Correlation of Arthroscopic Grading and Optical Coherence Tomography as Markers of Early Repair and Predictors of Later Healing Evident on MRI and Histomorphometric Assessment of Cartilage Defects Implanted with Chondrocytes Overexpressing IGF-I

Abstract

Objective

Injury of articular cartilage is common, and due to the poor intrinsic capabilities of chondrocytes, it can precipitate joint degradation and osteoarthritis (OA). Implantation of autologous chondrocytes into cartilaginous defects has been used to bolster repair. Accurate assessment of the quality of repair tissue remains challenging. This study aimed to investigate the utility of noninvasive imaging modalities, including arthroscopic grading and optical coherence tomography (OCT) for assessment of early cartilage repair (8 weeks), and MRI to determine long-term healing (8 months).

Design

Large (15 mm diameter), full-thickness chondral defects were created on both lateral trochlear ridges of the femur in 24 horses. Defects were implanted with autologous chondrocytes transduced with rAAV5-IGF-I, autologous chondrocytes transduced with rAAV5-GFP, naïve autologous chondrocytes, or autologous fibrin. Healing was evaluated at 8 weeks post-implantation using arthroscopy and OCT, and at 8 months post-implantation using MRI, gross pathology, and histopathology.

Results

OCT and arthroscopic scoring of short-term repair tissue were significantly correlated. Arthroscopy was also correlated with later gross pathology and histopathology of repair tissue at 8 months post-implantation, while OCT was not correlated. MRI was not correlated with any other assessment variable.

Conclusions

This study indicated that arthroscopic inspection and manual probing to develop an early repair score may be a better predictor of long-term cartilage repair quality following autologous chondrocyte implantation. Furthermore, qualitative MRI may not provide additional discriminatory information when assessing mature repair tissue, at least in this equine model of cartilage repair.

Keywords

optical coherence tomography MRI diagnostics cartilage repair repair AAV gene therapy therapeutic delivery

Introduction

Articular cartilage has poor regenerative and healing capacities, mainly due to its avascular and aneural nature.¹ Therefore, focal cartilage defects often lead to osteoarthritis (OA) with end-stage treatment being total joint replacement.^2,3 Surgical techniques to improve cartilage repair in focal defects are varied; however, autologous chondrocyte implantation (ACI), in which chondrocytes harvested from a non-weightbearing articular surface are re-implanted into defects, is likely the “gold standard” for repair of large cartilaginous lesions in human surgery.⁴ Although ACI has shown clinical success, full chondrogenesis, cell persistence, and the formation of hyaline cartilage have been problematic, thereby prompting investigation into genetic modification of chondrocytes ex vivo to improve intrinsic reparability. To augment and improve cartilage repair techniques, there is a need for increased understanding of the healing process over time.

Adeno-associated virus (AAV) is the most commonly used gene therapy vector due to its lack of pathogenicity, rare incidence of genomic integration, and long-term transgene expression. Although there are several known AAV serotypes, AAV-5 has been shown to effectively transduce equine chondrocytes.^5,6 Insulin-like growth factor (IGF)-I is a key anabolic and mitogenic factor in chondrocyte metabolism, and IGF-I supplementation in vitro⁷ and in vivo⁸ has been shown to increase extracellular matrix (ECM) production and decrease proteoglycan degradation.

Accurate assessment of cartilage injury and healing has been challenging due to the inability to see subsurface changes during arthroscopic examination and the destructive nature of cartilage biopsy procedures. Optical coherence tomography (OCT) and MRI are both relatively noninvasive imaging modalities that allow more in-depth evaluation of the cartilage subsurface. OCT measures the reflection of near-infrared light from biological tissues in real time, including articular cartilage.⁹ It yields high-spatial-resolution cross-sectional images (4-20 µm) at resolutions that are comparable to low-power histology.¹⁰ One of the greatest benefits of OCT is the in vivo applicability as the probe can easily be used in conjunction with arthroscopy to evaluate the health of the articular surface. MRI is another imaging modality that is frequently used to assess cartilage healing and provides a means to assess total joint health. Changes in the articular surface, including fissures, delamination, subchondral sclerosis, and focal loss, can be seen using standard MRI with cartilage-sensitive sequences.¹¹

OCT and MRI are attractive options for assessing cartilage repair over time as they can both be used to examine the microstructure of cartilage in a nondestructive manner. Historically, these techniques have been used mainly to assess cartilage degeneration and damage, with few studies investigating their efficacy in assessing repair tissue following ACI. Therefore, in this study, we aimed to compare cartilage repair in critically sized defects repaired with rAAV5-IGF-I-transduced autologous chondrocytes using different assessment techniques including OCT and arthroscopic examination performed 8 weeks post-implantation and MRI, gross examination, and histology performed 8 months post-implantation. We hypothesized that OCT and arthroscopic scores would be significantly correlated at 8 weeks and that MRI, gross scores, and histology scores would be significantly correlated at 8 months.

