Effects of early mobilization within 48 hours of injury in patients with incomplete cervical spinal cord injury

Abstract

Objective

To investigate the effects of early mobilization within 48 h of injury on motor function and walking ability in patients with incomplete cervical spinal cord injury (SCI).

Design

A retrospective observational study.

Setting

Intensive care unit or high care unit of a university hospital emergency center.

Participants

Of 224 patients with SCI having American Spinal Injury Association impairment scale grades C and D, 158 consecutive patients hospitalized for at least 3 weeks after injury were included.

Interventions

Patients were categorized into two groups: an early mobilization group in which patients were mobilized within 48 h of injury and a delayed mobilization group in which they were mobilized after 48 h of injury.

Outcome measures

The upper extremity motor score (UEMS), lower extremity motor score (LEMS), and Walking Index for Spinal Cord Injury II (WISCI II) were compared using propensity score matching analysis.

Results

Of the 158 patients who met the eligibility criteria, 32 were matched between the groups. There was a significant difference in the change in LEMS from the initial assessment to the assessment 2 weeks postoperatively in the early mobilization group (median 9 points vs. 3 points, p < 0.05). There were no significant differences in UEMS or WISCI II.

Conclusion

Early mobilization within 48 h may improve lower extremity motor function in patients with acute incomplete cervical SCI.

Introduction

In Japan, the number of middle-aged and older patients with spinal cord injury (SCI) due to minor trauma such as falls and falls on level ground is increasing (1), and is further expected to increase as the population ages. The goal of physical therapy (PT) in patients with SCI is to restore their pre-injury function and lifestyle by mobilizing them early after the injury, improving symptoms of lower extremity motor paralysis and reconstructing their gait. In recent years, acute care hospitals have emphasized early rehabilitation (2–5), particularly early mobilization, as hospital stays have become shorter (6). Patients wish to initiate the recovery process as soon as possible; therefore, demonstrating the impact of early mobilization is important (7). Prolonged duration between injury and rehabilitation has been associated with increased morbidity and decreased mobility and quality of life, suggesting that delayed rehabilitation may be detrimental (8–12). However, owing to factors in the health care delivery system (13), including shorter hospital stays (6) and ethical issues, no prospective studies have directly compared early and delayed mobilization (4, 5, 13). Consequently, the influence of early mobilization on motor function and walking ability in patients with incomplete cervical SCI remains unclear. Recently, propensity score-based analysis has been used in comparative studies between intervention and control groups with the same propensity score (14, 15). It is the most common method of comparison in observational studies, as it can adjust for confounding factors while estimating causal relationships (14, 15). In this method, the same propensity score is matched between the intervention and control groups for comparison; therefore, it is less susceptible to ethical issues. Therefore, this study aimed to determine the effect of early mobilization in patients with incomplete cervical SCI using propensity score matching analysis. We hypothesized that, compared to delayed mobilization, early mobilization would improve the recovery of motor function and walking ability in patients with incomplete SCI.

Materials and methods

Study design and setting

This retrospective observational study utilized propensity score matching analysisto compare motor function and walking ability between groups with early and delayed mobilization. This study was conducted in accordance with the Declaration of Helsinki and the Ethical Guidelines for Clinical Research Involving Human Subjects after obtaining approval from the Research Ethics Committee (Approval No. S22-010). With the option of opt-out, the study details were made publicly available on the Center's website, and patients were guaranteed the opportunity of refusal to participate in the study.

Participants

The participants were consecutive patients with American Spinal Injury Association impairment scale (AIS) grades C and D incomplete cervical SCI who were admitted to the intensive care unit or high care unit of our center from April 2018 to March 2022. The patients were selected retrospectively by evaluating their medical records. The follow-up period was 3 weeks after the injury. The exclusion criterion was transfer or discharge from the hospital within 3 weeks of the injury.

