PHYSIOLOGY AND BIOMECHANISMS

ORIGINAL ARTICLES
J SPORTS MED PHYS FITNESS 2010;50:19-24

CA Physiological profile of water polo players

in different competitive levels

DI
G. MELCHIORRI 1, 2, E. PADUA 1, F. SARDELLA 3, V. MANZI 1, V. TANCREDI 1, M. BONIFAZI 4,

E
M ®
Aim. The purpose of this study was to determine if there 1University of Rome Tor Vergata, Rome, Italy
are different physiological characteristics in water polo 2Don Gnocchi Foundation, Rome, Italy
players of three different competitive levels (national team, 3CONI – Istituto di Medicina e Scienza dello Sport, Rome, Italy
4University of Siena, Siena, Italy
NT, junior national team, NJ, and amateur club, AC).

A T
Methods. To better define the training levels of water polo
players competing at different levels, we administered a

V
shuttle swim test (SST, a specific test used by Italian coach-

R G H
es) and a classic speed-lactate test. The shuttle swim test is
based on actions at maximum intensity followed by incom-
plete recoveries. To compare the athletes, we used their
morphological characteristics, the speed during an incre-
shold speeds. The shuttle swim test for water polo provides
conditioning coaches and sport scientists with data on the

E I
sport-specific movement speed and the competition-specif-
mental test associated with fixed blood lactate concentrations ic fatigue resistance in each athlete.
of 2 mmol.1-1 (Aerobic Threshold, AT) and 4 mmol.1-1

R
KEY WORDS: Lactates - Heart rate - Exercise.

N
(Anaerobic Threshold, AnT), and the SST data (speed, heart
rate and lactate concentration).

I
Results. The heart rate at the end of the SST was 164±12

Y
beats.min-1 for NT, 166±10 beats.min-1 for NJ (no significant
difference, P>0.05) and 179±9 beats.min-1 for AC (signifi-
W ater polo is classified as a sport with alternate

M P
cantly different from NT and NJ, P<0.05). The AT and AnT
speeds were significantly higher in NT than in AC and NJ
(P<0.05). No significant differences were found in AT and
aerobic and anaerobic metabolic demands, with
the use of a high percentage of muscle mass and fair-

O
AnT speeds between AC and NJ (P>0.05). The mean speed
during the shuttle swim test was significantly higher in NT
ly high regional force.1, 2 As for other sports in the
same classification group, there is some difficulty in
proposing evaluation tests for this discipline. Such

C
and NJ than in AC (P<0.05). Lactate values were similar
in the NT, AC and NJ groups (P>0.05). The SST mean speed
was significantly correlated with the AT (P<0.01) and AnT
speeds (P<0.05) in three groups . In all groups no significant
correlation was found between SST blood lactate and AT or
AnT speeds (P>0.05).
Conclusion. Some differences were found among the senior
and junior professional and amateur water polo players in
both SST performance and Aerobic and Anaerobic Thre-
Received on December 11, 2009.
Accepted for publication on February 26, 2010.
Corresponding author: Prof. G. Melchiorri, University of Rome Tor
Vergata, Rome, Italy. E-mail: gmelchiorri@libero.it
tests should evaluate the physio-metabolic character-
istics of the athlete as well as the athletic conditioning
(specific training) and aspects of the typical techni-
cal abilities of the sport.3 The test parameters should
include speed, variations of direction and short paus-
es between one action and another (typical elements of
alternate aerobic-anaerobic sports), and metabolic
parameters indicative of the possible percentages of
metabolic demands. In addition, the test should be
easy to administer and useful for the coach.1, 3
The international literature contains few studies
dealing with water polo. However, we know that dur-
ing competitions a forward covers a mean distance of

Vol. 50 - No. 1 THE JOURNAL OF SPORTS MEDICINE AND PHYSICAL FITNESS 19

MELCHIORRI PHYSIOLOGICAL PROFILE OF WATER POLO PLAYERS

1050 m at an average speed of 1.35 m.s-1.4, 5 Game
analyses have demonstrated that water polo is an ‘inter-
mittent’ sport consisting of intense bursts of activity of
<15 seconds duration separated by lower intensity
intervals averaging <20 seconds duration.2
the athletes completed the entire test without any inter-
ruption. In total, 54 athletes were tested.
All the participating athletes were informed about the
test purposes and procedures, and they signed an
informed consent form.

