petrovsky#2jarfall03 12/29/03 2:18 pm page 402 curl with … · well as due to the machine itself....

Vol. 3, No. 4, Fall 2003 • The Journal of Applied Research402

i m p o rtant for people with disabilities, sincework may not be possible in a supine posi-tion. Abdominal exercise in the seatedposition with the 6 Second Abs machineappears to control timing better and is moremuscle specific than exercise in the supineposition.

INTRODUCTIONAbdominal exercise is a common type ofexercise because of body image and posturec o n c e rns. Weak abdominal muscles arealso associated with low back pain.1 , 2

Because of the importance of the abdomi-nal muscles, historically, the United StatesA rmy has assessed abdominal endurance intheir physical fitness test, as a means ofscreening applicants.3

Posture is part i c u l a r ly important forpeople with disabilities. When seated in awh e e l c h a i r, the abdominal and lower backmuscles are important in maintaining prop-er posture and preventing scoliosis, ky p h o-sis and lordosis.4 , 5 In addition, abdominalexercise can be useful in rehabilitation pro-grams to improve bowel and bladder func-tion in people with disabilities.6

The most common abdominal exe r c i s eis a partial curl. These exercises can be

Comparison Between an AbdominalCurl with Timed Curls on aPortable Abdominal MachineJerrold S. Petrofsky, PhD, JD*†

Julie Bonacci, BS†

Trinidad Bonilla, BS†

Rachel Jorritsma, BS†

Amy Morris, BS†

Abdul-Majeed Al-Malty, MS, PT*

*Department of Physical T h e ra py, Loma Linda Univers i t y, Loma Linda, Califo r n i a†Department of Physical T h e ra py, Azusa Pa c i fic Univers i t y, Azusa, Califo r n i a

KEY WO R D S : abdominal, exercise, exe r-tion, fatigue, strength.

ABSTRACTS even male and 4 female subjects we r eexamined to assess muscle use, while per-f o rming standard abdominal crunches ly i n gin the supine position versus abdominalc runches performed in the seated and supinepositions using an exercise device called the6 Second Abs. The results of the ex p e r i-ments showed that much of the work in asupine sit up was due to work against gr av i-t y. The addition of resistive bands, wh e nexercising in the supine position, caused anincrease in work accomplished duringabdominal exercise by at least 15-fold.F u rt h e r, significant exercise was accom-plished when using the machine in the seat-ed position facing forward. Mores i g n i fi c a n t ly, by rotating the trunk to theseated position, muscle use can be sw i t c h e dfrom the rectus abdominus, facing forwa r dto the obliques when the trunk is rotated tothe side, thereby increasing oblique wo r k15-fold over standard crunches. In additionto advantages of abdominal exercise in then o rmal population, this type of exercise is

Petrovsky#2JARFall03 12/29/03 2:18 PM Page 402

p e r f o rmed supine with the feet on the floorwith the heels 12 to 18 inches apart and theknees flexe d .7 Abdominals are flexed to liftthe shoulders and head off the floor to anangle of 30º. The arms can either becrossed on the chest or placed at the level ofthe head with the fingers next to the ears.The closest to the ideal position for maxi-mizing the use of the abdominal muscles isthe partial curl. Another common positionis with the knees flexed to 90º with the feeton a wall, which emphasize the rectusabdominus muscles.8 In an eff o rt to trainthe abdominal muscles, many machinesh ave been developed for athletic training,9

and rehabilitation after abdominal surg e ry1 0.In some of these exercise machines, abdom-inal exercise can be accomplished in theseated position.1 1 , 9 Other machines placethe subject in the supine position. W h e t h e ran abdominal exercise is performed by amachine in the seated position or on thef l o o r, there seems to be a great deal of va r i-ability in muscle activity during exe r c i s e .Substitution of other muscles is commonand hinders learning of only the abdominalm u s c l e s .1 2 In a recent publication Szasz eta l3 questioned the abdominal physical fi t-ness test used by the United States A rmybecause of substitution of hip flexors forabdominal muscles during standard supinec runches. The analysis of muscle use wa saccomplished through the surface elec-t r o myogram (EMG).1 3

The EMG, when measured by surfa c eelectrodes above an active muscle, repre-sents an interference pattern giving the sum-mation of activity of the underlying musclefi b e r s .1 4 The amplitude of the surface EMGis generally related to the tension deve l o p e din muscle.1 5 , 1 4 Therefore, the EMG has pro-vided a useful measure in assessing both theextent of muscle activity and musclefa t i g u e .1 6 , 1 7 , 1 8 , 1 9 D i fferent inve s t i gators havep u blished contradictory results. Somei nve s t i gators point to a linear relationshipb e t ween the amplitude of EMG and tensionin muscle during brief contraction.1 6 , 2 0

H oweve r, others point to, at least with cer-

tain muscles, such as the biceps, a slightlynon-linear relationship.2 1 While some of thevariation has been attributed to the type ofelectrode (needle or surface) or the size orposition of the electrodes2 2 , 2 3 much of thed i fference remains to be ex p l a i n e d .H oweve r, the EMG still provides the bestestimate of muscle activ i t y.

The type of exercise performed and theposition of the body affect both musclestrength and the EMG.1 3 Muscle position isi m p o rtant because it determines the amountof overlap between actin and myosin in themuscle myo fibrils and thereby aff e c t sstrength. Furt h e r, muscle position can aff e c tthe EMG, since as muscles contract underelectrodes, the contracting muscle eitherm oves toward or away from the electrodeschanging the amplitude of the EMG. Fo rthis reason, then, controlling muscle posi-tion is important both in determ i n i n gstrength and the extent of exercise, as we l las, the muscle use, as assessed by the EMG.

