A 26 year physiological description of a National Hockey League team

Applied Physiology, Nutrition, and Metabolism, August, 2008 by H.A. Quinney, Randy Dewart, Alex Game, Gary Snydmiller, Darren Warburton, Gordon Bell

Anaerobic power

From 1979 to 1994, all subjects performed a 30 s Wingate anaerobic power test using a resistance of 0.095 kg x kg body [mass.sup.-1] on a Monark cycle ergometer. From 1995 to 2005, the anaerobic power test was identical in set up, resistance setting, and starting instructions as the Wingate protocol, except that it was modified so that the players were required to perform 4 repeats of 5 s maximal effort sprints with 10 s of active recovery in a 60 s period. This modification was made to the traditional Wingate testing protocol in an attempt to be more sport specific and more closely simulate the intermittent, high-intensity skating patterns of actual game performance. Research in our laboratory has shown that the peak 5 s power output (PO) for hockey players was not different using these two different protocols. Thus, only the peak 5 s PO was used for comparison between years for our whole data set (26 years) to represent peak 5 s anaerobic power and was expressed in watts per kilogram (W x [kg.sup.-1]).

Cardio-respiratory fitness

From 1979 to 1983, maximal aerobic power was predicted from a submaximal [PWC.sub.170] protocol on a Monark cycle ergometer as described by Sjostrand (1947). These scores were eliminated from our data set because starting in 1984 to the end of the data collection, a metabolic measurement system was used to directly assess peak oxygen consumption (V[O.sub.2 peak]). The exercise testing protocol was conducted on a Monark cycle ergometer that increased power output in ~37 W increments until volitional exhaustion. V[O.sub.2 peak] was defined as the highest V[O.sub.2] attained during the test. Heart rate was measured with either an electrocardiograph (Cambridge, Model VS4) or a heart rate monitor (Polar Electro, Finland).

Statistical analyses

Descriptive statistics were used to describe the various physiological parameters reported in the results. Trend analysis was used to represent any change in the fitness parameters over the 26 year period. To do this, trend lines for each fitness parameter over time were generated and the coefficient of determination ([R.sup.2]) values are reported for each trend line. Note that some of the players in the data set were with the team for more than 1 year, but were retained in the data set for each season they were with the team. The seasons were also grouped according to successful or non-successful years and position (goal, forward, defense). Analysis of variance was used to evaluate differences between successful year criteria and the different positions and a Newman-Kuels multiple-comparison procedure was used to further investigate any significant F-ratio using Statistica 7.1 (Statsoft, Inc, Tulsa, Okla.). Only active players on the roster for each year were included in the analyses. The [alpha] level was set a priori at p < 0.05 for all analyses.

Results

Longitudinal changes in fitness

The trend in the mean age of the NHL players within a season fluctuated over the 26 year period (see Fig. 1) with an overall mean age of 24.5 y. Body mass, height, and BMI showed an increase over the 26 year period (Figs. 2, 3, and 4). Sum of skinfolds showed no consistent change over the period of the study (Fig. 5). The trend for absolute V[O.sub.2] peak showed an increase over the 26 years with the largest increases observed during the 1989-1993 period and then remained relatively constant since that time point (Fig. 6). Although relative V[O.sub.2 peak] showed increases between the years of 1989 and 1993, the overall trend showed no increase over the 26 year period, which is likely due to the increase in body mass during this same period of time (Fig. 7). Relative peak 5 s anaerobic PO exhibited an increase over the 26 year period (Fig. 8). Muscular endurance of the abdominals (curl-ups; Fig. 9) and combined grip strength (Fig. 10) showed only a small increasing trend, whereas trunk flexibility (sit and reach) remained relatively unchanged over the 26 year period (Fig. 11).