Introduction

Knee osteoarthritis (KOA) is a progressive, degenerative disease and one of the most common causes of physical disability1. Symptomatic KOA occurs in 20–30% of the elderly population aged ≥ 65 years2,3, and its prevalence is increasing worldwide, partly due to the aging population4. Pain is the most common reason that patients request medical care; however, due to the inability to halt the progression of osteoarthritis (OA), treatments focus on relieving pain and maintaining joint function5.

KOA is considered to be a whole-joint disease involving all structures in the knee joint. Recent studies have provided deeper insights into the pathology of OA at each structure, including the cartilage, synovium, subchondral bone, and muscles6. Traditionally, radiography was used to assess the severity of KOA, though discordance between the radiographic grading of KOA and the severity of pain has been reported7. Therefore, the use of magnetic resonance imaging (MRI) to evaluate the knee joint has increased, and correlations between knee pain severity and MRI findings such as effusion synovitis and bone marrow lesions (BMLs) have been reported8.

Patients with KOA often report anterior knee pain. MRI findings of patellofemoral (PF) osteophytes, effusion synovitis, and infrapatellar synovitis are associated with anterior knee pain9. Although anterior knee pain is induced by several factors10, its pathophysiology remains unclear.

The infrapatellar fat pad (IPFP) is an intraarticular and extra-synovial adipose structure of richly innervated tissue in the knee joint11. The IPFP increases the articular congruity, lubricating the joint and serving as a protective cushion to the joint6. The IPFP may act as an endocrine organ by secreting several catabolic cytokines12,13. Moreover, the IPFP is a known source of anterior knee pain11. The role of the IPFP in the knee joint is complicated and not fully understood.

Recent findings suggest that the IPFP plays a role in the pathogenesis of KOA14. Along with synovitis, IPFP fibrosis is one of the most common findings in patients with KOA15. Due to advances in ultrasound elastography technology, the elasticity of soft tissues can be evaluated. Elastography has been clinically utilized to assess several diseases, including breast cancer, chronic hepatitis16, rotator cuff tears17, and Achilles tendinopathy18. However, despite the fact that OA pathology involves soft tissue around the joint, IPFP elasticity has rarely been assessed. IPFP fibrosis may be actively involved in OA progression and pain generation19,20. As the use of ultrasound elastography has been recommended for the assessment of liver fibrosis21, elastography of the IPFP in patients with KOA was conducted to evaluate the elasticity, which presumably reflects fibrosis.

Therefore, this study clarifies the elastic properties of the IPFP in patients with symptomatic KOA, focusing on radiological OA stages and anterior knee pain. It was hypothesized that IPFP elasticity is associated with anterior knee pain, regardless of the stage of OA.

Methods

Patients

This cross-sectional study included a convenience sample of 120 patients diagnosed with KOA (Kellgren and Lawrence [KL] grade ≥ 2)22 who visited our hospital between July 2020 and December 2021. Patients with systemic inflammatory diseases (such as rheumatoid arthritis), mental handicaps or psychiatric conditions, or a history of surgery on the affected knee were excluded from the study. Patients with PF joint OA (KL grade of ≥ 3) were also excluded to minimize the effect of PF joint OA on the study results. This study was approved by the Ethical Review Board of Kochi Medical School (approval number 2020-11). A verbal explanation of the study was provided to each patient, and written informed consent was obtained from each patient prior to their participation in the study. This study was conducted in compliance with the Declaration of Helsinki.

Outline of the experimental protocol

On the first visit, the patients’ demographic data were assessed and knee radiography and pressure pain threshold (PPT) measurements of the IPFP were performed. Knee MRI and ultrasound elastography of the IPFP were conducted on a different day for the patients’ convenience. Partial excision biopsies of the IPFP were performed in patients who subsequently underwent knee surgery.

