Isolation of the Vastus Medialis ObliqueMuscle During Exercise

Edwin Mirzabeigi,* MD, Christopher Jordan, MD, JoAnne K. Gronley, DPT,Neal L. Rockowitz, MD, and Jacquelin Perry, MD

From the Pathokinesiology Laboratory, Rancho Los Amigos Medical Center,Downey, California

ABSTRACT

The purpose of this study was to selectively challengethe vastus medialis oblique muscle in comparison withthe vastus lateralis, the vastus intermedius, and thevastus medialis longus muscles by performing ninesets of strengthening exercises. These knee rehabili-tation exercises included isometric knee extension withthe hip at neutral, 30° external, and 30° internal rota-tion; isokinetic knee extension through full range; iso-kinetic knee extension in the terminal 30° arc; sidelyingipsilateral and contralateral full knee extension; andstand and jump from full squat. Electrical activity of thevastus medialis oblique, the vastus lateralis, the vastusintermedius, and the vastus medialis longus muscleswas measured in eight uninjured subjects. Our studyshowed that isometric exercises in neutral and externalrotation of the hip will challenge both the vastus medi-alis oblique and the vastus lateralis muscles. The re-sults suggest that the electromyographic activity of thevastus medialis oblique muscle was not significantlygreater than that of the vastus lateralis, the vastusintermedius, and the vastus medialis longus musclesduring the nine sets of exercises. Results suggest thatthe vastus medialis oblique muscle cannot be signifi-cantly isolated during these exercises.

The current literature provides us with mixed views re-garding the isolation of the vastus lateralis muscle versusthe vastus medialis oblique muscle during exercise as aconservative treatment of lateral subluxation of the pa-tella. The major muscles acting on the knee joint are thefour heads of the quadriceps femoris muscle. The quadri-ceps muscle acts in several ways: extension of the tibia onthe femur, external rotation of the tibia in the last degreesof extension, and acceleration of the knee in gait.

Lateral subluxation of the patella or frank lateral dis-location may be the result of trauma or abnormal muscu-loskeletal architecture. There are patients, however,whose musculoskeletal structures appear to be normal butwho still have lateral tracking of the patella. It is hypoth-esized that in some of these patients there is a relativeimbalance between the vastus medialis and the vastuslateralis muscles, with the large lateral muscle having thestronger pull.3, The more horizontally directed fibers ofthe vastus medialis oblique muscle function, for the mostpart, to counteract this lateral pull.13 Dysplasia or hypo-plasia of the vastus medialis muscle from congenital, trau-matic, or postsurgical causes may result in patellarsubluxation.3,11,14

A report by Henry and Crosland6 has suggested thatnonoperative treatment of patella subluxation or disloca-tion or both yields superior results to surgical correction ofthis problem. With this in mind, it would be desirable tohave some way to selectively strengthen the vastus medi-alis muscle, especially the vastus medialis oblique, andthus correct the muscle imbalance. There are a few reportsin the literature of attempts to isolate the vastus medialisand vastus medialis oblique muscles in therapy. Knight etal.10 compared straight-leg raising with knee extensionand actually found the vastus lateralis muscle more ac-tive. Two reports describe the use of EMG biofeedback andelectrical stimulation to selectively increase vastus medi-alis muscle strength.1,12 An EMG analysis of a rehabili-tation program for patients with a history of subluxationor dislocation or both that recommends “quad-sets” (exer-cises in which a supine patient with a pad under the distalfemur tightens the quadriceps muscle and maintains thetightness so that the heel comes off the table for a fewseconds) and straight-leg raising for increasing quadricepsstrength was unable to isolate the vastus medialis from thevastus lateralis muscle.17 Hanten and Schulthies5 suggestthat the vastus medialis oblique muscle may be selectivelyactivated by performing hip adduction exercises.

The purpose of this study was to test the hypothesis thatthe vastus medialis oblique muscle can be selectively chal-lenged by performing a set of nine exercises.

* Address correspondence and reprint requests to Edwin Mirzabeigi, MD,1041 E. Yorba Linda Boulevard, Suite 8, Placentia, CA 92670.

No author or related institution has received any financial benefit fromresearch in this study.

