Acta Piediatr 86: 385-90. 1997

Diurnal plasma vasopressin and urinary output in adolescents with monosymptomatic nocturnal enuresis G Lackgren', T Nevtus2 and A Stenberg' Section of Urology', Department of Pediatric Surgery, Uppsala University Children's Hospital, and Department of Padiatrics2, Uppsala University Children's Hospital, Uppsala, Sweden

Lackgren G, Neve6s T, Stenberg A. Diurnal plasma vasopressin and urinary output in adolescents with monosymptomatic nocturnal enuresis. Acta Paediatr 1997; 86: 385-90. Stockholm. ISSN 0803-5253 Plasma arginine vasopressin (AVP) levels, urinary flow and urine osmolality were investigated in a group of adolescents (20 boys and 5 girls), aged 11-21 y, with severe monosymptomatic nocturnal enuresis and a control group of healthy adolescents (I6M and 4F) with similar age- and sex-distribution. Half of the control group was investigated twice, with an interval of 6 months. AVP samples were taken every fourth hour in all adolescents and half of the control group were also investigated every second hour to achieve more samples during controlled sleep. After the study the enuretic group were put on long-term oral desmopressin (DDAVP). The difference between day and night values of AVP was significant for both groups, but there was no difference in the daylnight ratios of plasma-AVP. All the adolescents produced less urine while asleep, but the controls produced significantly more urine than the enuretics during day. The controls also had a significantly larger nocturnal elevation of urine osmolality than the enuretics, thus a tendency towards polyuria was found. We could not find any significant difference between responders to DDAVP treatment and non-responders in any of the parameters studied. AVP is secreted in a pulsatile fashion and with point hormone samples taken every fourth or second hour we were unable to find any difference in the diurnal AVP secretion between enuretics and normal controls. Adolescents, enuresis, vasopressin

G Liickgren, Section of Urology, Department of Paediatric Surgery, University Children's Hospital, S-751 85 Uppsala, Sweden

Primary nocturnal enuresis (PNE) is a common problem in childhood, with an estimated incidence of about 7-10% among 7-y-olds (1, 2). Although the natural history of nocturnal enuresis is benign, the condition leads to con- siderable social embarrassment and much suffering among those afflicted and their families.

The pathogenesis of PNE is still unknown, despite sev- eral decades of research and speculation. Recently, interest has been focused on the physiological nocturnal rise in arginine vasopressin (AVP) levels, first described by George et al. 1975 (3). Puri found low urinary levels of antidiuretic hormone in children with enuresis (4) and later studies by Rittig et al. (5) and Norgaard et al. (6) in 1989 indicated that the normal nocturnal rise in plasma AVP may be missing or deficient in enuretic children. Thus, the cause of PNE might be nightly polyuria; the enuretic child voids when urine production exceeds functional bladder capacity. This explanation provides a rationale for the treatment of PNE with the AVP analogue I-deamino-8-D- arginine vasopressin (DDAVP)-desmopressin (6).

The investigations of Rittig et al. have been repeated by other groups (7-9), but the results are not conclusive, and further confirmation is needed. Furthermore, some ques- tions can arise from the finding that AVP is released in a pulsatile fashion, and from the fact that nightly polyuria is common among children without PNE as well (10).

The aim of the present study was to investigate diurnal

plasma-AVP, urine osmolality and urinary output in adoles- cents with severe monosymptomatic nocturnal enuresis and in normal controls.

Materials and methods Twenty-five adolescents (20M and 5F, aged 11-21 y, mean age 13.5 y) with severe monosymptomatic PNE (minimum 3 wet nights per week) were included in the study. Children with urological or neurological abnormalities or with a history of urinary tract infections were excluded. All children included had, without success, tried different kinds of treatment, including enuresis alarms or antidepressants.

Before enlistment, informed consent was obtained from parents, patients and controls. Approval was also obtained from the hospital ethics committee.

The enuretic group had a pre-study period of 4 weeks and no antienuretic treatment was allowed during these weeks or during the study. Wet and dry nights were recorded during the last 2 weeks of this period. The pre-study investigations included urinary flow measurements, renal concentration test and a physical examination. Routine blood tests, including semm osmolality, creatinine, sodmm, potassium, haemoglobin and erythrocyte sedimentation rate were also carried out and a urine sample was examined for blood cells, bacteria, protein and glucose.

