Brain, Vol. 126, No. 3, 724-731,
March 2003
© 2003 Guarantors of Brain
doi: 10.1093/brain/awg066
Stroke patients have selective muscle weakness in shortened range
School of Physiotherapy, University of Sydney, Sydney, Australia
Correspondence to: Louise Ada, School of Physiotherapy, Faculty of Health Sciences, University of Sydney, PO Box 170, Lidcombe, NSW 1825, Australia E-mail: l.ada{at}cchs.usyd.edu.au
Received June 3, 2002. Revised September 26, 2002. Second revision October 15, 2002. Accepted October 16, 2002.
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Weakness is recognized as a major problem after stroke. This study examined the torque–angle curves of stroke individuals and compared them with those of neurologically normal controls to determine (i) if stroke patients were selectively weak when their muscles were placed in a shortened range and (ii) whether contracture influenced any selective weakness. This descriptive research study measured elbow flexor and extensor torque–angle curves and contracture. Twenty-two stroke subjects who had suffered a stroke 5 months to 6 years ago and 11 neurologically normal controls of similar age participated. Torque–angle curves of the elbow flexors and extensors were determined by measuring maximum isometric torque at 0, 20, 40, 60, 80, 100 and 120° of elbow flexion (0° being full elbow extension), where possible. Contracture of the elbow flexors and extensors was measured as the loss of passive elbow joint range of motion. Repeated measures analysis of variance revealed that the torque–angle curves of stroke subjects (with or without contracture) were significantly different from those of the control subjects for both the elbow flexors (P < 0.05) and extensors (P < 0.05). The stroke subjects appeared relatively weaker when the muscles were in their shortened range. This study confirms that selective weakness exists at short muscle lengths after stroke. The findings of this study help to explain why people after stroke have difficulty functioning when their muscles are in their shortened range. Therefore, strength training should be targeted specifically at muscles at their shortened lengths in order to promote the recovery of function after stroke.
Keywords: torque–angle curve; strength; stroke; weakness; contracture
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Impairments following brain damage can be classified as positive (e.g. spasticity and abnormal cutaneous reflexes) or negative (e.g. loss of strength and loss of dexterity). It is now recognized that, regardless of the presence of positive impairments, negative impairments remain the main obstacle to recovery of function following brain damage (Burke, 1988
). A common functional problem observed after stroke is that individuals are able to use their affected limbs reasonably well at certain joint angles, but not at others. For example, in the upper limb, it may be possible to use the affected arm in mid-range, but difficult when the affected arm is in the last 0–20° of elbow extension (Bohannon, 1991; Levin et al., 2000). In the lower limb, it is common to observe individuals after stroke having difficulty functioning when the affected knee is in the last 0–15° of extension (Carr and Shepherd, 1998). A recent study found that this clinical phenomenon in individuals after stroke could be attributed to loss of strength rather than loss of control (Ada et al., 2000).
It is known that force production in normal muscle is length-dependent. The exact shape of the curvilinear torque–angle curve primarily depends on physiological and biomechanical factors (for detailed review see Rassier et al., 1999). While it has been shown that the torque–angle curve is altered after training or chronic use of muscles at certain joint angles (Thepaut-Mathier et al., 1988; Kitai and Sale, 1989; Herzog et al., 1991; Weir et al., 1995), to date few studies have investigated the effect of neurological pathologies, such as stroke, on torque production across joint angles (Ada et al., 2000; Levin et al., 2000). The aim of this study, therefore, was to determine the effect of stroke on the elbow flexor and extensor torque–angle curves measured across the full range of motion. That is, we aimed to determine whether individuals after stroke present with selective weakness at particular joint angles. Our second aim was to determine whether contracture could contribute to any observed selective weakness. Overall, our aim was to produce findings that could be expected to guide rehabilitation.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Subjects
Individuals who had had one stroke resulting in hemiplegia were recruited for this study. The only exclusion criteria were severe cognitive or perceptual difficulties that would interfere with the comprehension of instructions. Twenty-two individuals (eight females and 14 males) who had suffered a stroke between 5 months and 6 years ago and who had a median age of 63 years (range 42–85 years) were accepted as subjects. Fifteen had had a stroke affecting their left side and seven had had a stroke affecting their right side. Scores on item 6 of the Motor Assessment Scale (Carr et al., 1985
) ranged from 1 to 6, i.e. no arm was paralysed and there was a wide range of functional ability in the affected arms. Scores on the Ashworth scale (Ashworth, 1964) ranged from 0 to 2, i.e. when hypertonia was present it was mild. In addition, 11 neurologically normal subjects (five females and six males) who had a median age of 61 years (range 52–76 years) were recruited to act as controls. The Human Ethics Committee, The University of Sydney and the Research Ethics Committee, South Western Sydney Area Health Service approved the experimental procedures and all subjects gave informed consent before data collection began.
