European Journal of Cardiovascular Medicine

Premature coronary artery disease (CAD) refers to CAD in persons less than 55 years of age in males and 60 years of age in females.1 The incidence of premature CAD has been reported to be 12%–16% in Indians as compared to 2% – 5% reported in Western populations.2,3 Half of the Coronary Vascular Disease-related deaths (i.e, 52% of CVDs) in India occur below the age of 50 years and about 25% of acute myocardial infarction (MI) in India occurs under the age of 40 years.4 This counts to numerous lives lost. Moreover being the productive age group and workforce of the country, this results in huge social and economic burden on the family and the country.

Till date, the reasons for the early occurrence of CAD in Indians is elusive and is a grey area of research. Many factors have been implicated but with inconsistent findings. Lipoprotein a (Lpa) and South Asian ethnicity are both recognized as Atherosclerotic Coronary vascular Disease (ASCVD) risk enhancers in the recently published 2018 cholesterol clinical practice guidelines.5 Recently, the US 2018 American Heart Association/American College of Cardiology Multisociety Guideline (AHA/ACC) has stated that once in a lifetime, measurement of Lp(a) is necessary for all individuals.6 With the importance of Lp(a) coming to the fore, the present study has focussed on Lp(a) levels in the concerned age group.

Lp(a) is a lipoprotein synthesised from liver. It resembles LDL lipoprotein with an extra apolipoprotein a moiety attached to apoplipoprotein B100. It can accumulate in the subendothelial space and cause atherosclerosis along with systemic inflammatory response.7,8 The blood values of Lp(a) exhibit a wide range from 0.1mg/dl to more than 300 mg/dl. Lp(a) concentrations are not influenced much by age, sex, fasting state, inflammation or lifestyle factors such as diet or physical activity or drugs.

The predominance of small dense low density lipoprotein (sdLDL) is currently accepted as a risk factor for cardiovascular disease (CVD) by the National Cholesterol Education Program (NCEPIII).9 sdLDL are believed to be particularly atherogenic due to increased susceptibility to oxidation, high endothelial permeability, decreased LDL receptor affinity and an increased interaction with matrix components. It is not reflected in the conventional lipid profile and can be a underlying cause in Premature CAD.10,11

The study has been designed to evaluate whether elevated Lp(a) and/or sdLDL can be associated with Premature CAD as the normally considered risk factors for CAD do not apply strictly to this group unlike others. Moreover the prevalence of elevated Lp(a) levels (> 20 mg/dl) in Indian population stands at 25-30% which needs to be explored in terms of its association in CAD as has been done in this study.

Objectives

• To evaluate Lp(a) and sdLDL in the study groups
• To determine the correlation between Lp(a) and sdLDL with premature CAD
• To determine the correlation of age of occurrence of CAD with Lp(a) and sdLDL
• To assess the predictive significance of the parameters in the study for premature CAD.

Study design: Case control study

Place of study: Department of Biochemistry, AIIMS Mangalagiri.

Study population and selection criteria

The study population were all the patients visiting Out patient Department (OPD), General Medicine, AIIMS Mangalagiri. All participating subjects were less than 55 years of age for males and less than 60 years of age for females with an event of CAD (stable/unstable angina, Myocardial infarction) within the last 1 month. Taking effect size as 15 from the previous studies,12 the study includes 80 cases of  diagnosed Premature CAD on the basis of detailed history, clinical examination, ECG changes, cardiac injury markers (Troponin T, CK-MB fraction) and/or echocardiography.

Exclusion criteria

Subjects with Diabetes mellitus, hypertension, chronic kidney disease, haemoglobinopathy or erythrocyte disorder were excluded from the study. Pregnancy, acute infections, any history of recent blood transfusion or intake of hypolidemic drugs for more than 1 month are also taken as an exclusion criteria.

80 age and sex matched controls are also included in the study. All control subjects were evaluated for ECG, chest X-ray, and serum analysis. They were classified as healthy subjects based on their normal physical examination results coupled with the absence of personal or family history and reasons for being suspected CAD. Other informations like smoking, hypertension, physical activity, age of the CAD event if any was elicited from the patients by a oral questionnaire.

Ethical Approval

The present study was approved by the Institutional Ethics Committee,  with reference no. AIIMS/MG/IEC/2022-23/221 dated 12-12-2022. Before being enrolled for the study, informed consent was obtained from the patients and controls to use their  clinical data for research. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Participant protection

Necessary ethical clearance was taken from the Institutional Ethics committee for the study. The primary investigator, coinvestigator or any laboratory technician under the supervision of investigators have  collected the blood samples for evaluation.

