Objective patterning of uroflowmetry curves in children with daytime and nighttime wetting.
ABSTRACT Pediatric uroflowmetry curve interpretation is incompletely standardized. Thus, we propose new, objective patterning.
Uroflowmetry curves were obtained in 100 children presenting with daytime incontinence or enuresis. Each curve was compared with a standard curve generated from a published nomogram and a new patterning method was formulated. Staccato and interrupted patterns were defined using International Children's Continence Society criteria. The remaining curves were divided by the deviation of the maximal flow rate from the median nomogram value as certain patterns, including tower-greater than 130%, not abnormal-70% to 130% and plateau-less than 70%. The correlation between the presenting symptom and patterns or other uroflowmetry parameters was evaluated. Six pediatric urologists also patterned the same curves subjectively.
All curves could be classified as 1 of the defined patterns using this method. Pattern distribution reflected the spectrum of presenting symptoms with more tower, interrupted and staccato patterns in children with daytime wetting than in those with monosymptomatic enuresis. Age adjusted voided volume was also smaller in the former group but post-void residual urine, and maximal and average flow rates did not correlate with presenting symptoms. Subjective patterning showed marked interobserver differences. When patterning applied by the current method was used as a reference, observer sensitivity for abnormal patterns inversely correlated with specificity.
Subjective uroflowmetry patterning is liable to personal bias. The proposed method enables objective patterning that complies with International Children's Continence Society standardization and clinical presentation.
- SourceAvailable from: Johan Vande Walle[show abstract] [hide abstract]
ABSTRACT: We updated the terminology in the field of pediatric lower urinary tract function. Discussions were held of the board of the International Children's Continence Society and an extensive reviewing process was done involving all members of the International Children's Continence Society as well as other experts in the field. New definitions and a standardized terminology are provided, taking into account changes in the adult sphere and new research results.The Journal of Urology 08/2006; 176(1):314-24. · 3.70 Impact Factor
- New England Journal of Medicine 06/2008; 358(22):2414-5. · 51.66 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: As the voiding habits of Iranian children differs from other children because of some cultural and religious considerations, we aimed to establish normal reference values of urinary flow rates in Iranian children between 7 to 14 years of age. Eight hundred and two uroflowmetry studies were performed on children with no history of a renal, urological, psychological or neurological disorder, between the ages 7 and 14. Five hundred twenty five studies from 192 girls and 335 boys were considered in this study excluding the staccato/interrupted voiding pattern or voided volume less than 20 ml. The voiding volume, the maximum and average urinary flow rates were extensively analyzed. The maximal and average urine flow rate nomograms were plotted for both girls and boys. Mean maximum urine flow rate was 19.9 (ml/sec) for boys and 23.5 (ml/sec) for girls with a mean voided volume of 142 (ml) for boys and 147 (ml) for girls. Flow rates showed a close association with voiding volume in both sexes. The maximum and average flow rates were higher in girls than in boys, and they showed a significant increase in flow rates with increasing age, where boys did not. The mean maximum urine flow rates (19.9 ml/sec for boys and 23.5 ml/sec for girls) were found to be higher in this study than other studies. Nomograms of maximal and average flow rates of girls and boys are presented in centile form, which can help the physician to evaluate the response to medical or surgical treatment and be useful for the screening of lower urinary tract disturbances in children, for a wide range of voided volumes.BMC Urology 02/2005; 5:3. · 1.69 Impact Factor
Objective patterning of uroflowmetry curves in children with
daytime and nighttime wetting.
Kanematsu, Akihiro; Johnin, Kazuyoshi; Yoshimura, Koji;
Okubo, Kazutoshi; Aoki, Katsuya; Watanabe, Masato;
Yoshino, Kaoru; Tanaka, Shiro; Tanikaze, Saburo; Ogawa,
The Journal of urology (2010), 184(4 Supplement 1): 1674-
© 2010 American Urological Association Education and
Research, Inc. Published by Elsevier Inc.
