Decreasing pressure injuries and acute care length of stay in patients with acute traumatic spinal cord injury

Abstract

Objectives

Identifying factors associated with the occurrence of pressure injuries (PI) during acute care and with longer length of stay (LOS), focusing on modifiable factors that can be addressed and optimized by the acute rehabilitation team.

Design

Prospective cohort study.

Setting

A single Level-1 trauma center specialized in SCI care.

Participants

A cohort of 301 patients with acute TSCI was studied.

Outcome measures

The primary outcome was the occurrence of PI during acute care stay. The secondary outcome was acute care LOS. Bivariate and multivariate logistic or linear regression analyses were performed to determine the association between non-modifiable factors and outcomes (PI of any stage and acute LOS), whereas bivariate and hierarchical multivariate logistic or linear regression analyses were used for modifiable factors.

Results

When controlling for the level and severity of the TSCI, the occurrence of pneumonia (OR = 2.1, CI = 1.1–4.1) was significantly associated with the occurrence of PI. When controlling for the level and severity of the TSCI, the occurrence of medical complications (PI, urinary tract infection and pneumonia) and lesser daily therapy resulted in significantly longer acute care LOS (P < .001).

Conclusions

Prevention of PI occurrence and the optimization of the acute care LOS represent crucial challenges of the acute rehabilitation team, as they are significantly associated with higher functional outcomes. Patients who develop pneumonia may benefit from more aggressive prevention strategies to reduce PI occurrence. Systematic protocols for the prevention of complications as well as greater volume of therapy interventions should be considered to optimize the acute care LOS.

Introduction

Following a traumatic spinal cord injury (TSCI), patients undertake a rehabilitation process typically consisting of three consecutive phases: acute hospitalization, intensive functional rehabilitation (IFR) and community integration. This continuum of care aims to improve patient quality of life by optimizing functional recovery.¹ Most functional recovery occurs within the first six months following the TSCI, which is when the acute care and intensive functional rehabilitation (IFR) take place.^2–5 The optimization of the clinical evolution of patients during the early process is thus crucial to promote efficient functional rehabilitation⁶ and ultimately improve long-term functional outcome.^7–10

Previous work showed that longer acute care length of stay (LOS) and the occurrence of medical complications during acute care negatively influenced the course of IFR and long-term functional outcome, even when adjusted for important confounding variables such as the level and severity of the TSCI.^7,11,12 One of the most common and preventable complications following TSCI is the occurrence of a pressure injury (PI), a serious complication that has the potential to interfere with functional, psychological and social well-being outcomes.^6,13,14 Compared to other rehabilitation phases, the acute hospitalization represents the period with the highest PI risk.^15,16 Indeed, the occurrence of altered level of consciousness, multiple concomitant traumatic injuries and severe neurological deficits represent some of the important factors leading to prolonged periods of immobility and decreased general health status, putting patients at higher risk of PI during acute care.¹⁷ Preventing PI during the acute care phase is of crucial importance, as its occurrence is associated with higher rates of medical complications, recurrence^18–21 and may interfere with rehabilitation, limiting the long-term functional outcome.^22–24

Previous work showed the impact of characteristics of the individual (age, comorbidities, etc.) and of the injury (level and severity of the TSCI, burden of associated traumatic injuries) on the LOS and on the occurrence of PI during acute care.^{15,17,20,25–31} Unfortunately, the acute medical and rehabilitation teams cannot modify these factors. On the other hand, the impact of various factors that can be addressed and optimized during the acute care remains poorly documented. The identification of such modifiable factors would aid the acute rehabilitation specialist in optimizing the rehabilitation process.

Thus, the aim of this study was to identify how the acute rehabilitation team can facilitate the subsequent rehabilitation process by decreasing PI incidence and acute care LOS following TSCI through the identification of (1) factors associated with the occurrence of PI, and (2) factors associated with longer acute care LOS, focusing on modifiable factors that can be addressed and optimized.

