Which functional tasks present the largest deficits for patients with total hip arthroplasty before and six months after surgery? A study of the timed up-and-go test phases

Xavier Gasparutto; Mathieu Gueugnon; Davy Laroche; Pierre Martz; Didier Hannouche; Stéphane Armand

doi:10.1371/journal.pone.0255037

. 2021 Sep 10;16(9):e0255037. doi: 10.1371/journal.pone.0255037

Which functional tasks present the largest deficits for patients with total hip arthroplasty before and six months after surgery? A study of the timed up-and-go test phases

Xavier Gasparutto ^1,^*, Mathieu Gueugnon ², Davy Laroche ², Pierre Martz ^3,⁴, Didier Hannouche ⁵, Stéphane Armand ¹

Editor: Peter Andreas Federolf⁶

PMCID: PMC8432811 PMID: 34506498

Abstract

Six to eight months after total hip arthroplasty, patients only attain 80% of the functional level of control groups. Understanding which functional tasks are most affected could help reduce this deficit by guiding rehabilitation towards them. The timed up-and-go test bundles multiple tasks together in one test and is a good indicator of a patient’s overall level of function. Previously, biomechanical analysis of its phases was used to identify specific functional deficits in pathological populations. To the best of our knowledge, this analysis has never been performed in patients who have undergone total hip arthroplasty. Seventy-one total hip arthroplasty patients performed an instrumented timed up-and-go test in a gait laboratory before and six months after surgery; fifty-two controls performed it only once. Biomechanical features were selected to analyse the test’s four phases (sit-to-stand, walking, turning, turn-to-sit) and mean differences between groups were evaluated for each phase. On average, six months after surgery, patients’ overall test time rose to 80% of the mean of the control group. The walking phase was revealed as the main deficiency before and after surgery (-41 ± 47% and -22 ± 32% slower, respectively). High standard deviations indicated that variability between patients was high. On average, patients showed improved results in every phase of the timed up-and-go test six months after surgery, but residual deficits in function differed between those phases. This simple test could be appropriate for quantifying patient-specific deficits in function and hence guiding and monitoring post-operative rehabilitation in clinical settings.

Introduction

Total hip arthroplasty (THA) is a common, cost-effective and usually highly successful surgical procedure to alleviate pain and improve motor function [1] (‘motor function’ will be referred to as ‘function’ throughout this paper to improve readability). Although THA improves function for patients with end-stage hip osteoarthritis (OA), patients have not fully recovered function six months and one year after surgery when compared to healthy control groups [2, 3]. On average, patient function increases from 70% of the level of healthy control groups before surgery to 80% six to eight months afterwards [2]. Understanding which domains of function (e.g. strength, balance or mobility) are most improved by THA and which retain the most significant post-surgical deficits could help to map this residual loss of function and guide patient-specific rehabilitation.

Several types of evaluation of function exist. Each covers different domains of functioning as defined by the World Health Organization’s International Classification of Functioning and Disability (WHO ICF) [4]. The most common tools are: (i) patient-reported outcomes (PROMs) [5] (e.g. the Hip disability and Osteoarthritis Outcome Score [6]) that evaluate multiple domains of function through self-administered questionnaires (e.g. personal factors, activities, participation); (ii) clinical evaluations that evaluate body function (e.g. hip range of motion); and, (iii) medical imaging that evaluates body structure (e.g. femoral offset). Clinical tests like the 6-minute walk test, the 30-second chair stand test or the Timed Up-and-Go (TUG) test—often described as performance-based tests [7]—are used to evaluate patients’ capacity as defined by the WHO ICF, i.e. what a person can do in a standardized evaluation setting [4].

The TUG test is a robust indicator of a patient’s overall level of motor function [8] as it groups together basic mobility skills needed for activities of daily living in a simple, practical test [9]. Participants start sitting in an armchair, stand up, walk three meters, turn back and return to sit in the armchair [9]. The test is recommended by the Osteoarthritis Research Society International [10] for assessing physical function in people diagnosed with hip OA, and it is a noted test of ambulatory activity covering the specific ICF codes of “changing basic body position” (d410), “walking” (d450) and “moving around” (d455), and of strength, agility and dynamic balance [10]. It is widely used to assess mobility in rehabilitation [11].

In the context of THA, pre-surgery TUG test results have been identified as predictive of post-surgery hospital length of stay [12], predictive of ambulatory status six months after a primary THA [13] and associated with a risk of post-surgery deep vein thrombosis [14]. However, as underlined by previous authors [15–17], using the TUG’s total time as the sole outcome provides limited information. Indeed, the analysis of the TUG’s phase, known as instrumented TUG (iTUG) [18], could provide a deeper understanding of a patient’s function before and after treatment. For instance, the iTUG test was shown capable of discriminating between patients with early-to-moderate Parkinson’s disease and healthy controls, whereas the total TUG time could not [18, 19]. The iTUG test also showed that slower total TUG times in a group of obese women compared to a control group were explained by the walking and turning phases, not by the sit-to-stand and stand-to-sit phases [17]. To the best of our knowledge, no studies of the TUG test’s different phases have been performed previously on patients who have end-stage OA or have undergone a THA. A study to identify the most difficult tasks facing patients before and after surgery could help to target patient-specific rehabilitation.

The present study’s objectives were: 1) to assess in which phases of the TUG test patients presented with significant functional deficits compared to a control group, before and after surgery; 2) to assess how these phases changed after surgery; and 3) to compare changes and deficits in total TUG time with changes and deficits in the functional outcomes of the different phases. To this end, we performed a biomechanical analysis of the TUG test for a group of patients before and six months after surgery, and their outcomes were compared with those of a control group.

Material & methods

Participants

This study included 71 THA patients and 52 controls. Participants’ characteristics are reported in Table 1. Surgeries were performed by two surgeons using three different approaches: a mini-invasive Rottinger approach (n = 38) [20], a mini-posterior approach (n = 29) [21] or a lateral approach (n = 4), with a dual-mobility cup implanted in all cases. The local ethics committee approved the study (Human Welfare Protection Committee EST I, in Dijon), and it was recorded as a clinical trial (NCT02042586), with all participants giving their written informed consent.

Table 1. Participants’ characteristics.

	Patients	Control Group	p-value
General Characteristics
Sex (M/F)	29/42	18/34	0.607
Age (years)	68.5 (8.9)	65.3 (8.6)	0.054
Height (cm)	162.8 (8.0)	163.4 (8.3)	0.703
Weight (kg)	75.6 (17.7)	66.9 (13.0)	0.001
OA Side (L/R)	(32/39)	-
Kellgren and Lawrence classification
II	14	-
III	31	-
IV	21	-
NA	5	-
Surgical Approach
Rottinger	38	-
Mini-posterior	29	-
Lateral	4	-

Open in a new tab

Sex distribution was tested using the χ² test, and differences in age, height and weight used Student’s t-tests.

NA = not available.

All the patients included in this study were diagnosed with primary hip OA, according to the American College of Rheumatology (ACR) criteria [22], and had stage II, III or IV hip OA according to the Kellgren and Lawrence classification. Mean patient age was 68.5 years old (SD = 8.9 years, range = 46–85 years), with 42 women and 29 men. All had undergone conservative treatment for at least three months before they were considered for a hip replacement, and all were able to understand simple instructions. Exclusion criteria included being younger than 25 years old or older than 85, previous hip surgery, bilateral hip OA, concomitant knee OA, pregnancy or breast-feeding, inflammatory disorders, rapidly progressive hip OA, hip dysplasia, neurological disease, motor neuron disease and all disorders that might interfere with or be worsened by a gait analysis. A control group of participants in the same age range was collected. Their exclusion criteria included OA in the lower limbs, neurological disease, motor neuron disease or disorders that might interfere with or be worsened by a gait analysis.

Measurement protocol

Participants were equipped with 35 reflective skin markers attached over their whole body according to the Conventional Gait Model [23]. Markers’ trajectories were measured at 100 Hz using an eight-camera optoelectronic system (MXT 40, Vicon, Oxford, UK), filtered at 6 Hz using a fourth-order Butterworth design and then occlusions were corrected using marker intercorrelations [24].

Participants were asked to perform the TUG test. They sat in an armchair with a seat at 47 cm off the ground, stood up, walked to a line on the ground three meters away, turned around and came back to sit in the chair [9] (Fig 1). The whole task was made at a self-selected speed, and participants were allowed to use the armrests. Instructions were as follows: “At the start signal, stand up from the chair without using the armrests, if possible, then walk towards the line in front of you, turn back at the line and come back to the chair and sit down, without using the armrests, if possible.” Participants performed ten TUG tests, and the fastest of the first three tests was retained for analysis. When sitting, the participant’s back could not touch the armchair’s backrest because of the motion capture equipment. Patients performed the test before surgery (M0) and six months after surgery (M6), whereas the control group performed the test only once.

Analysis

The TUG test was divided into four phases (Fig 1): sit-to-stand, walking (back and forth), turn and turn-to-sit [15]. The second turn and stand-to-sit phases were grouped together as a turn-to-sit phase due to the difficulty in discriminating between the end of the turn and the start of the sitting phase [15]. The beginning and end of the TUG test phases were identified using a custom algorithm based on an existing algorithm [25] (S1 File). Work by Beyea et al. [25] was adapted to identify the start and end of turns and the end of the TUG test. Indeed, when checked visually, Beyea’s algorithm identified timepoints outside of the phases for those events with the study population. The visual checks of the algorithm’s results were assessed using Mokka software [26] and specific plots (S1 and S2 Figs).

The time required to complete each phase and an appropriate set of each phase’s biomechanical features were selected for analysis. Feature were selected based on the literature [18, 19, 27–35], and grouped into two categories: 1) quality of movement (joint angles and distances) [36] and 2) speed (linear and angular velocities). We selected a total of 38 features covering the four phases (Table 2) in addition to the total TUG test time and the time for each phase. The reasons for choosing these features and their definitions are detailed in S2 File.

Table 2. Biomechanical features for each phase.

	Quality	Speed
Sit-to-Stand	Peak obliquity thorax	Peak vertical velocity thorax
	Range obliquity thorax	Peak vertical velocity pelvis
	Peak flexion thorax	Peak extension velocity of pathological hip
	Width base of support	Peak extension velocity of contralateral hip
	Length base of support
Walking	RMS obliquity thorax	Peak forward velocity thorax
	Range obliquity thorax	Mean forward velocity thorax
	Lateral RMS C7	Peak forward velocity pelvis
	Lateral range C7	Mean forward velocity pelvis
	Range of flex/ext pathological hip
	Range of flex/ext contralateral hip
Turn	Number of steps	Peak angular velocity thorax
	Side of turn	Mean angular velocity thorax
		Peak angular velocity pelvis
		Mean angular velocity pelvis
Turn-to-sit	Peak obliquity thorax	Peak vertical velocity thorax
	Range obliquity thorax	Peak vertical velocity pelvis
	Peak flexion thorax	Peak extension velocity of pathological hip
	Distance chair to start of turn	Peak extension velocity of contralateral hip
	Number of steps	Peak angular velocity thorax
		Mean angular velocity thorax
		Peak angular velocity pelvis
		Mean angular velocity pelvis

Open in a new tab

RMS = root mean square; flex/ext = flexion/extension.

Statistics

The difference in sex distribution was tested using a χ² test. The Kolmogorov–Smirnoff [37] test was used to assess the normality of the age, weight and height distributions among patients and controls. Differences between groups were then tested using unpaired Student’s t-tests.

A principal component analysis (PCA) was performed so that only the most relevant biomechanical features were retained for analysis (excluding phase completion times). To account for scale differences, features were centred and reduced by subtracting the mean and dividing by the standard deviation (SD). All three test groups (patients at M0, patients at M6 and the controls) were included in the analysis. The number of principal components was chosen using Cattell’s scree test [38]. One feature was selected per principal component and per phase based on the maximal cos2 value and if this value was above 0.20.

The statistical differences between groups for the selected features were assessed using a paired-sample Student’s t-test when comparing the patient groups and using an unpaired Student’s t-test when comparing patient and control groups. The Holm method was used to assess the level of statistical significance (α = 0.05) [39], and Cohen’s d was used to assess the effect size. To evaluate the average difference in patient function from control group function, the values of the features at M0 and M6 were reported as percentages of the control group mean. Lower function among patients than among the control group was described as a deficit. Depending on the feature, lower function could be a higher value (e.g. phase duration) or a lower value (e.g. peak vertical velocity of the thorax for the sit-to-stand phase). Changes between M0 and M6 were expressed as percentages of a feature’s value at M0 to express post-surgery increases or decreases in function. Furthermore, the percentage of patients with a similar level of function to the control group (i.e. within 1 SD of the control group’s mean) was also evaluated, as was the percentage of patients whose function increased between M0 and M6. Pearson correlations were calculated between the total TUG time and selected features for percentage differences to the control group and the percentage of change between M0 and M6. The level of significance was adjusted using the Holm method (α = 0.05) [39]. Only significant correlations were reported.

Finally, the percentages of each feature selected using PCA were averaged for each phase to give one overall figure per phase. Paired Student’s t-tests were performed to assess whether the average differences compared to the control group, or between M0 and M6, differed significantly between the TUG test’s different phases. To account for multiple testing, the Holm’s method was used to adjust the p-values (α = 0.05) [39].

Results

The patient and control groups did not differ significantly in sex, age or height but differed significantly in weight (Table 1).

Principal component analysis

The complete results of the PCA are presented in S1 Table. Cattell’s scree test selected three principal components, representing 56.4% of the total variance. The first principal component (39.1% of variance) was mainly associated with speed parameters, and the second (10.0% of variance) and third (7.3% of variance) with quality parameters.

The biomechanical features selected for each phase are presented in Table 3, and the average scores for those features across the three groups are reported in Fig 2 and Table 4. Each patient’s values are compared to the control group’s level in S3 Fig.

Table 3. Features selected using principal component analysis.

	Category	Feature	PC	Cos2
Sit-to-Stand	Quality	Peak flexion thorax	2	0.45
		Peak obliquity thorax obliquity	3	0.28
	Speed	Peak vertical velocity thorax	1	0.61
Walking	Quality	Range of flex/ext pathological hip	1	0.50
		Range obliquity thorax	3	0.30
	Speed	Peak forward velocity pelvis	1	0.86
Turning	Quality	Step number	1	0.21
	Speed	Mean angular velocity pelvis	1	0.66
Turn-to-Sit	Quality	Range obliquity thorax	2	0.66
		Distance chair to start of turn	1	0.23
	Speed	Peak angular velocity thorax	1	0.76

Open in a new tab

PC is the principal component associated with the feature; cos2 (є [0,1]) shows the importance of the feature considered on the associated principal component.

Fig 2 — Feature values representing higher levels of function are on the outer edge, and lower levels of function are on the inner edge. S2S = Sit-to-Stand; T2S = turn-to-sit.

Table 4. Mean values (standard deviation) for patients before surgery (M0), patients six months after surgery (M6) and the control group.

