- Email: [email protected]
Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study
THEKNE-02571; No of Pages 7 The Knee xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
The Knee
Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study Himabindu Mikkilineni a,⁎, Patricia B. Delzell a, Jack Andrish b, Jennifer Bullen c, Nancy A. Obuchowski c, Naveen Subhas a, Joshua M. Polster a, Jean P. Schils a a b c
Cleveland Clinic, Imaging Institute, 9500 Euclid Avenue, Cleveland, OH 44195, USA Cleveland Clinic, Department of Orthopedic Surgery, 9500 Euclid Avenue, Cleveland, OH 44195, USA Cleveland Clinic, Quantitative Health Sciences, 9500 Euclid Avenue, Cleveland, OH 44195, USA
a r t i c l e
i n f o
Article history: Received 5 April 2017 Received in revised form 8 December 2017 Accepted 26 January 2018 Available online xxxx Keywords: Ultrasound Hoffa fat pad Impingement Infrapatellar fat pad
a b s t r a c t Background: We sought to determine whether there are ultrasound parameters that differ between knees with symptomatic fat pad synovial impingement and asymptomatic knees. Methods: A prospective study was performed in patients with clinical signs and symptoms of fat pad synovial impingement and asymptomatic controls. Eleven symptomatic knees and 10 asymptomatic controls were evaluated. Ultrasound imaging was performed before and after exercise. Evaluated parameters included largest fat lobule compressibility, subjective assessment of vascularity, largest vessel diameter, and subjective assessment of dynamic fat pad motion during flexion and extension. Receiver operating characteristic (ROC) curve analysis was used to assess whether changes in these parameters were different between symptomatic and asymptomatic knees. Results: Change in the largest vessel diameter was greater and trended toward dilation in asymptomatic knees compared to symptomatic knees (mean: 0.83 vs. −0.02; P b 0.001). No significant differences were observed between symptomatic and asymptomatic knees with respect to preexercise versus post-exercise states in subjective assessment of vascularity (P = 0.131), fat pad motion (P = 0.115), or percent change of the largest fat lobule (P = 0.241). However, overall compressibility of the fat pad lobule was significantly diminished in the pre-exercise state in symptomatic knees compared to asymptomatic controls. Conclusions: This study demonstrated a statistically significant change in the largest vessel diameter from pre- to post-exercise states between symptomatic and asymptomatic knees, as well as abnormal pre-exercise fat lobule compressibility in symptomatic knees. These findings show promise that with further research, ultrasound could have clinical utility in diagnosing infrapatellar fat pad impingement. © 2018 Elsevier B.V. All rights reserved.
1. Introduction Knee pain is the most common complaint of patients presenting to the general orthopedic surgeon's office, with pain most frequently reported in the anterior knee. Recently, the introduction of taping techniques aimed at improving patellofemoral alignment has renewed interest in the role of fat pad synovial impingement as an etiology for anterior knee pain [1]. The infrapatellar fat pad (or Hoffa's fat pad) has a rich vascular and nervous supply [2], and it is currently believed that inflammation of the infrapatellar fat pad can occur after acute injury or repetitive microtrauma, leading to anterior knee pain. The etiology of impingement may arise from abnormalities related to patellofemoral maltracking [3–6]. Current nonoperative treatment methods for ⁎ Corresponding author at: 25875 Science Park Drive, AC116, Beachwood, OH 44122, USA. E-mail address: [email protected] (H. Mikkilineni).
https://doi.org/10.1016/j.knee.2018.01.008 0968-0160/© 2018 Elsevier B.V. All rights reserved.
Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
2
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
infrapatellar fat pad impingement include nonsteroidal anti-inflammatory drug therapy, taping, physical therapy, and localized anesthetic or steroid injections. If nonoperative therapy fails, surgical options such as partial fat pad resection or debridement may be considered [5]. Symptomatic infrapatellar fat pad impingement is currently diagnosed based on physical examination and magnetic resonance imaging (MRI) findings. However, on physical examination, distinguishing between infrapatellar fat pad irritation and patellar tendon pathology can be difficult [4,5]. MRI findings such as superolateral fat pad edema or static measurements to identify dynamic patellofemoral maltracking abnormalities have been validated by a few studies [6–11]. However, because ultrasound offers higher spatial resolution and is capable of real-time vascular assessment and dynamic evaluation, this modality may offer a better diagnostic approach for evaluating for infrapatellar fat pad impingement. Thus far, only one study has reported on the normal ultrasound appearance of the infrapatellar fat pad in young, asymptomatic patients [12]. To our knowledge, no studies have evaluated ultrasound parameters in the assessment of abnormal infrapatellar fat pads. In this study, we therefore sought to determine whether there are ultrasound parameters that differ between knees that are clinically diagnosed with fat pad impingement and asymptomatic controls. 