Desmopressin and oxybutynin in monosymptomatic nocturnal enuresis: a randomized, double-blind, placebo-controlled trial and an assessment of predictive factors
Roberto Del Gado, Department of Pediatrics, Second University of Naples, Via L. De Crecchio 4, Naples, 80131, Italy. e-mail: email@example.com
Study Type – Therapy (case series)
Level of Evidence 4
What's known on the subject? and What does the study add?
The desmopressin analogue to antidiuretic vasopressin is an evidence-based therapy but conflicting results are provided regarding the initial dose of oral desmopressin. Previous studies report the use of a combined therapy with desmopressin and oxybutynin to treat desmopressin-resistant monosymptomatic nocturnal enuresis. These studies show promising results, but they suffer from lack of randomization and lack of a placebo-controlled patient group and are of small sample size. In addition to this, no predictive factors of response to the combined therapy have been considered.
This study showed no significant difference between either a 120 µg or a 240 µg desmopressin initial dose with regard to degree of response. The study is the first randomized, double-blinded, placebo-controlled trial showing the efficacy of combination therapy with desmopressin plus oxybutynin for monosymptomatic nocturnal enuresis. Furthermore, bladder volume and wall thickness index, nocturnal polyuria and voiding latency were assessed as predictive factors of response to the therapy.
- • To assess the efficacy of desmopressin plus oxybutynin and compare two starting dosages of desmopressin (120 and 240 µg) in a randomized, double-blinded, placebo-controlled trial for children with monosymptomatic nocturnal enuresis (MNE) resistant to desmopressin.
- • The predictive factors of children with MNE responsive to desmopressin and combination therapy were also evaluated.
PATIENTS AND METHODS
- • Our sample included 206 patients aged between 6 and 13 (mean age 10.6 ± 2.9 years), 117 males. All patients were required to have MNE.
- • The patients were randomly divided into two groups: the first group was given oral melt 120 µg and the second group 240 µg, for 2 weeks.
- • All patients who had experienced failure of treatment with sublingually administered desmopressin alone were given either desmopressin plus 5 mg oxybutynin or desmopressin plus placebo in a randomized, double-blinded trial for 4 weeks.
- • As predictive factors, bladder volume and wall thickness index, nocturnal polyuria and voiding latency were considered.
- • There was no significant difference between the 120 µg and 240 µg patients in terms of response.
- • The oxybutynin group showed a higher rate of full and partial responses (45% success) compared with the placebo group (17% success), P < 0.01.
- • The responders to combined oxybutynin and desmopressin had significantly lower bladder volume and wall thickness index than the other patients.
- • Our findings highlight that anticholinergic agents may play an important role for a subset of children with enuresis who have a restricted bladder capacity and thickened bladder wall.
- • Ultrasonography-measured bladder variables can provide useful predictive clues for MNE.
- • Predictive factors can help to differentiate treatment subtypes and guide clinical management in primary nocturnal enuresis.
monosymptomatic nocturnal enuresis
bladder volume and wall thickness index.
Primary nocturnal enuresis is the leakage of urine while sleeping in children aged 5 or older. The International Children's Continence Society suggested using the term monosymptomatic nocturnal enuresis (MNE) to signify that children have problems only when asleep . Three major pathogenetic mechanisms have been established as crucial and they are nocturnal polyuria , detrusor overactivity  and an increased arousal threshold .
The desmopressin analogue to antidiuretic vasopressin is an evidence-based therapy (grade Ia evidence) . The ordinary dose of oral desmopressin lyophilisate is 120–240 µg. This dose is not influenced by body-weight or age, and the prescribing physician may choose to start with the higher dose and gradually taper down in the case of good effect or use the opposite strategy . The anti-enuretic effect is seen immediately. In the literature, conflicting results are provided regarding the initial dose of oral desmopressin lyophilisate. Vande Walle et al.  suggest starting with 120 µg, whereas Nevéus  suggests 240 µg. No internationally accepted protocol exists to assess the starter dosage of desmopressin lyophilisate in the case of MNE.
