Subjects
In a mountain village (Katashina Village, Gunma, Japan, population 4573, 2230 males, 2343 females in 2017), local medical examinations are administered annually to screen for lifestyle-related diseases. The total number of participants in 2017 was 946. Of these, we recruited subjects who were > 40 years old for a foot checkup by board-certified orthopedic surgeons, regardless of the presence of symptoms. Ultimately, 173 healthy volunteers participated in the checkup.
All of the individuals provided their written informed consent and responded to a baseline questionnaire, which asked for information, such as the age and gender. Of the 173 participants, a total of 169 (63 males, 106 females, average age: 66.0 ± 12.4 years old) who satisfied all of the examination criteria below were included in the present study, which was approved by our institutional review board. All procedures were performed according to the Declaration of Helsinki.
Measurement of the photographic HV angle (pHVA)
With a digital camera, we took photographs of the participants’ feet in the standing position from a 15° inclined angle relative to the vertical direction from a distance of 100 cm and then measured the pHVA (Fig. 1) [14, 15]. First, we drew a tangential line from the inside edge of the hallux (A) to the inside edge of the head part of the first metatarsal bone (B). We then drew a line of the same length as AB from point B in the direction of the heel. Where the line came into contact with the inside edge of the first metatarsal bone was defined as point C. Using an angle measurement application (hakarun®; onegland.net, Shizuoka, Japan), two examiners (HO and KT) measured the pHVA as the angle formed by AB and BC.
A previous study reported that pHVA values were lower than the radiographic HV angle (rHVA) values, with a mean difference of -5.3° (95% confidence interval [CI], -4.3 to -6.2) [15]. HV is generally defined as an rHVA of ≥ 20° [16]. In the present study, to prevent classifying participants with an rHVA < 20° into the HV group, we set the cut-off as a pHVA of 20° and classified subjects into the HV group (pHVA of 1 or both feet ≥ 20°) and no HV group (pHVA of both feet < 20°). In addition, to analyze the relationship between the severity of HV and the magnitude of postural sway, we calculated the total pHVA for both feet.
Hallux pain
We asked participants about pain in their right and left hallux or first metatarsal joint while standing on both feet. We assessed the subjective pain of both feet using the Numerical Rating Scale (NRS) (0–10), in which a higher score indicates greater pain, and we further calculated the total NRS score for both feet.
Postural sway
Using a force plate (WIN POD®; medicapteurs, Balma, France), we assessed participants’ static balance in standing position with 2-legged stance and eyes open [5, 17] (Fig. 2). We asked participants to stand on the force plate barefoot, with heels 10 cm apart, and to stare at a point 1.5 m away from the force plate for 30 seconds. From a safety perspective, measurements in a single-leg stance or with the eyes closed were not conducted. We recorded the total trajectory length acquired by the measurement of the gravity center fluctuation (mm), area of the center of pressure (COP; the area of an oval circumscribed to the gravity center fluctuation; mm2) and mediolateral and anteroposterior postural sway (mm). The mediolateral axis of the center of gravity was defined as the X-axis, and the anteroposterior axis of the center of gravity was defined as the Y-axis. The magnitude of the mediolateral and anteroposterior postural sway was expressed as the root-mean-square COP displacement, relative to the mean (mm) [7].
Measurement of the body muscle mass
Using a bioelectrical impedance analysis (Tanita MC-780A multi frequency segmental body composition analyzer; Tanita Co., Ltd., Tokyo, Japan), we investigated the trunk and lower limb muscle mass [18, 19]. The participants stepped onto the analyzer barefoot, each foot on 2 electrodes, and the examiner entered the participants’ information (age, gender and height). Once the body mass was assessed by the scale, the participants held 2 electrodes gripped in each hand during impedance measurement, which was performed for 20 seconds. Segmental muscle mass values were indicated on the digital screen (trunk, left and right lower limbs). We then calculated the total muscle mass of both lower limbs (lower limb muscle mass).
Statistical analyses
We assessed the intraclass correlation coefficients (ICCs1.1) and interclass correlation coefficients (ICCs2.1) for the procedure used to measure the pHVA. We performed univariate analyses using Welch’s t-test and the chi-square test, and compared the HV and no HV groups. Using Spearman’s correlation coefficient, we then analyzed the single correlation between the pHVA and magnitude of postural sway. After adjusting for potential confounders, we used a multiple linear regression model to further confirm the association between the presence of HV or pHVA and the magnitude of postural sway. A p-value of 0.05 was considered statistically significant. All statistical analyses were conducted using a statistical analysis system (IBM SPSS Statistics 26 software package; IBM Japan, Tokyo, Japan).
The sample size was calculated (α = 0.05, β = 0.2) before the present study. The minimum sample size was assessed using a statistical power analysis program (G*Power Version 3.1.; Universität Düsseldorf, Düsseldorf, Germany). The minimum sample size required for each group to achieve a statistical power of > 0.8 was 33 in the HV group and 79 in the no HV group [2, 5].