Extramedullary femoral alignment system in total knee arthroplasty: Accuracy in relation of severity and different types of varus

Abstract

Objectives

We assessed the accuracy of an extramedullary guide system for femoral component alignment in TKA.

Methods

We retrospectively analysed 87 total knee arthroplasties using a newly developed extramedullary guide system.

Results

Correct postoperative coronal plane femoral component, with deviation from neutral alignment by 3° or less, was found in 87% of the study population. The percentages were 100%, 91% and 79% for HKA of 0–3°, 3–10° and >10°respectively.

Conclusion

The correct use of the extramedullary guide system allows the restoration of the neutral mechanical axes of the lower limb, especially in cases of limited varus deformity.

1. Introduction

Total knee arthroplasty (TKA) is the first-choice treatment option for late-stage knee osteoarthritis (OA) as it has been shown to have excellent survivorship.¹ A postoperative lower-limb alignment that results in a neutral mechanical axis represents the goal of the various surgical techniques employed in most TKA.² Indeed, several studies have reported that early TKA failure is common if the implant is not aligned within 3° of the neutral mechanical axes.^3,4 Historically, intramedullary alignment (IM) has proven to be more accurate than extramedullary alignment (EM).⁵ Clinical evidence dating back to the late 1980s or early 1990s has documented the use of previous generations of EM instruments that used the anterosuperior iliac spine as reference, and used tensioning devices. In the past years, the application of the EM system has been reconsidered due to the risk of fat embolism and the substantial amount of blood loss in the femoral canal reaming following IM alignment.⁶ Furthermore, it has been demonstrated that there are no significant differences in terms of femoral cut accuracy between the new EM devices and IM alignment.⁵ Recently, a new EM device called Extra Medullary Alignment System (EMAS) (Adler Ortho, Cormano, Milan, Italy) has been developed. It does not require any specific cutting blocks, navigation or pre-operative computed tomography study. Furthermore, its efficacy and accuracy have been analysed in patients with inaccessible femoral canal due to extra-articular deformities or previous internal fixation.⁷ The encouraging results reported at middle follow-up in patients with extra-articular deformity lay the foundation of the extent of its use to routine cases. The primary aim of the present study was to analyse the accuracy of EMAS in reproducing a neutral distal femoral resection on the coronal and sagittal planes in patients affected by varus knee OA. The secondary aim was to evaluate if there are any differences in relation to varus severity and different varus knee phenotypes.

2. Materials and methods

The present study was approved by the institutional review boards of the authors' hospitals. Between January 2018 and February 2020, 125 patients underwent primary TKA for knee OA. In January 2020, the patients' files and the hospital's digital database were reviewed retrospectively. Excluded were patients with preoperative valgus alignment, extra-articular deformities, Blount disease and ipsilateral hip replacement, as well as patients without pre- and/or postoperative weight-bearing full-leg radiographs. In total, 73 patients (87 knees) comprising 46 women and 27 men were included in the study. The average patient age at the time of surgery was 68.0 years (range, 53–88). The average body mass index of the study cohort was 28.6 kg/m² (range, 24–36). The weight-bearing full-leg radiographs were obtained as described by Paley,⁸ with the subjects standing barefoot and the feet together in the “stand at attention” position while the patellae were oriented forward. In accordance to Bellemans et al.,⁹ the medial proximal tibial angle (MPTA), mechanical lateral distal femoral angle (mLDFA), joint line convergence angle (JLCA), anatomic-mechanical angle (AMA) and hip-knee angle (HKA) were determined based on the preoperative full-leg radiographs (Fig. 1). The patients were classified in accordance to the severity of varus, with a preoperative HKA of 0–3° (no malalignment) (n = 18), 3–10° (moderate malalignment) (n = 45) and >10° (severe malalignment) (n = 24), respectively. Next, the patients were classified in relation to preoperative tibial and femoral alignment. For the tibia, an MPTA ≤87° (n = 53) was considered varus and an MPTA > 87° (n = 34) was considered valgus. For the femur, an mLDFA ≤90° (n = 34) was considered varus and an mLDFA > 90° (n = 53) was considered valgus. Postoperative HKA, α and γ angle were determined based on the postoperative full-leg radiographs (Fig. 1). In accordance to the Knee Society roentgenographic evaluation form,¹⁰ the coronal and sagittal femoral component alignments were rated as “aligned” if the α and γ angle were 90° ± 3°, respectively. Conversely, the patients were classified as outliers if the coronal and sagittal malalignments were greater than 3°. Conventionally, positive values of α and γ angle correspond to the valgus and flexion alignment of the femoral component, respectively. Two authors (AA, LF) assessed all radiographs twice with an interval of at least 4 weeks to avoid any memory effect. The intraclass correlations for intraobserver and interobserver measurements were 0.91 and 0.88, respectively. All measured values were calculated up to one decimal place.

