Kinematic Gait Analysis After Primary Total Hip Replacement: A Systematic Review

Abstract

Total hip replacement (THR) is a surgical procedure indicated for patients affected by severe hip osteoarthritis. Although this technique has proved to be effective in relieving pain and restoring function, gait limitations may persist following this procedure. The objective of this systematic review was to evaluate gait kinematics after THR and compare the results with those of the pre-operative state and with healthy control individuals. PubMed/MEDLINE, Embase, Web of Science, CENTRAL and Scopus databases were searched until December 2019. Methodological quality and internal validity score of each study were assessed using the PEDro and the Newcastle–Ottawa scales. In all, ten studies met our inclusion criteria. Following THR, statistically significant improvements were seen in dynamic hip and knee range of motion of both the affected and the contralateral limb, single-limb support time symmetry, step length, stride length, walking speed and gait pattern. However, deficits were observed in all the previous parameters, as well as in hip adduction angle in comparison with healthy subjects. In conclusion, gait patterns improve after THR in comparison with the pre-operative state, although there are deficits relative to healthy individuals.

Electronic supplementary material

The online version of this article (doi:10.1007/s43465-020-00101-x) contains supplementary material, which is available to authorized users.

Introduction

Total hip replacement (THR) is a surgical procedure indicated for patients affected by severe hip osteoarthritis [1] and over 80% of subjects report satisfaction after this operation [2]. Even though THR has proved effective in relieving pain, restoring function and improving quality of life [3], functional limitations may persist after the surgery [4]. In this regard, patients report deficits in some activities such as walking and climbing stairs [5] and more than 25% are not able to go back to participating in sports they previously practiced [6]. Studies that examined gait in subjects with a THR have reported residual hip impairments related, among other matters, to a loss of terminal hip extension during late stance and reduced flexor and extensor moments of force [7]. In addition to these findings, several studies make reference to poor trunk control in the mediolateral direction [8]. This may be the outcome of a strategy adopted by patients, consisting of lateral bending toward the affected side to reduce the mechanical demand on weaker hip abductor muscles and facilitate control of balance in the frontal plane [9]. Such an action might induce overloading of the contralateral hip joint and lead to OA in that limb, or increase the risk of falls [10]. Additionally, as a result of pain and weakness in the hip operated upon, people often show a pattern of noticeable limping, or asymmetries in their gait [11]. When outcomes in walking gait after THR are compared with those of healthy subjects, lowered walking speeds and stride lengths have been identified, along with limitations on sagittal and coronal hip joint range of motion (ROM) [12]. However, pre-operative gait status, which is of relevance because of its association with postoperative functional status [13], was not considered in these studies. Furthermore, it remains unclear whether differences in surgical approach can have an impact on the amount of gait asymmetry after THR [14] or on recovery rates [15]. In this regard, three-dimensional optoelectronic movement (3DGA) measurement systems are the gold standard to analyse gait [16]. This technique consists of a series of retroreflective markers placed on the skin according to bony landmarks. Their spatial movement is detected by video cameras and that signal is translating to spreadsheet software, which forms the three-dimensional graphic [17, 18]. However, reviews assessing gait after primary THR [5, 12, 19] have not focused on this procedure as the principal instrument for analysing gait. Consequently, the present systematic review was undertaken with the purpose of synthesizing data from studies assessing gait kinematics after primary unilateral THR through 3DGA, and comparing it both with pre-operative status and healthy control subjects.

Methods

Study Protocol

A systematic review following the principles of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement was performed [20]. This systematic review was pre-registered in the International Prospective Register of Systematic Reviews (PROSPERO) [21].

Information Sources and Search Strategy

We concluded a literature search on December 2019 using PubMed/MEDLINE, Embase, Web of Science, CENTRAL and Scopus databases. Searches were restricted to English language, without limitations on dates of publication. Details of the search strategy for PubMed/MEDLINE are shown in Supplemental Appendix 1.

