2021 Volume 3 Issue 1 Article ID: 2019-0021-OA
Objectives: Increasing attention has been paid to pelvic incidence (PI) as a potential parameter related to low back pain. However, little knowledge exists regarding potential anthropometric landmarks specialized for the estimation of PI. This study aimed to examine the inter- and intra-examiner reliability of potential anthropometric landmarks applicable to estimate PI. Methods: Twenty healthcare workers were recruited as participants. Three were experienced physiotherapists for more than 5 years in clinical practice. Eight anatomical landmarks were selected: (1) the acromion, (2) the upper edge of the iliac crest, (3) the posterior superior iliac spine (PSIS), (4) the anterior superior iliac spine (ASIS), (5) the upper edge of the greater trochanter, (6) the coccyx, (7) the lateral joint space of the knee, and (8) the lateral malleolus. Photographs of the right-side view of the subjects were used to determine the two-dimensional (x, y) coordinates of the landmarks. Results: Most landmark measurements reached acceptable levels for intra-examiner (ICC1, 0.64 to 0.98) and inter-examiner reliability (ICC3, 0.71 to 0.97). However, as possible anatomical landmarks, the PSIS (ICC1 0.65, ICC3 0.48), acromion (ICC3 0.66), and coccyx (ICC1 0.64) tended to have relatively low ICCs. Conclusions: Our study suggests that potential anthropometric landmarks on the body surface examined on palpation have acceptable intra- and inter-examiner reliability; however, identifying the acromion, PSIS, and coccyx as anatomical landmarks using the measurement method in this study remain difficult to be considered reliable.
Much attention has been focused recently on pelvic incidence (PI) as the basis of analysis for sagittal spinopelvic alignment in spinal surgery1,2,3). Likewise, PI, which is a sacropelvic parameter defined as the angle formed between the line perpendicular to the sacral plate at its midpoint and the line connecting this point to the axis of the femoral head4) (Figure 1), is regarded as a potential parameter relating to low back pain (LBP). The rationale behind this is that PI has a sacral slope in the pelvis at a constant angle regardless of the change in posture and is fixed at puberty4,5). Such individual pelvic parameters might represent the extent of different mechanical stresses on the lumbar vertebrae depending on the PI value. In the case of a small PI, for example due to physiologic lumbar lordosis, compressive force on the intervertebral disc will increase. Conversely, Labelle et al.1) revealed the association between a large PI and the severity of spondylolisthesis; thus, a large PI may lead to lumbar hyperlordosis, resulting in increased stress to the facet joint that is involved in spondylolisthesis (Figure 2). Thus, PI is the critical angle not just for determining lumbar lordosis, but also for understanding the pelvic rotation, inclination, and sagittal balance6). Furthermore, PI as an individual-specific parameter has the potential to provide a new perspective for epidemiological studies as a determinant of LBP, although it is challenging to apply in practice. Examiners need to take a spine lateral X-ray in a standardized upright position for the direct measurement of PI angular parameters. X-ray exposure solely for epidemiological studies in healthy individuals is impractical and must be strictly limited.
Measurement of Pelvic Incidence
Difference in mechanical stress of the lumbar vertebrae depending on the PI value
In contrast, ergonomics has provided an anthropometric measurement method that is used worldwide. For example, the ISO 7250-17) serves as a guide on how to take anthropometric measurements and provides information to ergonomists and designers on the definitions of anthropometric landmarks on the body surface. The ISO guidelines, however, mainly focus on the size of specific body segments when applying the information to the design of the places where people work and live. Little knowledge exists on the potential anthropometric landmarks specialized for the estimation of PI, as well as its reliability in measuring anthropometric landmarks on the body surface.
Therefore, prior to the development of indirect surrogate indicators reflecting PI using anthropometric landmarks, we tried to examine the reliability of potential anthropometric landmarks that physiotherapists attach to the body surface based on anatomical points.
To confirm the reliability of anthropometric landmarks for representing anatomical points on the human body surface, 20 health care workers (11 males, 9 females, mean age 35.0 ± 10.7 years, height 163.7 ± 9.0 cm, and body mass index 20.8 ± 2.7 kg/m2) employed in Nagoya City West Medical Center were evaluated. The inclusion criteria were as follows: (i) participants who had no subjective symptoms of LBP, (ii) those who could come to the measuring room after day shift (between 17:00 and 18:00) for 2 consecutive days to control the effects of inter-day variation on their postures, and (iii) those who could maintain their static upright standing posture. Participants with severe obesity (waist circumference ≥100 cm) were excluded. We explained the experimental procedures to the participants and acquired their written informed consent. This research was approved by the Ethics Review Committee of Nagoya City University (60-16-0013) and Nagoya City West Medical Center (16-02-331-17).
