05) and the DHCA group significantly lower in body surface area than the rest ( p 0.001); other marked differences were not observed. Student s t test, analysis of variance, and among groups Fisher s exact test were used when applicable for statistical analysis. Calculation and analysis were performed with SPSS for Windows version 7.5 (SPSS Inc, Chicago, IL). Data for patient characteristics and S-100 values are shown as mean standard deviation.

3 Ann Thorac Surg ERB ET AL 2000;69: S-100 PROTEIN AFTER CHD REPAIR 1517 Fig 1. S-100 baseline values. p for healthy neonates (NB) versus all groups. (TGA cyanotic neonates with d-transposition of the great arteries; DHCA acyanotic neonates operated on in deep hypothermic cardiac arrest; ECC acyanotic neonates operated on with extracorporeal circulation but without DHCA; CoA neonates with aortic coarctation.) Results Baseline Values In the four groups of children undergoing operation, the S-100 levels collected on admission showed levels of approximately 1 g/l (Fig 1) that did not differ significantly among groups. The control group of healthy newborns, however, had surprisingly high S-100 levels in comparison with the children who were operated on (1.97 g/l, p versus all groups). Normal value in adults is less than 0.5 g/l. Perioperative Values Among the children undergoing operation, the TGA group showed the highest overall levels of the S-100 protein (Fig 2) after ECC and DHCA. There was a highly significant difference between the TGA group and all other groups on aortic cross-clamping ( p ), after aortic cross-clamping ( p 0.02), and after bypass ( p 0.001). Postoperatively in the intensive care unit, the TGA group still had the highest levels but were not statistically different from other groups ( p 0.06). Second highest overall values were seen in the DHCA group, these being the acyanotic neonates also operated on in DHCA. In the ECC and CoA groups, S-100 only showed a slight perioperative reaction. Among these three groups (DHCA, ECC, and CoA), no statistical difference was observed. The highest absolute levels were seen in 4 children with evidence of postoperative neurologic dysfunction. A marked rise of S-100 without evidence of neurologic dysfunction was not observed. Pathology One of the 33 neonates undergoing operation died (DHCA group). In the ECC and CoA groups, neurologic complications were not observed. Except for the cases stated below this also applies for the TGA and DHCA groups. Fig 2. Perioperative S-100 values. *p 0.04; x, p for cyanotic neonates with d-transposition of the great arteries (TGA) versus all other groups. (DHCA acyanotic neonates operated on in deep hypothermic cardiac arrest; ECC acyanotic neonates operated on with extracorporeal circulation but without DHCA; CoA neonates with aortic coarctation; Pre-op preoperative; x-clamp aortic cross-clamping; off-byp off bypass; post-op postoperative; 1. pod first postoperative day; 3. pod third postoperative day.) Acyanotic Heart Defects Two newborns in the DHCA group with evidence of postoperative cerebral dysfunction also had a marked rise of the S-100 protein. Both children underwent correction of truncus arteriosus communis (Fig 3). In child A there was chromosome translocation 7/22 present (7/22) (p22, a11). After the operation she experienced a gener- Fig 3. Two neonates with acyanotic congenital heart disease operated on in deep hypothermic cardiac arrest (DHCA) with postoperative neurologic deficit compared with their group. Child A had correction of truncus arteriosus and generalized seizure postoperatively. Child B had correction of truncus arteriosus, evidence of cerebral damage in cranial ultrasound, and generalized seizure on the seventh postoperative day. (pre-op preoperative; x-clamp aortic cross-clamping; off-byp off bypass; post-op postoperative; 1. pod first postoperative day; 3. pod third postoperative day; 7. pod seventh postoperative day.)

