Clinical Sciences 875

Routine Blood Parameters in Elite Soccer Players

Authors T. Meyer, S. Meister

Affiliation University of Saarland, Institute of Sports and Preventive Medicine, Saarbrücken, Germany

Key words Abstract fere with clinical decisions in soccer players.

●
▶ athletes ▼ Upper limits of the 95 % confidence intervals
●

Downloaded by: Universität Paderborn. Copyrighted material.

▶ exercise
Often blood screening is repeatedly done in elite
●
▶ reference range
athletes although training and competition can
●
▶ plasma volume
affect its results and normal ranges for highly
●
▶ variability
active individuals are widely unknown. This
study was conducted to provide reference data in
professional soccer players. 467 male soccer
players of the 2 highest German leagues were
observed over an entire season. Venous blood
sampling was conducted 4 times in a standard-
ized manner to determine complete blood count,
AST, ALT, CK, creatinine, urea, uric acid, choles-
terol, electrolytes, ferritin, CRP and TSH.
There were significant changes during the season
in hematocrit, creatinine, uric acid (decrease),
CK, AST, urea, sodium, potassium, magnesium,
and TSH (increase). Only the changes in hemato-
crit and CK were large enough to possibly inter-
exceeded population reference ranges slightly in
AST, urea, creatinine, and potassium. The mean
intraindividual coefficient of variation was below
10 % in complete blood count (except leukocytes),
creatinine, uric acid, total cholesterol, and all
electrolytes. There seems to be a profound effect
of elite soccer training and competition on hema-
tocrit (plasma volume) and CK only. Spontaneous
variability of most parameters is too small to jus-
tify repeated sampling only for routine screening.
Trial Registration: NCT00946855
Effect of Elite Soccer Training on Routine Labora-
tory Parameters (SOCCERLAB), http://www.clini
caltrials.gov/ct2/show/NCT00946855?term = soc
cerlab&rank = 1

Introduction This deficit of knowledge is not limited to the

accepted after revision
May 18, 2011
Bibliography
DOI http://dx.doi.org/
10.1055/s-0031-1280776
Int J Sports Med 2011; 32:
875–881 © Georg Thieme
Verlag KG Stuttgart · New York
ISSN 0172-4622
Correspondence
Prof. Tim Meyer
University of Saarland
Institute of Sports and
Preventive Medicine
Campus, Bldg. B 8-2
66123 Saarbrücken
Germany
Tel.: + 49/681/302 70400
Fax: + 49/681/302 4296
tim.meyer@mx.uni-saarland.de
▼ medical care of elite athletes. Physicians of all dis-
In elite athletes, venous blood sampling to screen
for deficits and unknown disease is frequently
done. However, there is some uncertainty in the
interpretation of abnormal values that might
occur in this particular population. Theoretically,
they can either be indicative of pathology or be
caused by the physiological demands of repeated
training and competition. It is well known that
acute physical exercise, i. e., within hours before
the blood sample is taken, can lead to profound
changes in plasma volume, leukocyte counts, and
lymphocyte subgroups [7,10, 19]. In addition,
blood chemistry can be changed [22, 27]. Even
exercise-induced changes in cardiac markers tro-
ponin and brain natriuretic peptide have been
reported [14, 29]. However, the knowledge about
“chronic” changes still mainly relies upon anec-
dotal evidence and studies with small sample
size [7, 17].
ciplines often have to deal with active individuals
who spend a considerable amount of time train-
ing. Although these ambitious non-professional
athletes are not as fit as elite ones, the problem of
interpreting their blood values is about the same.
But the current literature mainly deals with acute
effects of exercise on laboratory parameters, that
is within minutes or hours after exercise, although
this does not represent the typical situation when
these athletes are seeking medical advice.
To overcome this problem and to provide “refer-
ence values” for athletes during continuous long-
term high-intensity training, professional soccer
was chosen as a discipline that contains high mus-
cular demand of different kinds (from locomotor
activity, change of directions, jumping, and from
contacts) and is characterized by frequent compe-
titions [4, 34]. The recruited professional athletes
from the highest 2 German leagues represent a

Meyer T, Meister S. Routine Blood Parameters in … Int J Sports Med 2011; 32: 875–881
876 Clinical Sciences

Table 1 Number of participating soccer players per club (rows; A – R) and

were scheduled: baseline just prior to the teams’ first training
sampling time (columns; T0, T1, T2, T3) session (July 2008), one sample in October/November 2008, one
in February/March and one in April/May 2009. The number of
club T0 (n) T1 (n) T2 (n) T3 (n) T0–T3 (n) participating players per club and sampling time is given
A 21 19 in ●
▶ Table 1.

