1 Introduction

Achieving permanent fixation in revision knee arthroplasty is challenging due to the presence of bone defects and soft tissue insufficiency.¹ In knee surgeries, constraint revision prostheses, with which rotation and translational movement are restricted, are preferred in case of ligament failure causing instability. On the other hand, revision knee prostheses may be used in patients with significant metaphyseal bone defect. With both type of prostheses, intramedullary extensions, called stems, are used to increase stability, protect the metaphyseal bone in the joint's proximity, and transfer the load to the stronger diaphysis.^2–4 The use of stems increases the initial fixation strength of the prosthesis, provides better restoration of the mechanical alignment,^5–7 and increases the long-term survival of the prosthesis.^8–10

Two different methods are utilized in stem fixation. In the full cementation technique, the prosthesis and the stem are completely cemented, while in the hybrid fixation technique, a cemented component and an uncemented press fit stem filling the medullary canal are used. In the hybrid fixation technique, the stems that fill the diaphysis should have sufficient length to tightly fit into the medullary canal to ensure component stability.¹¹ The length of a press fit stem should be evaluated simultaneously with the canal diameter.¹² There is no clear clinical evidence yet on the optimal stem diameter and length. Many authors have reported that the canal filling ratio (CFR) is the criterion that best expresses the ideal combination of stem size and length.^2,5–9,13

The CFR was first used in an article by Parsley et al., in 2003⁵ which examined patients who received revision knee prostheses. The authors evaluated the effect of the stem CFR on the optimal mechanical alignment of the tibial component. The ratio was calculated by dividing the diameter of the tibia near the lower tip of the stem by the diameter of the endosteum adjacent to this region seen on the AP radiograph.⁵ In another study, Fleischman et al. utilized a three-dimensional CFR calculation method using the area of an ellipse.⁹ Lee et al. measured CFR using the method described by Parsley on AP and lateral femur and tibia radiographs in patients who underwent revision knee prosthesis surgery, and considered the higher value measured as the CFR.³

Today, no consensus or scientific evidence exists in the literature on the most accurate method for measuring CFR. It is thought that any consensus that might emerge over time from studies conducted on this matter will ensure that CFR measurement be considered a uniform and instrumental criterion. Therefore, the present study aims to determine a method for measuring tibial and femoral stem CFR on direct radiographs with the highest accuracy rate in patients who have undergone knee replacement surgery using modular revision components.

2 Patients and methods

This study was conducted in accordance with the Declaration of Helsinki and after approval was obtained from the institutional review board, the radiological data of 373 patients who underwent revision knee arthroplasty between 2005 and 2021 were retrieved from the hospital database. Those with a visible distal part of the femoral or tibial stem on computed tomography (CT) scans and those with radiographs suitable for radiographic measurement were included in the study after informed consent was obtained. Patients with impaired femoral or tibial anatomy due to a history of fracture or infection and those unsuitable for radiological and tomographic measurements were excluded. A total of 28 femoral and 27 tibial components in 29 knees of 29 patients (nine males, 20 females; mean age: 72.86 ± 6.32 years) were included in the study.

2.1

2.1 Measurement of the CFR on direct radiographs

The CFR of the femoral and tibial component stems of the patients was measured close to the stem tip, at the level where the stem is expected to tightly fit into the medulla, as described by Parsley.⁵ Intramedullary stem diameter of the femoral and tibial components in the AP and lateral planes were measured using the ExtremePacs software (ExtremePacs, Ankara, Turkey) on the radiographs of the patients (Fig. 1 and 2). At the same time, the widths of the tibial and femoral medullary canals were measured in line with the points where the diameter was measured (Fig. 3 and 4). Measurements were noted and CFR calculations were made based on the five different methods in the literature.

Close

Fig. 1 Measurement of the stem diameter based on Parsley's method on the AP radiograph.

Export to PPT

Close

Fig. 2 Measurement of the stem diameter based on Parsley's method on the lateral radiograph.

Export to PPT

Close

Fig. 3 Measurement of the width of the medullary canal from the region that corresponded to the stem tip.

Export to PPT

Close

Fig. 4 Measurement of the medullary canal width based on Parsley's method on the lateral radiograph.

Export to PPT

In the first method, as described by Parsley,⁵ CFR was calculated by dividing the diameter of the lower tip of the stem by the diameter of the medullary canal that was at the same level with the tip on AP radiographs.