Feature Categories	Features	Patients M0			Patients M6			Control Group			Group Comparison	Cohen’s d
Feature Categories	Features	Mean	SD	95%CI	Mean	SD	95%CI	Mean	SD	95%CI	Group Comparison	a	b	c
Total
Speed	Time (s)	14.9	4.1	13.9 to 15.8	12.9	2.8	12.2 to 13.5	10.7	2.1	10.2 to 11.3	a, b, c	1.2	0.7	0.8
Sit-to- Stand
Quality	Peak thorax flexion (deg)	39.3	10.0	37.0 to 41.6	38.1	8.9	36.0 to 40.1	38.6	10.4	35.7 to 41.4	-	-	-	-
Quality	Peak thorax obliquity (deg)	4.5	2.4	4.0 to 5.1	3.9	2.1	3.4 to 4.3	3.7	1.8	3.1 to 4.1	-	-	-	-
Speed	Peak vertical velocity thorax (m/s)	0.56	0.19	0.52 to 0.60	0.66	0.17	0.62 to 0.70	0.77	0.17	0.72 to 0.82	a, b, c	1.1	0.6	0.6
Speed	Duration (s)	1.3	0.4	1.2 to 1.4	1.1	0.3	1.0 to 1.2	1.0	0.2	0.9 to 1.0	a, b	0.9	0.4	-
Walk
Quality	Range of flex/ext pathological hip (deg)	28.3	7.8	26.5 to 30.2	35.0	7.1	33.4 to 36.7	43.6	5.1	42.2 to 45.0	a, b, c	2.2	0.9	1.3
Quality	Range thorax obliquity (deg)	13.6	5.9	12.2 to 15.0	11.5	3.9	10.6 to 12.4	8.7	3.3	7.8 to 9.7	a, b, c	1.0	0.4	0.7
Speed	Peak forward velocity pelvis (m/s)	0.91	0.20	0.86 to 0.96	1.02	0.21	0.97 to 1.07	1.19	0.20	1.14 to 1.24	a, b, c	1.4	0.6	0.8
Speed	Duration (s)	7.3	2.7	6.7 to 7.9	6.1	1.8	5.6 to 6.5	4.9	1.3	4.5 to 5.2	a, b, c	1.1	0.6	0.7
Turn
Quality	Number of steps	4.1	0.9	3.9 to 4.3	3.9	0.9	3.6 to 4.1	3.4	0.9	3.2 to 3.7	a	0.8	-	-
Speed	Mean angular velocity pelvis (deg/s)	74.8	19.2	70.3 to 79.3	83.1	21.1	78.2 to 88.1	96.3	20.3	90.1 to 101.8	a, b, c	1.1	0.4	0.6
Speed	Duration (s)	2.4	0.5	2.3 to 2.5	2.2	0.5	2.1 to 2.3	1.9	0.4	1.8 to 2.0	a, b, c	1.1	0.4	0.7
Turn-to-Sit
Quality	Range thorax obliquity (deg)	22.3	9.0	20.3 to 24.5	20.3	7.8	18.4 to 22.1	17.8	6.8	16.0 to 19.7	a	0.6	-	-
	Distance chair to start turn (m)	1.03	0.20	0.98 to 1.08	1.06	0.17	1.02 to 1.10	1.17	0.16	1.13 to 1.21	a, c	0.8	-	0.7
Speed	Peak angular velocity thorax (deg/s)	148.6	37.6	139.9 to 157.4	164.9	41.9	155.2 to 174.7	200.7	37.2	190.6 to 210.8	a, b, c	1.4	0.6	0.9
	Duration (s)	3.9	1.1	3.7 to 4.2	3.5	0.8	3.3 to 3.7	3.0	0.6	2.8 to 3.2	a, b, c	1.0	0.5	0.6

Open in a new tab

Significant differences between groups, as per the Holm method, are reported between patients at M0 and the control group as “a”, between patients at M0 and patients at M6 as “b”, and between patients at M6 and the control group as “c”. Cohen’s d was reported for significant differences.

Patients’ functional deficits compared to the control group

Deficits before surgery

The test phase with the largest mean deficit before surgery was the walking phase, at -41% (Table 5). Compared with the other phases, the walking phase deficit was statistically larger than the sit-to-stand, turning and turn-to-sit phases, as per the Holm method. The other phases did not differ significantly from each other. Only 8% of patients had a range of hip flexion–extension during walking similar to the control group’s level, whereas 31% to 69% of patients’ other features were at the control level (Table 6).

Table 5. Mean difference ± SD (95% confidence interval) between groups by phase expressed in percentages.

Tasks	Patients M0 vs CG			Patients M0 vs M6			Patients M6 vs CG
Tasks	Mean	SD	95%CI	Mean	SD	95%CI	Mean	SD	95%CI
Sit-to-Stand (%)	-21	± 46	(-31 to -10)	8	± 48	(-4 to 19)	-9	± 37	(-17 to 0)
Walk (%)	-41	± 47	(-52 to -30)	16	± 35	(8 to 24)	-22	± 32	(-30 to -15)
Turn (%)	-24	± 25	(-29 to -18)	8	± 26	(2 to 14)	-14	± 24	(-20 to -8)
Turn-to-Sit (%)	-27	± 38	(-36 to -19)	8	± 26	(2 to 14)	-16	± 32	(-23 to -8)

Open in a new tab

Table 6. Mean, SD and 95% confidence interval of the significant differences between groups, i.e. differences between patient and control groups and between patients before and six months after surgery.

Feature Categories	Features	Patients M0 vs CG					Patients M6 vs M0					Patients M6 vs CG
Feature Categories	Features	Mean	SD	95%CI	% > CG level	R total time	Mean	SD	95%CI	%Δ >0	R total time	Mean	SD	95%CI	% > CG level	R total time
Total
Speed	Time (%)	-39	38	-48 to -30	34	-	11	17	7 to 15	72	-	-20	26	-26 to -14	56	-
Sit to Stand
Quality	Peak thorax flexion (%)	2	26	-4 to 8	87	-	0	24	-6 to 6	38	-	-1	23	-7 to 4	86	-
Quality	Peak thorax obliquity (%)	-24	65	-39 to -9	75	-	-5	74	-22 to 12	59	-	-6	57	-19 to 8	83	-
Speed	Peak vertical velocity thorax (%)	-27	25	-33 to -21	41	0.59	26	37	18 to 35	73	0.63	-14	22	-19 to -9	61	0.65
Speed	Time (%)	-33	46	-44 to -22	46	0.71	8	33	1 to 16	70	0.57	-13	33	-21 to -6	68	0.62
Walk
Quality	Range of flex/ext pathological hip (%)	-35	18	-39 to -31	8	0.37	31	41	22 to 41	82	-	-20	16	-23 to -16	27	0.40
Quality	Range thorax obliquity (%)	-56	67	-71 to -40	49	-	5	42	-4 to 15	69	-	-32	45	-42 to -21	61	-
Speed	Peak forward velocity pelvis (%)	-24	17	-27 to -20	31	0.87	15	22	10 to 20	79	0.77	-14	18	-18 to -10	51	0.78
Speed	Time (%)	-49	55	-62 to -36	39	0.95	11	26	5 to 17	73	0.83	-25	38	-33 to -16	62	0.92
Turn
Quality	Number of steps (%)	-20	26	-26 to -14	69	-	4	28	-3 to 10	42	-	-12	27	-18 to -5	77	0.45
Speed	Mean angular velocity pelvis (%)	-22	20	-27 to -18	41	0.61	14	28	8 to 21	68	0.52	-14	22	-19 to -9	61	0.64
Speed	Time (%)	-28	28	-35 to -21	41	0.68	6	23	1 to 11	66	0.49	-17	26	-23 to -11	58	0.68
Turn to Sit
Quality	Range thorax obliquity (%)	-25	51	-37 to -14	69	-	3	31	-4 to 11	52	-	-14	44	-24 to -3	76	-
	Distance chair to start turn (%)	-12	17	-16 to -8	51	0.49	6	22	1 to 11	61	-	-9	15	-13 to -6	63	0.45
Speed	Peak angular velocity thorax (%)	-26	19	-30 to -22	31	0.73	13	23	8 to 18	72	0.67	-18	21	-23 to -13	46	0.78
	Time (%)	-31	37	-39 to -22	44	0.76	8	21	3 to 13	68	0.71	-16	27	-22 to -9	65	0.76

Open in a new tab

“% > CG” represents the number of patients with function similar to the control group’s level (within 1 SD of the mean) for that feature, and “% Δ > 0” represents the number of patients with an improvement in that feature’s function at M6. “R total time” represents the feature’s coefficient of correlation with total test time. Features with no significant differences between the groups considered are reported with light grey background.

Residual deficits six months after surgery

The largest mean residual deficit at M6 was for the walking phase (-22%, 95%CI: -30% to -15%, Table 5). Using the Holm method, this deficit was significantly greater than the sit-to-stand and turning phase deficits, but not greater than the turn-to-sit phase deficit. Only 27% of patients had a range of hip flexion–extension during walking similar to controls, whereas, for other features with significant differences, 46% to 65% of patients had a similar level of function to controls (Table 6).

Change in function six months after surgery

The walking phase showed the greatest mean improvement in function (16%, 95%CI: 8% to 24%, Table 5), although this was not statistically significant. The sit-to-stand, turning and turn-to-sit phases showed similar increases (8%, Table 5).

Total TUG times

Mean total TUG times were 14.9 ± 4.1 s and 12.9 ± 2.8 s for patients at M0 and M6, respectively, and 10.7 ± 2.1 s for the control group (Table 4). On average, patients were -39% (95%CI: -48% to -30%) slower than the control group before surgery, and they had improved their total time by 11% (95%CI: 7% to 15%) six months after surgery. However, they were still -20% (95%CI: -26% to -14%) slower than the control group (Table 6). These differences were all statistically significant. The majority of patients (72%) improved their total time between M0 and M6. At M0, 34% of patients had a functional level similar to the control group, rising to 56% at M6 (Table 6).

Associations between deficits and changes in features and deficits and changes in total TUG times (Table 6)

The deficits and changes in total TUG time were mainly associated with speed parameters, especially with the deficits and changes in the duration of the walking phase and with the peak forward velocity of the pelvis (Table 6). The deficits in the range of flexion–extension of the patient’s pathological hip during walking at M0 and M6, the deficit in the number of steps taken during the turning phase at M6, and the deficits in the distance from the chair to the start of the turn at M0 and M6 were the only quality parameters associated with the difference in total time that displayed moderate correlations (Table 6). The change in total TUG time between M0 and M6 was only associated with speed parameters. The highest correlations were with the change in the time of the walking phase and the change in peak walking speed.

Discussion

The present study assessed the change in function among a group of patients with THA, before and six months after surgery, and the functional deficits in comparison to a control group by the means of an instrumented TUG test. The test was divided into four phases [15], representing different activities of daily living (sit-to-stand, walking, turning and turn-to-sit); multiple biomechanical features quantifying the motor functions were selected using PCA and, the features for each phase were compared between groups. Finally, mean pre- to post-surgery changes and deficits in features, phase times and phase functional outcomes—between patients at those times and compared with the control group—were assessed to evaluate links between them and deficits and changes in total TUG time.

Patients’ functional deficits compared to the control group

Results suggested that the walking task represented the main limitation to patients with THA both before and six months after their surgery. The changes in walking revealed in this study were consistent with a recent literature review showing that six months post-surgery, gait speed and the pathological hip’s range of flexion–extension increased, although they remained lower than control group levels [3]. Our study’s results showed that the hip range of flexion–extension during walking seemed to be an appropriate feature of motor function with which to study the effects of THA. Indeed, this feature presented the lowest percentage of patients with a level of function similar to that of the control group, both before and after surgery (8% and 27%, respectively), as well as the greatest number of patients with a positive change in function after surgery (82%). This suggests that, despite their deficits six months after surgery, THA has a positive effect on patients’ sagittal hip kinematics during walking.

Changes in function six months after surgery

There were no significant differences in the changes between the phases, although the walking phase showed the largest mean increase in function. Thus, THA seems to improve patients’ motor functions homogeneously across the different phases of the TUG test.

Changes and deficits in total TUG time compared to changes and deficits in functional outcomes

The changes and deficits in total TUG time were mainly strongly to moderately correlated with the changes and deficits in the test’s speed features, but also with a few features representing the quality of movements (pathological hip ranges of motion at M0 and M6, number of steps during the turn at M6, and distances from the chair to the start of turn at M0 and M6). More specifically, the changes and deficits in total test time were strongly correlated with the changes and deficits in the duration of the walking phase, the TUG test’s longest phase. Total TUG test time could be a good indicator of speed parameters, but it does not seem to reflect comparably the quality of movements, such as the presence of compensatory movements (e.g. high range of thorax obliquity). Moreover, Caronni et al. [15] showed that an improvement in total TUG time was not associated with improvements in all the test’s phases. Thus, this feature’s utility seems limited in improving our understanding of a patient’s level of function across multiple domains.

Nevertheless, the residual deficit in total TUG time was 20% compared to the control group mean. This is consistent with the 20% residual deficit in function reported by Vissers et al. six to eight months post-surgery [2], and it supports a previous study suggesting that the total time of the TUG test could be a good indicator of the overall level of motor function [8].

Outcome variability

Although THA patients’ function improved in every phase of the TUG test after their surgery, on average, they nevertheless presented with deficits in each one of those phases when compared to the control group before and six months after their surgery. This supports previous studies showing improvements in function but an incomplete recovery six months after surgery [2]. However, the SDs of the percentage differences between groups were very high, e.g. the SD of the deficits in total TUG time at M0 was 38% and the mean deficit was 39%. These high SDs suggest great variability in patient profiles. Indeed, at M6, some patients had similar levels of function to the control group, whereas others showed no improvements in function (S3 Fig). Understanding the preoperative factors at the origin of those differences in profile could help to identify the patients at risk of having a lower level of function after surgery and improve patient rehabilitation.

The high variability resulting from the analysis of the TUG test’s phases suggests that it could be used to identify patient-specific levels of function. Identifying a functional deficit in a particular phase might help target the functional domain in which the patient is most limited. For example, a patient with a large post-surgical deficit in the sit-to-stand phase could be oriented towards specific strength training to reduce this limitation [33]. Such tests could also be used to monitor patients’ rehabilitation processes post-surgery.

The PCA showed that the largest part of the variance between groups was associated with speed features. This seems to indicate that patients improve the speed of their movements more than they improve their quality at six months after surgery. One explanation could be that the reduction in post-surgery pain leads to improved hip function and thus to increases in speed, but that the compensations developed before surgery, to avoid pain in the arthritic hip, might take more than six months to correct. Nevertheless, the range of thorax obliquity was significantly lower after surgery for the test’s walking and turn-to-sit phases, which indicates a decrease in compensatory movements during walking and less asymmetry during the sitting phase.

Surprisingly, the direction of the turn was not a relevant parameter for this pathology. Indeed, at M0 and M6, half the patients turned towards their pathological side and the other half towards their non-pathological side.

Implementation in clinical settings

The TUG test could be performed with a simpler, low-cost setup by using Inertial Measurement Units (IMUs), as previously done with other pathological populations [18]. Only three IMUs placed on the thorax, pelvis and thigh of the pathological hip would be necessary to measure all the features used in the present study. The iTUG test could be a simple and cost-effective means of assessing the functional deficits of patients with hip OA before and after THA, and it could help healthcare professionals to target patient-specific rehabilitation to close the gaps in function with asymptomatic controls. As recommended by the COSMIN taxonomy, the duration of the phases and the features should be assessed for reliability, validity and responsiveness before they are translated to clinical practice [40]. Recently, and with encouraging results, responsiveness to changes in multiple iTUG test parameters and the minimal detectable change in total TUG time have been evaluated among older adults (63.9 ± 6.1 years) undergoing a six-week physiotherapy programme [41]. Indeed, responsiveness to change was shown to have small to moderate effect sizes, and the minimal detectable change in total TUG time was 0.77 s. Moreover, Smith et al. [41] obtained positive feedback from patients and clinicians when implementing such an iTUG test in clinical settings, suggesting its good feasibility and acceptability. Nevertheless, further studies are needed to identify whether using such tests before surgery could help predict patients’ post-surgical function.