2. Materials and methods This prospective research study was approved by the institutional review board, and informed consent was obtained for all patients and controls. We sought to compare knees with signs and symptoms of infrapatellar synovial fat pad impingement to asymptomatic control knees using ultrasound evaluation. We compared knees under two conditions – before exercise and after exercise – to assess the influence of exercise on ultrasound imaging parameters among symptomatic and asymptomatic knees. Patients were recruited by the referring clinician. Patients eligible for inclusion were those aged 13–40 years who had experienced anterior knee pain for at least one month and had clinical signs and symptoms of infrapatellar fat pad impingement. Physical examination findings suggestive of infrapatellar fat pad impingement included a positive Hoffa test. A Hoffa test is performed by pressing two thumbs deep to either side of the patellar tendon just below the patella with the patient lying down and the knee bent; if pain is elicited as the patient straightens his or her leg, this is considered a positive sign for fat pad impingement. Exclusion criteria included a history of knee trauma, knee surgery, treatment for previous knee pain, abnormal radiographs or MRI scans (aside from infrapatellar fat pad findings on MRI), and physical examination findings suggestive of knee instability or internal derangement. Asymptomatic control subjects were recruited via flyer posting and were included if they lacked a history of knee pain, treatment for previous knee complaints, knee trauma, or knee surgery. The final study included a total of 21 knees: 11 symptomatic knees from 10 patients and 10 asymptomatic knees from five control subjects. 2.1. Ultrasound technique Ultrasound images were obtained by one of four musculoskeletally trained ultrasound technologists with four years of musculoskeletal ultrasound experience using the Siemens S2000 system (Siemens Medical Solutions, Erlangen, Germany). Grayscale, power Doppler, and dynamic images were acquired using a 14-MHz or 18-MHz linear transducer, depending on body habitus. To determine which ultrasound parameters should be evaluated in this study, we performed a small initial pilot study to assess the interobserver and intraobserver variability of ultrasound parameters in 0°, 45° and 90° of knee flexion. Based on the results of this pilot study, we determined that we would measure the following ultrasound parameters in study patients: anteroposterior distance of the largest fat lobule measured at 0° and 90° (Figure 1), from which a percent change was calculated to determine percent compressibility of the lobule; subjective assessment of the overall vascularity of the fat pad at 45°; largest vessel diameter in the superolateral fat pad at 45° (Figure 2); and subjective assessment of fat pad motion when the knee was moved from 0° to 90° and back to 0°. The overall vascularity of the fat pad was described based on number of vessels seen in the superolateral fat pad with the knee in 45° flexion, and categorized as none (zero vessels), mild (one to two vessels), moderate (three to five vessels), or hypervascular (six or more vessels). In all patients, the above parameters were measured before and after supervised exercise. After pre-exercise images were obtained, each patient exercised on a treadmill for a maximum of 15 min using a modified Bruce protocol [13] (Table 1) until he or she began experiencing knee symptoms. Every symptomatic patient in our study experienced symptoms within 15 min. As soon as symptoms occurred, patients were taken directly to the ultrasound suite across the hall. The control patients also exercised using this protocol for 15 min. The time between onset of symptoms in patients/end of exercise in controls and post-exercise ultrasound scanning was less than five minutes. Images were reviewed by a board-certified musculoskeletally trained radiologist with one year of experience. 2.2. Statistics Ultrasound parameters measured before and after exercise were reported. For each ultrasound parameter, a receiver operating characteristic (ROC) analysis was used to assess whether the change in the parameter with exercise differed for symptomatic and asymptomatic knees. A Wald test was used to assess the null hypothesis that the area under the ROC curve was 0.5, with the standard error adjusted to account for the clustered nature of the data (a patient could have both knees included in the study) [14]. A significance level of 0.05 was used. Percentile bootstrap 95% confidence intervals were computed for (1) the mean difference in the largest vessel Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
3
Figure 1. Fat pad compressibility. (A) Sagittal ultrasound image at the level of the patellar tendon with the knee in 90° flexion, demonstrating the anteroposterior distance of the largest fat lobule. (B) As the knee is moved into 0° flexion, the same fat lobule is observed in real time to flatten, and the anteroposterior distance of the lobule is measured again.
diameter for asymptomatic versus symptomatic knees in the pre-exercise state, (2) the mean difference in the change in compressibility of the largest fat lobule for asymptomatic versus symptomatic knees in the pre-exercise state, and (3) the mean difference in the change in compressibility of the largest fat lobule for asymptomatic versus symptomatic knees in the post-exercise state. Subjects were randomly sampled with replacement for each of 10,000 bootstrap samples. 3. Results Ultrasound parameters measured before and after exercise for asymptomatic (n = 10) and symptomatic (n = 11) knees are shown in Table 2. Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
4
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
Figure 2. Fat pad vascularity. Sagittal power Doppler ultrasound image at the superolateral aspect of the infrapatellar fat pad. The number of vessels and the largest vessel diameter were documented.
