| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Fédération de Biochimie,
2
Département d’Anésthésie-Réanimation, and
3
Service de Chirurgie Cardiaque, Hôpital Pitié-Salpêtrière, 47-83 Boulevard de l’Hôpital, F75651 Paris Cedex 13, France
a author for
correspondence: fax 33-1-42-17-76-16, e-mail jean-louis.beaudeux{at}psl.ap-hop-paris.fr
Cerebral injury is an important complication after
cardiovascular surgery. The neurological events usually are diagnosed
using rather crude psychometric tests and clinical observations
(1). Biological markers have been proposed to detect
cerebral damage during cardiac surgery, e.g., the creatine kinase BB
isoform or neuron-specific enolase (NSE). Because of its
neurospecificity, the 
isoform of enolase is of particular
interest. Measurements in biological fluids use -subunit-specific
immunoassays. Because of the abundance of the 
isoform of
enolase in erythrocytes, systemic NSE values may be falsely increased
by the frequent hemolysis that occurs during surgery
(2)(3).
S-100ß protein (S-100) is a member of a family of calcium-binding proteins present primarily in nervous tissue, where it is concentrated mainly in glial cells. Although the role of this protein in brain function and disease has not been elucidated conclusively, it has been ascertained that the appearance of this protein constituent of neural cells in biological fluids is a reliable indicator of active cell damage in the nervous system in different pathological conditions. Measurement of S-100 in the blood recently has been used successfully to monitor cerebral damage after cardiac surgery (4)(5)(6). S-100 is absent from red blood cells (RBCs), and plasma concentrations would not be influenced by hemolysis.
We examined the reliability of measurements of plasma S-100 in either in vivo or ex vivo hemolyzed blood samples. Blood samples were taken from 24 patients (19 males and 5 females; age, 63.4 ± 9.8 years, mean ± SD), who were undergoing coronary artery bypass grafting (CABG) with cardiopulmonary bypass (CPB). All patients gave informed consent for inclusion in the study. None presented a neurological complication in the pre- or postoperative period. Blood samples were taken at seven predetermined times: after induction of anesthesia, at the end of CPB, and 1, 2, 4, 6, and 12 h thereafter [time points control, t0, t1, t2, t4, t6, and t12, respectively]. Blood was collected in heparin-containing Vacutainer® tubes and centrifuged within 2 h (1000g for 10 min at 4 °C); the plasma samples were then frozen at -20 °C until use (within 2 months).
S-100 and NSE concentrations were determined with immunoluminometric sandwich assays on a LIA-mat 300 analyzer (Byk-Sangtec France Laboratories) with the manufacturer’s reagents. Briefly, 100-µL samples were incubated in tubes coated with anti-S-100ß chain for 60 min. After the tubes were washed, aminobutylethylisoluminol (ABEI)-labeled anti-S-100ß was added, and the tubes were incubated for 120 min. After a final washing step, the luminescence of ABEI oxidized in the presence of deuteroferriheme and hydrogen peroxide was measured. The NSE measurement used a one-step procedure: 25 µL of sample and 300 µL of ABEI-labeled anti-NSE were incubated in tubes coated with anti-NSE for 60 min. ABEI luminescence was then induced as described above. The amounts of S-100 and NSE in the samples were calculated using calibrations curves prepared with calibrators containing known concentrations of the proteins. Plasma hemoglobin was determined by derivative spectrophotometry as described previously (7).
Patients were divided into two groups according to the occurrence of
hemolysis during and after CABG: group 1 (n = 12) included
patients who presented with moderate pre- and postoperative hemolysis
(none or one of the seven blood samples with a plasma hemoglobin
concentration >350 mg/L); group 2 (n = 12) included patients who
presented with substantial hemolysis (three or more samples >350
mg/L). Hemoglobin plasma concentrations, which did not differ
significantly between the two groups after the induction of anesthesia,
were significantly higher in group 2 than in group 1 at
t0 (509 ± 43 vs 195 ± 20
mg/L; P <0.001) and up to 6 h thereafter (Table 1
). There were no significant differences between the two groups
of patients in either the epidemiological and clinical characteristics
or in the surgical conditions.
|
The plasma concentrations of S-100 and NSE are reported in Table 1
. In
group 1, both S-100 and NSE reached a maximum as soon as the end of CPB
(t0), and then decreased regularly
until 12 h; at this time, NSE concentrations were within the
reference interval, whereas plasma S-100 concentrations
remained high (0.35 ± 0.05 vs 0.10 ± 0.03 µg/L before
CPB). This evolution of both markers agrees with previous studies
performed on patients undergoing CABG with CPB who did not present with
neurological complications, and it probably reflects transient and
reversible brain tissue damage (5)(8).
