SHORT COMMUNICATION
The Effect of Lumbar Support on the Effectiveness
of Anti-G Straining Manuevers
Juha Oksa, Tero Linja, and Harri Rintala
OKSA J, LINJA T, RINTALA H. The effect of lumbar support on the
effectiveness of anti-G straining maneuvers. Aviat Space Environ
Med 2003; 74:886 –90.
Introduction: The ability of fighter pilots to perform efficient anti-G
straining maneuvers (AGSM) is vital for their G-tolerance. The aim of this
study was to evaluate if the use of a lumbar support that optimizes the
posture of the spine enhances the effectiveness of AGSM. Methods:
Eleven fighter pilots performed four AGSM training sessions with 4 –5
repetitions in each session, each lasting approximately 10 s. The sessions were done without lumbar support or using supports that were 7,
14, or 26 mm thick. During the sessions, the electromyogram (EMG)
data were recorded from eight muscles (rectus and biceps femoris,
gluteus maximus, erector spinae, rectus abdominis, obliquus externus,
latissimus dorsi, and pectoralis major). Results: In a single best AGSM,
average EMG (aEMG) increased 12%, 10%, and 14% and power spectrum area (PSA) increased 20%, 26%, and 44% while using 7-, 14-, and
26-mm supports, respectively, in relation to the condition without support. According to aEMG and PSA, effectiveness increased in 11 and 10
subjects, respectively, with the average increase being 12% (aEMG) and
25% (PSA). During the whole session (repeated AGSM), effectiveness
was best with support in 7 subjects and without support in 4 subjects. In
the total population, the average increase in the effectiveness during
repeated AGSM was 6% and 11% with aEMG and PSA, respectively.
Discussion: The use of lumbar support tended to increase the effectiveness of muscular work, especially during single AGSM, but also during
repeated AGSM.
Keywords: fighter pilot, G-tolerance, EMG, muscular work.
D
URING AERIAL COMBAT maneuvering, fighter
pilots are exposed to high ⫹Gz forces (1,7,9,10,12,13),
which tend to direct circulation toward the extremities
and cause “greyout,” “blackout,” or even loss of consciousness. These episodes are due to insufficient blood
perfusion to the eye and brain, which is caused by the
high hydrostatic pressure gradients due to high ⫹Gz
forces.
To prevent such episodes from occurring, fighter pilots use anti-G trousers to reduce venous pooling in the
lower extremities and increase arterial BP. They also
use anti-G straining maneuvers (AGSM). With a simultaneous and powerful contraction of the large muscle
groups, the diameter of dependent blood vessels is
reduced, thereby increasing peripheral resistance, venous return, and intrathoracic and intra-abdominal
pressure. The resulting increase in the arterial pressure
at head level induced by AGSM ensures sufficient cerebral perfusion (15). The effectiveness of the AGSM
depends on the ability of the pilot to contract the muscles as simultaneously as possible and on the strength
of his or her muscles. The stronger the muscles are the
more efficiently they reduce the diameter of the vessels
886
and increase arterial pressure; therefore, the more efficient they are in reducing the blood flow to the extremities and maintaining sufficient cerebral perfusion.
In an isometric contraction, skeletal muscle produces
its highest force level at its midlength (8). For example,
for abdominal muscles this means that their highest
force level would be observed when the spine is in an
upright position and the shoulders are bent only
slightly forward. Leaning the torso forward or backward would reduce the maximal force produced by the
abdominal muscles. The seat in the Hawk MK 51 aircraft causes the pilot to sit with torso bent forward; this
study was conducted to evaluate whether the use of a
lumbar support, to optimize the posture of the spine,
would enhance the effectiveness of an AGSM.
METHODS
Eleven male subjects (8 flight cadets and 3 flight
instructors) voluntarily participated the study. Their
mean (⫾SD) age was 23 ⫾ 1 yr, height 179 ⫾ 4 cm,
weight 76 ⫾ 7 kg, and body mass index (BMI) 24 ⫾ 2%.
