- Email: [email protected]
Bowel dysfunction in spinal-cord-injury patients
Bowel
dysfunction
in
spinal-cord-injury patients
Scott Glickman, Michael A Kamm
Introduction
Summary Background This study aimed to determine the prevalence, nature, and effects-both physical and psychological—of spinal-cord-injury (SCI) on bowel function. Methods 115 consecutive hospital outpatients (89 male, median age 38 years) with chronic SCI (median duration 62 months, range 9-491 months, 48% cervical, 47% thoracic, 5% lumbar) completed a questionnaire about pre and post injury bowel function, the Hospital Anxiety and Depression Scale (HADS), and self assessment of the impact of their disabilities and symptoms.
Findings Nausea,
diarrhoea,
constipation,
and
faecal
incontinence were all much more common (p<0·0001) after SCI. 95% of patients required at least one therapeutic method to initiate defaecation. Half the patients became dependent on others for toileting. 49% took more than 30 min to complete their toilet procedure. Bowel function was a source of distress in 54% of patients and this was significantly (p=0·005) associated with the time required for bowel management and frequency of incontinence (p=0·001). There was a highly significant correlation between the HADS scores and the time taken for bowel management. On a scale of 0 (for no perceived problem) to 10 (maximum perceived problem), patients rated their loss of mobility as a mean of 6·8 (SD 3·3) and their bowel management as 5·1 (SD 3·6).
Interpretation Bowel psychological problem
function is a in SCI patients.
major physical and
The initial focus in patients with traumatic spinal-cord injury (TSCI) is the irreversible motor paralysis, although sensory loss, spasticity, pain, and bladder and bowel dysfunction also require attention. It has been our impression that, once patients have adapted to their impaired mobility, one of their worst troubles was bowel dysfunction. In this study we investigated its magnitude.
Methods 115 consecutive outpatients attending the supraregional spinal unit at the Royal National Orthopaedic Hospital were interviewed. 109 of these had come for routine outpatient followup, five were using sports facilities on the hospital site, and one was taking part in a research study. We included patients of any age with a spinal injury at any level, provided that the injury had occurred more than 9 months previously. Since 96% of inpatients continue to be seen in the outpatient clinic, those interviewed were representative of all patients treated in the spinal clinic. Patients were interviewed by SG according to a 236-field questionnaire. All data were categorical apart from time since injury, which was a continuous variable. In addition to demographic data, information was recorded on the spinal injury (level, pattern, the time since injury, ambulation, ability to stand, whether patient actively exercised, and how much) and pre and post injury bowel disorders and symptoms. Bowel disorders before or after injury were registered only if the patient had received medical attention for them. Successful defaecation was defined as the passage of stool, and toilet dependency as the need for another person to assist-in transfer to a toilet, administering a rectal stimulant, or toilet hygiene. Incontinence was the passage of liquid or solid stool at an unwanted time. Methods used to promote defaecation were recorded. Patients also completed the Hospital Anxiety and Depression Scale (HADS) questionnaire.’ Visual analogue scales (0 for no perceived problem, 10 for maximum problem) were used to determine how troubled patients were by their various disabilities. For statistical analysis non-parametric McNemar tests or Wilcoxon matched pairs signed rank tests were performed to compare pre and post injury bowel disorders, methods of bowel stimulation, and bowel function. Chi-squared tests were used for associations between bowel function and clinical variables such as level and pattern of injury, ambulation, and the occurrence of spasms, spasticity, and autonomic dysreflexic attacks. Twosamplet tests were used to assess the relation between the presence of a post-injury bowel disorder and the HADS scores. Stepwise multiple regression analysis was used to evaluate the relation between the variables of bowel function and both emotional upset and HADS scores. A two-tailed p value of less than 0-05 was regarded as statistically significant.
