Concordance Between One-hour Pad Test and Subjective Assessment of Stress Incontinence

Concordance Between One-hour Pad Test and Subjective Assessment of Stress Incontinence

Female Urology Concordance Between One-hour Pad Test and Subjective Assessment of Stress Incontinence M. Liebergall-Wischnitzer, O. Paltiel, D. Hochne...

251KB Sizes 0 Downloads 16 Views

Female Urology Concordance Between One-hour Pad Test and Subjective Assessment of Stress Incontinence M. Liebergall-Wischnitzer, O. Paltiel, D. Hochner-Celnikier, Y. Lavy, D. Shveiky, and O. Manor OBJECTIVES

METHODS

RESULTS

CONCLUSIONS

To examine the concordance among the 1-hour pad test results, subjective questions regarding incontinence, and a quality-of-life questionnaire to assess the role of the pad test as a noninvasive measurement tool in clinical trials. The 1-hour nonstandard pad test is one of several quantitative tools used to measure urinary incontinence; however, its utility has been questioned. The study subjects were women participating in 2 clinical trials evaluating noninvasive interventions: circular muscle exercises versus pelvic floor muscle training for urinary incontinence. The quantity of urinary leakage according to the pad test and questions regarding subjective urinary leakage from the quality-of-life questionnaire were evaluated for all study subjects combined and in subgroups. A total of 731 clinical pad tests were evaluated from the 2 trials. Significant associations were found between several questions regarding subjective leakage and the pad test results in the study subgroups. A significant correlation was seen between the pad test results and the quality-of-life questionnaire scores (r ⫽ 0.14 before intervention and r ⫽ 0.42 after intervention in the combined studies; P ⬍ .05). The 1-hour pad test demonstrated concordance with subjective assessment tools for urinary incontinence and should be considered a part of the armamentarium for assessing the severity of this condition. UROLOGY 76: 1364 –1368, 2010. © 2010 Elsevier Inc.

T

he 1-hour clinical pad test is 1 of several quantitative tools recommended by the International Continence Society1 for assessing the extent of urinary incontinence. The International Urogynecological Association has recently recommended the pad test for clinical studies to measure urinary leakage.2 The test is used as an objective assessment to distinguish between continent and incontinent patients and as an evaluation instrument before and after interventions. A 1-g gain in pad weight is considered a positive result (ie, proof of incontinence).3,4 However, a wide variation has been seen in the 1-hour pad test results,5 although the reproducibility has been improved with standard bladder filling6 or ultrasound scanning.7 Despite its disadvantages, for lack of a better clinical, noninvasive quantitative tool, we chose the 1-hour pad test (nonstandard, without catheterization) as the main assessment tool in 2 randomized clinical trials: a pilot trial and a From the Henrietta Szold Hadassah-Hebrew University School of Nursing, Faculty of Medicine, Jerusalem, Israel; Braun Hebrew University Hadassah School of Public Health and Community Medicine, Faculty of Medicine, Jerusalem, Israel; and Department of Obstetric and Gynecology, Hadassah University Medical Center, Jerusalem, Israel Reprint requests: Michael Liebergall-Wischnitzer, R.N., Ph.D., Henrietta Szold Hadassah-Hebrew University School of Nursing, P.O. Box 12000, Jerusalem 91120 Israel. E-mail: [email protected] Submitted: March 18, 2010, received (with revisions): May 27, 2010

1364

© 2010 Elsevier Inc. All Rights Reserved

community-based trial, testing the effectiveness of circular muscle exercises versus pelvic floor muscle training for stress urinary incontinence (SUI) in women.8,9 We assumed that women who chose to participate in a study of exercise treatment of SUI would not easily accept invasive testing. The 1-hour clinical pad test was followed by completion of a questionnaire that included items regarding subjective urinary symptoms10-13 and quality of life.14 The present study’s objective was to examine the association between the results of the 1-hour clinical pad test and the findings from subjective questions regarding urinary incontinence and a quality-of-life questionnaire to assess the role of the pad test as a valid quantitative tool for use in clinical trials. The construct validity of a tool is determined by consistent associations between the tool and other variables that reasonably link together. Because we did not use a reference standard objective measurement of urinary leakage, we inferred construct validity when the results of the pad test showed concordance with subjective measures of SUI.

