Central Blood Pressure and Variability Evaluation

The AARDVARK (Aortic Aneurysmal Regression of Dilation: Value of ACE-Inhibition on RisK) trial was designed to investigate the hypothesis that an ACE-inhibitor (perindopril) would reduce growth rate of small AAAs in a three-arm randomised placebo-controlled trial. While this didn't show a significant effect, results from an in-trial sub-study (CAVE study) demonstrated a significant relationship between central BP variability and aneurysm growth rates. This requires confirmation, in a dedicated study.

Design:

This is a multicentre prospective longitudinal observational cohort study. Each patient will have a study visit every 4 months with a minimum of 6 clinical visits where the study team will collect demographic data; psychological data (short form questionnaires that can be independently filled in); and physical data. The study team will also access routinely collected NHS cross-sectional imaging for comparison throughout the study, as quality control. No additional procedures involving ionising radiation will take place.

Study timeline:

There will be a 10-month recruitment period; all subjects will attend visits for a period of 18-24 months. The data collection will continue for 24 months in total. There will be a 6-month period of data analysis and write up.

Recruitment:

A cohort of up to 175 patients will be recruited from two trusts in London comprising four hospitals - Imperial College Healthcare NHS Trust (ICH) at St Mary's and Charing Cross hospitals, and London North West University Healthcare Trust (LNWH) at Northwick Park and Hillingdon hospitals. All visits and measurements will take place at central sites (St Mary's and Northwick Park hospitals).

Data Collection:

1. Demographic data - e.g. current prescribed anti-hypertensive medication and other cardiovascular medications, height and weight, medical history, smoking status
2. Psychological data - three short-form questionnaires that can be independently filled in by participants. Questionnaires will constitute measures of anxiety and depression (Hospital Anxiety and Depression Score [HADS]), current medication adherence (modified Voils Adherence Score), and beliefs around medication and adherence (Beliefs about Medicines Questionnaire [BMQ]).
3. Physical data - sitting mean peripheral and central systolic and diastolic BP, visit-to-visit central systolic and diastolic BP variability measured as standard deviation, coefficient of variation and variation independent of the mean will be measured using a validated arm cuff-based method. Ambulatory BP parameters will be measured using a validated monitor. USS AAA measurement will take place at bedside in supine position with outer-to-outer (OTO) calliper placement.

Primary Endpoint Analysis:

The association between central BP variability measured as standard deviation and AAA growth will be analysed using regression models where BP variability is the exposure and AAA growth is the outcome.

Random effects multilevel models where level-1 units are AAA repeated measurements nested within patients (level-2 units) will be used to estimate AAA growth. To explore non-linearity in AAA growth appropriate modelling will be used eg. linear splines.

Secondary Endpoints Analyses:

Survival analysis techniques such as Cox proportional hazards models will be used to assess the relationship between central BP variability and time taken for the AAA to reach the threshold for intervention (5.5 cm) or rupture.

Associations between haemodynamic parameters and aneurysm growth rate will be investigated. For each of the other measurements of systolic and diastolic central BP variability (coefficient of variation (SD ÷ mean) and variation independent of the mean calculated as SD÷(mean x) where x is determined empirically by curve fitting) and for other BP and haemodynamic parameters studied, the same multilevel model or generalised linear models will be used to estimate AAA growth. All analyses will be adjusted for an a-priori list of confounding variables.