Effects of Ankle Dorsiflexion on Active and Passive Unilateral Straight Leg Raising

RICHARD L. GAJDOSIK, BARNEY F. L E V E A U , and RICHARD W. BOHANNON

The purpose of this study was to analyze the straight-leg-raising (SLR) maneuver while the ankle was fixed in dorsiflexion or relaxed in plantar flexion. Twenty-two healthy subjects underwent active and passive SLR with the ankle in each position. We used cinematography to document movement of the right lower limb and pelvis and electromyography to document hamstring muscle activity. Anal­yses of variance of the angles of maximum SLR and change in the pelvic position showed a significant F ratio (p = .01) among the active and passive trials. Post hoc analyses demonstrated significant differences (p = .01) between SLR with dorsiflexion and SLR with plantar flexion. The EMG activity among trials was not significantly different. The possible causes of the effects of dorsiflexion on SLR are discussed. We encourage clinicians to document and compare SLR with dorsiflexion and SLR with plantar flexion, and we recommend additional research to examine the relative influence of tissue structures on SLR.

Key Words: Exercise test, Leg, Muscles, Neurologic examination, Pelvis.

The unilateral straight-leg-raising (SLR) test is widely reported in the lit­erature as an indirect test for measuring hamstring muscle tightness and as an aid in the diagnosis of sciatica and nerve root irritation.1,2 Although interest in the SLR test has been widespread, a review of the literature revealed no stud­ies that examined the effects of ankle positions on SLR. Such studies are needed so that clinicians can have an objective understanding of the variables that may influence test results.

Traditionally, when the SLR test is used to measure hamstring muscle tight­ness, the ankle is relaxed in plantar flex­

ion (PF). No documentation exists, however, on the differences in SLR that may result if the ankle is held in dorsi­flexion (DF). If the position of the ankle influences measuring hamstring muscle tightness, as determined by SLR, then a standard ankle position should be re­quired for hamstring muscle tightness testing.

When the SLR test is used in the diagnosis of sciatica or disk prolapse, passive DF of the ankle near the limit of pain-free SLR is used as a qualifying test because ankle DF puts tension on the sciatic nerve and its roots.3-5 Studies of the movement of the sciatic nerve and its roots on SLR in cadavers have documented that movement of the nerve and roots diminishes progressively after 70 degrees of SLR, but tension generated along the nerve increases.6, 7

The possibility that ankle DF could limit SLR seems reasonable in light of these reports.

The purposes of our study were two­fold: 1) to analyze the effect of ankle DF on the angle of SLR in relation to the horizontal plane (SLR-horizontal), in relation to the pelvis (SLR-pelvis), and on the change in position of the pelvis in relation to the horizontal plane (pel­vis-horizontal) and 2) to examine the electromyographic activity of the ham­string muscles during both active and

passive SLR. We expected to find a sig­nificant difference in the amount of SLR with the ankle fixed in DF in com­parison with the amount of SLR with the ankle relaxed in PF. We expected no difference in the EMG activity of the hamstring muscles. These expectations stemmed from our unpublished obser­vations that ankle DF limits both active and passive SLR in normal subjects and that the loss of motion is independent of the EMG activity of the hamstring muscles.

METHOD

Subjects

Twenty-two healthy adults, 12 women and 10 men, with a mean and standard deviation for age, weight, and height of 28.0 ± 4.8 yr, 66.0 ± 9.9 kg, and 173.4 ± 8.0 cm, respectively, vol­unteered to participate in this study. Each subject reviewed and signed an informed consent form for the study, which was approved by the Committee on the Protection of the Rights of Hu­man Subjects of the University of North Carolina at Chapel Hill. Subjects had Normal muscle strength and range of motion of the back and lower extremi­ties and were not obese. They had no history of orthopedic or neurologic dis­orders.

Mr. Gajdosik is Associate Professor, Physical Therapy Program, University of Montana, Mis­soula, MT, and currently a doctoral candidate, De­partment of Anatomy, School of Medicine, Univer­sity of North Carolina at Chapel Hill. Direct all correspondence to 612 Hibbard Dr, Chapel Hill, NC 27514 (USA).