Study Selection

The first step was to eliminate any duplication of articles retrieved from the databases. The next was to have two reviewers independently analyze titles and abstracts, whereupon any papers which were not seen to meet the selection criteria were removed. Finally, full-text articles were reviewed to confirm they should be included in this review. The following inclusion criteria were applied: (1) include patients with primary unilateral hip prostheses following hip osteoarthritis; (2) involve both pre-operative and post-operative assessment; (3) incorporate a healthy control group; and (4) use 3DGA measurement systems for analyzing gait. In contrast, studies were discarded if they included patients with any previous prosthesis. Conferences, single-case studies, pilot studies and narrative and systematic reviews were also excluded.

Data Extraction and Quality Assessment

Various items of data were extracted for each study by two reviewers. These were: the author or authors, the year, participants (with details of their sex and age), the surgical approach used, pre- and post-operative assessment, gait analysis, outcome measures and principal results.

The methodological quality of the studies was assessed by two researchers using the PEDro scale [22] for the randomized controlled trials and the Newcastle–Ottawa Scale [23] for the case–control studies.

Data Synthesis

The search results showed that current literature contained heterogeneous information, precluding the performance of a meta-analysis. For this reason, we have applied a qualitative analysis.

Results

Study Selection

The search strategy retrieved 2257 records. After checks for duplication examination 925 studies remained. Of these, 893 were excluded on the basis of an assessment of their titles and abstracts. Thus, 32 publications were shortlisted and subjected to further evaluation, of which ten studies met the eligibility criteria and were included for qualitative analyses [14, 24–32]. Figure 1 Provides a diagram of the whole procedure for selecting studies.

Fig. 1.

Flow diagram of study selection process

Study Characteristics

In total, 237 subjects with THR (47.7% female) and 196 healthy control individuals (41.8% female) were covered by the investigations. The average age of the subjects with THR ranged between 47.3 [30] and 66.9 years [29], while that of controls varied from 42.1 [28] to 65.7 [29]. All of the studies used an optoelectronic camera video system to assess gait biomechanics and had both a pre-operative and a post-operative examination. Five investigations carried out a post-operative assessment after 6 months [24, 25, 27, 31, 32], whilst another six did so approximately 1 year after the intervention [24–26, 29–31]. Furthermore, postoperative examinations were performed at 3 weeks [25, 31], around 14 and 28 weeks, [28] 3 months, [14, 24, 25, 30, 31] and 6 months after THR. [30] Table 1 summarizes the main results of the selected studies.

Table 1.

