Non-traditional risk factors for myocardial infarction in adults under forty: A systematic review of emerging trends

Abstract

BACKGROUND

Myocardial infarction (MI) incidence is increasing in adults aged < 40 years, however, as many as 25% may occur with no traditional risk factors.

AIM

To look at non-traditional risk factors for early onset MI.

METHODS

Based on the guidance of PRISMA-2020, search of PubMed, EMBASE, Web of Science, Scopus and Cochrane Central Register of Controlled Trials between April 2015 and April 2020 for observational studies focusing on the association between non-traditional risk factors and MI in adults aged 18-40 years. Two reviewers independently screened studies, extracted data and checked quality using Newcastle-Ottawa Scale or Agency for Healthcare Research and Quality checklist. The protocol was registered in PROSPERO, No. CRD420251061098.

RESULTS

Thirteen studies (7 cohort, 4 case-control, 2 cross-sectional) from 11 countries met inclusion criteria with a sample size ranging from 154 participants in a pilot case-control study to 5.7 million people in a United States National Inpatient Sample analysis. Psychosocial factors showed consistent associations: Depression showed an MI risk 1.6-3.1-fold higher and being unpartnered was associated with a post-MI readmission risk that was 28%-31% higher. Autoimmune conditions had the greatest associations, with human immunodeficiency virus infection quadrupling odds of MI (4.06), and the risk of systemic lupus erythematosus doubling (2.12). Obstructive sleep apnea increased major adverse cardiovascular events by almost four times (hazard ratio = 3.87). Adhering to the Mediterranean diet was protective (odds ratio = 0.55). Accelerated biological aging (shortening of telomeres) separated young patients with MI from controls. Traditional risk factors did not account for up to 30% of MI cases in each of the cohorts. Most studies were of moderate to high quality, although causes of heterogeneity in design and age stratification of participants mixtures limited causality inference.

CONCLUSION

Non-traditional psychosocial, autoimmune, inflammatory, and lifestyle factors play an important role in the risk of MI in young adults. Integrating these factors into risk prediction models could improve the early identification of high-risk individuals and target prevention strategies for this vulnerable population.

Key Words: Acute myocardial infarction; Coronary artery disease; ST-elevation myocardial infarction; Young adult

Core Tip: Non-traditional risk factors such as psychosocial stressors, autoimmune disorders, biological aging and specific lifestyle and environmental exposures play a significant role in myocardial infarction risk and outcome among individuals aged < 40 years. Psychosocial factors like depression, low socioeconomic status, and unpartnered status were significantly associated with adverse events, especially among young women. Autoimmune diseases and inflammation indicators, as well as sleep disorders and unhealthy lifestyle patterns, additionally improved the risk.

INTRODUCTION

Coronary artery disease (CAD) and acute myocardial infarction (MI) are major global causes of morbidity and mortality for which the burden is rising among younger adults[1,2]. Although MI has been described as a disease of the elderly for several decades, recent epidemiological data suggest that the relative incidence of MI events among the young is increasing[3,4]. This trend is alarming as there are high rates of recurrent ischemic events, premature mortality and long-term socioeconomic impact with early-onset MI[5]. Long-standing risk factors including smoking, hypertension, diabetes, dyslipidemia, obesity, inactivity and positive family history of CAD have been the target of the traditional approaches of risk stratification and prevention[6,7]. Nevertheless, none of these traditional risk factors account for all of the burden of MI in young adults, with as many as 25% of premature cardiovascular disease (CVD) and 12% of acute MI cases occurring without recognized risk factors[8].

This gap in risk prediction has led to increasing recognition of non-traditional risk factors in the pathogenesis of MI among young adults[9]. Non-traditional risk factors are defined as exposures or conditions not routinely included in standard cardiovascular risk prediction models such as Framingham, Atherosclerotic Cardiovascular Disease (ASCVD), or Systematic Coronary Risk Evaluation[6,10]. These encompass a diverse range of factors, including chronic inflammatory and autoimmune diseases [e.g., systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), human immunodeficiency virus (HIV)], psychosocial stressors [e.g., depression, anxiety, low socioeconomic status (SES)], environmental exposures (e.g., air pollution, gut microbiota), sleep disorders (notably OSA), and certain lifestyle factors (such as recreational drug use, e-cigarettes, and cannabis)[8,11,12].

Identification and analysis of atypical risk factors are essential in youth because they frequently have atypical presentations, non-atherosclerotic etiologies, and are often underrepresented in clinical trials[13-15]. These unique risk profiles and etiologies are not sufficiently accounted for in current guidelines and risk prediction models, which have largely been extrapolated from older populations[6,16]. This calls for improved risk stratification methodology and prevention strategies that include traditional and non-traditional risk factors in this population[8,15].

