- Email: [email protected]
Effects of anal invasive treatment and incontinence on mental health and psychosocial functioning of adolescents with Hirschsprung's disease and low anorectal anomalies
Effects
of Anal Invasive Treatment and Incontinence on Mental Health and Psychosocial Functioning of Adolescents With Hirschsprung’s Disease and Low Anorectal Anomalies By Trond
H. Diseth,
Thore
Egeland,
and
Ragnhild
Emblem
Oslo, Norway
Background/Purpose: Recent studies of adolescents with Hirschsprung’s disease (HD) and low anorectal anomalies (LARA) showed persistent impairment of fecal control in both groups, but very different mental and psychosocial outcome. /Vethock To explore possible reasons for these differences, 19 adolescen% with HD (aged 10 to 20 years; median, 161 operated on by the Duhamel technique were compared with 17 adolescents with LARA (aged 12 to 20 years; median, 15). The 36 adolescents were assessed for treatment procedures, bowel function, and mental and psychosocial outcome by data collected from medical records, physical examination, semistructured interview, and standardized questionnaires. The parents of 30 adolescents were also interviewed and completed questionnaires. Resuks: Duration of anal invasive treatment procedure and current bowel function were associated with mental and psychosocial outcome. The treatment variable, duration of anal dilation, was the most significant predictor of the adolescents’s mental health (Rz = .41, P< .OlJ, whereas
YMF’TOMS OF FECAL incontinence are shameful and embarrassing and may lead to secretive and unsocial behavior. Hirschsprung’s disease (HD) and low anorectal anomalies (LARA) represent congenital intestinal malformations that may be associated with incontinence. In recent studies of adolescents with HD and LARA,1-3 we showed persistent impairment of fecal continence function in both groups but very different mental and psychosocial outcome. HD adolescents had no more psychopathology or psychosocial dysfunction than healthy peers,l whereas adolescents with LARA had a high frequency of mental and psychosocial problems and family difficulties.z~3 To explore possible reasons for the reported differences, we compare adolescents with HD and LARA with regard to bowel function. medical procedures, and family/ parental factors, and we evaluate the relationships be-
S
From the Division of Child and Adolescent Psychiatry, Hospital; Section of Medical Statistics, UnivemiQ Deparment of Pediatric Surgery, The National Hospital, Address reprint requests to l?H. Diseth, MD, Division Adolescent Psychiatry. The National Hospital, 0027 Oslo, Copyright 0 1998 by Wl3. Saunders Company
0022.3468/98/3303-0012$03.00/0 468
The National of Oslo; and Oslo, Norway. of Chdd und Norway.
chronic family difficulties and parental warmth together with the current bowel function variables, fecal and flatus continence function, best explained the variance in psychosocial outcome (R* = .77, P< .OOOl). Thus, the differences in treatment procedures and continence function between the HD and LARA groups may partially explain differences in mental and psychosocial outcome. Conc/us~ons: These findings suggest that anal dilatation and continence dysfunction may have negative impact on mental health and psychosocial functioning. IndicaCons for and ways of performing the procedure of dilation, and the treatment of persistent incontinence problems, are questioned. J Pediatr Surg 33:468-475. Copyright cz 1998 by W.B. Saunders Company. INDEX WORDS: Congenital intestinal malformations, sive anal procedure, continence function, mental psychosocial functioning.
invahealth,
tween these variables and mental and psychosocial outcome. We address both the medical and psychosocial aspects of incontinence problems and invasive anal procedures as they were reported by the adolescents and their parents. MATERIALS
AND
METHODS
Subjects The HD group consisted of 19 patients (13 boys, 6 girls), medirm age. 16 years (range, 10 to 20 years), operated on by the Duhamel technique.4 The parents of 13 adolescents participated. The LARA group consisted of 17 patients (11 boys, 6 girls). median age, 15 (range, 12 to 20 years), with anocutaneous Cn = 15) or anovestibular fistula (n = 2). All their parents participated. Sampling procedures and the representativity of the samples have been presentecllm3 and were found satisfactory. Sociodemographic characteristics. hospital admissions, and after-treatment procedures in HD aud LARA patients xe presented in Table 1. We found no significant differences between the two patient groups in gender, age, or sociodemographic characteristics. The diagnosis of HD was made significantly later than LARA (z = -2.25, P = .03, terminology explained in statistics section). The HD adolescents had expetienced significantly longer hospital admissIons than the LARA patients (z = ~3.37, P < ,001). Because of a residual septum, an extra procedure with crushing clamps applied through the anus with the child tied in the bed for 5 days, was performed in seven (37%) HD patients at an average of 1.5 years of age. Because of severe constipation, three (16%) .kxma/
ofPed!atric
Sw-gerK
Vol33,
No 3 (March),
1998: pp 468-475
EFFECTS
OF ANAL
Table
DILATATIONS
I. Sample
After-Treatment
AND
Characteristics, Procedures
INCONTINENCE
Hospital
in Adolescents
469
Admissions, With
and
HD and LARA Patients
Age tyr),
median
(range)
Boys, n t%) Family composrtron,
educatron
(yr), median
15 (12-20)
13 (68)
II
17 (89)
13 (761
2 (II)
4 (24)
(65)
13 (9-15)
10 (9-19)
12 (9-19)
12 (9-16)
n t%)
Rural Age of diagnoses
(d), median
(range)
Hospital admissions, medran (range) Number of hospital admissions Longest hospital admission Last hospital
admission
Type of after-treatment, Extra crushing clamps Daily dilatation Weekly lavage *f<
16 (10-20)
(range)
Mother Father Community, Urban
LARA n = 17
n 1%)
Both parents Srngle parent Parental
HDn=l9
(age in yr.1
9 (47)
8 (47)
10 (531
9 (53)
75 (I-1080)
2 (l-780)*
4(2-II) 19.5 (11-150) 2.5 (l-16)
4 (Z-19) 8 (3.45)t 1.5 ~0.1-10~
n (%) 7 (37) 0 (01 3 (16)
.05.
tf
< .Ol
*f
< ,001.
