Comparison of gait between stroke patients and healthy volunteers using knee-ankle-foot orthoses with adjustable knee flexion

Abstract

[Purpose] This study aimed to develop an easy-to-operate knee joint for knee-ankle-foot orthoses with four adjustable levels of knee flexion movement and ascertain how different ranges of motion in flexion affect the knee flexion angle during walking. [Participants and Methods] Eight stroke patients participated, for whom knee joints of knee-ankle-foot orthoses were constructed. During walking, the knee flexion angle was measured for the following four knee joint settings: free flexion, 15° of flexion, 30° of flexion, and fixed extension. These measurements were compared to those of eight healthy volunteers from a previous study. [Results] Gait analysis revealed that the knee flexion angle during mid-stance was significantly greater in stroke patients than in healthy volunteers when set to 30° of flexion and in free flexion. [Conclusion] Appropriate adjustment of the range of knee mobility using a knee joint developed for knee-ankle-foot orthoses can prevent disuse atrophy caused by using knee-ankle-foot orthoses with a fixed knee joint.

INTRODUCTION

Japanese stroke management guidelines published in 2021 recommend knee-ankle-foot orthoses (KAFOs) for stroke patients with poor weight-bearing due to impaired strength of the knee extensors and muscles around the hip joints¹⁾. However, Ohata reported decreased activity in the rectus femoris and hamstrings during walking after 1 month of knee immobilization with a KAFO²⁾. Studies of ankle-foot orthoses (AFOs) have raised concerns that immobilization of the ankle joint by an AFO can result in disuse atrophy of the tibialis anterior^{3, 4)}. In our previous study of 8 stroke patients who had required a KAFO during walking, we found increases in the vastus medialis activity ratio during loading response when the knee joint of the KAFO was set to allow 30° of flexion compared with when it was fixed in extension⁵⁾. The results suggest that performing walking exercises using a KAFO with such an adjustable knee joint may reduce the risk of disuse atrophy of muscles around the knee joint. However, that same study also showed excessive knee flexion in stance phase when the knee joint of the KAFO was allowed to move from 0° to 30° flexion. To address this problem, we have developed a KAFO knee joint with the following four settings—fixed in extension (fixation), 15° flexion range (15° movement), 30° flexion range (30° movement), and free flexion (free)—and we have demonstrated the effect of these settings on knee joint flexion angle during walking in healthy individuals⁶⁾. The objective of the present study was to perform gait measurements in recovering stroke patients in order to determine the differences in knee flexion angle between these patients and the healthy individuals from our previous study.

PARTICIPANTS AND METHODS

For recovering stroke patients, inclusion criteria were first occurrence of stroke; in the recovery phase, within 180 days of onset of stroke; and prescribed a KAFO with an adjustable knee joint. The exclusion criterion was inability to walk using a KAFO knee joint set to “free” with the assistance of a physical therapist. For volunteers, inclusion criteria were healthy undergraduate students aged 20 years or older and exclusion criteria were movement disorders and deformities in the lower limb joints.

The study participants were 8 recovering stroke patients (5 men, 3 women; age, 59.3 ± 10.1 years; height, 167.1 ± 8.8 cm; weight, 57.5 ± 10.3 kg) and 8 healthy volunteers (6 men and 2 women; age, 21.4 ± 0.4 years; height, 167.8 ± 7.4 cm; weight, 54.4 ± 8.6 kg). Each participant’s lower limb was molded to fabricate a personalized KAFO. The KAFO employed a knee joint in which the flexion range of motion could be adjusted to “fixation”, “15° movement”, “30° movement”, and “free” by first loosening the nut and then rotating the cam without using tools (Figs. 1 and 2). The foot was restricted to 0° of plantar flexion and free dorsiflexion by using a double Krenzak foot joint. A temporary fitting of the KAFO was done by the prosthetist for a compatibility check, and any incompatibilities were corrected. The knee joint was set by a physical therapist. The knee joint used in this study can be precisely set to four positions using a cam mechanism: “fixation”, “15° movement”, “30° movement”, and “free”.

