Prevalence and Risk Factors of Incontinence After Stroke
The Copenhagen Stroke Study
Background and Purpose The purpose of this study was to investigate in a community-based population the prevalence of both urinary (UI) and fecal (FI) incontinence and to analyze risk factors by means of multivariate analysis.
Methods Included were 935 acute stroke patients admitted consecutively during 19 months. We evaluated UI and FI using subscores of the Barthel Index during the hospital stay and at 6-month follow-up.
Results On admission, the proportions of patients with full UI, partial UI, and no UI were 36%, 11%, and 53%, respectively (8%, 11%, and 81% at 6-month follow-up). The proportions of patients with full FI, partial FI, and no FI on admission were 34%, 6%, and 60%, respectively (5%, 4%, and 91% at 6-month follow-up). By multivariate analysis, significant risk factors for UI and FI were age, severity of stroke, diabetes, and comorbidity of other disabling diseases.
Conclusions On admission in the acute state, almost half of an unselected stroke population have UI and/or FI. The proportion declines to one fifth (UI) and one tenth (FI) of the surviving patients at 6 months. Increasing age, stroke severity, diabetes, and other disabling diseases increase the risk of UI as well as FI.
Incontinence of urine and of feces after stroke are of significant importance both prognostically and socially for patients and for their caregivers. Early urinary incontinence after stroke implies a poor prognosis,1 2 3 4 and persisting urinary incontinence is known to increase the burden on caregivers.5
For the planning of stroke care, information on the prevalence of incontinence is essential. However, only a few community-based studies6 7 have reported the prevalence of incontinence with varying results. Brocklehurst et al6 found that within the first 2 weeks after stroke, urinary and fecal incontinence occurred in 39% and in 23%, respectively. However, the study included a relatively small number of patients. Wade and Hewer7 reported urinary incontinence in 44% and fecal incontinence in 31% in their study of 976 stroke patients, but only half of the patients were assessed. Other studies were performed in selected patients1 8 and included only a small number of patients.8
Furthermore, risk factor analysis is indispensable to find patients at higher risk of incontinence, to compare our prevalence data with other studies based on study populations biased by different risk factor distributions, and finally to speculate on underlying mechanisms. Previous studies have reported several factors to be risk factors for incontinence: severity of stroke,1 6 age,1 2 female sex,2 lower level of consciousness,1 2 severe motor paresis,1 8 dysphasia,8 and mental impairment.8 However, since these studies used univariate analyses only, it is not clear whether these factors have independent influence or if they exert their influence through associations with other factors. On the other hand, important risk factors could have been obscured by interactions.
The purpose of this study was to investigate in a large community-based study population the prevalence of both urinary and fecal incontinence and to analyze risk factors by means of multivariate analysis.
Subjects and Methods
This study is a part of the Copenhagen Stroke Study, which has been reported elsewhere.9 10 The study population was community based. First, Bispebjerg Hospital serves a well-defined community with 239 886 inhabitants within the city of Copenhagen. Second, hospital care is free, and a very high proportion of stroke patients are admitted to the hospital; in a neighboring area within Greater Copenhagen it was recently shown that 88% of all stroke patients are hospitalized. Third, Bispebjerg Hospital is the only hospital serving the region. Finally, all persons from the community who have an acute cerebrovascular disease that requires admission are referred to the neurological department of Bispebjerg Hospital.
The study population consisted of 935 acute stroke patients admitted consecutively to the neurological department of a hospital in Copenhagen, Denmark, during a 19-month period beginning March 1, 1992. Six hundred forty patients (68%) were admitted within the first 24 hours from stroke onset, 797 (85%) within 4 days, and 867 (93%) within the first week. After discharge, a follow-up examination was performed 6 months after stroke.
Diagnosis of Stroke
Stroke was diagnosed by a neurologist according to the World Health Organization criteria: rapidly developed clinical signs of focal disturbance of cerebral function, lasting more than 24 hours or leading to death, with no apparent cause other than vascular origin.11 Subarachnoid bleeding was not included. CT scan was performed with a Siemens Somatom DR scanner in 759 (81%) of the 935 patients included, and CT description was used to divide patients into subtypes (infarction or hemorrhage) (Table 1⇓).
