Reference distributions for apolipoproteins AI and B and the apolipoprotein B/AI ratios: a practical and clinically relevant approach in a large cohort

The two serum apolipoproteins in the highest concentrations, apolipoprotein (apo) AI and apolipoprotein B, and the apolipoprotein B/AI ratio are measured to assess clinical risk for atherosclerotic heart and peripheral vascular diseases. The study is based on a cohort of over 37,000 Caucasian individuals from northern New England measured in one laboratory by immunonephelometry using standardized reference materials. All samples received for protein analyses were accepted provided adequate identifying information was available. Laboratory and demographic information was entered into a single database for subsequent study. Our results show that for males without evidence of inflammation, values of apo AI change little through life. For females, however, values gradually increase until about 60 years of age then fall somewhat thereafter. Among adults, females have higher apo AI values on average, than males. Apo B values change significantly through life, increasing after the end of the second decade to a peak during the sixth decade, then falling thereafter. In the past, concern has been expressed that apo AI is an acute phase reactant (APR), thus complicating cardiovascular risk assessment. The effects of an APR (C-reactive protein >or=10 mg/L) on apo AI, but not on apo B, are measurable for both sexes, most noticeably beyond the age of 60 years in males and females. When values were expressed as age- and gender-specific multiples of the median (MoMs), the resulting distributions fit a log-Gaussian distribution well over a broad range. The size of the relatively homogenous cohort, by a standardized approach, provides a firm basis for comparison to preexisting reference intervals and for establishing a clinically useful and current reference interval for the three main apolipoprotein values.     

Factors behind the increase in cardiovascular mortality in Russia: apolipoprotein AI and B distribution in the Arkhangelsk study 2000

BACKGROUND: Cardiovascular mortality is markedly higher in Russia than in Western Europe and the US. Little is known about indicators of atherosclerotic risk in the Russian population. To our knowledge, this is the first study of apolipoprotein (apo) AI and B in Russia based on the WHO-IFCC standard. METHODS: We measured apo AI and B by immunoturbidimetric assay in 3694 men and women from Arkhangelsk, Russia, in 1999-2000. RESULTS: The age-related distribution of apo B was similar to that in other countries, whereas the apo AI profile was different. For men > or = 20 years, apo AI was considerably higher than in studies from other countries. Women had also relatively high apo AI concentrations, although the difference was not as pronounced as in men. The apo AI concentration was positively associated with age and lifestyle variables such as alcohol consumption and physical activity, and negatively associated with body mass index and self-reported myocardial infarction. gamma-Glutamyltransferase was positively associated with apo AI in both sexes. CONCLUSIONS: The apparently favorable apolipoprotein profiles contrast with official death statistics indicating high cardiovascular mortality in Russia. High apo AI might indicate excessive alcohol consumption.     

Reference distributions for alpha2-macroglobulin: a comparison of a large cohort to the world's literature

The lack of satisfactory methods for quantifying serum levels and a credible reference material has limited bedside use of serum alpha(2)-macroglobulin (alpha2M) measurements. Great strides have been made in the last few years. The remaining barrier to more relevant and cost effective use of serum protein data for diagnosis and prognosis is the availability of reliable reference intervals from birth to old age for both males and females. A total of 40 publications reporting reference intervals have been identified that meet the criteria used in our prior five studies, and these have been analyzed statistically. On average, previous small studies of these individual proteins agree with our life-long reference ranges over their constrained age ranges. This meta-analysis provides support for our reference ranges and places them in the perspective of previous publications.     

Reference values for apolipoproteins A-I and B in healthy subjects, by age.

We determined reference values of apolipoproteins A-I (apo A-I) and B (apo B) in serum from a population of 448 healthy subjects (265 men and 183 women, ages 18 to 61 years) by a kinetic immunonephelometric procedure. Frequency distributions of apo A-I were normal, whereas those of apo B were not and yielded asymmetrical curves. Thus, reference intervals for apo A-I were determined as mean +/- 2SD (1.08-1.89 g/L), but a nonparametric method was used for determining reference intervals for apo B (0.60-1.94 g/L). Apo B concentrations were significantly higher (P less than 0.001) in men than in women (0.63-2.01 g/L, mean 1.21 g/L; and 0.54-1.91 g/L, mean 1.08 g/L, respectively). No significant differences for apo A-I between men and women were observed. Concentrations of both proteins increased with age, but apo B increased more than apo A-I. We conclude that not only sex but also the age of the subjects must be considered in interpreting laboratory results for apolipoproteins.     

