#hypertrophic-cardiomyopathy #diagnosis #cardiac-mri #sudden-cardiac-death #risk-stratification

HCM: Diagnosis and Risk Assessment

Detail pages. concepts/HCM-Risk-SCD · concepts/HCM-AI-ECG-Scar · concepts/Late-Gadolinium-Enhancement · concepts/ECG-Ventricular-Hypertrophy

Definition

The diagnosis of hypertrophic cardiomyopathy is phenotypic: increased left ventricular wall thickness in one or more segments that cannot be explained by abnormal loading conditions or infiltration. In an adult the threshold is 15 mm, or 13 mm in a first-degree relative of an index case; in a child it is a z-score above 2. Risk assessment then answers a single question — what is the probability of sudden cardiac death over the next five years — using the HCM Risk-SCD calculator in adults and HCM Risk-Kids in children. The two European guidelines published a year apart disagree about how binding that number should be, which is the most consequential open question on this page. (sources/esc-cmp-2023, rating: very high; sources/VA-SCD-ESC-2022, rating: very high)

Abbreviations: CMR, cardiac magnetic resonance; ICD, implantable cardioverter-defibrillator; LGE, late gadolinium enhancement; LV, left ventricular; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; NSVT, non-sustained ventricular tachycardia; SCD, sudden cardiac death; VF, ventricular fibrillation.

Key Concepts

Diagnosis

Adult criterion: maximum LV wall thickness ≥15 mm in any segment, or ≥13 mm in a first-degree relative of an index case, unexplained by loading conditions (Class I). Paediatric criterion: wall thickness z-score >2. (sources/esc-cmp-2023, rating: very high)
Initial work-up comprises history, a three- to four-generation pedigree, ECG, echocardiography, CMR and first-line laboratory testing. (sources/esc-cmp-2023)
CMR is recommended at baseline (Class I, Level B). It identifies apical and lateral hypertrophy missed by echocardiography, apical aneurysm, and myocardial fibrosis — LGE is present in about 65% of patients. It is the non-invasive standard for measuring wall thickness for risk assessment; CT and PET are problem-solving alternatives. See concepts/Late-Gadolinium-Enhancement and concepts/ECG-Ventricular-Hypertrophy. (sources/esc-cmp-2023; sources/HCM-NEJM-2025, rating: very high)
Echocardiography is usually decisive for obstruction, but a normal resting gradient does not exclude it — provocation is required before the diagnosis of nonobstructive disease is accepted. (sources/HCM-NEJM-2025)
Gene-positive, phenotype-negative relatives require longitudinal follow-up rather than discharge; family testing also detects the phenocopies covered on concepts/HCM-Genetics-and-Phenotypes. (sources/HCM-NEJM-2025)
Machine learning is entering the diagnostic pathway but not the guideline. An explainable model can detect and localise LV scar from a 12-lead ECG with an F1 of 89% on external validation, which would matter most where CMR is unavailable; no guideline incorporates ECG-based scar detection, and the model cannot quantify scar. See concepts/HCM-AI-ECG-Scar. (sources/lvh-scar-aiecg-naturesr-2025, rating: high)

Invasive Haemodynamic Assessment

Indication and provocation sequence. Catheterisation is required when non-invasive data are inconclusive, to document a resting or provoked gradient ≥50 mmHg before septal reduction. Provoke in order with Valsalva, then post-extrasystolic potentiation, then isoproterenol — the most sensitive manoeuvre for labile obstruction. (sources/hemodynamics-circ-2012, rating: high)
Post-procedural assessment repeats both resting and provoked gradients; a residual provocable gradient ≥50 mmHg indicates an incomplete result. The technique itself — the Braunwald–Brockenbrough sign, transseptal versus retrograde access, and catheter entrapment — is set out in full on concepts/LVOTO. (sources/hemodynamics-circ-2012)

SCD Epidemiology

Incidence is 0.5–1% per year, approximately 1% before the ICD era, concentrated in children and adults under 30 and rare above 60. In some patients SCD is the first clinical manifestation. (sources/HCM-VA-FCVMed-2022, rating: high; sources/HCM-NEJM-2025, rating: very high)
In childhood the cumulative five-year incidence approaches 8–10%, and HCM is the leading cause of SCD in young people in North America. Ventricular fibrillation is the commonest terminal rhythm, and antiarrhythmic drugs do not clearly prevent it. (sources/HCM-VA-FCVMed-2022; sources/HCM-NEJM-2025)
Prophylactic ICD implantation has reduced disease-related mortality more than tenfold, and five- and ten-year HCM-related survival of 98% and 94% now approaches that of the general population. High-risk ICD use has brought annual SCD below 0.5%. (sources/HCM-VA-FCVMed-2022; sources/HCM-NEJM-2025)
The substrate that generates it — ion channel remodelling, myofilament calcium sensitivity and sarcoplasmic reticulum calcium overload — is on concepts/HCM-Mechanism.

