Heit, Michael MD, MSPH*; Blackwell, Linda RN, CCRC*; Thomas, Shannon MD*; Ouseph, Rosemary MD†
“Should I be concerned about urinary incontinence in my kidney transplant recipients who no longer require hemodialysis and are now voiding?” A nephrology colleague posed this question and inspired our study group to perform a MEDLINE search of the English medical literature from 1966 to present, using the key words “urinary incontinence” and “kidney transplantation,” which failed to produce any articles addressing this subject. Therefore, we designed a cross-sectional study to estimate the prevalence, severity, and impact that urinary incontinence has on activities of daily living in kidney transplant recipients.
Incontinence depends on a number of factors, including bladder volume. Bladder volume in turn depends on the effective filtration of blood through functioning kidneys. We hypothesized that poor kidney function, as measured by elevated serum creatinine after transplantation, would be associated with lower incontinence severity scores and a milder impact of incontinence on activities of daily living. If we were unable to prove this hypothesis, we hypothesized that previously established clinical and sociodemographic predictors of urinary incontinence1 in community-dwelling nontransplant recipients would be applicable to kidney transplant recipients as well. We tested this hypothesis by comparing established clinical and sociodemographic predictors of urinary incontinence1 from incontinent kidney transplant recipients and continent recipients.
MATERIALS AND METHODS
After allograft transplantation, nephrologists from the University of Louisville Nephrology Clinic provide healthcare services for kidney transplant recipients. The nephrologists medically manage 184 male and 135 female kidney transplant recipients. The inclusion criterion for our study was females aged 20–80 years. For this population, the background incontinence rate for community-dwelling nontransplant women is approximately 30%.2 A research nurse conducted face-to-face interviews with eligible kidney transplant recipients after their clinic visit. The same research nurse conducted telephone interviews with eligible kidney transplant recipients who were not scheduled for follow-up during the study period. All interviews were conducted from January 2001 through February 2002. Informed consent was obtained from all study participants regardless of the survey methodology used.
The continence status of all participants before and after kidney transplantation was ascertained by asking “Do you have problems with uncontrollable urine loss?” This question has been used in a large population-based epidemiological study of female urinary incontinence,3 in which a prevalence rate of 37.7% was reported. Patients who reported problems with uncontrollable urine loss before kidney failure but not after kidney transplantation were considered “pretransplant incontinent.” Patients who reported problems with uncontrollable urine loss after transplant but not before kidney failure were considered “post-transplant incontinent.” Patients who reported problems with uncontrollable urine loss before kidney failure and after kidney transplantation were considered “persistent incontinent.” All study participants who reported post-transplant incontinence regardless of pretransplant status were categorized as cases. All study participants who were continent after transplant regardless of pretransplant status were categorized as controls.
The type of urinary incontinence was ascertained from all patients with post-transplant incontinence. Patients who responded “yes” to the question, “Do you lose urine during sudden physical exertion, lifting, coughing, or sneezing?” were considered stress incontinent if in addition they responded “no” to “Do you experience such a strong and sudden urge to void that you leak before reaching the toilet?” Patients who responded “no” to the first question, and “yes” to the second were considered “urge incontinent.” Patients who responded “yes” to both questions were considered “mixed incontinent.” These questions have been validated against urodynamic testing results in a large sample of incontinent women.4
Patients characterized their urine leakage frequency as follows: 1 = less than once a month, 2 = a few times a month, 3 = a few times a week, and 4 = every day and/or night. They also reported the volume of urine leakage as follows: 1 = drops, 2 = small splashes, and 3 = more. Sandvik's Incontinence Severity Index was calculated as the product of the frequency and volume response values. The index ranged from 1 to 12, where values from 1 to 2 indicated slight incontinence, 3 to 6 indicated moderated incontinence, 7 to 9 indicated severe incontinence, and 10 to 12 indicated very severe incontinence. The construct validity of this 4-level severity index has been established by comparison with pad-weighing tests with a correlation coefficient of .54.5
The impact of incontinence on activities of daily living was assessed with the Incontinence Impact Questionnaire short form, which is highly correlated with the long form.6 The long form's Cronbach's α coefficients for the physical activity, travel, social relationships, and stress symptoms subscales are .87, .87, .90, and .90, respectively. The criterion validity of the long form was established by comparing Incontinence Impact Questionnaire scores with number of incontinent episodes with correlation coefficients of .46.