Potential anthropometric landmarks based on anatomical pointsThe authors discussed potential anthropometric landmarks applicable for the estimation of PI and eight anatomical points were selected: (1) the acromion, (2) the highest point of the upper edge of the iliac crest, (3) the most protruding point of the posterior superior iliac spine (PSIS), (4) the most protruding point of the anterior superior iliac spine (ASIS), (5) the highest point of the upper edge of the greater trochanter, (6) the lower end of the coccyx, (7) midpoint of the lateral joint space of the knee, and (8) the most lateral point of the lateral malleolus. Furthermore, the navel was defined as the origin (0, 0) of the two-dimensional (x, y) coordinate system with the x-axis as the anteroposterior direction and the y-axis as the vertical direction. The distance between two fixed points placed on the poles was not only used for adjusting the x-axis tilt, but also for defining it as the known length for calibration when measuring the coordinates for each picture (Figure 3). We referred to landmark definitions of (1) the acromion and (4) the ASIS provided by the ISO 7250-1, but the others applied the general palpation methods for clinical and physiotherapeutic treatment, as follows: (2) the upper margin of the iliac crest is defined as the highest position when the upper margin of the pelvis is palpated. As for (3) PSIS, palpation of the iliac crest was performed, and the palpation was gradually extended to the rear from the median to slightly outside. (5) The upper margin of the greater trochanter can be identified by palpating the largest bulge with the palm of the examiner and moving it upward until the bones could not be touched. (6) The coccyx, also known as the tailbone, can be palpated as a point at the lower end of the spine. (7) The lateral cleft of the knee joint can be palpated at the gap between the femur and tibia and defined as the center of a horizontal line of the gap between the anterior and posterior in the sagittal plane. (8) lateral malleolus The expanded lower end of the fibula situated on the lateral side of the leg at the ankle can be easily detected with palpation and direct observation.
Anthropometric landmarks in upright standing position
The examiners, three experienced male physiotherapists with over 5 years in clinical practice (mean age: 35.7±7.4 years), evaluated the participants in two sessions within more than a 24-hour interval. Laboratory conditions of both sessions were designated to be similar to confirm the intra-examiner reliability of the test and retest measurements.
In each session, the participants wore elastic spats and a thin T-shirt, as shown in Figure 3, and were requested to take a neutral body posture in an upright position. Then, the participants were asked to point to their navel using the index finger of the left hand and to take a posture of the upper right limb called “the clasp position” which is commonly used for the imaging of the entire spine for clinical diagnosis. They were asked to point to the left clavicle with their right index finger to make the right lateral measurement points visible, as shown in Figure 3.
The examiner located eight anthropometric landmarks on the body surface upon palpation and attached markers on them. To project the coccyx from the posterior of the coronal plane into the right side of the lateral plane, we used a solid ping-pong ball as a sphere marker. For other landmarks, flat circular reflective ones were used. After that, a photograph of the subject’s right lateral plane was taken using a digital camera (Power Shot S5, Canon Inc., Japan) fixed on a camera tripod at a height of 60 cm with the focus distance to the subject set to 3 m. The order in which the three examiners measured was randomly assigned to minimize the order effect bias.
Statistical analysisTo examine intra- and inter-examiner reliability of measurements, two-dimensional (x, y) coordinates of the landmarks in each picture were obtained using ImageJ ver. 1.48 software (NIH, Bethesda, MD, USA). The coordinate measurements (mm) taken at the center point of each marker were measured with the lengths of the rectangle made from each landmark and the navel. In general, PI is an angle that is defined on the sagittal plane; therefore, we focused on the analyses of anteroposterior and vertical directions on the sagittal plane, meaning that displacements on the coronal plane were ignored.