KURENAI : Kyoto University Research Information Repository
An objective patterning of uroflowmetry curves in
children with daytime and nighttime wetting
Akihiro Kanematsu1), Kazuyoshi Johnin2), Koji Yoshimura1), Kazutoshi Okubo1),
Katsuya Aoki3), MasatoWatanabe4), Kaoru Yoshino4), Shiro Tanaka5), Saburo
Tanikaze4) and Osamu Ogawa1)
1) Department of Urology, Kyoto Universtiy
2) Department of Urology, Shiga University of Medical Science
3) Department of Urology, Nara Medical University
4) Department of Urology, Aichi Children’s Health and Medical Center
5) Division of Clinical Trial Design & Management, Translational Research
Center, Kyoto University Hospital
Running head: Uroflowmetry patterning in children
Key Words: uroflowmetry, enuresis, incontinence, pediatric
*Manuscript (Submit in MS Word; include Title Page and Abstract; Tables and Figures should NOT be included but attached
Purpose: Interpretation of pediatric UFM curves is incompletely standardized.
Therefore, we propose a new objective patterning.
Materials and Methods: UFM curves were obtained from 100 children presenting with
daytime incontinence or enuresis. Each curve was compared with a standard curve
generated from a published nomogram, and a new patterning method was formulated.
First, Staccato and Interrupted patterns were defined using International Children’s
Continence Society (ICCS) criteria. Next, the remaining curves were divided by the
deviation of maximal flow rate (MFR) from the median value of the nomogram as
Tower (>130%), ‘Not abnormal’ (70–130%), and Plateau (<70%) patterns. The
correlation between the presenting symptom and the patterns or other UFM
parameters was evaluated. Finally, six pediatric urologists also patterned the same
Results: All of the curves could be classified to one of the defined patterns using this
method. The distribution of the patterns reflected the spectrum of presenting
symptoms, with more Tower, Interrupted, and Staccato patterns in children with
daytime wetting than in those with monosymptomatic enuresis. Age-adjusted voided
volume was also smaller in the former group, but post-void residual, MFR and
average flow rate were not correlated with presenting symptoms. Subjective
patterning showed marked inter-observer differences. When the patterning applied by
the present method was used as a reference, the sensitivity of the observers for
abnormal patterns was inversely correlated with their specificity.
Conclusion: Subjective UFM patterning is liable to personal bias. The proposed
method enables objective patterning that complies with ICCS standardization and
Uroflowmetry (UFM) is the most common diagnostic urodynamic procedure for
evaluating children with lower urinary tract problems.1,2 In adults and in children, it is
useful for assessing bladder outlet obstruction, and the maximal flow rate (MFR) is a
reliable parameter for that purpose. In addition, UFM is widely used for diagnosing
children with daytime and nighttime wetting, which are highly prevalent conditions
during childhood.3-6 It has been postulated that UFM can detect underlying bladder
problems by defining certain abnormal urination patterns. The International
Children’s Continence Society (ICCS) proposed standardizing the terminology for
such abnormal curves, and classified them into Bell (for normal), Tower, Plateau,
Staccato and Interrupted patterns.1 However, the patterning of UFM curves is
incompletely standardized and is largely at the physician’s subjective discretion. In
short, UFM is a first-line, non-invasive test to assess voiding function, but the
interpretation of the results is not scientifically standardized for evaluating pediatric
bladder problems, and some authors question its utility.7,8
Here, we attempted to develop an objective patterning method, starting by
comparing clinical UFM curves with corresponding standard curves. The method was
validated by reference to clinical symptoms and compared with subjective patterning
performed by six pediatric urologists.
