Effects of manual therapy on the infrapatellar fat pad: a randomized controlled trial

Abstract

[Purpose] This study aimed to elucidate the immediate effects of manual therapy on the infrapatellar fat pad. [Participants and Methods] Forty healthy university students were randomly assigned to either the manual therapy group or control group. The manual therapy group received a 3-min manual technique synchronized with knee extension, whereas the control group rested for 5 min. The primary outcome was the gliding velocity of the infrapatellar fat pad, which was assessed using ultrasonography and particle image velocimetry. The secondary outcome measure was the tibial angle of the patellar tendon. [Results] Group difference were observed in the deep-layer gliding velocity of the infrapatellar fat pad; however, consistent immediate changes were not observed following the intervention. No discernible alterations were identified in the superficial layer or in the patellar tendon tibial angle. [Conclusion] These findings indicate that the immediate effects of manual therapy on the infrapatellar fat pad are limited. Further studies with modified intervention conditions and clinical populations are required to clarify their clinical significance.

INTRODUCTION

The infrapatellar fat pad (IFP) is defined as the adipose tissue occupying the space formed by the inferior border of the patella, femoral condyles, tibial plateau, and patellar tendon at the anterior aspect of the knee joint¹⁾. Anatomically, it is located within the joint capsule but outside the synovium. This tissue is characterized by its low pain threshold, attributed to its dense innervation, including the presence of substance P^{2, 3)}. In addition, it has been reported that the IFP displays a two-layer structure composed of independent functional units, designated as the macro-chamber (superficial layer) and the micro-chamber (deep layer)⁴⁾. In light of these findings, the IFP has garnered attention not only as a structural support tissue or shock absorber but also as a potential source of pain arising from mechanical irritation associated with gliding disorders and fibrosis.

A plethora of studies have investigated conservative treatments, such as physical therapy, aimed at the improving muscle function and alignment. Additionally, taping has been studied as a means of reducing the load, and injections of local anesthetics or corticosteroids have been examined as treatments for musculoskeletal conditions. Surgical techniques considered for this procedure include arthroscopic partial meniscectomy, anterior compartment release, and adhesiolysis. However, these interventions vary significantly across studies in terms of patient populations, inclusion criteria, outcome measures, and follow-up periods. This variation has results in a marked lack of evidence, particularly evidence from randomized controlled trials. Consequently, there is no established gold standard guidelines for treatment selection, and clinical practice currently relies on case-by-case judgment based on empirical rules^{5, 6)}. In this context, manual therapy has emerged as a noninvasive treatment modality that offers several advantages, including reproducibility, minimal invasiveness, and the absence of any requirement for specialized equipment or environment. Moreover, it is regarded as a method that can directly address IFP dysfunction, such as enhancing the mobility of the joint capsule and surrounding soft tissues. To date, the effects of manual therapy on IFP dynamics and symptom improvement have not been sufficiently verified, and its efficacy and mechanism of action remain unclear. However, few studies have examined its immediate effects, although such information is crucial for conducting posttreatment evaluations, determining treatment plans, and deciding whether to continue the therapy. Conversely, manual therapy for the IFP is widely performed in clinical practice, and immediate pain relief and improvements in joint range of motion have been reported. These immediate effects may be attributable to improvements in inter-tissue gliding within the deep layer of the IFP, resulting in increased gliding velocity induced by manual therapy. Therefore, the immediate effects of a simple and reproducible manual therapy technique on the IFP gliding dynamics have not yet been sufficiently clarified using objective quantitative methods. Recent studies have also investigated the effects of manual therapy on the IFP, particularly in clinical populations, and have reported changes in IFP volume or mobility following intervention^{7, 8)}. However, many of these studies have employed highly specialized techniques or focused primarily on morphological changes and long-term outcomes. Accordingly, this interventional study aimed to evaluate the immediate effects of manual therapy on the IFP using objective ultrasound-based assessments and clarify their clinical significance. Furthermore, we objectively assessed the two-layer structure using particle image velocimetry (PIV), a method recently employed to analyze gliding speeds over the fat pad^{9, 10)}. This approach allowed us to evaluate changes in gliding properties as an indicator of this technique. Unlike static evaluations using magnetic resonance imaging (MRI), PIV offers the advantage of quantitatively assessing IFP gliding properties and dynamics by analyzing tissue displacement during joint motion with ultrasound imaging. Compared with conventional dynamic ultrasonographic motion analysis based on visual assessment or manual tracking, PIV provides objective quantitative measurements of tissue displacement and velocity with high reproducibility¹⁰⁾. This noninvasive method has proven effective in capturing tissue movement. As a secondary outcome measure, the patellar tendon tibial angle (PTTA) was also measured, as it is an indicator of IFP morphological changes^{11, 12)}.