Each subject performed four AGSM training sessions
which were used to test four conditions in balanced
order: without lumbar support and using a support 7,
14, or 26 mm thick (Respecta Ltd, Finland). The supports were made of slightly compressible material and
were custom designed for the ejection seat but not for
individual subjects. The support was adjusted so that its
lower edge was slightly below the level of the subject’s
iliac and altered the curvature of the spine, i.e., made it
closer to a standing posture with less forward bend
than usually produced by the seat. Measurement in one
subject showed that the 7-mm support was in contact
with the spine from L1 to L5, the 14-mm support from
From the Oulu Regional Institute of Occupational Health, Laboratory of Physiology, Oulu, Finland (J. Oksa); the Kuortane Sports
Institute, Kuortane, Finland (T. Linja); and the Finnish Air Force
Headquarters, Tikkakoski, Finland (H. Rintala).
This manuscript was received for review in October 2002. It was
revised in February and April 2003. It was accepted for publication in
April 2003.
Address reprint requests to: Juha Oksa, Ph.D., who is a specialized
research scientist, Oulu Regional Institute of Occupational Health,
Laboratory of Physiology, Aapistie 1, 90220 Oulu, Finland;
juha.oksa@ttl.fi.
Reprint & Copyright © by Aerospace Medical Association, Alexandria, VA.
Aviation, Space, and Environmental Medicine • Vol. 74, No. 8 • August 2003
LUMBAR SUPPORT & AGSM—OKSA ET AL.
L1 to T11, and the 26-mm support from L1 to T8; the
upper limit of contact varied slightly among subjects.
The sessions were carried out in a dismounted BAE
Hawk MK 51 jet trainer ejection seat without rudders.
During each session the subjects performed 4 –5 separate AGSMs of the type termed “L-1” (3,16). Each effort
lasted approximately 10 s with breaks of at least 1-min
between efforts. The subjects were instructed to perform as powerful an AGSM as possible and to contract
all the muscles simultaneously. The subjects’ flight experience with the Hawk MK 51 aircraft was on average
58 h for the cadets and 560 h for the instructors.
Before starting the sessions, an experienced instructor
reviewed the L-1 technique with the subjects, after
which they were allowed to practice the AGSM as much
as they wished sitting on a bench. The subjects trained
at their own pace and usually performed approximately
10 –15 repetitions. When they were placed in the ejection seat and preparing for a formal AGSM session, the
subjects tried the AGSM lightly a few times with the
lumbar support. To verify the straightening effect of the
lumbar support on the spine, the angle of the spine at
T6 was measured with a goniometer (mie, Medical
Research Ltd, UK), first in a normal standing position
(reference angle) and then in the seated position without and with all the supports. In the seated position, the
lower portion of the spine (e.g., L5 level) was in contact
with the seat or the lumbar support; thus, the effect of
the lumbar support on the angle of the lower back was
not possible to measure.
During the AGSM sessions, the effectiveness of muscle contractions were evaluated by measuring the surface electromyogram (EMG) from eight different muscles on the right hand side of the body. The muscles
were: m. rectus femoris, m. biceps femoris, m. gluteus
maximus, m. erector spinae (L 4 –5 level), m. latissimus
dorsi, m. rectus abdominis (L 4 –5 level), m. obliquus
externus, and m. pectoralis major. The measurements
were performed with a portable 8 channel EMG
(ME3000P8, Mega Electronics Ltd, Finland).
The EMG signals from the working muscles were
acquired with a sampling rate of 2000 Hz, using pregelled bipolar surface electrodes (M-OO-S, Medicotest,
Ølstykke, Denmark). The electrodes were placed over
the belly of the muscle (except at abdominal and lower
back region) and the distance between recording contacts was 2 cm. Ground electrodes were attached above
inactive tissue. The measured EMG signal was amplified 2000 times (preamplifier situated 6 cm apart from
the measuring electrodes) and the signal band between
20 and 500 Hz was full wave rectified and averaged
(aEMG) with a 10 ms time constant. From the averaged
data the percentile portion (PP) of each muscles’ activity from the total EMG was calculated. In the results
(Fig. 1), PP is expressed as pooled data from three body
regions: leg (rectus and biceps femoris and gluteus
maximus), lower torso (rectus abdominis, obliquus externus, and erector spinae), and upper torso (latissimus
dorsi and pectoralis major). The moment when each
muscle started working (recruitment order) was evaluated as well. To assess the frequency component of the
EMG, the power spectrum was estimated by a moving
Aviation, Space, and Environmental Medicine • Vol. 74, No. 8 • August 2003
Fig. 1. The proportional percentile of EMG activity in different body
parts during a single AGSM without and with the most efficient AGSM
using lumbar support. Values are mean ⫾ SE, n ⫽ 11.