Sample collection and Analysis

Blood samples were collected from the participants after their informed consent in writing. Sodium Fluoride blood collection tubes were used for the collection of blood sample meant for estimation of Fasting Blood Glucose (FBG) by Glucose oxidase method using kits from Roche (Germany). Whole blood will be collected in EDTA Blood collection tubes for determination of HbA1c at the same time. All the parameters will be assessed on the day of sample collection in clinical chemistry analyser Roche Cobas Pure (Make- Germany). Total cholesterol (TC) ,Triglycerides (TG) and High density lipoprotein cholesterol (HDL) were evaluated by standard kits from M/s Roche. Lp(a)  was evaluated by immunoturbidometric method using kit from Randox laboratories.13 The Randox Lp(a) kit is standardized to the WHO/IFCC reference material, SRM 2B. sdLDL was calculated by the formula sdLDL-C = 0.14*ln(TG)*LDL-C − 0.45*LDL-C + 10.88.14 This sdLDL-C formula integrates both triglyceride levels and LDL-C through a logarithmic transformation, allowing better reflection of triglyceride-driven LDL particle remodeling compared to formulas based only on LDL-C or fixed ratios. It offers a more physiologically responsive and population-adaptable estimation of small dense LDL without requiring specialized assays, making it particularly suitable for large clinical and epidemiological studies. Reliability was tested daily using internal controls. The same measurement technology was maintained  throughout the period of study.

Statistical Analysis

The normality of data was tested using Kolmogorov-Smirnov test and Shapiro Wilk test. The continuous variables was expressed as Mean ± Standard deviation (SD) and the non parametric parameters as median (minimum value- maximum value). Students T test was used to compare the 2 groups in case of parametric data and for non parametric data the comparisons were evaluated by Mann whitney test. Correlation between the groups and premature CAD and age of diagnosis of CAD was evaluated by spearmans correlation. Binary logistic regression was used to model the association between multiple variables and premature CAD. ROC curve was plotted for predictive significance. P value < 0.05 was considered statistically significant in all cases. SPSS statistical software version 24 was used for statistical evaluation.

Baseline characteristics

Of the total 160 subjects in our study (including cases and controls), 47.82% were males and 52.18% were females. The mean age in the control group was found to be 47.32 ± 11.57 years whereas the same for the case group was 52.18 ± 14.68 years for group I and 52.96 ± 09.51 years for group II. The average age of males in the study groups is 49.17± 10.67 (n=76) whereas the same for females is  49.01 ±10.32 (n=83). No significant difference in the age was observed between cases and controls (p value ≥ 0.05) [Table 1].

As observed in table no 2, other than total cholesterol, all parameters were significantly different in cases as compared to controls. sdLDL was significantly elevated (p value < 0.001)  in cases [48.78 (19.03-113.15)] in comparison to controls [35.83 (17.98- 59.79)]. Similarly, Lpa was significantly elevated in cases (p value < 0.001) as compared to controls [69.00 (5.0 – 231.0)] vs controls [32.30 (3.0- 185.56)].

As shown in table 3, CAD  is found to be significantly associated with the lipid indices. Also it is significantly associated with sdLDL (Spearman’s correlation coefficient, ρ= + 0.563) and Lpa ( ρ = + 0.307). Only total cholesterol and sdLDL were found to be correlated with age of diagnosis of CAD (p value < 0.05).

The binary logistic regression analysis revealed that several lipid parameters were significantly associated with the incidence of CAD, after controlling for the effects of other variables in the model (table 4). Specifically, for every one-unit increase, Triglycerides (OR = 1.010, 95% CI: 1.000-1.019, p = 0.039), ApoB (OR = 1.028, 95% CI: 1.009-1.048, p = 0.004), sdLDL (OR = 1.104, 95% CI: 1.046-1.166, p < 0.001), and Lp(a) (OR = 1.010, 95% CI: 1.003-1.017, p = 0.005) were associated with significantly increased odds of CAD. sdLDL emerged as the strongest predictor, indicating a 10.4% increase in the odds of CAD for each unit increase. Conversely, total cholesterol, LDL and Age did not show a statistically significant association with CAD incidence in this multivariable model.

Receiver Operating Characteristic (ROC) curve analysis was performed (Fig 1) to assess the discriminative ability of individual lipid markers in identifying CAD. The Area Under the Curve (AUC) values and their 95% confidence intervals are presented, with all markers showing statistically significant discriminative power (p < 0.001 for all, indicating AUC significantly greater than 0.5). sdLDL demonstrated the highest discriminative ability with an AUC of 0.825 (95% CI: 0.759-0.891), suggesting a "good" level of discrimination for CAD.