Dr. LeVeau is Professor and Chairman, Depart­ment of Physical Therapy, School of Allied Health Sciences, University of Texas, Health Sciences Cen­ter at Dallas, TX 75235.

Mr. Bohannon is Chief of Physical Therapy, Southeastern Regional Rehabilitation Center, Cape Fear Valley Medical Center, PO Box 2000, Fayette-ville, NC 28302.

This study was presented as a Research Platform Presentation at the Sixty-First Annual Conference of the American Physical Therapy Association, New Orleans, LA, June 19, 1985.

This article was submitted November 7, 1984; was with the authors for revision four weeks; and was accepted April 30, 1985.

1478 PHYSICAL THERAPY

RESEARCH

Instrumentation

We used a motor-driven 16-mm Bolex motion picture camera* with a 25-mm lens to film the position of the lower limb during SLR. The camera was mounted on a tripod at a height of 129.5 cm (51 in) and positioned 4.6 m (15 ft) from and perpendicular to the sagittal plane of the subject, who lay on a pad­ded table 79 cm (31 in) high. A plumb bob was placed within the filming field to provide a reference for angle meas­urements. Filming was performed at 24 frames/sec, and the developed film was analyzed with a Vanguard motion ana­lyzer.†

The raw EMG signal was amplified and processed by equipment designed and constructed by the Biomedical En­gineering Department, School of Medi­cine, University of North Carolina at Chapel Hill. The frequency response of the amplifier was 1 Hz to 1 kHz. The common mode rejection ratio was 90 dB, and the differential input imped­ance was 3 kMΩ. The raw EMG signal and one-second time-reset integrated EMG (IEMG) signal were recorded on a Honeywell Visicorder‡ light-sensitive oscillograph. A signal from a footswitch used to document the onset of SLR was also recorded on the Visicorder.

Before testing, the subject was posi­tioned prone, and the belly of the long head of the right biceps femoris muscle was identified by palpation during an isometric contraction at 90 degrees of knee flexion. The skin over the muscle belly was abraded according to Shackel's method, which reduced skin resistance below 20 kΩ.8 Then, 16-mm silver-silver chloride surface electrodes were affixed longitudinally 3 cm apart over the mus­cle belly; placement of the electrodes was confirmed with active isotonic con­tractions of hip extension and knee flex­ion.

We used an active isometric muscle contraction as the standard for comput­ing the percentage of IEMG activity of the long head of the biceps femoris mus­cle during the SLR trials. The subject in a standing position flexed the right knee to 90 degrees and extended the right hip so that the right thigh was aligned with

the left thigh. The subject held the po­sition for five seconds. No resistance was added to the weight of the limb. The IEMG activity over the second, third, and fourth seconds was averaged to de­termine the IEMG per second. The IEMG activity of each trial was com­puted as a percentage of the standard contraction for each subject.

A 3-point splint (wt = .32 kg) held the knee in full extension, and a rigid ankle-foot orthosis (wt = .33 kg) molded to 10 degrees of DF fixed the ankle in DF (Figure).

Procedure

After the EMG electrodes were in place, we marked the subject's skin to measure SLR-horizontal, SLR-pelvis, and the pelvis-horizontal angles. The marking procedure has been described in detail elsewhere.9 Marks included a line between the right anterior and pos­terior superior iliac spines, a point just distal to the greater trochanter, and a point over the lateral malleolus. We also marked the long axis of the fifth meta­tarsal to record the position of the ankle and the lateral aspect of the knee to document any change in knee extension (Figure).