In the second method, CFR was calculated on AP and lateral radiographs, according to the measurement and calculation method employed by Parsley and Lee et al.⁸ The greater of these measurements was accepted as the stem CFR.

In the third method, the area of the stem and the medullary canal near the stem tip, where we expect the prosthesis to tightly fit into the canal, was calculated based on the area of an ellipse as described by Fleischman et al.⁹ For the calculation of the ellipse area, the values previously measured on the AP and lateral radiographs were divided by two to obtain the radius values. The stem and medullary canal areas were calculated with the ellipse area calculation formula, while CFR was calculated by dividing the stem area by the medullary canal area.^7,9

For the fourth method, CFR was calculated by dividing the diameter of the lower tip of the stem by the diameter of the medullary canal at the same level, as measured on the lateral radiographs, using the measurement method applied by Parsley.⁵

Finally in the fifth method, CFR was measured using Parsley's method on the AP and lateral radiographs, as reported by Lee et al.⁸ The smaller of these measurements was accepted as the CFR of the stem.

2.2

2.2 Measurement of the CFR using computed tomography

The sagittal, axial, and coronal sections of the patients were viewed side by side on the same screen using the picture archiving and communication system (PACS) software. The area covered by the stem and the medullary area were measured from the level where we expected the stem to fit tightly into the medulla (Fig. 5 and 6). The real CFR valuesfor the femoral and tibial stems were calculated by dividing the area covered by the stem by the medullary area.

Close

Fig. 5 Measurement of the medullary region on the coronal CT section.

Export to PPT

Close

Fig. 6 Measurement of the ‘stem’ area on the coronal CT section.

Export to PPT

The coronal sections taken demonstrated that the medullary area, especially in the tibia, was irregularly shaped and quite wide (Fig. 7). Since calculating the CFR of a stem placed in this non-elliptical or non-circular area using the above method may not be suitable for the press fit stem placement technique, a second calculation was made by determining the largest stem size that would fit tightly into the medullary canal on the CT coronal sections (Fig. 8). In this calculation, the CFR was found by dividing the stem area by the area of the maximum stem that would fit tightly into the medullary canal.

Close

Fig. 7 It is noteworthy that the tibia has an irregular shape and a great width in the region that corresponded to the stem tip on the coronal CT section.

Export to PPT

Close

Fig. 8 Determination of the maximum ‘stem’ area that can fit into the medullary canal on the coronal CT section of the same patient.

Export to PPT

3 Results

The p values of the Durbin-Conover test are presented in Table 1. The mean, median, Shapiro-Wilk test W and Shapiro-Wilk test p values of the femoral stem CFR measurement methods are given in Table 2.

The Durbin-Conover pairwise comparison test revealed a statistically significant difference between CFR measurements based on the largest stem size that would tightly fit into the femoral medullary canal on CT and other radiological measurements (p < 0.05). The value closest to the median score (67.9) of the stem calculation according to this method using CT was that of the femoral stem CFR calculation (61.5) based on the ellipse area measurement method (Table 2).

The assumption of normality according to the Shapiro-Wilk test was met in all methods used for the measurement of femoral stem CFR (p > 0.05).

There was no statistically significant difference between the real femoral stem CFR on CT and the femoral stem CFR calculated using the ellipse area measurement method (Method 2) (p > 0.05).

The p values of the Durbin-Conover test applied to compare the measurement of the real CFR of the tibial stem and the measurement of the CFR based on the largest tibial stem size that would tightly fit into the medullary canal with other radiological measurements are presented in Table 3. The mean, median, Shapiro-Wilk test W and Shapiro-Wilk test p values of all tibial stem CFR measurements are given in Table 4.

There was a statistically significant difference among the actual CFR of the tibial stem detected on CT and other radiological measurements (p < 0.05). The value closest to the median score (44.8) of the stem calculation according to this method using CT was that of the tibial stem CFR calculation (53.5) based on the ellipse area measurement method (Table 4).

The assumption of normality according to the Shapiro-Wilk test was met in all methods used for the measurement of tibial stem CFR (p > 0.05).

There was no statistically significant difference between the CFR calculated based on the largest stem size that would tightly fit into the medullary canal on CT and the smaller CFR value obtained using Method 5 (p > 0.05).

