VOIDING PATTERN AND ACQUISITION OF BLADDER CONTROL FROM BIRTH TO AGE 6 YEARS A LONGITUDINAL STUDY

Transcription

1 /05/ /0 Vol. 174, , July 2005 THE JOURNAL OF UROLOGY Printed in U.S.A. Copyright 2005 by AMERICAN UROLOGICAL ASSOCIATION DOI: /01.ju e3 VOIDING PATTERN AND ACQUISITION OF BLADDER CONTROL FROM BIRTH TO AGE 6 YEARS A LONGITUDINAL STUDY U.-B. JANSSON,* M. HANSON, U. SILLÉN AND A.-L. HELLSTRÖM From the Department of Surgery (Urotherapeutic Unit, U-BJ, MH, A-LH) and Pediatric Uro-Nephrological Centre (US), Queen Silvia Children s Hospital and Sahlgrenska Academy, Faculty of Health and Caring Sciences, Institute of Nursing, University of Göteborg (A-LH), Göteborg, Sweden ABSTRACT Purpose: We describe the voiding pattern and acquisition of bladder control in healthy children up to age 6 years. Materials and Methods: We determined age for daytime and nighttime dryness, voiding patterns, voiding volumes and post-void residual volume per 4 hours individually and noninvasively every 3 months up to age 3 years and every 6 months up to age 6 years in 36 female and 23 male patients using 4-hour voiding observation and uroflowmetry/ultrasound. Results: Median age for attaining daytime and nighttime dryness was 3.5 and 4 years, respectively. No significant difference was found between girls and boys. All but 1 child attained daytime dryness an average of 10 months before attaining nighttime dryness. Bladder sensation was reported in 31%, 79% and 100% of patients at ages 2, 3 and 4 years, respectively. Median bladder capacity was 67 ml, 123 ml and 140 ml at years 1, 3 and 6, respectively. Median post-void residual volume was 5.5 ml, 0 ml and 2 ml at ages 1, 3 and 6 years, respectively. Conclusions: Today bladder control is acquired at a later stage despite earlier awareness of bladder function. The occurrence of bladder sensation from age 1.5 years motivates an earlier start with toilet training. Infants with small post-void residual volume at age 6 months or large bladder capacity will probably attain daytime dryness earlier than those with large post-void residual volume at age 6 months orsmall bladdercapacity. KEY WORDS: bladder, urination, reference values, pediatrics Urinary continence is of psychosocial importance for children as well as for adults. 1 However, the age at which a child can be expected to be dry varies from 5 months to 6 years according to different studies The main reasons for this variation in age are probably due to the definition used and the support given to the child. In the 1950s it was not unusual to make a child sit on a potty chairas soon as he orshe was able to sit. However, since the 1960s, there has been a gradual and more child oriented approach not to force the child to use a potty chairortoilet. To obtain a more comprehensive picture of bladder function in childhood and the acquisition of bladdercontrol, we followed healthy patients from birth to age 6 years. In a previous article longitudinal results were reported, focusing on voiding frequency, bladder capacity and post-void residual volume up to age 3 years. 12 MATERIALS AND METHODS Voiding pattern was observed, followed and described in 59 healthy pediatric patients (36 females and 23 males) every 3 months 1 week until age 3 years, and then every 6 months 1 month until age 6 years. Healthy infants (no history of urinary tract infection, no malformation of the urinary tract that could affect bladder function) were volunteered consecutively for the study by the parents after receiving information by the nurse at the initial visit to a pediatric health care clinic in Göteborg, Sweden, and then by the investigator. Submitted forpublication September23, Supported by the Solstickan Foundation, Stockholm, Sweden. Study received institutional ethics committee approval. * Correspondence: Department of Surgery (Urotherapeutic Unit), Queen Silvia Children s Hospital, Smörlottsgatan 6, plan 1, Göteborg, Sweden (telephone: ; FAX: ; [email protected]). 