Limitations

The present study’s main limitation is its lack of one-year and ongoing follow-up to assess changes in patient function in the long term, which is key to understanding the success of THA [42].

The patient population and the control group had a significant difference in terms of weight, but greater weight has been reported as a risk factor for hip OA [43], which could explain this.

The duration of test phases and the time differences between groups were relatively small, especially for the sit-to-stand and turning phases. Thus, those results should be taken with care as the difference could be within measurement accuracy of the algorithm. Moreover, data centring and reduction led to the selection of parameters with non-clinically meaningful differences. Indeed, the PCA identified parameters with differences that were within the range of measurement accuracy (1 degree), e.g. the peak thorax flexion and range of thorax obliquity during the sit-to-stand phase. Future studies could use measurement accuracy as a criterion during the feature selection.

The features measured in the present study were chosen based on previous ones, but other features might also be relevant when assessing levels of function in each phase. For example, we observed visually that patients moved within a continuum between two strategies in the turn-to-sit phase: those who seemed the most impaired walked right up to the chair, turned and then sat down, and those who seemed less impaired could turn and sit in one continuous movement, including one or more backward steps. These two strategies were similar to those described by Weiss et al. [35] among elderly patients: the distinct transition strategy and the overlapping transition strategy. The “distance chair to start of turn” feature was chosen with the assumption that a longer distance would imply that patients turned and sat simultaneously, i.e. using an overlapping transition strategy. This feature was continuous, as this study aimed to assess percentages of differences and change, but a categorical feature classifying patients’ strategies could be of interest to understand patients’ motor function.

Conclusion

This study showed that, on average, patients with THA presented deficits in all four phases of the TUG test when compared to a control group before and six months after surgery, even though their motor function improved in all those phases after surgery. Walking seemed to be the main limitation facing patients before and six months after surgery. Furthermore, the large variability in functional level observed between patients indicates that the iTUG test was able to capture that variety among this population. Thus, this test could be appropriate for evaluating patient-specific levels of function in clinical settings. A biomechanical analysis of the TUG test with a simple IMU-based measurement system could provide cost-effective, relevant standard information for identifying patients’ specific functional deficits and could help target and monitor patient-specific rehabilitation in clinical practice.

Supporting information

S1 Fig. Figures to check the phases of patients.

(PDF)

Click here for additional data file.^{(1.3MB, pdf)}

S2 Fig. Figures to check the phases of the control group.

(PDF)

Click here for additional data file.^{(488.9KB, pdf)}

S3 Fig. Comparison of patient’s features and times of phases with the level of the control group.

(PDF)

Click here for additional data file.^{(10MB, pdf)}

S1 Table. Results of the principal component analysis.

(PDF)

Click here for additional data file.^{(288.1KB, pdf)}

S1 File. Algorithm for the detection of the timed up-and-go phases.

(PDF)

Click here for additional data file.^{(169.2KB, pdf)}

S2 File. Features description and selection.

(PDF)

Click here for additional data file.^{(225.4KB, pdf)}

Data Availability

We have uploaded our data on Yareta, an online repository following the FAIR principles. The DOI of the repository is: 10.26037/yareta:vynrqmr3hvegrcnh2oemxkwwb4 https://yareta.unige.ch/#/home/detail/64001c26-3b96-4978-838a-ea9eaf511f49.

Funding Statement

This work was supported by the Division of Orthopaedics and Trauma Surgery at the University Hospitals of Geneva, the “Fondation pour la recherche ostéoarticulaire” of Geneva, the “Programme Hospitalier de Recherche Clinique” and the “conseil régional de Bourgogne Franche-Comté”. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Liu X, Zi Y, Xiang L, Wang Y. Total hip arthroplasty: a review of advances, advantages and limitations. International journal of clinical and experimental medecine. 2015;8: 27–36. [PMC free article] [PubMed] [Google Scholar]
2.Vissers MM, Bussmann JB, Verhaar JAN, Arends LR, Furlan AD, Reijman M. Recovery of Physical Functioning After Total Hip Arthroplasty: Systematic Review and Meta-Analysis of the Literature. Physical Therapy. 2011;91: 615–629. doi: 10.2522/ptj.20100201 [DOI] [PubMed] [Google Scholar]
3.Bahl JS, Nelson MJ, Taylor M, Solomon LB, Arnold JB, Thewlis D. Biomechanical changes and recovery of gait function after total hip arthroplasty for osteoarthritis: a systematic review and meta-analysis. Osteoarthritis and Cartilage. 2018;26: 847–863. doi: 10.1016/j.joca.2018.02.897 [DOI] [PubMed] [Google Scholar]
4.World Health Organization. How to use the ICF: A practical manual for using the International Classification of Functioning, Disability and Health (ICF). Geneva; 2013.
5.Lovelock TM, Broughton NS, Williams CM. The Popularity of Outcome Measures for Hip and Knee Arthroplasties. Journal of Arthroplasty. 2018;33: 273–276. doi: 10.1016/j.arth.2017.08.024 [DOI] [PubMed] [Google Scholar]
6.Nilsdotter AK, Lohmander LS, Klässbo M, Roos EM. Hip disability and osteoarthritis outcome score (HOOS)–validity and responsiveness in total hip replacement. BMC Musculoskeletal Disorders. 2003;4: 10. doi: 10.1186/1471-2474-4-10 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Dayton MR, Judd DL, Hogan CA, Stevens-Lapsley JE. Performance-Based Versus Self-Reported Outcomes Using the Hip Disability and Osteoarthritis Outcome Score after Total Hip Arthroplasty. American Journal of Physical Medicine and Rehabilitation. 2016;95: 132–138. doi: 10.1097/PHM.0000000000000357 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Gandhi R, Tsvetkov D, Davey JR, Syed KA, Mahomed NN. Relationship between self-reported and performance-based tests in a hip and knee joint replacement population. Clinical Rheumatology. 2009;28: 253–257. doi: 10.1007/s10067-008-1021-y [DOI] [PubMed] [Google Scholar]
9.Podsiadlo D, Richardson S. The Timed “Up & Go”: A Test of Basic Functional Mobility for Frail Elderly Persons. Journal of the American Geriatrics Society. 1991;39: 142–148. doi: 10.1111/j.1532-5415.1991.tb01616.x [DOI] [PubMed] [Google Scholar]
10.Dobson F, Bennell K, Hinman RS, Abbott JH, Roos EM. Recommended performance-based tests to assess physical function in people diagnosed with hip or knee osteoarthritis. Melbourne; 2013. doi: 10.1016/j.joca.2013.05.002 [DOI] [PubMed] [Google Scholar]
11.Caronni A, Sterpi I, Antoniotti P, Aristidou E, Nicolaci F, Picardi M, et al. Criterion validity of the instrumented Timed Up and Go test: A partial least square regression study. Gait & Posture. 2018;61: 287–293. doi: 10.1016/j.gaitpost.2018.01.015 [DOI] [PubMed] [Google Scholar]
12.Petis SM, Howard JL, Lanting BA, Somerville LE, Vasarhelyi EM. Perioperative Predictors of Length of Stay After Total Hip Arthroplasty. The Journal of Arthroplasty. 2016;31: 1427–1430. doi: 10.1016/j.arth.2016.01.005 [DOI] [PubMed] [Google Scholar]
13.Nankaku M, Tsuboyama T, Akiyama H, Kakinoki R, Fujita Y, Nishimura J, et al. Preoperative Prediction of Ambulatory Status at 6 Months After Total Hip Arthroplasty. Physical Therapy. 2012;93: 1–6. doi: 10.2522/ptj.20120016 [DOI] [PubMed] [Google Scholar]
14.Sasaki K, Senda M, Nishida K, Ota H. Preoperative time required for the timed “up and go” test in women with hip osteoarthritis could predict a deep venous thrombosis complication after total hip arthroplasty. Acta medica Okayama. 2010;64: 197–201. doi: 10.18926/AMO/40012 [DOI] [PubMed] [Google Scholar]
15.Caronni A, Picardi M, Aristidou E, Antoniotti P, Pintavalle G, Redaelli V, et al. How do patients improve their timed up and go test? Responsiveness to rehabilitation of the TUG test in elderly neurological patients. Gait & Posture. 2019;70: 33–38. doi: 10.1016/j.gaitpost.2019.02.010 [DOI] [PubMed] [Google Scholar]
16.Coulthard JT, Treen TT, Oates AR, Lanovaz JL. Evaluation of an inertial sensor system for analysis of timed-up-and-go under dual-task demands. Gait & Posture. 2015;41: 882–887. doi: 10.1016/j.gaitpost.2015.03.009 [DOI] [PubMed] [Google Scholar]
17.Cimolin V, Cau N, Malchiodi G, Aspesi V, Merenda V, Galli M, et al. Do wearable sensors add meaningful information to the Timed Up and Go test? A study on obese women. Journal of Electromyography and Kinesiology. 2019;44: 78–85. doi: 10.1016/j.jelekin.2018.12.001 [DOI] [PubMed] [Google Scholar]
18.Salarian A, Horak FB, Zampieri C, Carlson-Kuhta P, Nutt JG, Aminian K. iTUG, a Sensitive and Reliable Measure of Mobility. IEEE Transactions on Neural Systems and Rehabilitation Engineering. 2010;18: 303–310. doi: 10.1109/TNSRE.2010.2047606 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Zampieri C, Salarian A, Carlson-kuhta P, Aminian K, Nutt JG, Horak FB. The instrumented timed up and go test: potential outcome measure for disease modifying therapies in Parkinson’s disease. Journal of Neurology, Neurosurgery, and Psychiatry. 2010;81: 171–176. doi: 10.1136/jnnp.2009.173740 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Hansen BJ, Hallows RK, Kelley SS. The Rottinger approach for total hip arthroplasty: Technique and review of the literature. Current Reviews in Musculoskeletal Medicine. 2011;4: 132–138. doi: 10.1007/s12178-011-9093-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Moore AT. The self-locking metal hip prosthesis. Journal of Bone and Joint Surgery. 1957;39: 811–827. [PubMed] [Google Scholar]
22.Altman R, Alarcón G, Appelrouth D, Bloch D, Borenstein D, Brandt K, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hip. Arthritis & Rheumatism. 1991;34: 505–514. doi: 10.1002/art.1780340502 [DOI] [PubMed] [Google Scholar]
23.Baker R, Leboeuf F, Reay J, Sangeux M. The Conventional Gait Model—Success and Limitations. In: Müller B, Wolf SI, editors. Handbook of Human Motion. Springer International Publishing AG; 2017. pp. 1–19. doi: [DOI] [Google Scholar]
24.Gløersen Ø, Federolf P. Predicting Missing Marker Trajectories in Human Motion Data Using Marker Intercorrelations. PloS one. 2016. doi: 10.1371/journal.pone.0152616 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Beyea J, Mcgibbon CA, Sexton A, Noble J, Connell CO. Convergent Validity of a Wearable Sensor System for Measuring Sub-Task Performance during the Timed Up-And-Go Test. Sensors. 2017;17: 1–18. doi: 10.3390/s17040934 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Barre A, Armand S. Biomechanical ToolKit: Open-source framework to visualize and process biomechanical data. Computer Methods and Programs in Biomedicine. 2014;114: 80–87. doi: 10.1016/j.cmpb.2014.01.012 [DOI] [PubMed] [Google Scholar]
27.Janssen WGM, Bussmann HBJ, Stam HJ. Determinants of the sit-to-stand movement: A review. Physical Therapy. 2002;82: 866–879. doi: 10.1093/ptj/82.9.866 [DOI] [PubMed] [Google Scholar]
28.Abujaber SB, Marmon AR, Pozzi F, Rubano JJ, Zeni JA. Sit-To-Stand Biomechanics Before and After Total Hip Arthroplasty. Journal of Arthroplasty. 2015;30: 2027–2033. doi: 10.1016/j.arth.2015.05.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Aissaoui R, Dansereau J. Biomechanical analysis and modelling of sit to stand task: a literature review. Proceedings of the IEEE International Conference on Systems, Man and Cybernetics. 1999;1: 141–146. doi: 10.1109/icsmc.1999.814072 [DOI]
30.Alexander NB, Schultz AB, Warwich DN. Rising from a chair: Effects of age functional ability on performance biomechanics. Journals of Gerontology. 1991;46: 91–98. doi: 10.1093/geronj/46.3.m91 [DOI] [PubMed] [Google Scholar]
31.Nankaku M, Tsuboyama T, Kakinoki R, Kawanabe K, Kanzaki H, Mito Y, et al. Gait analysis of patients in early stages after total hip arthroplasty: Effect of lateral trunk displacement on walking efficiency. Journal of Orthopaedic Science. 2007;12: 550–554. doi: 10.1007/s00776-007-1178-2 [DOI] [PubMed] [Google Scholar]
32.Hollands KL, Hollands MA, Zietz D, Miles Wing A, Wright C, Van Vliet P. Kinematics of turning 180° during the timed up and go in stroke survivors with and without falls history. Neurorehabilitation and Neural Repair. 2010;24: 358–367. doi: 10.1177/1545968309348508 [DOI] [PubMed] [Google Scholar]
33.Schot PK, Knutzen KM, Poole SM, Mrotek LA. Sit-to-stand performance of older adults following strength training. Research Quarterly for Exercise and Sport. 2003;74: 1–8. doi: 10.1080/02701367.2003.10609058 [DOI] [PubMed] [Google Scholar]
34.Boonstra MC, Schreurs BW, Verdonschot N. The Sit-to-Stand Movement: Differences in Performance Between Patients After Primary Total Hip Arthroplasty and Revision Total Hip Arthroplasty With Acetabular Bone Impaction Grafting. Physical Therapy. 2011;91: 547–554. doi: 10.2522/ptj.20090376 [DOI] [PubMed] [Google Scholar]
35.Weiss A, Mirelman A, Giladi N, Barnes LL, Bennett DA, Buchman AS, et al. Transition between the Timed Up and Go turn to sit subtasks: is timing everything? Journal of the American Medical Directors Association. 2016;17. doi: 10.1016/j.jamda.2015.07.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Baker R, McGinley JL, Schwartz MH, Beynon S, Rozumalski A, Graham HK, et al. The Gait Profile Score and Movement Analysis Profile. Gait & Posture. 2009;30: 265–269. doi: 10.1016/j.gaitpost.2009.05.020 [DOI] [PubMed] [Google Scholar]
37.Massey FJ. The Kolmogorov-Smirnov Test for Goodness of Fit. Journal of the American Statistical Association. 1951;46: 68–78. doi: 10.1080/01621459.1951.10500769 [DOI] [Google Scholar]
38.Cattell RB. The Scree Test For The Number Of Factors. Multivariate Behavioral Research. 1966;1: 245–276. doi: 10.1207/s15327906mbr0102_10 [DOI] [PubMed] [Google Scholar]
39.Holm S. A Simple Sequentially Rejective Multiple Test Procedure. Scandinavian Journal of Statistics. 1979;6: 65–70. doi: 10.2307/4615733 [DOI] [Google Scholar]
40.Mokkink LB, Terwee CB, Patrick DL, Alonso J, Stratford PW, Knol DL, et al. The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: An international Delphi study. Quality of Life Research. 2010;19: 539–549. doi: 10.1007/s11136-010-9606-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Smith E, Cunningham C, Greene BR, Mccarthy U, Catherine P. Detecting subtle mobility changes among older adults: the Quantitative Timed Up and Go test. Aging Clinical and Experimental Research. 2020. doi: 10.1007/s40520-020-01733-7 [DOI] [PubMed] [Google Scholar]
42.Rylander JH. CORR Insights®: Gait Analysis Reveals that Total Hip Arthroplasty Increases Power Production in the Hip During Level Walking and Stair Climbing. Clinical orthopaedics and related research. 2019;477: 1848–1850. doi: 10.1097/CORR.0000000000000865 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Aresti N, Kassam J, Nicholas N, Achan P. Hip osteoarthritis. BMJ (Online). 2016;354: 1–10. doi: 10.1136/bmj.i3405 [DOI] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0255037.r001

Decision Letter 0

Peter Andreas Federolf

11 Feb 2021

PONE-D-20-38214

Which functional tasks present the largest deficits for patients with total hip arthroplasty before and 6 months after surgery? A study of the Timed Up-and-Go phases

PLOS ONE

Dear Dr. Gasparutto,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Specifically, both reviewers requested more details on several items in the methods section of the paper.