B 18 19 22
C 18 20 21 Blood sampling and processing
D 6 21 10 Samples were taken by an experienced phlebotomist (either the
E 2 14 15
team physician or one of the investigators) in the morning (08.00
F 16 23 21 25 10
to 11.00 a. m.) after overnight fasting from an antecubital vein in
G 15 24 20 1 1
the supine position. Samples were divided into 2.7 ml EDTA
H 21 24 25 24 11
I 9 blood for the blood count and 9 ml whole blood (serum gel
J 18 20 18 tubes). Whole blood was centrifugated within 20 min after sam-
K 15 21 1 pling and serum separated from the other compounds for all
L 24 21 5 17 further determinations. Players had to fill in questionnaires
M 19 about their present health condition and other factors possibly
N 21 21 interfering with the determination and/or interpretation of the
O 26 27 24 24 18 blood parameters.
P 23 19 They covered
Q 22 27 21
▶ infections and other internal diseases within the last 4 weeks
R 16 22 20
▶ intramuscular injections within the last 2 weeks
total 216 323 240 187 40

Downloaded by: Universität Paderborn. Copyrighted material.

large sample of one of the most competitive leagues in the world.
Therefore, these athletes might delineate the upper limit of train-
ing-induced changes in laboratory parameters. It was intended to
describe the course of common blood variables over one season
and, thus, to detect the effects of continuous physical strain.
Materials and Methods
▼ port of cooled EDTA blood and serum was either carried out by
All procedures were in accordance with ethical standards of the
Helsinki Declaration from 1975 and the journal guidelines as
outlined in [13]. The study was approved by the local ethics
committee (Ärztekammer of Saarland, Saarbrücken, Germany).
Manuscript preparation was made in accordance with the
STROBE statement.
Subjects
All clubs of the 2 first German professional soccer leagues
(males) and their team physicians were invited to participate in
this study by an official letter from the head organisation Deut-
sche Fußball Liga (DFL) and, in addition, by e-mail from the first
author. 17 of 36 clubs (plus 1 club from the third league which
went down one season before study onset and maintained its
training schedule) agreed to participate. This led to the inclusion
of 532 players (52 % of the registered players during the season
2008/09). To be eligible for professional soccer in Germany, all
players have to pass a defined pre-season medical screening
examination that is designed to rule out relevant internal and
orthopedic diseases. In addition to this, club physicians were
asked not to include players with known diseases (e. g., rheu-
matic diseases, diabetes mellitus). A questionnaire to be filled in
from all players before each sample completed the medical
screening procedures to ensure sufficient health status for par-
ticipation. Each player was informed about the requirements of
the study and they gave informed consent.
General design
Players were asked to give a venous blood sample at 4 times dur-
ing the 2008/09 soccer season. The following sampling times
▶ diets within the last 2 weeks
▶ drug and supplement intake over the last 2 weeks
▶ information about individual training on the sample days and
the 2 preceding days (team training information provided by
the medical staff)
▶ information about food/fluid intake on the sample day
These questionnaires were provided in German, English, French,
Spanish, and Italian. All samples were handled within a single
laboratory (underlying external quality control 4 times per year
according to German law as well as daily internal control). Trans-
car (investigators themselves) or by a commercial transport
service that guaranteed uninterrupted cooling.
Blood parameters and determination methods
From EDTA blood a complete blood count was made (Sysmex
K-1000, Sysmex, Norderstedt, Germany) including erythrocyte,
leukocyte, lymphocyte, and thrombocyte counts, hemoglobin,
and hematocrit. From serum, the following parameters were
determined (determination method; intra- and inter-assay coef-
ficient of variation in parentheses):
aspartate-amino-transferase [AST] – (enzyme activity measure-
ment, AST, MDH, optical test; 3.5 %; 5.3 %)
alanine-amino-transferase [ALT] – (enzyme activity measure-
ment, ALT, LDH, optical test; 3.5 %; 5.3 %)
creatine kinase [CK] – (enzyme activity measurement, CK, HK,
G6PDH, optical test; 3.5 %; 5.3 %)
creatinine – (Jaffé method, kinetic measurement; 3 %; 4.5 %)
urea – (urease-GLDH-UV-test; 3 %; 4.5 %)
uric acid – (uricase method according to Trinder; 2 %; 3 %)
total cholesterol – (entirely enzymatic method according to
Trinder; 3 %; 4.5 %)
HDL-cholesterol – (photometry, 2nd and 3rd generation; 3 %; 4.5 %)
LDL-cholesterol – (photometry, 3rd generation; 2 %; 3 %)
sodium – (potentiometry, ion-selective electrode; 1 %, 1.5 %)
potassium – (potentiometry, ion-selective electrode; 2 %; 3 %)
magnesium – (photometry of magnesium complexes, Calmagit;
2 %; 3 %)
ferritin – (enzyme immunoassay; 3.6 %; 4.3 %)
C-reactive protein – (immune turbidimetry; CRP; 5 %; 7.5 %)