Previous review studies suggest a significant role for non-classical risk factors-especially HIV, SLE, and obstructive sleep apnea-in MI risk in young adults but the generalizability of such associations has usually been limited by older data, more restricted scope or less systematic approaches[8]. The review is unusual by its systematic and comprehensive approach, study selection, data extraction and quality checking. A methodologically sound and focused synthesis brings important gaps in the current literature on risk for MI in young adults and highlights incorporation of non-traditional risk factors in risk assessment and dissemination aimed at young adults at risk for MI.

MATERIALS AND METHODS

This systematic review was completed following the PRISMA 2020 guidelines[17] and was prospectively registered with PROSPERO (registration ID: CRD420251061098). The review set out to collate and critically evaluate existing evidence for non-traditional risk factors for MI in young adults over 18 years of age but below 40 years, by employing a structured and transparent approach to study selection, data extraction and quality appraisal. Any modifications to the protocol were entered into PROSPERO with date, description and rationale.

Eligibility criteria and framework

The review was designed using the Population-Exposure-Outcome framework.

Population: Adults aged 18-40 years diagnosed with MI (including ST-elevation MI and non-ST-elevation MI). Adults aged under 40 years with MI were the primary target population. Given the limited number of strictly age-restricted studies, we also included studies with broader age ranges if relevant subgroup data could be extracted or if the majority of participants were under 40.

Exposure: Non-traditional risk factors, explicitly defined as exposures not routinely included in standard cardiovascular risk models (e.g., Framingham, ASCVD, Systematic Coronary Risk Evaluation). These include psychosocial stressors (e.g., depression, marital status, SES), autoimmune diseases (e.g., SLE, RA), substance use (e.g., recreational drugs, cannabis, amphetamines, alcohol), environmental exposures (e.g., air pollution, microplastics, periodontitis), sleep disorders (e.g., OSA), and novel biomarkers [e.g., lipoprotein(a), high-sensitivity C-reactive protein, elevated hair cortisol]. Moderately prevalent conditions such as obesity were classified as traditional risk factors and excluded from the primary analysis, while cannabis use was included as non-traditional.

Outcomes: The primary outcome was the occurrence of MI or quantitative effect estimates (odds ratios, hazard ratios, or relative risks) linking non-traditional exposures to MI in the specified age group.

Study design: Observational studies (cohort, case-control, cross-sectional), large registries, and studies providing stratified data for adults under 40.

Language: Studies published in English were included; non-English studies were considered if translation resources were available.

Exclusion criteria: Reviews, editorials, conference abstracts, animal studies, in vitro studies, and studies not primarily addressing MI or non-traditional risk factors in the target population were excluded. Studies focused exclusively on traditional risk factors (e.g., hypertension, low-density lipoprotein cholesterol, diabetes, obesity) or lacking stratified data for adults under 40 were also excluded.

Search strategy

A systematic search strategy was employed in the following electronic databases: PubMed/MEDLINE, EMBASE, Web of Science, Scopus and the Cochrane Central Register of Controlled Trials. Reference lists of included studies and relevant systematic reviews were also used as such. The last search was run on 28 April 2025. A search strategy based on a combination of controlled and free-text terms, adjusted to each database (e.g., Medical Subject Heading, Emtree) was created. Search terms were “myocardial infarction”, “acute coronary syndrome”, “young adults”, “non-traditional risk factors”, “psychosocial stress”, “autoimmune disease”, “substance abuse”, “environmental exposure”, and “biomarkers”. Boolean operators were used as appropriate.

Study selection

All records identified from the database search were imported into Rayyan systematic review management platform for deduplication. Titles and abstracts were independently screened by two reviewers to assess relevance according to the pre-specified inclusion and exclusion criteria. Full texts of potentially eligible articles were then retrieved and assessed in detail by the same reviewers. Discrepancies in study selection were resolved through discussion or consultation with a third reviewer. The study selection process was documented using the PRISMA 2020 flow diagram, which details the number of records identified, screened, excluded (with reasons), and included in the final review (PRISMA, 2020).

Data extraction

Data extraction was conducted independently by two reviewers using a standard form of data extraction that was designed for this review. Data extraction included study features [author(s), year, country, design, sample size], participant features (age, sex, comorbidity), exposure characteristics (type of non-traditional risk factor, method of measurement), end result measures (estimation outcomes, confidence intervals, adjustment variables), and follow-up time. Any discrepancies with data extraction were resolved by agreement or by referring to a third reviewer.

Risk of bias assessment

The risk of bias in all included cohort and case control studies was evaluated with the Newcastle-Ottawa Scale (NOS), which is an assessment of quality across domains of selection, comparability and outcome/exposure. The two included cross-sectional studies were evaluated using the Agency for Healthcare Research and Quality (AHRQ) methods checklist used to evaluate study quality based on 11 items. All studies were independently scored by two reviewers with disagreements resolved by consensus or third reviewer as needed. In NOS, articles with scores of 7-9 were described as high quality, 4-6 as moderate quality and 0-3 as low quality. In the AHRQ checklist, scores were considered high quality (8-11), moderate (4-7) and low quality (0-3).

