1 Introduction

Knee osteoarthritis (KOA) is a common degenerative joint disease that affects approximately 43 million people in the United States, especially in the elderly.^1,2 Oxford-UKA, as a minimally invasive surgical method in the treatment of KOA, has been developed over several decades. Compared to total knee arthroplasty (TKA), the Oxford-UKA has many theoretical advantages including the preservation of bone stock,³ more rapid recovery, and rehabilitation,^4,5 etc. Pandit H et al.⁶ reported that the survival rate of phase III Oxford UKA was 97% at seven years. Many factors have a certain effect on the postoperative results of Oxford-UKA, including the degree of pre-operative articular cartilage degeneration, the choice of surgical indications by the operator, and postoperative rehabilitation. In addition, the optimal choice of UKA prosthesis during operation and the post-operation limb alignment are the key factors that determine survival rate.^7,8

With developments in implant design and surgical instrumentation, the Phase III Oxford-UKA could offer five femoral size options (extra-small XS, small S, medium M, large L, and extra-large XS) and seven tibial size options (AA, A, B, C, D, E, F) to suit different patients. Available prosthesis sizes and specific instruments could provide more precise bone removal.⁹

The results from several studies have demonstrated that it is difficult for surgeons to select the optimal femoral component based on a single method.^10–12 Thus, there is a potential risk of selecting an over- or undersized component.

The oversized femoral component may be an important fact that contributes to the degeneration of the patellofemoral joint.¹³ Furthermore, the increased flexion in association with an “extended” femoral component may contribute to early loosening of the femoral component.¹⁴ However, a downsized femoral component that is too small in dimension can cause insufficient femoral condyle coverage, which may cause the accelerated wear of the posterior femoral condylar under high flexion of the knee. The excessive overhang of tibial component may cause soft tissue irritation and pain, associating with worse clinical outcomes at five years after surgery.¹⁵ By contrast, if the tibial component is too small may increase the risk of tibial component subsidence and loosen due to inadequately supported.¹⁵ Therefore, appropriate sizing of the UKA prothesis plays an essential role in the mid- and long-term survivorship of UKA patients.

In addition, the UKA prosthesis was designed according to the Western population. Different from the Western population, the average femoral size is smaller and the posterior femoral condyle offset ratio is larger in the Chinese population.¹⁶ Besides, the size of the tibial prosthesis is larger in Western patients compared to Chinese patients.¹⁷ Therefore, the prosthesis of Oxford-UKA that used in Chinese patients would be smaller theoretically. The distributions of UKA prosthesis based on Asian patients might be different from Western populations in general. But there are few studies to investigate the distributions of Oxford-UKA prosthesis in Asian patients.

The aims of our study are to investigate prosthesis distributions of Oxford-UKA in Chinese patients. Besides, the factors affecting the selection of femoral prosthesis will be summarized and analyzed. We hope that our data will be helpful to offer a reference for surgeons preoperatively.

2 Patients and methods

Between January 2010 and December 2019, 1909 patients (2307 knees) who accepted Oxford-phase III UKA were recruited for this study at the same institution. The prosthesis information of all patients, including the femoral, tibial, and matchings of femoral-tibial components were recorded in detail.

Additionally, from January 2014 to December 2019, the proportion of Oxford-UKA to all knee arthroplasties (Oxford-UKA, Sled UKA, and total knee arthroplasty) was also investigated in our center.

Due to general information (gender, height, and weight) were missing for some patients, 1937 knees undergoing Oxford-UKA were recorded based on gender and 711 cases were registered on the basis of gender, height, and weight. In this study, a patient with bilateral UKA was considered as two separate cases (Fig. 1). The cohort of 711 cases included 153 men, 558 women with an average age of 71.93 years (ranges: 51–93 years), average height of 160.28 cm (ranges:138–185 cm), average weight of 68.57 kg (ranges: 42.5–120 kg) and average body mass index (BMI) of 26.68 kg/m² (ranges: 18.56–51.11 kg/m²).