Evaluation of Neurological Deficits and ADL Functions
Neurological deficits were assessed on admission, the day after admission, weekly until death or end of rehabilitation, and finally 6 months after stroke with the SSS score.12 13 ADL was assessed weekly until death or end of rehabilitation with the BI.14
Evaluation of Incontinence
Urinary and fecal incontinence were assessed and categorized into three grades (full/partial/no continence) with subscores of the BI. A patient was classified as having full urinary/fecal continence when he/she was fully able to control the bladder/bowel and had no accidents. When a patient had only occasional incontinence, the patient was classified as having partial urinary/fecal incontinence. In the remaining cases the patient was classified as having full incontinence.
All the patients in this study using an external device (ie, catheter) for bladder control were found to be unable to manage the device independently and were classified as having full urinary incontinence. Use of suppository/enema for bowel control was not considered in our evaluation of fecal incontinence.
The following risk factors were considered: age, sex, current smoking, hypertension, DM, IHD, history of former stroke, and history of other disabling diseases. Current smoking was defined as current smoking of any kind of tobacco; hypertension was defined as either (1) under treatment with antihypertensive drugs at the time of admission or (2) hypertension diagnosed during the hospital stay; DM/IHD was defined as a history of DM/IHD or DM/IHD diagnosed during the hospital stay.
Statistical analysis was performed with the use of the SPSS15 computer program. The Mann-Whitney test was used for comparison of continuous data, and the χ2 test was used for comparison of noncontinuous data. In the risk factor analysis of incontinence, the logistic regression analysis was used, and the odds ratio was calculated for each significant factor. All variables of interest were tested with the backward procedure, and variables that had a probability value below .3 were then entered with the forward procedure to attain as much information as possible. The level of significance was set at P<.05.
The study was approved by the Ethics Committee of Copenhagen.
Basic characteristics of the 935 patients initially included in the study are presented in Table 1⇑. Of these 935, 68 patients were not evaluated by BI in the first week. There were no significant differences between the patients who were evaluated and who were not concerning age (mean age, 74 versus 76 years; U=27280.5; P=.31), sex (χ2=1.97; df=l; P=.16), or initial SSS score (mean SSS score on admission, 36 versus 39; U=27631.0; P=.57). One hundred eighty-eight patients (20%) died during the hospital stay. At the 6-month follow-up, 493 (73%) of the 678 survivors were examined. The nonresponders generally had more severe strokes (mean SSS score on admission, 38 versus 43; U=37088.0; P<.01) but were of similar age (mean, 73 versus 73 years; U=45099.5; P=.82).
Frequency of Urinary Incontinence
On admission 309 patients (36%) had full urinary incontinence, 95 (11%) had partial incontinence, and 463 (53%) had no urinary incontinence. At discharge 103 (15%) of the survivors had full urinary incontinence, 87 (13%) had partial incontinence, and 507 (72%) had no urinary incontinence. At the 6-month follow-up, 38 (8%) of the responders had full urinary incontinence, 56 (11%) had partial incontinence, and 399 (81%) had no urinary incontinence (Table 2⇓).
Prognosis of Urinary Incontinence
In patients with initial full urinary incontinence, 158 (52%) had died by discharge, 94 (30%) had full urinary incontinence, 31 (10%) had partial incontinence, and 24 (8%) had no urinary incontinence at discharge. By the 6-month follow-up, 185 (60%) had died, 29 (14%) had full urinary incontinence, 21 (10%) had partial incontinence, and 32 (16%) had no urinary incontinence (Table 3⇓).
In patients with initial partial urinary incontinence, 15 (16%) had died by discharge, 5 (5%) had full urinary incontinence, 40 (43%) had partial incontinence, and 33 (36%) had no urinary incontinence at discharge. By the 6-month follow-up, 24 (25%) had died, 2 (3%) had full urinary incontinence, 14 (22%) had partial incontinence, and 32 (50%) had no urinary incontinence (Table 3⇑).