Demographic data

Patient demographic data (sex, age, height, and weight) and the clinical characteristics of the affected knee were assessed. Knee pain maps have been used for the evaluation of anterior knee pain23. The knee pain map developed by Hill et al.24, which includes 12 areas separated by grid lines, was used in this study (Fig. 1). Area 1 represents the anterior superomedial aspect of the knee, while area 2 represents the anterior superior aspect of the knee. The anterior superolateral aspect of the knee is represented in area 3, the anterior medial aspect in area 4, the patella in area 5, and the anterior lateral aspect in area 6. Area 7 represents the patellar ligament, area 8 represents the pes anserinus, and area 9 represents the tibial tuberosity. The anterior inferolateral aspect of the knee is represented by area 10, the anterior inferomedial aspect by area 11, and the popliteal aspect by area 1225. The patella and patellar ligament (areas 5 and 7, respectively) were directly palpated and the patients were shown the locations of these areas on the knee pain map. Then, the patients walked at a comfortable speed and identified the painful areas of the knee while walking. If areas 5 or 7 were reported to be painful, the pain was categorized as anterior knee pain. Patients were subsequently divided into two groups based on the presence or absence of anterior knee pain.

Figure 1
figure 1

Knee pain map. A knee pain map of the right knee is shown. Patient-reported pain in areas 5 (patella) or 7 (patellar ligament) while walking is categorized as anterior knee pain.

Full size image

PPT measurement

To quantitatively assess IPFP tenderness, one author (YS) measured the PPT 1 cm below the inferior pole of the patella using a digital hand algometer (SDMEIC Electronics, Solna, Sweden) with a 1 cm2 probe. Pressure was applied at a rate of 30 kPa/s until the patient-reported pain. The PPT was measured five times at 20-s intervals, and the average of the middle three values was included in the study analyses. The intra-rater test–retest reliability was confirmed using data from 10 volunteers (intraclass correlation coefficient [ICC] (1, K): 0.97).

Imaging findings

The knee was radiographically examined in a standing position on the anteroposterior view, as well as on the lateral and skyline views. The severities of OA in the femorotibial (FT) and PF joints were assessed using the KL grade (0, none; 1, doubtful joint space narrowing with possible osteophytes; 2, possible joint space narrowing with definite osteophytes; 3, definite joint space narrowing, moderate osteophytes, some sclerosis; and 4, severe joint space narrowing, large osteophytes, marked sclerosis, bone deformities). Affected knees were scanned using 3 T MRI. The following image sequences were used: a T1-weighted three-dimensional (3D) spoiled gradient recall (SPGR) acquisition with fat saturation in the steady state (flip angle: 30°, repetition time: 16–18 ms, echo time: 3.3–4.7 ms, field of view: 16 cm, slice gap: 0 mm, slice thickness: 1.5 mm, sagittal images obtained at a slice thickness of 1.5 mm without an interslice gap) and a T2-weighted two-dimensional fast spin echo with fat suppression (flip angle: 90–142°, repetition time: 2500 ms, echo time: 100 ms, field of view: 16 cm, axial, sagittal, and coronal images obtained at a slice thickness of 4 mm with an interslice gap of 1 mm).

The Hoffa score, assessed using the MRI Osteoarthritis Knee Score26, was graded from 0 to 3 based on the degree of IPFP hyperintensity (0, normal; 1, mild; 2, moderate; and 3, severe). The effusion synovitis and BML scores were assessed using the whole-organ MRI score27. The effusion synovitis score was graded from 0 to 3 based on the estimated maximal distention of the synovial cavity (0, normal; 1, < 33% of the maximum potential distention; 2, 33–66% of the maximum potential distention; and 3, > 66% of the maximum potential distention). The BMLs were graded from 0–3 at all 15 areas around the knee joint based on the degree of bone hyperintensity (0, none; 1, < 25%; 2, 25–50%; and 3, > 50%). The BML score was calculated as the sum of the BML grades. All images were evaluated by a single observer (YS).

