DR Hampton
DR Hampton
DR Hampton
Dr K K Hampton
The case:
The CADF in two documents entitled Athlete Biological Passport Expert evaluation, evaluation of
blood profile 8PY2524M36 dated 14.06.2012 and in the Athlete Biological Passport Evaluation of the
athlete's argument, dated 12.05.2014 have alleged that the biological passport shows a number of
abnormalities that suggest the use of an illegal method. The rider has responded with reports of Dr
Douwe de Boer dated 25 August 2013 and Dr Locatelli Massimo dated 1 August 2013. I have been
The second abnormality is the elevated reticulocyte values from March until August2Oil as well as
from April 2012 until the end of the Giro ltalia. This is because the reticulocyte series was abnormal
but there was no suggestion that the haemoglobin or off scores were abnormal.
It is necessary for the CADF expert panel to propose a doping scenario for the observed
abnormalities. In the case of the Giro ltalia 2012 in their summary they propose that the likelihood
described methods being due to blood doping such as the use of blood transfusion is very high. In
the original document dated 14.06.2012 they make no proposal as to an illegal method used for
raising a reticulocyte count, then in the document dated 12.05.2014 when discussing the expertise of
Dr Locastef li Massimo they suggest that the raised reticulocyte values in 2011 and 2012 very likely
represent the effect of an erythropoietin stimulation that has been introduced in 2011 and also applied
during the competitive season of the following year 20'12.
Giro 2012
The case against the athlete is that that his haemoglobin did not consistently fall throughout the race.
The samples in question are as follows:
In support of their case they quote in the letter dated 14.06.2012 seven references. lt is my
contention that their hypothesis is not supported either specifically by the references they have quoted
or by the available scientific literature. Reference 1 is in a journal I have not yet been able to access,
there does not appear to be an electronic version of it and I am currently awaiting a paper copy or
electronic copy from the British Library so cannot comment further, but will be able to do so in a short
time.
Reference 2 is a paper by Schmitt et al, this looks at the haematocrit values in a number of settings.
The numbers are small and not really applicable to the rider in this case. The first group is 8 subjects
studied for 23 hours after a t hour bout of cycle exercise. The second group is 7 subjects subjected
to 20 minutes of head down tilt. The third group is 4 subjects in a standing position for 60 minutes.
The fourth group is 10 subjects doing a maximum exedion tests on a cycle ergometer on one
occasion and the only group of relevance in this paper is 4 elite cyclists who participated in a 10 day
races. They measured haemoglobin mass by a CO-rebreathing method in 7 elite cyclists. Samples
were taken pre race and at the end of the race only. They state plasma volume of blood volume
tended to increase. Haematocrit dropped and haemoglobin concentration dropped. The starting
haemoglobin in this group of only 7 riders was 15.8 +f 0.99/dL. At the beginning of the end 14.7 +l0.7 if subjects are normally distributed the mean of +l-
and the mean +/- 2 standard deviations will cover 99% of observations. The starting haemoglobin of
15.8 -1 standard deviation of 0.9 gives a figures of 14.9 which is comfortably lower than the final
study has very limited power to prove the hypothesis held by the CADF.
Reference 4 is another paper by Professor Schumacher. In this case there are 23 subjects tested,
pre and post are a 5 day race which is only a quarter the length of the tour of ltaly. Again a detailed
statistical analysis does little to support the case made by the CADF but I would note for later in the
report that the reticulocyte count in these athletes was 1.6% on average.
Reference 5 is a paper by Gavican et al, Stability of haemoglobin mass during a 6 day UCI pro tour
cycling race. Again there are only 6 professional cyclists in the active group. lt is only a 6 day race
and samples are only taken on stages 1, 3 and
6.
the fact that the haemoglobin must fall during a21 day stage race.
Reference 6 is a review by Fellman called a hormonal and plasma volume alterations following
exercise and heat accumulation and addresses a number of sports, in particular marathon running.
Its relevance to professional cycling is not immediately obvious.
Reference 7 is another paper by Fellman et al, but this time an original article. lt is called intracellular
hydration induced by a 7 day endurance race. Again, there are only 9 athletes involved. Bloods were
taken only at the beginning and at the end of the race and over 7 consecutive days the athletes
performed 3 separate sports, namely running, cycling and cross country skiing. The distances
covered were relatively shorl and the amount of time they exercised for was very variable. The
results of this study, however, are quite interesting. In the figures in the article namely figures 1, 2
and 3, there is a relationship that shows changes in total body water, extra cellular water and
intracellular water for each of the individual riders. In fact there are only 8 data points because one
athlete's data was lost. What is clear is that even in these 8 athletes in whom you only have 2 data
points, there is immense variation with some athletes showing very significant changes in
body
composition, in particular body water, whilst others show almost no change at all or even increases.