Methods

Chondrocyte Harvest, Transduction, and Implantation

Following Institutional Animal Care and Use Committee approval, 24 horses were randomly assigned to 1 of 3 treatment groups, including (1) implantation with rAAV5-IGF-I-transduced chondrocytes, (2) implantation with rAAV5-green fluorescent protein (GFP)-transduced chondrocytes, or (3) implantation with naïve (untransduced) chondrocytes. In preparation for autologous chondrocyte implantation, cartilage was harvested arthroscopically from the distal (non-weightbearing) medial and lateral trochlear ridges of a randomly selected talus of each horse under general anesthesia. A total of 2 g of cartilage was removed from trochlear ridges of the talus using a ring curette (5 mm) and Ferris-Smith rongeurs. Cartilage was placed in sterile phosphate-buffered saline with 100 units/ml of penicillin/streptomycin. Cartilage was then digested in 0.075% collagenase (Worthington Biochemical, Freehold, NJ) as previously described.¹² Following isolation, chondrocytes were expanded in Ham’s F12 medium (Gibco-Life Technology, Grand Island, NY) supplemented with 10% fetal bovine serum, 50 μg/ml ascorbic acid, 30 μg/ml α-ketoglutarate, 300 μg/ml l-glutamine, 100 units/ml penicillin/streptomycin, and 25 mM HEPES (Gibco-Life Technology, Grand Island, NY) to a total of 20 × 10⁶ cells and then stored in liquid nitrogen until the time of implantation.¹² Chondrocytes were thawed and transduced 48 hours prior to surgery in Opti-Mem (Invitrogen, Grand Island, NY) with rAAV5-IGF-I, rAAV5-GFP, or no virus at a dose of 10⁵ vg/cell for 2 hours.

rAAV5-IGF-I and rAAV5-GFP were generated by the Research Vector Core at the Children’s Hospital of Philadelphia in HEK293 cells. Prior to vector production, full-length equine IGF-I cDNA was amplified by polymerase chain reaction. Equine IGF-I and GFP were subcloned into the rAAV transfer plasmid pHpa-trs-SK using SacII and Not sites. The transgenes were flanked by inverted terminal repeats and under control of the cytomegalovirus promoter.

Autologous chondrocyte implantation was performed with horses under general anesthesia and placed in dorsal recumbency. Full-thickness chondral defects were made bilaterally on the mid-distal lateral trochlear ridge (LTR) of the femur. Defects were created arthroscopically with a 15-mm fluted spade-bit cutter with a sharpened perimeter skirt (Special Devices, Grass Valley, CA). One defect in each horse was implanted with chondrocytes (rAAV5-IGF-I-transduced, rAAV5-GFP-transduced, or naïve) in autologous fibrinogen prepared from previously collected plasma,¹³ while the other defect was implanted with fibrinogen alone. Fibrinogen was polymerized to fibrin using calcium-activated bovine thrombin (500 U/ml; Sigma, St Louis, MO), providing an adhesive clot. Joints were then put through repeated range of motion to verify graft retention. Perioperative pain management was achieved through use of epidural morphine and oral phenylbutazone administration. Horses were box stall rested for 4 weeks postoperatively, followed by 4 weeks of increasing amounts of walking. Horses were then box stall rested for 2 weeks after second-look arthroscopy 8 weeks post-implantation, following which they were allowed unlimited pasture turnout until the end of the study at 8 months post-implantation.

Optical Coherence Tomography

An OCT probe optimized for arthroscopic use was provided by the University of Pittsburgh (Cartilage Restoration Center, Pittsburgh, PA) with wavelength of 1305 ± 55nm, in-plane pixel size of 10 × 10 µm², slice thickness of 100 µm, and frame rate of 100 frames/s. OCT images were captured during second-look arthroscopy 8 weeks post-implantation. Two-dimensional image sequences were acquired by a computer during arthroscopic examination. Two sites of interest in each defect were imaged with the OCT probe, including the best repair tissue and worst repair tissue within the defect. These sites were selected by the 2 surgeons performing each surgery (K.F.O., A.J.N.). Four images were captured at each site including where the probe was initially placed (0°) and then following 90°, 180°, and 270° rotation of the probe at the same site. OCT images were evaluated using a semiquantitative scoring system modified from the methods described by te Moller et al.¹⁴ and Bear et al.¹⁵ ( Table 1 ). All images were then scored by 2 observers (K.F.O., S.A.C.) blinded to the treatment. Individual parameters including birefringence homogeneity, surface fibrillation, surface clefts (horizontal and vertical), erosion, and repair tissue thickness compared with native tissue were evaluated. The sum of these scores was used to produce a total best OCT score and a total worst OCT score. The average of these 2 scores was used to produce a total average OCT score used in data analysis.

Table 1.

Pathology Scoring System for Optical Coherence Tomography Images from Repair Tissue 8 Weeks Post-Implantation (0 = Normal OCT; 18 = Most Abnormal).

