The Journal of International Medical Research 2010; 38: 220 226 Increased Serum Malondialdehyde Level in Neonates with Hypoxic Ischaemic Encephalopathy: Prediction of Disease Severity E KIRIMI 1, E PEKER 1, O TUNCER 1, H YAPICIOGLU 2, N NARLI 2 AND M SATAR 2 1 Neonatology Unit, University of Yuzuncu Yil, Van, Turkey; and 2 Neonatology Unit, University of Cukurova, Adana, Turkey Increased serum level of malondialdehyde (smda) in neonates with hypoxic ischaemic encephalopathy (HIE) was evaluated as a possible criterion for determining HIE severity. Mean body weight and gestational age in a healthy control group of neonates (n = 63) and in neonates with HIE (n = 69) were statistically similar. Apgar scores at 1 and 5 min for the HIE group were significantly lower than for the control group. The mean smda level for the HIE group was significantly higher than the control group. Within the HIE group, the smda level for neonates with Sarnat s grade II and III was significantly higher than for those with Sarnat s grade I. There was a significant correlation between Sarnat s grading and the smda level. The smda level was significantly higher for neonates who died (n = 20) compared with those who survived (n = 49). In conclusion, the smda level was highest in neonates with HIE and correlated with HIE severity. The smda concentration could, therefore, be used as a criterion for predicting disease severity. KEY WORDS: MALONDIALDEHYDE; HYPOXIC ISCHAEMIC ENCEPHALOPATHY; NEWBORN; NEONATE; SARNAT GRADING; APGAR SCORE Introduction Perinatal asphyxia and its associated condition, hypoxic ischaemic encephalopathy (HIE), are devastating neonatal conditions. Although the condition can be slight in some infants, it becomes severe in most affected cases, leading to future disability and impaired quality of life. 1,2 Studies have resulted in an improved understanding of this condition and promising research has been conducted to determine the pathogenesis of the brain and other damaged organs. 1 3 Oxygen and free oxygen radicals play a crucial role in the development of cerebral tissue damage and malondialdehyde (MDA) is one of the key end-products of lipid peroxidation induced by reactive oxygen species. 1 5 Serum MDA (smda) concentration may change during HIE and could, therefore, be used as a predictor for determining HIE diagnosis, treatment and prognosis. 3 5 The aim of the present study was to determine smda concentration in neonates 220
who were found to have asphyxia and HIE during the perinatal period, in order to evaluate whether smda concentration could be used as an indicator for predicting HIE severity. Patients and methods PATIENTS Neonates with HIE (HIE group) from two centres, the Neonatal Intensive Care Unit, Medical Faculty of Cukurova University, Adana, Turkey, and the Neonatal Intensive Care Unit of the Medical Faculty of Yuzuncu Yil University, Van, Turkey, were included in the study, with follow-up conducted between January 1999 and July 2009. Over the same period, healthy neonates from the former centre were used as a control group. The study was approved by the Human Ethics Committee of the Medical Faculty of the Cukurova University. For the HIE group, written informed consent was obtained from the families of the neonates. For the control group, samples were collected during routine screening tests after obtaining family written informed consent. Neonates with hyperbilirubinaemia were excluded from the study, since this condition may lead to oxidative stress, thereby increasing MDA levels. 6 ASSESSMENTS For the HIE and control groups, the physical and clinical data collected included: sex, weight, delivery route, gestational age (calculated according to the last menstrual cycle and/or the New Ballard Score 7 ), 1- and 5-min Apgar scores, smda concentrations, and time of smda sample collection. In addition, for the HIE group, clinical grading according to Sarnat and Sarnat, 8 and patient outcomes were recorded. Two primary criteria were used to define asphyxia in neonates: 9 (i) a 5-min Apgar score < 6 points and (ii) umbilical cord blood with a ph < 7.2. Patients had to meet both criteria. Venous blood samples were collected into tubes containing ethylenediaminetetraacetic acid within 12 19 h post-partum from neonates in both the HIE and control groups. The samples were placed in a glass tube and centrifuged at 1500 g for 15 