Close

Fig 1 Flow diagram of statistical methods. A patient with bilateral Oxford-UKA was considered as two separate cases.

Export to PPT

3 Results

3.1

3.1 The prosthesis distribution information

According to 1909 patients (2307 knees), the results demonstrated that the distributions of femoral components as follows: extra-large (XL) 0, large (L) 1.08%, medium (M) 26.09%, small (S) 59.64%, extra-small (XS) 13.18% (Fig. 2). The distributions of tibial components were: F 0, E 0.69%, D 7.80%, C 19.59%, B 24.79%, A 34.16%, AA 12.96% (Fig. 2). The matchings of femur-tibial components were: L-E 0.52%, L-D 0.52%, M-E 0.17%, M-D 7.28%,M − C 16.60%, M − B 1.95%, M-A 0.13%, M-AA 0.04%, S-C 2.99%, S-B 22.67%, S-A 31.12%, S-AA 2.82%, XS-B 0.17%, XS-A 2.90% and XS-AA 10.10% (Fig. 3). The optimal matchings between tibial and femoral components were: XL with F; L with E; M with C and D; S with A and B. Patient's gender and height based on gender play important roles in the selection of femoral and tibial components (P＜0.05) (Table 1).

Close

Fig 2 The distributions of femoral and tibial component sizes based on 2307 knees undergoing Oxford-UKA.

Export to PPT

Close

Fig 3 The distributions of matching between femoral and tibial components based on 2307 knees undergoing Oxford-UKA.

Export to PPT

Table 1 The sizes distribution and affecting factors of femoral components based on 711 knees undergoing Oxford-UKA.

Independent variable	Knees	Large (L)	Medium (M)	Small (S)	Extra-Small (XS)	Fisher Value	P Value
Gender
Male	153	7	114	32	0	256.33	0.00
Female	558	0	70	433	55
Height (mm)
(Male)
＜170	69	2	37	20	0	13.08	0.02
≥170 to ＜175	57	2	56	9	0
≥175 to＜180	20	2	15	3	0
≥180	7	1	6	0	0
(Female)
＜155	119	0	5	84	30	54.36	0.00
≥155 to ＜165	374	0	45	306	23
≥165 to＜175	62	0	20	40	2
≥175	3	0	0	3	0

The results indicated that S-sized femoral component and A-sized tibial component were the most in female patients, accounting for 71.19% and 41.83% separately. Whereas, M-sized and C-sized were the most in male patients, about 76.33% and 43.50%. Gender was a crucial factor in selecting femoral and tibial components for the surgeon preoperatively (Table 2, Table 3, and Fig. 4).

Table 2 The distribution of femoral component sizes based on 1937 knees undergoing Oxford-UKA.

Gender	Knees	L	M	S	XS
Male	469	22 (4.69%)	358 (76.33%)	88 (18.76%)	1 (0.21%)
Female	1468	0	152 (10.35%)	1045 (71.19%)	271 (18.46%)

Table 3 The distribution of tibial component sizes based on 1937 knees undergoing Oxford-UKA.

Gender	Knees	E	D	C	B	A	AA
Male	469	15 (3.20%)	138 (29.42%)	204 (43.50%)	85 (18.12%)	26 (5.54%)	1 (0.21%)
Female	1468	0	26 (1.77%)	174 (11.85%)	412 (28.07%)	614 (41.83%)	242 (16.49%)

Close

Fig 4 The distribution of femoral and tibial component sizes in different genders (male and female) based on 1937 knees undergoing Oxford-UKA.

Export to PPT

3.2

3.2 The proportion of Oxford-UKA

From January 2009 to December 2019, the cases of Oxford-UKA have been increasing in our center. The number of cases in the last three years was approximately 18.5 times than the first three years. The average number of Oxford-UKA in our center was 326 cases per year from January 2014 to December 2019, accounted for 63.94% of all knee arthroplasties (Table 4,Fig. 5).

Table 4 The distribution of various knee arthroplasty cases in our center from 2009 to 2019.