This is confounded even further wh e np e r f o rming abdominal exercise in the seatedversus the supine position. Cert a i n ly, in theseated position, gr avity would increase theapparent strength of the muscle, as wo u l dthe contribution of other muscles associatedwith posture. In the supine position, indi-viduals work against gr avity whereas thepostural muscles are totally relaxed wh i l elying on the floor. In this respect, then,while abdominal exercise is performed ind i fferent positions with different types ofm a c h i n e s ,1 1 no studies have measured thed i fferences in abdominal muscle strength invarious body positions, and attempted toeliminate the effect of gr avity in strengthmeasurements.

Thus, some of the differences seen invarious studies between different abdominalmachines may be positionally related, aswell as due to the machine itself. Furt h e r,u n l i ke bicycle erg o m e t ry, where exe r c i s ecan be timed against a metronome ors p e e d o m e t e r, abdominal exercise can beva r i a ble in length depending on the wishesof the individual accomplishing the exercise.

The Journal of Applied Research • Vol. 3, No. 4, Fall 2003 403


The purpose of the present inve s t i ga t i o nwas 1) to first assess muscle strength and 2)through the use of EMG, to assess muscleuse during abdominal crunches in 2 diff e r-ent body positions (supine with the leg sstraight, and supine with the knees bent at90º) and compare this to a reg i m e n t e dmachine called the 6 Second Abs machine,with which timing and load to assess muscleuse and muscle training could be controlled,in both the supine and seated position.

SUBJECTSFour female and 7 male research subjectsp a rticipated in these experiments. Ta ble 1lists heights, weights, age, and sex of part i c-i p a n t s .

METHODS

6 Second Abs Machine The 6 Second Abs machine is a commercialexercise device (Savvier Inc, Indian We l l s ,Ca). The device consists of a rectangularplastic frame with rubber bands on theinside to adjust resistance. Resistance canbe increased in a number of different stagesso that it becomes increasingly more diffi-cult to compress the rectangle (Figures 1and 2). As shown in Figure 2, as the

machine was compressed to the first, secondand third click position with 3 diff e r e n tresistance bands, there was a linear increasein load. The upper part of the rectangle wa splaced under the subject’s arms (under thetriceps muscles bilaterally) or held in thep e r s o n ’s arms against the chest, while thebase of the rectangle was placed on top ofthe middle of the quadriceps muscles. Boththe upper and lower rectangles were padded.

Electromyogram The EMG was recorded through 2 bipolarv i nyl adhesive electrodes (silver silve r- c h l o-ride) with an active surface area of 0.5 cm2.The first electrode was placed over the bellyof the active muscle. The second electrodewas placed 2 cm distal to the active elec-trode. EMG was amplified using a 4-chan-nel EMG amplifier with a flat frequencyresponse from DC to 1000 Hz. The com-mon mode rejection ratio of the amplifi e rwas greater than 120 Db. The EMG wa sthen digitized at 2000 samples/second by aBiopac (Biopac Corp., Santa Barbara, Ca),12-bit analog to digital conve rt e r, and dis-p l ayed and stored on an IBM computer forlater analysis. To help hold the electrode inplace, a layer of collodian was appliedaround and on top of the electrodes, so thatthe electrodes would adhere to the skin and

Table 1. General Characteristics of Subjects

Subject Height (cm) Weight (kg) Age (yrs) Sex1 166 56.4 23 F2 174 71.4 22 F3 152.5 52.3 26 F4 171 70.9 24 F5 173 71.4 34 M6 173 72 23 M7 178 71 24 M8 182.5 87.3 29 M9 175 77.3 26 M10 188 95.5 34 M11 185.4 79.5 26 MMean 174.4 73.2 26.5SD 9.8 12.2 4.2



would not shift impedance during data col-lection. We have prev i o u s ly used this tech-nique over a period of as long as 4 hoursduring exercise, with no movement of thee l e c t r o d e s .1 7 The amplitude of EMG wa sassessed by digitizing and half wave rectify-ing the raw data and calculating the RMSvoltage of EMG.

Measurement of Abdominal MuscleStrengthIsometric strength of the abdominal muscleswas measured in both the seated and supinepositions. To accomplish this, initially sub-jects sat or lay with the hips at a 90º angleand a cotton strap was placed around thechest and connected to an isometric stringgauge transducer. The strain gauge was lin-ear from 0 to 200 kg of force. The outputof the transducer was amplified with a straingauge conditioner amplifier that had a ga i nof 1000 times and digitized in a Biopac 12bit analog to digital conve rt e r. It was dis-p l ayed and stored on an IBM computer.The output was stored and analy z e d, as theaverage strength over the middle of a 3 sec-ond contraction. Strength was measured by

an isometric contraction with various anglesat the hip to examine the relationshipb e t ween muscle strength and position of thehip. At least one minute was allowe db e t ween contractions to allow for recove ry.

PROCEDURESTwo series of experiments were perform e das outlined below.

Series 1: Affect of Gravity on Strength ofthe Abdominal Muscles at Different JointAngles.

The strength generated at the abdomi-nal muscles was assessed with gr av i t y,a gainst gr av i t y, and gr avity eliminated dur-ing a three second maximum eff o rt. T h i swas accomplished with subjects in the fol-l owing positions: seated, lying supine on thef l o o r, and floating in a seated position in ahy d r o t h e r a py pool. The hip angles we r emeasured using a Goniometer. The station-a ry arm was aligned with the midline of thet runk and the fulcrum placed over thegreater trochanter; the moving arm wa sangled towards the lateral epicondyle of thef e m u r. Strength was measured at 45º, 60º,75º, and 90º angles. During this contrac-tion, the subject’s arms were crossed ove rtheir chest while their feet remained on thef l o o r. In the supine position the subjectswere on an examining table with their hipsand knees at a 90º angle against a wall. T h es u b j e c t ’s trunk was then held stationary bythe same transfer belt and recording dev i c e ,


Figure 2. Relationship between click numberand force to move the 6 Second Absmachine with bands of 20, 35 and 50 lbsresistance. The results are the average of 3machines.

