Morphologic differences in knee bone cuts between Han Chinese and Mongolian Chinese and their correlation with current knee prostheses

INTRODUCTION

Total knee arthroplasty (TKA) is widely adopted to treat knee osteoarthritis and trauma, restoring normal knee function.[1] Beyond patient and surgical factors, the geometric matching of a properly designed knee prosthesis is crucial for clinical success.[2,3,4,5,6,7,8,9,10,11,12,13] If the prosthesis is smaller than the resected surface of the distal femur and proximal tibia, it may lead to subsidence and loosening. Conversely, if the prosthesis is larger than the resected surfaces, it may alter local soft tissue balance and cause impingement.[14] Therefore, the morphology of resected knee is vital for designing an optimal knee prosthesis.[14]

Several studies have indicated that most TKA prostheses, designed based on Caucasian anatomy, are not suitable for Asian patients.[2,12,13] Generally, Asians have smaller knees than Caucasians.[7,8,15] Mahfouz et al.[16] reported significant differences in three-dimensional knee morphology among Caucasian, African American and East Asian populations. Li et al.[15] found that ethnicity- and gender-related differences exist in the resected femur and tibia surfaces between Chinese and Caucasian populations. Even with additional prosthesis sizes and smaller size increments, these differences are not fully addressed due to disparities in aspect ratios between natural knees and artificial knee implants.[12] Compared with the knee morphology of Asian patients, the aspect ratio — defined as the mediolateral (ML) dimension divided by the anteroposterior (AP) dimension — of most current knee prostheses used for Asian patients is not suitable.[2,11,12,13] Several studies have indicated that ethnicity-specific prostheses could better match the knee morphology of Asian populations.[1,2,7,8,11,12,13,17,18] Therefore, understanding morphologic differences among ethnic groups is essential to improve the clinical outcomes of TKA. However, there is a lack of comparative knee morphological studies among varied Chinese ethnicities.

It is well known that there are significant morphological differences between male knees and female knees.[1,9,12,15,19,20] Growing evidence has indicated that females have smaller knee dimensions than males, and standard prostheses often cause mediolateral (ML) overhang in female patients.[9,15,19,20,21,22,23] Many studies have revealed that gender-specific knee prostheses are indispensable to better match the knee sizes of female and male patients.[1,8,10,11,12,13,24,25] Furthermore, currently available femoral and tibial prostheses do not perfectly match-fit all Chinese male and female patients.[8,12,14,15,21] The fit rates of femoral components were remarkably higher in both Chinese males and females when using gender-specific prostheses compared to standard options.[1] Therefore, gender differences should be considered in the design of knee prostheses for Chinese men and women. However, the dimensional and morphological differences in the knee joint between Han Chinese and Mongolian Chinese populations have yet to be determined.

In this study, we investigated the knee morphological differences in bone cuts by gender and ethnicity between Han Chinese and Mongolian Chinese populations. Additionally, we compared the dimensions of Chinese knees with those of typical knee prostheses. Furthermore, the morphological relationship between the femur and tibia was examined with respect to ethnicity and gender differences.

METHODS

A total of 74 normal knees were analysed in this study, including 44 normal Han Chinese knees (26 males, 18 females) and 30 normal Mongolian Chinese knees (16 males, 14 females). The average age of the participants was 36.0 ± 7.2 years (range 23–45). The Mongolian Chinese participants were recruited by experienced surgeons from the Inner Mongolia Autonomous Region of China, the primary area where Mongolian Chinese reside. The Han Chinese were recruited by experienced surgeons from central China, the primary area where Han Chinese reside. Participants were excluded if they had a history of congenital anomalies, femoral fractures, knee injuries or other knee pathologies. Informed consent was obtained from all volunteers. This study was approved by the ethics committees of the Second Affiliated Hospital of Inner Mongolia Medical University (reference no.: YKD2017021).

All knees were scanned using a helical computed tomography (CT) scanner (Light Speed 16, GE Medical System; General Electric Company, Cincinnati, OJ, USA) (120 kVp; 320 mA; 512 × 512 matrix; slice thickness, 0.625 mm). Participants were placed supine in the scanner, with both knees taped to the scanner platform in an extended position and the patella facing the ceiling. The scanning data were then imported into a medical imaging program (Mimics version 16.0; Materialise, Leuven, Belgium) for three-dimensional reconstruction of knee models. Bony cuts and measurements were conducted under the guidance of an experienced surgeon using Geomagic Studio version 12.0 (SculptCAD, Dallas, TX, USA). Morphologic data from five repeated measurements were used to calculate the average values for analysis.