HD patients and 10 (59%) of the LARA patients (Xt = 5.46, df 1, P = ,02) had weekly lavages instituted up to an average of 3 years of age. Ninety-four percent of the LARA patients were dilatated with a metalhc dilatator inserted into the anus one to two times daily by their parents up to median 2 years of age (range. 0 to 3 years) None of the HD or LARA patients had permanent colostomies.
Methods Anorectal function was assessed by clinical and physiological examination. Fecal continence function was graded according to Wingspread classnications rating four climcal states of continence: (1) clean, (2) stammg, (3) intermittent fecal soiling, and (4) constant soilmg. Flatus continence function was assessed by asking the patients to indicate their control of flatus on a 10~cm visual analogue scale (VAS) 1,3Anal canal manometry was performed with a 2.mm microtransducer measuring anal canal restmg pressure and increase in pressure during maximal squeeze.z.3 The assessment of mental health and psychosocial functioning of the adolescents is multimodal and has been described in detail elsewhere.1-3 The reliability and validity of the measures are well estab1ished.l Four kinds of outcome measures are used for mental health and psychosocial functionmg. One outcome measure for the adolescents’ mental health was based on a semistructured interview of the adolescents (Child Assessment Schedule [CAS]) (Hodges K, unpublished manuscript) providing scores of symptom complexes (CAS symptom score) analogous to psychiatric diagnoses in the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R). Two outcome measures for behavioral and emotional problems based on standardized questionnaires (Youth Self-Report [YSR16 and Child Behavior Checklist [CBCL]‘) providing combined total behavior and emotional problem scores were also included. The outcome measure for psychosocial functioning of the adolescents was based on the Children’s Global Assessment Scale (CGAS).8 Psychosocial functioning describes the capacity to function well in everyday life. at home, at school and with peers, and to cope with those stresses that are commonly encountered without undue
distress. The CGAS score was based on information from the child interview and ranges from 100 (excellent function) to 0 (most severe malfunction). Psychosocial problems of climcal signmcance are reflected by a score equal or less than 71 Parents were interviewed separately using a slightly modified version of the standardized, semistmctured interview Parental Account of Children’s Symptoms (PACSJ9 which includes ratings of maternal warmth and criticism toward the index child. A global assessment of Chronic Family Difficulties (CFD)lo.il was based on all parental information available about the previous and present family situation The parent interviews were supplemented with questions covering emotional and practical aspects of treatment procedures and medical follow-up. The adolescent and parent interviews were videotaped. The assessments of mterrater reliability (Kappa or intraclass correlation) of the psychiatric and psychosocial assessments were found satisfactory.1s3
Statistics Because of nonnormal distributions, results are reported as medians (ranges) unless otherwise stated. Nonparametric statistics were primarly used.rs For assessment of differences between independent groups. a X2 test with Yates correction (xc) or Fisher’s Exact test for small samples. was used for categorical variables and the Mann-Whitney U test (2) was used for continuous variables. Dependency between sets of variables are expressed through the nonparametric Spearman’s rank correlation coefficient (rs). Multiple regression analyses with a stepwise procedure were conducted to determine independent somatic and family/parental predictors contributing significantly to the explained variance of the adolescents. mental health (CAS. YSR. CBCL) and psychosocial function (CGAS).l? Careful checks of model assumptions Including mvestrgations of residual plots did not show any violation of the assumptions. A two-tailed P value less than .05 was considered statistically significant. Because large numbers of tests are being performed, most of the emphasis should be directed toward the results with low P values. However. the sample size is small as a consequence of the nature of the study. therefore, there is a limit to how small the P values can become. The potential problems related to the retrospective design and small samples are discussed later. All calculations were performed using the SPSS statistics program for WINDOWS. Informed wrttten consent was obtained from both the adolescents and the parents. The study was approved by the Regional Ethics Committee for Medical Research.
RESULTS
Bowel Function Fecal and flatus continence function; constipation, and anal canal pressures in the patient groups are given in Table 2. Between the incontinent patients in the HD and LARA groups (n = 6 v n = IO), we found no significant difference in severity of fecal incontinence (Fisher’s Exact test, P = .50). In the LARA group flatus continence function was significantly worse ($ = 7.09, df 1, P = .Ol). Although the fecal continence function of the LARA adolescents was significantly negatively correlated to squeeze pressure (rS= -0.55, P = .OOl) and flatus continence function was significantly negatively correlated to resting pressure (rS = -.57, P = .02), no significant correlations between continence function and the anal canal pressures were found in the HD group. Fecal
470
DISETH,
Table 2. Somatic, Mental, and Psychosocial Outcome Adolescents With HD and LARA
Grade Grade
function,
13 (68) function,
n 1%) 16 (84)
VAS 4-10 Median (range) Constipation, n (%I Anorectal pressures Resting pressure, Squeeze prassure, DSM-III-R
km
median (range) median (range)
diagnosis,
n (%I
compulsive
OpposItIonal CAS symptom
CBCL raw scores, median Total behavior score
Internalizing Externalizing
median score
3 (16)
8 (47)
70 (40-120) IO0 (50-200)
65 (35-90) 78 (20-250)
3 (16)
9 153j* 4 1241
I (5) 3 (16)
2 (12)
3 (O-33)
3 (18) 8 (0.35)t
(range)
1
(6)
(range) 9 (l-32)
score score
VSR raw scores, Total behavior
6 (35) 11 (65)t 5 (o-7)*
disorder
disorder score, median
lnternalizlng Externalizing
3 (161 0 (O-5) H*O)
Dysthymic disorder Major depresswe syndrome Overanxious disorder Obsessive
7 (41) 10 (591
6 (32)
Flatus continence VAS O-3
15 (2-95)
3 (O-III 2 (O-11)
5 (l-26) 3 (O-38)
24 (7-93)
27 (3-86)
(range)
score score
CGAS score, median (range) CFD scora, mean (SD)
86 (54-96) 1.9 (1.4)
8 (O-23) 5 (O-29) 72 (51.94)* 2.6 (1.7)
Parental Parental
0.4CO.7) 0.3 (0.6)
0.6 (0.8) 0.7 (0.8)
warmth, cntlcism,
6 (O-42) 8 (2-15)
mean (SD! mean (SD)
NOTE, Fecal continence Wingspread classification. on a IO-cm Visual Analogue Abbreviation: tal Disorders, *F-c .05. tP< *P<
DSM-III-R,
function was graded accordmg to the Flatus continence function was indicated Scale (VAS). Diagnostic
EMBLEM
age were significantly correlated to mental health or psychosocial functioning. Parental and Family Dificulties
n c%)
I 2-4
AND
LARA (n = 17)
HD (n = 13)
Fecal continence
for
EGELAND,
and Statistical
Manual
of Men-
3rd ed. Revised.