min within 15 min of collection. Serum samples were then either analysed within 1 h or stored at 20 C in a deep-freeze and analysed within 2 months of collection. Samples that exhibited haemolysis were excluded from the study. The smda concentration of the samples was measured using the thiobarbituric acid test as described by Wong et al; 10 MDA in aerobic conditions when incubated with thiobarbituric acid (Sigma-Aldrich, St Louis, MO, USA) at 95 C and ph 3.4 resulted in a pink-stained complex and its optical density was measured at a wavelength of 532 nm using a UV 1800 spectrophotometer (Shimadzu Corp., Kyoto Japan,). The resultant smda concentration was calculated as described previously 11 and expressed in nmol/ml. STATISTICAL ANALYSIS The results were expressed as mean ± SD. Statistical evaluation of the data was carried out using the SPSS statistical package, version 12.0 (SPSS Inc., Chicago, IL, USA) for Windows. A Student s t-test, the χ 2 -test and Pearson s correlation analysis were used to analyse the data. A P-value < 0.05 was considered to be statistically significant. Results Physical and clinical data for the HIE and control groups are shown in Table 1. The HIE and control groups consisted of 69 and 63 neonates, respectively. There were no statistically significant differences between the 221
TABLE 1: Physical and clinical data for neonates with hypoxic ischaemic encephalopathy (HIE) in comparison with a healthy control group of neonates HIE group Control group Statistical n = 69 n = 63 significance Gestational age (weeks), mean ± SD (range) 39.6 ± 1.7 39.0 ± 1.3 NS a (34 42) (34 42) Weight (g), mean ± SD (range) 3123 ± 633 3174 ± 242 NS a (1590 4350) (2800 3800) Sex, n Male 32 28 NS b Female 37 35 Route of delivery, n Vaginal delivery 37 39 NS b Caesarean section 32 24 Apgar score, mean ± SD (range) 1 min 2.0 ± 1.0 7.4 ± 1.2 P = 0.0001 a (0 4) (6 9) 5 min 4.1 ± 1.2 9.1 ± 0.5 P = 0.0001 a (2 8) (8 10) Blood sample collection time (h), 14.5 ± 1.5 14.9 ± 1.0 NS a mean ± SD (range) (12 19) (13 18) Serum malondialdehyde (nmol/ml), 6.17 ± 2.42 2.18 ± 1.17 P = 0.0001 a mean ± SD (range) (2.36 13.52) (0.68 4.2) NS, not statistically significant (P > 0.05). a Student s t-test; b χ 2 -test. two groups in terms of mean gestational age, weight, sex, delivery route or mean time of smda blood sample collection. The 1- and 5- min Apgar scores for the HIE group were significantly lower than the control group (both P < 0.0001). The smda concentrations for the HIE group were significantly higher compared with the control group (P < 0.0001). Cases with HIE were split into three categories according to Sarnat s grading; means for the physical and clinical data for each grade are shown in Table 2. Gestational age, weight, sex, route of delivery, 1-min Apgar scores and time of smda blood sample collection for the three groups were found to be statistically similar. The 5-min Apgar score was significantly lower for neonates in the grade II and III groups compared with the score for neonates in the grade I group (both P < 0.05). The smda concentration for neonates in the grade II and III groups was significantly higher than for neonates in the grade I group (both P < 0.05). Although a higher level of smda concentration was found in grade III neonates compared with grade II neonates this difference was not statistically significant. A significant correlation was found between Sarnat s grading and the smda level (r = 0.634, P = 0.01; Fig. 1). The HIE cases were also divided into those who survived and were discharged from hospital (n = 49) and those who died (n = 20) and means for the physical and clinical data in each category are shown in Table 3. Gestational age, weight, sex, route of delivery, 1-min Apgar scores and time of smda blood sample collection for the two 222
