Yes the Tour Hgb may be explainable. In fact, the Hgb may have behaved in a completely consistent manner. (Edited: Please note that this article is the 3rd in a 4 part series attempting to look at the blood values from both perspectives. The final article considers the Reticulocyte percentage which is likely a more telling blood parameter in terms of identifying doping.)

In Article 1 of this series, we made the case that Lance Armstrong’s blood values should have reacted in a similar way to both grand tours, but they did not appear to. In article 2, we used calculations to illustrate how a Hgb might have been expected to trend down in the second grand tour.

The majority of the discussion can be summarized by this graph:

In the Giro Hgb trends down significantly while it trends up slightly in the Tour.

Expert opinion seems consistent with the conclusion of inconsistent trends but offers opposing explanations for the difference.

One thought is that something natural like dehydration secondary to something like diarrhea could skew the Hgb upwards in second half of the Tour.

The opposing view is that blood doping could cause the rise and would also account for the low normal Retic.

In this article, we consider the possibility that we missed something on our first pass.

The explanation proposed here is that the the expected drop in Hgb did in fact take place. It may simply have started before the second grand tour and reached a physiologic limit before the end of the race.

The clue to this possibility is listening to the Retic. During the Tour de France the Retic was consistently at the lower limits of normal. Usually a low normal Retic indicates that the body has a high normal total number of RBCs.

The observation of a stable Retic was touched upon in the last article and provided the justification to assume that changes in Hgb were not from changes in total RBC but from changes in blood volume.

What was not considered was the data from a few weeks before the Tour when this low normal Retic steady state was first achieved.

As it turns out, the Retic hits 0.6 on 6/16 and basically stays there until the last Tour data point.

On 6/16 the Hgb happens to be 16, a high normal value.

Now consider this graph which compares the Hgb trend during the Giro with data starting on 6/16 and continuing through the Tour.

Is this the missing drop in Hgb?

The profile of each curve is nearly identical for the first 25 days. Despite the flattening of the Tour line the overall trend is still a significant drop in Hgb as expected.

From 6/16 through the end of the Tour Retic is stable in the low normal range allowing the assumption that any drop in Hgb should be volume expansion.

In terms of estimated volume expansion, by the start of the tour the Hgb has already dropped to 14.3 indicating a 12% increase in blood volume, i.e. not dehydrated.

After the Tour starts the Hgb continues to trend down as one would expect hitting a low of 13.7, i.e. not ruling out blood doping but  not really supporting it either. The drop from 16 to 13.7 is a substantial 17% expansion in volume. This value even exceeds the change during the first grand tour of 14%.

From that point on Hgb bounces around recovering to 14.5. Is this evidence of dehydration?

Or is this evidence that there is simply a physiological limit to volume expansion, and that this limit was reached before the end of the Tour?

Consider the graph, again taking into account the 6/16 data point;

Even with the flattening of the curve blood volume expansion is still 10%.

Do the Hgb trends support manipulation?

Dehydration?

Or did the drop in Hgb begin early because of an intense build up lead up to the Tour?

If the 6/16 data point is included, the Hgb values do look more consistent with the Giro data. Dehydration also appears less likely as the overall trend is consistent with volume expansion rather than contraction.

The one problem with this reinterpretation, is that now the persistent low normal Retic values extend all the way from 6/16 through the end of the Tour. We’ll attempt to deal with this observation in the next and final article of this series.
Part 4