0 Scandinavian University Press 1997. ISSN 0803-5253

386 G Lackgren et d. ACTA PEDIATR 86 ( I 997)

Tublv I. Results. Values shown are mean +- 1 SD

AVP concentration (pgiml) Enuretics Controls Significance

Enuretics Controls Significance

Enuretics Controls Significance

Urine osmolality (mOsrn/l)

IJrine production ( m l h i

1.43 2 0.5 1 0.87 2 0.36

826 t 178 780 t 260

47.3 2 13.8 64.5 2 26.6

Night

1.57 2 0.46 0.96 2 0.37

847 -C 181 964 t 207

27.9 2 12.7 29.3 t 10.1

Dayhight ratio

0.90 -C 0.46 0.84 2 0.20 p = 0.580

1.00 2 0.26 0.83 2 0.29 p = 0.064

1.26 -C 0.80 2.46 2 1.58 p=0.117

Dayhight diff.

-0.41 2 0.83* -0.18 2 0.29*

p = 0.240

-21 2 186 -184 * 195***

p = 0.014

17.5 t 12.8* 35.2 2 27.0" p = 0.008

Significances in the last column indicate that there are differences between day and night for this variable. * p c 0.05 * * * p < 0.001

A control group of 20 healthy adolescents (4F and 16M, aged 12-17 years, mean age 13.5 y) without enuresis was enlisted and examined in the same manner. To test the reproducibility of the AVP measurements, half of the control group (2F and 8M) was re-examined with regard to their AVP values after 6 months. In the comparisons between the enuretics and controls only data from the first occassion were used. In the second half ( n = 10) of the control group, AVP measurements were taken every second hour to obtain more stable AVP- measurements during controlled sleep. However only measurements cor- responding in time to those of the other groups were used for comparisons between enuretics and controls.

During the study, the enuretics were observed at the hospital for 2 consecutive days whereas the controls were observed for 24 h. The liquid intake of all the children was standardized to 25 mlkg per 24 h during the investigation. All urine was collected and osmolality as well as quantity were measured for every portion; for the enuretics, this was done on the second day by means of an indwelling catheter from which urine was collected every fourth hour. The urine samples at 04 : 00 h and 08 : 00 h in the enuretics and the first morning urine in the controls were used as noctur- nal urine.

Blood samples for AVP- measurements were taken every fourth hour in the enuretic group and in half of the control group and every second hour in the other half of the control group. The time between 10 : 00 h and 22 : 00 h on the first day was considered day and from : 00 h until 0600 on the second day was considered night. These definitions were in accordance with the observed sleep of the children. Since some of the children were awake and some were asleep at 08 : 00 h on the second day, the samples taken at this time were omitted from the comparisons of day- and night-time AVP secretion.

The blood samples were collected by means of a large-bore intravenous cannula inserted upon admission to the hos- pital. All the adolescents were observed to be soundly asleep while the night-time samples were taken, and in no instance were they awakened by the procedure. All the blood samples in all patients and controls were taken by the same nurse.

When AVP samples were taken, a minimum of 5 ml blood was collected in ice-cooled vacutainer tubes, con- taining 0.05 ml K3 EDTA, 0.38 moVl (17%) as ariticoagu- lant, and was centrifugated within less than 30 min. The plasma samples were immediately frozen and stored at - 70°C until assayed. All the samples for each group of children were analysed at the same time. The analyses were performed by radioimmunoassay as described by Rooke and Baylis (1 l), utilizing an AVP-specific antibody which has shown no cross-reactivity with related hormones including oxytocin.

Statistical methods Comparisons between the enuretic and control groups regarding urine production and urine osmolality were based on two-way ANOVA, with group as factor.

The time series of AVP data was reduced by computing the mean value for the nightly (00 : 00-08 : 00 h) and daily (10 : 00-22 : 00 h) data, and the difference between day and night. This approach was preferred to a repeated measures analysis of variance since partially missing data would have reduced the sample size to only 34 individuals. The daily and nightly AVP values and their difference was analysed using two-factor Anova. The SAS (1990) package was used for the analyses. Results have been interpreted as significant when p <0.05; however, the actual p values are reported in the Results section and Table 1.

Results The adolescents with enuresis were found to have an average of 4.9 wet nights per week during the pre-study period.

Routine laboratory tests were normal for all children, controls as well as enuretics. Plasma AVP levels, urine osmolality and urine production, as well as the dayhight ratio and day-night difference of these parameters, were compared (Table 1).