.Ol. ,001.
continence function was significantly related to age in the HD group (rs = -.58, P = .Ol) with better function in older age, and to gender in the LARA group (z = -2.21, P = .03) with better function in boys. Mental Health and Psychosocial Functioning There was a significant difference between the HD and LARA patient groups in prevalence of psychiatric diagnoses ($ = 4.03, df 1, P = .04) and CAS symptom score (z = -2.74, P < .Ol, Table 2). The LARA patients expressed significantly more fears, anxieties, and worries and more problems with mood, friends, and social activities. We found no differences in CAS scores between the continent and incontinent groups. There was also a significant difference in CGAS score (z = -2.16, P = .03) between the HD group and the LARA group (Table 2). The continent and incontinent patients differed in CGAS score (z = -2.15, P = .03). Neither sex nor
The economic, housing, and educational situations, and the physical and mental health of the family members were generally good in both the HD and LARA patients groups with no statistically differences between the groups (Table 1). There were no statistically significant differences in the family/parental variables between the two groups (Table 2) or between the continent and incontinent patients. In the HD families, chronic family difficulties were mainly minor somatic and mental health problems in the parents or the siblings and minor intrafamilial conflicts. In the LARA families, problems were mainly caused by inadequate network support, disagreement on childcare, and marital conflicts. Sixty-five percent of the LARA families and 38% of the HD families had experienced an early mother-child separation when the child was sent away to the National Hospital for further treatment. The mother often remained in the local hospital without further information about the congenital anomaly. Fifty-three percent of the LARA parents and 15% of the HD parents recalled the early mother-child separation and the hospital admissions during the first years, as traumatic. When asked what they had found most difficult when having a child with a congenital intestinal malformation, 56% of the LARA parents described how bad they felt having to inflict pain on the child by anal dilatations, and/or having to force the child to endure the intrusive procedures of dilatations, lavages, and enemas day in and day out. Thirty-one percent of the LARA parents recollected a power struggle during these procedures. They stated that this had influenced their relationship with the child in a negative way. During long periods of childhood their children shied away from physical and emotional contact. In contrast, 23% of the HD parents recalled problems related to the intrusive procedures of crnshing clamps and lavage. Fifty-nine percent of the LARA parents and only 23% of the HD parents reported that the child’s malformation had a negative influence on their marital relationship and family life. In the LARA families the anomaly often became a taboo and was difficult to talk about. Fifty-three percent of the LARA parents recalled that they became isolated from their family, friends, and social activities, whereas 85% of the HD parents experienced support from family and friends after having a child with HD. At the time of the interview, 53% of the LARA parents and 23% of the HD parents still expressed worries about the present somatic and psychosocial situation of their adolescents, and 35% of the LARA parents expressed
EFFECTS
OF ANAL
DILATATIONS
AND
INCONTINENCE
471
worries about the future of the adolescents concerning their adolescents’ possibilities for sexual relationships, marriage, or for becoming a parent.
patients was correlated to several mental health (CAS, YSR) scores (CAS areas scores for fear [rs = 57, P = ,021 and reality testing [rs = .51, P = .04], YSR total score [rs = .59, p = ,011, internalizing and externalizing scores, and several YSR subscales; withdrawn, anxiousldepressed, thought problems, delinquent behavior, aggressive behavior). The duration of weekly lavages showed no such correlations. Number and length of the hospital stays were not correlated to any variables of mental health and psychosocial functioning. To further investigate the relative importance of the independent treatment, somatic and family/parental variables in predicting psychosocial (CGAS) and mental (CA& YSR, CBCL) outcome, stepwise (forward) multiple regression analyses were conducted. With CGAS (psychosocial functioning) as the dependent variable, the following variables were entered: somatic correlates; diagnostic group, fecal continence function, flatus continence function, and family/parental correlates; chronic family difficulties (CFD), parental warmth, and parental criticism, as explanatory variables. Because duration of anal dilatation was not significantly associated to CGAS either in the total group or in the diagnostic groups, we did not enter this variable into the model. Chronic family difficulties was the strongest predictor of CGA& explaining 56% of the variance (B = -6.9, P < .OOOl). With parental warmth added, the explained variance increased to 67% (l3 = -03, P = .OOZ).Flatus continence function added further 7% units (B = -1.53, P = .006), and fecal continence function added at last 3% units (l3 = -4.57, P = ‘04) to the explained variance of CGA,?. The diagnostic group and the other parental variables did not significantly contribute to the explained variance of psychosocial functioning of the adolescents. With CAS symptom score (mental health) as the
Associations Between Somatic Variables, Parental/Family Variables, and Mental and Psychosocial Outcome The initial bivariate analyses indicated significant associations between diagnostic groups, duration of anal dilatation, bowel function, parental/family variables, and mental health and psychosocial adjustment (Table 3). When splitting the patient group into the two diagnostic groups, fecal incontinence was significantly correlated with poorer psychosocial functioning (+rs = -58, P = .Ol) and more parental criticism (rs = .63. P = .03) in the HD patients, whereas in the LARA group a number of significant correlations were found between the flatus continence function and several variables of mental health (CAS, YSR) and psychosocial functioning (CAS symptom score [rs = .73, P < .OOl], several CAS content areas scores; worries/anxieties, family, mood, expression of anger, YSR total score [rs = .61, P = .Ol], extemalizing and internalizing scores, several YSR subscales; anxious/depressed, social problems, aggressive behavior, and psychosocial functioning [ +rs = - .53, P = ,031). Examination of the relation between treatment procedures and outcome variables showed that the anal invasive procedures of extra crushing clamps and weekly lavage in the 10 HD patients were not significantly correlated to CAS or CGAS scores. But there was a significant difference in CBCL externalizing scores (z = -2.46, P = .Ol) and a trend toward difference with regard to YSR externalizing scores (z = - 1.65, P = .09) with higher externalizing scores in the group with extra invasive procedures. In contrast, the duration of anal dilatation in the LARA Table 3. Spearman
Rank-Sum Treatment
Diagnostic group: HD:LARA Duration of dilatationt
Correlation
Coefficients
Bowel Function
Describing
DUtmOtl of Dilatation
Fecal cant
Flatus ccmt.
