Risk of intracerebral haemorrhage in Chinese patients with atrial fibrillation on warfarin with cerebral microbleeds: the IPAAC-Warfarin study.

BACKGROUND AND PURPOSE: Cerebral microbleeds (CMBs), which predict future intracerebral haemorrhage (ICH), may guide anticoagulant decisions for atrial fibrillation (AF). We aimed to evaluate the risk of warfarin-associated ICH in Chinese patients with AF with CMBs. METHODS: In this prospective, observational, multicentre study, we recruited Chinese patients with AF who were on or intended to start anticoagulation with warfarin from six hospitals in Hong Kong. CMBs were evaluated with 3T MRI brain at baseline. Primary outcome was clinical ICH at 2-year follow-up. Secondary outcomes were ischaemic stroke, systemic embolism, mortality of all causes and modified Rankin Scale ≥3. Outcome events were compared between patients with and without CMBs. RESULTS: A total of 290 patients were recruited; 53 patients were excluded by predefined criteria. Among the 237 patients included in the final analysis, CMBs were observed in 84 (35.4%) patients, and 11 had ≥5 CMBs. The mean follow-up period was 22.4±10.3 months. Compared with patients without CMBs, patients with CMBs had numerically higher rate of ICH (3.6% vs 0.7%, p=0.129). The rate of ICH was lower than ischaemic stroke for patients with 0 to 4 CMBs, but higher for those with ≥5 CMBs. CMB count (C-index 0.82) was more sensitive than HAS-BLED (C-index 0.55) and CHA2DS2-VASc (C-index 0.63) scores in predicting ICH. CONCLUSIONS: In Chinese patients with AF on warfarin, presence of multiple CMBs may be associated with higher rate of ICH than ischaemic stroke. Larger studies through international collaboration are needed to determine the risk:benefit ratio of oral anticoagulants in patients with AF of different ethnic origins.

Rare variants and de novo variants in mesial temporal lobe epilepsy with hippocampal sclerosis.

OBJECTIVE: We investigated the role of rare genetic variants and of de novo variants in the pathogenesis of mesial temporal lobe epilepsy related to hippocampal sclerosis (MTLE-HS). METHODS: Whole-exome sequencing (WES) was performed in patients with MTLE-HS and their unaffected parents (trios). Genes or gene sets that were enriched with predicted damaging rare variants in the patients as compared to population controls were identified. Patients and their parents were compared to identify whether the variants were de novo or inherited. RESULTS: After quality control, WES data from 47 patients (26 female), including 23 complete trios, were available for analysis. Compared with population controls, significant enrichment of rare variants was observed in SEC24B. Integration of gene set data describing neuronal functions and psychiatric disorders showed enrichment signal on fragile X mental retardation protein (FMRP) targets. Twenty-one de novo variants were identified, with many known to cause neuropsychiatric disorders. The FMRP-targeted genes also carried more de novo variants. Inherited compound heterozygous and homozygous variants were identified. CONCLUSIONS: The genetic architecture underlying MTHE-HS is complex. Multiple genes carrying de novo variants and rare variants among FMRP targets were identified, suggesting a pathogenic role. MTLE-HS and other neuropsychiatric disorders may have shared biology.

The professional practice and training of neurology in the Asian and Oceanian Region: A cross-sectional survey by the Asian and Oceanian Association of Neurology (AOAN).

OBJECTIVE: To survey AOAN member countries regarding their organizational structure, postgraduate neurology training program, and resources for neurological care provision. METHODOLOGY: A cross-sectional survey using a 36-item questionnaire was conducted among country representatives to AOAN from August 2015 to August 2016. RESULTS: A total of 18/20 AOAN member countries participated in the survey. All the countries have organized association with regular meetings, election of officers and neurology training program. In 9/18 countries, professionals other than neurologists were eligible for affiliation. In 11/18 countries, prior Internal medicine training (or equivalent postgraduate housemanship) is prerequisite to neurology program. Recertification examination is not a practice, but submission of CME is required in 7/18 countries to maintain membership. 12/18 countries publish peer-reviewed journals with at least 1 issue per year. Subspecialty training is offered in 14/18 countries. The ratio of neurologist to population ranges from 1:14,000 to as low as 1:32 million with 9/18 having <1 neurologist per 100,000 population. 6/18 countries have at least 1 specialized center solely for neurological diseases. In government-funded hospitals, the lag time to be seen by a neurologist and/or obtain neuroimaging scan ranges from 1day to 3months. All except one country have several medical- and lay- advocacy or support groups for different neurological conditions. IMPLICATIONS: The data generated can be used for benchmarking to improve neurological care, training, collaborative work and research in the field of neurosciences among the AOAN member countries. The paper presented several strategies used by the different organizations to increase their number of neurologists and improve the quality of training. Sharing of best practices, academic networking, exchange programs and use of telemedicine have been suggested.