Results
Department of Medical Rehabilitation, Charing Cross Hospital, London, UK (S Glickman FRCS), and Medical Physiology Unit, St Mark’s Hospital, London (M A Kamm FRCP)
Correspondence to: Dr M A Kamm, St Mark’s Hospital, Northwick Park, Harrow, Middlesex HA1 3UJ, UK
Of the 115 patients 89 (77%) were male. Median age was 37-5 years (range 18-75) and median time since spinal cord injury was 62 months (9-491). The level of the spinal injury was cervical in 48%, thoracic in 47%, and lumbar in 5%. 26 patients were able to walk, with or without aids. 54 experienced dysreflexic attacks, daily in one-third. Table 1 shows the percentage of patients who at any time had received medical attention for a bowel disorder,
1651
Table 1: Percentage of patients (n=115) with preinjury and postinjury bowel disorders and symptoms
before or after injury. Constipation, diarrhoea, and nausea were all significantly more common after spinal injury. There was no significant association between the presence of any one of these disorders after injury and the level or pattern of injury, degree of mobility, and presence or absence of spasticity, spasms, or autonomic dysreflexic attacks. The 58 patients with at least one postinjury bowel disorder had higher mean scores for depression on the HADS than the 57 without a bowel disorder (15-8 [SD 2-7] vs 14-8 [2’1]; p=0-02t test). The scores for total HADS or HADS anxiety scales did not differ significantly between those with and without postinjury bowel disorder. Table 2 records details of pre and post injury bowel function. Although the spontaneous bowel frequency was diminished after spinal injury, the actual bowel frequency was similar because of the frequent use of bowel stimulants (table 3). After injury, patients took significantly longer to open their bowels and in the whole toileting procedure. 22% took more than half an hour to open their bowels, and 49% took more than half an hour for the whole toileting procedure. Unsuccessful attempts at defaecation were more common after spinal injury: less than half the patients defaecated each time they attempted to open their bowels. Half the patients were dependent on others for toileting. Incontinence was also significantly more common after spinal injury, attributable partly to loss of sphincter control and partly to the use of stimulant laxatives. Even those with infrequent incontinence were handicapped by its
unpredictability. Toilet dependency was significantly associated with the level of injury (p<0-0001), pattern of injury (p=0-0097), and ambulation (p<0-0001). Higher level and complete and were all more likely to lead lesions, non-ambulation, to toilet dependency. 77% of dependent patients had cervical lesions, 67% had complete motor and sensory loss, and 98% could
walk, even with aids. 72% of toilet-independent patients had a thoracic injury, and 41% of these patients had some walking capability. not
Ambulation was also associated with the total time patients took to complete the toilet procedure (p=0-005). 77% of patients with walking capability took 30 min or less to go to the toilet, compared with 44% of those without walking capability. 95% of patients used at least one method of assistance to achieve defaecation (table 3). Half the patients required manual stimulation, and two thirds manually evacuated. Techniques exclusive to spinally injured patients such as the initiation of spasms or abdominal stroking were used by a quarter of patients.
1652
54% of patients said that their bowel management was a source of emotional upset. Half of these patients took longer than 15 min to defaecate, and 65% took longer than 30 min to complete the toilet procedure. The association between bowel management as a source of emotional upet and each aspect of bowel function was assessed by stepwise multiple regression analysis. Emotional upset was signficantly associated with time taken for whole toilet procedure (r=0-28, p=0-005) and frequency of faecal incontinence (r=0’37, p=0’001), but not with other aspects of bowel function. 70% of patients who were not emotionally upset by their bowel function
factors such as the level of injury. No patient in this study had a complication of their bowel problem that required surgery, although elective surgical treatments of bowel dysfunction in spinally injured patients are becoming more important. These treatments include antegrade colonic lavage via an appendicostomy,2 colostomy formation,3 and the use of anterior sacral root stimulation.4,5 Improved management of bowel dysfunction in spinally injured patients may lead to improved well being. Stone et aP treated 20 spinal-cord-injury patients with a colostomy, 12 for intractable constipation, seven for perineal pressure sores, and one for rectal cancer. Colostomy reduced the time spent on bowel care from a mean of 99 to 18 min per day. Patients with chronic constipation all felt that their quality of life had been improved by colostomy. Several factors may be important in the good outcome in their study. In addition to less time and dependency required for defaecation, patients may have been given more control over this intimate
perceived problem, 10=maximum problem. Table 4: Average patient rating on visual analogue scales *0=no
took less than 15 min to defaecate and less than 30 min to complete the toilet procedure. A stepwise multiple regression analysis was done to determine which aspects of bowel function were significantly associated with increased HADS total, anxiety, and depression scores. All three HADS scores were significantly associated with the time taken for the whole toilet procedure (r=0-31, p=0-002; r=0-31, p=0-002; r=0-20, p=0-04, respectively). The longer the time taken for the whole toilet procedure the greater the HADS scores. No other aspect of bowel function was significantly associated with the HADS scores. Patients rated the magnitude of their bowel dysfunction above that of their urinary problems, and not far below their greatest problems of loss of mobility, impaired sexual function, and dependence (table 4).