0363-5465/99/2727-0050$02.00/0THE AMERICAN JOURNAL OF SPORTS MEDICINE, Vol. 27, No. 1© 1999 American Orthopaedic Society for Sports Medicine

50

MATERIALS AND METHODS

Subjects

Eight uninjured nonathletic volunteers participated inthis study (mean age, 26.5 years; SD, 4.23). All subjectshad no history of any previous knee injuries or injuries tothe thigh affecting quadriceps muscle function. Each sub-ject signed a statement of informed consent before partic-ipating in the study.

Instrumentation

For this study dynamic EMG was used. The electrodesconsisted of a pair of 50-mm, nylon-coated, stainless steelwires with 2-mm bared tips placed in each muscle tested.Accuracy of electrode placement in the desired muscle wasconfirmed by palpation of the tendon and visible contract-ing of the muscle during mild electrical stimulationthrough the wire. The EMG wires were inserted into thevastus medialis oblique, the vastus medialis longus, thevastus intermedius, and the vastus lateralis muscles.

The myoelectric signals were transmitted via anAM/FM telemetry system to the recording instrumenta-tion. Bandwidth of the system was 150 to 1000 Hz with aroll-off at the low end beginning at 200 Hz. A 60-Hz notchfilter excluded noise generated by the AC power. Theoverall system gain was 1000.

A single-axis, double parallelogram goniometer (fabri-cated in our department) was used to provide a continuousrecord of knee joint angles. A Cybex II isokinetic dyna-mometer (Lumex, Inc., Ronkonkoma, New York) was usedto measure maximal isometric and isokinetic knee exten-sion torque. The resulting data were stored on a seven-channel analog tape.

Procedure

Testing started by recording a resting EMG to establish athreshold for computer signal processing. Next, a series ofmanual muscle tests were performed to record the EMGduring maximum effort. Each subject performed one rep-etition of the following nine sets of exercises: 1) maximumisometric knee extension in 15° of flexion and the hip inneutral position; 2) maximum isometric knee extension in15° of flexion and 30° of external rotation at the hip; 3)maximum isometric knee extension in 15° of flexion and30° of internal rotation at the hip; 4) full arc (90° to 0°)maximum isokinetic knee extension at 60 deg/sec; 5) shortarc (30° to 0°) maximum isokinetic extension at 60 deg/sec;6) full flexion to full extension with a 10-pound weighthanging free while lying on the ipsilateral side, to cause avalgus stress on the knee; 7) full flexion to full extensionwith a 10-pound weight hanging free while lying on thecontralateral side, to cause a varus stress on the knee; 8)full squats going from a full flexed position to upright; 9)jump squats going from a full flexed position, and jumpingat the end of extension.

Data Analysis

The EMG signals were digitized at 2500 Hz per channel,rectified, and integrated over 0.02-second intervals. Nor-malized EMG values of the vastus medialis longus, thevastus intermedius, the vastus lateralis, and the vastus me-dialis oblique muscles for the nine exercises were obtainedby relating the EMG output signals to the EMG of themaximum voluntary contractions. The results were ex-pressed as percentages of the manual maximal muscle test.

Normalized EMG readings of the vastus medialis lon-gus, the vastus intermedius, the vastus lateralis, and thevastus medialis oblique muscles for the nine exerciseswere analyzed using a two-way analysis of variance withrepeated measures. Post hoc comparisons within the val-ues obtained for each of the muscles were made using theBonferroni multiple comparisons analysis. Significancewas defined as P , 0.5.

RESULTS

Maximum isometric knee extension with the hip in theneutral position resulted in a statistically significant dif-ference among the vastus lateralis, the vastus interme-dius, the vastus medialis oblique, and the vastus medialislongus muscles. (Table 1, test 1). During maximum iso-metric knee extension with the hip at 30° of externalrotation, the vastus lateralis muscle had a significantlyhigher mean EMG value than the vastus medialis obliquemuscle (Table 1, test 2). Maximum isometric knee exten-sions with the hip at 30° of internal rotation resulted in asignificantly higher mean EMG value for the vastus late-ralis muscle than those for the vastus medialis longus, thevastus intermedius, and the vastus medialis oblique mus-cles (Table 1, test 3).

