Abstract
Purpose:
This study evaluated the functional and clinical outcomes of the subjects with symptomatic osteochondral lesions on articular surface of patella, who were treated with autologous osteochondral transplantation (AOT) method.
Study plan:
AOT method was applied for the treatment of 14 subjects (eight men and six women; mean age 29.7 years; range 19–49 years) with symptomatic patellar osteochondral lesions between March 2008 and April 2013. After a mean follow-up period of 3.7 years (range 32–80 months), pre- and postoperative clinical and functional evaluations of the patients were performed using Visual Pain Scale (VPS), Lysholm Knee Scoring Scale, and Kujala Anterior Knee Pain Scale. Wilcoxon test was used for statistical evaluation of pre- and postoperative outcomes. Improvement of the lesions was assessed by magnetic resonance imaging (MRI) at year 1 postoperatively at the earliest.
Results:
The mean lesion size was 1.32 cm2 (range 0.8–1.8 cm2). The mean pre- and postoperative VPS values were calculated to be 75.5 ± 12.32 (range 46–92) and 17.57 ± 10.21 (range 0–40), respectively (p < 0.01). The mean pre- and postoperative Lysholm knee scores were 44.57 ± 9.35 (range 26–65) and 80 ± 6.9 (range 70–94), respectively (p < 0.01), and the mean pre- and postoperative Kujala anterior knee pain scores were 48.21 ± 7.78 (range 38–68) and 78.42 ± 7.06 (range 70–96), respectively (p < 0.01). MRI taken at year 1 postoperatively showed that the autograft bone tissue was sufficiently incorporated into the recipient site in all patients; an even articular surface was formed, but the thickness of the cartilage tissue was mostly uneven between the adjacency of the recipient site and the autograft, which caused no negative effect on clinical and functional outcomes.
Conclusions:
Despite the difference in thickness of the cartilage tissue between the recipient and the donor site, the AOT technique for the treatment of patellar osteochondral lesions resolves the symptoms of the patient and ensures an apparent functional and clinical improvement even if an articular surface could be created.
Introduction
About 11–23% of all osteochondral lesions in the knee joint are observed in the patella. 1,2 Because of the difficulties in their treatment, patellar osteochondral lesions are treated using nonsurgical methods by many orthopedists. These lesions, when not properly treated, cause early degenerative changes. 3
Different methods have been described for the surgical treatment of patellar osteochondral lesions; however, there is a limited number of studies on this subject in the literature. Debridement and abrasion represent the most commonly preferred surgical methods because of their easy application and less invasive nature. However, healing with the fibrous cartilage with poor biomechanical characteristics and subsequent occurrence of osteoarthritis led to development of different treatments. 4 Therefore, there has been a shift in preferences over the last 20 years toward the various treatment methods that allow obtaining a hyaline and hyaline-like cartilage tissue. 5 –13 Among these, the autologous osteochondral transplantation (AOT) and autologous chondrocyte implantation (ACI) have particularly become popular with the recent studies in the literature.
The AOT method is commonly used for nonpatellar osteochondral lesions in the knee joint as it ensures chondral and subchondral integration and allows for viable chondrocyte transfer. However, many studies still report some disadvantages like problems with donor site and discrepancies between the thickness of the transferred cartilage tissue and the thickness of the cartilage tissue in the recipient site. 14
We intended to use the AOT method, which is a commonly used treatment for osteochondral lesions, in patellar osteochondral lesions, and to compare the outcomes with those of other treatment methods in the literature and thus demonstrate the efficacy of treatment.
Materials and methods
In this retrospective cohort study, 14 subjects (eight men and six women; mean age: 29.7 years; range: 19–49) were included with symptomatic patellar osteochondral lesions, who were treated with the AOT method between March 2008 and April 2013. Right knees of nine patients and left knees of five patients were affected. According to the International Cartilage Repair Society, the patients who had grade 4a and grade 4b osteochondral lesions were included in this study.