CAS Symptom
.72ll
.31 21
5711 .35*
.37* .62§
.33*
29
Fecal incontinence Flatus incontinence
.67ll
YSR total CBCL total CGAS
§P< .Ol. IIPC ,001.
With
CGAS
.cl4
.30 .33
-.36$ -.35
.I2 .I6
.21
-.39* ~.48§
.I1 .I7
p.7411
.35 .26
.83ll .25 .44§
.32 .44*
HD and LARA
Farmiy Psychosocnl
CBCL Total
34
warmth critvzism
*A total set of family variables were only available Korrelation coefficients only for those who were tP < .05.
in Adolescents Outcome
YSR Total
w
CAS symptom
Parental Parental
Correlations
Mental and Psychosocv.4
-.5711 -.6Ol\
Parental Warmth
.53§ - .76\j
Parental Critusm
CFD
.34
.24
.06 .30
.05 19
.38* .31
.29 .45§
.31 .44* -.58/j
.58ll
.57§
for 30 adolescents. anal dliatated (n = 161.
Variables*
.34 -.781/ .76ll .52§
DISETH,
472
dependent variable, diagnostic group, duration of dilatation, fecal continence function, flatus continence function, CFD, parental warmth, and parental criticism were entered as explanatory variables. Duration of anal dilatation was the only significant predictor of CAS, explaining 28% of the variance of CAS in the whole group (l3 = 12.97, P = .008). Because anal dilatation was a treatment procedure primarly used in the LARA group. further stepwise multiple regression analyses were conducted to examine to what extent treatment and somatic variables contributed significantly to mental outcome in the LARA group. The mental health scores CAS symptom score, YSR total score, and CBCL total score were dependent variables entering the somatic correlates duration of anal dilatation, fecal continence, and flatus continence, into the model. The most significant independent predictor of the adolescents’ mental outcome as measured by the CAS symptom score was duration of anal dilatation, which explained 41% of the variance (B = 15.27, p = .005). With flatus continence function added, the explained variance increased to 58% (B = 2.10, P = .03). Entering the family/parental variables into the model did not significantly increase the explained variance of CAS . Duration of anal dilation was also the strongest independent predictor of mental health as measured by a youth self report; YSR total problem score (R2 = .47,l3 = 27.02, P = .002). Also for YSR total score, with flatus continence function added, the explained variance increased (R2 = .61, l3 = 3.14, P = .04). Duration of anal dilatation was the only significant predictor of mental health as measured by the questionnaire of another informant, the parents; CBCL total problem score (R2 = .24,B = 25.29, P = .04). For all dependent variables, the same significant predictors were selected regardless of whether log transformations were performed. However, the P values refer to the original scale. DISCUSSION
When analyzing possible reasons for the diverse outcome of mental health and psychosocial outcome in HD and LARA patients, we found that although both the HD and LARA groups experienced persistent impairment of fecal continence, the groups differed with regard to treatment procedures. such as type and length of invasive anal treatment regimes; in parent-child experiences, eg, age at diagnosis, postpartum mother-child separation, length of hospital stays, and painful experiences associated with treatment procedures; and in flatus continence function at the follow-up. There were no statistically significant differences in family or parental factors between the two groups. The duration of anal invasive treatment procedure and current bowel function were associated with mental and
EGELAND,
AND
EMBLEM
psychosocial outcome. The early treatment variable, duration of anal dilatation, was the most significant predictor of the adolescents’ mental health, whereas chronic family difficulties and parental warmth together with the current bowel function variables, fecal and flatus continence function, best explained the variance in psychosocial outcome at follow-up. The retrospective design makes it difficult to draw conclusions about cause and effect, and the relatively small number of patients in each group limits the generalizability of findings. However, statistical power considerations before the implementation of the study showed that the study was worthwhile, ie, chances were that interesting clinical differences would be detected with sufficient power. Furthermore, because of the centralization of treatment of these low-frequency congenital anomalies, the present sample is representive of the Norwegian patient population. The two patient groups had been treated at the same hospital, were assessed by the same methods, and were comparable with regard to gender, age, and sociodemographic characteristics. Our findings suggest some risk factors that may explain the difference in mental and psychosocial outcome for patients with HD and LARA: 1. The diagnosis of HD was generally made days to weeks after birth, and most parents experienced a healthy child in the neonatal period and had time to establish the important first relationship. When hospitalized, the mother could accompany the HD child. In the LARA group, the anomaly was usually noted soon after birth and immediately created parental worries and anxieties. This was often followed by a postpartum mother-child separation when the newborns were transferred to the National Hospital and were subjected operations and invasive anal treatment procedures. Parental reports based on the interviews indicate that this may have influenced the long-term parent-child relationship. 2. The two groups also differed in duration of hospital stays but not in number. Repeated hospitalizations in early childhood have been found to be associated with high rates of both short-term and long-term behavioral and emotional prob1ems.l“-16 Patients in our sample were born in the 1970s and early 1980s and at that time rooming-in facilities and round-theclock visiting for parents and siblings were not routine. However, neither number or duration of hospital stays nor age at diagnosis were correlated to long-term effects on mental health or psychosocial functioning of these adolescents. 3. Type and duration of invasive anorectal operation techniques and type of after-treatment regimes also differed between the two groups. Children with HD operated on at the median age of 7 months were
EFFECTS
OF ANAL
DILATATIONS
AND
INCONTINENCE
kept tied in the bed for 5 days after the operation with crushing clamps in the anus. In 37% of the HD children, an extra procedure with crushing clamps was performed at an average of 1.5 years of age. This “tying up” of the child was reported by their parents to be unpleasant for the child. In children with LARA, anal dilatation has been advocated to avoid anal stricture.17 Generally, no analgesia is used in this procedure. Anal dilatations were performed once to twice daily by the parents up to median 2 years of age (range, O-3 years). The parents reported that this had a traumatic impact on the child and affected the parent-child relationship creating a power struggle and parental guilt by inflicting pain on the child. Both the procedure and the child’s reaction to it, may have influenced the relationship between the child and the parents. This may explain the significant associations found in the present study between duration of anal dilatation and the adolescents’ present mental outcome (CAS, YSR, CBCL), and parental factors (CID, marital relationship) as reported in both low and high anorectal malformations.3 In the present study, the anal dilatation was correlated to fear, reality testing, withdrawal, anxious/depressed mood, and delinquent or aggressive behavior. The duration of anal dilatation in childhood was the most important predictor of the adolescents’ mental health at the follow-up. The difference in mental outcome in patients with HD and LARA suggests long-term negative implications when anal dilatations are performed for years. 4. Chronic family difficulties was the most important predictor of the adolescents’ psychosocial functioning. The LARA families experienced less network support and reported more disagreement in childcare and/or marital conflicts than the HD families. The LARA families reported problems with handling the child’s fecal soiling, and the anorectal anomaly had become a subject to be avoided. At follow-up we also found more worries about the present and future situation in the LARA group. The cross-sectional design of this part of the study does not allow us to draw conclusions. However, the parents’ experiences of long-lasting emotional and practical strain related to the child’s congenital disease may create risk for vicious circles with early strong ambivialent feelings in the mother creating tensions in the marital relationship. Such interactive effects are indicated by Ludman’s 3-year prospective study of full-term neonates undergoing surgery.l* The parents of sick babies were under greater general stress and had significantly more psychological problems than controls. This indi-
