The Relationship between Quadriceps Angle and Anterior Knee Pain syndrome'

The Relationship between Quadriceps Angle and Anterior Knee Pain syndrome' Doug Caylor, SPT, ATC2 Ryan Fites, SPT2 Teddy W. Worrell, EdD, PT, ATC3 A nterior knee pain syndrome (AKPS) is a significant clinical problem for the patient and the clinician (6, 20, 27). Various authors have attributed this pain to intrinsic and extrinsic factors (20). One extrinsic factor is malalignment of the patellofemoral joint. A common tool used to assess malalignment is the quadriceps angle (Q-angle). The Q-angle is defined as the acute angle between the line connecting the anterior superior iliac spine (ASIS) and the midpoint of the patella (representing the line of pull of the combined quadriceps femoris musculature) and the line connecting the tibial tubercle with the same reference point on the patella (8) (Figure 1). Though many authors have used the Q-angle measurement, few give detailed descriptions of their methods (1, 4, 13, 16, 20, 24). The reliability of the Q-angle measurement is also poorly documented. Only Horton and Hall (8) reported Q-angle reliability values in a study of seven subjects. They recorded intrarater reliability of at least.92 (exact values were not reported), and interrater reliability of.87. Although many authors have undertaken studies in which the Q-angle was measured, there seems to be no universally accepted method. A method in which two lines are drawn manually on the skin, with one line extending from the midpoint of the Anterior knee pain syndrome (AKPS) represents a significant challenge for patients and for clinicians. The purposes of this study were: 1) to determine the reliability of the Q-angle measurement, 2) to quantify Q-angle changes that occur with knee flexion, and 3) to determine if subjects with AKPS (N = 52) have a significantly different Q-angle than subjects without AKPS (N = 50). With the knee in an extended position, intratester Q-angle intraclass correlation coefficients (ICC) ranged from.84 to.90, and standard error of measurement (SEM) values ranged from 2.01 to 2.23'. Intertester Q-angle ICC was 33, and the SEM was 2.49". With the knee flexed, the intratester ICC was.83 for both testers, and SEM values ranged from 0.68 to 2.45". Intertester ICC and SEM were.65 and 3.50, respectively. No significant difference was found in intratester Q-angle values between the extended and flexed knee positions (p > 0.05). No significant difference in Q-angle was found between asymptomatic subjects (1 1. 1 2 5.5") and symptomatic subjects (12.4 & 5.1") (p = 0.07). Increased Q-angles were not responsible for AKPS in this group of patients. Other factors were hypothesized to be responsible for their symptoms. Key Words: patellofemoral pain, chondromalacia, quadriceps angle ' Presented at the Sports Physical Therapy Section Team Concept Meeting, December 1992, Newport Beach, CA. ' Student, Krannert Graduate School of Physical Therapy, University of Indianapolis, Indianapolis, IN ' Assistant professor and director of research, Krannert Graduate School ol Physical Therapy, University ol Indianapolis, 1400 E. Hanna Ave., Indianapolis, IN 46227 patella to the ASIS and the other line extending from midpatella to the tibial tubercle with patient positioning unspecified, as described by several authors (16. 20). Aglietti et al (1) used the same landmarks and suggested that the measurement be made with the subject supine with the knees extended. Brown et al (4) further stipulated that the toes be directed vertically and the extensor mechanism be relaxed. Hvid and Andersen (I 3) suggested that the feet be placed together for the measurement. O'Donoghue (24) used a less specific method, replacing the line from the ASlS to midpatella as described earlier with a line drawn "along the general alignment of the thigh to the center of the patella." Horton and Hall (8) measured the Q-angle in the standing position with the knees in full extension. They used a universal goniometer to measure the Q-angle using the ASIS. midpoint of the patella, and tibial tubercle as landmarks, with a string stretched between the ASIS and the midpoint of the patella to ensure alignment of the proximal arm of the gonionieter. Several authors assert that the Q-angle decreases with knee flexion (7, 10, 1 1). Hehne (7) demonstrated bv vector analysis that with knee flexion, the Q-angle must be zero. Hungerford and Barry (1 1) explained that as the knee begins to JOSPT Volume 17 Number I January 1993