Study included patients aged 18–55 years with a lesion size of above 0.8 cm 2 with persistent anterior knee pain without any alignment deformity. Patients who have osteochondral lesions more than 6 months and do not respond to other medical and physical treatments were included in this study. We did not include patients with lesion lesser than 0.8 cm2 and who had treatment for debridement and microfracture. We did the measurements of the lesions by the magnetic resonance imaging (MRI) examination. Also at the time of operation, we made the measurements and compared with the MRI measurements. There was no significant difference between the measurements. Exclusion criteria were age outside the range of 18–55 years, presence of lesion smaller than 0.8 cm2, anterior cruciate ligament or meniscus lesion, body- mass index (BMI) ≥30 kg/m2, active infection, patellar instabilities, patellar tilts, internal torsion of femoral neck, varus–valgus deformity, and systemic inflammatory arthritis such as rheumatoid arthritis.
Preoperative evaluation of the knee joint was conducted for all patients with a detailed examination. Patellofemoral instability, patellar tilt, patellar alignment, anterior cruciate ligament, and meniscus examinations were performed. Complete blood count, sedimentation rate, C-reactive protein, and rheumatoid factor were investigated in all patients to determine the presence of systemic inflammatory arthritis and infection.
Routine preoperative scans were taken from anterior–posterior, Merchant, and lateral knee in 30° flexion. Patellar alignment deformity was checked using Caton–Deschamps index and the varus–valgus deformities were inspected by the measurement of anatomical axis. Pathologies of the soft tissues like anterior cruciate ligament and meniscus lesions were investigated in all patients by MRI. Magnetic resonance (MR) images were also used to examine the site, size, and characteristics of patellar osteochondral lesion. MRI was ordered for all patients at year 1 postoperatively at the earliest. Thus, the degree of incorporation of the transferred graft to the recipient site and the surface of the treated joint were evaluated. In addition, healing potential of the donor sites was examined.
All patients were given spinal–epidural anesthesia. Then, the anterior cruciate ligament and meniscus were evaluated by routine arthroscopic examination in the supine position. Patellar osteochondral defect was also examined during the diagnostic arthroscopy and data on site, size, depth, shape of the lesion, and suitability of the donor site were obtained

A 24-year-old male patient’s arthroscopic image of the lesion 5 weeks after a sports injury. Osteochondral lesion of 1 cm2 in the central–medial facet of patella.
Based on the clinical, radiological, and arthroscopic examination, arthrotomy was performed by medial or lateral parapatellar incision beginning from the superior base of patella and ending in the inferior apex of patella by taking the affected articular surface as reference.
In this study, we used the single-use transplantation systems (OOTS), which allow for transplantation of grafts with a diameter of 6 or 8 mm. We did not prefer larger diameter transplantation systems because of the tilt in articular surface of the patella and the differences in cartilage thickness. Pretransplant removal procedure was applied in the site with the osteochondral lesion using a curette, surface area of the lesion was calculated, and the diameter and number of transplants were decided. We use the sulcus terminals of the ipsilateral femur condylar as a donor site. Prior to the transplantation, 0.8-mm Kirschner wires were advanced through the recipient sites until they reached a distance of 2 mm to the anterior cortex, and the depth of recipient site was determined. Based on these criteria, the recipient site was prepared with the aid of suitable diameter AOT systems.
Following the evaluation of intercondylar notch in flexion and extension, osteochondral grafts were obtained from the sites, such as intercondylar notch, superior lateral trochlea, and anteromedial of medial femoral condyle, where contact surface of femur with patella and tibia is smaller. 15 Diameter and depth of grafts to be taken were determined based on the diameter and depth measured at the recipient site. Graft was taken from a depth of 1 mm deeper than the depth measured at the recipient site to allow for a more strict detection and thus to prevent the potential for an early graft delamination. 16,17
Following the placement of osteochondral grafts, suitability of articular surface was checked (Figure 2(a) and (b)). A drain was not used for the patients considering that the hematopoietic stem cell activity from the donor site would contribute to recovery. 18

(a) Osteochondral lesion at patella medial surface. (b) After autogenous osteochondral transplantation.