Children who experience failure of or partial response to treatment with desmopressin can be treated with a combined therapy with desmopressin and an anticholinergic therapy . The rationale is based on the concept of potentially increasing the functional bladder capacity with anticholinergic therapy, thereby increasing the response to desmopressin.
The purpose of this study was to assess the efficacy of desmopressin plus oxybutynin in a randomized, double-blinded, placebo-controlled trial for children with monosymptomatic enuresis resistant to desmopressin. Furthermore, the paper is aimed at evaluating and comparing two starting dosages of desmopressin (120 and 240 µg). The predictive factors of children with MNE responsive to a desmopressin and combined therapy were also evaluated.
PATIENTS AND METHODS
A sample of 228 children (10.8 ± 2.76 years) with MNE was recruited from the Department of Pediatric Urology of the Second University of Naples. The parents of 10 children refused to participate in the study and 12 withdrew. Therefore, our sample included 206 patients aged between 6 and 13 (mean age 10.6 ± 2.9 years), 117 males and 89 females. The study protocol was approved by the local ethics committee and informed consent was obtained from all participants.
All patients were required to have MNE. Patients were excluded if there were any episodes of daytime incontinence, increased or decreased voiding frequency (voiding more than eight times per day or less than three times per day, respectively), encopresis or constipation. Constipation was defined by using combined definitions from the Rome II criteria and the Paris Consensus on Childhood Constipation Terminology Group .
All children underwent a general physical examination and a careful history was taken by a paediatric urologist, focusing on bladder and bowel habits. The lower urinary tract function was assessed non-invasively with uroflowmetry, residual urine measurements and a three-day frequency–volume chart. The bladder volume was measured by ultrasonography when the children claimed they were full and wished to void (first urge). The calculated bladder volume was compared with the expected bladder capacity of the patient and expressed as a percentage of it. Expected bladder capacity is estimated by the formula 30 + (age in years × 30) in millilitres [1,11,12]. Then, each patient was invited to void and ultrasonography was repeated so as to measure the postvoid residual volume and bladder wall (BW) thickness. This measurement was considered reliable if the bladder emptying resulted in more than 90% of the maximally full bladder. The BW thickness was measured transversely in the sagittal plane of the abdomen. Measurements were made perpendicular to the luminal surface, and the mean BW thickness was calculated as an average of three measurements, i.e. anterior, posterior and lateral, divided by 3. Care was taken not to include the vagina, rectum or peritoneal reflection of the bladder dome. The bladder volume and wall thickness index (BVWI) was calculated by using the equation BVWI (%) = (maximum measured bladder volume divided by measured mean BW thickness). This volume was then expressed as a percentage of the expected volume based on standard values. BVWI was classified as thick (less than 70), normal (70–130) or thin (more than 130) [13,14]. Scans were conducted by using the ESAOTE Technos (Genoa, Italy) ultrasound unit with a 5 MHz frequency probe.
Wet and dry nights were documented during 14 days before the treatment. In this period, nocturnal urine production was assessed by weighing diapers or sheet covers and nocturnal polyuria was considered as nocturnal urine production greater than 130% of the expected bladder capacity by age . Urine monitors, detecting the time of voiding, were placed beneath the bedclothes. Voiding latency was defined as the time elapsed between bedtime and enuresis.
The patients were randomly divided into two groups: the first group was given 120 µg and the second group 240 µg sublingual desmopressin, for 2 weeks. Children took sublingual desmopressin 30–60 min before going to sleep and they were asked not to drink more than 0.2 L fluid in the last 2 h before bedtime and during the night. Liberal water and solute intake during the day was recommended, especially during the morning and early afternoon hours.