Maximum isokinetic knee extension through full rangeresulted in no significant difference among the vastusmedialis longus, the vastus intermedius, the vastus late-ralis, and the vastus medialis oblique muscles at 0° to 15°(Table 1, test 4). Knee extension in the terminal 30° arcresulted in no significant difference among the vastusmedialis longus, the vastus intermedius, the vastus late-ralis, and the vastus medialis oblique muscles at 0° to 15°and 16° to 30°(Table 1, test 5).

Ipsilateral sidelying full knee flexion to extension with a10-pound weight at the ankle resulted in relativelygreater vastus intermedius muscle activity, but the EMGactivity was low in all the muscles tested (Table 1, test 6).As extension progressed from 90° to 0°, EMG activity ofthe vastus lateralis muscle was slightly greater than thatof the vastus medialis oblique muscle in all increments(3% to 6% manual maximal muscle test). Contralateralsidelying full knee flexion to extension again showed lowactivity among all groups from 90° to 0°. The vastus in-termedius muscle had the highest relative EMG activity,while lower activity was seen in the vastus lateralis mus-cle, the vastus medialis oblique muscle, and the vastusmedialis longus muscle (Table 1, test 7).

In standing from a full squat, there was no statisticalsignificance at 0° to 15°. At 16° to 30°, however, the vastusintermedius muscle demonstrated significantly higher

Vol. 27, No. 1, 1999 Isolated Exercise of the Vastus Medialis Oblique Muscle 51

EMG activity than the vastus medialis oblique muscle (Ta-ble 1, test 8). A jump from a full squat at 0° to 15°, and at 16°to 30° resulted in no statistically significant variationsamong EMG activity in all muscle groups (Table 1, test 9).

DISCUSSION

The purpose of this study was to determine whether wecould selectively challenge the vastus medialis obliquemuscle. Nine different exercises were performed by eightuninjured subjects in an attempt to discover a particularexercise or exercises useful in isolating the vastus media-lis oblique muscle. Thus, our study could clarify some ofthe current mixed views on isolation of muscle groupsduring exercise.

Maximum isometric knee extension at 15° of flexionwith the hip in neutral position (Table 1, test 1), and thehip rotated 30° externally and 30° internally (Table 1,tests 2 and 3, respectively), illustrate the highest overallEMG activity for the vastus lateralis muscle. Anatomi-cally, the vastus medialis oblique muscle lies superior tothe other muscles when the hip is externally rotated (Ta-ble 1, test 2). It would be expected that a muscle in thissuperior position would have an increase in EMG activity;however, there was a drop in the vastus medialis obliquemuscle’s EMG activity. In comparison, during internal hiprotation the vastus lateralis muscle lies superior to theother muscles, and did display the highest EMG activity(Table 1, test 3).

Consistent with the findings of Knight et al.,10 higherEMG activity occurred in the vastus lateralis muscleduring isometric knee extensions compared with the vas-

tus medialis oblique muscle. The isokinetic knee exten-sions did not specifically challenge the vastus medialisoblique muscle, contrary to previous beliefs (Table 1, tests4 and 5).17

As during the isometric exercises, the trend for thevastus lateralis muscle to exhibit the highest EMG duringisokinetic knee extension continued. Although there wereno statistically significant differences in the isokineticexercises, power calculations identified that a sample sizeof 20 subjects would have reached statistical significance.

In comparisons of the vastus lateralis and the vastusmedialis oblique muscles during sidelying full knee flexionto full knee extension, both the ipsilateral and contralat-eral sidelying positions generally support higher EMGactivity in the vastus lateralis muscle (Table 1, tests 6 and 7).These sidelying exercises did not isolate the vastus medialisoblique muscle. Neither exercise exceeded 30% to 40% man-ual maximal muscle test. Results of stand and jump from afull squat did not demonstrate any significant differencebetween the EMG activity of the vastus lateralis and thevastus medialis oblique muscles (Table 1, tests 8 and 9).

According to Lieb and Perry13 knee extension using thevastus lateralis muscle alone required 55 pounds of forcebut also produced lateral patellar subluxation. Only 10pounds of force on the vastus medialis oblique muscle wasrequired to put the patella back into its neutral position.Maintaining sufficient force on the vastus medialis obliquetendon to keep the patella centered in the intercondylargroove, these investigators found that the force required ofthe vastus lateralis muscle to fully extend the knee wasdecreased by 13%. This clearly demonstrates the vastus me-

TABLE 1Summarized Results of the Nine Sets of Exercises Presented as Percentages of the Average EMG Signal of a

Maximal Voluntary Contractiona

TestVastus muscle

Medialis longus Intermedius Lateralis Medialis oblique

Maximum isometric knee extension at 15° knee flexion1) Hip in neutral position 101 99b 129b 1002) Hip at 30° external rotation 78 83 96c 68a

3) Hip at 30° internal rotation 87 85 108d 85Maximum isokinetic knee extension

4) Full range, 16°–30° 88 90 107 790°–15° 73 77 91 66

5) Terminal 30° arc, 16°–30° 64 77 96 680°–15° 82 85 99 68

Sidelying6) Ipsilateral, 16°–30° 21 34 30 24

0°–15° 24 34 26 217) Contralateral, 16°–30° 22 32 19 22

0°–15° 28 36 26 23From full squat

8) Stand, 16°–30° 17 32e 22 15e

0°–15° 13 20 14 119) Jump, 16°–30° 43 54 61 48

0°–15° 93 57 61 60a Bold print indicates the highest response of the muscles for each exercise in which there was statistically significant differences

between muscles.b P 5 0.022.c P 5 0.021.d P 5 0.015.e P 5 0.026.

52 Mirzabeigi et al. American Journal of Sports Medicine

dialis oblique muscle’s advantageous angle of pull on thepatella compared with that of the vastus lateralis muscle.

Grabiner et al.4 used fatigue rates of the vasti musclesto evaluate a series of static and dynamic quadricepsstrengthening exercises to determine whether either thevastus medialis oblique or the vastus lateralis musclescould be selectively activated. These authors used changesin the median frequency of the EMG signal as their fa-tigue index. The vastus medialis oblique muscle and thevastus lateralis muscle were not found to differ signifi-cantly in their respective rates of fatigue, and thereforethe authors concluded that neither muscle could be selec-tively strengthened.

Also using muscle fatigue determined from EMG signalcharacteristics, Kaljumae et al.8 demonstrated that thefatigability of both the vastus medialis and the vastuslateralis muscles decreased after a 10-week training pro-gram on a bicycle ergometer. However, the decrease wasgreater in the vastus medialis than in the vastus lateralismuscle. These investigators implied that endurance may bea more important determinant of the biomechanical balancebetween vasti muscles than strength, because of the greatertraining effects seen in the vastus medialis muscle.

Doucette and Goble2 found that 84% of subjects withlateral patellar compression syndrome demonstrated im-proved patellar tracking after an extensive program thatincluded joint mobility exercises, stretching of the ili-otibial band and lateral retinaculum, and exercises de-signed to strengthen the vastus medialis oblique muscle.No evidence was reported to support the specificity of thevastus medialis oblique muscle exercises used. Significantincreases in iliotibial band flexibility were found in thepain-free group.

Reynolds et al.15 revealed that isometric contractionshowed low levels of EMG activity in both the vastusmedialis oblique and the vastus lateralis muscles duringthe last 30° of distal-segment stabilized extension, whensubluxation commonly occurs. They also suggested thatthe addition of adduction while performing knee extensionexercises might facilitate strengthening of the vastus medi-alis oblique muscle during the early stages of rehabilitation.Using indwelling wire electrodes to monitor muscle activity,Hanten and Schulthies5 found that the vastus medialisoblique muscle may be selectively activated by performingresisted hip adduction exercises. These authors found a 16%difference in vastus medialis oblique muscle activity with amean activation level that was 62% of that obtained duringmaximum isometric knee extension at 50°.

