Posterior cord syndrome: Demographics and rehabilitation outcomes

Abstract

Context/Objectives: To describe demographics, clinical characteristics, and functional outcomes of patients with incomplete spinal cord injuries and posterior cord syndrome (PCS).

Design: Five-year retrospective case series.

Setting: Spinal cord injury (SCI) rehabilitation unit at a Level 1 tertiary university medical center.

Participants: 9 patients with incomplete cord injuries diagnosed with PCS admitted to rehabilitation within the past 5 years.

Outcome measures: Functional Independence Measure (FIM) motor scores, length of stay (LOS), discharge disposition.

Results: Incidence of PCS was 2% with an average age of 62.0 years. The most common etiology for PCS was spinal cord compression from localized tumors (78%). Seven (78%) patients had paraparesis. All patients had an American Spinal Injury Association impairment scale (AIS) classification of AIS D. SCI-related complications most commonly included: neuropathic pain (78%), spasticity (44%), and neurogenic bladder (78%). Average LOS on the rehabilitation unit was 28 days. Average admission and discharge FIM motor scores were significantly improved (P = 0.001) from 41 to 65, respectively. Two-thirds (67%) of patients were able to walk at least 150 feet with a rolling walker prior to discharge. Most (78%) patients were discharged to home. Continence improved from admission to discharge from 22% vs 56% (bladder) and 67% vs 78% (bowel).

Conclusions: We can conclude that PCS most often results in paraparesis due to tumor compression. Typical SCI-related medical complications are encountered. These patients often experience significant functional improvements during SCI rehabilitation with the majority also having bladder and bowel continence allowing them to return home at discharge.

Introduction

It is estimated that the annual incidence of spinal cord injury (SCI) is approximately 54 per million in the U.S. with approximately 18,000 new cases each year.¹ SCI is associated with loss of strength, sensation and bladder/bowel control which can lead to a functional decline in mobility and self-care skills. Additionally, potential SCI-related medical complications such as pain, spasticity, pressure ulcers, and infections. This can lead to significant neurological and functional impairment, often requiring acute inpatient rehabilitation before returning to the community.

Posterior Cord Syndrome (PCS) is the least common of the SCI incomplete clinical syndromes, with an incidence of about 1%.^2,3 PCS is described as a selective dysfunction of the posterior columns of the spinal cord, resulting in impairments of proprioceptive and vibration sense, two-point discrimination, and to a more variable extent, sense of deep touch and pressure below the level of injury.^2,4

Despite the preservation of motor strength, the loss of proprioception and position sense in PCS can lead to significant compromise of mobility and self-care abilities. Nevertheless, there appears to be limited information in the literature on the functional outcomes and SCI-related medical complications associated with PCS.³ In general, patients with more incomplete injuries, including the various SCI syndromes have been associated with shorter rehabilitation length of stays and higher measurements of functional independence at discharge.^5–9 Common SCI-related secondary medical complications such as neurogenic bladder and bowel, neuropathic pain and spasticity can lead to the need for additional clinical care and patient education in order to address potential long-term issues.

The objectives of this study will be to describe the demographics, injury characteristics and functional outcomes of a series of patients with PCS admitted to an inpatient rehabilitation unit with a diagnosis of PCS. It is hoped that results of this study may better assist in understanding this patient population along with their medical and functional outcomes, which might further enhance clinical care and rehabilitation outcomes for this patient population.

Methods

Subjects

Data was reviewed from four hundred consecutive patients with SCI who were admitted to an acute rehabilitation unit of an urban tertiary care medical center, between January 1, 2012 and December 31, 2016 (5 years). Patients were examined and classified according to the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI)¹⁰ standards. They were further identified as having a diagnosis of PCS if their examination noted loss of proprioceptive and vibration sense with preservation of muscle strength, temperature and pain sensation caudal to the neurological level of injury.

Measures

Demographics, injury characteristic and functional outcome data were reviewed for each patient. Demographic characteristics included age, sex, and ethnicity. Injury characteristics included etiology of injury, neurological level of injury (NLOI) and ISNCSCI American Spinal Injury Association Impairment Scale (AIS) classification. Additionally, pre-existing and SCI-related medical complication information was collected through a retrospective chart review.

