Early exercise therapy in patients with severe traumatic spinal cord injury: is it feasible in the ICU?

Introduction

Following traumatic spinal cord injury (SCI), patients remain immobilized in the intensive care unit (ICU) and the wards for several weeks before they are transferred to rehabilitation [1]. Unfortunately, this places them at high risk for deconditioning and developing immobility-associated complications [2]. In addition, immobility during the acute stages after TSCI could potentially hinder adaptive neuroplasticity and limit long-term neurological recovery, while early mobilization/exercise could improve outcomes [3]. Until now, early exercise therapy (EET) had never been attempted in humans due to practical obstacles for bedridden patients and potential concerns for safety, especially for patients in the ICU. In this context, the PROMPT-SCI trial is the first trial designed to evaluate the safety and feasibility of EET in patients with acute severe TSCI (ClinicalTrials.gov: NCT04699474) [4]. In this Correspondence, we aim to report specifically on our patients who were hospitalized in the ICU.

As part of the PROMPT-SCI trial, 45 adult patients were recruited from a single Level-1 trauma center in Montreal, Canada, between April 2021 and August 2023. All had sustained a severe TSCI leading to an American Spinal Injury Association Impairment Scale (AIS) grade A, B or C injury. After immediate medical stabilization and resuscitation, MAP therapy was instituted (target MAP≈ 85 mmHg) and surgery was performed within 48 h to decompress the spinal cord and stabilize the spine. The intervention consisted of daily 30-min sessions of continuous passive in-bed leg cycling for 14 consecutive days, starting within 48 h of surgery based on the initiation criteria described in Table 1. During cycling, MAP, heart rate (HR), respiratory rate (RR) and blood oxygen saturation (SpO2) were monitored, and adverse events were noted. Sessions were stopped if patients requested termination, if vital signs fluctuated outside of the following ranges in a sustained fashion: MAP: 60–110 mmHg; HR: 40–140 bpm; SpO2: ≥ 90%, or if there were other signs of medical instability (e.g.hypo/hypertension, cardiac anomaly, etc.). Vasopressors were titrated to achieve a MAP≈ 85 mmHg and maintained at stable levels during sessions. After each session, complete neurological exams were performed to ensure safety.

Table 1.

Baseline characteristics of the 40 ICU patients who participated in the PROMPT-SCI trial

Characteristic	Mean ± SD / n (%)
Sex
Male	40 (88.9%)
Female	5 (11.1%)
Age	53.1 ± 18.4
BMI	26.1 ± 6.2
CCI
0	15 (33.3%)
≥ 1	30 (66.7%)
Living with:
Alone	5 (11.1%)
With someone	40 (89.9%)
AIS grade
A	27 (60.0%)
B	9 (20.0%)
C	9 (20.0%)
NLI
C1-C4	13 (28.9%)
C5-C8	13 (28.9%)
T1-T8	5 (11.1%)
T9-L1	14 (31.1%)
AIS motor score	30.4 ± 21.5
AIS light touch score	57.8 ± 28.8
AIS pinprick score	58.0 ± 29.1
Mechanism of injury
% Sports	1 (2.2%)
% Assault blunt	0 (0%)
% Fall	29 (64.4%)
% Transport	13 (28.9%)
% Other	2 (4.4%)
Timing of surgery (hours)	22.9 ± 14.2

i Impairment scale, NLI Neurological level of injury

Forty participants initiated the 14-day protocol in the ICU within 3 days of the SCI. Of these 40 participants, 33 (82.5%) managed to complete their first full session of cycling < 48 h after spine surgery, while the 7 remaining patients managed to do so the next day (< 72 h of surgery). Of these 7, 5 did not initiate cycling < 48 h post-op (2 were intubated and required a high degree of medical attention, 2 had scheduling conflicts due to investigations or therapeutic activities related to care and 1 refused due to fatigue/pain) and the other 2 initiated a session but it was interrupted prematurely (1 due to pre-existing abdominal pain that persisted with cycling and 1 due to prolonged SpO2 < 89% that resolved spontaneously after termination).

Over the remainder of the protocol, there were 366 potential sessions for our 40 ICU patients. Of these 366 sessions, 55 were not attempted (15.0%) and 5 (1.4%) were initiated but terminated prematurely (overall completion rate = 83.6%). Most frequent reasons for unattempted sessions were: medical instability as per caring team (n = 11), pressure injury (n = 8) and patient refusal due to pain/fatigue (n = 11). Reasons for stopping sessions were: persistent pre-existing abdominal pain or pain at surgical site (n = 3), prolonged SpO2 < 89% (n = 1) and conflicting schedule (n = 1; planned bronchoscopy). In addition, we report no major adverse event related to cycling.

Also, of our 40 ICU participants, 3 were mechanically ventilated during the protocol. While 2 initiated cycling only 72 h post-op, these 3 patients had a 100% completion rate with no deviation to the protocol thereafter.

In terms of average cardiorespiratory response to cycling, we observed no increase in HR (69.6 ± 14.5 rest vs. 69.9 ± 14.6 cycling; p = 0.351), a small increase in MAP (88.7 ± 11.4 rest vs. 89.6 ± 11.1 cycling; p = 0.037), a small increase in RR (20.3 ± 5.8 rest vs. 21.6 ± 5.1 cycling; p < 0.001) and a small decrease in SpO2 (97.0 ± 2.1 rest vs. 95.9 ± 2.9 cycling; p < 0.001), confirming that passive in-bed leg cycling is a form of low-intensity exercise, which possibly helps limit the neurological risk [5]. However, at the individual level, we do report significant increases in MAP > 10 mmHg (and up to 30 mmHg) with the initiation of cycling in the majority of our patients (n = 35), but these were never significant enough to warrant premature termination of cycling based on the stopping criteria described above. In addition, post-hoc analyses did not show any relation between these MAP increases and timing of sessions (first sessions vs later sessions). We also observed that 70% of MAP increases > 10 mmHg occurred in patients with neurological lesions above T7, which predispose to autonomic dyssynergia and dysreflexia.

In conclusion, this study supports the safety and feasibility of passive in-bed leg cycling in ICU patients with severe TSCI. We recommend close monitoring of MAP during sessions, especially for patients with lesions above T7.

Abbreviations

AIS

American spinal injury association impairment scale

EET

Early exercise therapy

HR

Heart rate

ICU

Intensive care unit

MAP

Mean arterial pressure

PROMPT-SCI

Protocol for rapid onset of mobilization in patients with traumatic spinal cord injury

RR

Respiratory rate

SCI

Spinal cord injury

SpO2

Blood oxygen saturation

TSCI

Traumatic spinal cord injury

Author contributions

AD was responsible for designing the study, collecting the data, performing the statistical analyses, interpreting the results and drafting the manuscript. DSK, ARD, DB, FB and YP were responsible for designing the study, interpreting the results and reviewing the manuscript. JMMT was responsible for designing the study, interpreting the results, reviewing the final manuscript and securing funding.

Funding

This research was funded by the Craig H. Neilsen foundation.

Availability of data and materials

The datasets that were generated during the course of this research are available from the corresponding author upon reasonable request.

Declarations

Ethical approval and informed consent

All patients provided informed written consent before enrollment. This study was approved by the Comité d’Éthique de la recherche du CIUSSS du Nord-de-L’Île de Montréal (“Mobilisation précoce suite à une lésion médullaire”, study #2020–1901, approved on March 12th, 2020) and was conducted in accordance with the principles of the Declaration of Helsinki.