Description of selected studies

Study	Participants (N; Age)	Surgical approach (N)	Gait analysis	Statistically significant results
				vs. preoperative state	vs. CON
Foucher et al. [25]	THR: 26; 59 ± 9 CON: 25; 54 ± 6	AN: 13 A-P: 13	6 walking trials: -4 at self-selected normal speed -2 at self-selected fast (CON and pre) or low (CON and post) speed	↑ 38% dynamic hip ROM at 12 months (p < 0.001)	↓ dynamic hip ROM at 12 months (p ≤ 0.045)
Foucher et al. [26]	THR: 28; 63.6 ± 7.1 CON: 25; 57.6 ± 7.7	L: 13 P: 15	6 walking trials: -2 at self-selected low speed -2 at self-selected normal speed -2 at self-selected fast speed	↑ dynamic hip ROM at 12 months (p < 0.012)	↓ dynamic hip ROM at 12 months (p < 0.04)
Foucher and Wimmer [31]	THR: 26; 59 ± 9 CON: 25; 54 ± 6	A:13 A-P: 13	6 walking trials: -4 at self-selected normal speed -2 at self-selected fast (CON and pre) or low (CON and post) speed	No differences in contralateral hip/knee gait kinematics	↓ dynamic knee (contralateral) ROM at 12 months (p = 0.015)
Horstmann et al. [32]	THR: 52; 58.0 ± 9.0 CON: 24; 54.0 ± 6.6	L: 52	ND	↑ dynamic hip ROM at 6 months (p < 0.001) ↑ dynamic hip (contralateral) ROM at 6 months (p = 0.005) ↑ dynamic knee (affected and contralateral) ROM at 6 months (p < 0.001)	↓ dynamic limb ROM at 6 months (p < 0.001) ↓ dynamic hip (contralateral) ROM at 6 months (p = 0.012) ↓ dynamic knee (contralateral) ROM at 6 months (p < 0.012) ↓ step length (p < 0.001)
Lavigne et al. [24]	THR: 24; 49.8 HR: 24; 49.6 CON: 14; 44.4	P: 24	5 walking trials at normal and fast speeds	ND	↑ walking speed at 12 months (p = 0.001) ↑ cadence at 12 months (p < 0.05)
Lugade et al. [14]	THR: 23; 56.9 ± 3.4–57.0 ± 7.3 CON: 10; 59.9 ± 5.3	A: 12 A-L: 11	5 walking trials at self-selected speed	↑ single limb support time symmetry in A THR at 6 weeks (p < 0.009) ↑ single limb support time symmetry in A THR (p < 0.001) and A-P THR (p = 0.017) at 4 months ↑step length in A THR (p = 0.017) at 4 months	↓ symmetry for step length in A-L THR at 6 weeks (p = 0.018) and 4 months (p = 0.012) ↑ pelvic drop at midstance of the affected limb in A THR (p = 0.002) and A-L THR (p = 0.032)
Naili et al. [29]	THR: 34; 66.9 ± 9.0 CON: 25; 65.7 ± 9.5	A-L: 34	5 walking trials at self-selected speed	↑ walking speed (p < 0.05) and normalized walking speed at 12 months (p < 0.001) ↑ stride length at 12 months (p < 0.001) ↑ gait pattern of operated and non-operated limbs at 12 months (measured by GDI)	↓ walking speed and normalized walking sped at 12 months (p < 0.05) ↓ stride length at 12 months (p < 0.001)
Nantel et al. [27]	THR: 10; 49 ± 7.5 HR: 10; 44.9 ± 8.5 CON: 10; 48.6 ± 6	P: 10	4 walking trials with normal cadence	No statistically significant results at 6 months	No statistically significant results at 6 months
Rösler and Perka [28]	THR: 26; 64.6 ± 7.7 CON: 20; 42.1 ± 13.5	L	Walking trials at “acceptable” speed	↓flexion and extension of the hip and ipsilateral knee (cranialization of the centre of rotation)	ND
Wesseling et al. [30]	THR: 14; 47.3 ± 11.8 CON: 18; 53.0 ± 5.0	A: 14	Walking trials at self-selected speed	HCF peak 1: ↑ hip adduction on at 6 (p = 0.036) and at 12 months (p = 0.006) ↓ ankle flexion at 6 months (p = 0.009) HCF peak 2: ↑ hip flexion at 3,6 and 12 months (p < 0.001) ↑ hip adduction at 12 months (p = 0.011) ↑ hip rotation at 3 (p = 0.011), 6 (p = 0.005) and 12 months (p = 0.002) ↑ hip rotation (contralateral) at 3 (p = 0.016), 6 (p = 0.038) and 12 months (p = 0.005) ↓ knee flexion at 3.6 and 12 months (p < 0.001) ↓ ankle flexion at 6 months (p = 0.027) and ↑ from 6 to 12 months (p = 0.004)	↓hip adduction angle

Age is given as mean ± standard deviation

↑ higher/improvement, ↓ lower, A anterior, A-L antero-lateral, A-P antero-posterior, CON control, GDI Gait Deviation Index, HR hip resurfacing, L lateral, N number of patients, ND no data, P posterior, ROM hip range of motion, THR total hip replacement

Study Quality

Over the ten studies included in this review, two were randomized controlled trials [24, 25] and the rest consisted of case–control designs [14, 26–32]. The randomized controlled trials gave a PEDro score of 8/10, while the Newcastle–Ottawa scores for the case–control studies ranged from 5/9 to 7/9. Tables 2, 3 present the methodological quality of the studies analyzed.