The IPFP area was measured by manually delineating the contour of the IPFP on each SPGR image using the software program OsiriX Lite (Pixmeo SARL, Bernex, Switzerland). A 3D reconstruction of the IPFP was created, and the IPFP volume was automatically calculated. A single observer (YS) graded the IPFP volume. The IPFP size was evaluated by dividing the IPFP volume by the body weight of each patient to adjust for the physique.

Ultrasound shear wave elastography (SWE)

Point SWE measurement of the IPFP was performed by one author (YS) using ultrasonography (ARIETTA E70; FUJIFILM Healthcare Corporation, Tokyo, Japan) with a C251 convex probe (1–5 MHz; FUJIFILM Healthcare Corporation). All patients were examined in the supine position at 20° of knee flexion, achieved using a rolled towel under the knee. B-mode axial images of the IPFP were visualized by placing the probe over the patellar ligament. The 10 × 15 mm2 region of interest28 was placed in the middle of the axial sectional view at the midpoint level of the IPFP. The IPFP in the region of interest was vibrated via acoustic radiation force impulses, and the shear wave speed was automatically measured. The depth of the region of interest was set at less than 6 cm to avoid acoustic radiation force impulse attenuation21. The elastography measurements were performed only when the percentage of the net amount of the effective shear wave velocity (VsN) was ≥ 50%. Measurements were performed five times, and the average of the middle three values was used in the analyses. The inter-rater test–retest reliability was confirmed using data from five participants (ICC (1, K): 0.83).

Histological IPFP fibrosis

Partial excision biopsies were performed at the central distal third of the IPFP in patients who underwent knee surgery after the study assessments. The samples were fixed in 10% formalin and embedded in paraffin, and three nonconsecutive sections containing the synovial membrane were obtained from the paraffin-embedded samples and stained with Masson’s trichrome. The lesion with the most extensive fibrosis among the three sections was detected using a microscope (BZ-X800; Keyence, Osaka, Japan) at 40 × magnification. IPFP fibrosis was evaluated using the ImageJ Fiji package29. The image of the lesion with the most extensive fibrosis was split into RGB channels, and image calculations were performed (blue image minus red image). The black area representing fibrosis was measured and expressed as a percentage.

Statistical analysis

Spearman’s rank correlation coefficient was used to compare the IPFP characteristics (size and elasticity) and the KL grade of the FT joint. Univariate analyses among the two groups—divided by the presence or absence of anterior knee pain—were conducted to compare patient age, sex, KL grade of the FT joint, Hoffa score, effusion synovitis score, BML score, IPFP size, and shear wave speed. The Fisher’s exact test and the Mann–Whitney U test were used because several parameters were non-normally distributed variables. A logistic regression analysis was conducted using factors with p-values ≤ 0.2 in the univariate analyses as explanatory variables after adjustment for sex and age. The KL grade is a widely used classification for KOA severity and has clinical significance; therefore, the logistic regression analysis was performed on the model with the KL grade added as an explanatory factor.

The correlation between the percentage of IPFP fibrosis and shear wave speed was examined using Pearson’s correlation coefficient to evaluate the reliability of SWE. The normality of the data was assessed via a visual inspection of the normality plot. Statistical analyses were performed using Bell Curve for Excel (Social Survey Research Information Co., Ltd., Tokyo, Japan) and EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan). A p-value < 0.05 was considered statistically significant.

Results

Patient characteristics

After the exclusion of 23 patients due to MRI examination difficulty (n = 1), IPFP compression by a ganglion (n = 1), mental handicap (n = 2), evaluation by 1.5 T MRI (n = 2), poor physical condition (n = 3), and PF joint KL grade ≥ 3 (n = 14), 97 patients were included in the final analysis. The SIGNA Architect 3.0 T MRI system (GE Healthcare, Chicago, IL, US) was used to evaluate 73 knees, and the Ingenia 3.0 T MRI system (Philips, Amsterdam, NL) was used to evaluate 24 knees. Seventy-three patients were women, and the mean patient age was 71.7 years (range 46–86 years) (Table 1). Patients with a KL grade of the FT joint of 4 had higher shear wave speeds (p < 0.001) than patients with other KL grades (Table 2).