Procedures
Subjects sat on a high-backed chair and a wide seat-belt held them firmly against the chair, preventing trunk movement (Fig. 1A). The affected forearm (stroke group) or the non-dominant forearm (control group) was firmly secured in a horizontal arm frame. The arm frame was attached to a height-adjustable table that allowed the arm to be placed at 90° of shoulder flexion. The proximal end of the arm frame was attached to the table by a movable joint containing a potentiometer so that the forearm could be rotated and elbow angle recorded. A load cell (Applied Measurement Australia Pty Ltd, Oakleigh, Victoria, Australia; rated capacity 250 N, linearity 0.03%) attached at the distal end of the arm frame recorded applied force. The product of the force (N) and the distance between the load cell and the elbow joint was calculated as elbow torque (Nm). Subjects faced a computer monitor on which elbow angle and torque signals could be displayed for feedback purposes.
|
Maximum voluntary isometric flexor and extensor torques were measured at up to seven angles (0, 20, 40, 60, 80, 100 and 120°) (Fig. 1B). Not all subjects’ elbow torques could be measured at all angles (Table 2), because of either loss of range due to structural changes or natural limitations of the range caused by the size of the trunk, so that the elbow could not be flexed to 120° without the apparatus contacting the trunk. First, passive elbow range was recorded. Maximum passive elbow extension was measured as the angle achieved when a torque of 1 Nm was applied to straighten the elbow while the subject relaxed. Similarly, maximum passive elbow flexion was measured as the angle achieved when a torque of 1 Nm was applied to bend the elbow. Flexor and extensor isometric strengths were then measured at the angles that were within each individual’s available range by the experimenter resisting a maximal voluntary contraction (Fig. 1A). Thepaut-Mathieu and colleagues found that isometric training in the lengthened position assisted a muscle to generate torque at short lengths (Thepaut-Mathieu et al., 1988
). Therefore, to exclude a possible order effect, testing was performed from short to long muscle lengths, i.e. from 120° to 0° for the flexors and from 0° to 120° for the extensors. The best of two 4 s maximum isometric contractions at each angle was recorded, first for the flexors and then for the extensors. Verbal encouragement and visual feedback were given during each trial to elicit a maximum contraction. An interval of at least 1 min was given between trials to prevent muscle fatigue.
|
Statistical analysis
Mean torque–angle curves were calculated for the control and stroke groups. To determine if there was any difference in the torque–angle curves between the two groups, repeated measures analysis of variance (ANOVA) was performed and the interaction between group and angle was examined. In addition, the position of the peak of the torque–angle curves was determined.
Immobilization of muscles in a shortened position results in a decrease in sarcomere number and an increase in connective tissue, i.e. contracture, which presents as a decrease in passive range and increase in stiffness. Subjects were assigned to a contracture group if they had loss of passive range in either flexion or extension. A contracture was operationally defined as >5° loss of elbow extension or >5° loss of elbow flexion. As expected, all the control subjects had no contracture, while 14 stroke subjects were assigned to a contracture stroke group and eight stroke subjects to a non-contracture stroke group. Table 1 shows characteristics of the stroke subjects in each group. Figure 2 presents data for three subjects: one control, one stroke without contracture and one stroke subject with contracture, measured within their available range. Mean torque–angle curves were calculated for the two stroke subgroups: non-contracture stroke and contracture stroke groups. ANOVA was performed to determine if there was any difference in the torque–angle curves between the control group and both the contracture stroke group and the control group, as well as between the non-contracture stroke group and the contracture stroke group. In addition, the position of the peak of the torque–angle curves was determined.
