Mavric based T2 mapping assessment of infrapatellar fat pad scarring in patients with total knee arthroplasty

Abstract

The infrapatellar fat pad (IPFP) has been implicated as a source of postoperative knee pain. Imaging the IPFP is challenging in patients with total knee arthroplasty (TKA) due to metallic susceptibility artifact. Multi-Acquisition Variable-Resonance Image Combination (MAVRIC)-Based T2 Mapping has been developed to mitigate this artifact and can generate quantitative T2 data. Objectives of this study were to (1) measure T2 values of the IPFP in patients with TKAs using a MAVRIC based T2 mapping technique and (2) determine if IPFP T2 values are related to the degree of fat pad scarring or clinical MRI findings. Twenty-eight subjects (10 M, 18 F, Age: 66 + 7.2 years [Mean ±SD]) undergoing clinical MRIs were sequentially recruited. Morphological imaging and quantitative T2 mapping sequences were performed on a clinical 1.5 Tesla scanner. The morphologic images were graded for the presence and severity of fat pad scarring and clinical outcomes. T2 values were calculated in the total fat pad volume, a normal ROI, and an abnormal ROI. T2 values were shortened in the total IPFP volume (p = 0.001) and within abnormal regions (p = 0.003) in subjects with more severe IPFP scarring. The difference between T2 values in normal – abnormal regions was greater in subjects with severe vs. no scarring (+1426.1%, p = 0.008). T2 values were elevated in patients with MRI findings of osteolysis (+32.3%, p = 0.02). These findings indicate that MAVRIC-based T2 Mapping may be used as a quantitative biomarker of post-operative IPFP scarring in individuals following TKA.

Graphical Abstract

MAVRIC-based T2 mapping was used to mitigate artifact and generate quantitative T2 values of the infrapatellar fat pad (IPFP) in individuals with total knee arthroplasty (TKA). T2 values were shortened in individuals with more severe IPFP scarring and elevated in individuals with MRI findings of osteolysis. These findings indicate that MAVRIC-based T2 Mapping may be used as a quantitative biomarker of post-operative IPFP scarring in individuals following TKA.

Introduction

The prevalence of total knee arthroplasty (TKA) in the United states is substantial, with over 600,000 TKAs performed annually ¹. TKA incidence is expected to increase considerably in the future, with estimates of a 401% increase in projected total annual US use for primary TKA by 2040 ². While post-operative patient satisfaction is generally high for TKA, studies have reported that 3–5% of TKAs fail each year ^3,4 and 7–20% of patients report function loss and pain ⁵. Further research is needed to elucidate anatomic and biological factors that may contribute to or help predict TKA failure.

The infrapatellar fat pad (IPFP), an adipose tissue structure located in the anterior compartment of the knee, has been implicated in the progression of knee osteoarthritis and as a potential source of postoperative knee pain due to its substantial innervation, high proportion of substance-p containing fibers, and proximity to the synovial lining ^6–9.

TKA surgery can lead to chronic inflammation of the IPFP that can develop into scar tissue over time ¹⁰. Fibrosis of the IPFP can present clinically as anterior knee pain, stiffness, range of motion loss, and functional limitations ^6,11–13.

Magnetic resonance imaging (MRI) is the most common imaging modality to evaluate IPFP pathology such as fibrosis and inflammation ⁶. Previous studies have demonstrated indicators of IPFP abnormalities based on MRI appearance ^14,15; however these were qualitative assessments and may be susceptible to reader bias and subjectivity ¹⁶. The use of quantitative MRI (qMRI), specifically T2 mapping, can provide objective and quantitative measures of the inherent magnetic properties of the tissue ¹⁷. T2 is the MRI transverse relaxation time constant that has been associated with tissue water content and overall organization ¹⁸. To date, T2 mapping has been used to evaluate the IPFP in subjects following arthroscopic surgery, where shortened T2 values were observed in post-arthroscopic knees as compared to contralateral unoperated knees ¹⁹. However, application of T2 mapping techniques to individuals with TKA is challenging as metallic susceptibility generates significant in-plane and through-plane image distortions ²⁰ when using traditional fast-spin-echo based acquisitions²¹. Nevertheless, newer three dimensional (3D) multi-spectral imaging (MSI) techniques, such as multi-acquisition variable-resonance image combination (MAVRIC), can mitigate these artifacts and allow for enhanced visualization of soft tissue structures surrounding metallic implants ²². Recently, MAVRIC has been modified to permit MAVRIC-based T2 mapping ²³ and its clinical feasibility has been assessed in individuals with total hip arthroplasty (THA) ²². However, this novel pulse sequence has yet to be applied to individuals with TKA.