Optical Coherence Tomography Score Parameter	Qualifications
Birefringence homogeneity
0	Normal
1	Slight loss
2	Moderate loss
3	Severe to complete loss
Surface fibrillation
0	Normal
1	Mild
2	Moderate
3	Severe
Surface clefts (horizontal and vertical)
0	None
1	Focal
2	Multifocal
3	Generalized
Erosion
0	None
1	Mild
2	Moderate
3	Severe
Anomaly below surface (subsurface clefts)
0	None
1	Focal
2	Multifocal
3	Generalized
Repair tissue thickness
0	75%-100%
1	50%-74%
2	25%-49%
3	0%-24%

Arthroscopic Examination

Short-term healing was assessed at 8 weeks post-implantation using arthroscopic second-look examination with grading of defects by 2 blinded observers (A.J.N., K.F.O.) ( Table 2 ). Repair tissue was evaluated for defect fill, smoothness of repair tissue, whiteness of repair tissue, peripheral integration, and degree of subchondral bone attachment. The sum of these scores was used to produce a total second-look arthroscopy score.

Table 2.

Pathology Scoring System for Arthroscopic and Gross Tissue Pathology (0 = Normal Cartilage; 20 = Most Abnormal) and for Histological Tissue Pathology (0 = Normal Cartilage; 31 = Most Abnormal).

Arthroscopic and Gross Tissue Pathology Score Parameter	Qualifications (%)
Defect fill
0	100
1	>75; >100
2	51-75
3	26-50
4	0-25
Amount of smooth repair tissue covering defect
0	100
1	>75
2	51-75
3	26-50
4	0-25
Amount of white tissue
0	100
1	>75
2	51-75
3	26-50
4	0-25
Graft-recipient tissue integration
0	100
1	>75
2	51-75
3	26-50
4	0-25
Subchondral bone attachment
0	100
1	>75
2	51-75
3	26-50
4	0-25
Histological Healing Score Parameter	Qualifications (%)
Defect fill
0	100
1	> 75; > 100
2	51-75
3	26-50
4	0-25
Chondrocyte predominance
0	All
1	67-99
2	34-66
3	1-33
4	None
Perilesional cloning
0	None
1	Seldom
2	Occasional
3	Frequent
Lesion-perilesion tissue integration
0	Complete
1	Gap on 1 side
2	Gap on both sides
Subchondral bone attachment
0	100
1	>75
2	51-75
3	26-50
4	0-25
Surface fibrillation
0	None
1	Slight fibrillation
2	Moderate fibrillation
3	Severe fibrillation
Tidemark reformation
0	Complete
1	51-99
2	1-50
3	None
Toluidine blue staining
0	91-100
1	76-90
2	51-75
3	26-50
4	0-25
Collagen type II predominance
0	> 90
1	76-90
2	51-75
3	26-50
4	0-25

MRI

Horses were sacrificed with a barbiturate overdose 8 months post-implantation. Hind limbs were sectioned at the level of the mid-femur and mid-tibia. Skin, subcutaneous tissue, muscle, and fat surrounding the stifle joint were removed prior to placement in the MRI. MRI of the cadaveric equine stifles was performed using transverse, sagittal, and coronal plane fast spin-echo imaging (TE: 36 ms, TR: 3,000 ms, echo train length (ETL): 7, bandwidth (BW): 244 Hz/px, acquisition matrix: 512 ×352, field of view: 17 cm, slice thickness: 2.2-2.5 mm, slice spacing: 0 mm, number of excitations (NEX) 3-4). Images were obtained using a quadrature knee coil on a 1.5T clinical field strength MRI (1.5T Toshiba Vantage Atlas). Following image acquisition, all MRI studies were read by 1 experienced veterinary radiologist who did not have prior knowledge of the surgical procedure (Sarah L. Pownder). The images were scored based on a previously utilized modified Magnetic Resonance Observation of cartilage Repair Tissue (MOCART) scoring system used in translational models of cartilage repair.¹⁶ The transverse plane fast spin-echo images were assessed for degree of fill, surface morphology, repair tissue signal intensity, and subchondral bone morphology (Table 3; Supplementary Fig. S1). Sagittal images were used to correlate with sagittal plane histology sections. As the defect was large (1.5 cm), it was divided into three 5-mm regions (proximal, mid, distal) and scored independently. This method of evaluation was chosen as some repairs demonstrated excellent fill at the periphery with relatively no fill in the center. The sum of these scores was used to produce a total MRI score.

Table 3.

Scoring System Used for MRI of Repair Tissue 8 Months Post-Implantation (0 = Normal MRI; 17 = Most Abnormal).