Discussion
study demonstrates that bowel function is a major physical and psychological problem in patients with spinal cord injury. To our knowledge there are no other data that characterise the prevalence of these disorders in an unselected population of spinally injured patients. Constipation, diarrhoea, and nausea were all common. Most patients required drugs and many required personal assistance to achieve defaecation. For many patients bowel function took up a considerable part of the day. Major faecal incontinence was common and often unpredictable. Patients with high spinal injuries had the greatest difficulties and they were also the least mobile. This study relied on patients’ memories for preinjury data, and may therefore contain an element of recall bias; however, we believe that this effect is unlikely to account for the major differences observed between pre and post injury bowel function. Most patients did achieve a normal bowel frequency with drugs and manual stimulation. Of greater importance was the time and dependency required to achieve this function. Bowel dysfunction was perceived by patients as one of their most important disabilities. Patients with the greatest disability in bowel function also scored the highest on the HADS anxiety and depression scales. It was not possible to determine whether the bowel disability led to psychological distress, or whether those with greater anxiety and depression developed greater problems with bowel function. Psychological distress and impaired bowel function may also be independent sequelae correlating best with other This
activity. After spinal-cord injury the bowel loses part of its extrinsic autonomic innervation. Normal function of the sacral parasympathetic supply seems to be crucial to the maintenance of colonic function and normal defaecation. Patients with spinal-cord injury have slow colonic transit, especially in the left colon and rectum.6,7 Spinally injured patients do not have the normal postprandial increase in colonic motility, and have decreased bowel wall compliance.8 Patients with higher spinal injury also lose the sympathetic gut innervation, which is predominantly inhibitory,9 but gut function in these patients is also likely to be influenced by other factors such as greater impairment of mobility. As well as visceral impairment, there is loss of the somatic innervation to the external anal sphincter. Diarrhoea produced by laxatives, or spontaneous irregular colonic contraction, can therefore lead to major faecal incontinence. In summary, we have documented a high prevalence of physical and psychological morbidity in relation to bowel in Bowel function spinal-cord-injury patients. management should feature prominently in the care of these patients. We thank the staff at the Royal National Orthopaedic Hospital in Stanmore for allowing their patients to be studied, and Victoria Hill and Caroline Dore for statistical assistance.
References 1 2
3
Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 1983; 67: 361-70. Hill J, Scott S, MacLennan I. Antegrade-enemas for the treatment of severe idiopathic constipation. Br J Surg 1994; 81: 1490-91. Stone JM, Wolfe VA, Nino-Murcia M, Perkash I. Colostomy as treatment for complications of spinal cord injury. Arch Phys Med Rehabil 1990; 71: 514-18.
4
5
6 7 8
9
MacDonagh RP, Sun WM, Smallwood R, Forster D, Read NW. Control of defaecation in patients with spinal injuries by stimulation of sacral anterior nerve roots. BMJ 1990; 300: 1494-97. Frost F, Hartwig D, Jaeger R, Leffler E, Wu Y. Electrical stimulation of the sacral dermatomes in spinal cord injury: effect of rectal manometry and bowel emptying. Arch Phys Med Rehabil 1993; 74: 696-701. Devroede G, Rahan P, Duguay C, Tètreault L, Akoury H, Perey B. Traumatic constipation. Gastroenterology 1979; 77: 1258-67. Menardo G, Bausano G, Corazziari E, et al. Large-bowel transit in paraplegic patients. Dis Colon Rectum 1987; 30: 924-28. Glick ME, Haldeman S, Meshkinpour H. The neurovisceral and electrodiagnostic evaluation of patients with thoracic spinal cord injury. Paraplegia 1986; 24: 129-37. MacIntyre AS, Thompson DG, Day S, Burnham WR, Walker ER. Modulation of human upper intestinal nutrient transit by a beta adrenoreceptor mediated pathway. Gut 1992; 33: 1062-70.