Evolution of intracranial atherosclerotic disease under modern medical therapy.

OBJECTIVE: Understanding how symptomatic intracranial atherosclerotic disease (ICAD) evolves with current medical therapy may inform secondary stroke prevention. METHODS: In a prospective academic-initiated study, we recruited 50 patients (mean age = 63.4 ± 9.0 years) with acute strokes attributed to high-grade (≥70%) intracranial atherosclerotic stenosis for 3-dimensional rotational angiograms before and after intensive medical therapy for 12 months. Treatment targets included low-density lipoprotein ≤ 70mg/dl, glycosylated hemoglobin (HbA1c) ≤ 6.5%, and systolic blood pressure ≤ 140 mmHg. We analyzed infarct topography and monitored microembolic signal in recurrent strokes. The reference group was a published cohort of 143 ICAD patients. RESULTS: Overall, the stenoses regressed from 79% at baseline (interquartile range [IQR] = 71-87%) to 63% (IQR = 54-74%) in 1 year (p < 0.001). Specifically, the qualifying lesions (n = 49) regressed (stenosis reduced >10%) in 24 patients (49%), remained quiescent (stenosis same or ±10%) in 21 patients (43%), and progressed (stenosis increased >10%) in 4 patients (8%). There was no difference in intensity of risk factor control between groups of diverging clinical or angiographic outcomes. Higher HbA1c at baseline predicted plaque regression at 1 year (odds ratio = 4.4, 95% confidence interval = 1.4-14.5, p = 0.006). Among the 6 patients with recurrent strokes pertaining to the qualifying stenosis, 5 patients had solitary or rosarylike acute infarcts along the internal or anterior border zones, and 2 patients showed microembolic signals in transcranial Doppler ultrasound. INTERPRETATION: A majority of symptomatic high-grade intracranial plaques had regressed or remained quiescent by 12 months under intensive medical therapy. Artery-to-artery thromboembolism with impaired washout at border zones was a common mechanism in stroke recurrence.

Case-control association study of polymorphisms in the voltage-gated sodium channel genes SCN1A, SCN2A, SCN3A, SCN1B, and SCN2B and epilepsy.

High-frequency action potentials are mediated by voltage-gated sodium channels, composed of one large α subunit and two small ß subunits, encoded mainly by SCN1A, SCN2A, SCN3A, SCN1B, and SCN2B genes in the brain. These play a key role in epilepsy, with the most commonly mutated gene in epilepsy being SCN1A. We examined whether polymorphisms in the above genes affect epilepsy risk in 1,529 epilepsy patients and 1,935 controls from four ethnicities or locations: Malay, Indian, and Chinese, all from Malaysia, and Chinese from Hong Kong. Of patients, 19 % were idiopathic, 42 % symptomatic, and 40 % cryptogenic. We genotyped 43 polymorphisms: 27 in Hong Kong, 28 in Malaysia, and 12 in both locations. The strongest association with epilepsy was rs3812718, or SCN1A IVS5N+5G>A: odds ratio (OR) = 0.85 for allele G (p = 0.0009) and 0.73 for genotype GG versus AA (p = 0.003). The OR was between 0.76 and 0.87 for all ethnicities. Meta-analysis confirmed the association (OR = 0.81 and p = 0.002 for G, and OR = 0.67 and p = 0.007 for GG versus AA), which appeared particularly strong for Indians and for febrile seizures. Allele G affects splicing and speeds recovery from inactivation. Since SCN1A is preferentially expressed in inhibitory neurons, G may decrease epilepsy risk. SCN1A rs10188577 displayed OR = 1.20 for allele C (p = 0.003); SCN2A rs12467383 had OR = 1.16 for allele A (p = 0.01), and displayed linkage disequilibrium with rs2082366 (r (2) = 0.67), whose genotypes tended toward association with SCN2A brain expression (p = 0.10). SCN1A rs2298771 was associated in Indians (OR = 0.56, p = 0.005) and SCN2B rs602594 with idiopathic epilepsy (OR = 0.62, p = 0.002). Therefore, sodium channel polymorphisms are associated with epilepsy.