Table 2.

Levels of evidence and PEDro scores for randomized controlled studies

Study	PEDro score distribution
	1	2	3	4	5	6	7	8	9	10	11	Total PEDro score
Lavigne et al. [24]	●	●	●	●	●	●	─	●	─	●	●	8
Foucher et al. [25]	●	●	●	●	●	●	●	─	─	●	●	8

1. Eligibility criteria; 2. Random allocation; 3. Concealed allocation; 4. Baseline comparability; 5. Blind subjects; 6. Blind therapists; 7. Blind assessors; 8. Adequate follow-up; 9.Intention-to-treat analysis; 10. Between-group comparisons; 11. Point estimates and variability. A “●” indicates a “yes” score, and a dash indicates a “no” score

Table 3.

Levels of evidence and Newcastle–Ottawa Scale scores for case–control studies

Study	Newcastle–Ottawa Scale
	Selection				Comparability	Exposure			Total score
	1	2	3	4	1	1	2	3
Foucher et al. [26]	★	★		★	★★	★	★		★★★★★★★ (7)
Lugade et al. [14]	★	★	★	★	★★	★			★★★★★★★ (7)
Nantel et al. [27]	★	★	★	★	★★	★			★★★★★★★ (7)
Wesseling et al. [30]	★			★	★★	★		★	★★★★★★ (6)
Foucher and Wimmer [31]	★			★	★★	★			★★★★★ (5)
Horstmann et al. [32]	★			★	★★	★			★★★★★ (5)
Naili et al. [29]	★	★		★	★	★			★★★★★ (5)
Rösler and Perka [28]	★		★		★	★	★		★★★★★ (5)

Selection (maximum ★★★★): 1) Is the case definition adequate?, 2) Representativeness of the cases, 3) Selection of controls, 4) Definition of controls; Comparability (maximum ★★): 1) Comparability of cases and controls on the basis of the design or analysis; Exposure (maximum ★★★): 1) Ascertainment of exposure; 2) Cases and controls: same ascertainment method, 3) Cases and controls: same non-response date

Discussion

Summary of Main Results

This systematic review analysed gait patterns in persons following primary THR. The studies examined showed improvements in dynamic hip and knee ROM of both the affected and the contralateral limb, single-limb support time symmetry, step and stride length, walking speed and gait pattern after THR. Nevertheless, when compared with healthy controls, deficits were still shown in dynamic lower limb ROM, hip adduction angle, step and stride length, walking speed and gait symmetry.

Implications for Clinical Practice

Regarding the differences relative to healthy subjects, walking speeds and spatio-temporal parameters were in many cases described as lower when compared with healthy individuals [29, 33–35], although Lavigne et al. [24] showed a higher walking speed and cadence in THR subjects at 6 and 12 months follow-up. This discrepancy could be explained by the familiarity with the 10-m walking test in THR patients, who performed it pre, 3, 6, and 12 months, since healthy controls were evaluated only once. In addition, Nantel et al. [27] did not find such differences and concluded that THR patients had reached standard values within 6 months after surgery, even though spatio-temporal parameters were reported to be lower in relation to those of healthy subjects 10-years post-intervention [36]. Apart from that, ROM is a crucial outcome in the recovery process after surgery. A significant lower ROM in comparison with healthy subjects has been observed following THR [25, 26, 31, 32, 34, 35, 37], which could be related to a load asymmetry between the operated and the healthy limb [37] and goes along with the characteristic gait deviation [14, 29]. Furthermore, it has been proved that Gait Deviation Index (GDI) [38] is positively associated with hip abduction and hip flexion strength [39], which is reduced after THR surgery [24, 27, 40–43]. In this regard, the reduction in the volume of gluteus minimus showed by most patients [29] has been identified as a factor involved in the gait deviation, and contributes to higher hip joint loads up to 3 months after the intervention [44]. Likewise, the atrophy of this muscle has been pointed as a predictor of the weakness of the gluteus medius [45], which is the dominant hip abductor [46].