Statistical analysis

Since heterogeneity in design, populations and exposures is anticipated, narrative synthesis was carried out. Results were compiled thematically by type of non-orthodox risk factor (e.g., psychosocial, environmental, inflammatory, autoimmune, lifestyle) and summarized with effect estimates and adjustment for confounders. The findings were summarised in narrative and tabular form. Then PRISMA flow diagram was used to describe the selection process of the study, and tables were designed to show the characteristics of the studies (participants characteristics, exposure details, and outcomes, and quality assessment).

Addressing bias

Publication bias was taken into account, as studies showing significant associations are arguably more likely to get published. Selection bias arising from differing definitions of the term “young” and severity of MI was also discussed as a limitation.

RESULTS

A systematic literature search was conducted in four major electronic databases, namely, PubMed/MEDLINE, EMBASE, Cochrane Library and Web of Science, yielding 1055 studies. After duplicate removal there were 971 studies left for title and abstract screening. The full text review resulted in the inclusion of 13 studies in the final synthesis. The flowchart of the study selection is shown in Figure 1.

Figure 1  PRISMA-2020 flow diagram for systematic review.

Study selection and characteristics

Thirteen observational studies[18-30] met the inclusion criteria, encompassing diverse designs: Cohort studies (n = 7), case-control studies (n = 4) and cross-sectional analyses (n = 2). These studies spanned 11 countries, including high-income (United States, Sweden, Italy) and low/middle-income regions (India, China, South Korea), reflecting global heterogeneity in risk factor profiles. Sample sizes ranged from 154 participants in a pilot case-control study[21] to 5.7 million in a United States National Inpatient Sample analysis[18]. While most studies focused on adults aged 18-55, five provided stratified data for populations under 45 (Table 1)[19-21,28,30].

Table 1 Study characteristics and participant demographics.

Ref.	Country	Study design	Sample size	Age range	Sex distribution	Comorbidities	MI classification
Krittanawong et al[18], 2020	United States	Retrospective cohort (NIS)	5764755 (< 55 years), 1149185 AMI	< 55 (mean 45.1 AMI)	66.6% men, 33.4% women (AMI)	Obesity, smoking, HTN, diabetes, HIV, SLE, OSA, RA	STEMI, NSTEMI
Zhu et al[19], 2024	United States	Prospective cohort (VIRGO)	2979	18-55 (median 48)	67.2% women, 32.8% men	HTN, diabetes, obesity, smoking, alcohol abuse, prior CVD, COPD	STEMI, NSTEMI, EF < 40%
Dreyer et al[20], 2021	United States	Prospective cohort (VIRGO)	2979	18-55 (mean 47.1)	67.4% women, 32.6% men	HTN, diabetes, obesity, smoking, prior AMI, COPD	STEMI, NSTEMI
Gupta et al[21], 2020	India	Case-control	154 (77 MI, 77 controls)	18-45 (mean 35.3 MI)	65 men/12 women (MI), 58 men/19 women (controls)	Excluded smokers, diabetics, BMI > 35	AWMI, IWMI, LWMI
Cho et al[22], 2019	South Korea	Population-based cohort	2705090	20 + (stratified)	Not split, large population	HTN, diabetes, hypercholesterolemia, SES, depression	AMI (ICD-10)
Fan et al[23], 2019	China	Prospective cohort	804	18-85 (mean 57.5)	82.6% men	HTN, diabetes, hyperlipidemia, prior MI/PCI, smoking	ACS (STEMI/NSTEMI/UA)
Smolderen et al[24], 2015	United States, Spain, Australia	Cohort (VIRGO)	3572	18-55 (median 48)	67.1% women, 32.9% men	HTN, diabetes, obesity, smoking, hypercholesterolemia, prior AMI, CHF	STEMI, NSTEMI
Turati et al[25], 2015	Italy	Case-control	760 cases, 682 controls	19-79 (median 61 MI)	76.3% men (MI)	HTN, diabetes, hyperlipidemia, BMI	Non-fatal AMI
Orth-Gomér et al[26], 1986	Sweden	Case-control	210 (89 MI, 121 controls)	18-45 (mean 39.5)	89 men (MI), 121 men (controls)	HTN, diabetes, hyperlipidemia, smoking	MI (survivors < 45)
Zhao et al[27], 2023	China	Cross-sectional	8103	18-99 (mean 50.3)	53.5% women, 46.5% men	HTN, diabetes, obesity	Not MI-specific
Head et al[28], 2019	United States	Cross-sectional autopsy (PDAY)	2651	15-34 (mean 24.8)	75% men, 25% women	Excluded major comorbidities	Subclinical atherosclerosis
Lu et al[29], 2022	United States	Case-control (VIRGO/NHANES)	2264 AMI, 2264 controls	18-55 (median 48)	68.9% women, 31.1% men	HTN, diabetes, hypercholesterolemia, obesity, depression, low income, family history	STEMI, NSTEMI, type 1
Jariwala et al[30], 2022	India	Retrospective, multicenter	3656 (< 45 years PCI)	< 45 (mean men 37.4, women 41.1)	69.2% men, 30.8% women	HTN, diabetes, overweight, dyslipidemia, family history, smoking, alcoholism	ACS (STEMI/NSTEMI/UA)