The subject was then positioned su­pine on an examining table and the 3-point splint was secured to the knee. With the splint in place, the subject performed five active SLR practice trials (one trial per minute); the rhythm of a metronome aided the subject to com­

plete each trial in about three seconds. The practice trials also provided a warm-up period to decrease the poten­tial for increases in the angle of SLR that might result from measuring re­peated trials from a cold start.10

After each subject completed the practice trials, we secured the left thigh to the table with a cloth strap and put the switch under the right heel to be activated when SLR began. We col­lected data on four trials: 1) active SLR with the ankle relaxed in PF, 2) active SLR with the ankle fixed in DF, 3) passive SLR with the ankle relaxed in PF, and 4) passive SLR with the ankle fixed in DF. For the active trials, we told the subject to complete each trial at the speed learned in the practice trials and to stop when "firm resistance" was felt. For the passive trials, one investigator raised the limb at a similar speed until he felt "firm resistance," and the subject expressed that full SLR had been reached. The four trials were random­ized for each subject to lessen the poten­tial for systematic influence of repeated trials.

We obtained measurements of the an­gles of SLR-horizontal and pelvis-hori­zontal and of the angles of the knee and ankle (in relation to the lower limb) from the film by measuring before limb motion began and again after full SLR had been reached. We determined in-tratester reliabilities of measuring the joint angles at the knee and ankle by measuring two positions each from a

Figure. Subject undergoing active straight leg raising. Note 3-point splint holding knee in extension, ankle-foot orthosis holding ankle in dorsiflexion, skin markings, surface electrodes, and footswitch.

*Model HR 16, Bolex International SA, Yver-don, Switzerland.

† Vanguard Instrument Corp, Walt Whitman Rd, Melville, Long Island, NY 11746.

‡ Honeywell Test Instrument Div, Denver, CO 80217.

Volume 65 / Number 10, October 1985 1479

TABLE 1 Mean, Standard Deviation, Range, and Mean Difference for Straight-Leg-Raising Anglesa

and Change in Position of Pelvis (N = 22)

Trials SLR-horizontal

Active PF Active DF Passive PF Passive DF

SLR-pelvis Active PF Active DF Passive PF Passive DF

Change in pelvis-horizontal Active PF Active DF Passive PF Passive DF

81.2 72.1 85.3 75.3

49.8 42.7 51.9 45.2

31.4 29.4 33.4 30.1

s

7.6 9.2 9.5

10.1

8.2 9.5

10.4 10.7

5.2 5.5 3.9 4.5

Range

67.9-98.7 55.9-85.6

63.2-100.0 55.4-96.1

34.1-65.2 19.6-57.8 26.4-68.4 22.2-65.9

24.8-43.0 22.7-43.3 27.4-43.8 22.9-41.1

(±s)b

9.1 (±7.5)

10.1(±5.1)

7.1 (±4.2)

6.7(±4.3)

2.0(±5.8)

3.3(±4.1)

TABLE 2 Mean, Standard Deviation, and Range of Electromyography Activity for Long Head of Biceps Femoris Musclea and Total Time of Straight-Leg-Raising Trials (N = 22)

Trials

Active PFb

Active DFC

Passive PF Passive DF

20.8 19.4 13.1 11.0

s

29.2 27.8 25.9 17.5

Range

0.0-126.6 0.0-109.0 0.0- 92.6] 00. 66.8 J

Total Time(±s)

3.0(±0.7) 2.8(±0.6) 4.3(±1.0) 3.9(±0.8)

random sample of 11 subjects (n = 22). Correlation coefficients (Pearson r) were found to be .99 and .99, respectively. We did not calculate correlation coeffi­cients for measuring SLR-horizontal and the pelvis-horizontal angles because they were previously reported high.9

We determined the angle of the SLR-pelvis by subtracting the change in the angle of the pelvis-horizontal from the angle of SLR-horizontal (SLR-pelvis = SLR-horizontal-pelvis-horizontal). The change in the position of the pelvis-horizontal angle was calculated as the difference in the measurements of the angle of the pelvis at the starting posi­tion and at the end of SLR.

The time to complete each trial was calculated by determining the number of film frames from the time the heel left the table to the time the greatest SLR angle was reached and dividing by the film rate. This time was also meas­ured on the Visicorder recording from the footswitch marks. From these meas­

urements, we then determined the total IEMG activity during each trial. After the baseline noise reading was sub­tracted from the IEMG signals, we cal­culated the average IEMG per second. This value was then compared with the average IEMG per second for each sub­ject's standard active isometric contrac­tion.