In our study, none of the nine sets of exercises specifi-cally isolated the vastus medialis oblique muscle. How-ever, some of the exercises were more effective than othersin challenging the quadriceps muscle as a whole. Theexercise that best challenged all heads of the quadricepsmuscle was isometric knee extension with the femur inneutral alignment. Neither internal nor external rotationproduced EMG activity in the vastus medialis obliquemuscle that was as great as that recorded in the neutralposition. During the isokinetic exercises there was a con-sistent trend for lower EMG activity in the vastus medi-alis oblique muscle than in the vastus lateralis muscle. In

our sample this appears to put the vastus medialis obliquemuscle at a disadvantage during isokinetic exercises.Tests 6 through 8 demonstrated that the vasti muscleshad a maximum manual maximal muscle testing value ofonly 36%. These tests did not challenge the vasti muscles.This clearly shows that these tests are not specific enoughto challenge the vastus medialis oblique muscle or thevastus lateralis muscle. Insall7 states “. . . it is by nomeans certain that the commonly prescribed exercises doselectively strengthen this [the vastus medialis oblique]muscle and, in fact, exercises are perhaps least effectivewhen symptoms are secondary to malalignment.”

CONCLUSION

Our study suggests that isometric exercises in neutral andexternal rotation will challenge the vastus medialisoblique and the vastus lateralis muscles. However, fornone of the exercises in this study was the activity in thevastus medialis oblique muscle or the vastus medialislongus muscle significantly greater than the activity of thevastus lateralis muscle.

Despite previous beliefs that terminal knee extensionsincrease EMG activity in the vastus medialis obliquemuscle,16 our findings suggest greater EMG activity forthe vastus lateralis muscle. Further investigation maydevelop other sets of exercises that isolate the vastusmedialis oblique muscle.

REFERENCES

1. Bohannon RW: Effect of electrical stimulation to the vastus medialismuscle in a patient with chronically dislocating patellae. Phys Ther 63:1445–1447, 1983

2. Doucette SA, Goble M: The effect of exercise on patellar tracking in lateralpatellar compression syndrome. Am J Sports Med 20: 434–440, 1992

3. Fox TA: Dysplasia of the quadriceps mechanism: Hypoplasia of the vastusmedialis muscle as related to the hypermobile patella syndrome. Surg ClinNorth Am 55: 199–226, 1975

4. Grabiner MD, Koh TJ, Miller GF: Fatigue rates of vastus medialis obliqueand vastus lateralis during static and dynamic knee extension. J OrthopRes 9: 391–397, 1991

5. Hanten WP, Schulthies SS: Exercise effect on electromyographic activityof the vastus medialis oblique and vastus lateralis muscles. Phys Ther 70:561–565, 1990

6. Henry JH, Crosland JW: Conservative treatment of patellofemoral sublux-ation. Am J Sports Med 7: 12–14, 1979

7. Insall J: Patellar pain. [Current concepts review.] J Bone Joint Surg 64A:147–152, 1982

8. Kaljumae U, Hanninen O, Airaksinen O: Knee extensor fatigability andstrength after bicycle ergometer training. Arch Phys Med Rehabil 75:564–567, 1994

9. Kettelkamp DB, DeRosa GP: Biomechanics and functional role of thepatellofemoral joint. Instr Course Lect 25: 27–31, 1976

10. Knight KL, Martin JA, Londeree BR: EMG comparison of quadricepsfemoris activity during knee extension and straight leg raises. Am J PhysMed 58: 57–67, 1979

11. Leach RE: Malalignment syndromes of the patella. Instr Course Lect 25:49–54, 1976

12. LeVeau B, Rogers C: Selective training of the vastus medialis muscleusing EMG biofeedback. Phys Ther 60: 1410–1415, 1980

13. Lieb FJ, Perry J: Quadriceps function. An anatomical and mechanicalstudy using amputated limbs. J Bone Joint Surg 50A: 1535–1548, 1968

14. Mariani PP, Caruso I: An electromyographic investigation of subluxation ofthe patella. J Bone Joint Surg 61B: 169–171, 1979

15. Reynolds L, Levin TA, Medeiros JM, et al: EMG activity of the vastusmedialis oblique and the vastus lateralis in their role in patellar alignment.Am J Phys Med 62: 61–70, 1983

16. Smillie IS: Injuries of the Knee Joint. Third edition. Baltimore, Williams &Wilkins Co., 1962, pp 3–5

17. Wild JJ Jr, Franklin TD, Woods GW: Patellar pain and quadriceps reha-bilitation. An EMG study. Am J Sports Med 10: 12–15, 1982

Vol. 27, No. 1, 1999 Isolated Exercise of the Vastus Medialis Oblique Muscle 53