INTRADIALYTIC EXERCISE: A LARGE-SCALE NATIONWIDE IMPLEMENTATION STUDY

Abstract BACKGROUND AND AIMS Intradialytic exercise (IDE) is recommended for HD patients. However, this recommendation is mostly based on research conducted under optimal instead of real-world conditions, limiting generalizability and its implementation into routine clinical care. This study aims to analyze implementation outcomes of IDE in real-world conditions at a large-scale, nationwide level. METHOD After a pilot experience in a single unit, all NephroCare Portugal dialysis clinics were invited to initiate an IDE program. A national coordinator was nominated, and a nurse and a doctor were selected in each unit as local coordinators. A simple exercise protocol was designed to be easily applied by current dialysis staff (though some units benefited from exercise science student internships). The IDE protocol includes a bout of aerobic exercise (cycle ergometer) and lower limb resistance exercises (ankle weights). The RE-AIM (reach, effectiveness, adoption, implementation and maintenance) framework was used to study clinical implementation over the first year. For each RE-AIM dimension, specific implementation outcomes were adapted to IDE. Effectiveness measures included safety (incidence of intradialytic adverse events over 1 year) and physical function at baseline and at 1 year (sit to stand 30, 8-foot up & go, handgrip strength, sit to stand 5 and single leg stance). For safety measures, IDE group was compared to a group of patients that refused IDE. Physical function measures were only applied to IDE patients, and so comparisons were made between low- and high-frequency exercise groups. RESULTS Adoption: 21 dialysis units (58.3%). Reach: 1270 eligible patients (55.8%). Main reasons for noneligibility were physical/cognitive incapacity (50.8%) and cardiovascular risk (34.9%). Eligible patients were younger (P < 0.001) and had a better health status (lower comorbidity index: P < 0.001; lower prevalence of diabetes: P < 0.001; lower fat tissue index: P < 0.001; and a higher lean tissue index: P < 0.001). 811 (63.9%) committed to the intervention. Compared with non-IDE patients, IDE patients were younger (<0.001), had a lower dialysis vintage (<0.001), a lower comorbidity index (P < 0.001) and a higher lean tissue index (P = 0.035). Implementation: adherence to exercise sessions was 75.0% ± 19.7%. In 77% of the 50 356 HD treatments, exercise was performed as prescribed. Non-performed exercise sessions (n = 9768) are mostly justified by patient refusal (61.5%) and pain (8.4%). Patients performed 2.2 ± 0.6 exercise sessions/week achieving 86.3 ± 29.0 min/week. Maintenance (setting level): none of the clinics interrupted IDE. Maintenance (patient level): attrition rate was 57.2% mainly due to voluntary withdrawal (52.4%). Comparing to voluntary withdrawals, completers were mainly males (P < 0.001) and had a lower dialysis vintage (P = 0.007), and higher lean tissue index (P = 0.023). Effectiveness: IDE (n = 347) and non-IDE patients (n = 394) were compared. Total incidence of adverse events was lower in IDE, but no significant differences were found (P = 0.808). The individual analysis of each adverse event demonstrated no significant differences for cramps, hypotension, needle dislodgement and other adverse events. Overall physical function improved in IDE (P < 0.001), despite a slight reduction in handgrip strength (P = 0.001). A within-group (low- versus high-frequency exercise) effect was observed, highlighting a deterioration in handgrip strength in the low-frequency group (P = 0.002), whereas no change was observed in the high-frequency group (P = 0.097). CONCLUSION Large-scale implementation of IDE is a realistic and safe way to promote physical activity in HD patients with benefits on physical function. Yet, to optimize its generalizability, strategies to increase patients´ acceptability and long-term adherence are needed.