UBC Faculty Research and Publications

Clinical usefulness of lipid ratios to identify men and women with metabolic syndrome: a cross-sectional… Gasevic, Danijela; Frohlich, Jiri; Mancini, GB J; Lear, Scott A Oct 10, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12944_2014_Article_1141.pdf [ 628.41kB ]
Metadata
JSON: 52383-1.0223727.json
JSON-LD: 52383-1.0223727-ld.json
RDF/XML (Pretty): 52383-1.0223727-rdf.xml
RDF/JSON: 52383-1.0223727-rdf.json
Turtle: 52383-1.0223727-turtle.txt
N-Triples: 52383-1.0223727-rdf-ntriples.txt
Original Record: 52383-1.0223727-source.json
Full Text
52383-1.0223727-fulltext.txt
Citation
52383-1.0223727.ris

Full Text

RESEARCH Open AccessClinical usefulness of lipid ratios to identify menand women with metabolic syndrome:Keywords: Metabolic syndrome, Lipid ratios, Triglyceride-to-high-density-lipoprotein-cholesterol, Low-density-Gasevic et al. Lipids in Health and Disease 2014, 13:159http://www.lipidworld.com/content/13/1/159CanadaFull list of author information is available at the end of the articlelipoprotein-cholesterol-to-high-density-lipoprotein-cholesterol, Non-high-density-lipoprotein-cholesterol-to-high-density-lipoprotein-cholesterol, and total cholesterol-to-high-density-lipoprotein-cholesterol* Correspondence: dga4@sfu.ca1Department of Biomedical Physiology and Kinesiology, Simon FraserUniversity, 2600-515 West Hastings, Vancouver, British Columbia V6B 5K3,a cross-sectional studyDanijela Gasevic1*, Jiri Frohlich2,3, GB John Mancini4 and Scott A Lear5,3AbstractBackground: Waist circumference, a metabolic syndrome (MetSy) criterion, is not routinely measured in clinicalpractice making early identification of individuals with MetSy challenging. It has been argued that ratios ofcommonly measured parameters such as lipids and lipoproteins may be an acceptable alternative for identifyingindividuals with MetSy. The objective of our study was to explore clinical utility of lipid ratios to identify men andwomen with MetSy; and to explore the association between lipid ratios and the number of MetSy components.Methods: Men and women (N = 797) of Aboriginal, Chinese, European, and South Asian origin (35–60 years),recruited across ranges of body mass index (BMI), with no diagnosed cardiovascular disease (CVD) or onmedications to treat CVD risk factors were assessed for anthropometrics, family history of CVD, MetSy components(waist circumference, blood pressure, glucose, triglycerides (TG), high-density-lipoprotein-cholesterol (HDL-C)),low-density-lipoprotein-cholesterol (LDL-C), nonHDL-C, and health-related behaviours.Results: Mean levels of lipid ratios significantly increased with increasing number of MetSy components in menand women (p < 0.05). After adjustment for age, ethnicity, smoking, alcohol consumption, physical activity, familyhistory of CVD and BMI, (and menopausal status in women), all lipid ratios were associated with the number ofMetSy components in men and women (Poisson regression, p < 0.001). Compared to the rest of the lipid ratios(ROC curve analysis), TG/HDL-C was best able to discriminate between individuals with and without MetSy(AUC = 0.869 (95% CI: 0.830, 0.908) men; AUC = 0.872 (95% CI: 0.832, 0.912) women). The discriminatory power ofTC/HDL-C and nonHDL-C/HDL-C to identify individuals with MetSY was the same (for both ratios, AUC = 0.793(95% CI: 0.744, 0.842) men; 0.818 (95% CI: 0.772, 0.864) women). Additionally, LDL-C/HDL-C was a good marker forwomen (AUC = 0.759 (95% CI: 0.706, 0.812)), but not for men (AUC = 0.689 (95% CI: 0.631, 0.748)). Based on amultiethnic sample, we identified TG/HDL-C cut-off values of 1.62 in men and 1.18 in women that were best ableto discriminate between men and women with and without MetSY.Conclusions: Our results indicate that TG/HDL-C is a superior marker to identify men and women with MetSycompared to TC/HDL-C, LDL-C/HDL-C, and nonHDL-C/HDL-C.© 2014 Gasevic et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.Gasevic et al. Lipids in Health and Disease 2014, 13:159 Page 2 of 10http://www.lipidworld.com/content/13/1/159BackgroundMetabolic syndrome is a cluster of metabolic abnormalitiesassociated with type 2 diabetes [1,2], cardiovascular mor-bidity and mortality, and all-cause mortality [3,4]. It isalarming that the prevalence of metabolic syndrome is highand on the rise in both developed and developing countries[5,6]. Early identification and treatment of individuals withmetabolic syndrome is imperative to prevent debilitatingconsequences associated with its development. Regardlessof the approach for diagnosing metabolic syndrome, cen-tral obesity, as measured by waist circumference, is one ofthe main criteria for diagnosing metabolic syndrome[7-10]. However, waist circumference is not routinely mea-sured in primary care [11], which makes early identificationof individuals with metabolic syndrome challenging.It has been argued that ratios of commonly measuredparameters such as lipids and lipoproteins may be anacceptable alternative for identifying individuals withmetabolic syndrome [12-17]. Evidence shows that lipidratios perform better than individual lipids in predictingcardiovascular risk [18-21]. Identifying a ratio to serve asa quick and simple tool for identifying individuals at in-creased cardiometabolic risk may decrease complexity ofand increase efficiency in identifying and monitoringthose at risk; especially as electronic medical records be-come more commonplace or if useful biochemical ratioswere to be reported in routine laboratory test results.However, research on clinical usefulness of lipid ratios toidentify individuals with metabolic syndrome is scarceand mostly limited to specific population groups such asGhanian [17], Korean [14,16], Spanish [12], and Turkish[13]. Furthermore, some of the available studies researchedlimited number of ratios [13,17] or provided no cut-offsfor lipid ratios to