Once the spinal anesthetic had worn off, isometric exercises were performed by describing to the patients in their bed. Joint motions were initiated at day 2 postoperatively under the supervision of a physiotherapist with the aid of a continuous passive motion machine, allowing 0° extension and 90° flexion. After 4–6 weeks, patients were allowed to progress to full weight-bearing. Vigorous activities such as running, contact sports, and jumping were not allowed until 6 months after surgery.
Wilcoxon signed-rank test was used to evaluate the results of pre- and postoperative Visual Pain Scale (VPS), Lysholm Knee Scoring Scale, and Kujala Anterior Knee Pain Scale. Statistical analysis was performed using SPSS version 18 package program. A p-value of <0.01 was considered to be statistically significant.
Results
The mean surgery duration was 75 min (range 65–90 min), and the mean incision length was 7 cm (range 6–8 cm). A medial parapatellar incision was used in eight patients and lateral parapatellar incision was used in six patients. The mean lesion size was measured to be 1.32 cm2 (range 0.8–1.8 cm2). Based on the size of lesions, a minimum of two and a maximum of three AOT grafts of 6 and 8 mm were combined and transferred to the recipient site (Table 1).
Demographic data, arthrotomy type, the lesion size, diameter, and the number of the transferred OOTSs.
OOTS: single-use transplantation systems.
Osteochondritis dissecans was detected in the medial femoral condyle of a patient and thus, an 8-mm osteochondral autograft was transferred within the same session. Furthermore, lateral tilt was detected in two patients, and lateral retinacular release was performed.
The functional scores of the patients are provided in Tables 2 and 3. The mean pre- and postoperative VPS values were calculated to be 75.5 ± 12.32 (range 46–92) and 17.57 ± 10.21 (range 0–40), respectively (p < 0.01). The mean pre- and postoperative Lysholm knee scores were 44.57 ± 9.35 (range 26–65) and 80 ± 6.9 (range 70–94), respectively (p < 0.01). The mean pre- and postoperative Kujala anterior knee pain scores were 48.21 ± 7.78 (range 38–68) and 78.42 ± 7.06 (range 70–96), respectively (p < 0.01).
The statistical values of functional scores pre- and postoperatively.
GASPreO: GAS PreOperative; GASPostO: GAS PostOperative; LSYHOLMPreO: LSYHOLM PreOperative; LSYHOLMPostO: LSYHOLM PostOperative; KUJALAPreO: KUJALA PreOperative; KUJALAPostO: KUJALA PostOperative.
Preoperative and postoperative statistical analysis with Wilcoxon Test.
aPositive according to sort
bNegative according to sort
Recipient and donor sites were inspected by postoperative MRI examinations. Sufficient incorporation of autograft into the recipient site was observed in all patients. None of the patients experienced graft delamination. Articular cartilage surface was even, but there were discrepancies in cartilage thickness between the donor and the recipient sites in most of the patients. It was determined that the cartilage thickness of the transferred autograft was less than the cartilage thickness in patellar surface. However, this caused no negative effects on clinical and functional outcomes. None of the patients experienced unevenness of the articular cartilage surface (Figure 3(a) and (b)). At the follow-up examinations and control MRI, we did not have any complications at the donor sites.

(a) MR imaging in the axial plane and chondral delamination in the left knee. (b) MR imaging on the axial plane taken at month 19 postoperative control of the patient. MR: magnetic resonance.
