The present study found that a half marathon elicited substantial elevations in DOMS and markers of muscle damage (CK) and inflammation (CRP). This suggests that in recreationally trained runners, a half-marathon is an appropriate exercise challenge for investigating the effect of a dietary intervention on protection against EIMD and promotion of recovery.
Immediately after the race, there was evidence of a possibly harmful effect of BJ on DOMS in comparison to PLA. The possibly harmful effect of BJ on post-race DOMS was mirrored by a likely harmful effect on CRP at 24 h post-race and a possibly harmful effect at 48 h post-race. This dual rise in DOMS and CRP indicates that runners consuming the BJ may have suffered more muscle damage during the half-marathon than those consuming the energy-matched PLA drink, although this was not reflected by a clear difference between groups in the change in CK from pre-race to any time point post-race. The lack of a clear difference in CK between groups may be partly explained by the large differences in CK response among our runners and the small sample size. Large inter-individual variability in the release of CK from muscles in response to protocols designed to induce muscle damage has been widely reported in the literature [28, 29]. Moreover, evidence indicates that the degree of elevation in circulating CK may not directly correlate with the extent of muscle damage or loss of force [30, 31]. Therefore, circulating CK may not be a particularly sensitive marker for detecting small differences between individuals in the extent of muscle damage.
Our findings of a small, possibly harmful increase in DOMS immediately post-race and a likely harmful moderate increase in CRP at 24 h post-race were unexpected, because we originally hypothesised that the polyphenols in BJ would exhibit anti-inflammatory effects and possibly protect muscles against soreness. To our knowledge no other studies have reported on the effects of BJ on EIMD, however, our results contrast with evidence that juice and extracts of tart cherry protect against inflammation induced by long distance running [2, 14], intermittent running , and high intensity cycling [4, 8]. The effect of tart cherry beverages/supplements on DOMS is mixed with some studies reporting no effect [2, 8] and others reporting a reduction [9, 14, 15], although we are unaware of any studies that have reported an increase. The disagreement between our study and the studies of tart cherry is difficult to explain, but could reflect differences in the quantity and type of polyphenols present in the juices. In-house analysis of the BJ revealed that our dosing strategy seemed to supply greater quantities of total phenols (≈1500 mg/d) and anthocyanins (≈160 mg/d) than the tart cherry supplementation studies in runners (ranges reported in the literature; total phenols 991–1200 mg/d; total anthocyanin 66–80 mg/d) [2, 14, 15]. Although, we originally hypothesised that a high intake of polyphenols might protect against exercise-induced inflammation and soreness, it is not inconceivable that it could produce the opposite effect. The production of reactive oxygen species (ROS) by exercising muscles has been proposed to act as a brake on muscle contraction thereby limiting muscle damage during prolonged periods of contraction . BJ polyphenols may have enhanced intramuscular antioxidant protection to the extent that it reduced the braking effect of ROS on muscle contraction thus enabling the participants to run harder, but at the expense of generating more muscle damage, soreness and inflammation . We are unaware of any studies reporting improved performance coupled with increases in markers of soreness or inflammation in response to supplementation with polyphenols, however, Cobley et al.  found that acute supplementation with the antioxidant N-acetylcysteine improved performance, but elevated muscle damage in recreationally trained runners completing the Yo-Yo Intermittent Recovery Test after a muscle damaging intermittent shuttle run test. Unexpected effects of antioxidant micronutrients on muscle recovery have also been reported . Close et al.  found that supplementation with 1 g of ascorbic acid 2 h before and daily for 14 days after downhill running delayed the recovery of muscle function.
This study has several limitations. First, we did not measure any functional markers of muscle strength, thus it is impossible to determine whether the small increase in DOMS and moderate increase in CRP that we observed resulted in a greater loss of muscle force or slower recovery of muscle strength. Second, the timing of the assessment of muscle soreness after the race varied slightly between runners because they had to make their way through a busy finishing area to our mobile laboratory for assessment. It is possible that this small variation in time lapse (approximately 5 min) may have influenced the post-race assessment of DOMS. Third, we only followed our participants for 2 days after the half marathon. Whilst this is consistent with two studies of tart cherry and long distance running [2, 15], it is possible that 2 days may have been insufficient to fully capture the effects of BJ on CRP and CK both of which had not returned to pre-race levels at the end of the study. Fourth, except for prohibiting the use of antioxidant vitamins and NSAIDs, we placed no restrictions on the diet of our participants. We asked participants to complete a food diary for the duration of the race, but a number in both groups failed to complete their diaries so we were unable to accurately assess whether runners in either group changed their diet in any way that may have substantially influenced our results. It is possible that standardising the diets of the participants and restricting their intake of foods rich in polyphenols may have enhanced our ability to detect more effects of the BJ intervention. Fifth, the use of capillary blood sampling techniques limited the number of biomarkers that we could measure. The collection of larger blood samples would have enabled us to measure a more comprehensive battery of inflammatory markers and a range of markers of oxidative stress, which may have proved informative. Sixth, we adjusted the effects of our intervention for body mass, but this may not have adequately accounted for differences in lean and fat mass, which could have contributed to the observed effects. Finally, whilst our sample size was similar to comparable studies, the statistical analysis revealed a number of unclear outcomes indicating that a greater sample size was needed to increase the precision of our estimates of treatment effects.