Serum creatinine levels are obtained from kidney transplant recipients at each clinic visit as a measure of kidney function. The creatinine level at the time of the face-to-face interview was used as our measure of kidney function for recipients who were interviewed face to face. The creatinine level at the time of the last clinic visit was used as our measure of kidney function for recipients who were interviewed by telephone.
All participants were asked to complete 24-hour bladder diaries for 3 consecutive days. The mean 24-hour urine output provided an indirect and noninvasive measure of renal function after transplantation. They also provided objective data for identifying predictors of continence status after kidney transplant. Established clinical and sociodemographic predictors of urinary incontinence1 in community-dwelling nontransplant recipients, including age, body mass index, vaginal parity, and smoking history, were extracted from patient charts and entered onto data sheets.
Associations among post-transplant continence status and continuous, normally distributed clinical, sociodemographic, and 3-day 24-hour bladder diary data were tested by using Student t tests (two-sided). Associations among post-transplant continence status and non-normally distributed clinical, sociodemographic, and 3-day 24-hour bladder diary data were tested by using Mann-Whitney U tests. Associations between post-transplant continence status and categorical data were tested with χ2 tests with Yates continuity correction. Odds ratios with 95% confidence intervals (CIs) were reported when two-by-two cross tabulations were made. Associations among kidney function, as measured by serum creatinine, incontinence severity index scores, and impact of incontinence on activities of daily living were assessed with Spearman's correlation coefficients.
Normality of the sample estimates were evaluated by several methods, including a comparison of the shape of the frequency distribution (histogram) of the sample estimate with a normal curve, comparison of the mean and median values of the sample estimate for similarities, calculation of the quotient obtained when the skewness statistic for the sample estimate was divided by its standard error (values less than 2 suggest normality), and the Kolmogorov-Smirnov and Shapiro-Wilk tests of normality. Variances of the sample estimates were compared by using Levene's test for equality of variances. A P value less than .05 for the null hypothesis of equal variances assumed allowed us to reject the null hypothesis and use separate variance t tests for continuous normally distributed sample estimate comparisons. Pooled variance t tests were used for continuous normally distributed sample estimate comparisons when the P value was greater than .05 and the null hypothesis was accepted.
Not all study participants completed and returned their 24-hour bladder diaries during the study period. Therefore, differences in clinical and sociodemographic data between diary completers and noncompleters were compared with Student t tests, Mann–Whitney U tests, and χ2 tests with Yates Continuity correction to test for nonresponse bias. Missing data for body mass index (1%) was imputed as the mean for this variable. Missing data for duration in years from transplantation to interview (2%) was imputed as the median for this variable. Data were entered into SPSS for Windows V11.0 (SPSS, Chicago, Ill) for analysis. P values less than .05 were considered significant. The Human Studies Committee at our institution approved this study.
