Inter-rater Reliability of the GNRB Arthrometer

Although clinical examination is commonly used to diagnose an anterior cruciate ligament (ACL) injury, the method is subjective. For many years, the arthrometer KT1000 has been used to verify clinical results, mostly due to its simplicity. While the KT1000 is no longer available for purchase, other arthrometers have been introduced, one being the Genoroub (GNRB). The latter has several automatic steps, has been marketed to have a higher reproducibility, and able to detect partial tears.

While the KT1000 and GNRB have previously been compared in several studies, it's mostly been compared in healthy subjects. The current study aim to analyse the results in a clinical environment on injured objects and compare the results both before and after ACL reconstruction. The main hypothesis was that the GNRB offers high inter-rater reliability between two examinors and can replace the KT1000 in the everyday clinical setting.

Methods Consecutively, 20 patients > 18 years of age, 12 men and 8 women, with an ACL injury scheduled for surgical reconstruction, were included. All patients were screened between August 2020-2021 at the NU Hospital Group in Trollhattan, Sweden. To be eligible, the patients had to present with a first time ACL injury verified by clinical examination with or without an magnetic resonance imaging, a healthy contralateral knee and to not have sustained multi-ligament injuries to the knee. Minor collateral ligament injuries and meniscal and cartilage injuries were included.

All patients received both spoken and written information before signing an informed consent. The study was approved by the Swedish ethical committee.

All patients were examined both pre- and postoperatively, in accordance with local standard protocol, including range of motion, knee laxity measurements (manual Lachman test, pivot shift test) The participants were examined starting with the healthy/uninjured knee. Who of the two examinors started the testing was randomly decided, however alternated at the follow-up. Clinical examination and KT1000 testing were only performed by examiner 2, given the need for experience with the KT1000 for accurate results.

Statistically, a side-to-side difference (SSD) of 3 mm was set to indicate ACL injury according to general concensus. For the dichotomous values, an SSD of ≥3 mm was considered a yes to ACL injury and \<3mm a no, respectively, for both the GNRB and the KT1000.

A power analysis was performed estimating a difference of 1 mm between examiners with an SD of 1 mm, resulting in 17 patients needed to reach a power of 80% with a significance level of 0.05.

As for the inter-rater reliability, the intra-correlation coefficient (ICC) was calculated and considered poor (\<0.5), moderate (0.5-0.75), good (0.75-0.9) or excellent (>0.9).