Figure 1. Subject in the seated position usingthe 6 Second Abs machine.

Petrovsky#2JARFall03 12/29/03 2:18 PM 05

while a research assistant supported theirhead on a pillow.

The third position was measured in thewater to eliminate gr av i t y. (Figure 3) Eachsubject was required to wear a swimsuit andenter a pool up to 4 feet of wa t e r. Wa t e rtemperature was maintained at 36ºC. Eachsubject was placed in a chair with 2 researchassistants stabilizing the chair. A thirdresearch assistant recorded angle measure-ments on the back with a Goniometer andstrength was measured

Series 2: Muscle Use During VariousAbdominal ExercisesEMG was measured over the left and rightrectus abdominus and left and right obl i q u emuscles during the following exercises:

The first position was the seated posi-tion, in which the upper part of the rectangleof the machine was placed under each sub-j e c t ’s arms (under the triceps muscles), andthe base of the rectangle was placed over themiddle of the quadriceps muscles. T h es u b j e c t ’s arms were locked to avoid arm(biceps) substitution, since the device isdesigned to be a pure abdominal exe r c i s e r.In different experiments, subjects eitherexercised by flexing the abdominals andt h e r e by, bending over or decreasing theangle at the hips by raising the legs off ofthe floor.

In the second position, the abdominalmachine was positioned over the left quadri-ceps only, and the right arm was placed ove r

the handles. This twisted the body to the leftand performed the oblique side-bendingbout. The contraction of the device wa sthen accomplished in this position in orderto gain invo l vement of the oblique musclegroup.

In the third position, the standard curlwas performed while laying supine on thefloor with the hip and knees bent to approx-i m a t e ly 45º with arms folded and crossedover the chest. No device was used, and thesubject lifted their torso high enough for hisor her inferior border of the scapula to clearthe floor.

The final two exercises were perform e da gainst a wall with the knees and hips at 90ºangles. One exercise was performed usingthe 6 Second Abs machine (Figure 4) andthe other exercise was performed withoutthe machine. The subject curled straight upand then slow ly returned to the start, highenough to clear the inferior border of thescapula off the floor.

Statistical AnalysisStatistical analysis invo l ved the calculationsof means standard deviations and T tests.The level of significance was P< 0 . 0 5 .

RESULTS

Series 1 – The Relationship BetweenStrength and Angle of the HipFigure 5 shows the relationship betwe e nstrength (in pounds) and the angle of thehip for subjects in the seated position, ly i n gon the floor, and standing in wa t e r. Each


Figure 4. Subject doing an abdominal 6-sec-ond crunch.

Figure 3. Subject undergoing testing for lowerback strength in the pool.


seated position, or opposing the contractionof the muscles, as is seen when subjects arein the laying position. In the wa t e r, forexample, the average strength with the hipangle at 90º was 50.9±30.2 pounds (23.1kg). The strength was also linearly reducedsuch that the hip angle of 45º to ave r a g estrength was 21.9±14.9 pounds (9.95 kg).The slope of the line relating strength toangle was signifi c a n t ly steeper with subjectssitting in the pool, when compared to thesubjects sitting against the wall (P<0.01), orlying on the floor (P<0.01).

Series 2 – Muscle Use During VariousTypes of Abdominal ExercisesThe results of this series of experiments are

point in Figure 5 represents the mean for 10subjects (6 male, 4 female) ± the appropri-ate standard deviations (SD). The standardd eviations are large because the male andfemale groups have been added together inthis particular figure. The highest strengthwas found when subjects were in the seatedposition with the hip angle at 90º. In thisposition strength averaged 57. 9±29.9pounds (26.3 kg) As the hip angle wa sd e c r e a s e d, strength was almost reduced in alinear manner such that when the hip anglewas 45º, the strength averaged 49.7±29.7pounds (22.5 kg). The lowest strengthrecorded was when subjects were laying onthe floor (Figure 5). In this position sub-jects had to contract their abdominal mus-cles against gr av i t y. Because of theinfluence of gr avity on the body, the reduc-tion in strength was dramatic (Figure 5).When lying on the floor at a hip angle of90º, the maximum strength was 20±12.4pounds (9.1 kg), demonstrating a reductionwell over 70% from strength measurementsin the seated position. In this position, thestandard deviations were large as well. Fo rexample, at a hip angle of 45º, strengthaveraged 8.9±12.9 pounds (4.0 kg).I n t e rm e d i a ry between these measurementsin the chair and floor were measurements inthe wa t e r. In the wa t e r, subjects were neu-t r a l ly bu oyant such that there were no gr av i-tational effects either aiding themeasurement of strength, as was seen in the

Figure 5. The strength of the abdominal mus-cles during flexion measured at 90º, 75º, 60º,and 45º at the hip when seated, supine andin the water.