FIGURE 1. Quadriceps angle (Q-angle) landmarks. (From (Moss RI, DeVita P, Dawson MI: A biomechanical analysis of patellofemoral stress syndrome. I Athl Train 27(1):64-69, 1992, with permission). flex, the tibia internally rotates and the Q-angle decreases. The lateral movement of the tibial tubercle, due to the "screw home" mechanism, causes the Q-angle to increase during the last 30" of extension (1 1). No data were provided to support these assertions (7, 11). In addition, no research report was found that quantified the degree of change in the Q-angle that occurs with knee flexion. Moss et al (23) found significantly greater Q-angles in subjects with patellofemoral stress syndrome compared with asymptomatic high school athletes. Analysis of patellar forces in cadavers and in vivo has confirmed that high Q-angles increase compressive forces to a point that may result in pathological changes (21). Insall et al (1 6), in a prospective study, found that chondronialacia patellae was more common in patients with larger Q-angles. However, they did not report their method of assessing the Q-angle. Messier et al (20) reported that Q- angles were larger in a group with patellofemoral pain. Messier et al suggested that the Q-angle is a strong discriminator between noninjured runners and runners with patellofemoral pain. Huberti and Hayes (9). in a cadaver study, reported that increases in patellofemoral contact pressure occur with Q- angles both above and below normal values. Huberti and Hayes concluded that both high and low Q- angles should be considered potential causes of chondromalacia. In addition, Brown et a1 (4) reported that one indication for surgical patellar realignment for subluxation of the patella was an abnormally high Q- angle. Brown et al demonstrated that the correction of the Q-angle to 10" or less correlated with good to excellent results from the surgery. The reliability of the Q-angle measurements were not, however, reported in any of these studies (4, 9, 16, 20, 2 1, 23). A need exists for a reliable clinical tool to assess Q-angles. Therefore, the purposes of this study were: I) to determine the reliability of the Q-angle measurement, 2) to quantify Q-angle changes that occur with knee flexion, and 3) to determine if subjects with anterior knee pain have a significantly different Q-angle than subjects without anterior knee pain. The research hypotheses were: 1) the Q-angle can be measured accurately by two inexperienced testers, 2) the Q-angle will decrease with knee flexion, and 3) the Q-angle of patients with anterior knee pain will be greater than in subjects without anterior knee pain. METHODS Subjects Twenty subjects (9 male, 17 female) without a history of anterior knee pain or previous knee surgery volunteered to participate in the reliability study ( x age = 24.5 years, x weight = 146 Ib, 2 height = 68 in). In addition, these subjects also served as the asymptomatic group for comparison with a group with AKPS. Both knees of each subject were measured (N = 52). Fifty subjects ( I 8 male, 32 female) who were diagnosed by a physician as having AKPS and who were currently receiving physical therapy for this diagnosis, were recruited for the descriptive study (% age = 23 years, 2 weight = 150 Ib, x height = 67 in). A consent form that was approved by our university human subjects committee was signed by all subjects prior to participation. Q-Angle Measurement Procedure In the reliability study, Q-angle measurements were taken with the subjects in two standing positions with body weight distributed equally on both feet. In Position 