DSP Cardiomyopathy Risk Calculator – Desmoplakin-Related Ventricular Arrhythmia Risk

Clinical calculation support · Verify results before clinical use

5-Year Ventricular Arrhythmia Risk Assessment

Enter baseline clinical and imaging variables for an eligible patient with a pathogenic or likely pathogenic DSP variant.

— 5-year risk

Eligibility

The published model was developed for patients with a pathogenic/likely pathogenic DSP variant and no sustained ventricular arrhythmia before enrolment (primary prevention setting).

Pathogenic or likely pathogenic DSP variant

Confirmed genetic test result classified as P/LP in the DSP gene.

Required

Yes

No / uncertain variant

Previous sustained ventricular arrhythmia

Any sustained VA before the baseline assessment used for this calculation.

Required

Yes

Clinical and imaging variables

Use measurements from the baseline evaluation. All five model predictors are required.

Sex

Female sex is coded as 1 in the published prognostic index.

Required

Female

Male

History of non-sustained ventricular tachycardia (NSVT)

Documented NSVT on ECG monitoring before or at the baseline evaluation.

Required

Yes

24-hour premature ventricular contraction (PVC) count

Enter the total number of PVCs recorded over 24 hours, not PVC percentage or PVCs per hour.

Required

Total PVC count

PVCs / 24 h

The equation uses ln(total 24-hour PVC count). A value of zero cannot be entered because ln(0) is undefined.

Left ventricular ejection fraction (LVEF)

The model predictor is LVEF less than 50% (strictly below 50%).

Required

LVEF

Enter a value from 0 to 100%. Exactly 50% is coded as no for the LVEF <50% predictor.

Moderate or severe right ventricular systolic dysfunction

Defined as RVEF <40% on CMR, or RV fractional area change (FAC) <33% on echocardiography when CMR is unavailable.

Required

Yes

—5-year risk

Predicted risk of first sustained ventricular arrhythmia

—

Clinical interpretation

Prognostic index and model components

Predictor	Model value	Contribution

This is a model-estimated 5-year risk of a first sustained VA event, not an individual certainty or a direct estimate of sudden cardiac death. Interpret in the context of the patient's full clinical profile and shared decision-making.

About the DSP Risk Score

The DSP risk score estimates the 5-year risk of a first sustained ventricular arrhythmia in patients carrying a pathogenic or likely pathogenic DSP variant who have not experienced sustained VA before the baseline assessment. It was developed using the international DSP-ERADOS registry.

Model equation

5-year risk = 1 − (0.929)^exp(PI)

PI = 0.735 × Female sex + 0.592 × NSVT + 0.581 × (LVEF <50%) + 1.71 × Moderate/severe RV dysfunction + 0.199 × ln(24-hour PVC count) − 1.787.

Dichotomous predictors are coded as 1 when present and 0 when absent. The calculator reports the model-derived probability as a percentage.

Predictors and operational definitions

Predictor	Definition used in the model	Coefficient
Female sex	Female = 1; male = 0	0.735
NSVT	History present = 1; absent = 0	0.592
LVEF <50%	Strictly below 50% = 1	0.581
Moderate/severe RV dysfunction	RVEF <40% on CMR or RV FAC <33% on echocardiography = 1	1.710
24-hour PVC count	Natural logarithm of total PVC count over 24 hours	0.199
Intercept	Constant	−1.787

Risk categories

Predicted 5-year risk	Risk group
<5%	Low
5% to 20% (inclusive)	Intermediate
>20%	High

Risk group boundaries are presented as operational, non-overlapping intervals for display: exactly 5% is classified as intermediate and exactly 20% remains intermediate.

Evidence and Limitations

The original model included 471 patients, with 385 in the development cohort and 86 in the external validation cohort. The reported c-statistic was 0.782 in development and 0.791 in the original external validation cohort.

A further independent international validation published in 2026 included 450 patients and reported a c-statistic of 0.719 (95% CI 0.706–0.733), supporting external validity while showing that discrimination may vary across cohorts.

Important limitations

Use only in patients with a pathogenic/likely pathogenic DSP variant and no sustained VA before baseline; this is a primary-prevention model.
It is not validated for patients with prior sustained VA, non-DSP genotypes, or unclassified/benign DSP variants.
The 24-hour PVC input is an absolute count, not a percentage burden. The model applies the natural logarithm of the count.
Zero PVC count cannot be evaluated directly because the published logarithmic term is undefined at zero.
The score estimates sustained VA risk; it is not a standalone ICD indication and should not replace specialist assessment or shared decision-making.
History of myocarditis-like episodes, LGE, age, and other clinical features are not direct inputs in this published model.

References

Carrick RT, Gasperetti A, Protonotarios A, et al. A novel tool for arrhythmic risk stratification in desmoplakin gene variant carriers. European Heart Journal. 2024;45(32):2968–2979. DOI: 10.1093/eurheartj/ehae409.

Gasperetti A, Bauce B, Protonotarios A, et al. External validation of the DSP-risk score for prediction of clinically significant ventricular arrhythmias in primary prevention patients with desmoplakin cardiomyopathy-associated genetic variants. Heart Rhythm. 2026;23(10):e2585–e2594. DOI: 10.1016/j.hrthm.2026.04.038.

Arbelo E, Protonotarios A, Gimeno JR, et al. 2023 ESC Guidelines for the management of cardiomyopathies. European Heart Journal. 2023;44(37):3503–3626. DOI: 10.1093/eurheartj/ehad194.

Clinical disclaimer: This calculator is intended for educational and clinical decision-support purposes in appropriately selected patients. It provides a prognostic estimate and does not diagnose disease, determine an ICD indication, or replace clinical judgement. ICD decisions require individualized assessment of phenotype, arrhythmic history, imaging, comorbidities, patient preferences, and current professional guidance.