Reconsidering the Gold Standard for the Assessment of Spastic Equinus Foot

Abstract

Introduction:

To determine whether differences exist in the measured range of motion (ROM) by the goniometer, equinometer, and 3-dimensional (3D) gait analysis in different settings (under general anesthesia (GA) or in the clinic) among patients with cerebral palsy.

Methods:

A prospective comparative cohort study was conducted at our orthopaedics center, where 15 patients with diplegic cerebral palsy (30 limbs) were evaluated, all of whom had Gross Motor Function Classification System (GMFCS) level I and II. ROM was measured by (1) goniometer under GA (benchmark), (2) goniometer analysis during clinical examination, (3) equinometer under GA, (4) equinometer in the clinic, and (5) 3D gait analysis. ROM was measured during both knee flexion and extension. Subgroup analysis based on the GMFCS level was performed.

Results:

Nine patients were male with GMFCS level I and a mean age of 14.3 (SD=7.2) years. Statistically significant differences were noted between all studied measurement techniques in terms of ROM, dorsiflexion, and plantarflexion during both knee flexion and extension. The GMFCS level was an effect modifier of ROM measurements.

Discussion:

The ROM parameters during both knee flexion and extension differed from one measurement device/technique to another. The GMFCS level played a notable effect-modifying role on the ROM parameters.

Equinus deformity of the ankle joint is one of the most frequently encountered deformities in patients with cerebral palsy (CP), accounting for 71 to 99% of affected feet.¹ This deformity occurs as a result of the inability to adequately maintain ankle dorsiflexion that would allow proper contact with the supporting surface. In turn, this leads to an imposed and maintained tension on the Achilles tendon.² This is associated with a shortening of the gastrocnemius muscle. The resultant spasticity of the calf muscles leads to limitation in the range of motion (ROM). Nevertheless, in diagnosing equinus foot, their methods varied: 14% depended exclusively on measuring foot dorsiflexion beyond plantigrade with a straightened knee and a neutral hindfoot while 86% used the ‘Silfverskjöld test’ along with the aforementioned technique.³

Spasticity occurs because of increased intensity of the stretch reflex which is further exacerbated by movement velocity or speed. The clinical assessment of spasticity is not conclusive given the factor that velocity plays in this process; hence, the results are considered subjective.⁴ This highlights the need and the importance of instrumental assessment of spasticity as it provides a more objective and valid interpretation of the subject's ROM as compared with clinical assessment.

The measurement of ROM, particularly ankle dorsiflexion (DF), can be performed by a wide variety of static and dynamic instruments, including radiographs, dynamometers, 3-dimensional (3D) gait analysis,⁵ goniometers, and many others. The goniometer is the benchmark in measuring ROM and in guiding therapeutic decisions in clinical practice.⁶ In our previous research, we designed and validated another instrument called the equinometer in measuring ROM among healthy individuals.⁷ It has shown comparable results with the goniometer.

The benchmark for evaluating equinus foot remains examination under general anesthesia (GA); however, no research has been conducted to determine whether examination in the awake state would result in different results, particularly in patients who are not severely affected (with Gross Motor Function Classification System [GMFCS] level I or II). Thus, we carried out this investigation to compare examination under GA with that in the awake state (by using equinometer, goniometer, and 3D gait analysis) while hypothesizing that both settings provide similar, nonstatistically different, values.

Methods

This study followed the Strengthening the Reporting of Observational studies in Epidemiology guidelines for the reporting of observational studies.

Participants

The prospective cohort study was performed on 15 patients with the diagnosis of spastic CP at our center between October 1, 2020, and September 30, 2022. Patients were recruited at the special outpatient clinic for CP. Only patients with GMFCS levels I and II, aged 6 to 40 years, who were scheduled for surgery on at least one foot to increase dorsiflexion were included in this study. The exclusion criteria were age <6 or >40 years, GMFCS level ≥ III, lack of cooperation, or if there was no informed consent to participate in this study. The study protocol was approved by the Institutional Review Board of our hospital (S-487/2019) and registered in the German Register for Clinical Studies (DRKS, Number: DRKS00018350). Written informed consent was taken from each participant. Baseline data, including age, sex, diagnosis, and GMFCS level were collected and recorded. We followed the Strengthening the Reporting of Observational studies in Epidemiology guidelines in the conduct and reporting of this research.

