Pre-operative Functional Cardiovascular Reserve Is Associated with Acute Kidney Injury after Intervention

Saratzis A, Shakespeare J, Jones O, Bown MJ, Mahmood A, Imray
NIHR Leicester Cardiovascular Biomedical Research Unit, University of
Leicester, Leicester, UK. Electronic address:

Eur J Vasc Endovasc Surg. 2017 May;53(5):717-724. doi:
10.1016/j.ejvs.2017.01.014. Epub 2017 Mar 18.

BACKGROUND: Acute kidney injury (AKI) is a common complication after endovascular
intervention, associated with poor short and long-term outcomes. However, the
mechanisms underlying AKI development remain poorly understood. The impact of
pre-existing cardiovascular disease and low cardiovascular reserve (CVR) in AKI
is unclear; it remains unknown whether AKI is primarily related to pre-existing
comorbidity or to procedural parameters. The association between CVR and AKI
after EVAR was therefore assessed.
METHODS: This is a case control study. From a database of 484 patients, 292
undergoing elective endovascular aneurysm repair (EVAR) of an infrarenal
abdominal aortic aneurysm (AAA) in two tertiary centres were included. Of these,
73 patients who had developed AKI after EVAR were case matched, based on
pre-operative estimated glomerular filtration rate (eGFR; within
5 mL/min/1.73 m2) and age, with patients who had not developed AKI.
Cardiopulmonary exercise testing (CPET) was used to assess CVR using the
anaerobic threshold (AT). Development of AKI was defined using the Kidney Disease
Improving Outcomes (KDIGO) guidance. Associations between CVR (based on AT
levels) and AKI development were then analysed.
RESULTS: Pre-operative AT levels were significantly different between those who
did and did not develop AKI (12.1±2.9 SD vs. 14.8±3.0 mL/min/kg, p < .001). In
multivariate analysis, a higher level of AT (per 1 mL/min/kg) was associated with
a lower odds ratio (OR) of 0.72 (95% CI, 0.63-0.82, p < .001), relative to AKI
development. A pre-operative AT level of < 11 mL/min/kg was associated with
post-operative AKI development in adjusted analysis, with an OR of 7.8 (95% CI,
3.75-16.51, p < .001). The area under the curve (receiver operating
characteristic) for AT as a predictor of post-operative AKI was 0.81 (standard
error, 0.06, 95% CI, 0.69-0.93, p < .001).
CONCLUSIONS: Poor CVR was strongly associated with the development of AKI. This
provides pathophysiological insights into the mechanisms underlying AKI.