|
|
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Torque–angle curves of the elbow flexors and extensors for the control and the stroke subjects are represented in Fig. 3A and B. Group means, standard deviations and number of subjects measured at each angle are presented in Table 2. For both the flexors and the extensors, the stroke subjects appeared selectively weak in their shortened range. However, comparison of the torque–angle curves between the groups is difficult since they are confounded by weakness. Therefore, to be able to compare the torque–angle curves between subjects of varying strength, each subject’s data were normalized by expressing the individual data points as a percentage of his or her own peak maximum voluntary torque (Huijing et al., 1989
; van der Linden et al., 1998).
|
For both the flexors (Fig. 4A) and the extensors (Fig. 4B), the stroke subjects still appeared relatively weak in their shortened range. However, given that over half the stroke group had a contracture and were not able to be measured at 0° and/or 120° (Table 2), it remained difficult to compare the curves statistically. Therefore, to be able to compare the torque–angle curves between subjects with varying amounts of available range, each subject’s data were normalized so that percentages of maximum torque values were determined at 0, 25, 50, 75 and 100% of available range.
|
The graphs of the torque–angle curves of the data rescaled for peak maximum torque and available range for the flexors and the extensors are shown in Fig. 5. In the flexors, the torque–angle curve of the stroke group was significantly different from the torque–angle curve of the control group [F(4,124) = 6.8, P < 0.001] (Fig. 5A). The elbow flexors of the stroke group appeared relatively weaker than those of the control group at 0% of available range and relatively stronger at 75–100% of available range. Similarly in the extensors, the torque–angle curve of the stroke group was significantly different from the torque–angle curve of the control group [F(4,124) = 7.3, P < 0.001] (Fig. 5B). The elbow extensors of the stroke group appear relatively weaker than the control group at 0–50% of available range but not much different at 100% of available range. In the flexors, the peak of the curve occurs at 30% of available range for the controls and 45% for the stroke group. In the extensors, the peak occurs at 55% of available range for the controls and 85% for the stroke group. In both cases, the peak of the stroke group curves has moved towards a more lengthened position of the muscle.
|
The graphs of the torque–angle curves of the stroke group, subdivided into non-contracture and contracture groups and rescaled for peak maximum torque and available range for the flexors and the extensors, are shown in Fig. 5C and D. In the flexors (Fig. 5C), the torque–angle curves of the non-contracture stroke group [F(4,68) = 4.22, P = 0.004] and the contracture stroke group [F(4,92) = 7.32, P < 0.001] were both significantly different from the torque–angle curve of the control group but not from each other [F(4,80) = 1.73, P = 0.15]. The elbow flexors of both the stroke groups appeared relatively weaker than those of the control group at 0% of available range but only the contracture group appeared relatively stronger at 75–100% of available range. Similarly, in the extensors (Fig. 5D) the torque–angle curves of the non-contracture stroke group [F(4,68) = 5.72, P < 0.001] and the contracture stroke group [F(4,92) = 6.97, P < 0.001] were significantly different from those of the torque–angle curve of the control group but not from each other [F(4,80) = 0.97, P = 0.43]. The elbow extensors of both the stroke groups appeared relatively weaker than the control group at 0–25% of available range but not much different at 100% of available range. In the flexors, the peak of the curve occurred at 30% of available range for the controls and 45% for the non-contracture stroke group and 50% for the contracture stroke group. In the extensors, the peak occurred at 56% of available range for the controls and 75% for the non-contracture stroke group and 90% for the contracture stroke group.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The major finding of this study is that, even after adjustment for weakness and available range, both the flexor and extensor torque–angle curves of the stroke group were significantly different from those of the control group. Compared with the control curves, the curves for the stroke subjects had shifted so that the stroke subjects were relatively stronger when both the flexor and extensor muscles were in their lengthened range and relatively weaker when the muscles were in their shortened range. A similar pattern of selective weakness can be seen in the raw data of torque–angle curves of stroke patients in recent studies (Nadeau et al., 1997
; Levin et al., 2000). These findings provide concrete evidence of the clinical observation that stroke patients have particular difficulty in generating torque when muscles are in their shortened range, e.g. in straightening the elbow to reach forwards, or in fully extending the knee in order to weight-bear through the leg. Furthermore, it was found that selective weakness was present in both the contracture and the non-contracture stroke group and was not exaggerated in the contracture group. Therefore, it does not appear that contracture was a major contributing factor to selective weakness. Of course, contracture is a problem in its own right since it interferes with function. Not only does it decrease the available range in which to move but the increased stiffness of muscles pose an additional problem in that more force needs to be generated in the antagonist to produce movement within the available range.
