Pragmatic Referral Profiles for Perioperative Comprehensive Geriatric Assessment (CGA) in Elective Abdominal Aortic Aneurysm (AAA) Repair

Introduction

Elective abdominal aortic aneurysm (AAA) repair is increasingly performed in older adults living with frailty and multimorbidity. Advancing technique, particularly endovascular aneurysm repair (EVAR) has widened access for those not previously considered for open surgery, with EVAR rising from 23% to 66% of repairs between 2022 and 2025, total volume nearly tripling within the South-East Wales Vascular Network (SEWVN). Despite national guidance and evidence supporting perioperative Comprehensive Geriatric Assessment (CGA) in Vascular Surgery, no structured referral pathway existed within SEWVN to identify those most likely to benefit from multidisciplinary optimisation and shared decision making.

Methods

A retrospective service evaluation analysed National Vascular Registry (NVR) and routinely collected perioperative data from 204 adults undergoing primary elective AAA repair (96 open; 108 EVAR) within SEWVN between 2022 and 2025. Frailty-associated vulnerabilities, including delirium risk, medication burden, nutritional status and functional capacity were assessed alongside comorbidities and physiological reserve. Univariate analyses identified predictors of prolonged LOS (>7 days open; >2 days EVAR) and other adverse postoperative events. Referral profiles integrated statistical, clinical and national evidence, then applied retrospectively.

Results

Age 70 or over (odds ratio [OR] 3.00), chronic lung disease (OR 2.98) and high-risk cardiopulmonary exercise testing (CPET; OR 3.32) showed the strongest associations with prolonged LOS; coexisting vulnerabilities compounded risk. Postoperative delirium (7.4%; 15/204) more than doubled mean LOS in both cohorts (open: 22.2 vs 11.1 days; EVAR: 5.8 vs 3.2 days). Profiles identified 36 adults missed by an age-alone threshold; across the cohort, identifying modifiable risk factors enables CGA optimisation, while non-modifiable vulnerabilities inform anticipatory planning and shared decision making.

Conclusions

Routinely collected data can stratify patients for targeted CGA referral, directing limited resource to greatest impact. These criteria provide a local data informed framework for targeted patient selection, ensuring CGA reaches those most likely to benefit from perioperative optimisation.