Longitudinal Quantitative Ultrasonic Analysis of Patellar Tendon in a Collegiate Athlete After Bilateral Debridement: A Case Report

The patient was a National Collegiate Athletic Association Division I female basketball player (age = 20 years, height = 190.5 cm, mass = 87 kg) who began participating in organized sport at 7 years of age and competitive travel sport at 10 years of age. The athlete first reported knee pain at 13 years of age and was diagnosed with bilateral Osgood-Schlatter disease.⁸  As a freshman in high school, she was diagnosed by a physician with bilateral patellar tendinitis. Knee pain persisted as the patient entered collegiate athletics.

To determine a definitive diagnosis of the patellar tendon, magnetic resonance imaging (MRI) of both knees was obtained during the patient's freshman year in college. The images showed normal meniscal structures in both knees but focal lesions in the patellar tendons. The patient elected to undergo conservative treatment; however, her knee pain persisted, and MRI was repeated. Radiologic reports indicated central tendinopathy of the proximal portion of each patellar tendon with partial detachment of the deep fibers. The patient began a presurgical rehabilitation program for 3 weeks and participated in practice until the day before surgery. Bilateral open patellar debridement surgery, similar to that described by Bahr et al,⁹  followed by platelet-rich plasma therapy and Tenex Health TX procedures were performed. The intent of the surgery was complete removal of the degenerated and detached fibers to stimulate growth of new and more normal fibers and restore the patellar tendons. After surgery, the patient underwent a guided rehabilitation program for 12 months.

The patient's knee was immobilized in 0° of flexion at all times for 4 weeks except during rehabilitation and imaging sessions. Typical postoperative pain medications were prescribed by the attending physician to reduce postoperative inflammation. The attending athletic trainer (A.R.) implemented a modified anterior cruciate ligament rehabilitation protocol. An outline of the timeline, clinical goals, and sample rehabilitation interventions is given in Table 1.

The outcome of intratendinous tissue structure (peak spatial frequency radius [PSFR]) was measured using ultrasonography. The PSFR reflects compaction of the striated pattern of the collagen fascicles.⁵  The patient's knee was imaged bilaterally approximately every week for the first 2 months, every 3 weeks from 2 to 6 months after surgery, and every other month thereafter. The first imaging session (week 1) was conducted 5 days after the surgical procedure (week 0). The patient's knee was imaged on the same day of the week and at the same time of day for each session. Throughout the first 2 months, the patient missed 1 imaging session because of a schedule conflict and travel.

Ultrasound images were collected using a commercial research ultrasound machine (Vantage 256; Verasonics, Inc) with a 3- to 12-MHz (center frequency = 8 MHz) linear array transducer. Ultrasound settings were 2 transmit foci of 6 and 12 mm and an imaging depth of 30 mm. The transducer was oriented parallel to the tendon to minimize hypoechoic artifacts. Longitudinal images were collected below the knee between the patella and tibial tuberosity in the middle of the tendon. Images were acquired at 0°, 30°, 60°, and 90° of available knee flexion. Victorian Institute of Sport-P (VISA-P) scoring was collected on the same day as the imaging session and integrated as a qualitative and progressive metric to assess the patient's perceived level of pain and function.¹⁰ 

After the imaging sessions were completed, longitudinal images were extracted and processed using custom MATLAB (The MathWorks, Inc) algorithms. A polygonal region of interest (ROI) was placed around the entire tendon thickness, beginning approximately 1 cm distal to the inferior pole of the patella and spanning approximately 2.5 cm in length (Figure 1). All possible 32- × 32-pixel kernels (corresponding to a 2.1-mm-long by 1.2-mm-wide rectangle) that fit within the ROI were analyzed in the Fourier domain.⁵  The PSFR was averaged over all kernels of the ROI to obtain a single value bilaterally for each knee-joint angle and imaging session. Weekly changes in VISA-P scores and PSFR were calculated. Simple linear regression was used to compare VISA-P scores as a function of weeks after surgery.

