Interobserver reliability of the pivot shift test: A modified classification improves agreement

TL;DR

This study shows that a modified classification of the pivot shift test improves agreement among doctors when assessing knee instability in ACL injuries.

Contribution

The novel contribution is a modified pivot shift test classification that increases interobserver reliability by combining grades II and III into a single high-grade category.

Findings

The modified classification achieved substantial agreement (kappa 0.73) compared to fair agreement (kappa 0.39) with the classic classification.

Combining grades II and III into a high-grade category improved agreement from poor/fair to substantial.

The modified classification is recommended for clinical and research use due to better reliability.

Abstract

The pivot shift test evaluates the anterolateral rotational instability of the knee in patients with anterior cruciate ligament (ACL) injuries. The aim of this study was to evaluate the interobserver reliability of the classic pivot shift test and a modified classification.

An interobserver reliability study involving 4 observers and 17 patients with high suspicion of ACL injury. Observers were blind to diagnostic images and independently evaluated each patient. Kappa–Fleiss was used to assess the pivot shift test agreement between observers. Interobserver reliability was calculated for the classic classification (grades I, II and III), as well as for a modified classification in low‐grade (I) and high‐grade pivot shift (II and III). Kappa agreement was categorised as poor (<0.00), slight (0.00–0.20), fair (0.21–0.40), moderate (0.41–0.60), substantial (0.61–0.80) and perfect (0.81–1.00).

Patients had mean age of 29 years (range: 24–32) with similar sex distribution (male: 53%). The global interobserver reliability assessment favoured the modified (0.73) over the classic classification (0.39). Agreement for the classic categories (negative: 0.87, grade I: 0.61, grade II: 0.06 and grade III: 0.34), was surpassed by the modified approach (negative: 0.87, low‐grade: 0.61 and high grade: 0.73).

Our study demonstrated that the modified pivot shift test grading improves interobserver reliability compared to the classic classification. By combining the grade II and III categories into a single high‐grade category, we transformed poor and fair agreement into substantial agreement. The modified pivot shift classification can be utilised in both clinical practice and research, particularly due to its clinical implications for decision‐making in patients with grade II and III pivot shifts, which may be similar.

Level I.