Reliability of Measurements Obtained by Use of an Instrument Designed to Indirectly Measure Iliotibial Band Length

Transcription

1 Reliability of Measurements Obtained by Use of an Instrument Designed to Indirectly Measure Iliotibial Band Length William E. Melchione, PT' M. Scott Sullivan, MS, PT2 M usculoskeletal examination of the lower extremity often involves assessment of iliotibial band (I'TB) length (4,6,7,lO, 12.20). Ober's test is a commonly used method to indirectly assess ITB length (4,6,7, lo). The test was originally designed for use on patients with sciatica (13). Today, Ober's test is commonly used in the examination of patients with anterior knee pain and ITB Friction Syndrome (6,7,11,12,14,18). McConnell states that a tight ITB may contribute to the develop ment of anterior knee pain by pulling the patella laterally during knee flexion (9); she recommends assessment of ITB length through the use of Ober's test (10). Others have implicated a shortened ITB in conditions of subluxing patella (3) and ITB Friction Syndrome in runners (7,11, ). These authors also advocate the use of Ober's test in their assessments. The same authors (7,10,11,12,14,18) indicate that patients with anterior knee pain and ITB Friction Syndrome complain of knee pain during activities of daily living. Presumably, these authors utilize this test because they assume it indicates whether the ITB is tight during functional activities. There are no data to support or refute this assumption. In performing Ober's test, the Ober's test is widely used to indirectly assess iliotibial band length on patients with painful conditions of the lower extremity. The validity of Ober's test is questioned on several accounts: there is no method for standardizing the position of the pelvis during the measurement; the scale used to describe the results is nominal and appears to have been arbitrarily determined; and, finally, the reliability of judgments made using Ober's test has not been reported. The purposes of this study were to develop a method for quantifying an indirect measurement of iliotibial band length and to examine the intratester and intertester reliability of measurements obtained. lliotibial band lengths of 10 patients (N = 10) with anterior knee pain were measured twice by each of two examiners. Data were analyzed using the intraclass correlation coeficient (KC) and standard error of the measurement (SEM). The ICC values were 0.94 and 0.73 for intratester and intertester reliability, respectively. The SEM values were I" and 2" for intratester and intertester reliability, respectively. Measurements obtained with the modified method are reliable when taken on young patients with anterior knee pain. Key Words: motion measurement, iliotibial band, knee pain ' Supervisor of Sports Medicine, Department of Physical Therapy, Chippenham Medical Center, 681 Hioaks Road, Richmond, VA ' Assistant Professor, Department of Physical Therapy, School of Allied Health Professions, Medical College of Virginia, Virginia Commonwealth University, Richmond, VA examiner positions the subject on his or her side, with the side to be tested facing upward. The examiner flexes the contralateral hip and knee just enough to obliterate the lumbar lordosis, then flexes the ipsilateral knee to 90" and abducts and extends the ipsilateral hip so that it is in line with the trunk. At this point, the examiner allows the force of gravity to fully adduct the extremity. According to Ober, the hip will remain "more or less" abducted if the ITB is shortened (1 3). Ober's description of the test is somewhat unclear. For example, he states that during the procedure the examiner steadies the pelvis but does not describe how this is done (1 3). Additionally, he does not describe a method of quantifying the results of the test. These factors may lead to variations in the test procedure and inconsistencies in interpretation of the results. Furthermore, his conclusion that hip abduction indicates an abnormal result is unsubstantiated by data and appears to have been arbitrarily chosen. Finally, no studies have examined the reliability of judgments made by clinicians concerning ITB tightness through use of the test. Wadsworth et al have described a "modified Ober's test" to assess ITB length quantitatively (1 9). Pelvic stabilization is ensured in this test by supporting the subject's pelvis

