Risk factors of cartilage regeneration in patients who underwent high tibial osteotomy combined with microfracture

Background: One of the option for treating unicompartmental osteoarthritis (OA) is high tibial osteotomy (HTO). HTO which can shift the mechanical axis to the lateral compartment, can performed with microfracture which provides cells capable of producing cartilage. And this procedure is one of the common combinations that orthopaedic surgeons are in practice recently. The purpose of this study was to evaluate the degree of regeneration of cartilage after performing microfracture (MF) with high tibial osteotomy (HTO) after 2- year follow up and identify the factors influencing the regeneration of cartilage after the procedure. We also evaluated whether the regenerated cartilage status affects clinical outcomes. Methods: A total of 81 cases that underwent second-look arthroscopy at the time of plate removal after MF and HTO with a minimum two-year follow-up were included. The patients were divided into two groups according to femoral cartilage regeneration. Multivariable logistic regression analyses were performend to identify independent factors that influence cartilage regeneration. We also compared differences in functional outcomes between the two groups. Results: On the femoral side, grade I cartilage regeneration was found in seven (8.6%), grade II in 19 (23.5%), grade III in 26 (32.1%), and grade IV in 29 cases (35.8%), resulting in 26 (32.1%) patients belonging to well regenerated group (group A) and 55 (67.9%) to the poorly regenerated group (group B). Among factors, the size of the cartilage lesion (P=.011) and the presence of kissing lesions (P=.027) significantly affected cartilage regeneration. There were no statistically significant differences between group A and group B in terms of KSS and WOMAC scores. Conclusions: A large cartilage defect and the presence of kissing lesions are associated with poor cartilage regeneration after combined MF and HTO surgery. However, the quality of regenerated cartilage does not affect functional outcomes in patients with MF and HTO.

regeneration, risk factor Background One of the option for treating unicompartmental osteoarthritis (OA) is high tibial osteotomy (HTO) [1]. As the literature has shown comparable results for both UKA and HTO [2,3], the latter has its own advantages in preserving native joints. Because of its simplicity, medial opening wedge HTO (OWHTO) has been increasingly used in recent years [4,5].
Cartilage damage of the knee joint due to OA, is a challenging task for orthopaedic surgeons. While performing HTO, concurrent arthroscopic procedures like shaving, microfracture, drilling, and abrasion chondroplasty are introduced [6]. The microfracture is simple and the most common way to bring bone marrow cells to the articular surface for the regeneration of cartilage [6]. High tibial osteotomy which can shift the mechanical axis to the lateral compartment, can performed with microfracture which provides cells capable of producing cartilage. This procedure is one of the common combinations that orthopaedic surgeons are in practice recently. Meanwhile, there are a few studies that have examined the nature of regenerated cartilage after microfracture together with HTO [7,8,9]. Their results are conflicting with regard to the success of this combination in cartilage regeneration and functional outcomes.
In this study, we evaluted the status of regenerated cartilage based on second-look arthroscopic findings in patients with medial OA of the knee, treated with combination of arthroscopic microfractures (MF) and HTO. The purpose of this study was to evaluate the degree of regeneration of cartilage after performing microfracture (MF) with high tibial osteotomy (HTO) after 2-year follow up and identify the factors influencing the regeneration of cartilage after the procedure. We also analyzed whether regenerated cartilage affects clinical outcomes. We hypothesized that the regeneration of cartilage is affected by several factors and that regenerated cartilage affects clinical outcomes. Microfracture was done in both femoral and tibial side with ICRS grade III and IV. Among them, 137 patients underwent the plate removal and second-look arthroscopy. We excluded the patients with post-traumatic arthritis, combined ligament surgery or combined femoral osteotomy, or plate removal and patients who were followed-up less than 2 years, which left 81 patients. At the time of second-look arthroscopy, we evaluated the status of cartilage regeneration and reviewed patient records to analyze the factors affecting cartilage regeneration. All patients were operated at a single institute by two senior orthopaedic surgeons specialized in knee. A flowchart displaying the inclusion of patients in the study is shown in Figure 1.