The intra-class correlation coefficients (ICC1) of each coordinate show the test-retest reliability of each examiner, between initial test (first session) and retest (second session) measurements of the eight anthropometric landmarks obtained from three examiners. Upon taking the single measurement of the 3 examiner’s measurements, the inter-class correlation coefficients (ICC3) of each coordinate indicate agreement between examiners. The ICCs were calculated with a 95% confidence interval (CI). As for the ICC criteria, we defined ICC <0.70 as “not acceptable,” 0.71–0.79 as “acceptable,” 0.80–0.89 as “very good,” and >0.90 as “excellent” reliability8,9). All statistical analyses were performed using R version 2.8.1 (R Foundation for Statistical Computing, Vienna, Austria)10).
Table 1 shows the intra- and inter-examiner reliability of body landmarks attached to the body surface. Most landmark measurements indicated a more than “acceptable” test-retest reliability (ICC1 point estimate range, 0.64 to 0.98) for each examiner, although the y-coordinate of the PSIS for examiner B and that of the coccyx for examiner C were reported to have relatively low ICCs (underlined). Likewise, the inter-examiner reliability (ICC3) indicated “acceptable” levels of consistency between examiners, except for the y-coordinate of the PSIS from the second session (ICC3=0.48) and the x-coordinate of the acromion from the first session (ICC3=0.66).
| Intraclass correlation coefficients (ICC1) within sessions | Interclass correlation coefficients (ICC3) between examiners | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Examiner A | Examiner B | Examiner C | First session | Second session | ||||||
| x-axis | y-axis | x-axis | y-axis | x-axis | y-axis | x-axis | y-axis | x-axis | y-axis | |
| (1) Acromion | 0.92 | 0.85 | 0.80 | 0.91 | 0.84 | 0.95 | 0.66 | 0.86 | 0.89 | 0.91 |
| (0.81–0.97) | (0.67–0.94) | (0.56–0.92) | (0.80–0.96) | (0.64–0.93) | (0.89–0.98) | (0.42–0.83) | (0.73–0.94) | (0.74–0.95) | (0.81–0.96) | |
| (2) Upper edge of the iliac crest | 0.89 | 0.79 | 0.86 | 0.94 | 0.91 | 0.9 | 0.92 | 0.83 | 0.86 | 0.80 |
| (0.75–0.96) | (0.54–0.91) | (0.68–0.94) | (0.84–0.97) | (0.78–0.96) | (0.76–0.96) | (0.84–0.97) | (0.68–0.93) | (0.72–0.94) | (0.63–0.91) | |
| (3) Posterior superior iliac spine (PSIS) | 0.91 | 0.92 | 0.88 | 0.65 | 0.92 | 0.70 | 0.85 | 0.75 | 0.88 | 0.48 |
| (0.79–0.97) | (0.81–0.97) | (0.73–0.95) | (0.30–0.85) | (0.80–0.97) | (0.34–0.87) | (0.71–0.93) | (0.55–0.89) | (0.77–0.95) | (0.20–0.73) | |
| (4) Anterior superior iliac spine (ASIS) | 0.84 | 0.95 | 0.83 | 0.83 | 0.91 | 0.90 | 0.87 | 0.92 | 0.93 | 0.89 |
| (0.63–0.94) | (0.87–0.98) | (0.62–0.93) | (0.62–0.93) | (0.78–0.96) | (0.77–0.96) | (0.74–0.94) | (0.83–0.96) | (0.85–0.97) | (0.78–0.95) | |
| (5) Upper edge of the greater trochanter | 0.85 | 0.79 | 0.92 | 0.75 | 0.91 | 0.91 | 0.89 | 0.71 | 0.90 | 0.8 |
| (0.66–0.94) | (0.54–0.91) | (0.81–0.97) | (0.47–0.90) | (0.78–0.96) | (0.78–0.96) | (0.78–0.95) | (0.49–0.87) | (0.79–0.96) | (0.62–0.91) | |
| (6) Coccyx | 0.94 | 0.91 | 0.95 | 0.77 | 0.94 | 0.64 | 0.95 | 0.84 | 0.95 | 0.77 |
| (0.85–0.98) | (0.78–0.96) | (0.89–0.98) | (0.49–0.90) | (0.85–0.98) | (0.29–0.84) | (0.90–0.98) | (0.69–0.93) | (0.90–0.98) | (0.58–0.90) | |
| (7) Lateral joint space of the knee | 0.77 | 0.87 | 0.85 | 0.87 | 0.86 | 0.92 | 0.72 | 0.77 | 0.79 | 0.87 |
| (0.52–0.90) | (0.71–0.95) | (0.66–0.94) | (0.71–0.95) | (0.70–0.94) | (0.80–0.96) | (0.51–0.87) | (0.59–0.89) | (0.62–0.90) | (0.74–0.94) | |
| (8) Lateral malleolus | 0.90 | 0.98 | 0.92 | 0.98 | 0.89 | 0.99 | 0.76 | 0.97 | 0.82 | 0.97 |
| (0.77–0.96) | (0.95–0.99) | (0.80–0.97) | (0.96–0.99) | (0.74–0.95) | (0.99–1.00) | (0.56–0.89) | (0.94–0.99) | (0.66–0.92) | (0.93–0.99) | |
Data are shown as point estimates with 95% confidence intervals of intraclass correlation coefficients (ICCs). The ICC1 data indicate association between initial test (first session) and retest (second session) measurements of the eight anthropometric landmarks obtained from three examiners. The ICC3 reliability estimates were calculated by taking the average of the 3 examiners’ measurements. The x-axis and y-axis correspond to the anteroposterior and vertical directions, respectively. The underlined values indicate ICC data <0.70, defined as “not acceptable” by Landis and Koch criteria.