During the treatment, enuresis episodes were evaluated by nocturnal enuresis records. Failure included both a partial response and a non-response to desmopressin therapy. Non-response was defined as a 0–49% decrease and partial response as a 50–89% decrease in the number of wet nights per week .
A dose of 240 µg sublingual desmopressin was then administered, at bedtime, for 2 weeks, to all children who failed to respond to the 120 µg desmopressin therapy during the first part of the study.
All patients who had experienced failure of treatment with desmopressin alone, at the dosage of 240 µg per night for 2 weeks, were given either sublingual desmopressin plus oxybutynin 5 mg not modified-release tablets or sublingual desmopressin plus placebo in a randomized, double-blinded trial for 4 weeks.
A stratified randomization procedure was used to allocate patients into the two treatment groups. The strata formed were based on the initial degree of failure of desmopressin single therapy, and patients were assigned as non-responders or partial responders. With randomization within each stratum, we ensured that the distributions of non-responders and partial responders were balanced within the two treatment arms.
The Pearson chi-squared test was used to evaluate the difference in prevalences. A Kruskal–Wallis test allowed us to assess differences between means. Significance was set at 0.01. A logistic regression model was used to investigate the relationship between response at the end of treatment and possible predictive variables. The logistic regression model included the response at the end of treatment as the dependent variable. The following factors were considered: nocturnal diuresis, voiding latency, BVWI and bladder capacity. All analyses were conducted by using Statgraphics Centurion XVI.I (Warrenton, VA, USA).
In total, 206 children were randomly divided into two groups of 101 and 105 children. At the baseline, patients had a median of 6.6 wet nights per week.
As far as the degree of response is concerned, 86 patients (42%) fully and 85 patients (41%) partially responded to desmopressin therapy, regardless of the dose. There was no significant difference between the 120 µg and 240 µg desmopressin patients in terms of response (odds ratio [OR] 1.44; 95% CI 0.83–2.48; P= 0.19) (Fig. 1). Among the 66 (65%) patients who failed to respond to 120 µg desmopressin therapy, 10 partial and three non-responders (overall 20%) became full responders during the treatment with 240 µg desmopressin (Fig. 1). No statistical difference was seen between these children and the 120 µg desmopressin responders with regard to nocturnal diuresis (P= 0.45). The 240 µg desmopressin full responders showed a higher voiding latency compared with the others (P < 0.01). The 120 and 240 µg desmopressin full responders showed a significantly higher nocturnal urine production than the desmopressin partial and non-responders (P < 0.01) (Table 1). In the 120 µg desmopressin groups, night-time polyuria was found in 88% of full responders, in 57% of partial responders and in 22% of non-responders (OR 5.96, 95% CI 1.81–19.66, and OR 6.03, 95% CI 1.85–20.76 respectively; P < 0.01). In the 240 µg desmopressin groups, night-time polyuria was found in 91% of full responders, in 59% of partial responders and in 26% of non-responders (OR 27.13, 95% CI 5.92–44.38, and OR 22.75, 95% CI 5.28–49.76 respectively; P < 0.01).