In patients who initially had no urinary incontinence, 10 (2%) had died by discharge, 1 (0%) had full urinary incontinence, 12 (3%) had partial incontinence, and 95% had no urinary incontinence at discharge. By the 6-month follow-up, 31 (7%) had died, 6 (2%) had full urinary incontinence, 19 (5%) had partial incontinence, and 310 (86%) had no urinary incontinence (Table 3⇑).
Risk Factor Analysis of Initial Urinary Incontinence
Table 4⇓ shows results of the univariate risk factor analysis of initial urinary incontinence. Patients with initial urinary incontinence were significantly older and more frequently women, more often had DM and comorbidity of other disabling diseases, and were less frequently current smokers than patients without. The lesions in patients with urinary incontinence were significantly more frequently a hemorrhage, were larger in size, and more often involved cerebral cortex than those in patients without urinary incontinence. Patients with urinary incontinence also had significantly lower scores on the initial BI and SSS.
In multivariate logistic regression, age, lesion diameter, DM, hypertension, other disabling diseases, and initial SSS score were significant risk factors for urinary incontinence (Table 5⇓). According to this model, a 10-year increase in age increases the risk 1.7 times, compared with 1.2 times for a 10-mm increase in lesion size, 3.4 times for DM, and 2.4 times for comorbidity of other disabling diseases. A 10-point increase in the initial SSS score decreases the risk to approximately one third, and hypertension halves the risk.
Frequency of Fecal Incontinence
On admission 296 patients (34%) had full fecal incontinence, 51 (6%) had partial incontinence, and 520 (60%) had no fecal incontinence. At discharge 84 (12%) of the survivors still had full fecal incontinence, 40 (6%) had partial incontinence, and 573 (82%) had no fecal incontinence. At the 6-month follow-up, 24 (5%) of the responders had full fecal incontinence, 20 (4%) had partial incontinence, and 449 (91%) had no fecal incontinence (Table 2⇑).
Prognosis of Fecal Incontinence
In patients with initial full fecal incontinence, 158 (53%) had died by discharge, 78 (27%) had full fecal incontinence, 17 (6%) had partial incontinence, and 41 (14%) had no fecal incontinence at discharge. By the 6-month follow up, 185 (63%) had died, 22 (11%) had full fecal incontinence, 12 (6%) had partial incontinence, and 39 (20%) had no fecal incontinence (Table 6⇓).
In patients with initial partial fecal incontinence, 12 (24%) had died by discharge, 2 (4%) had full fecal incontinence, 14 (28%) had partial incontinence, and 22 (44%) had no fecal incontinence at discharge. By the 6-month follow-up, 19 (37%) had died, none had full fecal incontinence, 1 (3%) had partial incontinence, and 18 (60%) had no fecal incontinence (Table 6⇑).
In patients who initially had no fecal incontinence, 13 (3%) had died by discharge, 1 (0.2%) had full fecal incontinence, 7 (1%) had partial incontinence, and 494 (96%) had no fecal incontinence at discharge. By the 6-month follow-up, 36 (7%) had died, 2 (1%) had full fecal incontinence, 6 (2%) had partial incontinence, and 365 (90%) had no fecal incontinence (Table 6⇑).
Risk Factor Analysis of Initial Fecal Incontinence
Table 7⇓ shows results of the univariate risk factor analysis for initial fecal incontinence. Patients with fecal incontinence in the first week after stroke were significantly more frequently women and more often had a history of former stroke and comorbidity of other disabling diseases than patients without fecal incontinence. The lesions in patients with fecal incontinence were significantly more often a hemorrhage, were larger in size, and more often involved the cerebral cortex than those in patients without fecal incontinence. Patients with fecal incontinence also had significantly lower scores on the initial BI and SSS.