Surgical procedure and rehabilitation
The subjects were operated in the supine position. Before performing HTO, a diagnostic arthroscopy was performed in all patients. Associated unrepairable meniscal tear was treated with partial or subtotal meniscectomy until the peripheral rim was stabilized. The cartilage defect was noted and categorized by ICRS grading. After arthroscopic shaving, the cartilage defect was debrided by using a small ring curette until an intact homogeneous peripheral rim was obtained. The central defect was then microfractured by a perforating owl with a 3 to 4 mm distance between the holes. After releasing the tourniquet pressure, we confirmed active bleeding from the microfracture holes before completing the surgery. Through anteromedial incision, a biplanar medial opening wedge osteotomy was carried out using a saw blade directed to the tip of the fibular head. Under fluoroscopic guidance, the opening of the deformity correction was targeted to achieve three to five degrees of a mechanical valgus postoperatively. The osteotomy was fixed with a MediFix locking plate (Medyssey., Jechon, Korea). The postoperative physiotherapy protocol consisted of a range of motion and muscle-strengthening exercises from the first postoperative day on, without weight-bearing mobilization, accompanied by the help of crutches from the second day, partial weight-bearing mobilization after 4 weeks, and full weight-bearing mobilization after 8 weeks. A knee brace was used until full weight bearing was initiated.
When the osteotomy was healed completely, at about two years after index surgery, the plate was removed and second-look arthroscopy performed at the same time to examine the regenerated cartilage's characteristics. The cartilage regeneration was then classified according to the Cartilage Repair Assessment of ICRS Cartilage Injury Evaluation Package by the same orthopaedic surgeon who performed the previous osteotomy surgery. To investigate influential factors on cartilage regeneration, the patients were divided into two groups according to their cartilage status on second-look arthroscopy: group A (Well regenerated: grades I or II, Figure 2) and group B (Poorly regenerated: grades III or IV, Figure 3). The patients, radiological and articular factors affecting cartilage regeneration were then assessed in those two groups. Functional assessment was carried out using

Knee Society Score (KSS) and the Western Ontario and McMaster Universities
Osteoarthritis Index score (WOMAC), recorded preoperatively and at the last follow-up to compare the group A and group B.

Assessment of factors
We retrospectively assessed patient factors, including age, sex, body mass index (BMI),

Statistical analysis
Statistical analysis was carried out using SPSS Statistics for Windows (version 22; IBM, Armonk, NY). A univariate and multivariate logistic regression model was employed to analyze the factors affecting cartilage regeneration. Postoperative functional improvement was assessed using paired sample t tests. To compare functional outcomes between the two groups, an independent sample t-test was calculated. A P value < .05 was considered statistically significant.

Results
The demographic data are shown in Table 1. There were 73 patients (22 male and 51 females) with eight bilateral cases for a total of 81 cases.
The distribution of cartilage lesions on femoral and tibial articular surfaces as per ICRS classification at the time of index surgery is shown in

Assessment of factors
In terms of influential factors (patients, radiological and articular factors) for cartilage regeneration on the femoral side, univariate logistic regression analysis is shown in Table   4. and multivariate logistic regression analysis is shown in Table 5.
The two most important factors affecting cartilage regeneration were the size of cartilage lesions (≥ 2.5 cm2 ) (P = .011) and the presence of a kissing lesion (P = .027).

Functional Assessment
The mean preoperative KSS and WOMAC scores were 67.5 ± 11.2 and 33.1 ± 15.4, respectively. At the most current follow-up, KSS and WOMAC scores improved to 87.4 ± 9.2 and 11.4 ± 7.7, respectively. The pre-and postoperative values' difference was significant at the last follow-up (P < .001). There was no statistically significant difference between the groups A and B in terms of postoperative functional scores (KSS, P = .65; WOMAC, P = .89).