Thus, the hypothesis that all athletes perform similarly which is implicit in the case made by the
CADF, is disproved by this paper even though I do not believe it is strictly relevant to the case in
question.
By far the most relevant paper is the paper by Roberto Coscetti et al, Haematological and iron
metabolism parameters in professional cyclists during the Giro ltalia 3 weeks stage race. In this study
the doctors took samples from all 9 riders of the Liquigas-Cannondale team on 3 occasions only,
namely at day -1 pre race, day 12 and day 22 during the race. The 2011 Giro ltalia was 3 weeks long
as usual and covered 3,524.Skm. The 9 athletesfinished 3'd, 57th,6oth 73d, g2nd, g3d, 1o4th, 13gth,
142no within the 159 cyclists, so a fair range of performances. In the summary, the authors say that
the haemoglobin red cells and haematocrit decreased during the race was stabilisation in the second
half but final values were lower than baseline. lf, however, one looks at the figure 2, which shows for
the 8 riders in whom data was available the comparison of the day -1 with day 12 and day 22
haemoglobins, it can be seen that between day -1 and day 12 all8 riders had a small but significant
fall in haemoglobin. Between day 12 and 22 there was on average still a fall in haemoglobin but
looking at the specific points it can be seen that 4 of the riders' haemoglobin fell further. In 3 of the
riders there was a small increase at day 22backtowards the starting levels but not attaining them, but
in 1 rider the day 22 haemoglobin increased such that it was similar or even above the start in
haemoglobin. Again this shows that even amongst a group of highly trained professional cyclists
there is considerable variation in the physiological responses to a 3 week 21 stage tour event and that
the use of group means rather than individual data points is mostly misleading.
What is missing amongst all this data is appropriate scientific evidence to support the hypothesis
made by the expert group that anything other than a persistent fall in haemoglobin is abnormal.
There is no study performed in professional cyclists over a 3 week stage race with sufficienfly
frequent sampling to show that their hypothesis is correct. lt is in my opinion conjecture. I have
reviewed the evidence that they have quoted and that together with the paper of Coscetti leads me to
believe that there is considerable variation amongst individual athletes and that whilst in a population
the overall means would be consistent with their hypothesis that plasma volume increases, this is not
always so in individual riders. This is because statistics apply to populations not to individuals and the
CADF argument is largely statistical in the absence of appropriate scientific evidence.
Finally, in the document of 12.05.14 no additional scientific evidence in support of their argument is
introduced. They again quote two papers, namely the review by Fellman and the paper by Schmidt.
They note the findings of the Corsetti paper as raised by Dr Locatelli Massimo but for some
inexplicable reason dismissed this, although in my opinion it is the most appropriate comparator
paper.
Using the formula for determining changes in plasma volume that is quoted in Mitton Rubini et al
paper, reference 9, which is that changes in plasma volume can be calculated by
=
PVr
PVz
x
(1 - Hctl)
(1
-Hctz)
Hblx100
Hbz
It is possible to calculate the changes in plasma volume that would account for the different
haemoglobins and haematocrits in the 4 samples in question inthe 2012 Giro. Calling the sample on
3 May sample 1, 14 May sample 2, 20 May sample 3, and 24 May sample 4, then a comparison of
sample 1 versus sample 2 shows that the plasma volume, assuming the haemoglobin mass stays
constant, has increased by 7%. Sample 1 versus sample 3 it has increased by 2o/o. Sample 1 versus
sample 4 shows a decrease of 12% and a comparison of sample 3 versus sample 4 shows a
decrease of 14o/o. lf one assumes that plasma volume in a 75kg man is 3 litres, then the rider
weighed at the time of the race 63k giving him a plasma volume of 2.5litres, a 12o/o change in plasma
volume, that is the difference between samples 1 and 4 is a total difference in plasma volume of onty
302m1.
