Prevalence of depression, fatigue, and sleep disturbances in patients with myelopathy: Their relation with functional and neurological recovery

Abstract

Objectives

To observe the prevalence of fatigue, depression, and sleep disturbance in patients with myelopathy and their correlation with neurological and functional recovery.

Patients and methods

Study conducted in a university tertiary research hospital with 127 patients with myelopathy (92 males) admitted to neurorehabilitation unit between January 2010 and December 2013. Mean age was 32.71 ± 13.08 years (range 15–65 years), and mean duration of injury was 76.22 ± 82.5 days (range 14–365 days). Functional status and impairments were assessed using Barthel Index and Spinal Cord Independence Measures. Depression, fatigue, and sleep disturbances were assessed using Hospital Anxiety and Depression Scale, Fatigue Severity Scale, and Pittsburgh Sleep Quality Index scales, respectively. Neurological recovery was assessed using American Spinal Injury Association's impairment scale.

Results

Forty-four out of 104 (42%) patients had borderline or confirmed depression, 36/108 (33%) had significant fatigue, and 62/106 (58%) had significant sleep disturbances at admission. Significant correlation was observed between change in fatigue and depression scores with change in functional status scores (P < 0.05) but no correlation (P > 0.05) between change in sleep disturbance scores and functional status score and neurological recovery (P > 0.05). Similarly, change in fatigue and depression scores had no correlation with neurological status improvement. Fatigue, depression, and sleep disturbance scores showed significant improvement, that is, admission vs. discharge scores (P < 0.05) with significant correlation between improvement in all three variables (P < 0.05).

Conclusions

Study variables showed significant improvement in the present study with myelopathy patients but not necessarily correlating with functional and neurological recovery.

Introduction

Depression is a mood disorder characterized by a sense of inadequacy, despondency, decreased activity, pessimism, anhedonia, and sadness where these symptoms severely disrupt and adversely affect the person's life.¹ It is prevalent in about 25% of patients with myelopathy and impairment and disability associated with depression is equal to that attributed to cardiovascular disease, and greater than that caused by other chronic physical disorders such as hypertension, diabetes mellitus, and arthritis.^2,3

“Fatigue is a total body feeling and experience, encompassing physical, cognitive and emotional dimensions; an odious and unpleasant experience which causes distress; a chronic and unrelenting phenomenon and a subjective experience dependent upon an individual's perceptions.”⁴ In contrast to 4–9% of the general population, fatigue is present in up to 57% of patients with myelopathies and is more common in patients with incomplete injuries.⁵

Patients with myelopathies commonly have excessive daytime sleepiness and disturbed sleep, including reduced rapid eye movement (REM) sleep, increased need for sleep medications, snoring, and sleep-apnea syndrome.^6–8 Sleep disturbances such as sleep disordered breathing or obstructive sleep-apnea syndrome occur in 40–45% of patients.⁹

With increasing survival of myelopathy patients due to better medical and surgical management in the initial stages, prevalence of these conditions is set to increase. Identification and management of secondary complications and comorbidities assume an important role in this scenario. All three conditions—depression, fatigue, and sleep disturbances—can be considered as significant comorbidities affecting patients with myelopathies to a large extent and tend to interfere with the subsequent lifestyle, quality of life, and well-being. Fatigue and depression are among those comorbidities, which are identified as research priorities by people with myelopathy themselves.¹⁰ The objective of this study is to observe the prevalence of these conditions in patients with myelopathies who are admitted to rehabilitation department in various stages (acute, subacute, and chronic) and their correlation with neurological and functional recovery in these patients during hospital stay. We also wanted to observe if any correlation exists between these conditions with each other.