The mean percent change of the largest fat lobule was 63% for both pre-exercise and post-exercise states in asymptomatic knees; for symptomatic knees, the mean percent changes were 32% for the pre-exercise state and 50% for the post-exercise state (Table 2). In the pre-exercise state, the percent compressibility from 90° to 0° of the largest fat lobule was greater in asymptomatic knees than in symptomatic knees (63% vs. 32%; 95% confidence interval [CI] for the difference: 12%, 55%). In the post-exercise state, the percent compressibility from 90° to 0° of the largest fat lobule was slightly greater in asymptomatic knees than in symptomatic knees (63% vs. 50%; 95% CI for the difference: 0%, 28%). From pre-exercise to post-exercise, the symptomatic knees' compressibility increased by 18% on average, whereas there was no change on average (0%) in the asymptomatic knees (P = 0.241). In the pre-exercise state, the largest vessel diameter tended to be larger overall in the symptomatic knees than in the asymptomatic knees (1.4 mm vs. 0.9 mm; 95% CI for the difference: 0.2, 0.8). However, after exercise, there was a statistically significant greater change in the vessel diameter of the asymptomatic knees compared to the symptomatic knees (0.83 vs. −0.02; P b 0.001; Table 2). As shown in Figure 3, the vessel diameter decreased after exercise in more than half of the symptomatic knees, whereas the vessel diameter increased to some degree in all but one of the asymptomatic knees. No statistically significant difference was observed between symptomatic and asymptomatic knees with respect to the subjective assessment of vascularity before and after exercise (P = 0.131; Table 3). In 60% (6/10) of asymptomatic knees compared to 27% (3/11) of symptomatic knees, the overall vascularity increased. The majority of symptomatic knees (64%) had vascularity (i.e. number of vessels) that did not change between the pre-exercise and post-exercise states. There was also no significant difference between groups in the subjective assessment of fat pad motion before and after exercise (P = 0.115). Both symptomatic and asymptomatic groups had knees that showed decreased, increased, or the same dynamic motion without an identifiable trend in either group between the pre-exercise and post-exercise states. 4. Discussion In 1904, Albert Hoffa first suggested that a syndrome of infrapatellar fat pad impingement characterized by inflammatory hyperplasia mixed with regions of hemorrhage and fibrosis could be a cause of anterior knee pain occurring after acute trauma or repetitive microtrauma [3]. More recently, impingement of the superolateral aspect of the fat pad between the patellar tendon and lateral femoral
Table 1 Modified Bruce treadmill protocol, three-minute stages. Stage
Speed (mph)
Grade (%)
1 2 3 4 5
2.5 3.4 4.2 5.0 5.5
12 14 16 18 20
Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
5
Table 2 Ultrasound results before and after exercise in asymptomatic and symptomatic knees. Ultrasound result
Mean largest vessel diameter (mm) Mean percent change of largest fat lobule (%) Overall vascularity, n (%) None Mild Moderate Hypervascular Dynamic observations of motion, n (%) None Mild Moderate Severe Not obtained
Asymptomatic knees (n = 10)
Symptomatic knees (n = 11)
Pre-exercise
Post-exercise
Pre-exercise
Post-exercise
0.88 63
1.71 63
1.36 32
1.35 50
0 (0) 8 (80) 2 (20) 0 (0)
0 (0) 3 (30) 6 (60) 1 (10)
0 (0) 8 (73) 3 (27) 0 (0)
0 (0) 6 (55) 4 (36) 1 (9)
0 (0) 3 (30) 6 (60) 0 (0) 1 (10)
0 (0) 5 (50) 5 (50) 0 (0) 0 (0)
0 (0) 5 (45) 6 (55) 0 (0) 0 (0)
0 (0) 4 (36) 5 (46) 2 (18) 0 (0)
condyle has been described in such patients. Several MRI studies have shown this impingement to be associated with increased T2/edema-like signal in the superolateral fat pad, which is postulated to be associated with patellar maltracking [1,6–11]. However, to our knowledge, no previous studies have identified ultrasound parameters that could be used to assess the symptomatic infrapatellar fat pad. Our pilot study assessed the infrapatellar fat pad using four parameters: anteroposterior distance of the largest fat lobule at 0° to 90° with calculated percent change to determine percent compressibility of the lobule; subjective assessment of the overall vascularity of the fat pad at 45°; largest vessel diameter in the superolateral fat pad at 45°; and subjective assessment of fat pad motion when the knee was moved between 0° and 90°. In our study, pre-exercise fat pad lobule compressibility was diminished in symptomatic knees as compared to asymptomatic knees (32% vs. 63%); we postulate that this diminished compressibility is due to an underlying degree of chronic fibrosis. After
Figure 3. Vessel diameter change of fat pads from pre-exercise to post-exercise state. Change from before exercise to after exercise in the largest vessel diameter for symptomatic and asymptomatic knees. Values above zero on the y-axis indicate knees in which the largest vessel diameter increased after exercise; values below zero indicate a decrease after exercise.
Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
6
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
Table 3 Changes in ultrasound results with exercise in asymptomatic and symptomatic knees. Ultrasound result
Asymptomatic knees (n = 10)
Symptomatic knees (n = 11)
P
Mean change in the largest vessel diameter (mm) Mean change in the percent change of largest fat lobule from pre- to post-exercise state (%) Overall vascularity, n (%) Decreased Stayed the same Increased Dynamic observations of motion, n (%) Decreased Stayed the same Increased Not obtained
0.83 0
−0.02 18
b0.001 0.241 0.131
0 (0) 4 (40) 6 (60)
1 (9) 7 (64) 3 (27)
3 (30) 5 (50) 1 (10) 1 (10)
2 (18) 5 (45) 4 (36) 0 (0)
0.115