For each patient and control the ratios for the mean of the day values and the mean of the night values of plasma AVP

ACTA PEDIATR 86 (1997) Plasma vasopressin in nocturnal enuresis 387

7 -

6 -

5 -

4:

0 ,

0 0 0

0 8 0 - 0 O

Fig. I . Diurnal plasma-AVF' in adolescents with monosymptomatic noc- turnal enuresis. Box graph, indicating loth, 25th, 5Oth, 75th and 90th percentiles.

A V P 2 ' 5 ' ' ' ' ' ' ' ' ~ ' ' ' ' ' ' '

pg/m' 2.25- 0

2 - 0 0

1 7 5 0

1

0 0

0 0 c

.25 ' . 1 08day1 12 16 20 22 00 04 08day2

Fig. 2. Diurnal plasma-AVP in normal controls. Box graph, indicating loth, 25th, 50th, 75th and 90th percentiles.

were calculated. When these ratios were compared (Fig. 5), no significant differences were seen between the enuretic children and the control group. Both in the enuretic group and among the controls, the difference between day and night values of AVP was significant (p=O.O21 for

Fig. 4 . Diurnal plasma-AVP in normal controls. Graph showing half of the control group, with measurements every second hour. Box graph, indicat- ing loth, 25th, 50th, 75th and 90th percentiles.

enuretics: p = 0.012 for controls) with a higher nocturnal level (Figs 1 and 2). Half the control group was studied twice with a 6 months' interval (Fig. 3). The AVP values did not differ significantly in any of the sampling points between the first and the second sampling period, nor was there any difference in dayhght ratios. A circadian rhythm of plasma-AVP with a significant night-time increase was seen when the sampling was carried out more frequently (Fig. 4).

There was no significant change in urine osmolality between day and night in the enuretic group ( p = 0.48). There was a significant nocturnal elevation of urine osmolality for the control group ( p < 0.001). The change in urine osmolality between day and night was significantly larger for the control group than for the enuretic group ( p = 0.014) .The dayhight ratios are shown in Fig. 6.

The urine production of both enuretics and controls showed a distinct circadian rhythm, with both groups producing significantly less urine ( p < 0.001) while asleep. The urine production for the two groups was similar

388 ti Lackgreii er d. ACTA PBDL4TR 86 (1997)

2 -

1 -

,5 -

0

+-

I t

0- Enuretics Controls

5 -

4 -

3 -

2I 1 I

0

0

-8- -

Enuretics Controls

Fig. 5. Dayhight ratios of AVP secretion in enuretics and controls. Box graph. indicating 10th. 25th. 50th. 75th and 90th percentiles.

Fig. 7. Dayhight ratios of urine production in enuretics and controls. Box graph, indicating 10th. 25th, 50th, 75th and 90th percentiles.

at night ( p = 0.70), but during daytime the controls pro- duced significantly more urine than the enuretics ( p = 0.006). The change in urine production between day and night was larger for the controls ( p = 0.0076). There was no difference in daylnight ratio between the two groups (Fig. 7).

After the study. the enuretic group was subjected to long- term treatment with oral Desmopressin. Eleven members were full responders (0-1 wet night per week), five were intermediate responders ( I -3 wet nights per week) and nine were non-responders (> 3 wet nights per week) during treatment. The plasma-AVP, urine osmolality and unne

0 1 3 c

,4 “i

0

F

,2 ‘ L Enuretics Controls

Fig. 6. Dayhight ratios of urine osrnolality in enuretics and controls. Box graph, indicating 10th. 25th. 50th. 75th and 90th percentiles.

production at night did not differ significantly between responders and non-responders (Fig. 8).

Discussion Although the aetiology of primary nocturnal enuresis is still unknown, it is most likely multifactorial, involving hereditary factors, disturbances of the sleep pattern andlor increased nightly urine production.

It is known that the urine production decreases during sleep (5 , 12) but Poulton and Hinden found in an early study (13) that some children with PNE had a significant night-time polyuria. In 1980 Puri (4) found that children with PNE had low urinary levels of antidiuretic hormone and later Rittig et al. demonstrated a significant decrease in nocturnal plasma-AVP in enuretics compared with normal subjects ( 5 ) . Since then a number of studies have been performed aimed at throwing more light on the possible role of the antidiuretic hormone (AVP) and night-time polyuria in the pathogenesis of PNE (5 , 7-9, 14, 15).

Several of these studies may support the idea of a lack of circadian rhythm in patients with PNE, but the results of different studies on AVP-secretion and urine production have been somewhat conflicting. In all these studies dif- ferent criteria of patient selection have been used. Only occasionally was fluid intake standardized and in most of the studies no control groups were investigated. Further- more, the definitions of day and night vary between different studies. All these factors may partly account for the discordant results.

It has been clearly shown that treatment with synthetic long acting vasopressin (DDAVP) has a beneficial effect on enuresis, which supports the theory of a deficient action of AVP and/or other factors leading to an increase in urine

ACTA PZDIATR 86 (1997) Plasma vasopressin in nocturnal enuresis 389

0 ' I R AVP IR AVP NR AVP R U-osm IR U-osm NR U-osm R U-prod 1R U-prod NR U-prod

Fig. 8. Dayhight ratios of AVP secretion, urine production and urine osmolality in enuretics before treatment with oral desmopressin (DDAVP). R, responders, IR, intermediate responders, NR, non-responders. Box graph, indicating loth, 25th, 5Oth, 75th and 90th percentiles.

output which exceeds the functional bladder capacity of the enuretic child. The key question which, however, has not been answered is why the bladder empties during sleep, and the question of casuality is also still not solved. As has been suggested by Watanabe and his colleagues (lo), bladder distension might in itself be a stimulus for AVP secretion. Since the enuretic child voids as soon as his bladder is full while the non-enuretic child keeps his or her bladder dis- tended for a longer time before urinating, the nightly stimulus for AVP release might be greater in the latter than in the former. Thus, a lower nocturnal AVP level in the enuretic child could be the result of PNE as well as the cause.

It is evident that some children with PNE have a noc- turnal polyuria (5, 15), and it has been shown that the beneficial effect of DDAVP for those children who respond to the treatment is coupled to a reduction of this polyuria (17). Recently Wood et al. (1 8) found a tendency towards lower AVP-values in non-responders. In our group of adolescents with severe PNE the response to DDAVP- treatment was satisfactory and in a previously published study (19) we found that 50% were full responders to oral DDAVP-treatment. We were, however, unable to find any difference in AVP-secretion, urine osmolality or nocturnal urinary production between the responders and the non- responders. It is, however, a small number of patients in each group that may account for these results.

In the present study we did not find any difference in day/ night ratio of AVP-secretion between enuretics and controls. However, we found a significant increase in AVP-levels between day and night in both groups and thus, no difference between the two groups in the diurnal variation of AVP could be detected. When the 08 : 00 h AVP values were counted as day, as in the study by Rittig et d. (5 ) , the comparisons yielded the same statistical differences as when these values were omitted.

Since our patient and control groups are quite small, although larger than in most of the other studies (3-5),

power calculations were carried out. According to these, the probability of finding statistically significant differ- ences of mean values within groups provided the true difference is 30% of the larger value is approximately 75% and 90% for controls and enuretics, respectively. For the unpaired comparisons between enuretics and con- trols, however, the corresponding probability is only around 40%. Thus, small differences between enuretic children and healthy controls cannot be excluded by this study.

We found a signifcant difference between enuretics and controls in the change in urine osmolality between day and night ( p 5 0.001). Polyuria and a decreased urine osmol- ality may be present during the first hours of sleep but a difference from normal controls in the total nocturnal urine production may therefore not be obvious in some cases. All the adolescents in the study were on a standardized liquid intake and the adolescents with enuresis did not increase their urinary production over night compared to normal controls.

Most, if not all, of the hormones of the adenohypophysis are released in a pulsatile fashion (20), and observations made as early as 1975 indicated the same mode of secretion for AVP as well. In a pioneer study Rubin and his col- leagues (21) showed AVP to be episodically secreted nightly in adult men. Hormone levels were measured every 20 min and values were found to vary greatly between individuals and also between samples from the same individual. In the study performed by Wood et al. (1 8) the same mode of AVP secretion was found in enuretic children.

In all these studies on AVP secretion in enuretics the AVP comparisons are based on the results from point hormone samples with time intervals no shorter than 4 h, and in some of the studies (8,9, 15), AVP values from only a single blood sample were used for comparison . It is quite clear that measurements of AVP-secretion by point hor- mone samples with even 2-h intervals might be too crude to

ACTA PlEDIATR 86 (1997) 390 G Lackgreii er d.

demonstrate any differences between day and night even in normal controls.