Table 1 Patient characteristics (n = 97).
Full size table
Table 2 Patient characteristics according to KL grade of FT joint (n = 97).
Full size table

Association of IPFP size and elasticity with radiographic severity

The IPFP was significantly smaller (r = − 0.28; p = 0.006) and stiffer (r = 0.45; p < 0.001) as the KL grade severity of the FT joint increased (Spearman’s rank correlation coefficient).

Predisposing factors associated with anterior knee pain

Forty-one patients (42.3%) had anterior knee pain and exhibited significantly lower PPTs than patients without anterior knee pain (mean PPT: 419.5 ± 146.5 kPa vs. 498.9 ± 144.5 kPa; p = 0.006; Mann–Whitney U test). The shear wave speed was significantly associated with anterior knee pain in the univariate analyses (Table 3). Logistic regression analyses indicated that shear wave speed was the only significant explanatory variable of anterior knee pain (Table 4). The shear wave speed was also associated with PPT (r = − 0.20; p = 0.046; Pearson’s correlation analysis).

Table 3 Univariate analyses for predisposing factors associated with anterior knee pain (n = 97).
Full size table
Table 4 Logistic regression analysis for predisposing factors associated with anterior knee pain (n = 97).
Full size table

Association between the percentage of maximal fibrotic lesions and shear wave speed

Fourteen IPFP samples obtained from 14 patients were histologically evaluated. The mean patient age was 73.6 years (range 59–84 years), and the KL grades of the FT joint were 3 (n = 3) and 4 (n = 11). The mean percentage of fibrosis was 52.8% (range 29.5–88.8%) (Fig. 2). The percentage of maximal fibrosis tended to increase as the shear wave speed increased, though the correlation was not significant (r = 0.51; p = 0.062; Pearson’s correlation analysis) (Fig. 3).

Figure 2
figure 2

Histological infrapatellar fat pad fibrosis evaluation. (A) The infrapatellar fat pad (IPFP) lesion with the most extensive fibrosis of that in three nonconsecutive slice sections is detected in each sample using a microscope at 40 × magnification (scale bar = 500 µm). (B) The image of the IPFP lesion with the most extensive fibrosis is split into RGB channels. The black area representing the fibrosis is measured and expressed as the percentage. Image calculations are conducted using ImageJ Fiji package.

Full size image
Figure 3
figure 3

Association between the percentage of maximal fibrotic lesions and shear wave speed in the infrapatellar fat pad. The correlation between the percentage of maximal fibrosis and the shear wave speed is evaluated using 14 infrapatellar fat pad specimens (correlation coefficient: 0.51; p = 0.062).

Full size image

Discussion

In this study, smaller and stiffer IPFPs were associated with radiographic evidence of femorotibial osteoarthritis (FT-OA) progression, and patients with anterior knee pain had stiffer IPFPs regardless of the stage of FT-OA. This is the first study to evaluate the relationship between IPFP elasticity and anterior knee pain.

During SWE, the target tissue is vibrated using acoustic radiation force impulses, and the generated shear wave propagation speed is measured16. SWE provides quantitative evaluations and is less dependent on the examiner than other ultrasound elastography methods. In contrast, SWE results may vary due to the heterogeneous acoustic properties of tissues and refraction of the shear wave16. In this study, the IPFP was evaluated at a unified position and axis to achieve an accurate and consistent examination. In addition, only values with a VsN ≥ 50% were considered reliable. The VsN is a reliability index, and a cutoff value of 50% is widely utilized for evaluating the shear wave speed when staging liver fibrosis28,30. Although no significant correlation was observed between the percentage of maximal fibrotic lesions and shear wave speed, the percentage of fibrosis tended to increase as the shear wave speed increased. Takahashi et al.31 reported a correlation between liver elasticity evaluated by ultrasound elastography and histological liver fibrosis, which partly supports the results of the current study.