exercise, compressibility increased in the symptomatic knees to a more normal percentage (and to a percentage similar to that seen in asymptomatic knees, 50%). If the impingement is a function of inflammation, then compressibility may be increasing due to increased blood flow and edema, as exercise normally increases blood flow. In our study, an increase in vessel diameter from pre-exercise to post-exercise states was more prevalent in asymptomatic knees than in symptomatic knees. Interestingly, the vessel diameter was initially larger overall in pre-exercise symptomatic knees than in asymptomatic knees (1.4 mm vs. 0.8 mm); this finding correlates with previous findings of dilated vessels in the infrapatellar fat pad in patients with fat pad impingement [15]. Assessment of the vessels after exercise, however, did not show an increase in size in the symptomatic knees as one would expect in any exercised tissue. After exercise, vessels in the asymptomatic knees dilated to an average of 1.7 mm (average increase, 0.8 mm), whereas vessels in the symptomatic knees remained at an average vessel diameter of approximately 1.4 mm, with more than half of the symptomatic knees actually demonstrating constriction (Figure 3). Perhaps this is secondary to the microscarring and hypertrophy of the fat pad preventing dilation. Alternatively, the vessel may in some way already be damaged by the intrinsic fat pad pathology, preventing further dilation. No significant differences were observed between the two groups in subjective assessments of overall vascularity or fat pad motion. One patient in our clinical practice outside of the study was thought to have severe infrapatellar fat pad impingement; this patient demonstrated marked asymmetric changes in fat pad compressibility and vessel diameter, as well as a marked decrease in overall blood flow with diffusely constricted vessels after exercise. These vessels were observed in real time to reperfuse the fat pad as the symptoms resolved after a period of rest. Lack of vessel dilation in the symptomatic patients included in our study raises the possibility of an ischemic etiology for their pain. This would have therapeutic implications that differ from current standards, including treatment with topical vasodilators. Clearly, larger prospective studies focusing on the variable blood flow to the fat pad and including post-treatment evaluation would be helpful to further elucidate our findings. The major limitation of this study was its small sample size. This precluded statistical adjustment for potential confounders in the comparison of asymptomatic and symptomatic knees. Additionally, the lack of statistically significant differences between groups for several findings may have reflected a true lack of difference or may have been the result of the small study population. A larger sample size would provide greater potential for evaluating fat pad compressibility, as there are inherent limitations to ultrasound that limit our ability to “see” the entire fat pad because of overlying bony structures. Therefore, true volume measurements of the fat pad are not feasible with ultrasound. Given the histologic alteration of the fat pad in impingement syndrome, advanced ultrasound software techniques such as elastography, which evaluates relative stiffness of tissues, and shear wave imaging, which directly measures the shear wave velocities of tissues, would be interesting to evaluate. A recent study has shown normal tissue-specific parameters of the fat pad, including elastography [12], and future studies should build on these findings to evaluate differences in stiffness of normal and symptomatic fat pads. Although lack of histologic correlation makes interpretation of these parameters difficult, fat pad elastography and/or shear wave imaging could potentially be used for histologic correlation before fat pad partial excision surgery. Our study also raises the possibility of developing other treatment options for infrapatellar fat pad impingement, such as topical vasodilators. Once the ultrasound parameters identified in this report are validated in larger-scale studies, these factors could be used to evaluate the efficacy of these additional treatment options. In conclusion, this study provides a starting point for assessing various parameters of the symptomatic infrapatellar fat pad with ultrasound. The pre-exercise changes in fat pad compressibility and the postexercise changes in vessel dilation observed in this study provide insight into the possible pathologies occurring in this syndrome, including intrinsic fat pad scarring, vessel damage, and subsequent soft tissue ischemia. Ultrasound may prove to be a valuable tool for future research into the diagnosis and treatment follow-up of fat pad impingement.
Conflicts of interest The authors have no conflicts pertinent to this work to report. Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
7
Acknowledgments We would like to acknowledge the many contributors to this work, including our musculoskeletal ultrasound technology staff. Special thanks to our technical writer, Megan Griffiths, for her critical review of this manuscript. Preliminary findings of this research were presented at the American Institute of Ultrasound in Medicine meeting in Las Vegas, Nevada, in April 2014. This research was not supported by any external funding. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References [1] Campolo M, Babu J, Dmochowska K, Scariah S, Varughese J. A comparison of two taping techniques (Kinesio and McConnell) and their effect on anterior knee pain during functional activities. Int J Sports Phys Ther 2013;8:105–10. [2] Kohn D, Deiler S, Rudert M. Arterial blood supply of the infrapatellar fat pad. Anatomy and clinical consequences. Arch Orthop Trauma Surg 1995;114:72–5. [3] Hoffa A. The influence of the adipose tissue with regard to the pathology of the knee joint. JAMA 1904;43:795–6. [4] LLopis E, Padrón M. Anterior knee pain. Eur J Radiol 2007;62:27–43. [5] Dragoo JL, Johnson C, McConnell J. Evaluation and treatment of disorders of the infrapatellar fat pad. Sports Med 2012;42:51–67. [6] Jibri Z, Martin D, Mansour R, Kamath S. The association of infrapatellar fat pad oedema with patellar maltracking: a case–control study. Skeletal Radiol 2012;41: 925–31. [7] Subhawong TK, Eng J, Carrino JA, Chhabra A. Superolateral Hoffa's fat pad edema: association with patellofemoral maltracking and impingement. AJR Am J Roentgenol 2010;195:1367–73. [8] von Engelhardt LV, Tokmakidis E, Lahner M, David A, Haage P, Bouillon B, et al. Hoffa's fat pad impingement treated arthroscopically: related findings on preoperative MRI in a case series of 62 patients. Arch Orthop Trauma Surg 2010;130:1041–51. [9] Saddik D, McNally EG, Richardson M. MRI of Hoffa's fat pad. Skeletal Radiol 2004;33:433–44. [10] Jacobson JA, Lenchik L, Ruhoy MK, Schweitzer ME, Resnick D. MR imaging of the infrapatellar fat pad of Hoffa. Radiographics 1997;17:675–91. [11] Chung CB, Skaf A, Roger B, Campos J, Stump X, Resnick D. Patellar tendon-lateral femoral condyle friction syndrome: MR imaging in 42 patients. Skeletal Radiol 2001;30:694–7. [12] Vera-Pérez E, Sánchez-Bringas G, Ventura-Ríos L, Hernández-Díaz C, Cortés S, Gutiérrez M, et al. Sonographic characterization of Hoffa's fat pad. A pilot study. Rheumatol Int 2017;37:757–64. [13] Bruce RA, Kusumi F, Hosmer D. Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular disease. Am Heart J 1973;85:546–62. [14] Obuchowski NA. Nonparametric analysis of clustered ROC curve data. Biometrics 1997;53:567–78. [15] Kumar D, Alvand A, Beacon JP. Impingement of infrapatellar fat pad (Hoffa's disease): results of high-portal arthroscopic resection. Arthroscopy 2007;23:1180–6.
Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
Contents lists available at ScienceDirect
The Knee
Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study Himabindu Mikkilineni a,⁎, Patricia B. Delzell a, Jack Andrish b, Jennifer Bullen c, Nancy A. Obuchowski c, Naveen Subhas a, Joshua M. Polster a, Jean P. Schils a a b c
Cleveland Clinic, Imaging Institute, 9500 Euclid Avenue, Cleveland, OH 44195, USA Cleveland Clinic, Department of Orthopedic Surgery, 9500 Euclid Avenue, Cleveland, OH 44195, USA Cleveland Clinic, Quantitative Health Sciences, 9500 Euclid Avenue, Cleveland, OH 44195, USA
a r t i c l e
i n f o
Article history: Received 5 April 2017 Received in revised form 8 December 2017 Accepted 26 January 2018 Available online xxxx Keywords: Ultrasound Hoffa fat pad Impingement Infrapatellar fat pad
a b s t r a c t Background: We sought to determine whether there are ultrasound parameters that differ between knees with symptomatic fat pad synovial impingement and asymptomatic knees. Methods: A prospective study was performed in patients with clinical signs and symptoms of fat pad synovial impingement and asymptomatic controls. Eleven symptomatic knees and 10 asymptomatic controls were evaluated. Ultrasound imaging was performed before and after exercise. Evaluated parameters included largest fat lobule compressibility, subjective assessment of vascularity, largest vessel diameter, and subjective assessment of dynamic fat pad motion during flexion and extension. Receiver operating characteristic (ROC) curve analysis was used to assess whether changes in these parameters were different between symptomatic and asymptomatic knees. Results: Change in the largest vessel diameter was greater and trended toward dilation in asymptomatic knees compared to symptomatic knees (mean: 0.83 vs. −0.02; P b 0.001). No significant differences were observed between symptomatic and asymptomatic knees with respect to preexercise versus post-exercise states in subjective assessment of vascularity (P = 0.131), fat pad motion (P = 0.115), or percent change of the largest fat lobule (P = 0.241). However, overall compressibility of the fat pad lobule was significantly diminished in the pre-exercise state in symptomatic knees compared to asymptomatic controls. Conclusions: This study demonstrated a statistically significant change in the largest vessel diameter from pre- to post-exercise states between symptomatic and asymptomatic knees, as well as abnormal pre-exercise fat lobule compressibility in symptomatic knees. These findings show promise that with further research, ultrasound could have clinical utility in diagnosing infrapatellar fat pad impingement. © 2018 Elsevier B.V. All rights reserved.
1. Introduction Knee pain is the most common complaint of patients presenting to the general orthopedic surgeon's office, with pain most frequently reported in the anterior knee. Recently, the introduction of taping techniques aimed at improving patellofemoral alignment has renewed interest in the role of fat pad synovial impingement as an etiology for anterior knee pain [1]. The infrapatellar fat pad (or Hoffa's fat pad) has a rich vascular and nervous supply [2], and it is currently believed that inflammation of the infrapatellar fat pad can occur after acute injury or repetitive microtrauma, leading to anterior knee pain. The etiology of impingement may arise from abnormalities related to patellofemoral maltracking [3–6]. Current nonoperative treatment methods for ⁎ Corresponding author at: 25875 Science Park Drive, AC116, Beachwood, OH 44122, USA. E-mail address: [email protected] (H. Mikkilineni).
https://doi.org/10.1016/j.knee.2018.01.008 0968-0160/© 2018 Elsevier B.V. All rights reserved.
Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
2
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
infrapatellar fat pad impingement include nonsteroidal anti-inflammatory drug therapy, taping, physical therapy, and localized anesthetic or steroid injections. If nonoperative therapy fails, surgical options such as partial fat pad resection or debridement may be considered [5]. Symptomatic infrapatellar fat pad impingement is currently diagnosed based on physical examination and magnetic resonance imaging (MRI) findings. However, on physical examination, distinguishing between infrapatellar fat pad irritation and patellar tendon pathology can be difficult [4,5]. MRI findings such as superolateral fat pad edema or static measurements to identify dynamic patellofemoral maltracking abnormalities have been validated by a few studies [6–11]. However, because ultrasound offers higher spatial resolution and is capable of real-time vascular assessment and dynamic evaluation, this modality may offer a better diagnostic approach for evaluating for infrapatellar fat pad impingement. Thus far, only one study has reported on the normal ultrasound appearance of the infrapatellar fat pad in young, asymptomatic patients [12]. To our knowledge, no studies have evaluated ultrasound parameters in the assessment of abnormal infrapatellar fat pads. In this study, we therefore sought to determine whether there are ultrasound parameters that differ between knees that are clinically diagnosed with fat pad impingement and asymptomatic controls. 2. Materials and methods This prospective research study was approved by the institutional review board, and informed consent was obtained for all patients and controls. We sought to compare knees with signs and symptoms of infrapatellar synovial fat pad impingement to asymptomatic control knees using ultrasound evaluation. We compared knees under two conditions – before exercise and after exercise – to assess the influence of exercise on ultrasound imaging parameters among symptomatic and asymptomatic knees. Patients were recruited by the referring clinician. Patients eligible for inclusion were those aged 13–40 years who had experienced anterior knee pain for at least one month and had clinical signs and symptoms of infrapatellar fat pad impingement. Physical examination findings suggestive of infrapatellar fat pad impingement included a positive Hoffa test. A Hoffa test is performed by pressing two thumbs deep to either side of the patellar tendon just below the patella with the patient lying down and the knee bent; if pain is elicited as the patient straightens his or her leg, this is considered a positive sign for fat pad impingement. Exclusion criteria included a history of knee trauma, knee surgery, treatment for previous knee pain, abnormal radiographs or MRI scans (aside from infrapatellar fat pad findings on MRI), and physical examination findings suggestive of knee instability or internal derangement. Asymptomatic control subjects were recruited via flyer posting and were included if they lacked a history of knee pain, treatment for previous knee complaints, knee trauma, or knee surgery. The final study included a total of 21 knees: 11 symptomatic knees from 10 patients and 10 asymptomatic knees from five control subjects. 