We also conducted strata analyses of Table 1, shown in eTable 1A for females and eTable 1B for males, indicating that no significant sex difference can be found between each measurement item.
| Intraclass correlation coefficients (ICC1) within sessions | Interclass correlation coefficients (ICC3) between examiners | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Examiner A | Examiner B | Examiner C | First session | Second session | ||||||
| x-axis | y-axis | x-axis | y-axis | x-axis | y-axis | x-axis | y-axis | x-axis | y-axis | |
| (1) Acromion | 0.94 | 0.91 | 0.96 | 0.66 | 0.95 | 0.84 | 0.97 | 0.78 | 0.99 | 0.87 |
| (0.80–0.98) | (0.71–0.97) | (0.87–0.99) | (0.17–0.90) | (0.84–0.99) | (0.53–0.95) | (0.91–0.99) | (0.52–0.93) | (0.96–0.99) | (0.70–0.96) | |
| (2) Upper edge of the iliac crest | 0.80 | 0.66 | 0.68 | 0.96 | 0.90 | 0.92 | 0.75 | 0.93 | 0.81 | 0.76 |
| (0.37–0.95) | (0.10–0.91) | (0.12–0.92) | (0.84–0.99) | (0.64–0.98) | (0.70–0.98) | (0.42–0.93) | (0.79–0.98) | (0.54–0.95) | (0.43–0.93) | |
| (3) Posterior superior iliac spine (PSIS) | 0.96 | 0.92 | 0.91 | 0.68 | 0.88 | 0.71 | 0.79 | 0.42 | 0.76 | 0.52 |
| (0.85–0.99) | (0.72–0.98) | (0.68–0.98) | (0.11–0.91) | (0.60–0.97) | (0.17–0.92) | (0.49–0.94) | (0.01–0.80) | (0.44–0.93) | (0.11–0.84) | |
| (4) Anterior superior iliac spine (ASIS) | 0.82 | 0.97 | 0.76 | 0.80 | 0.87 | 0.88 | 0.88 | 0.87 | 0.79 | 0.77 |
| (0.42–0.96) | (0.89–0.99) | (0.28–0.95) | (0.37–0.95) | (0.55–0.97) | (0.55–0.97) | (0.68–0.97) | (0.67–0.98) | (0.49–0.94) | (0.47–0.94) | |
| (5) Upper edge of the greater trochanter | 0.86 | 0.84 | 0.97 | 0.84 | 0.98 | 0.90 | 0.86 | 0.89 | 0.95 | 0.91 |
| (0.54–0.96) | (0.49–0.96) | (0.90–0.99) | (0.50–0.96) | (0.94–0.99) | (0.65–0.98) | (0.66–0.96) | (0.70–0.97) | (0.87–0.99) | (0.74–0.98) | |
| (6) Coccyx | 0.89 | 0.92 | 0.88 | 0.88 | 0.93 | 0.71 | 0.86 | 0.65 | 0.93 | 0.71 |
| (0.67–0.97) | (0.75–0.98) | (0.63–0.96) | (0.63–0.96) | (0.77–0.98) | (0.25–0.91) | (0.67–0.96) | (0.31–0.88) | (0.77–0.98) | (0.25–0.91) | |
| (7) Lateral joint space of the knee | 0.81 | 0.93 | 0.75 | 0.86 | 0.96 | 0.89 | 0.60 | 0.77 | 0.74 | 0.89 |
| (0.41–0.95) | (0.74–0.98) | (0.25–0.94) | (0.50–0.96) | (0.85–0.99) | (0.61–0.97) | (0.20–0.88) | (0.45–0.94) | (0.41–0.93) | (0.69–0.97) | |
| (8) Lateral malleolus | 0.88 | 0.98 | 0.95 | 0.99 | 0.96 | 0.96 | 0.73 | 0.96 | 0.80 | 0.97 |
| (0.59–0.97) | (0.93–1.00) | (0.83–0.99) | (0.96–1.00) | (0.84–0.99) | (0.88–0.99) | (0.39–0.92) | (0.89–0.99) | (0.51–0.95) | (0.91–0.99) | |
Data are shown as point estimates and 95% confidence intervals of intraclass correlation coefficients (ICCs). The ICC1 data indicate association between initial test (first session) and retest (second session) measurements of the eight anthropometric landmarks obtained from three examiners. The ICC3 reliability estimates were calculated by taking the average of the 3 examiners’ measurements. The x-axis and y-axis correspond to anteroposterior and vertical directions, respectively.