The stroke subjects in this study were tested, on average, 2 years after their stroke. This has allowed time for adaptations, both neural and mechanical, to occur. Given that contracture did not appear to be a significant factor in the presence of selective weakness, the question of other possible causes is raised. Although the findings from this study cannot answer this question directly, it is interesting to speculate on the possible mechanism underlying this phenomenon. Generally, loss of strength in stroke individuals can be attributed to reduction in the number, firing frequency and/or recruitment order of motor units, and to characteristics of contractile properties and specific changes in nerve conduction velocity (for reviews see Bourbonnais and Vanden Noven, 1989; Ng and Shepherd, 2000). Since the mechanisms underlying loss of strength at different muscle lengths after stroke have not been investigated, we have reviewed the mechanisms behind the length-dependence of active force production in normal muscle (Rassier et al., 1999) to hypothesize on possible mechanisms after stroke. Neurologically normal subjects achieve maximal torque by recruiting all available motor units and all units fire at the rate necessary for fusion of twitches. While there is currently no information regarding the number of motor units recruited during maximal voluntary contractions at different muscle lengths, there is information about motor unit firing rates. During maximum voluntary contractions at short muscle lengths, there is a reduction in twitch duration. Therefore, in order to achieve fusion of twitches, neurologically normal subjects increase motor unit firing rates (Rack and Westbury, 1969; Gandevia and McKenzie, 1988; Bigland-Ritchie et al., 1992; Christova et al., 1998; Connelly et al., 1999). Since reduced motor unit firing rates have been observed after stroke (Rosenfalck and Andreassen, 1980; Dengler et al., 1990; Gemperline et al., 1995), it is possible that the selective weakness observed at short muscle lengths in the present study can be attributed to difficulty in producing the higher motor unit firing rates required to produce fusion of twitches, and therefore maximum torque, at short muscle lengths. Since this mechanism is primary rather than secondary, it could also account for the common clinical observation of selective weakness early after stroke, suggesting that the mechanism is not a long-term adaptation.
Clinical implications
Early after stroke, patients present to rehabilitation with significant muscle weakness. The results of this study show that muscles are selectively weak in their shorter lengths. This selective weakness needs to be addressed during rehabilitation. It has now been shown that intervention aimed at strengthening muscle results in improved function after stroke (e.g. Bütefisch et al., 1995; Sharp and Brouwer, 1997). Strengthening exercises targeted specifically at short muscle lengths may improve function even further by enhancing force production at these muscle lengths. For example, the quadriceps muscles can be exercised in their shortened range by doing isometric knee extensor exercises in long sitting followed by ascent and descent of small steps. In the upper limb, the elbow extensors can be exercised in their shortened range by straightening the elbow with the arm supported on a friction-free table followed by closing and opening a drawer.
The findings from this study may also help to resolve the question of whether the upper limb flexors or extensors are more affected following stroke. Clinical observation suggests that extensors are more affected than the flexors (Lance and McLeod, 1981; Ryerson and Levit, 1997), whereas experimental evidence suggests the opposite (Colebatch et al., 1986; Andrews and Bohannon, 2000). Colebatch and colleagues found that the ratio of elbow extensor/flexor strength was smaller for control subjects (55%) than for stroke subjects (71%) when strength was tested at 90° of elbow flexion (Colebatch et al., 1986). Therefore, it was concluded that the flexors were relatively more affected than the extensors. When strength was tested at 100° of elbow flexion in the present study, the findings were similar. However, at 20° of elbow flexion we found that the ratio of elbow extensor/flexor strength was greater for the control subjects (102%) than for the stroke subjects (61%), suggesting that the extensors were relatively more affected than the flexors. This means that the relative strength of the flexors versus the extensors after stroke is dependent upon the angle at which strength is tested. In addition, the clinical observation that the upper limb extensors are weaker than the flexors may be the result of the extensors being relatively more affected than the flexors in the part of range which is particularly important for function, i.e. near full elbow extension while reaching.