473
cates that a combination of stress factors rather than just the anxiety associated with the birth of a sick baby may contribute to the increased incidence of maternal depression. 5. The associations between less parental warmth/ more criticism and poorer psychosocial outcome of the children, are in line with previous findings.‘9-z1 We found high parental warmth to be one important predictor of the adolescents’ psychosocial functioning. This finding is in accordance with the documented protective nature of parental warmth. In this sample, many factors may have contributed to the parent’s Ievel of warmth and criticism. Chronic illness may lead to heightened maternal empathy toward the child,2z and strongly protective and emotionally involved parenting may be psychologically adaptive for children under stress. In the HD group, findings suggest that this may have contributed to counteract risk factors such as early separation. hospital admissions. and invasive treatment procedures. 6. The association between severity of physical impairment (fecal and flatus incontinence) and psychosocial outcome also corresponds to findings in other studieszz We found flatus incontinence to be one of four important predictors of CGAS and the two diagnostic groups differed in flatus continence function. Uncertainty about flatus control created many additional worries and preoccupations for the adolescents. To hide the unpleasant smell, they reported use of extensive amounts of perfume and/or they smoked cigarettes. Flatus incontinence was significantly correlated to poorer resting pressure, thus emphazising the importance of applying surgical techniques, preserving the internal anal sphincter. Our findings suggest that invasive anal treatment procedures and continence dysfunction may have negative impact on mental health and psychosocial functioning for patients with anorectal malformations. This exploratory study cannot document the mechanisms by which anal dilatation or incontinence affect mental and psychosocial outcome, but the findings raise interesting considerations. Possible consequences of daily dilatations performed for prolonged periods and of chronic fecal continence dysfunction may be addressed on the basis of developmental theories. The 2- to 3-years-of-age period is a critical age for the important developmental tasks of self control and autonomy. The ability of a child to control defecation at the end of this anal period represents psychosocial maturation and is associated with the building of autonomy and pride. Intrusive procedures in the anal region, a sensitive zone in this developmental phase, may be experienced as scary; they evoke protest and also
474
DISETH,
muscle contractions. Thus, both the anal dilatation and the persistent problems with continence may influence the child’s natural development of self-esteem, autonomy, and social activity, factors so important for mental health and psychosocial functioning. Concerning anal dilatation, one of the authorities in the field, Albert0 Pefia statesl’: “The parents are encouraged to perform the dilatations themselves twice a day at home. But if they have difficulties, it can be done at the hospital. The process of dilatation continues until the desired size is reached, then the colostomy may be closed, but the process of dilatation continues but gradually tapering the frequency over months.” The LARA parents in our study had performed daily anal dilatations for 2 to 3 years at home, and the dilatation had become a chronic stressor for both the child and the parents. One may question whether the recommended procedure of anal dilatation is based on empirical or documented effects. Penal’ noted that “the rationale behind dilatations consists of appreciation of the fact that the anus and the rectum are surrounded by muscle structures and therefore remain closed at rest, particularly in cases with good muscles. Thus, if the patient is not dilatated, the anus will tend to heal closed or very narrowly, provoking an anal stricture.” However, the effect of anal dilatation or the significance of the length of dilatations on outcome for patients with anorectal anomalies have not been documented. Our findings indicate that the use of anal dilatation needs to be reevaluated. Less embarrassing and painful procedures and more supportive interventions (eg, use of anesthetics, parent substitutes to perform the procedures, teaching relaxation techniques, and/or to have psychosocial experts work with the child and the parents) should be considered. To achieve optimal quality of life overall, the challenge is to find a compromise between the physically optimal procedures and procedures that are not psychosocially detrimental. In our study, a significant number of the HD and LARA patients had fecal incontinence problems (32% v 59%) and flatus incontinence problems (16% v 65%). In another study,z all adolescents with high anorectal anomalies had persistent fecal dysfunction, but only 19% had a permanent colostomy. One may ask whether too few of the patients with incontinence problems have been presented with the option of alternative procedures at an early age, eg, rectal irrigation procedures, antegrade colonic irrigation through appendixstomy, physiotherapeu-
EGELANDV
AND
EMBLEM
tic or biofeedback training, or permanent colostomy. Offering a permanent colostomy may have been delayed or avoided for fear of potentially serious psychosocial consequences. A colostomy may represent a chronic disability and loss of the hope for a perfect cure, but it may also reduce the practical and emotional strain of soiling and invasive treatment procedures, both for the child and the parents. In a controlled study of psychosocial adjustment to stoma surgery for childhood inflammatory bowel disease, Lask et alZ3 found no evidence that stoma surgery in childhood is psychologically harmful for psychosocial adjustment, self esteem, or quality of life. They concluded that as long as children are well prepared and followed-up carefully and sensitively, then stoma surgery should not be deferred for fear of adverse psychological consequences. However, although imperious not controlled defecation is an important complaint also in inflammatory bowel disease, the two patient groups are not quite comparable, and this may limit the generalizability of the findings in Lask’s study. This study of adolescents with congenital intestinal malformations demonstrates that the anatomic defect and its operative treatment is only one determinant for the long-term outcome. The duration of invasive procedures, the continence dysfunction. and the family/parental functioning, all seem to be associated with mental and psychosocial outcome of the adolescents. Performing invasive treatment procedures may have physical and mental implications, and findings suggest that the indications for and ways of performing these procedures, need to be reassessed, addressing both the medical and psychosocial aspects and outcome. Persistent incontinence into adolescence represents an outcome with negative psychosocial impact, and a colostomy may represent an alternative procedure to attain fecal control. The findings that chronic family difficulties and parental strain and emotional problems expressed as less warmth and more criticism, have an adverse effect on the adolescents’ psychosocial outcome, should be taken into consideration early in treatment, as well as the inclusion of psychosocial experts in the health care teams3 that treat patients with congenital intestinal malformations and their families. ACKNOWLEDGMENTS The authors thank Professor Inger Helene Vandvik for scientific advice, SonJa Heyerdahl, MD. PhD, for statistical support. and Medinnova for financial support.
