Morphological characteristics of the infrapatellar fat pad

The relationship between the morphological characteristics of the infrapatellar fat pad (IFP) and joint deformity has yet to be fully elucidated. Therefore, the purpose of this study was to clarify the morphological characteristics of the IFP and to identify the relationships between morphological characteristics of the IFP and degenerative grade of the articular surface of the patella. This investigation examined 41 legs from 25 Japanese cadavers. The IFP length, width, and volume were measured. It was categorized into three types: Type I, IFP proximal located on medial and lateral sides of the patella; Type II, the IFP proximal only located medially; and Type III, absence of the IFP proximal. Articular surfaces were graded as macroscopically intact or mildly altered (Grade I), moderately (Grade II), or severely (Grade III). Grade III was significantly more frequent than Grades I or II in Type III. IFP volume was significantly larger in Type I than in Types II or III. A negative correlation was found between the degenerative grade of the articular surface of the patella and IFP volume. It was suggested that a relationship between the degenerative grade of the articular surface of the patella and the IFP volume.

Relationship between degenerative grade of the articular surface of the patella and the classification of the IFP. In Type I, Grade I was seen in 9 knees (22%), Grade II in 1 knee (3%), and Grade III in 5 knees (12%). In Type II, Grade I was seen in 7 knees (17%), Grade II in 10 knees (24%), and Grade III in 3 knees (7%). In Type III, Grade I was seen in 1 knee (3%), Grade II in 0 knees (0%), and Grade III in 5 knees (12%). There was significant difference degenerative grade of the articular surface and the classification of the IFP (p = 0.003). In Type I, residual analysis indicated there were significantly lower in Grade II than in Grades I and III (adjusted residual − 2.2). In Type II, residual analysis indicated there were significantly higher in Grade II than in Grades I and III (adjusted residual 3.3). In Type III, residual analysis indicated there were significantly higher in Grade III than in Grades I and II (adjusted residual 2.9) ( Table 3).
The correlation of between degenerative grade of the articular surface of the patella and IFP volume. Negative correlation was found between degenerative grade of the articular surface of the patella and IFP volume (r = − 0.314, p = 0.045) (Fig. 1).