fast Fourier transformation (FFT window, 1024 points).
From the power spectra, mean power frequency (MPF),
median frequency (MF), zero crossing rate (ZCR), and
power spectrum area (PSA) were calculated. The parameters MPF, MF, and ZCR are conventionally used
for assessing the fatigue of working muscles. In a fatigued condition, there is a decline in these variables,
i.e., a shift to lower frequencies (14); therefore, they
were used to indicate possible fatiguing effects of the
AGSM sessions. It has been shown that during isometric muscle contraction there is a rather linear relationship between the force produced and EMG recorded
(11). In the literature the relationship between PSA and
force production is not as clearly reported as in the case
of aEMG. However, as PSA was found to be sensitive in
a similar fashion as aEMG, both parameters aEMG and
PSA, were chosen to indicate the effectiveness of muscle
contraction in this study. All the variables mentioned
above were analyzed for the whole AGSM session and
for the most efficient single maneuver (defined as yielding the highest PSA value) in each session without and
with the support.
After the measurements the subjects were asked to
evaluate which sitting position felt the most natural and
whether the AGSM was more efficient to do without or
with the support. If the subject replied to the latter
question “with the support” then he was asked with
which support.
The data are mainly expressed as individual values,
with the exception of Fig. 1, where the data are pooled.
Those data were tested with Student’s two-sample t-test
considering the results from the position without the
support as reference. The level of significance was set at
0.05. However, due to the limited number of subjects
used in this study (n ⫽ 11) the statistical power is not
sufficient for conclusive statements, rather statistics
give support to the observed trends in physiological
findings.
RESULTS
The use of 7, 14, and 26 mm supports brought the
angle of the spine 2°, 3°, and 4°, respectively, closer to
887
LUMBAR SUPPORT & AGSM—OKSA ET AL.
TABLE I. POWER SPECTRUM AREA (␮V s⫺1) DURING THE
WHOLE TRAINING SESSION WITHOUT SUPPORT AND
WHILE USING 7, 14, AND 26 mm THICK LUMBAR SUPPORTS.
TABLE III. AVERAGE EMG (␮V) OF EIGHT MUSCLES DURING
WHOLE TRAINING SESSION WITHOUT AND WHILE USING
7, 14, AND 26 mm THICK LUMBAR SUPPORTS.
Subject
Without
7 mm
14 mm
26 mm
% Difference
Subject
Without
7 mm
14 mm
26 mm
% Difference
1
2
3
4
5
6
7
8
9
10
11
13.8
17.5
19.3
29.6
29.1
5.9
10.6
24.5
17.0
53.7
28.4
13.6
20.9
13.1
22.9
34.8
4.63
15.3
23.1
14.8
50.0
28.3
12.7
22.1
15.9
19.00
35.50
7.1
11.6
17.1
11.8
57.6
30.5
19.3
15.3
18.1
24.3
42.2
3.9
21.4
17.0
10.5
48.5
32.7
28
21
⫺6
⫺18
31
17
50
⫺6
⫺13
7
13
1
2
3
4
5
6
7
8
9
10
11
90
115
121
157
159
53
76
135
98
196
134
90
126
99
135
174
40
99
126
87
191
132
93
129
104
126
176
54
84
107
82
205
139
116
114
114
142
191
40
116
103
75
180
147
22
11
–6
–11
17
2
34
–7
–13
4
9
The most efficient training session has been marked with bold and the
% difference indicates the difference between the most efficient session and the reference. In the case the reference is the most efficient,
the difference is between reference and the second best session.
