BACKGROUND: Patellofemoral pain is hypothesized to result in less joint coordination variability. The ability to relate coordination variability to patellofemoral pain pathology could have many clinical uses; however, evidence to support its clinical application is lacking. The aim was to determine if vector coding's coupling angle variability, as a measure of joint coordination variability, was less for runners with patellofemoral pain than healthy controls as is commonly postulated.

METHODS: Nineteen female recreational runners with patellofemoral pain and eleven healthy controls performed a treadmill acclimation protocol then ran at a self-selected pace for 15min. 3-D kinematics, force plate kinetics, knee pain and rating of perceived exertion were recorded each minute. Data were selected for the: pain group at the highest pain reached (pain≥3/10) in a non-exerted state (exertion<14/20), and; non-exerted healthy group from the eleventh minute. Coupling angle variability was calculated over several portions of the stride for six knee-ankle combinations during five non-consecutive strides.

FINDINGS: 46 of 48 coupling angle variability measures were greater for the pain group, with 7 significantly greater (P<.05).

INTERPRETATION: These findings oppose the theory that less coupling angle variability is indicative of a pathological coordinate state during running. Greater coupling angle variability may be characteristic of patellofemoral pain in female treadmill running when a larger threshold of pain is reached than previously observed. A predictable and directional response of coupling angle variability measures in relation to knee pathology is not yet clear and requires further investigation prior to considerations for clinical utility.