2 RESEARCH STUDY and aligning it with a lined, plexiglass backrest affixed to a table. Measurements of hip adduction are determined by use of a gravity goniometer. Repeated, indirect measures of ITB length, taken by the same examiner and performed in various positions of knee flexion, demonstrated intraclass correlation coefficients (ICC) ranging from 0.84 to 0.93 when taken on healthy subjects. The results of Wadsworth et al's study are not able to be generalized when considering measurement of ITB length in a patient population. In addition to the problems of poor standardization and lack of quantitative results, Ober's test utilizes a lower extremity position which rarely occurs during functional activities. There is no evidence supporting the notion that tension on the ITB in the position described by Ober is directly related to tension on the ITB in functional positions. During Ober's test, in which the knee is flexed to 90, the patella is positioned distal to the anatomical position. According to a review of the literature by Brattsrom, at 90" of knee flexion, most of the compressive force at the patellofemoral joint is concentrated at the lateral facets where they contact the lateral surface of the intercondylar groove of the femur (2). In this position, the lateral femoral condyle mechanically blocks lateral glide of the patella. Brattstrom indicates that the patella is situated at the opening of the intercondylar groove in the beginning range of flexion (2). This would explain why more lateral glide can be elicited during physical examination when the knee is extended than when it is flexed (2). Since the ITB is attached to the patella (8). it is probably under greater tension during Ober's test than it would be if tested in positions that allowed for some lateral glide of the patella. A modified test utilizing a slightly flexed knee position, which occurs during the gait cycle, would not increase tension on the ITB be- yond what would normally occur as a result of the lower extremity position during gait. We believe that this modification of Ober's test is a more valid indication of the effect of ITB length on hip adduction during functional activities than Ober's test. The purposes of this study were to modify the lower extremity position during Ober's test to mimic a lower extremity position that occurs during gait, to develop a method for quantifying results of the modified test, and to examine the reliability of measurements obtained with the new test. METHOD Subjects Ten consecutive patients (five males and five females) between the Most of the compressive force at the patellofemoral joint is concentrated at the lateral facets. ages of 16 and 43 years (x age = 23.8 yrs) who were experiencing anterior knee pain and were being treated by one of the investigators (WEM) were solicited to volunteer for this study. Anterior knee pain may be caused by a variety of pathologies, though the symptoms may be present prior to any evidence of pathology (3,6,9,20). Therefore, no attempt was made to classify these patients by pathology per se. None of the patients had a history of direct trauma or surgery to the lower extremity, and none were diagnosed with a meniscus or knee ligament injury. All subjects signed a consent form approved by the Committee on Research Involving Human Subjects at the State University of New York at Stony Brook. Examiners Two physical therapists with 3-4 years of experience primarily in an outpatient orthopaedic setting participated as examiners in this study. Fach was instructed in the use of the instrument developed for this study, and each underwent at least 15 practice trials measuring healthy subjects. Instrumentation A specially designed pelvic level was fabricated for this study to aid the examiner in controlling the position of the pelvis during the test. This level consisted of a 2 by 8-in transparent plastic ruler and two spirit levels. A spirit level is a small level used to determine grade, when placed on a string, in construction. One spirit level was affixed parallel to the long edge of the ruler. A second spirit level was affixed to the ruler perpendicular to the first level. When the pelvic level is placed on a subject, one level lies in the sagittal and transverse planes, and one lies in the sagittal and coronal planes. The pelvic level was positioned on the patient with a Velcro'" strap so that it was aligned with the posterior superior iliac spines (PSISs) (Figure 1). A fluid-filled inclinometer and two 12-in universal goniometers were utilized in this study. Each instrument was designed with markings at 1 " increments. One of the two universal goniometers was chosen to be used only on the knee joint. A groove was cut out on one arm so that the inclinometer could be placed directly on the distal lateral thigh. The other goniometer was used to determine hip position. Procedure The modified Ober's test utilized the hip and knee positions that Volume 18 Number 3 September 1993 JOSPT