In multivariate logistic regression analysis, age, DM, lesion size, other disabling disease, and initial SSS score were significant influencing factors (Table 5⇑). According to this model, a 10-year increase in age and a 10-mm increase in lesion size results in a 1.5- and a 1.3-fold increased risk, respectively; DM and comorbidity of other disabling diseases double the risk; and a 10-point increase in the initial SSS score decreases the risk to approximately one third.
Relationship Between Urinary and Fecal Incontinence
A large overlap of patients with urinary incontinence and patients with fecal incontinence was found. Eighty-four percent of patients with initial urinary incontinence also had fecal incontinence, and 98% of patients with initial fecal incontinence also had urinary incontinence. There was a high Spearman's rank correlation (r=.84, P<.0l) between subscores of initial urinary and fecal incontinence.
The frequencies of urinary and fecal incontinence found in this study are similar to what has been found in other community-based studies. In the present study, the proportions of stroke patients with urinary incontinence on admission and at 6-month follow-up were 47% and 19%, respectively. Wade and Hewer7 reported that 44% of 531 stroke patients examined in the first week after stroke had urinary incontinence, and the proportion decreased to 11% after 6 months. Brocklehurst et al6 studied 135 two-week survivors from stroke and found that 39% of the patients had urinary incontinence. After 6 months, 12% still had urinary incontinence.
The proportions of stroke patients with fecal incontinence on admission and at 6-month follow-up in our study were 40% and 9%, respectively. In the study by Wade and Hewer,7 the frequencies of fecal incontinence in the first week and 6 months after stroke were 31% and 7%, respectively. Brocklehurst et al6 found that fecal incontinence occurred in 23% within the first 2 weeks, and 3% still had fecal incontinence after 6 months.
At 6-month follow-up, 27% of the survivors were not examined, and these patients generally had more severe strokes. The risk factor analysis in our study revealed that the severity of stroke increases the risk of both urinary and fecal incontinence. Therefore, our frequency data at 6-month follow-up are considered to be somewhat underestimated.
In our study we did not discriminate between incontinence that began before or after stroke. In a hospital-based study of conscious stroke patients, Barer1 found that 2.5% had had urinary incontinence before stroke. Westle et al16 reported from a community-based study of 3809 residents aged 65 years and older that 5.8% of men and 7.6% of women often have difficulty holding urine until reaching a toilet. They reported the prevalence of fecal incontinence to be 8.5% in men and 7.8% in women. Based on these figures, we can estimate that preexisting incontinence explains only a small part of incontinence after stroke.
Our study reports the first multivariate risk factor analysis of urinary and fecal incontinence after stroke that reveals risk factors not reported before in univariate analyses: age and DM for fecal incontinence.
The multiple logistic regression analysis found age, severity of stroke (initial SSS score and diameter of lesion), DM, hypertension, and comorbidity of other disabling diseases to be significant risk factors for urinary incontinence.
All risk factors for urinary incontinence found in this study except hypertension are in conformity with previously known risk factors reported from univariate analyses1 2 6 8 in stroke patients or with risk factors revealed by a multivariate analysis in general elderly populations.16 These factors, except hypertension, are well in accordance with the three major mechanisms of urinary incontinence after stroke analyzed by Gelber et al17 : (1) disruption of the neuromicturition pathways; (2) incontinence due to stroke-related cognitive and language deficits; and (3) concurrent neuropathy or medication use. Our finding that hypertension halves the risk of urinary incontinence after stroke could possibly be attributed to pharmaceutical effects of hypertension therapy or to its association with other covariates in the logistic regression model. The following findings support the latter. First, there is an association between hypertension and DM in our study population. In patients with hypertension, the prevalence of DM was almost twice as high (24% versus 13%). Such an association between covariates of a regression model can cause an artifact that requires causal interpretation to be differentiated. Second, hypertension no longer had a significant influence in multiple logistic regression analysis when DM was removed from the covariates and the analysis was redone. However, DM continued to be a significant risk factor when hypertension was removed.
The multiple logistic regression analysis found age, DM, severity of stroke (initial SSS score and diameter of lesion), and comorbidity of other disabling diseases to be significant risk factors for fecal incontinence.