Please submit your revised manuscript by Mar 28 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

We look forward to receiving your revised manuscript.

Kind regards,

Peter Andreas Federolf

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

3.Thank you for stating the following in the Competing Interests section:

"I have read the journal's policy and the authors of this manuscript have the following competing interests: Pierre Martz is a paid consultant for XNOV and SERF."

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: General

The study investigated changes in an instrumented TUG test in patients before and 6 months after total hip replacement and compared them to healthy controls. They predefined biomechanical features according to the literature and used a PCA analysis to reduce the number of features. The selected features were then compared between groups. However, the description of the biomechanical features and especially the statistical analysis section is insufficient and makes it difficult completely follow and understand the analysis and results. If clarified, the paper presents interesting and clinically relevant data of a simple test that could be used in clinical practice to assess the functional outcome after total hip replacement.

Major comments:

Analysis/Table 2: The biomechanical features are insufficiently explained. Abbreviations are not explained (table 2). How was angular velocity of thorax/pelvis defined? In a plane/resultant? With respect to what point/segment? How was side of turn defined? How was pathological/contralateral defined in controls?

Statistics

How were the biomechanical features reduced?

You state that you kept one feature per component and per subtask. What do you mean by component? PC component? Why do you keep Peak angular velocity thorax and peak vertical velocity thorax in turn to sit which are both associated with PC1 (table 3)?

Bonferroni correction: what is the number of tests?

It is unclear why calculate the percentage of variation. From my understanding of the results, this is not presented anywhere and not related to the aims of the study. Please clarify.

Results

You often write about a deficit. But it is not explained in the methods how this deficit was defined.

Numbers in tables, figures and text are repeatedly different from each other even though they should be same (i.e. Fig.2 and Table 4 values for control group).

Why do you use different thresholds for presenting p-values (i.e. Table 4 and Fig 2. once p<0.001, once p<0.003)?

You write about a statistically larger deficit in some phases. It is difficult to understand how you derive these results. It would be helpful to directly show the results of the statistical comparison between subtasks (incl p-values and corresponding threshold of Bonferroni correction)

Figure 2: The figure caption states that Fig 2 presents an overview about the features selected by the PCA. However, it also contains the duration of the different TUG phases. From my understanding, these were not included in the PCA. Please clarify. Moreover, an explanation on how the points in the plot were selected is missing (sometimes larger number is inside, sometimes outside. How was position of point for M6 defined?).

Minor comments:

Abstract

Line 38: Do you mean “improve”? What parameter? Shorter duration of the phases?

Line 39-41: You conclude that the iTUG could be instrumented with a wearable sensor in a clinical setting. However, from your paper, it’s not clear how that would be achieved. Please revise.

Introduction

Line 47/48: This should be “6 months and 1 year after THA”

Lines 61-62: Please write “test” after each test (6minute walk test, 30s chair stand test or TUG test), or only after TUG test.

Lines 78ff: Please specify what parameters are measured/analyzed in the iTUG.

Methods

Lines 107-109: You state that all patients had either stage II, III or IV hip OA, but in Table 1 K/L classification is missing for 4 patients. Why are these data missing?

Line 118: Please change to “destabilised” (you used British English spelling in the rest of the manuscript).

Line 119-120: Please change to “between groups” (or “between populations”)

Line 131: There is a space missing between “chair” and “without”

Lines 133ff: Why did you perform 10 TUGs but only analyze the fastest trial of the first 3 TUGs? Please explain.

Line 147/S1: Is there any information on what S1 shows? The figure is not self-explanatory, it’s not clear what the sticks indicate.

Line 175: Please change to “one global figure per phase”.

Line 176: I think you mean “development” instead of “evolution”. Please change throughout the manuscript (evolution and evolve).

Results (there are no line numbers after page 14, so comments were made on paragraphs)

Table 4: Please check that you use the same precision for mean and SD. Was the SD of mean angular velocity pelvis (turn) and peak angular velocity thorax (turn to sit) really below 1?

Line 206: Table 6 appears before Table 5 in the text. Please change

Line 209: This should be “Bonferroni correction”

Table 5: What do bold numbers indicate? What is parameter corr. total time?

Table 6: Please specify in legend, what parameter is meant by percentage.

Table 5/6: Were the 95% CI of the difference really that much smaller than the SD of the difference?

P.14, paragraph 1, last sentence (49-86%): why do you here include all parameters and in previous paragraph only those with significant differences?

Discussion

P.15, paragraph 2: what scores of the patients? This implies to me that you compared PC scores, but from the results it’s not what you did. Please clarify.

P.15, paragraph 3, last sentence: You state that this (hip flex/ext range walking phase) has a strong positive effect. However, only 20% of patients reach level of controls. Isn’t this contradictory?

P.17, variability of outcomes: From the presented results, I don’t understand how you analysed variability.

P.17, paragraph 1: were there any patients that had decreased performance 6months postoperative?

P.17, paragraph 2: please change “strong post-surgery deficit” to “large post-surgery deficit”

P.18, paragraph 1: Please change to “three synchronized IMUs…”

P.18, paragraph 2: you state that a simple 1 IMU/smartphone provides enough parameters to assess the function of the patients. However, from your analysis this is not valid. The presented parameters were not measured with IMUs and would need several sensors (data on pelvis, hip, thorax…). Please elaborate on how such a sensor should be set up according to your results.

P.18, paragraph 2, last 4 lines: Please use either “a test” or “tests”. Please change to “However, further studies are needed…”

P.18, paragraph 4. Please change to “duration of the phases” instead of “time of the phases”.

P.19, paragraph 1: You state that there was a study on minimal detectable changes of iTUG parameters. How do the difference between pre and post patients compare to these minimal detectable changes? Please elaborate.

P.20, paragraph 1: You abbreviated timed-up and go test as TUG test, please change.

Reviewer #2: The study explores functional deficits before and its change after total hip arthroplasty with an instrumented timed up-and-go test, analyzing phase specific differences between patients and a control group. The topic is of clinical relevance and the sample size adequate to produce meaningful results. The relevance and experimental setup are well described and introduced. Limitations are within the study design, where no post-measurement of the control group is available, resulting in multiple testing and a reduction in statistical power. Further methodological shortcomings, including the variable selection, reduce the significance of the results. In some cases, false or non-ideal statistical tests are used and in other cases (e.g. PCA), methods are not well enough described to judge the correctness of the results. Combined with further language and typing errors, major revisions are necessary to improve the manuscript. Nevertheless, the study has the potential to come to good and meaningful results.

To improve the study/manuscript following main issues should be revised:

1. The preselection of variables is arbitrary and not theory or hypothesis driven. The sole explanation are previous studies, who investigated other disorders and included further variables which have not been considered in the current study (e.g. spatiotemporal variables, such as cadence, stride length, length of single support…). Ideally the variables should be based on known limitations in hip OA patients, identified in previous studies. Furthermore, the joint angles over the complete range of motion (ROM) are unspecific, making conclusions about deficits for specific movements or walking phases difficult. For example, dividing hip ROM in flexion and extension would help to interpret whether deficits during walking are more related to the late stance phase (restricted hip extension), indicating problems to generating step length or swing phase (restricted hip flexion), which might result in higher fall risk due to lower foot clearance. It is unclear why peak thorax obliquity was only analyzed during sit2stand/turn2sit and not walking/turning and why peak thorax obliquity was not directly investigated of the pathological side? Conversely, what is the rationale behind investigating different mean velocities of the pelvis and thorax. In which scenario do the authors expect to find notable differences in the mean velocities of these segments?

2. PCA:

The PCA might be a good approach to identify important variables. However, insufficient clarity and scarce description make it impossible to judge the results. Especially in the presence of differently scaled variables the standardization and re-scaling is of utmost importance for reliable results. It is unclear how the variables were centered and what is meant by “reduced”. Please clarify.

It is not described if both groups were included in the PCA analysis.

Why was the time not included? Wouldn’t that give important information which quality features are related to the time? Doing so might also spare the correlations, which further complicate the analysis.

The choice of variables based on the cos² values is inconsistent. For example, in quality during walking “Range of flexion contralateral hip (0.47)” and “Range obliquity thorax (0.30)” have higher cos² value than “RMS obliquity thorax (0.27)”. The selection method or the description needs clarification.

3. The Kolmogorov-Smirnoff test is inappropriate to compare the patient characteristics. Age distribution can be the same in a group of children and adults. The interesting information is, whether the patient and control group are of similar age. Therefore, t-tests are needed, and the K-S tests only serves to test the assumption of normal distribution for the t-tests. For weight, a non-parametric test like the Mann-Whitney U test is necessary.

4. Number of tests and p-value correction:

The fact that the study design does not allow for repeated measures of variance and the downside of multiple testing is a problem. The current approach with the Bonferroni method is very conservative: if I count right 16*3 t-test were calculated for the variables plus 4 for the phases, resulting in a significance level of 0.05/52 <= 0.001, which conversely inflates the false negative (type II error) risk.

I would recommend controlling the FDR with the Benjamini-Hochberg correction or the Holm method.

To further strengthen the study effect sizes with CI could be calculated.

5. Overall deficit in percent:

The approach to calculate the overall deficit only on the significantly different variables within a phase is questionable. What if there was no significant variable within a phase? Has this phase gets no overall rating? What about cases where significant differences exist only pre- or post-surgery? Is the variable only included in one of the overall scores but not in the other?

Overall score should include all important variables (already identified by the PCA) and differences tested afterwards.

Further minor revisions:

Note: These corrections are not exhaustive since it can be expected that major changes to the manuscript will follow the main revisions. Therefore, the corrections are rather examples hinting towards deficits requiring attention.

1. Further remarks:

• “52 aged-matched controls” if they were age-matched, there would not be a 3y difference in age.

• “pregnancy or breast-feeding patients” relevant in this age group?

• “The patients and control groups did not differ in sex, […]” as pointed out earlier, the K-S test does only test distribution differences and gives no insight about mean group differences.

• Table 4: Please revise/double-check. SD of Mean angular velocity pelvis (deg/s) (turn) and Peak angular velocity thorax (deg/s) (turn2sit) at M6 appear wrong.

Use the same number of digits for a variable mean and SD.

• Table 5 is inconsistent. Sometime coloring is false (e.g. SD from Range thorax obliquity (%) during walking pre-post comparison). Also reporting of correlations is inconsistent (sometimes left out despite sig. diff and sometimes displayed without sig. diff.)

• “This deficit was statistically larger than the other phases (p < 0.05) but not with the Bonferonni correction.“ Reporting „significant“ results, which are insignificant after p-value correction goes against the logic of doing a correction.

• “[…] selected by PCA and the scores of the patients […]” implies that PCA scores were compared, which is not the case.

• “[…] presence of a limp (indicated by the range of thorax obliquity).” Limping might not be very well represented by the thorax ROM. It does not reveal if the trunk lean was one-sided. Furthermore, thorax ROM might not be a good measure of limping because there are probably ways of limping that don’t involve an excessive trunk lean. Letter is a good strategy to unload the hip and is a known compensatory movement for weak hip abductors or hip pain. For limping, however, a better and more direct measure would be the single support phase.

• The section “Implementation in clinical settings” might be well-meant, however, has too little relation to the results.

• “The reliability, validity and responsiveness should be assessed for the time of the tasks and for the features of this study before using them in clinical practice, […]” Limitations should relate to the study.

• Last paragraph of Limitations: “Surprisingly, the PCA identified parameters…” is not a limitation and belongs in the discussion.

2. Variable names:

The variables should be named more precisely.

“Lateral distance between feet”. Is that step width? There is no lateral distance between the feet.

“Only 8% of patients had a range of hip flexion extension during walking […]” variable name in tables implies only hip flexion was analyzed. If the variable comprises hip flexion/extension it should be named accordingly.

3. Language and typos

Language and wording are not quite yet on a publication level and should get revised by a native speaker or expert.

Examples for awkward phrasing, wording, grammar, and typos:

• “Exclusion criteria included a previous surgery to the hip” � previous hip surgery.

• “hospital length of stay after surgery” � hospitalization?

• “interfere or be destabilised by gait analysis” � interfere with or be worsened by …

• “(i.e. above the mean of the control group minus one standard deviation)” � within 1SD of the control group

• “Evolution” is the wrong word to describe the change/difference between pre/post

• “pace” unusual in this context. Speed is probably more appropriate.

• “TUG test was not performed previously for patients with end stage” � previously in patients.

• “up-an-go” � up-and-go.

In conclusion, there is much room for improvement and besides the major issues all the little errors throughout the manuscript and tables indicate that the manuscript is not ready for publication. Nevertheless, the study might provide very useful information to the scientific community.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Sep 10;16(9):e0255037. doi: 10.1371/journal.pone.0255037.r002

Author response to Decision Letter 0

30 Mar 2021

Dear editor and reviewers,

We would like to thank the reviewers for their thorough analysis of our paper and the highly relevant observations made on the method, presentation of results, and points to clarify. These comments helped us increase significantly the quality of the paper as well as clarifying the outcome of the study.

The following document gives a detailed answer of the comments of each reviewers. Our answers are written after each comments of the reviewers.

We hope that these answer will satisfy the reviewers and we remain at your disposal for any additional clarifications,

Best regards,

The authors

Reviewer #1:

General

- The statistical analysis and its description was modified based on the comments of reviewer 1 and 2 and we added a supporting information file (S4) to give details on the computation of the biomechanical features and their selection.

Major comments:

- A supplementary material that defines the various features and gives more details in the selection was added and mentioned in the manuscript.

L 152: “The choice of the features and their definitions are detailed in Supporting Information S4.”

We believe that including the full definition in the manuscript would impede the readability of the paper. For the control group, the features of both hips were averaged, thus the “pathological” and “contralateral” features are equal.

Statistics

How were the biomechanical features reduced?

- The description of the PCA was modified as follow to give more details:

L 157: “A principal component analysis (PCA) was performed to keep only the most relevant biomechanical features in the analysis (excluding the time of the phases). The feature were centred and reduced by subtracting the mean and dividing by the standard deviation, to account for the difference of scales. All three groups (patients at M0, patients at M6 and control group) were included in the analysis.”