3 7, (which was aligned with the thigh and the midaxillary line). while the examiner placed the hip in 5" of extension. The examiners attempted to i manually control for hip rotation throughout the maneuver; however, no measure of hip rotation was obtained. The assistant removed the I goniometer to allow the examiner to complete the test maneuver, then zeroed the gravity goniometer on a known, level surface. The subject " was instructed to relax his or her involved lower extremity. The examiner then placed the knee in 5" of flexion, fully abducted the lower extremity, then allowed the force of gravity to adduct the extremity until the hip could not adduct any further. The assistant placed the gravity FIGURE 1. The pelvic level in the testing position. Black arrows indicate the spirit levels. One level is in the goniometer on the distal lateral sagittal and coronal planes, and the other is in the sagittal and transverse planes. thigh and read and recorded the an- occur during the push-off phase of the gait cycle (ie., 5" of knee flexion and 5" of hip extension) (5). This position was chosen for two reasons: 1) it occurs during a common functional activity, and 2) in this position, the ITB is likely to be under some degree of tension when the hip is adducted. The latter assumption is based on the anatomy of the ITB, with its attachments to the tensor fascia lata and gluteus maximus proximally and the lateral patella and Gerdy's tubercle distally (8). The testing procedure was performed twice by two examiners, each serving as an assistant to the other. The subjects' PSISs were palpated and marked with a grease pencil. The pelvic level was then strapped to the subject's pelvis and aligned with the marks over the PSISs. The universal goniometer with the cut-out was strapped to the lateral aspect of the involved lower extremity with Velcro straps, the arms being aligned with the midshafts of the thigh and leg. The subjects were positioned on their uninvolved side with the uninvolved hip, knee, and lumbosacral spine flexed to the point at which the lumbar lordosis appeared flattened. The pelvis was positioned so that the coronal and sagittal planes were aligned with a true horizontal. An assistant then measured sagittal plane hip position of the involved lower extremity with a goniometer gle of hip adduction (Figure 2). The gravity goniometer was facing away from the examiner so that he was blinded to the results of the measurement. The examiner and the assistant then switched roles. Each examiner performed the maneuver two times in an alternating order. The pelvic level was not removed and reapplied FIGURE 2. The inclinometer was posrtioned on the distal lateral thigh by the assistant while the examiner supported the position of the lower extremity.

4 RESEARCH STUDY each time measurements were repeated; however, its position was checked and adjusted prior to each measurement. Each examiner, when serving as assistant, privately recorded his measure on a separate data sheet. Data Analysis The intraclass correlation coefficient (ICC) (2,l) was used because it produces a coefficient of agreement while accounting for random effects of examiners (1 6). Standard error of the measure (SEM) was also calculated (1). RESULTS The ICC (2,l) for measurements taken by the same examiner was.94, with an SEM of 1 ". The ICC (2.1) for measurements obtained by different examiners was.73, with an SEM of 2" (Table). DISCUSSION Since the time Ober described his test for ITB length, there have been tremendous advances in medical technology and sophistication of diagnostic testing. Yet, in the area of physical examination, clinical signs that have a long tradition of use are slow to lose their popularity among clinicians. A positive Ober's test is one such clinical sign. Despite the fact that the procedure is poorly standardized and the definition of an abnormal finding a p pears to have been arbitrarily chosen, Ober's test is frequently used and recommended. The modifications proposed in this study allow the examiner to control the position of TABLE. Reliability of indirect measurements of iliotibial band length. the pelvis and to quantify the measurements obtained with ratio-scaled data (1 5). We believe that this ratio-scaled measurement is preferable to the nominal scale described by Ober for the following reasons: 1) it offers a more precise indirect assessment of ITB length; 2) it may be used to study the variability of the measurement; and 3) it has potential to be a more sensitive indicator of change in the measurement. Although the actual length of the ITB is difficult to measure directly, the measurement of hip adduction while the ITB is under tension allows for an indirect assessment of ITB length. Some degree of un Sol? tissue swelling due to inflammation or other factorsl such as pressure on the tissue due to an osteophyfe, may inhibit hip adduction. certainty regarding the validity of this test may always be present. Soft tissue swelling due to inflammation or other factors, such as pressure on the tissue due to an osteophyte, may inhibit hip adduction. In these circumstances, the amount of hip adduction measured could mistakenly be attributed to the length of the ITB. Nevertheless, we believe this clinical measurement represents the best available