Metabolic syndrome — the 5 criteria and what each one signals.

Where the criteria came from

Gerald Reaven first described "Syndrome X" in his 1988 Banting Lecture, proposing that insulin resistance was the common upstream driver of a cluster of cardiovascular risk factors — glucose intolerance, hypertension, hypertriglyceridemia, and low HDL cholesterol [Reaven 1988]. The clinical packaging came later. The ATP-III (Adult Treatment Panel III) report from the U.S. National Cholesterol Education Program codified five operational criteria in 2001, updated in 2005 [NCEP ATP-III 2002]. The IDF (International Diabetes Federation) issued its own definition in 2005, making waist circumference an obligatory criterion and adjusting cutoffs by ethnicity [Alberti 2009 harmonized].

The 2009 harmonized definition from the IDF, AHA (American Heart Association), NHLBI, WHF, IAS, and IASO removed the requirement for central obesity and accepted three of any five criteria. That is the version used in most modern epidemiology and the version this article uses below.

The five criteria, what each one signals

Three of the following five criteria establish a metabolic syndrome diagnosis under the 2009 harmonized definition:

• Waist circumference elevated by population-specific cutoff (≥102 cm in men / ≥88 cm in women is the U.S. cutoff; IDF uses lower cutoffs for South Asian, East Asian, and other populations).
• Triglycerides ≥150 mg/dL (1.7 mmol/L), or on drug treatment for elevated triglycerides.
• HDL cholesterol <40 mg/dL (1.0 mmol/L) in men, <50 mg/dL (1.3 mmol/L) in women, or on drug treatment.
• Blood pressure ≥130 mmHg systolic and/or ≥85 mmHg diastolic, or on antihypertensive treatment.
• Fasting glucose ≥100 mg/dL (5.6 mmol/L), or on glucose-lowering treatment.

What each one is actually pointing at:

Waist circumference is a surrogate for visceral adipose tissue (VAT) — the metabolically active fat surrounding abdominal organs. VAT is endocrinologically distinct from subcutaneous fat: it drains directly into the portal circulation, delivering free fatty acids and adipokines to the liver, and it produces a different inflammatory cytokine signature. Two people with identical BMI can have substantially different VAT, and the one with more VAT carries more cardiometabolic risk per kilogram [Després 2008]. Waist circumference catches the compartment that BMI misses.

Elevated triglycerides reflect hepatic VLDL overproduction driven by free fatty acid flux from VAT and by carbohydrate-driven de novo lipogenesis. In insulin resistance, the liver fails to suppress VLDL output appropriately in the fed state, and triglyceride-rich lipoproteins accumulate in circulation. Elevated fasting triglycerides on a routine panel are one of the cleanest non-invasive readouts of impaired hepatic insulin action.

Low HDL cholesterol tracks with the same lipoprotein remodeling. In a high-triglyceride state, cholesteryl ester transfer protein (CETP) shuttles cholesterol out of HDL particles in exchange for triglycerides; the resulting triglyceride-enriched HDL is then catabolized faster, lowering circulating HDL-C. The drop in HDL is therefore not an independent disease — it is a coupled readout of the same lipoprotein dysregulation that elevates triglycerides.

Elevated blood pressure in the metabolic syndrome context is driven by several converging mechanisms: insulin's vascular effects (normally vasodilatory, blunted in insulin resistance), increased sympathetic nervous system activity, sodium retention, and the renin-angiotensin-aldosterone system shifts that accompany visceral adiposity. The pressure threshold in the metabolic syndrome criteria (130/85) is deliberately set below the standard hypertension threshold (140/90 historically, now lower under newer guidelines) — the syndrome captures pre-hypertension that is part of the cluster.

Elevated fasting glucose reflects hepatic insulin resistance specifically. When hepatic insulin signaling is impaired, gluconeogenesis fails to suppress appropriately overnight, and morning fasting glucose climbs. A fasting glucose between 100 and 125 mg/dL is the prediabetes range — biological insulin resistance with intact compensatory beta-cell function. Above 126 mg/dL on two readings is diabetes.

Insulin resistance is the upstream physiology. The five criteria are five different downstream readouts of the same fire.

Why the cluster matters more than any single number

Any individual criterion is informative but limited. Many adults with an isolated elevated triglyceride reading or an isolated borderline blood pressure do not have meaningful insulin resistance. The diagnostic value of the metabolic syndrome construct is that the cluster is mechanistically coherent — visceral adiposity, hepatic insulin resistance, lipoprotein dysregulation, vascular dysfunction, and impaired glucose handling are not independent problems that happen to co-occur, they are coupled readouts of a shared upstream physiology [Eckel 2010].

The clinical implication is that a patient meeting three of five criteria has a different prognosis than a patient with three independent risk factors that happen to add up to a similar Framingham score. The clustering pattern itself carries information about the underlying biology — and about how the patient is likely to respond to interventions that target the upstream driver versus interventions that target each downstream marker in isolation.