While previous studies have examined qualitative indicators of IPFP inflammation and fibrosis ¹⁵, there are limited data on non-invasive quantitative assessments of post-surgical IPFP signal characteristics ^19,24. In addition, no quantitative analysis of the IPFP in knees with TKA is currently available, and the relationship between T2 values and magnitude of IPFP scarring is unknown. The absence of T2 mapping quantitative evaluations in knees with TKA along with the pain and functioning implications of the IPFP highlights the need for MAVRIC-based T2 analysis of the IPFP in individuals with TKA.

Therefore, the objectives of this study were: (1) evaluate T2 values of the IPFP in subjects with TKAs using a MAVRIC-based T2 mapping technique and (2) determine if T2 values relate to the degree of IPFP scarring. We hypothesized that shorter IPFP T2 values are associated with an increased severity of IPFP scarring.

Methods

Study Cohort

This was a case-control study (Level III evidence). This study received local Institutional Review Board approval and informed written consent was obtained from all subjects prior to participation. Initial power analyses to detect 10% differences in mean IPFP T2 values at 80% power and 5% level of significance indicated 26 knees were needed. 76 individuals with TKAs undergoing clinical MRIs were identified through an institutional database (Epic) and sequentially recruited. 34 of those individuals consented to participate in quantitative imaging. Individuals who had contraindications to MRI were excluded. Of the original 34 subjects, 4 individuals declined participation following consent and prior to quantitative imaging, and 2 were found to have resected fat pads and were thus removed from the study. In summary, 28 of the original 34 subjects (10 males, 18 females, age 66 ± 7 years) were included in the final evaluation. Experimental and control groups were selected based on the level of IPFP fibrosis. The control group was defined as individuals with no IPFP scarring, as identified on MRI images. Three experimental groups were defined as: mild IPFP scarring, moderate IPFP scarring, and severe scarring, as determined by a board-certified radiologist with extensive MRI reading experience.

MRI Protocol

MRIs were acquired using clinical 1.5 Tesla scanners (DVMR 450) with an 8-channel phased-array knee coil (GE Healthcare, Waukesha, WI). Imaging consisted of axial, coronal, and sagittal two dimensional (2D) fast-spin-echo (FSE) morphologic imaging, as well as 3D morphologic isotropic MAVRIC-SL imaging, and quantitative MAVRIC-based T2 mapping imaging (Table 1).

Table 1.

MRI protocol for scanning knee arthroplasty at 1.5T

	Imaging Series
Variable	2D FSE	3D Iso-MAVRIC	MAVRIC T2 MAPPING
FOV	Axial, Coronal Sagittal	Sagittal	Sagittal
In-plane Frequency Direction	Ax: S to I Cor: R to L Sag: A to P	A to P	S to I
Receiver Bandwidth (kHz)	±125	±125	±125
Flip Angle (°)	90	110	100
Repetition Time (ms)	4505	3500	3500
Echo Time (ms)	27	8.8	TE1: 9.8; TE2: 90
Echo Train length	16	48	40
Acquisition Matrix (mm)	512 x (256–320)	224 x 224	256 x 192
Field of View (cm)	20	20–22	21–22
Slice Thickness (mm)	3–3.5	1.2	3.5

Morphological and Quantitative Imaging Analyses

A board-certified radiologist with extensive experience of reading MRI of arthroplasty evaluated the morphologic images for the presence of fat pad fibrosis, assigned a global fibrosis severity as: none, mild, moderate or severe, and manually segmented three regions of interest (ROIs) from the source T2 mapping images for quantitative analysis (Figure 1) ²⁵. Regions were defined as: 1) The full volume of the IPFP; 2) a circular ROI (3.4 mm³) within normal tissue in a mid-sagittal image, defined as a region of healthy tissue displaying no scarring; and 3) a circular ROI (3.4 mm³) within abnormal tissue in a mid-sagittal image, identified by hypointense regions of the IPFP.

