Goldenberg Risk Score (ICD Outcome) Calculator

Clinical calculation support · Verify results before clinical use

Risk factor assessment

Enter the patient's clinical findings. The score counts five published predictors; it does not calculate an individual annual ICD-event probability.

0/5Factors
01

Age >70 years

Enter age in completed years. Exactly 70 years does not meet this criterion.

+1 point
Years
02

NYHA functional class III–IV

Corresponds to NYHA functional class >II in the original study.

+1 point
03

QRS duration >120 ms

Enter the measured QRS duration. Exactly 120 ms does not meet this criterion.

+1 point
Milliseconds (ms)
04

Atrial fibrillation present

The original study considered AF as the baseline rhythm at enrollment.

+1 point
05

Blood urea nitrogen (BUN) >26 mg/dL

Use BUN in mg/dL. The published score factor was BUN >26 and <50 mg/dL; BUN ≥50 mg/dL identifies a separate very-high-risk subgroup in the study.

+1 point*
mg/dL
!

Serum creatinine — applicability check

This is not a sixth score factor. It is collected to flag the original study's separately analyzed very-high-risk subgroup.

No points
mg/dL
0Score / 5
Goldenberg clinical risk score
0 risk factors
No listed risk factors are present based on the current entries.
Interpretation of the original study

In the MADIT-II analysis, patients with zero score factors did not show a detectable all-cause mortality benefit from ICD therapy during the study follow-up. This score alone does not determine ICD eligibility.

This is a research-derived risk stratification score, not a standalone ICD eligibility tool. Apply the evidence to the appropriate population and interpret it alongside current guidelines, competing non-arrhythmic mortality, comorbidities, patient preferences, and clinical judgment.

About the score

The Goldenberg clinical risk score was developed from the MADIT-II population to stratify patients with prior myocardial infarction and left ventricular dysfunction by risk of all-cause mortality, and to examine whether the survival benefit associated with primary-prevention ICD therapy differed across risk groups.

Each of the five selected clinical factors contributes one point. The published study excluded a prespecified very-high-risk (VHR) subgroup from the main score analysis, defined by BUN ≥50 mg/dL and/or serum creatinine ≥2.5 mg/dL; that subgroup was evaluated separately.

Score components

FactorCriterionPoints
Age>70 years+1
NYHA classClass III–IV (>II)+1
QRS duration>120 ms+1
Atrial fibrillationPresent at baseline+1
BUN>26 and <50 mg/dL in the main score cohort+1

What the study found

  • Zero risk factors: no detectable all-cause mortality benefit from ICD therapy was observed in this analysis.
  • One or two factors: the observed survival benefit was strongest in these groups.
  • Three or more factors: the estimated benefit was attenuated and not statistically significant in the reported analysis.
  • VHR subgroup: BUN ≥50 mg/dL and/or serum creatinine ≥2.5 mg/dL; these patients were analyzed separately and should not be interpreted using the main cohort's score results.

Important: The categories above describe observed treatment-effect findings in the original study. They are not calibrated annual probabilities of appropriate ICD therapy, and they should not be translated into a percentage risk from the total score.

References

Goldenberg I, Vyas AK, Hall WJ, et al.; MADIT-II Investigators.Risk stratification for primary implantation of a cardioverter-defibrillator in patients with ischemic left ventricular dysfunction. Journal of the American College of Cardiology. 2008;51(3):288–296. https://doi.org/10.1016/j.jacc.2007.08.058. PMID: 18206738.
Primary study recordPubMed abstract and indexed bibliographic details: PubMed 18206738.
Clinical disclaimer: This calculator supports educational risk assessment and does not replace comprehensive clinical evaluation or guideline-based shared decision-making. The score was derived from a specific post-MI, left ventricular dysfunction trial population; generalization to other populations may be inappropriate. It should not be used by itself to recommend for or against ICD implantation.