MATERIAL AND METHODS Women eligible to participate in both trials were aged 20-65 years, with a positive pad test result of at least 1 g of urinary 0090-4295/10/$36.00 doi:10.1016/j.urology.2010.05.048

leakage. Women with SUI complaints but without a demonstrated leakage of ⱖ1 g on the pad test; those with genital prolapse of grade 3 or greater; pregnant or breastfeeding women; those within 12 weeks of delivery, 6 weeks of abortion, or 6 months of pelvic surgery; and those with any previous surgery for urinary incontinence, previous pelvic radiotherapy, or with a significant comorbidity that limited physical activity were excluded.8,9 All participants underwent quantitative and qualitative assessments of urinary leakage.

Quantitative Assessment The 1-hour clinical pad test was performed in the pilot study according to the International Continence Society instructions, which include not voiding for 2 hours before the pad test and to wear a weighed pad.8,9 The participants were asked to sit and drink 500 mL of sodium-free water. Next, the women were instructed to walk for 30 minutes. The walk included going up and down 24 steps. On returning to the clinic, the participants were instructed to perform several activities, including standing and sitting 10 times, coughing vigorously 10 times, running in place for 1 minute, picking up an object from the floor 5 times, and putting their hands under water for 1 minute. After the activities were completed, the pad was reweighed to measure the amount of urinary leakage. From the results of the pilot study, the instructions were modified for the community-based trial. The participating women were instructed to avoid caffeine drinks for 2 hours before the test, to drink low-sodium (9 mg/L) water instead of sodium-free water, and to add 5 jumping jacks as a part of the physical activities. The 1-hour clinical pad test is intended to duplicate the stimuli of everyday life.3 All nurses involved in performing the pad test had received previous training from the primary investigator. No urodynamic tests were performed.

Qualitative Assessment The participants completed health and urinary leakage questionnaires. Questions regarding urinary leakage were determined from variables used in other validated questionnaires that had been partially modified to meet the needs of a Hebrewspeaking population10-13 and were pretested in the pilot study.9 Additionally, after Hebrew validation, we administered a quality-of-life questionnaire.14 The scoring was determined from 22 items, each measured on a -point scale and subdivided into 3 subscores: (a) avoidance and limiting behavior score— 8 items; (b) psychosocial impact score—9 items; and (c) social embarrassment score—5 items.14

Ethical Issues The Hadassah-Hebrew University institutional review board (Helsinki Committee) approved both trials, and all women provided written informed consent after receiving an explanation of the study goals and procedures.

Statistical Analysis The associations between the pad test results and subjective assessments of urinary leakage (amount, level of annoyance, and frequency of leakage in general and during the past week) were examined using linear models. The associations between the pad test results and quality-of-life scores were evaluated using Pearson’s correlation coefficient, and confidence intervals were calculated using Fisher’s Z-transformation. Separate analUROLOGY 76 (6), 2010

Figure 1. Flow chart of pad test administration.

Table 1. Demographic and urinary leakage data before intervention Variable

Pilot Study

Age group (n) 20-50 y 28 (48.3) 51-65 y 30 (51.7) Stress incontinence (n) 57 (96.6) Mixed incontinence (n) 29 (50) Pad test (g) Mean ⫾ SD 10.4 ⫾ 15.6 Range 1-106.1

Community Study 147 (61.5) 92 (38.5) 237 (98.8) 102 (42.6) 11.5 ⫾ 15.3 1-139.5

Data in parentheses are percentages.

yses were conducted for each study (pilot study and communitybased trial) and of measurements taken before and after the intervention (the latter by intervention group). Finally, the results for the 2 studies were combined to obtain overall measures of association. In the combined analysis, the linear models were adjusted for study type, and the correlation coefficients and confidence intervals were combined using inverse variance weights.