Functional outcomes included rehabilitation length of stay (LOS), discharge disposition, and functional status as assessed via the Functional Independence Measure (FIM).¹¹ The primary outcome measure of the study was the Functional Independence measure (FIM). The FIM motor and FIM cognitive subscales were also assessed. FIM change scores were calculated based on the difference between rehabilitation admission and discharge. FIM efficiency scores were calculated by dividing change scores by the respective duration of stay in rehabilitation. Lastly, the FIM efficiency was calculated as the change in FIM total between rehabilitation, admission and discharge, divided by the length of stay (LOS) in the rehabilitation care setting.

FIM motor scores range from 13 to 91 with higher scores reflecting greater levels of independence. Self-care scores reflect level of independence in eating, grooming, bathing, upper and lower body dressing, and toileting. Sphincter control scores reflect level of independence with bladder and bowel management. Mobility scores reflect level of independence with regards to bed/chair/wheelchair transfers, toilet transfers, and tub/shower transfers.

Statistical Methods

Demographic, injury, and FIM information were summarized by medians, ranges, and interquartile ranges (IQR). The total FIM and the cognitive and motor subscales were summarized separately at admission and discharge. A repeated measures ANOVA was used to characterize the improvement of the total, cognitive, and motor FIM between admission and discharge. An ANOVA model was used to determine differences in the FIM efficiency scores.

Results

Incidence, demographics and injury characteristics

Of the 400 patients admitted for SCI inpatient rehabilitation during this 5-year review, nine patients (2.3%) were diagnosed with posterior cord syndrome and included in this study. Average age at onset of individuals with PCS was 62.0 years (range 49–76 years). There were 5 (55%) male and 4 females and ethnicity included: Caucasian (n = 4, 44%), African American (n = 4, 44%), and Asian (n = 1, 11%) (Table 1).

Table 1. Demographic characteristics.

Variable	Category	Posterior cord (N = 9)
Age (median) (IQR)		62.0 (52–69)
Sex	Male (%)	55%
	Female	44%
Ethnicity	African American	44%
	Caucasian	44%
	Asian	11%

The most common etiology for PCS was noted to be due to spinal cord compression from tumors (n = 7, 78%) located in the cervical (n = 1), thoracic (n = 5), and lumbar (n = 1) regions (Table 2). Other etiologies included thoracic epidural abscess (n = 1) and cervical spinal stenosis (n = 1). Seven (78%) patients developed paraparesis vs. two (22%) with tetraparesis. The neurological level of injury at admission for patients with SCI syndromes ranged from C7 to L1 with two-thirds (67%) having a thoracic level. All patients (100%) with PCS were classified as AIS D, motor incomplete, by ISNCSCI at admission and at discharge from rehabilitation.

Table 2. SCI etiology and functional changes.

Patient	Age/Sex	Etiology	NLOI/AIS	Rehab LOS	FIM Change	Disposition
1	58 M	Thoracic epidural abscess	T8 ASIA D	6	51	Home
2	52 F	Intramedullary spinal cord ganglioma	T7 ASIA D	29	35	SNF
3	76 M	Cervical epidural tumor from metastatic lung adenocarcinoma	C7 ASIA D	45	8	SNF
4	69 M	Thoracic epidural tumor due to B-cell lymphoma	T5 ASIA D	28	20	Home
5	62 F	Cervical myelopathy due to severe cervical spinal stenosis	C8 ASIA D	28	49	Home
6	49 M	Thoracic cord impingement due to metastatic prostate cancer	T11 ASIA D	34	22	Home
7	70 F	Thoracic cord impingement due to intradural meningioma	T11 ASIA D	23	44	Home
8	69 M	Thoracic cord impingement due to metastatic NSCLC	T4 ASIA D	22	14	Home
9	49 F	Spinal cord ependymoma	T3 ASIA D	34	35	Home

Commonly noted SCI-related complications for these individuals included: neuropathic pain (78%), spasticity (44%), neurogenic bladder (78%), neurogenic bowel (33%), and urinary tract infection (UTI) (33%).