help guide physicians in identifyingindividuals at cardiometabolic risk [14,15]. Therefore,the objective of this study was to explore the clinicalusefulness of lipid ratios to identify men and womenwith metabolic syndrome. Lipid ratios researchedinclude: total cholesterol-to-high-density-lipoprotein-choles-terol (TC/HDL-C), triglyceride-to-high-density-lipoprotein-cholesterol (TG/HDL-C), low-density-lipoprotein-cholesterol-to-high-density-lipoprotein-cholesterol (LDL-C/HDL-C), andnon-high-density-lipoprotein-cholesterol-to-high-density-lipoprotein-cholesterol (nonHDL-C/HDL-C). In addition,given that the number of metabolic syndrome componentshas been shown to predict incident cardiovascular disease(CVD) and type 2 diabetes [22], we additionally exploredthe association of lipid ratios with the number of metabolicsyndrome components among men and women comingfrom the ethnically diverse population.ResultsParticipants were equally represented across sexes andethnicity. Compared to women, men had lower levels ofHDL-C and significantly higher levels of total cholesterol,LDL-C, nonHDL-C, triglycerides, lipid ratios, glucose,blood pressure, and waist circumference; also, alcoholconsumption was more prevalent among men than amongwomen (Table 1). Similarly, the average number of meta-bolic syndrome components and prevalence of metabolicsyndrome were higher in men than in women.Mean levels of lipid ratios were on average higher inmen than in women and significantly increased with in-creasing number of metabolic syndrome components inboth men and women (Figure 1, Additional file 1: TableS1). In both men and women, mean levels of lipid ratiosin individuals with zero metabolic syndrome componentswere significantly lower from their counterparts with 1 ormore metabolic syndrome components (p < 0.05). In menand women, except for LDL-C/HDL-C ratio, there was asignificant difference between mean levels of lipid ratiosacross number of metabolic syndrome components(p < 0.05). However, in both men and women, no signifi-cant difference was found between mean levels of LDL-C/HDL-C of participants with 3 and those with 4/5 meta-bolic syndrome components (p > 0.05). Furthermore, theresults of partial correlation analyses (age and ethnicityadjusted) reveal that in both men and women, comparedto the rest of the lipid ratios, TG/HDL-C showed strongercorrelations with waist circumference, blood pressure andblood glucose (Additional file 2: Table S2). In addition, theresults of Poisson regression analyses indicate that, afteradjustment for age, ethnicity, smoking, alcohol consump-tion, physical activity, family history of CVD and bodymass index (BMI) (and menopausal status in women), alllipid ratios were significantly associated with the numberof metabolic syndrome components in both men andwomen (Table 2).Using ROC curve analyses, we plotted sensitivity over1-specificity for each of the ratios in both men andwomen (Figure 2). Compared to the rest of the lipid ra-tios, TG/HDL-C was best able to discriminate betweenapparently healthy men and women with and withoutmetabolic syndrome (AUC = 0.869 (95% CI: 0.830, 0.908)for men; AUC= 0.872 (95% CI: 0.832, 0.912) for women).The discriminatory power of TC/HDL-C and nonHDL-C/HDL-C to identify individuals with metabolic syndromewas the same (for both ratios, AUC= 0.793 (95% CI:0.744, 0.842) for men; 0.818 (95% CI: 0.772, 0.864) forwomen). In addition, based on Hosmer and Lemeshow’scriteria [23] LDL-C/HDL-C was an “acceptable” marker todiscriminate between women with and without metabolicsyndrome (AUC = 0.759 (95% CI: 0.706, 0.812)), howeverthis was not the case for men (AUC = 0.689 (95% CI:0.631, 0.748)). In both men and women, the analysis thattested the statistical significance of the differences be-tween areas under the curve revealed significant differencebetween areas under the curves for all the pairs of lipid)))47)Gasevic et al. Lipids in Health and Disease 2014, 13:159 Page 3 of 10http://www.lipidworld.com/content/13/1/159Table 1 Distribution of risk factors in men and womenMen 380Age (years) 46.8 ± 8.7EthnicityAboriginal 81 (21.3%)Chinese 100 (26.3%European 97 (25.5%)South Asian 102 (26.8%Current smokers 44 (11.6%)Current consumers of alcohol 128 (33.7%Physical activity (min/week) 226 (100, 4Family history of cardiovascular disease 168 (44.2%Waist circumference (cm) 92.6 ± 11.2ratios (Figure 2). The exception is the pair of TC/HDL-Cand nonHDL-C/HDL-C for which AUCs were the same.Lipid ratio cut-off values, with their respective sensitivitiesand specificities, for identifying individuals with metabolicsyndrome are provided in Table 3.DiscussionThe purpose of this study was to explore the clinical utilityof lipid ratios to identify men and women with metabolicsyndrome. Additionally we explored the association bet-ween lipid ratios and number of metabolic syndromecomponents. Our results indicate that increases in lipidratios are significantly associated with increase in thenumber of metabolic syndrome components in both menand women after adjusting for age, ethnicity, smoking sta-tus, alcohol consumption, physical activity, family historyof CVD, and BMI (and menopausal status in women).Body mass index (kg/m2) 27.6 ± 4.3TC (mmol/L) 5.25 ± 0.97HDL-C (mmol/L) 1.13 ± 0.29LDL-C (mmol/L) 3.30 ± 0.85TC/HDL-C 4.91 ± 1.46LDL/HDL-C 3.07 ± 1.01nonHDL-C 4.12 ± 0.99nonHDL-C/HDL-C 3.81 (2.93, 4.6TG 1.48 (0.99, 2.2TG/HDL-C 1.34 (0.82, 2.2Systolic BP (mmHg) 117 (110, 124Diastolic BP (mmHg) 79.2 ± 9.7Glucose (mmol/L) 5.30 (5.00, 5.6With MetSy 119 (31.3%)Number of MetSy components 1.84 ± 1.40Categorical variables presented as n (%), and sex differences were explored using Cdistributed or median (25%, 75%) if skewed. Sex differences explored using t-test aTC – total cholesterol, HDL-C – high-density lipoprotein cholesterol, LDL-C – low-densitTG/HDL-C – ratio of TG and HDL-C, nonHDL-C/HDL – ratio of nonHDL-C