MRI Variable	Qualifications
Degree of repair fill
0	67%-100%
1	34%-66%
2	1%-33%
3	Exposed bone
Surface of repair
0	Flush
1	Fibrillation
2	Damage, 50%
3	Damage ≥50%
4	No cartilage
Structure of repair
0	Homogenous
1	Not homogenous
2	No cartilage
Signal intensity of repair
0	Normal
1	Mild hyperintense
2	Moderate hyperintense
3	No cartilage
Subchondral bone
0	Normal
1	Depressed
Subchondral bone
0	Normal
1	Proud
Subchondral bone
0	Intact
1	Mild granulation/sclerosis
2	Moderate granulation/sclerosis
3	Severe granulation/sclerosis

Gross and Histopathology

Cartilage repair tissue was graded by the same two blinded observers (K.F.O., A.J.N.) using the same semiquantitative scoring system used at arthroscopy ( Table 2 ). The sum of these scores was used to produce a total gross pathology score. An osteochondral block 5 mm in width, 15 mm in height, and 25 mm in length was collected from the middle of each defect using an oscillating saw. The block extended 5 mm beyond the proximal and distal edges of the graft so that native tissue could also be assessed. Osteochondral blocks were fixed in 4% paraformaldehyde (PFA) and decalcified in 10% ethylenediaminetetraacetic acid (EDTA). Following decalcification, blocks were embedded in paraffin, sectioned at 6 μm, and stained with hematoxylin and eosin (H&E) to evaluate morphology and toluidine blue to evaluate proteoglycan content. A semiquantitative scoring system ( Table 2 ) was used to assess histological healing. Briefly, osteochondral specimens were scored for defect fill, predominance of chondrocyte cell type, perilesional cloning, graft-recipient tissue integration, subchondral bone attachment, surface fibrillation, tidemark reformation, toluidine blue staining, and collagen type II predominance. The sum of these scores was used to produce a total histology score.

Statistical Analysis

Each horse in the study provided multiple observations: one chondral defect filled with (1) rAAV5-IGF-I chondrocytes, (2) rAAV5-GFP chondrocytes, or (3) naïve chondrocytes, and the contralateral defect filled with fibrin vehicle. To account for possible correlation between observations made in the same horse, a mixed-effects model was used for analysis. Horse was treated as a random effect and treatment was treated as a fixed effect. Multiple comparisons for differences in parameters for each interaction were made with a Tukey’s post hoc test. Correlation analysis was performed using Spearman’s rho correlation coefficient. Statistical analysis was performed using Stata (StataCorp, College Station, TX). The level of significance was set at P < 0.05.

Results

Horses

Of the 24 horses included in the study, there were 13 Thoroughbreds and 11 Quarter Horses with a mean age of 2.9 years (range, 2-4 years). The study population included 11 castrated male horses and 13 female horses with a mean weight of 506 kg (range, 396-532 kg).

Arthroscopic Examination

Second-look arthroscopy was performed at 8 weeks post-implantation to evaluate short-term healing and carry out OCT. At 8 weeks, all defects were at least partially filled with repair tissue ( Fig. 1 ). Implantation of defects with rAAV5-IGF-I-transduced chondrocytes (5.38 ± 0.94) yielded repair tissue that was significantly improved when compared with rAAV5-GFP-filled (11.13 ± 1.52; P = 0.001), naïve chondrocyte-filled (8.5 ± 1.27; P = 0.029), and fibrin-filled (10.79 ± 0.74; P < 0.001) defects ( Table 4 ).

Figure 1.

Representative of arthroscopic, OCT (best and worst), gross, histologic, and MRI images from horses treated with fibrin, rAAV5-IGF-I chondrocytes, rAAV5-GFP chondrocytes, or naïve chondrocytes. All images in 1 treatment group are from the same horse. Arthroscopic images show the defect in the lateral trochlear ridge at 8 weeks post-implantation. The defects repaired with rAAV5-IGF-I-transduced chondrocytes appear to have the best repair tissue compared with the other treatment groups. OCT images were obtained from the defects during arthroscopy 8 weeks post-implantation with both the “best” and “worst” area of the repair tissue imaged. Sagittal plane proton density fast spin echo MRI images oriented proximal-distal (left to right) noting the patella in each field of view for orientation. In the fibrin defect, note the lack of continuity at the proximal and distal interfaces (arrows). In the rAAV5-IGF-I defect, the depth of the subchondral sclerosis is shown (arrowheads). In the rAAV5-GFP defect, a margin of proud bone is visible (dotted arrow and seen in corresponding histology). In the naïve chondrocyte defect, a relatively uniform articular margin with a focus of subchondral bone inhomogeneity (white bracket) is shown. Gross pathology images and photomicrographs of osteochondral sections collected show the same defects 8 months post-implantation demonstarting better healing in the rAAV5-IGF-I-treated defects. OCT = optical coherence tomography.

Table 4.

Total Scores for Repair Tissue Post-Implantation.