Group classification

Early mobilization was defined as follows: early was defined as within 48 h of injury (16, 17), and mobilization was defined as getting out of bed and transfer to a wheelchair (12, 18). Patients were divided into two groups: an early mobilization group, in which patients were mobilized within 48 h of injury by the medical staff, such as physical therapists or nurses, and a delayed mobilization group, in which patients were mobilized 48 h after injury. The time of mobilization was determined by retrospectively evaluating the medical records.

Outcome measures

Basic information includes age, sex, pre-injury activity level (modified Rankin scale [mRS] score), AIS grade, presence of extremity trauma, cardiac disease, renal disease, cerebrovascular disease, diabetes mellitus, orthostatic hypertension, or dementia, Charlson comorbidity index (CCI) score, number of days from injury to the start of PT, and the number of days to start walking. The mRS score before the injury was interpreted as no assistance if it was less than 2 and as assistance if greater than 3. Motor function was assessed by examining upper extremity motor scores (UEMS) of 5 key muscles (C5: elbow flexors; C6: wrist extensors; C7: elbow extensors; C8: finger flexors; T1: little finger abductor muscles) of both upper extremities and the lower extremity motor scores (LEMS) of 5 key muscles (L2: hip flexors; L3: knee extensors; L4: ankle dorsiflexors; L5: toe extensors; S1: ankle plantar flexors) of both lower extremities, according to the International Standards for Neurological Classification of Spinal Cord Injury developed by the American Spinal Injury Association (19). The Walking Index for Spinal Cord Injury II (WISCI II) level was used to assess walking ability. Changes in scores from the first PT session to the 3 week were quantified. All information was extracted retrospectively from the medical records.

Rehabilitation

All patients received 40–60 min of PT five times a week from the physiotherapist in charge of the patients. The patients’ independence improved through rehabilitation approaches, such as getting out of bed, automatic lower limb movement, basic movement practice, and gait practice using a walking aid. The physician determined the time of the first rehabilitation appointment according to each patient's condition. Daily occupational and speech therapies were prescribed to the participants as needed.

Statistical analysis

All data are presented as medians (interquartile range) with or without normality. The effects of early mobilization were compared using the unpaired t-test, Mann-Whitney's U test, χ-square test, or Fisher's exact probability test for each examination item between the early and delayed mobilization groups. To examine the effect of early mobilization on motor function and walking ability, we calculated the probability of early separation using covariates as potential confounders. Next, we used nearest-neighbor matching to match the early and delayed mobilization groups in a ratio of 1:1 for approximate probabilities. Variables used to calculate the propensity scores were age (20, 21), sex (22), pre-injury activity level (23), AIS grade (24, 25), number of days from injury to the start of PT (11), presence of extremity trauma (26, 27), presence of dementia (28), CCI score (29), presence of orthostatic hypotension (30, 31), initial UEMS, and LEMS (11, 19, 32) as they could affect early mobilization and outcomes according to previous studies. The tolerance (caliper) for matching was set at ±20% of the standard deviation of the propensity score (33). The effect size (r) was then calculated to assess thedifference between the two groups. The effect size (r) was interpreted as 0.1 for small, 0.3 for medium, and 0.5 for large (34). The statistical analysis software JMP version 16 (SAS Institute, Cary, NC) was used, and the significance level was set at 5%.

Results

Characteristics of the participants

A total of 224 patients met the eligibility criteria. Of these, 158 patients were included in the analysis, excluding 66 patients who were transferred or discharged within 3 weeks of injury. The early and delayed mobilization groups comprised 100 (63%) and 58 (37%) patients, respectively (Figure 1). A comparison between the two groups before propensity score matching showed that the early mobilization group had significantly less time from injury to the start of PT, a lower probability of diabetes mellitus (Table 1), a significantly shorter time from injury to the start of walking, and a significantly greater change in LEMS (Table 2). When assessing the LEMS of each key muscle, the changes in L3, L4, and L5 were significantly greater in the early mobilization group (Table 2). However, no significant differences in the UEMS of key muscles or WISCI II levels were observed between the two groups (Table 2).

Figure 1.

Flowchart of participant selection. AIS, American Spinal Injury Association impairment scale.