289 Selection criteria were mothers of patients up to 2 months old who were willing to come to the clinic regularly for the next 6 years. Methods used were 4-hour voiding observation for as long as the patients were in diapers, 12 and uroflow measurement plus ultrasound when they had become dry. The parents were interviewed at each observation about demographic data and patient voiding habits at home. The normal daily routine was followed as much as possible. All observations were made at a preschool situated close to the home of the patient. Information was obtained about voided volumes, post-void residual volume and maximum urinary flow in the olderchildren. A total of 618 observations were made in 57 of 59 children (34 of 36 girls and 23 of 23 boys). Two patients participated at only 1 observation, and, therefore, were excluded from the analyses. At ages 0 to 3 years 459 observations (mean 8, median 8, range 3 to 11) were made and analyzed. 12 In the second part of the study, when the children were 3.5 to 6 years old, 159 observations were made every 6 months in 43 of the original group (24 girls and 19 boys), with 1 to 6 observations each (mean 3, median 3). The number of patients varied at each observation because of the longitudinal design. As a result, different numbers of cases are presented in different analyses and the percentage of the actual number is reported. Some children also became dry before age 3 years, and, therefore, some results after age 3 years include different numbers of patients. Dropouts were mainly due to parents having difficulty getting time off from work. Acquisition of bladder control was studied by interviewing the parents at each visit. The questions that were asked related to whether the child was dry during the day and night always (every day/night), often (2 to 6 days/nights weekly), sometimes (1 day/night weekly) ornever(less than 1 day/night

2 290 VOIDING PATTERN AND BLADDER CONTROL FROM BIRTH TO AGE 6 YEARS TABLE 1. Ages of 46 children at achievement of daytime and nighttime dryness Pt Age (yrs) Daytime Dryness Nighttime Dryness No. Girls (%) No. Boys (%) No. Girls (%) No. Boys (%) 2 3 (11) 0 (0) 0 (0) 0 (0) (43) 3 (17) 0 (0) 2 (11) 3 17 (61) 7 (39) 4 (14) 4 (22) (79) 16 (89) 8 (29) 10 (56) 4 27 (96) 16 (89) 18 (64) 11 (61) (100) 17 (94) 20 (71) 12 (67) 5 28 (100) 18 (100) 21 (75) 15 (83) (75) 16 (89) 6 24 (86) 16 (89) When data regarding age of achieving dryness were unavailable on 1 observation mean value for age was considered. weekly). A total of 46 parents of 28 girls and 18 boys reported data relating to these questions. For descriptive purposes mean, median, standard deviation and range were used. For all correlation analyses the Spearman rank correlation (r) was used. For comparisons through time the Wilcoxon signed rank test was used. Comparisons between groups were made using the Mann- Whitney U test. All tests were 2-tailed and were conducted at the 5% significance level. RESULTS Daytime continence. Median age foracquisition of dryness during the day was 3.5 years (mean 3.25, range 1.75 to 5.5). Girls were reportedly dry during the day at a median age of 3 years (mean 3.25, range 1.75 to 5.5) and boys at 3.5 years (4, 2.5 to 4.5) but the difference was not significant (p 0.33). More than half (52%) of the children (61% of girls, 39% of boys) were always dry during the day at age 3 years, and this percentage increased to 93% (96% of girls, 89% of boys) at age 4 years and to 100% at age 5 years (table 1). All of the girls but 1 were always dry during the day at age 4 years, and all of the boys but 1 at age 4.5 years (table 1). Nighttime continence. Median age forbeing dry at night was 4 years (mean 4, range 2.25 to 6). No significant difference between girls and boys was found (p 0.98). At age 3 years 20% of the children (17% of girls, 24% of boys) were always dry during the night, while 71% (75% of girls, 65% of boys), 88% and 98% were always dry during the night at ages 4, 5 and 6 years, respectively (table 1). Urinary continence. All of the children became dry during the day before becoming dry at night, except for 1 boy who always was dry at night 1 year before becoming dry during the day. Reported median interval between dryness during the day and dryness at night was 4 months (mean 10, range 0 to 42). Median interval between sometimes and always being dry during the day was 1 year (mean 1, range 0.25 to 3.75), while the corresponding number for being dry at night was 2 years (2, 0.25 to 3.75). All of the children who participated at the last observation at age 6 years were always dry during the day and night, except for1 boy who wet the bed once weekly. Use of potty chair or toilet. Questions about the use of a potty chairortoilet were asked beginning at age 1 year. Table 2 shows the proportion of children always (daily) using a potty chair/toilet from ages 1.5 (2%) to 6 years. Awareness of