RESULTS A total of 731 pad tests were performed on 496 women (100 women from 2001 to 2002 in the pilot study and 396 women from 2004 to 2006 in the community-based trial). Of these, 35 (pilot study) and 151 (communitybased trial) women did not enter the intervention trials because they had not met the inclusion criteria of a pad test result of ⱖ1 g or had prolapse ⬎2°, or met other exclusion criteria. A total of 59 (pilot study) and 240 (community-based trial) women met the inclusion criteria and participated in the interventions. A total of 59 women from the pilot study and 176 from the community-based trial were retested after 12 weeks of intervention. The remainder did not return for the postintervention pad test and thus only provided 1 observation to the study (Fig. 1). Of the participants, 28 (48.3%) in the pilot study and 147 (61.5%) in the community-based trial were aged 1365

Table 2. Associations between pad test and subjective questions regarding urinary incontinence before and after intervention* Combined Studies Before Intervention Pad Test Amount* Few drops/small amount Moderate amount Large amount Annoyance Not at all/seldom Sometimes Often/very often Frequency ⱖ1⫻/mo/once every 2 mo ⱖ1⫻/wk ⱖ1⫻/d Leakage during past week Never Once-twice Three times or more

Results Patients (n) Mean ⫾ SD

Community-Based Trial After Intervention Paula PFMT Patients (n) Mean ⫾ SD Patients (n) Mean ⫾ SD

152 (50.8) 123 (41.1) 17 (5.7)

7.6 ⫾ 9.1† 13.4 ⫾ 14.2 16.7 ⫾ 14.5

74 (82.2) 16 (17.7)

2.2 ⫾ 6.3‡ 6.9 ⫾ 10.0

60 (72.2) 23 (27.8)

2.2 ⫾ 3.8‡ 4.3 ⫾ 5.1

16 (5.4) 97 (32.7) 187 (62.0)

7.0 ⫾ 8.8 9.7 ⫾ 11.7 11.3 ⫾ 12.7

35 (38.5) 42 (46.1) 14 (15.4)

0.6 ⫾ 0.8† 2.7 ⫾ 5.5 9.8 ⫾ 14.0

23 (27.7) 33 (39.8) 27 (32.5)

1.2 ⫾ 2.6† 1.8 ⫾ 3.0 5.2 ⫾ 5.7

20 (7.0) 86 (30.0) 121 (63.1)

6.5 ⫾ 6.2‡ 8.1 ⫾ 10.5 12.3 ⫾ 13.3

25 (28.1) 46 (51.7) 18 (20.2)

1.0 ⫾ 1.8 3.0 ⫾ 6.9 6.1 ⫾ 11.2

26 (31.3) 29 (34.9) 28 (33.7)

1.2 ⫾ 2.8‡ 2.7 ⫾ 3.4 4.2 ⫾ 5.7

25 (8.7) 133 (46.3) 129 (44.9)

5.2 ⫾ 6.0† 8.9 ⫾ 10.6 13.1 ⫾ 13.6

30 (33.3) 48 (53.3) 12 (13.3)

0.6 ⫾ 1.1† 2.3 ⫾ 5.0 11.9 ⫾ 14.5

21 (25.9) 41 (50.6) 19 (23.5)

1.2 ⫾ 2.5‡ 2.5 ⫾ 4.4 5.2 ⫾ 5.0

Paula, circular muscle exercises; PFMT, pelvic floor muscle training. Data in parentheses are percentages. * Some categories were combined owing to distribution of participants’ responses. † P ⬍ .01. ‡ P ⬍ .05.

20-50 years and 30 (51.7%) and 92 (38.5%) were aged 51-65 years, respectively (Table 1). SUI was reported by 97% (pilot study) and 99% (community-based trial) of the participants, and 50% (pilot study) and 43% (community-based trial) of the participants reported mixed incontinence (ie, also had feelings of urgency). The mean pad test results were 10.4 ⫾ 15.6 g in the pilot study and 11.5 ⫾ 15.3 g in the community-based trial before the intervention. Two outliers with baseline results of 106.2 g from the pilot study and 138.5 g from the communitybased trial were not included in the results (Table 1). In the combined studies, significant associations were found between the pad test results and questions regarding the leakage amount and leakage frequency during the past week before the intervention. For example, of the women reporting leakage of “a few drops” or “a small amount,” the mean pad test result was 7.6 ⫾ 9.1 g compared with 16.7 ⫾ 14.5 g among those reporting “a large amount” (P ⬍ .01; Table 2). After the interventions, in the community-based trial, all subjective questions showed significant associations with the pad test results in both exercise groups, with the exception of the frequency question in the circular muscle exercises group (Table 2). Before the intervention, the analysis on the results from both studies noted a statistically significant correlation of 0.14 between the pad test results and the qualityof-life total score (P ⬍ .05). Similar correlations were observed for the scores of avoidance and limiting behavior and for psychosocial impact. The combined analysis of measurements taken after the intervention showed substantial correlation coefficients between the pad test 1366