A
Physiological measurements during game play indi-
cate a cumulative effect of the repeated sequences of Pre-test procedures
activities and suggest there is a high metabolic demand

C
on the athletes.2, 6 Mean lactate levels in Italian athletes

I
are between 7 and 9 mmol/L.4, 6 The previously cited
articles and the more detailed study by Smith in 1998
Familiarization sessions were implemented during
the 2 weeks prior to the beginning of the experimen-
tal procedures. The testing sessions took place at the

D
2 demonstrated the importance of the aspects of “fight”, same hour of the training sessions (afternoon) to avoid
“stop and go”, variation of the swim direction, repeat- possible circadian influences on the parameters under

E
ed swim passages from the horizontal to the vertical investigation (±2 hours). The athletes abstained from
position in the water during swimming, and intense alcohol and caffeinated beverages and refrained from
work by the lower limbs in positioning and specific heavy training in the 24 hours before the experimen-

M ®
technical gestures.
Few authors have attempted to identify the physio-
logical characteristics contributing to success in water
tal sessions. The athletes consumed their last meal at
least 3 h before the test, and a report of nutrient con-
tent was taken to ensure that a sufficient carbohydrate

A T
polo before the changes in the rules of the game by
comparing elite water polo players and average-level

V
players. In these studies an important limiting factor to
intake during the week before the testing was con-
sumed. The athletes were then fitted with a heart rate
transmitter and recorder (Polar Electro S610, Kempele,

H
achieve high performance capability in water polo

R
players were the repeated sprint ability (RSA).7 After

G
the introduction of new game rules in 2006, more than
Finland) to assess changes in heart rate every 5 s
throughout the tests. To measure the individual blood
lactate concentration during the tests, we used an

E I
in the past,8 in water polo can be hypothesized that enzymatic-amperometric method (SUPER GL,
the ability to increase the frequency of such high-pow- DiaSys, Germany) involving the collection of
microsamples of capillary blood (20 µL) from the

IN YR
er actions would improve the performance of the teams
already competing at high levels.
Therefore, the purpose of this study was to identi-
fy the physiological determinants of performance in
earlobe and immediate addition to the solution sup-
plied by the manufacturer in appropriate test tubes. The
subjects were well motivated, and throughout testing

M P
modern water polo. It was hypothesized the exis- sessions, verbal encouragements were granted to
tence of a marked different physiological character- induce maximal effort. Before the beginning of the
istics among water polo players that competing at tests, athletes performed a codified 15 minutes warm-

O
different levels. up followed by a 5 minutes recovery period. All test-
ing procedures took place in an acclimatized swim-

C
Materials and methods
Subjects
Male athletes competing at three different levels
were tested. They were recruited from the Italian
national water polo team (group NT, national senior),
Italian junior national team (group NJ, national junior)
and an amateur team (group AC, amateur club). All the
national athletes (NT and NJ) belonged to professional
teams . The athletes of the national senior team were
27.9±2.1 years old, those of the NJ group 23.5±1.7
years, and those of the AC group 24.8±1 years. All
ming pool where temperature (18-20 °C) and humid-
ity (45-50%) were electronically controlled. All the
tests were undertaken in November 2008, when play-
ers were involved in the first part of the championship
and were at a good fitness level.
Procedures
Testing sessions were administered in random order,
with at least four days between testing sessions, and
consisted of:
— an incremental swim test over 200 m to construct
the speed-lactate curve;9
— an SST test with actions at maximum intensity

20 THE JOURNAL OF SPORTS MEDICINE AND PHYSICAL FITNESS March 2010

PHYSIOLOGICAL PROFILE OF WATER POLO PLAYERS MELCHIORRI

followed by incomplete recoveries for anaerobic thresh- ity of this testing procedure has been reported else-
old assessment.10 where.10

Incremental swim test Statistical analysis

The athletes performed an incremental test over 200 The results are expressed as means±standard devi-