The role of the Ala746Thr variant in the ATP13A2 gene among Chinese patients with Parkinson's disease.

The association between idiopathic Parkinson's disease (PD) and the ATP13A2 (PARK9) Ala746Thr variant, associated with Kufor-Rakeb syndrome, is controversial. We investigated this association in 69 patients with early onset PD (EOPD; ≦50 years of age), 192 patients with late onset PD (LOPD; >50 years of age), and 180 healthy controls in the Chinese population in Hong Kong. The presence of the Ala746Thr variant in the ATP13A2 locus was examined in all participants. We detected the heterozygous Ala746Thr variant in one healthy control (0.6%), one patient with EOPD (1.4%, p=0.50), and one patient with LOPD (0.5%, p=0.96). We suggest that the ATP13A2 Ala746Thr variant is not a common risk factor for PD in the Chinese population in Hong Kong.

Two-stage genome-wide association study identifies variants in CAMSAP1L1 as susceptibility loci for epilepsy in Chinese.

In the majority of patients, epilepsy is a complex disorder with multiple susceptibility genes interacting with environmental factors. However, we understand little about its genetic risks. Here, we report the first genome-wide association study (GWAS) to identify common susceptibility variants of epilepsy in Chinese. This two-stage GWAS included a total of 1087 patients and 3444 matched controls. In the combined analysis of the two stages, the strongest signals were observed with two highly correlated variants, rs2292096 [G] [P= 1.0 × 10(-8), odds ratio (OR) = 0.63] and rs6660197 [T] (P= 9.9 × 10(-7), OR = 0.69), with the former reaching genome-wide significance, on 1q32.1 in the CAMSAP1L1 gene, which encodes a cytoskeletal protein. We also refined a previously reported association with rs9390754 (P= 1.7 × 10(-5)) on 6q21 in the GRIK2 gene, which encodes a glutamate receptor, and identified several other loci in genes involved in neurotransmission or neuronal networking that warrant further investigation. Our results suggest that common genetic variants may increase the susceptibility to epilepsy in Chinese.

Gene-wide tagging study of the association between ABCC2, ABCC5 and ABCG2 genetic polymorphisms and multidrug resistance in epilepsy.

AIM: To determine the association between polymorphisms of the multidrug transporter genes ABCC2, ABCC5 and ABCG2, and drug resistance in epilepsy by genotyping comprehensive sets of tagging SNPs. MATERIALS & METHODS: A total of 25 tagging SNPs from ABCC2, ABCC5 and ABCG2 genes were genotyped in a total of 590 Han Chinese epilepsy patients (262 drug resistant and 328 drug responsive). Genotype and allele distributions in drug-responsive and drug-resistant patients were compared. RESULTS: Genotype distributions of all the selected SNPs were consistent with Hardy-Weinberg equilibrium. None of the polymorphisms, either genotype or allele distributions, were significantly associated with drug resistance. For each gene, no haplotypes of over 1% frequency, and that included all SNPs of the gene, were associated with drug resistance. CONCLUSION: This gene-wide tagging study revealed no association between ABCC2, ABCC5 and ABCG2 genetic polymorphisms and multidrug resistance in epilepsy.

Failure to detect association between polymorphisms of the sodium channel gene SCN1A and febrile seizures in Chinese patients with epilepsy.

Arecent study in Caucasians found an association between the single nucleotide polymorphism (SNP) of SCN1A, IVS5N +5 G>A (rs3812718), and febrile seizures (FS). We examined whether this and other tagging SNPs of SCN1A were associated with an increased risk of FS in Han Chinese. A total of 728 Han Chinese patients with focal epilepsy were recruited: 97 had a history of FS (58% male, mean age 35 ± 12 years) and 631 did not (50% male, mean age 40 ± 15 years). Genotyping was performed for IVS5N +5 G>A and seven other tagging SNPs selected from the HapMap database. Genotyping was also performed in 848 ethnically matched population controls (50% male, mean age 37 ± 17 years). There was no statistically significant difference in either allele or genotype frequency of any of the SNPs studied between epilepsy patients with and without FS, and between epilepsy patients with FS and controls. The results do not suggest that SCN1A SNPs are susceptibility factors for FS in Han Chinese.

