Home > Clinical Guides > Nocturnal Enuresis: The Management of... > Tricyclic medication and the management...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

National Clinical Guideline Centre (UK). Nocturnal Enuresis: The Management of Bedwetting in Children and Young People. London: Royal College of Physicians (UK); 2010. (NICE Clinical Guidelines, No. 111.)

15Tricyclic medication and the management of bedwetting

15.1. Introduction

What are they? The tricyclic drugs have been used for treating bedwetting for many years. The need for close follow up and the potential for serious cardiac consequences in overdose mean they are considered for use in specialist centres only.

How do they work? Tricyclics have significant anticholinergic effects and thus have similar properties to Oxybutynin (see anticholinergics). They also have additional central effects which are not well understood but can be beneficial in preventing bedwetting in a group of children who have not responded to first line treatments.

How is it given? Imipramine is only available as tablets. To minimise side effects it is best started as a low dose and increased fortnightly to the maximum dose allowed for the age of the child. The single daily dose should be given around 3 hours before sleep. A course of treatment should last for 3 months maximum before reducing the dose slowly and stopping it for a week or so to assess progress.

Side effects and contraindications. Most children tolerate this medication without experiencing side effects. The main side effects are dry mouth, gastrointestinal symptoms and occasional behavioural changes. These resolve when the medication is stopped. The tricyclics have the potential to interact with other long term medications e.g. for epilepsy and this should be checked before starting treatment. Overdosage can cause serious cardiac arrhythmias (abnormalities of heart rhythm) and death. Tricyclics are contraindicated in children with a family history of early cardiac death or who have any evidence of cardiac disease.

15.2. What is the clinical and cost effectiveness of tricyclic medication for children and young people under 19 years who have bedwetting?

15.2.1. Evidence review

The studies included in the review had varying dosages of imipramine given, based on age or weight of the patient, with younger children being given 25 mg imipramine and older children being given 50 mg imipramine.

15.2.1.1. Imipramine compared to placebo

Fourteen randomised controlled trials compared imipramine to placebo; Agarwala (1968) 130, Attenburrow (1984) 131, Batislam (1995) 132, Drew (1966) 133, Fournier (1987) 81, Harrison (1970) 134, Hodes (1973) 135, Khorana (1972) 136, Kolvin (1972) 103, Manhas (1967) 137, Martin (1971) 138, Poussaint (1965) 139, Smellie (1996) 140, and (1964) 141.

Table 15-1Imipramine compared to placebo - Clinical summary of findings

OutcomeImipraminePlaceboRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights64/171 (37.4%)12/168 (7.1%)RR 4.81 (1.67 to 13.89)271 more per 1000 (from 48 more to 915 more)LOW
Number of children who had >80% improvement at the end of treatment16/35 (45.7%)5/27 (18.5%)RR 2.47 (1.03 to 5.89)272 more per 1000 (from 6 more to 905 more)VERY LOW
Number of children who showed >50% improvement in the number of dry nights27/45 (60%)10/39 (25.6%)RR 1.27 (0.06 to 27.63)69 more per 1000 (from 241 fewer to 1000 more)VERY LOW
Mean number of wet nights per week at the end of treatment912-MD −2.5 (−5.74 to 0.74)LOW
Mean number of wet nights per week at the end of treatment (no SD)129100-not pooledVERY LOW
Mean number of wet nights per 2 weeks during treatment2929-MD −2.3 (−4.19 to − 0.41)LOW
Mean number of wet nights during 26 nights of treatment5757-MD −6.3 (−8.6 to −4)LOW
Mean number of wet nights per week at follow up912-MD −1.5 (−4.85 to 1.85)LOW
Mean number of wet nights per week at follow up (no SD)3527-not pooledVERY LOW
Number of children who dropped out2/32 (6.3%)0/32 (0%)RR 5 (0.25 to 100.21)0 more per 1000 (from 0 fewer to 0 more)VERY LOW

15.2.1.2. Low dose imipramine compared to placebo

One randomised controlled trial Martin (1971) 138 compared 10 mg imipramine to placebo. The usual stated dosage (in the BNF) for imipramine in the treatment of nocturnal enuresis 25 mg imipramine for younger children and 50 mg imipramine for older children. It was therefore considered that a dosage of 10 mg imipramine compared to placebo should be evaluated separately from the usual higher dosage of imipramine compared to placebo.

Table 15-2Low dose imipramine compared to placebo - Clinical summary of findings

OutcomeLow dose ImipraminePlaceboRelative risk (95% CI)Absolute effectQuality
Mean number of wet nights during 26 nights of treatment5757-MD −3.1 (−5.1 to −1.1)VERY LOW

15.2.1.3. Low dose imipramine compared to high dose imipramine

One randomised controlled trial, Martin (1971) 138, compared 10 mg imipramine to 25 mg imipramine. The usual stated dosage for imipramine (in the BNF) in the treatment of nocturnal enuresis 25 mg imipramine for younger children and 50 mg imipramine for older children. It was therefore considered that the comparison of 10 mg imipramine to 25 mg imipramine should be evaluated separately.

Table 15-3Low dose imipramine compared to high dose imipramine - Clinical summary of findings

OutcomeLow dose ImipramineHigh dose ImipramineRelative risk (95% CI)Absolute effectQuality
Mean number of wet nights during 26 nights of treatment5757-MD 3.2 (1.3 to 5.1)VERY LOW

15.2.1.4. Imipramine compared to desmopressin

Two randomised controlled trials Vertucci (1997) 123 and Lee (2005) 25 compared imipramine to desmopressin.

Table 15-4Imipramine compared to desmopressin - Clinical summary of findings

OutcomeImipramineDesmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who dropped out7/48 (14.6%)3/49 (6.1%)RR 2.38 (0.65 to 8.68)84 more per 1000 (from 21 fewer to 468 more)VERY LOW

Table 15-5Imipramine compared to desmopressin - Clinical summary of findings

OutcomeImipramineDesmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights19/28 (67.9%)25/29 (86.2%)RR 0.79 (0.59 to 1.06)181 fewer per 1000 (from 353 fewer to 52 more)VERY LOW
Number of children who had 0–1 wet nights per month3/25 (12%)9/26 (34.6%)RR 0.35 (0.11 to 1.13)225 fewer per 1000 (from 308 fewer to 45 more)VERY LOW
Mean number of wet nights per week at the end of treatment2526-MD 1.4 (0.55 to 2.25)VERY LOW
Mean number of wet nights per week at the end of treatment (no SD)2829-not pooledVERY LOW
Mean number of wet nights per week after treatment with imipramine and desmopressin (separate treatments) (no SD)2829-not pooledVERY LOW

15.2.1.5. Imipramine compared to enuresis alarm

Two randomised controlled triails, Fournier (1987) 81 and Kolvin (1972) 103 compared imipramine to enuresis alarm treatment.

Table 15-6Imipramine compared to alarm - Clinical summary of findings

OutcomeImipramineAlarmRelative risk (95% CI)Absolute effectQuality
Number of children who had >80% improvement in the number of dry nights at the end of treatment16/35 (45.7%)17/32 (53.1%)RR 0.86 (0.53 to 1.4)74 fewer per 1000 (from 250 fewer to 212 more)VERY LOW
Number of children who dropped out1/8 (12.5%)1/8 (12.5%)RR 1 (0.07 to 13.37)0 fewer per 1000 (from 116 fewer to 1000 more)VERY LOW
Mean number of wet nights per week at the end of treatment (no SD)4340-not pooledVERY LOW
Mean number of wet nights per week at the end of follow up (no SD)3532-not pooledVERY LOW

15.2.1.6. Imipramine compared to imipramine combined with enuresis alarm

One randomised controlled trial Fournier (1987) 81 compared imipramine to imipramine with an enuresis alarm.