Table 1.

Participant characteristics.

	Entire cohort		p-value	Propensity score-matched cohort		p-value
	Early mobilization n = 100	Delayed mobilization n = 58		Early mobilization n = 32	Delayed mobilization n = 32
Age (years)	73 (65–79)	74 (64–81)	0.69	74 (66–80)	73 (62–80)	0.75
Sex (%)			0.14			0.59
Male (n)	78 (78.0)	39 (67.2)		21 (65.6)	23 (71.9)
Female (n)	22 (22.0)	19 (32.8)		11 (34.4)	9 (28.1)
Pre-injury mRS score (%)			0.20			1.00
mRS score ≤2	95 (95.0)	52 (89.7)		31 (96.9)	31 (96.9)
mRS score >2	5 (5.0)	6 (10.3)		1 (3.1)	1 (3.1)
AIS grade (%)			0.60			1.00
C	44 (44.0)	28 (48.3)		15 (46.9)	15 (46.9)
D	56 (56.0)	30 (51.7)		17 (53.1)	17 (53.1)
Days from injury to start of PT	1 (1–2)	3 (2–3)	<0.05	2 (1–3)	2.5 (2–3)	0.17
Extremity trauma (%)	3 (3)	1 (1.7)	0.62	1 (3.1)	1 (3.1)	1.00
Cardiac disease (%)	16 (16)	6 (10.3)	0.32	3 (9.4)	1 (3.1)	0.30
Renal disease (%)	9 (9)	2 (3.4)	0.19	0 (0)	0 (0)	1.00
Cerebrovascular disease (%)	14 (14)	7 (12.1)	0.73	4 (12.5)	1 (3.1)	0.16
Diabetes mellitus (%)	19 (19)	17 (29.3)	<0.05	6 (18.6)	10 (31.3)	0.25
Dementia (%)	6 (6)	1 (1.7)	0.21	0 (0)	1 (3.1)	0.31
CCI score	0 (0–1)	0 (0–1)	0.72	0 (0–1)	0 (0–1)	0.89
Orthostatic hypotension (%)	17 (17)	6 (10.3)	0.25	4 (12.5)	4 (12.5)	1.00
Admission UEMS	26 (12–37)	23 (12–34)	0.44	29 (13–37)	24 (15–31)	0.69
C5 average of left and right	4.0 (2.5–5.0)	3.5 (2.5–4.5)	0.38	3.5 (2.5–4.5)	3.5 (1.6–4.9)	0.82
C6 average of left and right	2.5 (1.0–4.0)	2.0 (1.0–4.0)	0.67	2.8 (1.1–4.0)	2.0 (1.0–3.9)	0.52
C7 average of left and right	2.5 (1.5–4.0)	2.5 (0.5–3.6)	0.42	2.5 (1.5–3.9)	2.5 (0.5–3.4)	0.58
C8 average of left and right	2.5 (1.0–4.0)	2.0 (1.0–4.1)	0.72	2.5 (1.1–4.0)	2.5 (0.1–4.4)	0.67
T1 average of left and right	2.0 (0.5–4.0)	1.5 (0.5–4.5)	0.81	2.5 (1.0–4.0)	1.5 (0.0–4.5)	0.37
Admission LEMS	34 (17–46)	33 (21–45)	0.97	34 (18–45)	34 (13–45)	0.78
L2 average of left and right	3.0 (1.5–4.5)	3.0 (1.5–4.0)	0.91	2.8 (1.1–4.4)	3.0 (1.5–4.0)	0.83
L3 average of left and right	3.0 (1.0–4.5)	3.0 (1.0–4.0)	0.96	3.3 (1.0–4.5)	2.5 (1.0–4.4)	0.59
L4 average of left and right	3.0 (1.0–4.5)	2.5 (1.0–5.0)	0.91	2.8 (1.1–4.5)	2.5 (0.6–4.5)	0.68
L5 average of left and right	3.5 (1.5–4.5)	4.0 (1.5–5.0)	0.74	3.0 (1.6–4.4)	4.0 (1.3–5.0)	0.64
S1 average of left and right	4.0 (2.0–5.0)	4.0 (1.5–5.0)	0.98	3.8 (2.1–5.0)	3.5 (1.5–5.0)	0.77
Admission WISCI II (%)			0.68			1.00
0	91 (91.0)	55 (94.8)		31 (96.9)	31 (96.9)
1	1 (1.0)	0 (0.0)		0 (0.0)	0 (0.0)
6	1 (1.0)	1 (1.0)		0 (0.0)	0 (0.0)
7	1 (1.0)	0 (0.0)		0 (0.0)	0 (0.0)
13	2 (2.0)	0 (0.0)		0 (0.0)	0 (0.0)
17	2 (2.0)	0 (0.0)		0 (0.0)	0 (0.0)
20	2 (2.0)	2 (2.0)		1 (3.1)	1 (3.1)