and HDL-C, LDWomen 417 p value47.5 ± 8.9 0.2410.70498 (23.5%)118 (28.3%)99 (23.7%)102 (24.5%)35 (8.4%) 0.13398 (23.5%) 0.001) 208 (95, 424) 0.269187 (44.8%) 0.85785.1 ± 12.2 <0.001Compared to TC/HDL-C, LDL-C/HDL-C, and nonHDL-C/HDL-C, TG/HDL-C was shown to be a better clinicalmarker to discriminate between individuals with and with-out metabolic syndrome. These results were consistent formen and women.The results of our study are in line with those basedon Spanish, Korean, and Japanese populations where in-creases in lipid ratios were shown to be associated withincrease in the number of metabolic syndrome compo-nents [12,14,15]. Our study extends the previous fin-dings by reporting the positive association between lipidratios and the number of metabolic syndrome compo-nents in a multiethnic population; this relationship per-sisted after adjusting for ethnicity and other factorsknown to be associated with metabolic syndrome.The observed strong relationship between lipid ratiosand the number of metabolic syndrome components27.3 ± 5.3 0.4985.23 ± 1.03 0.7761.43 ± 0.35 <0.0013.17 ± 0.92 0.0463.86 ± 1.23 <0.0012.35 ± 0.91 <0.0013.80 ± 1.05 <0.0013) 2.68 (2.01, 3.45) <0.0012) 1.17 (0.83, 1.63) <0.0014) 0.86 (0.54, 1.22) <0.001) 115 (106, 126) 0.02575.5 ± 9.2 <0.0010) 5.10 (4.80, 5.40) <0.001104 (24.9%) 0.0451.63 ± 1.22 0.026hi-square test. Continuous variables presented as mean ± SD if normallynd Mann U Whitney test for normally distributed and skewed data, respectively.y lipoprotein cholesterol, TC/HDL-C – ratio of TC and HDL-C, TG – triglycerides,L-C/HDL-C – ratio of LDL-C and HDL-C, MetSy – metabolic syndrome.ABCDFigure 1 Mean levels of lipid ratios across number of metabolicsyndrome components in men and women (1A, 1B, 1C and 1D).All means adjusted for age and ethnicity and presented as mean (95%CI). *geometric means. Bonferroni-corrected pairwise comparisons(0 vs. 1; 0 vs. 2; 0 vs. 3; 0 vs. ≥4; 1 vs. 2; 1 vs. 3; 1 vs. ≥4; 2 vs. 3; 2 vs. ≥4;and 3 vs. ≥4 ): Men: Except for the pair 3 vs. ≥4 for LDL-C/HDL-C, allother pairwise comparisons of means of TC/HDL-C, TG/HDL-C,LDL-C/HDL-C and nonHDL-C/HDL-C across the number of metabolicsyndrome components were significant at p < 0.05. Women: Except forthe pair 3 vs. ≥4, all other pairwise comparisons of means of TC/HDL-C,TG/HDL-C, LDL-C/HDL-C and nonHDL-C/HDL-C across the number ofmetabolic syndrome components were significant at p < 0.05.TG/HDL-C: triglyceride-to-high-density-lipoprotein-cholesterol, TC/HDL-C:total cholesterol-to-high-density-lipoprotein-cholesterol, LDL-C/HDL-C:low-density-lipoprotein-cholesterol-to-high-density-lipoprotein-cholesterol, nonHDL-C/HDL-C: non-high-density-lipoprotein-cholesterol-Gasevic et al. Lipids in Health and Disease 2014, 13:159 Page 4 of 10http://www.lipidworld.com/content/13/1/159may suggest lipid ratios as potential simple tools for earlyidentification of individuals with constellation of cardio-metabolic abnormalities. However, the actual lipid ratiocut-offs to guide clinicians in identifying individuals withmetabolic syndrome are rarely reported. The studies avail-able looked at a limited number of lipid ratios and per-tained to specific groups such as Spanish population [12]or a population of Ghanian women [17]. We extend theavailable findings by reporting lipid ratio cut-off values foridentifying men and women with metabolic syndrome.According to our results, in both men and women,TG/HDL-C was shown to be a better marker to identifyindividuals with metabolic syndrome compared to TC/HDL-C, LDL-C/HDL-C, and nonHDL-C/HDL-C. Basedon a multiethnic sample, we identified TG/HDL-C cut-offvalues of 1.62 in men and 1.18 in women that were bestable to discriminate between men and women with andwithout metabolic syndrome. These numbers were com-parable but slightly lower than those reported by Corderoet al. [12] who clinically assessed a large working popula-to-high-density-lipoprotein-cholesterol.tion of Spanish men and women and reported TG/HDL-Table 2 The association between lipid ratios and thenumber of metabolic syndrome componentsMen WomenExp (B) (95% CI) Exp (B) (95% CI)TC/HDL-C 1.262 (1.197, 1.330) 1.278 (1.204, 1.355)ln_TG/HDL-C 1.875 (1.680, 2.092) 1.797 (1.603, 2.016)LDL-C/HDL-C 1.236 (1.147, 1.332) 1.358 (1.243, 1.483)ln_nonHDL-C/HDL-C 2.861 (2.258, 3.626) 2.633 (2.109, 3.286)All Poisson regression models adjusted for age, ethnicity, smoking, alcoholconsumption, physical activity, family history of cardiovascular disease, andBMI. Models for women additionally adjusted for menopause status. Allmodels significant at p < 0.001. TC/HDL-C – ratio of total cholesterol andhigh-density lipoprotein cholesterol, TG/HDL-C – ratio of triglycerides andhigh-density lipoprotein cholesterol, nonHDL-C/HDL – ratio of non-high-densitylipoprotein cholesterol and high-density lipoprotein cholesterol, LDL-C/HDL-C –ratio of low-density lipoprotein cholesterol and high-density lipoproteincholesterol.Aa,b, c,e, fBa,b, c,e, fFigure 2 Receiver operating characteristic curves for evaluating the usefulness of lipid ratios to identify men (A) and women (B) withmetabolic syndrome. The diagonal line indicates a test with an area under the receiver operating characteristic curve of 0.5. The areas underthe curve for TC/HDL-C and nonHDL-C/HDL-C overlapped, as the discriminatory power of TC/HDL-C and nonHDL-C/HDL-C to identify individualswith metabolic syndrome was the same. The difference between areas under the curve (p ≤ 0.01): aTG/HDL-C vs. TC/HDL-C. bTG/HDL-C vs.nonHDL-C/HDL-C. cTG/HDL-C vs. LDL-C/HDL-C. dTC/HDL-C vs. nonHDL/HDL-C. eTC./HDL-C vs. LDL-C/HDL-C. fnonHDL-C/HDL-C vs. LDL-C/HDL-C.TG/HDL-C: triglyceride-to-high-density-lipoprotein-cholesterol, TC/HDL-C: total cholesterol-to-high-density-lipoprotein-cholesterol, LDL-C/HDL-C:low-density-lipoprotein-cholesterol-to-high-density-lipoprotein-cholesterol, nonHDL-C/HDL-C: non-high-density-lipoprotein-cholesterol-to-high-density-lipoprotein-cholesterol.Table 3 Results of receiver-operating curve analyses featuring the thresholds of lipid ratios with their respectivesensitivities and specificities for diagnosing metabolic syndrome in men and womenMen WomenLipid ratio Criterion value (Sensitivity, specificity) Criterion value (Sensitivity, specificity)TG/HDL-C 1.62 (84.0%, 80.1%) 1.18 (70.2%, 88.2%)TC/HDL-C 5.05 (73.1%, 71.6%) 3.91 (81.7%, 68.7%)LDL-C/HDL-C* 3.79 (40.2%, 88.5%) 2.53 (67.6%, 74.8%)nonHDL-C/HDL-C 4.05 (73.1%, 71.6%) 2.91 (79.8%, 70.9%)*According to the AUC, LDL-C/HDL-C showed to be a good marker of metabolic syndrome in women but not in men.Gasevic et al. Lipids in Health and Disease 2014, 13:159 Page 5 of 10http://www.lipidworld.com/content/13/1/159Gasevic et al. Lipids in Health and Disease 2014, 13:159 Page 6 of 10http://www.lipidworld.com/content/13/1/159C values of 2.75 and 1.65 as cut-offs for identifying menand women with metabolic syndrome, respectively. Theobserved differences in cut-offs reported in the abovementioned study and the ones reported in this study aremost likely due to a difference in population groupsstudied.It has recently been reported that TG/HDL-C predictsthe development of CVD as effectively as the diagnosedmetabolic syndrome [24]. Similarly, TG/HDL-C wasfound to predict coronary heart disease and CVD mor-tality as well as or even better than does metabolicsyndrome in men [25]. However, the detection of an ele-vated TG/HDL-C is not meant to replace metabolic syn-drome diagnosis in clinical practice; and it should ratherbe considered as a simple tool to quickly identify pa-tients at increased cardiometabolic risk for whom fur-ther risk evaluation and clinical intervention are needed[25]. Indeed, it has recently been reported in the studyin young adults that, compared to metabolic syndromediagnostic, TG/HDL-C may be able to identify a greaternumber of individuals at risk; however a use of a meta-bolic syndrome diagnostics approach identifies individ-uals with an accentuated cardiometabolic risk profile[26]. While metabolic syndrome diagnosis may providemore comprehensive approach to identifying presentcardiometabolic risk factors in individuals, this diagnos-tics is challenged by the fact that waist circumference,one of the integrative components of metabolic syn-drome, is not commonly measured. Indeed, it has re-cently been reported that WC is routinely measured byonly 6% of primary care physicians [11]. If translated, itmay mean that more than 90% of primary care physi-cians would not be able to diagnose individuals withmetabolic syndrome given that one of its components isnot routinely measured. In contrast, both TG and HDL-Care routinely measured in clinical practice, and TG/HDL-Ccould be readily calculated (or provided by the laboratory)to serve as a quick tool to identify men and women atincreased cardiometabolic risk. This approach to useTG/HDL-C to identify individuals at risk for whom fur-ther care is needed may reduce time and complexity forinitial diagnosis of people with constellation of cardiomet-abolic abnormalities.Other research groups have also proposed TG/HDL-Cas a potential, simple tool to identify patients at in-creased risk for CVD. The evidence shows a strong as-sociation between TG/HDL-C and insulin resistance asmeasured by glucose clamp [27,28], modified insulinsuppression test [29], and homoeostasis model assess-ment index [29,30]. This association was also found tobe independent of waist circumference [31]. Further,others found TG/HDL-C to be an independent predictorof future type 2 diabetes mellitus [25,32] and its relatedmicrovascular complications [33]; coronary heart disease[34,35]; major cardiovascular events including overalldeath, myocardial infarction, and unstable angina thatrequired revascularization [36] and those including an-gina pectoris, myocardial infarction, myocardial revas-cularization, and fatal or nonfatal stroke [37]; and firstcoronary event irrespective of BMI [38]. The strong rela-tionship between TG/HDL-C and CVD may be foundedin the atherogenic properties of TG/HDL-C. It has beenreported that increase in TG/HDL-C is significantly as-sociated with decrease in LDL particle size and increasein fractional esterification rates of cholesterol in plasmadepleted of apoB-lipoproteins; hence the proposed namefor TG/HDL-C being atherogenic index of plasma [39].In light of available evidence, including the results of ourstudy, it seems reasonable to conclude that TG/HDL-Cmay serve as a useful marker to identify individuals atincreased risk of CVD.Several limitations to our study should be considered.This study is a cross-sectional design, thus longitudinalstudies are needed to explore whether the associationbetween TG/HDL-C and metabolic syndrome persists orchanges over time in men and women. Further, theM-CHAT participants were recruited across a range ofBMI, so this study may not be representative of thegeneral population. However, recruiting people acrossranges of BMI allows for the opportunity to explore theassociations between lipid ratios and metabolic syndromein the population with a range of body sizes. Also, the re-sults of our analyses indicate a significant association bet-ween lipid ratios and number of metabolic syndromecomponents independent of BMI. Moreover, our analyseswere adjusted for age, ethnicity, smoking, status, alcoholconsumption, physical activity, family history of CVD andBMI (and menopausal status in women) but not for dietknown to influence plasma TG levels. Of importance, TG/HDL-C cut-points for identifying individuals at risk mayvary across ethnicity and race [40,41]. Also, TG/HDL-Cwas shown not to be a reliable risk marker in individualsof South Asian [41] and African American origin [42,43].While statistical power did not allow us to additionallystratify our analyses by ethnicity, given