Methodology

Patients

We reviewed data that were collected prospectively on 301 patients with acute TSCI admitted to a single Level I trauma center between April 2010 and October 2018. The inclusion criteria were: (1) being 18 years and older, (2) having sustained an acute TSCI from level C1 to L1 requiring surgical management, and (3) having a neurological deficit with an American Spinal Injury Association impairment scale (AIS) grade A to D as determined from the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI).³² Patients were excluded if they (1) were pregnant, (2) had ankylosing spondylitis, (3) had a pre-existing major neurological condition (e.g.: stroke, Parkinson disease, multiple sclerosis, etc.), (4) died during the acute care hospitalization. Patients with ankylosing spondylitis were excluded as this condition is associated with a more complex surgical management, more medical complications and a different outcome when compared to other TSCI patients.^33–35 Patient who died were excluded from our study as death occurred within the first 24–48 h upon admission due to neurogenic shock or respiratory/hemodynamic instability and thus the management of those were prioritized over decreasing pressure injury occurrence. This study was approved by the institution’s ethics committee and all patients were enrolled on a voluntary basis and provided informed consent. A research assistant who was not involved in the study design and data analysis performed the data collection.

Data collection

Information on socio-demographic factors and patient characteristics such as age, sex, presence of diabetes and tobacco consumption (past or active smoking vs. non-smoking) were compiled. The burden of comorbidities was also calculated using the Charlson Comorbidity Index (CCI).³⁶ CCI scores were then dichotomized into scores of less than 2 vs 2 or more, as a steep increase in financial burden is seen in patients with comorbidity scores of 2 or more.³⁶

Data collection also included delay from trauma to admission and surgery. The delay of admission designated the interval of time between the trauma and time of admission (in hours). The delay to surgery was defined as the interval of time between the injury and time of incision (in hours).

Within the first 72 h upon admission, the neurologic evaluation was performed based on the ISNCSCI, which includes a standardized motor, sensory and rectal examination.³² The neurologic level of the injury (NLI) was defined as the most caudal level with preserved normal sensation and motor function. Then, the NLI was stratified for high tetraplegia (C1 to C4), low tetraplegia (C5 to C8) and paraplegia (T1–L1). The severity of the TSCI was determined by the ISNCSCI where injuries are classified as being neurologically ‘complete’ or ‘incomplete’ depending on sacral sparing³² and was stratified as AIS grade A, AIS B, AIS C and AIS D.³²

During the acute care hospitalization, data regarding the occurrence of the most common TSCI medical complications (pneumonia, urinary tract infection and PI)³⁷ were collected. Pneumonias were diagnosed using clinical features and confirmed by a radiologist using chest X-rays.³⁸ Urinary tract infections were diagnosed using criteria from the 2006 Consortium for Spinal Cord Medicine Guidelines for healthcare providers.³⁹

Total time spent in therapy (physical and occupational therapy) during the acute care stay was also compiled (in hours) and daily therapy time was calculated by dividing total time spent in therapy by the number of days of acute care LOS. Daily therapy time was used rather than the total number of days of therapy as in Canada, patients will benefit from therapy as long as they are hospitalized. In accordance with current guidelines,⁴⁰ in addition to our PI prevention protocols that are applied upon arrival at our emergency room as described below, therapy interventions are initiated as early as the first post-operative day. At our SCI-specialized Level-1 trauma center, surgery generally occurs within 72 h post-trauma.^41,42 Passive mobilizations of the four limbs if the patient is tetraplegic or lower limbs if the patient is paraplegic are performed by the physiotherapist six times a week on the first post-operative day and during the following fourteen days. The main purpose of these mobilizations is to promote skin integrity by increasing the range of motion, improving patient mobility and preventing contractures.^43,44 In addition, patient positioning and seating is addressed and optimized to prevent secondary complications. Thereafter, therapy interventions include exercises focusing on increasing muscle strength, range of motion and trunk stability, all of which are essential prerequisites to transfers and other functional activities that will be further addressed during IFR. Also, every patient admitted to our SCI-specialized Level-1 trauma center benefit from a physiatry consultation. The consultation is done as early as possible, and thus typically occurs within the first week upon admission. During acute care, the role of the physiatrist include preliminary neuro-functional prognosis, the establishment of the subsequent rehabilitation plan, prescribing interventions that will promote recovery and decrease secondary complications, educating patients and families about the rehabilitation process and encouraging their participation in discharge planning discussions and finally ensuring patients have met all IFR pre-requisites (increasing strength, joint mobility, endurance and postural control^45–50).⁴⁰

Variables were classified as being (1) non-modifiable (age, sex, diabetes, CCI, tobacco consumption, AIS grade and NLI) or (2) modifiable (admission delay, surgery delay, occurrence of medical complications (pressure injuries, urinary tract infection and pneumonia) and daily therapy time). Although tobacco consumption is traditionally considered a modifiable risk factor, in our study it was assessed at time of injury so up until then it was not modifiable and could have hypothetically contributed to outcomes during the acute hospitalization. Modifiable factors were our main independent variables and were controlled for possible cofounding non-modifiable factors.