Materials and Methods
Clinical data sets: UFM curves were obtained from 100 consecutive children aged
6–15 years old who presented at Shiga Medical Center for Children for daytime
incontinence and/or enuresis. Patients with spinal lesions, major urinary tract
anomalies detectable by screening ultrasound or urinary tract infection were
There were 41 children with daytime incontinence with or without enuresis (Group
A), 28 with non-monosymptomatic enuresis (non-MNE) without daytime
incontinence (i.e. children with nocturnal enuresis with daytime symptoms like
urgency, frequnecy, or holding but no episode of daytime wetting, Group B), and
31 with monosymptomatic enuresis (MNE, Group C). All tests were performed at full
bladder, on their first office visit.
The age-adjusted capacity of the children was calculated using published formula
for Japanese children, (capacity) = (age +2) × 25.9 Residual urine volume was
calculated from ultrasound images using the formula Q=abc/2.
Formulation of the patterning method (Figure 1). Step 1: Generation of tentative
standard curves: Tentative standard uroflowmetry curves were generated with a
shape similar to our unpublished data in healthy adults, and fitting to average
and maximal flow rate for corresponding voided volume in a published pediatric
nomogram.10, 11 Step 2: All of the clinical curves were scanned and digitized by
GraphCel, a free on-line software (http://t_kobo.at.infoseek.co.jp/grp/), for further
analysis by Microsoft Excel (Microsoft, Redmond, WA). The standard curves were
also transformed to Excel files. Step 3: Each curve was directly compared with the
corresponding standard curve by overlapping, and deviation from the standard curve
was considered abnormal.
As a result, the following patterning method was formulated. First, Staccato and
Interrupted patterns were defined by the ICCS criteria.1 We grouped them together as
one group in further analysis, as differentiation between the two patterns was
considered to be difficult owing to the indefinite duration of interruption. Next, the
other curves were classified by deviation of the MFR from the median value of the
nomogram as Tower, ‘Not abnormal’, and Plateau patterns.
Validation of the patterning method. First, each clinical group was stratified by
standard parameters, i.e.? voided volume, MFR, average flow rate (AFR), and post-
void residual (PVR). Second, each group was stratified by the new patterns, setting
cutoff lines of MFR for defining Tower and Plateau patterns. Finally, six pediatric
urologists subjectively patterned the same curves. The urologists were informed about
age and sex but not about the clinical presentation of the patients. Assuming the
patterning obtained with the present method as a reference, we assessed
agreement of the subjective and objective patterning by Kappa coefficients and
95% confidence intervals for each pair of six raters and the objective patterning.
Statistical analysis was performed with Statview software (Abacus, Burlington, MA)
and SAS software (SAS, Cary, NC). P values <0.05 were considered as significant.
Demographic characteristics. The demographic characteristics of the 100 children
undergoing UFM are shown in Table 1. There was no statistical difference in the sex
ratio, but the age distribution was significantly shifted to a younger age in Group A
Relationship between standard parameters of uroflowmetry and presenting
symptoms (Figure 2). The age-adjusted voided volume was smaller in children with
daytime incontinence than those without. PVR, MFR and AFR, as a single parameter,
were not correlated with the presenting symptoms.
Relationship between the new uroflowmetry patterning and the presenting symptoms
(Figure 3). Using the present method, all of the curves could be classified to one of
the patterns. We have tentatively set multiple cutoff lines by changing it from 50-
70%, and 130-150% respectively. Finally, we set the cutoff values of MFR for
defining Tower and Plateau patterns at 70 and 130%, since it seemed to
represent average result of the raters, and the distribution of the patterns reflected
the spectrum of presenting symptoms, with more Tower, Interrupted, and Staccato
patterns for daytime incontinence and non-MNE, and less in those with MNE (Figure
Comparison of subjective and objective patterning (Figure 4). The six examiners
who performed subjective patterning judged that 4–14% (mean, 10.5%) of the curves
were ‘unclassifiable’. The judgment for the remaining curves showed marked inter-
observer differences. There was a wide range of inter-observer difference, from
0.21-0.64. The range of agreement rate between the raters and the objective
patterning was narrower, from 0.37-0.55 (Table 2). When the patterning applied
using the present method was set as the reference, the sensitivity among the observers
for abnormal patterns was inversely correlated with their specificity (Figure 4). In 30
cases, 3 raters or more made the same diagnosis, discrepant from the objective
patterning. Majority of these, 22/30 (73.3%) were related with discrimination of
Tower and Plateau pattern from normal pattern. (three examples are shown in
This paper presents a new objective patterning method complying with the ICCS
standardization and reflecting clinical symptoms.