mRS, modified Rankin scale; AIS, American Spinal Injury Association Impairment Scale; PT, physical therapy; CCI, Charlson Comorbidity Index; UEMS, Upper Extremity Motor Score; LEMS, Lower Extremity Motor Score; WISCI II, Walking Index for Spinal Cord Injury II; C5, elbow flexors; C6, wrist extensors; C7, elbow extensors; C8, finger flexors; T1, little finger abductor muscles; L2, hip flexors; L3, knee extensors; L4, ankle dorsiflexors; L5, toe extensors; S1, ankle plantar flexors.

Table 2.

Relationship of early mobilization with motor function and walking ability.

	Entire cohort		p-value	Effect size (r)	Propensity score-matched cohort			Effect size (r)
	Early mobilization n = 100	Delayed mobilization n = 58			Early mobilization n = 32	Delayed mobilization n = 32	p-value
Days to start walking	3 (2–7)	7 (4–11)	<0.05	0.39	3 (2–8)	9 (4–11)	<0.05	0.38
UEMS	7 (1–11)	5 (0–12)	0.17	0.11	8 (2–11)	4 (0–10)	0.08	0.22
C5	0.5 (0.0–1.0)	0.0 (0.0–1.0)	0.37	0.07	0.5 (0.0–1.0)	0.0 (0.0–1.0)	0.33	0.12
C6	0.5 (0.0–1.5)	0.5 (0.0–1.0)	0.35	0.07	0.5 (0.0–1.4)	0.3 (0.0–0.9)	0.19	0.16
C7	0.5 (0.0–1.4)	0.5 (0.0–1.2)	0.66	0.04	0.5 (0.0–1.4)	0.3 (0.0–1.0)	0.31	0.13
C8	0.5 (0.0–1.5)	0.5 (0.0–1.5)	0.87	0.01	0.5 (0.0–1.0)	0.0 (0.0–1.5)	0.52	0.08
T1	0.5 (0.0–1.1)	0.0 (0.0–1.5)	0.89	0.01	0.5 (0.0–1.0)	0.0 (0.0–1.5)	0.57	0.07
LEMS	6 (2–14)	3 (0–8)	<0.05	0.20	9 (2–15)	3 (0–6)	<0.05	0.25
L2	0.5 (0.0–1.4)	0.3 (0.0–1.0)	0.10	0.13	0.5 (0.0–1.5)	0.0 (0.0–1.0)	0.12	0.20
L3	0.5 (0.0–1.5)	0.0 (0.0–1.0)	<0.05	0.18	0.5 (0.0–1.5)	0.3 (0.0–1.0)	0.22	0.15
L4	0.5 (0.0–1.5)	0.0 (0.0–1.0)	<0.05	0.18	0.5 (0.0–1.5)	0.3 (0.0–1.0)	0.22	0.15
L5	0.5 (0.0–1.5)	0.0 (0.0–1.0)	<0.05	0.21	1.0 (0.0–1.5)	0.0 (0.0–0.5)	<0.05	0.40
S1	0.5 (0.0–1.5)	0.0 (0.0–1.0)	0.10	0.13	0.8 (0.0–1.5)	0.0 (0.0–1.0)	0.10	0.21
WISCI II	0 (0–17)	0 (0–14)	0.61	0.04	1 (0–20)	0 (0–15)	0.22	0.15

Note: UEMS, LEMS, each Key muscle (C5-T1, L2-S1), and WISCI indicate the amount of change.

