Background

Low back pain (LBP) is characterized as pain from below the last ribs to the gluteal margin [1,2,3,4]. It can be classified according to the duration of symptoms as acute LBP, lasting up to six weeks, subacute, from six to 12 weeks, or chronic, lasting more than 12 weeks [2]. More than 85% of cases do not have a specific cause and, therefore, are diagnosed as non-specific LBP of musculoskeletal origin [3]. LBP has affected millions of people over the years, both in high-income countries and in middle- and low-income countries [5, 6]. Treatment recommendations for LBP vary depending on their classification [1, 3, 4, 7, 8]. Physical activity, manual therapy, and supervised exercise are recommendations for patients with acute and subacute LBP [1, 4, 8, 9]. For the treatment of chronic non-specific LBP, the guidelines recommend exercise as the first line of treatment, without specifying the best type of exercise[1,2,3,4, 9].

Studies have already been conducted to understand the biomechanical behavior and the changes in the body regions involved during episodes of LBP [10,11,12,13]. Biomechanical changes in patients with LBP are not exclusively related to the trunk. The proximity of the hip joint to the lumbar region directed biomechanical investigations to the hip [14,15,16,17,18,19], and several studies on the topic were carried out. These studies investigated the association between hip range of motion and non-specific LBP[19], compared the lower limb muscle strength of patients with LBP to the strength of healthy individuals [15], performed kinematic analysis of the hip during sitting and lifting movements in patients with LBP [14], and investigated the activation of hip muscles in patients with LBP during the standing position [16], movements in the sagittal plane[18]and in the prone hip test [17]. These biomechanical studies[14,15,16,17,18,19] showed different results, therefore a summary of their findings would facilitate the understanding of the role of the hip joint in non-specific LBP. Thus, the present study aimed to systematically review observational studies that used biomechanical hip assessment in patients with non-specific LBP.

Methods

Study design

Systematic review written according to the guidelines of the Preferred Report Items for Systematic Review and Meta-Analyses (PRISMA) and prospectively registered in the International Prospective Register of Systematic Reviews (CRD42020213599).

Inclusion criteria

We included observational studies, i.e., cross-sectional, cohort, and case–control studies, that performed biomechanical hip assessment in patients with acute, subacute, and chronic non-specific LBP, by measuring muscle strength, range of motion, kinematics, muscle activation, balance, or posture. Patients of both sexes should be over 18 years of age. Studies with pregnant women and patients diagnosed with LBP due to nerve root compromise and severe causes, such as neoplasms, inflammatory diseases, infections, and traumas [3] were not included. In studies that presented data from patients with non-specific LBP and another diagnosis, only data from patients with non-specific LBP were extracted. Eligible studies had to be published in full in peer-reviewed scientific journals.

Search strategy

The search was performed in the following databases: PubMed, Embase, Cinahl, and Sportdiscus on February 22nd, 2024. The descriptors used were extracted from the Medical Subject Heading (MeSH) and divided into four blocks: 1) type of study, 2) LBP, 3) hip, and 4) biomechanical assessment (Additional file 1). The descriptors were combined to perform the searches with OR between the terms of each block and AND between the blocks. There was no restriction on language and date of publication.

Study selection

Two independent reviewers (GZP and CMNC) conducted the selection process of the studies, first considering the title and abstract, and then the full reading of the study. Disagreements between reviewers were resolved initially by discussion and, in case of persistence, a third reviewer (GCM) reached a consensus.

Data extraction

Data were extracted by two independent reviewers (GZP and GCM) using a customized spreadsheet. The spreadsheet contained bibliometric data such as date of publication of the study, country, language, and authors; objectives of the study; personal and clinical characteristics of the patients such as age, sex, and duration of pain; sample size; description of the type of study; biomechanical variables and description of the assessment; as well as results. Disagreements between reviewers were resolved initially by discussion and, in case of persistence, a third reviewer (CMNC) reached a consensus.