Table 15-7Imipramine compared to imipramine and alarm - Clinical summary of findings

OutcomeImipramineImipramine and alarmRelative risk (95% CI)Absolute effectQuality
Number of drop outs at end of trial0/8 (0%)0/8 (0%)not poolednot pooledLOW
Mean number of wet nights per week at follow-up (no SDs)88-not pooledVERY LOW

15.2.1.7. Imipramine compared to desmopressin combined with oxybutynin

One randomised controlled trial Lee (2005) 25, compared imipramine to desmopressin combined with oxybutynin.

Table 15-8Imipramine compared to desmopressin and oxybutynin - Clinical summary of findings

OutcomeImipramineDesmopressin and oxybutyninRelative risk (95% CI)Absolute effectQuality
Number of children who dropped out7/48 (14.6%)3/48 (6.3%)RR 2.33 (0.64 to 8.49)84 more per 1000 (from 23 fewer to 472 more)VERY LOW

Table 15-9Imipramine compared to desmopressin and oxybutynin - Clinical summary of findings

OutcomeImipramineDesmopressin and oxybutyninRelative risk (95% CI)Absolute effectQuality
Number of children who had 0–1 wet nights per month3/25 (12%)9/26 (34.6%)RR 0.35 (0.11 to 1.13)225 fewer per 1000 (from 308 fewer to 45 more)VERY LOW
Mean number of wet nights per week at the end of treatment2526-MD 1.43 (0.45 to 2.41)VERY LOW

15.2.1.8. Amitriptyline compared to placebo

One randomised controlled trial, Poussaint (1966) 142 compared amitriptyline to placebo.

Table 15-10Amitriptyline compared to placebo - Clinical summary of findings

OutcomeAmitriptylinePlaceboRelative risk (95% CI)Absolute effectQuality
Mean number of wet nights per week at the end of treatment (no SD)2525-not pooledLOW

15.2.1.9. Amitriptyline compared to desmopressin

One randomised controlled trial Burke (1995) 122 compared amitriptyline to desmopressin.

Table 15-11Amitriptyline compared to desmopressin - Clinical summary of findings

OutcomeAmitriptylineDesmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights3/14 (21.4%)1/17 (5.9%)RR 3.64 (0.42 to 31.27)156 more per 1000 (from 34 fewer to 1000 more)MODERATE
Number of children who dropped out of the trial0/14 (0%)3/17 (17.6%)RR 0.17 (0.01 to 3.06)146 fewer per 1000 (from 174 fewer to 363 more)MODERATE
Mean number of wet nights per week at the end of treatment1417-MD −1.4 (−2.95 to 0.15)MODERATE
Mean number of wet nights per week at the end of follow up1417-MD 0.1 (−1.67 to 1.87)MODERATE

15.2.1.10. Amitriptyline compared to enuresis alarm

One randomised controlled trial, Danquah (1975) 104 compared amitriptyline to an enuresis alarm.

Table 15-12Amitriptyline compared to alarm - Clinical summary of findings

OutcomeAmitriptylineAlarmRelative risk (95% CI)Absolute effectQuality
Mean number of wet nights per week at the end of treatment (no SD)1010-not pooledVERY LOW
Median number of days to arrest1010-not pooledVERY LOW

15.2.1.11. Amitriptyline compared to amitriptyline combined with desmopressin

One randomised controlled trial, Burke (1995) 122 compared amitriptyline to amitriptyline combined with desmopressin.

Table 15-13Amitriptyline compared to amitriptyline and desmopressin - Clinical summary of findings

OutcomeAmitriptylineAmitriptyline and desmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights3/14 (21.4%)5/14 (35.7%)RR 0.6 (0.18 to 2.04)143 fewer per 1000 (from 293 fewer to 371 more)MODERATE
Number of children who dropped out of the trial0/14 (0%)3/14 (21.4%)RR 0.14 (0.01 to 2.53)184 fewer per 1000 (from 212 fewer to 327 more)MODERATE
Mean number of wet nights per week at the end of treatment1414-MD 0 (−1.64 to 1.64)MODERATE
Mean number of wet nights per week at the end of follow up1414-MD −1.2 (−3.46 to 1.06)MODERATE

15.2.1.12. Nortriptyline compared to placebo

One randomised controlled trial Lake (1968) 143, compared nortriptyline to placebo.

Table 15-14Nortriptyline compared to placebo - Clinical summary of findings

OutcomeNortriptylinePlaceboRelative risk (95% CI)Absolute effectQuality
Mean number of wet nights per week at the end of treatment (no SDs)5454-not pooledVERY LOW

15.2.1.13. Imipramine compared to placebo for children with bedwetting

One randomised controlled trial compared imipramine to placebo for children with bedwetting, Tahmaz (2000) 129

Table 15-15Imipramine compared to placebo for children with bedwetting - Clinical summary of findings

OutcomeImipraminePlaceboRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights4/12 (33.3%)1/12 (8.3%)RR 4 (0.52 to 30.76)249 more per 1000 (from 40 fewer to 1000 more)VERY LOW
Number of children who had >90% improvement in the number of dry nights7/14 (50%)5/23 (21.7%)RR 2.3 (0.9 to 5.86)282 more per 1000 (from 22 fewer to 1000 more)VERY LOW
Number of children who had 50 to 90% improvement in the number of dry nights5/14 (35.7%)8/23 (34.8%)RR 1.03 (0.42 to 2.52)10 more per 1000 (from 202 fewer to 529 more)VERY LOW
Number of children who relapsed at 6 months9/11 (81.8%)3/6 (50%)RR 1.79 (0.55 to 5.76)395 more per 1000 (from 225 fewer to 1000 more)VERY LOW

15.2.1.14. Imipramine compared to desmopressin for children with bedwetting

One randomised controlled trial Lee (2005) 25 compared imipramine to desmopressin for children with bedwetting.

Table 15-16Imipramine compared to desmopressin - Clinical summary of findings

OutcomeImipramineDesmopressinRelative risk (95% CI)Absolute effectQuality
Number of children who had 0–1 wet nights per month3/23 (13%)14/23 (60.9%)RR 0.21 (0.07 to 0.65)481 fewer per 1000 (from 213 fewer to 566 fewer)LOW
Mean number of wet nights per week at the end of treatment2323-MD 1.3 (0.38 to 2.22)VERY LOW

15.2.1.15. Imipramine compared to oxybutinin for children with bedwetting

Two randomised controlled trials, Esmaelli (2008) 128 and Tahmaz (2000) 129 compared imipramine to oxybutynin for children with bedwetting. See GRADE table in chapter 14.

15.2.1.16. Imipramine compared to enuresis alarm for children with bedwetting

One randomised controlled trials, Wagner (1982) 110 compared imipramine to enuresis alarm treatment for children with bedwetting.

Table 15-17Imipramine compared to alarm - Clinical summary of findings

OutcomeImipramineAlarmRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights4/12 (33.3%)10/12 (83.3%)RR 0.4 (0.17 to 0.93)500 fewer per 1000 (from 58 fewer to 691 fewer)VERY LOW
Number of children who relapsed at 6 months4/4 (100%)5/10 (50%)RR 1.8 (0.93 to 3.48)400 more per 1000 (from 35 fewer to 1000 more)VERY LOW
Mean number of wet nights per week at end of treatment (no SDs)1212-not pooledVERY LOW

15.2.1.17. Imipramine compared to imipramine combined with oxybutinin for children with bedwetting

Two randomised controlled trials, Esmaelli (2008) 128 and Tahmaz (2000) 129 compared imipramine to imipramine combined with oxybutynin for children with bedwetting.

See GRADE table in chapter 14.

15.2.1.18. Imipramine compared to desmopressin combined with oxybutinin for children with bedwetting

One randomised controlled trial Lee (2005) 25, compared imipramine to desmopressin combined with oxybutynin for children with bedwetting.