Outcomes and Examination

All participants underwent an examination of the ankle joint ROM by three methods: goniometer, equinometer, and 3D gait analysis. Goniometer and equinometer measurements were performed at 2 times: under clinical conditions immediately before surgery in an examination room and at the second time point in the operating room under GA.

The 3D gait analysis was performed before surgery. The period was up to 6 months before surgery and measurement under clinical conditions because it was not possible to plan it otherwise because of the COVID-19 pandemic. The examination room consisted of an examination table, a universal plastic goniometer (Manufacturer GIMA), and the equinometer. Each participant was examined by one rater with experience in handling the equinometer. In addition, the examiner received preparatory training on how to report measurements of hindfoot movement with the goniometer.

Goniometer Measurements

Measurements were performed on 30 feet, with participants lying in the supine position on the examination table. The detailed measurement method was provided previously,⁷ and the accuracy of this tool has already been proven.⁸ This procedure was performed under clinical examination (in the clinic) and under GA immediately before the planned surgery. The measurement under GA was set as the benchmark or reference measurement.

Equinometer Measurements

The Equinometer (Supplemental Digital Content 1, http://links.lww.com/JG9/A325) was developed⁹ and validated⁷ by our department. The detailed description of the components and the protocol used for the equinometer were previously described.⁷ The measurement was performed by one experienced examiner. At least three values were recorded, and the average value was calculated.

Instrumental 3D Gait Analysis

The 3D gait analysis was performed in the gait laboratory ‘Motion Laboratory’ in our center. The accuracy of this tool was previously established.¹⁰ Reflective markers were applied to the skin using bony landmarks and recorded in three dimensions by an optical measurement system. Gait analysis was performed on each of the 15 participants without walking aids or orthoses. The DF and plantarflexion (PF) of the ankle joint during several gait cycles were recorded as numbers by the software and noted and stored in a database. Only the values necessary for the study were extracted from all the data of the recording. These were the maximum dorsiflexion of the ankle joint, which was reached during stance phase. The maximum values during the swing phase were not considered because the measurement conditions were too different. These extracted values were then compared with the values of the DF of the benchmark.

Statistical Analysis

SPSS was used for data analysis, and the Prism software (GraphPad version 8) was used to make figures. Categorical variables were presented as numbers and percentages while numerical variables were presented as means and standard deviations. Each measurement method was compared with the benchmark. The paired Student t-test was used to determine the differences in the means of two groups. All investigated methods were compared with the same benchmark goniometer in each comparison/test.

Results

Patients' Characteristics

Fifteen patients with CP (30 limbs) with spastic diplegia were included, of whom nine were male. The mean age of patients during the time of measurement was 14.3 (SD=7.2) years. Nine patients had GMFCS level I, and the remaining six patients (40%) had GMFCS level II.

Comparison Between Measurement Scales Regarding Range Of Motion

Range of Motion During Knee Extension

In knee extension, statistically significant differences in ROM were noted between the different measurement methods (Figure 1A). As compared with the benchmark, equinometer in the clinic showed significantly higher ROM (42.2° versus 40.4°, P = 0.033). However, goniometer in the clinic (37.8) and equinometer under GA (38.7°) had significantly lower ROM, whereas 3D gait analysis had the lowest ROM of all (23.3°).

Figure 1.

Graph showing the difference between the equinometer, goniometer, and 3D gait analysis during knee extension in terms of (A) range of motion, (B) maximum dorsiflexion, and (C) maximum plantarflexion.

Regarding maximal DF, no differences were observed between the benchmark and both equinometer measurements. Meanwhile, the measurements from both the goniometer in the clinic (0.9° versus 3.9°, P < 0.001) and the 3D gait analysis (10.6° versus 3.9°, P < 0.001) were significantly different from the benchmark (Figure 1B).