Patients and methods

This prospective descriptive study included patients with myelopathy admitted to neurological rehabilitation department between January 2010 and December 2013 in a tertiary university research hospital in India. Prior approval for the study was taken from the institute's ethics committee. Informed consent was obtained from all the patients to participate in the study. Patients with confirmed myelopathy (who had undergone CT or MRI scan imaging) of traumatic and nontraumatic origin with monophasic insult were included. Patients medically fit to participate in rehabilitation program were included in the study. Patients with multiple sclerosis, having cardiovascular illness, unstable vertebral injuries, requiring ventilatory support were excluded. There were three sources of patient recruitment: (1) patients reporting in outpatient department of neurological rehabilitation department, (2) patients admitted and referred from department of neurology, and (3) patients admitted and referred from the department of neurosurgery. All the patients meeting the inclusion criteria were recruited in the study. Their functional status and impairments were assessed using Barthel Index (BI) and Spinal Cord Independence Measures (SCIM). Depression, fatigue, and sleep disturbances in the patients were assessed using Hospital Anxiety and Depression Scale (HADS), Fatigue Severity Scale (FSS), and Pittsburgh Sleep Quality Index (PSQI) scales, respectively. All scales were used both at admission and discharge. Other details regarding diagnosis, duration of injury, and neurological level of injury (using American Spinal Injury Association (ASIA) impairment scores) were also recorded.

Outcome measures

The BI is a 10-item ordinal scale (range: 0–100) with ratings for feeding, moving from wheelchair to bed and return, grooming, transferring to and from a toilet, bathing, walking on a level surface, going up and down stairs, dressing, and continence of the bowels and bladder. BI has adequate inter-rater agreement, internal consistency, and constructs validity in myelopathy patients.¹¹

The SCIM is the only comprehensive ability rating scale that has been designed specifically for patients with spinal cord lesions. The instrument focuses on performing everyday tasks, and captures the economic burden of disability as well as the impact of disability on the patient's overall medical condition and comfort with validity and reliability.¹²

The HADS was developed by Zigmond and Snaith in 1983 to identify “caseness” (possible and probable) of anxiety disorders and depression among patients in nonpsychiatric hospital clinics. It is divided into an anxiety subscale (HADS-A) and a depression subscale (HADS-D), both containing seven intermingled items. Patients scoring up to 7 points are considered to be not having any anxiety or depression, scores between 8 and 10 suggestive of borderline illness whereas scores of 11 or more suggestive of moderate to severe anxiety and depression.¹³

The FSS is a 9-item measure of the severity of fatigue. It requires the participant to choose the degree of agreement on a 7-point ordinal scale ranging from 1 (strongly disagree) to 7 (strongly agree). Scores are calculated by deriving an arithmetic mean. Cutoff scores of over 4 are indicative of significant fatigue. The FSS has acceptable reliability with regard to internal consistency, test–retest reliability, and validity for use in persons with SCI.¹⁴

The PSQI consists of 19 self-rated questions, which assess a wide variety of factors relating to sleep quality, including estimates of sleep duration and latency and of the frequency and severity of specific sleep-related problems. These 19 items are grouped into 7-component scores, each weighted equally on a 0–3 scale. The 7-component scores are then summed to yield a global PSQI score, which has a range 0–21 with higher scores indicating worse sleep quality. A score of >5 indicates impaired sleep quality.¹⁵

ASIA impairment scale (AIS) was used to assess the neurological status/recovery in all cases. Patients were assessed by the senior residents in the unit well acquainted and trained in using AIS.

Rehabilitation program

It included medical treatment, management of complications along with physical therapy, occupational therapy, providing orthoses for upper and lower limbs, counseling for psychological issues such as low mood, aggression, coping strategies for nociceptive and neuropathic pain, insomnia, and fatigue. Medications for sleep disturbance were not routine but if other measures such as activity scheduling and progressive muscle relaxation would fail, patients were given medications such as benzodiazepines and zolpidem. Physical therapy consisted of a range of motion exercises, strengthening exercises for limbs and trunk, gait training in the form of conventional gait training, and body weight support treadmill training (as and when indicated), balance assessment and training (Biodex Balance Master), and electrotherapy. Occupational therapy included activity of daily living training, functional ability training, fine motor hand skills training, coordination exercises, sensory, and proprioceptive reeducation. Assistive adaptive devices, barrier-free environment at home and working place, and modification orientation training were provided to the patients. Quadriplegic patients showing distal upper limb and hand recovery were trained with hand robot (Amadeo). Orthotic management included providing patients with limb orthoses. The majority of patients required short lower limb orthoses (ankle foot orthoses). Knee gaiter, wrist cock-up splints, resting splints, and tenodesis splints were also provided to patients for prevention/correction of limb deformities.