Patients in group 2 exhibited similar kinetics for plasma S-100: maximal values were observed at the end of CPB (3.12 ± 0.49 vs 2.87 ± 0.50 µg/L in group 1; not significant), and concentrations from t1 to t12 did not differ significantly from those observed in group 1. In contrast, as soon as t0 and until t6, NSE concentrations were significantly higher in group 2 than in group 1; the maximal value was observed at t0. In both groups, we found positive correlations between S-100 and NSE (P <0.001), plasma hemoglobin and NSE (P <0.001), and to a lesser extent, plasma hemoglobin and S-100 (P <0.01). This suggests that the increases in the three markers were strongly related to the CPB. However, the large increase in NSE in the immediate postoperative period probably was attributable to the conjunction of both a specific release of this marker by cerebral tissue (which also led to the increase in S-100 plasma concentrations) and the release by blood cells because of CPB-induced hemolysis.
To confirm this hypothesis, we next studied the effect of in vitro
hemolysis on both S-100 and NSE plasma measurements by adding lysed
RBCs to normal plasma. After centrifugation and removal of plasma and
white cells by aspiration, the RBCs were washed three times in saline
and lysed by freezing. Red cell debris was removed by centrifugation,
and the supernatant was added to plasma to obtain hemoglobin
concentrations of 100–800 mg/L (final dilution of the plasma, 1:2).
The initial hemoglobin concentration of the three plasmas used in this
experiment ranged from 40 to 85 mg/L. As expected, we observed a
significant increase in NSE concentrations, reaching more than fourfold
the initial concentration at the maximal loading point (Fig. 1
). Moreover, we found a significant relationship between the
increase in NSE concentration and the hemoglobin (Hb) concentration, at
least in our experimental conditions: [NSE (µg/L)] = 1.84 x
10-2 [Hb (mg/L)]. Such a relationship between
the NSE and Hb concentrations has been described previously
(9)(10) and might allow correction of NSE values
to determine the specific amount of NSE released by neural tissue
damage, by subtraction of the amount of NSE produced by RBC lysis
(8). In contrast, plasma S-100 concentrations were not
affected by the addition of RBC lysate, even at a final Hb
concentration of 800 mg/L. We can thus affirm that hemolysis neither
modified plasma S-100 concentrations nor interfered with the S-100
immunoluminometric assay. This result agrees with the study of Gao et
al. (9), who did not observe any interference by in vitro
hemolysis on S-100 concentrations when measured by an immunoradiometric
assay.
|
In conclusion, this report indicates that hemolysis, occurring in vivo or mimicked in vitro by the addition of a RBC lysate, by itself falsely increases NSE plasma concentrations but does not interfere with S-100 plasma determinations. We conclude that S-100 protein may be a more accurate biological marker of cerebral injury than NSE in the diagnosis of neurological complications during and after cardiovascular surgery.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J. Reisinger, K. Hollinger, W. Lang, C. Steiner, T. Winter, E. Zeindlhofer, M. Mori, A. Schiller, A. Lindorfer, K. Wiesinger, et al. Prediction of neurological outcome after cardiopulmonary resuscitation by serial determination of serum neuron-specific enolase Eur. Heart J., January 1, 2007; 28(1): 52 - 58. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. F. Laterza, V. R. Modur, D. L. Crimmins, J. V. Olander, Y. Landt, J.-M. Lee, and J. H. Ladenson Identification of Novel Brain Biomarkers Clin. Chem., September 1, 2006; 52(9): 1713 - 1721. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, S-100; 
, NSE. Values are means ± SE (bars)
of three separate experiments.