The importance of the human spine’s sagittal alignment, particularly in the lumbar region, has been argued from the physiological and pathophysiological points of view [1]. Lumbar curvature misalignment, particularly increased lumbar lordosis (LL) or hypolordosis, can, in the long term, lead to spinal instability and development of disorders such as low back pain and spondylolisthesis [2]. In this regard, some researchers examined the lumbar lordosis curvature (LLC), and some studies have evaluated the morphology and orientation of LL in asymptomatic individuals [3,4]. Moreover, some authors have raised the effect of LL morphology and its upper and lower segments on each other [5], resulting in some compensatory mechanisms and changes in the alignment of the upper or lower limb through a kinetic chain [6]. Therefore, explicit knowledge of the normal LLC values is important to better understand abnormal LL alignment or hyperkyphotic posture that can lead to postural misalignments or spinal disorders.

Previous studies have established mean values and dispersion of the spine’s lumbar curve and stressed that these anthropometrical parameters vary because of human diversity [3,7–10]. In addition, this diversity is possibly caused by the difference in the number of subjects, sex, and age diversity in these studies [3,7]. The inclusion of participants in different sex and age groups often results in a relatively low number of volunteers; therefore, the data related to the degree of lumbar curvature in these studies could be skewed by outliers because of limitations in their subject recruitment.

To the best of our knowledge, no large-scale study has analyzed a full range of ages in both sexes. In addition, various invasive and noninvasive techniques and tools have been used to evaluate the LLC. This diversity in methods could result in discrepancies among previous studies, causing confusion about what constitutes normal LL [11]. Most studies have assessed the LLC using invasive devices such as radiological methods [12,13], and their results could not be generalized to those measuring LLC using noninvasive tools. Nonetheless, this method is not ideal for studies with large samples because of the high cost and participants’ exposure to ionizing radiation. Thus, given ethical issues in using X-rays and the fact that X-rays are mainly indicated for severe abnormalities or structural abnormalities, the results cannot be comparable with those obtained using noninvasive devices, which mainly are applied to measure nonstructural or positional abnormalities. Accordingly, the use of a low-cost, valid, and reliable noninvasive tool such as Flexicurve (KF60; Shanghai Kearing Stationary Co. Ltd., Shanghai, China) is encouraged [14].

Measuring the LLC in a standing position is challenging. Given the importance of the lumbar region in establishing an acceptable LLC value, assessing it in a larger population and different age groups in both sexes is necessary. However, the use of a safe and valid measurement tool to establish the LLC norm may be used as a reference for further studies, physicians, and healthcare professionals and for comparison and evaluation whether their measurements were with the normal range.

The primary aim of this study was to establish normal LLC values measured by Flexicurve and investigate the association of sex, age, and body mass index (BMI) with LLC in the Iranian population aged 5–85 years. The secondary aim was to compare the LLC values between different age, sex, and BMI groups to understand the possible association of these variables with the LLC. We hypothesize that a significant difference in the LLC exists between the different sex, age, and BMI groups.

This study was based on the results of a cross-sectional study of asymptomatic volunteers from Iran. All participants provided informed consent to participate in the study. This study was conducted in compliance with the principles of the Declaration of Helsinki and was approved by the Sports Sciences Research Institute of Iran.

Table 1 presents the demographic characteristics of the participants. A significant three-way interaction was found between sex, age, and BMI (F(2, 20)=2.944, p=0.0001). Therefore, a two-way ANOVA was performed to examine the effect of sex and age groups on LLC (Table 2). A significant interaction was found between the effects of sex and age groups on LLC (F(2,7)=3.908, p=0.0001). Simple main effect analysis showed that female LLC was significantly higher than male LLC by approximately 7.5° across the eight age groups. However, no difference was found between sexes in participants aged <10 years (Table 2, Fig. 1A). Age was found to be associated with LLC (F=19.557, p=0.0001). However, this association was significant only in children (aged <10 years) and older people (age >70 years) compared with other age groups (Fig. 1A). Moreover, age-related trends in LLC were illustrated in the present study and others for men (Fig. 1B) and women (Fig. 1C) separately.

Simple main effects analysis for BMI showed that people with obesity had significantly higher LLC than people with normal BMI when they were categorized into adolescent and adult groups (p<0.05); however, no differences were found between BMI groups of younger participants (p=0.875) (Fig. 2). Moreover, sex, age, and BMI showed significant association with LLC, particularly moderate (r=0.31, p=0.001) and small (r=0.13, p=0.001), (r=0.2, p=0.001), respectively.

This study revealed that the normal LLC ranges from 10.2° to 74.9° with a mean of 42.34°±13° for the whole Iranian population, 38.57°±11.44° for the male population, and 46°±13.38° for the female population. This study showed a constant increase in LLC with increasing age up to approximately 60 years with sex differences, with women having higher LLC than men. However, from this point, the LLC starts to decrease gradually, and these values in men decrease to the value at the age of 5. The female LLC starts to decrease at the age of 60; however, values are still higher than male LLC. As a whole, a small significant correlation was found between age and LLC, which agrees with previous results [3,17].

Our finding also agrees with the results of other studies using radiographic methods to measure lumbosacral lordosis in boys and girls aged 5–20 years [20] and men and women aged 20–29 years [21]. However, decreases were observed in isolated groups, e.g., aged 40–49 years. This was interpreted as a casual occurrence.