PARTICIPANTS AND METHODS

This study was conducted as a parallel-group randomized controlled trial with single-center blinding of the participants, assessors, and analysts. The sample size was calculated using G*Power version 3.1 (Heinrich Heine University Düsseldorf, Düsseldorf, Germany), and the required sample size was determined to be 40 participants. As no prior data were available regarding the immediate effects of manual therapy on IFP gliding velocity, an analysis of variance (ANOVA) model was assumed at the planning stage as a simplified and conservative approximation. Accordingly, a moderate effect size (0.4), which is generally considered to represent a clinically meaningful magnitude of effect in intervention studies, was adopted, with a statistical power of 0.80 and a significance level of 5%. In the final statistical analysis, a generalized linear mixed model (GLMM) was applied, and effect sizes were expressed as η².

Measurements were obtained from the dominant leg. Participants with a history of dominant knee trauma or surgery were excluded. Patients with a history of trauma or surgery to any joint other than the dominant knee joint were excluded, provided that the surgery or trauma had occurred within the past year. Questionnaires were disseminated to students affiliated with a medical universities. These questionnaires explicitly delineated the exclusion criteria. The study objectives and participation requirements were meticulously explained. The recruitment period was from March 1 to May 31, 2025, and lasted for a total of 2 months. The study’s sample population comprised 48 healthy university students (42 men and six woman) who applied for the study. Prior to participation, all participants received comprehensive verbal and written explanations of the study’s purpose, methods, and ethical considerations. Written informed consent was obtained from all participants. The present study was conducted in accordance with the Declaration of Helsinki and approved by the Tokyo Ariake Medical University of Medical and Health Sciences Ethics Review Committee (Approval No.: 432).

Based on previous studies describing the IFP dynamics during knee motion, manual therapy targeting the IFP was performed with the participant in the supine position and the knee in a relaxed position^{10, 13)}. The therapist applied manual traction to the IFP in the anteroinferior direction consistent with previous reports. The magnitude of the applied force was adjusted individually and delivered at a high intensity within a range that did not provoke pain or discomfort based on participant feedback. No mechanical device was used to quantify the applied pressure. Therefore, the force magnitude was standardized according to a pain-free threshold rather than an absolute value. All interventions were performed by a single experienced examiner. Inter-operator reliability was not assessed in this study (Fig. 1). The control group was placed in the supine position on a bed for a duration of 5 min. The intervention and measurements were conducted in a dedicated evaluation room. An ultrasound diagnostic apparatus and a Biodex system were positioned at the bedside within the room to facilitate rapid pre- and post-intervention assessments. The control group was also subjected to identical environmental conditions to ensure standardization of the measurement protocols.

Fig. 1.

Manual therapy.

From the flexed position of the knee joint, the IFP is extended so that it is pulled anteriorly and inferiorly. IFP: Infrapatellar fat pad.

Following registration, the participants were randomly assigned to either the manual therapy group or the control group using a stratified block randomization table. A randomization table was prepared by a researcher who was not involved in the intervention or measurements.

Of the 48 participants who had initially registered, four were excluded from the study because of inadequate blinding procedures, and an additional four were excluded from the subsequent analysis because of corrupted image data. Specifically, these participants disclosed information regarding their group allocation to other participants during the study period, which was identified through post-measurement interviews. To preserve the integrity of participant blinding and avoid potential contamination effects, these participants were excluded from the analysis according to the predefined exclusion criteria. In the final analysis, data from a total of 40 participants were examined: 36 men and four women, with an average age of 20.5 ± 0.8 years, an average weight of 64.5 ± 9.0 kg, and an average height of 170.6 ± 0.1 cm. These participants were then equally assigned to the two groups, as shown in Table 1. Furthermore, the ultrasound images of one participant were not accurately depicted, which rendered PTTA analysis difficult; this case was treated as a missing value (Fig. 2).

Table 1. Demographic data for all participants.

Characteristic	Manual therapy (n=20)	Cotrol (n=20)	p-value
Sex (Male:Female)†	20:00	16:04	-
Age (years)†	20.7 ± 0.5	20.4 ± 0.9	0.24‡
Height (cm)†	172.8 ± 0.1	168.3 ± 0.1	0.05‡
Weight (kg)†	66.4 ± 8.1	62.7 ± 9.7	0.19‡
BMI (kg/m2)†	22.2 ± 2.1	22.1 ± 2.5	0.84‡

^†: Data were presented as the mean ± standard deviation. ^‡: Student t-test. BMI: body mass index.