| Intraclass correlation coefficients (ICC1) within sessions | Interclass correlation coefficients (ICC3) between examiners | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Examiner A | Examiner B | Examiner C | First session | Second session | ||||||
| x-axis | y-axis | x-axis | y-axis | x-axis | y-axis | x-axis | y-axis | x-axis | y-axis | |
| (1) Acromion | 0.77 | 0.83 | 0.84 | 0.77 | 0.94 | 0.71 | 0.87 | 0.87 | 0.95 | |
| (0.82–0.98) | (0.37–0.93) | (0.50–0.95) | (0.53–0.95) | (0.36–0.93) | (0.79–0.98) | (0.40–0.90) | (0.69–0.96) | (0.69–0.96) | (0.87–0.99) | |
| (2) Upper edge of the iliac crest | 0.92 | 0.97 | 0.90 | 0.64 | 0.90 | 0.90 | 0.92 | 0.74 | 0.92 | 0.72 |
| (0.76–0.98) | (0.89–0.99) | (0.69–0.97) | (0.84–0.99) | (0.64–0.98) | (0.68–0.97) | (0.81–0.98) | (0.45–0.91) | (0.80–0.98) | (0.42–0.91) | |
| (3) Posterior superior iliac spine (PSIS) | 0.89 | 0.92 | 0.88 | 0.64 | 0.93 | 0.71 | 0.86 | 0.65 | 0.98 | 0.59 |
| (0.67–0.97) | (0.75–0.98) | (0.63–0.96) | (0.13–0.89) | (0.77–0.98) | (0.25–0.91) | (0.57–0.96) | (0.31–0.88) | (0.93–0.99) | (0.23–0.85) | |
| (4) Anterior superior iliac spine (ASIS) | 0.84 | 0.89 | 0.86 | 0.81 | 0.92 | 0.89 | 0.91 | 0.81 | 0.97 | 0.84 |
| (0.52–0.95) | (0.65–0.97) | (0.57–0.96) | (0.47–0.95) | (0.74–0.98) | (0.66–0.97) | (0.77–0.97) | (0.58–0.94) | (0.93–0.99) | (0.63–0.95) | |
| (5) Upper edge of the greater trochanter | 0.84 | 0.72 | 0.90 | 0.68 | 0.87 | 0.91 | 0.93 | 0.59 | 0.90 | 0.53 |
| (0.54–0.96) | (0.26–0.91) | (0.68–0.97) | (0.19–0.90) | (0.62–0.96) | (0.70–0.97) | (0.81–0.98) | (0.23–0.85) | (0.75–0.97) | (0.16–0.83) | |
| (6) Coccyx | 0.94 | 0.91 | 0.96 | 0.66 | 0.95 | 0.84 | 0.96 | 0.78 | 0.99 | 0.87 |
| (0.80–0.98) | (0.71–0.97) | (0.87–0.99) | (0.17–0.89) | (0.84–0.98) | (0.53–0.95) | (0.91–0.99) | (0.52–0.93) | (0.96–1.00) | (0.70–0.96) | |
| (7) Lateral joint space of the knee | 0.73 | 0.69 | 0.91 | 0.83 | 0.77 | 0.90 | 0.83 | 0.62 | 0.84 | 0.79 |
| (0.29–0.92) | (0.21–0.90) | (0.72–0.98) | (0.51–0.95) | (0.38–0.93) | (0.70–0.97) | (0.61–0.95) | (0.27–0.87) | (0.63–0.95) | (0.53–0.93) | |
| (8) Lateral malleolus | 0.89 | 0.96 | 0.88 | 0.96 | 0.76 | 0.99 | 0.77 | 0.95 | 0.80 | 0.94 |
| (0.67–0.98) | (0.87–0.99) | (0.64–0.97) | (0.85–0.99) | (0.35–0.93) | (0.96–1.00) | (0.51–0.93) | (0.88–0.99) | (0.51–0.94) | (0.84–0.98) | |
Data are shown as point estimates and 95% confidence intervals of intraclass correlation coefficients (ICCs). The ICC1 data indicate association between initial test (first session) and retest (second session) measurements of the eight anthropometric landmarks obtained from three examiners. The ICC3 reliability estimates were calculated by taking the average of the 3 examiners’ measurements. x-axis and y-axis correspond to anteroposterior and vertical directions, respectively.