The tibial mechanical axis was defined as the line connecting the centre of the knee to the centre of the ankle.[26] The proximal tibia was cut perpendicular to the tibial mechanical axis, with a resection depth of 6 mm below the medial plateau and a 7° posterior slope[2] [Figure 1a]. The tibial mediolateral (tML) dimension was defined as the longest medial-lateral line on the proximal tibial cut surface. This line was drawn parallel and collinear to the femoral epicondylar axis, defined by connecting the medial sulcus of the medial epicondyle, as described by Uehara et al.[17] The tibial middle anteroposterior (tAP) dimension was taken as the length of a line passing through the midpoint of the tML line and perpendicular to it. The tibial medial anteroposterior (tMAP) and tibial lateral anteroposterior (tLAP) dimensions were defined as the lengths of lines drawn parallel to the tAP line and passing through the posterior-most points of the medial and lateral tibial condyles, respectively [Figure 1b]. The tibial aspect ratio (tML/tAP) was calculated to analyse the shape of the knee.

Figure 1.

(a–d) Diagrams show the cuts and measurements of the proximal tibia and distal femur. fAP: femoral anteroposterior, fLAP: femoral lateral anteroposterior, fMAP: femoral medial anteroposterior, fML: femoral mediolateral, tAP: tibial middle anteroposterior, tLAP: tibial lateral anteroposterior, tMAP: tibial medial anteroposterior, tML: tibial mediolateral

The femoral anatomic axis was defined as a line connecting the centre point of the transepicondylar introcession and the centre point of the intramedullary canal at the distal third of the femur.[12] The distal femur was cut with a resection depth of 9 mm above the lowest point of the medial condyle and angled 6° valgus relative to the anatomic axis[12] [Figure 1c]. The femoral mediolateral (fML) axis was defined by the most prominent points of the medial and lateral femoral condyles, with the femoral anteroposterior (fAP) axis set perpendicular to the ML axis. The fML dimension was measured on the distal femoral cut surface along the ML axis, while the fAP dimension was measured as the total width of the lateral condyle along the AP axis.[12] The femoral medial anteroposterior (fMAP) and femoral lateral anteroposterior (fLAP) dimensions were taken as the widest parts of the medial and lateral condyles on the distal femoral cut surface along the AP axis [Figure 1d]. The femoral aspect ratio (fML/fAP) was calculated to analyse the shape of the knee.

The measured femoral and tibial values were then compared with the corresponding morphology of four commonly used prostheses in China: Scorpio and Duracon (Stryker Howmedica Osteonics, Allendale, NJ, USA), PFC sigma (DePuy-Johnson and Johnson, Warsaw, IN, USA), and Nexgen (Zimmer, Warsaw, IN, USA). Various sizes of prostheses were obtained based on a previous study by Cheng et al.[12] An appropriate TKA size was chosen for each knee joint. All tibial components of these prostheses were symmetrical in the ML direction.[12]

Statistical analysis was performed using IBM SPSS Statistics version 22.0 (IBM Corp, Armonk, NY, USA). The measured dimensions were presented as mean ± standard deviation. Data analysis was conducted using the t-test, independent sample t-test and Pearson’s correlation coefficient. A linear regression analysis was conducted to determine the correlations between the femoral/tibial ML and AP dimensions. A P value < 0.05 was considered statistically significant. The Pearson correlation coefficient (r) was categorised as weak (r ≤ 0.3), moderate (0.3 < r ≤ 0.7), strong (0.7 < r ≤ 0.9) or excellent (r > 0.9).

RESULTS

The measurements of proximal tibial morphology are summarised in Table 1. For both Mongolian Chinese and Han Chinese, the tML and tAP dimensions in males were significantly larger than those in females (P < 0.05). Significant differences in size and shape were observed for males and females in the two populations.

Table 1.

Average values of the tibia and femur morphology measurement (mm).

Parameter	Mean±SD
	Mongolian Chinese male	Mongolian Chinese female	Han Chinese male	Han Chinese female
Tibial mediolateral (tML)	81.09±4.06	70.69±3.08	78.52±4.30	70.39±3.15
Tibial anteroposterior (tAP)	52.93±2.56	47.98±1.89	52.37±3.13	46.22±3.07
Tibial medial anteroposterior	54.26±2.35	48.06±1.68	53.69±3.39	47.69±2.66
Tibial lateral anteroposterior	48.89±3.61	45.42±3.39	49.66±3.88	44.66±2.59
Aspect ratio (tML/tAP), %	153.26±4.22	147.39±5.31	150.06±4.86	152.81±10.98
Femoral mediolateral (fML)	76.21±4.35	65.56±2.26	73.64±3.52	64.53±2.49
Femoral medial anteroposterior	53.93±2.18	49.35±1.98	52.97±1.81	50.18±1.83
Femoral lateral anteroposterior	53.39±2.15	48.71±2.02	52.48±1.80	49.73±1.85
Femoral anteroposterior (fAP)	68.15±3.48	60.74±1.90	66.14±4.01	61.74±3.84
Aspect ratio (fML/fAP), %	111.87±4.28	107.97±3.49	111.50±4.14	104.73±4.84