In conclusion, we have produced experimental evidence of selective weakness at short muscle lengths in patients after stroke. This evidence is consistent with the commonly observed clinical phenomenon of difficulty contracting affected muscles in their shortened range. The clinical implication of this finding is that muscle strengthening exercises, placing particular emphasis on the shortened range, may lead to an improved functional outcome after stroke.
|
Acknowledgements |
|---|
We wish to thank the stroke sufferers who contributed their time to this study and Professor Simon Gandevia for his advice on the original manuscript.
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Ada L, Canning C, Dwyer T. Effect of muscle length on strength and dexterity after stroke. Clin Rehabil 2000; 14: 55–61.
Andrews AW, Bohannon RW. Distribution of muscle strength impairments following stroke. Clin Rehabil 2000; 14: 79–87.
Ashworth B. Preliminary trial of carisoprodol in multiple sclerosis. Practitioner 1964; 192: 540–2.[ISI][Medline]
Bigland-Ritchie BR, Furbush FH, Gandevia SC, Thomas CK. Voluntary discharge frequencies of human motoneurons at different muscle lengths. Muscle Nerve 1992; 15: 130–7.[CrossRef][ISI][Medline]
Bohannon RW. Relationship between active range of motion deficits and muscle strength and tone at the elbow in patients with hemiparesis. Clin Rehabil 1991; 5: 219–24.
Bourbonnais D, Vanden Noven S. Weakness in patients with hemiparesis. [Review]. Am J Occup Ther 1989; 43: 313–9.[ISI][Medline]
Burke D. Spasticity as an adaptation to pyramidal tract injury. [Review]. Adv Neurol 1988; 47: 401–23.[Medline]
Bütefisch C, Hummelsheim H, Denzler P, Mauritz K-H. Repetitive training of isolated movements improves the outcome of motor rehabilitation of the centrally paretic hand. J Neurol Sci 1995; 130: 59–68.[CrossRef][ISI][Medline]
Carr JH, Shepherd RB. Neurological rehabilitation: optimizing motor performance. Oxford: Butterworth-Heinemann; 1998.
Carr JH, Shepherd RB, Nordholm L, Lynne D. Investigation of a new motor assessment scale for stroke patients. Phys Ther 1985; 65: 175–80.
Christova P, Kossev A, Radicheva N. Discharge rate of selected motor units in human biceps trachii at different muscle lengths. J Electromyogr Kinesiol 1998; 8: 287–9.[CrossRef][ISI][Medline]
Colebatch JG, Gandevia SC, Spira PJ. Voluntary muscle strength in hemiparesis: distribution of weakness at the elbow. J Neurol Neurosurg Psychiatry 1986; 49: 1019–24.[Abstract]
Connelly DM, Rice CL, Ross MR, Vandervoort AA. Motor unit firing rates and contractile properties in tibialis anterior of young and old men. J Appl Physiol 1999; 87: 843–52.
Dengler R, Konstanzer A, Hesse W, Wolf W, Struppler A. Abnormal behaviour of single motor units in central weakness. In: Berardelli A, Benecke R, Manfredi M, Marsden CD, editors. Motor disturbances II. London: Academic Press; 1990. p. 379–84.
Gandevia SC, McKenzie DK. Activation of human muscles at short muscle lengths during maximal static efforts. J Physiol (Lond) 1988; 407: 599–613.