Figure 6. The average EMG amplitude of 10subjects is shown during abdominal exerciseat either 20 (1), 35 (2), or 55 (3) pounds ofresistance using the abdominal machine orperforming free abdominal crunches (trian-gles) for the left and right rectus muscles.Because of differences between individuals,EMG amplitudes have been normalized as apercent of the EMG amplitude during a max-imal voluntary effort. The ordinate isexpressed as a percent of the maximumEMG generated by the abdominal muscles.This corresponds to a percent of muscleactivity. Using the 6 Second Abs method,exercise against the heaviest workload (3)the average subject used 90% of their musclepower. The graph also illustrates the relation-ship between workload and normalized EMGactivity, when subjects exercised in the sittingand side sitting position, performed freeabdominal crunches, 6 Second Abs crunch-es, and seated leg raises (see methods).



s h own in Figures 2, 6, 7, 8, and 9. Eachfigure shows the average results of 10 sub-jects performing standard abdominalc runches and using the 6 Second A b smachine: in the seated position with thebody turned sideways to the left by approx i-m a t e ly 30º; with the seated leg raise abdom-inal crunch (only in some of the studies);and the 6 Second Abs machine abdominalc runches. Because of the number of gr a p h sin each panel, standard deviations have beenomitted. For example, Figures 6 and 7 showthe peak EMG activity for the right rectus

abdominus muscles, (Figure 6), left rectusabdominus, the right oblique muscles (fi g-ure 7), and left rectus muscles. The num-bers shown represent the average peak EMGa c t ivity of 10 subjects, exercising with loadsof 20 (1), 35 (2), and 55 (3) pounds wh i l eusing the abdominal machine. The cru n c hsubjects simply lay on the floor and per-f o rmed standard abdominal cru n c h e s .Therefore, only one data point is shown forthe crunch in the figures, since there was noway to adjust the load except by the amountof work necessary to move the body aga i n s tgr av i t y. As can be seen in these figures, thegreatest muscle activity for the right and leftrectus muscles was during the 6 Second A b smethod. Standard abdominal crunches useda p p r ox i m a t e ly 25% of the muscle activ i t y

Figure 7. The average EMG amplitude of 10subjects during abdominal exercise at either20 (1), 35 (2), or 55 (3) pounds of resistanceusing the 6 Second Abs machine or doingfree abdominal crunches (triangles) for theleft and right oblique muscles is illustrated.EMG amplitudes, because of differencesbetween individuals, have been normalizedas a percent of the EMG amplitude during amaximal voluntary effort. The ordinate isexpressed as a percent of the maximumEMG generated by the abdominal muscles.This corresponds to a percent of muscleactivity. The 6 Second Abs machine duringexercise against the heaviest workload (3)the average subject used 90% of their musclepower. Graphs show the relationshipbetween workload and normalized EMGactivity when subjects exercised in the sittingand side sitting position, performed freeabdominal crunches, 6 Second Abs crunch-es, and seated leg raises (see methods).

Figure 8. The total work done during theabdominal exercise against loads of 20 (1),35 (2), and 55 (3) pound bands for exerciseusing the abdominal machine in the sittingand side sitting position, during the horizontal6 Second Abs method and the sitting legraise positions for the left and right rectusmuscles. The data are compared againsttotal muscle work during standard abdominalcrunches laying horizontal on the floor, withthe legs and hips at 90º against the wall(approximately 45º crunch). Each bar repre-sents the mean of 10 subjects. EMG isexpressed in relative work units (the productof the duration of the exercise and the aver-age EMG activity).




for the right rectus abdominals, when com-pared to the 6 Second Abs machine usedlying on the floor. These differences bypaired T test were significant (P<0.01). T h erectus muscle activity as assessed by theEMG during side sitting exercises or legraises and even in the seated position wa sstill higher than the EMG activ i t y, wh i c hwas measured when the exercise wa saccomplished by a standard floor abdominalc runch. . For the right and left rectus mus-cles, muscle use during a crunch was equalto muscle use while performing side sittingand leg raises with the 6 Second A b smachine using the lowest setting (20pounds). Therefore, even using the lightestbands in the machine, the 6 Second A b smachine provided as good as or a betterexercise than exercising while laying on the

floor or performing abdominal crunches. The best wo r kout for the oblique mus-

cles was in the seated position while rotat-ing the trunk was rotated. Even using thel owest strength band, oblique muscle usewas 4 times that of standard floor cru n c h e s(P<0.01). Using the strongest band, muscleuse was 6 times greater than standardc runches. The exercise was even more spe-c i fic when rotated to only one side, in wh i c ho n ly one abdominal set received heavy exe r-cise (Figure 5).

H oweve r, these figures do not tell thefull story. The average time during the mus-cle activity while performing standardabdominal crunches, with the subject ly i n gon the floor, was 2.1±. 34 seconds for thegroup of 10 subjects. Whereas the 6 SecondAbs method, performed in the seated posi-tion, the horizontal position, or the sideposition, is purp o rted to have an ave r a g etime of 6 seconds, but in fact the ave r a g etime was 5.4 seconds for the 6 Second A b smethod. These differences were signifi c a n t(P<0.01). It is important to recognize thatsince the exercise was performed over alonger period of time using the 6 SecondAbs versus standard abdominal cru n c h e s ,s i m p ly looking at the average EMG ampli-tude during the method does not tru ly repre-sent the amount of work being perform e dduring exercise. Therefore, the data is rep-resented as a measure of total work per-f o rmed by multiplying the EMG amplitudeand the duration of the exercise (Figures 7and 8). For example, the work of the rightrectus abdominus muscle (the work load inr e l a t ive units for work against the 20, 35,and 55 pound bands) in the sitting, side sit-ting, 6 Second Abs method (45º cru n c h ) ,and leg raise method are shown as the ave r-age for all 10 subjects. Two diff e r e n tc runches are shown in this figure, the stan-dard crunch with the subject laying in a hor-izontal position on the floor, and a4 5 º - c runch position. The 45º crunch, wh i l enot using the 6 Second Abs machine, wa sp e r f o rmed with the hip at an angle of 90ºand the knees at 90º, in a similar position to