A
m to construct the speed-lactate curve. They swam
the crawl 200 m five times, respectively at 60, 70, 80,

C
90, 100% of maximal speed (recorded one week before
ation. Before using parametric tests, we verified the
normality of the distributions using the Shapiro-Wilk
W-test. Pearson’s product moment correlation coeffi-

DI
over the same distance) with 10, 10, 20 and 30 minutes
of “passive” rest between repetitions.9
In four days preceding 200 meters test athletes didn’t
have hard training. Blood lactate concentrations were
cient (and the corresponding 95% CI) was used to exam-
ine correlations between parameters. According to
Hopkins 12 magnitude for correlation coefficients were
considered as trivial (r<0.1), small (0.1<r<0.3) mod-

E
plotted against swim speeds and individual blood lac- erate (0.3<r<0.5), large (0.5<r<0.7), very large
tate concentration profiles (speeds at 2 and 4 mmol.l-l) (0.7<r<0.9) and nearly perfect (r>0.9) and perfect
were identified via exponential interpolation.11 The (r=1). To establish whether construct validity existed

M ®
highest HR measured during the maximal incremen-
tal test was used as maximum reference value (HRmax).
the test variables was analyzed with one-way analysis
of variance (ANOVA); group comparisons were
checked for homogeneity of variance. Post-hoc analy-

A
ses were carried out with the Tukey HSD test.

T
Shuttle Swim Test
Significance was set at 0.05 (P≤0.05). SPSS (SPSS

V
The shuttle swim test consisted of two divided series Inc., Chicago, IL, USA) was used for all statistical

H
of 120-m swims at maximum speed for a total of 240 calculations.

R
m. The subjects swam seven legs per series: one of 40
m, two of 20 m and four of 10 m. After a recovery of

E IG
90 seconds, they swam the second series with the same
legs (40, 20, 20, 10, 10, 10, 10 m). The recovery time
between one leg and the next one was equal to the
Results

Some differences in body weight and percent of

IN YR
time spent swimming the previous leg. The space in
which the series were performed was delimited by
two lane lines 10 m apart, so that the athlete was forced
to touch the lane line and to restart in the opposite
body fat mass were found. Senior athletes have high-
er body mass and less percent of fat mass than amateur
group (88.5±10.3 kg and 18.0±3% for senior and
83.5±15.2 kg and 25.0±5.0% for amateur, P<0.05) . No

M P
direction when swimming fractions longer than 10 m;
therefore, the 20-m and 40-m legs were performed by
repeating the 10-m course several times in a stop and
statistical difference was found in fat mass between
senior and junior group (17.0±4.0% of fat mass for

O
junior athletes, P>0.05). Differences was recorded for
go manner. A typical water polo action, the “balance body weight between senior and junior subjects
movement”, was used by all athletes to stop and restart (88.5±10.3 kg and 85.7±8.5 kg respectively, P<0.05).

C
from the lane line. The lactate concentration was mea-
sured before the test (basal value) at the end of the
first 120-m series (within 1 minute) and at the end of
the second 120-m series (at the 4th minute of recovery,
the time that gave significantly higher lactate values in
previous lactate curve tests). The swim times during the
tests were measured by the experimenter (present study
first author) with the start and stop indications given by
other experimenter situated at the marks delimiting
the distance. The athletes were requested to touch the
lane lines with their hands. Performance times was
measured in seconds with a digital chronometer
(EPSAN, Lelystad, Netherlands). Reliability and valid-
Statistical significance was found in difference between
junior and amateur percent of fat mass (17.4±4.0%
and 25±5% respectively, P<0.05). No statistical dif-
ferences was found for height between three groups
(186.6±6.9 cm for senior, 186.6±6.9 cm for amateur
and 185.8±5.2 cm for junior, P>0.05). NT and NJ dif-
fered significantly in age (27.9±2.1 years and 23.5±1.7
years respectively, P>0.05). AC athletes (24.8±1 years)
are younger than senior group, (P<0.05). No differ-
ence was found in age between junior and amateur
group (Table I). Speeds during the incremental test
associated with fixed blood lactate concentrations of
2 mmol.1-1 (Aerobic Threshold, AT) and 4 mmol/L-1