Nocturnal polyuria might still be one of the causes of enuresis, as indicated by the differences in urine osmolality between enuretics and controls. A difference in AVP secre- tion between enuretics and non-enuretics might still be the case, but proving or disproving this hypothesis would call for continuous monitoring of AVP. and other factors may certainly play a role in the aetiology of primary nocturnal enuresis. i l ~ . k ~ ~ ~ ~ ~ . ( e ~ ~ e m m r s . - - T h e investigators are greatly indebted to Ms Bent Lilja for expert technical assistance. and to the research department at Ferring AB Malmo. Sweden, for help with the AVP analyses.This project has bcen supported by grants from Her Royal Highness the Crown Princess Louise’\ Fund for Scientific Research and Ollie and Elof Eriksson’s Foundation for Scientific Research.

References 1, Ferguswn DM. €lomood LJ. Nocturnal enuresis and behavioral

problems in adolescence: a 15-year longitudinal study. Pediatrics 1994: 94: 663-8

2. Klackenberg G. Nocturnal enuresis in a longitudinal perspective. Acta Paediatr Scand 1981: 70: 453-7

3. George CPL, Messerle FH, Genest J, et al. Diurnal variation of plasma vasopressin in man. J Clin Endocrinol Metab 1975: 41: 332

4. Puri VN. Urinary levels of antidiuretic hormone in nocturnal enuresis. Ind Pediatr J Ind Acad Pediatr 1980; xvii: 675-6

5. Rittig S, Knudsen UB, N0rgaard JP, Pedersen EB. Djurhuus JC. Abnormal Diurnal Rhythm of Plasma Vasopressin and Urinary Output in Patients with Enuresis. Am J Physiol 1989; 256: 6M-71

6. Nsrgaard JP, Rittig S, Djurhuus JC. Nocturnal Enuresis: an Approach to Treatment Based on Pathogenesis. J Pediatr 1989:114 Suppl: 705- 10

7. Obara K, Takahashi H, Takeda M, Sato S. The effect of desmopressin (DDAVP) in patients complaining of nocturia. Nippon Hinyokika Gakkai Zasshi 1993: 84 (12): 2131-6

8. Shu SG, Lii YP, Chi CS. The efficacy of intranasal DDAVP therapy in children with nocturnal enuresis. Chung Hua I Hsueh Tsa Chih (Taipei) 1993: 52 (6): 368-71

9. Wille S, Aili M. Hanis A, et al. Plasma and urinary levels of vasopressin in enuretic and nonenuretic children. Scand J Urol Nephrol 1994; 28: 119-22

10. Watanabe H. Kawauchi A. Nocturnal Enuresis: Sockdl Aspects and Treatment Perspectives in Japan. Scand J Urol Nephrol 1994; Suppl 163: 29-38

11. Rooke P, Baylis PH. A new Sensitive Radioimmunoassay for plasma arginine vasopression. J lmmunoassay 1982; 3 (2): 115-31

13. Mills JN. Diurnal rhythm in urine flow. J Physiol 1951; 113: 528 13. Poulton EM, Hinden E. Classification of enuresis. Arch Dis Child

1953; 28: 392 14. Evans JH. Meadow SR. Desmopressin for bed wetting: length ol’

treatment, vasopressin secretion, and response. Arch Dis Child 1992:

15. Steffens J, Netzer M, Isenberg E, Alloussi S, Ziegler M. Vasopressin deficiency in primary nocturnal enuresis. Eur Urol 1993; 24: 366-70

16. Kawauchi A, Watanabe H, Kitamori T, Imada N, Ohne T. The possibility of centripetal stimulation from the urinary bladder for vasopressin excretion. J Kyoto Pref Univ Med 1993; 102: 747-52

17. Nmgaard JP, Jonler M, Rittig S, Djurhuus JC. A pharmacodynamic study of desmopressin in paticnts with nocturnal cnurcsis. 3 Urol

18. Wood CM, Butler RJ, Penney MD, Holland PC. Pulsatile release of arginine vasopressin (AVP) and it’s effect on response to desmopressin in enuresis. Scand J Urol Nephrol 1994 Suppl 163: 93-101

19. Stenberg A, Lackgren G. Desmopressin Tablets in the Treatment of Severe Nocturnal Enuresis in Adolescents. Pediatrics 1993; 94 (6): 841-6

20. Van Cauter E. Diurnal and ultradian rhythms in human endocrine function: a minireview. Horn Res 1990; 34: 45-53

21. Rubin RT, Poland RE, Revessoud F, Gouin PR, Tower BB. Anti- diuretic hormone: episodic nocturnal secretion in adult men. Endocr Res Comm 1975: 2: 459-69

67 (2): 184-8

1995; 153 (6): 1984-6

Received Feb. 15, 1996. Accepted in revised form Dec. 4, 1996