TABLE 2: Physical and clinical data for neonates with hypoxic ischaemic encephalopathy graded according to Sarnat s grading 8 Grade I Grade II Grade III n = 16 n = 31 n = 22 Gestational age (weeks), mean ± SD 39.1 ± 2.8 39.8 ± 1.0 39.9 ± 1.4 Weight (g), mean ± SD 3114 ± 899 3225 ± 615 2985 ± 379 Sex, n Male 7 18 7 Female 9 13 15 Route of delivery, n Vaginal delivery 9 19 9 Caesarian section 7 12 13 Apgar score, mean ± SD 1 min 2.3 ± 0.4 2.1 ± 0.7 1.8 ± 1.5 5 min 5.1 ± 1.2 4.1 ± 1.0 a 3.3 ± 0.7 a Blood sample collection time (h), mean ± SD 14.4 ± 1.6 14.7 ± 1.5 14.4 ± 1.4 Serum malondialdehyde (nmol/ml), mean ± SD 3.46 ± 0.95 6.47 ± 1.21 a 7.72 ± 2.82 a a P < 0.05 versus grade I (Student s t-test). All other comparison were not statistically significant (P > 0.05). 14 Serum malondialdehyde (nmol/ml) 12 10 8 6 4 2 0 Grade I (n = 16) Grade II (n = 31) Grade III (n = 22) FIGURE 1: Serum malondialdehyde concentrations for neonates with hypoxic ischaemic encephalopathy, showing a statistically significant correlation between serum malondialdehyde concentrations and Sarnat s grading: 8 r = 0.634, P = 0.01 (mean values shown as bold horizontal lines; minimum and maximum outliers shown as error bars; one case extreme outlier with Sarnat s grade III hypoxic ischaemic encephalopathy shown with a circle) 223
TABLE 3: Physical and clinical data for neonates with hypoxic ischaemic encephalopathy categorized according to those that survived and were discharged from hospital and those that died Discharged Died Statistical n = 49 n = 20 significance Gestational age (weeks), mean ± SD 39.5 ± 1.8 40.0 ± 1.5 NS a Weight (g), mean ± SD 3145 ± 686 3068 ± 489 NS a Sex, n Male 25 7 Female 24 13 NS b Route of delivery, n Vaginal delivery 27 10 Caesarian section 22 10 NS b Apgar score, mean ± SD 1 min 1.9 ± 0.8 2.1 ± 1.3 NS a 5 min 4.3 ± 1.1 3.7 ± 1.3 P = 0.046 a Blood sample collection time (h), mean ± SD 14.7 ± 1.5 14.5 ± 1.5 NS a Serum malondialdehyde (nmol/ml), mean ± SD 5.67 ± 1.83 7.41 ± 3.19 P = 0.031 a NS, not statistically significant (P > 0.05). a Student s t-test; b χ 2 -test. groups were found to be statistically similar. The 5-min Apgar score was significantly lower for neonates who died compared with those who survived (P = 0.046), whereas the smda concentration was significantly higher in the group that died versus those who survived (P = 0.031). Discussion Research has focused on the evaluation of free radical activity in HIE, and HIE disease severity and prognosis may be best evaluated once this free radical activity has been evaluated. Free oxygen radicals interact with membrane lipids resulting in their degradation and the release of lipid peroxidation end-products, the one known to be most common being MDA. 12 14 By measuring the concentration of MDA, researchers have evaluated oxidative stress and the extent to which free oxygen radicals have resulted in cell damage. 4,5,12 15 To date, it is not completely understood whether MDA concentration is directly correlated with HIE disease severity. One of the most interesting experimental studies was reported by Horakova et al., 16 who reported that the MDA level did not increase in rats that had been given stobadine and then killed just after ischaemia, whereas the level was very high in rats subjected to reperfusion following ischaemia. In a diffuse head injury animal model, Hsiang et al. 17 studied the time course of the increase in MDA level in various regions of the brain following cerebral damage. In the frontal, parietal and brain stem regions, the MDA level peaked 4 h following trauma. The MDA level was also found to differ between various cerebral regions. With the exception of a number of experimental studies, 16 to the authors knowledge, no studies have been conducted to explain the severity of cerebral damage related to HIE as manifested by changes in 224
the MDA level in neonates. Dorman et al. 18 demonstrated that in Sprague-Dawley rats with bromethalin-induced brain lipid peroxidation and cerebral oedema, the MDA level increased significantly and was correlated with the severity of clinical findings. Histopathologically, one possible interpretation for this could be that the greater the number of cell membranes damaged by lipid peroxidation in the brain, the higher the observed level of MDA would be. In the present study, the correlation between the severity of clinical findings and the smda level was investigated in neonates with HIE. When the clinical findings were classified according to Sarnat s grading, the smda level for neonates with Sarnat s grade I was significantly lower than for neonates with grades II and III. Although there was no significant difference between the smda level for neonates with grades II and III, the level found in grade III neonates was higher than found in grade II neonates. The results showed a significant correlation between the severity of cerebral damage (as measured by Sarnat s grading) and the smda