Treatment Group	Total Arthroscopy Score	Total OCT Score	Total MRI Score	Total Gross Score	Total Histology Score
rAAV5-IGF-I	5.38 ± 0.94^a	5.11 ± 0.35^b	7.75 ± 1.01^a	4.25 ± 0.88^a	12.38 ± 1.30^a
rAAV5-GFP	11.13 ± 1.52^b	5.37 ± 0.37^b	6.52 ± 0.31^a	5.75 ± 0.31^ab	15.88 ± 1.64^b
Naïve chondrocyte	8.5 ± 1.27b^b	3.41 ± 0.46^a	5.94 ± 0.73^a	6.88 ± 0.58^b	18.50 ± 1.49^b
Fibrin	10.79 ± 0.74^b	5.58 ± 0.25^b	5.57 ± 0.34^a	6.75 ± 0.42^b	18.58 ± 0.68^b

Data presented as mean ± SEM. Different letters denote significant differences between groups, n = 8 (rAAV5-IGF-I, rAAV5-GFP, naïve chondrocyte), n = 24 (fibrin).

OCT = optical coherence tomography.

Optical Coherence Tomography

Arthroscopic application of the OCT probe was successful in all horses 8 weeks post-implantation. Defects implanted with naïve chondrocytes had significantly better total scores (3.41 ± 0.46) than those implanted with fibrin (5.58 ± 0.25; P < 0.001), rAAV5-IGF-I (5.11 ± 0.35; P = 0.002), and rAAV-GFP (5.37 ± 0.37; P < 0.001) ( Table 4 ).

MRI

MRI was performed on stifles collected from horses immediately following humane euthanasia. There were no significant differences in MRI scores between any of the treatment groups. Although fibrin-filled defects had the lowest (best) MRI scores, this was not significant ( Table 4 ).

Gross Pathology Scores

Gross pathology scoring of repair tissue filled with rAAV5-IGF-I chondrocytes had significantly lower (better) total pathology scores (4.25 ± 0.88), compared with defects repaired with naive chondrocytes (6.88 ± 0.58; P = 0.003) and fibrin alone (6.75 ± 0.42; P = 0.001). Although rAAV5-IGF-I-implanted defects had better total pathology scores compared with defects treated with rAAV5-GFP chondrocytes (5.75 ± 0.31), this difference was not significant (P = 0.058) ( Table 4 ).

Histology

Histological scoring at 8 months post-implantation revealed significantly improved total healing in rAAV5-IGF-I-treated defects (12.38 ± 1.30) compared with defects treated with rAAV5-GFP chondrocytes (15.88 ± 1.64; P = 0.033), naïve chondrocytes (18.50 ± 1.49; P = 0.001), or fibrin alone (18.58 ± 0.68; P < 0.001) ( Table 4 ). There was no significant difference in total histology scores in defects treated with rAAV5-IGF-I or rAAV5-GFP.

Correlation between Assessment Modalities

Using the Spearman’s rho correlation coefficient, we found significant correlation between OCT and arthroscopy total scores (rho = 0.3, P = 0.043) and between arthroscopy total scores and histology total scores (rho = 0.349, P = 0.015). Histology total scores and gross tissue total scores were strongly correlated (rho = 0.435, P = 0.002) ( Fig. 2 ). MRI total scores were not significantly correlated with any other outcome variable. OCT total scores were not significantly correlated with gross pathology or histology total scores.

Figure 2.

Spearman’s rho correlation coefficients of (A) arthroscopy and OCT scores, (B) arthroscopy and gross pathology scores, (C) arthroscopy and histopathology scores, and (D) gross pathology and histopathology scores. OCT = optical coherence tomography.

Discussion

In this study, we investigated the usefulness of OCT and MRI assessment of cartilage repair tissue. We found that OCT and arthroscopic scoring of early repair tissue at 8 weeks post-implantation were significantly correlated. However, arthroscopic scoring was correlated with both gross and histological scoring of repair tissues performed at 8 months post-implantation, while OCT data were not. This may suggest that visual inspection and manual probing of repair tissue is a better predictor of long-term cartilage repair quality following ACI. Interestingly, we also found that MRI scores were not correlated with any other assessment variable.

Several studies^9,15,17
-19 have evaluated OCT for the assessment of early cartilage degeneration and damage; however, few studies have investigated the use of OCT to evaluate early repair tissue in chondral defects. Chu et al.²⁰ demonstrated that OCT was very accurate at identifying early cartilage degeneration, with both quantitative and qualitative OCT being strongly correlated with arthroscopy. Similarly, the same research group found that OCT had good agreement with histopathology (κ = 0.80) on osteochondral cores collected from the tibial plateaus of osteoarthritic human knees.⁹ In this study, assessment of repair tissue 8 weeks post-implantation also showed moderate and significant correlation between OCT and arthroscopy score data. Despite this correlation, at 8 months post-implantation, arthroscopic scoring continued to be correlated with gross and histologic scoring, while OCT scores were not. Arthroscopic evaluation allows visual assessment of superficial and perilesional tissue as well as subjective assessment of tissue firmness and attachment through manual probing of the tissue, while OCT can provide information about the subsurface tissue including the presence of clefts and tears. It is possible that the quality of the superficial layer of repair tissue, including the perilesional interface, is most important in long-term success, as this area is the most vulnerable to trauma and shearing.