Discussion
The most important finding of our study is that although Cartilage Repair Assessment grade III or IV cartilage regeneration could be found in about one third of all cases, there were no functional benefits compared to poor cartilage regeneration. The main focus of this research was to determine the factors associated with cartilage regeneration and we found that cartilage lesions ≥2. 10-year follow-up and reported that cartilage lesions >2 cm2 and a preoperative mechanical axis which bisected the total width of the tibia more than 25% were associated with poor results. But BMI and partial meniscectomy was not contributing factors [13].
According to the patient factors, consistent with Bae et al.
[13], we could not find any effect of BMI on cartilage regeneration, contrary to Kumagai et al. [11] and Kim et al. [12].
The relatively low BMI of our study sample might have been the reason why there was no association with cartilage regeneration. In a study of arthroscopic abrasion arthroplasty combined with HTO, Akizuki et al. showed that age was the only factor influencing tibial cartilage [14]. Though some studies have demonstrated a better cartilage regeneration in younger individuals, especially < 35 years, our investigation revealed no age effect since we had included patients with high-grade OA, all of them above 40 years of age [6].
A large varus deformity as determined by the preoperative HKA angle is also one of the contributing factors for the regeneration of the cartilage in most studies. Kumagai et al. and Koshino et al. have also shown that mature regeneration was observed more frequently in the knees with more than 1° of mechanical valgus angulation after osteotomy than in those with less than 1° [15]. However, Tsukada and Wakui recently reported no significant differences in the ratio of cartilage repair between overcorrected (defined as knees with mechanical valgus angulation ≥8°) and moderately corrected (defined as knees with mechanical valgus angulation ≤8°) knees [16]. Our study did not show any association between clinical outcome and cartilage regeneration with pre-and postoperative HKA angles as well as the correction angle. The goal of our corrective osteotomy was to achieve three to five degrees valgus angulation and hence most of our patients had overcorrected knees. It is well known that an open wedge HTO tends to increase, and a closed wedge HTO tends to decrease the postoperative slope [17,18].
Although the tibial slope increased after surgery, neither the preoperative tibial slope nor its change affected cartilage regeneration postoperatively.
Meniscal root tear is commonly associated with high-grade degenerative arthritis. Menisci are important for resisting axial load by generating hoop stresses. Meniscal root tear alters the knee's biomechanics [19,20,21,22]. A root tear of the meniscus is almost equivalent to total meniscectomy, increasing tibiofemoral contact pressure as demonstrated by Allaire et al. [19]. Takahashi et al. showed that tears in the posterior root of the medial meniscus were independent risk factors for cartilage degeneration as shown on T1p MRI scans [14]. Though statistically not significant, most cases with poor cartilage regeneration in our study had a root tear of the medial meniscus. Krych et al. reported that non-operative management of the medial meniscus root tear is associated with worsening arthritis and poor functional outcomes at a five-year follow-up [23]. Most of the root tears in our study were managed with either partial or subtotal meniscectomy to obtain a stable peripheral rim. Sterett et al. reported that patients with a tear of the medial meniscus at the time of chondral resurfacing were 9.2 times more likely to undergo TKA than those without tear [24]. But similar to Bae et al.[13], our study also showed that meniscectomy in general does not affect cartilage regeneration.
Among all factors, the most significant were the size of the cartilage lesion and the presence of a kissing lesion. In our study, a cartilage defect ≥ 2.5 cm2 was the most consistent factor associated with poor cartilage regeneration. Moreover, when the kissing lesions is present, it can possibly affect cartilage regeneration by disturbing the stable position of marrow stem cells during the regeneration process. Like in our study, most investigations assessing cartilage regeneration have shown that a larger cartilage defect is associated with poor cartilage regeneration [11,13,24]. Similar to Schuster et al.[25], our study revealed that there were no statistically significant differences in functional outcomes between well regenerated and poorly regenerated groups. This can be interpreted that postoperative functional improvement is mainly due to the medial compartment's unloading after HTO.

Limitations
This study had several limitations. First, it was a retrospective study with limited cases.
Second, tibial cartilage regeneration was not taken into account. Third, cartilage factors were not treated in detail. Fourth, we did not compare with the control group which patients who had undergone HTO without MF.  The values are the number of cases.   Figure 1 Flowchart of this study