Personalised Multicomponent Exercise Programme in Peripheral Arterial Disease

Peripheral arterial disease (PAD) is characterised as an atherosclerotic disease, most common in the lower limbs (aortoiliac, femoropopliteal, and infrapopliteal arterial segments), which causes a decrease in blood flow to the areas adjacent to and posterior to the affected area. Intermittent claudication (IC) is the most common symptom in this disease that appears with exertion and relieves with rest, causing fatigue, cramps, discomfort, or pain in the lower limbs due to limited blood flow to the affected muscles. Supervised physical exercise has emerged as the first line of intervention in improving the symptoms of intermittent claudication and disease progression, and in the last decade there has been an exponential increase in the use of wearable technologies to monitor dose-response. However, the approach used is still simplistic because it is not personalised. In other words, patients with similar diagnoses and symptoms get the same treatment, without personalising the stimulus according to their exercise responses and level of adaptation. With this in mind, this study aims to monitoring the real-time response of a multicomponent exercise programme (cardiovascular and resistance training) to personalise the dose-response, and use artificial intelligence models to gather and analyse vast amounts of data towards grouping/differentiating based on individual responses. The main hypothesis is that a supervised multicomponent exercise programme will improve the functional capacity of patients with PAD in a cluster personalised approach.

Framework: Continuous monitoring of training allows for a better understanding and adjustment of the proposed objectives, based on the physiological responses provided. With the use of wearables increasing significantly and emerging as the main trend since 2016, near-infrared spectroscopy (NIRS) has been gaining emphasis as method for assessing muscle oxygen saturation (SmO2) and is an important tool in exercise monitoring. In the field of PAD, the use of NIRS is also fundamental as it allows the oxygen available in the muscle to be visualised in real time. NIRS technology makes it possible to observe the response to exercise in detail, in terms of deoxygenation and reoxygenation, which are essential analysis processes since PAD is characterised by oxygen insufficiency and the responses are highly individualised. In conjunction, the use of high-precision accelerometry (1600 Hz) helps to identify all the movements made by analysing the responses in the different axes. Based on the considerable gap in the literature regarding the effect of continuous monitoring and personalisation of the dose-response in this population, this study aims to assess the effect of a multicomponent cardiovascular and resistance training programme, personalised, and continuously monitored through heart rate (HR) and SmO2. The experimental design comprises the following phases: i) programme design; ii) sample recruitment; iii) initial evaluation of the experimental group (Supervised exercise training) and the control group (Usual care) (M0 - 0 week); iv) personalisation of the exercise load; v) programme implementation; and vi) final evaluation of both groups (M1 - 12 weeks after). Sample: Patients with PAD will be recruited at the Local Health Unit of Trás-os-Montes and Alto Douro, by the Angiology and Vascular Surgery Service. Participation in the study will be voluntary and written informed consent will be obtained from each patient. After the baseline assessments patient will be randomised to 1 of 2 groups, using a random number program with blocking to assure that the group sizes do not differ by \>2. Intervention: There will be two study arms: i) intervention arm - 12 weeks of supervised exercise (cardiovascular and resistance training) three times a week; and ii) comparator arm - a usual care group that will be instructed according to standard lifestyle modification but without specific recommendations about the exercise programme. Power and sample size calculations: Power analysis performed in G\*Power software (using ANOVA repeated measures within-between factors), version 3.1.9.7 (Institut für Experimentelle Psychologie, Düsseldorf, Germany) for an effect size of 0.25, an α of 0.05, and a power of 0.80 (1-ß) showed that 34 patients would be needed to achieve statistical significance. Considering the potential loss to follow-up, an attrition rate of 10% will be added to the calculated n, totalling 38 patients to be recruited. Statistical analysis: Statistical analyses will be carried out using JASP software, version 0.18.1 (JASP Team, 2023) and the figures produced in GraphPad Prism software, version 8.0.1 (GraphPad Software, San Diego, California).