2.1. Ultrasound technique Ultrasound images were obtained by one of four musculoskeletally trained ultrasound technologists with four years of musculoskeletal ultrasound experience using the Siemens S2000 system (Siemens Medical Solutions, Erlangen, Germany). Grayscale, power Doppler, and dynamic images were acquired using a 14-MHz or 18-MHz linear transducer, depending on body habitus. To determine which ultrasound parameters should be evaluated in this study, we performed a small initial pilot study to assess the interobserver and intraobserver variability of ultrasound parameters in 0°, 45° and 90° of knee flexion. Based on the results of this pilot study, we determined that we would measure the following ultrasound parameters in study patients: anteroposterior distance of the largest fat lobule measured at 0° and 90° (Figure 1), from which a percent change was calculated to determine percent compressibility of the lobule; subjective assessment of the overall vascularity of the fat pad at 45°; largest vessel diameter in the superolateral fat pad at 45° (Figure 2); and subjective assessment of fat pad motion when the knee was moved from 0° to 90° and back to 0°. The overall vascularity of the fat pad was described based on number of vessels seen in the superolateral fat pad with the knee in 45° flexion, and categorized as none (zero vessels), mild (one to two vessels), moderate (three to five vessels), or hypervascular (six or more vessels). In all patients, the above parameters were measured before and after supervised exercise. After pre-exercise images were obtained, each patient exercised on a treadmill for a maximum of 15 min using a modified Bruce protocol [13] (Table 1) until he or she began experiencing knee symptoms. Every symptomatic patient in our study experienced symptoms within 15 min. As soon as symptoms occurred, patients were taken directly to the ultrasound suite across the hall. The control patients also exercised using this protocol for 15 min. The time between onset of symptoms in patients/end of exercise in controls and post-exercise ultrasound scanning was less than five minutes. Images were reviewed by a board-certified musculoskeletally trained radiologist with one year of experience. 2.2. Statistics Ultrasound parameters measured before and after exercise were reported. For each ultrasound parameter, a receiver operating characteristic (ROC) analysis was used to assess whether the change in the parameter with exercise differed for symptomatic and asymptomatic knees. A Wald test was used to assess the null hypothesis that the area under the ROC curve was 0.5, with the standard error adjusted to account for the clustered nature of the data (a patient could have both knees included in the study) [14]. A significance level of 0.05 was used. Percentile bootstrap 95% confidence intervals were computed for (1) the mean difference in the largest vessel Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
3
Figure 1. Fat pad compressibility. (A) Sagittal ultrasound image at the level of the patellar tendon with the knee in 90° flexion, demonstrating the anteroposterior distance of the largest fat lobule. (B) As the knee is moved into 0° flexion, the same fat lobule is observed in real time to flatten, and the anteroposterior distance of the lobule is measured again.
diameter for asymptomatic versus symptomatic knees in the pre-exercise state, (2) the mean difference in the change in compressibility of the largest fat lobule for asymptomatic versus symptomatic knees in the pre-exercise state, and (3) the mean difference in the change in compressibility of the largest fat lobule for asymptomatic versus symptomatic knees in the post-exercise state. Subjects were randomly sampled with replacement for each of 10,000 bootstrap samples. 3. Results Ultrasound parameters measured before and after exercise for asymptomatic (n = 10) and symptomatic (n = 11) knees are shown in Table 2. Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
4
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
Figure 2. Fat pad vascularity. Sagittal power Doppler ultrasound image at the superolateral aspect of the infrapatellar fat pad. The number of vessels and the largest vessel diameter were documented.
The mean percent change of the largest fat lobule was 63% for both pre-exercise and post-exercise states in asymptomatic knees; for symptomatic knees, the mean percent changes were 32% for the pre-exercise state and 50% for the post-exercise state (Table 2). In the pre-exercise state, the percent compressibility from 90° to 0° of the largest fat lobule was greater in asymptomatic knees than in symptomatic knees (63% vs. 32%; 95% confidence interval [CI] for the difference: 12%, 55%). In the post-exercise state, the percent compressibility from 90° to 0° of the largest fat lobule was slightly greater in asymptomatic knees than in symptomatic knees (63% vs. 50%; 95% CI for the difference: 0%, 28%). From pre-exercise to post-exercise, the symptomatic knees' compressibility increased by 18% on average, whereas there was no change on average (0%) in the asymptomatic knees (P = 0.241). In the pre-exercise state, the largest vessel diameter tended to be larger overall in the symptomatic knees than in the asymptomatic knees (1.4 mm vs. 0.9 mm; 95% CI for the difference: 0.2, 0.8). However, after exercise, there was a statistically significant greater change in the vessel diameter of the asymptomatic knees compared to the symptomatic knees (0.83 vs. −0.02; P b 0.001; Table 2). As shown in Figure 3, the vessel diameter decreased after exercise in more than half of the symptomatic knees, whereas the vessel diameter increased to some degree in all but one of the asymptomatic knees. No statistically significant difference was observed between symptomatic and asymptomatic knees with respect to the subjective assessment of vascularity before and after exercise (P = 0.131; Table 3). In 60% (6/10) of asymptomatic knees compared to 27% (3/11) of symptomatic knees, the overall vascularity increased. The majority of symptomatic knees (64%) had vascularity (i.e. number of vessels) that did not change between the pre-exercise and post-exercise states. There was also no significant difference between groups in the subjective assessment of fat pad motion before and after exercise (P = 0.115). Both symptomatic and asymptomatic groups had knees that showed decreased, increased, or the same dynamic motion without an identifiable trend in either group between the pre-exercise and post-exercise states. 4. Discussion In 1904, Albert Hoffa first suggested that a syndrome of infrapatellar fat pad impingement characterized by inflammatory hyperplasia mixed with regions of hemorrhage and fibrosis could be a cause of anterior knee pain occurring after acute trauma or repetitive microtrauma [3]. More recently, impingement of the superolateral aspect of the fat pad between the patellar tendon and lateral femoral
Table 1 Modified Bruce treadmill protocol, three-minute stages. Stage
Speed (mph)
Grade (%)
1 2 3 4 5
2.5 3.4 4.2 5.0 5.5
12 14 16 18 20
Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
5
Table 2 Ultrasound results before and after exercise in asymptomatic and symptomatic knees. Ultrasound result
Mean largest vessel diameter (mm) Mean percent change of largest fat lobule (%) Overall vascularity, n (%) None Mild Moderate Hypervascular Dynamic observations of motion, n (%) None Mild Moderate Severe Not obtained
Asymptomatic knees (n = 10)
Symptomatic knees (n = 11)
Pre-exercise
Post-exercise
Pre-exercise
Post-exercise
0.88 63
1.71 63
1.36 32
1.35 50
0 (0) 8 (80) 2 (20) 0 (0)
0 (0) 3 (30) 6 (60) 1 (10)
0 (0) 8 (73) 3 (27) 0 (0)
0 (0) 6 (55) 4 (36) 1 (9)
0 (0) 3 (30) 6 (60) 0 (0) 1 (10)
0 (0) 5 (50) 5 (50) 0 (0) 0 (0)
0 (0) 5 (45) 6 (55) 0 (0) 0 (0)
0 (0) 4 (36) 5 (46) 2 (18) 0 (0)
condyle has been described in such patients. Several MRI studies have shown this impingement to be associated with increased T2/edema-like signal in the superolateral fat pad, which is postulated to be associated with patellar maltracking [1,6–11]. However, to our knowledge, no previous studies have identified ultrasound parameters that could be used to assess the symptomatic infrapatellar fat pad. Our pilot study assessed the infrapatellar fat pad using four parameters: anteroposterior distance of the largest fat lobule at 0° to 90° with calculated percent change to determine percent compressibility of the lobule; subjective assessment of the overall vascularity of the fat pad at 45°; largest vessel diameter in the superolateral fat pad at 45°; and subjective assessment of fat pad motion when the knee was moved between 0° and 90°. In our study, pre-exercise fat pad lobule compressibility was diminished in symptomatic knees as compared to asymptomatic knees (32% vs. 63%); we postulate that this diminished compressibility is due to an underlying degree of chronic fibrosis. After
Figure 3. Vessel diameter change of fat pads from pre-exercise to post-exercise state. Change from before exercise to after exercise in the largest vessel diameter for symptomatic and asymptomatic knees. Values above zero on the y-axis indicate knees in which the largest vessel diameter increased after exercise; values below zero indicate a decrease after exercise.
Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
6
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
Table 3 Changes in ultrasound results with exercise in asymptomatic and symptomatic knees. Ultrasound result
Asymptomatic knees (n = 10)
Symptomatic knees (n = 11)
P
Mean change in the largest vessel diameter (mm) Mean change in the percent change of largest fat lobule from pre- to post-exercise state (%) Overall vascularity, n (%) Decreased Stayed the same Increased Dynamic observations of motion, n (%) Decreased Stayed the same Increased Not obtained
0.83 0
−0.02 18
b0.001 0.241 0.131
0 (0) 4 (40) 6 (60)
1 (9) 7 (64) 3 (27)
3 (30) 5 (50) 1 (10) 1 (10)
2 (18) 5 (45) 4 (36) 0 (0)
0.115
exercise, compressibility increased in the symptomatic knees to a more normal percentage (and to a percentage similar to that seen in asymptomatic knees, 50%). If the impingement is a function of inflammation, then compressibility may be increasing due to increased blood flow and edema, as exercise normally increases blood flow. In our study, an increase in vessel diameter from pre-exercise to post-exercise states was more prevalent in asymptomatic knees than in symptomatic knees. Interestingly, the vessel diameter was initially larger overall in pre-exercise symptomatic knees than in asymptomatic knees (1.4 mm vs. 0.8 mm); this finding correlates with previous findings of dilated vessels in the infrapatellar fat pad in patients with fat pad impingement [15]. Assessment of the vessels after exercise, however, did not show an increase in size in the symptomatic knees as one would expect in any exercised tissue. After exercise, vessels in the asymptomatic knees dilated to an average of 1.7 mm (average increase, 0.8 mm), whereas vessels in the symptomatic knees remained at an average vessel diameter of approximately 1.4 mm, with more than half of the symptomatic knees actually demonstrating constriction (Figure 3). Perhaps this is secondary to the microscarring and hypertrophy of the fat pad preventing dilation. Alternatively, the vessel may in some way already be damaged by the intrinsic fat pad pathology, preventing further dilation. No significant differences were observed between the two groups in subjective assessments of overall vascularity or fat pad motion. One patient in our clinical practice outside of the study was thought to have severe infrapatellar fat pad impingement; this patient demonstrated marked asymmetric changes