Outcome variables

The primary outcome was the occurrence of at least one PI of any stage as defined by the National Pressure Ulcer Advisory Panel (NPUAP) during the acute care hospitalization at our Level I SCI-specialised trauma center.⁵¹ We collected information on both stage and localization of each pressure injury. Protocols for prevention of PI at our institution aim at maintaining skin integrity, decrease pressure, shear and friction, control moisture level and optimize nutrition, hydration and mobility.⁸ A risk assessment using the Braden Scale⁵² is conducted in the first 24 h upon admission and once per week afterward. Upon arrival at our emergency room, patients are transferred on a foam stretcher pad with a viscoelastic polymer gel mattress (Blue Cloud™; Batrik Medical Manufacturing, Montreal, Canada) and undergo log roll mobilizations once every 2 h until spinal stabilization surgery. Following surgery, patients are cared for on a therapeutic, low air loss mattress (Versacare A.I.R.^® Surface; Rom-Hill, Mississauga, Canada), with repositioning every 2 h and regular skin care/assessment.

The LOS, defined as the delay in days between admission and discharge from acute care was the secondary outcome variable.

Statistical analysis

We described our cohort using mean ± standard deviation for continuous variables, and proportions or percentages for categorical variables. As the LOS and total daily therapy variables were not normally distributed, the log base 10 transformation was used for the LOS and the square root transformation was used for the total daily therapy in order to reduce skewness. Similarly, surgery delay was dichotomized into <24 h delay and ≥24 h delay; and admission delay was dichotomized into <12 h delay and ≥12 h delay which corresponded to the median of our cohort and was deemed clinically relevant.⁵³

All non-modifiable variables were first examined using bivariate logistic regressions to assess the association of each factor with the occurrence of PI. A hierarchical multivariate logistic regression analysis with all non-modifiable factors that were significantly associated with the occurrence of PI in bivariate analysis was then done to determine the final model of non-modifiable factors that will be controlled for in the second step.

Unadjusted coefficients for modifiable factors were estimated using bivariate logistic regression analyses. Adjusted coefficients were estimated using a hierarchical multivariate logistic regression analyses where modifiable factors were controlled for the non-modifiable factors identified in the first step. The association between the independent variables and the occurrence of PI was expressed in terms of odds ratio (OR) with 95% confidence interval (CI).

We used the same analytic strategy for LOS as the dependent variable but by using linear rather that logistic regression analyses. The association between the independent variables and the LOS was expressed in terms of beta (β) coefficients with significance levels. We used IBM SPSS Statistics Version 25 software package (IBM Corp., Armonk, NY, USA) for our statistical analyses.

Results

Among the 534 patients hospitalized between April 2010 and October 2018, 233 patients were excluded: 80 patients died during the acute care hospitalization, 147 patients refused to be part of the study, five had pre-existing major neurological condition (three strokes, two hemiplegia) and one patient was missing information regarding PI occurrence. A total of 301 patients were thus included in the analyses (Table 1). The mean age was 50.76 ± 18.42 years old.

Table 1. Patients characteristics- non-modifiable and modifiable factors (N = 301).

Non-modifiable factors		Presence of PI (%)
		No	Yes	P value
Age	less than 75	91.9	89.2	0.5
Sex	Male	80.9	86.2	0.3
Tobacco status	Yes, currently or in the past	60.2	52.3	0.4
Charlson comorbidity index (CCI)	less than 2	92.4	90.8	0.7
Diabetes	No	89.0	90.8	0.7
Level of injury	C1–C4	33.6	61.5	<.001*
	C5–C8	24.9	15.4
	T1–L1	41.5	23.1
AIS grade	A	36.8	55.4	0.1
	B	10.8	12.3
	C	17.7	16.9
	D	34.6	15.4
Modifiable factors
Urinary tract infection (UTI)	Yes	16.9	24.6	0.2
Pneumonia	Yes	19.5	43.1	<.001*
Admission delay (h)	<12 h	39.4	41.4	0.8
Surgery delay (h)	<24 h	53.6	48.3	0.5
Daily therapy time (h/day)	Mean (± SD)	1.0 (0.3)	1.0 (0.2)	0.8

H: hours.

SD: standard deviation.

*P is significant if < 0.05.