1653
dysfunction
in
spinal-cord-injury patients
Scott Glickman, Michael A Kamm
Introduction
Summary Background This study aimed to determine the prevalence, nature, and effects-both physical and psychological—of spinal-cord-injury (SCI) on bowel function. Methods 115 consecutive hospital outpatients (89 male, median age 38 years) with chronic SCI (median duration 62 months, range 9-491 months, 48% cervical, 47% thoracic, 5% lumbar) completed a questionnaire about pre and post injury bowel function, the Hospital Anxiety and Depression Scale (HADS), and self assessment of the impact of their disabilities and symptoms.
Findings Nausea,
diarrhoea,
constipation,
and
faecal
incontinence were all much more common (p<0·0001) after SCI. 95% of patients required at least one therapeutic method to initiate defaecation. Half the patients became dependent on others for toileting. 49% took more than 30 min to complete their toilet procedure. Bowel function was a source of distress in 54% of patients and this was significantly (p=0·005) associated with the time required for bowel management and frequency of incontinence (p=0·001). There was a highly significant correlation between the HADS scores and the time taken for bowel management. On a scale of 0 (for no perceived problem) to 10 (maximum perceived problem), patients rated their loss of mobility as a mean of 6·8 (SD 3·3) and their bowel management as 5·1 (SD 3·6).
Interpretation Bowel psychological problem
function is a in SCI patients.
major physical and
The initial focus in patients with traumatic spinal-cord injury (TSCI) is the irreversible motor paralysis, although sensory loss, spasticity, pain, and bladder and bowel dysfunction also require attention. It has been our impression that, once patients have adapted to their impaired mobility, one of their worst troubles was bowel dysfunction. In this study we investigated its magnitude.
Methods 115 consecutive outpatients attending the supraregional spinal unit at the Royal National Orthopaedic Hospital were interviewed. 109 of these had come for routine outpatient followup, five were using sports facilities on the hospital site, and one was taking part in a research study. We included patients of any age with a spinal injury at any level, provided that the injury had occurred more than 9 months previously. Since 96% of inpatients continue to be seen in the outpatient clinic, those interviewed were representative of all patients treated in the spinal clinic. Patients were interviewed by SG according to a 236-field questionnaire. All data were categorical apart from time since injury, which was a continuous variable. In addition to demographic data, information was recorded on the spinal injury (level, pattern, the time since injury, ambulation, ability to stand, whether patient actively exercised, and how much) and pre and post injury bowel disorders and symptoms. Bowel disorders before or after injury were registered only if the patient had received medical attention for them. Successful defaecation was defined as the passage of stool, and toilet dependency as the need for another person to assist-in transfer to a toilet, administering a rectal stimulant, or toilet hygiene. Incontinence was the passage of liquid or solid stool at an unwanted time. Methods used to promote defaecation were recorded. Patients also completed the Hospital Anxiety and Depression Scale (HADS) questionnaire.’ Visual analogue scales (0 for no perceived problem, 10 for maximum problem) were used to determine how troubled patients were by their various disabilities. For statistical analysis non-parametric McNemar tests or Wilcoxon matched pairs signed rank tests were performed to compare pre and post injury bowel disorders, methods of bowel stimulation, and bowel function. Chi-squared tests were used for associations between bowel function and clinical variables such as level and pattern of injury, ambulation, and the occurrence of spasms, spasticity, and autonomic dysreflexic attacks. Twosamplet tests were used to assess the relation between the presence of a post-injury bowel disorder and the HADS scores. Stepwise multiple regression analysis was used to evaluate the relation between the variables of bowel function and both emotional upset and HADS scores. A two-tailed p value of less than 0-05 was regarded as statistically significant.