Functional outcomes

Median LOS was 11 days (range 6–15) on acute care and 28 days (range 6–45) in rehabilitation. Average admission and discharge FIM motor scores were 28 and 56, respectively (Table 3, Figs. 1–3). Patients at discharge had significantly higher total and subscale FIM scores. On average, the FIM scores were 27 points higher at discharge than admission (SE = 6, P < 0.001). Similar results were obtained when assessing the motor (Diff = 22, SE = 6, P = 0.001) and cognitive FIM scores (Diff = 5, SE = 2, P = 0.023). This information is presented graphically in Figs. 1–3. Ceiling effects, or values of 6 or 7 on each of the FIM items contributing to the FIM subscales, were present in 2 subjects at admission and 2 additional subjects (i.e. 4 total) at discharge for the cognitive subscale.

Figure 2.

Change in FIM motor scores from admission to baseline. Each narrow line represents a single subject while the average over the entire cohort is represented by the thick gray line.

Table 3. Length of stay, functional outcome, and disposition.

Characteristic	Time	Median [IQR]
Age		62.0 [52.0, 69.0]
LOS (Acute) (days)		11.0 [7.0–11.0]
LOS (Rehab) (days)		28.0 [23.0–34.0]
FIM Efficiency		0.9 [0.5, 1.6]
FIM Total	Admission	69.0 [60.0, 74.0]
FIM Total	Discharge	100.0 [86.0, 105.0]
FIM Total Change (Avg)		27 (P < 0.001)
FIM Motor	Admission	41.0 [37.0, 44.0]
	Discharge	65.0 [54.0, 71.0]
FIM Motor Change (Avg)		22 (P = 0.001)
FIM Cognitive	Admission	29.0 [28.0, 31.0]
FIM Cognitive Change (Avg)	Discharge	33.0 [33.0, 35.0] 5 (P = 0.023)
Discharge Disposition	Home Skilled Facility	78% 22%

Notes: Characteristics are summarized by medians and interquartile ranges (IQR).

Figure 1.

Change in total FIM scores from admission to baseline. Each narrow line represents a single subject while the average over the entire cohort is represented by the thick gray line.

Figure 3.

Change in FIM cognitive scores from admission to baseline. Each narrow line represents a single subject while the average over the entire cohort is represented by the thick gray line.

Continence at admission and discharge was 22% vs 56% (for bladder) and 67% vs 78% (for bowel). Six (67%) patients were able to walk at least 150 feet with a rolling walker at time of discharge. Seven of the nine patients (78%) were discharged to home, two to a skilled facility.

Discussion

Posterior cord syndrome has a somewhat rare occurrence which, no doubt, relates to the paucity of PCS outcome-related literature. The incidence of PCS noted in this study (about 2%) is similar to that cited in past studies.^2,3 Despite this low occurrence rate, individuals with PCS can have significant functional impairments and SCI-associated medical complications, and often require admission for acute inpatient rehabilitation.

The posterior columns of the spinal cord contain long ascending fibers carrying sensory stimuli relating to the direction of movement and position of joints, vibration, and to some extent pressure and touch. Such stimuli from the sacral, lumbar and lower thoracic areas of the body ascend in the fasciculus gracilis (more medial) tract, which terminates in the gracilis nuclei in the rostral cervical cord and medulla. The stimuli from the upper thoracic area of the body and the upper extremities ascend up the fasciculus cuneatus (more lateral) tract which terminates in the cuneatus nuclei, also in the rostral cervical cord and medulla. These tracts then synapse with respective fibers that cross the midline and ascend up the medical lemniscal tracts in the brainstem, ultimately terminating in the posterior thalamic nuclei.

This study notes that individuals with PCS are older than the general SCI population, with a median age of 62 years. The difference in age found between the general SCI population and patients with PCS is likely a reflection of the underlying etiologies associated with PCS. SCI secondary to tumors and spinal stenosis have been primarily seen in older individuals, and they accounted for 89% of all SCI syndrome cases in this study.^12–16 Additionally, nearly half were female, which is also more consistent with the non-traumatic etiologies that led to these cases of PCS.