Relationship of apolipoproteins Al, B and lipoprotein Lp(a) to hepatic function of liver recipients during the early post-transplant period

Abstract. To evaluate serum apo AI, apo B, Lp(a) and the ratio of unesterified cholesterol to total cholesterol as markers of hepatic synthetic capacity after orthotopic liver transplantation, serial measurements of these variables were performed on post-transplant days 1, 3, 5, 7, 10 and 14 in 70 patients. Liver function was assessed by a quantitative dynamic test based on the hepatic conversion of lidocaine to monoethylglycinexylidide (MEGX). Patients were divided into two groups on the basis of clinical and laboratory findings, those with evidence ( n = 46) and those without evidence ( n = 24) of hepatic dysfunction. Apo AI levels fell in both groups to day 5, but then began to increase in the group with good hepatic function, a highly significant ( P & 0001) positive correlation being found with the results of the MEGX test on post-transplant days 7, 10 and 14. The ratio of unesterified cholesterol to total cholesterol rose in both groups from days 1 to 7 and then began to fall in the group without hepatic dysfunction; a highly significant ( P < 0.001) negative correlation was observed with the results of the MEGX test on days 10 and 14. Apo B levels rose in both groups from days 1 to 10, with no significant differences between the two groups; no correlation was observed with the results of the MEGX test on any study day. In contrast, serum Lp(a) concentrations fell in both groups to day 5, but then began to increase in patients with good hepatic function; a highly significant ( P <0001) positive correlation with the results of the MEGX test was recorded on post-transplant day 14. We conclude that serial measurements of serum apo AI, Lp(a) and the ratio of unesterified cholesterol to total cholesterol are of value in monitoring the restoration of hepatic synthetic capacity in patients after orthotopic liver transplantation.     

Reference distributions for complement proteins C3 and C4: a comparison of a large cohort to the world's literature

The lack of credible reference materials and satisfactory methods for quantifying serum levels has limited the bedside use of complement protein (C3 and C4) measurements. However, great technological strides have been made in the last few years. The remaining barrier to a more relevant and cost-effective use of serum protein data for diagnosis and prognosis is the availability of reliable reference intervals from birth to old age for both males and females. Fifty-one publications reporting reference intervals were identified that meet the criteria used in our prior four studies, and these were analyzed statistically. Previous small studies with constrained age ranges agree, on average, with our larger series of life-long reference ranges. This meta-analysis provides support for our reference ranges and places them in the context of previous publications.     

Measurement of serum apolipoprotein B by radioimmunoassay

Abstract. A sensitive and specific double antibody radioimmunoassay for the major apolipoprotein (apo B) of human serum very low density lipoprotein (VLDL) and low density lipoprotein (LDL) is described. Using anti-LDL and anti-apo B antibodies the immunoreactivity of LDL and apo B were compared. Human LDL and its isolated apo B were not immunologically identical when each antiserum was used with its homologous label; a population of antibodies was selected which reacted with antigenic sites unique to the antigen itself as well as to those which were common to the closely related protein. When the heterologous label was used with either antiserum, a population of antibodies directed against antigenic sites shared by the LDL and apo B molecules was selected.
Apo B in sera samples can be measured using either anti-LDL or anti-apo B antibodies provided that intact LDL was used for preparation of the iodinated tracer and standard. Serum apo B levels in healthy normolipi-daemic males and females were 0.93 ±0.25 g/l (range 0.58–1.39) and 0.90 ± 0.15 g/l (range 0.58–1.12), respectively. The total cholesterol and apo B, and phos-pholipid and apo B concentrations for both males and females were significantly correlated (P<0.05). In another normolipidaemic population ( n = 52), total serum apo B values correlated positively with LDL cholesterol ( r= 0.92, P< 0.001).
Apo B was measured in sera from patients with abetalipoproteinaemia, familial hypercholesterolaemia and Tangiers disease. Apo B was not detected in the serum of subjects with abetalipoproteinaemia, while the apo B level in the familial hypercholesterolaemic subjects was significantly elevated (range 3.26–4.94 g/l) compared to normals (P<0.001). Serum apo B (0.80 g/l) of the subject with Tangier disease was within the normal range.     