Figure 1.

Sagittal MAVRIC-SL MRI of a patient with TKA displaying ROIs evaluated in the study: A) full IPFP volume (red outline), B) Regin of normal IPFP tissue (blue circle), C) Region of abnormal IPFP tissue (green circle).

MRI reports were acquired through a hospital-wide database (Epic) and picture achieving and communication system (PACS) imaging systems. The presence [Y/N] of the following clinical outcomes were obtained from the reports: osteolysis, synovitis, joint effusion, marrow edema, and fibrous membrane formation. These findings were re-confirmed by the radiologist who assigned fibrosis severity.

T2 values were calculated on a pixel-by-pixel basis using a mono-exponential fitting (equation 1) of signal intensity (SI) to echo time (TE) using custom MATLAB software (MATLAB, Mathworks, Natick, MA).

$$T_2=\frac{T{E}_2-T{E}_1}{\ln (S{I}_1)-\ln (S{I}_2)}$$ Equation 1

where TE₁ and TE₂ are the times and SI₁ and SI₂ are the corresponding signal intensities at each respective echo time. Values for statistical analysis were capped at 250 ms to minimize measurement error ²³, as has been previously performed ²².

Statistical Methods

Continuous variables were reported as means and standard deviations (SD) or median and interquartile range (IQR) depending on distribution. Normality of data was tested by the Anderson-Darling test. One-way analysis of variance (ANOVA) tests with Tukey adjusted post-hoc analyses were performed to evaluate the following relationships: (1) full volume IPFP T2 values (mean, median, and SD) by IPFP scarring severity, (2) normal ROI and abnormal ROI T2 values (mean, median, and SD) by IPFP scarring severity, and (3) within subject IPFP T2 differences (mean normal ROI – mean abnormal ROI) by IPFP scarring severity. Non-parametric ANOVAs (Kruskal-Wallis) were used for any data that did not fit the criteria of normality. Paired t-tests were performed to compare within subject T2 values of normal vs. abnormal regions. Categorical variables were presented as frequencies and percentages. Dichotomous variables (Y/N) were created for MRI findings (osteolysis, synovitis, joint effusion, marrow edema, fibrous membrane formation, revision status post MRI) and Tukey adjusted t-tests were performed to compare T2 values between affected and non-affected groups. Linear mixed models were performed to evaluate T2 values with length of implantation (defined as years between primary TKA surgery and imaging). Analyses controlled for age and sex. Statistical measures were then re-evaluated separately by sex. In all cases, statistically significant differences were considered at p < 0.05. Statistical analysis was performed using MATLAB (ver. 2020b Mathworks, Natick, MA) and JMP (Pro 14, SAS Institute).

Results

Study Cohort

A total of 28 subjects with unilateral TKA (10 males, 18 females, age 66 ± 7 years, mean ± SD) were enrolled (Table 2). Six subjects had revision TKA surgery prior to imaging. Twelve subjects underwent revision surgeries scheduled for after the MRI. Average time from most recent arthroplasty to imaging was 4.9 ± 3.5 years. Six subjects had MRI findings of osteolysis and 10 subjects displayed the presence of synovitis. For two subjects, abnormal regions were not able to be located within the fat pad and therefore measurements of abnormal ROI T2 values were not obtained.

Table 2.

Demographic and clinical information for the full cohort and organized by IPFP scarring severity