One hundred of 123 eligible kidney transplant recipients agreed to participate in our study and completed the face-to-face interview (n = 79) or telephone interview (n = 21) for a response rate of 81.3% (Figure 1). The mean age, body mass index, and vaginal parity of all study participants was 49.5 years (range 20–76 years; 95% CI 47.1, 51.8), 29.4 kg/m2 (range 17.3–43.9 kg/m2; 95% CI 28.1, 30.6), and 1.8 (95% CI 1.5, 2.1), respectively. Seventy-five percent were non-Hispanic white, 24% were non-Hispanic African American, and 1% were Asian/Pacific Islander. Forty-six percent of interviewed kidney transplant recipients admitted smoking tobacco products. The mean serum creatinine level was 1.5 mg/dL (95% CI 1.4, 1.7). The mean duration in time from kidney transplantation to incontinence was 2.3 years (median 1 year; 95% CI 1.2, 3.3), and the mean duration in time from incontinence to interview was 2.1 years (median 1.1 years; 95% CI 1.3, 2.9), for incontinent study participants. The mean duration in time from kidney transplantation to interview was 4.6 years (95% CI 3.9, 5.3) for all study participants and 4.28 years (95% CI 3.3, 5.3) for incontinent study participants.
Thirty-nine (39%) kidney transplant recipients reported urinary incontinence at some point in their life. Six (6%) kidney transplant recipients were categorized as persistent incontinent, and 22 (22%) were categorized as post-transplant incontinent. Therefore, 28 (28%) kidney transplant recipients reported urinary incontinence at the time of interview and were categorized as cases. Only 4 (14.3%) of 28 kidney transplant recipients who reported urinary incontinence at the time of the interview reported seeking care for uncontrollable urine leakage from a doctor after transplantation.
Sixty-one (61%) kidney transplant recipients denied urinary incontinence at some point in their life. Eleven (11%) kidney transplant recipients were categorized as pretransplant incontinent. Therefore, 72% of kidney transplant recipients denied urinary incontinence at the time of the interview and were categorized as controls. Only 4 (23.5%) of 17 kidney transplant recipients who reported urinary incontinence before transplantation reported seeking care for uncontrollable urine leakage from a doctor before transplantation.
A history of urinary incontinence before kidney failure was not predictive of post-transplant incontinence. Post-transplant urinary incontinence was reported in 6 (35.3%) of 17 study participants who reported pretransplant urinary incontinence, compared with 22 (26.5%) of 83 study participants who did not report pretransplant urinary incontinence (P = .5, odds ratio 1.3; 95% CI 0.6, 2.8; Table 1).
Stress, urge, and mixed incontinence were reported in 32.1%, 25%, and 42.9% of cases, respectively. The mean Sandvik's Incontinence Severity Index score was 5.1 (median 3.5; 95% CI 3.6, 6.6), indicative of moderate disease. The mean Incontinence Impact Questionnaire short form score was 12.9 (median 4.8; 95% CI 6.7, 18.9). Eight (28.6%) patients reported that their symptoms were worse, 15 (53.6%) patients reported no change, and 5 (17.8%) patients reported their symptoms were better since the onset of incontinence. The mean number of pads worn per day by incontinent kidney transplant recipients was 1.0 (median 0; mode 0; 95% CI 0.4, 1.6).
Renal function, as measured by serum creatinine after transplantation, was not associated with incontinence severity (ρ = −.09, P = .6) or impact of urinary incontinence on activities of daily living (ρ = −.07, P = .7). Previously established clinical and sociodemographic predictors of urinary incontinence in community-dwelling nontransplant recipients, including age, body mass index, vaginal parity, and smoking history, were similar in continent and incontinent kidney transplant recipients. Creatinine levels and duration from transplantation to interview were similar as well (Table 2).
Twenty-five kidney transplant recipients completed their 3-day 24-hour bladder diaries as requested. Three-day 24-hour bladder diaries were completed by 8 (28.6%) of incontinent kidney transplant recipients and 17 (23.6%) of continent kidney transplant recipients. Diary completers did not differ from diary noncompleters with respect to age, vaginal parity, tobacco use, pre- or post-transplant continence status, serum creatinine level, or duration in years from transplantation to interview. Diary completers had lower body mass indices than diary noncompleters (27.2 ± 6.4 versus 30.1 ± 6.3 kg/m2; P = .048).
