RESUMO
BACKGROUND: Despite widespread use of return to sport testing following anterior cruciate ligament reconstruction, studies suggest inadequacy in current testing criteria, such as limb symmetry index calculations, to determine athletes' readiness to return to play. Recurrence quantification analysis, an emerging non-linear data analysis tool, may reveal subtle neuromuscular differences between the injured and uninjured limb that are not captured by traditional testing. We hypothesized that isokinetic torque curve data of the injured limb would demonstrate lower determinism and entropy as compared to the uninjured limb. METHODS: 102 patients (44 M, 58F, 10 ± 1 months post-anterior cruciate ligament reconstruction) underwent isokinetic quadriceps strength testing using a HumacNorm dynamometer. Patients completed maximum effort knee extension and flexion at 60°/sec. Data were post-processed with a MATLAB CRQA Graphical User Interface and determinism and entropy values were extracted. Paired-sample t-tests (α = 0.05) were used to compare data from the injured and uninjured limb. FINDINGS: Determinism and entropy values in the torque curves were lower in the injured limb than the uninjured limb (p < 0.001). Our findings indicate there is less predictability and complexity present in the torque signals of injured limbs. INTERPRETATION: Recurrence quantification analysis can be used to assess neuromuscular differences between limbs in patients who have undergone anterior cruciate ligament reconstruction. Our findings offer further evidence that there are changes to the neuromuscular system which persist following reconstruction. Further investigation is needed to establish thresholds of determinism and entropy values needed for safe return to sport and to evaluate the utility of recurrence quantification analysis as a return to sport criterion.