Full Cohort
N (% female)	28 (64%)
Age (years)	66 ± 7
Revision prior to imaging, N (%)	5 (18%)
Revision after imaging, N (%)	12 (43%)
Length of Implantation (years)	4.9 ± 3.5
Osteolysis, N (%)	6 (21)
Synovitis, N (%)	10 (36)
Joint effusion, N (%)	4 (14)
Marrow edema, N (%)	3 (11)
Fibrous membrane formation, N (%)	3 (11)
Distribution of Outcomes by IPFP scarring severity
Variable	None	Mild	Moderate	Severe
N (% Female)	3 (67)	14 (79)	4 (50)	7 (43)
Age (years)	65 ± 8	67 ± 6	66 ± 8	67 ± 8
Revision prior to imaging, N (%)	1 (33)	2 (14)	1 (25)	1 (14)
Revision after imaging, N (%)	2 (67)	6 (43)	2 (50)	2 (29)
Length of Implantation (years)	7.6 ± 1.0	4.2 ± 3.0	6.4 ± 1.0	4.4 ± 4.0
Osteolysis, N (%)	2 (67)	2 (14)	0 (0)	2 (29)
Synovitis, N (%)	2 (67)	5 (36)	1 (25)	2 (29)
Joint effusion, N (%)	0 (0)	2 (14)	0 (0)	2 (29)
Marrow edema, N (%)	0 (0)	1 (7)	1 (25)	1 (14)
Fibrous membrane formation, N (%)	0 (0)	2 (14)	0 (0)	1 (14)

Age and Length of Implantation expressed as mean ± SD

Quantitative Analysis

Between subject comparisons

Total IPFP Volume

A significant difference of mean T2 values of the total IPFP volume was found between the different IPFP scarring classes (p = 0.001). Specifically, mean T2 values of the total IPFP volume were shorter in subjects with severe scarring as compared to those with moderate (−19.9% p = 0.02), mild (−22.6%, p < 0.001), and no scarring (−20.2%, p = 0.03, Table 3, Figure 2A, 2B). Similarly, significant differences in median T2 values were found between the different IPFP scarring severity (p = 0.001). Median T2 values of the total IPFP volume were found to be shorter in subjects with severe scarring vs. moderate (−22.2%, p = 0.02) and mild scarring (−24.2%, p < 0.001, Table 3 and Figure S1).

Table 3.

Mean and median T2 values for the total IPFP volume, abnormal regions, and normal regions for each level of IPFP scarring severity

Variable		None	Mild	Moderate	Severe
IPFP Total Volume T2 (ms)	Mean	111.6 ± 6.3	115.1 ± 3.3	111.2 ± 5.5	89.1 ± 4.2*
	Median	110.1 ± 7.1	114.1 ± 3.6	111.3 ± 6.1	86.5 ± 4.6**
IPFP Abnormal ROI T2 (ms)	Mean	140.7 ± 12.6+	101.3 ± 5.2++	85.9 ± 8.5	72.1 ± 6.5
	Median	122.9 ± 10.6+++	97.3 ± 5.3++	83. ± 9.1	71.7 ± 6.9
IPFP Normal ROI T2 (ms)	Mean	144.1 ± 12.0	136.2 ± 4.9	129.0 ± 8.1	125.6 ± 6.1
	Median	129.8 ± 10.7	130.7 ± 5.4	128.9 ± 9.2	126.8 ± 7.0

Notes:

^*= Mean T2 of IPFP Total Volume was significantly shorter than all other mean T2 values of differing scarring severity;

^**= Median T2 of IPFP Total Volume was significantly shorter than median T2 values of moderate and mild scarring

⁺= Mean T2 of abnormal ROIs was significantly longer than all other abnormal ROI mean T2 values of differing scarring severity

⁺⁺= Mean and Median T2 of abnormal ROIs from individuals with mild scarring was significantly longer than abnormal ROI mean T2 values of severe scarring

⁺⁺⁺= Median T2 of abnormal ROIs from individuals with no scarring was significantly longer than abnormal ROI mean T2 values of individuals with moderate and severe scarring

Figure 2.

A) Representative T2 maps of the IPFP in four individuals with TKA (76 yr. old female with no scarring, 57 yr. old female with mild scarring, 57 yr. old female with moderate scarring, and 62 yr. old male with severe scarring), displaying shortening of T2 with increasing levels of IPFP scarring. B) Box plot of mean T2 values of IPFP full volume regions vs. IPFP fat pad scarring. T2 values of abnormal regions were shorter in individuals with severe scarring compared to those with moderate, mild, and no scarring. For each box, the central mark indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points not considered outliers, and the outliers are plotted individually using the red ‘+’ marker symbol