Fig. 1.

a) Pre-operative full-leg radiographs; b) α angle detected on post-operative full-leg radiographs; and c) γ angle detected on post-operative lateral radiographs.

2.1. Surgical technique

The patients were operated on using a surgical technique described previously.⁷ In brief, all surgeries were performed via a medial parapatellar approach. A tourniquet was used in each case. All patients were administered the same regional anaesthesia. The average surgical time was 80 min (range, 65–110). Postoperatively, the patients underwent the hospital's standard recovery protocol. The operated knees were passively moved immediately after surgery. Standing was allowed on the first day post-op. The patients were discharged to the rehabilitation department within 7 days from surgery.

All patients received a Genus Fixed Bearing cruciate retaining knee prosthesis (Adler Ortho, Milan, Italy), with femoral resection performed by the senior authors (PC and MF) using the EMAS EM alignment guide.

The EMAS is characterized by a spacer with two extensible paddles (lateral and medial) and a removable frontal tower. After proximal tibia osteotomy, the device is inserted with the knee in full extension and with its two paddles completely retracted. It is possible to adjust the two paddles' height by acting on the medial and lateral screw holes available on the front of the instrument. The instrument set includes +2 mm, +4 mm, +7 mm and +10 mm shims to be added to the back of the device in case the bone removed from the tibia was thicker than 10 mm. The joint is aligned and balanced by adjusting the two paddles' thickness, while soft tissue releases are performed if necessary. An extra-medullary rod allows the surgeons to verify the lower limb alignment. After the achievement of the right joint tension and axis, a cutting block can be connected to the device to perform the distal femoral resection with the knee in flexion. Subsequently, through a similar procedure, the joint is balanced in flexion, and a sizing instrument attached to the EMAS′ frontal tower will guide the surgeon to select the right size of the 4-in-1 cutting block and position it in the correct external rotation, with 1 mm difference between the two paddles’ thickness being equal to one degree of femoral component rotation. However, the surgeon always has the option to check the femoral implant rotation by referring to the Whiteside line and the transepicondylar or posterior condylar axes.

2.2. Statistical analysis

Categorical variabilities were expressed in numbers of observations and percentages. Continuous variabilities were expressed by mean and standard deviation.

Data from two groups were compared using Student's t-test or Satterthwaite t-test for continuous variables and Chi-squared or Fisher's exact test for categorical variables. Data from three groups were compared using Fisher's exact test for categorical variables and one-way analysis of variance employing Games-Howell post-hoc comparison tests for continuous variables. Linear regression was used to assess the relationship between continuous variables. Data were collected using Excel (Microsoft, Redmond, WA, USA). Statistical analyses were performed using Stata/SE 15.1 (StataCorp, College Station, Texas, USA). Two-sided p < 0.05 was considered statistically significant.

3. Results

The mean postoperative HKA was 2.1° ± 2.2° (range, 0–8.4°), with 22 knees (25.3%) considered maligned due to postoperative HKA greater than 3°. The mean postoperative α angle was −0.3° ± 2.1° (range, −7.0°–4.5°), with 11 (12.6%) knees classified as maligned. For the γ angle, these values were 1.66° ± 1.433° (range, −2.8°–4.4°) and 8 (9.2%), respectively.

We did not report any case of valgus alignment over 5° in either group. A single case of varus alignment over 5° was observed in the group with moderate preoperative malalignment. We did not report any case of hyperextension malalignment (γ < −3°) of the femoral component in either group. We found no malalignment over 5° of flexion in either group (Fig. 2).