NIS: National inpatient sample; VIRGO: Variation in recovery: Role of gender on outcomes of young AMI patients (a major multinational study); PDAY: Pathobiological determinants of atherosclerosis in youth (study); NHANES: National Health and Nutrition Examination Survey; MI: Myocardial infarction; HTN: Hypertension; HIV: Human immunodeficiency virus; SLE: Systemic lupus erythematosus; OSA: Obstructive sleep apnea; RA: Rheumatoid arthritis; STEMI: ST-elevation myocardial infarction; NSTEMI: Non-ST-elevation myocardial infarction; CVD: Cardiovascular disease; COPD: Chronic obstructive pulmonary disease; EF: Ejection fraction; AMI: Acute myocardial infarction; BMI: Body mass index; AWMI: Anterior wall myocardial infarction; IWMI: Inferior wall myocardial infarction; LWMI: Lateral wall myocardial infarction; PCI: Percutaneous coronary intervention; ACS: Acute coronary syndrome; UA: Unstable angina.

Participant demographics and clinical profiles

Young adults with MI exhibited distinct demographic profiles across studies. Women constituted 33%-68% of cohorts, with higher representation in United States-based studies[19,24]. Traditional risk factors like hypertension (28%-64%), smoking (12%-60%), and obesity (20%-51%) remained prevalent but failed to explain up to 30% of MI cases in cohorts excluding these factors[21,28]. Notably, 13% of autopsied young adults (15-34 years) had accelerated coronary atherosclerosis unaccounted for by traditional risk factors[28]. Table 1 presents the study characteristics and participant demographics.

Psychosocial and socioeconomic risk factors

Psychosocial stressors consistently emerged as significant predictors of MI incidence and adverse post-MI outcomes in young adults. Two large United States cohort studies[19,20] found that unpartnered status (single, divorced, or widowed) was associated with a 28%-31% increased risk of 1-year all-cause readmission after MI [adjusted hazard ratio (HR) = 1.28-1.31], with unpartnered women facing the highest risk. Depression, measured by Patient Health Questionnaire-9, was also more prevalent in women with MI (39%) than in men (22%) and independently increased MI risk [adjusted odds ratio (OR) = 1.64-3.09][24,29]. Low SES, defined by income or education, amplified risk synergistically with depression, contributing to a 47% higher incidence of MI in South Korean adults [HR = 1.47, 95% confidence interval (CI): 1.36-1.60][22]. Across studies, the association between depression and MI was consistent, though the magnitude of effect varied by sex and measurement method.

Inflammatory and biological aging markers

Accelerated biological aging, as measured by telomere length, was strongly associated with MI in young adults, particularly in those without traditional risk factors. Gupta et al[21] found that MI patients had telomeres nearly seven times shorter than controls (mean T/S ratio 0.115 vs 0.792, P < 0.0001), with the difference most pronounced among women. However, this association was reported in a small case-control study, and longitudinal risk estimates were not available.

Autoimmune and chronic inflammatory disorders

Autoimmune diseases demonstrated some of the strongest associations with MI risk. Krittanawong et al[18] reported that HIV infection quadrupled the odds of MI (OR = 4.06, 95%CI: 3.48-4.71), and SLE doubled the risk (OR = 2.12, 95%CI: 1.89-2.39) in young adults. Interestingly, RA was associated with a slightly reduced risk (OR = 0.83, 95%CI: 0.76-0.89), possibly reflecting treatment effects or surveillance bias. These findings were consistent across large administrative datasets, but stratified data for those under 40 were limited.

Environmental and lifestyle factors

Environmental and occupational exposures also contributed to MI risk. OSA was associated with a 3.9-fold higher risk of major adverse cardiovascular and cerebrovascular events after one year (HR = 3.87, 95%CI: 1.20-12.46)[23], though the overall association was more modest and not limited to young adults. Night shift work and low job autonomy explained 5% of MI risk variance in Swedish men under 45[26]. Adherence to a Mediterranean diet was protective, halving the risk of non-fatal MI (OR = 0.55, 95%CI: 0.40-0.75)[25]. Short sleep duration (≤ 6 hours) worsened cardiovascular health profiles and increased the odds of hypertension and dyslipidemia[27], though the outcome was not MI specifically. Exposure details and outcomes of every study is presented in Table 2.

Table 2 Exposure details and outcomes.