I have eluded before to the evidence that there is variation between different individuals, even within
the same cycling team or in the same endurance event in their physiological response to exercise. In
a cycle team this is compounded further as many of the riders will have different roles and will be
performing at maximal exertion at different times in the race, for instance sprint stages are more
common in the first week when sprinters and the lead-out men will be exerting themselves maximally
and will often perform slightly less intensively during individual time trials and mountain stages. In the
second week riders who are neither sprinters nor GC contenders will either try and get in a breakaway
or be charged with chasing down a breakaway and thus their period of maximal effort may well fall in
the second week while GC contenders and climbers will only be exerting themselves maximally
during individual time trials and mountain stages and often will be protected by the team during other
stages. Consequently, one might well expect the observed differences between different people
within the same team depending on their specific role.
The rider in question had performed extremely well in the 2011 Giro. He had finished 6th overall, his
best ever result in a grand tour. He had won the white jersey and he had finished only 1 1 minutes 28
seconds behind the overallwinner. fhe2012 Giro ran from 5th -27th May.
Date
Result
5 May
Stage
6 May
Stage 2
Finished 42no
7 May
Stage 3
Finished 33'o
8 May
Rest day
9 May
Stage 4
10 May
Stage 5
Finished 47th
11 May
Stage 6
51"t
12May
Stage 7
13 May
Stage 8
17'.l.
14 May
Stage 9
60"'
15 May
Stage 10
gth
16 May
Stage
11
56th
17 May
Stage 12
24tn
18 May
Stage 13
36th
19 May
Stage 14
16th
20May
Stage 15
gtn
21May
Rest day
22May
Stage 16
27|h
23 May
Stage 17
30th
24May
Stage 18
'121"1
25May
Stage 19
26 May
Stage 20
27 May
Stage 21
Team trial
Mountain stage
Finished 28th
gth
Won by rider
1sth - 9 min 10 sec adrift
The rider overall finished 1sth in 2012 and did not ride the Tour
de France or spain that year. He was
not in the top 10 and was at the end 19 min and 58 seconds adrift of the
winner. lf we specifically
consider the period from the 21"t, 22nd, 23d, 24th May, that is the 4
days prior to the rider winning
stage 19, a climb, and the day after 4th sample on 24 May was taken, we
can see that on day 21 there
was no strenuous exercise as it was a rest day. on stage 16 the rider
was nearly 9 minutes adrift, the
next day 10 minutes adrift of the winner. The next day 121't in a sprint
finish. At this time the rider
was 20th overall and 12 minutes and 53 seconds adrift of the leader. lt
was clear at this stage that 1)
he could not win overall and 2) had not been performing at maximal
intensity in the 4 days leading up
to the mountain stage 19 on the 25th. Thus, this would further support
the reason why his plasma
volume was returning to normal as his exercise had not been as strenuous
as since he could no
longer win the tour he had decided on a strategy of a single maximal
effort to win a single stage which
he successfully did on stage 19 on 25 May, although he only won this
by 3 seconds. Following this,
again although there were only 2 stages left it can be seen that his
effort was again not maximal,
losing 9 minutes and 1O seconds on stage 20 and 3 minutes 2g seconds
on stage 21 . Consequenly,
I would argue that as the change in plasma volume is related
to exercise intensity and is reversible,
then the rider was not at maximal effort in the 4 days leading up to his
win in stage 19 and therefore
there would have been a lessening of the expansion of his plasma volume.
As we have seen in the
papers quoted above, haemoglobins can increase as the
expansion in plasma volume decreases in
the second half of the tour and it is my opinion that the increase in his haemoglob
in on 24 May is due
to a reduction in plasma volume rather than an increase in red cell mass due
to an illegal method
such as transfusion.
lf the rider had been transfused prior to sample 4 on 24 May then according
to the letter of
Damsgaard et al reference 10, which investigates the effects of blood withdrawal
and blood reinfusion
on haemoglobin and reticulocyte count amongst other variables, then reinfusion should
be associated
with not only a significant increase in haemoglobin and haematocrit but a significant fall in
reticulocyte
count. In the experts' opinion, they suggest that the reticulocyte count has not fallen significanfly
because there has not been time but in this letter following reinfusion on day O the reticulocyte
count
is significantly different at P<0.05 on days 1, 3,7 and 14. Consequenfly the suppression
in
would note that although in the second part of this argument I will have to address
the raised
1 there
were 3 samples, mean haemoglobin 15.33. Considering these
34 samples together the mean
haemoglobin is 15.39. From sample 43 dated 14.3.2011 the reticulocyte
count increases, for the rest
ot2011 there were a total of 10 samples with a mean haemoglobin of
15.16 and in 2ol2upto sample
62 including samples 56, 57, 58 and 59 taken at the time of the Giro
2012there are 17 samples with a
mean haemoglobin of 15'04. Overall, in the 17 samples in this time period
the mean haemoglobin is
15'11' Consequently, it can be very easily seen that the mean haemoglobin
before the increase in
reticulocyte counts was persistently higher than the mean haemoglobin
aftenruards. Since the
purpose of giving an illegal method such as using recombinant
erythropoietin is not to increase a
reticulocyte count but to increase the red cell mass, which is reflected
by an increase in haemoglobin
as well as an increase in reticulocyte count, in order to increase
the oxygen carrying capacity of the
blood, if the rider was using recombinant erythropoietin throughout this period
he was clearly
unsuccessful as although his reticulocyte count was increased his haemoglobin
was not, and hence
there would be no increase in performance due to increased haemoglobin
and hence increased
oxygen carrying capacity. Consequently, alternative explanations
for his increase reticulocyte counts
should be sought and I refer to the report by Dr Douwe de Boer.