level, and this approach could be useful to evaluate the prognosis of neonates with HIE. Apart from clinical grading, the neonates with HIE were also divided into two groups according to whether they died or whether they survived and were discharged from hospital. The smda level for neonates who died was significantly higher than for those who survived. Thus, the smda level could potentially be used as a criterion for evaluating the prognosis and mortality of neonates with HIE. Of the neonates included in the present study, those who were discharged are still being followed up. The presence and the severity of neurological deficits in these neonates will be evaluated, and the correlation between neurological deficits and smda levels measured during the neonatal period will be assessed. The prognosis of the neonates with HIE will then be prospectively predicted from the beginning of the neonatal period. Conflicts of interest The authors had no conflicts of interest to declare in relation to this article. Received for publication 7 August 2009 Accepted subject to revision 12 August 2009 Revised accepted 27 November 2009 Copyright 2010 Field House Publishing LLP References 1 Evans DJ, Levene MI: Hypoxic ischaemic injury. In: Textbook of Neonatology, 3rd edn (Rennie JM, Roberton NRC, eds). Edinburgh: Churchill Livingstone, 1999; pp 1235 1251. 2 Hill A, Volpe JJ: Neurological and neuromuscular disorders. In: Neonatology: Pathophysiology and Management of the Newborn, 5th edn (Avery GB, MA Fletcher, MH MacDonald, eds). Philadelphia: Lippincott Williams & Wilkins, 1999; pp 196 203. 3 Bågenholm R, Nilsson UA, Kjellmer I: Formation of free radicals in hypoxic ischemic brain damage in the neonatal rat, assessed by an endogenous spin trap and lipid peroxidation. Brain Res 1997; 773: 132 138. 4 Fulia F, Gitto E, Cuzzocrea S, et al: Increased levels of malondialdehyde and nitrite/nitrate in the blood of asphyxiated newborns: reduction by melatonin. J Pineal Res 2001; 31: 343 349. 5 Ikeda T, Choi BH, Yee S, et al: Oxidative stress, brain white matter damage and intrauterine asphyxia in fetal lambs. Int J Dev Neurosci 1999; 17: 1 14. 6 Yi it S, Yurdakök M, Kilin K, et al: Serum malondialdehyde concentration in babies with hyperbilirubinaemia. Arch Dis Child Fetal Neonatal Ed 1999; 80: F235 F237. 7 Ballard JL, Khoury JC, Wedig K, et al: New Ballard Score, expanded to include extremely premature infants. J Pediatr 1991; 119: 417 423. 8 Sarnat HB, Sarnat MS: Neonatal encephalopathy following fetal distress. A 225
clinical and electroencephalographic study. Arch Neurol 1976; 33: 696 705. 9 Volpe JJ: Neurology of the Newborn, 3rd edn. Philadelphia: WB Saunders, 1995; pp 21 369. 10 Wong SH, Knight JA, Hopfer SM, et al: Lipoperoxides in plasma as measured by liquid-chromatographic separation of malondialdehyde-thiobarbituric acid adduct. Clin Chem 1987; 33: 214 220. 11 Cortas NK, Wakid NW: Determination of inorganic nitrate in serum and urine by a kinetic cadmium-reduction method. Clin Chem 1990; 36: 1440 1443. 12 Buonocore G, Zani S, Perrone S, et al: Intraerythrocyte nonprotein-bound iron and plasma malondialdehyde in the hypoxic newborn. Free Radic Biol Med 1998; 25: 766 770. 13 Huguet F, Guerraoui A, Barrier L, et al: Changes in excitatory amino acid levels and tissue energy metabolites of neonate rat brain after hypoxia and hypoxia ischemia. Neurosci Lett 1998; 240: 102 106. 14 Temesvári P, Karg E, Bódi I, et al: Impaired early neurologic outcome in newborn piglets reoxygenated with 100% oxygen compared with room air after pneumothorax-induced asphyxia. Pediatr Res 2001; 49: 812 819. 15 Kumar A, Ramakrishna SV, Basu S, et al: Oxidative stress in perinatal asphyxia. Pediatr Neurol 2008; 38: 181 185. 16 Horakova L, Ondrejickova, Uraz V, et al: Short cerebral ischemia and subsequent reperfusion and treatment with stobadine. Experientia 1992; 48: 872 874. 17 Hsiang JN, Wang JY, Ip SM, et al: The time course and regional variations of lipid peroxidation after diffuse brain injury in rats. Acta Neurochir (Wien) 1997; 139: 464 468. 18 Dorman DC, Côté LM, Buck WB: Effects of an extract of Gingko biloba on bromethalin induced cerebral lipid peroxidation and edema in rats. Am J Vet Res 1992; 53: 138 142. Author s address for correspondence Dr Erdal Peker Neonatology Unit, Yuzuncu Yil University Hospital, 65100 Van, Turkey. E-mail:pekererdal@hotmail.com 226