Methodological quality assessment

The methodological quality of the included studies was assessed using the Epidemiological Appraisal Instrument [20]. This instrument consists of 43 questions divided into five scales: 1) description, with 17 questions; 2) subject selection, with seven questions; 3) measurement quality, with 10 questions; 4) data analysis, with seven questions; and 5) generalization of results, with two questions. Each question was scored on a scale of 0 to 2, where 0 is “no” or “not informed”, 1 is “maybe” or “partial”, and 2 is “yes”. Questions not applicable to the type of study were disregarded. The scale score was calculated by adding the scores of each question and dividing the total by the number of questions applicable to the type of study used in the assessment. Case–control studies were evaluated with 38 questions, cohort studies with 39 questions, and cross-sectional studies with 34 questions. The methodological quality of the studies was classified as low when the studies had scores between 0 and 0.65, moderate when the studies had scores between 0.7 and 1.35, and high when the studies had scores between 1.4 and 2 [21]. The scale score and the methodological quality classification were performed by two independent reviewers (GZP and GCM). In case of disagreement, a third reviewer (CMNC) reached a consensus.

Data analysis

The results extracted from the studies were presented descriptively: mean and standard deviation per group for the case–control and cross-sectional studies and by effect size and confidence interval for the cohort studies. When these data were not presented in the study, two emails with the request were sent to the authors within seven days. If there was no response, the data were presented as provided in the study. The results were grouped by type of biomechanical assessment (range of motion, strength, kinematics, and electromyography) and, later, by type of observational study (cross-sectional, case–control, and cohort) and classification of LBP (acute, subacute, and chronic). In studies with more than one type of biomechanical assessment, the outcomes were presented separately in the text considering the biomechanical assessment (some studies were cited more than once throughout the text). Due to the heterogeneity found in the biomechanical hip assessments in patients with LBP, it was not possible to group the results into meta-analyses.

Results

Study selection

The search strategy returned 338 studies: 116 duplicates were excluded and 123 were excluded after reading the titles and abstracts. Of the 99 studies for full-text reading, 47 were excluded: 11 because the participants did not present LBP, nine because they were not observational studies, 12 because they did not perform biomechanical hip assessment, two because they included pregnant women, eight because they included participants under 18 years of age, and five because they were abstracts presented at conferences. The authors of the abstracts were contacted to clarify whether the studies were published in a scientific journal. As a result, one study was added after contact with the authors. Another study was added after manual search. In total, 54 studies were included for data extraction (Fig. 1).

Fig. 1
figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PRISMA) flow diagram summarizing study selection processes

Full size image

Methodological quality

None of the studies presented poor methodological quality (Table 1). Of the studies on range of motion, eight [22,23,24,25,26,27,28,29] presented moderate methodological quality and one [30] had high methodological quality. Three studies on strength [31,32,33] had moderate methodological quality and one [34] had high methodological quality. Of the studies on kinematics, 13 [35,36,37,38,39,40,41,42,43,44,45,46,47] presented moderate methodological quality, and three [48,49,50] had high methodological quality. Of the studies on electromyography, five [17, 18, 27, 51, 52] presented moderate methodological quality, and two [53, 54] had high methodological quality. Finally, of the studies that evaluated more than one outcome, 14 [16, 55,56,57,58,59,60,61,62,63,64,65,66,67] presented moderate methodological quality, and four [68,69,70,71] presented high methodological quality.

Table 1 Methodological quality of the included studies
Full size table

Characteristics of the included studies

Of the 54 studies included, nine assessed range of motion [22,23,24,25,26, 28,29,30, 72], 16 assessed kinematics [35,36,37,38,39,40,41,42,43, 45,46,47,48,49,50, 68], four strength [31,32,33,34], seven electromyography [17, 18, 27, 51,52,53,54], and 18 assessed more than one outcome (six assessed kinematics and range of motion [44, 56, 59, 60, 63, 69], three range of motion and electromyography [58, 65, 66], two strength and range of motion [67, 71], one strength, range of motion, and electromyography [16], one electromyography and kinematics [55], two electromyography and kinematics [57, 61], one electromyography, kinematics, and strength [62] and two kinematics and strength[64, 70].

Results of the included studies

Nine inquiries regarding missing descriptive data were sent via email to the authors. Only one study author responded that the requested data were no longer available. Thus, no additional data provided by the authors were included in this systematic review. The objectives of the included studies, characteristics of the sample, outcomes assessed, and the main results are shown in Table 2.