You state that you kept one feature per component and per subtask. What do you mean by component? PC component?

- Yes, we mean PC component, it was added to the manuscript for clarity (L 162).

Why do you keep Peak angular velocity thorax and peak vertical velocity thorax in turn to sit which are both associated with PC1 (table 3)?

- Indeed, both features are associated with PC1. The turn-to-sit phase is composed of two movements, turning and sitting, thus although we grouped those two movements in one phase, we wanted to keep a feature per movement. However, we did this for the speed features but not quality feature. This is questionable and thus we removed the feature “peak vertical velocity of thorax” from the analysis for methodological consistency.

Bonferroni correction: what is the number of tests?

- For the differences in features between groups (Table 4), there were 16 tests per group comparison (one for each feature) and 3 group comparisons. The number of tests was 48 leading to a corrected p-value of 0.05/48 = 0.001.

For the comparison of the differences between groups at phase level, there were 4 tests per group comparison and 3 group comparisons. The number of tests was 12 leading to a corrected p-value of 0.05/12 = 0.004.

Reviewer 2 suggested that we use the Holm method instead of the Bonferroni method. We removed the mentions to Bonferroni and used the Holm method instead (L 166).

It is unclear why calculate the percentage of variation. From my understanding of the results, this is not presented anywhere and not related to the aims of the study. Please clarify.

- What we meant by “percentage of variation” was the change in patient’s function between M0 and M6 expressed in percentage of M0. These results are presented in the results, in the section “Evolution of function six months after surgery”. The manuscript was modified as follow to avoid misunderstandings:

L 172: “The change between M0 and M6 was expressed in percentage of the feature value at M0 to express the increase or decrease in function after surgery.”

Results

You often write about a deficit. But it is not explained in the methods how this deficit was defined.

- An explanation of the deficits was added to the method in the Statistics section:

L 169: “A lower function of the patients with respect to the control group was described as a deficit. Depending on the feature, a lower function can be a higher value (e.g. duration of phase) or a lower value (e.g. peak vertical velocity of the thorax for the S2S).”

Numbers in tables, figures and text are repeatedly different from each other even though they should be same (i.e. Fig.2 and Table 4 values for control group).

- The consistency of the numbers in the text, figures and tables were checked.

Why do you use different thresholds for presenting p-values (i.e. Table 4 and Fig 2. once p<0.001, once p<0.003)?

- This is a typo, the correct value was p < 0.001, since the Holm method is now used, it was corrected in Fig. 2 as p < Holm correction.

- As suggested by reviewer 2, we are now using the Holm correction, thus we removed mentions to the Bonferroni correction. To clarify, we mentioned in the manuscript, that the deficits were significant with the Holm method (L 214, L 230 and L 236). The Holm method as a varying level of p-value to assess significance of the test and it seemed that reporting both number would impede readability of the manuscript. However, if the reviewer sees fit, we can report both values for the tests of the differences between phases.

- Indeed, the duration of the phases was not included in the PCA, we changed slightly the title of Fig2 as follow: “Overview of the duration of the phases and features selected by the PCA for the three groups […]”

Moreover, an explanation on how the points in the plot were selected is missing (sometimes larger number is inside, sometimes outside. How was position of point for M6 defined?).

To avoid having lines crossing each other and make the figure easier to read, we chose to have the control group values on the outside and M0 value on the inside. In this way, an improvement of patients is visually seen as their score by going toward the outside part of the plot. In some cases a lower value will be an improvement (e.g. shorter duration of phase), in other cases a higher value will be an improvement (e.g. higher peak linear velocity of pelvis).

A short explanation was added to the legend of Fig.2:

Lxx: “Feature values representing higher function are on the outer edge and lower function values are on the inner edge.”

Minor comments:

Abstract

Line 38: Do you mean “improve”? What parameter? Shorter duration of the phases?

- Yes, we changed to ‘improved’. This statement is related to the results with “one global figure per phase” (Table 5). We kept this statement general for the abstract due to the word limitations but if the reviewer thinks it is needed, we can add some more details.

Line 39-41: You conclude that the iTUG could be instrumented with a wearable sensor in a clinical setting. However, from your paper, it’s not clear how that would be achieved. Please revise.

- Three inertial measurement units taped to the thorax, pelvis and thigh on the side of surgery could be used to compute the same features that we computed in this paper. This was mentioned in the discussion in the section “Implementation in clinical settings” (L 327).

The mention of the Inertial Measurement Units was removed from the abstract to avoid confusion, but kept in the discussion. The abstract was modified as follow:

L 39: “This simple test could be relevant to quantify patient-specific deficits of function to guide and follow post-operative rehabilitation in clinical settings.”

Introduction

Line 47/48: This should be “6 months and 1 year after THA”

- The manuscript was changed accordingly.

Lines 61-62: Please write “test” after each test (6minute walk test, 30s chair stand test or TUG test), or only after TUG test.

- The manuscript was changed accordingly.

Lines 78ff: Please specify what parameters are measured/analyzed in the iTUG.

- The parameters measured/analysed in the iTUG depend on the studies. The point was to say that the iTUG provides biomechanical parameters that can lead to a deeper understanding of the pathologies that are studied. The relevant parameters will differ from pathology to pathology.

Methods

Lines 107-109: You state that all patients had either stage II, III or IV hip OA, but in Table 1 K/L classification is missing for 4 patients. Why are these data missing?

- Those four patients performed X-rays outside of the Hospital and didn’t bring the X-rays during their visits. It was not possible to evaluate their OA stage with the K/L classification. If the reviewer sees it appropriate, we can mention it in the manuscript.

Line 118: Please change to “destabilised” (you used British English spelling in the rest of the manuscript).

- The manuscript was changed accordingly.

Line 119-120: Please change to “between groups” (or “between populations”)

- The manuscript was changed accordingly.

Line 131: There is a space missing between “chair” and “without”

- The manuscript was changed accordingly.

Lines 133ff: Why did you perform 10 TUGs but only analyze the fastest trial of the first 3 TUGs? Please explain.

- The standard way of performing the TUG test is to perform two tests and keep the fastest time (Podsialdo et al. 1991). Ten measurements were performed for a secondary study that would evaluate the variability of the test. We kept only the first 3 for this analysis to be close to the initial definition of the TUG test and, because for some patients, the measurement of the first or second trial failed. If the reviewer sees it appropriate, we can mention it in the manuscript.

- Podsiadlo D, Richardson S. The Timed “Up & Go”: A Test of Basic Functional Mobility for Frail Elderly Persons. J Am Geriatr Soc. 1991;39: 142–148.

Line 147/S1: Is there any information on what S1 shows? The figure is not self-explanatory, it’s not clear what the sticks indicate.

- We added comments on the first figure of S1 and S2 to guide the reader. If the reviewer sees fit we can add further explanations.

Line 175: Please change to “one global figure per phase”.

- The manuscript was changed accordingly.

Line 176: I think you mean “development” instead of “evolution”. Please change throughout the manuscript (evolution and evolve).

- We modified “evolution” by “change” throughout the manuscript.

Results (there are no line numbers after page 14, so comments were made on paragraphs)

- Line numbers were added to the whole document.

Table 4: Please check that you use the same precision for mean and SD.

- This was corrected in the table.

Was the SD of mean angular velocity pelvis (turn) and peak angular velocity thorax (turn to sit) really below 1?

- No, this is a typo.

Line 206: Table 6 appears before Table 5 in the text. Please change

- The manuscript was changed accordingly, the number of Table 6 was changed to 5 and conversely.

Line 209: This should be “Bonferroni correction”

- The method to assess the significant differences was changed, the “Holm method” is now mentioned.

Table 5: What do bold numbers indicate? What is parameter corr. total time?

- The bold numbers were used to outline results during our analysis and were removed from the manuscript.

The parameter “corr. total time” is the coefficient of correlation of the considered feature with the total time of the TUG. It was changed to “R total time” in the table and a description was added to the legend.

Table 6: Please specify in legend, what parameter is meant by percentage.

- The definition of the percentages was done in the Statistics section of the method:

L 168: “To evaluate the average difference of the patients’ function with the control group, the values of the features at M0 and M6 were reported in percentage of the mean of the control group. A lower function of the patients with respect to the control group was described as a deficit. Depending on the feature, a lower function was either a higher value (e.g. duration of phase) or a lower value (e.g. peak vertical velocity of the thorax for the S2S). The change between M0 and M6 was expressed in percentage of the feature value at M0 to express the increase or decrease in function after surgery. Furthermore, the percentage of patients at the level of the control group (i.e. within 1 standard deviation of the control group) was evaluated, as well as the percentage of patients with an increase in function between M0 and M6. The percentages of differences with the control group and the percentage of change between M0 and M6 were correlated with the differences and change of the total time.”

The legend of the table was modified as follow for clarity:

“Table 5: Mean difference ±SD (95% confidence interval) between groups at phase level expressed in percentages”

Table 5/6: Were the 95% CI of the difference really that much smaller than the SD of the difference?

- Yes, we used the formula 95%CI = mean ± 1.96 * SD / sqrt(n) with n the number of patients. As we had 71 patients it reduces the expected range of the 95% CI.

As an example, for the values of the S2S comparison of patients M0 vs CG, the mean and SD of the difference for the 71 patients are respectively -30% and 37% which lead to:

- A lower bound of: -30 – 1.96 * 37 / sqrt(71) = -39

- An upper bound of: -30 + 1.96 * 37 / sqrt(71) = -21

P.14, paragraph 1, last sentence (49-86%): why do you here include all parameters and in previous paragraph only those with significant differences?

- This is a mistake, we modified the manuscript as follow:

L 230: “Only 27% of patients had a range of hip flexion-extension during walking at the level of the controls while the other significant features had 46% to 62% of patients at this level (Table 5).”

Discussion

P.15, paragraph 2: what scores of the patients? This implies to me that you compared PC scores, but from the results it’s not what you did. Please clarify.

- With “scores of the patients”, we meant the values of the features for the patients. As this might be misleading, we modified the manuscript as follow:

L 259: “To that end, the TUG test was divided into four phases [15] representing different daily living activities (sit-to-stand, walking, turning and turn-to-sit), multiple features were selected by PCA and finally, the features of the patients were compared before surgery and at 6 months after surgery and to a control group for each phase.”

- We meant that this was the feature with the largest number of patients with an increase (82%). We modified this part of the discussion as follow:

L 269: “This feature had the lowest number of patients at the level of the control groups before and 6 months after surgery (8% and 27% respectively). However, this feature showed the highest number of patients with a positive change in function 6 months after surgery (82%). This suggests that, despite the deficits 6 months after surgery, THA has a positive effect on the sagittal hip kinematics during walking.”

P.17, variability of outcomes: From the presented results, I don’t understand how you analysed variability.

- The SD values on the percentage of deficits at feature level and phase level were very high. The SD values were often at the same level as the mean and even larger than the mean for some cases (Table5 and 6). In our opinion, this indicates a high variability in the profiles of the patients.

This idea was clarified in the manuscript:

L 299: “However, the SD levels of the percentage of differences between groups where very high, e.g. the SD of the deficits of the total TUG time at M0 was 38% while the mean deficits was 39%. These high levels of SD suggests the high variability of profiles in patients. Indeed, some patients were at the level of control groups while others did not improve their function at M6 (Supporting Information S6).”

P.17, paragraph 1: were there any patients that had decreased performance 6 months postoperative?

- Yes, some patients had decreased performance in multiple feature. We did not see patients with a clear decrease on all feature. The supplementary file S6 shows the values of the features at M0 and M6 of all patients compared to the control groups. As examples, patient 22 had a decrease in the duration of most phases and patient 42 had a decrease on most of the features. On the contrary, patient 7 had an increase for most of the features.

P.17, paragraph 2: please change “strong post-surgery deficit” to “large post-surgery deficit”

- The manuscript was changed accordingly.

P.18, paragraph 1: Please change to “three synchronized IMUs…”

- The manuscript was changed accordingly.

- To obtain the same features we would need IMUs on thorax, pelvis and thigh of the pathological side. However, this setup could be cumbersome in clinical settings and we suggest that a setup with one IMU would have more chances to be used in such settings. We expect that one IMU on the thorax could provide sufficient information to assess the function of the patients, but it could not provide the same features as the one described in this study. We removed this part from the manuscript as it is more a speculation and because the present paper does not have results supporting this claim.

P.18, paragraph 2, last 4 lines: Please use either “a test” or “tests”. Please change to “However, further studies are needed…”

- The manuscript was changed accordingly.

P.18, paragraph 4. Please change to “duration of the phases” instead of “time of the phases”.

- The manuscript was changed accordingly.

- The sentence we used could lead to misinterpretation. The study by Smith et al. (2020) reported in the paper assessed only the MDC of the total time evaluated measured with a commercial iTUG system but did not report MDC for the parameters of the TUG. However, they reported the effect size “to quantify the change observed in each of the [i]TUG parameters”. We modified the sentence in the manuscript as follow:

L 336: “Recently, the responsiveness to change of multiple parameters of the iTUG and minimal detectable change of the total time have been evaluated among older adults (63.9 ± 6.1 years) undergoing a six-week physiotherapy program with encouraging results [36]. Indeed, the responsiveness to change was shown with small to moderate effect sizes and the MDC of the total time was 0.77s.”

P.20, paragraph 1: You abbreviated timed-up and go test as TUG test, please change.

- The manuscript was changed accordingly.

Reviewer #2

The study explores functional deficits before and its change after total hip arthroplasty with an instrumented timed up-and-go test, analyzing phase specific differences between patients and a control group. The topic is of clinical relevance and the sample size adequate to produce meaningful results. The relevance and experimental setup are well described and introduced. Limitations are within the study design, where no post-measurement of the control group is available, resulting in multiple testing and a reduction in statistical power. Further methodological shortcomings, including the variable selection, reduce the significance of the results. In some cases, false or non-ideal statistical tests are used and in other cases (e.g. PCA), methods are not well enough described to judge the correctness of the results. Combined with further language and typing errors, major revisions are necessary to improve the manuscript. Nevertheless, the study has the potential to come to good and meaningful results.

Major Revisions: To improve the study/manuscript following main issues should be revised:

1. The preselection of variables is arbitrary and not theory or hypothesis driven. The sole explanation are previous studies, who investigated other disorders and included further variables which have not been considered in the current study (e.g. spatiotemporal variables, such as cadence, stride length, length of single support…). Ideally the variables should be based on known limitations in hip OA patients, identified in previous studies. Furthermore, the joint angles over the complete range of motion (ROM) are unspecific, making conclusions about deficits for specific movements or walking phases difficult. For example, dividing hip ROM in flexion and extension would help to interpret whether deficits during walking are more related to the late stance phase (restricted hip extension), indicating problems to generating step length or swing phase (restricted hip flexion), which might result in higher fall risk due to lower foot clearance. It is unclear why peak thorax obliquity was only analysed during sit2stand/turn2sit and not walking/turning and why peak thorax obliquity was not directly investigated of the pathological side? Conversely, what is the rationale behind investigating different mean velocities of the pelvis and thorax. In which scenario do the authors expect to find notable differences in the mean velocities of these segments?

- A table was added to supporting information with more details on the choice of the features.

The spatiotemporal parameters of gait were not selected due to the very short distance of walking that was below or equal to 3m. The target for turn was at 3m from the table, but, when considering that some participants started turning before the target and some started walking while standing up, the distance of walking was often below 3m. Previous studies showed that 5 steps are required to have stable variability of gait parameters [1] and a modified TUG with 7m of walking was recommended to assess spatiotemporal parameters during iTUG [2]. However a distance of 2.4m to 3m was found sufficient to measure gait speed [3].