Study ID	Non-traditional risk factors assessed	Measurement methods	Outcome measures	Effect estimates (OR/HR/RR with 95%CI)	Covariates adjusted for
1	HIV, SLE, OSA, RA	ICD-9 codes (NIS)	AMI risk	HIV: OR = 4.06 (3.48-4.71); SLE: OR = 2.12 (1.89-2.39); OSA: OR = 1.16 (1.12-1.20); RA: OR = 0.83 (0.76-0.89)	Age, sex, race, BMI, DM, HTN, HLD, CKD, smoking
2	Marital/partner status, depression, social support, stress	Structured interview, PHQ-9, Social Support Inventory, Perceived Stress Scale	1-year all-cause readmission	HR = 1.31 (1.15-1.49) unpartnered vs partnered; adjusted HRs for demo/SES/clinical/psychosocial	Age, sex, race, SES, clinical, psychosocial
3	Depression, social support, stress, SES	PHQ-9, Social Support Inventory, PSS, SES data	1-year all-cause readmission	HR = 1.28 (1.15-1.42) unpartnered vs partnered; adjusted HRs	Age, sex, race, MI severity, comorbidities, psychosocial
4	Telomere length (biological aging)	qPCR for telomere length	Telomere length in MI vs controls	Mean T/S ratio MI 0.115 vs controls 0.792 (P < 0.0001); shorter in MI	Age, gender, BMI
5	Socioeconomic status, depression	Insurance premium, ICD-10 codes	AMI incidence	HR low SES vs high 1.16 (1.14-1.19); HR depression vs none 126 (1.21-1.31)	Age, sex, comorbidities
6	OSA	Polygraphy (AHI ≥15)	MACCE, unstable angina	HR = 1.55 (0.94-2.57) MACCE; HR = 3.87 (1.20-12.46) MACCE after 1 years	Age, sex, BMI, HTN, diabetes, prior MI/PCI
7	Depression, psychosocial stress	PHQ-9, PSS, interview	Prevalence of depressive symptoms at AMI	Women: 39% PHQ-9 ≥ 10, men: 22%; adjusted OR for women 1.64 (1.36-1.98)	Age, sex, SES, comorbidities
8	Mediterranean diet adherence	Food Frequency Questionnaire, MDS	Non-fatal AMI	MDS ≥ 6: OR = 0.55 (0.40-0.75); per point: OR = 0.91 (0.85-0.98)	Age, sex, BMI, HTN, diabetes
9	Type A behavior, psychosocial work, education	Jenkins Activity Survey, work environment survey	MI risk, variance explained	Work monotony/poor discretion 5% variance; type A 2%; education NS	Age, sex, education
10	Sleep duration	Self-report, AHA CVH score	Ideal CVH, BP, glucose, cholesterol	≤ 6 hours sleep: OR = 1.38 (1.15-1.67) for non-ideal CVH	Age, sex, BMI, comorbidities
11	Unexplained (likely non-traditional) risk	Autopsy, lesion quantification	Subclinical atherosclerosis	OR high-growth group per year age: 1.125 (1.063-1.190)	Age, cholesterol, BMI, HbA1c, CRP
12	Depression, low income, family history	PHQ-9, interview, lab, SES	First AMI, PAFs	Women: Depression OR = 3.09 (2.37-4.04); low income OR = 1.79 (1.28-2.50)	Age, sex, SES, comorbidities
				Men: Depression OR = 1.77 (1.15-2.73)
13	Hypothyroidism, CTD, RHD, takayasu, SCAD, OCP use	Medical record, interview, labs	Prevalence, in-hospital outcomes	Non-traditional RFs rare; no adjusted effect estimates; in-hospital mortality 1.77%-2%	Age, sex

OR: Odds ratio; HR: Hazard ratio; RR: Relative risk; CI: Confidence interval; HIV: Human immunodeficiency virus; SLE: Systemic lupus erythematosus; OSA: Obstructive sleep apnea; RA: Rheumatoid arthritis; BMI: Body mass index; DM: Diabetes mellitus; HTN: Hypertension; PHQ-9: Patient Health Questionnaire-9; SES: Socioeconomic status; MACCE: Major adverse cardiovascular/cerebrovascular event; MDS: Mediterranean Diet Score; CVH: Cardiovascular health; CTD: Connective tissue disorder; RHD: Rheumatic heart disease; SCAD: Spontaneous coronary artery dissection; OCP: Oral contraceptive pill; PAF: Population attributable fraction; PSS: Perceived Stress Scale; HLD: Hyperlipidemia; CKD: Chronic kidney disease; MI: Myocardial infarction; qPCR: Quantitative polymerase chain reaction; ICD-10: International classification of diseases, 10^th revision; AMI: Acute myocardial infarction; PCI: Percutaneous coronary intervention; SES: Socioeconomic status; NS: Not significant; AHA: American Heart Association; BP: Blood pressure; HbA1c: Hemoglobin A1c; CRP: C-reactive protein; RFs: Risk factors.