For both populations, tML dimension was positively correlated with tAP dimension in all groups, while the tML/tAP aspect ratio was negatively correlated with tAP dimension across all participant groups [Figure 2a and b]. Under a given tAP dimension along the regression curves [Figure 2a], Mongolian Chinese males had larger tML dimensions than Han Chinese males, and both Mongolian and Han Chinese males had larger tML dimensions than females. Furthermore, both Mongolian and Han Chinese males had larger aspect ratios than females at the same tAP values [Figure 2b].

Figure 2.

Graphs show the effects of ethnicity and gender on resected proximal tibia and distal femur in Mongolian Chinese males (MC-M), Mongolian Chinese females (MC-F), Han Chinese males (HC-M) and Han Chinese females (HC-F), as well as a comparison of the femoral and tibial values with four types of prostheses. Correlations between (a) the tibial mediolateral (tML) and tibial middle anteroposterior (tAP) dimensions; (b) tAP and tML/tAP aspect ratio; (c) the femoral mediolateral (fML) and femoral anteroposterior (fAP) dimensions; and (d) fAP and fML/fAP aspect ratio.

For both populations, the dimensions and aspect ratios of the proximal tibia were compared with four conventional tibial prostheses [Figure 2a and b]. These prostheses had undersised tML and smaller tAP dimensions with overhang in tML and larger tAP dimensions [Figure 2a]. Compared to the aspect ratios of Mongolian and Han Chinese knees, only one prosthesis showed a similar change in aspect ratio among the four conventional tibial prostheses, but the rate of change was different between the two populations [Figure 2b].

The measurements of distal femoral morphology are summarised in Table 1. For both Mongolian Chinese and Han Chinese, the fML and fAP dimensions in males were significantly larger than those in females (P < 0.05). Significant differences in fAP and fML dimensions were observed in both populations, for both males and females. Mongolian Chinese females had a larger aspect ratio than Han Chinese females (P < 0.05).

For both populations, fML dimension was positively correlated with fAP dimension in all conditions, while the fML/fAP aspect ratio was negatively correlated with fAP dimension across all participant groups [Figure 2c and d]. Under a given fAP dimension along the regression curves, significant differences in size and shape were observed between Mongolian Chinese and Han Chinese [Figure 2c]. Both Mongolian Chinese and Han Chinese male knees were larger than those of their female counterparts [Figure 2c]. Furthermore, significant differences in aspect ratio were found between Mongolian Chinese and Han Chinese females [Figure 2d].

For both populations, the dimensions and aspect ratios of the femur were compared with four conventional femoral prostheses [Figure 2c and d]. Two of the femoral prostheses had a larger fML dimension for Mongolian Chinese males and Han Chinese males [Figure 2c]. Compared to the aspect ratios of Mongolian and Han Chinese knees, two of the femoral prostheses showed a similar change in aspect ratio for both males, but no designs showed a similar change in aspect ratio for females [Figure 2d].

Figure 3 shows the morphologic relationship between the tibia and femur in Mongolian Chinese and Han Chinese knees. The correlation between the tML and fAP dimensions is shown in Figure 3a, while the correlation between the tML and fML dimensions is shown in Figure 3b. The fML and fAP dimensions were positively correlated with tML dimension (Mongolian Chinese: r = 0.66 for fAP; Han Chinese: r = 0.66 for fAP; Mongolian Chinese: r = 0.97 for fML; Han Chinese: r = 0.88 for fML). As the tML dimension increased, the fML and fAP dimensions also increased. Both Mongolian Chinese and Han Chinese males had larger dimensions than their female counterparts.

Figure 3.

Graphs show the morphological relationship between the tibia and femur in Mongolian Chinese males (MC-M), Mongolian Chinese females (MC-F), Han Chinese males (HC-M) and Han Chinese females (HC-F). Correlations between the tibial mediolateral (tML) dimension and (a) femoral anteroposterior (fAP) and (b) femoral mediolateral (fML) dimensions.