Fig. 2.

Participants’ progress throughout the trial.

The gliding velocity in the IFP before and after manual therapy intervention was the primary evaluation item. Gliding velocity was evaluated using dynamic ultrasound imaging (LOGIQ e Smart Ultrasound, GE Healthcare, Tokyo, Japan, 12ML, B-mode). Ultrasound imaging was used to measure tissue velocity during isokinetic knee flexion and extension performed at an angular velocity of 45° per second. This experiment was conducted using an isokinetic dynamometer (Biodex System 3, Biodex Medical Systems, Shirley, NY, USA) (Fig. 3). During the measurement, the participants performed a knee flexion–extension task with a range of motion from 90° knee flexion to full extension (0°). All measurements were conducted under constant parameters, with a gain setting of 62, a dynamic range of 75, and a frame rate of 85 fps. The acquired images were then analyzed to determine the gliding velocities. To ensure consistency in the ultrasonographic measurements, probe placement, orientation, and pressure were carefully controlled. The ultrasound probe was positioned using the distal pole of the patella and patellar tendon as anatomical landmarks and maintained at a fixed position throughout the measurement. The probe pressure was minimized and standardized by applying only the minimum force necessary to obtain a clear image, thereby avoiding excessive compression of the underlying tissues. In addition, to enhance measurement reliability, all ultrasound imaging was performed by an examiner with more than six years of experience in musculoskeletal ultrasonography, who underwent a standardized 20-minute training session prior to data collection to confirm probe handling and image acquisition procedures. PIV analysis was performed using the Flow PIV software (Library Inc., Tokyo, Japan), which entailed the calculation of the mean of two measurements and the use of this value. Measurements were performed twice, and the mean values were used for the analysis. This was performed to reduce random measurement variability while maintaining feasibility during dynamic ultrasonographic assessment. In addition, the number of measurements was determined based on previous ultrasound-based studies that evaluated the IFP dynamics¹⁰⁾. The analysis targeted the mean velocity and peak velocities for the superficial and deep layers within the region of interest (ROI). According to the report by Kitagawa et al.¹⁴⁾, the two layers were distinguished in the following manner. For the purpose of PIV evaluation at each layer, the region of interest (ROI) was defined by referring to the prior study by Nakanishi et al¹⁰⁾. The superficial layer commenced at 75 pixels distal to the patellar apex, extended 30 pixels deep from the patellar tendon base in the vertical direction, and encompassed 150 pixels distal in the horizontal direction. The deep layer was analyzed using the same range, centered on the high-intensity septum in the IFP (Fig. 4).

Fig. 3.

Ultrasound depiction.

Dynamic ultrasound image during isokinetic knee movement using the Biodex System.

Fig. 4.

Areas of interest in the particle image velocimetry.

The superficial part (Sp) was delineated as a reference point situated 75 pixels from the patellar pool (Pp), the vertical was set at 30 pixels from the patellar tendon (Pt), and the horizontal was designated at 150 pixels. The deep part (Dp) was defined as a reference point positioned at the septal, with a comparable range.

As a secondary outcome measure, changes in the PTTA, which is the angle formed between the patellar tendon and its tibial attachment in the knee extension position, were assessed before and after the intervention using ultrasonography. The ultrasound imaging device and settings used in this study were identical to those used for the gliding assessment. Measurements were performed twice, and the mean values were calculated for analysis. The evaluation was conducted by the same researcher responsible for the gliding measurements (Fig. 5).

Fig. 5.

Patellar tendon tibial angle (PTTA).

Angle formed by the patellar tendon and tibial attachment.

Pre-intervention scores were incorporated as covariates to adjust for individual differences prior to the intervention. Subsequently, a generalized linear mixed model with the two groups and pre- and post-intervention as factors was used to examine the main effects and interactions. Statistical analyses were conducted using SPSS Statistics version 30 with the significance level set at less than 5%.

RESULTS

Regarding IFP gliding velocity, a comparison of mean and peak velocities between the manual therapy and control groups before and after the intervention revealed significant between-group effects for mean velocity (p=0.04, η²=0.05) and peak velocity (p=0.001, η²=0.12) in the deep layer (Tables 2 and 3). However, no time effect was observed, and no significant differences were found in any parameters of the superficial layer.

Table 2. Gliding speed and PTTA before and after intervention for each group (Mean ± SD).

