Keywords
alignment - tibiofemoral angle - preoperative templating - total knee arthroplasty
- reliability
Introduction
Total knee arthroplasty (TKA) has evolved as one of the most effective surgical procedures
in the management of painful, symptomatic osteoarthritis (OA) of knee following failed
conservative treatment. The primary aim of TKA is to improve knee function and range
of movement, decrease pain, and restore the mechanical axis.[1]
[2]
[3]
Many factors influence the successful outcomes of TKA including patient's selection,
pre- and postoperative rehabilitation programs, surgical experience, and prosthesis
design.[4]
[5]
[6]
[7]
[8]
[9] In more recent practice, a great deal of focus has been on preoperative templating
to aid surgical planning and bone and soft tissue releases to achieve optimal limb-length
alignment.[7]
[10]
[11]
[12] Digital preoperative templating has been found to be particularly helpful in allowing
surgeons to plan their surgical approach, bone osteotomies, anticipate and deal with
challenging intraoperative encounters, predict implant sizes and overall, reduce the
risk of premature implant failure.[7]
[10]
[11]
[12]
Routine anteroposterior (AP) long leg alignment and lateral and skyline radiographs,
with a standard magnification have historically been used for preoperative TKA templating.
In most recent studies, a large emphasis has been placed on the importance of attaining
a neutral coronal alignment (CA) of the knee. Various measurements have been established
and described in the literature.[13]
[14]
[15] Despite this, determining the ideal CA poses a great challenge for reconstructive
knee surgeons.[16]
[17]
[18]
[19]
[20]
We describe a new linear measurement that can be utilized to analyze the CA on long
leg alignment radiographs, which has never been described in the literature. We feel
this approach is more simple, quick and can be easily calculated, making it a valuable
tool for preoperative templating for TKA, for orthopaedic and early career surgeons.
Additionally, this measurement can complement other established measurements used
by knee surgeons, increasing their confidence in accurately completing osteotomies.
Patients and Methods
Study Design
Local ethical committee approval was obtained for this project. A retrospective evaluation
of Radiology Information System and Picture Archiving and Communication System was
performed to identify 50 consecutive cases referred to knee clinics over a 1-year
period with knee pain. The imaging acquired included weight bearing, long leg alignment
AP views of the lower limb including all three joints of the ipsilateral hip, knee,
and ankle articulations. All images were calibrated to represent real-life measurements;
hence, a marker was not required. Patients with a history involving surgery of the
knee, hip or ankle, and lower limb, infection, or tumors were excluded.
Image Analysis
All radiographs were analyzed by senior author (RB), musculoskeletal radiologist with
over 10 years of experience, and one radiology registrar. In addition, one reader
repeated measurements after a cooling period to evaluate for intraobserver reliability.
Calculation of the Linear Coronal Knee Offset
The linear coronal knee offset (LCKO) is measured on weight bearing lower limb AP
radiographic imaging. For utmost precision, the patella must be aligned in the AP
projection in the center of the femoral condyles with both ankle in neutral position.
In addition, it is crucial to verify that the images are not rotated, and that patients
do not have any flexion deformities as such conditions could potentially impact the
accuracy of the measurements. The LCKO is measured by first, drawing a lateral line
tangential to greater trochanter and lateral malleolus. A second perpendicular line
is drawn from the lateral edge of lateral tibial plateau to the first line. The distance
of the perpendicular line corresponds to the LCKO. ([Figs. 1] and [2]).
Fig. 1 Schematic showing measurement of linear coronal knee offset (LCKO). Line a is lateral
line tangential to greater trochanter and lateral malleolus. Line b (LCKO) which is
perpendicular line is drawn from the lateral edge of lateral tibial plateau to the
line a.
Fig. 2 Long leg films showing measurement of linear coronal knee offset (red line) in three
cases.
Methodology
Data Collection
Data collected included patient demographics, indication, standard radiological measurement
of the anatomical tibiofemoral angle (TFA), and a LCKO measurement calculated for
each patient. TFA of 5 to 7 degrees of valgus was considered as normal, less than
5 degrees was varus, and over 7 degrees was valgus orientation of the knee. Data was
recorded on a Microsoft Excel data spreadsheet and SPSS 24.0 software (SPSS Inc. Chicago,
Illinois, United States). Analysis of variance was used for statistical analysis.
Kappa correlation was used to assess intra- and interobserver reliability. The intraclass
correlation coefficient (ICC) is usually between 0 and 1 (below 0.5 indicate poor,
0.5 to 0.75 being moderate, 0.75 to 0.9 good, and over 0.9 suggests excellent reliability).
Results
Among 50 consecutive cases (100 lower limbs), there was a male predominance (30 males
and 20 female). Based on the TFA, there were three cohorts. There were more males
in the normal and valgus cohort with equal proportion in the varus cohort. The mean
age of entire cohort patients was 36.3 years (range: 12–80 years). Descriptive statistics
of three cohorts are described in [Table 1]. The LCKO was statistically significant between the three cohorts. The mean LCKO
in normal cohorts was 0.24cm (99% confidence interval [CI]: −0.068–0.58), varus was
−0.6 (99% CI: 1.84–0.65,) and valgus was 1.72cm (99% CI: 1.38–2.1; [Table 1]) There was good intra- and interobserver reliability (Kappa of 0.8 and 0.8, respectively).