DISCUSSION

The most important finding of the present study was that Mongolian Chinese have larger sizes (ML and AP dimensions) and shapes (ML/AP aspect ratio) of the proximal tibia and distal femur compared to Han Chinese. Significant differences were found in ML and AP dimensions, as well as ML/AP aspect ratios, between the two populations and most conventional knee prostheses. In this study, the measured AP and ML dimensions of Han Chinese knees were generally consistent with those of previous studies[8,12,15,21,27] [Table 2]. The differences in dimensions were primarily attributed to variations in measurement methods. For example, the measured fAP dimension of Han Chinese knees was 10 mm larger than that reported by Li et al.,[15] who did not include the anterior condyle thickness. The tML and tAP measurements from Yang’s study[8] were smaller than those of our study, possibly due to the different cut methods used.

Table 2.

Comparison of the morphological data of the current study with those reported in literature.

Study	Population	fML (mm)	fAP (mm)	tML (mm)	tAP (mm)	Prostheses (compared)
Loures et al.,[5] 2016	Brazilian	77.7±4.9 (M) 67.8±4.0 (F)	70.3±4.7 (M) 61.5±4.9 (F)	79.8±5.8 (M) 69.6±4.3 (F)	53.9±6.1 (M) 46.0±4.0 (F)	More than a quarter of patients unsatisfied
Miyatake et al.,[4] 2016	Japanese			76.4±3.2(M) 68.3±2.9(F)		Genesis II and Persona better than NexGen
Erkocak et al.,[6] 2015	Turkish			77.1±5.1(M) 68.7±3.6(F)	47.6±3.8(M) 40.9±3.1(F)	Mismatched
Chung et al.,[7] 2015	Korean	76.1±4.0(M) 67.9±3.3(F)	67.2±3.9 (M) 61.1±3.2 (F)			Mismatched
Li et al.,[15] 2014	Caucasian	74.6±3.9(M) 65.4±1.4(F)	59.6±3.2 (M) 55.4±2.8 (F)	79.4±4.3(M) 70.2±2.7(F)	49.5±2.9(M) 45.2±2.3(F)
Yue et al.,[1] 2014	Chinese					Morphological or gender-specific prosthesis (better)
Yang et al.,[21]2014	Chinese	79.0±5.0(M) 71.2±4.3(F)	66.8±4.0 (M) 61.3±3.3 (F)
Yang et al.,[19] 2014	Chinese			77.3±4.7(M) 71.1±3.7(F)	48.5±4.0 (M) 44.7±3.5 (F)	Mismatched
Li et al.,[15] 2014	Chinese	72.7±3.8 (M) 64.4±2.6 (F)	56.5±2.5 (M) 52.8±2.6 (F)	77.4±3.3 (M) 69.1±2.8 (F)	49.6±2.4 (M) 44.2±2.3 (F)
Lim et al.,[9] 2013	Korean	81.5±5.7 (M) 76.7±3.71 (F)	59.0±4.01 (M) 58.4±3.10 (F)	80.6±6.31 (M) 70.0±3.45 (F)
Chaichankul et al.,[11] 2011	Thai	70.15±3.87 (M) 59.91±3.75 (F)		74.44±3.44 (M) 64.95±3.45 (F)	50.15±3.09 (M) 43.23±2.57 (F)	Mismatched
Cheng et al.,[12] 2009	Chinese	74.4±2.9 (M) 66.8±3.1 (F)	66.6± 2.4 (M) 61.0±2.7 (F)	76.4±2.8 (M) 68.8±4.6 (F)	51.3±2.0 (M) 45.7±1.9 (F)	Mismatched
Lonner et al.,[23] 2008	American	76.92 (M) 67.49 (F)	62.27 (M) 56.32 (F)
Kwak et al.,[2] 2007	Korean			76.1±4.0 (M) 67.64±3.12 (F)	48.2±3.3 (M) 43.2±2.3 (F)	Mismatched
Uehara et al.,[17] 2002	Janpanese			77.9±4.1 (M) 69.5±3.4 (F)	54.1±3.0 (M) 49.2±2.9 (F)	Mismatched
Present study	Mongolian Chinese	76.21±4.35 (M) 65.56±2.26 (F)	68.15±3.48 (M) 60.74±1.90 (F)	81.09±4.06 (M) 70.69±3.08F)	52.93±2.56 (M) 47.98±1.89 (F)	Mismatched
Present study	Han Chinese	73.64±3.52 (M) 64.53±2.49 (F)	66.14±4.01 (M) 61.74±3.84 (F)	78.52±4.30 (M) 70.39±3.15F)	52.37±3.13 (M) 46.22±3.07 (F)	Mismatched