The patient consistently adhered to the rehabilitation program with specific postoperative rehabilitation milestones as follows: no pain with walking at 3 months, walking up and down stairs without pain by 7 months, and progression to functional and position-specific drills in individual workouts with a tolerable level of pain by 9 months. She progressed more quickly with the left than the right limb. During position-specific drills at 9 months after surgery, she experienced substantial bilateral joint effusion (MRI-confirmed chondromalacia patellae) and regressed in rehabilitation. The effusion was managed using nonsteroidal anti-inflammatory drugs and cryotherapy. The patient was cleared to progress to full activity (Table 1) by the physician at the end of 10 months after surgery (2 months before the season started). By 12 months after surgery (corresponding to the start of the competitive season), the patient participated in 1.5 hours of practice daily.

For the first week after surgery, the patient was limited to 30° of knee flexion. Imaging at 60° of knee flexion was performed during weeks 2 and 3. Imaging at 90° of knee flexion was performed at 4 weeks after surgery and every imaging session thereafter. The PSFR value increased slightly with increased knee flexion (Table 2). By 1 year after surgery, PSFR increased across all knee-joint angles (Table 2; Figure 2). Monthly percentage changes in PSFR values are shown in Table 2. Fluctuations in PSFR were most variable at 30° of knee flexion during the first 3 months after surgery but stabilized by 15 weeks after surgery. At 60° of knee flexion, PSFR was equivalent between limbs after the first month after surgery, followed by an acute divergence around 15 weeks after surgery. Despite the fluctuation in PSFR of the left patellar tendon, the difference between limbs was minimal by 6 months. The PSFR at 90° of knee flexion trended negatively for the left limb at 2 months after surgery and did not recover until 4 months after surgery. Another acute decline in PSFR occurred nearly 5 months after surgery, which did not resolve until approximately 3 months later.

The VISA-P scores increased consistently yet slowly throughout the first 2 months of rehabilitation, followed by an acute decrease around 3 months (Figure 3). The scores continued to increase throughout the remainder of the study but plateaued around 7 months at a score of 37 and a final score of 36 (normal = 100, tendinopathy = 80).

In this exploration case report, we investigated longitudinal changes in intratendinous tissue structure as determined from SFA of ultrasound B-mode images. The PSFR provided a quantitative metric for tracking tendon structure throughout rehabilitation and the return-to-play progression after bilateral open patellar tendon surgery (debridement, platelet-rich plasma therapy, and Tenex procedures). Overall, PSFR increased slightly over the 12 months in the more-extended knee-joint positions. The patient's pain and function improved throughout the study, but the former persisted and the latter remained low despite intensive rehabilitation.

The addition of regular imaging in this case study was designed to quantify the progression of structural adaptation in vivo. Given that PSFR is a measure of tissue organization, we expected to see an increase in PSFR over time as the patient healed. We did observe an overall increase in PSFR throughout the 12-month rehabilitation and return-to-play progression, indicating an increase in the overall organization of the collagen fascicles. This finding supports the idea that quantitative image analysis may be useful for monitoring rehabilitation progress.

Previous researchers¹¹  who used PSFR did not investigate the effects of various knee-joint angles on PSFR and only studied 90° of knee flexion. We noted that PSFR appeared to depend on the knee-joint angle, as a minimal change in PSFR was evident at 90° of knee flexion compared with other knee-joint angles. This observation could be attributed to the fact that the collagen in the patellar tendon is under tension and more aligned in a lengthened position as the knee is flexed, which is consistent with earlier PSFR results with isometric loading¹²  and the basic tendon response to lengthening.¹³  Moreover, the fluctuations, especially at 0° and 30° of knee flexion, could have been due to immature, unaligned collagen fascicles that matured throughout the healing process and rehabilitation program.