A visual inspection and palpation of a body segment are generally used for the analysis of a body segment alignment for clinical diagnosis or as a basis of anthropometric measurements with traditional goniometry for the ergonomic design of products or workplaces. There are a few reliability studies focusing on the lumbar curves in the standing position on visual inspection11,12), lumbar spinal levels by palpation13), or features of anthropometric data collected using three-dimensional body scanners14). However, despite its widespread use, research to confirm the reliability of the landmarks identified by palpation has been limited, especially with regards to the available landmarks targeted for PI estimation.
This study examined the reliability of eight possible anthropometric landmarks on anatomical points of the human body surface for the estimation of PI using such surrogate anthropometric landmarks. In general, manual spinal palpation studies11,12,13) have common features such as poor inter-examiner reliability between therapists, despite “acceptable” or “very good” agreements of intra-examiner reliability within therapists. Upon observing the data in the current study, all landmarks except the acromion, PSIS, and coccyx demonstrated acceptable inter- and intra-examiner reliability, indicating that experienced physiotherapists can identify the anatomical points and mark them on a stable body surface. In contrast, some misalignments assumed as a palpation discrepancy have occurred.
The y-coordinate of the PSIS for examiner B indicates a “not acceptable” reliability, meaning that the identification of the PSIS could be prone to misinterpretation by different examiners. One reason for the poor reliability might be the difficulty during palpation due to minimal bone swelling. Moriguchi et al.15) showed that some anatomical landmarks such as the ASIS or greater trochanter had larger discrepancies upon palpation in both intra- and inter-rater reliability. Upon palpation of the PSIS, it might be difficult to confirm the fossae lumbales laterales, called dimples of Venus, which is a cue to identify the PSIS under the fascia and ligaments. Thus, the ambiguous definition of the PSIS might lead to measurement error. Another possible explanation is the difficulty of anatomical landmark palpation owing to abdominal fat15). Severely obese participants (waist circumference ≥100 cm) were excluded from the present study, due to difficulty in identifying the landmark under abdominal fat, especially in the coronal plane. Furthermore, Kouchi and Mochimaru16) pointed out that skilled examiners have significantly smaller landmarking errors than the less-skilled examiners for many landmark measurements. Our results also indicated similar trends. Furthermore, a systematic review of the intra- and inter-examiner reliability of palpation for PSIS asymmetry does not indicate substantial levels of inter-examiner reliability supporting clinical utility17). This may be due to uncertainties in the projection between planes. The PSIS, in general, is palpated on the coronal plane, although the coordinates of the PSIS used in this study were projected to the sagittal plane. Abdominal subcutaneous fat thickness can also distort the coordinates of the PSIS due to increased difficulty in its identification upon palpation. The PSIS is one of the most prominent pelvic landmarks; however, it may be unsuitable to use the PSIS as a landmark in the sagittal plane.