REFERENCES 1, Diseth TH, BJcWnland K. Novik T. et al: Bowel function. mental health and psychosocial function in adolescents with Hirschsprung’s disease. Arch Dis Child 76.100-106, 1997 2. Diseth TH. Emblem R: Somatic function, mental health and psychosocial adjustment of adolescents with anorectal anomalies. J Pediatr Surg 31:638-643. 1996
3. Diseth TH, Emblem R, Vandvik IH: Adolescents with anorectal malformations and their famihes. Family Systems Medicine 13:215231,1995 4. Duhamel B: A new operation for the treatment of Hirschsprung’s disease. Arch Dis Child 35:38, 1960 5. Stephens FD, Smith ED: Classification, identification, and assess-
EFFECTS
OF ANAL
DILATATIONS
AND
INCONTINENCE
ment of surgical treatment of anorectal anomalies. Pedratr Surg Int 1:200-2051986 6. Achenbach TM~ Edelbrock CS: Manual for the Youth Self Report and Profile. Burlington. VT, University of Vermont. Department of Psychiatry. 1987 7. Achenbach TM: Manual for the Child Behavior Checklist and 1991 Profile. Burlington, VT, Umversity of Vermont, Department of Psychiatry, 1991 8. Shaffer D, Gould US, Brasic J, et al: Children’s Global Assessment Scale (CGAS). Psychopharm Bull 21:747-748, 1985 9. Taylor EA (ed): The overactive child, in Clinics in Developmental Medtcine 97. Oxford, England, Blackwell Scientific, 1986 10. Vandvik IH, Hoyeraal HM, Fagerhm H: Chronic family drfficulties and stressful life events in recent onset juvenile arthritis. J Rheumatol 16:1088-1092, 1989 11 Bjomstad P, Lindberg H, Spurkland I: Unge hjerter i faresonen. (Young hearts at risk). Oslo. Norway, Tano AS, 1990 12. Altman DG: Practical Statistics for Medical Research. London. England, Chapman and Hall, 1991 13. Kleinbaum D. Kupper L, Muller K: Applied regression analysis and other multivariable methods (2nd ed). Boston, MA, PWS-KENT Publishing Company, 1988 14. Douglas JWB: Early hospital admission and later disturbances of behavior and learning. Dev Med Child Neural 17:456-480. 1975 15. Quinton D, Rutter M: Early hospita1 adnusston and later
475
disturbances of behaviour: An attempted rephcation of Douglas’ findings. Dev Med Child Neurol 18:447-459, 1976 16. Graham P: Prevention. m Rutter M. Taylor E, Hersov L: Child and Adolescent Psychiatry: Modem Approaches. Oxford, England, Blackwell Scientific Publications. 1994 17. Peria A: Atlas of Surgical Management of Anorectal Malformations. New York, NY, Springer-Verlag, 1990 18. Ludman L, Lansdown R, Spitz L: Factors associated with developmental progress of full term neonates who required intensive care. Arch Dis Child 64:333-337, 1989 19. Rutter M. Yule B, Quinton D, et al: Attainment and adjustment in two geographical areas: III. Some factors accounting for area differences. Br J Psychiatr 126:520-533, 1975 20. Quinton D, Rutter M: Family pathology and child psychiatric disorder: A four-year prospective study. in Nicol AR (ed): Longitudinal Studies in Chtld Psychology and Psychtatry. Chichester, Wiley, 1985 21. Richman N, Stevenson J. Graham P: Preschool to school-A behavioural study. London. England, Academic Press. 1982 22. Garralda ME: Chronic physical illness and emotional disorder in childhood Where the brain’s not involved. there may still be problems. Br J Psychiatr 164:8-10, 1994 23. Lask B, Jenkins J. Nabarro L. et al: Psychosocial sequelae of stoma surgery for inflammatory bowel disease m childhood. Gut X1257-1260, 1987
of Anal Invasive Treatment and Incontinence on Mental Health and Psychosocial Functioning of Adolescents With Hirschsprung’s Disease and Low Anorectal Anomalies By Trond
H. Diseth,
Thore
Egeland,
and
Ragnhild
Emblem
Oslo, Norway
Background/Purpose: Recent studies of adolescents with Hirschsprung’s disease (HD) and low anorectal anomalies (LARA) showed persistent impairment of fecal control in both groups, but very different mental and psychosocial outcome. /Vethock To explore possible reasons for these differences, 19 adolescen% with HD (aged 10 to 20 years; median, 161 operated on by the Duhamel technique were compared with 17 adolescents with LARA (aged 12 to 20 years; median, 15). The 36 adolescents were assessed for treatment procedures, bowel function, and mental and psychosocial outcome by data collected from medical records, physical examination, semistructured interview, and standardized questionnaires. The parents of 30 adolescents were also interviewed and completed questionnaires. Resuks: Duration of anal invasive treatment procedure and current bowel function were associated with mental and psychosocial outcome. The treatment variable, duration of anal dilation, was the most significant predictor of the adolescents’s mental health (Rz = .41, P< .OlJ, whereas
YMF’TOMS OF FECAL incontinence are shameful and embarrassing and may lead to secretive and unsocial behavior. Hirschsprung’s disease (HD) and low anorectal anomalies (LARA) represent congenital intestinal malformations that may be associated with incontinence. In recent studies of adolescents with HD and LARA,1-3 we showed persistent impairment of fecal continence function in both groups but very different mental and psychosocial outcome. HD adolescents had no more psychopathology or psychosocial dysfunction than healthy peers,l whereas adolescents with LARA had a high frequency of mental and psychosocial problems and family difficulties.z~3 To explore possible reasons for the reported differences, we compare adolescents with HD and LARA with regard to bowel function. medical procedures, and family/ parental factors, and we evaluate the relationships be-
S
From the Division of Child and Adolescent Psychiatry, Hospital; Section of Medical Statistics, UnivemiQ Deparment of Pediatric Surgery, The National Hospital, Address reprint requests to l?H. Diseth, MD, Division Adolescent Psychiatry. The National Hospital, 0027 Oslo, Copyright 0 1998 by Wl3. Saunders Company
0022.3468/98/3303-0012$03.00/0 468
The National of Oslo; and Oslo, Norway. of Chdd und Norway.