			Manual therapy (n=20)	Control (n=20)
Average speed (cm/sec)	Superficial	pre	0.54 ± 0.14	0.53 ± 0.78
		post	0.57 ± 0.92	0.55 ± 0.67
	Deep	pre	0.78 ± 0.11	0.83 ± 0.06
		post	0.81 ± 0.12	0.85 ± 0.49
Peak speed (cm/sec)	Superficial	pre	0.84 ± 0.99	0.87 ± 0.06
		post	0.86 ± 0.11	0.88 ± 0.05
	Deep	pre	1.02 ± 0.10	1.07 ± 0.05
		post	1.01 ± 0.11	1.08 ± 0.53
PTTA (°)		pre	35.0 ± 4.42	36.8 ± 5.14
		post	37.2 ± 6.61	36.6 ± 6.23

PTTA: patellar tendon tibial angle.

Table 3. Fixed effect and interaction effect. (Manual therapy group, n=20; Control group, n=20).

		Time		Group		Interaction
		p-value	η2	p-value	η2	p-value	η2
Average speed	Superficial	0.27	0.02	0.55	0.01	0.86	0.01
	Deep	0.39	0.01	0.04	0.05	0.91	0.01
Peak speed	Superficial	0.42	0.01	0.21	0.02	0.94	0.01
	Deep	0.87	0.01	0.001	0.12	0.51	0.01
PTTA		0.45	0.08	0.63	0.03	0.36	0.01

Results of generalized linear mixed model analyses for each outcome measure, showing fixed effects of time, group, and their interaction with p-values and effect sizes (η²). PTTA: patellar tendon tibial angle.

For PTTA, no significant differences were observed in terms of time, between-group effects, or interactions. However, a moderate effect size (η²=0.08) was observed over time (Tables 2 and 3).

DISCUSSION

Manual therapy for IFP is frequently performed in clinical practice, and it is empirically known to provide immediate pain relief and improve joint range of motion. However, few studies have provided objective reports evaluating the effectiveness of manual therapy for IFP, nor have these studies clearly demonstrated its scientific basis. Moreover, some techniques that have been employed thus far depend excessively on the practitioner’s expertise or require highly skilled methods. Standardization of techniques and objective verification of their effects have not yet reached a sufficient level of advancement. This study addressed this clinical challenge by employing a reproducible and simple technique based on previously reported IFP dynamics. The present study investigated the immediate effects of manual therapy on IFP gliding velocity and PTTA in students enrolled in the Department of Medical Science.

Substantial disparities between the groups were evident in the mean and peak velocities of the deep layer. However, no interaction effects were observed, and no substantial disparities were identified in any of the superficial components. Given the absence of main effects or interaction effects in the PTTA, the present intervention method appears to exert limited immediate effects on IFP. It should be noted that the manual therapy intervention in the present study was limited to a single 3-minute application. Therefore, the findings of limited immediate effects should be interpreted within this context, and the duration-dependent effects of manual therapy on the IFP gliding dynamics cannot be excluded. Longer or more repeated interventions may produce different or more pronounced effects, which should be explored in future studies. When comparing the present findings with those of previous studies, several methodological differences should be considered, including the intervention duration, participant characteristics, and outcome measures. Previous studies investigating interventions targeting the IFP have predominantly focused on patient populations, particularly individuals with knee osteoarthritis, and have often employed longer or repeated intervention protocols. Kitano et al. reported changes in IFP-related outcomes following repeated low-intensity pulsed ultrasound therapy in patients with knee osteoarthritis¹¹⁾, while Kitagawa et al. demonstrated improvements in IFP flexibility after a structured physical therapy program involving multiple sessions⁸⁾. In addition, Okita et al. reported alterations in IFP volume and mobility during quasi-static knee extension after manual therapy in patients with knee osteoarthritis⁷⁾. More recently, Nakanishi et al. showed that isometric quadriceps exercise influences the local microcirculation within the IFP in female patients with knee osteoarthritis¹⁵⁾.

In contrast, the present study examined the immediate effects of a single, short-duration manual therapy intervention applied for 3 min in a non-patient population and evaluated the dynamic IFP gliding properties using quantitative ultrasound-based analysis. These differences in intervention duration, participant characteristics, and outcome measures may contribute to the discrepancies between the present findings and those of previous studies and should be considered when interpreting the limited immediate effects observed in the present study. In this context, the present study should be regarded as a preliminary investigation conducted in a non-patient population to establish baseline information on immediate IFP gliding responses in the absence of overt pathology.

A meticulous examination is imperative to avoid overinterpretation of the findings, especially concerning the intergroup disparities evident in the deep layer and modest effect size observed for the time effect in PTTA. The variations in the deep layer were presumably attributable to differences in group mean levels following the intervention. However, these variations do not necessarily indecate clinical efficacy.