Data Analysis

We computed descriptive data for the angles of SLR-horizontal, SLR-pelvis, the change in the angle of the pelvis-horizontal, and the percentage of IEMG activity for each trial. Changes in the angles of the knee and ankle were also computed.

Differences between trials for the an­gles and the IEMG data were treated with a one-way analysis of variance (ANOVA) for repeated measures. Dun­can new multiple range tests were used for all post hoc analyses.11

RESULTS

Table 1 shows descriptive data for the SLR angles, change in the position of the pelvis-horizontal angle, and the mean differences between SLR with an­kle DF and SLR with ankle PF. Table 2 shows the summary data of the average percentage of EMG activity of the biceps femoris muscle across trials.

The mean and standard deviation of the initial position of the ankle for all PF trials was 35.9° ± 6.7° (90° = 0°, neutral). Plantar flexion increased 2.4° ± 5.0° from the initial position to the end of SLR. For the SLR trials with DF, we attempted to secure the ankle at 10 degrees of DF, but 10 degrees of DF could not be maintained by most sub­jects. Consequently, the initial mean an­gle for DF was 4.3° ± 3.6°. Dorsiflexion decreased slightly (0.5° ± 2.5°) from the initial position to the end of SLR. The knee remained relatively fixed in exten­sion with a 1.5° ± 1.4° mean increase in flexion for all trials.

The ANOVA summary data for all SLR angles and EMG activity are shown in Tables 3 and 4, respectively. The ANOVA revealed significant F ratios (p = .01) for SLR-horizontal and SLR-pelvis angles and for the change in the pelvis-horizontal angle among trials. Percentage of EMG variation among trials was not significantly different.

Duncan post hoc analyses demonstra­ted significant differences between SLR with DF and SLR with PF for SLR-horizontal and SLR-pelvis angles (Tab. 5). No differences existed between the active and passive trials within the con­ditions of DF and PF. The change in the pelvis-horizontal angle showed that active SLR with PF was not significantly different from active or passive SLR with DF. Only passive SLR with PF was different from active and passive SLR with DF.

DISCUSSION

The results of this study clearly dem­onstrated that SLR was less with DF than with PF during both active and passive tests. Comparisons by post hoc analyses showed no significant differ­ences between active and passive SLR with PF or with DF. We believe that changes in the positions of the ankle and knee were not major factors in differ­ences between DF and PF trials because the changes were minimal and consist­ent across trials.

a Measured in degrees. b Mean difference between straight leg raising (SLR) with ankle dorsiflexion (DF) and SLR

with ankle plantar flexion (PF).

a Data analyzed were the percentage of IEMG from comparison with an active voluntary isometric contraction.

b Plantar flexion. c Dorsiflexion.

1480 PHYSICAL THERAPY

RESEARCH

In analyzing the EMG data, we chose to compare the average percentage of IEMG per second of the long head of the biceps femoris muscle with an active isometric contraction. We did this be­cause we expected minimal EMG activ­ity and because submaximal isometric contractions have been shown to be more reliable than maximal contrac­tions.12 The standard should be easy to replicate in future studies. Our finding that EMG activity of the biceps femoris muscle was minimal was consistent with the findings of Norton and Sahrmann13

that passive SLR elicits negligible EMG activity in the hamstring muscles and those of Moore and Hutton14 that the angle of SLR is not dependent on the level of EMG activity of the hamstring muscles. The low average percentage of EMG activity in our study, both with DF and with PF, supports the suggestion that some other factors, such as the rest­ing tension of muscles and other tissues, may influence SLR.2