Incontinent kidney transplant recipients recorded a greater daily fluid intake on their 3-day 24 hour-bladder diaries than continent recipients (2,951.9 ± 1,228.6 mL versus 1,854.6 ± 614.8 mL, P = .04). We found no significant difference in mean voided volumes between incontinent and continent kidney transplant recipients who completed their diaries (271.0 ± 97.2 mL versus 340.1 ± 145.5 mL, P = .2). Therefore, incontinent kidney transplant recipients who completed their diaries recorded a greater number of voids per day than continent kidney transplant recipients who completed their diaries (10.8 ± 5.9 versus 6.7 ± 1.6, P = .01). There was a significant difference in recorded leaks per day between incontinent and continent kidney transplant recipients who completed their diaries as expected (3.4 ± 3.1 [median 2.8] versus 0.9 ± 2.6 [median 0], P = .01 for median). Six of 17 “continent” kidney transplant recipients recorded urine loss on their 3-day 24-hour bladder diaries. Renal function, as measured indirectly by 24-hour urine output in kidney recipients who completed their diaries, was not associated with continence status (incontinent kidney transplant recipients 2,696.9 ± 1,198.8 mL versus continent kidney transplant recipients 2,259.3 ± 957.9 mL, P = .34).
The prevalence and severity of urinary incontinence affecting kidney transplant recipients was similar to community-dwelling nontransplant recipients according to the results of a large population-based epidemiological study from Norway.2 The prevalence rate was somewhat smaller than the rate reported for community-dwelling nontransplant recipients from the United States (37%), where the question used to identify cases for this study was drawn.3 Kidney transplant recipients reported more urge incontinence, both pure and mixed, than community-dwelling nontransplant recipients from Norway, who reported more stress incontinence.2 We were unable to prove our hypothesis that poor kidney function, as measured by elevated serum creatinine after transplantation, was associated with lower incontinence severity scores and a milder impact of incontinence on activities of daily living. None of the previously established clinical or sociodemographic predictors of urinary incontinence1 in community-dwelling nontransplant recipients were applicable to kidney transplant recipients despite similarities in the prevalence and severity of disease.
We found an association between mean daily fluid intake and continence status in kidney transplant recipients who completed their diaries. Incontinent kidney transplant recipients drank approximately 1 L of fluid more per day than continent recipients. Additionally, the mean daily fluid intake of incontinent kidney transplant recipients varied approximately twice as much as continent recipients. Although it is impossible to prove causality in a cross-sectional study, it would seem prudent to advise new transplant recipients to limit their daily fluid intake to the recommended levels7 (6 to 8 8-oz glasses of fluid per day) because they may predispose themselves to leakage through greater daily fluid intakes, which can now be filtered at the glomerular level8 en route to their urinary bladder. It is incorrect to generalize these findings, yet it is reasonable to advise all women of the association between 24-hour fluid intake and continence status because restricting daily fluid intake to recommended levels is an effective coping strategy for urinary incontinence with negligible side effects.9
Van der Weide et al8 surveyed 63 kidney transplant recipients and 74 nontransplant controls and found urinary incontinence reported in 14.3% of kidney transplant recipients and 18.9% of controls. Forty-nine percent of kidney transplant recipients voided 7 times or more per day compared with 23% of the control group. Sixty-two percent of transplant recipients voided twice or more per night compared with only 19% of the control group. These differences persisted after adjusting for volume of fluid intake. Based on the results of their study, the authors concluded that urinary frequency was related to decreased bladder capacity after kidney transplantation. The major limitation of their study was the lack of frequency volume charts to objectively compare bladder capacity, urinary frequency, and fluid intake in kidney transplant recipients and controls.8
We did not find an association between bladder capacities, as measured by mean voided volumes, and continence status in kidney transplant recipients who completed their diaries. It is likely, therefore, that the increased mean daily fluid intake seen in our incontinent kidney transplant recipients resulted in their increased number of voids per day compared with continent kidney transplant recipients who completed their diaries. Based on our findings, we are unable to attribute the high prevalence of urinary frequency7 and urge incontinence in kidney transplant recipients to decreased bladder capacity, as measured by mean voided volume. However, 51 diaries from continent and incontinent kidney transplant recipients would need to be compared to validate our conclusions because our power to detect a 20% relative difference in bladder capacity, as measured by mean voided volume, was only 29% with an α significance level of .05.