UEMS, Upper Extremity Motor Score; LEMS, Lower Extremity Motor Score; WISCI, Walking Index for Spinal Cord Injury II.

UEMS of the 5 key muscles (C5, elbow flexors; C6, wrist extensors; C7, elbow extensors; C8, finger flexors; T1, little finger abductor muscles)

LEMS of the 5 key muscles (L2, hip flexors; L3, knee extensors; L4, ankle dorsiflexors; L5, toe extensors; S1, ankle plantar flexors).

Comparison of factors between the two groups after propensity score matching

Propensity score matching was performed in 32 patients each in the early and delayed mobilization groups. The comparison of factors between the two groups after propensity score matching showed no significant differences (Table 1). Figure 2 shows the PT received by both groups.

Figure 2.

Total number of physical rehabilitation sessions received by participants in the early and delayed mobilization groups after propensity score matching.

Motor function and walking ability with and without early mobilization

After propensity score matching, the number of days to start walking was 3 and 9 in the early and delayed mobilization groups, respectively; it was significantly less in the early mobilization group, and the effect size (r) was moderate at 0.38. The LEMS changed by 9 and 3 in the early and delayed mobilization groups, respectively; the change in LEMS was significantly greater in the early mobilization group, and the effect size (r) was moderate at 0.25. Among the LEMS key muscles, only L5 showed a significant change in the early mobilization group, while the remaining key muscles showed a trend toward improvement but without a significant difference (Table 2). The average left-right frequencies of admission of the key muscles in early and delayed mobilization are shown in Figure 3. A ceiling effect was observed in all key muscles of the lower extremities in the early mobilization group. In contrast, although there was no significant difference between UEMS and WISCI II level, the UEMS changed by 8 and 4 in the early and delayed mobilization groups, respectively, while the WISCI II level changed by 1 and 0 in the early and delayed mobilization groups, respectively, indicating a trend of improvement in the early mobilization group (Table 2). WISCI II level at the first time tended to be concentrated in 0 (unable to walk).

Figure 3.

Frequency of left-right averages of admission key muscles in the early and delayed mobilization after propensity score matching.*: ceiling effect (>15%).

Discussion

The effects of early mobilization in patients with acute incomplete cervical SCI have not been reported due to ethical issues (4, 5). In this study, we compared the effects of early and delayed mobilizations using propensity score matching analysis, in which factors affecting early mobilization and outcomes were used as covariates. The results revealed that early mobilization within 48 h of injury improved motor function. The LEMS change in the early and delayed mobilization groups was 9 and 3, respectively, significantly higher in the early mobilization group. In a report on distribution-based minimal clinically important differences between patients with acute and subacute SCI (35), estimates based on effect sizes for small changes in LEMS between patients with AIS grades C and D cervical SCI ranged from 1.8 to 2.4 points and estimates based on effect sizes for large changes in LEMS ranged from 4.6 to 5.9 points. Both groups exceeded the limit of small changes, indicating a certain degree of improvement regardless of the timing of mobilization. However, only the early mobilization group showed a change exceeding 6 points, an estimate based on the effect size for large changes. Therefore, the results suggest that early mobilization from bed within 48 h of injury is clinically beneficial and leads to significant improvements in motor function.