In this report, we showed that MFR and AFR, standard parameters of UFM, do not
reflect the clinical symptoms by themselves (Figure 2). The only factor reflecting the
symptom was age-adjusted voided volume, which should be better represented by a
flow–void chart. The poor utility of MFR and AFR is because of the presence of
Staccato and Interrupted patterns in many children. Although these patterns manifest
the worst voiding pattern, fractionated peaks and relatively long voiding time mean
they are classified at an average level of MFR and AFR. Initial definition of these
patterns leaves single-peaked curves, which are more manageable by a conventional
parameter, namely MFR. Indeed, MFR, which is considered to be ‘the most relevant
variable’ according to the ICCS standardization committee,1 can be fully utilized to
differentiate Tower and Plateau patterns from normal uroflow curves.
The first point of the present study is that UFM patterns alone, in the absence of
clinical information, may reflect the clinical symptoms of children with daytime and
nighttime wetting (Figure 3). Daytime incontinence, which is considered to be a
manifestation of overactive bladder or dysfunctional voiding, is more associated with
Tower, Staccato or Interrupted patterns than MNE. Although such features have been
postulated as the rationale to perform UFM as first-line examination for children with
incontinence and enuresis, the reliability of UFM has been questioned because of the
lack of an objective standard.7,8 Our findings may reconcile the present consensus of
the clinical utility of UFM and the concurrent skepticism against it.
The second point is that subjective patterning is liable to personal bias (Table 2).
When the patterning using the present method was used as a reference, the sensitivity
of the observers for abnormal patterns was inversely correlated with their specificity
(Figure 4). This suggests that each physician may have internally consistent cutoffs
for patterning, which may not be consistent with that of other physicians. Without
standardization of the evaluation method, we can neither discuss usefulness or
uselessness of UFM for diagnosing and treating pediatric overactive bladder or
dysfunctional voiding, nor can we compare UFM data with other parameters
such as bladder wall thickness or external sphincter electromyography.
In this study, we deliberately avoided using the term ‘Bell’ to define the normal
pattern because the term ‘Bell’ may frequently result in over-interpretation of subtle
changes in the curves. For example, a curve nearly identical to the standard curve
could be interpreted as a Plateau or Tower (Figure 5B and 5C). Referring to a
standard nomogram is sufficient to avoid such confusion.
There is one remaining question regarding this study, which is the lack of normal
control. The cutoff line for MFR in the present patterning method was defined rather
arbitrarily, to translate the physicians’ judgments to a mathematical standard. The
proposed cutoff may differentiate MNE from daytime incontinence and non-MNE,
but it is unclear whether it can differentiate between dry and wet children. Hence, we
cannot conclude from the present result whether the 38.7% ‘abnormal’ patterns found
in MNE patients are higher or identical with that of completely dry children. This
point is particularly important because it is still unclear whether pure MNE children
have normal bladder function or not, as enuretic children do exhibit bladder
overactivity at night, causing cortical arousal.12,13 Further validation of the present
results on an epidemiological basis would further elucidate the true
pathophysiology of enuretic children, and the usefulness of UFM for their
Here, we formulated an objective patterning of UFM, complying with ICCS
standardization and clinical presentation of enuresis. This patterning may replace the
current subjective UFM patterning, which is liable to personal bias. Further validation
of this new patterning is needed with reference to a population-based control.
