Sensitivity of the Athlete Blood Passport for blood doping could be improved by including total haemoglobin mass (Hb, but this measure may be unreliable immediately following strenuous exercise. We examined the stability of Hbfollowing ultra-endurance triathlon (3.8 km swim, 180 km bike, 42.2 km run). 26 male sub-elite triathletes, 18 Racers and 8 Controls, were tested for Hbusing CO re-breathing, twice 1-5 days apart. Racers were measured before and 1-3 h after the triathlon. Controls did no vigorous exercise on either test day. Serum haptoglobin concentration and urine haemoglobin concentration were measured to assess intravascular haemolysis. There was a 3.2% (p<0.01) increase in Racers' Hbfrom pre-race (976 g±14.6%, mean ±% coefficient of variation) to post-race (1 007 g±13.8%), as opposed to a 0.5% decrease in Controls (pre-race 900 g±13.9%, post-race 896 g±12.4%). Haptoglobin was 67% (p<0.01) reduced in Racers (pre-race 0.48 g / L±150%, post-race 0.16 g / L±432%), compared to 6% reduced in Controls (pre-race 1.08 g / L±37%, post-race 1.02 g / L±37%). Decreased serum haptoglobin concentration in Racers, which is suggestive of mild intravascular blood loss, was contrary to the apparent Hbincrease post-race. Ultra-endurance triathlon racing may confound the accuracy of post-exercise Hbmeasures, possibly due to splenic contraction or an increased rate of CO diffusion to intramuscular myoglobin.