Table 15-18Imipramine compared to desmopressin and oxybutynin - Clinical summary of findings

OutcomeImipramineDesmopressin and oxybutyninRelative risk (95% CI)Absolute effectQuality
Number of children who had 0–1 wet nights per month3/23 (13%)14/22 (63.6%)RR 0.2 (0.07 to 0.62)509 fewer per 1000 (from 242 fewer to 591 fewer)LOW
Mean number of wet nights per week at the end of treatment2322-MD 1.07 (0.06 to 2.08)VERY LOW

15.2.1.19. Imipramine for children with monosymptomatic nocturnal enuresis

One observational study, Monda (1995) 144 considered imipramine for children with monosymptomatic nocturnal enuresis. Children had 1 mg/kg imipramine, increased to 1.5 mg/kg if still wetting after 2 weeks, and was given 30 to 45 minutes before going to bed.

15.2.1.20. Imipramine compared to placebo for children with severe wetting

One randomised controlled trial compared imipramine to placebo for children with severe wetting, Hagglund (1964) 145.

Table 15-19Imipramine compared to placebo for children with severe wetting - Clinical summary of findings

OutcomeImipraminePlaceboRelative risk (95% CI)Absolute effectQuality
Number of children who achieved >90% improvement in the number of dry nights3/7 (42.9%)0/8 (0%)RR 7.88 (0.48 to 130.28)0 more per 1000 (from 0 fewer to 0 more)VERY LOW

15.2.1.21. Imipramine and placebo compared to placebo for children with severe wetting

One randomised controlled trial compared imipramine and placebo to placebo for children with severe wetting only, Forsythe (1969) 146.

Table 15-20Imipramine and placebo compared to placebo - Clinical summary of findings

OutcomeImipramine and placeboPlaceboRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights1/76 (1.3%)1/85 (1.2%)RR 1.12 (0.07 to 17.57)1 more per 1000 (from 11 fewer to 199 more)VERY LOW
Number of children who achieved greater than 50% improvement in the number of dry nights22/76 (28.9%)21/85 (24.7%)RR 1.17 (0.7 to 1.95)42 more per 1000 (from 74 fewer to 235 more)VERY LOW

15.2.1.22. Imipramine and placebo compared to nortriptyline, placebo for children with severe wetting

One randomised controlled trial compared imipramine and placebo to nortriptyline and placebo for children with severe only wetting, Forsythe (1969) 146.

Table 15-21Imipramine and placebo compared to nortriptyline and placebo - Clinical summary of findings

OutcomeImipramine and placeboNortriptyline and placeboRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights1/76 (1.3%)1/86 (1.2%)RR 1.13 (0.07 to 17.78)2 more per 1000 (from 11 fewer to 201 more)VERY LOW
Number of children who achieved greater than 50% improvement in the number of dry nights22/76 (28.9%)34/86 (39.5%)RR 0.73 (0.47 to 1.14)107 fewer per 1000 (from 209 fewer to 55 more)VERY LOW

15.2.1.23. Nortriptyline and placebo compared to placebo for children with severe wetting

One randomised controlled trial compared nortriptyline and placebo to placebo for children with severe only wetting, Forsythe (1969) 146.

Table 15-22Nortriptyline and placebo compared to placebo - Clinical summary of findings

OutcomeNortriptyline and placeboPlaceboRelative risk (95% CI)Absolute effectQuality
Number of children who achieved 14 consecutive dry nights1/86 (1.2%)1/85 (1.2%)RR 0.99 (0.06 to 15.55)0 fewer per 1000 (from 11 fewer to 175 more)VERY LOW
Number of children who achieved greater than 50% improvement in the number of dry nights34/86 (39.5%)21/85 (24.7%)RR 1.6 (1.02 to 2.52)148 more per 1000 (from 5 more to 375 more)VERY LOW

15.2.2. Side effects of tricyclics for the treatment of bedwetting

15.2.2.1. Imipramine compared to placebo

Five randomised controlled trials, Agarwala (1968) 130, Attenburrow (1984) 131, Batislam (1995) 132, Manhas (1967) 137 and Martin (1971) 138 compared imipramine to placebo. All studies considered 25 mg imipramine.

One randomised controlled trial, Martin (1971) 138 considered low dose (10 mg) imipramine compared to placebo.

Table 15-23Imipramine compared to placebo - Clinical summary of findings

OutcomeImipraminePlaceboRelative risk (95% CI)Absolute effectQuality
Number of children with anxiety4/57 (7%)1/57 (1.8%)RR 4 (0.46 to 34.7)54 more per 1000 (from 10 fewer to 607 more)VERY LOW
Number of children with lethargy4/9 (44.4%)0/12 (0%)RR 11.7 (0.71 to 192.98)0 more per 1000 (from 0 fewer to 0 more)VERY LOW
Number of children with sleep disturbances3/57 (5.3%)3/57 (5.3%)RR 1 (0.21 to 4.75)0 fewer per 1000 (from 42 fewer to 199 more)VERY LOW
Number of children with dizziness1/29 (3.4%)0/29 (0%)RR 3 (0.13 to 70.74)0 more per 1000 (from 0 fewer to 0 more)LOW
Number of children with giddiness2/29 (6.9%)1/27 (3.7%)RR 1.86 (0.18 to 19.38)32 more per 1000 (from 30 fewer to 680 more)VERY LOW
Number of children with dizziness and dry mouth1/9 (11.1%)0/12 (0%)RR 3.9 (0.18 to 85.93)0 more per 1000 (from 0 fewer to 0 more)LOW
Number of children with gastrointestinal8/16 (50%)0/12 (0%)RR 13 (0.82 to 205.24)0 more per 1000 (from 0 fewer to 0 more)VERY LOW
Number of children with upset stomach2/9 (22.2%)0/12 (0%)RR 6.5 (0.35 to 120.8)0 more per 1000 (from 0 fewer to 0 more)VERY LOW
Number of children with abdominal pain4/86 (4.7%)1/84 (1.2%)RR 2.89 (0.46 to 18.13)23 more per 1000 (from 6 fewer to 206 more)MODERATE
Number of children with abdominal pain and epistaxis1/29 (3.4%)0/27 (0%)RR 2.8 (0.12 to 65.93)0 more per 1000 (from 0 fewer to 0 more)VERY LOW
Number of children with vomiting and drowsiness leading to withdrawal1/9 (11.1%)0/12 (0%)RR 3.9 (0.18 to 85.93)0 more per 1000 (from 0 fewer to 0 more)LOW
Number of children with vomiting, sweating and sickness1/9 (11.1%)0/12 (0%)RR 3.9 (0.18 to 85.93)0 more per 1000 (from 0 fewer to 0 more)LOW
Number of children with anorexia1/9 (11.1%)0/12 (0%)RR 3.9 (0.18 to 85.93)0 more per 1000 (from 0 fewer to 0 more)LOW
Number of children with weight loss0/57 (0%)2/57 (3.5%)RR 0.2 (0.01 to 4.08)28 fewer per 1000 (from 35 fewer to 108 more)LOW
Number of children with constipation3/9 (33.3%)0/12 (0%)RR 9.1 (0.53 to 156.72)0 more per 1000 (from 0 fewer to 0 more)VERY LOW

Table 15-24Low dose imipramine compared to placebo - Clinical summary of findings

OutcomeLow dose imipraminePlaceboRelative risk (95% CI)Absolute effectQuality
Number of children with anxiety2/57 (3.5%)1/57 (1.8%)RR 2 (0.19 to 21.44)18 more per 1000 (from 15 fewer to 368 more)VERY LOW
Number of children with sleep disturbances5/57 (8.8%)3/57 (5.3%)RR 1.67 (0.42 to 6.65)36 more per 1000 (from 31 fewer to 299 more)VERY LOW
Number of children with abdominal pain1/57 (1.8%)1/57 (1.8%)RR 1 (0.06 to 15.6)0 fewer per 1000 (from 17 fewer to 263 more)VERY LOW
Number of children with weight loss2/57 (3.5%)2/57 (3.5%)RR 1 (0.15 to 6.86)0 fewer per 1000 (from 30 fewer to 205 more)VERY LOW

15.2.2.2. Low dose imipramine compared to high dose imipramine

One randomised controlled trial, Martin (1971) 138 considered low dose (10 mg) imipramine compared to high dose imipramine (25mg).