Figure 9. This figure illustrates the total workdone during the abdominal exercise againstloads of 20 (1), 35 (2), and 55 (3) poundbands for exercise using the abdominalmachine in the sitting and side sitting posi-tion, during the horizontal 6 Second Absmethod and the sitting leg raise positions forthe oblique muscles. This data is comparedagainst total muscle work during standardabdominal crunches laying horizontal on thefloor and with the legs and hips at 90ºagainst the wall (approximately 45º crunch).Each bar represents the mean of 10 subjects.EMG in this figure is expressed in relative workunits, which is the product of the duration ofthe exercise and the average EMG activity.


the 6 Second Abs method, for basis of com-parison. As can be seen in these figures, theaverage workload of the subjects when com-paring the 6 Second Abs method to the 45ºc runch was about 6 times as high. (P< 0 . 0 1 )When comparing the total work perform e dby the rectus abdominal muscles using the 6Second Abs method to a standard abdominalc runch, the work was as least 10 times ashigh for the subjects. Thus for all musclesexamined (the right or left rectus abdominusor the right or left obliques) work loadswere substantially higher by about 15 foldwhen using the 6 Second Abs method com-pared to standard abdominal cru n c h e s( Figures 2,7,8,9).

The true work of the abdominal mus-cles, in the seated position with the tru n kr o t a t e d, was also higher than figure 6 show s .In this position, the oblique muscles wo r ke das much as 20 times harder in the seatedposition with the 6 Second Abs machinewhen compared to the standard cru n c h( Figure 9). This difference was signifi c a n t(P< 0 . 0 1 ) .

DISCUSSIONThe abdominal muscles have an import a n tfunction in maintaining posture while stand-ing and sitting. For example, Snigders eta l2 4 s h owed that in healthy subjects, duringunconstrained sitting, both the internal andex t e rnal oblique muscles were ve ry activecompared to standing as assessed by EMG.In most of the subjects, the activity of theo blique abdominals was signifi c a n t ly lesswhen sitting on a soft car seat than wh e ncompared to sitting on a hard office chair.Their data seemed to indicate that theo blique abdominals help to stabilize thebase of the spine, in part i c u l a r, the sacroiliacjoint, while sitting or standing. As such,strengthening the abdominal muscles seemsto be important in alleviating back andpelvic pain during prolonged standing ands i t t i n g .2 4 , 2

Despite higher body weight, abdominalmuscles are we a ker in women prone to backpain and back injury due to obesity.2 5 T h u s

i m p a i rment in this muscle group, predispos-es them to a number of other medical prob-l e m s .

C e rt a i n ly, for someone in a wh e e l c h a i r,such as a stroke patient or a patient with anincomplete spinal cord injury, where theabdominal muscles are part i a l ly active, pos-ture becomes a crucial component for main-taining stability while seated in thewh e e l c h a i r. Proper posture can also help toavoid scoliosis, kyphosis and lordosis.5 A n dyet, for people with disabilities, abdominalexercise has always been difficult to accom-plish. In fact, even for the non-disabl e dpopulation, the abdominal muscles havebeen hard to isolate with any specific exe r-cise due to substitution of other muscles,such as the hip flex o r s .3 This is especiallyt rue for what is termed the abdominalc runch, a type of exercise performed lay i n gin the supine position and flexing the tru n k .

In an eff o rt to specifi c a l ly train abdom-inal flexors, a number of studies have com-pared various types of abdominal trainingd evices to standard abdominal cru n c h e s .9

M a ny of the commercially ava i l a ble dev i c e skeep the subject in the seated or supineposition. The results are somewhat contra-d i c t o ry and va r i a ble. Howeve r, log i c a l ly,there cert a i n ly would be advantages and dis-a d vantages to exercising the abdominalmuscles in the seated position (as would bedone with a commercial exercise dev i c esuch as a Universal Gym) versus the supineposition. In the seated position, the hips andpelvic area are supported on a seat, use ofthe hip flexors is minimal and therefore, theabdominal flexors are easier to isolate. A scited above, co-contraction of the hip flex-ors has been a significant problem in assess-ing abdominal muscle strength.3 I ncontrast, in the supine position, it is easy tosubstitute the hip flexors, such as the psoasmajor or iliopsoas, for the abdominals, andit would be much more difficult to isolatemuscles if the hips are allowed to move ins everal degrees of freedom. Gravity is acomplicating fa c t o r, when changing fromone body position to another.



As reported under results, musclestrength was almost 3-fold higher in theseated position than in the supine position.When measuring strength in a pool, ton egate the effect of gr av i t y, the true strengthof the muscles can be assessed. Thus in thetherapeutic pool, when measuring strength,the true strength is measured and the actualstrength of the abdominal flexors was abouth a l f way between that measured in the seatedversus the supine position. The gr e a t e s tstrength of the muscle was with the lowe rback and hips at an angle of 90º. This isp a rt i c u l a r ly important since many abdomi-nal exercises usually start at 90º, in eitherthe supine or sitting positions.

The fact that muscle strength va r i e swith joint angle is cert a i n ly not new. W h e nmuscle is passive ly changed in length, theoverlap between the actin and myosin in themuscle myo fibrils changes and alters themaximum strength developed in muscle.2 6

This is called the length tension relation-ship. Howeve r, unlike muscles such as thebiceps and handgrip, where peak strengthoccurs about halfway between relaxationand full flex i o n ,1 4 the peak strength of thesemuscles occurs with the hips at 90º and theback straight. A ny lesser angle reducedstrength furt h e r.