Despite it is not the objective of this review, in view of all the above and because of the strong impact that muscle state has on gait function, exercise programmes to enhance strength should be included as early as possible for THR patients to improve postural stability and gait efficiency [47, 48]. Nevertheless, authors speculate that post-operative gait might be influenced by pre-intervention gait status [26]. Some studies have confirmed improvements in post-operative walking speed and stride length after a peri-operative exercise programme, in comparison with conventional care regimes [49, 50]. Furthermore, despite the fact that gait symmetry is achieved postoperatively, the lower ROM values presented for the contralateral limb following THR must be taken into account. Thus, the evaluation of the joints adjacent to those affected and joints of the contralateral limb should be considered by physiotherapists in the design of rehabilitation programs.

In regard to recovery times, positive results were reported practically immediately after the intervention. Thus, improvements in single limb support time symmetry and hip flexion and rotation appeared at 6 weeks [14] and 3 months [30] evaluations. Nevertheless, greater improvements were concluded from 6 months [32], with the best results appearing approximately one year after the surgery [25, 26, 29, 30].

Strengths and Limitations

As far as we know, this systematic review is the only one that focuses on gait assessment through 3DGA. Although this fact could be a limitation as a result of the restriction of other gait assessments, we consider it strength, since 3DGA is the most widespread technique when it comes to motion analysis [16], allowing gait evaluation in dynamic activities such as walking and running with many successive gait cycles [51]. Therefore, this technique provides objective information about joint motions or kinematics and time and distance variables or spatiotemporal data [52], and it has demonstrated excellent test–retest reliability, as well as being proved a very valuable instrument in movement disorder analysis and follow-up [53]. Besides that, other strengths of this review reside in requiring as inclusion criteria, an evaluation of the pre-operative state and the existence of a healthy control group, offering a specific perspective of this topic. Thus, the existence of healthy control subjects allows establishing the differences between a standard gait and that following a total hip replacement. In addition, we have only included studies about primary THR under the belief that they may present a gait pattern closer to that of healthy subjects in contrast to individuals with a bilateral hip prosthesis. However, a recent study [54] has concluded that bilateral patients have a kinematic profile closer to normative patterns when compared with unilateral cases, although they reported higher pain levels.

Nevertheless, this review has several limitations. First, the studies were heterogeneous in THR approaches, which may be determinant in the recovery. Second, it is possible that our inclusion criteria may discard some studies that have potential relevance in this area. Third, the focus of this systematic review allows only obtaining kinematic data, and no kinetic parameters were evaluated. Fourth, none of the studies have mentioned any surgery-related complication, as the sciatic and femoral nerves injuries, which are a frequent side-effect of this type of operations [55]. Fifth, none of the studies have assessed gait function in the context of functional daily life activities, such as stair climbing, which biomechanics are frequently impaired in individuals following a THR [56].

Conclusion

This systematic review concludes improvements in dynamic hip and knee range of motion of both affected and contralateral limbs, single-limb support time symmetry, step and stride length, walking speed and gait patterns. However, there are deficits in comparison with healthy individuals.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author Contributions

LM and HO contributed to the study conception and design. LM, MAR and HO extracted data from the studies and evaluated the inclusion and exclusion criteria. LM and HO assessed the methodological quality of the studies. All the authors contributed to the interpretation of the data and drafted the manuscript.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Compliance with Ethical Standards

Conflict of interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Ethical standard statement

This manuscript has been approved by all the authors and represents honest work. This manuscript has not been published and is not under consideration for publication elsewhere.

Informed consent

For this type of study formal consent is not required.