Effects of statins on progression of subclinical brain infarct.

BACKGROUND: Subclinical brain infarct (SBI) is associated with subsequent stroke and cognitive decline. A longitudinal epidemiological study suggests that statins may prevent development of SBI. We investigated the effects of statins upon development of brain infarct by performing a post-hoc analysis of the Regression of Cerebral Artery Stenosis (ROCAS) study. METHODS: The ROCAS study is a randomized, double-blind, placebo-controlled study evaluating the effects of simvastatin 20 mg daily upon progression of asymptomatic middle cerebral artery stenosis among stroke-free individuals over 2 years. A total of 227 subjects were randomized to either placebo (n = 114) or simvastatin 20 mg daily (n = 113). The number of brain infarcts as detected by MRI was recorded at baseline and at the end of the study. The primary outcome measure was the number of new brain infarcts at the end of the study. RESULTS: Among the 227 randomized subjects, 33 (14.5%) had SBI at baseline. At the end of the study, significantly fewer subjects in the active group (n = 1) had new brain infarcts compared with the placebo group (n = 8; p = 0.018). The new brain infarcts of subjects in the active group were subclinical. Among the placebo group, the new brain infarcts of 3 subjects were symptomatic while those of the remaining 5 subjects were subclinical. Among putative variables, multivariate regression analysis showed that only the baseline number of SBIs (OR = 6.27, 95% CI 2.4-16.5) and simvastatin treatment (OR = 0.09, 95% CI 0.01-0.82) independently predicted the development of new brain infarcts. CONCLUSIONS: Consistent with findings of the epidemiological study, our study suggests that statins may prevent the development of a new brain infarct.

Statins for asymptomatic middle cerebral artery stenosis: The Regression of Cerebral Artery Stenosis study.

BACKGROUND: The progression of cerebral atherosclerosis increases the risk of stroke and vascular events. Given the known benefits of statins in retarding coronary and carotid atherosclerosis progression, we studied the effects of statins on asymptomatic middle cerebral artery (MCA) stenosis progression. METHODS: We conducted a randomized, double-blind, placebo-controlled study to evaluate the effects of simvastatin on the progression of MCA stenosis among stroke-free individuals who had mild to moderately elevated LDL cholesterol (3.0-5.0 mmol/l). Two hundred and twenty-seven subjects were randomized to either placebo (n = 114) or simvastatin 20 mg daily (n = 113). The severity of MCA stenosis at baseline and at the end of the study was graded by MRA into normal, minimal (<10%), mild (10-49%), moderate (50-90%) and severe (>90%). The primary outcome was the change in grading of MCA stenosis over 2 years. RESULTS: At the end of the study, the LDL cholesterol level decreased by 1.43 and 0.12 mmol/l for the active and placebo groups, respectively (p < 0.001). There was no significant difference in the proportion of patients having stable, progressive and regressive MCA stenosis between the placebo (72, 22 and 6%) and active groups (78.6, 15.5 and 5.8%). The all-cause mortality was significantly lower in the active group (n = 0) relative to the placebo group (n = 7, p = 0.014). Any clinical events were also lower in the active group (n = 5) than in the placebo group (n = 13, p = 0.052). CONCLUSIONS: Simvastatin 20 mg daily had no apparent effect upon the evolution of asymptomatic MCA stenosis over 2 years.

Effects of valproate or lamotrigine monotherapy on the reproductive endocrine and insulin-related metabolic profile in Chinese adults with epilepsy: a prospective randomized study.

Cross-sectional studies have suggested that valproate treatment may be associated with hyperinsulinemia and hyperandrogenism in women. Few prospective data are available. We evaluated the reproductive endocrine and insulin-related metabolic parameters in men and women with untreated epilepsy randomized to valproate (n=44) or lamotrigine (n=37) monotherapy for 12 months. On treatment, there was no significant difference in fasting serum insulin concentrations between the two groups. In women (n=40), there was no significant difference between the two groups in change from baseline in serum total testosterone, dehydroepiandrosterone sulfate, luteinizing hormone, or follicle-stimulating hormone. In men (n=41), follicle-stimulating hormone concentration significantly decreased in patients taking valproate compared with those on lamotrigine as early as 3 months after treatment. Greater attention should be paid to investigate the potential impact of valproate on reproductive function in men.