Sixty-five patients (21.6%) developed at least one PI during the acute care hospitalization. The most common localization was the sacrum (n = 53; 81.5%), followed by ‘other’ (chin, big toe, back, ankle, elbows, proximal fibula, trochanter) (n = 9; 13.8%) and heels (n = 3; 4.6%). The overall distribution of severity of PI was Stage I (n = 19; 29.2%), Stage II (n = 41; 63.1%), Stage III (n = 2; 3.1%), Stage IV (n = 2; 3.1%), Stage X (0), suspected deep tissue PI (n = 1; 1.5%), undetermined (0).

Bivariate analyses of non-modifiable factors were all non-significant (P > 0.08) except for the AIS grade and NLI, which were significantly associated with the occurrence of PI (P = 0.02 and P < .001, respectively) (Table 2). The final model of hierarchical multivariate logistic regression of non-modifiable factors associated with the occurrence of PI included two predictors: AIS grade and NLI (P < .001). The occurrence of pneumonia was significantly associated with the occurrence of PI in bivariate analyses and when adjusted for AIS grade and NLI (Table 3).

Table 2. Non-modifiable factors associated with the occurrence of pressure injuries: bivariate analysis.

Non-modifiable factors			95% CI
		OR	Lower	Upper
Age	Less than 75	1d	–	–
	75 and more	1.38	0.55	3.44
Sex	female	–	–	–
	male	1.47	0.68	3.18
Tobacco status	Yes, currently or in the past	1d	–	–
	No	1.30	0.72	2.35
	Unknown or refused to answer	1.72	0.66	4.48
Charlson Comorbidity Index (CCI)	less than 2	1d	–	–
	2 or more	1.23	0.47	3.24
Diabetes	No	1.22	0.48	3.10
	yes	1d	–	–
AIS grade*	A	3.39	1.58	7.28
	B	2.56	0.91	7.19
	C	2.15	0.84	5.47
	D	1d	–	–
Level of Injury*	C1–C4	3.29	1.69	6.40
	C5–C8	1.11	0.47	2.64
	T1–L1	1d	–	–

1^d: reference category.

*P is significant if < 0.05.

Table 3. Modifiable factors associated with the occurrence of pressure injuries.

Modifiable factors			95% CI			95% CI
		OR	Lower	Upper	ORadjusted**	Lower	Upper
Urinary tract infections	No	1d	–	–	1d	–	–
	Yes	1.60	0.83	3.09	1.06	0.51	2.23
Pneumonia*	No	1d	–	–	1d	–	–
	Yes	3.13	1.74	5.62	2.13	1.10	4.11
Surgery delay	<24h	1d	–	–	1d	–	–
	≥24h	1.24	0.69	2.22	1.63	0.82	3.23
Admission delay	<12h	1d	–	–	1d	–	–
	≤12h	0.92	0.51	1.67	1.21	0.63	2.31
Daily therapy time (h/day)	0.78	0.12	5.12	0.39	0.43	3.46

OR: odds ratio.

1^d: reference category.

*P is significant if < 0.05.

**adjusted for AIS grade and level of injury.

The mean LOS was 29.3 ± 21.2 days. Analyses were done to identify factors that correlated with a longer acute care LOS. Bivariate analysis of non-modifiable factors was all non-significant (P > 0.07) except for the AIS grade and NLI, which were significantly associated with the LOS (P < .001). The final model of hierarchical multivariate linear regression analysis included two predictors: AIS grade and NLI (P < .001). The occurrence of PI, pneumonia and urinary tract infections as well as lower daily therapy time were significantly associated with longer acute care LOS in bivariate analyses and when adjusted for AIS grade and NLI (P < .001) (Table 4).

Table 4. Modifiable factors associated with acute care LOS.

Modifiable factors	Unadjusted coefficients		Adjusted coefficients**
	Beta	P value	Beta	P value
Pressure injuries	0.30	<.001*	0.20	<.001*
Urinary tract infections	0.33	<.001*	0.29	<.001*
Pneumonia	0.44	<.001*	0.37	<.001*
Surgery delay	0.04	0.52	0.11	0.11
Admission delay	−0.14	0.02*	−0.09	0.14
Daily therapy time (h/day)	−0.39	<.001*	−0.43	<.001*

*P is significant if < 0.05.

**adjusted for AIS grade and NLI.

Discussion

The optimization of the clinical evolution of TSCI patients during acute care is crucial to promote efficient IFR⁶ and ultimately improve functional outcome.^7–10 Since the prevention of PI and the optimization of the acute care LOS represent serious challenges during the acute rehabilitation phase,⁷ our results shed light on how the medical and rehabilitation teams may better contribute to facilitate the subsequent rehabilitation process.