Risk Stratification and ICD Selection

HCM Risk-SCD is the adult tool (Class I, Level B) for patients aged 16 and over. It estimates five-year SCD probability from age, NSVT, unexplained syncope, maximum wall thickness, family history of SCD, left atrial diameter and LVOT gradient. HCM Risk-Kids is the validated paediatric equivalent. See concepts/HCM-Risk-SCD. (sources/esc-cmp-2023, rating: very high)
Risk must be reassessed at first evaluation and then every 1–3 years or whenever clinical status changes (Class I, ESC 2022 — an explicit new requirement). Neither tool should be applied to elite athletes or to metabolic and infiltrative disease such as Fabry or Noonan. (sources/VA-SCD-ESC-2022, rating: very high; sources/esc-cmp-2023)
The two European guidelines grade the same threshold differently. ESC 2023 makes an ICD at five-year risk ≥6% Class IIa, with Class IIb at 4–6% and at low risk on extensive LGE (≥15%) or LVEF <50%. ESC 2022 makes the same ≥6% threshold Class I and adds a Class IIa tier at 4–6% with any of LGE ≥15% of LV mass, LVEF <50%, an abnormal blood-pressure response to exercise, apical aneurysm or a sarcomeric variant. Threshold tables and calculator mechanics are on concepts/HCM-Risk-SCD. (sources/esc-cmp-2023; sources/VA-SCD-ESC-2022)
Secondary prevention is Class I for survivors of cardiac arrest or sustained ventricular tachycardia with haemodynamic compromise. (sources/esc-cmp-2023; sources/VA-SCD-ESC-2022)
AHA/ACC 2024 treats the calculator as one element of shared decision-making (Class IIa) and states explicitly that a prespecified threshold should not be the sole arbiter. Its primary-prevention markers are family history of SCD, wall thickness ≥30 mm, apical aneurysm, unexplained syncope, multiple or prolonged NSVT, extensive LGE and LVEF <50%, with reassessment every 1–2 years when none are present. (sources/HCM-AHA-2024, rating: very high; sources/HCM-NEJM-2025)
The Maron registry quantifies the trade-off. Among 527 patients with at least one risk marker and a primary-prevention ICD, 15.6% received device-terminated VF or VT; only 0.3% of patients without risk markers and without an ICD died suddenly. AHA/ACC criteria are the more sensitive and ESC criteria the more specific. (sources/HCM-NEJM-2025)
A sarcomeric variant alone does not justify an ICD independent of a validated risk tool, and an LV apical aneurysm is not accepted as a sole indication in Europe. Catheter ablation is Class IIb for recurrent symptomatic sustained monomorphic VT or recurrent ICD shocks refractory to antiarrhythmic drugs. (sources/esc-cmp-2023; sources/VA-SCD-ESC-2022)

Contradictions / Open Questions

Paediatric risk scores carry adult-derived assumptions. HCM Risk-Kids is validated, but the underlying physiology — small cavity, high heart rate, growth-related changes in wall thickness and z-scores — means an adult-derived five-year probability model is being applied to a moving anatomical target. The ESC 2022 Class IIa threshold at a Risk-Kids score ≥6% borrows the adult cut-point. (sources/VA-SCD-ESC-2022, rating: very high)
A phenotype-based definition leaves the preclinical carrier outside every risk pathway. Diagnosis requires wall thickness above a threshold, so a genotype-positive relative with normal wall thickness has no diagnosis to code, no calculator input beyond family history, and no guideline mandate for an ICD — even though electrical remodelling is already present. The definition and the risk model share the same blind spot. See concepts/HCM-Mechanism. (sources/esc-cmp-2023; sources/HCM-VA-FCVMed-2022, rating: high)
Provocation is required but not standardised. Valsalva, post-extrasystolic potentiation and isoproterenol do not provoke the same gradient in the same patient, and the treatment threshold of 50 mmHg is applied to whichever manoeuvre happens to be positive. Neither the risk models nor the septal-reduction trials specify which provocation defines obstruction, so patients can meet or miss criteria depending on local practice. (sources/hemodynamics-circ-2012, rating: high; sources/esc-cmp-2023)
Sensitivity and specificity are being traded without an outcome trial. The Maron registry shows that AHA/ACC criteria identify more patients who will need device therapy and ESC criteria identify fewer who will not, but no trial has randomised patients to one framework and measured death. The choice between them is a value judgement presented as a guideline recommendation. (sources/HCM-NEJM-2025, rating: very high)
The calculator has never been tested against a treated population. HCM Risk-SCD was derived before widespread myosin inhibition and before septal reduction was offered at today's volume, both of which change the haemodynamic and symptomatic trajectory without necessarily changing arrhythmic risk. A five-year probability estimated from a historical cohort is being applied to patients whose haemodynamics have been altered by therapy. (sources/esc-cmp-2023; sources/HCM-NEJM-2025)

Connections

Part of entities/HCM — the parent overview
The variants and phenocopies being diagnosed — concepts/HCM-Genetics-and-Phenotypes
The substrate behind the risk — concepts/HCM-Mechanism
Treatment of obstruction and arrhythmia — concepts/HCM-Management
Calculator mechanics and validation — concepts/HCM-Risk-SCD
Imaging and ECG surrogates — concepts/Late-Gadolinium-Enhancement, concepts/ECG-Ventricular-Hypertrophy and concepts/HCM-AI-ECG-Scar
Outflow tract obstruction and its measurement — concepts/LVOTO
Devices — entities/ICD and concepts/ICD-Programming
Wider context — concepts/Sudden-Cardiac-Death and concepts/Phenotypic-Approach-to-Cardiomyopathy
Family follow-up — concepts/Cascade-Family-Screening