the establishedethnic differences in the way body fat is accumulated[44,45] and ethnic differences in cardiometabolic risk[46,47], we believe that further research featuring ethnic-specific analyses is warranted.ConclusionsIn conclusion, our study shows strong positive associ-ation between lipid ratios and metabolic syndrome inapparently healthy men and women (without CVD andnot on CVD-related medications) drawn from an eth-nically diverse population. TG/HDL-C appeared to be asuperior marker compared to TC/HDL-C, LDL-C/HDL-C,and nonHDL-C/HDL-C, and it is useful for identifyingGasevic et al. Lipids in Health and Disease 2014, 13:159 Page 7 of 10http://www.lipidworld.com/content/13/1/159both men and women with metabolic syndrome. Namely,TG/HDL-C readings of 1.62 or greater in men and 1.18 orgreater in women can help primary care physicians easilyidentify individuals at increased risk for CVD. Early identi-fication of individuals at risk would allow for an early im-plementation of lifestyle and medication strategies, whileTG/HDL-C could be further used to evaluate the successof such strategies.MethodsApparently healthy men and women were recruited aspart of the Multi-Cultural Community Health AssessmentTrial (M-CHAT) designed to compare body fat distribu-tion of Aboriginal, Chinese, and South Asian populationsto that of a European population and explore how fatdistribution relates to CVD risk factors [48]. Details on re-cruitment of study participants have already been pub-lished [48]. In brief, eligible participants were apparentlyhealthy individuals between ages of 30 and 65 of Aborigi-nal, Chinese, European and South Asian origin residing inthe Vancouver mainland. To ensure a range of body fatmass across groups, participants were recruited from thefollowing BMI (kg/m2) ranges: 18.5-24.9 (low range), 25.0-29.9 (middle range), and 30 and over (upper range). Giventhat it was challenging to identify Aboriginal men with aBMI of less than 25, Aboriginal men of any BMI were re-cruited. Similarly, due to a difficulty in identifying Chineseindividuals with a BMI of 30 kg/m2 and higher, the targetfor the upper BMI range was changed to a BMI of 28 andhigher. Individuals with a fluctuation in weight that wasgreater than 2.5 kg three months prior to the assessmentdate, and those with diagnosed CVD or on medications totreat CVD risk factors, were not eligible for the study.Ethnicity was self-reported. All study participants pro-vided informed consent for the study, and the study wasapproved by the Simon Fraser University Research EthicsBoard.All participants were assessed for socio-demographicsand risk factors. Following the standard protocols forblood sample collection, shipment, and processing, fast-ing blood samples were collected from each participantat the research site and assessed for lipids and glucose atSt. Paul’s Hospital in Vancouver, Canada. Glucose, TC,TG, and HDL-C were determined using standard proto-cols by ADVIA 1650 analyzer (Bayer Health Care, LLC.,New Jersey, USA). Friedwald formula was used to calcu-late LDL-C [49]. Height and weight were measured, andBMI was calculated as weight in kilograms divided byheight in metres squared. Waist circumference repre-sented an average of two measurements taken upon ex-piration at the mid point between the low rib marginand iliac crest. Blood pressure was recorded as mean of5 measurements taken after 10 minutes seated rest inthe left arm using BpTRU model BPM-200 oscillometricoffice blood pressure monitor (VSM MedTech Ltd.,Coquitlam, British Columbia). Smoking status (currentsmokers/non-smokers), alcohol consumption (currentlyconsume alcohol/do not consume alcohol) and familyhistory of CVD (yes/no) were determined by self-report.Women self-reported their menopausal status by choos-ing from the following options (women were provided adefinition for each of the options): pre-menopausal(regular menstrual cycles), peri-menopausal (menopausaltransition with symptoms such as menstrual irregu-larity), post-menopausal (end of menstrual cycles, atleast 12 months since the last menstrual period), or hys-terectomy (surgical removal of the uterus). Menopausalstatus variable used as a confounder in regressionanalyses consisted of two categories, pre-menopausal(pre-menopausal and peri-menopausal) and menopausal(post-menopausal and hysterectomy). Physical activitywas measured using the Modifiable Physical Activityquestionnaire previously used in multi-ethnic popula-tions and determined by self-report [50,51]. A personwould qualify as having metabolic syndrome if he/shehad three out of 5 risk factors present (hypertension,hypertriglyceridaemia, lowered HDL-C, hyperglycaemia,and central obesity) using the most recent harmonizeddefinition of the metabolic syndrome where ethnic spe-cific thresholds for waist circumference were used [10].Statistical analysesOnly participants with complete data for all metabolic syn-drome components were included to the study (N = 797).Considering significant sex differences in body fat distribu-tion and cardiometablic risk [52-54], all statistical analyseswere performed separately for men and women. Cate-gorical variables were presented as counts and percent-ages, and differences between men and women wereexplored using Chi-square test. Continuous variables werepresented as means ± standard deviation (SD) if normallydistributed, or medians with 25th and 75th percentiles ifnot normally distributed. Differences between men andwomen in the distribution of continuous variables wereexplored using t-test and Mann U Whitney test for nor-mally distributed and skewed data, respectively.Age and ethnicity adjusted levels of lipid ratios acrossthe number of metabolic syndrome components weredetermined using general linear modelling (Bonferroni-corrected post hoc comparisons). In our apparentlyhealthy multiethnic study population, there was a limitednumber of