voiding. Since bladdersensation is difficult to evaluate in children and is only regarded as relevant in toilet trained, cooperative children 4 years or older, 13 the questions asked were related to patient behavior/expressions of the voiding process. Distinctions were made based on whether the children were aware of when they had voided ( I have voided/i am wet ), were aware of when they were actually voiding ( I am voiding ) orhad a sensation of needing to void ( I will void ). The children expressing I have voided reportedly did so up to age 2.5 years, those expressing I am voiding did so up to age 3.5 years and those expressing I will void did so up to age 4 years (table 3). Signs of urgency. The definition of urgency is a strong desire to void accompanied by fear of leakage 13 but the question asked was whether the parent had observed signs of urgency. The occurrence of signs of urgency was reported in 1 child at age 1.75 years. At age 3 years 6 of 14 children reportedly displayed signs of urgency, compared to 17 of 32 at age 3.5 years and 18 of 36 at age 6 years. Voiding postponement. Voiding postponement was considered when the child tried to ease the need to void by using different inhibiting behaviors. The parents did not observe these behaviors before any child was 2 years old. Voiding postponement appeared to become more common after age 3 years and was most frequently reported at age 3.5 years, when 18 of 32 children exhibited signs of voiding postponement. At age 6 years 16 of 36 children displayed signs of voiding postponement. The most common behaviors due to voiding postponement were rushing, walking on the spot, hand pressing the urethra and telling. Leakage. Anotherpart of bladdercontrol is the ability to store urine without leakage. Urinary loss in a child who is not reliably dry is difficult to estimate. According to parent answers, a few children had involuntary small amounts (a few drops) of urinary loss. The occurrence of leakage was first reported at age 1.75 years in 1 child. From ages 2.75 to 6 years 13% to 30% (from 2 to 8 children) reportedly had leakage at each age. No tendency toward a decrease or increase was observed. Urinary flow measurement. Flowmetry was performed beginning at age 3.5 years when possible. If there was more than 1 void during the observation period, the flow with the largest volume was chosen for the report. The most common curves were bell and tower-shaped at all ages. 14 The other flow curves were staccato or plateau in shape, and they only appeared on 1 occasion. No bladder volume was less than 50% of the corresponding median value for age. Bladder capacity. Bladdercapacity was defined as maxi- TABLE 2. Reported daily potty chair/toilet use in children 1 to 6 years old Pt Age (yrs) No. Girls Daily Use (%) No. Boys Daily Use (%) 1 0 (0) 0 (0) (0) 0 (0) 2 3 (11) 0 (0) (39) 1 (6) 3 18 (64) 6 (33) (86) 14 (78) 4 26 (93) 14 (78) (93) 16 (89) 5 27 (96) 16 (89) (96) 17 (94) 6 28 (100) 18 (100)

3 VOIDING PATTERN AND BLADDER CONTROL FROM BIRTH TO AGE 6 YEARS 291 TABLE 3. Reported expressions of bladder sensation in children 1 to 4 years old Pt Age (yrs) I Have Voided I Am Voiding I Will Void mum voided volume plus residual urine volume during 4 hours up to age 3 years. From age 3.5 years urinary flow volume plus residual urine on 1 occasion was reported. Median volume of bladder capacity increased primarily during the first 3 years of life, most obviously between years 2 and 3. From ages 3 to 6 years there was a slight increase in voided volumes from a median of 123 to 140 ml (table 4). There was a significant change in bladdercapacity at 2 intervals, the first from ages 0.5 to 1 year and the second from 2 to 2.5 years (table 4). Median bladdercapacity was 88 ml (mean 108, range 20 to 329) when children were dry during the day and 96 ml (103, 20 to 211) when they were dry at night. A significant negative correlation (r 0.33, p 0.03) was observed between bladder capacity and age at daytime dryness. This finding indicates that the smallerthe bladdercapacity the laterthe children became dry, and the larger the bladder capacity the earlier they became dry. No significant correlation was noted between age at nighttime dryness and bladder capacity (r 0.08, p 0.65). Post-void residual urine. Estimated post-void residual volume decreased slowly with time and demonstrated large variations in every age group. Median volume decreased from 5 ml at age 6 months