Fig. 2.

a) The distribution of femoral α angle (coronal alignment) in the three groups; b) the distribution of femoral γ angle (sagittal alignment) in the three groups.

When stratifying postoperative HKA by preoperative varus severity, there were significant differences, with a higher proportion of malalignment in the preoperative more severely malaligned knee (Table 1). An exception was the comparison between the groups with no and moderate preoperative HKA malalignment, which did not show significant difference (p = 0.160). When correlating the coronal alignment of femoral component with preoperative HKA, there were significant between-group differences, except between the groups with moderate and severe HKA malalignment (p = 0.368) (Table 2). No significant association was found between postoperative femoral coronal plane malalignment and preoperative HKA severity.

Table 1.

Comparison of postoperative HKA alignment according to preoperative HKA varus alignment severity.

			Post-hoc comparison (p-values)
			Group B		Group C
	Postop-HKA [Degree°]*	Malalignment**	Degree	Malalignment	Degree	Malalignment
Global p-value	<0.001	<0.001
Preoperative malalignment:
Aligned (n = 19)	0.42 ± 0.69	0 (0.0%)	<0.01	0.160	<0.001	<0.001
Moderate (n = 44)	1.82 ± 1.89	7 (15.9%)			<0.001	<0.001
Severe (n = 24)	3.94 ± 2.21	15 (62.5%)

*Expressed as mean ± standard deviation; ** expressed as n (%).

Table 2.

Comparison of coronal plane femoral component alignment according to preoperative HKA varus alignment severity.

			Post-hoc comparison (p-values)
			Group B	Group C
	α [Degree]*	Malalignment**	Degree	Degree
Global p-value	0.024	0.655
Preoperative malalignment:
Aligned (n = 19)	0.77 ± 0.79	0 (0.0%)	0.001	0.003
Moderate (n = 44)	−0.45 ± 2.06	4 (9%)		0.368
Severe (n = 24)	−1.0 ± 2.5	7 (21%)

*Expressed as mean ± standard deviation; ** expressed as n (%).

For postoperative femoral sagittal alignment, there was no significant association between prosthetic flexion alignment and preoperative varus HKA malalignment when measured on a continuous scale (p = 0.074) and on a categorical scale (p = 0.170) (Table 3).

Table 3.

Comparison of sagittal plane femoral component alignment according to preoperative HKA varus alignment severity.

	γ [Degree]*	Malalignment**
Global p-value	0.074	0.170
Preoperative malalignment:
Aligned (n = 19)	1.65 ± 0.89	0 (0.0%)
Moderate (n = 44)	1.60 ± 1.69	5 (11.4%)
Severe (n = 24)	1.78 ± 1.3	3 (12.5%)

*Expressed as mean ± standard deviation; ** expressed as n (%).

The different types (varus versus valgus) of preoperative tibia and femoral alignment and their association with postoperative HKA and femoral component alignment in the coronal and sagittal plane are reported in Table 4.

Table 4.

Comparison of postoperative alignment according to preoperative femoral and tibial morphology.

		HKA				Femur coronal plane				Femur sagittal plane
		Pre op [Degree]*	p-value	Malalignment**	p-value	α [Degree]*	p-value	Malalignment**	p-value	γ [Degree]*	p-value	Malalignment**	p-value
MPTA	Varus (n = 53)	8.6 ± 4.43	<0.001	18 (34.0%)	0.04	−0.9 ± 2.25	<0.001	8 (15%)	0.60	1.92 ± 1.34	0.04	8 (15%)	0.05
	Valgus (n = 34)	4.9 ± 3.37		4 (11.8%)		0.55 ± 1.45		3 (8.8%)		1.25 ± 1.50		0 (0.0%)
mLDFA	Varus (n = 34)	10.49 ± 3.98	<0.001	17 (50.0%)	<0.001	−1.55 ± 2.24	<0.001	8 (23.5%)	0.03	2.07 ± 1.21	0.02	5 (14.7%)	0.30
	Valgus (n = 53)	5.01 ± 3.21		5 (9.4%)		0.45 ± 1.58		3 (5.7%)		1.40 ± 1.51		3 (5.7%)

*Expressed as mean ± standard deviation; ** expressed as n (%). Abbreviations: MPTA: medial proximal tibial angle; mLDFA: mechanical lateral distal femoral angle.