What then, might have limited SLR with the ankle fixed in DF? One poten­tial factor may be that DF puts added tension on the sciatic nerve and related structures.4, 5 Such increased tension in the sciatic nerve and its roots has been reported at the end of SLR.6, 7 Dorsiflex-ion before SLR may take up slack in the sciatic nerve and manifest the increased tension by limiting the angle of SLR. In addition, structures not normally sug­gested as limiting SLR could affect SLR, such as the skin, subcutaneous connec­tive tissue, and the enveloping deep fas­cia of the posterior aspect of the entire lower limb. Markee et al have described variations in the hamstring muscles that could decrease SLR with DF, such as extra muscle fascicles extending from the semitendinosus muscle to the fascia on the back of the thigh and from the long head of the biceps femoris muscle to the sural fascia.15 Also, fascial con­nections between the gastrocnemius muscle and the hamstring muscles in the popliteal region may allow DF to increase the passive tension of the ham­strings, and this increased passive ten­sion could limit SLR. Although several explanations can be suggested, the spe­cific causes of the effects of DF on SLR are beyond the scope of this study. Ad­ditional anatomical studies of SLR with the ankle relaxed and with it held in DF are needed to examine the relative con­tributions of the various tissue structures to limiting SLR.

TABLE 3 Summary of Analysis of Variance for Straight-Leg-Raising Angles and Change in Pelvis Between Trials

Source of Variation

SLR-horizontal Subjects Trials Error Total

SLR-pelvis Subjects Trials Error Total

Change in pelvis-horizontal

Subjects Trials Error Total

df

21 3

63 87

21 3

63 87

21 3

63 87

SS

5200.13 2331.00 1768.05 9299.18

7243.43 1161.80 751.65

9156.88

1138.17 204.02 805.93

2148.12

MS

247.63 777.00 28.06

106.89

344.93 387.27

11.93 105.25

54.20 68.01 12.79 24.69

F

27.69

32.46

5.32

Critical F Value (.01)

4.13

4.13

4.13

TABLE 4 Summary of Analysis of Variance for Standardized Electromyographya of the Long Head of the Biceps Femoris Muscle During Straight Leg Raising

Source of

Variation

Subjects Trials Error Total

df

21 3

63 87

SS

37581.43 1578.14

16947.53 56107.10

MS

-789.59 526.05 269.01 644.91

F

1.96b

Critical F Value (.01)

4.13

TABLE 5 Summary of Duncan post hoc Analyses of Differences of Straight-Leg-Raising Angles and Change in Position of Pelvis Between Trials

Angles

SLR-horizontal Shortest significant ranges

Rank order of meansa

SLR-pelvis Shortest significant ranges

Rank order of meansa

Change in pelvis-horizontal Shortest significant ranges

Rank order of meansa

ADF 72.1

ADF 42.7

ADF 29.4

Trial

(2) 4.3

PDF 75.3

(2) 2.8

PDF 45.2

(2) 2.9

PDF 30.1

(3) 4.4

APF 81.2

(3) 2.9

APF 49.8

(3) 3.0

APF 31.4

(4) 4.6

PPF 85.2

(4) 3.0

PPF 51.9

(4) 3.1

PPF 33.4

a Data analyzed were the percentage of IEMG from comparison with an active voluntary isometric contraction.

bNS.

a Key: ADF = active straight leg raising (SLR) with dorsiflexion (DF), PDF = passive SLR with DF, APF = active SLR with relaxed plantar flexion (PF), PPF = passive SLR with relaxed PF. Note: Any means not underscored by the same line are significantly different (p = .01). Any means underscored by the same line are not significantly different.

Volume 65 / Number 10, October 1985 1481

Clinical Implications

Clinicians should be aware that ankle DF limits the angle of both active and passive SLR. Because the difference be­tween DF and relaxed PF was greatest for SLR-horizontal (about 10°), and measuring this angle is probably the most common method of measuring SLR, clinicians should standardize the testing procedure and document the po­sition of the ankle during the test. Meas­uring the angle of SLR with DF may provide an additional test for evaluating SLR, and by comparing the results to SLR with PF, additional information about the flexibility of SLR may be ob­tained.