The most efficient training session has been marked with bold and the
% difference indicates the difference between the most efficient session and the reference. In the case the reference is the most efficient,
the difference is between reference and the second best session.
the reference angle measured during standing. All of
the subjects reported that the posture felt more natural
and the perceived efficacy of the AGSM was better with
lumbar support; 9 subjects judged the 26-mm and 2
subjects the 14-mm support to be the best in increasing
the perceived efficacy of the AGSM.
Based on PSA results during the whole session, the
use of lumbar support increased the effectiveness of
muscular work in 7 subjects (Table I). The average
increase in the effectiveness was 11%. In 4 subjects the
effectiveness was best without the support.
During the single most efficient AGSM, the effectiveness of muscular work increased by 20%, 26%, and 44%
while using 7, 14, and 26 mm thick supports, respectively. The average increase in the effectiveness (regardless of which support was used) was 25%. The effectiveness increased in 10 subjects (Table II).
Based on aEMG results during the whole session, the
use of lumbar support increased the effectiveness of
muscular work in 7 subjects (Table III). The average
increase in the effectiveness was 6%. In 4 subjects the
effectiveness was best without the support.
During the single most efficient AGSM the effective-
ness of muscular work increased by 12%, 10%, and 14%
while using 7, 14, and 26 mm thick supports, respectively. The average increase in the effectiveness (regardless of which support was used) was 12%. The effectiveness increased in all 11 subjects (Table IV).
Fig. 1 illustrates the proportion of percentile EMG
activity representing different body parts during single
best AGSM. A nonsignificant tendency can be seen that
the use of lumbar support enhanced the contractility of
the muscles in the torso (Fig. 1).
There were no differences in the recruitment order of
the muscles between sessions, rather they contracted
quite simultaneously. Parameters MPF, MF, and ZCR
did not show any systematic differences between conditions.
TABLE II. POWER SPECTRUM AREA (␮V s⫺1) DURING
SINGLE AGSM WITHOUT SUPPORT AND WHILE USING
7, 14, AND 26 mm THICK LUMBAR SUPPORTS.
TABLE IV. AVERAGE EMG (␮V) DURING SINGLE AGSM
WITHOUT SUPPORT AND WHILE USING 7, 14, AND
26 mm THICK LUMBAR SUPPORTS.
DISCUSSION
The results of this study show that the use of lumbar
support tends to enhance the effectiveness of muscular
work, especially during a single AGSM and to a lesser
extent during repeated AGSMs. One factor that might
explain the increase in the effectiveness of muscular
work during single AGSM is that with the aid of lumbar
Subject
Without
7 mm
14 mm
26 mm
% Difference
Subject
Without
7 mm
14 mm
26 mm
% Difference
1
2
3
4
5
6
7
8
9
10
11
19.0
21.5
21.6
32.4
35.6
7.0
18.9
33.0
22.6
43.7
33.1
12.0
27.5
13.2
40.3
50.9
10.5
20.6
35.6
18.4
39.0
30.3
13.9
24.6
29.8
29.3
61.5
9.6
25.2
16.4
13.9
52.3
41.3
26.5
17.9
26.1
28.9
51.1
3.5
48.0
20.9
21.4
47.5
24.0
28
22
28
20
42
33
61
7
⫺5
16
20
1
2
3
4
5
6
7
8
9
10
11
123
129
136
179
182
57
118
161
114
196
144
88
150
101
189
204
75
126
171
102
191
133
94
135
142
157
254
68
136
99
90
205
150
126
122
137
157
204
39
192
116
115
190
113
2
14
4
5
28
24
39
6
1
4
4
The most efficient training session has been marked with bold and the
% difference indicates the difference between the most efficient session and the reference. In the case the reference is the most efficient,
the difference is between reference and the second best session.
888
The most efficient training session has been marked with bold and the
% difference indicates the difference between the most efficient session and the reference. In the case the reference is the most efficient,
the difference is between reference and the second best session.