Gene-wide tagging study of association between ABCB1 polymorphisms and multidrug resistance in epilepsy in Han Chinese.

AIMS: It remains controversial whether polymorphisms of the multidrug resistance gene ABCB1 are associated with pharmacoresistance in epilepsy. To further study the potential association, we genotyped a broad set of tagging SNPs, and explored whether any associations were affected by other host factors. We correlated any association with cerebral mRNA expression of ABCB1. MATERIALS & METHODS: A total of 12 tagging and candidate SNPs of ABCB1 were genotyped in 464 Chinese epilepsy patients (270 drug responsive, 194 drug resistant). Genotype and allele distributions in drug-responsive and drug-resistant patients were compared. ABCB1 mRNA was quantified by real-time PCR in brain samples resected from 20 patients with drug-resistant epilepsy. Its level was compared between patients with different genotypes of ABCB1 SNPs found to be associated with drug resistance. RESULTS: The intronic polymorphism rs3789243 (p = 0.009 for allele analysis) and the coding polymorphism 2677G/T/A (p = 0.02), and haplotypes containing them, were associated with drug resistance. The 2677G/T/A genotypes remained significantly associated with drug resistance after multiple logistic regression and correction for multiple comparisons. The associations with drug resistance were found in males (p = 0.004 for rs3789243 and p = 0.0007 for 2677T/A>G) but not females, and in patients with localization-related (p = 0.006 for rs3789243 and p = 0.01 for 2677T/A>G) but not idiopathic-generalized epilepsy. ABCB1 mRNA levels did not correlate with genotypes. CONCLUSION: In Chinese epilepsy patients, the ABCB1 intronic polymorphism rs3789243 and the coding polymorphism 2677, and haplotypes containing them, may be associated with drug resistance, without an effect on mRNA expression. There was preliminary evidence of interactions between these polymorphisms and gender and epilepsy syndrome.

Effects of statins on the progression of cerebral white matter lesion: Post hoc analysis of the ROCAS (Regression of Cerebral Artery Stenosis) study.

Arteriosclerotic related cerebral white matter lesion (WML) is associated with increased risk of death, stroke, dementia, depression, gait disturbance, and urinary incontinence. We investigated the effects of statins on WML progression by performing a post hoc analysis on the ROCAS (Regression of Cerebral Artery Stenosis) study, which is a randomized, double-blind, placebo-controlled study evaluating the effects of statins upon asymptomatic middle cerebral artery stenosis progression among stroke-free individuals. Two hundreds and eight randomized subjects were assigned to either placebo (n = 102) or simvastatin 20 mg daily (n = 106) for 2 years. Baseline severity of WML was graded visually into none, mild, and severe. Volume (cm3) of WML was determined quantitatively at baseline and at end of study using a semi-automated method based on MRI. Primary outcome was the change in WML volume over 2 years. After 2 years of follow-up, there was no significant change in WML volume between the active and the placebo group as a whole. However, stratified analysis showed that for those with severe WML at baseline, the median volume increase in the active group (1.9 cm3) was less compared with that in the placebo group (3.0 cm3; P = 0.047). Linear multivariate regression analysis identified that baseline WML volume (beta = 0.63, P < 0.001) and simvastatin treatment (beta = -0.214, P = 0.043) independently predicted change in WML volume. Our findings suggest that statins may delay the progression of cerebral WML only among those who already have severe WML at baseline.

Multidrug resistance in epilepsy and polymorphisms in the voltage-gated sodium channel genes SCN1A, SCN2A, and SCN3A: correlation among phenotype, genotype, and mRNA expression.