Incontinent kidney transplant recipients reported a moderately severe degree of urinary incontinence yet reported only a minimal impact of urinary incontinence on activities of daily living. We hypothesize that the impact of urinary incontinence on activities of daily living, as measured by a disease-specific quality-of-life instrument is dependent on comorbid conditions. Kidney transplant recipients report major improvements in quality of life after surgery when hemodialysis is no longer required. These improvements may modify the impact of urinary incontinence on activities of daily living in kidney transplant recipients. Our study group plans on testing this hypothesis by comparing the impact of urinary incontinence on activities of daily living in kidney transplant recipients to nonrecipients matched for incontinence severity.
A history of urinary incontinence before kidney failure was not predictive of post-transplant incontinence. However, we had a 9% probability of being correct when drawing this conclusion based on the 33% increased rate of post-transplant urinary incontinence reported by pretransplant incontinent women compared with pretransplant continent women. We would need to interview 454 incontinent kidney transplant recipients to increase this probability to 80%. Because we cannot predict who is at risk, counseling about the possibility of urinary incontinence after renal transplant should be extended to all women undergoing the procedure. Our study provides more evidence that only a small proportion of incontinent women seek care from a physician. This is especially discouraging for this cohort of women who report a moderately severe degree of disease and see a nephrologist monthly for post-transplant care. This suggests that for kidney transplant recipients the impact of urinary incontinence on activities of daily living may be a stronger predictor of care seeking than incontinence severity. However, large knowledge gaps in our understanding of incontinence care-seeking behaviors remain.10 Interventions can be designed to increase the percentage of women who seek continence care as these knowledge gaps are filled.
There are several limitations of our study that must be considered before our conclusions can be accepted. Our power to detect significant associations among measures of renal function, incontinence severity, and impact of incontinence on activities of daily living or determine whether established clinical and sociodemographic predictors of incontinence1 in community-dwelling nontransplant recipients were applicable to kidney transplant recipients was limited by the number of incontinent kidney transplant recipients interviewed. We could not justify a sample size calculation because we were unable to draw from previously published work that focused on urinary incontinence in kidney transplant recipients.
As a group, kidney transplant recipients have more comorbidity than age-matched community-dwelling nontransplant recipients.11,12 We did not want these differences in comorbidity to confound the relationship between transplant and continence status, if one existed. Therefore, we designed our study to minimize confounding bias by comparing only incontinent kidney transplant recipients with continent kidney transplant recipients because both groups would be expected to have similar comorbidity, whether measured or not.
Another limitation of our study was missing diary data that may have introduced nonresponse bias into our study. We felt that diary completers were representative of our entire kidney transplant cohort because important sociodemographic and clinical data were similar, except for body mass index, when comparing kidney transplant recipients who completed their bladder diaries to kidney transplant recipients who did not complete their diaries. Diary completers as a group may be healthier than diary noncompleters, but incontinent noncompleters are unlikely to be heavier than continent noncompleters because body mass index was not related to continence status in kidney transplant recipients. Measurement bias threatens the validity of our study because mixed survey methodologies were used to collect data. Sensitive issues, such as urinary incontinence, may be easier for women to discuss with a female clinical research coordinator face to face compared with a telephone interview. Our study group felt that the additional power achieved by conducting telephone interviews with eligible kidney transplant recipients who were not scheduled for follow-up during the study period outweighed the threats to validity when mixed survey methodologies are used. We minimized this threat by having the same clinical research coordinator interview study participants face to face and by telephone.