Consistency and heterogeneity across studies

Overall, the association between psychosocial stress, depression, and MI was consistent across studies, with stronger effects observed in women. Findings for inflammatory and autoimmune markers were also robust, although some heterogeneity was noted in the strength of association and the populations studied. For example, while HIV and SLE consistently increased MI risk, the association for RA was inverse in the largest cohort. Evidence for environmental and lifestyle factors was generally supportive but sometimes limited by study design or outcome definitions (e.g., sleep duration studies assessed cardiovascular health, not MI directly). The role of telomere length as a marker of biological aging was supported by a single small study, and further research is needed to confirm this association. Table 3 presents the summary of evidence strength.

Table 3 Summary table of evidence strength.

Risk factor category	Number of studies	Direction of association	Effect estimates (range)	Strength of evidence
Depression/psychosocial	5	Positive (risk ↑)	OR = 1.64-3.09, HR = 1.28-1.31	Strong
Low socioeconomic status	3	Positive (risk ↑)	HR = 1.16-1.47	Moderate
Autoimmune (HIV, SLE)	2	Strong positive (risk ↑)	OR = 2.12-4.06	Strong
Rheumatoid arthritis	1	Negative (risk ↓)	OR = 0.83	Emerging
Telomere length	1	Positive (risk ↑, shorter TL)	T/S ratio 0.115 vs 0.792	Emerging
Obstructive sleep apnea	1	Positive (risk ↑)	HR = 1.55-3.87	Moderate
Mediterranean diet	1	Negative (risk ↓)	OR = 0.55	Moderate
Short sleep duration	1	Positive (risk ↑) (CV health)	OR = 1.38	Emerging
Occupational stress	1	Positive (risk ↑)	5% variance explained	Emerging

Strength of evidence: Strong factors show consistent findings in multiple large studies; moderate factors show consistent findings in ≥ 2 studies or 1 large study; emerging factors show single or small studies, or inconsistent results. OR: Odds ratio; HR: Hazard ratio; HIV: Human immunodeficiency virus; SLE: Systemic lupus erythematosus; TL: Telomere length; T/S: Telomere-to-single-copy gene ratio; CV: Cardiovascular.

Subgroup and sensitivity analyses

Subgroup analyses revealed notable effect modification by sex, age, SES, and comorbidities. In a large administrative database, HIV and SLE were strongly associated with MI risk in young adults, but stratified data for those under 40 were not available[18]. Two cohort studies found that unpartnered status was linked to higher 1-year readmission rates after MI, especially among women, though the association was attenuated after psychosocial adjustment and only one study found a significant sex-marital status interaction[19,20]. Age-stratified analysis showed telomere length was especially reduced in MI patients aged 31-45, with the difference most pronounced in females[21]. Combined low SES and depression conferred a substantially higher MI risk[22]. The protective effect of a Mediterranean diet was more evident in those with lower BMI or without hypertension[25]. Sleep duration analyses showed that short sleep (≤ 6 hours) was associated with non-ideal cardiovascular health, though not specifically MI[27]. Details of these analyses are presented in Table 4.

Table 4 Subgroup and sensitivity analyses.

Study ID	Subgroup (e.g., sex, age, region)	Effect estimates (OR/HR/RR with 95%CI)	Notes
1	Age (< 40 vs 40-55), sex	HIV: OR = 4.06 (3.48-4.71); SLE: OR = 2.12 (1.89-2.39); OSA: OR = 1.16 (1.12-1.20) for MI risk in young adults	No stratified data < 40; large administrative database; cross-sectional analysis of hospitalizations
2	Sex (women vs men), marital status (unpartnered vs partnered)	HR = 1.31 (1.15-1.49) unpartnered vs partnered for 1-year readmission; unpartnered women: 37.6% readmission vs unpartnered men: 26.8%	No significant sex-marital status interaction (P = 0.69); effect attenuated after psychosocial adjustment
3	Sex, marital status	HR = 1.28 (1.15-1.42) unpartnered vs partnered for 1-year readmission; unpartnered women had highest readmission	Significant sex-marital status interaction; psychosocial factors partially mediate risk
4	Age group (18-30 vs 31-45), sex	Telomere length shorter in MI patients aged 31-45 vs controls (P < 0.05); females with MI had shorter telomeres than female controls (P < 0.01)	No longitudinal MI risk data; small sample; case-control design
5	SES and depression combined	HR = 1.47 (1.36-1.60) for low SES + depression vs high SES, no depression (AMI risk)	Large population; no stratification < 40; depression by ICD codes; median 11.6 years follow-up
6	Time (≤ 1 year vs > 1 year), OSA status	HR for MACCE after 1 year in OSA: 3.87 (1.20-12.46); overall HR = 1.55 (0.94-2.57) for OSA vs non-OSA	Mean age 575; not exclusive to young adults; median 1-year follow-up
7	Sex (women vs men)	Adjusted OR for depressive symptoms at AMI: 1.64 (1.36-1.98) women vs men	Data collected at AMI admission only; no MI risk prediction; age up to 55; no follow-up
8	BMI (< 25 vs ≥ 25), hypertension status	Stronger inverse association of mediterranean diet with AMI in BMI < 25 and normotensive individuals	Case-control; median age 61; not restricted to young adults; no follow-up
9	Education level, sex	Men with high education had higher type A and work strain; women with low education had more type A behavior	Small sample; only men in main analysis; older data; case-control; no follow-up
10	Sleep duration categories (≤ 6 hours, 7 hours, 8 hours, ≥ 9 hours)	OR = 1.38 (1.15-1.67) for short sleep (≤ 6 hours) and non-ideal CVH	Cross-sectional; broad age range; outcome is CV health, not MI; no follow-up
11	Age (per year increase), high-risk subgroup vs low-risk	OR = 1.125 (1.063-1.190) per year age for high-growth atherosclerosis group	Cross-sectional autopsy study; subclinical outcome; no direct MI data; no follow-up
12	Sex, AMI subtype (type 1 vs others)	Women: Depression OR = 3.09 (2.37-4.04), men OR = 1.77 (1.15-2.73); family history stronger in men	Case-control; age 18-55; no exclusive < 40 data; robust adjustment; no follow-up
13	Sex	Non-traditional RFs rare (e.g., hypothyroidism, CTD, SCAD); no significant sex differences	Retrospective; in-hospital outcomes only; no effect estimates for non-traditional RFs; no follow-up