In addition, in other muscles, such asthe biceps, the length tension relationship isve ry sharp; tension varies by more then50% over the normal range of motion of themuscle. In contrast, strength varies only asmall amount for the abdominal muscles,over a wide range of motion in the supine orsitting position. Norm a l ly a muscle, such asthe rectus abdominus, would show a gr e a tdeal of variation in strength with positiondue to the “long arrangement” of fibers inthis muscle. As muscle contracts, ove r l a pincreases between actin and myosin andstrength is reduced at decreasing jointangles maximally, if the muscle fibers are inthe long arrangement. The body norm a l lyminimizes the length tension relationship byplacing muscle fibers in a pennate arr a n g e-ment. Here, in the sitting or supine posi-

tions, the change in strength is small withdecreasing joint angles. The diff e r e n c eb e t ween the observed and expected resultsm ay be due to gr av i t y.

In the seated position, as the trunk isbent forwa r d, progr e s s ive ly more upperbody weight is removed from the spine andbody weight, rather than being support e dthrough the bones in the spine, is support e dthrough the lower back muscles. If thel ower back muscles relax, the weight of theupper body now provides a force to causef l exion of the trunk. This force adds to thestrength of the abdominal muscles giving atotals force that equals;Muscle Force for flexion= active force ofthe abdominals + passive weight of theupper body

Thus at a 45º angle at the hip, 70% ofthe body weight is born by the lower backmuscles. Therefore the measured musclestrength would be:muscle strength = abdominal active musclestrength + 70% upper body we i g h t

Therefore, while muscle strength maydecrease at reduced hip angles, in the seatedposition, gr avity progr e s s ive ly adds toapparent strength. By simply relaxing theback extensors, apparent strength canincrease by 70% of the body weight. T h eoverall result is, when sitting, joint angle hasve ry little affect on strength, as demonstrat-ed by the following form u l a :S = Ms+ Bm* C o s AS=muscle strength for the abdominal flex-ors, Ms= true strength of the abdominalf l exors at a given back angle, A= angle ofback relative to hip,Bm= body mass for upper body

When in the supine position, a similare ffect is seen. With the back at a 90º angle,full body weight is born by the abdominalmuscles when flexing the trunk upwards, sothat strength is reduced by upper bodyweight. As the angle of the trunk isr e d u c e d, more weight is born by the spine,and therefore, it becomes easier to flex thespine at decreasing angles. But decreasing



muscle length also reduces muscle strength.Therefore the length tension curve reacts tojoint angle less than expected.

The overall result is that strengthdecreases little with joint angle when sittingas demonstrated by the following form u l a :

S = Ms- Bm* C o s ( 9 0 - a )

S=muscle strength for the abdominal flex-ors, Ms= true strength of the abdominalf l exors at a given joint angle,A= angle of back relative to hip

In wa t e r, there is no effect of gr av i t y.Therefore, the true length tension relation-ship is seen, establishing the true strength ofthe abdominal flexors at a given joint angleat a given joint angle.

The fact that muscle strength varies solittle, over a wide degree of trunk angles,demonstrates no disadvantage to exe r c i s i n gthe rectus and oblique muscles at hip anglesof 10º, 20º, or 30º. The eff e c t ive wo r k l o a dseems to be the same and therefore, the jointposition does not need to be compensatedfor the workload produced on the exe r c i s ed evice. Thus, it would seem that a dev i c e ,which provides a linear change in force withangle of the lower back, would be appropri-ate for exercise. But the opposite is tru e .Muscle strength we a kens quickly withdecreasing hip angle. Therefore, a progr e s-s ive exercise device such as the 6 SecondAbs machine provides a better wo r kout thana constant weight device, since in order tog ive a constant wo r kout, an increasing loadwould compensate for the effect of gr av i t yon muscle strength. In this respect, simplec runches cause relative load to decreasewith joint angle, due to the effect of gr av i t y,and thereby, decrease the wo r k l o a d, as theabdominals are flexed in the seated orsupine positions. Since the wo r k l o a dincreased as the machine is compressed pastthe first, second, and third click positions( Figure 2), the decreased load due to gr av i t yis compensated for by increasing the load in

order to maintain a good wo r kout. T h edecrease in muscle strength during a 30ºc runch (Figure 5) was about 40%.H oweve r, gr avity added 30% to apparentstrength during a reduction of 30º in hipangle (Figure 5). Since the increase in forceto compress the machine to the third posi-tion was almost double the first position( Figure 2), gr avity was more than compen-sated for.

When ab roller devices have been com-pared to standard abdominal crunches, thisrelationship has been emphasized dramati-c a l ly. For example, when Demont et al9

compared 2 abdominal roller trainingd evices to a standard abdominal cru n c h ,t h ey found little difference in work or mus-cle use. This is not surprising since theroller devices, cause the trunk to flex ove rmore of an angle, and would still prov i d ethe same relative work as an abdominalc runch. This first series of experiments inthis study yielded similar results, as show nin the strength diagram (Figure 5). T h eresults of the studies by Demont et al9 a r enot surprising. This also confi rms otherfindings that show moderate repetitive loadsbuild endurance but not strength, while themost eff e c t ive modality for bu i l d i n gstrength is high load and low repetition.2 7 , 2 8

If the load is too light or too heav y, trainingis inhibited. Thus a small Ab roller or otherd evice would use too light a load for properstrength training, while large Ab machinesp r o b a bly use too heavy a load. Furt h e r,multiple sets must be used to make strengthtraining eff e c t ive .2 9