Table 15-25Low dose imipramine compared to high dose imipramine - Clinical summary of findings

OutcomeLow dose imipramineHigh dose imipramineRelative risk (95% CI)Absolute effectQuality
Number of children with anxiety2/57 (3.5%)4/57 (7%)RR 0.5 (0.1 to 2.62)35 fewer per 1000 (from 63 fewer to 113 more)VERY LOW
Number of children with sleep disturbances5/57 (8.8%)3/57 (5.3%)RR 1.67 (0.42 to 6.65)36 more per 1000 (from 31 fewer to 299 more)VERY LOW
Number of children with abdominal pain1/57 (1.8%)1/57 (1.8%)RR 1 (0.06 to 15.6)0 fewer per 1000 (from 17 fewer to 263 more)VERY LOW
Number of children with weight loss2/57 (3.5%)0/57 (0%)RR 5 (0.25 to 101.89)0 more per 1000 (from 0 fewer to 0 more)VERY LOW

15.2.2.3. Imipramine compared to desmopressin

One randomised controlled trial, Vertucci (1997) 123 considered imipramine compared to desmopressin.

Table 15-26Imipramine compared to desmopressin - Clinical summary of findings

OutcomeImipramineDesmopressinRelative risk (95% CI)Absolute effectQuality
Number of children with pallor, restlessness and cold extremities1/57 (1.8%)0/57 (0%)RR 3 (0.12 to 72.13)0 more per 1000 (from 0 fewer to 0 more)VERY LOW

15.2.2.4. Amitriptyline compared to placebo

One randomised controlled trial, Poussaint (1966) 142 considered amitriptyline compared to placebo.

Table 15-27Amitriptyline compared to placebo - Clinical summary of findings

OutcomeAmitriptylinePlaceboRelative risk (95% CI)Absolute effectQuality
Number of children who became irritable7/16 (43.8%)5/16 (31.3%)RR 1.4 (0.56 to 3.49)125 more per 1000 (from 138 fewer to 779 more)VERY LOW
Number of children who were calmer2/16 (12.5%)0/16 (0%)RR 5 (0.26 to 96.59)0 more per 1000 (from 0 fewer to 0 more)VERY LOW
Number of children who were drowsy3/16 (18.8%)0/16 (0%)RR 7 (0.39 to 125.44)0 more per 1000 (from 0 fewer to 0 more)VERY LOW
Number of children with fatigue1/16 (6.3%)1/16 (6.3%)RR 1 (0.07 to 14.64)0 fewer per 1000 (from 59 fewer to 859 more)VERY LOW
Number of children with stomach ache1/16 (6.3%)5/16 (31.3%)RR 0.2 (0.03 to 1.53)250 fewer per 1000 (from 304 fewer to 166 more)VERY LOW
Number of children with lower appetite1/16 (6.3%)1/16 (6.3%)RR 1 (0.07 to 14.64)0 fewer per 1000 (from 59 fewer to 859 more)VERY LOW

15.2.2.5. Nortriptyline compared to placebo

One randomised controlled trial, Lake (1968) 143 compared nortriptyline to placebo.

Table 15-28Nortirpyline compared to placebo - Clinical summary of findings

OutcomeNortriptylinePlaceboRelative risk (95% CI)Absolute effectQuality
Headache, aching arms and sore tummy1/54 (1.9%)0/54 (0%)RR 3 (0.12 to 72.05)0 more per 1000 (from 0 fewer to 0 more)VERY LOW

15.2.2.6. Imipramine

Two observational studies, Bain (1973) 147 and Goel (1974) 148 considered the side effects of imipramine and amitriptyline.

15.2.2.7. Imipramine compared to placebo for children with bedwetting

One randomised controlled trial, Tahmaz (2000) 129 compared imipramine to placebo.

Table 15-29Imipramine compared to placebo - Clinical summary of findings

OutcomeImipraminePlaceboRelative risk (95% CI)Absolute effectQuality
Number of children with dry mouth or nausea3/14 (21.4%)4/16 (25%)RR 0.86 (0.23 to 3.19)35 fewer per 1000 (from 192 fewer to 548 more)VERY LOW

15.2.2.8. Imipramine compared to oxybutynin for children with bedwetting

One randomised controlled trial, Tahmaz (2000) 129 compared imipramine to oxybutynin.

Table 15-30Imipramine compared to oxybutynin - Clinical summary of findings

OutcomeImipramineOxybutyninRelative risk (95% CI)Absolute effectQuality
Number of children with dry mouth or nausea3/14 (21.4%)4/16 (25%)RR 0.86 (0.23 to 3.19)35 fewer per 1000 (from 192 fewer to 548 more)VERY LOW

15.2.2.9. Imipramine compared to imipramine and oxybutynin for children with bedwetting

One randomised controlled trial, Tahmaz (2000) 129 compared imipramine to imipramine and oxybutynin.

Table 15-31Imipramine compared to imipramine and oxybutynin - Clinical summary of findings

OutcomeImipramineImipramine and oxybutyninRelative risk (95% CI)Absolute effectQuality
Number of children with dry mouth or nausea3/14 (21.4%)4/23 (17.4%)RR 1.23 (0.32 to 4.71)40 more per 1000 (from 118 fewer to 646 more)VERY LOW

15.2.2.10. Imipramine for children with monosymptomatic nocturnal enuresis

One observational study, Monda (1995) 144 considered imipramine for children with monosymptomatic nocturnal enuresis. Children had 1 mg/kg imipramine, increased to 1.5 mg/kg if still wetting after 2 weeks, and was given 30 to 45 minutes before going to bed.

15.2.3. Network Meta-Analysis

Tricyclic medication was amongst the interventions included in a network meta-analysis of interventions used for nocturnal enuresis. The summary of results of this analysis is presented in chapter 24 and a detailed description of the analysis is presented in appendix F. If studies did not meet the inclusion criteria of the network meta-analysis protocol as stated in appendix F they were not included in the network meta-analysis

15.2.4. Health economic evidence review

Given the lack of published evidence assessing the cost-effectiveness of different interventions, including tricyclics, used in the treatment of bedwetting, the GDG identified this area as high priority for original economic analysis. Therefore, a cost-utility analysis was undertaken where costs and quality-adjusted life-years (QALYs) were considered from a UK National Health Service and Personal Social Services perspective. The time horizon for the analysis was 13 years, modelling patients from the time they entered at age 7 years until they reached age 20.

A summary of the analysis is provided below. The full report is presented in appendix G.

Summary of results

The results of the probabilistic sensitivity analysis are summarised in table 15-32 in terms of mean total costs and mean total QALYs and mean net benefit for each treatment sequence, where each mean is the average of 20,000 simulated estimates. The option with the greatest mean net benefit is the most cost-effective at a specified threshold (for example, £20,000). The percentage of simulations where each strategy was the most cost-effective gives an indication of the strength of evidence in favour of that strategy being cost-effective.

Table 15-32. Basecase probabilistic sensitivity analysis results.

Table 15-32

Basecase probabilistic sensitivity analysis results.

The results of the incremental analysis in the probabilistic analysis, excluding dominated and extendedly dominated strategies, are presented in table 15-33.