Statistical analysis

Data were collected and managed with SPSS version 15 (SPSS Inc., Chicago, IL, USA), and analysis was also done on the same software. Data were tested for normal distribution, and suitable parametric or nonparametric tests were used. For comparison of BI and SCIM at admission and discharge, paired t-test was used, and for comparison of nonparametric data such as FSS, PSQI, and HADS, Wilcoxon rank-sum was used. For finding correlation between the change in HADS, FSS, and PSQI scores and functional improvement, Spearman's rho was used. For testing the change between admission and discharge in ASIA scores, McNemar–Bowker test was used and for comparing ASIA improvement and improvement in functional scores and study variables, Mann–Whitney test was used. Statistical significance was considered at P-value < 0.05.

Results

A total of 127 patients meeting the criteria were included in the study. The mean age was 32.71 ± 13.08 years (range 15–65 years). Ninety-two patients were males (72.4%) and 35 were females (27.6%). Patients stayed in the rehabilitation for a mean of 39.91 ± 25.43 days (range 14–130 days). Mean duration of lesion was 76.22 ± 82.5 days (range 14–365 days). Most common neurological level of lesion was lumbosacral seen in 45 patients (35.4%) followed by lower dorsal level (D7–D12 level) in 41 patients (32.3%), cervical level in 23 patients (18.1%), and upper dorsal (D1–D6 level) in 17 patients (13.4%).

Sixty-three patients (49.6%) had traumatic etiology and 64 patients (50.4%) had a variety of nontraumatic myelopathies. Among the nontraumatic myelopathies group, 29 patients had acute transverse myelitis (22.8%), 5 had spinal tuberculosis (3.5%), 13 had primary spinal tumors (10.2%), 7 had ossified posterior longitudinal ligament/ligamentum flavum (5.5%), 8 had prolapsed intervertebral disc (6.3%), and 2 patients had spinal vascular disease (1.6%).

All the patients were assessed within 24–48 hours of admission in the rehabilitation unit by a senior resident in-charge of the respective wards using the outcome measures. Similar outcome measures were used to assess patients within 24 hours of discharge. Changes were noted down in the scores for comparison and analysis later on.

The mean scores for BI, SCIM, HADS, FSS, and PSQI at admission and discharge are given in Table 1. There was a significant difference at admission and discharge for all these variables indicating an improvement in functional status, fall in levels of depression and fatigue, and better sleep quality in these patients’ admission vs. discharge. Though the average HADS (depression) score was 6.93 at admission, 44 out of 104 (42%) at admission had scores of 8 or more on the HAD scale indicating borderline or confirmed depression. Similarly, 36 out of 108 (33%) at admission had FSS scores of 4 or more indicating significant fatigue. Poor sleep quality defined as scores higher than 5 was seen in 62 out of 106 patients (58%) at admission.

Table 1.

Mean scores of functional indices and study variables at admission and discharge

	Mean scores (at admission)	Mean scores (at discharge)	P-value
Barthel Index	29.88 ± 18.89	55.45 ± 23.1	<0.001
Spinal Cord Independence Measures	29.27 ± 15.11	53.51 ± 21.42	<0.001
Hospital Anxiety and Depression Scale (depression)	6.87 ± 4.28	4.54 ± 3.72	<0.001
Fatigue Severity Scale	3.12 ± 1.51	2.17 ± 1.22	<0.001
Pittsburgh Sleep Quality Index	6.75 ± 4.5	4.89 ± 3.75	<0.001

The ASIA scores at admission and discharge are provided in Table 2. There was a significant change in the ASIA scores at admission and discharge (P < 0.001), showing significant neurological recovery during stay in the rehabilitation unit.

Table 2.

ASIA scores at admission and discharge

ASIA at discharge
ASIA at admission
	A	B	C	D	Total
A	47	6	4	1	58 (47.1%)
B	0	10	2	6	18 (14.6%)
C	0	0	20	16	36 (29.2%)
D	0	0	0	11	11 (8.9%)
Total	47 (38.2%)	16 (13%)	26 (21.1%)	34 (27.6%)	123

Spearman correlation coefficients were calculated to determine the strength of the associations between continuous variables, because the study variables were not normally distributed. (Kolmogorov–Smirnov test).