The results in the early mobilization group showed that only the LEMS of L5 had significantly increased, while that of other key muscles showed a trend toward improvement but without significant differences. The LEMS of each key muscle are reportedly strongly associated with standing retention and gait, but the role of L5 is unknown (20, 36, 37). It is unlikely that only L5 was improved by early mobilization. The L5 may have significantly improved due to the higher admission score of 4.0 in the delayed mobilization group. The reasons for the lack of differences in each key muscle in the early mobilization group could be due to the ceiling effect, with grade 5 motor function in key muscles having an incidence of 15% (38) or higher in the LEMS of all key muscles. The significant improvement in LEMS and the trend of improvement in each key muscle suggest that early mobilization promoted improvement of the entire lower extremities rather than a specific key muscle. Early mobilization resulted in no difference in the UEMS or its key muscles because the same approach can be taken with or without early mobilization.

A systematic review of patients receiving intensive care (39) showed that in addition to early mobilization, rehabilitation using joint mobilization, basic movement, and gait exercises could improve muscle strength (40–42). The review included three studies, which were characterized by more active rehabilitation in the early mobilization group, with a higher frequency of intervention, such as twice daily. Therefore, the results of previous studies cannot be directly compared with those of the present study, which used the same rehabilitation protocol. We provided 40–60 min of PT five times a week to all participants and found a significant difference in the improvement in motor function depending on whether the patients were mobilized early or late. The number of days to start walking was also significantly different, suggesting that early mobilization may have facilitated the transition to standing, weight bearing, and gait practice as the next movement task. An SCI study in rats showed that early weight-bearing and walking exercises after injury significantly improved motor function (43, 44). This is because the spinal cord is inherently plastic during the early post-injury period, and early exercise promotes functional recovery (45). These findings suggest that early exercise promotes functional recovery.

Lower extremity motor function is the most crucial factor affecting gait (20, 46), and improvement in walking ability is expected to accompany improvement in lower extremity motor function. This study showed a trend toward improvement in walking ability with early mobilization but without a significant difference. Approximately 3 weeks after injury, there is a marked improvement in physical function (47), but it is difficult to determine the effects on movement ability (48) when the LEMS is low. The rate of gait acquisition has been reported to decrease in older people (48, 49). In addition, WISCI II assessment may not have fully captured the improvement in walking ability due to a limited number of relevant levels. The WISCI II is a 21-level grading scale from 0 (unable to walk) to 20 (independent walking without a cane) for assessing the walking ability of patients with SCI. It has excellent reliability and validity and is a theoretical construct (50, 51). However, the patients may have only considered a few levels, which could have hindered the comprehensive assessment of their walking improvement. Of note, in the acute phase, gait exercises are often performed with a walker with wheels and cannot be assessed using WISCI II. Further, even if the independence of the walker with wheels improves, this is not reflected in the WISCI II level. Additionally, even if the WISCI II level does not change, it undeniably can affect the walking speed and efficiency. These factors suggest that early mobilization may not have improved walking ability.

This study has some limitations, including the small sample size (52) and possible bias in background factors because of the single-center nature of the study. Second, the time of mobilization could have been affected by various factors, including physical condition and postoperative pain status, and therefore, the unknown confounding factors should have been assessed more effectively. Further, this was a retrospective study that failed to address potential bias. In this study, patients in the early mobilization group started to walk significantly earlier, suggesting the increase in activity associated with early mobilization. However, because we could not investigate the time of mobilization and the protocol of rehabilitation during the day, we cannot comment on the effect of prolonged mobilization or of changes in the protocol of rehabilitation associated with mobilization. Therefore, the results of this study should be interpreted in light of these limitations. Further studies, including prospective studies at multiple centers, are necessary to clarify the effect of early mobilization and the causes of delayed mobilization.

Conclusion

In this study, we compared the motor function and walking ability between the early and delayed mobilization groups using propensity score matching analysis to clarify the effect of early mobilization in patients with incomplete cervical SCI. The results showed that early mobilization within 48 h accelerated the onset of walking and significantly improved lower extremity motor function. In contrast, there was no significant improvement in upper extremity motor function or walking ability. In the future, it is necessary to examine further the factors beneficial to lower extremity motor function following early mobilization.

Disclaimer statements

Contributors None.

Funding None.

Conflicts of interest Authors have no conflict of interests to declare.