means of indirectly assessing ITB length. The clinician using this test must make inferences based on clusters of signs and symp toms, not from the results of this test alone. Indirect measurements of ITB length obtained with a fluid-filled inclinometer and the pelvic level de- scribed in this study demonstrate good to excellent reliability (1 7). Both intertester and intratester reliability were investigated. The intertester reliability indicates the objectivity of a measurement (1 5) and, therefore, makes the results easy to generalize to examiners with similar backgrounds to those participating in this study. The SEM describes the range in which a single subject's true score could be expected to lie when measurement error is considered (1). At a 95% confidence interval, the true score for the patient would lie within f 2 SEM (f 4") when measurement is obtained on similar patients by examiners with similar backgrounds to those participating in this study. A change of 4" or less could be attributed to measurement error. When measurements are taken on an individual by the same examiner, the SEM is 1 ". At a 95% confidence interval, the true score would be expected to lie within f 2". One limitation of this method of indirectly assessing the length of the ITB is the need for two individuals. Because the fluid-filled goniometer was not strapped to the lateral thigh of the subject, it must be placed on the thigh of the subject by a second examiner. Other limitations of this study include the small number of subjects and the use of only two examiners. These limitations notwithstanding, we believe that the measurement described in this paper is worthy of further study. There is a need for continued research in this area, as few studies have attempted to quantify measurements of ITB length. Measurement of large samples of healthy individuals is needed so that normative values of ITB length may be established. This will aid the clinician in determining whether an individual has a shortened ITB. CONCLUSION The authors have designed a new method to indirectly quantify Volume I8 Number 3 September 1993 JOSPT

5 the length of the ITB using a modified version of Ober's test. A pelvic level, a fluid-filled goniometer, and two universal goniometers were used to maintain subject position and to record the measurement. Repeated measurements obtained by the described method demonstrated good reliability between testers and excellent reliability within testers when taken on patients with anterior knee pain. JOSPT ACKNOWLEDGMENTS The authors wish to thank Barbara Bitzer, PT, for her assistance with this project. REFERENCES I. Anastasi A: Psychological Testing, New York: Macmillan Publishing Company, Brattstrom H:Anatomy andphysiology of the patello-femoral joint. Acta Orthop Scand 685: 14-24, Fulkerson lp, Hungerford DS: Disorders of the Patellofemoral loint, pp 91. Baltimore: Williams & Wilkins, Crady IF, O'Conner K1, Bender I: Iliotibial band syndrome. 1 Am Podiatr Med Assoc 76: , lnman VT, Ralston HI, Todd F: Human Walking, Baltimore: Williams & Wilkins, lrrgang I: Associated patholgies. In: Mangine RE fed), Physical Therapy of the Knee, pp New York: Churchill Livingstone, /ones DC, lames SL: Overuse injuries of the lower extremity: Shin splints, iliotibial band friction syndromes, and exertional compartment syndromes. Clin Sports Med 6: , Kaplan EB: The iliotibial tract: Clinical and morphological significance. I Bone loint Surg 4OA: , McConnell 1: The management of chondromalacia patellae: A long term solution. Aust I Physiother 32: , McConnell I: McConnell Patellofemoral Treatment Plan. Course Notes. Andover, MA, I. Noble CA: lliotibial band friction syndrome in runners. Am I Sports Med 8: , Noble HB, Hajek MR, Porter M: Diagnosis and treatment of iliotibial band tightness in runners. Phys Sportsmed 10:67-74, Ober FR: The role of the iliotibial band and fascia lata as a factor in the causation of low back disabilities and sciatica. I Bone loint Surg 18: , Renne lw: The iliotibial friction syndrome. I Bone joint Surg 57A: , Rothstein /M: Measurement in Physical Therapy, New York: Churchill Livingstone, Shrout PE, Fleiss jl: lntraclass correlations: Uses in assessing rater reliability. Psycho1 Bull 86: , Fleiss 11: The Design and Analysis of Clinical Experiments, pp New York: lohn Wiley & Sons, Inc., Sutker AN, lackson DW, Pagliano IW: lliotibial band syndrome in distance runners. Phys Sportsmed 9:69-73, Wadsworth CT, Altman BA, Holscher ID, Ragsdale RL, Nielsen DH: Influence of knee flexion angles on iliotibial band tension measured indirectly with a gravity goniometer. Presented at the Combined Sections Meeting of The American Physical Therapy Association, Orlando, FL, lanuary 3 1 -February 3, Wallace LA, Mangine RE, Malone T: The knee. In: Could la, Davies CI (eds): Orthopaedic and Sports Physical Therapy, pp St. Louis: C.V. Mosby Company, 1985