OR: Odds ratio; HR: Hazard ratio; HIV: Human immunodeficiency virus; RR: Relative risk; SLE: Systemic lupus erythematosus; OSA: Obstructive sleep apnea; MI: Myocardial infarction; AMI: Acute myocardial infarction; ICD: Implantable cardioverter-defibrillator; MACCE: Major adverse cardiac and cerebrovascular events; BMI: Body mass index; CV: Cardiovascular; RFs: Risk factors; CTD: Connective tissue disease; SCAD: Spontaneous coronary artery dissection.

Risk of bias assessment

A summary of risk of bias assessments, including individual scores, quality classification, and key limitations for each study, is provided in Table 5. The risk of bias for all 13 included studies was assessed using the NOS for cohort and case-control studies and the AHRQ tool for cross-sectional studies. Most studies were rated as moderate-to-high quality (NOS score ≥ 5 or AHRQ score ≥ 7), supporting the overall reliability of the evidence. However, common limitations included residual confounding, retrospective data collection, and incomplete adjustment for potential confounders. Several studies also had limitations related to exposure measurement or lack of stratified data for younger age groups.

Table 5 Risk of bias assessment.

Study ID	Study design	Quality tool	Score	Quality classification	Key limitations
1	Cohort	NOS	7	High	Administrative data, coding errors
2	Cohort	NOS	7	High	Residual confounding, observational design
3	Cohort	NOS	6	Moderate	Retrospective data, potential selection bias
4	Case-control	NOS	5	Moderate	Small sample size, cross-sectional design
5	Cohort	NOS	8	High	Limited exposure detail
6	Cohort	NOS	7	High	Short follow-up, single center
7	Cohort	NOS	7	High	Potential residual confounding
8	Case-control	NOS	6	Moderate	Recall bias, dietary assessment
9	Case-control	NOS	5	Moderate	Small sample, self-reported exposures
10	Cross-sectional	AHRQ	7	Moderate	Self-report, cross-sectional design
11	Cross-sectional	AHRQ	6	Moderate	Surrogate outcome, autopsy data
12	Case-control	NOS	7	High	Potential selection bias
13	Cohort	NOS	5	Moderate	Retrospective design, limited non-traditional risk data

NOS: Not otherwise specified; AHRQ: Agency for healthcare research and quality.

DISCUSSION

This systematic review emphasizes on the increasing importance of non-traditional risk factors in the pathogenesis of MI under the age of 40 years. While the increase in many classic risk factors, such as hypertension, diabetes, obesity and dyslipidemia, has led to increased rates of premature CAD, a significant proportion of young MI patients present with other risk factors or additional risk factors[15,28]. Of great note, recent large registries such as YOUNG-MI are reporting that 10%-20% of young MI patients lack standard modifiable risk factors, and thus the need to identify emerging contributors[15,31].

Psychosocial and socioeconomics factors

Psychosocial factors such as depression, stress, social isolation, and low SES were identified as major non-traditional risk factors in young MI patients[19,24,29]. Several studies such as the VIRGO and YOUNG-MI registry cohort groups found that depression is very common in young MI patients, especially among women, with an adjusted odds ratio for depressive symptoms as high as between 1.6 to over 3[12,24,32]. Unpartnered status was also found to increase the risk of a poor outcome post-MI (HR = 1.3) particularly in women[19,20]. These findings are consistent with the study by the International Hearty Wellbeing Program (INTERHEART), which showed that psychosocial stress is an important risk factor for MI globally[33]. Mechanistically, psychosocial stress may induce MI through neuroendocrine activation, endothelial dysfunction and pro-inflammatory pathways[12,34]. Sex-specific analyses showed that women have a greater burden of psychosocial risk factors - resulting in greater relative risk than men[29].