Gemperline JJ, Allen S, Walk D, Rymer WZ. Characteristics of motor unit discharge in subjects with hemiparesis. Muscle Nerve 1995; 18: 1101–14.[CrossRef][ISI][Medline]
Herzog W, Guimaraes AC, Anton MG, Carter-Erdman KA. Moment–length relations of rectus femoris muscles of speed skaters/cyclists and runners. Med Sci Sports Exerc 1991; 23: 1289–96.[Medline]
Huijing PA, van Lookeren Campagne AA, Koper JF. Muscle architecture and fibre characteristics of rat gastrocnemius and semimembranosus muscles during isometric contractions. Acta Anat (Basel) 1989; 135: 46–52.[ISI][Medline]
Kitai TA, Sale DG. Specificity of joint angle in isometric training. Eur J Appl Physiol Occup Physiol 1989; 58: 744–8.[CrossRef][ISI][Medline]
Lance JW, McLeod JG. A physiological approach to clinical neurology. 3rd ed. London: Butterworth-Heinemann; 1981.
Levin MF, Selles RW, Verheul MH, Meijer OG. Deficits in the coordination of agonist and antagonist muscles in stroke patients: implications for normal motor control. Brain Res 2000; 853: 352–69.[CrossRef][ISI][Medline]
Nadeau S, Gravel D, Arsenault AB, Bourbonnais D, Goyette M. Dynamometric assessment of the plantarflexors in hemiparetic subjects: relations between muscular gait and clinical parameters. Scand J Rehabil Med 1997; 29: 137–46.[ISI][Medline]
Ng SSM, Shepherd RB. Weakness in patients with stroke: implications for strength training in neurorehabilitation. [Review]. Phys Ther Rev 2000; 5: 227–38.
Rack PM, Westbury DR. The effects of length and stimulus rate on tension in the isometric cat soleus muscle. J Physiol (Lond) 1969; 204: 443–60.
Rassier DE, MacIntosh BR, Herzog W. Length dependence of active force production in skeletal muscle. [Review]. J Appl Physiol 1999; 86: 1445–57.
Rosenfalck A, Andreassen S. Impaired regulation of force and firing pattern of single motor units in patients with spasticity. J Neurol Neurosurg Psychiatry 1980; 43: 907–16.[Abstract]
Ryerson S, Levit K. Functional movement re-education. New York: Churchill Livingstone; 1997.
Sharp SA, Brouwer BJ. Isokinetic strength training of the hemiparetic knee: effects on function and spasticity. Arch Phys Med Rehabil 1997; 78: 1231–6.[CrossRef][ISI][Medline]
Thepaut-Mathieu C, Van Hoecke J, Maton B. Myoelectrical and mechanical changes linked to length specificity during isometric training. J Appl Physiol 1988; 64: 1500–5.
van der Linden BJ, Koopman HF, Grootenboer HJ, Huijing PA. Modelling functional effects of muscle geometry. J Electromyogr Kinesiol 1998; 8: 101–9.[CrossRef][ISI][Medline]
Weir JP, Housh TJ, Weir LL, Johnson GO. Effects of unilateral isometric strength training on joint angle specificity and cross-training. Eur J Appl Physiol Occup Physiol 1995; 70: 337–43.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. I.E. Renner, H. Woldag, and H. Hummelsheim Central Compensation at Short Muscle Range Is Differentially Affected in Cortical Versus Subcortical Strokes Stroke, August 1, 2006; 37(8): 2076 - 2080. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Mehrholz, K. Rutte, and M. Pohl Jump training is feasible for nearly ambulatory patients after stroke Clinical Rehabilitation, May 1, 2006; 20(5): 406 - 412. [Abstract] [PDF]
|
|
|
|
 |
|
 E. G. Cruz, H. C. Waldinger, and D. G. Kamper Kinetic and kinematic workspaces of the index finger following stroke Brain, May 1, 2005; 128(5): 1112 - 1121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. H. McCrea, J. J. Eng, and R. Willms Phenol Reduces Hypertonia and Enhances Strength: A Longitudinal Case Study Neurorehabil Neural Repair, June 1, 2004; 18(2): 112 - 116. [Abstract] [PDF]
|
|
|
|
|
|
N. Byl Dr. Byl Responds Neurorehabil Neural Repair, March 1, 2004; 18(1): 9 - 11. [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||