Year		2009	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019
Oxford-UKA		13	17	32	68	222	212	275	320	435	331	382
	Medial	–	–	–	–	–	0	0	0	9	63	90
Sled-UKA
	Lateral	–	–	–	–	–	10	33	31	58	58	70
TKA		–	–	–	–	–	31	25	28	42	22	24
Percentage（%）		–	–	–	–	–	83.79	82.58	84.43	79.96	69.83	67.49

Close

Fig 5 The cases of Oxford-UKA in our center from 2009 to 2019.

Export to PPT

4 Discussion

In order to better know the prosthesis distributions of Oxford-UKA in Chinese patients. Our study investigated distributions of femoral, tibial, and matchings of femoral-tibial components. The data demonstrated that the patients who accepted Oxford-UKA were mainly women, approximately 75.79%, the S-sized femoral component was predominated, about 71.19%. Whereas the male patients accounted for 24.21%, and M-sized femoral components were the most, approximately 76.33%. The ratio of male and female patients who accepted Oxford-UKA surgery was about 1.0:3.1 in our study.

Data from registries and meta-analyses demonstrated that the percentage of women undergoing TKA is consistently between 55% and 77%.^20–22 Conversely, the gender distribution of patients undergoing UKA is variable compared with TKA, ranging from 38% to 83%.^23–31 This variability indicated that the selection criteria of Oxford-UKA are inconsistent for different surgeons. It is noted that tibial and femoral component sizes are different in female and male patients. Compared to male patients, the sizes of the femoral condyle and tibial plateau are smaller in female patients. This fact can be explained that the average height of men is higher than that of women in our series. Our results also suggested that the patient's height based on gender had a significant difference in selecting femoral components. Interestingly, a study from Lustig S et al. holds different views.³² Their study suggested that the patients' height only correlated with both femoral and tibial component sizes in male patients, not in female patients. They explained this difference was that less size variability used in female patients. They also noted that there was no necessary relationship between tibial and femoral component sizes for male patients. There was no finding in the female group. According to Fawzy E. et al.,¹⁰ using the patients' height based on gender preoperative to predict femoral component size was correct in 75%, acceptable in 25%, and was never unacceptable. In short, the patient's gender and height based on gender are two important factors that should be considered preoperative in order to select optimal femoral and tibial components for patients.

Different from the Phase I and phase II Oxford-UKA prosthetic components, there is only one size femoral component i.e. medium. Phase III has five sizes of femoral components, including extra-large (XL), large (L), medium(M), small(S), extra-small (XS). From the perspective of knee anatomy, different sizes could better match the femoral prosthesis. In our study, lacking femoral XL, femoral L matches tibial D and E, femoral M matches tibial C and D, femoral S matches tibial A and B, femoral XS matches tibial AA. Compared to Western patients, Chinese patients are relatively shorter. In this study, our results demonstrated that the M-sized femoral component was the most commonly used for male patients, the S-sized was predominated in female patients. Totally, the most common size was S-size (59.64% cases) in our study. Malhotra R et al.³³ investigated the distributions of femoral component sizing of Oxford-UKA in Indian patients reported that the most common size was XS for females and S for males. However, the most common size in Fawzy et al.’s study was M-size (54% cases).¹⁰ Obviously, the distribution of femoral components differs significantly between Asia patients and Western populations. According to our study, it is very necessary to prepare the S- and XS-sized femoral component for Chinese female patients (89.65% cases) and the M- and S-sized for Chinese male patients (95.09% cases) preoperatively.