Although OCT images have an axial resolution of 10 µm, allowing for more detail than MRI and ultrasound, proteoglycan content and cellular detail cannot be seen.¹⁰ In addition, the OCT probe allows for assessment of tissue to a depth of 1.0 to 1.5 mm, which limits its applicability for evaluation of thicker cartilage and assessment of subchondral bone. This was especially relevant in this study as cartilage on the equine lateral trochlear ridge generally has a thickness of 2.0 mm or more.²¹ Hence, we were unable to see the depth of the subchondral attachment or tidemark using OCT, which may have limited its efficacy for evaluation of repair tissue. In addition, the signal-to-noise ratio of the OCT machine used in this study was low, which limits interpretation of deeper repair tissue. More recently, polarization-sensitive OCT devices have been used more frequently as they provide more detailed assessment of tissues.²² It should be noted that previous studies evaluating OCT in the horse in joints with thinner articular cartilage such as the metacarpophalangeal¹⁴ and intercarpal joint²³ have found that OCT was useful for examining the entire cartilage layer to the level of the subchondral bone.

As previously mentioned, few studies have investigated the utility of OCT for evaluating cartilage repair tissue. Han et al.²⁴ showed that OCT was highly correlated to histology in chondral lesions repaired with chondrocyte implantation in rabbits. In the rabbit study, histological evaluation of the repair tissue was performed immediately following OCT. While we found minimal correlation between OCT and histology, this is likely because the 2 assessments were separated by a 6-month healing period. Performing OCT at the end of the 8-month study period would have been interesting but was not feasible in this study. Further investigation is required, but it is possible that early in the repair process arthroscopy may be a more useful predictor for long-term repair outcome than OCT.

Equine cartilage closely approximates human cartilage as it is of similar thickness, and the LTR has similar load-bearing characteristics to the human femoral condyle.²¹ In addition, the biology and healing characteristics of equine cartilage are similar to man, whereas more commonly used small animal models, such as rabbits, often have exuberant and complete healing.²¹ For these reasons, the horse is becoming an increasingly common animal model for cartilage repair. Currently, there are no standardized quantitative systems for OCT of equine cartilage. In this study, we used a combination of previous subjective scoring methods described by te Moller et al.¹⁴ and Bear et al.¹⁵ that included evaluation of different morphologic characteristics and birefringence. A characteristic multilaminar pattern is seen in normal cartilage due to OCT form birefringenence.²⁵ Loss of birefringence, a phenomenon created by highly organized tissue causing back-reflection of polarized light in specific banding patterns, is one of the first signs noted in OA cartilage.²⁶ Currently, little is known about how OCT form birefringence changes during the cartilage repair process. Although subjective analysis of OCT images was easy to perform, the development of quantitative systems for equine cartilage would be useful in the development of this imaging technique in clinical research.

MRI provided a global evaluation of the joint and the repair tissue, including alterations of the subchondral bone. Despite this, MRI scores were not significantly different between any of the treatment groups. In addition, we found that MRI scores were not correlated with the other short-term (arthroscopy and OCT) or long-term (gross pathology and histopathology) assessment modalities used in this study. The MOCART classification system used in this study is the most commonly used system for evaluating cartilage repair.²⁷ Blackman et al.²⁸ found that MRI scans obtained 6 months postoperatively were most closely correlated with clinical outcome. This is partially the result of normal components of the cartilage healing process, such as subchondral edema, joint effusion, and hyperintense signal, being considered abnormal MRI findings.²⁸ In the equine model, it is possible that active healing is still occurring 8 months postoperatively such that MRI studies still have an abnormal appearance across all groups.

Quantitative MRI, including T2 mapping to evaluate collagen content and orientation, and T1rho (T1ρ), to evaluate proteoglycan distribution, are useful techniques to objectively assess the biochemical content and structure of the ECM.²⁹ Unfortunately such quantitative imaging was not possible in this study due to the limitations of the magnet. Use of a higher strength magnet (3.0 T or higher) would have also provided more detailed images due to the increased signal-to-noise ratio.³⁰ Previous studies²⁰ have shown correlation between T2 mapping values and OCT; however, correlation was only found for superficial T2 mapping, rather than deep T2 mapping.

Study limitations need to be considered when examining the reported results. We did not quantify transduction efficacy in all transduced cultures and instead relied on preliminary experiments using the same vector in which consistent transduction efficacy was achieved. Although surgeons were blinded to treatment limb at second-look arthroscopy, it would be theoretically possible for surgeons to recall which limb received treatment and which limb received fibrin control 8 weeks prior based on unique horse color or appearance. In addition, surgeons subjectively selected “best” and “worst” repair tissue sites before performing OCT, which could introduce some selection bias. Finally, a high-field magnet with specialized sequences to evaluate cartilage would have provided superior information regarding repair and perilesional tissue.

