# Identification of characteristics of patients who choose operative versus non-operative management after ACL injury: a latent class analysis

## TL;DR

This study identifies three distinct groups of patients with ACL injuries, showing differences in their likelihood of choosing surgery and returning to sports.

## Contribution

The study introduces a novel method using latent class analysis to identify subgroups of ACL injury patients based on clinical, psychological, and social factors.

## Key findings

- Three patient subgroups were identified with differing probabilities of undergoing ACL reconstruction.
- Class 2 patients had nearly twice the probability of choosing ACLR compared to Class 3.
- Class 3 patients showed the highest return to preinjury activity levels despite lower surgical rates.

## Abstract

After an anterior cruciate ligament (ACL) rupture, characteristics of patients that may benefit from having ACL reconstruction (ACLR) versus undergoing non-surgical treatment are still largely not known. Identifying patient subgroups may help clinicians and researchers better understand the unique needs of individuals within common clinical profiles. The primary aim of this study was to identify subgroups of patients with an ACL injury, based on patient characteristics, psychological and social factors, and the extent of the initial injury, and to evaluate subgroup characteristics based on undergoing ACLR versus non-surgical management. The secondary aim was to compare subgroups on their knee function and return to sports outcomes.

A total of 275 participants (age 15–40, 48% male) with an acute ACL injury treated with usual care, either ACLR or not, from the Natural Corollaries and Recovery after ACL injury (NACOX) multicenter longitudinal cohort study were included. At two years after injury, 169 (62%) had undergone ACLR. The analysis uncovered 3 latent classes (entropy = 0.91, Akaike’s information criterion (AIC) = 12125.2, Bayesian information criteria (BIC) = 12302.4, sample size adjusted BIC = 12147.0, Vuong-Lo-Mendell-Rubin (VLMR) likelihood ratio test=-6085.8, VLMR p-value = 0.5). Participants in Class 2 (n = 155) had 1.8 times (hazards ratio (HR):1.814, 95% confidence interval (CI) 1.293–2.545, p < 0.001) higher probability for having an ACLR compared to Class 3 (n = 98). There were no statistically significant differences in patient reported knee function (International Knee Documentation Committee (IKDC) score) at one- or two-years follow-up between the classes. A higher proportion of participants in Class 3 returned to same or higher activity level compared to Class 1 (n = 22) and 2.

The three distinct subgroups of patients after ACL rupture differed in probability to undergo ACLR and return to activity after injury at one and two years after injury. There was no difference between subgroups in patient-reported knee function at one or two years after injury. Clinicians should consider the differentiating characteristics between subgroups along with the goals of their patients when engaging in shared decision-making about surgical or nonsurgical management.

Three patient subgroups were uncovered with significant differences in the decision to undergo ACL reconstruction or non-operative management, as well as in their return to preinjury activity level by two years. Individuals in Class 2, who were the youngest and had the highest pre-injury activity level, self-reported knee function, and motivation to return to sport had a nearly 2 times higher probability for selecting to undergo ACL reconstruction. Individuals in Class 1 made up a small percentage of the subgroups, were predominately male, injured during contact sports, and had the lowest motivation to return to sport and knee self-efficacy, whereas individuals in Class 3 were the oldest, had the highest percentage of females, lowest percentage of contact injuries, primarily participated in IKDC level III sports preinjury and had the highest motivation to return to preinjury activity levels and the highest proportion of participants who did return to the same or higher activity level.

Three patient subgroups were uncovered with significant differences in the decision to undergo ACL reconstruction or non-operative management, as well as in their return to preinjury activity level by two years.

Individuals in Class 2, who were the youngest and had the highest pre-injury activity level, self-reported knee function, and motivation to return to sport had a nearly 2 times higher probability for selecting to undergo ACL reconstruction.

Individuals in Class 1 made up a small percentage of the subgroups, were predominately male, injured during contact sports, and had the lowest motivation to return to sport and knee self-efficacy, whereas individuals in Class 3 were the oldest, had the highest percentage of females, lowest percentage of contact injuries, primarily participated in IKDC level III sports preinjury and had the highest motivation to return to preinjury activity levels and the highest proportion of participants who did return to the same or higher activity level.