Maximal PF was significantly different in the equinometer in the clinic (−37.7° versus −36.5°, P = 0.033) and 3D gait analysis (−12.7° versus −36.5°, P < 0.001) from the benchmark. Meanwhile, the degree of maximal PF was not different between the benchmark and both the goniometer in the clinic and the equinometer under GA (Figure 1C).

Range of Motion During Knee Flexion

The differences in ROM during knee flexion between the different methods are illustrated in Figure 2A. As compared with the benchmark, both the goniometer in the clinic (42.6° versus 45.2°, P = 0.002) and equinometer during GA (43.1° versus 45.2°, P = 0.033) had significantly lower ROM, respectively. Although equinometer in the clinic resulted in slightly higher ROM than the benchmark, this difference did not reach statistical significance (P = 0.12).

Figure 2.

Graph showing the difference between the equinometer, goniometer, and 3D gait analysis during knee flexion in terms of (A) range of motion, (B) maximum dorsiflexion, and (C) maximum plantarflexion.

Regarding maximal DF, the goniometer in the clinic was significantly different from the benchmark (5.1° versus 6.4°, P = 0.033), whereas no differences were observed between the benchmark and both equinometer measurements (Figure 2B).

Maximum PF was significantly different between the benchmark and both the goniometer in the clinic (−38.8° versus −37.5°, P = 0.033) and the equinometer under GA (−38.8° versus −37.7°, P = 0.033); however, no difference was noted from the equinometer in the clinic (Figure 2C).

Stratified Range of Motion Analysis by the GMFCS Level

GMFCS Level I During Knee Extension

Twelve limbs were analyzed (Figure 3). No differences in passive DF were noted between the benchmark and equinometer both under GA (P = 0.12) and in the clinic (P = 0.12), respectively. However, the difference between the benchmark and goniometer in the clinic resulted in a significant increase in the degree of DF in the positive direction (P = 0.033). Meanwhile, the difference between the benchmark and the 3D gait analysis resulted in a shift in the negative direction (P = 0.002).

Figure 3.

Graph showing the difference in range of motion between the equinometer, goniometer, and 3D gait analysis during knee extension in patients with cerebral palsy with Gross Motor Function Classification System level I. GMFCS = Gross Motor Function Classification System

The same was observed regarding maximum PF. No differences were noted between the benchmark and the equinometer measurements both under GA and in the clinic (P = 0.12). Similarly, the goniometer in the clinic resulted in a difference from the benchmark in the positive direction (P = 0.033), whereas the difference between it and the 3D gait analysis was observed in the negative direction (P < 0.001).

GMFCS Level I during Knee Flexion

Twelve limbs were analyzed (Supplemental Digital Content 2, http://links.lww.com/JG9/A326). No differences were noted between the benchmark and all remaining methods regarding both maximum DF and PF among patients with spastic CP with GMFCS level I.

GMFCS Level II during Knee Extension

Eighteen limbs with GMFCS level II were analyzed (Supplemental Digital Content 3, http://links.lww.com/JG9/A327). The benchmark resulted in a difference in maximum DF from the goniometer in the clinic (P < 0.001) and the 3D gait analysis (P = 0.002). However, no notable differences were noted in DF between the benchmark and both equinometer measurements.

Both the equinometer in the clinic (P = 0.002) and the 3D gait analysis (P < 0.001) resulted in difference in maximum PF from the benchmark. However, the measurements of maximum PF were not different between the benchmark and both the goniometer in the clinic and the equinometer under GA.

GMFCS Level II during Knee Flexion

The benchmark resulted in differences in maximum DF from the goniometer in the clinic (P = 0.033) and the equinometer under GA (P < 0.001) (Supplemental Digital Content 4, http://links.lww.com/JG9/A328). However, no differences were noted in DF between the benchmark and the equinometer in the clinic (P > 0.05).

Regarding maximum PF, both the equinometer in the clinic (P = 0.033) and the goniometer in the clinic (P = 0.033) resulted in difference from the benchmark. However, the measurements of maximum PF were not different between the benchmark and both the goniometer in the clinic and the equinometer under GA.