Meyer T, Meister S. Routine Blood Parameters in … Int J Sports Med 2011; 32: 875–881

thyroid-stimulating hormone [TSH] – (enzyme immunoassay;
2.5 %; 4.6 %)
Reference ranges were adopted from the national standard text-
book [36]. No such ranges are presented for total cholesterol and
its subfractions because target values (in the context of cardio-
vascular risk management) are more meaningful than popula-
tion means for the evaluation of blood lipids. Certain parameters
were excluded from analysis whenever one of the following con-
founders was present:
infection – leukocytes, CRP, ferritin
intramuscular injection – CK
supplementation of iodine, intake of thyroxine – TSH
intake of allopurinol – uric acid
iron supplementation – ferritin
magnesium supplementation – magnesium
creatine intake – creatinine
Statistics
The statistical analysis was performed with the package Statis-
tica 6.1. Data are presented as medians and quartiles (in figures
for longitudinal analyses) or 95 % confidence intervals (for
descriptive purposes). Lacking normal distribution for several
parameters (Kolmogorov-Smirnov test) indicated this procedure
to be adequate. Longitudinal analyses were conducted by use of
a Friedman ANOVA. Within this analysis an α-error of p < 0.05
was considered statistically significant. To quantify variability of
each parameter, the mean intraindividual coefficient of variance
(CV) was calculated. It represents the arithmetic average of all
individual CVs (defined as standard deviation of a given variable
divided by its mean value). It was intended to include as many of
the professional soccer players in Germany as possible to come
as close to a complete sample as possible. A calculation of the
necessary sample size was, thus, not carried out.
Results
▼ Repeated standardized blood sampling has never been con-
There were considerably different numbers of subjects for the 4
sampling times (● ▶ Table 1). The main reasons for this inequality
were scheduling difficulties (at the onset of the preparation
period when many squads were still incomplete; similarly
towards the end of the season when several travelling duties and
coaches’ training preferences interfered), injuries, and non-com-
pliance with the study requirements (e. g., deviation from the
required one training sessions on the day prior to sampling).
Results from non-compliant subjects were strictly eliminated
from all analyses. ●▶ Fig. 1 provides a flow sheet that illustrates
Compliance flow sheet
n = 532
Competition on pre-sampling day or early training on
sampling day
n = 522
Chronic disease
n = 520
Rehabilitation process after injury
n = 494
Less or more than one team training on day before
sampling
n = 467
Fig. 1 Flow sheet indicating the reasons for the elimination of players
from statistical analysis.
Meyer T, Meister S. Routine Blood Parameters in … Int J Sports Med 2011; 32: 875–881
Clinical Sciences 877
the reasons for eliminations. Anthropometric data of the remain-
ing 467 players (45 % of the registered players during the season
2008/09) were: age 24.9 ± 4.4 years; height 1.83 ± 0.07 m; weight
78.7 ± 9.6 kg; BMI 23.6 ± 6.9 kg/m2.
Median values, 95 % confidence intervals,
intraindividual variability
●▶ Fig. 2 displays the course of erythrocyte count, hemoglobin,
and hematocrit values together with 95 % confidence intervals.
CK results are shown in ●▶ Fig. 3. All other parameters are listed
in ●
▶ Table 2 together with their mean intraindividual coefficient
of variation (CV) between T1 and T2 as well as between T1, T2,
and T3 (number of eligible players for T1 and T2 obviously higher
than for all 3 sampling dates from T1 to T3; therefore slightly
higher mean CVs can be expected for T1–T3). The mean CVs for
erythrocytes, hemoglobin, and hematocrit were 2.6, 2.2, and 2.4
(T1 and T2), and 2.8, 2.5, and 2.7 (T1, T2, and T3), respectively.
For CK, CVs were 36.7 (T1 and T2) and 39.9 (T1, T2, and T3).
Confidence intervals of T0–T3 overlap widely with each other
and with the population reference range. T1 might serve as the
best reference sampling date because the number of players was
Downloaded by: Universität Paderborn. Copyrighted material.
the highest.
Systematic changes over the season
Over the course of the soccer season statistically significant
changes were detected in certain parameters: platelets, AST, cre-
atinine, urea, uric acid, sodium, potassium, and magnesium. The
effect size of these changes, however, was small for all variables.
Information about the longitudinal development is provided
within the figures and in the second last column of ● ▶ Table 2
together with the number of players eligible for this analysis (at
all sampling dates).
Discussion
▼
ducted over an entire competitive season in such a large number
of elite professional soccer players. The results of this study indi-
cate that routine blood parameters are robust against the con-
founding influences of regular intense training and competition.
To avoid misinterpretation of abnormal blood results, soccer-
specific reference values have to be used for very few parame-
ters: CK (vastly different), AST, urea, and creatinine (all with only
modest deviations from population reference values). A careful
application of these 95 % confidence intervals in similar team
sports (handball, basketball, field hockey) seems possible. The
results of this study may further help in demarcating a normal
range for routine blood parameters in highly active male ath-
letes who consult physicians and for whom routine blood values
are determined to rule out disease. They can thus be relevant for
any physician consulted by a highly active patient who shows
suspicious blood parameters.
An important reason for high CK values in soccer players are the
game’s specific movements with its stop-and-go character and
many direction changes which impose high eccentric biome-
chanical strain on the working muscles leading to microinjuries
and the release of CK from the cytosol [25, 39]. Typical soccer
training and competition is obviously more prone to such mus-
cle damage than training in other disciplines like swimming [24]
or cycling. Wide 95 % confidence intervals up to 1 200 IU/l are
partly due to single individuals with overproportionate CK
878 Clinical Sciences

5.8 1 400
1 327 U/I
5.6 5.71
5.07 5.63 5.64 1 200 308 U/I
1 217 U/I
5.56 107 U/I
4.51 4.92 4.91 331 U/I
5.4
Erythrocytes [×106/µL]

4.38 4.45 4.93 125 U/I

4.42 1 000 1 051 U/I 1 031 U/I
183 U/I 320 U/I
5.2 83 U/I 107 U/I
800

CK (U/L)
5.0

4.8 600

4.6 400

4.4
200
4.2
T0 T1 T2 T3 **
0
[n=215] [n=320] [n=226] [n=186] T0 T1 T2 T3
[n=211] [n=292] [n=221] [n=158]
17.5
17.0 Fig. 3 Course of creatine kinase (CK) during the season (sampling times
16.9 from baseline T0–T3). Open squares and hatched areas indicate median
16.5 15.4 16.8
15.1 16.6 and 95 % CI for each sampling date (n given at the x-axis). Open circles
13.5 16.5
Hemoglobin [g/dL]

16.0 13.7 15.1 and whiskers stand for median and quartils of those players eligible
14.9 13.4
13.6 for longitudinal analysis (n = 26) revealing a significant trend towards

Downloaded by: Universität Paderborn. Copyrighted material.

15.5
increased values over the season (p = 0.001; ** for sampling dates with
15.0 significantly different values from T0; Wilcoxon test). Numbers in the
figure refer to median (middle), upper (top) and lower (bottom) border of
14.5 the 95 % confidence interval.
14.0
13.5 II) muscle fibres might also tend to higher CK values [40]. In this
13.0 study 15 % of the players with CK values above 800 U/l were of
T0 T1 T2 T3
[n=215] [n=320] [n=226] [n=186] African origin (in contrast to 5.4 % of the entire study sample).
Interestingly, recent findings within a mixed athlete population
55
51.0 led to very similar reference intervals: 82–1 083 U/l [24] under-
46.0
lining the potential applicability of our results for athletes from
41.4 47.7 47.5
47.1 related sport disciplines.
50 42.7 42.6
43.0
39.0 38.2 The most frequently given reason for increased urea values in
Hematocrit [%]

38.9
elite athletes is their training volume associated with some
45 degree of glyconeogenesis that leads to degradation of structural
or functional proteins [12, 38]. Those amino acids that are not
utilized for energy metabolism are mainly eliminated through
40 the kidneys after metabolization to urea. Too low fluid intake
and kidney disease as alternative explanations seem unlikely –
*** *** *** even more without indications from medical history.
35 Slightly higher creatinine values than in the non-athletic popu-
T0 T1 T2 T3
[n=215] [n=320] [n=226] [n=186] lation are in line with existing literature [1]. Reasons for high
creatinine concentrations include increased muscle mass lead-
Fig. 2 Course of erythrocytes (top left), hemoglobin (top right), and ing to more creatinine turnover than usual [2, 35] and creatine
hematocrit (bottom) during the season (sampling times from baseline
intake [15]. It may be assumed that our recording of supplemen-
T0–T3). Open squares and hatched areas indicate median and 95 % CI
tal creatine intake was not complete because current reports
for each sampling date (n given at the x-axis). Open circles and whiskers
stand for median and quartils of those players eligible for longitudinal about creatine intake in athletic populations indicate more fre-
analysis (n = 39 for each parameter). A significant trend over the season quent use than self-reported in our study. Although not listed as
was detected for hematocrit only (p = 0.035; *** for sampling dates with a prohibited substance, creatine might enhance performance in
significantly different values from T0; Wilcoxon test). Erythrocytes and sports with repeated high-intensity character like soccer [16].
hemoglobin did not show significant changes during the season (p = 0.15 Given the present critical discussion about doping issues in elite
and p = 0.16, respectively). Numbers at the top of the hatched areas refer sport, some players might have been reluctant to admit creatine
to median (middle), upper (top) and lower (bottom) border of the 95 %
use (reported prevalence in the present study: only 0.6 %). How-
confidence interval.
ever, neither from the questionnaires nor from additional club
physician information any indications for creatine use of rele-
vant scale have arisen.
increases. However, it is well known that higher permeability of There is no obvious explanation for the observed systematic
muscle cell membranes exists in certain individuals increase in potassium from T0 to all sample times during the
[11, 26, 27, 31]. Some ethnic groups seem to be more frequently season. It is well known that potassium levels are increased
affected, and athletes with high percentages of fast twitch (type acutely during and after exercise [18]. But little is known about

Meyer T, Meister S. Routine Blood Parameters in … Int J Sports Med 2011; 32: 875–881
 Clinical Sciences 879

the long-term effect of continued high-intensity training on

Table 2 Reference range, median values, 95 % CI, intra-class CV, and results from testing for longitudinal change (α-error for significance of trend, n for the number of players eligible for this analysis at each of the sampling resting values. An increased concentration of Na-K-ATPase in the

11
39
40
40

40

12
40
40

40
40
39

40

40
31

12
40
cell membranes has been observed in athletes but this finding

n
Trend would rather be compatible with lower potassium resting levels

0.119
0.032
α-error
[20]. On the other hand, the frequent use of electrolyte drinks

0.02
0.46

0.02

0.14
0.02
0.27

0.05
< 0.01
0.01

0.22

< 0.01
0.01

0.11
0.01
and improved food quality compared to off-season might favour
higher potassium blood levels. Although “seasonal pseudohy-
T1–T3

perkalemia” has been reported as a result of exposing whole

11.7

11.5
11.5

29.3
20.3
17.7
16.6

21.7
13.4

1.5
8.5
9.2
6.3

6.2
4.3
6.9
CV

blood samples to temperature changes [33] this confounding

effect should have been widely avoided by centrifugating all
T1–T2

samples timely at the site of sampling.

10.7
10.5

20.5
18.8
10.0

16.8
15.6

15.3
11.8

1.6
8.1
9.0
6.6

5.8
3.9
6.5
CV

It is noteworthy that there is obviously a considerable increase in

plasma volume from the ongoing training process [3]. This obser-
187
187
184

187

187
167

153
186
187
152
186
187
187

152
187
187

vation is consistent with an increase in total erythrocyte mass but

n

an even larger increase in plasma volume which has been

observed after endurance training [30]. Although soccer is not a
0.75–1.30

30.7–81.4

0.78–0.98

0.73–3.50
1.00–3.49
3.81–5.18
125–238
142–331

138–144
59–162

17–155
3.2–8.6

4.0–7.2

pure endurance sport, players have to cover distances up to

95 % CI

18–63
13–49

24–55
T3

12 000 m in a whole soccer match interspersed by high-intensity

periods [34]. As a result of ongoing training and competition, con-
Median

siderable changes in endurance capacity over a typical soccer sea-

1.04

49.3

1.00
4.34
0.87

1.66 son have been described [21]. It is likely that such seasonal
177

108
210

141
5.4

5.5
29
23

54
37

Downloaded by: Universität Paderborn. Copyrighted material.

fluctuations in endurance capacity can explain haematological
changes similar to endurance athletes.
239

240
240
240
157

166
226
240
240

240

163

235
240
240
212
240

On average, professional soccer players seem to have a favoura-

n

ble blood lipid profile. In this study, low total cholesterol was
0.77–1.31

28.3–85.8

1.00–5.89
3.86–5.27
0.79–1.01

0.58–4.68

accompanied by an average LDL to HDL ratio of 2.03 (at T1, the

118–244
139–308

137–148
56–166

18–157
3.7–9.0

4.0–7.4
95 % CI

20–63
13–53

25–53
T2

sampling date with the most subjects). This contrasts somewhat

to recently published data from 504 American football players of
the same age who had 10 mg/dl higher total cholesterol and a
Median

LDL/HDL ratio of above 2.3 [37]. Factors contributing to this dif-

1.04

49.0

1.00
4.43
0.89

1.63
176

104
212

142
5.4

5.7

ference might be a preferred high body weight for some posi-

30
24

58
36

tions within an American football team and the resulting

322

323
323
323
246

246
320
323
323

323
321

244
323
323
298

322

different nutritional behaviour as well as the greater importance

n

of endurance for soccer players. It is well known that endurance

0.81–1.33

31.6–82.1

1.00–5.39
3.68–5.06
0.75–0.99

0.61–3.64

training elicits a modest HDL-enhancing effect [6].

123–237
145–316

137–148
59–172

18–154
3.6–8.2

3.9–7.2
95 % CI

20–60
13–46

26–53

Limited intraindividual variability (CV) for some parameters

T1

means that there is little fluctuation within a given athlete due

to the ongoing training and competition alone. Therefore, they
Median

should not be determined repeatedly during one season solely

1.05

51.7

1.00
4.32
0.87

1.61
170

105
217

142
5.7

5.5
30
24

36

58

for screening purposes (without suspicious complaints or other

clinical signs). Although such screening is common in many
215

216
216
216

178
164
215
216
216

216
216

176

215
216
216
212

clubs, based on the results from this study it is clearly advisable

n

to limit the range of parameters to values other than blood

count, creatinine, uric acid, total cholesterol, and electrolytes
0.83–1.35

33.2–77.6

1.00–4.14
3.46–4.74
0.78–0.98

0.66–3.67
128–252
137–339

137–146
60–171

14–174
3.1–9.5

4.2–7.8

(CV below 10 %). Such a restriction of blood screening to clini-

95 % CI

25–52
18–57
14–51
T0

cally meaningful items – like iron deficiency and anemia – is in

accordance with current literature [8, 9]. Although it is a statisti-
cal approach, deviations from the 95 % confidence intervals
Median

1.06

51.9

1.00
3.93
0.88

1.66

might be used to indicate findings which necessitate further

176

104
222

140
5.9

5.8
35
26
24

60

investigations. With the lack of long-term studies targeting at

clinically meaningful endpoints this is the only reasonable way
0.84–1.24
Ref. range

140–360

135–145
4.0–10.0

0.75–1.0
18–360
3.6–8.2

3.6–4.8

0.4–4.2

to proceed.
19–44

< 5.1
< 50
< 50

Limitations of the Study

total cholesterol [mg/dL]
HDL cholesterol [mg/dL]
LDL cholesterol [mg/dL]

▼
Within applied studies there is always some kind of trade-off
creatinine [mg/dL]
platelets [x103/μL]

uric acid [mg/dL]

Mg + + [mmol/L]

between external and internal validity. In the present investiga-

WBC [x10³/μL]

Na + [mmol/L]

ferritin [μg/L]
urea [mg/dL]

K + [mmol/L]

TSH [mIU/L]

tion, our chosen procedures to handle corrections for changes in

CRP [mg/L]
AST [U/L]
ALT [U/L]

plasma volume and to control sampling circumstances might

dates).

represent such trade-off. In daily medical care of athletes, physi-

cians typically do not correct for plasma volume changes when

Meyer T, Meister S. Routine Blood Parameters in … Int J Sports Med 2011; 32: 875–881
880 Clinical Sciences
longitudinally comparing routine blood results. This is why it
was decided to present uncorrected measurements. However, a
correction according to Dill and Costill ([5]; not presented)
revealed no relevant differences from the reported results. In
addition, blood screening in professional soccer players is often
carried out with less rigorous control of sampling time, training
schedule and nutrition than in the present study. But we felt that
only strict elimination of all samples taken from players without
sufficient compliance to study requirements guaranteed compa-
rability to population reference values. Obviously, this proce-
dure reduced the number of samples for our calculations
considerably. To some degree, control of compliance relied upon
the truth of questionnaire answers, of course. Particularly, the
documentation of supplementation might have been incom-
plete in some cases.
It is impossible to completely isolate chronic effects of training
and competition on blood values from acute ones. In the present
study, it cannot be excluded that changes in some parameters
represent rather acute effects arising from training or competi-
tion during the days before sampling than chronic ones from the
accumulation of such loads. This is particularly evident for CK
values. Also, T0 was not free from acute influences because sev-
eral players might have trained on an individual basis on the
days prior to their first club training session.
In recent years, there has been an increasing awareness of sub-
stance abuse, i. e., doping, in elite sport. There is never a com-
plete certainty that study results from elite athletes like in the
present study have not been affected by doping substances/pro-
cedures. But we found no indications for such misbehaviour.
None of the participating players had ever tested positive. In
addition, there were no suspicious blood values; e. g., high liver
enzymes in combination with low HDL cholesterol (sometimes
seen under anabolic steroids) or frequent elevated values for
hemoglobin or hematocrit. Hemoglobin values above 17 g/dl
were present in 4 players at T0, in 5 at T1, and in 0 at T2 and T3,
respectively. However, reticulocytes were not determined.
Together with other hematological indices their number and its
changes can be indicative of recent erythropoietin abuse or
blood doping [23, 28].
Reference values for the general population were taken from the
most common laboratory textbook in Germany. However, for
single blood parameters there might be slightly deviating refer-
ence ranges in other populations [32] due to different genetic
make-up or other socio-ecological circumstances. However, they
do not relevantly affect our interpretation of the study results
because the number of non-Caucasians was low in our sample
and all these professional players typically live under comforta-
ble circumstances.
It can be concluded that there is a profound effect of elite soccer
training and competition on plasma volume and CK only. Devia-
tions in all other values can be interpreted similar to the general
population even in highly active athletes. There are a number of
parameters that should not be repeatedly determined for rou-
tine screening purposes without clinical indication in athletes.
Acknowledgements
▼ after eccentric and concentric contractions. Muscle Nerve 1986; 9:
This study was funded by the Deutsche Fußball Liga (DFL; Ger-
man Football League) which served as a sponsor only (no influ-
ence on study conduction). There has been no further funding.
We would like to thank all team physicians of the participating
Meyer T, Meister S. Routine Blood Parameters in … Int J Sports Med 2011; 32: 875–881
clubs together with their medical staff who took care that on-
site conduction of this study worked perfectly. Parts of this manu-
script have been presented in oral form at the German Congress
of Sports Medicine 2009 in Ulm, Germany.
References
1 Banfi G, Del Fabbro M, Lippi G. Creatinine values during a competitive
season in elite athletes involved in different sport disciplines. J Sports
Med Phys Fitness 2008; 48: 479–482
2 Banfi G, Del Fabbro M, Lippi G. Relation between serum creatinine and
body mass index in elite athletes of different sport disciplines. Br J
Sports Med 2006; 40: 675–678
3 Banfi G, Del Fabbro M, Mauri C, Corsi MM, Melegati G. Haematological
parameters in elite rugby players during a competitive season. Clin
Lab Haematol 2006; 28: 183–188
4 Bangsbo J, Norregaard L, Thorso F. Activity profile of competition soc-
cer. Can J Sport Sci 1991; 16: 110–116
5 Dill DB, Costill DL. Calculation of percentage changes in volume of
blood, plasma, and red cells in dehydration. J Appl Physiol 1974; 37:
247–248
6 Durstine JL, Grandjean PW, Davis PG, Ferguson MA, Alderson NL, DuBose
KD. Blood lipid and lipoprotein adaptations to exercise: a quantitative
analysis. Sports Med 2001; 31: 1033–1062
7 El-Sayed MS. Effects of exercise and training on blood rheology. Sports
Med 1998; 26: 281–292
Downloaded by: Universität Paderborn. Copyrighted material.
8 Fallon KE. Clinical utility of blood tests in elite athletes with short
term fatigue. Br J Sports Med 2006; 40: 541–544
9 Fallon KE. Utility of hematological and iron-related screening in elite
athletes. Clin J Sport Med 2004; 14: 145–152
10 Gabriel H, Schwarz L, Born P, Kindermann W. Differential mobilization
of leucocyte and lymphocyte subpopulations into the circulation dur-
ing endurance exercise. Eur J Appl Physiol 1992; 65: 529–534
11 Haralambie G. Neuromuscular irritability and serum creatine phos-
phate kinase in athletes in training. Int Z Angew Physiol 1973; 31:
279–288
12 Haralambie G, Berg A. Serum urea and amino nitrogen changes with
exercise duration. Eur J Appl Physiol 1976; 36: 39–48
13 Harriss DJ, Atkinson G. Update – Ethical Standards in Sport and Exer-
cise Science Research. Int J Sports Med 2011; 32: 819–821
14 Herrmann M, Scharhag J, Miclea M, Urhausen A, Herrmann W, Kinder-
mann W. Post-race kinetics of cardiac troponin T and I and N-terminal
pro-brain natriuretic peptide in marathon runners. Clin Chem 2003;
49: 831–834
15 Jackson KA, O’Rourke KM, Kark A, Kennedy GA. Artefactual elevation of
creatinine due to creatine water supplements. Med J Aust 2010; 193:
616–617
16 Law YL, Ong WS, GillianYap TL, Lim SC, Von Chia E. Effects of two and
five days of creatine loading on muscular strength and anaerobic
power in trained athletes. J Strength Cond Res 2009; 23: 906–914
17 Malm C, Ekblom O, Ekblom B. Immune system alteration in response
to increased physical training during a five day soccer training camp.
Int J Sports Med 2004; 25: 471–476
18 Marcos E, Ribas J. Kinetics of plasma potassium concentrations during
exhausting exercise in trained and untrained men. Eur J Appl Physiol
1995; 71: 207–214
19 McDonald CA. Enzyme levels after running. JAMA 1981; 246: 40–41
20 McKenna MJ, Harmer AR, Fraser SF, Li JL. Effects of training on potas-
sium, calcium and hydrogen ion regulation in skeletal muscle and
blood during exercise. Acta Physiol Scand 1996; 156: 335–346
21 McMillan K, Helgerud J, Grant SJ, Newell J, Wilson J, Macdonald R, Hoff
J. Lactate threshold responses to a season of professional British youth
soccer. Br J Sports Med 2005; 39: 432–436
22 Meyer T, Gabriel HHW, Rätz M, Müller HJ, Kindermann W. Anaerobic
exercise induces moderate acute phase response. Med Sci Sports Exerc
2001; 33: 549–555
23 Morkeberg J, Belhage B, Ashenden M, Borno A, Sharpe K, Dziegiel MH,
Damsgaard R. Screening for autologous blood transfusions. Int J Sports
Med 2009; 30: 285–292
24 Mougios V. Reference intervals for serum creatine kinase in athletes.
Br J Sports Med 2007; 41: 674–678
25 Newham DJ, Jones DA, Edwards RH. Plasma creatine kinase changes
59–63
26 Nicholson GA, Morgan G, Meerkin M, Strauss E, McLeod JG. The creatine
kinase reference interval. An assessment of intra- and inter-individual
variation. J Neurol Sci 1985; 71: 225–231