athlete blood passport do not flag microdose epo doping, reference 1 1, there is
a study that shows
that small doses of erythropoietin given frequently can be given without triggering abnormalities
of the
athlete's biological passport. Ten subjects were given twice weekly intravenous irrjections of
erythropoietin for up to 12 weeks in 5 treatment groups, the results of which are shown in table
1 . lt
can be seen that there are significant increases in haemoglobin mass, there is a 10% increase
in
haemoglobin from 15.53 at baseline to 16.43 g/dl at the end, whereas the reticulo,cyte count
starts at
0.63 and finishes at 0.66 in these 9 athletes across the whole period; thus, in this study the use of
microdosing erythropoietin increased haemoglobin and red cell mass and hence oxygen carrying
capacity without increasing reticulocyte count. The CADF expert panel, however, were suggesting
in
this athlete that erythropoietin has increased reticulocyte count without increase in haemoglobin.
Certainly' my clinical expertise in the therapeutic use of erythropoietin to treat patir-.nts with anaemias,
is that therapeutic doses of erythropoietin of either human recombinant or pegylated human
significant increases in the haemoglobin. Indeed, if the erythropoietin only increased the reticulocyte
count and did not increase the haemoglobin, it would have found no clinical use whatsoever. Similar
points are made in the article in Blood in 2011 called Blood doping and its detection by Jackleman
and Lungbe which is reference 12. Again this suggests that the use of erythropoiertin either in full
dose or smaller so called microdosing, is used to increase the haemoglobin not thr-. reticulocyte count,
Summary:
I have reviewed the athlete's biological passport, the scientific evidence presented by the expert panel
and the additional scientific evidence from the Corsetti paper which is perhaps mone relevant to a 3-
week stage race and conclude that it is not possible to be comforlably satisfied thart the athlete's
increase in haemoglobin in the sample on 24 May 2}12was due to the use of an illegal method as
there is considerable variation between athletes which could account for this and arlso changes in
plasma volume are not consistent between athletes or across all the stages of the race, and in part
are dependent on race strategy and periods of exceptional exertion and periods of suboptimal
exertion.
I have also reviewed the data from the athlete's biological passport on his reticulooyte count and
related this to his haemoglobin. Although his reticulocyte count is increased the heremoglobin actually
falls during this period which is not consistent with the effects of erythropoietin even at therapeutic or
microdosing and therefore not consistent with an increase in performance by use qf an illegal method.
Dr Douwe de Boer has provided an alternative explanation that temporally fits with the changes
observed.
10
References:
1'
2.
3.
4'
5'
6'
7'
Garvican LA, Eastwood A, Martin DT, Ross MLR, Gripper A, Gore CJ. Stability
of
haemoglobin mass during a 6 day UCI ProTour cycling race. Clin. J. Med.
Off. J. Can.
Acad. Spor Med. 2010 May;20(3):200_a.
Fellmann N. Hormonal and plasma volume alternations following enduriance
exercise. A
brief review. Sports med. Auckl. Nz. 1992 Jan;13(1)37_a9.
8.
9'
10'
11.
12'
for anti-doping strategies. '1006, Haematologica, the Haematology Jourrral, 2005; g1(7).
Michael Ashenden, Clare E Gough, Andrew Garnham, Christopher J Gore,
Ken Sharpe.
Current markers of the Athlete Blood Passport do not flag microdose EpO doping.
Euro J
Appl Physiot. DOt 1 0. 1 007/s00421-O1j_1867_6.
Wolfgang Jelkmann and Carsten Lundby. Blood doping and its detection.
Btood.
September 2011, volume 11g, number g.
11
12