Table 15-33. Incremental analysis of basecase probabilistic results with dominated and extendedly dominated sequences removed.

Table 15-33

Incremental analysis of basecase probabilistic results with dominated and extendedly dominated sequences removed.

Results of the basecase probabilistic analysis indicate treatment sequences that included imipramine were never cost-effective.

The GDG was concerned that alarms, despite their cost-effectiveness, may not be an appropriate intervention for all children. For this group of children, a strategy of starting and maintaining desmopressin with or without the addition of an anticholinergic until sustained dryness is achieved is considered cost-effective. Imipramine as a first line intervention or as longer term treatment was not cost-effective in this scenario, as desmopressin based strategies were either less costly and more effective (thus dominating imipramine-based sequences) or had a more favourable ICER (thus extendedly dominating imipramine-based sequences).

A series of sensitivity analyses were undertaken to test some of the assumptions feeding into the model and none of these affected the cost-effectiveness of the alone compared to no treatment. Furthermore, imipramine-based treatment sequences never became cost-effective in any sensitivity analysis undertaken.

The data for imipramine which was fed into the model was not particularly promising, in that the odds ratio of imipramine compared to no treatment from the network meta-analysis crossed 1 and were thus not statistically significant. In addition, despite imipramine’s very small acquisition cost, the BNF 149 states that a consultation with a health care professional must take place every 3 months before further courses of treatment can be pursued. The combination of non-significant effectiveness results and ongoing monitoring costs are likely to contribute to imipramine’s poor performance in the cost-effectiveness analysis.

The economic analysis conducted and presented here represents the first undertaken to assess the cost-effectiveness of interventions used in the treatment of children with bedwetting. And although the analysis is directly applicable to decision making in the UK NHS, it has some potentially serious limitations, some of which may significantly impact the overall conclusions that can be drawn. The main limitations of the analysis are related to the fact that assumptions had to be made in the absence of evidence. Some of these key assumptions centre around:

  • treatment effectiveness being independent of age
  • health care resource use having been estimated by GDG
  • utility weights having been estimated by GDG

A full discussion of these can be found in appendix G.

15.2.5. Evidence statements

The evidence statements are organised by population included in studies and intervention. A number of different tricyclic antidepressants drugs have been used in the studies and the GDG wished these to be reported separately.

Studies included children with bedwetting and possible daytime urinary symptoms

Imipramine

The studies included in the review had varying dosages of imipramine given, based on age or weight of the patient, with younger children being given 25 mg imipramine and older children being given 50 mg imipramine.

Agarwala (1968) 130, Hodes (1973) 135, Khorana (1972) 136, Manhas (1967) 137, Poussaint (1965) 139, Smellie (1996) 140
  • Six studies showed that children treated with imipramine were more likely to achieve 14 consecutive dry nights compared to children treated with placebo. Relative risk 5.06, 95% CI 2.84, 8.99. Children had an age range of 5 to 16 years and had 2 to 12 weeks of treatment.
Kolvin (1972) 103
  • One study showed that children treated with imipramine were more likely to have an > 80% improvement in the number of dry nights compared to children treated with placebo. Relative risk 2.47, 95% CI 1.03, 5.89. Children had a mean age of 9 years and 4 months and had 2 months of treatment.
  • One study showed that children treated with placebo had 0.52 fewer wet nights per week at follow up compared to children treated with imipramine. Children had a mean age of 9 years and 4 months and had 2 months of treatment. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Kolvin (1972) 103
  • One study showed there was no significant difference in the number of children who had a > 80% improvement in the number of dry nights between children treated with imipramine and children treated with an enuresis alarm. Relative risk 0.86, 95% CI 0.53, 1.4. Children had a mean age of 9 years and 4 months and had 2 months of treatment.
  • One study showed that children treated with an enuresis alarm had 1.05 fewer wet nights per week at follow up compared to children treated with imipramine. Children had a mean age of 9 years and 4 months and had 2 months of treatment. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Batislam (1995) 132, Manhas (1967) 137
  • Two studies showed that children treated with imipramine were more likely to have a >50% improvement in the number of dry nights compared to children treated with placebo. Relative risk 2.35, 95% CI 1.27, 4.34. Children had an age range of 5 to 18 years and had 4 weeks of treatment.
Attenburrow (1984) 131
  • One study showed there was no significant difference in the number of wet nights per week at the end of treatment between children treated with imipramine and children treated with placebo. Mean difference −2.5, 95% CI −5.74, 0.74. Children had a median age of 7 years and had 7 weeks of treatment.
  • One study showed there was no significant difference in the number of wet nights per week at follow up between children treated with imipramine and children treated with placebo. Mean difference −1.5, 95% CI −4.85, 1.85. Children had a median age of 7 years and had 7 weeks of treatment.
Drew (1966) 133, Fournier (1987) 81, Harrison (1970) 134, Kolvin (1972) 103, Smellie (1996) 140, Treffert (1964) 141
  • Six studies showed that children treated with imipramine had 0.4 to 4 fewer wet nights per week at the end of treatment compared to children treated with placebo. Children had an age range of 5 to 18 years and had 20 nights to 2 months of treatment. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Agarwala (1968) 130
  • One study showed that children treated with imipramine had fewer wet nights per 2 weeks during treatment compared to children treated with placebo. Mean difference −2.3, 95% CI −4.19, −0.41. Children had an age range of 6 to 12 years and had 2 to 4 weeks of treatment.
Martin (1971) 138
  • One study showed that children treated with imipramine had fewer wet nights during 26 nights of treatment compared to placebo. Mean difference −6.3, 95% CI −8.6, −4. Children had an age range of 5 to 15 years and had 26 nights of treatment. Children had an age range of 5 to 15 years and had 26 nights of treatment.
Harrison (1970) 134
  • Two studies showed there was no significant difference in the number of children who dropped out of the trial between children treated with imipramine and children treated with placebo. Relative risk 5.00 95% CI 0.25, 100.20. Children had an age range of 6 to 18 and had 20 nights.
Vertucci (1997) 123
  • One study showed there was no significant difference in the number of children who achieved 14 consecutive dry nights between children treated with imipramine and children treated with desmopressin. Relative risk 0.79, 95% CI 0.59, 1.06. Children had a mean age of 10 years and had 3 weeks of treatment.
  • One study showed that children treated with desmopressin had 1.8 fewer wet nights per week at the end of treatment compared to children treated with imipramine. Children had a mean age of 10 years and had 3 weeks of treatment. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study showed that children treated with imipramine first had 0.7 fewer wet nights per week after children had been treated with both drugs compared to children treated with desmopressin first. Children had a mean age of 10 years and had 3 weeks of treatment. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Lee (2005) 25
  • One study showed there was no significant difference in the number of children who had 0 to 1 wet nights per month between children treated with imipramine and children treated with desmopressin. Relative risk 0.35, 95% CI 0.11, 1.13. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed that children treated with desmopressin had fewer wet nights per week at the end of treatment compared to treatment with imipramine. Mean difference 1.4, 95% CI 0.55, 2.25. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed there was no significant difference in the number of children who dropped out of the trial between children treated with imipramine and children treated with desmopressin. Relative risk 2.38, 95% CI 0.65, 8.68. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed children continue to have a decrease in the number of wet nights at 1 month, 3 months and 6 months in treatment with both imipramine or desmopressin treatment. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed there was no statistically significant difference in the number of children who had 0 to 1 wet nights per month at the end of treatment between children treated with imipramine and children treated with desmopressin and oxybutinin. Relative risk 0.35, 95% CI 0.11, 1.13. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed that children treated with desmopressin and oxybutinin had fewer wet nights per week at the end of treatment compared to children treated with imipramine. Mean difference 1.43, 95% CI 0.45, 2.41. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed there was no statistically significant difference in the number of children who dropped out between children treated with imipramine and children treated with desmopressin and oxybutinin. Relative risk 2.33, 95% CI 0.64, 8.49. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed children continue to have a decrease in the number of wet nights at 1 month, 3 months and 6 months in treatment with both imipramine or desmopressin combined with oxybutynin treatment. Children had a mean age of 7.8 years and were treated for 6 months.
Fournier (1987) 81, Kolvin (1972) 103
  • Two studies showed that children treated with imipramine had 0 to 0.6 fewer wet nights per week at the end of treatment compared to children treated with an enuresis alarm. Children had a mean age of 8.5 (Fournier 1987) and 9 years and 4 months (Kolvin 1972) and had 6 weeks (Fournier (1987) and 2 months (Kolvin 1972) of treatment. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Fournier (1987) 81
  • One study showed there was no difference in the number of children who dropped out of the trial between children treated with imipramine and children treated with an enuresis alarm. Relative risk 1, 95% CI 0.07, 13.37. Children had a mean age of 8.5 and had 6 weeks of treatment.
  • One study showed that children treated with imipramine and an enuresis alarm had 0.9 fewer wet nights per week at follow up compared to children treated with imipramine. Children had a mean age of 8 years and 5 months and had 6 weeks of treatment. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study showed there was no difference in the number of children who dropped out between children treated with imipramine and children treated with imipramine and an enuresis alarm. There were no drop outs from either treatment group. Children had a mean age of 8 years and 5 months and had 6 weeks of treatment.
Low dose imipramine compared to placebo