27 Noakes TD. Effect of exercise on serum enzyme activities in humans.
Sports Med 1987; 4: 245–267
28 Parisotto R, Gore CJ, Hahn AG, Ashenden MJ, Olds TS, Martin DT, Pyne
DB, Gawthorn K, Brugnara C. Reticulocyte parameters as potential
discriminators of recombinant human erythropoietin abuse in elite
athletes. Int J Sports Med 2000; 21: 471–479
29 Scharhag J, George K, Shave R, Urhausen A, Kindermann W. Exercise-
associated increases in cardiac biomarkers. Med Sci Sports Exerc
2008; 40: 1408–1415
30 Schmidt W, Prommer N. Effects of various training modalities on blood
volume. Scand J Med Sci Sports 2008; 18 (Suppl 1): 57–69
31 Schwartz PL, Carroll HW, Douglas JSJ. Exercise-induced changes in
serum enzyme activities and their relationship to max VO2. Int Z
Angew Physiol 1971; 30: 20–33
32 Sharpe K, Hopkins W, Emslie KR, Howe C, Trout GJ, Kazlauskas R, Ash-
enden MJ, Gore CJ, Parisotto R, Hahn AG. Development of reference
ranges in elite athletes for markers of altered erythropoiesis. Haema-
tologica 2002; 87: 1248–1257
33 Sinclair D, Briston P, Young R, Pepin N. Seasonal pseudohyperkalaemia.
J Clin Pathol 2003; 56: 385–388
Meyer T, Meister S. Routine Blood Parameters in … Int J Sports Med 2011; 32: 875–881
Clinical Sciences 881
34 Stolen T, Chamari K, Castagna C, Wisloff U. Physiology of soccer – an
update. Sports Med 2005; 35: 501–536
35 Swanimathan R, Major P, Snieder H, Spector T. Serum creatinine and
fat-free mass (lean body mass). Clin Chem 2000; 46: 1695–1696
36 Thomas L. Labor und Diagnose (Laboratory and diagnosis). TH Books,
Frankfurt 2008
37 Tucker AM, Vogel RA, Lincoln AE, Dunn RE, Ahrensfield DC, Allen TW,
Castle LW, Heyer RA, Pellman EJ, Strollo PJ Jr, Wilson PW, Yates AP.
Prevalence of cardiovascular disease risk factors among National Foot-
ball League players. JAMA 2009; 301: 2111–2119
38 Urhausen A, Kindermann W. Biochemical monitoring of training. Clin
J Sport Med 1992; 2: 52–61
39 Warren GL, Lowe DA, Armstrong RB. Measurement tools used in the
study of eccentric contraction-induced injury. Sports Med 1999; 27:
43–59
40 Wong ET, Cobb C, Umehara MK, Wolff GA, Haywood LJ, Greenberg T,
Shaw ST Jr. Heterogeneity of serum creatine kinase activity among
racial and gender groups of the population. Am J Clin Pathol 1983;
79: 582–586
Downloaded by: Universität Paderborn. Copyrighted material.