A total of 21.6% of patients developed at least one pressure injury during the acute-care hospitalization, which is similar to rates reported in other SCI-specialised acute care centers.^{11,14,15,19,31,54} The most common localization was the sacrum, which is also consistent with previous work.⁵⁵

Our results show that more severe TSCI (AIS grade) and higher NLI were significantly associated with the occurrence of PI during the acute care hospitalization. This is consistent with previous studies^{15,20,26,31,56} showing that higher NLI and motor complete TSCI result in greater mobility limitations, impaired microvascular response due to decreased autonomic control and more sensory deficits,^55,57 all of which predisposes to the occurrence of PI.¹⁵ Also, they are at higher risk of developing nutritional deficiencies due to dysphagia,² greater muscle mass loss and increased fat mass below the level of injury which further increases PI risk.^58,59 Finally, patients with a higher NLI are more prone to respiratory compromise with decreased airway secretion clearance predisposing them to pneumonia which in turn, leads to higher PI risk.^20,60

We identified that, when controlled for AIS grade and NLI, patients who develop pneumonia during the acute care hospitalization had a 2-fold increased risk of developing a PI. The clinical interpretation of this result is however complex, as these two factors are most likely inter-related. Indeed, risk factors of PI and pneumonia in the SCI population are both related to the level and severity of the injury.⁶¹ It was also suggested in previous work^15,60 that hypoxia and hypercapnia caused by pneumonia compromises tissue oxygen thus, predisposing patients to PI.⁶² In addition, the occurrence of pneumonia may lead to an impaired inflammatory response as well as decreased mobility, further increasing PI risks.¹⁵ Indeed, it has been shown that the presence of pneumonia leads to changes in inflammatory mediators, such as increase in plasma TNF-a, and decrease in urine TNF-a, GM-CSF, and IL-15, compromising patients’ immunity, predisposing them to PI occurrence.^60,63 In our cohort, information regarding timing of the occurrence of medical complications during acute care was collected as of 2014. As of 2014, for the patients who developed both a pneumonia and a pressure injury during their acute care LOS, pneumonia preceded pressure injury occurrence in all cases, thus supporting previous work. In light of our results, more aggressive prevention strategies should be considered for patients with higher NLI, more severe TSCI and sustaining (or at higher risk of developing) pneumonia. Along the same lines, future studies may focus on the development of new PI prevention strategies, as despite the presence of specialized care, PI incidence remains worrisome. Perhaps further screening protocols could be used to stratify TSCI patients according to their risk of developing PI during acute care.

When adjusted for AIS grade and NLI, the occurrence of medical complications (PI, urinary tract infection and pneumonia) was significantly associated with a longer acute care LOS. Our results support previous work,^{6–9,22–24,64–66} while highlighting the importance of medical complication prevention and early transfer to SCI-specialized trauma centers. Indeed, SCI-specialized trauma centers generally comprise a coordinated dedicated multidisciplinary team who better understands the specifics of comprehensive management of TSCI.^67,68 Timely initiation of specific rehabilitation protocols, the accurate estimation of the outcome and the appropriate prediction of the resources required for rehabilitation represent key factors related to specialized trauma centers fostering prevention of acute complications following TSCI.