results and overall quality of life, with subscores of 0.370.42 (P ⬍ .05; Table 3).

COMMENT To assess the role of the pad test as a valid tool in measuring female SUI we analyzed our pad test experience and assessed the concordance between the pad test results and subjective urinary question results and responses regarding female incontinence using a quality-oflife questionnaire. The pad test was performed on 496 women and served as both a useful diagnostic tool for confirming SUI and as a quantitative test to assess urinary leakage before and after an intervention. To the best of our knowledge, this is 1 of the largest reported databases using the 1-hour clinical pad test and evaluating its performance characteristics. The test was easy to perform and relatively inexpensive and was considered an acceptable tool by the women tested. In addition to the advantages of this tool, it has some disadvantages. A total of 35 women (35%) were excluded from our pilot study and 147 (31.1%) excluded from the community-based trial. These women reported symptoms of urinary leakage and were interested in entering the studies but did not show urinary leakage with the 1-hour pad test using a threshold of 1 g.3 These findings were similar to those from other studies in which 30% of women reported that the pad test results, followed by ultrasonography, did not represent their normal leakage.15 A 1-g threshold excluded some incontinent women from the study, because they did not meet the inclusion criteria. An option exists of repeating the test, and we did so in some cases, but the problem remains if, with the UROLOGY 76 (6), 2010

0.56 (0.25, 0.77)* 0.06 (⫺0.32, 0.42) Abbreviations as in Table 2. * P ⬍ .05.

0.10 (⫺0.03, 0.23) 0.01 (⫺0.24, 0.27)

0.07 (⫺0.036, 0.194)

0.14 (0.02, 0.27)* 0.13 (⫺0.13, 0.38)

0.14 (0.026, 0.256)

0.39 (0.03, 0.66)* 0.00 (⫺0.37, 0.37)

0.31 (0.11, 0.48)* 0.34 (0.13, 0.51)*

0.26 (0.06, 0.44)* 0.34 (0.14, 0.51)*

0.39 (0.25, 0.52)

0.37 (0.24, 0.51)

0.40 (0.27, 0.54) 0.32 (⫺0.05, 0.61) 0.15 (⫺0.24, 0.49) 0.14 (0.01, 0.26)* 0.25 (0.00, 0.48)*

0.14 (0.03, 0.258) 0.14 (0.01, 0.27)*

Total quality of life Paula PFMT Avoidance and limiting behavior Paula PFMT Psychosocial impact Paula PFMT Social embarrassment score Paula PFMT

0.16 (⫺0.10, 0.41)

0.15 (0.43, 0.273)

0.44 (0.10, 0.69)* 0.09 (⫺0.30, 0.44)

0.27 (0.07, 0.44)* 0.40 (0.20, 0.56)*

0.42 (0.29, 0.56) 0.30 (0.09, 0.48)* 0.39 (0.20, 0.56)*

Paula and PFMT Combined Community Study Quality of Life Scale Score

Pilot Study

Community Study

Combined

Pilot Study

After Intervention Before Intervention

Table 3. Estimated correlation coefficients and 95% confidence intervals for association between pad test and quality of life scores