chronic family difficulties and parental warmth together with the current bowel function variables, fecal and flatus continence function, best explained the variance in psychosocial outcome (R* = .77, P< .OOOl). Thus, the differences in treatment procedures and continence function between the HD and LARA groups may partially explain differences in mental and psychosocial outcome. Conc/us~ons: These findings suggest that anal dilatation and continence dysfunction may have negative impact on mental health and psychosocial functioning. IndicaCons for and ways of performing the procedure of dilation, and the treatment of persistent incontinence problems, are questioned. J Pediatr Surg 33:468-475. Copyright cz 1998 by W.B. Saunders Company. INDEX WORDS: Congenital intestinal malformations, sive anal procedure, continence function, mental psychosocial functioning.
invahealth,
tween these variables and mental and psychosocial outcome. We address both the medical and psychosocial aspects of incontinence problems and invasive anal procedures as they were reported by the adolescents and their parents. MATERIALS
AND
METHODS
Subjects The HD group consisted of 19 patients (13 boys, 6 girls), medirm age. 16 years (range, 10 to 20 years), operated on by the Duhamel technique.4 The parents of 13 adolescents participated. The LARA group consisted of 17 patients (11 boys, 6 girls). median age, 15 (range, 12 to 20 years), with anocutaneous Cn = 15) or anovestibular fistula (n = 2). All their parents participated. Sampling procedures and the representativity of the samples have been presentecllm3 and were found satisfactory. Sociodemographic characteristics. hospital admissions, and after-treatment procedures in HD aud LARA patients xe presented in Table 1. We found no significant differences between the two patient groups in gender, age, or sociodemographic characteristics. The diagnosis of HD was made significantly later than LARA (z = -2.25, P = .03, terminology explained in statistics section). The HD adolescents had expetienced significantly longer hospital admissIons than the LARA patients (z = ~3.37, P < ,001). Because of a residual septum, an extra procedure with crushing clamps applied through the anus with the child tied in the bed for 5 days, was performed in seven (37%) HD patients at an average of 1.5 years of age. Because of severe constipation, three (16%) .kxma/
ofPed!atric
Sw-gerK
Vol33,
No 3 (March),
1998: pp 468-475
EFFECTS
OF ANAL
Table
DILATATIONS
I. Sample
After-Treatment
AND
Characteristics, Procedures
INCONTINENCE
Hospital
in Adolescents
469
Admissions, With
and
HD and LARA Patients
Age tyr),
median
(range)
Boys, n t%) Family composrtron,
educatron
(yr), median
15 (12-20)
13 (68)
II
17 (89)
13 (761
2 (II)
4 (24)
(65)
13 (9-15)
10 (9-19)
12 (9-19)
12 (9-16)
n t%)
Rural Age of diagnoses
(d), median
(range)
Hospital admissions, medran (range) Number of hospital admissions Longest hospital admission Last hospital
admission
Type of after-treatment, Extra crushing clamps Daily dilatation Weekly lavage *f<
16 (10-20)
(range)
Mother Father Community, Urban
LARA n = 17
n 1%)
Both parents Srngle parent Parental
HDn=l9
(age in yr.1
9 (47)
8 (47)
10 (531
9 (53)
75 (I-1080)
2 (l-780)*
4(2-II) 19.5 (11-150) 2.5 (l-16)
4 (Z-19) 8 (3.45)t 1.5 ~0.1-10~
n (%) 7 (37) 0 (01 3 (16)
.05.
tf
< .Ol
*f
< ,001.
HD patients and 10 (59%) of the LARA patients (Xt = 5.46, df 1, P = ,02) had weekly lavages instituted up to an average of 3 years of age. Ninety-four percent of the LARA patients were dilatated with a metalhc dilatator inserted into the anus one to two times daily by their parents up to median 2 years of age (range. 0 to 3 years) None of the HD or LARA patients had permanent colostomies.