The inter-examiner reliability (ICC3) of the acromion in the first session did not reach an acceptable reliability level. The acromion is the most lateral point of the lateral edge of the spine (acromial process) of the scapula, projected vertically to the surface of the skin, as defined in the ISO standard7). In general, the acromion is widely used as a reliable anatomical point that is easy to distinguish in ergonomic studies evaluated through electromyography of the trapezius muscles or assessed through the kinematics of the lumbar and cervical regions18). Nevertheless, the low reliability may be due to fluctuations in the upright standing posture differing from those of the attached landmark on the body surface. The participants were requested to take a neutral body posture in an upright position; however, there was a slight forward or backward tilt in some cases when comparing the photos of the first and second sessions. Deviation of the x-coordinates between the first and second photos on the sagittal plane might pertain to the neutral body posture when they were instructed. Furthermore, when photographing anatomical points on the sagittal plane of subjects, they were merely instructed to point to the left clavicle with the right index finger to make the points visible, as shown in Figure 3. The instructions were not strictly controlled between measurement days. Hence, this might lead to the error of the x-coordinate of the acromion being dependent on the position of the clavicle where the subject indicated.
As for the measurement of the coccyx, which is the final bone in the vertebral column, the y-coordinate measured by examiner C had low reliability. The shape of the coccyx, a small triangular bone at the base of the spinal column, may affect the reliability. The palpation of the coccyx could depend on the examiner’s experience and skill, such as those who tend to palpate the tip of the coccyx (fourth coccygeal vertebra), or those who tend to palpate around the coccygeal cornu or first coccygeal vertebra. However, the coccyx has been relatively neglected in anatomical research19), at the present time, which leads to difficulty in discussing the reason in more detail. For the reasons mentioned above, we can conclude that the PSIS, acromion, and coccyx are not suitable as stable indicators.
Strength and limitation of studyThis is the first study to examine the inter- and intra-examiner reliability of potential anthropometric landmarks applicable to estimate PI. A standardized photography procedure with specified focal length, angle of view, and height of lens was established, and the landmarks were strictly measured from the photographs with high accuracy. Despite this, some limitations persist. First, this study design cannot clearly identify the source of errors derived either from random error on landmark detection upon palpation, from fluctuations of upright standing posture, or from a systematic error due to lens distortion of a photograph. Second, the reliability of landmarks measured upon palpation does not assure generalizability, particularly for assessing overweight or obese patients, due to limited participants. Third, the sample size in this study was relatively small and subgroup analyses were not applied, meaning that confounding factors were not addressed. Fourth, the focal length and the vertical and horizontal angles of view were fixed when photographing. Theoretically, it can be assumed that quite few measurement errors will occur between repeated photos of the same subject, but the presence of body sway during quiet standing may affect this. This study did not account for such stability, and the effects remain unclear. Finally, the only instruction for controlling the standing location was to take a neutral body posture in an upright position and to point their navel using the index finger of the left hand, in addition to “the clasp position” that is ordinary used when imaging X-ray of the entire spine for clinical diagnosis. Such instructions might lead to some biases, such as inverting the acromion under such postures, compared to the acromion of participants keeping their Frankfurt plane horizontal.
Our study suggests that potential anthropometric landmarks on the body surface examined upon palpation have acceptable inter- and intra-examiner reliability; however, identifying anatomical landmarks of the acromion, PSIS, and coccyx using the measurement method in this study remains too difficult to be considered reliable. This evidence will provide a basis for an easy and simple way to estimate PI indirectly without X-rays, contributing the field of occupational and public health in the future.
This study was approved by the Institutional Review Board of Nagoya City University Graduate School of Medical Sciences (No. 60-16-0013) and Nagoya City West Medical Center (16-02-331-17).
Written informed consent was obtained from all participants in this study.
The authors declare no conflict of interests for this study.
We would like to express our gratitude to the staff (physiotherapists, occupational therapists, speech-language-hearing therapists, pharmacists, and nurses) of Nagoya City West Medical Center, who willingly agreed to participate in this study. We would like to thank Editage (www.editage.com) for English language editing.
TE and SY designed and developed the study protocol and measurement method. SK, TU, and SY collected the data. SY analyzed the data and wrote the first draft of the manuscript. TE supported the analysis and interpreted the data. MK and TE supervised SY with respect to ergonomics and occupational health. KA, SK, TU, and SY considered the interpretation of data from the physical therapist’s point of view. AI established the foundation for the smooth progress of research as the head of the data collection organization. TE was responsible for the study of PI and assisted with the drafting of the manuscript. All the authors interpreted the data, contributed to the writing of the manuscript, revised it critically for important intellectual content, and agreed with the final version and the findings.