participants with 4 (n = 66) and 5 (n = 18)metabolic syndrome components; therefore, we com-bined these two categories. Poisson regression analysiswas used to explore the association between lipid ratiosand the number of metabolic syndrome components. Allmodels were adjusted for age, ethnicity, smoking, alcoholconsumption, physical activity, family history of CVDbetween lipid measures and metabolic syndrome, insulin resistance andGasevic et al. Lipids in Health and Disease 2014, 13:159 Page 8 of 10http://www.lipidworld.com/content/13/1/159and BMI. Models for women were additionally adjustedfor menopausal status.The clinical utility of lipid ratios to identify individualswith metabolic syndrome was explored using receiver-operating characteristic curve (ROC) analysis. Plots ofsensitivity (true positives) versus 1 minus specificity (falsepositives) were constructed in both men and women foreach lipid ratio. Area under the curve (AUC) was calcu-lated to explore which lipid ratio showed highest accuracyin predicting metabolic syndrome. AUC is a measure ofdiscrimination, and AUC of 0.5, 0.6 ≤AUC< 7, 7 ≤AUC<0.8, 0.8 ≤AUC< 0.9, and ≥ 0.9 corresponds to no discri-mination, poor, acceptable, excellent, and outstandingdiscrimination, respectively [23]. Optimal cut-off point ofeach lipid ratio to identify individuals with metabolic syn-drome corresponded to a maximum value of Youdenindex that was calculated as sensitivity + specificity – 1[55,56]. Additionally, the statistical significance of the dif-ference between areas under the curves was tested usingthe methods of De Long et al. [57]. Analyses were per-formed using Statistical Package for Social Sciences (SPSS)version 19 except for the comparisons of ROC curves thatwere performed using MedCalc statistical software.P values of less than 0.05 were considered statisticallysignificant.Additional filesAdditional file 1: Table S1. Mean levels (95% CI) of lipid ratios acrossincreasing number of metabolic syndrome components.Additional file 2: Table S2. Partial correlations (age and ethnicityadjusted) between lipid ratios and metabolic syndrome components inmen (A) and women (B).AbbreviationsAUC: Area under the curve; BMI: Body mass index; CVD: Cardiovasculardisease; LDL-C/HDL-C: Low-density-lipoprotein-cholesterol-to-high-density-lipoprotein-cholesterol; M-CHAT: Multi-Cultural Health Assessment Trial;nonHDL-C/HDL-C: non-high-density-lipoprotein-cholesterol-to-high-density-lipoprotein-cholesterol; ROC: Receiver-operating characteristic curve;SD: Standard deviation; SPSS: Statistical Package for Social Sciences;TC/HDL-C: Total cholesterol-to-high-density-lipoprotein-cholesterol;TG/HDL-C: Triglyceride-to-high-density-lipoprotein-cholesterol.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionDG designed the study, performed statistical analyses and data interpretationand wrote the manuscript. GBJM participated in the design of the study,data interpretation, and helped draft the manuscript. JF critically assessed themanuscript and participated in the study discussion. SAL participated in thedesign of the study, data interpretation and study discussion. All authorsread and approved the final manuscript.Author details1Department of Biomedical Physiology and Kinesiology, Simon FraserUniversity, 2600-515 West Hastings, Vancouver, British Columbia V6B 5K3,2Canada. Department of Pathology and Laboratory Medicine, University ofBritish Columbia, Vancouver, BC V6T 2B5, Canada. 3Division of Cardiology,Providence Health Care, Healthy Heart Program, 1081 Burrard Street,adiponectin. Circ J 2010, 74:931–937.15. Kawamoto R, Tabara Y, Kohara K, Miki T, Takayama S, Abe M, Katoh T,Ohtsuka N: Relationships between lipid profiles and metabolic syndrome,insulin resistance and serum high molecular adiponectin in Japanesecommunity-dwelling adults. Lipids Health Dis 2011, 10:79.16. Kim SW, Jee JH, Kim HJ, Jin SM, Suh S, Bae JC, Kim SW, Chung JH, Min YK,Lee MS, Lee MK, Kim KW, Kim JH: Non-HDL-cholesterol/HDL-cholesterol isa better predictor of metabolic syndrome and insulin resistance thanVancouver, BC V6Z 1Y6, Canada. 4Department of Medicine, University ofBritish Columbia, 2775 Laurel Street, Vancouver, British Columbia V5Z 1M9,Canada. 5Faculty of Health Sciences, Simon Fraser University, Burnaby, BritishColumbia V5A 1S6, Canada.Received: 24 April 2014 Accepted: 28 September 2014Published: 10 October 2014References1. Ford ES, Li C, Sattar N: Metabolic syndrome and incident diabetes.Diabetes Care 2008, 31(9):1898–1904.2. Gupta AK, Prieto-Merino D, Dahlof B, Sever PS, Poulter NR, ASCOT Investigators:Metabolic syndrome, impaired fasting glucose and obesity, as predictorsof incident diabetes in 14120 hypertensive patients of ASCOT-BPLA:comparisons of their relative predictability using a novel approach.Diabetic Medicine 2011, 28(8):941–947.3. Galassi A, Reynolds K, He J: Metabolic syndrome and risk of cardiovasculardisease: a meta-analysis. Am J Med 2006, 119(10):812–819.4. Gami AS, Witt BJ, Howard DE, Erwin PJ, Gami LA, Somers VK, Montori VM:Metabolic syndrome and risk of incident cardiovascular events anddeath: a systematic review and meta-analysis of longitudinal studies.JACC 2007, 49(4):403–414.5. Mohan V, Deepa M: The metabolic syndrome in developing countries.Diabetes Voice 2006, 51:15–17.6. Ford ES, Giles WH, Mokdad AH: Increasing prevalence of the metabolicsyndrome among US adults. Diabetes Care 2004, 27:2444–2449.7. International Diabetes Federation: The IDF consensus worldwide definition ofthe metabolic syndrome. [https://www.idf.org/webdata/docs/MetS_def_update2006.pdf]8. Balkau B, Charles MA: Comment on the provisional report from the WHOconsultation. European Group for the Study of Insulin Resistance (EGIR).Diabetes Medicine 1999, 16(5):442–443.9. National Cholesterol Education Program (NCEP) Expert Panel on Detection,Evaluation, and Treatment of High Blood Cholesterol in Adults (AdultTreatment Panel III): Third report of the National Cholesterol EducationProgram (NCEP) Expert Panel on Detection, Evaluation, and Treatment ofHigh Blood Cholesterol in Adults (Adult Treatment Panel III) Final Report.Circulation 2002, 106(25):3143.10. Alberti KGMM, Eckel RH, Grundy SM, Grundy SM, Zimmet PZ, Cleeman JI,Donato KA, Fruchart J-C, James PT, Loria CM, Smith SC Jr: Harmonizing themetabolic syndrome. A joint interim statement of the InternationalDiabetes Federation Task Force on Epidemiology and Prevention;National Heart, Lung, and Blood Institute; American Heart Association;World Heart Federation; International Atherosclerosis Society; andInternational Association for the Study of Obesity. Circulation 2009,120:1640–1645.11. Gupta M, Singh N, Tsiqoulis M, Kaiij M, Hirjikaka S, Quan A, Teoh H, Verma S:Perceptions of Canadian primary care physicians towards cardiovascularrisk assessment and lipid management. Can J Cardiology 2012,28(1):14–19.12. Cordero A, Laclaustra M, Leon M, Casasnovas JA, Grima A, Luengo E,Ordonez B, Bergua C, Bes M, Pascual I, Alegria E, on behalf of the MESYASRegistry Investigators: Comparison of serum lipid values in subjects withand without the metabolic syndrome. Am J Cardiol 2008, 102:424–428.13. Onat A, Can G, Kaya H, Hergenc G: “Atherogenic index of plasma” (log10triglyceride/high-density lipoprotein-cholesterol) predicts high bloodpressure, diabetes, and vascular events. J Clin Lipid 2010, 4:89–98.14. Kimm H, Lee SW, Lee HS, Shim KW, Cho CY, Yun JE, Jee SH: Associationsapolipoprotein B/apolipoprotein A1. Int J Cardiol 2013, 168:2678–2683.17. Arthur FKN, Adu-Frimpong M, Osei-Yeboah J, Mensah FO, Owusu L:Prediction of metabolic syndrome among postmenopausal GhanianGasevic et al. Lipids in Health and Disease 2014, 13:159 Page 9 of 10http://www.lipidworld.com/content/13/1/159women using obesity and atherogenic markers. Lipids Health Dis 2012,11:101.18. Wang TD, Chen WJ, Chien KL, She-Yi Su SS, Hsu HC, Chen MF, Liau CS, LeeYT: Efficacy of cholesterol levels and ratios in predicting future coronaryheart disease in a Chinese population. Am J Cardiol 2001, 88(7):737–743.19. Ingelsson E, Schaefer E, Contois JH, McNamara JR, Sullivan L, Keyes MJ,Pencina MJ, Schoonmaker C, Wilson PWF, D’Agostino RB, Vasan RS: Clinicalutility of different lipid measures for prediction of coronary heart diseasein men and women. JAMA 2007, 298(7):776–785.20. McQueen MJ, Hawken S, Wang X, Ounpuu S, Sniderman A, Probstfield J,Steyn K, Sanderson JE, Hasani M, Volkova E, Kazmi K, Yusuf S, for theINTERHEART study investigators: Lipids, lipoproteins, and apolipoprteins asrisk markers of myocardial infarction in 52 countries (the INTERHEARTstudy): a case–control study. Lancet 2008, 372:224–233.21. Eliasson B, Cederholm J, Eeq-Olofsson K, Svensson AM, Zethelius B,Gudbjornsdottir S, on behalf of the National Diabetes Register: Clinicalusefulness of different lipid measures for prediction of coronary heartdisease in type 2 diabetes. Diabetes Care 2011, 34(9):2095–2100.22. Klein BEK, Klein R, Lee KE: Components of the metabolic syndrome andrisk of cardiovascular disease and diabetes in Beaver Dam. Diabetes Care2002, 25(10):1790–1794.23. Hosmer DW, Lemeshow S: Applied Logistic Regression. 2nd edition. Hoboken,NJ: John Wiley & sons, Inc; 2000:156–164.24. Salazar MR, Carbajal HA, Espeche WG, Aizpurua M, Sisnieguez CE, March CE,Balbin E, Stavile RN, Reaven GM: Identifying cardiovascular disease riskand outcome: use of the plasma triglyceride/high-density lipoproteincholesterol concentration ratio versus metabolic syndrome criteria.J Intern Med 2013, 273:595–601.25. Vega GL, Barlow CE, Grundy SM, Leonard D, DeFina LF: Triglyceride-to-high-density-lipoprotein-cholesterol ratio is an index of heart diseasemortality and of incidence of type 2 diabetes mellitus in men. J InvestigMed 2014, 62:345–349.26. Murquia-Romero M, Jimenez-Flores JR, Sinqrist-Flores SC, Espinoza-CamachoMA, Jimenez-Morales M, Pina E, Mendez-Cruz AR, Villalobos-Molina R,Reaven GM: Plasma triglyceride/HLD-cholesterol ratio, insulin resistance,and cardiometabolic risk in young adults. J Lipid Res 2013, 54(10):2795–2799.27. Stein E, Kushner H, Gidding S, Falkner B: Plasma lipid concentrations innondiabetic African American adults: association with insulin resistanceand the metabolic syndrome. Metabolism 2007, 56(7):954–960.28. Karelis AD, Pasternyk SM, Messier L, St-Pierre DH, Lavoie JM, Garrel D,Rabasa-Lhoret R: Relationship between insulin sensitivity and thetriglyceride-HDL-C ratio in overweight and obese postmenopausalwomen: a MONET study. Appl Physiol Nutr Metab 2007, 32(6):1089–1096.29. McLaughlin T, Reaven G, Abbasi F, Lamendola C, Saad M, Waters D, Simon J,Krauss RM: Is there a simple way to identify insulin-resistant individualsat increased risk of cardiovascular disease? Am J Cardiol 2005,96(3):399–404.30. Fan X, Liu EY, Hoffman VP, Potts AJ, Sharma B, Henderson DC: Triglyceride/high-density lipoprotein cholesterol ratio: a surrogate to predict insulinresistance and low-density lipoprotein particle size in nondiabeticpatients with schizophrenia. J Clin Psych 2011, 72(6):806–812.31. Kang H-T, Yoon J-H, Kim J-Y, Ahn S-K, Kinton JA, Koh S-B, Kim J-K: Theassociation between the ratio of triglyceride to HDL-C and insulinresistance according to waist circumference in rural Korean population.Nutr Metab Cardiovasc Dis 2012, 22(12):1054–1060.32. Haedegh F, Hatami M, Tohidi M, Sarbakhsh P, Saadat N, Azizi F: Lipid ratiosand appropriate cut off values for prediction of diabetes: a cohort ofIranian men and women. Lipids Health Dis 2010, 9:85.33. Zoppini G, Negri C, Stoico V, Casati S, Pichiri I, Bonora E: Triglyceride-high-density lipoprotein cholesterol is associated with microvascularcomplications in type 2 diabetes mellitus. Metabolism 2012, 61(1):22–29.34. Jeppesen J, Hein HO, Suadicani P, Gyntelberg F: Relation of high TG-lowHDL cholesterol and