Bloomer 2005.

Methods	Randomised controlled trial (parallel design)
Participants	Setting: laboratory; USA n = 20 weight‐trained men (10 in the experimental group; 10 in the placebo group) Antioxidant group: mean age 24 (SEM 1.1) years Placebo group: mean age 26.2 (SEM 2.0) years Inclusion/exclusion criteria All participants had been weight training their lower body for a minimum of 12 months prior to testing and demonstrated a minimum strength of 1.5 times their body weight in the barbell back squat exercise. All participants were free of the orthopaedic and metabolic conditions that would have affected the variables of measurement.
Interventions	Intervention Astaxanthin ‐ BioAstin; 1732 mg safflower oil; haematococcus algae extract (contains 4 mg astaxanthin and 480 mg lutein) Placebo 1732 mg safflower oil 2 capsules per day Duration 3 weeks before and 96 hours after
Outcomes	PRIMARY Delayed onset muscle soreness in the dominant leg during knee extension using a 10 cm visual analogue scale where 0 is "no pain" and 10 is "unbearable pain" SECONDARY Muscle performance 1 RM concentric strength in the knee extension Isometric knee extensor action was performed using the modified York barbell (York, P.A. knee extension/flexion machine. The cable length was adjusted so that the knee was at 90 degrees flexion. Mean dynamic force was determined in the knee extension exercise using a Body‐Solid knee extension machine interfaced with a Fitrodyne dynamometer (Fitronic, Bratislava, Slovakia).
Exercise type	York knee extension machine 10 sets of 10 repetitions at 85% of 1 RM
Sources of funding	The study was supported by Cyanotech Corp. Kailua‐Kona, HI, and IMAGINNutrition, Inc. Laguna Niguel, CA
Notes	Authors were contacted on 3 October 2013 to request raw data for delayed onset muscle soreness and maximal voluntary isometric contraction and responded on 1 November 2013
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	No details in manuscript Authors were contacted on 24 May 2016; response: "Likely via coin flip or random number selection"
Allocation concealment (selection bias)	Low risk	No details in manuscript Authors were contacted on 24 May 2016; response: "Blinding code retained by person not associated with research and/or provided in sealed envelope"
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Double‐blind
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants successfully completed testing with 100% compliance
Selective reporting (reporting bias)	High risk	No published protocol available All outcomes reported at all time points Adverse effects of antioxidant supplementation were not reported
Other bias	Unclear risk	No details on whether participants were asked to refrain from using other supplements or anti‐inflammatory medication