Validation of Visual Examination During Stair Descent and Single Leg Squat Following Knee Arthroscopy: A Pilot Study
Translated title
Validering af visuel inspektion ved trappegang og et bens squat efter knæartroskopi: Et pilot studie
Author
Hansen, Jakob
Term
4. semester
Education
Publication year
2016
Submitted on
2016-06-02
Pages
16
Abstract
Formål: At undersøge om fysioterapeuter pålideligt kan vurdere bevægelseskvalitet ud fra simple videooptagelser under krævende opgaver som at gå ned ad trapper og lave en enbensknæbøjning, og hvordan disse vurderinger stemmer overens med 3D‑bevægelsesanalyse. Metode: Tre artroskopipatienter og 12 raske personer udførte trappenedstigning og enbensknæbøjning. Vi indsamlede 3D‑bevægelsesdata for krop, bækken, knæ og fod (kinematik) samt knæbelastning (kinetik), herunder det eksterne knæadduktionsmoment (KAM, en indikator for belastning på knæets inderside/mediale side) og knæets fleksions/ekstensionseffekt. To‑dimensionelle videoer blev uafhængigt bedømt af tre fysioterapeuter. Vi beregnede konsistens for samme bedømmer over tid (intra‑rater‑reliabilitet) og enighed mellem bedømmere (inter‑rater‑reliabilitet) med AC1‑aftalestatistikken, hvor højere værdier betyder større enighed. Enigheden mellem de visuelle vurderinger og 3D‑målingerne blev også beregnet med AC1. Resultater: Når den samme fysioterapeut vurderede videoerne to gange, var enigheden næsten perfekt (AC1 = 0,91–0,94). Mellem fysioterapeuter var enigheden lav i gennemsnit (kombineret middel AC1 = 0,29) og varierede meget på tværs af par (bedste par middel ≈ 0,55; laveste ≈ 0,09). Sammenligningen mellem visuelle vurderinger og 3D‑målinger gav en moderat gennemsnitlig enighed (kombineret middel AC1 = 0,50) med stor variation. For de enkelte bedømmere spændte enigheden med 3D fra god (AC1 ≈ 0,72) til ringe (AC1 ≈ −0,04). Enigheden mellem de samlede visuelle vurderinger og KAM var kun let til moderat, hvilket tyder på, at visuelle vurderinger ikke afspejler den mediale knæbelastning særlig godt. Konklusion: Visuel video‑vurdering var meget konsistent for den enkelte bedømmer, men kun let konsistent mellem bedømmere ved trappenedstigning og enbensknæbøjning. Enigheden med 3D‑bevægelsesanalyse var moderat og afhængig af bedømmeren. Overordnede visuelle vurderinger stemte ikke godt overens med KAM og ser derfor ikke ud til at fange den mediale knæbelastning.
Purpose: This study examined whether physiotherapists can reliably judge movement quality from simple video during demanding tasks like walking downstairs and performing a single‑leg squat, and how those judgments compare with laboratory‑grade 3D motion analysis. Methods: Three arthroscopy patients and 12 healthy volunteers performed stair descent and single‑leg squat tasks. We collected 3D motion capture to calculate movement of the trunk, pelvis, knee and foot (kinematics) and knee loading measures (kinetics), including the external knee adduction moment (KAM, an indicator of load on the inner/medial side of the knee) and knee flexion/extension power. Two‑dimensional videos were independently rated by three physiotherapists. We assessed consistency when the same rater scored videos more than once (intra‑rater reliability) and agreement between different raters (inter‑rater reliability) using the AC1 agreement statistic, where higher values indicate better agreement. Agreement between visual ratings and 3D measures was also calculated with AC1. Results: When the same physiotherapist rated the videos twice, agreement was almost perfect (AC1 = 0.91–0.94). Between physiotherapists, agreement was low on average (combined mean AC1 = 0.29) and varied widely across pairs (best pair mean ≈ 0.55; lowest ≈ 0.09). Comparing visual ratings with 3D motion capture produced a moderate average agreement (combined mean AC1 = 0.50) with substantial variation. For individual raters, agreement with 3D ranged from good (AC1 ≈ 0.72) to slight (AC1 ≈ −0.04). Agreement between overall visual ratings and KAM was only slight to moderate, suggesting visual judgments did not reflect medial knee loading well. Conclusion: Visual video assessment was very consistent within raters but only slightly consistent between raters for stair descent and single‑leg squat. Agreement with 3D motion capture was moderate and depended on the rater. Overall visual ratings did not align well with KAM, indicating they may not capture medial knee loading.
[This abstract was generated with the help of AI]
Documents