Despite general agreement between our findings with previously reported values of LLC [7–9,22], LLC values varied in some investigators. These variations and differences in LLC between studies are due to the diversity in age range, measurement methods or tools, sample size, and BMI [3,7–10,23–36] (Tables 3–5, Fig. 3). Nevertheless, our finding can be compared favorably with the results of two similar studies that measured LLC using similar measuring methods and tools [3,37]. Our LLC values were smaller than those reported by Youdas et al. [3]; however, the difference was not considerable. Moreover, Link et al. [37] used a flexible ruler and reported a mean value of 34.41°±9.85° in 61 English men aged 20–30 years, which is lower than our values in this age category (41.06°±12.87°). The reason for these differences between the two studies could be the wide LLC ranges in the study by Link et al. [37] (18.92°–62.96°) compared with our value (38.99°–43.13°). In studies that used surface measurement tools to measure LLC, our values are close to those reported by Norton et al. [23] who used a computerized three-dimensional electromechanical digitizer (Metrecom; FARO Technologies Inc., Lake Mary, FL, USA). On the contrary, LLC values from a study that used Spinal Mouse (IDIAG, Fehraltorf, Switzerland) were lower considerably (male, 17.3°±0.5°; female, 29.6°±0.7°) [8]. Moreover, studies that used radiological devices reported a variety of LLC values measured from the superior endplate of T12/L1 and S1 using the Cobb method. Some of them [7,24] agreed, whereas others [25,26,38] did not approve the values obtained using Flexicurve (Fig. 3).

This study also compared the LLC values between different age, sex, and BMI groups to understand the possible association of these variables with the LLC in the Iranian population. In this study, women had larger LLCs than men by approximately 7.5° across all age groups. In agreement with our results, Youdas et al. [3] reported that men have lower LLC than women by approximately 6.5°. Comparing other studies with surface or radiological measurements, inconsistent results regarding sex effects on LLC were noted, however, in major studies, women showed higher LLC values than men with a difference of 0.2°–12.3°. Despite debates concerning the significant influence of sex on LLC, a recent systematic review and meta-analysis showed convincing evidence for the sex differences in LLC, indicating that women have larger LLC, possibly because of the greater sacral slope, than men [11]. Moreover, the greater LLC for women could be due to differences in the vertebral shapes and the anatomical and functional capacity differences that could affect biomechanical factors in the standing position [9].

In addition, to sex, age was also a conflicting variable that could affect the LLC. The difference in LLC between age groups was significant in this study. Significant differences in LLC were found between children (age <10 years) and older people (age >70 years) with other age groups (middle-aged group). The LLC values for the youngest and oldest groups were significantly lower than the others (middle-aged groups). This may depend on the orientation of the pelvis, lumbosacral joint, and sacrum with increasing age [39]. In addition, the intervertebral disk may flatten, particularly in the middle lumbar segments, because of degenerative processes and might reduce the LLC in the old age [11]. These age-related decreases in LLC in older adults could have been associated with recruiting compensation mechanisms in the lower limb (e.g., increased knee flexion angle) to maintain the sagittal global balance [40]. However, the actual effect of age might not be clearly defined until a longitudinal study is completed. Compared with studies investigating the LLC in 10-year increments of age, the age-related trends for LLC in the present study were very similar to that in the study by Youdas et al. [3], particularly for women [3,30,33,40] (Fig. 1). The main reason might be the use of a similar tool (Flexicurve) to measure LLC in both studies.

Regarding the BMI, adolescents and adults with obesity had significantly higher LLC than those with normal BMI; in contrast, these variables were not significantly different among children. Fig. 2 illustrates that the mean LLC was higher in normal and obese groups. This is in agreement with a small positive correlation between BMI and LLC in this study. However, contradictory results have been reported with either positive or negative correlations between BMI and LLC. As mentioned before, these contradictory results in the literature might be due to the dissimilarity in measurement techniques and tools, subjects, sex, and age diversity.

A limitation of this study is its cross-sectional design. Because the study involved a one-time measurement of exposure and outcome, causal relationships could not be derived from the data analysis. In addition, while the Flexicurve technique is reliable, it cannot be used to treat pathological conditions. The results of this study can only be compared with other studies that employed the same patient positioning and measurement tools. Furthermore, this study was conducted on a specific population, so the findings may not be generalizable to other countries or ethnicities.

However, the cross-sectional design is very useful for establishing normative data for specific or multiple variables at one time across a large population, including different age groups and sexes. A notable strength of this study is the large sample size and broad population coverage. The use of an accurate, simple, and inexpensive tool to measure the variable of interest allowed us to make robust conclusions regarding sex-, age-, and BMI-related trends and values in LLC.

This study showed a significant difference in LLC between different age groups in both sexes. These differences were more obvious in children (age <10 years) and older people (age >70 years) in comparison with other age groups. Therefore, in using our LLC values as normative data, age and sex should be considered particularly when using them as a comparative norm for children and older people of both sexes. In this study, we used a simple, inexpensive, and reliable method and tools to measure and establish normative values for the Iranian normal population aged 5–85 years. The LLC norm established using a safe and valid measurement tool can be a valuable reference when comparing measurements obtained in further studies by physicians and healthcare professionals.