Regarding the hip ROM of flexion-extension, this parameters was selected as it is commonly used in THA literature [4]. We agree that dividing the ROM in flexion and extension would be an interesting to understand patient’s limitation but that it would be more appropriate for a specific study focused on gait.

Regarding thorax and pelvis velocities, iTUG studies sometimes use angular/linear velocities of the pelvis [5] (or waist [6]), and sometime velocities of the thorax [2]. We chose to take both in the analysis and to identify the most relevant in this specific study with PCA.

1. Sangeux M, Passmore E, Graham HK, Tirosh O. The gait standard deviation, a single measure of kinematic variability. Gait Posture. 2016;46: 194–200. doi:10.1016/j.gaitpost.2016.03.015

2. Salarian A, Horak FB, Zampieri C, Carlson-kuhta P, Nutt JG, Aminian K. iTUG , a Sensitive and Reliable Measure of Mobility. IEEE Trans neural Syst Rehabil Eng. 2010;18: 303–310.

3. Stuck AK, Bachmann M, Füllemann P, Josephson KR, Stuck AE. Effect of testing procedures on gait speed measurement: A systematic review. PLoS One. 2020;15. doi:10.1371/journal.pone.0234200

4. Bahl JS, Nelson MJ, Taylor M, Solomon LB, Arnold JB, Thewlis D. Biomechanical changes and recovery of gait function after total hip arthroplasty for osteoarthritis : a systematic review and meta-analysis. Osteoarthr Cartil. 2018;26: 847–863. doi:10.1016/j.joca.2018.02.897

5. Mellone S, Tacconi C, Chiari L. Validity of a Smartphone-based instrumented Timed Up and Go. Gait Posture. 2012;36: 163–165. doi:10.1016/j.gaitpost.2012.02.006

6. Zakaria NA, Kuwae Y, Tamura T, Minato K, Kanaya S. Quantitative analysis of fall risk using TUG test. Comput Methods Biomech Biomed Engin. 2015;18: 426–437. doi:10.1080/10255842.2013.805211

2. PCA:

It is not described if both groups were included in the PCA analysis.

-The description of the PCA was modified as follow to give more details:

The duration of the phases are the most commonly reported parameters in iTUG studies. We chose beforehand to keep them in the analysis to be able to compare our results to the literature. Thus, the PCA guided only the selection of biomechanical features.

For the quality during walking, the selected feature was “Range obliquity thorax (0.30)” and not “RMS obliquity thorax (0.27)”. The bold font is a typo in the Supplementary file. The “range of flexion contralateral hip” does have a higher cos2 than the range of obliquity of the thorax but it is on the same PC as the “range of flexion of pathological hip” which is why it was excluded. In the first version of the manuscript we selected one feature per component apart for the turn-to-sit phase that regroups two movements. However, as suggested by reviewer 1, we removed the peak vertical velocity of the thorax for consistency in the method.

After checking the PCA results the feature “Distance chair to start of turn” of the turn to sit phase was added to the analysis to have a consistent selection of parameters.

- Unpaired Student’s t-tests were added to table 1 to assess the differences between groups. However, we do not understand why the test for the weight should be non-parametric. Could the reviewer develop on this point?

4. Number of tests and p-value correction:

I would recommend controlling the FDR with the Benjamini-Hochberg correction or the Holm method.

- The Holm method was used instead of the Bonferroni correction and all data updated. As expected, this correction included more significant differences (see Table 4):

- Difference in range of thorax obliquity during walking between M0 and M6

- Difference in duration of turn between M0 and M6

- Difference in range of thorax obliquity during turn to sit between M0 and CG

At phase level, we obtained the following modifications:

- Walking had a significantly larger deficit than the three other phases at M0

- Walking had a significantly larger deficit than the S2S and turn phases at M6

These new results strengthen the conclusion of the first version of our analysis, i.e. walking seems to present the largest deficit before and 6 months after surgery for patients undergoing THA.

To further strengthen the study effect sizes with CI could be calculated.

Effect sizes and 95% Confidence Interval were added to Table 4.

5. Overall deficit in percent:

Overall score should include all important variables (already identified by the PCA) and differences tested afterwards.

- From our point of view, if the feature was not significantly different, there was no deficit/difference to investigate which is why we only took into account the significant features. If there was no significant variable, then there would be no percentage of deficit/difference. However, this method will increase the average deficit of the considered phase.

The point of view of the reviewer seems to be more robust from a methodological standpoint. We implemented these recommendations in the paper and modified the Statistics section of the method as follow:

L 179: “Finally, the percentages of the features selected with PCA were averaged for each phase to have one global figure per phase.”

Further minor revisions:

1. Further remarks:

• “52 aged-matched controls” if they were age-matched, there would not be a 3y difference in age.

- The mention “aged-matched” was removed from the manuscript.

• “pregnancy or breast-feeding patients” relevant in this age group?

- It is not relevant for the patients of this study. However, during the recruitment of patients, the range of age was between 25 and 85 years old which made this criterion relevant. For this new version of the manuscript we mentioned the age at the start of the exclusion criteria:

L 110: “Exclusion criteria included patients younger than 25 years old or older than 85 years old”

If the reviewer prefers, we can remove the mention of “pregnancy or breast-feeding patients” from the manuscript.

• “The patients and control groups did not differ in sex, […]” as pointed out earlier, the K-S test does only test distribution differences and gives no insight about mean group differences.

- An unpaired Student’s t-test was added to the analysis and did not show significant differences.

• Table 4: Please revise/double-check. SD of Mean angular velocity pelvis (deg/s) (turn) and Peak angular velocity thorax (deg/s) (turn2sit) at M6 appear wrong.

- The consistency of the numbers in the text, figures and tables were checked an updated when needed.

• Use the same number of digits for a variable mean and SD.

- The manuscript was changed accordingly.

- The consistency was checked and results were modified accordingly.

- The manuscripts now refers only to the Holm method.

• “[…] selected by PCA and the scores of the patients […]” implies that PCA scores were compared, which is not the case.

- This was not the case indeed. The “score” here relates to the features and not to the values of the PCA. This seems to be a formulation issue, we modified the sentence:

L 260: “multiple features were selected by PCA and finally, the features of the patients were compared …”

- We changed “limp” to “compensatory movements” to be more general.

• The section “Implementation in clinical settings” might be well-meant, however, has too little relation to the results.

- This study was a first step toward the development of clinical tools for the measurements of function. Thus, we believe that this section can be of interest to provide a broader context and expected applications of the study to the reader. However, if the reviewer thinks this is section should be left out we can remove it from the paper.

- This part was moved to the section “Implementation in clinical settings”.

• Last paragraph of Limitations: “Surprisingly, the PCA identified parameters…” is not a limitation and belongs in the discussion.

- This paragraph was moved to the section “Variability of outcomes” where the PCA is discussed.

2. Variable names:

The variables should be named more precisely.

“Lateral distance between feet”. Is that step width? There is no lateral distance between the feet.

- This parameter is the distance between the midpoint of the 2 markers of the right and left foot in the lateral direction of the lab reference frame when the patient is sitting in the chair. It was changed to “Width base of support”. A thorough description of the features was added in Supporting Information S4.

- This variable takes into account hip flexion and extension. It was modified throughout the manuscript.

3. Language and typos

Language and wording are not quite yet on a publication level and should get revised by a native speaker or expert.

Examples for awkward phrasing, wording, grammar, and typos:

• “Exclusion criteria included a previous surgery to the hip” � previous hip surgery.

- The manuscript was changed accordingly.

• “hospital length of stay after surgery” � hospitalization?

- This term is commonly used in medical journals, it represents the time from the day of admission to the day of discharge. See reference [12] in the manuscript: Petis et al. (2016).

- Petis SM, Howard JL, Lanting BA, Somerville LE, Vasarhelyi EM. Perioperative Predictors of Length of Stay After Total Hip Arthroplasty. J Arthroplasty. 2016;31: 1427–1430. doi:10.1016/j.arth.2016.01.005

• “interfere or be destabilised by gait analysis” � interfere with or be worsened by …

- The manuscript was changed accordingly.

• “(i.e. above the mean of the control group minus one standard deviation)” � within 1SD of the control group

- The manuscript was changed accordingly.

• “Evolution” is the wrong word to describe the change/difference between pre/post

- We modified “evolution” by “change” throughout the manuscript.

• “pace” unusual in this context. Speed is probably more appropriate.

- We changed “pace” by “speed” as suggested.

• “TUG test was not performed previously for patients with end stage” � previously in patients.

- The manuscript was changed accordingly.

• “up-an-go” � up-and-go.

- The manuscript was changed accordingly.

Attachment

Submitted filename: Response to reviewers.docx

Click here for additional data file.^{(48.2KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0255037.r003

Decision Letter 1

Peter Andreas Federolf

5 May 2021

PONE-D-20-38214R1

Which functional tasks present the largest deficits for patients with total hip arthroplasty before and 6 months after surgery? A study of the Timed Up-and-Go phases

PLOS ONE

Dear Dr. Gasparutto,

Please submit your revised manuscript by Jun 19 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Peter Andreas Federolf

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: No

Reviewer #2: No

**********

6. Review Comments to the Author

Reviewer #1: The manuscript was greatly improved and the performed analysis is much clearer now. Even though there is so much information and data presented in the manuscript that it’s not always easy to follow the presented data. The description of the statistical methods still needs some minor improvements to make it clearer to the reader what tests were performed. There are also still some grammatical errors (use of singular/plural) and the manuscript should be checked in that regard again.

Abstract:

The performed test should be named “timed up-and-go test” in the abstract and manuscript text and not only “timed up-and-go”. (lines 24, 30 and 136)

Line 26: It should say: “biomechanical analysis of its phases was previously used…”

Introduction:

Line 68: This should be changed to “people diagnosed with hip OA”. Hip arthroplasty is not a diagnosis.

Methods

Using a t test to test for differences in sex distribution is not very common. I think Chi-square test would be more appropriate to check whether the patient and control group had different distribution of sex.

Line 178: Which correlation coefficient was calculated? Pearson? It appears that only significant correlations were reported in the results. What was the level of significance for the correlation coefficients?

Line 179: Do you mean “the percentage differences of the features selected by PCA”?

Line 180-182: It’s not entirely clear what you tested here. The sentence implies that you used paired t tests to test for differences between patients and controls which was not the case. This should be clarified. Maybe something like this would be clearer (if this is indeed the intended meaning): “Paired Student’s t tests were performed to assess whether the average differences with respect to the control group or between M0 and M6 differed significantly between the different phases of the TUG .”

Results:

Line 232: It is unclear what is meant with “other significant features”. Do you mean “features with significant differences”?

Line 235: Please change to “with the Holm method.”

Line 248: correlation of deficit in total and peak walking speed not presented in table 6 (it would be logical but peak walking speed is never presented throughout the paper).

Line 249: “the step numbers” should be changed to “number of steps”.

Discussion

Line 284: “time of the walking phase” should be changed to “duration of the walking phase”

Lines 281-290: It’s confusing that you’re talking about correlations between total time and features but actually mean the correlations between the change in total time and change in features (which is what the title implies and what is presented in the results). For a better understanding for the readers it should be specified in the text as well.

Lines 321-324: It is unclear what the authors want to say with that. PCA selects features with the highest variations but that does not mean that there need to be significant differences between groups. So, it’s no surprise to me that some of the selected features are not significantly different between groups. The second point about the measurement accuracy is important but should be discussed in a different context (i.e. are the statistically significant differences also clinically meaningful).

Line 363: there is a comma too much “…and change, but a categorical feature…”

General points on grammar and use of abbreviations. The list might not be complete so it should be checked again

Singular/plural of nouns are not always used correctly:

o Line 158: the features were…

o Line 301: the mean deficit was…

Verbs are not always in the correct form (3rd person singular -> is/was/suggests…; 3rd person plural -> are/were/suggest)

o Line 252: should be “the change… was associated…”

o Line 264: “the walking task represents the main limitation”

o Line 285/286: “total time could be good indicator… but does not seem…”

o Line 301: “these high levels of SD suggest…”

o Line 316: “the reduction of pain… leads…”

Introduced abbreviations are not consistently used or later abbreviated or never introduced

o OA: line 86, line 332, line 349

o TUG: line 367, line 372

o THA: line 73, line 296, line 366

o SD: abbreviation should be introduced at first use (line 108), but then used in text (i.e. lines 159, 175)

o S2S (line 172) is never explained in text (only in figures) and otherwise not used in abbreviated form in the text. For better readability I’d suggest writing it out here (sit to stand?) and not use the abbreviation

o MDC in line 340 is not introduced

Reviewer #2: Thank you for implementing the changes. Following are responses and further remarks.

Authors:

“The spatiotemporal parameters of gait were not selected due to the very short distance of walking that was below or equal to 3m. The target for turn was at 3m from the table, but, when considering that some participants started turning before the target and some started walking while standing up, the distance of walking was often below 3m. Previous studies showed that 5 steps are required to have stable variability of gait parameters [1] and a modified TUG with 7m of walking was recommended to assess spatiotemporal parameters during iTUG [2]. However a distance of 2.4m to 3m was found sufficient to measure gait speed [3]. Regarding the hip ROM of flexion-extension, this parameters was selected as it is commonly used in THA literature [4]. We agree that dividing the ROM in flexion and extension would be an interesting to understand patient’s limitation but that it would be more appropriate for a specific study focused on gait. Regarding thorax and pelvis velocities, iTUG studies sometimes use angular/linear velocities of the pelvis [5] (or waist [6]), and sometime velocities of the thorax [2]. We chose to take both in the analysis and to identify the most relevant in this specific study with PCA.”

Well, many TUG studies investigate the spatiotemporal parameters and, in my opinion, it makes sense. For example, the number of steps, step time, single support length, step length etc. can be very informative in terms of patient agility and potential limping.

How where the kinematics calculated for the different phases? Was it the mean of all steps within the phase for each side or both sides? Furthermore, only the fastest trial was chosen. Why not the mean of the 10 trials? Especially, as you mention the study of Sangeux [1], who did not really investigate the variance of consecutive steps but rather number of trials. They recommend 6 trials for kinematics. Based thereon, the results of the current study would be questionable as well, given that only one TUG trial was analyzed. Given that the gait kinematics most likely deviate from normal around the phase transitions.

[2] did not report differences in spatiotemporal parameters between the TUG (3m) and iTUG (7m), despite the time. A recommendation was not given. Only the conclusion that spatial parameters are less reliable due to the method (accelerometer/gyroscope).

[4] reported ROM probably out of convenience from max. flexion and extension. The included studies, such as, Klausmeier V. (2010) and Mayr E. (2009) report the flexion/extension separately, which makes a lot of sense.

[5] and [6] are accelerometer/smartphone studies and the velocity site is most likely defined by the limited application possibilities.

In conclusion, the variable choice remains inconclusive, and a good rationale is missing. On the one hand, the authors identify the most relevant variables using the PCA but on the other hand, cherry pick and limit their pre-selection to variables which clearly stem and are defined by the convention/limitations of accelerometer or IMU systems.

Two approaches would be conclusive:

1. Use ALL variables previously used in TUG tests to be able to compare results to accelerometer based iTUGs and to identify the most relevant.