Vol. 50 - No. 1 THE JOURNAL OF SPORTS MEDICINE AND PHYSICAL FITNESS 21

MELCHIORRI PHYSIOLOGICAL PROFILE OF WATER POLO PLAYERS

TABLE I.—Characteristics of the athletes. The lactate values were similar in the NT, AC and NJ
Senior Amateur Junior groups (Table II). Correlations among aerobic fitness
(N.=18) (N.=18) (N.=18) performances and SST parameters are shown in Table
III.
Height (cm) 186.6±6.90 186.6±6.90 185.8±5.2
Body mass (kg) 88.5±10.3 83.5±15.2 85.7±8.5

CA
Fat mass ( % of body weight) 18±2 25±5 17±4 Discussion

This is the first study that examined the physiolog-

I
TABLE II.—Physiological characteristics for Senior, Amateur and ical determinants of performance in different agonis-
Junior water polo players (mean±SD).
tic level water polo player. In Italy, there are four

D
Senior Amateur Junior national water polo divisions (A1, A2, B, C). A1 is
(N.=18) (N.=18) (N.=18)
the highest division with the best athletes. All the

E
AT speed 2 mmol/l (m/s) 1.31±0.08 1.21±0.11*S 1.20±0.07**S Italian senior (NT) and junior (NJ) national team mem-
AnT speed 4 mmol/l (m/s) 1.37±0.07 1.27±0.10*S 1.28±0.06*S* bers played in the A1 division, while the amateur club
SST first test blood lactate players (AC) competed in the C division. Therefore, the

M ®
concentration (mmol/l)
SST second test blood lactate
concentration (mmol/l)
6.0±2.5

9.2±2.5
6.9±1.9

11.0±2.5
5.7±1.2

9.7±2.1
groups differed in both training time (average 13 hours
per week for NT and NJ vs. 7 hours for AC) and tech-
nical/competitive level.

VA T
SST Mean speed (m/s) 1.82±0.07 1.63±0.09**S,J 1.76±0.09
AT: aerobic threshold; AnT: anaerobic threshold; SST: shuttle swim test; S: senior;
J: junior; C: amateur. Asterisks followed by a letter indicate significant differences
The two tests were used to asses differences in gen-
eral (speed-lactate curve) and specific (shuttle swim
test, SST) swimming capacity. The main finding of this

H study was that NT show higher 2 and 4 mmol.1-1 thresh-

between groups. *P<0.05; **P<0.01.

R
old speeds than in NJ or AC (P<0.05). Furthermore, the

G
mean speed during the SST was significantly higher in
TABLE III.—Pearson correlation (r) between aerobic fitness perfor-

E I
NT and NJ than in AC (P<0.01).
mance and shuttle swim parameters.
Although the age difference, longer sports experience
At speed 2 mmol/L AnT speed 4 mmol/L and higher time of training between NT and NJ or AC

IN YR
(m/s) (m/s)
could explain the higher AT and AnT speeds. It is sur-
R CI (95%) R CI (95%) prising that AC and NJ did not differ significantly in
upper-lower upper-lower
these variables (Table II). Since the NJ players showed
good competitive abilities at the 2008 junior world

M P
SST Mean speed (m/s) 0.52* (0.74 -0.24) 0.56# (0.75 -0.29)
SST blood lactate championship, where they won the silver medal, and
concentration (mmol/l) -0.28ns (-0.55 -0.05) -0.31a (-0.58 -0.01) trained for the same time per week as the NT group, it

O
is possible that the longer career of the NT players
AT: aerobic threshold; AnT: anaerobic threshold; SST: shuttle swim test. *P<0.01;
#P<0.05; aP=0.06. was the cause of the difference between NJ and NT.
However, the aerobic and anaerobic threshold speeds