to 0 ml at age 3 years (table 5). From ages 3 to 6 years median post-void residual volume remained at 0 ml, except at 6 years when it was 2 ml. Volume range increased with age, measuring 0 to 18 ml at 3 years and 0 to 68 ml at 6 years. The boy with 68 ml post-void residual had a voided volume of 258 ml but voided volumes of 240 to 260 ml in other children were observed without post-void residual urine (table 5). Decrease in post-void residual volume with time was not significant at any age interval (table 5). Median post-void residual volume was 0 ml (mean 2, range 0 to 13) when a child was reportedly dry during the day, and 0 ml (3, 0 to 42) when the child was dry at night. No significant correlation was found between post-void residual volume and age at daytime dryness (r 0.09, p 0.57). Correlation between post-void residual volume and age at nighttime dryness was positive but not significant (r 0.19, p 0.23). However, a significant positive correlation was found between post-void residual volume at age 0.5 year and age at daytime dryness, indicating that the residual urine volume at 0.5 year may predict age at daytime dryness (r 0.35, p 0.02). The smallerthe post-void residual volume at age 0.5 yearthe earlier the child became dry during the day, and the larger the post-void residual at 0.5 yearthe laterthe child became dry during the day. Emptying problems. According to parent answers, emptying problems such as difficulty starting were rare and were not reported before ages 4 to 5 years. With the exception of 1 child at age 4 years and 1 at 6 years squeezing was not reported. Soiling and encopresis. The prevalence of soiling and encopresis was low. One child experienced soiling sometimes (once weekly) at age 3.5 years and another had the same problem at age 5.5 years. However, 5 children experienced soiling sometimes at age 6 years or older. Nine children had constipation more than once (5 girls and 4 boys) but they became continent during the day and at night at 3 and 4.5 years, respectively. Bedwetting/enuresis. At age 6 years only 1 child (a boy) wet the bed sometimes (once weekly). The numberof children who woke up during the night to void increased with age from 2 at 3.5 years to 12 at 6 years. DISCUSSION Bladder control, defined as always being dry, was acquired at a later stage compared to most previous studies, in which complete daytime dryness varied from 20% to 99% with a weighted mean of 40% (median 39%) at age 2 years. 3 9, 15 In our study almost no patient was dry at this age. One reason for the difference is that all of the previous studies are old, which suggests differences in the age at which toilet training is started. Since the introduction of disposable diapers and the general acceptance of the view that children decide when they are ready to be dry, the age for starting training has increased. According to these latter studies, acquiring bladder control was only regarded as a matter of maturation. 2, 5, 6 However, the fact that training can accelerate this maturation is clearfrom otherstudies. 7, 8, 11, 16 It is interesting that even if the frequency figures differ at ages 2 and 3 years between previous studies and the present one, the percentage of children with full daytime control at age 4 years is similar between previous studies 9 and ours (94% and 93%, respectively). This finding suggests that most children are mature enough for independent volitional control of voiding at this age. Handling toileting with the support of a parent or other caregiver is probably possible at an earlier stage in most children. Pt Age (yrs) TABLE 4. Bladder capacity/voided volumes plus residual urine in healthy children 0 to 6 years old Mean Ml Bladder Capacity (SD) Median Ml Bladder Capacity (range) Observed Change Mean Change (ml) (32/22) 55 (19) 50 (26 98) 1 52 (32/20) 70 (29.5) 67 (23 166) 49 (30/19) (24/14) 68 (34.5) 59.5 (21 176) 36 (23/13) (23/12) 78 (31.2) 70 (27 165) 26 (17/9) (16/8) 106 (61.3) 91.5 (26 329)* 20 (13/7) (12/7) 128 (72.4) 123 (23 300) 12 (9/3) (19/16) 121 (65.7) 97 (20 270)* 14 (8/6) (18/8) 109 (59.4) 104 (34 289)* 24 (16/8) (12/10) 145 (92.4) (24 348) 16 (10/6) (12/5) 114 (74.4) 86 (39 270) 14 (9/5) (12/9) 124 (50.9) 115 (60 249) 13 (8/5) (18/15) 161 (101.2) 140 (30 452)* 19 (11/8) Values at ages 0.5 to 3 years are based on 4-hour voiding observation, and values after age 3 years are based on 1 void. * Maximum values at ages 2.5, 3.5, 4 and 6 years are for same boy who was dry at age 2.5 years and had a mean post-void residual volume at age 0.5 year of 3 ml (median 1.5, range 0 to 10). p Value