Results
Department of Medical Rehabilitation, Charing Cross Hospital, London, UK (S Glickman FRCS), and Medical Physiology Unit, St Mark’s Hospital, London (M A Kamm FRCP)
Correspondence to: Dr M A Kamm, St Mark’s Hospital, Northwick Park, Harrow, Middlesex HA1 3UJ, UK
Of the 115 patients 89 (77%) were male. Median age was 37-5 years (range 18-75) and median time since spinal cord injury was 62 months (9-491). The level of the spinal injury was cervical in 48%, thoracic in 47%, and lumbar in 5%. 26 patients were able to walk, with or without aids. 54 experienced dysreflexic attacks, daily in one-third. Table 1 shows the percentage of patients who at any time had received medical attention for a bowel disorder,
1651
Table 1: Percentage of patients (n=115) with preinjury and postinjury bowel disorders and symptoms
before or after injury. Constipation, diarrhoea, and nausea were all significantly more common after spinal injury. There was no significant association between the presence of any one of these disorders after injury and the level or pattern of injury, degree of mobility, and presence or absence of spasticity, spasms, or autonomic dysreflexic attacks. The 58 patients with at least one postinjury bowel disorder had higher mean scores for depression on the HADS than the 57 without a bowel disorder (15-8 [SD 2-7] vs 14-8 [2’1]; p=0-02t test). The scores for total HADS or HADS anxiety scales did not differ significantly between those with and without postinjury bowel disorder. Table 2 records details of pre and post injury bowel function. Although the spontaneous bowel frequency was diminished after spinal injury, the actual bowel frequency was similar because of the frequent use of bowel stimulants (table 3). After injury, patients took significantly longer to open their bowels and in the whole toileting procedure. 22% took more than half an hour to open their bowels, and 49% took more than half an hour for the whole toileting procedure. Unsuccessful attempts at defaecation were more common after spinal injury: less than half the patients defaecated each time they attempted to open their bowels. Half the patients were dependent on others for toileting. Incontinence was also significantly more common after spinal injury, attributable partly to loss of sphincter control and partly to the use of stimulant laxatives. Even those with infrequent incontinence were handicapped by its
unpredictability. Toilet dependency was significantly associated with the level of injury (p<0-0001), pattern of injury (p=0-0097), and ambulation (p<0-0001). Higher level and complete and were all more likely to lead lesions, non-ambulation, to toilet dependency. 77% of dependent patients had cervical lesions, 67% had complete motor and sensory loss, and 98% could
walk, even with aids. 72% of toilet-independent patients had a thoracic injury, and 41% of these patients had some walking capability. not
Ambulation was also associated with the total time patients took to complete the toilet procedure (p=0-005). 77% of patients with walking capability took 30 min or less to go to the toilet, compared with 44% of those without walking capability. 95% of patients used at least one method of assistance to achieve defaecation (table 3). Half the patients required manual stimulation, and two thirds manually evacuated. Techniques exclusive to spinally injured patients such as the initiation of spasms or abdominal stroking were used by a quarter of patients.
1652
54% of patients said that their bowel management was a source of emotional upset. Half of these patients took longer than 15 min to defaecate, and 65% took longer than 30 min to complete the toilet procedure. The association between bowel management as a source of emotional upet and each aspect of bowel function was assessed by stepwise multiple regression analysis. Emotional upset was signficantly associated with time taken for whole toilet procedure (r=0-28, p=0-005) and frequency of faecal incontinence (r=0’37, p=0’001), but not with other aspects of bowel function. 70% of patients who were not emotionally upset by their bowel function
factors such as the level of injury. No patient in this study had a complication of their bowel problem that required surgery, although elective surgical treatments of bowel dysfunction in spinally injured patients are becoming more important. These treatments include antegrade colonic lavage via an appendicostomy,2 colostomy formation,3 and the use of anterior sacral root stimulation.4,5 Improved management of bowel dysfunction in spinally injured patients may lead to improved well being. Stone et aP treated 20 spinal-cord-injury patients with a colostomy, 12 for intractable constipation, seven for perineal pressure sores, and one for rectal cancer. Colostomy reduced the time spent on bowel care from a mean of 99 to 18 min per day. Patients with chronic constipation all felt that their quality of life had been improved by colostomy. Several factors may be important in the good outcome in their study. In addition to less time and dependency required for defaecation, patients may have been given more control over this intimate
perceived problem, 10=maximum problem. Table 4: Average patient rating on visual analogue scales *0=no
took less than 15 min to defaecate and less than 30 min to complete the toilet procedure. A stepwise multiple regression analysis was done to determine which aspects of bowel function were significantly associated with increased HADS total, anxiety, and depression scores. All three HADS scores were significantly associated with the time taken for the whole toilet procedure (r=0-31, p=0-002; r=0-31, p=0-002; r=0-20, p=0-04, respectively). The longer the time taken for the whole toilet procedure the greater the HADS scores. No other aspect of bowel function was significantly associated with the HADS scores. Patients rated the magnitude of their bowel dysfunction above that of their urinary problems, and not far below their greatest problems of loss of mobility, impaired sexual function, and dependence (table 4).