Discussion
The cartilage tissue that forms the articular surface of the patella is hyaline cartilage and has a poor capacity for self-repair. 19 The main goal in the treatment of patellar osteochondral lesions is to ensure healing with hyaline, the original articular cartilage, and hyaline-like cartilage tissue without disrupting the biomechanics of patellofemoral joint. Today, treatment for patellar osteochondral lesions is still a matter of debate. Besides the treatment methods such as debridement, abrasion chondroplasty, and microfracture which ensure healing with the low-quality fibrous cartilage, other methods are also used including AOT, ACI, allograft transplantation, perichondrial, periosteal tissue transplantation, and treatment methods that enable healing with a cartilage that is close to original articular cartilage using biomaterials. 5,6,12,13,19 –25
Easy-to-use and cost-effective methods like debridement and abrasion have been used for the treatment of patellar osteochondral lesions for a long period of time. However, studies showed the recurrence of lesions in patellofemoral joint in the mid- to long-term and frequent occurrence of patellofemoral degeneration following these treatments. 26,27
Microfracture is a relatively popular method in the treatment of osteochondral lesions; however, it is rarely performed for patellar lesions because of the difficulties with its application. As a disadvantage, it ensures healing with the fibrous cartilage. 28 –30 In animal experiments, histopathological studies also demonstrated that stem cell stimulation takes place after microfracture, but healing occurs with fibrous cartilage. 24,31
Some retrospective studies showed that the long-term outcomes of the treatment with microfracture was not successful. Kai Mithoefer et al. applied microfracture on knees of 48 patients who were then followed up for 24 months and used international knee scoring scale, daily life scores, and MRI to evaluate the outcomes. This study showed that the symptoms recurred in the short-term, but the recurrence associated with the BMI in the long-term and also MR images showed the recurrence of the cartilage damage in the long-term. 32
Difficulties in arthroscopically assisted techniques complicate the application in patellar osteochondral lesions. In addition, the positive effect of stem cell stimulation in the microfracture is limited in the patella.
Although rare, autologous soft tissue transplants are used to obtain hyaline and hyaline-like cartilage tissue in articular surface repairs. Among these, the periosteal and perichondrial grafts represent the most commonly used applications. 33 –36 Studies demonstrated that the treatment with perichondrial graft led to symptom reduction in the early period, but the success rate decreased over the long-term. 34,35
Periosteal graft is a more preferable and promising method in the treatment of patellar osteochondral lesions because of both the chondriogenic and osteogenic properties. 36 Lorentzon et al. 37 applied subchondral debridement and drilling in 26 patients with patellar osteochondral lesions, followed by treatment with the periosteal graft and obtained excellent outcome in 17 patients (65%), good outcome in 8 patients (31%), and poor outcome in 1 patient (4%) after a mean follow-up period of 42 months. However, this treatment method has not become widespread because of the insufficient number of studies in the literature. On the other hand, ACI, the most popular method in the treatment of patellar osteochondral lesions in recent years, has become the commonly used method as it enables healing with a hyaline and hyaline-like cartilage tissue. This treatment method was first used in clinical setting by Peterson et al. 38,39 in 1987. The first-generation ACI techniques were associated with a high incidence of complications due to the periosteal graft used for the surface closure. Later ACI techniques used double membrane derived from type 1 and type 3 porcine collagen to close the surface of defect and resulted in no hypertrophy of the membrane and associated complications. 40
Peterson et al. 38 conducted a case series studies of 61 patients where they treated femoral condyle and patellar cartilage lesions with the ACI method and published the results of a mean follow-up of 7.4 years. According to this study, they obtained good–excellent outcomes after 5–11 years of follow-up of 51 patients (84%), and within the same study, they performed arthroscopy after 54 months on average in 11 patients and demonstrated healing with hyaline cartilage in 8 patients (73%) and with fibrocartilage in 3 patients (27%).
In another study, Peterson et al. 39 published the 2- to 10-year results of 58 patients with osteochondritis dissecans of the knee joint who were treated with the ACI method and achieved good–excellent outcome in 91% of the patients.
Some studies found a considerable amount of complications following the ACI method. Study of Micheli et al. 41 demonstrated a high incidence of adhesions, arthrofibrosis, and hypertrophic changes following the ACI method and the other studies also demonstrated a need for an additional surgery in 17–57% of the cases for the above reasons. 42 –44 Another complication following the ACI method is delamination. In the literature, the mean incidence of delamination after ACI is 14%. 41 –44 This complication occurs within the first year of surgery and is detected by the control MRI exams. 45 Once it is detected, the implanted cartilage tissue must be removed and chondrocyte implantation must be repeated.
The ACI method still has many problems to be solved because of the need for a two-stage surgery, the additional cost of tissue culture laboratories, and the complications resulting from the nature of bio-tissues used.
AOT has been the preferred method for a long time, especially in the treatment of patellar lesions where both chondral and subchondral defects are observed. The greatest advantages of this method can be applied in a single session, allows transfer of viable chondrocyte-containing hyaline cartilage along with the bone tissue, ensures a robust fixation, and is cost-effective compared to the ACI method. Disadvantages include the limited donor site availability, uncertainty of healing status of the donor site in large lesions, and incompatibility between the cartilage thickness of donor and recipient sites.
Astur et al. 46 treated a case series of 33 patients with patellar osteochondral lesions using the AOT method and published the functional outcomes by Lysholm score, Fulkerson score, and Kujala knee score. Following a mean follow-up period of 30.2 months, pre- and postoperative Lysholm scores were 57 and 81, pre- and postoperative Fulkerson scores were 54 and 80, and the pre- and postoperative Kujala knee scores were 54 and 75. Three patients (9%) developed arthrofibrosis, and no other complication was observed. All these scores were statistically significant (p < 0.05). The same study showed good incorporation of the osteochondral grafts into the recipient site by 83% in MRI controls at 6 months postoperatively, which increased to 100% in MRI controls at 1 year postoperatively.
AOT was applied to patellar osteochondral lesions in a total of 14 subjects in our study and a control MRI was taken at 1 year at the earliest in all patients. Thus, we observed good incorporation of the grafts at the recipient site in all subjects. None of our patients experienced graft delamination. While the cartilage tissue thickness was mostly observed to be different between the recipient and the donor sites, no unevenness was detected on the articular surface
Cohen et al. 47 conducted a study in a case series of 17 patients with patellar osteochondral lesions who were treated by the AOT method with a mean follow-up period of 19.8 months and determined and published the pre- and postoperative functional outcomes by Lysholm, Fulkerson, and Kujala knee scores. They reported the pre- and postoperative Lysholm scores as 54.5 and 75.7 (p < 0.05), pre- and postoperative Fulkerson scores as 52.5 and 78.4 (p < 0.05), and pre- and postoperative Kujala scores as 49 and 74 (p < 0.05), respectively. No patient experienced a complication in this study.
In our study, the mean pre- and postoperative Lysholm knee scores were 44.57 ± 9.35 and 80 ± 6.9, respectively, and the mean pre- and postoperative Kujala anterior knee pain scores were 48.21 ± 7.78 and 78.42 ± 7.06, respectively. Pre- and postoperative statistical results of both scoring systems were calculated as p < 0. 01 and thus, a significant difference was detected. These values were similar with those of other studies in the literature. No complications were experienced by the patients in our study.
This study has some limitations. One of these limitations was the small number of patients. This was caused by the rare occurrence of the condition, selection of subjects with isolated patellar osteochondral lesions to standardize the study and conduct the study by a single surgeon at a single study site. Other limitations include the short duration of study and the absence of comparison with a control group. Another limitation to the study is to include a 49-year-old patient as the degenerative signs increases after 40 years . But this patient did not have any MR images about degeneration. We believe that there is a need for a study conducted in larger number of patients for a longer period of time by comparison with a control group for a more precise determination of efficacy of treatment with AOT on patellar osteochondral lesions.
Conclusion
Despite the difference in thickness of the cartilage tissue between the recipient site and the donor site, the AOT technique for the treatment of patellar osteochondral lesions resolves the symptoms of the patient and ensures an apparent functional and clinical improvement even if an articular surface could be created.
Footnotes
Authors’ note
All operations were carried out in Orthopaedics and Traumatology Clinic, Beypazarı Government Hospital, Turkey.
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.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