2. Use evidence-based variables, which were previously identified to be relevant for THA patients.

Authors: “- Unpaired Student’s t-tests were added to table 1 to assess the differences between groups. However, we do not understand why the test for the weight should be nonparametric. Could the reviewer develop on this point?”

The K-S test is used to test the assumption of normal distribution a t-test. A significant result in the K-S test indicates a non-normal distribution and a violation of the t-test assumption. In this case a non-parametric test is warranted.

Commonly the scientific community assumes that the researchers know and use the appropriate assumption tests for their statistical analysis and usually do not ask to report the assumption test results. The K-S test results can be omitted in the table, but it should be indicated that a non-parametric test is used to compare the weight.

Authors: “If the reviewer prefers, we can remove the mention of “pregnancy or breast-feeding patients” from the manuscript.”

If the authors think it is relevant, it can remain in the manuscript.

Reviewer:

• The section “Implementation in clinical settings” might be well-meant, however, has too little relation to the results.

Authors:

Reviewer: Since it was not subject of the current investigation, I would exclude this part.

Reviewer:

• “hospital length of stay after surgery” hospitalization?

Authors:

- This term is commonly used in medical journals, it represents the time from the day of admission to the day of discharge. See reference [12] in the manuscript: Petis et al. (2016).

-Petis SM, Howard JL, Lanting BA, Somerville LE, Vasarhelyi EM. Perioperative Predictors of Length of Stay After Total Hip Arthroplasty. J Arthroplasty. 2016;31: 1427–1430. doi:10.1016/j.arth.2016.01.005

Reviewer: It remains a clumsy wording and weakens the readability but since it is minor issue the authors may choose to stick to their wording.

Further minor corrections:

Methods:

“The level of statistical significance was controlled with the Holm method (� = 0.05)”

Better: To account for multiple testing the Holm’s method was used to adjust the p-values.

L119: Student t-tests change to Student’s t-test

P17L235: “although not statistically significant with the Holm.”

Was it significant different without the Holm correction? Either it is significant or not. Either the authors acknowledge the method or not.

P17L235: a similar increase.

P18L269ff: “This feature had the lowest number of patients at the level of the control groups before and 6 months after surgery (8% and 27% respectively).”

Hard to comprehend. Please rephrase.

P19L281: with a strong

P19L285ff.: “The total time could be a good indicator of speed parameters but do not seem to reflect as well the quality of movement such as the presence of compensatory movements (e.g. high range of thorax obliquity).”

Check for errors, maybe rephrase.

P20L305: “help to identify patients at risks of lowering their level of function after surgery” Appears wrong: implies that patients actively lower their function but it’s something that happens to them.

P20L307: “high variability observed with the analysis of the TUG”…the variability is a result of the analysis and not an observation of the analysis. Please revise the sentence/wording.

P22L348ff.: “The patient population and control group had a significant difference in term of weight, but a higher weight was reported as a risk factor for hip osteoarthritis [43] which could explain this difference.”

Is not really a limitation, at least not how it is stated. Weight might be a limitation in terms of different kinematics or measurement inaccuracies due to marker placement or skin artifacts.

P22L360ff:

“The feature “distance chair to start of turn” was chosen with the assumption that a longer distance would imply that the patients turned and sat at the same time, i.e. with an overlapping transition strategy. This feature was continuous, as this study aimed at assessing percentages of differences and change, but, a categorical feature classifying the strategy of the patients could be of interest to understand the patient’s function.”

Why is this a limitation?

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

PLoS One. 2021 Sep 10;16(9):e0255037. doi: 10.1371/journal.pone.0255037.r004

Author response to Decision Letter 1

28 Jun 2021

Reviewer #1

The manuscript was greatly improved and the performed analysis is much clearer now. Even though there is so much information and data presented in the manuscript that it’s not always easy to follow the presented data. The description of the statistical methods still needs some minor improvements to make it clearer to the reader what tests were performed. There are also still some grammatical errors (use of singular/plural) and the manuscript should be checked in that regard again.

- The paper was proofread by a native English speaker: Darren Hart of “publish-or-perish.ch”, a professional proofreader.

Abstract:

The performed test should be named “timed up-and-go test” in the abstract and manuscript text and not only “timed up-and-go”. (lines 24, 30 and 136)

- The manuscript was modified accordingly.

Line 26: It should say: “biomechanical analysis of its phases was previously used…”

- The manuscript was modified accordingly.

Introduction:

Line 68: This should be changed to “people diagnosed with hip OA”. Hip arthroplasty is not a diagnosis.

- The manuscript was modified accordingly.

Methods

- The t-test was changed to a Chi-square test, there is still no significant differences.

- Yes, we used Pearson correlations and reported only the significant correlations. The Holm method was used to adjust the level of significance. The manuscript was modified as follow:

L177: “Pearson correlations were calculated between the total TUG time and selected features for percentage differences to the control group and the percentage of change between M0 and M6. The level of significance was adjusted using the Holm method (� = 0.05) [39]. Only significant correlations were reported.”

Line 179: Do you mean “the percentage differences of the features selected by PCA”?

- The manuscript was modified for clarity. See previous comment.

- Thank you, your sentence is clearer. The manuscript was modified as suggested.

Results:

Line 232: It is unclear what is meant with “other significant features”. Do you mean “features with significant differences”?

- Yes, this is what we meant. The manuscript was modified as suggested.

Line 235: Please change to “with the Holm method.”

- The manuscript was modified accordingly.

Line 248: correlation of deficit in total and peak walking speed not presented in table 6 (it would be logical but peak walking speed is never presented throughout the paper).

- Indeed, in this study, the feature corresponding to the peak walking speed is the peak forward velocity of the pelvis. This was changed in the manuscript.

L247: “The deficits and changes in total TUG time were mainly associated with speed parameters, especially with the deficits and changes in the duration of the walking phase and with the peak forward velocity of the pelvis (Table 6).”

Line 249: “the step numbers” should be changed to “number of steps”.

- The manuscript was modified accordingly.

Discussion

Line 284: “time of the walking phase” should be changed to “duration of the walking phase”

- The manuscript was modified accordingly.

- The title and section were modified as follow:

L 245:“ Associations between deficits and changes in features and deficits and changes in total TUG times (Table 6)

- The PCA discussion was removed and the point about measurement accuracy was added in the limitation section:

L 354: “The duration of test phases and the time differences between groups were relatively small, especially for the sit-to-stand and turning phases. Thus, those results should be taken with care as the difference could be within measurement accuracy of the algorithm. Moreover, data centring and reduction led to the selection of parameters with non-clinically meaningful differences. Indeed, the PCA identified parameters with differences that were within the range of measurement accuracy (1 degree), e.g. the peak thorax flexion and range of thorax obliquity during the sit-to-stand phase. Future studies could use measurement accuracy as a criterion during the feature selection.”

Line 363: there is a comma too much “…and change, but a categorical feature…”

- The manuscript was modified accordingly.

General points on grammar and use of abbreviations. The list might not be complete so it should be checked again

Singular/plural of nouns are not always used correctly:

o Line 158: the features were…

- The manuscript was modified accordingly.

o Line 301: the mean deficit was…

- The manuscript was modified accordingly.

Verbs are not always in the correct form (3rd person singular -> is/was/suggests…; 3rd person plural -> are/were/suggest)

o Line 252: should be “the change… was associated…”

- The manuscript was modified accordingly.

o Line 264: “the walking task represents the main limitation”

- The manuscript was modified accordingly.

o Line 285/286: “total time could be good indicator… but does not seem…”

- The manuscript was modified accordingly.

o Line 301: “these high levels of SD suggest…”

- The manuscript was modified accordingly.

o Line 316: “the reduction of pain… leads…”

- The manuscript was modified accordingly.

Introduced abbreviations are not consistently used or later abbreviated or never introduced

o OA: line 86, line 332, line 349

- The manuscript was modified accordingly.

o TUG: line 367, line 372

- The manuscript was modified accordingly.

o THA: line 73, line 296, line 366

- The manuscript was modified accordingly.

o SD: abbreviation should be introduced at first use (line 108), but then used in text (i.e. lines 159, 175)

- The manuscript was modified accordingly.

o MDC in line 340 is not introduced

- MDC stands for minimal detectable change. The abbreviation was removed from the manuscript.

-----------------------------------------------------------------------------

Reviewer #2:

Thank you for implementing the changes. Following are responses and further remarks.

Authors:

[5] and [6] are accelerometer/smartphone studies and the velocity site is most likely defined by the limited application possibilities.

Two approaches would be conclusive:

1. Use ALL variables previously used in TUG tests to be able to compare results to accelerometer based iTUGs and to identify the most relevant.

2. Use evidence-based variables, which were previously identified to be relevant for THA patients.

- We respectfully disagree with reviewer 2 on this matter, which was interestingly not raised by reviewer 1. As stated in our revised manuscript, we have added a supplementary material describing specifically the choice and computation of the features (S4). The approach suggested by the reviewer (“taking ALL variables previously used in TUG test”) does not seem to be realistic or relevant for the present paper since, as an example, parameters used to assess the upper limbs of patients with Parkinson disease are not applicable for patients with total hip arthroplasty. Moreover, this suggestion would require a specific study dedicated to the choice of variables through a systematic review of literature, which is not within the scope of the present paper. The second approach suggested is to use “evidence-based variables […] relevant for THA patients”. We believe that this is in accordance with the approach we followed, since all the features have specific references describing their previous use in the literature (S4). Since our paper is the first iTUG study with THA patients, not all references are related to THA patients but we chose mainly studies with cohorts of elderly patients which will compare somewhat with our own cohort. We would appreciate your opinion on this issue. We can add more references to S4 but it does not seem realistic for us to start this study from scratch at this point.

- It seems that there was a misunderstanding with the KS-test. The values reported in Table 1 are two-sample KS test used to compare the groups. When testing for normal distribution with the one-sample KS test, the parameters of both groups are normal, thus, the t-test is appropriate in all cases. Moreover, as recommended by reviewer 1, a Chi-square test was used to check whether the patient and control group had different distribution of sex. The KS-test column was removed from the table and the manuscript modified as follow:

L154: “The difference in sex distribution was tested using a χ2 test. The Kolmogorov–Smirnoff [23] test was used to assess the normality of the age, weight and height distributions among patients and controls. Differences between groups were then tested using unpaired Student’s t-tests.”

Authors: “If the reviewer prefers, we can remove the mention of “pregnancy or breast-feeding patients” from the manuscript.”

If the authors think it is relevant, it can remain in the manuscript.

- This mention was kept in the manuscript.

Reviewer:

• The section “Implementation in clinical settings” might be well-meant, however, has too little relation to the results.

Authors:

Reviewer: Since it was not subject of the current investigation, I would exclude this part.

- We would like to insist on this point which was not raised by reviewer 1. This sections seems relevant to us as it provides a broader context to the present study and the direction of future research on this topic. Could the editor decide on whether we should include this section?

Reviewer:

• “hospital length of stay after surgery” hospitalization?

Authors:

- This term is commonly used in medical journals, it represents the time from the day of admission to the day of discharge. See reference [12] in the manuscript: Petis et al. (2016).

-Petis SM, Howard JL, Lanting BA, Somerville LE, Vasarhelyi EM. Perioperative Predictors of Length of Stay After Total Hip Arthroplasty. J Arthroplasty. 2016;31: 1427–1430. doi:10.1016/j.arth.2016.01.005

Reviewer: It remains a clumsy wording and weakens the readability but since it is minor issue the authors may choose to stick to their wording.

- The term “hospital length of stay” has been kept in the manuscript.

Further minor corrections:

- The paper was proofread by a native English speaker: Darren Hart of “publish-or-perish.ch”, a professional proofreader. Your suggestion were implemented before the final proofreading.

Methods:

“The level of statistical significance was controlled with the Holm method (alpha = 0.05)”

Better: To account for multiple testing the Holm’s method was used to adjust the p-values.

- Thank you, this sentence was modified accordingly.

L119: Student t-tests change to Student’s t-test.

- The manuscript was modified accordingly.

P17L235: “although not statistically significant with the Holm.”

Was it significant different without the Holm correction? Either it is significant or not. Either the authors acknowledge the method or not.

- This mention was removed from the manuscript:

L233: “The walking phase showed the greatest mean improvement in function (16%, 95%CI: 8% to 24%, Table 5), although this was not statistically significant.”

P17L235: a similar increase.

- The manuscript was modified accordingly.

P18L269ff: “This feature had the lowest number of patients at the level of the control groups before and 6 months after surgery (8% and 27% respectively).”

Hard to comprehend. Please rephrase.

- The manuscript was modified as follow:

L 273: “Indeed, this feature presented the lowest percentage of patients with a level of function similar to that of the control group, both before and after surgery (8% and 27%, respectively), as well as the greatest number of patients with a positive change in function after surgery (82%).”

P19L281: with a strong

- The manuscript was modified accordingly.

Check for errors, maybe rephrase.

- The manuscript was modified as follow:

L 289: “Total TUG test time could be a good indicator of speed parameters, but it does not seem to reflect comparably the quality of movements, such as the presence of compensatory movements (e.g. high range of thorax obliquity).”

- The manuscript was modified as follow:

L 308: “Understanding the preoperative factors at the origin of those differences in profile could help to identify the patients at risk of having a lower level of function after surgery and improve patient rehabilitation.”

P20L307: “high variability observed with the analysis of the TUG”…the variability is a result of the analysis and not an observation of the analysis. Please revise the sentence/wording.

- The manuscript was modified as follow:

L 311: “The high variability resulting from the analysis of the TUG test’s phases suggests that it could be used to identify patient-specific levels of function.”

Is not really a limitation, at least not how it is stated. Weight might be a limitation in terms of different kinematics or measurement inaccuracies due to marker placement or skin artifacts.

- From our point of view, the ideal control group would be healthy patients matched in age, sex and weight to have the least differences in term of population characteristics. Cimolin et al. (2019) showed a decreased time of the TUG test for severely obese women (BMI of 41.1 kg/m2) when compared to healthy controls (BMI of 22.8 kg/m2). The difference in BMI is nothing comparable in our study (BMI of 28.8 kg/m2 and 25.2 kg/m2 for patients and controls respectively) but it could be a source of differences in the features and thus, a limitation of the present study.

P22L360ff:

Why is this a limitation?

- The limitation here is the choice of the feature. It underlines that different features could have been selected to assess the iTUG. Here we gave an example of the fact that two features can represent the same phenomenon: the two different strategies to perform the turn-to-sit task.

PLoS One. doi: 10.1371/journal.pone.0255037.r005

Decision Letter 2

Peter Andreas Federolf

9 Jul 2021

Which functional tasks present the largest deficits for patients with total hip arthroplasty before and six months after surgery? A study of the Timed Up-and-Go test phases

PONE-D-20-38214R2

Dear Dr. Gasparutto,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Peter Andreas Federolf

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: The authors have addressed all my comments. Some of the proof reader’s corrections resulted in uncommon wordings, especially regarding the use of the possessive ‘s (although probably grammatically correct). They’re very minor remarks, so the authors can choose to go back to their original wordings or stay with the current version.

Some examples:

Line 84: The study’s objective (“the objective of the study” is more common)

Line 119: Marker trajectories (instead of markers’ trajectories) is commonly used

Line 121: using a fourth-order Butterwort filter (instead of design)

Line 197: each phase’s duration (“duration of each phase” is more common)

Line 293: in all the test’s phases (“all the phases of the test” is more common)

Reviewer #2: Thank you for implimenting the changes. Even if I disagree in some cases the study adds to the outcome evaluation of THA even if two important planes of motion were neglected.