Table 1. Characteristics of the patients
| N (%)||35 (35)||46 (46)||20 (19)||51 (43)||49 (41)||18 (15)||13 (21)||15 (24)||33 (33)|
|Age, mean (sd), years||10 (1.35)||9 (1.45)||8 (1.85)||10 (1.24)||10 (1.98)||9 (1.95)||8 (1.16)||9 (1.39)||12 (1.76)|
|Gender, n|| || || || || || || || || |
|Mean bladder capacity by age, % (sd)||94.2 (15.8)||82 (13.6)||55 (14.7)||93.0 (16.5)||83 (15)||53 (19.4)||54.2 (22.6)||76 (26)||86 (21.8)|
| P > 0.01|| P > 0.01|| P < 0.01|| P > 0.01|| P > 0.01|| P < 0.01|| P < 0.01|| P > 0.01|| P > 0.01|
|Nocturnal urine production, mean (sd), mL||397 (127)||254 (136)||172 (121)||397 (134)||273 (128)||165 (119)||202 (114)||179 (124)||168 (143)|
| P < 0.01|| P > 0.01|| P > 0.01|| P < 0.01|| P > 0.01|| P > 0.01|| P < 0.01|| P > 0.01|| P > 0.01|
|Mean BVWI (sd)||108 (13)||81 (17)||47 (11)||106 (15)||79 (18)||49 (12)||44 (16)||57 (14)||71 (18)|
| P > 0.01|| P > 0.01|| P < 0.01|| P > 0.01|| P > 0.01|| P < 0.01|| P < 0.01|| P > 0.01|| P > 0.01|
|Voiding latency, h||6.9 (1.3)||6.8 (1.8)||6.8 (1.3)||7.5 (1.6)||6.7 (1.9)||6.8 (0.8)||6.9 (0.9)||6.8 (1.7)||6.7 (1.1)|
| P > 0.01|| P > 0.01|| P > 0.01|| P < 0.01|| P > 0.01|| P > 0.01|| P > 0.01|| P > 0.01|| P > 0.01|
The oxybutynin group showed a higher rate of full and partial responses (45% success) compared with the placebo group (17% success) (OR 0.24; 95% CI 0.10–0.56; P < 0.01) (Fig. 1). Out of the 61 patients treated with oxybutynin plus desmopressin, the full responders showed a significantly lower BVWI and mean bladder capacity by age and a significantly higher nocturnal urine production than the partial and non-responders (P < 0.01). Desmopressin full responders produced, during the night, more urine than the full responders to the combined therapy (P < 0.01) (Table 1). Among the patients who failed to respond to desmopressin therapy, the partial responders showed a higher rate of both full and partial responses to combined therapy, compared with the non-responders (51% and 33% success rate respectively) (OR 0.31; 95% CI 0.20–0.59; P < 0.01).
Using the full sample, nocturnal diuresis (OR 1.86; 95% CI 1.30–2.50; P < 0.01; R 0.88) significantly predicted an increased response to desmopressin while BVWI (OR 1.65; 95% CI 1.53–2.75; P < 0.01; R 0.81) and mean bladder capacity by age (OR 1.74; 95% CI 1.45–2.64; P < 0.01; R 0.76) predicted an increased response to oxybutynin plus desmopressin.
The desmopressin analogue to antidiuretic vasopressin is an evidence-based therapy (grade Ia evidence). A Cochrane review of 75 trials demonstrated success in approximately two-thirds of children with MNE . Other reports document even lower success rates [15,16], probably indicating that patient and selection centre play a major role in response rates. These data are in agreement with our findings which showed a full response in 42% and a partial response in 41% of patients regardless of the desmopressin dose.
The study of Lottmann et al.  showed that lower doses of desmopressin were associated with a greater reduction in wet nights. However, as a feature of the study design, patients who responded well to low doses were excluded from the group of patients for whom medication was augmented. In this way, the group receiving higher doses of desmopressin was consequently selected for children with more treatment-resistant primary nocturnal enuresis.
In our study, instead, patients were randomly divided into two groups, each of which was given either 120 or 240 µg sublingual desmopressin for 2 weeks. Our data show no significant difference between the 120 µg and 240 µg desmopressin group with regard to degree of response, and these results are in agreement with those of Nevéus et al.  who found no significant difference between the two doses of desmopressin.
The above mentioned study described a group of bedwetting children not responding to the 0.4 mg desmopressin but reliably dry when the dose was doubled. It was hypothesized that either these high dose responders voided with a full bladder or, at least, urine production was a major determinant of the time of bladder emptying. However, the sample under investigation was too small for conclusions to be drawn .
Our study included a sample of patients who failed to respond to 120 µg desmopressin therapy but responded to 240 µg desmopressin. Between the 120 and the 240 µg desmopressin responders a difference was observed with regard to voiding latency but not with regard to nocturnal diuresis (Table 1). These data highlight that patients with a late episode of enuresis need a higher dose of desmopressin.