4 Discussion

In revision total knee replacement surgery, modular stems should be used in cases with insufficient bone stock to support the prosthesis.² Using diaphyseal stems to fill the canal allows for significant diaphyseal cortical contact. Increased cortical contact reduces micromovements of the knee prosthesis, increasing component stability and providing appropriate mechanical alignment.^2,6,7,12,13 Biomechanical studies have shown that knee prosthesis applications using a stem that does not sufficiently fill in the diaphysis do not provide better tibial component stability than stemless applications.⁴ To date, there has been no clear guideline on which modular stem to use in revision knee prosthesis surgery. Parsley et al. reported that better alignment was achieved in cases with a CFR greater than 0.85,⁵ while Gililland et al. stated that compression of the diaphyseal stem along a segment longer than 4 cm provided effective stability.¹⁴ Several studies have reported that CFR can be used as a criterion to determine the ideal combination of stem size and length.^7,8,13 However, no scientific consensus has been reached on the gold standard measurement method for CFR.

During measurements made using axial CT sections, we observed that the shape of the femoral canal was generally similar to an ellipse, while the tibial canal was generally irregular in shape and wide. Therefore, in CFR measurement of the femoral stem, measurements made by dividing the cross-sectional area of the stem near the stem tip by the femoral canal area will be the most accurate. The irregularity and wideness of the canal of the tibia suggests that calculating the CFR with the same method would not be suitable for the press fit stem application logic. Therefore, it seems more appropriate to calculate the CFR by dividing the cross-sectional area near the stem tip by the maximum thickness stem area that can be compressed into this region in the CT sections of the tibia. No statistically significant correlation could be established between results obtained with the actual CFR measurement using CT applied for the tibial stem and the CFR measurement methods applied for the femoral stem, calculated according to the maximum thickness of the stem cross-sectional area that can be compressed into the medullary canal, and the results obtained with the radiological CFR measurement methods performed using direct radiographs. Ultimately, we concluded that the most accurate method for measuring the CFR of the femoral stem on direct radiographs was that in which the CFR was calculated using the ellipse area measurement method (Method 2).

We also concluded that the most accurate method for measuring the CFR of the tibial stem was measuring the stem diameter and medullary canal width on the AP and lateral radiographs using Parsley's method, with the smaller value accepted as the CFR (Method 5).

Canal filling ratio is a criterion that entered the literature in the early 2000s. With few studies on the subject, its clinical use has not been fully widespread. However, significant results have been obtained in few studies. Using their own measurement method, Parsley et al. found that the tibial alignment was better in the AP plane in cases where the CFR of the cementless tibial stem was 0.85 or more.⁵ Nakasone et al. showed that with cementless press fit modular stems that tightly fit in the diaphysis, component alignment gradually improved as the CFR rises above 0.85 in the femur and tibia.⁶ Lee et al. stated that a CFR of 0.85 and above in the femoral and tibial stems positively affected implant survival.⁸ Canal filling ratio measurement methods used in previous studies, alignment and implant survival rates, and the valuable information obtained with the measurement methods used will continue to serve researchers in the future.

The present study is the first to determine the most accurate roentgenographic measurement method for calculating CFR. The limited number of patients, different designs of the implants, different lengths of the tibial and femoral stems, and the different cross-sections of the stems used (Fig. 9) are the factors that may be considered limitations since that may affect the CFR measurements and thus the statistical results.

Close

Fig. 9 Views of different ‘stem’ sections on axial CT sections.

Export to PPT

5 Conclusions

In patients who underwent surgery using hybrid type press fit revision knee prostheses, the most accurate method for measuring the CFR of femoral stems on direct radiographs was the CFR calculation using the ellipse area measurement method (Method 2). The most accurate method for measuring the CFR of tibial stems was determined to be Method 5, which took the smaller value of the measurements of the stem diameter and medullary canal width on AP and lateral radiographs using Parsley's method.

Institutional Ethical Committee Approval

Institutional Ethical Committee Approval was obtained with approval no: 78/551.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Authors contribution

Osman Cimen, MD: Conceptualization, formal analysis, investigation, methodology, resources, software, writing original draft.

Mehmet Akif Kaygusuz, Prof: Conceptualization, data curation, project administration, supervision, validation, review and editing (writing).