Discussion

In this study, we noted significant differences in ROM parameters (both dorsiflexion and plantar flexion) during knee extension and flexion between all measurement methods. Our study highlights that without anesthesia, ROM measurements become lower with the use of goniometer and equinometer. In addition, ROM measurement by the equinometer is remarkably higher under clinical conditions than under GA. This could potentially be explained by the presence of midfoot break which may result in a clinically notable reduction in maximum ankle DF with a notable increase in maximum midfoot DF.¹¹

Almost all ROM parameters differed markedly at all positions (knee flexion and extension) based on the setting in which the goniometer was used. The use of the goniometer in the outpatient clinic resulted in different measurements from that in the operating room when the patient is under GA. This has been previously described in the literature. Standard goniometry (under GA) results in better measurement accuracy as compared with when patients are fully conscious. For clinical research purposes, the use of the standard goniometer is preferred because it leads to improved reproducibility of ROM parameters.¹² This finding highlights the importance of standardizing the technique in which the goniometer is used to be able to provide accurate outcome comparison and to assess the efficacy of surgical interventions used to treat equinus deformity.¹³

The equinometer⁹ was designed to fills the gaps associated with the goniometer, such as the low reliability and the inability to manually control velocity, given the variability in the muscle tone of CP children.¹⁴ However, the ‘equinometer’ makes it easy to measure the ROM in different positions by using well-designated landmarks which eases the positioning of the foot in a neutral position for proper measurement. In our latest research, we validated the use of this device in clinical practice, and given its considerably high reliability and accuracy, we recommended its use as an alternative method to the standard goniometer as we noted no significant difference in the measured ROM parameters between both parameters.⁹ Unlike our previous observation, in this study, we noted significant differences in ROM measurements (ie, PF) between both the equinometer and goniometer during knee flexion. This discrepancy could be related to other factors that are individual-related. For instance, in our previous research, we compared the goniometer with the equinometer among healthy individuals with normal gait; however, in this study, we included patients with CP, and their clinical findings during evaluations could have attributed to this difference.

However, it is important to mention that during both knee flexion and extension, maximum DF and PF did not differ between the standard goniometer and the equinometer measurement in the clinic except for PF during knee extension. This highlights that the equinometer in the clinic can provide reproducible results to the standard goniometer in clinical settings. In addition, 3D gait analysis has been described to potentially alter the decision-making process in patients with CP based on the measured ROM parameters.¹⁵ In our study, 3D gait analysis resulted in markedly different DF and PF from the benchmark goniometer.

As previously mentioned, the difference in ROM measurements could be attributed to individual-related factors.⁶ In our study, we investigated one of these factors, which is the GMFCS level. For example, during knee flexion, no notable differences were noted regarding DF and PF between all measurement methods among patients with CP with GMFCS level I. However, in patients with GMFCS level II, notable differences in ROM measurements were observed in maximum DF and PF. This should be further investigated in future research. Notably, the effect-modifying role of GMFCS level ceased regarding DF during knee extension, where the differences in ROM measurements were observed in both subgroups (patients with GMFCS level I and level II).

These findings highlight the importance of putting the type and setting of ROM into account when diagnosing patients with equinus deformity because the notable changes observed among them could lead to the misdiagnosis of equinus foot, where a cutoff value of ≤ 5° dorsiflexion is used to clinically define this abnormality.¹⁶

Study Limitations and Recommendations for Future Research

We highlight the discrepancy in the reported ROM parameters based on the type and timing of measurement. This can potentially affect the surgical decision-making process and subsequently alter patients' outcomes. Our sample size might not have been enough to limit the occurrence of random errors. Patients' age⁷ and other factors⁶ should be considered for their effect-modifying role. Finally, we did not perform between-method comparison; therefore, future research will focus on the comparison between all methods in a network approach.

In summary, our study highlights that the setting in which the ROM measurement method is applied results in different measurements and that GMFCS plays a notable effect-modifying role in this regard.

Supplementary Material