Inflammatory diseases and the biological aging

Accelerated biological aging as manifested by reduced telomere length was linked to MI in young adults without traditional risk factors[21]. This finding supports the hypothesis that premature cellular senescence is a contributing factor for early atherosclerosis and MI[35]. Although limited by small sample size and the cross-sectional design, these results are consistent with larger evidence for a role for telomere attrition in cardiovascular risk. Recent reviews also strongly emphasize the emerging role of biomarkers such as lipoprotein(a), although there is currently some inconsistency in findings and these require prospective validation[36]. Emerging evidence from population-based and clinical studies indicates that chronically elevated cortisol, when measured through hair cortisol concentration (HCC), is strongly implicated as a risk factor for MI[37,38]. Hair cortisol provides a retrospective index of cumulative cortisol exposure over months, thus overcoming the limitations of diurnal serum or salivary cortisol, which vary with acute stressors. In the large cross-sectional study by Faresjö et al[37] individuals with prior cardiovascular events exhibited significantly higher HCC levels compared to those without such events and HCC positively correlated with major cardiometabolic risk factors such as hypertension, central obesity, and dyslipidemia. Unlike traditional risk markers, HCC captures the biopsychosocial dimension of cardiovascular risk, offering additional prognostic insight into stress-mediated cardiometabolic pathology.

Autoimmune and chronic inflammatory disorders

Autoimmune diseases, such as SLE and HIV infection were strongly linked to the risk of MI in young adults[15]. Chronic systemic inflammation and immune dysregulation in these conditions increase the rate of atherosclerosis and thrombosis[39,40]. The inverse association shown with RA in one study may be related to treatment or to bias in surveillance and needs further study[18]. These findings highlight the importance of cardiovascular risk assessment and management in the young patient with autoimmune disease[16].

Lifestyle, environmental and sleep factors

OSA was linked with a nearly 4-fold higher risk of major adverse cardiovascular events after 1 year in young patients with acute coronary syndrome[23]. Mechanisms can include intermittent hypoxia and sympathetic activation[41]. Short sleep duration was also associated with impaired CVD health parameters such as hypertension and dysglycemia, although direct outcomes of MI were not measured[27]. Adherence to a Mediterranean diet was protective against risk of MI, reducing risk by around 45%[25]. These findings are consistent with the importance of lifestyle and environmental factors in the development of early-onset MI; as are recent studies of air pollution and long covid[42,43].

Consistency, heterogeneity and limitations

In various articles, the link between psychosocial stress, depression and MI was independent, especially with regard to women[24,29]. Inflammatory and autoimmune markers also displayed strong associations but the strength of effect varied[18]. Evidence for environmental and lifestyle factors was quite supportive but sometimes limited by study design or outcome definitions[23,27]. Limitations of included studies include heterogeneity in terms of design, exposure definitions and age ranges, with some studies including adults up to 55 years[28]. Many used cross sectional or administrative data which may cause bias and limit causal inference[17]. Little longitudinal data and stratified analyses for adults strictly under 40 were limited. Small sample sizes and the reliance of International Classification of Diseases coding in some cohorts may further limit the generalizability.

Clinical implications

These findings have important clinical implications to risk assessment and prevention on young adults. Current risk models, like the ones suggested by the American College of Cardiology/American Heart Association, do not adequately take into account non-traditional risk factors[16]. Incorporation of factors related to the psyche, autoimmunity and lifestyle into risk prediction tools could lead to better identification of high-risk individuals in early stages[31,33]. We recommend development and validation of a non-traditional risk score prototype specific to young adults, which may be used to devise targeted preventive strategies and interventions.

Future directions

Future research efforts should focus on prospective cohort studies in adults under 40 characterized in detail about psychosocial, biological, autoimmune, and environmental exposures. Integration of novel biomarkers, genetics information and sex-specific studies will improve the risk stratification. Interventional studies addressing modifiable, non-traditional risk factors, including depression, sleep disorders and environmental exposures, are warranted to assess their effectiveness on MI prevention and outcomes in the population.

CONCLUSION

This systematic review shows that non-traditional risk factors such as psychosocial stressors, autoimmune disorders, biological aging and specific lifestyle and environmental exposures play a significant role in MI risk and outcome among individuals aged < 40 years. Psychosocial factors like depression, low SES and unpartnered status were significantly associated with adverse events, especially among young women. Autoimmune diseases and inflammation indicators, as well as sleep disorders and unhealthy lifestyle patterns, additionally improved the risk. The evidence highlights considerable heterogeneity in the risk profiles and important sex differences. However, the current literature has limitations such as variations in study design, age stratification, and no longitudinal data. Moving forward, adding non-traditional risk factors to clinical risk assessment and prevention strategies is important in early identification and better management of young adults at risk for MI. Future studies are needed that aim at prospective validation of novel biomarkers and the creation of sex-specific risk prediction models that can be used to guide targeted interventions in this population.