The IFP exhibits distinct structural and functional characteristics in its superficial layer, also referred to as the macro-chamber, and its deep layer, designated as the micro-chamber¹⁶⁾. In the deep layer, collagen fibers are arranged in a thick configuration, thereby playing a supportive role. By contrast, the superficial layer, on the other hand, contains fine, highly mobile lobular structures believed to contribute to gliding properties under dynamic conditions. Moreover, PIV analysis demonstrated that IFP gliding velocity during knee flexion and extension was greater in the deep layer than in the superficial layer¹⁰⁾. Furthermore, studies employing MRI-derived three-dimensional models have corroborated substantial volume alterations in the deep-layer region during knee extension¹⁷⁾. These findings suggest that the deep layer of the IFP may reflect dynamic characteristics in response to joint movement.

Although the observed between-group differences in deep structures and the effect size of PTTA in this study cannot be entirely ruled out as being potentially related to these anatomical considerations, the results do not consistently support an intervention effect. Consequently, the findings of this study may instead suggest that “the effect of manual therapy on IFP using this technique is limited”. It is imperative that further largescale studies be conducted employing rigorous verification methods. Only then can a meaningful discussion be presented regarding the clinical significance of this phenomenon.

Although no statistically significant difference was observed in PTTA, a tendency toward increased angles was noted in the manual therapy group. Further investigation is necessary to determine the clinical significance of these finding. PTTA has been adopted as an evaluation index for morphological changes in the IFP owing to its high reliability and reproducibility under identical conditions. In this study, PTTA measurement at a constant knee extension angle was proposed as a quantitative indicator of alterations in tissue softness and gliding properties. This method is clinically useful because it enables the direct assessment of structural and functional changes in joint components and supporting tissues by comparing measurements before and after treatment while maintaining a constant joint angle. PTTA is known to vary with knee flexion angle and quadriceps tension, rendering it susceptible to multifactorial influences during dynamic measurements^{18, 19)}. Conversely, evaluation at a fixed joint angle is expected to more purely reflect changes in local flexibility and gliding properties induced by the intervention. The present study was conducted to evaluate the hypothesis that manual therapy improves IFP flexibility, resulting in changes in the angle of tendon tension transmission via the patellar apex and tibial tuberosity, there by increasing PTTA. Evaluations were conducted under static conditions. This measurement strategy is regarded as effective in capturing subtle structural changes otherwise difficult to detect using conventional range-of-motion or subjective indicators. Furthermore, PTTA offers the advantage of objectively quantifying subtle structural changes that are difficult to capture using conventional range-of-motion measurements or subjective assessments, making it a promising evaluation metric for future IFP research. While the present study did not detect changes following a 3-min intervention, the development of more efficacious intervention methods may demonstrate the utility of PTTA.

To date, interventional studies on manual therapy for IFP are extremely limited^{7, 8)}. The techniques employed in previous studies frequently required highly skilled and advanced methods, and the effects of these interventions were often limited, making it challenging to conclude that they demonstrated sufficient clinical utility. In light of these findings, the present study adopted a more streamlined and reproducible intervention technique that centered on the dynamic characteristics of the IFP. However, this approach did not yield statistically significant results. The potential for future research lies in extending the intervention duration or refining the technique, which may have more discernible effects on IFP dynamics. Furthermore, the results suggest a potential discrepancy between patients’ subjective perceptions of immediate symptom changes in clinical settings and objective measurements of indicators, such as gliding ability. Consequently, future research should prioritize integrating subjective outcomes with objective assessments.

The present study had a few limitations. First, a marked sex imbalance was observed among participants, with female participants being underrepresented. This imbalance arose as an unintended consequence of the randomization and blinding procedures in the context of a limited sample size, resulting in a higher proportion of female participants being allocated to the control group. Although this sex imbalance is recognized as a limitation, the randomization process was appropriately implemented and did not indicate a flaw in the study design. Importantly, excluding female participants post-hoc would have disproportionately reduced the control group sample size and potentially compromised the randomized group structure. Therefore, rather than applying exclusion criteria after randomization, sex was statistically adjusted in the generalized linear mixed model, and the influence of sex on the study outcomes was considered to be adequately controlled.

Second, the study population was restricted to healthy individuals. Consequently, its direct efficacy in cases with clinical issues, such as pain or restricted range of motion, remains unclear. Therefore validation through clinical implementation is imperative.

Funding

None of the authors received any funding for this study.

Conflict of interest

The authors declare no conflict of interest.