Aviation, Space, and Environmental Medicine • Vol. 74, No. 8 • August 2003
LUMBAR SUPPORT & AGSM—OKSA ET AL.
support the spine is “forced” to a more upright position
and the torso is straighter. In a more upright position
the abdominal, back, and chest muscles are closer to
their midlength. At their midlength, muscles are able to
produce force to a greater extent than when they are
shortened or lengthened (8). In an isometric contraction
the relationship between EMG and the force produced
has been found to be close to linear (2). When the
posture is upright it might be expected that the muscles
in the torso would contract more forcefully than when
the posture is bent, thus producing a larger amount of
EMG activity. This is supported by the observation in
the results that there is a tendency (nonsignificant) to
see more EMG activity in the muscles of the torso when
the lumbar support is being used. This shift of EMG
activity toward the torso can be considered beneficial in
terms of effectiveness of AGSM while flying Hawk 51
MK aircraft, because in that plane type anti-G trousers
are used but anti-G vests are not. Therefore, an efficient
contraction of the upper body muscles is the only
means of increasing intra-abdominal and intrathoracic
pressure, which in turn increases the systemic arterial
BP.
According to these results, approximately 2/3 of the
total EMG activity comes from the upper body (lower
and upper torso) and 1/3 from the lower body. Similar
kinds of results have been obtained by Eiken et al. (5,6).
They found that while performing AGSM during a high
⫹G-load more EMG activity comes from the upper
body (ca. 60% MVC, m. rectus abdominis) than from the
lower body (ca. 35– 45% MVC, m. vastus lateralis). In a
similar fashion, Cornwall and Krock (4) found that
during fatiguing high ⫹G-loading the EMG amplitude
diminished less in the upper body than in the lower
body. These phenomena are beneficial in terms of efficient AGSM. High arterial pressure during ⫹G-loading
is a key factor in ensuring sufficient blood perfusion to
the brain (15). Strong and simultaneous contraction of
the upper body muscles increase the intra-abdominal
and intrathoracic pressure, thus increasing arterial pressure and effectiveness of AGSM.
The effect of the lumbar support during repeated
AGSM was not as clear as it was during single AGSM.
The reason for this might be explained through learning. The pilots learned to perform AGSM in a HW 51
MK seat or in a similar posture. While participating in
this study, it was the first time that the pilots performed
AGSM with a lumbar support, which changes the posture they normally use and affects the motor performance of AGSM. It might well be that more thorough
training (now just a few trials) with the lumbar supports prior to participating in the study would have
resulted in even more efficient muscular work during
AGSM sessions. However, even the magnitude of enhancement of the effectiveness of muscular work found
in this study will most likely be of importance in reducing the strain and fatiguing effects of ⫹G-loading during actual flight tasks.
On the other hand, because the results were obtained
in the laboratory (1-G) environment, the direct application of these results to high ⫹G environment may not
be justified. It is possible that that due to changes in
Aviation, Space, and Environmental Medicine • Vol. 74, No. 8 • August 2003
posture, the effectiveness of muscular work in a high
⫹G environment may be different than what the results
from the present study indicate.
All the subjects felt that the sitting posture and the
effectiveness of AGSM were better with a lumbar support. When comparing which support the subjects perceived as the most effective in increasing their AGSM
efficiency and which support the EMG results showed
to be the most efficient, four of the subjects voted for the
same support that the “objective” EMG showed to be
the most efficient. In general, the subjects identified the
beneficial effect of the lumbar support but were not
very accurate in specifying which one was the best
(presuming that the EMG result indicates the most efficient AGSM). This is not surprising since the subjects
were doing such an evaluation for the first time. Presumably, with training the accuracy of such evaluation
would improve.
The EMG parameters used in evaluating the effectiveness of muscular work during AGSM gave very
similar results. Both in the single and repeated AGSM,
the PSA parameter gave percentile higher figures than
aEMG. This difference may be due to different ways of
calculating those parameters—aEMG is a mere mean of
EMG amplitude, whereas PSA is the total area of the
spectra and also takes into account the frequency component of EMG. In the frequency parameters of EMG,
we found no systematic shift to lower frequencies in
any of the studied conditions. This indicates that the
type and duration of the AGSM sessions performed in
this study did not produce measurable fatigue to the
working muscles.
