As a result of the ever-increasing use of imaging and clinical chemistry, symptom-oriented research has lost ground in many areas of clinical medicine. In emergency medicine, the importance of symptom-oriented research is obvious, as the three major tasks (triage, work-up and disposition) are still under-investigated. Scientific progress is closely linked to the analysis of readily available information, such as the patients’ symptoms.

A decade ago, there were more questions than answers. Therefore, we describe the state of the evidence and the importance of symptoms for decisions at triage, during work-up and for disposition. Recent advances in each field focusing on symptoms as predictors of outcome and/or diagnosis are shown. Finally, future directions of research regarding novel triage tools, efficient work-up and evidence-based disposition are discussed. Symptom-oriented research has been a driver for medical progress for centuries, and re-focusing on patient-centred clinical research will strengthen this field in the future in order to support smarter medicine.

INTRODUCTION: We sought to identify baseline and periprocedural variables affecting hospital length of stay (LoS) in patients undergoing transcatheter aortic valve implantation (TAVI).

METHODS: Data on 304 consecutive patients undergoing TAVI at a single centre between August 2008 and December 2017 were collected prospectively. All patients underwent a complete clinical, echocardiographic and laboratory evaluation including a comprehensive frailty assessment at baseline. LoS was defined as the number of in-hospital days after the TAVI procedure during the index hospitalisation until the time the patient left the hospital for home or a rehabilitation clinic.

RESULTS: The mean LoS was 10.4 ± 7.1 days (median 8, interquartile range 5–12) with a significant trend towards shorter LoS over time (p <0.001). Patients discharged directly home were more likely to have shorter LoS (p = 0.007). All periprocedural complications were significantly associated with prolonged LoS (p <0.05 for all). Multivariate analysis showed an independent association between LoS and emergency admission (beta 3.24 ± 1.56, p = 0.039), baseline gait speed (beta: 0.39 ± 0.16, p = 0.018), baseline serum C-reactive protein (CRP, beta 0.14 ± 0.04, p = 0.001) and subclavian access (beta 8.27 ± 2 .9, p = 0.005). Gait speed and serum CRP remained significant determinants of LoS even after adjustment for periprocedural complications and patients’ discharge destination.

CONCLUSION: Baseline gait speed and serum CRP are significant independent determinants of LoS after TAVI.