The maximum Youden's Index for sdLDL is 0.582, occurring at a cutoff of 42.85 mg/dL. At this cutoff, the Sensitivity is 0.696 (69.6%) and the Specificity is (1−0.114= 0.886) 88.6%. Similarly the maximum Youden's Index for Lpa is 0.329, which occurs at a cutoff of 115 mg/dl. At this cutoff, the Sensitivity is 0.405 (40.5%) and the Specificity is (1−0.076=0.924) 92.4% (Table 6). The results indicate that sdLDL is a substantially better individual discriminator for PCAD than Lpa, as reflected by its higher Youden's Index.

Table 1: Demographic data in the study groups

Table 2:  Biochemical characteristics of study groups

Data are expressed as the mean ± SD or median (minimum value – maximum value).

#The mean difference is significant at the level of 0.05/0.001 amongst controls and cases.

Table 3: Correlation analysis of parameters with CAD  and age of diagnosis of CAD in the study groups

**Correlation is significant at 0.01 level

*Correlation is significant at 0.05 level

Table 4: Binary Logistic Regression Analysis for Association with CAD Incidence

**Correlation is significant at 0.01 level

*Correlation is significant at 0.05 level

Table 5: Area Under the Curve

Table 6: ROC curve analysis of sdLDL and Lp(a) for PCAD prediction

This study proposes an alternative parameter for evaluation of premature CAD as they do not follow the conventional picture of lipid profile as in older CAD. Conventional lipid profile parameters such as total cholesterol and LDL-C often fail to capture the true atherogenic burden in patients with premature CAD. This is because they measure the cholesterol content but not the number, size, or functional quality of lipoprotein particles.15 Importantly, atherogenic lipoproteins like small dense LDL and genetically determined Lp(a), which are strong contributors to premature CAD, are not included in routine lipid testing. Therefore, reliance solely on the conventional lipid profile may underestimate risk in younger individuals, highlighting the need for particle-based lipid markers in this high-risk population.

Small dense fraction of LDL has higher tendency to get oxidized LDL which enhance pro-inflammatory genes resulting in recruitment of monocytes into the sub endothelial space. Oxidized LDL is taken up by monocytes leading to the formation of foam cells which are basis of atherosclerosis. Lipid profile is almost same in most patients of CAD and fails to explain the high morbidity and mortality of CAD.16,17 Lp(a) promotes atherogenesis through its LDL-like particle that delivers cholesterol to the arterial wall, while its apolipoprotein(a) moiety interferes with fibrinolysis, thereby enhancing thrombogenic potential. These dual pro-atherogenic and pro-thrombotic properties make elevated Lp(a) a particularly important determinant of premature CAD risk.18.

As shown in table 1, the mean age in the control group was found to be 47.32 ± 11.57 years whereas the same for the cases was 52.18 ± 14.68 years for group I and 52.96 ± 09.51 years for group II.There was no significant difference in the age groups. The age group in the study is in conformity with the study by Higashioka M et al.19 Increased enrolment of females in this study might be because of the reason that most of them are housewives and can attend the OPD in office hours.

Table no 2 presents a comprehensive comparison of lipid profile parameters, including small dense LDL (sdLDL) and lipoprotein(a) [Lp(a)], between cases of PCAD and controls in this Indian population. Notably, cases exhibited significantly higher values for triglycerides, LDL cholesterol, ApoB, sdLDL, and Lp(a) compared to controls, all with p-values < 0.05, indicating robust statistical associations. The findings are consistent with the observations by a metanalysis by Tian X et al where they observed lpa amongst PCAD patients as higher than 50mg/dl.20 Chen et al in their study observed sdLDL in PCAD patients as above 30 mg/dl. 21 The difference in the two values may be attributed to two different ethnic populations with different dietary patterns. Bohra et al in their study on PCAD patients in Rajasthan observed similar values. 22 These findings underscore the importance of sdLDL and Lp(a) as independent risk factors for premature CAD, alongside conventional markers like LDL and triglycerides. The significant elevation in sdLDL and Lp(a) among cases suggests their potential utility in early risk stratification and management of CAD in this demographic, highlighting the need for routine assessment of these parameters in clinical practice for the Indian population. In contrast, total cholesterol (TC) and HDL cholesterol levels did not differ significantly between cases and controls, further reinforcing the value of advanced lipid markers over traditional metrics in identifying high-risk individuals for premature CAD.