4 292 VOIDING PATTERN AND BLADDER CONTROL FROM BIRTH TO AGE 6 YEARS Pt Age (yrs) TABLE 5. Post-void residual volume in healthy children 0 to 6 years old Mean Ml Residual Vol (SD) Median Ml Residual Vol (range) Observed Change Mean Change (ml) (32/22) 6 (4.6) 5 (0 22) 1 52 (32/20) 7 (6.4) 5.5 (0 33) 49 (30/19) (24/14) 6 (6.8) 3.5 (0 26) 36 (23/13) (23/14) 7 (7.6) 4 (0 24) 27 (17/10) (16/8) 7 (11.3) 2.5 (0 50) 20 (13/7) (12/8) 3 (5.3) 0 (0 18) 13 (9/4) (19/16) 6 (13.2) 0 (0 56) 15 (8/7) (18/8) 2 (4) 0 (0 19) 24 (16/8) (12/10) 4 (5.6) 0 (0 20) 16 (10/6) (12/5) 1 (2.4) 0 (0 10) 14 (9/5) (12/9) 3 (3.6) 0 (0 12) 13 (8/5) (18/15) 6 (12.3) 2 (0 68) 19 (11/8) p Value The same relationship applies to nighttime dryness, which occurs later than daytime control. In previous studies 8% to 41% of children (weighted mean 21%) were dry at night at age 2 years, 15 whereas none were dry in our study. At age 4 years the mean number of children attaining complete nighttime dryness according to Berk and Friman was 85% (range 69% to 99%), 15 while the numberin ourstudy was only 71%. The present study suggests that bladder control in Scandinavian countries today is acquired at a later stage compared to previously. Even in the relatively recent study by Bloom et al the mean age foracquisition of bladdercontrol was 2.4 years (SD 0.6), 17 compared to the median of 3.5 years in our series. The study of Swiss children by Largo, 6 which used a longitudinal approach, revealed a frequency of complete dryness during the day that was more in line with our results, even if 20% of patients were already completely dry at age 2 years. This latter finding probably indicates an earlier start to toilet training, training that is more or less nonexistent in Swedish children today. In the present study all children used a potty chair/toilet from age 4.5 years. At the same age all of the children usually reported a need to void ( I will void ). The children started to report that they had voided or were just in the process of voiding from age 15 months. This finding means that they were already observant about their voiding functions from the age of 18 months, and, therefore, it makes sense to introduce toilet training from this age, which is in line with what Brazelton recommended in Exhibition of urgency, voiding postponement behaviors and leakage of urine, signs frequently seen in bladder function problems, was observed in children in the process of attaining bladder control. Urgency was most common after ages 3.5 to 4 years, when half of the children displayed signs of urgency, compared to the 20% rate reported by Hellstrom et al in an epidemiological study of 7-year-old school entrants. 18 Signs of voiding postponement behaviors were seen often at age 3.5 years and later. Signs of urine leakage were also commonly seen in 23% of patients (range 13% to 30%) afterachievement of bladdercontrol, which is similarto the findings of Bloom et al. 17 Bladder capacity increased with time, primarily during the second and third year, doubling from ages 3 months to 3 years and tripling at age 6 years. The small increase in bladdervolumes afterage 3 years was probably due to the fact that capacity was estimated from uroflowmetry, which was often measured on only 1 occasion, and, therefore, can be suspected of not being representative of individual maximal bladder capacity. Interindividual variation at various ages was also remarkable and increased with age. Therefore, it is difficult to establish so-called normal values forbladdercapacity in children, especially after toilet training. This concept has also been emphasized by Mattsson and Lindstrom in a study of Swedish schoolchildren. 19 Once children have acquired bladder control they are able to empty the bladder when they decide it is convenient, and seldom when maximum or functional capacity is reached. However, it was noted in this series that patients with an initially large bladder capacity tended to have a large capacity at the end of the study, while those with an initially small capacity tended to have a small capacity. Furthermore, high bladder capacity could also be used as a predictor of early daytime dryness. The question appears to be whether an increase in age when attaining bladder control influences the risk of functional voiding problems developing at a later stage. An increase in voiding problems has been reported in recent years. 