In conclusion, the prevalence and severity of urinary incontinence affecting female kidney transplant recipients is similar to community-dwelling nontransplant recipients.2 Incontinent kidney transplant recipients also are similar to incontinent community-dwelling nontransplant recipients in that they seek care from physicians in a minimal number of cases. Measures of kidney function and previous established clinical and sociodemographic predictors of urinary incontinence1 in community-dwelling nontransplant recipients were not associated with severity of urinary incontinence, impact of incontinence on activities of daily living, or continence status in female kidney transplant recipients. Incontinent kidney transplant patients may predispose themselves to leakage through greater daily fluid intake, which can now be filtered at the glomerular level. Urinary incontinence in kidney transplant recipients has a minimal impact on activities of daily living despite a moderately severe degree of urinary incontinence. We hypothesize that the impact of urinary incontinence on activities of daily living is dependent on comorbid conditions, in this case transplant status.
1.Brown JS, Nyberg LM, Kusek JW, Burgio KL, Diokno AC, Foldspang A, et al. National Institute of Diabetes and Digestive Kidney Diseases International Research Working Group on Bladder Dysfunction. Proceedings of the National Institute of Diabetes and Digestive and Kidney Diseases International Symposium on Epidemiologic Issues in Urinary Incontinence in Women [review]. Am J Obstet Gynecol 2003;188:S77–88.
2.Hannestad YS, Rortveit G, Sandvik H, Hunskaar S of the Norwegian EPINCONT study. Epidemiology of Incontinence in the County of Nord-Trondelag. A Community-based epidemiological survey of female urinary incontinence: the Norwegian EPINCONT study. Epidemiology of Incontinence in the County of Nord-Trondelag. J Clin Epidemiol 2000;53:1150–7.
3.Diokno AC, Brock BM, Brown MB, Herzog AR. Prevalence of urinary incontinence and other urological symptoms in the noninstitutionalized elderly. J Urol 1986;136:1022–5.
4.Sandvik H, Hunskaar S, Vanvik A, Bratt H, Seim A, Hermstad R. Diagnostic classification of female urinary incontinence: an epidemiological survey corrected for validity. J Clin Epidemiol 1995;48:339–43.
5.Sandvik H, Seim A, Vanvik A, Hunskaar S. A severity index for epidemiological surveys of female urinary incontinence: comparison with 48-hour pad-weighing tests. Neurourol Urodyn 2000;19:137–45.
6.Uebersax J, Wyman J, Shumaker S, McClish D, Fantl A. Short forms to assess life quality and symptom distress for urinary incontinence in women: the Incontinence Impact Questionnaire and Urogenital Distress Inventory. Neurourol Urodyn 1995;14:131–9.
7.Valtin H. “Drink at Least Eight Glasses of Water a Day.” Really? Is There Scientific Evidence for “8 × 8” [review]? Am J Physiol Regul Integr Comp Physiol 2002;283:R993–1004.
8.Van der Weide MJ, Hilbrands LB, Bemelmans BL, Meuleman EJ, Frederiks CM. Lower urinary tract symptoms after renal transplantation. J Urol 2001;166:1237–41.
9.Engberg SJ, McDowell BJ, Burgio KL, Watson JE, Belle S. Self care behaviors of older women with urinary incontinence. J Geront Nurs 1995;21:7–14.
10.Hunskaar S, Arnold EP, Burgio K, Herzog AR. Epidemiology and Natural History of Urinary Incontinence [review]. Int Urogynecol J Pelvic Floor Dysfunct 2000;11:301–19.
11.Rebollo P, Ortega F, Baltar JM, Badia X, Alvarez-Ude F, Diaz-Corte C, et al. Health related quality of life (HRQOL) of kidney transplanted patients: variables that influence it. Clin Transplant 2000;14:199–207.
12.Matas AJ, Halbert RJ, Barr ML, Helderman JH, Hricik DE, Pirsch JD, et al. Life satisfaction and adverse effects in renal transplant recipients: a longitudinal analysis. Clin Transplant 2002;16:113–121.