Lesser daily therapy time was also associated with a significant longer acute care LOS. Lesser daily therapy time generally occurs in individuals with greater morbidity, higher vulnerability to medical complications, higher psychological distress and/or pain,⁶⁹ which may also contribute to longer LOS.^7,11 This result supports previous work showing that a higher quantity of rehabilitation therapies is associated with shorter LOS.⁷⁰ However, while previous work focused on the subacute to chronic period,⁷⁰ our study is the first to show this relationship during the acute care hospitalization, where the TSCI population presents very distinctive characteristics related to their acute state and postoperative status. Previous studies have also showed a positive relationship between quantity of rehabilitation therapies during IFR and functional outcome.^70,71 Thus, perhaps by increasing the volume of rehabilitation interventions during acute care we could not only decrease LOS but also potentially improve functional outcome early in the rehabilitation process as suggested by a recent study conducted on patients with acute conditions.⁷² Accordingly, in addition to prevention of complications, rehabilitation interventions during acute care should pay attention to providing adequate pre-requisites for IFR, for instance working on endurance and balance training. Early repetitive training may also optimize motor skill learning, which can be highly valuable in the early recovery process.^44,50 Future studies should investigate these benefits and the safety and cost-effectiveness of such rehabilitation strategies. Finally, although therapeutic interventions follow current guidelines,⁴⁰ the actual duration of therapy a patient receives during acute care depends on several factors. Indeed, sometimes logistical or medical problems occur, causing a decrease in therapy time. For example, the occurrence of medical complications, as well as incontinence and neurogenic bowel management may disrupt compliance with the planned therapy schedule.⁶⁹ Similarly, therapy sessions may be missed due to medical tests or psychosocial issues.⁶⁹ At our institution, patients are typically scheduled to get one hour of PT and about 30 min of OT per day. However, due to logistical and medical problems, therapies are often shortened or missed. Thus, patients get on average one hour of therapy total (OT and PT) per day, including indirect interventions such as charting, far less than planned. Thus, we believe that rather than modifying existing interventions or focusing on specific interventions, we should instead prioritize reducing the number of therapies shortened or missed during acute care. To do so, emphasis should be put on optimizing logistical approaches to planning therapies, patient transportation and patient, therapists and equipment preparation, while emphasizing the importance of punctuality. As suggested by Hammond et al., such strategies could include more staff at key moments during the day such as during the morning routine, more efficient scheduling of medical tests and better screening and management of secondary complications at admission and throughout the acute care stay.⁶⁹ Consequently, a future study should evaluate the impact of missed therapies on functional outcome at discharge from acute care. Subsequently, future work should evaluate the impact of proposed strategies on the number of missed therapies during acute care.

Study limitations

Our study should be interpreted in the context of specific study limitations. First, it was conducted in a single Level-1 trauma center, which may limit external validity as SCI-specialized trauma centers’ specificities may vary from one country to another. Information regarding other potential factors was not collected, such as the occurrence of spasticity, which may influence the occurrence of PI during inpatient rehabilitation⁷³ and mean arterial pressure as hypotension has been shown to be a strong predictor of pressure injury in acute care.⁷⁴ Finally, our study focused on physical and occupational therapy only, thereby neglecting the time spent with other caregivers who also actively participate in the acute rehabilitation process and PI prevention. Future work should consider evaluating the impact of all members of the rehabilitation team.

This study showed a higher incidence of PI stage 2 as compared to stage 1 (63.1% vs. 29.2% of stage 2 and 1 respectively). As a stage 1 represents a lesser severity of the skin breakdown, it could have been expected to find it in a higher percentage. However, this result is consistent with other studies^75,76 and may reflect challenges in diagnosing stage 1 PI, which may be underestimated.^77,78 Nevertheless, our study showed a PI incidence similar to previous studies among the acute TSCI population.^{11,14,15,19,31,55,76}

Conclusion

Preventing the occurrence of pressure injury (PI) and decreasing the acute care length of stay (LOS) are crucial challenges for the acute rehabilitation team, as these factors significantly impact subsequent rehabilitation outcomes. This study aimed at identifying modifiable factors, that the rehabilitation team may address, associated with the occurrence of PI and longer acute care LOS. Our results suggest that patients sustaining pneumonia may be at higher risk of developing a PI, independently of the level and severity of the traumatic spinal cord injury (TSCI). Thus, greater attention towards the prevention of PI should be given to individuals at higher risk of pneumonia, which both may be related to higher level and severity of the injury. Developing new protocols for PI prevention is highly recommended as the incidence of PI remains worrisome despite management in a trauma specialized in SCI care.

Occurrence of medical complications (PI, urinary tract infection and pneumonia) were associated with longer acute care LOS. We recommend early transfer to a specialized trauma center, to ensure appropriate and aggressive medical complications prevention strategies. Since higher daily therapy time in physical and occupational therapy was associated with decreased acute care LOS, we would suggest that additional therapy interventions may be considered to promote effective transfer to subsequent rehabilitation phases and ultimately higher functional outcomes. However, future studies should evaluate the safety as well as cost-effectiveness of such rehabilitation strategy. Finally, we believe that rather than modifying existing interventions or focusing on specific interventions, we should instead prioritize reducing the number of therapies shortened or missed during acute care.

Disclaimer statements

Contributors None.

Funding This works was supported by Fonds de recherche du Québec – Santé (FRQS) under Grant #35370 and part of this data was collected through the Rick Hansen Spinal Cord injury Registry.

Conflicts of interest The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.