As shown in Table 3, the Spearman’s correlation analysis demonstrates that sdLDL have the strongest positive correlation with PCAD (+0.563, p<0.01) among all lipid parameters assessed and is also significantly correlated with the age of diagnosis of CAD (+0.160, p<0.05). This robust association surpasses that observed for conventional lipids such as LDL (+0.263), total cholesterol (+0.282), and Lp(a) (+0.307), emphasizing the pivotal role of sdLDL in the pathogenesis and earlier onset of premature CAD in the studied population. The findings closely mirror prior evidence, notably the results reported by Bansal et al.,23 where elevated sdLDL and a lower sdLDL index effectively distinguished individuals at higher risk for CAD. Importantly, these results further reinforce the proposition that advanced lipid markers like sdLDL should be prioritized as sensitive predictors of premature CAD, especially in populations where traditional lipid metrics may underrepresent risk.

Table 4 presents the binary logistic regression analysis for factors associated with the incidence of PCAD, revealing that sdLDL (OR 1.104, 95% CI 1.046–1.166, p<0.001) and Lp(a) (OR 1.010, 95% CI 1.003–1.017, p=0.005) are strong, independent predictors of CAD in this population. The findings are in line with those observed by Choi S et al.24 They observed a OR of 2.312( CI 1.512-3.537) between sdLDL and CAD. Notably, ApoB and triglycerides also emerge as significant risk markers, whereas traditional lipids such as total cholesterol and LDL failed to reach statistical significance. This might be due to several reasons. First, traditional markers such as total cholesterol and LDL-C may lose significance when more specific and atherogenic lipoprotein fractions (e.g., ApoB, sdLDL, and Lpa) are included in the same model, as these reflect particle number and quality rather than bulk cholesterol content. Second, collinearity between lipid variables may reduce the apparent independent effect of LDL-C and total cholesterol. Third, age, though a well-established risk factor, may not emerge as significant in this study due to the restricted age range of participants, relatively small sample size, or the stronger contribution of lipid-related factors in this cohort. Taken together, these findings suggest that particle-based parameters such as sdLDL, ApoB, and Lp(a) may be more sensitive markers of CAD risk in this population than conventional lipid indices.

The present study demonstrates that small dense LDL (sdLDL) and lipoprotein(a) [Lp(a)] possess strong discriminative power for premature CAD among the hospital population in South India, with an optimal sdLDL cutoff of 42.85 mg/dL (AUC 0.825) and Lp(a) cutoff of 115 mg/dL (sensitivity 40.5%, specificity 92.4%). The sdLDL threshold closely aligns with recent Indian studies, such as Bohra et al.22, which reported similar values (>40 mg/dL) for predicting CAD risk in young adults, further validating its use as a clinically useful marker. The significantly higher Lp(a) cutoff of 115 mg/dL determined by Youden’s index compared to the median value of 69 mg/dL (range 5.0–231.0) in the study likely reflects the need to optimize the balance between sensitivity and specificity for clinical discrimination of premature CAD. While the median represents the central tendency of Lp(a) levels in the cohort, the Youden’s index identifies the threshold that best distinguishes cases from controls, which often lies above the median in skewed distributions. It may also reflect underlying genetic determinants, diet, and high prevalence of metabolic syndrome among this group pf population. International research indicates that Lp(a) levels can vary considerably between populations, and higher thresholds have been identified in Asian Indian cohorts due to distinctive atherogenic risk profiles and genetic diversity.25 Thus, our findings not only reinforce the importance of sdLDL as a sensitive marker, but also highlight the necessity for population-specific Lp(a) reference ranges when evaluating CAD risk among Indians.

sdLDL and Lp(a) are associated with the incidence of PCAD and may serve as a promising biomarker for PCAD risk prediction. However, further research is warranted to validate these findings in larger cohorts and explore the clinical utility of sdLDL and Lp(a) in risk stratification and preventive interventions for CAD in this population.

Candidates could be advised for evaluation of sdLDL along with lipid profile. If found high sdLDL, such cases should be considered as high risk cases and monitored accordingly or preventive strategies can be introduced. This will help to reduce the mortality and morbidity of  CAD along with a substantial increase in quality of life and less financial burden on the society.

1.       Patel J, et al. Premature coronary artery disease, risk factors, clinical presentation, angiography and interventions: Hospital based registry. Indian Heart J. 2022;74(6):461–468

2.       Mammi MVI, Pavithran P, Rahman PA, et al. Acute MI in North Kerala: A 20-year hospital-based study. Indian Heart J. 1991;43:93-96.