1 (with the knee extended), the subjects were standing in a comfortable manner. In Position 2 (with the knee flexed), the subjects stood alternately with one heel on a 1 %-in heel lift (x knee flexion = 24.3"). The center of the patella and the tibial tubercle were located manually and marked with an ink pen. The ASIS was located, and the subject was instructed to hold one end of a string at the ASIS. The other end of the string was taped just inferior to the midpoint of the patella using a horizontally placed strip of athletic tape, such that the pen mark was still visible. The center of the goniometer was placed at the midpoint of the patella. One arm of the goniometer was aligned with the string leading to the ASIS, and the other arm of the goniometer was aligned with the tibial tubercle. The center of the goniometer was covered, preventing the researcher from seeing the reading. Two measurements were taken by the researcher and recorded by a second person. Both knees of each subject were measured twice by Tester 1 in Position 1 (extended) and Position 2 (flexed) (Trial 1). All ink marks were then removed using rubbing alcohol. Tester 2 then measured the same subject twice in both positions as de- Volume 17 Number I January 1993 JOSPT

scribed above. Then, both testers repeated this procedure (Trial 2) for a total of 16 measurements per subject. All measurements were taken in one testing session. In the descriptive study, Q-angles were measured as described in the reliability study for Position 1 only. Each subject was measured only once by one tester. Data were collected by testers at different clinical sites. ICC SEM (') p-value Q-angle (') (+So) Extended Position Intratester (#l).84 2.23 0.07 10.77 + 5.57 Intratester (#2).90 2.01 0.10 12.76 + 6.36 Intertester 33 2.49 0.00 12.43 + 6.05 Flexed Position Intratester (#l).83 0.68 0.43 9.30 f 1.67 Intratester (#2).83 2.45 0.04 10.43 + 5.93 Intertester.65 3.50 0.31 10.80 5.93 TABLE 1. Reliability of Q-angle measurements (N = 52). Data Analysis Reliability Study Intratester re- liability was determined by comparing the mean of the two Q-angle measurements of Trial 1 to the mean of the two Q-angle measurements of Trial 2 using a repeated measures ANOVA (N = 52). Intertester reliability was determined by comparing the first measurement for both extended and flexed positions of each tester during Trial 1 using a repeated measures ANOVA. Intraclass correlation (ICC 2.1). as described by Shrout and Fleiss (28), and the standard error of the measurement (SEM) were determined for both intratester and intertester reliability (3). A dependent t-test compared Q-angle values for Position 1 and Position 2. The probability level was set at p < 0.05. The mean and standard deviation were calculated for the knee flexion angle in Position 2. Descriptive Study An independ- ent t-test was used to determine if a significant difference existed between the Q-angle of patients with AKPS and subjects without AKPS. The probability level was set at p < 0.05. RESULTS Reliability Study For Position 1 (extended), intratester ICC and SEM values ranged from.84 to.90 and 2.01 to 2.'B0, respectively. Intertester ICC was.83 and the SEM was 2.49" for the extended position (Table 1). In Position 2 (flexed), the intratester ICC was.83 for both testers, and the SEM ranged from.68 to 2.45". The intertester ICC was.65, Several authors assert that the Q-angle decreases with knee flexion. and the SEM was 3.50" for the flexed position (Table 1). No significant difference was found in intratester Q-angle values between test Position 1 and test Position 2 for either tester (Table 2). The mean knee flexion angle was 24.3 + 8.2". Q-angle values (') (5D) p-valw Tester 1 Position 1 9.74 f 5.97 0.31 Position 2 11.17 + 5.45 Tester 2 Position 1 12.43 f 6.05 0.26 Position 2 10.43 f 5.93 TABLE 2. Mean Q-angle values and probability level lor Position I (extended) and Position 2 (flexed). Descriptive Study Q-angle values for both groups are reported in Table 3. Q-angle values represent interrater measurements. There