The paper can be published.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

PLoS One. doi: 10.1371/journal.pone.0255037.r006

Acceptance letter

Peter Andreas Federolf

2 Sep 2021

PONE-D-20-38214R2

Which functional tasks present the largest deficits for patients with total hip arthroplasty before and six months after surgery? A study of the Timed Up-and-Go test phases

Dear Dr. Gasparutto:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Peter Andreas Federolf

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Fig. Figures to check the phases of patients.

(PDF)

Click here for additional data file.^{(1.3MB, pdf)}

S2 Fig. Figures to check the phases of the control group.

(PDF)

Click here for additional data file.^{(488.9KB, pdf)}

S3 Fig. Comparison of patient’s features and times of phases with the level of the control group.

(PDF)

Click here for additional data file.^{(10MB, pdf)}

S1 Table. Results of the principal component analysis.

(PDF)

Click here for additional data file.^{(288.1KB, pdf)}

S1 File. Algorithm for the detection of the timed up-and-go phases.

(PDF)

Click here for additional data file.^{(169.2KB, pdf)}

S2 File. Features description and selection.

(PDF)

Click here for additional data file.^{(225.4KB, pdf)}

Attachment

Submitted filename: Response to reviewers.docx

Click here for additional data file.^{(48.2KB, docx)}

Data Availability Statement

[pone.0255037.ref001] 1.Liu X, Zi Y, Xiang L, Wang Y. Total hip arthroplasty: a review of advances, advantages and limitations. International journal of clinical and experimental medecine. 2015;8: 27–36. [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref002] 2.Vissers MM, Bussmann JB, Verhaar JAN, Arends LR, Furlan AD, Reijman M. Recovery of Physical Functioning After Total Hip Arthroplasty: Systematic Review and Meta-Analysis of the Literature. Physical Therapy. 2011;91: 615–629. doi: 10.2522/ptj.20100201 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref003] 3.Bahl JS, Nelson MJ, Taylor M, Solomon LB, Arnold JB, Thewlis D. Biomechanical changes and recovery of gait function after total hip arthroplasty for osteoarthritis: a systematic review and meta-analysis. Osteoarthritis and Cartilage. 2018;26: 847–863. doi: 10.1016/j.joca.2018.02.897 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref004] 4.World Health Organization. How to use the ICF: A practical manual for using the International Classification of Functioning, Disability and Health (ICF). Geneva; 2013.

[pone.0255037.ref005] 5.Lovelock TM, Broughton NS, Williams CM. The Popularity of Outcome Measures for Hip and Knee Arthroplasties. Journal of Arthroplasty. 2018;33: 273–276. doi: 10.1016/j.arth.2017.08.024 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref006] 6.Nilsdotter AK, Lohmander LS, Klässbo M, Roos EM. Hip disability and osteoarthritis outcome score (HOOS)–validity and responsiveness in total hip replacement. BMC Musculoskeletal Disorders. 2003;4: 10. doi: 10.1186/1471-2474-4-10 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref007] 7.Dayton MR, Judd DL, Hogan CA, Stevens-Lapsley JE. Performance-Based Versus Self-Reported Outcomes Using the Hip Disability and Osteoarthritis Outcome Score after Total Hip Arthroplasty. American Journal of Physical Medicine and Rehabilitation. 2016;95: 132–138. doi: 10.1097/PHM.0000000000000357 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref008] 8.Gandhi R, Tsvetkov D, Davey JR, Syed KA, Mahomed NN. Relationship between self-reported and performance-based tests in a hip and knee joint replacement population. Clinical Rheumatology. 2009;28: 253–257. doi: 10.1007/s10067-008-1021-y [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref009] 9.Podsiadlo D, Richardson S. The Timed “Up & Go”: A Test of Basic Functional Mobility for Frail Elderly Persons. Journal of the American Geriatrics Society. 1991;39: 142–148. doi: 10.1111/j.1532-5415.1991.tb01616.x [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref010] 10.Dobson F, Bennell K, Hinman RS, Abbott JH, Roos EM. Recommended performance-based tests to assess physical function in people diagnosed with hip or knee osteoarthritis. Melbourne; 2013. doi: 10.1016/j.joca.2013.05.002 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref011] 11.Caronni A, Sterpi I, Antoniotti P, Aristidou E, Nicolaci F, Picardi M, et al. Criterion validity of the instrumented Timed Up and Go test: A partial least square regression study. Gait & Posture. 2018;61: 287–293. doi: 10.1016/j.gaitpost.2018.01.015 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref012] 12.Petis SM, Howard JL, Lanting BA, Somerville LE, Vasarhelyi EM. Perioperative Predictors of Length of Stay After Total Hip Arthroplasty. The Journal of Arthroplasty. 2016;31: 1427–1430. doi: 10.1016/j.arth.2016.01.005 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref013] 13.Nankaku M, Tsuboyama T, Akiyama H, Kakinoki R, Fujita Y, Nishimura J, et al. Preoperative Prediction of Ambulatory Status at 6 Months After Total Hip Arthroplasty. Physical Therapy. 2012;93: 1–6. doi: 10.2522/ptj.20120016 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref014] 14.Sasaki K, Senda M, Nishida K, Ota H. Preoperative time required for the timed “up and go” test in women with hip osteoarthritis could predict a deep venous thrombosis complication after total hip arthroplasty. Acta medica Okayama. 2010;64: 197–201. doi: 10.18926/AMO/40012 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref015] 15.Caronni A, Picardi M, Aristidou E, Antoniotti P, Pintavalle G, Redaelli V, et al. How do patients improve their timed up and go test? Responsiveness to rehabilitation of the TUG test in elderly neurological patients. Gait & Posture. 2019;70: 33–38. doi: 10.1016/j.gaitpost.2019.02.010 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref016] 16.Coulthard JT, Treen TT, Oates AR, Lanovaz JL. Evaluation of an inertial sensor system for analysis of timed-up-and-go under dual-task demands. Gait & Posture. 2015;41: 882–887. doi: 10.1016/j.gaitpost.2015.03.009 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref017] 17.Cimolin V, Cau N, Malchiodi G, Aspesi V, Merenda V, Galli M, et al. Do wearable sensors add meaningful information to the Timed Up and Go test? A study on obese women. Journal of Electromyography and Kinesiology. 2019;44: 78–85. doi: 10.1016/j.jelekin.2018.12.001 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref018] 18.Salarian A, Horak FB, Zampieri C, Carlson-Kuhta P, Nutt JG, Aminian K. iTUG, a Sensitive and Reliable Measure of Mobility. IEEE Transactions on Neural Systems and Rehabilitation Engineering. 2010;18: 303–310. doi: 10.1109/TNSRE.2010.2047606 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref019] 19.Zampieri C, Salarian A, Carlson-kuhta P, Aminian K, Nutt JG, Horak FB. The instrumented timed up and go test: potential outcome measure for disease modifying therapies in Parkinson’s disease. Journal of Neurology, Neurosurgery, and Psychiatry. 2010;81: 171–176. doi: 10.1136/jnnp.2009.173740 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref020] 20.Hansen BJ, Hallows RK, Kelley SS. The Rottinger approach for total hip arthroplasty: Technique and review of the literature. Current Reviews in Musculoskeletal Medicine. 2011;4: 132–138. doi: 10.1007/s12178-011-9093-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref021] 21.Moore AT. The self-locking metal hip prosthesis. Journal of Bone and Joint Surgery. 1957;39: 811–827. [PubMed] [Google Scholar]

[pone.0255037.ref022] 22.Altman R, Alarcón G, Appelrouth D, Bloch D, Borenstein D, Brandt K, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hip. Arthritis & Rheumatism. 1991;34: 505–514. doi: 10.1002/art.1780340502 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref023] 23.Baker R, Leboeuf F, Reay J, Sangeux M. The Conventional Gait Model—Success and Limitations. In: Müller B, Wolf SI, editors. Handbook of Human Motion. Springer International Publishing AG; 2017. pp. 1–19. doi: [DOI] [Google Scholar]

[pone.0255037.ref024] 24.Gløersen Ø, Federolf P. Predicting Missing Marker Trajectories in Human Motion Data Using Marker Intercorrelations. PloS one. 2016. doi: 10.1371/journal.pone.0152616 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref025] 25.Beyea J, Mcgibbon CA, Sexton A, Noble J, Connell CO. Convergent Validity of a Wearable Sensor System for Measuring Sub-Task Performance during the Timed Up-And-Go Test. Sensors. 2017;17: 1–18. doi: 10.3390/s17040934 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref026] 26.Barre A, Armand S. Biomechanical ToolKit: Open-source framework to visualize and process biomechanical data. Computer Methods and Programs in Biomedicine. 2014;114: 80–87. doi: 10.1016/j.cmpb.2014.01.012 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref027] 27.Janssen WGM, Bussmann HBJ, Stam HJ. Determinants of the sit-to-stand movement: A review. Physical Therapy. 2002;82: 866–879. doi: 10.1093/ptj/82.9.866 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref028] 28.Abujaber SB, Marmon AR, Pozzi F, Rubano JJ, Zeni JA. Sit-To-Stand Biomechanics Before and After Total Hip Arthroplasty. Journal of Arthroplasty. 2015;30: 2027–2033. doi: 10.1016/j.arth.2015.05.024 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref029] 29.Aissaoui R, Dansereau J. Biomechanical analysis and modelling of sit to stand task: a literature review. Proceedings of the IEEE International Conference on Systems, Man and Cybernetics. 1999;1: 141–146. doi: 10.1109/icsmc.1999.814072 [DOI]

[pone.0255037.ref030] 30.Alexander NB, Schultz AB, Warwich DN. Rising from a chair: Effects of age functional ability on performance biomechanics. Journals of Gerontology. 1991;46: 91–98. doi: 10.1093/geronj/46.3.m91 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref031] 31.Nankaku M, Tsuboyama T, Kakinoki R, Kawanabe K, Kanzaki H, Mito Y, et al. Gait analysis of patients in early stages after total hip arthroplasty: Effect of lateral trunk displacement on walking efficiency. Journal of Orthopaedic Science. 2007;12: 550–554. doi: 10.1007/s00776-007-1178-2 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref032] 32.Hollands KL, Hollands MA, Zietz D, Miles Wing A, Wright C, Van Vliet P. Kinematics of turning 180° during the timed up and go in stroke survivors with and without falls history. Neurorehabilitation and Neural Repair. 2010;24: 358–367. doi: 10.1177/1545968309348508 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref033] 33.Schot PK, Knutzen KM, Poole SM, Mrotek LA. Sit-to-stand performance of older adults following strength training. Research Quarterly for Exercise and Sport. 2003;74: 1–8. doi: 10.1080/02701367.2003.10609058 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref034] 34.Boonstra MC, Schreurs BW, Verdonschot N. The Sit-to-Stand Movement: Differences in Performance Between Patients After Primary Total Hip Arthroplasty and Revision Total Hip Arthroplasty With Acetabular Bone Impaction Grafting. Physical Therapy. 2011;91: 547–554. doi: 10.2522/ptj.20090376 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref035] 35.Weiss A, Mirelman A, Giladi N, Barnes LL, Bennett DA, Buchman AS, et al. Transition between the Timed Up and Go turn to sit subtasks: is timing everything? Journal of the American Medical Directors Association. 2016;17. doi: 10.1016/j.jamda.2015.07.005 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref036] 36.Baker R, McGinley JL, Schwartz MH, Beynon S, Rozumalski A, Graham HK, et al. The Gait Profile Score and Movement Analysis Profile. Gait & Posture. 2009;30: 265–269. doi: 10.1016/j.gaitpost.2009.05.020 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref037] 37.Massey FJ. The Kolmogorov-Smirnov Test for Goodness of Fit. Journal of the American Statistical Association. 1951;46: 68–78. doi: 10.1080/01621459.1951.10500769 [DOI] [Google Scholar]

[pone.0255037.ref038] 38.Cattell RB. The Scree Test For The Number Of Factors. Multivariate Behavioral Research. 1966;1: 245–276. doi: 10.1207/s15327906mbr0102_10 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref039] 39.Holm S. A Simple Sequentially Rejective Multiple Test Procedure. Scandinavian Journal of Statistics. 1979;6: 65–70. doi: 10.2307/4615733 [DOI] [Google Scholar]

[pone.0255037.ref040] 40.Mokkink LB, Terwee CB, Patrick DL, Alonso J, Stratford PW, Knol DL, et al. The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: An international Delphi study. Quality of Life Research. 2010;19: 539–549. doi: 10.1007/s11136-010-9606-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref041] 41.Smith E, Cunningham C, Greene BR, Mccarthy U, Catherine P. Detecting subtle mobility changes among older adults: the Quantitative Timed Up and Go test. Aging Clinical and Experimental Research. 2020. doi: 10.1007/s40520-020-01733-7 [DOI] [PubMed] [Google Scholar]

[pone.0255037.ref042] 42.Rylander JH. CORR Insights®: Gait Analysis Reveals that Total Hip Arthroplasty Increases Power Production in the Hip During Level Walking and Stair Climbing. Clinical orthopaedics and related research. 2019;477: 1848–1850. doi: 10.1097/CORR.0000000000000865 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0255037.ref043] 43.Aresti N, Kassam J, Nicholas N, Achan P. Hip osteoarthritis. BMJ (Online). 2016;354: 1–10. doi: 10.1136/bmj.i3405 [DOI] [PubMed] [Google Scholar]

PERMALINK

Which functional tasks present the largest deficits for patients with total hip arthroplasty before and six months after surgery? A study of the timed up-and-go test phases

Xavier Gasparutto

Mathieu Gueugnon

Davy Laroche

Pierre Martz

Didier Hannouche

Stéphane Armand

Roles

Abstract

Introduction

Material & methods

Participants

Table 1. Participants’ characteristics.

Measurement protocol

Fig 1. The four phases of the timed-up and go test.

Analysis

Table 2. Biomechanical features for each phase.

Statistics

Results

Principal component analysis

Table 3. Features selected using principal component analysis.

Fig 2. Overview of each phase’s duration and the features selected using PCA for all three groups: Patients before surgery (PM0), patients six months after surgery (PM6) and Control Group (CG).

Table 4. Mean values (standard deviation) for patients before surgery (M0), patients six months after surgery (M6) and the control group.

Patients’ functional deficits compared to the control group

Deficits before surgery

Table 5. Mean difference ± SD (95% confidence interval) between groups by phase expressed in percentages.

Table 6. Mean, SD and 95% confidence interval of the significant differences between groups, i.e. differences between patient and control groups and between patients before and six months after surgery.

Residual deficits six months after surgery

Change in function six months after surgery

Total TUG times

Associations between deficits and changes in features and deficits and changes in total TUG times (Table 6)

Discussion

Patients’ functional deficits compared to the control group

Changes in function six months after surgery

Changes and deficits in total TUG time compared to changes and deficits in functional outcomes

Outcome variability

Implementation in clinical settings

Limitations

Conclusion

Supporting information

Data Availability

Funding Statement

References

Decision Letter 0

Peter Andreas Federolf

Roles

Author response to Decision Letter 0

Decision Letter 1

Peter Andreas Federolf

Roles

Author response to Decision Letter 1

Decision Letter 2

Peter Andreas Federolf

Roles

Acceptance letter

Peter Andreas Federolf

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases