Trunk alignment in different standing positions in healthy subjects and stroke patients -a comparative study with a simple method for the everyday practice.: Trunk alignment in healthy and stroke subjects

ABSTRACT

Objectives: Weight-bearing (WB) on the lower extremities is an important outcome parameter in the rehabilitation of poststroke hemiparesis. However, the patients often regain this ability by compensatory movement patterns.

Methods: Our goal was to characterize with a simple method the trunk alignment of healthy subjects and stroke patients (n = 17 for both groups) during standing and following lateral weight shift (WS). To describe trunk alignment, five markers were placed on the subjects’ back, and the angles of the trunk at both sides were defined by the lines drawn from the posterior angle of the acromion and the iliac crest on the same body side to the seventh thoracic spinal process. Weight distributions on the lower extremities during standing and lateral WS were determined with a force platform.

Results: The patients had significantly limited WB capacity on their paretic limb, which was accompanied with significant asymmetry in the trunk alignment during standing and following WS to the paretic side.

Discussion: Our results show that this patient population tends to use abnormal compensatory movement patterns to optimize weight shifting, and changes of trunk alignment play a key role in this. This should be taken into consideration during rehabilitation.

KEYWORDS:

• Trunk alignment
• hemiparesis
• weight distribution
• rehabilitation
• stroke
• asymmetry

1. Introduction

Controlled trunk stability and mobility including weight shifting are essential components of balance for everyday activities.^1,2 Even slight impairments of these functions may lead to functional asymmetry.³ The optimal alignment of the trunk during quiet standing (as starting position: SP) and following lateral weight shift (WS) requires normal sensory and motor processing and proper interaction between the musculoskeletal and nervous systems.^4,5 Postural impairments are common in stroke patients, and there is a correlation between asymmetry in weight-bearing (WB) capacity and the severity of stroke.^1,2,6,7 The majority of individuals having undergone stroke bear less weight on the paretic limb in standing position,^8–11 which is due to the alterations in muscle tone and strength,¹² the impaired sense of trunk position,¹³ poor postural control,¹⁴ and/or perceptual problems.^15,16 These impairments significantly interfere with balance and everyday activities;¹⁷ therefore, physiotherapists focus not only on improving postural control but also on avoiding inappropriate compensatory strategies.^13,18 Most studies determine functional mobility or investigate asymmetry in relation to the weight distribution on the lower extremities in stroke patients.^8,9,19–22 Sophisticated methods have been applied to underscore the importance of trunk alignment during SP and WS in healthy subjects. It has been shown that trunk displacement has a very important role in postural control.^23–26 However, according to our knowledge, there isn’t a sound rationale regarding a standardized postural compensation that occurs during weight shifting in the healthy controls, although therapists are often emphasizing that there should be an elongation on the weight-bearing side (WBS) of the trunk while practicing WS. Furthermore, the exact angles and their changes during these body positions have not been investigated either in healthy subjects or in patients with movement disorders. Therefore, our aim was to characterize trunk alignment with a simple and reproducible method during quiet standing and following voluntary lateral WS in the frontal plane in healthy subjects and patients with stroke. The result of this study might provide a clearer picture about the postural compensation to widen the knowledge how the trunk alignment is changing following lateral, voluntary WS in order to specify the treatment interventions. Our hypothesis was that in healthy subjects following lateral WS in standing, the trunk alignment would change, i.e. the WBS would be elongated, and there would be a proportional shortening on the non-weight-bearing side (NWBS) compared to the SP and that this would be altered in stroke patients.

2. Methods

2.1. Participants

Seventeen subjects with hemiparesis due to a single cerebrovascular accident between 3 and 45 months previously (right/left: 10/7, male/female: 9/8, age: 59 [SEM: 2.9] years, weight: 78 [2.7] kg, height: 167 [1.8] cm) and 17 healthy controls (male/female: 7/10, age: 60 [2.5] years, weight: 75 [4.0] kg, height: 168 [2.5] cm) participated in the study. The anthropometric parameters of the two groups were comparable. Regarding the patients’ physical function, it was not analyzed in a large detail, but the patients were eligible for the study if they were able to meet the following conditions: (1) at least 3 months had passed since their cerebrovascular event, (2) if they had hemiparesis, (3) if they were able to stand independently, and (4) if they were free of any musculoskeletal or neurological disorder other than the cerebrovascular event. The control subjects were selected (1) according to their age (age-matched participants) and (2) if they were free of any musculoskeletal or neurological disorder. Participation was voluntary. All of the subjects gave their informed consent prior to participation in the study, which was approved by the local Institutional Ethics Committee and conformed to the Declaration of Helsinki in all respects. The manuscript conforms to the STROBE guidelines.

2.2. Markers and angles determination

To enable the description of the trunk alignment during the tasks, five markers (circular patches of 4 cm diameter, white with black point in the middle) were placed on the subjects’ back (Figure 1). The placement was partially based on earlier studies,^25,27,28 as follows:

Figure 1. Presentation of marker setup.

Marker 1 and 2: bilaterally and symmetrically on the posterior angle of the acromion.

Marker 3: on the spinal process of the seventh thoracic vertebra (the apex of kyphosis).

Marker 4 and 5: bilaterally and symmetrically at the level of the iliac crests.

The angles of the trunk (acromion–vertebra–iliac crest angles; AVIAs) on both sides in the different positions were determined with the use of angular dimension tool of Corel Draw (Corel Corp., Canada).

2.3. Ground contact force determination

All subjects were tested on a force platform (ZWE-PII Stabilometer; 50 × 50 cm; Elektro-Bionika LTD, Budapest, Hungary) to determine the contact forces (as a measure of WB capability) of both legs. The measurements were preceded by verbal instruction and demonstration of the actual task. Initially, the subjects were instructed to stand on the platform with their feet 10 cm apart with their arms by their side and to gaze at the wall 3 m in front of them.^21,29 Then, the subjects were asked to stand putting equal body weight on their lower extremities (SP). Next, the subjects were instructed to shift as much weight as possible onto the right leg without lifting off the left foot (WS position), then return to SP. The next task was to shift as much weight as possible onto the left leg without lifting off the right foot (WS position). This sequence was repeated 3 times, and every position was held for 3 s. All of the subjects could perform the tasks.

In the WS position, the two sides were designated as WBS and NWBS. The WBS was the loaded side while the NWBS was the unloaded side. Since no significant differences were found between the three trials, the means of the WS values were analyzed. Regarding SP, the mean of the first three measurements was used for further analyses. Ground contact forces were determined for each leg in all positions and expressed as percentage of body weight (relative ground force: RGF). The movement of the patients was also recorded with a video camera, and the positions of the markers were analyzed offline.

2.4. Data and statistical analyses

Asymmetry indices (AIs) were introduced based on a formula of symmetry index.³⁰ These indices provide measures of RGFs (Asymmetry Indices of Weight bearing: AIW) and AVIAs (Asymmetry Indices of Angles: AIA) on both sides (see below). The data underlying the study are available at the corresponding author.

2.4.1. Data collection and analysis for SP

Two categories were defined according to RGF: the more loaded (MS) and the less loaded (LS) sides for both groups. In the control group, all the subjects had more weight on MS than on the LS. However, since all of the patients had more load on their paretic side compared to the non-paretic side, therefore the MS was the non-paretic side, while the LS was the paretic side in all cases.

The asymmetry index of WB in starting position (AIW-SP) was determined as follows:

$\begin{array}{rl}& AIW-SP=\left(\left[RG{F}_{MS}-RG{F}_{LS}\right]\ast 100\right)/\\ & \left(0.5\left[RG{F}_{MS}+RG{F}_{LS}\right]\right)\end{array}$

An asymmetry index for AVIAs in starting position (AIA-SP) was also calculated:

$\begin{array}{rl}& AIA-SP=\left(\left[AVI{A}_{MS}-AVI{A}_{LS}\right]\ast 100\right)/\\ & \left(0.5\left[AVI{A}_{MS}+AVI{A}_{LS}\right]\right)\end{array}$

For both indices, 0 indicates perfect symmetry; a positive score indicates a higher value on the MS side, while a negative score indicates a lower value on the MS side.

2.4.2. Data collection and analysis for WS

Four categories were introduced for WS. Two are based on the side of WB (WBS vs. NWBS) and the categories determined during SP (LS vs. MS):

In the case of WS to the LS: WBS-LS and NWBS-MS

In the case of WS to the MS: WBS-MS and NWBS-LS

Regarding the WB ability, AIs of RGF were introduced for both WS positions (WS to the LS and MS):

$\begin{array}{rl}& AIW-WS-LS=\left(\left[RG{F}_{WBS-LS}-RG{F}_{NWBS-MS}\right]\ast 100\right)/\\ & \left(0.5\left[RG{F}_{WBS-LS}+RG{F}_{NWBS-MS}\right]\right),\mathrm{a}\mathrm{n}\mathrm{d}\end{array}$

$\begin{array}{rl}& AIW-WS-MS=\left(\left[RG{F}_{WBS-MS}-RG{F}_{NWBS-LS}\right]\ast 100\right)/\\ & \left(0.5\left[RG{F}_{WBS-MS}+RG{F}_{NWBS-LS}\right]\right),\end{array}$

Since RGF was always higher on the WBS compared to the NWBS, these values were positive. Regarding the analysis of the trunk angles in WS position, AIs were introduced as follows¹:

$\begin{array}{rl}& AIA\mathrm{\_}WS\mathrm{\_}LS=\left(\left[AVI{A}_{WBS-LS}-AVI{A}_{NWBS-MS}\right]\ast 100\right)/\\ & \left(0.5\left[AVI{A}_{WBS-LS}+AVI{A}_{NWBS-MS}\right]\right),\mathrm{a}\mathrm{n}\mathrm{d}\end{array}$

$\begin{array}{rl}& AIA-WS-MS=\left(\left[AVI{A}_{WBS-MS}-AVI{A}_{NWBS-LS}\right]\ast 100\right)/\\ & \left(0.5\left[AVI{A}_{WBS-MS}+AVI{A}_{NWBS-LS}\right]\right),\end{array}$

where positive scores represent greater angle on the WBS, while “0” represents perfect symmetry.

To determine the degree of trunk elongation and shortening on both the WBS and NWBS and using SP as the baseline, the elongation index (EI) was introduced²:

$EI-WBS/NWBS=\left(AVI{A}_{WBS/NWBS}-AVI{A}_{SP}\right)$

A positive value means elongation, and a negative value means shortening on the given side.

Data are presented as mean ± SEM. Since the data sets were not normally distributed, the nonparametric Mann–Whitney U test was used to compare the two groups, and the Wilcoxon matched pairs signed-rank test was applied to perform the within-subject comparisons. To characterize correlations between the different parameters, Spearman’s rho was calculated. Statistical analyses were performed in STATISTICA for Windows 12.0 (Statistica Inc., Tulsa, Oklahoma, USA). A P value lower than 0.05 was considered significant.

3. Results

3.1. Analysis of the SP

Regarding WB, both groups exhibited significant asymmetry between the two sides, but it was more pronounced in the patient group (Figure 2A); thus, the AIW-SP value also indicated significant difference between the two groups (signed with * in Table 1).

Table 1. The mean (95% confidence intervals) of asymmetry indices of weight distribution (AIW) and trunk angles (AIA) in healthy subjects and stroke patients.

	AIW		AIA
SP
Control	7.1 ± (4.5–9.6)		−2.4 ± (−6.6–1.9)
Patients	25.6 ± (4.2–16.7)*		4.5 ± (0.2–8.7)*
WS	MS	LS	MS	LS
Control	141 ± (125.5–157.3)	125 ± (105.8–144.4)^#	−1.7 ± (−5.32–1.83)	4.7 ± (−1.63–10.93)
Patients	125 ± (106.8–143.2)	82 ± (62.7–103.2)*^#	−0.5 ± (−6.17–5.21)	−7.6 ± (−13.31 to −1.99)*

SP: Starting position; WS: (lateral) weight shift; MS: more loaded side; LS: less loaded side.

* and ^# denote significant differences (p < 0.05) compared to control group and MS, respectively.

Figure 2A. Relative ground force (RGF)in starting position (SP) in patients and controls.

*P < 0.05 compared to control group. ^#P < 0.05 compared to contralateral side.

Figure 2B. Relative ground force (RGF) in weight shift position in patients and controls.

*P < 0.05 compared to control group. ^#P < 0.05 compared to contralateral side.

As for the AVIAs, no significant differences were found between the two groups on either side; however, the patient group was characterized by significantly smaller AVIAs on the LS than on the MS (P < 0.05) (signed with ^# in Figure 3A). Therefore, the AIA-SP also showed significant difference between the two groups (signed with * in Table 1).

Figure 3. Trunk angles (AVIAs) on both sides in starting and weight shift positions in patients and controls. *P < 0.05 compared to control group. ^#P < 0.05 compared to contralateral side (MS: more loaded side; LS: less loaded side WS: weight shift).

3.2. Analysis of WS position

Regarding the WS position, while the control subjects shifted slightly less weight on their LS than on the MS, the patients shifted significantly less weight on the LS than on the MS. (Figure 2B)Thus, both groups were characterized by less asymmetry in the WS position when shifting the weight to LS than to MS, but the degree of the asymmetry was significantly lower in the patients when shifting weight to LS than in the control subjects (Table 1).

Regarding the changes of AVIAs in WS position on the MS, there were no significant differences between the WBS and NWBS and between the two groups (Figure 3B). On the contrary, in the case of WS on the LS, the stroke patients showed a perfectly opposite trend in the angles at the two sides compared to the control group, i.e. the AVIA values were significantly higher on the NWBS than the WBS (Figure 3C). Therefore, the AI for the angles in WS to the MS did not reveal significant asymmetries in either group, while in WS to the LS, showed an opposite pattern between the patients (decrease) and controls (increase); therefore, significant differences were observed between the two groups (Table 1). Regarding the EIs of the trunk only, patients showed significant shortening on the WBS and elongation on the NWBS in WS to the LS (Figure 4).

Figure 4. Elongation indices on both sides in starting and weight shift positions in patients and controls.*P < 0.05 compared to control group. ^#P < 0.05 compared to contralateral side (MS: more loaded side/LS: less loaded side; WBS: weight-bearing side; non-weight-bearing side).

As shown in Figure 5, the difference between a patient and a control subject is well observable.

Figure 5. Changes of trunk alignment during weight shift in a healthy control subject (top) and a stroke patient (bottom). SP: The starting position (quiet standing); L: shift to the left; R: shift to the right.

3.3. Correlation analyses

No significant correlations were found between the different RGFs in either group. As for the correlation between AVIAs and RGFs, only a few significant correlations were found (Table 2). The correlation analysis of AIs and EIs, however, revealed significant negative correlation between the WB asymmetry in SP and in WS, and significant positive correlation between WB asymmetry in WS and trunk angles asymmetry in WS in the patient group when shifting the weight to the LS (Table 3). Furthermore, significant correlations were observed between trunk angles asymmetry in SP and in WS in both groups and on both sides. Since the EIs and AIs were calculated from the AVIAs, in most cases, there were significant correlations between these indices.

Table 2. Correlations between relative ground force (RGF) and acromion–vertebra–iliac crest angle (AVIA) values.

	RGF-SP			RGF-WS		AIW-SP
	MS	LS		MS	LS	MS	LS
RGF-WS
Control	0.14		0.30
Patients	−0.36		0.43
AVIA-SP
Control	−0.10		−0.16	−0.03	0.44
Patients	0.53*		−0.04	−0.32	0.09
AVIA-WS-WBS
Control	0.02		−0.21	−0.10	0.32
Patients	0.45		0.11	−0.17	0.29
AVIA-WS-NWBS
Control	−0.09		0.55*	0.20	0.11
Patients	−0.17		−0.44	−0.09	−069*
AIW-WS
Control						−0.07		−0.35
Patients						−0.22		−0.52*

AIW: Asymmetry Indices of Weight bearing; SP: starting position; WS: (lateral) weight shift; MS: more loaded side; LS: less loaded side; WBS: weight-bearing side; NWBS: non-weight-bearing side.

*Significant degree of correlation.

Table 3. Correlations between the asymmetry/elongation indices.

	AIW-SP		AIW-WS		AIA-SP		AIA-WS		EI-WBS
	MS	LS	MS	LS	MS	LS	MS	LS	MS	LS
AIW-WS
Control	−0.07	−0.35
Patients	−0.22	−0.52*
AIA-SP
Control	−0.33		−0.26	−0.19
Patients	0.36		−0.17	−0.42
AIA-WS
Control	−0.19	0.25	−0.25	0.12	0.70*	−0.89*
Patients	0.30	−0.43	−0.08	0.64*	0.80*	−0.79*
EI-WBS
Control	0.20	0.12	−0.11	−0.02	−0.53*	−0.50*	0.13	0.81*
Patients	−0.009	−0.25	0.23	0.35	0.11	−0.15	0.62*	0.66*
EI-NWBS
Control	0.27	−0.27	0.20	−0.19	−0.90*	0.95*	−0.91*	−0.97*	0.30	−0.66*
Patients	−0.36	0.42	0.20	−0.63*	−0.93*	0.90*	−0.94*	−0.95*	−0.33	−0.39

AIW: Asymmetry Indices of Weight bearing; AIA: Asymmetry Indices of Angles; EI: Elongation Indices; SP: starting position; WS: weight shift; WBS: weight-bearing side; NWBS: non-weight-bearing side.

*Significant degree of correlation.

4. Discussion

Our study revealed major differences in trunk alignment between healthy subjects and stroke patients. It was symmetrical in healthy subjects during quiet standing, while, in contrast with our hypothesis, only nonsignificant shortening was detected in this group on the NWBS in WS position to the LS. In patients, the larger trunk angles on the NWBS compared to the WBS in WS to the LS revealed an altered (i.e. compensatory) strategy of trunk movement. The few correlations between trunk angles and RGFs suggest that trunk angles do not necessarily reflect WS ability only but also the compensatory strategy. The significant correlations between angles asymmetry in SP and WS in both groups and on both sides show that subjects with a higher degree of asymmetry during SP will have a higher degree of asymmetry also in WS. The strong correlations between EIs and AIAs suggest that EIs might be used as parameters for the characterization of trunk alignment in different WS positions.

During SP, the expected body weight on a single limb would be 50% of the total body weight, and the control group showed only slight RGF asymmetry. Regarding the WS ability in our healthy group (82–86%), it was lower than in an earlier study (95% in individuals between 58 and 70 years of age),³¹ which might be due to the involvement of older subjects in our study. The significant inverse correlation (R = −0.75) between the age and WS percentages in the healthy group suggests slightly impaired postural stability in the older subjects.^3,8,32 Regarding the patient group, in agreement with earlier studies,^1,8,9,32,33 the non-paretic side had greater WB ability compared to the paretic one during SP; thus, the degree of asymmetry in the RGF was significantly higher than in the control group (Table 1).

The alterations of trunk alignment in SP and in different WS positions indicate that the stroke patients were not able to elongate the affected side; therefore, they performed WS in a compensatory manner. Surprisingly, asymmetry in WB and trunk angles (AIW and AIA) correlated significantly (R = 0.64) only in patients shifting weight to the paretic side (LS), suggesting that asymmetry in WB is not necessarily linked to the asymmetries of trunk alignment. Since WS to the paretic side is a challenging task for these patients, it is assumed that the abnormal elongation/shortening strategy might compensate for the impaired WB ability and may be applied as a protective strategy against falling.

Stroke patients are unable to perform satisfactory extension of the lower extremity to support the body weight leading to passive standing.¹ This passive WB position together with lateral flexion of the trunk to the WBS can help the patients to perform the required task. The weak abductors and adductors on the affected side and/or the deficient interaction or coordination between these muscles can also contribute to the observed impairments.³⁴ Furthermore, the overactivity of the unaffected side and the stiffness of the paretic side can constrain WS toward either side, which the patients may try to compensate by trunk movements.^21,34,35

A major focus of rehabilitation in hemiparesis is to increase the patients’ ability to bear weight on the paretic limb.^5,7,9,21,22 However, so far, no particular attention has been paid to such patients’ trunk alignment in different standing positions. Our results suggest that a quantitative analysis of WB in itself is not enough to build up a correct treatment plan, since trunk alignment may also contribute to the movement performance. With more information about the pathological postural alignment associated with stroke, the therapist is better equipped to determine the state of the patients and to establish appropriate interventions. For instance, in patients with shortening on the WB side, attention must be paid to the correction of this alignment, that is elongation should be facilitated on the loaded side.

The study had several limitations. One important limitation might be that the results are based on a relatively small sample of stroke patients. This fact has to be taken into consideration when interpreting such findings as moderate correlations between WB and the trunk angles. Another limitation is that we focused only on the description of WB and trunk alignment in these groups, but in this study, we did not evaluate the subjects’ physical functions (motor control, coordination, strength, walking speed, and stroke severity). Thus, these results cannot be generalized to the entire population of stroke patients with hemiparesis. Furthermore, we did not evaluate the risk of falling and its correlation to trunk alignment and weight distribution in this study; e.g. if the hemiparetic limb utilization leads to increased or decreased fall and if patients with moderate impairments can have full WB on the affected side. The amount of WB can be different in stroke patients with severe, moderate, or slight impairment. Future studies should include the measurements of physical functions and the risk of falling. Clearly, studies with larger samples are needed with the abovementioned, extended measurements and training paradigms.

5. Conclusions

To our knowledge, this study is the first attempt to give an exact characterization of trunk alignment during quiet standing and lateral WS in healthy and stroke patients. Obviously, this study was not designed to substantiate ultimate conclusions; its aim was rather to point out the practical importance of trunk alignment. The results revealed that the stroke patients’ WB ability is deficient and that they tend to compensate for this deficiency with abnormal trunk alignment. Therefore, the measurement of WB ability in itself is an unsatisfactory proxy of therapeutic success, and attention must also be paid to how the individual patient performs WS, and incorporate this information in the therapeutic plan. Our simple method can help to reach that end by allowing the therapist to numerically describe patients’ trunk alignment in an easier way.

Disclosure of interest

The authors report no conflicts of interest.

Acknowledgments

The authors wish to thank Csilla Keresztes and Janos Wodala for the linguistic correction of the manuscript.