3.       Negus BH, Williard JE, Glamann DB, et al. Coronary anatomy and prognosis of young asymptomatic survivors of myocardial infarction. Am J Med. 1994;96:354-358.

4.       Murray CJL, Lopez AD. Global comparative assessments in the health sector. Geneva: World Health Organization; 1994.

5.       Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol: a report of the ACC/AHA Task Force. J Am Coll Cardiol. 2018;30423393.

6.       Stone NJ, Blumenthal RS, Lloyd-Jones D, Grundy SM. Comparing primary prevention recommendations: a focused look at US and European guidelines on dyslipidemia. Circulation. 2020 Apr 7;141(14):1117-1120.

7.       Shiffman D, et al. Single variants can explain the association between coronary heart disease and haplotypes in the apolipoprotein(a) locus. Atherosclerosis. 2010;212:193-196.

8.       Lanktree MB, Anand SS, Yusuf S, Hegele RA, Investigators S. Comprehensive analysis of genomic variation in the LPA locus and its relationship to plasma lipoprotein(a) in South Asians, Chinese, and European Caucasians. Circ Cardiovasc Genet. 2010;3:39-46.

9.       National Cholesterol Education Program Expert Panel. Third report of the NCEP expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) final report. Circulation. 2002;106(25):3143-3421.

10.    De Graaf J, Hak-Lemmers HL, Hectors MP, et al. Enhanced susceptibility to in vitro oxidation of the dense low-density lipoprotein subfraction in healthy subjects. Arterioscler Thromb. 1991;11:298-306.

11.    Lichtenstein AH, Chung M, Lau J, Balk EM. Systematic review: association of LDL subfractions with cardiovascular outcomes. Ann Intern Med. 2009;150:474-484.

12.    Gambhir JK, Kaur H, Gambhir DS, Prabhu KM. Lipoprotein(a) as an independent risk factor for CAD in patients below 40 years of age. Indian Heart J. 2000;52(4):411-415.

13.    Wyness S, Genzen JR. Performance evaluation of five lipoprotein(a) immunoassays on the Roche cobas c501 chemistry analyzer. J Appl Lab Med. 2021;6(3):856-868.

14.    Han T, Piao Z, Yu Z, et al. An equation for calculating small dense LDL cholesterol. Lipids Health Dis. 2024;23:366.

15.    Campos H, Blijlevens E, McNamara JR, et al. Low density lipoprotein particle size and coronary artery disease. Arterioscler Thromb. 1992;12(2):187-195.

16.    Hulthe J, Bokemark L, Wikstrand J, Fagerberg B. The metabolic syndrome, LDL particle size, and atherosclerosis: the AIR study. Arterioscler Thromb Vasc Biol. 2000;20:2140-2147.

17.    Koenig W, Sund M, Froehlich M, et al. C-reactive protein predicts future risk of coronary heart disease in healthy middle-aged men: MONICA Augsburg Cohort Study. Circulation. 1999;99:237-242.

18.    Stein JH. Lipoprotein Lp(a) excess and coronary heart disease. Arch Intern Med. 1997;157(11):1170.

19.    Higashioka M, Sakata S, Honda T, et al. Association of small dense LDL-C and coronary heart disease in subjects at high cardiovascular risk. J Atheroscler Thromb. 2021;28(1):79-89.

20.    Tian X, et al. Association between lipoprotein(a) and premature ASCVD: a meta-analysis. Front Cardiovasc Med. 2024;PMC11086656.

21.    Chen S, et al. Novel lipid biomarkers and ratios as risk predictors for premature CAD. Front Cardiovasc Med. 2023;PMC10710552.

22.    Bohra A, Tiwari P, Agarwal S, Jain S, Modh H. Advanced lipid markers as predictors of premature CAD in young Indians. J Acad Med Pharm. 2025;7(2):1071-1076.

23.    Bansal SK, Agarwal S, Daga MK. Conventional and advanced lipid parameters in premature CAD patients in India. J Clin Diagn Res. 2015 Nov;9(11):BC07.

24.    Choi S, Kim WG, Park K, et al. Small dense LDL and lipoprotein(a) as independent risk factors for early-onset CAD. Korean Circ J. 2024;54(1):23-32.

Banerjee D, Wong EC, Palaniappan L, Shin J, Fortmann SP. Racial and ethnic variation in lipoprotein(a) levels among Asian Indian and Chinese patients. J Lipids. 2011;2011:1-6.