Table 2 Results of the included studies
Full size table

Range of motion

Biomechanical assessments of range of motion were performed using goniometer [26, 58, 60, 65, 67, 71], inclinometer [22, 24, 25, 28,29,30, 66], and Thomas [23, 29, 66, 67, 71], Ober [71] and Straight leg raise test [58, 71]. The most investigated movement was total hip rotation[22, 25, 28, 30], followed by internal rotation [24, 26, 30, 71] and external hip rotation [22, 26, 71]. The results suggested a reduction in the total hip range of motion in patients with acute and chronic LBP [22, 25, 28, 30] and a reduction in internal rotation [24, 26, 30] and external hip rotation in patients with subacute and chronic LBP [22, 26]. Two studies identified a reduced range of motion between the dominant and non-dominant lower limbs [30] and between men and women with subacute and chronic LBP [26]. Only one study [63] showed that patients with acute LBP had a higher range of hip rotation compared to healthy individuals.

Strength

Biomechanical strength assessments were performed using the manual muscle test [32, 54], pressure meter [31, 33], force transducer [16] and hand-held dynamometer [34, 64, 67, 70, 71]. The main movement tested was hip abduction [16, 32,33,34, 64, 67, 70, 71]. The results showed that the hip abductor and extensor muscles were weaker in patients with acute, subacute, and chronic LBP compared to healthy individuals [31,32,33,34].

Kinematics

Biomechanical kinematic assessments were performed using 3-dimensional systems [35, 37, 39, 41, 42, 45, 46, 57, 59, 69], force platform [36, 37, 44, 55, 62, 70], cameras [38, 40, 43, 44, 64] and sensors [49, 50, 56, 60, 63, 68]. Hip movements were assessed during functional activities such as sit-to-stance-to-sit [36], sit-to-stand [37, 49], walking [46, 56, 59, 65], among others [35, 40, 44, 49, 57, 60, 64, 68, 69]. The results showed that during sit-to-stand, patients with chronic LBP had limited trunk flexion [37] and patients with subacute LBP had reduced task execution time [36, 49]. During walking, patients with chronic LBP showed a reduction in gait speed [46, 56] and step distance [65] compared to healthy individuals.

Electromyography

Electromyography was used to assess the gluteus maximus [17, 18, 27, 51, 53,54,55, 57, 65, 66], hamstrings [17, 18, 52, 53, 58, 65] and gluteus medius [16, 27, 54, 62] muscles. Muscle activation was performed in various activities, from exercise [27, 51] to functional activities such as sit-to-stand and stand-to-sit [52, 53] and standing [16, 55, 62]. The results were very different between the studies as the muscles were tested in different ways. Two studies tested the gluteus medius and maximus muscles during exercise [27, 51] and found no differences between patients with chronic LBP and healthy individuals. Two other studies [52, 53] tested the hamstring muscles in similar positions (semi-sitting position [53] and sit-to-stand and stand-to-sit [52]) and obtained different results. One study [53] evaluated the semi-sitting position during an isometric contraction and found no difference between patients with chronic LBP and healthy individuals, while other study [52] evaluated the sit-to-stand and stand-to-sit dynamically and observed decreased activation of the time to peak in the hamstring muscles of patients with chronic LBP compared to healthy individuals.

Discussion

The objective of this systematic review was to summarize the results of observational studies that performed biomechanical assessments in patients with non-specific LBP. The 54 studies included in the review used the outcomes range of motion, kinematics, strength, and electromyography for biomechanical assessment. The most common assessments were range of motion and kinematics. Patients with LBP, regardless of the duration of symptoms, showed a significant reduction in hip range of motion, especially total hip rotation [22, 25, 28, 30] and internal hip rotation [19, 24, 26, 30], even with the use of different assessment tools [19, 24,25,26, 28,29,30]. Range of motion can be assessed in different ways. The goniometer [26, 58, 60, 65, 67, 71] and inclinometer [22, 24, 25, 28,29,30, 66] are the most routine, despite having a measurement error between 3 and 5 degrees [73, 74]. Thus, there may be differences in range of motion, such as between dominant and non-dominant lower limbs [30] and between men and women with subacute and chronic LBP [26], that are not greater than the measurement error in studies assessing this outcome. However, the studies included in this systematic review found a significant reduction in total hip rotation in a variety of patients with LBP from participants of various sports [28, 30] to helicopter pilots[25] compared to healthy individuals. Patients with subacute and chronic LBP showed a significant reduction in internal hip rotation compared to healthy individuals [24, 26, 30], and one study showed an association between LBP and reduced internal hip rotation [19].