Correlation between study variables and functional improvement is given in Table 3.

Table 3.

Correlation of change in study variables with functional improvement

	Change in BI	Change in SCIM
Change in FSS	ρ = 0.166	ρ = 0.283
	P = 0.085	P = 0.031
Change in PSQI	ρ = 0.187	ρ = 0.247
	P = 0.055	P = 0.062
Change in HADS (depression)	ρ = 0.221	ρ = 0.290
	P = 0.024	P = 0.027

ρ = Spearman's rho.

* Bold value indicates statistical significance.

Improvement in fatigue, sleep, and depression scores was compared with each other. We observed significant correlation between fatigue and depression (P = 0.003), fatigue, and sleep disturbance (P = 0.009). Similarly, significant correlation between sleep disturbance and depression was observed (P = 0.001).

We tried to observe the correlation between neurological recovery (ASIA scores) with fatigue, depression, and sleep disturbances statistically. There was no significant correlation between neurological recovery with fatigue (P = 0.997), sleep disturbances (P = 0.694), and depression (P = 0.242).

Discussion

The results show that there was significant improvement in functional status of patients during the period of inpatient rehabilitation. Improvement in functional status did not correlate significantly with improvement in sleep but correlated with decrease in fatigue and depression. An earlier study in the same setup reported no significant correlation between functional recovery and depression in myelopathy patients.¹⁶ These different observations could be attributed to the fact that the current study has relatively larger sample size and more patients had incomplete injuries. Previous studies have reported prevalence of depression over the first year after myelopathies between 23 and 38%.^17–19 A recent study has reported much higher prevalence of depression (63.9%) within the first 6 months following SCI.²⁰ Another recent study has reported 15% of patients with SCI of less than 1 year having major depressive disorder.²¹ Kennedy and Rogers²² demonstrated that in the acute phase of injury, depression scores initially decrease followed by an increase at around 18 weeks, which peaks at around 48 weeks and the longer stay of subjects in the hospital setting, the more likely they are to become depressed. In the current study, we did not try to observe the correlation between prevalence of depression and duration of illness and length of stay in hospital as this was not the objective of the study. The same study showed that functional improvement peaked at 1–3 months post-admission in hospital. The authors reported that functional independence was negatively correlated with depression unlike our study. The positive correlation between functional recovery and depression observed in our study could be attributed to the fact that most of these patients had “reactive depression” occurring as a result of myelopathy and ensuing disability. With patients becoming more and more independent for their activities of daily living following rehabilitation, their depression scores improved. Counseling including coping strategies and antidepressant medications (in some patients) would have played some role in improvement in depression.

A large study with 4976 subjects at 1-year post-spinal cord injury reported that functional improvement was negatively correlated to somatic symptoms of depression.²³ One other study with small sample size reported that functional status indicators significantly improved at discharge as did depression scores, but not significantly.²⁴ A study of 244 people with spinal cord injury revealed that only ASIA D group had higher scores on depression scales,²⁵ whereas our study showed insignificant correlation between neurological recovery and depression scores.

Fatigue following myelopathies is said to be of two types: (i) physical fatigue that occurs in muscles partially or totally paralyzed below the level of weakness and (ii) chronic fatigue, which is a debilitation condition with physical, psychological, and age-related components.²⁶ Less exercise, poor diet, intake of pain medications, and alcohol intake are some of the factors reported to be associated with more severe fatigue scores in myelopathy patients.²⁷ In a study of 41 patients with SCI, Wijesuriya et al.²⁸ found no correlation between fatigue and age, level of injury, and completeness of lesion. Only duration of injury negatively correlated with fatigue scores. Present study also found no correlation between improvement in neurological status and change in fatigue scores, although significant correlation was observed between fatigue and functional abilities. We hypothesize that this correlation could be because of the psychological component as well as improvement in the activity of daily living with various rehabilitation strategies adopted. Depression, pain, effect of medications, poor sleep quality, posture, diet, and effort required to accomplish tasks are factors most consistently associated with fatigue in spinal cord injured patients.²⁹ Nearly one-third patients in the present study reported significant fatigue (FSS ≥ 4), which is lower as compared with some earlier studies, which have reported fatigue prevalence to be more than 50% in this patient population.^30,31 It has to be mentioned though these studies used different outcome measures for fatigue assessment and prevalence when compared with our study. Jensen et al.³² reported fatigue in more than two-third SCI patients with a healthy sample size. Authors observed that fatigue was negatively correlated with social integration and psychological functioning.

