The focus of the present analysis was on the comparison of overall upper body posture between people with and without low back pain, taking particular account of their gender, age and BMI. For this purpose, video raster stereography was used to evaluate not only the individual angles but also the entire spine, shoulder and pelvic area in the sagittal, frontal and transverse planes. Constitutional parameters such as the trunk length or shoulder and pelvic distance were excluded from the analysis.

Although the group comparison of all test subjects showed isolated significant differences in the frontal plane in the spine and shoulder areas, and in the transverse plane in the shoulder and pelvic areas, these were found to be not clinically relevant. These differences are predominantly within the range of measurement error or are very marginally outside of it, therefore, this observation appears to be a natural variance in posture. Similar results and conclusions can be seen in the gender-specific analysis. While the women predominantly showed significant rotational differences in all three areas (standard deviation of rotation, shoulder and pelvis rotation), the male participants only showed significance in the frontal trunk decline (frontal plane) and in the pelvis rotation (transverse plane).

Based on the available data, the previous findings of reviews by Chun et al. [12] and Chaléat-Valayer et al. [13] which showed that people with LBP tend to have a smaller lumbar lordotic angle (LLA) compared to age-matched healthy individuals could not be confirmed in the present study. In addition, our study could not confirm that there was a slight difference in the sagittal alignment of the spine and pelvis between patients with chronic low back pain and the control subjects. There were significant, albeit small, differences in the transverse and frontal planes but not in the sagittal plane. However, since the majority of CT and MRI scans are obtained with the patient in a supine position, and X-rays as well as whole spine scans are conducted while the patient is standing, including back scans, comparing the results of sagittal spine alignment using different methods may lead to misleading interpretations [22, 23].

The subsequent regression analysis analysed the influence of the independent variables group (LPB/no-LBP), age, BMI, weight, height and sex (male/female) with regard to the evaluation parameters for each area. Here, the independent variable group showed an effect on only a few dependent variable correlations (frontal trunk decline, scapular rotation and pelvis rotation). Thereby, the variables of sex and age dominated in the parameters of the sagittal plane. Other influencing factors such as age, BMI, height or weight also seem to have a significant influence on upper body posture more frequently than group affiliation. Age is mostly associated with gender and not with the group membership. Overall, sex appears to be the main influencing factor for upper body posture. Taking into account the regression coefficient and the clinical classification of this, a clinically meaningful consequence in relation to gender is only obtained for the sagittal parameters, i.e. the kyphosis and lordosis angles. Accordingly, the kyphosis angle was found to be 8.52° and the lordosis angle 16.11° lower in men than in women. With regard to the kyphosis angle, the BMI can also be classified as a relevant influencing factor. f the BMI increased by a value of 1, then the kyphosis angle was found to increase by 3.32°. As the BMI is calculated from the height and weight, these significant p-values can be classified as follows: if the height increases by 10 cm, then the kyphosis angle increases by 6.14° and if the weight increases by 1 kg, then angle decreases by 0.74°. Furthermore, in the sagittal plane, the aforementioned angles are the only clinically relevant results in the spinal region.

Previous analyses of the data from the no-LBP group by Ohlendorf et al. [20] yielded similar results as both the lordosis and kyphosis angles increased significantly with age regardless of gender. Women generally have higher values than men, being approximately 18° for the lordosis and 6° for the kyphosis angles [20]. Abrisham et al. [19] also found in 403 EOS imaging data that the mean lordosis angle was greater in women than in men in all age groups. Murrie et al. [24] also found no difference in lumbar lordosis between 27 LBP and 29 no-LBP subjects using magnetic resonance imaging (MRI). However, they also confirmed that lumbar lordosis is significantly more pronounced in women and people with a higher body mass index. Nonetheless, they were unable to detect any significant age-related changes. They attributed this to the genetic aspect that the female pelvis differs anatomically from the male. Mirzashahi et al. [15] also investigated the role of spinopelvic parameters as risk factors for non-specific lumbar spine pain using radiographic images. The comparison of 148 LBP and 148 controls showed that the female sex, higher BMI, smoking and blue-collar jobs were associated with a higher risk of non-specific LBP, whereby lumbar lordosis, sacral slope and pelvic tilt were all greater in LPB patients [15].

The positive correlations between BMI and the kyphosis angle have also been demonstrated in other studies. Increased body weight results in increased body mass; this creates a caudally directed load under the influence of gravity and, thus, favours a forward leaning posture. The kyphotic posture is reinforced by the anterior tilt of the body, shifting the body’s centre of gravity anteriorly [6, 25]. In this thematic context, Do Nascimento et al. [26] studied 25 volunteers (20w/5m) aged 18 to 40 years, 10 of whom had a BMI of 18 to 25 kg/m² and 15 of whom had a BMI ≤ 30 kg/m². Using visual inspection and the Balance System (Biodex), the overweight volunteers were found to have increased protrusion of the head, hyperkyphosis of the thoracic spine and hyperlordosis of the lumbar spine.

According to Souza et al. [27], the abdomen is displaced forward in overweight people which leads to an anterior shift of the body’s centre of gravity, resulting in increased lordosis and anteversion of the pelvis. The thoracic kyphosis also increases in this context. Increased body weight primarily affects the spine as it acts as a scaffold against gravity in an attempt to hold the person upright. The increased axial load caused by the increased body mass leads to attempts by the spine to compensate by adopting kyphotic, lordotic or scoliotic postures [26]. This results in overloading of the bones and joints and overstretching of the ligaments [28]. However, this could only be partially proven in the study data presented here as the lordosis angle was not significantly subject to this influence.

Limitations must also be taken into account in this study. When measuring people with a higher BMI, skin displacements or a thicker dermis, as found in obese patients, must be taken into account as these can lead to inaccuracies in the measurements. However, Drerup et al. [29], describe that the reproducibility of the marking of the vertebra prominens and the lumbar dimples (DL/DR) is 1 mm. The sensitivity of this method is 98% and the specificity 84% according to Asamoah et al. [30] compared to X-ray.

In order to be able to draw more precise conclusions and better comparisons with the back parameters, X-ray images should be included in future analyses. These would allow a better overview to be formed of the skeletal condition of the patient and, if necessary, information about any prolapse present. It would also be desirable to record the success of the therapy, i.e. data from the back scan as part of the therapeutic treatment, in order to gain a better understanding of the significance of pain or discomfort and its relation to the body’s posture. Pain sensitivity is an important issue that can severely affect a patient’s physical and mental state. It is, therefore, all the more important to pay close attention to this topic and to investigate it in the future.