One paper considered 10 mg imipramine compared to a placebo. The usual stated dosage for imipramine in the treatment of nocturnal enuresis 25 mg imipramine for younger children and 50 mg imipramine for older children. It was therefore considered that a dosage of 10 mg imipramine should be evaluated separately from the usual higher dosage of imipramine.

Martin (1971) 138
  • One study showed that children treated with 10 mg imipramine had fewer wet nights during 26 nights of treatment compared to children treated with placebo. Mean difference −3.1, 95% CI −5.1, −1.1. Children had an age range of 5 to 15 years and had 26 nights of treatment.
  • One study showed that children treated with 25 mg imipramine had fewer wet nights during treatment compared to children treated with 10 mg imipramine. Relative risk 3.2, 95% CI 1.3, 5.1. Children had an age range of 5 to 15 years and had 26 nights of treatment.
Amitriptyline
Poussaint (1966) 142
  • One study (containing two trials) showed that children treated with amitriptyline had 1.4 and 1.5 fewer wet nights per week at the end of treatment compared to children treated with placebo. Children had an age range of 5 to 15 years and had treatment for 4 or 8 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Burke (1995) 122
  • One study showed there was no statistically significant difference in the number of children who achieved 14 consecutive dry nights between children treated with amitriptyline and children treated with desmopressin. Relative risk 0.6, 95% CI 0.18, 2.04. Children had a mean age of 8.6 to 8.9 years and had treatment for 16 weeks.
  • One study showed there was no statistically significant difference in the mean number of wet nights per week at the end of treatment between children treated with amitriptyline and children treated with desmopressin. Mean difference −1.4, 95% CI −2.95, 0.15. Children had a mean age of 8.6 to 8.9 years and had treatment for 16 weeks.
  • One study showed there was no statistically significant difference in the mean number of wet nights per week at follow up between children treated with amitriptyline and children treated with desmopressin. Mean difference 0.1, 95% CI −1.67, 1.87. Children had a mean age of 8.6 to 8.9 years and had treatment for 16 weeks.
  • One study showed there was no statistically significant difference in the number of children who dropped out of the trial between children treated with amitriptyline and children treated with desmopressin. Relative risk 0.17, 95% CI 0.01, 3.06. Children had a mean age of 8.6 to 8.9 years and had treatment for 16 weeks.
  • One study showed there was no statistically significant difference in the number of children who achieved 14 consecutive dry nights between children treated with amitriptyline and children treated with amitriptyline combined with desmopressin. Relative risk 0.6, 95% CI 0.18, 2.04. Children had a mean age of 8.6 to 8.9 years and had treatment for 16 weeks.
  • One study showed there was no difference in the mean number of wet nights per week at the end of treatment between children treated with amitriptyline and children treated with amitriptyline combined with desmopressin. Mean difference 0, 95% CI −1.64, 1.64. Children had a mean age of 8.6 to 8.9 years and had treatment for 16 weeks.
  • One study showed there was no statistically significant difference in the mean number of wet nights per week at follow up between children treated with amitriptyline and children treated with amitriptyline combined with desmopressin. Mean difference −1.2, 95% CI −3.46, 1.06. Children had a mean age of 8.6 to 8.9 years and had treatment for 16 weeks.
  • One study showed there was no statistically significant difference in the number of children who dropped out of the trial between children treated with amitriptyline and children treated with amitriptyline combined with desmopressin. Relative risk 0.14, 95% CI 0.01, 2.53. Children had a mean age of 8.6 to 8.9 years and had treatment for 16 weeks.
Danquah (1975) 104
  • One study showed that children treated with enuresis alarms had 0.8 fewer wet nights per week compared to children treated with amitriptyline. Children had a mean age of 10.4 years and had treatment for 7 weeks. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study showed that children treated with an enuresis alarm stopped bedwetting 4.4 days earlier than children treated with amitriptyline. Children had a mean age of 10.4 years and had treatment for 7 weeks.
Nortriptyline compared to placebo
Lake (1968) 143
  • One study showed children treated with nortriptyline had 0.83 fewer wet nights per week during treatment compared to children treated with placebo. Children had an age range of 5 to 12 years and had 2 weeks of treatment. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.