Abnormal and normal ROIs

A significant difference in mean T2 values of abnormal IPFP ROIs was found between the different IPFP scarring severity classes (p = 0.003). Mean T2 values from abnormal ROIs were (1) shorter in subjects with severe scarring compared to subjects with mild scarring (−28.9%, p = 0.01) and no scarring (−48.8%, p < 0.001) (2) shorter in subjects with moderate scarring vs. no scarring (−39.0%, p = 0.009), and (3) shorter in mild vs. no scarring (−28.0%, p = 0.03) (Table 3 and Figure 3A). Similarly, a significant difference in median T2 values was also found by IPFP scarring severity (p = 0.013). Specifically, median T2 values from abnormal ROIs were shorter in subjects with severe scarring vs. no scarring (−41.6%, p = 0.003), severe vs. mild scarring (−26.3%, p = 0.04), and moderate vs. no scarring (−32.4%, p = 0.04) (Table 3 and Figure S2). No significant differences in mean or median T2 values were found among normal ROIs between groups (mean normal T2: p= 0.58; median normal T2: p = 0.84) (Table 3 and Figure 3B).

Figure 3.

A) Box plot of mean T2 values of IPFP abnormal regions vs. IPFP fat pad scarring. T2 values of abnormal regions were shorter in individuals with moderate or severe scarring compared to those with mild or no scarring. For each box, the central mark indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points not considered outliers, and the outliers are plotted individually using the red ‘+’ marker symbol. B) Box plot of mean T2 values of IPFP normal regions vs. IPFP fat pad scarring. No differences were found between scarring groups.

Within subject comparisons

Within subject differences between mean T2 values from normal ROIs and abnormal ROIs differed significantly by fat pad scarring class (p = 0.04). The mean difference between normal ROIs and abnormal mean T2 values was greater in subjects with severe scarring vs. no scarring (+1426.1%, p = 0.008) (Figure 4). Further, mean T2 values were shorter in abnormal ROIs vs. normal ROIs within subjects (−30.4%, p < 0.001). T2 value standard deviations in the total IPFP volume, abnormal, and normal regions did not differ with fat pad scarring.

Figure 4.

Box plot of the difference between normal and abnormal T2 values within subjects vs. severity of IPFP fat pad scarring. The difference was greater in individuals with severe scarring compared to those with no scarring. For each box, the central mark indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points not considered outliers.

Association with Morphologic Features

Mean T2 values from abnormal ROIs were greater in subjects who exhibited osteolysis on morphologic images (+32.3%, p = 0.02) (Figure 5). No differences in T2 values were detected based on the presence or absence of synovitis, joint effusion, marrow edema, fibrous membrane formation, or revision status. Further, length of implantation was not a significant predictor of T2 values.

Figure 5.

Box plot of mean T2 values of IPFP abnormal regions vs. osteolysis. T2 values of abnormal regions were shorter in individuals with MRI findings of osteolysis. For each box, the central mark indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points not considered outliers.

Subsex analysis:

Significant differences in total volume and abnormal ROI IPFP T2 values were found based on fat pad scarring severity, (total volume mean T2: p = 0.009, abnormal mean T2: p = 0.03), but only within female subjects (Figure 6). Specifically, females with severe scarring had shorter total IPFP volume mean T2 values than females with mild (−24.0%, p = 0.004) and no scarring (−26.6%, p = 0.01) (Figure 6A). Similarly, differences of total volume median T2 values were shorter in females with severe scarring vs. females with mild (−25.8%, p = 0.003) and no scarring (−30.3%, p = 0.007). Further, abnormal ROI mean T2 values were shorter in females with severe scarring compared to females with no scarring (−46.7%, p = 0.016), and shorter in females with moderate vs. no scarring (−44.2%, p = 0.04) (Figure 6C). Similarly, abnormal ROI median T2 values were shorter in females with severe scarring as compared to no scarring (−47.4%, p = 0.02). No differences were found in male subjects (Figure 6B, 6D). In both females and males, mean T2 values were shorter in abnormal ROIs vs. normal ROIs within subjects (females: −30.4%, p < 0.001, males: −30.4%, p < 0.001). Similarly, median T2 values were shorter in abnormal ROIs vs. normal ROIs within subjects (females: −29.9%, p < 0.001, males: −29.9%, p < 0.001).

Figure 6.

Box plot of mean total IPFP volume T2 values and mean abnormal region T2 values vs. IPFP fat pad scarring separated by sex. For each box, the central mark indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points not considered outliers, and the outliers are plotted individually using the red ‘+’ marker symbol.