OBJECTIVES: Many antiepileptic drugs (AEDs) prevent seizures by blocking voltage-gated brain sodium channels. However, treatment is ineffective in 30% of epilepsy patients, which might, at least in part, result from polymorphisms of the sodium channel genes. We investigated the association of AED responsiveness with genetic polymorphisms and correlated any association with mRNA expression of the neuronal sodium channels. METHODS: We performed genotyping of tagging and candidate single nucleotide polymorphisms (SNPs) of SCN1A, 2A, and 3A in 471 Chinese epilepsy patients (272 drug responsive and 199 drug resistant). A total of 27 SNPs were selected based on the HapMap database. Genotype distributions in drug-responsive and drug-resistant patients were compared. SCN2A mRNA was quantified by real-time PCR in 24 brain and 57 blood samples. Its level was compared between patients with different genotypes of an SCN2A SNP found to be associated with drug responsiveness. RESULTS: SCN2A IVS7-32A>G (rs2304016) A alleles were associated with drug resistance (odds ratio = 2.1, 95% confidence interval: 1.2-3.7, P=0.007). Haplotypes containing the IVS7-32A>G allele A were also associated with drug resistance. IVS7-32A>G is located within the putative splicing branch site for splicing exons 7 and 9. PCR of reverse-transcribed RNA from blood or brain of patients with different IVS7-32A>G genotypes using primers in exons 7 and 9 showed no skipping of exon 8, and real-time PCR showed no difference in SCN2A mRNA levels among genotypes. CONCLUSION: Results of this study suggest an association between SCN2A IVS7-32A>G and AED responsiveness, without evidence of an effect on splicing or mRNA expression.

LRRK2 Gly2385Arg mutation and clinical features in a Chinese population with early-onset Parkinson's disease compared to late-onset patients.

The frequency of LRRK2 Gly2385Arg mutation in Hong Kong Chinese with early-onset (age < or =45 years) Parkinson's disease was identified and compared with late-onset patients (age >50 years) and controls. The mutation prevalence were 8.8, 8.3, and 0% for early-onset, late-onset, and controls, respectively. The mean age of onset among LRRK2 G2385R carriers was 42.7 years old for early-onset compared to 74.3 for late-onset patients. LRRK2 G2385R mutation appears to be as prevalent among early-onset as late-onset patients.

PARK2 mutations and clinical features in a Chinese population with early-onset Parkinson's disease.

Our aim was to characterise PARK2 mutations and clinical features in Hong Kong Chinese with early-onset Parkinson's disease. Subjects were recruited from two major hospitals. Detailed data included demographics, age of onset, duration of disease, neurological manifestations, complications and disease severity. Genetic analysis for PARK2 mutations was performed. Thirty-four patients were recruited (mean age of onset = 39 years; mean duration of disease = 10 years). Seven patients reported a family history. The salient clinical manifestations were resting tremor (33/34), bradykinesia (33/34), rigidity (30/34), postural instability (20/34), good response to L-dopa (33/34), asymmetry at onset (31/34) and sleep benefit (12/34). Motor complications were reported in a significant number of patients, and depression was the most common nonmotor complication. Five patients were identified to have PARK2 mutations. Two sisters were compound heterozygotes for an insertion and a deletion, a novel and rare 1 bp insertion/nonsense mutation c1378_1379insG (exon 12) and the entire deletion of exon 7. Another patient was homozygous for the entire deletion of exon 6. Two carriers were identified, one with a T1321C (Cys441Arg) missense mutation in exon 12 and another with a snp within intron 4. Our study reviewed a higher prevalence of PARK2 mutations in Chinese than that previously documented. A compound heterozygous mutation within two sisters with significant differences in age of onset and phenotypic manifestations suggest that modifier affects may be present in this family.

Association between ABCB1 C3435T polymorphism and drug-resistant epilepsy in Han Chinese.

There is accumulating evidence to suggest that overexpression of efflux drug transporters at the blood-brain barrier, by reducing antiepileptic drug (AED) accumulation in the seizure foci, contributes to drug resistance in epilepsy. P-glycoprotein, encoded by the ABCB1 gene, is the most studied drug transporter. There are conflicting data as to whether the CC genotype of the ABCB1 3435C>T polymorphism is associated with drug resistance in Caucasian patients with epilepsy. We investigated this association in ethnic Chinese. ABCB1 3435C>T was genotyped in 746 Han Chinese patients with epilepsy and 179 controls. Patients with drug-resistant epilepsy were more likely to have the TT genotype compared with those with drug-responsive epilepsy (16.7% vs 7.4%, odds ratio=2.5, 95% confidence interval=1.4-4.6, P=0.0009). Our results contrast with those of studies of Caucasians, and highlight the complexity of the possible role of this polymorphism in AED response in different ethnic populations.