Reference distributions for the positive acute phase proteins, alpha1-acid glycoprotein (orosomucoid), alpha1-antitrypsin, and haptoglobin: a comparison of a large cohort to the world's literature

Limiting bedside use of positive acute phase protein measurements (alpha1-acid glycoprotein (orosomucoid), alpha1-antitrypsin, and haptoglobin) has been the lack of satisfactory methods for quantifying serum levels and a credible reference material. Great strides have been made in the last few years. The remaining barrier to more relevant and cost-effective use of serum protein data for diagnosis and prognosis is the availability of reliable reference intervals from birth to old age for both males and females. Sixty publications reporting reference intervals have been identified which meet the criteria used in our prior two studies, and these have been analyzed statistically. Previous small studies of these individual proteins agree on average, over their constrained age ranges, with our life-long reference ranges. This meta-analysis provides support for our reference ranges and places them in the perspective of previous publications.     

Reference distributions for serum iron and transferrin saturation: a comparison of a large cohort to the world's literature

The appropriate clinical use of serum iron and transferrin saturation (TSAT) requires satisfactory reference intervals from birth to old age, and for males and females. This study identified 54 publications from 1974 to 2001 that met the criteria used in three prior meta-analyses, and these were analyzed statistically. A summary of our review is presented along with our reference population data on these measurements. This analysis places previous publications in perspective and suggests possible reasons for the observed differences. Previous studies of the individual analytes, serum iron, transferrin, and TSAT values agree with the reference ranges presented in this study, although the entire experience over time and between sexes has not been available before. Our 95% reference ranges are somewhat broader than those of the smaller studies, but they agree well with those of the larger ones.     

Quantification of human serum apolipoprotein AI by enzyme immunoassay

We developed a quantitative assay for apolipoprotein AI (apo AI) in human serum, using a "sandwich"-type enzyme-linked immunosorbent assay. Diluted serum samples were pipetted into the wells of polystyrene microtiter plates that had been previously coated with purified rabbit anti-human apo AI antibodies. After incubation for 2 h and washing, antibodies conjugated to horseradish peroxidase (EC 1.11.1.7) were added and incubated for 2 h; after further washing, the bound enzyme was assayed by oxidation of o-phenylenediamine. Assay conditions were optimized for the incubation time and the amounts of coating antibodies and conjugate. Assay sensitivity is about 0.5 ng of apo AI, with a working range of 1 to 14 ng, similar to that of radioimmunoassays for human apo AI. The standard curves for apo AI in serum or HDL and for purified apo AI were parallel. Delipidation, heat treatment, or addition of detergents did not affect the amount of immunoassayable apo AI in human serum. The intra- and interassay CVs were 4 and 8%, respectively. Results for 100 serum samples compared well with those by immunonephelometry (r = 0.94).     

Automated immunoturbidimetric analysis of six plasma apolipoproteins: correlation with radial immunodiffusion assays

We measured six apolipoproteins (apo AI, AII, B, CII, CIII, and E) by turbidimetric method using an automatic discrete biochemical analyzer and commercially available antisera. The turbidimetric method was compared with the single radial immunodiffusion method. Linearity for serum apolipoprotein assay by the automated turbidimetric method was better than by the single immuno-diffusion method. The linearity by the turbidimetric method was 2.5 G/L for AI, 1.0 G/L for AII, 4.5 G/L for B, 0.12 G/L for CII, 0.3 G/L for CIII, and 0.12 G/L for E. The presence of high concentrations of bilirubin (up to 0.15 G/L) and hemoglobin (up to 50 G/L) interfered with apolipoprotein measurement. Comparison of the immunoturbidimetric and the single radial immunodiffusion (SRID) methods showed excellent coefficients of correlation, r = 0.963, 0.896, 0.846, 0.936, 0.972, and 0.937 for apo AI, AII, B, CII, CIII, and E, respectively. Reference ranges for the six apolipoproteins were determined by using sera from 450 healthy subjects and were 1.4 +/- 0.3 G/L for AI and 0.3 +/- 0.01 for E. The observed levels of AII (P less than 0.001), B (P less than 0.01), and CIII (P less than 0.01) were significantly higher in males. The serum levels of apo B, CII, and E showed a gradual increase with age which was more prominent in females than in males. The levels of apo AI, AII decreased significantly over an 11 day period in 22 patients with myocardial infarction.     

Pediatric reference intervals for lipids and apolipoproteins on the VITROS 5,1 FS Chemistry System

OBJECTIVES: Lipid biomarkers are integral in the assessment of dyslipidemia and cardiovascular risk, conditions that have become increasingly prevalent in pediatric populations. A comprehensive set of pediatric reference intervals for traditional or recently established lipid analytes is not currently available. DESIGN AND METHODS: 525 outpatient samples from a pediatric population were categorized into five age groups ranging from 0 to 20 years of age. Groups were further partitioned by gender. Serum or plasma samples were analyzed on the VITROS 5,1 FS Chemistry System for cholesterol and triglycerides by dry-film methods, direct HDL-C and LDL-C by selective detergent elimination, and apolipoproteins AI and B by immunoturbidimetry. Reference intervals were established by non-parametric methods at the 2.5th and 97.5th percentiles. RESULTS: Lipid levels show age- and gender-related differences, particularly during the first year of life and in young adults following puberty. Concentrations of total cholesterol, LDL-C, and apo B were lowest in the 12 months after birth and remained relatively constant throughout childhood, but decreased for males in early adulthood. Triglyceride levels increased gradually throughout childhood and adolescence, and along with cholesterol, the upper limits of these intervals exceeded the recommended concentrations of lipid levels in children. For HDL-C and apo AI, no age- or sex-related differences were found until after puberty when values for males decreased slightly. CONCLUSIONS: Our current reference intervals in children and adolescents provide an important update for lipid markers and suggest earlier incidence of hypercholesterolemia when compared to previous ranges. Increased profiling of lipids is anticipated, and these will aid in the early assessment of cardiovascular risk in pediatric populations.     

Polymorphisms at the 5'-end of the apolipoprotein AI gene and severity of coronary artery disease.

Elevated HDL-cholesterol (C) and apo AI are associated with decreased coronary artery disease (CAD) risk. We determined distributions of two MspI polymorphisms of the apo AI gene, associated in other studies with increased HDL-C, among 644 patients aged < or = 65 years in relation to circulating lipids and CAD severity assessed angiographically. The rare allele distributions at both sites were in Hardy-Weinberg equilibrium in these patients but the base changes were not associated with HDL-C and apo AI levels. However, patients homozygous for the -75 bp substitution were more likely to have one or more significantly diseased vessels (> 50% luminal obstruction)(OR: 4.75, 95%CI: 1.10- 20.46) as also were patients with the rare +83 bp alleles (OR: 2.56, 95%CI: 1.13-5.81). While there was an additive effect of the two polymorphisms to have severe CAD (OR: 6.33, 95%CI: 1.33-30.02), the polymorphism at +83 bp remained significant in predicting CAD severity after adjusting for other variables in a logistic regression analysis (OR: 2.95, 95%CI: 1.26-6.90), which was also strongly associated with the positive family CAD history (P = 0.009). We conclude that patients with these base changes in this Australian coronary population do not have increased HDL-C and apo AI levels but do have more severe CAD.     

Immunoturbidimetric method for routine determinations of apolipoproteins A-I and B

A simple immunoturbidimetric method for quantifying apolipoproteins (apo) A-I and B in serum or plasma is described. A special reagent formulation, including large amounts of suitable detergents, obviates the need for a sample blank even with grossly lipemic specimens. The assay is rapid, easily automated, and thus convenient for routine work. For both apo A-I and apo B, the assay range is about 0.2-3.5 g/L. The performance characteristics were assessed with discrete (Optimate and Olli CD) and centrifugal analyzers (Cobas Fara and IL Monarch 2000). Average analytical recovery was 101.5% for apo A-I and 99.4% for apo B. Dilution tests showed found/expected ratios of 101.2% (apo A-I) and 101.0% (apo B). Overall precision (CV) ranged from 1.4% to 3.3% for apo A-I and from 1.1% to 8.3% for apo B. Comparisons with commercially available rate nephelometry, radial immunodiffusion, and immunoturbidimetric methods gave good correlations (r greater than or equal to 0.938). Using the immunoturbidimetric method, we also established the relationships between apolipoproteins and lipids and determined the reference intervals. We conclude that the proposed method is suitable for routine use in clinical laboratories.     

Enzyme-linked immunosorbent assay to measure apolipoproteins AI and B secreted by a human hepatic carcinoma cell line (Hep G2).

We describe an enzyme-linked immunosorbent assay (ELISA) to measure apolipoproteins AI and B secreted by Hep G2 cells and in cell homogenates. These assays utilize commercially available polyclonal antibodies, affinity-purified to improve their specificity, thereby achieving a dramatic increase in the sensitivity of the assay. These affinity-purified antibodies were also more sensitive than a series of monoclonal antibodies tested. We achieved a sensitivity of 0.4 ng in the apo AI assay, and a sensitivity of 5 ng in the apo B assay. By these methods, we measured secretion rates by Hep G2 cells of 358 +/- 41 ng/mg cell protein/hr for apo B and 137 +/- 8 ng/mg cell protein/hr for apo AI. These assays also allowed the measurement of intracellular apolipoproteins and thus can be used to facilitate investigations of human lipoprotein metabolism in cell culture systems.     

血清apo A1、apo B、apo E与特应性皮炎的相关性

目的探讨血清载脂蛋白(apo)A1、apo B、apo E水平与特应性皮炎(AD)的相关性。方法选取230例AD患者及100名无过敏性疾病史的体检健康者,检测血清apo A1、apo B、apo E、总免疫球蛋白E(IgE)及过敏原[特异性免疫球蛋白E(sIgE)]水平。采用Logistic回归分析评估各项目及其与AD发病风险的关系。结果AD组和正常对照组女性血清apo A1水平均高于男性(P<0.05)。AD组女性血清apo A1水平高于正常对照组女性(P<0.05)。AD组男性血清apo B水平低于正常对照组男性(P<0.01)。sIgE阳性者男女比例与总IgE>200 kU/L者比较,差异无统计学意义(P>0.05)。以7岁为年龄分组界限,≤7岁组与>7岁组sIgE阳性率和总IgE>200 kU/L比例差异均无统计学意义(P>0.05)。apo A1、apo B与apo E均呈正相关(r值分别为0.190、0.287,P<0.01)。apo A1与apo B无相关性(r=0.030,P>0.05)。apo A1、apo B和apo E均与年龄呈正相关(r值分别为0.191、0.417、0.143,P<0.01)。Logistic回归分析结果显示,apo A1可增加AD的发病风险[比值比(OR)=7.346,95%可信区间(CI)为2.349~22.977],但与总IgE和sIgE无关;apo B校正年龄、性别后与AD发病风险无关,但可增加总IgE>200 kU/L的风险(OR=6.694,95%CI为1.739~25.762)。apo E无论是否校正年龄和性别均与AD发病风险、总IgE和sIgE无关。结论apo A1可能是独立于IgE介导的参与AD发病的危险因子。apo B受年龄、性别等因素的影响较大,与AD发病风险的关系尚不明确。apo E与AD的发病风险无关。     

应用UPLC-MS/MS 检测上海市成年健康人群血清维生素B6 和B7 参考区间的建立

目的 建立上海市健康成年人血清维生素B6 和B7 的参考区间。方法 选取2021 年9 月~2022 年2 月上海市1 059 例健康成年体检者作为研究对象,其中男性580 例﹑女性479 例,年龄≥ 18 岁。将受试者按年龄分青年组(18~44 岁)、中年组(45~59 岁)和老年组(≥ 60 岁),采用超高效液相色谱- 串联质谱法(UPLC-MS/MS)建立维生素B6 和B7的参考区间,并对参考区间进行验证。结果 剔除12 例离群值后,1 047 例成年人血清维生素B6 和B7 水平均呈非正态分布(均P<0.01);血清维生素B6[3.64(2.76~5.15)ng/ml vs 3.34(2.55~4.59)ng/ml] 和B7[0.22(0.14~0.34)ng/ml vs 0.20(0.13~0.33)ng/ml] 水平在不同性别组间的差异有统计学意义(Z=2.624,P<0.01;Z=2.196,P<0.05),且男性均高于女性;3 个年龄组间血清维生素B6 和B7 水平的差异均无统计学意义(Z=0.345~1.817,均P>0.05)。因此,合并年龄组后,维生素B6 参考区间为男性1.50~19.78 ng/mL,女性1.47~20.31 ng/ml;维生素B7 参考区间为男性0.03~1.25 ng/ml,女性0.03~1.41 ng/ml。对建立的参考区间进行验证,结果均符合标准。结论 初步建立了上海市成年人血清维生素B6和B7 参考区间,各地区实验室应按不同年龄和不同性别建立适合本地区的参考区间。     

DNA polymorphisms of apolipoprotein B and AI/CIII genes and response to gemfibrozil treatment.