Fig. 1.

Knee joint used in this study.

Fig. 2.

Four conditions for the knee–ankle–foot orthosis knee joint.

The measurement method was the same as previously reported⁶⁾. The participants wore their personalized KAFO and walked for approximately 10 m in a straight line. Measurements were made with the KAFO knee joint set first to “fixation”, followed by “15° movement”, “30° movement”, and “free”, and gait measurements were taken after three practice runs using each setting. The measurements for the stroke patients were taken at a walking speed where the patients could walk safely without using a cane while being assisted from behind by a physical therapist and after it was determined that there was no risk of fall even with the “free” knee joint setting. The measurements for the healthy volunteers were taken at a walking speed where the volunteers could walk comfortably without using a cane or the assistance of a physical therapist.

Using an adhesive elastic bandage, two inertial sensors (AMWS020B; ATR-Promotions, Kyoto, Japan) were affixed to the front of the lower leg of the KAFO-wearing limb and the front of the thigh. Gait measurement software (Sensor Controller, ATR-Promotions) was used to perform measurements with a 1,000-Hz sampling frequency. Then, data-analysis software (Sensor Data Analyzer; ATR-Promotions) was used to compute the knee flexion angle from the sagittal angular acceleration of the lower leg and thigh. Data for three to five gait cycles were extracted from the recorded data. The initial contact point was defined by the appearance of a spike-like waveform in vertical acceleration recorded by the inertial sensor affixed to the lower leg. After normalizing the time of each extracted gait cycle to 100%, the values of the gait cycles were averaged. The maximum knee flexion angles were measured during loading response, mid-stance, terminal stance, pre-swing, and swing⁷⁾. The Mann–Whitney U test was used to compare the maximum flexion angles during each phase of the walking cycle in stroke patients and healthy volunteers under the four knee joint settings. All statistical analyses were performed using JSTAT ver. 16.1, and statistical significance was set at a p-value of <0.05.

This study complies with the ethical standards of the 1964 Declaration of Helsinki and its subsequent revisions. This study was approved by the Niigata University of Health and Welfare ethics committee (approval number: 19021-230414). The study was explained in detail using an explanatory document and written informed consent was obtained from all participants before their inclusion in the study.

RESULTS

Table 1 presents the clinical evaluation results for the 8 stroke patients, while Fig. 3 shows the mean knee flexion angles of the 8 stroke patients and 8 healthy volunteers under the four different knee joint settings. The maximum knee flexion angles were as follows: loading response, 3–12%; mid-stance, 13–31%; terminal stance, 32–50%; pre-swing phase 51–62%; and swing phase, 63–100%. During mid-stance, the knee flexion angle was significantly greater in the stroke patients than in the healthy volunteers under the “30° movement” and “free” settings (p<0.05). In the pre-swing and swing phases, the knee flexion angle was significantly greater in the healthy volunteers than in the stroke patients under the “free” setting (p<0.05) (Table 2).

Table 1. Clinical characteristics of the 8 stroke patients and 8 healthy volunteers.

	Number	Age	Sex	Height	Weight	Disease type	Paralyzed side	Stroke onset to measurement date	Brunnstrom stage	Functional Independence Measure
Stroke patients	1	71	Woman	160	52.5	Hemorrhage	Right	106	Ⅲ	82
	2	60	Man	163	52.3	Hemorrhage	Right	57	Ⅲ	67
	3	76	Man	158	38.9	Hemorrhage	Right	86	Ⅲ	56
	4	47	Man	176	66.1	Infarction	Right	98	Ⅲ	63
	5	54	Man	180	72	Hemorrhage	Left	112	Ⅳ	102
	6	54	Woman	167	54	Infarction	Left	67	Ⅳ	93
	7	62	Man	175	60	Infarction	Left	92	Ⅲ	43
	8	50	Woman	158	64	Hemorrhage	Right	163	Ⅲ	98
	Average	59.3		167.1	57.5			97.6		75.5
	Standard deviation	10.1		8.8	10.3			32.3		21.5
Healthy volunteers	1	21	Man	173	61
	2	21	Man	170	56
	3	21	Woman	158	44
	4	21	Man	175	65
	5	23	Man	169	49
	6	21	Man	160.2	44.8
	7	22	Woman	160	50
	8	21	Man	177	65
	Average	21.4		167.8	54.4
	Standard deviation	0.7		7.4	8.6

Fig. 3.