In the present inve s t i gation, as citeda b ove, when comparing a 6 Second A b smachine to standard abdominal cru n c h e s ,the EMG activity using the rectus abdomi-nus muscles was substantially higher usingthe 6 Second Abs machine in the supineposition than for standard crunches. Sincethis is a progr e s s ive resistance dev i c e ,decreasing joint angle increases the wo r k-load and increases muscle use. The wo r k-load with 30º of flexion was at least 15-foldhigher with the 6 Second Abs machine than



with standard abdominal crunches. T h i smeans that strength training can progr e s smuch faster and more effi c i e n t ly than withother devices. Quantifying the effect ofworkload on the effi c i e n cy of training is dif-ficult to accurately establish but it can beassumed that there would be signifi c a n ta d vantages in using such a machine. Basedon studies of AB crunches by the UnitedStates A rmy,3 there may be a 100-folda d vantage in strength training with progr e s-s ive resistance devices, such as with thed evice examined here. Furt h e r, by using aclicking device, the variability in the timingof abdominal crunches can be reduced.Although individuals were told to performthe standard abdominal crunches slow ly,i n ev i t a bly exercise was performed over abrief and va r i a ble period of time, ave r a g i n go n ly 1 or 2 seconds per crunch. With ac l i c ker in the machine, it was much easier tomaintain proper timing and spend approx i-m a t e ly 6 seconds per exercise cycle. Bycontracting the muscles over a heavy loadand over a long period of time, the totalwork accomplished was much higher for therectus abdominus muscles in the supine posi-tion when compared to standard cru n c h e s .

E ven in the sitting position, wo r k l o a don the rectus abdominus muscles was sub-s t a n t i a l ly higher with the 6 Second A b smachine than performing standard cru n c h e slying in the supine position. Howeve r, ascan be seen from the strength diagr a m( Figure 5), since muscle strength, due togr av i t y, is much higher in the seated posi-tion, much higher resistances must be addedto the machine to exercise eff e c t ive ly in theseated position when compared to thesupine position. Howeve r, given this restric-tion of using higher workloads in the seatedposition, the level of muscle activity andtotal work gain was substantially higher thanthat of supine cru n c h e s .

For someone with disabilities, supinec runches may be a total impossibilitybecause of the level of disability; the onlytype of exercise that might be accomplishedm ay be from the seated position.

I n d ividuals that are gr o s s ly ove r weight ori n d ividuals that have muscle weakness maynot be able to get into the supine position onthe floor, but they could exercise from aseated position and still receive a betterwo r kout than lying on the floor and try i n gto perform a standard crunch. Furt h e r, ifthe trunk is rotated to a 30º angle and exe r-cise is performed in the seated position, atleast a 15-fold increase in total work isaccomplished on the oblique muscles, andthe rectus abdominus muscles. T h e r e f o r e ,even for someone without disabilities, themachine seems to be ve ry specific in theseated position, simply flexing the trunk for-ward; the main muscles used are the rectusabdominus. Whereas, when rotating thet runk the main muscles used are theo bliques. Therefore, in the seated position,the machine can specifi c a l ly train the rectusabdominus versus the oblique muscles with-out interference from the hip flexors.

P r evious studies have shown that usinga resistance exercise program to trainabdominal muscles while performing stan-dard abdominal crunches or using some ofthe large abdominal machines has been inef-f e c t ive in changing girth and skin foldmeasurements over the abdominal area.3 0 I nthe past, exercise programs that we r edesigned to improve the appearance andfunction of the abdominal muscles we r eo n ly moderately eff e c t ive or totally ineff e c-t ive .3 1 , 3 2 H oweve r, the failure in training wa sp r o b a bly due to a failure in technique andequipment. The present inve s t i gation onlyexplored muscle use and total work duringbouts of 10 contractions. In the same stud-ies, specific muscle use of the abdominalscould not even be isolated during brief, letalone sustained wo r ko u t s .3 The implicationis that the 6 Second Abs machine is a ve rye ff e c t ive device for training endurance andmuscle strength in the abdominal muscles.Studies historically have pointed to the fa c tthat strength training requires overload ofmuscles in a standard resistance progr a m .2 6

Without sustaining a workload at a heav yl evel, no significant training will be seen. If



muscles can be substituted, then overload isd i fficult to establish and this is why prev i-ous wo r kout machines have been ineff e c t ivein abdominal training. Here, with the 6Second Abs machine subjects could achieveove r l o a d, making it an eff e c t ive trainingd evice. The additive effect of timing of theexercise through a clicker on muscle train-ing, by forcing a progr e s s ive load aga i n s tthe muscle and allowing the body to flexwithout rotation (maintaining form), helpsto reduce substitution of muscles and pro-vides a more regulated wo r kout. The fa c tthat a variety of resistance bands are ava i l-a ble, from ve ry weak bands for people withdisabilities to ve ry strong bands for the ath-lete, make it more useful than large exe r c i s emachines, in which weight can only bechanged in 10 pound increments, and theminimal load is high due to the inertia ofthe machine. In a large exercise machine,by using a load to exercise against, less timeis needed for a wo r kout, since the wo r ko u tis conducted at a higher level of activ i t y.Fi n a l ly, it is human nature to work less asfatigue ensues. By using a clicker to setpace and load, subjects are forced to wo r khard with this machine as they fa t i g u ewhereas with crunches, as fatigue approach-es, people work faster and at less of anangle of flex i o n .

REFERENCES1. Deyo RA. Conservative therapy for low back

pain. Distinguishing useful from useless therapy.JAMA. 1983;250:1057-1062.

2. Donchin M, Woolf O, Kaplan L, Flooman Y.Secondary prevention of low-back pain. A clini-cal trial. Spine 1990;15:1317-1320.

