Spinal Fluid Clotting Activity: a New Method of Evaluating Neonatal Brain Damage

Pediatr. Res. 16: 412-415 (1982) Spinal Fluid Clotting Activity: a New Method of Evaluating Neonatal Brain Damage BERNARD DALENS,""' MARIE-JOSEPHE BEZOU, MAURICE COULET, JEAN-PIERRE HABERER, A N D G U Y VANNEUVILLE Department of Anesthesiology [B. D., J.-P. H., G. V.] and Department of Hematology [M.-J. B, M.C.], Hotel-Dieu, Clermont-Ferrand. France Introduction The relationships between cerebrospinal fluid disturbances and neonatal cerebral injury have been emphasized by several studies (1, 6, 8, 9, 10, 11, 12, 13), including ours (2, 3, 4, 5). In particular, we have previously reported close correlations of enzyme levels in and with brain damage as by at year age (4); were suggested to delineate high-risk from low-risk infants. A preliminary rep0rt Graeber and Stuart (7) described a procoagulant activity of CSF in relation to cerebral injury. This examine the 'lotting us activity of neonates according to our methodology of evaluating brain damage (4). After several attempted assays, two kinds of tests were retained: (1) activated ~ a r t i athromboplastin l time, which explored the early stages of coagulation, and (2) thrombelastography. which realized a complete analysis of coagulation, allowing for subsequent reexaminations as well. MATERIALS AND METHODS This study was realized with neonates admitted to our neonatology center and intensive care unit. Eighty-six infants, whose major final diagnosis is shown on Table 1, were examined. Apgar scores, daily physical examinations and electroencephalographic patterns were systematically recorded. Apgar scores at 1 min were lower than 3 in 24 cases, greater than 6 in 42 cases and intermediate in 20 cases. Neurologic examinations were quite normal in 28 cases and revealed deep abnormalities in 21 cases (prolonged coma, complete lack of primitive reflexes, seizures); they were variously (but no longer than the seventh day of life) disturbed in the 37 remaining cases. Electroencephalograms revealed paroxysmal patterns in 15 cases, normal tracings in 41 cases and nonparoxysmal abnormalities in 30 cases. Lumbar punctures for diagnostic evaluation of cerebrospinal fluid (CSF) were done from birth to the 25th day of life. CSF, 0.8-1.2 ml, was collected in plastic tubes and assayed within 20 min (in every case). Protein, cell counts, cultures, enzyme determinations [lactate dehydrogenase E.C. 1.1.2.7.; hydroxybutyrate dehydrogenase; creatine kinase EC. 2.7.3.2.; glutamic oxaloacetic transaminase E.C. 2.6.1.1.1 were performed using standard techniques. For the evaluation of CSF clotting activity, two mixtures were realized: (1) 0.3 ml amount of normal pooled and platelet poor plasma (NPPP) were mixed to a blank (0.3 ml amount of Michaelis buffer): control mixture, and (2) 0.3 ml amount of NPPP mixed to the same volume of CSF: assay mixture. Measurement of activated partial thromboplastin time (APTT). Control as well as assay mixtures in 0.1 ml amounts, were incubated at 37OC for 3 min, and then respectively mixed to 0.1 ml of cephalin (1/10° dilution) and 0.1 ml of a suspension of kaolin (C.K. test Stago). Then 0.1 ml amount of calcium chloride (CaC12:0.025 mole/liter) was added and coagulation time measured at 37°C. Thrombelastography, Control and assay mixtures, in 0.3 ml amounts, were distributed into different cups and placed for 3 min in a water bath, N~~~ 0.06 ml of calcium chloride ( C ~ C ~ ~ : O , O ~ mole/liter) was added, respectively, in the two cups and recording procedure of thrombelastography performed for 60 min. Two measurements (14) were made from the tracings (Fig. 1): ( I ) the reaction time r, which elapsed between recalcification of the plasma and the point at which the straight line divided into two lines. This line was measured (mm) from the start to this point (film speed:2 mm/min) and (2) the maximum amplitude (ma), which was the widest point of the two lines (mm). Three lots of samples were delineated in relation to the time from birth to day 3 (Dlelapsed since birth: (1) samples ~ 3 1 (2) , those collected between day 4 and day 7 (D4-D7), and (3) those collected later than day 7 ( D 8 - ~ 2 5 ) . All assays were performed in duplicate (and a difference no greater than 3% had been observed). The results (means and S.D.) were expressed in % of assays versus control parameters (APTT, r and ma) and were analyzed with Student's t test. Differences were considered significant when P < 0.01. RESULTS Clinical assessment of infants. The eighty-six neonates were divided into three classes in relation to brain damage according to clinical (Apgar scores, physical examinations and EEG tracings) and biologic [enzyme determinations following our previously reported (4) methodology] patterns: (1) low-risk class (21 cases): infants with normal clinical patterns and low enzyme levels, (2) high-risk class (23 cases): infants demonstrating severe clinical abnormalities (at least two of the three clinical patterns severely disturbed) with high enzyme levels, and (3) mild-risk class (42 cases): infants demonstrating either a discrepancy between clinical and biochemical patterns or revealing mild abnormalities of clinical as well as biologic evaluations. CSF clotting activity (Table 2). The subclasses delineated by Apgar scores (<3, >6, intermediate), physical examinations (normal, deeply abnormal, slightly disturbed) and EEG tracings (normal, paroxysmal, variously disturbed) did not reveal significant differences of clotting parameters. Conversely, the three classes of infants delineated in relation to brain damage (low, high and mild-risk classes) demonstrated significant differences. APTT (Fig. 2). All mean values were below 85%, i.e., all CSF samples demonstrated a positive clotting activity versus a blank, even later than day 8. Main features of the measurements were: (1) quite constant values in every sample of low-risk infants, (2) significantly lower values in (Dl-D3) samples of high-risk infants in comparison to low-risk ones, and (3) lack of significant differences in (D8-D25) samples of the three classes of infants. 412 413 NEONATAL BRAIN DAMAGE The low number of (D3-D7) samples of low-risk infants did not authorize to conclude that the smaller APTT values of highrisk infants were significantly different. Reaction time r (Fig. 3). All mean r values were below 75%, i.e., they all revealed a procoagulant activity of CSF samples. Lowrisk and mild-risk classes did not show significant differences at any date of collection. High-risk infants demonstrated significantly smaller r values in (Dl-D3) samples, and significantly greater ones (i.e., less procoagulant activity) in (D4-D7) as well as (D8D25) samples; in this group, the differences observed according to t h e time elapsed since birth were significant. Maximum amplitude ma (Fig. 4). The three classes of infants demonstrated significant differences in the three lots of samples. In high-risk infants, the low (D I-D3) ma values (negative clotting activity) w (...truncated)