**1. Introduction**

Carbohydrate (CHO) is an important source of energy throughout strenuous prolonged exercise. Premature fatigue during prolonged exercise is linked with depletion of carbohydrate stores (i.e., blood glucose and liver and muscle glycogen stores). Thus, carbohydrate consumption before and during exercise improves exercise performance compared with a fasted condition [1,2]. Muscle glycogen concentrations are directly correlated to time to fatigue during moderately strenuous exercise ranging from 60%–80% of maximal oxygen uptake (VO2max) [1]. Thus, endurance and high-intensity intermittent exercise will be adversely affected by reduced glycogen stores. During soccer matches, this would most likely occur in the second half of a game [3–5]. Soccer players with lower levels of muscle glycogen cover less distance and run at lower speeds during the last 15 min of a match [6]. Total number of sprints and markers of acceleration and deceleration capacity are reduced in the last 15 min of the

normal duration of a soccer match [7]; therefore, research that targets maintaining these performance outcomes during a soccer match is important.

The glycemic index (GI) differentiates types of carbohydrates based on how fast they cause an increase in blood glucose concentrations [8]. Some studies have indicated consumption of low-GI foods prior to exercise may improve exercise performance compared to high-GI foods [9,10]. Low-GI foods cause a lower insulemic response compared to high-GI foods. Insulin inhibits fat oxidation during exercise [11]; therefore, consumption of low-GI foods might allow increased utilization of fats, lower carbohydrate usage, and preservation of glycogen stores [12–14]. Advantages of carbohydrate ingestion with different GIs prior to prolonged endurance exercises are well documented [15–17]; however, further studies need to address the possible impact of foods with different GIs on high-intensity intermittent exercise, important for many team sports (e.g., soccer, hockey, rugby). It is important to note that, during low to moderate intensity intervals (e.g., rest and recovery times) of high-intensity intermittent exercise, a considerable amount of energy needed for exercising muscles is provided by fat oxidation [18,19]. Our previous studies have shown some metabolic benefits (i.e., lower insulin levels, higher fat oxidation, lower carbohydrate oxidation and reduced lactate levels) when low-GI meals are consumed before interval treadmill exercise programmed to simulate the repeated high-intensity intervals of a typical soccer match, but performance specific to soccer is difficult to evaluate on a treadmill [20–22]. In these previous studies, boiled lentils were compared to high-GI foods (i.e., mashed potatoes with egg whites added to match for protein), but these meals may not be typical before matches for soccer players (or other athletes involved in sports with high-intensity intervals). Endurance athletes often consume sport nutrition bars [23] and surveys of youth soccer players indicate that about 37% consume food, such as sport nutrition bars, up to 1 h before games in an attempt to improve performance [24]; however, the effectiveness of a sport nutrition bar for soccer performance has never been evaluated. From a practical point of view, using sport nutrition bars (high-CHO) can be considered when time is limited before the start of competition. Therefore, the purpose of the current study was to evaluate low- and high-GI sport nutrition bars, consumed before and at half time on metabolism and performance during a soccer-specific field test, which incorporates skills important for soccer performance (i.e., agility running, ball dribbling, kicking accuracy, and ball heading) [25]. We hypothesized that a low-GI sport nutrition bar would be superior to a high-GI sport nutrition bar to improve performance and metabolic responses when consumed before and during a simulated soccer match.

#### **2. Materials and Methods**

#### *2.1. Participants*

Eight male recreational soccer players participated in this study. Their mean ± standard deviation for age, body mass, and predicted maximal oxygen uptake values were 30 ± 7 years, 76.6 ± 8.6 kg, and 56.5 ± 2.5 mL/kg/min, respectively. The University of Saskatchewan Biomedical Research Ethics board approved the study protocol, and all participants signed a consent form before the study began. The approval number is 12–33. The approval date was February 21, 2012.