was no significant difference in Q-angle values between the asymptomatic and symptomatic groups (p = 0.07). DISCUSSION Reliability Study The results of the reliability study demonstrated that two inexperienced testers, after 3 hours of training, were able to accurately reproduce Q-angle measurements (intratester) for the extended and flexed positions. The intertester reliability (.83) was acceptable for the extended position but not the flexed position (.65). Our reliability data are similar to the intrarater reliability (.92) and interrater reliability values (.87) reported by Hall and Horton (8). In the flexed position, intertester reliability was low and SEM was large (3.5 "). We believe that this occurred because of difficulty experienced in aligning the goniometer on the flexed knee. Variations in patellar position in the transverse and sagittal planes appeared to alter goniometer placement. Thus, more error was made when two different testers attempted to determine the Q-angle in the flexed position. JOSPT Volume 17 Number I January 1993

.. -. --. - - - - -. - Croup Q-angle (') Q-angle range (') pvalue Symptomatic ( N = 50) 12.4 f 5.1 2 to 24.07 Asymptomatic (N = 52) 11.1 + 5.5-6 to 24 TABLE 3. Mean Q-angle values, ranges, and probability level ior symptomatic and asymptomatic subjects A dependent t-test did not reveal a significant change in Q-angle for the extended or flexed position for either tester. We can only speculate that more knee flexion was necessary for a significant decrease in the Q- angle to occur. Inconsistencies were noted during Q-angle measurements with knee flexion, eg., one tester noted an increase in Q-angle while the other tester noted a decrease in Q-angle with knee flexion (Table 2). Even though acceptable ICC values were present for both testers, measurement error (SEM) is associated with the measurement. Because of the low intertester reliability of Q- angle measurement with the knee in the flexed position, we could not accurately determine if the Q-angle decreases with knee flexion in the descriptive study because both testers collected data at different clinical locations. Therefore, for the descrip tive study of subjects having AKPS, only the extended position data were used to assess the Q-angle. We modified Horton and Hall's (8) Q-angle assessment method to include taping of the string at the midpoint of the patella for convenience. We used a flexible 6-in plastic goniometer to increase conformity to the knees. The asymptomatic subjects had an average Q-angle of 1 1.1 2 5.5" (intertester value). Descriptive Study Results of this study revealed that subjects with AKPS did not have a significantly different Q-angle than asymptomatic subjects. The average Q-angle for the symptomatic subjects was 12.4 + 5.1 ", and the average Q-angle for the asymptomatic group was 1 1.1 + 5.5". The symptomatic Q-angle was 1.3" larger, which appears very small. However, this difference approaches significance at p = 0.07. Moss et al (23) reported that a static Q-angle difference of 1.go was statistically significant at p = 0.0 1. Moss et al reported that symptomatic subjects had an average Q-angle of 17.1 ", and asymptomatic subjects had an average Q-angle of 15.2". Our data are in conflict with many studies that state that an increased Q-angle is a predisposing factor to AKPS (4, 9, 15, 16, 20, 21). Direct comparisons of Q-angle values found in this study with other studies reporting normal and abnormal Q-angle values are not possible because of different methods of measuring Q-angle and lack of reported reliability coefficients. In addition, no universally accepted, normal Q-angle value exists. Horton and Hall (8) measured 50 men and 50 women and found an average Q- angle of 15.8 + 4.5" for women and 11.2 + 3.0" for men. Insall et al (16) reported an average Q-angle of 14" for 50 subjects, regardless of gender. Percy and Strother (27) considered 14 and 17 " normal for males and females, respectively. Paulos et al (26) considered a Q-angle above 15" to increase the tendency for