20 However, it is not yet known whether these findings are substantial or are simply the result of improved awareness in parents. One explanation for a relationship between late bladder control and functional disturbances might be a longer period of incoordination in infancy. Another important consideration is that sitting on a potty chairortoilet with support to the thighs and the feet facilitates relaxation of the pelvic floormuscle, forcomplete bladderemptying. 21 One or 2 such voids per day may decrease the risk of urinary tract infection compared to using a diaper to void in. A large number of factors probably affect the development of functional disorders as well as the time at which a child achieves bladdercontrol. CONCLUSIONS Bladder control is acquired at a later stage today. The occurrence of awareness of a need to void from age 1.5 years suggests an earlier start for toilet training. Infants with a small post-void residual volume at age 0.5 year and those with large bladder capacity will probably become dry during the day earlier than those with large post-void residual volume at age 0.5 yearorsmall bladdercapacity. REFERENCES 1. Hagglof, B., Andren, O., Bergstrom, E., Marklund, L. and Wendelius, M.: Self-esteem before and after treatment in children with nocturnal enuresis and urinary incontinence. Scand J Urol Nephrol Suppl, 18: 79, Muellner, S. R.: Development of urinary control in children: some aspects of the cause and treatment of primary enuresis. JAMA, 172: 1256, Brazelton, T. B.: A child-oriented approach to toilet training. Pediatrics, 29: 121, Oppel, W. C., Harper, P. A. and Rider, R. V.: The age of attaining bladder control. Pediatrics, 42: 614, Klackenberg, G.: Expectations and reality concerning toilet training. Acta Paediatr Scand, suppl., 224: 85, Largo, R. H. and Stutzle, W.: Longitudinal study of bowel and bladder control by day and night in the first six years of life. I: Epidemiology and interrelations between bowel and bladder control. Dev Med Child Neurol, 19: 598, 1977

5 VOIDING PATTERN AND BLADDER CONTROL FROM BIRTH TO AGE 6 YEARS devries, M. W. and devries, M. R.: Cultural relativity of toilet training readiness: a perspective from East Africa. Pediatrics, 60: 170, Takahashi, E.: Investigation of the age of release from the diaper environment. Pediatrician, suppl., 14: 48, Stenhouse, G.: Toilet training in children. N Z Med J, 101: 150, Largo, R. H., Molinari, L., von Siebenthal, K. and Wolfensberger, U.: Does a profound change in toilet-training affect development of bowel and bladder control? Dev Med Child Neurol, 38: 1106, Bakker, E. and Wyndaele, J. J.: Changes in the toilet training of children during the last 60 years: the cause of an increase in lower urinary tract dysfunction? BJU Int, 86: 248, Jansson, U.-B., Hanson, M., Hanson, E., Hellström, A.-L. and Sillén, U.: Voiding pattern in healthy children 0 to 3 years old: a longitudinal study. J Urol, 164: 2050, Norgaard, J. P., van Gool, J. D., Hjalmas, K., Djurhuus, J. C. and Hellstrom, A.-L.: Standardization and definitions in lower urinary tract dysfunction in children. International Children s Continence Society. BrJ Urol, 81: 1, Hansson, S., Hjälmås, K., Jodal, U. and Sixt, R.: Lower urinary tract dysfunction in girls with untreated asymptomatic or covert bacteriuria. J Urol, 143: 333, Berk, L. B. and Friman, P. C.: Epidemiologic aspects of toilet training. Clin Pediatr, 29: 278, Smeets, P. M., Lancioni, G. E., Ball, T. S. and Olivia, D. S.: Shaping self-initiated toileting in infants. J Appl Behav Anal, 18: 303, Bloom, D. A., Seeley, W. W., Ritchey, M. L. and McGuire, E. J.: Toilet habits and continence in children: an opportunity sampling in search of normal parameters. J Urol, 149: 1087, Hellstrom, A. L., Hanson, E., Hansson, S., Hjalmas, K. and Jodal, U.: Micturition habits and incontinence in 7-year-old Swedish entrants. Eur J Pediatr, 149: 434, Mattsson, S. and Lindstrom, S.: Diuresis and voiding patterns in healthy schoolchildren. Br J Urol, 76: 783, Vande Walle, J., Theunis, M., Renson, C., Raes, A. and Hoebeke, P.: Commercial television bladder dysfunction. Acta Urol Belg, 63: 105, Wennergren, H. M., Oberg, B. E. and Sandstedt, P.: Importance of leg support for relaxation of the pelvic floor muscles. A surface electromyograph study in healthy girls. Scand J Urol Nephrol, 25: 205, 1991