In conclusion, regardless of which parameter is
used (PSA, aEMG, subjective evaluation), the results
show that the use of lumbar support enhances the
effectiveness of muscular work, especially during single AGSM, but also during repeated AGSM in ⫹1 Gz
condition.
REFERENCES
1. Bain B, Jacobs I, Buick F. Electromyographic indices of muscle
fatigue during simulated air combat maneuvering. Aviat Space
Environ Med 1994; 65:193– 8.
2. Bigland B, Lippold OCJ. The relation between force, velocity and
integrated electrical activity in human muscles. J Physiol 1954;
123:214.
3. Burton RR, Leverett SD Jr, Michaelson ED. Man at high sustained ⫹Gz acceleration: a review. Aerosp Med 1974; 45:
1115–36.
4. Cornwall M, Krock L. Electromyographic activity while performing the anti-G straining maneuver during high sustained acceleration. Aviat Space Environ Med 1992; 63:971–5.
5. Eiken O, Kölegård R, Lindborg B, et al. Protection against
increased gravitational (G) forces afforded by the hydrostatic anti-G suit Libelle® is not adequate for use in a 9 G
aircraft. Stockholm, Sweden: Swedish Defence Research
Agency; 2001. Scientific Report: FOI-R– 0085–SE, ISSN 1650 –
1942.
6. Eiken O, Kölegård R, Lindborg B, et al. The effect of muscular
straining on G-protection during assisted pressure breathing.
Stockholm, Sweden: Swedish Defence Research Agency; 2001.
Scientific Report: FOI-R– 0230 –SE, ISSN 1650 –1942.
7. Froom P, Ribak J, Tendler Y, et al. Spondylolisthesis in pilots: a
follow-up study. Aviat Space Environ Med 1987; 58:588 –9.
8. Gordon AM, Huxley AF, Julian FJ. The variation in isometric
tension with sarcomere length in vertebrate muscle fibres.
J Physiol 1964; 1:170 –92.
889
LUMBAR SUPPORT & AGSM—OKSA ET AL.
9. Hämäläinen O, Vanharanta H. Effect of Gz forces and head movements on cervical erector spinae muscle strain. Aviat Space
Environ Med 1992; 63:709 –16.
10. Hämäläinen O. Flight helmet weight, ⫹Gz forces, and neck muscle strain. Aviat Space Environ Med 1993; 64:55–7.
11. Moritani T, DeVries H. Neural factors versus hypertrophy in the
time course of muscle strength gain. Am J Phys Med 1979;
58:115–30.
12. Oksa J, Hämäläinen O, Rissanen S, et al. Muscle strain during
aerial combat maneuvering exercise. Aviat Space Environ Med
1996; 67:1138 – 43.
13. Oksa J, Hämäläinen O, Rissanen S, et al. Muscle fatigue caused by
repeated aerial combat maneuvering exercise. Aviat Space Environ Med 1999; 70:556 – 60.
14. Petrofsky JS, Lind AR. The influence of temperature on the amplitude and frequency components of the EMG during brief
and sustained isometric contractions. Eur J Appl Physiol 1980;
44:189 –200.
15. Wood EH. Development of anti-G suits and their limitations.
Aviat Space Environ Med 1987; 58:699 –706.
16. Wood EH, Lambert EH, Baldes EJ, et al. Effects of acceleration in
relation to aviation. Fed Proc 1946; 5:327– 44.
ERRATA
In the March 2003 Table of Contents, the following article was omitted:
“Hypothermia and Local Cold Injuries in Combat and Non-Combat Situations—The Israeli Experience” by D. S. Moran, Y. Heled, Y. Shani, and Y.
Epstein, pages 281–284. We apologize to the authors for this error.
In the August 2000 issue, the article “The Role of the Flight Surgeon in
Greece” pages 851–3, please note that the first author’s name should read
Vaseleios Hriskos (not Chriskos) throughout the manuscript and in the
Table of Contents. We apologize to the author for this error.
890
Aviation, Space, and Environmental Medicine • Vol. 74, No. 8 • August 2003