Mortality and CVD risk per cluster count

The pivotal meta-analysis on this question is Mottillo and colleagues in JACC (2010), pooling 87 prospective cohort studies covering 951,083 patients. Metabolic syndrome (by NCEP or revised NCEP definitions) was associated with approximately 2.35-fold increased risk of cardiovascular disease, 2.40-fold increased CVD mortality, and 1.58-fold increased all-cause mortality compared with no metabolic syndrome [Mottillo 2010]. The signal was robust across sex, age strata, and study design.

Risk scales with the number of criteria met. In NHANES and other population cohorts, individuals meeting four or five criteria carry roughly 2-3× the all-cause mortality of individuals meeting zero, with intermediate risk at one to three criteria. The relationship is approximately monotonic; there is no obvious threshold below which clustering is "safe."

The relationship to incident type 2 diabetes is even stronger than the cardiovascular relationship. Adults with metabolic syndrome at baseline have approximately 5-fold higher risk of developing type 2 diabetes over follow-up [Ford 2008]. The cluster is functionally a pre-diabetes diagnosis with a richer phenotype.

Intervention hierarchy — lifestyle to surgery

The intervention evidence stack is reasonably mature and ordered roughly by intensity, evidence quality, and effect size.

Lifestyle modification — the foundation. The Diabetes Prevention Program (DPP) randomized 3,234 adults with impaired glucose tolerance to intensive lifestyle, metformin, or placebo. The lifestyle arm — targeting 7% weight loss and 150 minutes/week of moderate activity — reduced incident diabetes by 58% over an average of 2.8 years, outperforming metformin (31% reduction) [Knowler 2002 DPP]. The lifestyle effect durably persisted in long-term follow-up. Whatever else the intervention hierarchy includes, lifestyle is the floor, not the ceiling.

Metformin — the cheap, well-tolerated second line. Metformin reduces hepatic glucose output through partial inhibition of mitochondrial complex I and activation of AMPK signaling. In DPP, it reduced incident diabetes by 31% versus placebo. Its safety profile across decades of use is among the best in metabolic medicine. It is appropriate for prediabetes and for adjunctive use in metabolic syndrome when lifestyle alone does not produce sufficient response.

GLP-1 receptor agonists — the modern second-to-third line. Semaglutide, tirzepatide, and the next-gen incretins (covered in our pipeline review) produce 15-25% weight loss in obesity, with substantial improvements across every metabolic syndrome criterion — waist circumference, triglycerides, HDL, blood pressure, and fasting glucose all move in the right direction. The cardiovascular outcome trials (SUSTAIN-6, SELECT, SURMOUNT) demonstrated MACE (major adverse cardiovascular events) reduction, not just biomarker improvement [Lincoff 2023 SELECT]. These agents are increasingly the appropriate next step when lifestyle alone is insufficient and weight-related morbidity is significant. See our broader piece on the GLP-1 era for the prescribing landscape.

Bariatric/metabolic surgery — the most efficacious option. Roux-en-Y gastric bypass and sleeve gastrectomy produce 25-30% mean weight loss at 1-2 years and durable remission of type 2 diabetes in roughly 30-60% of patients depending on baseline disease severity, with concomitant resolution or improvement across every metabolic syndrome criterion. The Swedish Obese Subjects (SOS) cohort demonstrated long-term mortality reduction in surgical versus matched non-surgical patients [Sjöström 2007]. Surgery sits at the top of the efficacy hierarchy and carries correspondingly distinct surgical, nutritional, and long-term follow-up considerations.

Statins and antihypertensives — for the downstream markers. These remain core therapies for the individual risk factors — elevated LDL, elevated blood pressure — when they meet treatment thresholds. They do not address the upstream insulin resistance driving the cluster, but they do reduce the cardiovascular event rate associated with the cluster.

Where the diagnosis falls short

The metabolic syndrome label has critics, and the critiques are worth taking seriously. The ADA and EASD jointly published a critique in 2005 arguing that the clustering does not exceed what would be expected from the known correlations among the criteria, that the diagnostic cutoffs are arbitrary, and that the label adds little to the management of any individual criterion already targeted by existing guidelines [Kahn 2005].

The defense is that the cluster pattern carries information about a shared biology — visceral adiposity-driven insulin resistance — that treating any single criterion in isolation misses. Whether the diagnostic label changes clinical care or simply re-describes risk factors that would be addressed anyway is a fair argument.

The practical position: do not over-weight the binary "yes/no" diagnosis. Track the criteria as continuous variables. Watch the cluster pattern. Target the upstream physiology — visceral adiposity and insulin resistance — when the cluster is present, rather than chasing each downstream marker independently.

A tiered framework

We do not write protocols. We write frameworks that work between a patient and a clinician.