In summary, this study found that arthroscopic assessment of early repair tissue in chondral defects may be the most useful predictor of long-term cartilage repair, and that OCT could be used to complement arthroscopy by providing subsurface, microscopic detail about the tissue. Certainly, OCT may be useful in clinical cartilage repair research in models with thinner cartilage, as it can increase our understanding of in vivo subsurface healing while avoiding destructive biopsy procedures. In this way, enhancing our understanding of in vivo cartilage repair can direct development of future therapies.

Supplemental Material

sj-docx-1-car-10.1177_19476035231154508 – Supplemental material for Correlation of Arthroscopic Grading and Optical Coherence Tomography as Markers of Early Repair and Predictors of Later Healing Evident on MRI and Histomorphometric Assessment of Cartilage Defects Implanted with Chondrocytes Overexpressing IGF-I

Supplemental material, sj-docx-1-car-10.1177_19476035231154508 for Correlation of Arthroscopic Grading and Optical Coherence Tomography as Markers of Early Repair and Predictors of Later Healing Evident on MRI and Histomorphometric Assessment of Cartilage Defects Implanted with Chondrocytes Overexpressing IGF-I by Sarah A. Ciamillo, Sarah L. Pownder, Hollis G. Potter, Darko Stefanovski, Alan J. Nixon and Kyla F. Ortved in CARTILAGE

Footnotes

Acknowledgments and Funding

We would like to thank Dr Constance Chu for generously loaning the OCT probe. The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This study was funded by NIH 5R01-AR055373 (A.J.N.).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval

Ethical approval was provided by the Institutional Animal Care and Use Committee.

ORCID iD

Kyla F. Ortved

Supplementary material for this article is available on the Cartilage website at .

References

Mankin

HJ.

The response of articular cartilage to mechanical injury. J Bone Joint Surg Am. 1982;64(3):460-6. doi:10.1007/978-1-4471-5451-8_96.

Strauss

Goodrich

Chen

Hidaka

Nixon

AJ.

Biochemical and biomechanical properties of lesion and adjacent articular cartilage after chondral defect repair in an equine model. Am J Sports Med. 2005;33(11):1647-53. doi:10.1177/0363546505275487.

Brown

Johnston

Saltzman

Marsh

Buckwalter

JA.

Posttraumatic osteoarthritis: a first estimate of incidence, prevalence, and burden of disease. J Orthop Trauma. 2006;20(10):739-44.

Kon

Gobbi

Filardo

Delcogliano

Zaffagnini

Marcacci

Arthroscopic second-generation autologous chondrocyte implantation compared with microfracture for chondral lesions of the knee: prospective nonrandomized study at 5 years. Am J Sports Med. 2009;37(1):33-41. doi:10.1177/0363546508323256.

Ortved

Begum

Stefanovski

Nixon

AJ.

AAV-mediated overexpression of IL-10 mitigates the inflammatory cascade in stimulated equine chondrocyte pellets. Curr Gene Ther. 2018;18(3):171-9. doi:10.2174/1566523218666180510165123.

Begum

Ortved

Nixon

AJ.

AAV-5 provides more efficient transgene expression in chondrocytes grown in adherent and suspension culture. In: Orthopaedic Research Society 56th Annual Meeting, Seattle, WA, 2010 Mar 6-9.

Tyler

JA.

Insulin-like growth factor 1 can decrease degradation and promote synthesis of proteoglycan in cartilage exposed to cytokines. Biochem J. 1989;260(2):543-8.

Fortier

Lust

Mohammed

Nixon

AJ.

Coordinate upregulation of cartilage matrix synthesis in fibrin cultures supplemented with exogenous insulin-like growth factor-I. J Orthop Res. 1999;17(4):467-74. doi:10.1002/jor.1100170403.

Chu

Lin

Geisler

Chu

Pan

Arthroscopic microscopy of articular cartilage using optical coherence tomography. Am J Sports Med. 2004;32(3):699-709.

10.

Pan

Xie

Chu

CR.

Hand-held arthroscopic optical coherence tomography for in vivo high-resolution imaging of articular cartilage. J Biomed Opt. 2003;8(4):648-54. doi:10.1117/1.1609201 [doi].

11.

Argentieri

Burge

Potter

Magnetic resonance imaging of articular cartilage within the knee. J Knee Surg. 2018;31(2):155-65. doi:10.1055/S-0037-1620233.

12.

Nixon

Lust

Vernier-Singer

Isolation, propagation, and cryopreservation of equine articular chondrocytes. Am J Vet Res. 1992;53(12):2364-70.

13.

Dresdale

Rose

Jeevanandam

Reemtsma

Bowman

Malm

Preparation of fibrin glue from single-donor fresh-frozen plasma. Surgery. 1985;97(6):750-5.