Apolipoprotein B XbaI polymorphism and apolipoprotein AI/CIII SstI polymorphism have been found to be associated with variations in serum lipoprotein levels. We investigated whether these gene polymorphisms are involved in determining the lipid-modulating action of gemfibrozil. Of the 221 male subjects with hyperlipidemia studied, 121 responded well to the treatment with more than a 25% reduction in the non-high-density lipoprotein cholesterol level, whereas 100 were nonresponders. Among responders, but not nonresponders, homozygosity for the apolipoprotein B X2 allele (XbaI site present) and heterozygosity for the apolipoprotein AI/CIII S2 allele (SstI site present) were associated with elevated baseline serum low-density lipoprotein cholesterol and triglyceride levels, respectively. However, the hypolipidemic effect of gemfibrozil among the responders was independent of these gene polymorphisms. These data indicate that common polymorphisms of the apolipoprotein B and apolipoprotein AI/CIII gene loci influence serum lipid levels by mechanisms that are amenable to an intervention with gemfibrozil.     

Different VLDL apo B, and HDL apo AI and apo AII metabolism in two heterozygous carriers of unrelated mutations in the lipoprotein lipase gene

BACKGROUND: Lipoprotein lipase (LPL) deficiency has been suggested as a cause of low HDL-cholesterol (HDL-C) plasma levels, by a mechanism that involves an enhanced catabolism of HDL apolipoprotein (apo) AI. To verify the role of 2 different LPL gene mutations on HDL metabolism, we studied the in vivo turnover of the apo AI and apo AII in heterozygous carriers of LPL deficiency. METHODS: Apo AI and AII kinetics were studied by a 10-h primed constant infusion of 5,5,5-2H3-leucine approach in 2 carriers, 1 man (patient 1) and 1 woman (patient 2), and 5 control subjects. The rates of HDL apolipoproteins production (PR) and catabolism (FCR) were estimated using a one-compartment model-based analysis. RESULTS: Both carriers had low HDL-C plasma levels and only patient 1 was hypertriglyceridemic. VLDL apo B was 4-times slower in patient 1 as compared to patient 2. The FCRs of apo AI in both carriers was within the range of the controls (0.200, 0.221 and 0.211+/-0.051 day(-1), respectively). Apo AII FCR in patient 1 was about 20% lower than the mean of the control group whereas being normal in patient 2. Apo AI PR in patient 1 (9.20 mg kg(-1) day(-1)) was below the lowest value in controls (range, 10.52-13.24 mg kg(-1) day(-1)) whereas in patient 2 it was normal. Apo AII PR in both patients was similar to controls. CONCLUSION: The heterozygous carriers of 2 different mutations in the LPL gene had different VLDL apo B FCR, and from normal to slightly low HDL apolipoprotein FCR and PR. These results disagree with the putative enhanced apo AI FCR in LPL deficient patients and suggest the need to reconsider the effects of LPL activity on HDL metabolism.     

Elevated lipoprotein (a) and apolipoprotein B to AI ratio in South Asian patients with ischaemic stroke

BACKGROUND: Stroke is a continuing cause of excess cardiovascular disease (CVD) mortality amongst migrants from the Indian subcontinent (South Asians) living in Britain. However, little is known about the dyslipidaemia associated with stroke in South Asians. In particular, the highly atherogenic lipoprotein (a) [Lp(a)] and high apolipoprotein (Apo) B to AI ratio are emerging risk factors for CVD. METHODS: Using a case-control study, we investigated features of the dyslipidaemia in South Asian patients with stroke compared with South Asian subjects with no history of clinically detectable stroke. We studied 55 consecutive South Asian patients with ischaemic stroke (confirmed on computerised scan of the brain) and 85 controls. RESULTS: The stroke patients were significantly older than controls (65.2 vs. 59.8 years, p = 0.001), but were similarly matched for male gender (63.6 vs. 61.2%), smoking habit (20.7 vs. 18.1%) and presence of type 2 diabetes (25.5 vs. 19.3%). There were no differences between serum total cholesterol (p = 0.07) and high-density lipoprotein cholesterol (p = 0.08) between the groups, but stroke patients had higher serum triglycerides (p = 0.005). Mean [95% confidence interval (CI)] Apo B to AI ratio was higher amongst stroke patients [1.0 (0.9-1.0) vs. 0.7 (0.7-0.75), p < 0.001]. Similarly, geometric mean serum Lp(a) was significantly higher (p = 0.037) in stroke patients [19.9 mg/dl (14.0-28.5)] vs. controls [15.1 mg/dl (11.4-20.1)]. On logistic regression, stroke was independently associated with age and Apo B to AI ratio (p < 0.01). CONCLUSION: The present study suggests that Lp(a) and the Apo B to AI ratio are associated with ischaemic stroke in South Asians. A prospective analysis is needed to elucidate the role of Lp(a), Apo B and AI as risk factors for ischaemic stroke in this population, as well as the effects of intervention.