4 1518 ERB ET AL Ann Thorac Surg S-100 PROTEIN AFTER CHD REPAIR 2000;69: Fig 4. Two neonates after correction of d-transposition of the great arteries (TGA) and neurologic deficit compared with their group. Child C showed severe electroencephalographic changes and had multiple cerebral seizures after the operation. Child D had preoperative and postoperative evidence of discrete cerebral damage. (pre-op preoperative; x-clamp aortic cross-clamping; off-byp off bypass; post-op postoperative; 1. pod first postoperative day; 3. pod third postoperative day.) alized seizure and showed a cerebral ventricular widening. The child died on the first postoperative day. Child B showed unremarkable perioperative S-100 levels, but there was evidence of cerebral damage on cranial ultrasound. Only after a seizure on the seventh postoperative day was there a marked rise of S-100. Neurologic follow-up shows a serious retardation. Cyanotic Heart Defects In the TGA group, there were 2 children with postoperative evidence of cerebral damage; 1 had a marked rise of the S-100 protein. The other child with evidence of preoperative and postoperative cerebral dysfunction showed no particularly marked elevation of S-100 in comparison to the rest of the group (Fig 4). Child C showed severe EEG changes and had multiple cerebral seizures after the operation. Cephalic ultrasound was normal, and S-100 levels were very high. Postoperative neurologic outcome under anticonvulsive medication was adequate (normal EEG and computed tomographic scan with unremarkable pediatric neurologic examination). Child D had preoperative and postoperative evidence of cerebral damage in the form of periventricular hyperdensity (before) and cerebral bleeding (subependymal) with delayed awakening (after). Seizures did not occur. The S-100 levels did not exceed those of the rest of the group. Late postoperative pediatric neurologic examination and EEG showed no abnormalities. Comment The three main observations of this study are as follows: S-100 levels in neonates, regardless of state of health, are higher than in healthy adults. Moreover, children on the day of birth have significantly higher serum S-100 levels than older neonates, which implies a time-related course of this protein after birth, even under physiologic conditions. This study, solely performed on neonates undergoing cardiac operation, shows that there is a perioperative reaction of the S-100 protein after ECC. Significant rises compared with preoperative values could only be seen in newborns operated on in DHCA. The TGA group showed the highest overall values even though DHCA times and all other criteria were comparable to those of the acyanotic children also operated on in DHCA. Looking at the children with high levels or evidence of cerebral damage, a marked rise of the S-100 protein primarily seems to be the result of cerebral seizures, because the children who did not experience seizures even in the presence of cerebral damage had inconspicuous S-100 levels. The state of brain development may be a possible explanation for our findings. Development of the human central nervous system is not complete until about 1 year of age [12]. It is a complicated process involving neuronal development, maturation, and also cell death of excess neurons. Also, the blood brain barrier (BBB) is not completely functional until about 6 months after birth, which means proteins (eg, S-100) that normally, ie, in later life, do not pass the BBB can be found in the blood. Moreover, as the S-100 protein probably also plays an important role in neuronal cell proliferation and regulation, levels in the cerebrospinal fluid are bound to be physiologically high in neonates. This would explain the relatively high levels on the day of birth and among our operated on children when compared with baseline values of adults. Because oxygen saturation also seems to be an important factor for the normal development of the BBB, the higher S-100 values seen postoperatively in the TGA group might be explained with a lack of oxygen saturation caused by the cyanotic heart defect of these children, and, therefore, a less developed BBB. In our study, intracerebral blood flow and oxygen saturation were not measured and can therefore not be taken into account. We did have 4 children with neurologic problems, 2 experiencing seizures early, 1 with late seizures, and 1 with no seizures but a definite neurologic problem. Except for child A, who had severe cardiac failure perioperatively, we have no clear explanation for the mechanism of these findings. All children with seizures showed very high S-100 levels. We think this is because of a passing injury to the BBB, provoking a higher permeability for the S-100 protein. Why two of the children with a cerebral problem did not show high S-100 levels is unclear. In our opinion this makes the S-100 protein relatively unreliable as a general marker for cerebral damage perioperatively. One of these two children had a late seizure (seventh postoperative day) and only subsequently a marked rise of the S-100. Seizures consistently provoked a higher spike in serum S-100 levels, maybe because of increased permeability of the BBB. One can speculate about whether cell death or damage are also involved. Theoretically, an intact BBB could keep serum levels at normal values even if there is hypoxic or any other cerebral damage with higher intracerebral S-100 levels. Even if low S-100 levels do not rule out

5 Ann Thorac Surg ERB ET AL 2000;69: S-100 PROTEIN AFTER CHD REPAIR 1519 cerebral problems, high levels were always associated with seizures and EEG changes. This makes the postoperative S-100 detection in the serum of operated on children a potential screening method in the intensive care unit. Postoperative seizures are often overlooked, and continuous EEG is a time-consuming and cumbersome examination that is not performed as a postoperative routine. Further studies are warranted. Higher patient numbers could give more statistical power. A future goal in our study group is to establish a neuropsychological status of the children who underwent operation when they are older and then correlate it to the measured S-100 values. Kits for detection of the S-100 protein were kindly provided by Sangtec Medical, Bromma, Sweden. References 1. Bellinger DC, Jonas RA, Rappaport LA, et al. Developmental and neurologic status of children after heart surgery with hypothermic circulatory arrest or low-flow cardiopulmonary bypass. N Engl J Med 1995;332: Jonas RA. Hypothermia, circulatory arrest, and the pediatric brain. J Cardiothorac Vasc Anesth 1996;10: Usui A, Kato K, Abe T. S-100 protein in blood and urine during open heart surgery. Clin Chem 1989;35: Johnsson P. Markers of cerebral ischemia after cardiac surgery. J Cardiothorac Vasc Anesth 1996;10: Lindberg L, Ölsson AK, Anderson K, Joegi P. Serum S-100 protein levels after pediatric cardiac operations: a possible new marker for postperfusion cerebral injury. J Thorac Cardiovasc Surg 1998;116: Abdul-Khaliq H, Blasig IE, Baur MO, Hohlfeld M, Alex- Meskhishvilli-V, Lange PE. Release of the cerebral protein S-100 into the blood after reperfusion during cardiac operations in infants: is there a relation to oxygen radicalinduced lipid peroxidation? J Thorac Cardiovasc Surg 1999; 117: Westaby S, Johnsson P, Parry AJ. Serum S-100 protein: a potential marker for cerebral events during cardiopulmonary bypass. Ann Thorac Surg 1996;61: Kumar P, Dhital K, Hossein-Nia M, Patel S, Holt D, Treasure T. S-100 protein release in a range of cardiothoracic surgical procedures. J Thorac Cardiovasc Surg 1997;113: Baeckstroem M, Joennsen H, Bergh C, Blomquist S, Alling C, Johnnson P. Early release of S-100 after cardiac surgery is associated with neuropsychological outcome [Abstract]. Ann Thorac Surg 1998;66:S Roach GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes after coronary bypass surgery. N Engl J Med 1996;335: Fano G, Biocca S, Fulle S, Mariggio MA, Belia S, Callisano P. The S-100: a protein family in search of a function. Prog Neurobiol 1995;46: Rodier PM. Developing brain as a target of toxicity. Environ Health Perspect 1995;103(Suppl 6):73 6.

SURGERY FOR CONGENITAL HEART DISEASE

SURGERY FOR CONGENITAL HEART DISEASE SERUM S-100 PROTEIN LEVELS AFTER PEDIATRIC CARDIAC OPERATIONS: A POSSIBLE NEW MARKER FOR POSTPERFUSION CEREBRAL INJURY L. Lindberg, MD, PhD a A-K. Olsson, MD, PhD a