14.

te Moller

NCR

Brommer

Liukkonen

Virén

Timonen

Puhakka

, et al. Arthroscopic optical coherence tomography provides detailed information on articular cartilage lesions in horses. Vet J. 2013;197(3):589-95. doi:10.1016/j.tvjl.2013.05.031 [doi].

15.

Bear

Szczodry

Kramer

Coyle

Smolinski

Chu

CR.

Optical coherence tomography detection of subclinical traumatic cartilage injury. J Orthop Trauma. 2010;24(9):577-82. doi:10.1097/BOT.0b013e3181f17a3b.

16.

Murata

Ishikawa

Sunaga

Saito

Sogawa

Nakayama

, et al. Osteochondral regeneration of the femoral medial condyle by using a scaffold-free 3D construct of synovial membrane-derived mesenchymal stem cells in horses. BMC Vet Res. 2022;18(1):53. doi:10.1186/S12917-021-03126-Y.

17.

Nebelung

Marx

Brill

Arbab

Quack

Jahr

, et al. Morphometric grading of osteoarthritis by optical coherence tomography–an ex vivo study. J Orthop Res. 2014;32(10):1381-8. doi:10.1002/jor.22673 [doi].

18.

Cernohorsky

Kok

Bruin

Brandt

Faber

Tuijthof

, et al. Comparison of optical coherence tomography and histopathology in quantitative assessment of goat talus articular cartilage. Acta Orthop. 2014;86(2):1-7. doi:10.3109/17453674.2014.979312.

19.

Chu

Izzo

Irrgang

Ferretti

Studer

RK.

Clinical diagnosis of potentially treatable early articular cartilage degeneration using optical coherence tomography. J Biomed Opt. 2007;12(5):051703. doi:10.1117/1.2789674 [doi].

20.

Chu

Williams

Tolliver

Kwoh

Bruno 3rd

Irrgang

JJ.

Clinical optical coherence tomography of early articular cartilage degeneration in patients with degenerative meniscal tears. Arthritis Rheum. 2010;62(5):1412-20. doi:10.1002/art.27378 [doi].

21.

McIlwraith

Fortier

Frisbie

Nixon

Equine models of articular cartilage repair. Cartilage. 2011;2(4):317-26. doi:10.1177/1947603511406531.

22.

Goodwin

Workman

Thambyah

Vanholsbeeck

Impact-induced cartilage damage assessed using polarisation-sensitive optical coherence tomography. J Mech Behav Biomed Mater. 2021;117:104326. doi:10.1016/J.JMBBM.2021.104326.

23.

Viren

Huang

Saarakkala

Pulkkinen

Tiitu

Linjama

, et al. Comparison of ultrasound and optical coherence tomography techniques for evaluation of integrity of spontaneously repaired horse cartilage. J Med Eng Technol. 2012;36(3):185-92. doi:10.3109/03091902.2012.663054 [doi].

24.

Han

Chu

Adachi

Usas

Huard

, et al. Analysis of rabbit articular cartilage repair after chondrocyte implantation using optical coherence tomography. Osteoarthritis Cartilage. 2003;11(2):111-21. doi:S1063458402908629 [pii].

25.

Liu

Harman

Giattina

Stamper

Demakis

Chilek

, et al. Characterizing of tissue microstructure with single-detector polarization-sensitive optical coherence tomography. Appl Opt. 2006;45(18):4464-79.

26.

Bear

Williams

Chu

Coyle

Chu

CR.

Optical coherence tomography grading correlates with MRI T2 mapping and extracellular matrix content. J Orthop Res. 2010;28(4):546-52. doi:10.1002/jor.20998 [doi].

27.

Marlovits

Singer

Zeller

Mandl

Haller

Trattnig

Magnetic resonance observation of cartilage repair tissue (MOCART) for the evaluation of autologous chondrocyte transplantation: determination of interobserver variability and correlation to clinical outcome after 2 years. Eur J Radiol. 2006;57(1):16-23. doi:S0720-048X(05)00288-3 [pii].

28.

Blackman

Smith

Flanigan

Matava

Wright

Brophy

RH.

Correlation between magnetic resonance imaging and clinical outcomes after cartilage repair surgery in the knee: a systematic review and meta-analysis. Am J Sports Med. 2013;41(6):1426-34. doi:10.1177/0363546513485931 [doi].

29.

Hirose

Nishioka

Nakamura

Oniki

Yamashita

Mizuta

T1ρ and T2 mapping of the proximal tibiofibular joint in relation to aging and cartilage degeneration. Eur J Radiol. 2012;81(10):2776-82. doi:10.1016/J.EJRAD.2011.11.019.

30.

Novakofski

Pownder

Koff

Williams

Potter

Fortier

LA.

High-Resolution methods for diagnosing cartilage damage in vivo. Cartilage. 2016;7(1):39-51. doi:10.1177/1947603515602307.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.17 MB