Table 1
Descriptive statistics of the linear coronal knee offset (LCKO) among the three study
cohorts of patients
|
Parameters
|
Valgus knee
|
Normal
|
Varus knee
|
|
Number of patients
|
58
|
20
|
22
|
|
Average age (years)
|
35.37
|
30.16
|
43.42
|
|
Maximum age (years)
|
80
|
80
|
80
|
|
Minimum age (years)
|
12
|
16
|
12
|
|
Male
|
36
|
13
|
11
|
|
Female
|
22
|
7
|
11
|
|
Mean
|
1.72
|
0.24
|
-0.6
|
|
Standard deviation (cm)
|
1.10
|
0.45
|
2.06
|
|
Standard error of mean (SEM)
|
0.13
|
0.10
|
0.44
|
|
90% confidence interval (CI)
|
1.52–1.95
|
0.07–0.42
|
−1.35–0.16
|
|
95% confidence interval (CI)
|
1.47–2
|
0.04–0.45
|
−1.51–0.32
|
|
99% confidence interval (CI)
|
1.38–2.1
|
−0.068–0.580
|
−1.84–0.65
|
|
Minimum
|
0
|
−0.5
|
−6
|
|
Median
|
1.6
|
0
|
−0.3
|
|
Maximum
|
4.40
|
1.30
|
3.00
|
Discussion
Frontal lower limb malalignment (FLLM) is strongly interrelated with progression of
knee OA, and the measurement of FLLM is helpful in apprehending the progression as
well as guiding the management of knee OA.[1]
[2]
[3]
[4] The importance of achieving a neutral alignment of the knee has been well acknowledged
in the literature through multiple clinical and biomechanical studies, and has been
proven to balance soft tissue forces, improve knee function, reduce implant failure
and revision rates.[5]
[6]
[7] Establishing the precise CA for TKA, however, remains to be one of the greatest
challenges for reconstructive knee surgeons.[7]
[8]
[9]
[10] To yield a near successful alignment, accurate preoperative planning is essential,
and with the modern advanced use of digital software imaging systems, the measurement
of FLLM has become swifter and more straightforward in producing satisfactory patient
outcomes.[11]
[12]
[13]
Standing whole lower limb plain radiographs form the benchmark technique is evaluating
FLLM in the weight bearing axis as well as other contributing angles that may lead
to a deformity.[4] Different measurements and techniques have been described in the literature which
include, the valgus correction angle (VCA), Mikulicz line, and mechanical mTFA.
The mTFA has historically been the gold-standard technique in calculating FLLM.[14] This is done by drawing a straight line from the center of the knee to the center
of the femoral head (mechanical axis of the femur), and from the center of the knee
to the ankle (mechanical axis of the tibia). Subsequently, the medial angle formed
by the mechanical axis of the femur and tibia is known as the hip–knee–ankle angle
(HKA), and measures 0/180 degrees in normal knees.[15]
[16]
[17]
Although commonly utilized, many factors influence the HKA, yielding inaccurate and
variable results. These include patient height, pelvic width, femoral head deformities,
or inadequate radiographs.[16]
[18]
[19]
[20]
[21] In these situations, the VCA can instead be used. This is measured by determining
the angle between two lines—the angle between a straight line connecting the center
of the femoral head to the center of the distal femur and a second line along the
anatomical long axis of the tibia. The resultant angle is approximately 6 ± 1 degrees
of valgus with the mTFA. A normal VCA is interpreted as a value between 173 and 175
degrees, where a greater value indicates a varus deformity and a lesser value indicated
a valgus deformity. However, comparable to the mTFA, many studies have proved the
unreliable results of the VCA, which can significantly be affected by axial limb rotation
and flexion deformities, with high variability and low interobserver reliability in
multiple studies.[22]
[23]
[24] Moreover, both the latter are angular measurements, hence, can be difficult to analyze
and interpret, especially for those surgeons early within their careers.
The novel LCKO measurement we feel provides a simpler method in assessing FLLM, with
high interobserver reliability. Moreover, it can easily identify a normal, varus,
or valgus knee deformity with a subsequent increase in deformity classification dependent
on the values. We also found this measurement easier and quicker to calculate than
conventional methods and can be used as an adjunct to other measurements to categorize
knee deformities as well as aid with surgical planning, especially for early career
surgeons.
We hope this uncomplicated technique can be used in future clinical studies and aid
clinicians with preoperative planning for conventional TKA, high tibial osteotomies,
as well as analyzing postoperative robotic knee procedures that preoperative utilize
long leg alignment films for referencing.
Limitations
There are a few limitations that we can associate with this study. First, this being
a retrospective study with a relatively small sample size that can reduce the robustness
of the findings. Despite this, the LCKO was found to possess high ICC with a score
of 0.8 and thus increases our confidence in the findings.
Conclusion
This study described an innovative method to measure coronal lower limb malalignment.
The LCKO is easier to calculate with good intra- and inter-observer reliability. This
can complement traditional methods to evaluate angular knee deformities.
Larger, multicenter studies will be beneficial to underpin our findings and support
decision-making process in patients undergoing TKA achieve mechanical limb alignment.
