Accumulating evidence indicates that oculomotor functions are disrupted at the time of concussive injury or shortly thereafter. Rapid, reliable instruments to assess oculomotor functions, particularly on the field of play, have been developed for this purpose. These instruments are used in athletic populations, sometimes in isolation, to either make or exclude a diagnosis of concussion. The King-Devick (KD) is a measure of rapid number naming used in the evaluation of sports-related concussion (SRC). Recent data from the Canadian Football League and Rugby Union suggest that the KD should not be used as a standalone measure of SRC. The purpose of this study was to examine the diagnostic utility of the KD in professional ice hockey players from the National Hockey League (NHL).
The King-Devick Test (KD): The KD assesses oculomotor functions and information processing speed. Identifying numbers scanning from left to right, players must quickly name a series of numbers in three visual displays that become increasingly disorganized. KD interpretations guidelines identify an acute performance score ≥1 seconds slower than baseline as a Test Failure, which may indicate that a concussion has occurred. ≥1 errors during acute evaluation = test failure. Participants & Procedures: NHL players suspected of having a concussion by medical staff and/or exhibited visible signs of possible concussion following a hit to the head or upper torso were evaluated with the KD. Players who were evaluated and not diagnosed with concussion served as Active controls. A small group of non-Active control players was also tested for inclusion in the present study for comparison. Data were extracted if the player was English-speaking and had valid baseline and acute KD evaluations. Concussed players were only included if they had an acute KD assessment within 2 days of injury. Data Analytic Strategy: 3x2 (Group x Time) ANOVA was used to examine baseline to acute test score changes between concussed athletes, Active controls, and non-Active controls. We assessed within-group change from baseline to acute evaluation with follow-up paired sample t-tests. Chi-square analyses were used to evaluate significant group differences in acute test failure. We calculated Sensitivity, Specificity, PPV and NPV for a series of cut points. Receiver operating characteristics (ROC) curve analysis with Youden’s statistic was used to identify optimal cut point.
1605 players were evaluated with the KD at baseline. ANOVA revealed a significant group by time interaction, F(2,137) = 8.77, p <0.001, partial eta2 = 0.11. Concussed players revealed a decline in performance from baseline to acute evaluation t(52) = 3.05, p <0.01, d=0.42, while Active controls significantly improved, t(75) = 2.05, p <0.05, d=0.24. This difference in performance between concussed players and Active controls was found irrespective of whether the baselines were obtained 1 year or 2 years prior to post-injury testing. No significant change between baseline and acute testing was observed for non-Active controls. ROC curve analysis showed an optimal cut point approaching a 1-sec increase from baseline the acute evaluation.
Consistent with previous studies, our data suggest that the KD is useful in differentiating concussed and not concussed athletes acutely. However, the relatively low predictive values indicate that a decline in KD performance should not be used as a standalone measure to diagnose concussion. The fact that a difference in performance between concussed players and Active controls was found regardless of whether baselines were obtained 1 or 2 years prior to post-injury testing calls into question if baselines should be required on a yearly basis. A next step is to examine whether the KD adds unique variance in predicting a diagnosis of concussion when used in combination with the SCAT5 or other measures that are commonly used as part of a comprehensive suspected concussion evaluation.
- The KD is useful in differentiating concussed and not concussed athletes acutely
- KD performance should not be used as a standalone measure