The 3D size of the IPFP was previously evaluated by measuring its maximum area on a midsagittal MRI image32. In this study, the IPFP volume shown on a 3D-reconstructed image was used for a more accurate assessment of the IPFP size. Furthermore, the IPFP size was evaluated as volume/weight based on the correlation between the IPFP volume and body weight33. A smaller IPFP is associated with the radiographic progression of FT-OA. Several studies have compared the IPFP volume of patients with KOA to healthy controls34,35. Fontanella et al. reported a decrease in the IPFP volume as KOA progressed, though the differences in IPFP volume were not significant35. The differences between the results of the current study and previous findings35 may be owing to the relatively larger number of patients, weight adjustment for the IPFP volume, and high proportion of end-stage OA observed in the current study. The IPFP was found to be stiffer as radiographic progression of FT-OA was observed. Favero et al. reported that the thickness of the interlobular septa and fibrosis in the adjacent synovial membrane were higher in patients with KOA than in healthy controls15. The shear wave speed was estimated to reflect these IPFP changes; therefore, the results of the current study are consistent with previous findings.

MRI provides semiquantitative data regarding the IPFP in studies regarding knee pain. The inconsistent findings of previous studies36,37,38 may be due to differences in pain measurements or KOA severity. Hence, anterior knee pain associated with the IPFP was evaluated in this study. Subjective anterior knee pain evaluations were supported by quantitative PPT data. IPFP fibrosis is common in patients with KOA, suggesting that IPFP fibrosis plays an essential role in OA pathophysiology. Quantitative assessments of IPFP fibrosis via MRI are difficult. Therefore, SWE was used in this study. The combination of ultrasound elastography and MRI provides a multifaceted evaluation of the IPFP.

Patients with KOA and anterior knee pain have stiffer IPFPs in this study; however, anterior knee pain is not correlated with IPFP size. Onuma et al. reported that irreversible IPFP fibrosis resulted in sprouting of calcitonin gene-related peptide-positive nerve fibers and prolonged pain in rats20. Similarly, the results of the current study suggest that IPFP inflammation induces changes in the nerve endings, and concomitant fibrosis was inferred by SWE, resulting in a significant association between the shear wave speed and anterior knee pain. Furthermore, the surface of the IPFP adjacent to the synovium had severe fibrosis in several specimens, synovitis can be a primary cause of IPFP fibrosis. Therefore, the IPFP evaluation using ultrasound elastography can help diagnose painful IPFP in patients with KOA. A recent study that included contrast-enhanced MRI demonstrated that the increased IPFP synovium volume was significantly associated with knee pain39. Ultrasound is more convenient and cost-effective than MRI and is also noninvasive. The results of the current study will contribute to the development of treatments for painful IPFP in patients with KOA.

This study has several limitations. First, the cross-sectional nature of this study precludes any inference regarding the causality between IPFP characteristics and anterior knee pain. Nevertheless, a significant association between IPFP elasticity and anterior knee pain was identified in this study, suggesting the possibility that the causality be resolved in the future. Second, many patients had a KL of grade 3 or 4; therefore, these results may not apply to patients with early-stage KOA. Third, the effects of other disorders and treatments that may affect knee pain, such as spinal disorders and painkillers, cannot be ruled out. Fourth, other potential pain factors, such as sensitization and psychosocial factors, were not assessed in this study. Fifth, the severity of knee pain was not assessed. It is likely that stiffer IPFPs are found in severely painful knees. In addition, the same MRI model was not used to evaluate all of the patients. However, no significant differences in patient sex, age, height, or weight were noted between the groups evaluated using each MRI system. Last, a histological evaluation was not performed for all patients. Despite these limitations, this study provides useful insights regarding the IPFP elasticity and anterior knee pain in patients with KOA.

In conclusion, patients with KOA and anterior knee pain have stiffer IPFPs regardless of FT-OA grade, indicating that ultrasound elastography is useful for the diagnosis of painful IPFP in patients with KOA.