1. Neveus, T., von Gontard, A., Hoebeke, P. et al.: The standardization of
terminology of lower urinary tract function in children and adolescents: report
from the Standardization Committee of the International Children's
Continence Society. J Urol, 176: 314, 2006
2. Neveus, T.: The new International Children's Continence Society's
terminology for the pediatric lower urinary tract--why it has been set up and
why we should use it. Pediatr Nephrol, 23: 1931, 2008
3. Sureshkumar, P., Jones, M., Cumming, R. et al.: A population based study of
2,856 school-age children with urinary incontinence. J Urol, 181: 808, 2009
4. Chung, J. M., Lee, S. D., Kang, D. I. et al.: Prevalence and associated factors
of overactive bladder in Korean children 5-13 years old: a nationwide
multicenter study. Urology, 73: 63, 2009
5. Kajiwara, M., Inoue, K., Usui, A. et al.: The micturition habits and prevalence
of daytime urinary incontinence in Japanese primary school children. J Urol,
171: 403, 2004
6. Kajiwara, M., Inoue, K., Kato, M. et al.: Nocturnal enuresis and overactive
bladder in children: an epidemiological study. Int J Urol, 13: 36, 2006
7. Jensen, K. M., Nielsen, K. K., Kristensen, E. S. et al.: Uroflowmetry in
neurologically normal children with voiding disorders. Scand J Urol Nephrol,
19: 81, 1985
8. Dogan, H. S., Akpinar, B., Gurocak, S. et al.: Non-invasive evaluation of
voiding function in asymptomatic primary school children. Pediatr Nephrol,
23: 1115, 2008
9. Hamano, S., Yamanishi, T., Igarashi, T. et al.: Evaluation of functional
bladder capacity in Japanese children. Int J Urol, 6: 226, 1999
10. Kajbafzadeh, A. M., Yazdi, C. A., Rouhi, O. et al.: Uroflowmetry nomogram
in Iranian children aged 7 to 14 years. BMC Urol, 5: 3, 2005
11. Okubo, K., Yoshimura, Kanematsu, A., Nishiyama, H., Ogawa, O.: Average
shapes of uroflow curves of adult men. Abstract presented in Annual Meeting
of International Continence Society, 2009
12. Yeung, C. K., Diao, M., Sreedhar, B.: Cortical arousal in children with severe
enuresis. N Engl J Med, 358: 2414, 2008
13. Hjalmas, K., Arnold, T., Bower, W. et al.: Nocturnal enuresis: an international
evidence based management strategy. J Urol, 171: 2545, 2004
Figure 1. Fitting with standard curves. The examined lines are shown in black and the
corresponding standard curves are shown in grey. Typical ‘Not abnormal’ (A),
‘Interrupted’ or ‘Staccato’ (B), ‘Plateau’ (C) and ‘Tower’ (D) patterns are shown.
Figure 2. Relationship between presenting symptoms and standard UFM parameters
A. Voided volume, expressed as the percentage to age-adjusted standard capacity
B. Post-void residual volume.
C. Maximal flow rate (MFR), as stratified by the reference nomogram.
D. Average flow rate (AFR), as stratified by the reference nomogram.
Stratification of voiding patterns according to the presenting symptoms. More Tower,
Staccato, and Interrupted patterns were found in Group A compared with Group C.
Subjective patterning was compared with the new patterning as a reference. A
negative relationship was found between the sensitivity for abnormal patterns (the
rate at which a rater judge ‘abnormal’ curve by the present patterning, as
abnormal) and the specificity for abnormal patterns (the rate at which a rater
judge ‘normal’ curve by the present patterning, as normal).
In some cases, collective discrepancy was found between the objective and
subjective patternings. Standard curves are superimposed and shown as grey
A. The curve was judged as ‘Bell’ pattern by all examiners, but classified as
‘Plateau’ with the objective patterning.
B. The curve was judged as ‘Tower’ by all examiners, but classified as ‘Not
abnormal’ with the objective patterning.
C. The curve was judged as ‘Plateau’ by 3 examiners, but classified as ‘Not
abnormal’ with the objective patterning.