4. Discussion

The main findings of this study were that the coronal and sagittal femoral component alignments were respectively obtained in 87% and 90.8% of the study population. In particular, in patients with neutral preoperative mechanical axes, EMAS was able to correctly align the femoral component in all patients. However, in presence of preoperative severe varus deformity, the rate of correctly aligned postoperative HKA axes decreased from 100% to 79%. The present study revealed that EMAS’ ability to achieve the same optimal results declines in low and moderate deformity.

Recently, Castellarin et al. reported results obtained with the EMAS technique on a consecutive series of 303 knees.¹¹ However, the authors focused on the clinical results obtained with a mobile bearing knee instead of the accuracy of EMAS. Our findings are consistent with those of the most recent literature presenting results on EM instrumentation.⁵ Baldini et al. compared an EM instrumentation similar to ours with standard IM instrumentation, and found that the femoral coronal alignment was within 2° of the neutral mechanical axes in 86% of the EM group with a 6° mean of preoperative varus.¹² On the other hand, Matsumoto et al. reported an optimal femoral coronal alignment (within 3°) in 98% of patients using a similar EM instrument with a 5° mean of preoperative varus.¹³ However, both author groups had used a set of instruments that required preoperative templating measurements of the inter-femoral head centre distance. Therefore, more preparation and preoperative tests are necessary compared to EMAS. Seo et al.¹⁴ and Ku et al.¹⁵ reported 90.6% and 80.1% correct coronal alignments, respectively using a EM instruments.

In the present study, the sagittal alignment was as accurate as the coronal. Through the use of EMAS, we were able to report the highest percentage of sagittal alignment compared with other research.^12,13 A few studies have been conducted about the role of sagittal alignment in clinical results. Dennis et al.¹⁶ and Gromov et al.¹⁷ reported that a sagittal aligned femoral component was related to better functional outcome and longer survival rate. We did not report any case of extension (more than 3°) and flexion (more than 5°) of the femoral component independently from the severity of varus.

Originally, the present study analysed the efficacy of EMAS in relation to different types of varus. Regarding femoral component alignment in the coronal plane, we reported a higher occurrence of malalignment in cases of varus femur. Moreover, analysing our results it pointed out that in case of mLDFA >90°, there is a tendency to varus positioning the femoral component probably due to hypoplasia of medial femoral condyle or femoral shaft bowing.

On the other hand, regarding sagittal femoral malalignment, it has been reported a higher percentage of sagittal flexion of femoral component in case of varus tibia. Indeed, it has been demonstrated that more the tibia is in varus, the medial resection of tibial cut might be insufficient. For these reasons, a more distal tibial cut might increase the flexion gap imbalance.^18,19 Therefore, an intentional sagittal flexion of the femoral component during TKA can be a useful strategy to decrease the flexion gap.²⁰

The present study has several limitations. First, this was a retrospective study, and no control group was included. Second, we included only varus osteoarthritic knees, so our conclusions cannot be generalized to other types of deformity. Finally, all operations were performed by experienced knee arthroplasty surgeons, and the results might reflect the experience of the surgeons. The aim of the present study was not focused on evaluating the patients’ clinical results.

Despite these limitations, the present study demonstrated that EMAS is a reliable and effective EM alignment guide, at least as accurate as the latest EM alignment system with all the advantages compared to IM alignment systems, without requiring supplementary preoperative tests or long operation times.

5. Conclusion

The correct use of EMAS allows the restoration of the neutral mechanical axes of the lower limb, especially in cases of limited varus deformity. EMAS was the least accurate in terms of coronal alignment in cases of mLDFA >90°. Regarding sagittal alignment, we reported an increase of flexion of femoral component if MPTA<87°.

Authors' contributions

A.A. and A.P.G. were involved in the design of the study, A.A. performed the data collection; A.A. and A.P.G. analysed the data, all authors interpreted the data, A.A. and A.P.G. wrote the article. All authors read, reviewed, and approved the manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.