Differences in T2 values from abnormal ROIs with osteolysis were also found in females but not in males. Abnormal ROI T2 values were longer in females with osteolysis compared to females without osteolysis (−61.0%, p < 0.001). No significant differences were detected for the clinical outcomes of synovitis, joint effusion, marrow edema, fibrous membrane formation, revision status, or time from surgery to imaging.

Discussion

In this study, a MAVRIC-based T2 mapping technique was used to evaluate T2 values of the IPFP in individuals post TKA. T2 values of the full IPFP volume and in abnormal ROIs were shorter in individuals with more severe IPFP scarring. In addition, the difference between T2 values in normal vs. abnormal regions was greater in subjects with more severe IPFP scarring. Further, T2 values were prolonged in subjects with MRI evidence of osteolysis. Previous MRI studies have been qualitative in nature and have described the appearance of IPFP fibrosis as linear regions of low signal intensity ^14,26, and identified indicators of various IPFP abnormalities such as Hoffa disease, post-surgical fibrosis, intracapsular chondroma, synovitis, and shear injury ²⁷. In addition, the nature of prior quantitative studies of the IPFP have used histologically measured IPFP tissue architecture, size, and composition from individuals post TKA ^28,29 or ACL surgery ⁸ and calculated IPFP T2 values in post-arthroscopic knees ¹⁹. However, prior quantitative studies were based on morphologic imaging alone to assess IPFP volume or required invasive tissue sampling and few studies have quantitatively evaluated the tissue using non-invasive imaging. Our study builds upon previous findings by quantifying IPFP T2 values in patients with TKA using a novel MAVRIC-based T2 mapping technique which circumvents the prior limitation of metal susceptibility artifact.

Shortening of mean T2 values in the total IPFP volume as well as in abnormal ROIs was observed with increased IPFP scarring severity. Similar findings were reflected by median T2 values emphasize that this reduction in T2 is maintained even as the effect of outliers is removed. Our results are consistent with a previous examination of IPFP T2 values which found that decreased mean T2 values were associated with fibrosis of the IPFP in knees with prior arthroscopic surgery as compared to control knees ¹⁹. Reduced T2 values may be explained by increased tissue organization and collagen content due to scar formation with lower tissue water content ³⁰. Histological findings confirm the presence of increased tissue collagen representing fibrosis in the IPFP in post-operative knees (8,27). Dense tissue structures lead to increased proton spin-spin interaction, more rapid dephasing of proton spins, thus shorter T2 characteristics ³¹. These results align with what one would expect with a pixel intensity evaluation. On a T2-weighted MRI, areas of fibrosis will display lower pixel intensity. The novelty and value of using a multi-spectral T2 mapping technique is that T2 values represent inherent magnetic properties of the tissue whereas pixel values, representing signal intensity, vary depending on the sequence acquisition parameters, coil sensitivity profiles, magnetic field inhomogeneities, and radio-frequency amplification settings ³². The results of this study confirm MAVRIC-based T2 mapping as an effective technique to quantify fibrosis in the IPFP.

Notably, the greater magnitude of T2 differences between abnormal regions vs. normal regions within individuals with severe scarring, compared to those with normal tissue may indicate increased tissue heterogeneity in the IPFP with increased scarring severity. Previous studies have observed indicators of increased tissue heterogeneity in post-operative and disease degraded knees ^33,34. A greater increase in the standard deviation of IPFP 3D MRI signal intensity, i.e. greater heterogeneity, was observed in progressor vs. non-progressor knees and in osteoarthritic (OA) vs. healthy knees ³³. Further, in a cohort study of 874 subjects, an increased IPFP MRI signal SD was associated with worse knee pain and greater cartilage loss after two years of follow-up ³⁴. However, the SD of the T2 values did not differ among groups in the current study. The observation of shorter T2 values in abnormal regions with increased scarring severity despite no differences between normal regions may indicate that IPFP scarring is more severe in localized regions within the fat pad.