lateral patellar malposition. In yet another study, Kolowich et al (1 7) reported that 10" is considered normal for all subjects. In addition, Hvid et al (1 4) reported Q-angle values above 20 and 1.5" for women and men, respectively, to be pathological. Aglietti et al (1) reported that asymptomatic subjects had an average Q-angle of 15". while subjects with chondromalacia had an average Q-angle of 20". The reliability of the Q-angle measurements in this study was not reported. Gender was equally distributed between the reliability study (65% female) and the descriptive study (64% female). This is in agreement with several studies that report AKPS is more common in females (12, 24, 25, 33). In females, the Q- angle is generally greater than in males because of greater pelvic width and greater degree of valgus alignment at the knee (6, 25). Hvid and Andersen (1 3) reported that in 29 patients with patellofemoral complaints, both Q-angle and degree of internal hip rotation were found to - --- - No significant difference in Q-angle exists between asymptomatic and symptomatic gmups. be higher in women than in men. Horton and Hall (8) theorized that because of their increased Q-angle, women may be at a greater disposition for developing patellofemoral problems. Yates and Grana (33) reported that 76% of the patients with knee pain in their study were female and concluded that patellofemoral problems are more common in young women. Many authors have postulated that factors other than high Q-angles may be the cause of AKPS (2, 5, 12, 18, 19, 22, 27, 29-32). These factors include excessive pronation (1 9, 27, 29, 30), iliotibial band tightness (27), lack of hamstring flexibility (2, 12, 22, 27, 31), vastus medialis weakness (2, 18, 22, 29, 32). femoral sulcus angle (5). and patellar tilt, rotation, and glide (19). We did not Volume 17 Number I January 1993-JOSPT

evaluate these other possible contributing factors to AKPS in our study. We can only hypothesize that factors other than high Q-angles were contributing to the symptoms of AKPS. In addition, to our knowledge, the relationship between Q-angles and abnormal patella position (rotation, tilt, and glide) has not been determined. Further research concerning the relationship between Q-angle and patella position is needed. Apparently, AKPS is a multiplefactor problem, at least in the present symptomatic population. Clinically, we have observed subjects with large Q-angles without AKPS, and this study supports the concept that AKPS can exist in a patient population without significantly higher Q- angles than an asymptomatic group. CLINICAL RELEVANCE When evaluating patients with AKPS, a complete lower extremity evaluation should be performed. This evaluation should assess strength (quadriceps, especially VMO control), flexibility (iliotibial band and hamstrings), foot position (pronation), and patellar position (Qangle, tilt, glide, and rotation). Patients with AKPS should be evaluated in terms of multiple predisposing factors, and, thus, a multiple-factor therapeutic intervention should be implemented based on the evaluative findings. CONCLUSION We have found 1) that the Q- angle can be measured reliably by two inexperienced measurers with the knee in an extended position, 2) that the Q-angle did not significantly change with 24.3' of knee flexion. and 3) that no significant difference in Q-angle exists between asymptomatic and synlptomatic groups. We hypothesize that in this group of pa- tients with AKPS, other etiological factors were responsible for their symptoms. Therefore, when evaluating patients with AKPS, a complete assessment of lower extremity alignment, flexibility, and strength should be performed. JOSPT - -- REFERENCES Aglietti P, lnsall IN, Cerulli C: Patellar pain and incongruence. I: Measurements of incongruence. Clin Orthop 176:217-224, 1983 Antich TI, Randall CC, Westbrook RA, Morrissey MC, Brewster CE: Evaluation of knee extensor mechanism disorders: Clinical presentation of 1 12 patients. I Orthop Sports Phys Ther 8:248-254, 1986 Baumgartner