3. Szasz A, Zimmerman A, Frey E, Brady D,Spalletta. An electromyogrphical evaluation ofthe validity of the 2-minute sit-up section of theArmy Physical Fitness Test in measuring abdomi-nal strength and endurance. Mil Med.2002;167:950-953.

4. Warden SJ, Wajswelner H, Bennell KL. 1999Comparison of abshaper and conventionally per-formed abdominal exercises using surface elec-tromyography. Med Sci Sports Exerc31:1656-1664.

5. Youdas JW, Garrett TR, Harmsen S, Suman VJ,Carey JR. Lumbar lordosis and pelvic inclination

of asymptomatic adults. Phys Ther.1996;76:1066-1081.

6. Ryerson, S. Neurological Rehabilitation. In:Umphred, D, Ed. Stroke St. Louis, Mo: MosbyInc; 1995:681-721.

7. Hall CM, Brody LT. Therapeutic exercise:Moving toward function. Lippincott Philadelphia,PA: Williams &Wilkins; 1999:109-133.

8. Nieman, DC. Exercise testing and prescription: ahealth related approach. 4thed Mountain View,CA: Mayfield Pub Co; 1999:22-39.

9. Demont RG, Lephart SM, Giraldo JL,Giannantonio FP, Yuktananandana P, Fu FH.Comparison of two abdominal training deviceswith an abdominal crunch using strength andEMG measurements. J Sports Med Phys Fitness.1999;39:253-258.

10. Hemingway AE, Herrington L, Blower AL.Changes in muscle strength and pain inresponseto surgical repair of posterior abdominal wall dis-ruption followed by rehabilitation. Br J SportsMed. 2003;37:54-58.

11. Sands WA, McNeal JR. A kinematic comparisonof four abdominal training device and a tradition-al abdominal crunch. J Strength Cond Res.2002;16:135-141.

12. Miller MI, Medeiros JM. Recruitment of internaloblique and transverses abdominis muscles dur-ing the eccentric phase of the curl-up exercise.Phys Ther.198767:1213-1217.

13. Andersson EA, Ma Z, Thorstensson A. RelativeEMG levels in training exercises for abdominalsand hip flexor muscles. Scan J Rehab Med.1998;30:175-183.

14. Petrofsky JS. Computer analysis of the surfaceEMG during isometric exercise. Comp Biol Med1980;10:83-95.

15. Petrofsky JS, Laymon M. The relationshipbetween muscle temperature, MUAP conductionvelocity and the amplitude and frequency com-ponents of the surface EMG during isometriccontractions. Journal Orthop. Neurol Surgery. Inpress.

16. Bigland B, Lippold, O. The relation betweenforce, velocity and integrated EMG. J Physiol.1954;123:214-224.

17. Petrofsky JS, Dahlms, T, Lind AR. Power spec-trum of the EMG during static exercise.Physiologist. 1975;18:350.

18. Lindstrom L, Magnusson R, Petersen I. Muscularfatigue and action potential conduction velocitychanges studied with frequency analysis of EMGsignals. Electromyography. 1970;10:341-356.

19. Karlsson S, Gerdle B. Mean frequency and signalamplitude of the surface EMG of the quadricepsmuscles increase with increasing torque—a studyusing the continuous wavelet transform. JElectromyogr Kinesiol. 2001;11:131-140.



20. Petrofsky JS, Lind AR. Frequency analysis of thesurface EMG during sustained isometric contrac-tions exercise. Europ J Physiol 1980;43:173-182.

21. Komi PV, Buskirk ER. Reproducibility of elec-tromyographic measurements with inserted wireelectrodes and surface electrodes.Electromyography. 1970;10:357-367.

22. Alon G. High voltage stimulation. Effects ofelectrode size on basic excitatory responses.Phys Ther. 1985;65:890-895.

23. Alon G, Kantor G, Ho HS. Effects of electrodesize on basic excitatory responses and on select-ed stimulus parameters. J Orthop Sports PhysTher. 1994;20:29-35.

24. Snijders CJ, Bakker MP, Vleeming A, StoeckartR, Stam HJ. Oblique abdominal muscle activityin standing and in sitting on hard and soft seats.Clin Biomech. 1995;10:73-79.

25. Hulens M, Vansant G, Lysens R, Claessens AL,Muls E, Brumagne S. Study of differences inperipheral muscle strength of lean versus obesewomen: an allometric approach. Int J ObesRelat Metab Disord. 2001;25:676-681.

26. Astrand PO, Rodahl K. Physiology of WorkCapacity and Fatigue. New York, NY: McGrawHill; 2001:137-159.

27. Berger RA, Harris MW. Effects of various repeti-tive rates in weight training on improvements instrength and endurance. J Assoc Phys MentRehabil. 1966;20:205-207.

28. Faigenbaum AD, Westcott WL, Loud RL, LongC. The effects of different resistance training pro-tocols on muscular strength and endurance devel-opment in children. Pediatrics.1999;104:1154-1182.29.

29. Rhea MR, Alvar BA, Ball SD, Burkett LN.Three sets of weight training superior to 1 setwith equal intensity for eliciting strength. JStrength Cond Res. 2002;16:525-529.

30. Thomas TR, Ridder MB. Resistance exerciseprogram effects on abdominal function andphysique. J Sports Med Phys Fitness.1989;29:45-8.

31. Katch FI, Clarkson PM, Kroll W, et al. Effectsof sit-up exercise training on adiposecell sizeand adiposity. Res Q Ex Sport. 1984;55:242-247.

32. Noland M, Kearney JT. Anthropometric and den-sitometric response of women to specific andgeneral exercise. Res Q. 1978;49:322-8.



petrovsky#2jarfall03 12/29/03 2:18 pm page 402 curl with … · well as due to the machine itself....

Documents