in fat pad compressibility and vessel diameter, as well as a marked decrease in overall blood flow with diffusely constricted vessels after exercise. These vessels were observed in real time to reperfuse the fat pad as the symptoms resolved after a period of rest. Lack of vessel dilation in the symptomatic patients included in our study raises the possibility of an ischemic etiology for their pain. This would have therapeutic implications that differ from current standards, including treatment with topical vasodilators. Clearly, larger prospective studies focusing on the variable blood flow to the fat pad and including post-treatment evaluation would be helpful to further elucidate our findings. The major limitation of this study was its small sample size. This precluded statistical adjustment for potential confounders in the comparison of asymptomatic and symptomatic knees. Additionally, the lack of statistically significant differences between groups for several findings may have reflected a true lack of difference or may have been the result of the small study population. A larger sample size would provide greater potential for evaluating fat pad compressibility, as there are inherent limitations to ultrasound that limit our ability to “see” the entire fat pad because of overlying bony structures. Therefore, true volume measurements of the fat pad are not feasible with ultrasound. Given the histologic alteration of the fat pad in impingement syndrome, advanced ultrasound software techniques such as elastography, which evaluates relative stiffness of tissues, and shear wave imaging, which directly measures the shear wave velocities of tissues, would be interesting to evaluate. A recent study has shown normal tissue-specific parameters of the fat pad, including elastography [12], and future studies should build on these findings to evaluate differences in stiffness of normal and symptomatic fat pads. Although lack of histologic correlation makes interpretation of these parameters difficult, fat pad elastography and/or shear wave imaging could potentially be used for histologic correlation before fat pad partial excision surgery. Our study also raises the possibility of developing other treatment options for infrapatellar fat pad impingement, such as topical vasodilators. Once the ultrasound parameters identified in this report are validated in larger-scale studies, these factors could be used to evaluate the efficacy of these additional treatment options. In conclusion, this study provides a starting point for assessing various parameters of the symptomatic infrapatellar fat pad with ultrasound. The pre-exercise changes in fat pad compressibility and the postexercise changes in vessel dilation observed in this study provide insight into the possible pathologies occurring in this syndrome, including intrinsic fat pad scarring, vessel damage, and subsequent soft tissue ischemia. Ultrasound may prove to be a valuable tool for future research into the diagnosis and treatment follow-up of fat pad impingement.
Conflicts of interest The authors have no conflicts pertinent to this work to report. Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008
H. Mikkilineni et al. / The Knee xxx (2018) xxx–xxx
7
Acknowledgments We would like to acknowledge the many contributors to this work, including our musculoskeletal ultrasound technology staff. Special thanks to our technical writer, Megan Griffiths, for her critical review of this manuscript. Preliminary findings of this research were presented at the American Institute of Ultrasound in Medicine meeting in Las Vegas, Nevada, in April 2014. This research was not supported by any external funding. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References [1] Campolo M, Babu J, Dmochowska K, Scariah S, Varughese J. A comparison of two taping techniques (Kinesio and McConnell) and their effect on anterior knee pain during functional activities. Int J Sports Phys Ther 2013;8:105–10. [2] Kohn D, Deiler S, Rudert M. Arterial blood supply of the infrapatellar fat pad. Anatomy and clinical consequences. Arch Orthop Trauma Surg 1995;114:72–5. [3] Hoffa A. The influence of the adipose tissue with regard to the pathology of the knee joint. JAMA 1904;43:795–6. [4] LLopis E, Padrón M. Anterior knee pain. Eur J Radiol 2007;62:27–43. [5] Dragoo JL, Johnson C, McConnell J. Evaluation and treatment of disorders of the infrapatellar fat pad. Sports Med 2012;42:51–67. [6] Jibri Z, Martin D, Mansour R, Kamath S. The association of infrapatellar fat pad oedema with patellar maltracking: a case–control study. Skeletal Radiol 2012;41: 925–31. [7] Subhawong TK, Eng J, Carrino JA, Chhabra A. Superolateral Hoffa's fat pad edema: association with patellofemoral maltracking and impingement. AJR Am J Roentgenol 2010;195:1367–73. [8] von Engelhardt LV, Tokmakidis E, Lahner M, David A, Haage P, Bouillon B, et al. Hoffa's fat pad impingement treated arthroscopically: related findings on preoperative MRI in a case series of 62 patients. Arch Orthop Trauma Surg 2010;130:1041–51. [9] Saddik D, McNally EG, Richardson M. MRI of Hoffa's fat pad. Skeletal Radiol 2004;33:433–44. [10] Jacobson JA, Lenchik L, Ruhoy MK, Schweitzer ME, Resnick D. MR imaging of the infrapatellar fat pad of Hoffa. Radiographics 1997;17:675–91. [11] Chung CB, Skaf A, Roger B, Campos J, Stump X, Resnick D. Patellar tendon-lateral femoral condyle friction syndrome: MR imaging in 42 patients. Skeletal Radiol 2001;30:694–7. [12] Vera-Pérez E, Sánchez-Bringas G, Ventura-Ríos L, Hernández-Díaz C, Cortés S, Gutiérrez M, et al. Sonographic characterization of Hoffa's fat pad. A pilot study. Rheumatol Int 2017;37:757–64. [13] Bruce RA, Kusumi F, Hosmer D. Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular disease. Am Heart J 1973;85:546–62. [14] Obuchowski NA. Nonparametric analysis of clustered ROC curve data. Biometrics 1997;53:567–78. [15] Kumar D, Alvand A, Beacon JP. Impingement of infrapatellar fat pad (Hoffa's disease): results of high-portal arthroscopic resection. Arthroscopy 2007;23:1180–6.
Please cite this article as: Mikkilineni H, et al, Ultrasound evaluation of infrapatellar fat pad impingement: An exploratory prospective study, Knee (2018), https://doi.org/10.1016/j.knee.2018.01.008