Subjects with MRI findings of osteolysis had increased T2 values in abnormal ROIs. Elevated T2 values may be due to inflammatory reactions associated with periprosthetic osteolysis. Wear debris from total joint replacements can lead to inflammatory conditions including increased NF-kB signaling and cytokine expression, which display as increase water content within imaged tissues and have a concomitant increase in T2 values. This in turn leads to hyperactive osteoclast activity and subsequent bone breakdown ^35,36. Increased T2 values have been observed in the infrapatellar fat pad and other adipose tissues with increased inflammation.^37,38 A higher degree of IPFP abnormalities was associated with higher T2 values and inflammatory markers in the IPFP of individuals with ACL injuries ³⁷. We note that T2 prolongation due to edema or inflammation can compete with T2 shortening due to scarring complicating the clinical interpretation of T2 values. Further, a study of rodent anterior cruciate ligament transection (ACLX) found T2* values of the IPFP increased during disease progression after ACLX, which was attributed to edematous change during the inflammatory process ³⁸.

Subsex analyses revealed that differences in T2 values with fibrosis and osteolysis persisted for females but not males. Prior work with murine models has indicated that females may be more predisposed to inflammatory responses to wear debris from implants ^39,40. Female mice displayed the highest rate and severity of metal sensitivity lymphocyte responses after induction of metal delayed type hypersensitivity (metal-DTH) ³⁹. Further, in an evaluation of material dependent responses to implant wear, female mice with cobalt-chromium-molybdenum (CoCrMo) debris displayed significantly more inflammatory osteolysis than males and females with implant grade conventional polyethylene (UHMWPE) ⁴⁰. However, it should be noted that the sex-specific findings within the current study may be due to limited enrollment of males compared to females.

Measures of T2 values did not relate to revision status or time from initial surgery to imaging (length of implantation). Indeed, these clinical outcomes are multifactorial and are affected by multiple anatomic regions not limited to the IPFP. In some cases, severe fibrosis was observed in anatomic regions of the knee but not in the IPFP (Figure 7).

Figure 7.

Representative image slice from the knee of a 63-year-old female. Fibrosis in the IPFP (red outline) was moderate but severe fibrosis was displayed in knee capsule (green outline). Differentiating the joint capsule from the fat pad was based on location. The capsule was located just anterior to the distal femoral and proximal tibial components and posterior to the fat pad.

There were some limitations to this study. First, tissue samples of the IPFP were not collected, thus histological confirmation of fibrosis was not performed. Second, we did not have access to patient-reported outcomes regarding pain and functioning, so direct relationships to clinical symptoms could not be established. Third, the limited sample size and enrollment of males may have obscured some findings within our sub-sex analysis. However, to the authors knowledge this study is the first to perform quantitative MAVRIC-based T2 mapping in individuals with total knee arthroplasty and demonstrates the feasibility of using T2 mapping as a quantitative biomarker of post-TKA fibrosis and inflammation. Future work is required to confirm relationships with histology and patient-reported outcomes and to further examine sex-based differences in T2 values of the IPFP with clinical outcomes. Finally, only 2 echo times, 9.8 and 90 ms, were used for T2 calculation and it is possible that these TEs may be too short for to characterize long T2 species within soft tissue structures. A prior study found that capping values at 250 ms would minimize error in the resulting summary statistics ²³, which was utilized in a prior study ²², as well as the current study. We found that 14 of our 28 subjects had voxels with T2 > 250 ms present, in which these voxels represented only a small percentage (4.4 ± 5.4%) of the IPFP volume.

In summary, this study used quantitative MAVRIC-based T2 mapping to evaluate the IPFP within individuals with TKA and found that T2 values were shorter with increased scarring severity and longer with MRI findings of osteolysis. This work provides evidence that MAVRIC-based T2 Mapping may be used as a quantitative biomarker of post-operative IPFP scarring in individuals following TKA. Future work may involve a retrieval analysis for direct histologic confirmation of specific tissue structure alterations and degradation indicated by T2 values. Longitudinal monitoring of IPFP T2 values may provide a means to assess the development and progression of IPFP scarring and osteolysis. Osteolysis can lead to massive bone loss and subsequent implant failure, resulting in extensive and expensive revision procedures ⁴¹. T2 mapping may also be further applied in individuals with TKA to evaluate the severity of synovial reactions ²². The potential use of T2 mapping as a tool to monitor the onset and progression of pathology may help to predict and/or prevent implant failure.