LDL cholesterol to the incidence of ischemic heartdisease. An 8-year follow-up in the Copenhagen Male Study. ArteriosclerThromb Vasc Biol 1997, 17:1114–1120.35. Hadaegh F, Khalili D, Ghasemi A, Tohidi M, Sheikholeslami F, Azizi F:Triglyceride/HDL-cholesterol ratio is an independent predictor forcoronary heart disease in a population of Iranian men. Nutr MetabCardiov Dis 2009, 19(6):401–408.36. Casella-Filho A, Hueb W, Jonke V, Santos R, Favarato D, Lima E, Segre A,Cury Rezende P, Larossa Garzillo C, Chacra APM, Ramires J, Kalil-Filho R: Thetriglyceride/HDL ratio remains as an independent plasma lipid marker ofmajor cardiovascular events in diabetic patients 10-year follow-up fromthe Medicine Angioplasty or Surgery Study II (MASS II). JACC 2012,59(13s1):E1663–E1663 (abstract).37. Salazar MR, Carbajal HA, Espeche WG, Aizpurua M, Maciel PM, Reaven GM:Identification of cardiometabolic risk: visceral adiposity index versustriglyceride/HDL cholesterol ratio. AM J Med 2014, 127:152–157.38. Cordero A, Andres E, Ordonez B, Leon M, Laclaustra M, Grima A, Luengo E,Moreno J, Bes M, Pascual I, Civeira F, Pocovi M, Alegria E, Casasnovas JA,on behalf of the MESYAS (Metabolic Syndrome Active Subjects) Studyinvestigators: Usefulness of triglycerides-to-high-density lipoproteincholesterol ratio for predicting the first coronary event in men. Am JCardiol 2009, 104:1393–1397.39. Dobiasova M, Frohlich J: The plasma parameter log (TG/HDL-C) as anatherogenic index: correlation with lipoprotein particle size andesterification rate inapoB-lipoprotein-depleted plasma (FERHDL).Clin Biochem 2001, 34(7):583–588.40. Li C, Ford ES, Menq YX, Mokdad AH, Reaven GM: Does the association ofthe triglyceride to high-density lipoprotein ratio with fasting seruminsulin differ by race/ethnicity? Cardiovasc Diabetol 2008, 7:4.41. Gasevic D, Frohlich J, Mancini GB, Lear SA: The association betweentriglyceride to high-density-lipoprotein cholesterol ration and insulinresistance in a multiethnic primary prevention cohort. Metabolism 2012,61(4):583–589.42. Sumner AE, Finley KB, Genovese DJ, Criqui MH, Boston RC: Fastingtriglyceride and the triglyceride-HDL cholesterol ratio are notmarkers of insulin resistance in African Americans. Arch Int Med 2005,165(12):1395–1400.43. Knight MG, Goedecke JH, Ricks M, Evans J, Levitt NS, Tulloch-Reid MK,Sumner AE: The TG/HDL-C ratio does not predict insulin resistance inoverweight women of African descent: a study of south African, AfricanAmerican, and west African women. Ethn Dis 2011, 21(4):490–494.44. Lear SA, Humphries KH, Kohli S, Chockalingam A, Frohlich JJ, BirminghamCL: Visceral adipose tissue accumulation differs according to ethnicbackground: results of the Multicultural Community Health AssessmentTrial (M-CHAT). Am J Clin Nutr 2007, 86(2):353–359.45. Kohli S, Sniderman AD, Tchernof A, Lear SA: Ethnic-specific differencesin abdominal subcutaneous tissue compartments. Obesity 2010,18(11):2177–2183.46. Kurian AK, Cardarelli KM: Racial and ethnic differences in cardiovasculardisease risk factors: a systematic review. Ethn Dis 2007, 17(1):143–152.47. Chiu M, Austin PC, Manuel DG, Tu JV: Comparison of cardiovascular riskprofiles among ethnic groups using population health surveys between1996 and 2007. CMAJ 2010, 182(8):E301–E310.48. Lear SA, Birmingham CL, Chockalingam A, Humphries KH: Study design ofthe multicultural community health assessment trial (M-CHAT): Acomparison of body fat distribution in four distinct populations. Ethn Dis2006, 16:96–100.49. Friedwald WT, Levy RI, Fredrickson DS: Estimation of the concentration oflow-density lipoprotein cholesterol in plasma, without use of thepreparative ultracentrifuge. Clin Chem 1972, 18:499–502.50. Kriska AM, Knowler WC, LaPorte RE, Drash AL, Wing RR, Blair SN, Bennett PH,Kuller LH: Development of questionnaire to examine relationship ofphysical activity and diabetes in Pima Indians. Diabetes Care 1990,13:401–411.51. Pereira MA, Kriska AM, Joswiak ML, Dowse GK, Collins VR, Zimmet PZ,Gareeboo H, Chitson P, Hemrai F, Purran A: Physical inactivity and glucoseintolerance in the multiethnic island of Mauritius. Med Sci Sports Exerc1995, 27:1626–1634.52. Larsson B, Bengtsson C, Bjorntorp P, Lapidus L, Sjostrom L, Syardsudd K,Tibblin G, Wedel H, Welin L, Wilhelmsen L: Is abdominal body fatdistribution a major explanation for the sex difference in the incidenceof myocardial infarction? The study of men born in 1913 and the studyow women, Goteborg. Sweden. Am J Epidemiology 1992, 135(3):266–273.53. Freedman DS, Jacobsen SJ, Barboriak JJ, Sobocinski KA, Anderson AJ,Kissebah AH, Sasse EA, Gruchow HW: Body fat distribution and male/female differences in lipids and lipoproteins. Circulation 1990,81:1498–1506.54. He H, Ni Y, Chen J, Zhao Z, Zhong J, Liu D, Yan Z, Zhang W, Zhu Z: Sexdifferences in cardiometabolic risk profile and adiponectin expression insubjects with visceral obesity. Transl Res 2010, 155(2):71–77.55. Youden WJ: Index for rating diagnostic tests. Cancer 1950, 3:32–35.56. Schisterman EF, Perkins NJ, Liu A, Bondell H: Optimal cut-point and itscorresponding Youden Index to discriminate individuals using pooledblood samples. Epidemiology 2005, 16:73–81.57. DeLong ER, DeLong DM, Clarke-Pearson DL: Comparing the areas undertwo or more correlated receiver operating characteristic curves:a nonparametric approach. Biometrics 1988, 44:837–845.doi:10.1186/1476-511X-13-159Cite this article as: Gasevic et al.: Clinical usefulness of lipid ratios toidentify men and women with metabolic syndrome: a cross-sectionalstudy. Lipids in Health and Disease 2014 13:159.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionGasevic et al. Lipids in Health and Disease 2014, 13:159 Page 10 of 10http://www.lipidworld.com/content/13/1/159Submit your manuscript at www.biomedcentral.com/submit

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0223727/manifest

Comment

Related Items