UROLOGY 76 (6), 2010

repeated test, the threshold has not been achieved. Ryhammer et al16 suggested that in questionable cases, a urine tracing method be used to avoid the predetermined cutoff level, and Arvonen et al17 recommended additional tests under specific conditions for those women who experienced urinary leakage that could not be demonstrated. To enhance our ability to study the incontinent women, we added an additional provocation in the community-based trial, namely 5 jumping jacks, to stimulate leakage.9 The ratio between the tested women and eligible women changed from 1:1.5 in the pilot study to 1:1.6 in community-based trial; thus, resulting in decreases in the length of the study’s measurement and its costs. This relation might have also resulted from a deviation from the International Continence Society recommendations used in the pilot study,1 by asking the participants to avoid caffeinated drinks 2 hours before the test in the community-based trial. In contrast, other trials required drinking avoidance for 2-4 hours before the test.18 Also, we changed the protocol in the communitybased trial to include low-sodium water (9 mg/L) instead of sodium-free water,19 in addition to added jumping jacks, as reported in another trial.20 We found these modifications contributed to the standardization of the test and increased the number of potentially eligible women. Other researchers have modified the pad test by adding an ultrasound scan before conducting the pad test to measure the functional bladder capacity and showed good test-retest reliability.7 It is clear that standardization is lacking and acceptable guidelines are needed.5,21 We found a significant association between the subjective questions regarding urinary leakage and the pad test results in both studies after the intervention, but it was not as strong as the correlation found between subjective reports of leakage frequency and volume and the pad test results that was reported by Jackson et al12 (r ⱖ 0.5). However, their population entered the study after urodynamic evaluation, which could have contributed to the stronger correlation.12 Abdel-fattah et al22 showed a correlation of r ⫽ 0.48 between a leakage severity scale and the pad test results. The difference between the results might have resulted from their use of a general scale regarding urinary incontinence versus our specific questions regarding the amount, frequency, and so forth and their use of the standard pad test. We found a significant, but modest, correlation between the pad test results and the scores of the qualityof-life questionnaire, a validated measure of the quality of life with incontinence,22,23 before the intervention (r ⫽ 0.143) and a substantially significant correlation after the intervention (r ⫽ 0.422). The smaller correlation coefficient observed before the intervention versus after stemmed from the truncated range that characterized the data before the intervention, because only women with pad test results of ⱖ1 g were recruited for the study. It could be questioned whether the pad test is needed as a quantitative tool or whether questionnaires could 1367

used alone, such as was concluded by Abdel-fattah et al.22 The pad test is 1 of few quantitative tools used for assessing urinary incontinence, and, despite its low reproducibility, no superior, noninvasive clinical tool is available. In women with SUI after urodynamic evaluation, Franco et al24 described low correlations between the pad test results and a variety of questionnaires, with the greatest correlation of 0.177 found between the pad test results and quality-of-life questionnaire results. In our studies, the population did not undergo urodynamic testing, which might have influenced the correlation results. In contrast to the findings from Abdel-fattah et al,22 we have demonstrated that the pad test is a useful tool, allowing quantitative comparisons before and after intervention. We believe it can serve as 1 element of the armamentarium that includes a detailed patient history and validated questionnaires that include questions of symptoms and quality-of-life measures. We agree with Nygaard4 that physiologic and subjective outcomes should be measured in clinical trials because every tool can contribute to the clinical information. Listening to women’s complaints and assessing their condition using a number of different tools could assist in tailoring suitable treatments to them.

CONCLUSIONS The 1-hour pad test should be 1 of several assessment instruments used to measure female urinary leakage when used together with patient history and subjective questionnaires regarding urinary incontinence and quality of life. References 1. Abrams P, Cardozo L, Fall M, et al. The standardization of terminology of lower urinary tract function: report from the Standardization Sub-Committee of the International Continence Society. Neurourol Urodyn. 2002;21:167-178. 2. Ghoniem G, Stanford E, Kenton K, et al. Evaluation and outcome measures in the treatment of female urinary stress incontinence: International Urogynecological Association (IUGA) guidelines for research and clinical practice. Int Urogynecol J Pelvic Floor Dysfunct. 2008;19:5-33. 3. Abrams P, Blaivas J, Stanton S, et al. The standardization of terminology of lower urinary tract function. The International Continence Society Committee on Standardization of Terminology. Scand J Urol Nephrol Suppl. 1988;114:5-19. 4. Nygaard I. Physiologic outcome measures for urinary incontinence. Gastroenterology. 2004;126:S99-105. 5. Soroka D, Drutz HP, Glazener CM, et al. Perineal pad test in evaluating outcome of treatments for female incontinence: a systematic review. Int Urogynecol J Pelvic Floor Dysfunct. 2002;13:165175. 6. Lose G, Walter S. Pad weighing tests. Methods Clin Urodyn. 1991; 1:17-35.