Studies included children with bedwetting only Imipramine

Wagner (1982) 110
  • For children with bedwetting one study showed there was no significant difference in the number of children who achieved 14 consecutive dry nights between children treated with imipramine and children treated with placebo. Relative risk 4.00, 95 % CI 0.52, 30.76. Children had an age range of 6 to 16 years and had 14 weeks of treatment.
Tahmaz (2000) 129
  • One study showed there was no significant difference in the number of children who had 90% improvement in the number of dry nights between children treated with imipramine and children treated with placebo. Relative risk 2.30 95% CI 0.90, 5.86. Children had a mean age of 9.44 years and had 3 months of treatment.
  • One study showed there was no significant difference in the number of children who had 50% to 90% improvement in the number of dry nights between children treated with imipramine and children treated with placebo. Relative risk 1.03, 95% CI 0.42, 2.52. Children had a mean age of 9.44 years and had 3 months of treatment.
  • One study showed there was no significant difference in the number of children who relapsed at 6 months between children treated with imipramine and children treated with placebo. Relative risk 1.79, 95% CI 0.55, 5.76. Children had a mean age of 9.44 years and had 3 months of treatment.
  • Two studies showed there was no significant difference in the number of children who achieved 50% to 90% improvement in the number of dry nights between children treated with imipramine and children treated with oxybutinin. Relative risk 0.95 95% CI 0.37, 2.45. Children had a mean age of 9.44 years and had 3 months of treatment.
  • One study showed there was no significant difference in the number of children who dropped out of the trial between children treated with imipramine and children treated with oxybutynin. Relative risk 0.86 95% CI 0.48, 1.55. Children had a mean age of 9.44 years and had 3 months of treatment.
  • One study showed there was no significant difference in the number of children who achieved 50% to 90% improvement in the number of dry nights between children treated with imipramine and children treated with imipramine and oxybutinin. Relative risk 1.43 95% CI 0.53, 3.83. Children had a mean age of 9.44 years and had 3 months of treatment.
  • One study showed more children relapsed at 6 months after treatment with imipramine compared to children treated with imipramine and oxybutinin. Relative risk 2.86, 95% CI 1.08, 7.53. Children had a mean age of 9.44 years and had 3 months of treatment.
Lee (2005) 25
  • One study showed that children treated with desmopressin were more likely to achieve 0 to 1 wet nights per month compared to children treated with imipramine. Relative risk 0.21, 95% CI 0.07, 0.65. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed that children treated with desmopressin had fewer wet nights per week at the end of treatment compared to children treated with imipramine. Mean difference 1.3, 95% CI 0.38, 2.22. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed that children treated with desmopressin and oxybutinin were more likely to achieve 0 to 1 wet nights per month compared to children treated with imipramine. Relative risk 0.02, 95% CI 0.07, 0.62. Children had a mean age of 7.8 years and were treated for 6 months.
  • One study showed children treated with desmopressin and oxybutinin had fewer wet nights per week at the end of treatment compared to children treated with imipramine. Mean difference 1.07, 95% CI 0.06, 2.08. Children had a mean age of 7.8 years and were treated for 6 months.
Tahmaz (2000) 129, Esmaelli (2008) 128
  • Two studies showed there was no significant difference in the number of children who achieved 14 consecutive dry nights between children treated with imipramine and children treated with oxybutinin. Relative risk 0.94, 95% CI 0.48, 1.84. Children in Tahmaz (2000) had a mean age of 9.44 years and had 3 months of treatment, children in Esmaelli (2008) had a mean age of 8.9 years and had 1 month of treatment.
  • Two studies showed there was no significant difference in the number of children who achieved 14 consecutive dry nights between children treated with imipramine and children treated with imipramine and oxybutinin. Relative risk 0.94, 95% CI 0.48, 1.84. Children in Tahmaz (2000) had a mean age of 9.44 years and had 3 months of treatment, children in Esmaelli (2008) had a mean age of 8.9 years and had 1 month of treatment.
Esmaelli (2008) 128
  • One study showed that children treated with oxybutinin had fewer wet nights per week during treatment compared to children treated with imipramine. Mean difference 1, 95% CI 0.02, 1.98. Children had a mean age of 8.9 years and had 1 month of treatment.
  • One study showed that children treated with imipramine and oxybutinin had fewer wet nights per week during treatment compared to children treated with imipramine. Mean difference 1, 95% CI 0.02, 1.98. Children had a mean age of 8.9 years and had 1 month of treatment.
Wagner (1982) 110
  • One study showed that more children treated with an enuresis alarm achieved 14 consecutive dry nights compared to children treated with imipramine. Relative risk 0.4, 95% CI 0.17, 0.93. Children had a mean age of 7.9 years and had 14 weeks of treatment.
  • One study showed that children treated with an enuresis alarm had 2.17 fewer wet nights per week at the end of treatment compared to children treated with imipramine. Children had a mean age of 7.9 years and had 14 weeks of treatment. No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
  • One study showed there was no significant difference in the number of children who relapsed at 6 months between children treated with imipramine and children treated with an enuresis alarm. Relative risk 1.8, 95% CI 0.93, 3.48. Children had a mean age of 7.9 years and had 14 weeks of treatment.

Studies included children with monosymptomatic nocturnal enuresis

Imipramine
Monda (1995) 144

Studies included children with severe bedwetting

Imipramine
Hagglund (1964) 145
  • For children with severe wetting one study showed there was no significant difference in the number of children who had >90% improvement in the number of dry nights between children treated with imipramine and children treated with placebo. Relative risk 7.88, 95% CI 0.48, 130.28. Children had an age range of 4 to 14 years.
Forsythe (1969) 146
  • For children with severe wetting one study showed there was no significant difference in the number of children who achieved 14 consecutive dry nights between children treated with imipramine and placebo and children treated with placebo. Relative risk 1.12, 95% CI 0.07, 17.57. Children had an age range of up to 15 years and had 8 weeks of treatment.
  • For children with severe wetting one study showed there was no significant difference in the number of children who had >50% improvement in the number of dry nights between children treated with imipramine and placebo and children treated with placebo. Relative risk 1.17, 95% CI 0.70, 1.95. Children had an age range of up to 15 years and had 8 weeks of treatment.
  • For children with severe wetting one study showed there was no significant difference in the number of children who achieved 14 consecutive dry nights between children treated with imipramine and placebo and children treated with nortriptyline and placebo. Relative risk 1.13, 95% CI 0.07, 17.78. Children had an age range of up to 15 years and had 8 weeks of treatment.
  • For children with severe wetting one study showed there was no significant difference in the number of children who had >50% improvement in the number of dry nights between children treated with imipramine and placebo and children treated with nortriptyline and placebo. Relative risk 0.73, 95% CI 0.47, 1.14. Children had an age range of up to 15 years and had 8 weeks of treatment.
Nortriptyline
Forsythe (1969) 146
  • For children with severe wetting one study showed there was no significant difference in the number of children who achieved 14 consecutive dry nights between children treated with nortripyline and placebo and children treated with placebo. Relative risk 0.99, 95% CI 0.06, 15.55. Children had an age range of up to 15 years and had 8 weeks of treatment.
  • For children with severe wetting one study showed children treated with nortriptyline and placebo were more likely to have >50% improvement in the number of dry nights compared to children treated with placebo. Relative risk 1.60, 95% CI 1.02, 2.52. Children had an age range of up to 15 years and had 8 weeks of treatment.

Side effects for tricyclics

The side effects are extracted from RCTs or observational studies and listed by individual tricyclic.

Imipramine
Martin (1971) 138
Attenburrow (1984) 131
Agarwala (1968) 130
Manhas (1967) 137
Manhas (1967) 137, Martin (1971) 138
  • Two randomised controlled trials showed there was no statistically significant difference in the number of children with abdominal pain between children treated with imipramine and children treated with placebo. Relative risk 2.89, 95% CI 0.46, 18.13. Children had an age range of 5 to 18 years and had between 20 nights and 3 months of treatment.
Batislam (1995) 132
  • One randomised controlled trial showed there was no statistically significant difference in the number of children with gastrointestinal problems between children treated with imipramine and children treated with placebo. Relative risk 13, 95% CI 0.82, 205.24. Children had an age range of 5 to 18 years and had between 20 nights and 3 months of treatment.
Vertucci (1997) 123
Bain (1973) 147
  • One observational trial showed there was an increase in cases of imipramine poisoning, in 1968 and 1970. In 1968, 17 cases of poisoning were reported, by 1970 there were 36 cases. The study reported one author collected the reason for 20 deaths in children from imipramine poisoning. Only one of these was from a drug prescribed for the child who died from nocturnal enuresis.
Goel (1974) 148
  • One observational trial showed there were 60 cases of amitriptyline and imipramine poisoning in children between January 1966 and July 1973. 16 of which were from the medication prescribed for the child poisoned for the treatment of nocturnal enuresis. The study reported the cases of poisoning from amitriptyline and imipramine prescribed for the treatment of nocturnal enuresis. The study reported the cardiovascular features of poisoning (prescribed for both nocturnal enuresis and depression, the study did not separate out the results for the two groups). From amitriptyline poisoning 24 children had sinus tachycardia, 2 children had sinus arrhythmia, 2 children had ventricular premature systole, 0 children had conduction disturbances, 1 child had hypotension and 1 child had cardiorespiratory arrest. From imipramine poisoning 12 children had sinus tachycardia, 2 children had sinus arrhythmia, 1 child had ventricular premature systole, 2 children had conduction disturbances, 2 children had hypotension and 2 children had cardiorespiratory arrest. The study also reported neurological and atropinic features of poisoning, from amitriptyline 36 patients had drowsiness, 17 had agitation and / or restlessness, 16 had ataxis, 5 had mydriasis, 9 had vomiting, 8 had flushing of the face, 1 had coma, 6 had convulsions, 4 had hyperrefexia, 2 had retention of urine, 3 had hallucinations, 1 had dysarthria and 2 had nystagmus. From imipramine 12 patients had drowsiness, 7 had agitation and / or restlessness, 1 had ataxis, 8 had mydriasis, 3 had vomiting, 3 had flushing of the face, 2 had coma, 2 had convulsions, 1 had hyperreflexia, 2 had retention of urine, 0 had hallucinations, 1 had dysarthria and 0 had nystagmus.
Tahmaz (2000) 129
Monda (1995) 144
Low dose imipramine compared to high dose imipramine
Martin (1971) 138
Amitriptyline
Poussaint (1966) 142
Nortriptyline
Lake (1968) 143

One randomised controlled trial showed there was no statistically significant difference in the number of children with a sore tummy between children treated with nortriptyline and children treated with placebo. Relative risk 3, 95% CI 0.12, 72.05. Children had an age range of 5 to 15 years and had 2 weeks of treatment.