This is supported by a pharmacodynamic study  which shows a clear dose–response relationship for the duration of urinary-concentrating action. For bioactivity throughout the night 120 µg is preferable, but for some children the urinary-concentrating effect did not last for 11 h, suggesting suboptimal dosing (which may be due either to poor absorption or to a high excretion rate).
Norgaard et al.  showed that about half of enuretic episodes in children occur during the early hours of the night, suggesting that a pharmacological action lasting for as little as 3 h may be sufficient. If a duration of action were needed for longer (throughout the night), then doses of ≥120 µg desmopressin would be required.
We suggest starting with the dose of 120 µg desmopressin and then, if the patient is therapy resistant or only a partial responder, the dose can be doubled, but only if the reappearance of diluting capacity and increased diuresis in the morning is documented.
The use of the combined therapy with desmopressin and oxybutynin in treating desmopressin-resistant MNE has been reported and has shown promising results in previous studies, which suffer, however, from lack of randomization and lack of a placebo-controlled patient group and are of small sample size [7,19–22]. The effectiveness of this approach has been corroborated by a randomized placebo-controlled study which showed the efficacy of a combined therapy with desmopressin plus a long-acting anticholinergic agent in treating MNE . Unfortunately, it did not identify possible predictive factors that would predict a response to combination therapy for monosymptomatic desmopressin non-responders. Our study is the first randomized, double-blinded, placebo-controlled trial showing the efficacy of combination therapy with desmopressin plus oxybutynin for MNE and has a significantly greater sample size than previous trials [9,23]. Furthermore, the response rate for combination therapy was similar to the study by Austin et al.  who used desmopressin and long-acting tolterodine, indicating the same efficacy of these different anticholinergic agents.
One of the pharmacological challenges in the management of PNE is predicting who will respond to desmopressin and who to combination therapy. Functional bladder capacity has been demonstrated to be a reliable predictor of response to desmopressin; children with larger capacities are more likely to become successful responders [22,24]. In a small group of selected patients who received combination therapy and achieved dryness, Nevéus  observed that the children tended to have small bladders, as well as high levels of urine production and low renal-concentrating capacity.
Previous studies [13,14] have shown the high predictive value of BVWI for treatment response to desmopressin. They considered normal BVWI (70–130) as highly predictive of normal bladder function and good treatment response to desmopressin. Our study confirms these data and, furthermore, it focuses on the importance of BVWI as a predictive factor for treatment response to combined oxybutynin and desmopressin. A BVWI of less than 70 is indicative of a small bladder capacity with a thick BW so that the combined therapy can help to dampen down uninhibited detrusor contractions.
Moreover, our findings highlight that in full responders to combined therapy, both detrusor overactivity and nocturnal polyuria play an important pathogenetic role, whereas in the desmopressin full responders nocturnal urine production is greater than in all the other patients and therefore is considered as the main pathogenetic factor. In fact, among the 120 µg and 240 µg desmopressin full responders, there was the largest percentage of nocturnal polyuria, and compared with partial and non-responders the difference was statistically significant.
Our study identifies a group of 33 patients who are non-responders both to desmopressin and to combined therapy. These patients present neither detrusor overactivity nor nocturnal polyuria as a predominant pathogenetic factor. This makes it difficult to shape the treatment specifically and target the underlying dysfunction of these patients.
Our findings highlight that anticholinergic agents may play an important role for a subset of children with enuresis who have a restricted bladder capacity and thickened BW. These data show that ultrasonography-measured bladder variables can provide useful predictive clues for MNE and help to differentiate treatment subtypes and guide clinical management in primary nocturnal enuresis. No difference between the 120 µg and 240 µg desmopressin dose was shown as far as the degree of response is concerned.
CONFLICT OF INTEREST