In addition, PSFR appeared to detect acute changes (Figure 2) in intratendinous structure due to acute trauma (eg, during week 12, the patient slipped while exiting a vehicle) and increased tissue loading (eg, during week 25, she returned to sport-specific training). Consistent with previous literature¹⁴  in which tendon morphology was characterized, we did not demonstrate any relationship between changes in PSFR and VISA-P scores. Changes in PSFR may be more indicative of tendon loading than clinical presentations of pain,¹¹  although this needs to be validated in a larger sample size.

The patient's pain and function increased throughout rehabilitation and the return-to-play progression (Table 1). The volume and load of exercises intended to stress the tendon were dictated by ongoing pain. The patient did not perform warmups before practice with the team and instead warmed up on a stationary bicycle. In addition, she did not participate in an entire practice or in consecutive drills. She performed a selection of drills that were chosen for her and totaled approximately 1.5 hours. With regard to specific measures of her pain and function, we noted 3 acute decreases in VISA-P scoring at weeks 12 and 25. The decrease in VISA-P at week 12 coincided with a fall. The second decrease around week 25 occurred after a week of inactivity. It is well known that optimal loading is an important factor in tendon health in rehabilitation.¹⁵  This period of inactivity at the 6-month mark may have negatively affected her progress in rehabilitation. The third decrease in VISA-P occurred as the patient began to progress to sport-specific skill workouts, strength training in the weight room, and daily rehabilitation. This final acute decrease possibly occurred because of the increasing physical demands on the knees despite progressive rehabilitation.

The patient's final VISA-P score of 39 (an increase of 34 from baseline) was considerably lower than a normal score of 100 and a score of 80 indicating tendinopathy. This was a substantial increase (Figure 3) and greater than the minimal detectable change for improvement.¹⁶  One contributing factor to the prolonged recovery could have been the bilateral surgical procedure. Maffulli et al¹⁷  reported that patients who underwent bilateral surgery for patellar tendinopathy returned to sport later than their counterparts who underwent unilateral surgery.¹⁷  It is also possible that the chronicity of pain contributed to the poorer outcome than the outcome reported earlier,¹⁴  as this patient had clinical evidence of tendinopathy for nearly a decade before this investigation. Researchers¹⁸  found that more athletic exposure increased the risk of young dancers sustaining knee injuries, which is consistent with this patient's athletic history. The intensity of the patient's participation and her high level of competition, even in the presence of continuing pain, may have prevented proper recovery of the tendon before the study and also influenced the poorer outcome. Despite the low VISA-P score, she did return to full participation in sport, albeit at a substantially reduced activity level with limited involvement in competitions.

Spatial frequency analysis has been related to mechanical properties of the tissue and shown to differentiate between pathologic and healthy tendons.^5,6,11  In this exploration case report, we observed that SFA provided objective metrics to the progression of tendon healing after patellar tendon surgery in a female Division I collegiate athlete. Given the unique nature of this patient's condition, including chronic pain and athletic participation, these findings should not be generalized. We provide initial observations that SFA may be a useful complement to clinical treatments that address both the structural and functional recovery of injured tissue. The PSFR supplies an additional objective measure of the tendon structure that can potentially aid clinicians in making rehabilitation decisions. Because a different type of collagen is laid down in an unorganized fashion at the site of injury, those fibers need to be able to withstand the stresses that will be applied. This is particularly important, as the sites of initial injury are often the sites of reinjury. For clinicians without access to these image-analysis methods, our findings could support the concept of load management and its implications on tissue structural adaptations. However, we reported only a single case and do not know how changes in PSFR will influence rehabilitation progression in a broader population. Based on our observations, we suggest that structural changes in the patellar tendon after surgery and during rehabilitation of an athlete with chronic tendinopathy are not necessarily related to pain. Although tendon structure and pain are not necessarily related, the objective information gained from SFA measures provides the clinician and patient with confidence that the tendon can be progressively loaded and withstand stresses during activity. Future researchers should investigate the relationships among structural adaptations, PSFR, reinjury risk, and specific joint function for a more complete view of recovery.