It is difficult to determine the ideal size of the femoral component not only preoperatively but also intraoperatively when a UKA prosthesis is implanted with a minimally invasive approach.¹⁰ There were two methods that commonly used for predicting the sizes of femoral components previously, template prediction and gender combined height prediction. Especially for template prediction, the reliability of prediction and the relationship with the surgeon's experience are still under debate. According to Bothra V et al.,¹¹ the template prediction lacks reliability about the place for preoperative radiological templating in UKA surgery. The results had no significant difference with the surgeon's experience. However, the results from Kasis AG et al.¹² holds a different opinion. They assessed the precision and accuracy of the templating system for UKA. The results revealed that the accuracy of the template prediction was closely related to the experience of observers. In addition, there are still some errors in using the tibial components to predict the femoral components. The reason is that the optimal tibial component is closely related to the depth of the vertical osteotomy during the operation, which depends on the experience of surgeons. The study from Fawzy E et al.¹⁰ reported that the template prediction was correct in 67%, height alone was correct in 56% and height based on gender was correct in 75%. So, they suggested that using both to assess femoral component sizes in order to reduce errors. Additionally, their results also indicated that the learning curve of surgeons was also essential. The junior surgeon got the sizes right in about 52% of cases, whereas the senior surgeon was approximately 67%.¹⁰ Fawzy E et al.¹⁰ also revealed that the patient's height based on gender was a better selection in predicting femoral components, but height alone about 56%. The optimal femoral component is cruciate for obtaining good long-term survivorship in UKA patients. An oversized femoral component may contribute to the degeneration of the patellofemoral joint¹³ and early loosening of the femoral component.¹⁴ In our study, the two sizes of tibial components (A and B) accounted for the higher proportion, 34.16%, and 24.79% respectively. This result also coincides with the highest proportion of S-sized femoral components (59.64%) because femoral S and tibial A, B are optimal matchings.

Certainly, an appropriate size of the tibial component also plays an essential role in promising survivorship of Oxford-UKA. According to Chau R. et al.,¹⁵ excessive overhang (≥3 mm) has a significant relationship with worse clinical outcomes at 5 years after surgery. In addition, under-hang of tibial component with an increased risk of tibial component subsidence and loosening. The main possible reason is that the load is transmitted mainly through the cancellous bone rather than the stronger cortical bone.¹⁵ In Oxford-UKA manufacturer, reaching the medial cortex or overhang slightly is recommended.³⁴

Several limitations to our study are identified. Firstly, all operations were performed by one surgeon, which may contribute to a certain deviation in the results. Secondly, there were no XL-sized femoral component and F, E-sized tibial components in this study. Even if these limitations, our study offers the primary distribution maps of femoral, tibial, and matchings of tibial-femoral components in Chinese patients. These results can provide a reference for researchers and clinicians to select an appropriate Oxford-UKA prosthesis preoperatively. In addition, our study also pointed out that the phase III Oxford-UKA components had specific distribution characteristics in Chinese patients.

5 Conclusion

In our center, the UKA is a major surgical procedure for the treatment of patients with end-stage osteoarthritis, limited to a single compartment of the knee. In Chinese patients, the size of femoral components is mainly small (S) for women, medium (M) for men. The tibial components of female patients are mainly A and B, whereas C is predominant for male patients. The more commonly used matching forms are S-A and S-B. These findings of our studies could offer a better reference for surgeons in selecting an optimal UKA prosthesis.

Acknowledgement

Neil.

Funding source

This study was supported by grants from Shanghai Health System Advanced Suitable Technology Promotion Project Plan (NO.2019SY060); Shanghai Municipal Commission of Health and Family Planning (NO. 201840187); Shanghai Municipal Science and Technology Commission (NO.18411969800); Shanghai Municipal Health Commission (NO.201940249).

Ethics approval and consent to participate

N/A.

Consent for publication

All authors have read the final version and give consent for the article to be published.

Availability of data and materials

The datasets used and analyzed during the current study are available from the first or corresponding author on reasonable request.

CRediT authorship contribution statement

Fang-xing Wang: Conceptualization, Data curation, Investigation, Software, Writing – original draft. Hua-ming Xue: Funding acquisition, Validation, Writing – review & editing, Supervision. Tong Ma: Funding acquisition, Validation, Writing – review & editing, Supervision. Tao Yang: Supervision. Tao Wen: Supervision. Yi-hui Tu: Conceptualization, Funding acquisition, Validation, Writing – review & editing, Supervision.