TA: Norm-referenced measurement: Reliability. In: Safrit and Woods (eds), Measurements Concepts in Physical Education and Exercise Science, pp 45-72. Champaign, IL: Human Kinetics, 1989 Brown DE, Alexander AH, Lichtman DM: The Elmslie-Trillat procedure: Evaluation in patellar dislocation and subluxation. Am I Sports Med 12:104-108, 1984 Fulkerson lp, Hungerford DS: Disorders of the Patellofemoral loint, Baltimore: Williams & Wilkins, 1990 Heckmann TP: Conservative versus postsurgical patellar rehabilitation. In: Mangine RE (ed), Physical Therapy of the Knee, pp 127-129. New York, NY: Churchill Livingstone, 1988 Hehne H: Biomechanics of the patellofemoral joint and its clinical relevance. Clin Orthop 258:73-85, 1990 Horton MC, Hall TL: Quadriceps femoris muscle angle: Normal values and relationships with gender and selected skeletal measures. Phys Ther 69:897-901, 1989 Huberti HH, Hayes WC: Patellofemoral contact pressures: The influence of Q-angle and tendofemoral contact. I Bone loint Surg 66A:715-724, 1984 Hughston IC, Walsh WM, Puddu C: Patellar Subluxation and Dislocation, Philadelphia: WB Saunders Co., 1984 Hungerford DS, Barry M: Biomechanics of the patellofemoral joint. Clin Orthop 144:9-15, 1979 Hunter LY: Aspects of injuries to the lower extremity unique to the female athlete. In: Nicholas /A, Hershman EB (eds), The Lower Extremity and Spine in Sports Medicine, pp 90-1 1 1. St. Louis, MO: CV Mosby Co., 1986 Hvid I, Andersen LI: The quadriceps angle and its relation to femoral torsion. Acta Orthop Scand 53:577-579, 1982 Hvid I, Andersen LI, Schmidt H: Chondromalacia patellae: The relationship to abnormal patellofemoral joint mechanics. Acta Orthop Scand 5:661-666, 1981 lnsall I, Ballough P, Barstein A: Proximal "tube" realignment of the patella for chondromalacia patellae. Clin Orthop 144:63-69, 1979 lnsall I, Falvo KA, Wise DW: Chondromalacia patellae: A prospective study. 1 Bone loint Surg 58A: 1-8, 1976 Kolowich PA, Paulos LE, Rosenberg TD, Farnsworth S: Lateral release of the patella: Indications and contraindications. Am I Sports Med 18:359-365, 1990. Le Veau BF, Rogers C: Selective training of the vastus medialis muscle with EMG. Physiother 60: 14 10-14 15, 1980 McConnell I: The management of chondromalacia patellae: A long term solution. Aust / Physiol 32:2 15-223, 1986 Messier SP, Davis SE, Curl WW, Lowery RB, Pack RI: Etiologic factors associated with patellofemoral pain in runners. Med Sci Sports Exerc 23: 1008-1015, 1991 Minns RI, Birnie AIM, Abernethy PI: A stress analysis of the patella, and how it relates to patellar articular cartilage lesions. / Biornech 12:699-7ll, 1979 Moskwa CA, Nicholas /A: Musculoskeletal risk factors in the young athlete. Phys Sportsmed 17:49-59, 1989 Moss RI, DeVita P, Dawson ML: A biomechanical analysis of patellofemoral stress syndrome. I Athl Train 27:64-69, 1992 O'Donoghne DH: Treatment of acute dislocations of the patella. In: Funk FI /r (ed), Symposium on the Athlete's Knee. St. Louis, MO: C.V. Mosby Co., 1980 Outerbridge RE: Further studies on the etiology of chondromalacia patellae. I Bone loint Surg 468: 179-190, 1964 Paulos L, Rusche K, lohnson C, Noyes FR: Patellar malalignment: A treatment rationale. Phys Ther 60: 1624-1632, 1980 Percy EC, Strother RT: Patellalgia. Phys Sportsmed 13:43-59, 1985 Shrout PE, Fleiss TL: lntraclass correlations: Uses in assessing rater reliability. Psych01 Bull 86:420-428, 1979 Subotnic S: The foot and sports medicine. 1 Orthop Sports Phys Ther 2:53-54. 1980 JOSPT Volume I7 Number I Januar); 1993

30. Tiberio D: The effect of subtalar joint chanical power in jogging. I Biomech thop Sports Phys Ther 11:535-532, pronation on patellofemoral mechan- 16:91-97, 1983 1990 ics: A theoretical model. I Orthop 32. Woodall W, Welsh I: A biomechanical 33. Yates C, Crana W: Patellofemoral pain: Sports Phys Ther 9: 142-153, 1987 basis for rehabilitation programs in- A prospective study. Orthopedics 3 1. Winter D: Moments of force and me- volving the patellofemoral joint. I Or- 9:663-667, 1986 Volume 17 Number 1 January 1993 JOSPT