Decision-making preceding transcatheter aortic valve implantation in frail older adults : Vulnerable autonomy, novel frailty scoring and clinical outcomes important to treatment strategy. A mixed method study

Introduction: Transcatheter aortic valve implantation (TAVI), also known as transcatheter aortic valve replacement (TAVR), is a new technique for treating severe, symptomatic aortic stenosis. The mean age for TAVI patients is over 80 years and most of the patients have comorbidities and frailty. Some patients may be too frail and have a short life expectancy even after the intervention, and will benefit more from a palliative approach. Established surgical scores have limitations in determining risk among candidates for TAVI. Assessment of frailty might help to estimate the mortality risk and identify patients likely to benefit from treatment. On the other hand, there is a risk of ageism and undertreating older adults. How can we select the right patients for the procedure? Patient autonomy is a leading principle in bioethics and a basis for shared decision-making. In the light of the increasing focus on patient-centred care, this project has explored TAVI patients’ experience of the decision-making process preceding intervention. This thesis consists of three studies focusing on the decision-making process prior to TAVI. Paper 1 focuses on the patient perspective, paper 2 takes the doctors’ viewpoint and paper 3 includes both perspectives. Aims: The aim of paper one was to explore conditions for autonomous choice as experienced by older adults who recently underwent transcatheter aortic valve implantation. The aim of paper two was to develop a frailty score to guide the decision for TAVI. The aim of paper three was to examine baseline characteristics and clinical outcomes important to older adults and their doctors to enhance shared decision-making prior to transcatheter aortic valve implantation. Materials and methods: We conducted a mixed method study, with one qualitative sub-study (paper 1) and two quantitative sub-studies (papers 2 and 3). All patients underwent TAVI due to severe and symptomatic aortic stenosis. The qualitative study involved semi-structured interviews of a purposive sample of ten older adults after the procedure. Analysis was by systematic text condensation. In paper 2 we conducted a prospective observational study in 82 patients ≥70 years accepted for TAVI from 2013 to 2015 and 65 patients ≥ 80 years (from a concomitant study on delirium) accepted from 2011 to 2013, giving a total of 147 patients. Prior to the procedure, a geriatric assessment (GA) was completed in 142 patients (missing data for calculating frailty score in five patients). Based on this, an eight-element frailty score with a 0–9 (least frail to most frail) scale was developed. In paper 3 we conducted a prospective, observational study of 82 TAVI patients ≥70 years (the last cohort of study 2), with two-year follow-up focusing on baseline frailty status (including cognitive deficits) and outcome measures important for shared decision-making prior to the procedure. Results: In paper 1, the median age of the included patients was 83.5 years (range 73-89 years). Even when choice seemed difficult or lacking, TAVI patients deliberately took the chance presented to them by taking into account risk assessment, ambivalence and fate. They regarded declining the treatment as worse than accepting the risk related to the procedure. The experience of being carefully advised by their doctor formed the basis of autonomous trust. This trust mitigated ambivalence about the procedure and risks. TAVI patients claimed that it had to be their decision and expressed feelings consistent with self-empowerment. Despite this, choosing the procedure as an obligation to their family or passively accepting it were also reported. In paper 2, patients had a mean age of 83 (SD 4) years, and 54% were women. The novel GA frailty score predicted two-year mortality in Cox analysis, also when adjusted for gender, age and logistic EuroSCORE (HR 1.75, 95% CI: 1.28–2.42, P < 0.001). A ROC curve analysis indicated that a GA frailty cut-off score of ≥ 4 predicted two-year mortality with a specificity of 80% (95% CI: 73%–86%) and a sensitivity of 60% (95% CI: 36%–80%), and the area under the curve was 0.81 (CI 0.71–0.90). All-cause two-year mortality was 11%. In paper 3, mean age was 83 years (SD 4.7) and 48% were women. Fifteen patients (18%) had a Mini Mental Status Examination (MMSE) score below 24 points at baseline, indicating cognitive impairment or dementia, while five patients had an MMSE below 20 points. At baseline and six months, mean New York Heart Association (NYHA) class was 2.5 (SD 0.6) and 1.4 (SD 0.6) (p