Discussion
study demonstrates that bowel function is a major physical and psychological problem in patients with spinal cord injury. To our knowledge there are no other data that characterise the prevalence of these disorders in an unselected population of spinally injured patients. Constipation, diarrhoea, and nausea were all common. Most patients required drugs and many required personal assistance to achieve defaecation. For many patients bowel function took up a considerable part of the day. Major faecal incontinence was common and often unpredictable. Patients with high spinal injuries had the greatest difficulties and they were also the least mobile. This study relied on patients’ memories for preinjury data, and may therefore contain an element of recall bias; however, we believe that this effect is unlikely to account for the major differences observed between pre and post injury bowel function. Most patients did achieve a normal bowel frequency with drugs and manual stimulation. Of greater importance was the time and dependency required to achieve this function. Bowel dysfunction was perceived by patients as one of their most important disabilities. Patients with the greatest disability in bowel function also scored the highest on the HADS anxiety and depression scales. It was not possible to determine whether the bowel disability led to psychological distress, or whether those with greater anxiety and depression developed greater problems with bowel function. Psychological distress and impaired bowel function may also be independent sequelae correlating best with other This
activity. After spinal-cord injury the bowel loses part of its extrinsic autonomic innervation. Normal function of the sacral parasympathetic supply seems to be crucial to the maintenance of colonic function and normal defaecation. Patients with spinal-cord injury have slow colonic transit, especially in the left colon and rectum.6,7 Spinally injured patients do not have the normal postprandial increase in colonic motility, and have decreased bowel wall compliance.8 Patients with higher spinal injury also lose the sympathetic gut innervation, which is predominantly inhibitory,9 but gut function in these patients is also likely to be influenced by other factors such as greater impairment of mobility. As well as visceral impairment, there is loss of the somatic innervation to the external anal sphincter. Diarrhoea produced by laxatives, or spontaneous irregular colonic contraction, can therefore lead to major faecal incontinence. In summary, we have documented a high prevalence of physical and psychological morbidity in relation to bowel in Bowel function spinal-cord-injury patients. management should feature prominently in the care of these patients. We thank the staff at the Royal National Orthopaedic Hospital in Stanmore for allowing their patients to be studied, and Victoria Hill and Caroline Dore for statistical assistance.
References 1 2
3
Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 1983; 67: 361-70. Hill J, Scott S, MacLennan I. Antegrade-enemas for the treatment of severe idiopathic constipation. Br J Surg 1994; 81: 1490-91. Stone JM, Wolfe VA, Nino-Murcia M, Perkash I. Colostomy as treatment for complications of spinal cord injury. Arch Phys Med Rehabil 1990; 71: 514-18.
4
5
6 7 8
9
MacDonagh RP, Sun WM, Smallwood R, Forster D, Read NW. Control of defaecation in patients with spinal injuries by stimulation of sacral anterior nerve roots. BMJ 1990; 300: 1494-97. Frost F, Hartwig D, Jaeger R, Leffler E, Wu Y. Electrical stimulation of the sacral dermatomes in spinal cord injury: effect of rectal manometry and bowel emptying. Arch Phys Med Rehabil 1993; 74: 696-701. Devroede G, Rahan P, Duguay C, Tètreault L, Akoury H, Perey B. Traumatic constipation. Gastroenterology 1979; 77: 1258-67. Menardo G, Bausano G, Corazziari E, et al. Large-bowel transit in paraplegic patients. Dis Colon Rectum 1987; 30: 924-28. Glick ME, Haldeman S, Meshkinpour H. The neurovisceral and electrodiagnostic evaluation of patients with thoracic spinal cord injury. Paraplegia 1986; 24: 129-37. MacIntyre AS, Thompson DG, Day S, Burnham WR, Walker ER. Modulation of human upper intestinal nutrient transit by a beta adrenoreceptor mediated pathway. Gut 1992; 33: 1062-70.
1653