Adaptation of the Charlson Comorbidity Index for risk prediction and implantable defibrillator decision making in the older adult
J B Mortiboys, R Crapper, J T Middleton, F Carr, N LewisAbstract
Introduction
A retrospective study of older adults aged ≥ 65 years who underwent implantable cardiac defibrillator (ICD) implant at our tertiary cardiac centre was undertaken to evaluate risks and benefits in this specific cohort. We used a modified Charlson Comorbidity Index (CCI) score to assess whether this could identify patients who may be at risk of more harm than benefit from ICD implant based on mortality, adverse events and anti-tachycardia therapies.
Methods
All patients aged ≥ 65 years who underwent ICD or cardiac resynchronisation therapy defibrillator (CRT-D) implant between 01/01/2015 and 31/01/2020 were included in the study and followed-up until 01/01/2025. Device and medical records at the time of implant, device follow-up data and mortality data were used to analyse the cohort and formulate CCI scores. We then modified the CCI score to set the baseline age at 65 years to better reflect true comorbidity within the group. Therefore, for the age component of the score, patients aged 65-69 years scored 0, 70-79 years 1 point and ≥ 80 years 2 points, as opposed to 2, 3 and 4 points respectively on the unmodified score. Data was analysed using Microsoft Excel and SPSS V. 29.0.2.0.
Results
252 patients aged ≥ 65 years underwent device implant (163 CRT-D, 89 ICD) and were included in the study. Median age at implant was 74 years, with majority male sex (86%), ischaemic aetiology (67%) and primary prevention indication (72%). A 54% mortality rate (N = 136) was observed during a median follow-up of 5.51 years. Of those, 70% (N = 95) died without receiving any appropriate anti-tachycardia therapy from their device. 71 patients (28%) received appropriate therapy during follow-up, with a median 2.5 year survival post therapy. Of the 181 patients who received no therapy, 32 (18%) experienced direct device related complications, with 66% (N = 21) requiring a second procedure as a result. When our modified CCI score was applied, a score ≥ 5 was associated with a significantly higher risk of mortality than a score ≤ 2 (figure 1, hazard ratio 1.4 vs. 0.6, p < 0.001). 55% of patients with an age-adjusted CCI score ≥ 5 died without receiving anti-tachycardia therapy, compared to 17% in those with a score ≤ 2. Device related complication rates were high, yet similar in both groups (figure 2).
Conclusions
In our cohort, we found high 5-year mortality alongside high incidence of death without appropriate anti-tachycardia therapy, as well as high complication rates. Patients with age-adjusted CCI scores ≥ 5 had significantly higher rates of mortality, with lower rates of appropriate anti-tachycardia therapies compared to those with scores ≤ 2. This highlights the need for careful patient selection in this population. The age-adjusted CCI has the potential to be incorporated into pre-device decision making and counselling processes. Future work should produce a dedicated multivariable tool to guide selection for ICDs in older adults.