1368

7. Paick JS, Ku JH, Shin JW, et al. Significance of pad test loss for the evaluation of women with urinary incontinence. Neurourol Urodyn. 2005;24:39-43. 8. Liebergall-Wischnitzer M, Hochner-Celnikier D, Lavy Y, et al. Paula method of circular muscle exercises for urinary stress incontinence—a clinical trial. Int Urogynecol J Pelvic Floor Dysfunct. 2005;16:345-351. 9. Liebergall-Wischnitzer M, Hochner-Celnikier D, Lavy Y, et al. Randomized trial of circular muscle versus pelvic floor training for stress urinary incontinence in women. J Womens Health. 2009;18: 377-385. 10. Hannestad Y, Rortveit G, Sandvik H, et al. 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;53:1150-1157. 11. Ishiko O, Hirai K, Sumi T, et al. The urinary incontinence score in the diagnosis of female urinary incontinence. Int J Gynecol Obstet. 2000;68:131-137. 12. Jackson S, Donovan J, Brookes S, et al. The Bristol female lower urinary tract symptoms questionnaire: development and psychometric testing. Br J Urol. 1996;77:805-812. 13. Sandvik H, Espuna M, Hunskaar S. Validity of the incontinence severity index: comparison with pad-weighing tests. Int Urogynecol J Pelvic Floor Dysfunct. 2006;17:520-524. 14. Patrick DL, Martin ML, Bushnell DM, et al. Quality of life of women with urinary incontinence: further development of the incontinence quality of life instrument (I-QOL). Urology. 1999;53: 71-76. 15. Simons A, Yoong W, Buckland S, et al. Inadequate repeatability of the one-hour pad test: the need for a new incontinence outcome measure. Br J Obstet Gynaecol. 2001;108:315-319. 16. Ryhammer AM, Djurhuus JC, Laurberg S. Pad testing in incontinent women: a review. Int Urogynecol J Pelvic Floor Dysfunct. 1999;10:111-115. 17. Arvonen T, Fianu-Jonasson A, Tyni-Lenne R. Effectiveness of two conservative modes of physical therapy in women with urinary stress incontinence. Neurourol Urodyn. 2001;20:591-599. 18. Henalla SM, Kirwan P, Castleden CM, et al. The effect of pelvic floor exercises in the treatment of genuine urinary stress incontinence in women at two hospitals. Br J Obstet Gynaecol. 1998;95: 602-606. 19. Peattie AB, Plevnik S, Stanton SL. Vaginal cones: a conservative method of treating genuine stress incontinence. Br J Obstet Gynaecol. 1988;95:1049-1053. 20. Morkved S, Bo K, Fjortoft T. Effect of adding biofeedback to pelvic floor muscle training to treat urodynamic stress incontinence. Obstet Gynecol. 2002;100:730-739. 21. Martin JL, Williams KS, Sutton AJ, et al. Systematic review and meta-analysis of methods of diagnostic assessment for urinary incontinence. Neurourol Urodyn. 2006;25:674-684. 22. Abdel-fattah M, Barrington J, Youssef M. The standard 1-hour pad test: does it have any value in clinical practice? Eur Urol. 2004;46: 377-380. 23. Rett MT, Simoes JA, Herrmann V, et al. Management of stress urinary incontinence with surface electromyography-assisted biofeedback in women of reproductive age. Phys Ther. 2007;87:136142. 24. Franco AV, Lee F, Fynes MM. Is there an alternative to pad tests? Correlation of subjective variables of severity of urinary loss to the 1-h pad test in women with stress urinary incontinence. BJU Int. 2008;102:586-590.

UROLOGY 76 (6), 2010