Discussion
This study intended to clarify the morphological characteristics of IFP (shape, length, width, volume) using Japanese cadavers and the relationship between morphological characteristics of IFP (shape, volume) and degenerative grade of the articular surface of the patella. To the best of our knowledge, no detailed anatomical studies of the relationship between morphological characteristics of the IFP and degeneration of the articular surface of the patella have been reported previously.
In this study, Type I was seen in 15 knees (37%), Type II in 20 knees (49%), and Type III in 6 knees (14%). The IFP superior medial extension was seen in 35 knees (86%), IFP superior lateral extension in 15 knees (37%),  www.nature.com/scientificreports/ and loop type in 1 knee (3%). In previous studies, the existence of the IFP proximal located on the medial and lateral sides of the patella has been reported 3,20 . In a study using 36 knees from fresh cadavers 20 , IFP superior medial extension was seen in 100%, IFP superior lateral extension in 83%, and loop type in 11%. In a study using 43 knees from formalin-fixed bodies 3 , IFP superior medial extension was seen in 81%, IFP superior lateral extension in 65%, and loop type in 9.3%. The result was that the frequencies of IFP superior lateral extension and loop type were low. The reason for this might be due to differences in specimens (ethnic background, cadaver fixation method) and methods of dissection. In this study, Grade III was significantly more frequent than Grades I or II in Type III. And IFP volume was significantly larger in Type I than in Types II or III. In addition, weak negative correlation was found between degenerative grade of the articular surface of the patella and IFP volume. In previous studies, Cowan et al., 27 reported that individuals with PFJ OA had a larger IFP than non-OA controls, and IFP volume was directly related to PFJ OA pain. In contrast, Fontanella et al., 17 reported that a progressive decrease of IFP depth, femoral and tibial length and an increase of hypointense signal were observed in MRI of moderate and end-stage OA compared to non-osteoarthritic controls suggesting an important role of inflammation derived fibrosis and dimensions in OA pathology. And other data have suggested that a protective role of the IFP in knee OA, with a larger IFP being associated with reduced development and progression of knee pain and cartilage volume loss over 28,29 . In addition, Steidle-klok et al., 30 reported that Independent of the ambiguous role of the IFP in knee osteoarthritis, the size of the IFP does not appear to be related to knee pain. Interestingly, different findings on the relationship between IFP volume and degeneration of the PFJ have been reported, although the subjects, analysis methods, and study designs differ. In this study, it was suggested that a relationship to the degenerative grade of the articular surface of the patella and the IFP volume. This appears a quite interesting observation that must be deepen in the future. Some limitations should be considered when interpreting the findings from this investigation. First, there are evaluation items. This study doesn't evaluate the femoral trochlea surface. In addition, degeneration of the articular surface of the patella often be accompanied by osteophytes 31 , which would seem to affect the proximal IFP. But this study doesn't evaluate osteophyte formation. And patella baja has been reported to cause patellar ligament shortening, fibrosis of the Hoffa's fat pad, and other complications 31 . But this study doesn't evaluate the patellar alignment. In the future, there is a need to investigate to correlate the IFP volume with both degeneration of joint surface (the patellar surface and the femoral trochlea surface) and osteophyte formation in the patellofemoral joint. Furthermore, an in vivo prospective study using ultrasound or MRI examination is needed including patellar alignment. Second, this study was to clarify the morphological characteristics of IFP in Japanese cadavers, but bilateral and unilateral knees are mixed. This study does not adequately reflect the Japanese knees. In addition, as all cadavers were from Japanese individuals, whether the present findings apply to individual from other ethnicities is unclear. Further studies are required to evaluate variations according to ethnic origin. Measurement conditions. In the dissection procedure, an isolated knee specimen was prepared by cutting the distal one-third of the femur and the central part of the tibia, and the skin, subcutaneous tissue, and tensor fasciae lateral were removed. For the IFP body, the synovium and meniscus were incised along the anterior edges of the medial collateral ligament (MCL) and lateral collateral ligament (LCL) ( Fig. 2A,B). Then, in a position of knee flexion, the quadriceps femoris and patella were inverted from proximally to distally, and the ligamentum mucosum was incised (Fig. 2C). The IFP was detached from the anterior surface of the tibia, and the patellar tendon was resected with the patellar tendon attached to the patella and part of the rough surface of the tibia. The meniscus and transverse knee ligament attached to the IFP were then carefully removed. The presence or absence of the IFP proximal was confirmed between the insertional tendons of the vastus medialis/vastus lateralis, the patellar retinaculum, and the synovium (Fig. 3). Regarding the removal of the synovium, only synovium that was not adherent to the IFP proximal and IFP body was carefully removed (Fig. 3A). Quadriceps muscle fiber bundles were removed along the shape of the IFP proximal (Fig. 3B). Part of the tibial tuberosity was carefully removed from the posterior part to remove the most distal end of the IFP body. After measuring the IFP superior medial extension and superior lateral extension and IFP body, the IFP was removed from the patella, patellar tendon, and patellar retinaculum for IFP volume measurement. With reference to previous studies 11 , these values weren't corrected by body size. Morphological measurement was performed by one examiner, and the IFP superior medial extension and superior lateral extension, IFP body length, width, and IFP volume were measured (Fig. 4). The specimen was photographed from behind using a digital camera (Finepix F600EXR; Fujifilm, Tokyo, Japan) with the patella joint surface facing forward without tilting. At the time of measurement, perpendicular lines were drawn in the following parts: the upper end of the patellar articular surface; the most end of the IFP superior medial extension and superior lateral extension, the lower end of the patellar articular surface, the horizon passing through the distal end of the IFP body, and the medial and lateral ends of the IFP body. IFP superior medial extension and superior lateral extension length was measured as the distance between the horizontal line passing through the upper end of the patellar joint surface and the horizontal line passing through the lower end of the patellar joint surface when the IFP was connected to the suprapatellar fat pad (SFP). When the IFP did not bind to the SFP, this distance was taken as the distance between the horizontal line passing through the most end of the IFP proximal and the horizontal line passing through the lower end of the patellar articular surface (Fig. 4A). IFP superior medial extension and superior lateral extension width was measured at the distance of the horizontal line connecting the medial and lateral ends of the IFP superior medial extension and superior lateral extension length and was measured at three points of the IFP superior medial extension and superior lateral extension length (Fig. 4B). IFP body length was measured as the distance between the horizontal line passing through the lower end of the patellar joint surface and the horizontal line passing through the most distal end of the IFP body. The IFP body width is the distance of the line connecting the perpendicular lines passing through the inner and outer ends of the IFP body. Image analysis software (Image J; NIH, Bethesda, MD, USA) was used for the measurement, and the mean ± standard deviation from three measurements was calculated. Before measuring volume, the IFP was soaked in a preservation solution (water: 78%, alcohol: 16%, glycerin: 4%, phenol: 2%) for 10 min to prevent the IFP from drying out. For the IFP volume, a 500-ml cyl- www.nature.com/scientificreports/ inder (SANPLATEC, Osaka, Japan) was used, and water displacement was recorded in 2.5-ml intervals, and the mean ± standard deviation from three measurements was calculated. When the IFP superior medial extension and SFP were connected, the SFP was removed at the upper end of the patellar joint surface. To prevent a decrease in IFP volume due to drying of the IFP, specimens were immersed in the preservation solution at 30-min intervals for 10 min, and the entire experiment was performed within 3 h. Knees were classified into 3 types by the morphological characteristics of IFP superior medial extension and superior lateral extension. Type I was an IFP located on the medial side (superior medial extension) and lateral side (superior lateral extension) of the patella (Fig. 5A,a). Type II was an IFP located only on the medial side (superior medial extension) of the patella (Fig. 5B,b). Type III was absence of the IFP on both superior medial extension and superior lateral extension (Fig. 5C,c).

Statistical analysis.
The Shapiro-Wilk test was used to assess normality of the different variables. Fisher's exact test was used for comparisons of classifications of the IFP by sex and laterality, and to compare differences in degenerative grade of the articular surface for each type of category. Residual analysis was used for the post hoc test. Comparisons of IFP superior medial extension length, width (distal, intermediate, proximal) between Types I and II were made using unpaired t-tests. Comparisons of IFP body length, width, and IFP volume in each type were made with one-way repeated-measures analysis of variance, and Tukey's method. Comparisons of IFP superior medial extension and superior lateral extension length, width (distal, intermediate, proximal) among Type I were made with paired t-tests. Comparisons between degenerative grade of the articular surface of

Perspective
In this study, we clarified the morphological characteristics of the IFP using Japanese cadavers and the relationship between morphological characteristics of the IFP and degenerative grade of the articular surface of the patella. The IFP superior medial extension was seen in 35 knees (86%), IFP superior lateral extension in 15 knees (37%), and loop type in 1 knee (3%). It was suggested that a relationship between the degenerative grade of the articular surface of the patella and the IFP volume.  www.nature.com/scientificreports/

Data availability
The datasets generated and/or analysed during the current study are not publicly available due to limitations of ethical approval involving the patient data and anonymity but are available from the corresponding author on reasonable request.