F: female, fAP: femoral anteroposterior, fML: femoral mediolateral, M: male, tAP: tibial middle anteroposterior, tML: tibial mediolateral

Several studies have revealed that Asian knees are generally smaller than Western knees[15,16,28,29,30] [Table 2]. Compared with Brazilian patient knees,[5] the measured fML and fAP dimensions of Han Chinese knees were smaller, while those of Mongolian Chinese and Brazilians were similar. Compared with American knee measurements,[23] the measured fML dimension of Han Chinese knees in this study was smaller, but the measured fAP dimension was larger. In contrast, the measured fML dimension of Mongolian Chinese knees was closer to that of Americans. Therefore, the dimensions of Han Chinese knees were significantly different from those of Western knees, but the dimensions of Mongolian Chinese and Western knees were similar.

Compared with Korean patient knees,[2,9] the measured fML dimension of Han Chinese knees in this study was smaller, but the fAP dimension was larger; the measured tML and tAP dimensions of Han Chinese knees were also larger. Compared with Thai normal knees,[11] the measured ML and AP dimensions of Chinese knees were larger. These results indicate that morphological variations exist among different ethnicities within the Asian population. The present study demonstrated that even Han Chinese and Mongolian Chinese populations exhibit distinct morphological features. Interestingly, our measurement data revealed that the size (AP and ML dimensions of the proximal tibia and distal femur) and shape (tibial and femoral aspect ratios) of knees showed significant differences based on gender and ethnicity. Therefore, knee size and shape differences between Han Chinese and Mongolian Chinese should be considered, especially when selecting suitable prostheses for TKA.

Consistent with previously reported studies,[7,9,12,14,15,21,22,23,29] our results demonstrated that both Han Chinese females and Mongolian Chinese females had narrower femoral condyles than males for a given fAP dimension. This may explain the observation that females tend to experience more ML overhang than males when using current TKA prostheses.[10,11,12,15,21] For a given tAP dimension, Han Chinese and Mongolian Chinese females had narrower tibial platforms than males. The narrower tibial platform implies a potential for ML overhang or AP undercoverage when downsizing the prosthesis. These finding may have implications for soft tissue balancing and postoperative tibial positioning. Therefore, the design of femoral and tibial prostheses should account for gender variations.

In term of mismatch between Chinese knees and the four selected TKA prostheses, the measured femoral data showed a larger fML dimension across all ranges of the fAP dimension, with this discrepancy being more pronounced in females. For the aspect ratio (fML/fAP) and fAP dimension, only one prosthesis showed a decreasing femoral aspect ratio with increasing fAP dimension, although the rate of change differed between Han Chinese and Mongolian Chinese. The remaining prostheses showed no change in the femoral aspect ratio with increasing fAP dimension. Similarly, our tibial data showed a decreasing aspect ratio (tML/tAP) with increasing tAP dimension, similar to observations in other studies.[15,25] However, most current prostheses have a constant aspect ratio. We found that the tML dimension was undersized with the smaller tAP dimension and exhibited overhang with the larger tAP dimension. This was more evident in Han Chinese and Mongolian Chinese male knees. Therefore, suitable knee prostheses should be selected based on ethnicity and gender differences.

In this study, we found that tML dimension was positively correlated with both fML and fAP dimensions in Han Chinese and Mongolian Chinese, a trend similar to that reported by Cheng et al.[12] The measured results suggest that it may be important to consider the tibia and femur as a whole when designing prostheses. Therefore, the tML and fAP dimensions should be considered key factors in designing proper prostheses for Han Chinese and Mongolian Chinese populations.

This study has some limitations. First, only 74 normal knees were analysed, and therefore, these results provide only general guidelines for gender-specific and ethnicity-specific knee prostheses. As our sample size is small, more samples should be investigated in future studies. Second, the current study evaluated the knees of healthy Han Chinese and Mongolian Chinese individuals without considering patients with inflammatory arthritis or rheumatoid conditions, thereby limiting the applicability of our findings to those with knee degeneration. Lastly, only four types of knee prostheses were evaluated for mismatch comparison; future studies should assess a larger range of prostheses. Further investigation is needed to evaluate the clinical impact of implant designs based on these ethnic differences.

In conclusion, the results of the present study indicate that Mongolian Chinese have larger sizes and different shapes of the proximal tibia and distal femur compared to Han Chinese. Significant differences were found in ML and AP dimensions, as well as ML/AP aspect ratios between the two populations and conventional knee prostheses. These findings suggest that variations in ethnicity and gender should be considered in the development of anatomic knee prostheses.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.