DiPiro et al.³³ observed disabling fatigue in 10.3% and moderate to severe depression in 21.2% in a study with 652 myelopathy patients. Disabling fatigue and use of support of other people for ambulation were found to increase the odds of having depression in this group. Lidal et al.³⁴ reported that prevalence of fatigue was 25% among patients who had lived with SCI for more than 20 years. Same study reported direct correlation between high depression scores with high fatigue scores, which is in agreement with our study. Anton et al.³⁵ reported high prevalence of fatigue in ASIA complete and motor-complete myelopathy patients, 1-year post-lesion. They also observed significant correlation between fatigue and depression. Another study with myelopathy patients with a two and a half-year duration of lesion observed significant correlation between fatigue and depression.³⁶ Interestingly, myelopathy patients indulging in heavy physical activity report lesser fatigue.³⁷ In other words, individuals who are at a higher level of functioning have lesser fatigue. Same study also reported that individuals who were even mildly active had lesser depression scores. Similar observations have been made in our study. Some other studies have reported fatigue correlates not only with depression but also with pain in individuals with myelopathy.^38,39

Higher prevalence of poor sleep quality has been demonstrated in myelopathy patients.⁴⁰ Sleep disturbances are reported to be due to the injury itself per se, decreased melatonin levels after the injury, more common in tetraplegics and incomplete injuries, due to pain and spasm and in patients with bladder incontinence.⁴¹ Difficulty in falling asleep, night-time awakenings, sleep apnea, snoring, daytime sleepiness, and using prescription pills for inducing sleep are all common in this population.⁴² Periodic limb movements and restless leg syndrome have also been observed in these patients, particularly ASIA A individuals.⁴³ Older patients tend to have fewer sleep dysfunctions when compared with younger group and duration of lesion and age at onset are not related with sleep dysfunctions.⁴⁴ Patients with sleep disturbance/insomnia (28.2%) and depressive symptoms (25.2%) tend to have disturbing lesion-related neuropathic pain.⁴⁵ A study of 100 patients who had completed more than a year after SCI showed that poor sleep quality was present in about 38% of the patients, and there was no correlation between sleep quality and level of injury, which agrees with our data.⁴⁶

We believe that our study is unique in bringing out correlation between improvement in sleep quality, fatigue, and depression scores during in-patient rehabilitation. Our patients significantly improved in all three parameters, and improvement in fatigue scores was significantly correlated with improvement in sleep quality and depression scores. Similarly, improvement in sleep quality scores was significantly correlated to improvement in depression scores in these patients. We hypothesize that there is correlation between all three variables in myelopathy patients, and it is impossible to pin point which variable triggers or exacerbates the other. Strategies including physical and psychosocial rehabilitation are complimentary and contribute to the improvement in depression, fatigue, and sleep disturbances individually and collectively. This study did not assess the effect of improvement in depression or fatigue (with intervention in the form of counseling or pharmacological treatment) on better and positive functional outcome. Further study is recommended to observe this causal relationship between the study variables.

Limitations of the study

The study had relatively smaller sample size considering multiple variables. The study was conducted with heterogeneous group of myelopathy patients in terms of etiopathology and duration of illness.

The data cannot be generalized as it was a single-center study. The study could have been more informative in terms of sociodemographic information such as education, marital, and vocational status of the patients.

Conclusions

High prevalence of fatigue, depression, and sleep disturbance was observed in the myelopathy patients. But these variables showed significant recovery in the present study when admission and discharges scores were compared. All patients showed significant neurological and functional recovery at the time of discharge. Each of these variables was found to be statistically insignificantly correlated with neurological recovery, but improvement in fatigue and depression significantly correlated with functional improvement. Improvement in all three variables significantly correlated with each other.

Disclaimer statements

Contributors All authors/co-authors have contributed to this research and the preparation of the manuscript.

Funding No funding (including for equipments and medications) was required or used for conducting this study.

Conflicts of interest None.

Ethics approval Institute's ethics committee approval was obtained.