In addition to measuring SLR-hori­zontal, clinicians may choose to meas­ure the angle of SLR-pelvis as recently suggested in another study by R.W.B.9

Because ankle DF limited this angle by about 7 degrees in our study, clinicians should consider the position of the ankle when using this technique as well.

CONCLUSION

The angle of SLR with the ankle fixed in DF was less than the angle of SLR with the ankle relaxed in PF for both active and passive SLR. The loss of mo­tion was apparently unrelated to the EMG activity of the long head of the biceps femoris muscle. Although in­creased passive tension of anatomical structures crossing the posterior aspect of the lower limb was implicated as caus­ing the loss of motion, the specific lim­iting structures were unclear. Additional studies are needed to examine the rela­tive contribution of the anatomical structures that may limit SLR. Clini­cians should be aware that ankle DF limits SLR. Moreover, clinicians should consider SLR with DF as an additional SLR test and compare the results of the test with those of SLR with PF.

Acknowledgments. We thank O.W. Hensen, Jr, PhD, Department of Anat­omy, University of North Carolina at

Chapel Hill, for his support of the study, and Alice Workman, Chief, Occupa­tional Therapy Department, Southeast­ern Regional Rehabilitation Center, Cape Fear Valley Medical Center, Fay-etteville, NC, for constructing the ankle-foot orthosis. Special thanks to the phys­ical therapists at the North Carolina Me­morial Hospital, Chapel Hill, NC, who volunteered as subjects.

REFERENCES 1. Gajdosik R, Lusin G: Hamstring muscle tight­

ness: Reliability of an active-knee-extension test. Phys Ther 63:1085-1088, 1983

2. Bohannon RW, Gajdosik R, LeVeau BF: Con­tribution of pelvic and lower limb motion to increases in the angle of passive straight leg raising. Phys Ther 65:474-476, 1985

3. Fajersztajn J, cited in Woodhall B, Hayes GJ: The well-leg-raising test of Fajersztajn in the diagnosis of ruptured intervertebral disc. J Bone Joint Surg [AM] 32:786-792, 1950

4. Breig A, Troup JDG: Biomechanics consider­ations in the straight-leg-raising test: Cadaveric and clinical studies of the effects of medial hip rotation. Spine 4:243-250, 1979

5. Troup JDG: Straight-leg-raising (SLR) and the qualifying tests for increased root tension: Their predictive value after back and sciatic pain. Spine 6:526-527, 1981

6. Chamley J: Orthopaedic signs in the diagnosis of disc protrusion with special reference to the straight-leg-raising test. Lancet 1:186-192, 1951

7. Goddard MD, Reid JD: Movements induced by straight-leg-raising in the lumbo-sacral roots, nerves and plexus, and in the intrapelvic sec­tion of the sciatic nerve. J Neurol Neurosurg Psychiatry 28:12-18, 1965

8. Shackel B: Skin drilling, a method for diminish­ing galvanic skin potentials. Am J Psychol 72:114-121,1959

9. Bohannon RW: Cinematographic analysis of the passive straight-leg-raising test for ham­string muscle length. Phys Ther 62:1269-1274, 1982

10. Atha J, Wheatley DW: The mobilising effects of repeated measurement on hip flexion. Br J Sports Med 10:22-25, 1976

11. Duncan DB: Multiple-range and multiple-F tests. Biometrics 11:1-42, 1955

12. Yang JF, Winter DA: Electromyography relia­bility in maximal and submaximal isometric contractions. Arch Phys Med Rehabil 64:417-420, 1983

13. Norton BJ, Sahrmann SA: The effect of stretch­ing procedures on EMG activity in the ham­string muscles. Abstract. Phys Ther 61:686, 1981

14. Moore MA, Hutton RS: Electromyographic in­vestigation of muscle stretching techniques. Med Sci Sports Exerc 12:322-329, 1980

15. Markee JE, Logue JT, Williams M, et al: Two-joint muscles of the thigh. J Bone Joint Surg [AM] 37:125-142, 1955

1482 PHYSICAL THERAPY