Our finding that age and DM have an independent negative influence on fecal incontinence after stroke apart from stroke severity and disabilities in ADL is original. The former could be explained by an association between aging and lower anal sphincter pressure18 and the latter by an abnormal internal-anal-sphincter function in DM patients with fecal incontinence.19 To our knowledge, no other studies have reported on risk factors for fecal incontinence after stroke based on a community-based population.
The poor prognosis of stroke patients with early urinary incontinence regarding mortality is in accordance with former studies.1 2 4 Barer1 found in 362 hospital-based conscious stroke patients that 6-month mortality increased rapidly with increasing severity of initial urinary incontinence. Wade and Hewer2 also reported higher 6-month mortality among stroke patients with initial urinary incontinence.
This study shows the natural history of urinary and fecal incontinence after stroke and offers basic data that are comparable with other studies and that can be compared with a future therapeutic trial. Since physicians reportedly pay inadequate attention to urinary incontinence problems in stroke patients,20 21 more attention should be paid to stroke patients with the risk factors revealed by this study: age, stroke severity, DM, and comorbidity of other disabling disease.
On admission in the acute state, nearly half of the patients with stroke have urinary and/or fecal incontinence, and 6 months later the proportion decreases to one fifth and one tenth, respectively. Age, DM, severity of stroke, and comorbidity of other disabling diseases are the common independent risk factors for both urinary and fecal incontinence. This study suggests that stroke patients with these risk factors need extra attention and care for incontinence.
Selected Abbreviations and Acronyms
|ADL||=||activities of daily living|
|IHD||=||ischemic heart disease|
|SSS||=||Scandinavian Stroke Scale|
This study was supported by grants from the Danish Health Foundation, the Danish Heart Foundation, Ebba Celinders Foundation, and the Gangsted Foundation.
- Received May 21, 1996.
- Revision received August 23, 1996.
- Accepted August 25, 1996.
- Copyright © 1997 by American Heart Association
Barer DH. Continence after stroke: useful predictor or goal of therapy? Age Ageing.. 1989;188:183-191.
Wade DT, Hewer RL. Outlook after an acute stroke: urinary incontinence and loss of consciousness compared in 532 patients. QJM New Series.. 1985;56:601-608.
Gladman JRF, Harwood DMJ, Barer DH. Predicting the outcome of acute stroke: prospective evaluation of five multivariate models and comparison with simple methods. J Neurol Neurosurg Psychiatry.. 1992;55:347-351.
Taub NA, Wolfe CDA, Richardson E, Burney PGJ. Predicting the disabilities of first-time stroke sufferers at 1 year: 12 month follow-up of a population-based cohort in southeast England. Stroke.. 1994;25:352-357.
Flaherty JH, Miller DK, Coe RM. Impact on caregivers of supporting urinary function in noninstitutionalized, chronically ill seniors. Gerontologist.. 1992;32:541-545.
Wade DT, Hewer RL. Functional abilities after stroke: measurement, natural history and prognosis. J Neurol Neurosurg Psychiatry.. 1987;50:177-182.
Borrie MJ. Urinary incontinence after stroke: a prospective study. Age Ageing.. 1986;15:177-181.
Report of the WHO Task Force on Stroke and Other Cerebrovascular Disorders: Stroke—1989: recommendations on stroke prevention, diagnosis, and therapy. Stroke.. 1989;20:1407-1431.
Scandinavian Stroke Study Group. Multicenter trial of hemodilution in ischemic stroke: background and study protocol. Stroke.. 1985;16:885-890.
SPSS. Statistical Package for Social Sciences: Windows Version, Release 6.0. Chicago, Ill: SPSS Inc; 1992.
Gelber DA, Good DC, Laven LJ, Verhulst SJ. Causes of urinary incontinence after hemispheric stroke. Stroke.. 1993;24:378-382.
Henriksen T. Incontinence after stroke. Lancet.. 1991;338:1335. Letter.
Benbow S, Sangster G, Barer D. Incontinence after stroke. Lancet.. 1991;338:1602-1603. Letter.