C
(Anaerobic Threshold, AnT), and the shuttle swim
test data are reported in Table II. Mean heart rate at the
end of the SST was 164±12 beats.min-1 for NT and
166±10 beats.min-1 for NJ (no significant difference,
p>0.05). In AC subjects we recorded 179±9 beats.min-
1 (significantly different from NT and NJ, P<0.05).
The speeds at AT and AnT were significantly high-
er in NT than in AC and NJ, (P<0.05). The last two
groups did not differ significantly in these variables
(P>0.05). The average speeds during the SST were
significantly higher in NT and NJ than in AC (P<0.05).
are only two of more variables of interesting to study
differences in swimming quality among different ago-
nistic level water polo players. Therefore, as proposed
for other team sports,13-17 we used a test (SST) that
more closely resembles the demands of water polo
match. The shuttle swim test was chosen because it
has been used for a long time by Italian coaches is
easily administered and has good reliability and valid-
ity.10 The values for both the first and second series
were in the range (7-9 mmol.l-1) near to these record-
ed during games.2, 4-6, 8
The SST blood lactate concentration was higher
after the second series than after the first (P<0.05,

22 THE JOURNAL OF SPORTS MEDICINE AND PHYSICAL FITNESS March 2010

PHYSIOLOGICAL PROFILE OF WATER POLO PLAYERS MELCHIORRI

2.2 imal speed, as in the SST, swimming economy is less

important.20 During the shuttle test, the mean time to
cover the distance was significantly correlated with
2.2 the competitive level. The mean time for the senior
players was significantly different from that of the

A
amateur group but not from that of the junior nation-
2.2 al group competing at the same level. The highest
mean speed of the national senior players and the low-

2.2

IC est mean speed of the amateurs could represent a good

performance index for water polo players. The lactate
levels during the SST did not differ among the three

2.2

E
40
D 20
Distance (m)
10
groups. However, in contrast to continuous exercise, in
which the blood lactate concentrations are lower but
closely match the muscle lactate concentrations, there
is a low correlation between the muscle and blood lac-
tate concentrations when subjects perform repeated
AC

M ® NJ NS

Figure 1. Mean speed for each distance (40, 20 and 10m) during SST in
intense exercise.21 The lack of a significant difference
in blood lactate concentration at the end of the SST
could reflect this condition.

A
NJ, NT and AC.

T
The results suggest that different training approach-
es should be adopted for senior and junior players or

V
Figure 2). May be a “cumulative effect” of sprint with

H
an incomplete recovery can explain higher lactate con-

R
centration after the second series. The heart rate differed
when the objective is to increase the competitive lev-
el. In junior players, more extensive swimming at the
2 and 4 mmol/L-1 thresholds is necessary to increase the

G
among the three groups (P<0.05). The mean value for general swimming fitness.

E I
all three groups during the SST (170±10 beats.min-1) The results of this study and our previous experience
was apparently higher than the match value.18, 19 The with the shuttle swim test suggest that it is easy to use

IN YR
higher mean heart rate in AC (179±9 beats.min-1) than
in the other two groups was probably due to greater car-
diovascular fatigue, and the different weekly training
time could explain this difference. There was also a sig-
and relevant to the physiological indices of match
competition. The SST for water polo provides condi-
tioning coaches and sport scientists with data on the
sport-specific movement speed and the competition-

M P
nificant difference in mean speed between NT and AC specific fatigue resistance in each athlete. By analysing
and between NJ and AC (P<0.01). Shuttle test better the maximal and repeated-effort performances in the
can reflect the competitive/technical differences test, the conditioning coach can evaluate the training

O
between athletes or the different levels of specific strategy for individual athletes. In conclusion, our data
training due to more matches played during the year in support the use of the shuttle swim test for objective

C
NT and NJ (national league games and international
games).
The senior group would be expected to have a more
economical swimming technique. This would also
explain the differences in the speed-lactate curve.20
In the procedure to define the speed-lactate curve, the
athletes swam the 200 m distance three times, the first
time slowly. The lactate levels measured in NJ and
AC after the slow leg were higher than in NT, indi-
cating that the national junior and amateur players
have a less economical swimming technique than the
senior players and generally poorer swimming train-
ing. When the test involves short distances and max-
assessment of the training status of water polo players.
References
1. Dal Monte A., Faina M. Valutazione dell’atleta. Turin: Unione
Tipografico-Editrice Torinese (UTET); 1999
2. Smith HK. Applied physiology of water polo. Sports Med 1998;26:317-
34.
3. Gore C.J. Physiological tests for elite athletes. 2000; Human Kinetics,
Champaign, IL, USA
4. Sardella F, Alippi B, Rudic R, Castellucci G, Bonifazi M. Analisi
fisiometabolica della partita. La Tecnica del Nuoto (Quaderno Centro
Studi e Ricerche F.I.N.) 1992;19:21-4.
5. Rudic R, D'Ottavio S, Bonifazi M, Alippi B, Gatta G, Sardella F. Il
modello funzionale nella pallanuoto. La Tecnica del Nuoto 1999;26:
21-4.