NCGC network meta-analysis (see appendix F)

For children with bedwetting and possible daytime symptoms
  • The NCGC NMA showed there was a statistically significant difference in the number of children who achieved a full response between children treated with imipramine and no treatment / placebo. Relative risk 6.149, 95% CI 3.100, 8.537. Children had an age range of 5 to 17 years and treatment for a minimum of 8 weeks.
  • The NCGC NMA showed there was a statistically significant difference in the number of children who achieved a full response between children treated with amitriptyline and no treatment / placebo. Relative risk 9.514, 95% CI 6.906, 9.677. Children had an age range of 5 to 17 years and treatment for a minimum of 8 weeks.
For children with bedwetting only
  • The NCGC NMA showed there was no statistically significant difference in the number of children who experienced a recurrence of bedwetting at 6 months between children treated with combined imipramine and oxybutynin and no treatment / placebo. Relative risk 0.011, 95% CI 0.0001, 2.764. Children had an age range of 5 to 17 years and treatment for a minimum of 8 weeks.
  • The NCGC NMA showed there was no statistically significant difference in the number of children who experienced a recurrence of bedwetting at 6 months between children treated with imipramine and no treatment / placebo. Relative risk 4.566, 95% CI 0.277, 52.54. Children had an age range of 5 to 17 years and treatment for a minimum of 8 weeks.
  • The NCGC NMA showed there was no statistically significant difference in the number of children who achieved a full response between children treated with imipramine and no treatment / placebo. Relative risk 2.259, 95% CI 0.513, 6.172. Children had an age range of 5 to 17 years and treatment for a minimum of 8 weeks.
  • The NCGC NMA showed there was no statistically significant difference in the number of children who achieved a full response between children treated with combined imipramine and oxybutynin and no treatment / placebo. Relative risk 4.188, 95% CI 0.561, 8.737. Children had an age range of 5 to 17 years and treatment for a minimum of 8 weeks.

For estimates of treatment effect relative to other active comparators, please see section 24.4 in chapter 24.

15.2.6. Health economic evidence statements

NCGC economic evaluation (see appendix G)

  • Intervention sequences that include imipramine are not cost-effective in the treatment of children with bedwetting as they are more costly and less effective than alternative intervention sequences such as ones starting with alarm and moving to combined alarm and desmopressin or starting with desmopressin and moving on the alarm or combined desmopressin and anticholinergic. This evidence has potentially serious limitations and direct applicability.

15.2.7. Evidence to recommendations

Relative values of different outcomes

The GDG considered the children, young people and parents or carers starting treatment for bedwetting were seeking an outcome of sustained dryness. A number of different outcomes were used to capture this: the outcome of 14 consecutive dry nights, reduction in wet nights and the mean number of wet nights allow evaluation of the effectiveness of treatment. Follow up rates, where available, indicated sustained dryness.

Trade off between clinical benefit and harms

The GDG were concerned about the potential side effects of tricyclic antidepressants and their danger in overdose.

Economic considerations

Imipramine was shown not to be a cost-effective first line intervention for the treatment of bedwetting. First line treatment with alarm or desmopressin is likely to be less costly and more effective than offering imipramine.

Imipramine is not considered to be a cost-effective intervention and therefore should not be used early in the treatment of bedwetting. If, however, patients have not responded to any other treatments or they are deemed unsuitable for various reasons, imipramine could be considered as a possible alternative.

Quality of evidence (this includes clinical and economic)

The evidence from studies of direct comparisons was generally poor with many older studies having wide confidence intervals which lacked follow up data. There were questions over the lengths of treatment for both imipramine, which may take longer to be effective than other drug treatments and for enuresis alarm comparisons where the length of treatment was insufficient to see the full effect of the treatment.

Other considerations

The GDG used the evidence from direct comparisons, the network meta-analysis and the health economic evidence to inform their recommendations.

While the research studies have used various tricyclic antidepressants the GDG considered that imipramine was the tricyclic of choice to use in children and young people. It is the drug most commonly used for this indication and there is therefore more experience of its use and the case fatality rate is considered to be higher with other tricyclics.

One study used a lower dose of imipramine (10mg) than currently recommended and although 25mg was more effective, the lower dose did result in fewer wet nights compared to placebo. This might indicate that lower doses are worth trying if imipramine is being used.

The GDG considered that from clinical experience there was a role for the use of tricyclic antidepressants, particularly for children with daytime symptoms, although some children and young people with bedwetting only do also respond.

The GDG considered children, young people and families and carers required information to understand the use of imipramine and developed a recommendation on information for children, young people and families from professional and personal experience.

The GDG considered however that a trial of imipramine should only be instigated by healthcare professionals with appropriate expertise in using imipramine. The GDG considered that children and young people started on imipramine require careful follow up and review because of the side effects profile to ensure that children and young people are only continuing to take the drug if it is effective. Imipramine should be slowly withdrawn if side effects are experienced or if there is no improvement in bedwetting after 2 weeks at maximum dose for age.

Children and young people who respond well require medical follow up at 3 monthly intervals along with slow withdrawal of medication ensuring that the dose of imipramine is kept as low as possible to maintain dryness.

15.2.8. Recommendations

15.2.8.1.

Do not use tricyclic antidepressants as a first line treatment for bedwetting in children and young people. [1.14.1]

15.2.8.2.

If offering a tricyclic antidepressant, imipramine should be used for the treatment of bedwetting in children and young people. [1.1342]

15.2.8.3.

Consider imipramine for children and young people with bedwetting who:

  • have not responded to all other treatments and
  • have been assessed by a healthcare professional with expertise in the management of bedwetting that has not responded to an alarm and/or desmopressin.[1.14.3]
15.2.8.4.

If offering imipramine for bedwetting, inform the child and young person and their parents or carers:

  • that many children and young people, but not all, will experience a reduction in wetness
  • how imipramine works
  • that it should be taken before bedtime
  • that the dose should be increased gradually
  • about relapse rates( for example, more than two out of three children and young people will relapse after a 3-month course of imipramine)
  • that the initial treatment course is for 3 months and further courses may be considered
  • about the particular dangers of imipramine overdose, and the importance of taking only the prescribed amount and storing it safely.[1.14.4]
15.2.8.5.

Perform a medical review every 3 months in children and young people who are using repeated courses of imipramine for the management of bedwetting.[1.14.5]

15.2.8.6.

Withdraw imipramine gradually when stopping treatment for bedwetting in children and young people.[1.14.6]

Copyright © 2010, National Clinical Guideline Centre.

Apart from any fair dealing for the purposes of research or private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

The rights of the National Clinical Guideline Centre to be identified as Author of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act, 1988.

Cover of Nocturnal Enuresis
Nocturnal Enuresis: The Management of Bedwetting in Children and Young People.
NICE Clinical Guidelines, No. 111.
National Clinical Guideline Centre (UK).

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...