The etiology of PCS, in this study, was most often related to spinal cord tumors as mentioned, however, we also noted cases associated with spinal stenosis and with epidural abscess. In each of these cases, it is suspected that SCI was caused by these nearby masses compressing the posterior aspect of the spinal cord with resultant cord edema or occlusion of the posterior spinal artery (PSA). Prior literature has noted PCS to be linked with a variety of causes^17–21 where interruption of PSA blood flow is often seen. The PSA supply the posterior 20–30% of the spinal cord, including the posterior columns, the posterior aspect of the dorsal horns, and a border zone partially involving the corticospinal and anterolateral tracts. The anatomy of the vasculature accounts for the profound proprioceptive deficits and the variation in weakness and sensory loss typically seen in PCS.²²

The etiologies (cancer, epidural abscess or spinal stenosis) seen in this study comprise those associated with non-traumatic (rather than traumatic) SCI which have been shown to make up about one-third of all SCI.^14,23 Individuals with non-traumatic SCI are more likely to be older and with more incomplete neurological involvement, as was seen with individuals in this study. Completeness of injury and neurological level of injury are important factors in predicting functional outcome after SCI.^24–26 Incomplete spinal cord injuries tend to indicate a more favorable neurological and functional prognosis. All patients in this study had motor incomplete (AIS D) injuries which indicate potential for more favorable functional outcomes. The neurological level of injury (NLI) were most often thoracic, which correlated to the vertebral levels of tumor locations and is consistent with available literature as thoracic NLI has been cited as being the most common level for cancer-related SCI.

Medical complications seen in this study included: neuropathic pain (78%), spasticity (44%), neurogenic bladder (78%), neurogenic bowel (33%), and urinary tract infection (33%). These complications are representative of common SCI-associated complications in patients with SCI.²⁷

Functional outcomes following SCI syndromes

An important objective of acute SCI rehabilitation is to maximize patient functional outcomes in mobility, self-care and bladder/bowel control. The FIM is a functional assessment tool that provides the measurement of these categories. In this study, results show an increasing trend in the FIM scores during the rehabilitation LOS of PCS patients. Even though the FIM measures do not have a symmetric distribution, the statistical analysis proceeded as if this assumption were true. Statistical methodology assuming the data originated from a gamma or log-normal distribution, which are appropriate for skewed distributions, were attempted but did not find any meaningful differences compared to the analysis presented. Consequently, the simplest analysis was utilized to enhance the interpretability of the results.

The average rehabilitation LOS for patients with SCI has continued to decline over the years.^1,25,28 It has been reported that the average rehabilitation LOS for the general SCI population is 45 days.¹ This study shows that patients with PCS have shorter LOS than the general SCI population (28 vs 45 days). This is likely secondary to the higher percentage of incomplete injuries found in individuals diagnosed with an SCI syndrome. Additionally, the dual etiology of SCI and cancer may have contributed to a more challenging rehabilitation management course for the majority of these patients, as cancer-related issues (diagnostic work-up, cancer pain, fatigue, depression) may have contributed to longer acute care and rehabilitation LOS.

Functional improvements were noted in the patients allowing for successful discharge disposition following rehabilitation. The majority of patients (67%) in this study were ambulatory to a distance of at least 150 feet prior to discharge from rehabilitation. In addition, bowel and bladder continence improved from admission to discharge (67% to 78% and 22% to 56%, respectively). 78% of patients in this study were discharged home, indicating that significant functional gains were achieved during admission to allow a high level of independence after comprehensive rehabilitation.

Conclusions

The rare occurrence of posterior cord syndromes has led to a paucity of literature regarding functional outcomes research. A greater understanding of the overall incidence, functional impairment, rehabilitation functional outcomes and medical complications in these patients is necessary and may allow for enhanced functional expectations. Future research needs to further define long-term outcomes following discharge.

Limitations of this study

The retrospective nature of this study limited the available data collected regarding functional outcomes to solely admission and discharge measures. In the future, a prospective study would allow more information to be gathered including functional ambulatory and ADL status at the time of injury or diagnosis and then throughout the hospital course, including during inpatient rehabilitation and at follow up visits after discharge. This would allow tracking of changes to FIM scores and other measures over time to see if functional improvements could be maintained after rehabilitation programs. A larger sample size, perhaps including patients from multiple institutions, would further improve data quality and accuracy of outcome measures.

Disclaimer statements

Contributors None.

Funding None.

Conflicts of Interest None.