Average values of knee joint angles in the 8 stroke patients and 8 healthy volunteers.

Table 2. Comparison of knee flexion angle between stroke patients and healthy adults under four conditions.

Range	Conditions	8 stroke patients		8 healthy volunteers		p-value
		Median	Interquartile range	Median	Interquartile range
Loading response	Fixation	3.1	1.1	3.5	2.5
	15° movement	6.8	8.6	11.0	1.9
	30° movement	9.1	12.8	11.2	3.5
	Free	19.4	14.9	8.0	5.1
Mid-stance	Fixation	3.7	0.5	2.2	1.5
	15° movement	10.3	6.3	10.4	2.9
	30° movement	21.1	7.8	11.5	4.4	*
	Free	23.7	17.8	9.2	3.9	*
Terminal stance	Fixation	2.7	2.5	5.4	5.4
	15° movement	7.0	1.7	12.7	5.3
	30° movement	11.8	6.0	9.2	6.1
	Free	9.2	15.1	10.9	3.4
Pre-swing	Fixation	3.7	1.6	8.5	7.2
	15° movement	12.8	9.9	17.9	3.8
	30° movement	14.8	21.5	29.8	6.1
	Free	14.5	27.4	41.3	7.5	*
Swing	Fixation	7.2	1.9	6.9	4.0
	15° movement	15.9	6.3	15.5	3.1
	30° movement	32.3	3.8	30.4	6.6
	Free	47.2	7.0	54.0	8.3	*

Flexion direction is positive. *Significant difference according to the Mann–Whitney U test (p<0.05).

DISCUSSION

In the healthy volunteers, the knee flexion angle was about 10° during loading response and subsequently decreased during mid-stance under the “fixation”, “15° movement”, “30° movement”, and “free” settings. In contrast, the stroke patients showed an increased knee flexion angle during mid-stance compared with loading response under the “30° movement” and “free” settings, suggesting an excessively flexed knee compared with normal gait. According to physical therapy guidelines in Japan, task-oriented exercise with variable difficulty, which involves repeating tasks similar to the target movements, is useful for improving activities of daily living and mobility⁸⁾. For patients with stroke, fine adjustment of the difficulty level may be required. For example, under the “30° movement” or “free” setting, if the flexion angle of the knee joint becomes larger during mid-stance than during loading response, the setting should be changed to “fixation” or “15° movement”.

This study has some limitations. First, walking speed was not controlled in this study, so differences in walking speed between stroke patients and healthy volunteers may have affected the results. The age difference between the stroke patient group and the healthy control group may have had a significant impact on the results, and the small sample size may have limited the statistical power. Also, the knee joint flexion angles, but not the actual muscle activities, were measured and analyzed. However, our previous study showed greater increases in the vastus medialis activity ratio during loading response when the knee joint was at 30° movement than when it was at fixation⁵⁾. Furthermore, a study of AFOs showed increases in the tibialis anterior activity ratio in patients with stroke after using an AFO with plantarflexion resistance for 2 months⁹⁾. Taken together, disuse atrophy resulting from the use of KAFOs with a fixed knee joint as reported in previous studies^2,3,4) can be prevented by appropriately adjusting the mobility range using the knee joint developed for KAFOs.

The significance of this study is that the proposed knee joint, when used in gait practice with a KAFO, allows the flexion range of motion of the joint to be gradually increased as the weight-bearing capability of the lower limb improves, which would enable task-oriented practice similar to walking with an ankle-foot orthosis, which is the target movement for the next stage.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

The authors declare no conflict of interest.