Vol. 50 - No. 1 THE JOURNAL OF SPORTS MEDICINE AND PHYSICAL FITNESS 23

MELCHIORRI PHYSIOLOGICAL PROFILE OF WATER POLO PLAYERS

6. Rodriguez FA. Physiological testing of swimmers and water polo
players in Spain. In: Miyashita M, Mutoh Y, Richardson AB, editors.
Medicine and science in aquatic sports. Med Sport Sci. Basel, Karger,
1994;39:172-7.
7. Melchiorri G, Castagna C, Sorge R, Bonifazi M. Game-activity and
Blood Lactate in men’s Elite Water Polo Players. J Strength
Conditioning Res [In press].
15. Spencer M, Bishop D, Dawson B, Goodman C. Physiological and
metabolic responses of repeated-sprint activities specific to field-
based team sports. Sports Med 2005;35:1025-44.
16. Impellizzeri FM, Marcora SM, Castagna C, Reilly T, Sassi A, Iaia
FM et al. Physiological and performance effects of generic versus
specific aerobic training in soccer players. Int J Sports Med 2006;
27:483-92.

A
8. Olbrecht J. The Science of Winning. Luton: Swimshop; 2000. 17. Hollander AP, Dupont SHJ, Volkerijk SM. Physiological strain during
9. S Melchiorri G, Manzi V, Padua E, Sardella F, Bonifazi M. Shuttle competitive water polo games and training. In: Miyashita M, Mutoh
Swim Test for water polo players: validity and reliability. J Sports Y, Richardson AB, editors. Medicine and Science in Aquatic Sports.

C
Med Phys Fitness 2009;49:327-30. Med Sport Sci. Basel: Karger 1994;39:178-85.
10. Banister EW. Modeling elite athletic performance. In: Green H, 18. Platanou T, Geladas N. The influence of game duration and playing

DI
McDougal J, Wenger H, editors. Physiological testing of elite athle-
tes. Champaign (IL): Human Kinetics; 1991. p. 403-4244.
11. Hopkins WG. Measures of reliability in sports medicine and science.
Sports Med 2000;30:1-15.
12. Bangsbo J. The physiology of soccer--with special reference to inten-
se intermittent exercise. Acta Physiol Scand Suppl 1994;619:1-155.
position on intensity of exercise during match-play in elite water polo
players. J Sports of Sports Sciences 2006;24:1173-81.
19. Maglischo EW. Swimming fastest. Champaign, IL: Human Kinetics;
1993.
20. Krustrup P, Mohr M, Steensberg A, Bencke J, Kjaer M, Bangsbo
J. Muscle and blood metabolites during a soccer game: implica-

E
13. Hoffman JR, Epstein S, Einbinder M, Weinstein Y. The influence of
aerobic capacity on anaerobic performance and recovery indices in
basketball players. J Strength Cond Res 1999;13:407-11.
tions for sprint performance. Med Sci Sports Exerc 2006;38:1165-
74.
21. Helgerud J, Engen LC, Wisloff U, Hoff J. Aerobic endurance trai-

M ®
14. Bishop D, Lawrence S, Spencer M. Predictors of repeated-sprint abi- ning improves soccer performance. Med Sci Sports Exerc 2001;33:
lity in elite female hockey players. J Sci Med Sport 2003;6:199-209. 1925-31.

VA T
R G H
E I
IN YR
M P
CO

24 THE JOURNAL OF SPORTS MEDICINE AND PHYSICAL FITNESS March 2010