Kinematics has been extensively studied in patients with LBP, especially during functional activities such as sit-to-stand [36, 37, 49], walking [38, 56, 65] and lifting [35, 40]. Although the current kinematic assessment aims to assess common day-to-day movements [75], no studies have been found that performed kinematic assessment during everyday functional activities that generate pain in patients with LBP, such as putting on shoes or pants [76]. In general, patients with chronic LBP showed a reduction in the execution time of functional activities and range of motion, which indicates that they use different strategies than healthy individuals to perform the same functional activities [77]. The human movement system has the ability to adapt and use new strategies in the short and long term [75]. These strategies adopted by patients with LBP may be the result of motor adaptations to avoid painful movements during the execution of tasks [78, 79]. This was observed in a previous systematic review [80], that showed “moderate” strength of evidence for reduction of gait preference velocity and “high” strength of evidence for decrease in stride distance in patients with LBP compared to healthy individuals.

The gluteus maximus [17, 18, 27, 51, 53, 54] and hamstring muscles [17, 18, 52, 53] were the most evaluated in studies that used electromyography as a form of assessment. In addition to being the main extensor muscles of the hip [81], they are superficial, which facilitates electromyographic assessment [51]. During the electromyographic assessment, standardization is recommended such as using more than one channel per muscle group and normalizing the value obtained by the maximum voluntary contraction [82, 83]. However, in patients with LBP, the recommendation is to perform normalization by submaximal voluntary contraction, reducing the chance of interference from the pain intensity [17, 52]. The studies included in this systematic review [27, 51] did not report the patients' pain intensity, therefore it is not possible to know if there was interference during normalization. However, the gluteus maximus and hamstring muscles had greater electromyographic activation in patients with chronic LBP compared to healthy individuals during the prone hip extension task [17], which may indicate that patients with chronic LBP have difficulty maintaining a stable pelvic lumbar region [17, 84].

In this systematic review, 9 studies [16, 31,32,33,34, 64, 67, 70, 71] evaluated the strength of hip muscles in patients with LBP. Despite the good reliability between the hand-held dynamometer and the isokinetic dynamometer [85] and the difference in cost between the devices, the hand-held dynamometer is still inaccessible to many health professionals [86,87,88]. The cost of the manual dynamometer may be one of the reasons why the included studies assessed strength using a manual muscle test [32] and pressure meter [31, 33]. The results of this assessment corroborate a recent systematic review [15], in which patients with acute, subacute, and chronic LBP presented weakness of the abductor [32,33,34, 71] and extensor muscles [31, 34, 67, 71] of the hip compared to healthy individuals.

Although LBP affects millions of people worldwide [5, 6] and some studies seek to understand how the hip behaves in the presence of LBP [14,15,16,17,18,19], this is the first systematic review that summarizes the main findings of biomechanical hip assessments in patients with LBP, considering the type of assessment, the objective of each study, and its result. The results of this systematic review allow an overview of what is expected in the hip assessment of patients with LBP, directing clinicians to more accurate assessments and researchers to new studies that investigate the causes of LBP in a specific population or risk factors in an asymptomatic population. Future research may determine how much the biomechanical outcomes of the hip can be modified during the treatment of patients with LBP, as this question remains unanswered [89]. On the other hand, the heterogeneity of the biomechanical assessments and styles of reporting presented a challenge in this systematic review. Although the methodological quality of the included studies was moderate or high, the results were not always presented clearly.

Conclusion

The studies that evaluated the hip biomechanics of patients with LBP are of moderate and high methodological quality. Range of motion is lower in the total, internal, and external hip rotation movements of patients with LBP compared to healthy individuals. The strength of the hip abductor and extensor muscles is lower in patients with LBP compared to asymptomatic individuals. In the kinematic assessment, patients with LBP adopt strategies to reduce speed and change hip flexion movements compared to asymptomatic individuals during functional activities. Patients with LBP submitted to electromyographic assessment presented shorter activation time of the hip muscles and greater amplitude of electromyographic activity compared to healthy individuals. Therefore, greater attention should be given to hip assessment and management during the treatment of these patients.