Methods Anorectal function was assessed by clinical and physiological examination. Fecal continence function was graded according to Wingspread classnications rating four climcal states of continence: (1) clean, (2) stammg, (3) intermittent fecal soiling, and (4) constant soilmg. Flatus continence function was assessed by asking the patients to indicate their control of flatus on a 10~cm visual analogue scale (VAS) 1,3Anal canal manometry was performed with a 2.mm microtransducer measuring anal canal restmg pressure and increase in pressure during maximal squeeze.z.3 The assessment of mental health and psychosocial functioning of the adolescents is multimodal and has been described in detail elsewhere.1-3 The reliability and validity of the measures are well estab1ished.l Four kinds of outcome measures are used for mental health and psychosocial functionmg. One outcome measure for the adolescents’ mental health was based on a semistructured interview of the adolescents (Child Assessment Schedule [CAS]) (Hodges K, unpublished manuscript) providing scores of symptom complexes (CAS symptom score) analogous to psychiatric diagnoses in the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R). Two outcome measures for behavioral and emotional problems based on standardized questionnaires (Youth Self-Report [YSR16 and Child Behavior Checklist [CBCL]‘) providing combined total behavior and emotional problem scores were also included. The outcome measure for psychosocial functioning of the adolescents was based on the Children’s Global Assessment Scale (CGAS).8 Psychosocial functioning describes the capacity to function well in everyday life. at home, at school and with peers, and to cope with those stresses that are commonly encountered without undue
distress. The CGAS score was based on information from the child interview and ranges from 100 (excellent function) to 0 (most severe malfunction). Psychosocial problems of climcal signmcance are reflected by a score equal or less than 71 Parents were interviewed separately using a slightly modified version of the standardized, semistmctured interview Parental Account of Children’s Symptoms (PACSJ9 which includes ratings of maternal warmth and criticism toward the index child. A global assessment of Chronic Family Difficulties (CFD)lo.il was based on all parental information available about the previous and present family situation The parent interviews were supplemented with questions covering emotional and practical aspects of treatment procedures and medical follow-up. The adolescent and parent interviews were videotaped. The assessments of mterrater reliability (Kappa or intraclass correlation) of the psychiatric and psychosocial assessments were found satisfactory.1s3
Statistics Because of nonnormal distributions, results are reported as medians (ranges) unless otherwise stated. Nonparametric statistics were primarly used.rs For assessment of differences between independent groups. a X2 test with Yates correction (xc) or Fisher’s Exact test for small samples. was used for categorical variables and the Mann-Whitney U test (2) was used for continuous variables. Dependency between sets of variables are expressed through the nonparametric Spearman’s rank correlation coefficient (rs). Multiple regression analyses with a stepwise procedure were conducted to determine independent somatic and family/parental predictors contributing significantly to the explained variance of the adolescents. mental health (CAS. YSR. CBCL) and psychosocial function (CGAS).l? Careful checks of model assumptions Including mvestrgations of residual plots did not show any violation of the assumptions. A two-tailed P value less than .05 was considered statistically significant. Because large numbers of tests are being performed, most of the emphasis should be directed toward the results with low P values. However. the sample size is small as a consequence of the nature of the study. therefore, there is a limit to how small the P values can become. The potential problems related to the retrospective design and small samples are discussed later. All calculations were performed using the SPSS statistics program for WINDOWS. Informed wrttten consent was obtained from both the adolescents and the parents. The study was approved by the Regional Ethics Committee for Medical Research.
RESULTS
Bowel Function Fecal and flatus continence function; constipation, and anal canal pressures in the patient groups are given in Table 2. Between the incontinent patients in the HD and LARA groups (n = 6 v n = IO), we found no significant difference in severity of fecal incontinence (Fisher’s Exact test, P = .50). In the LARA group flatus continence function was significantly worse ($ = 7.09, df 1, P = .Ol). Although the fecal continence function of the LARA adolescents was significantly negatively correlated to squeeze pressure (rS= -0.55, P = .OOl) and flatus continence function was significantly negatively correlated to resting pressure (rS = -.57, P = .02), no significant correlations between continence function and the anal canal pressures were found in the HD group. Fecal
470
DISETH,
Table 2. Somatic, Mental, and Psychosocial Outcome Adolescents With HD and LARA
Grade Grade
function,
13 (68) function,
n 1%) 16 (84)
VAS 4-10 Median (range) Constipation, n (%I Anorectal pressures Resting pressure, Squeeze prassure, DSM-III-R
km
median (range) median (range)
diagnosis,
n (%I
compulsive
OpposItIonal CAS symptom
CBCL raw scores, median Total behavior score
Internalizing Externalizing
median score
3 (16)
8 (47)
70 (40-120) IO0 (50-200)
65 (35-90) 78 (20-250)
3 (16)
9 153j* 4 1241
I (5) 3 (16)
2 (12)
3 (O-33)
3 (18) 8 (0.35)t
(range)
1
(6)
(range) 9 (l-32)
score score
VSR raw scores, Total behavior
6 (35) 11 (65)t 5 (o-7)*
disorder
disorder score, median
lnternalizlng Externalizing
3 (161 0 (O-5) H*O)
Dysthymic disorder Major depresswe syndrome Overanxious disorder Obsessive
7 (41) 10 (591
6 (32)
Flatus continence VAS O-3
15 (2-95)
3 (O-III 2 (O-11)
5 (l-26) 3 (O-38)
24 (7-93)
27 (3-86)
(range)
score score
CGAS score, median (range) CFD scora, mean (SD)
86 (54-96) 1.9 (1.4)
8 (O-23) 5 (O-29) 72 (51.94)* 2.6 (1.7)
Parental Parental
0.4CO.7) 0.3 (0.6)
0.6 (0.8) 0.7 (0.8)
warmth, cntlcism,
6 (O-42) 8 (2-15)
mean (SD! mean (SD)
NOTE, Fecal continence Wingspread classification. on a IO-cm Visual Analogue Abbreviation: tal Disorders, *F-c .05. tP< *P<
DSM-III-R,
function was graded accordmg to the Flatus continence function was indicated Scale (VAS). Diagnostic
EMBLEM
age were significantly correlated to mental health or psychosocial functioning. Parental and Family Dificulties
n c%)
I 2-4
AND
LARA (n = 17)
HD (n = 13)
Fecal continence
for
EGELAND,
and Statistical
Manual
of Men-
3rd ed. Revised.
.Ol. ,001.
continence function was significantly related to age in the HD group (rs = -.58, P = .Ol) with better function in older age, and to gender in the LARA group (z = -2.21, P = .03) with better function in boys. Mental Health and Psychosocial Functioning There was a significant difference between the HD and LARA patient groups in prevalence of psychiatric diagnoses ($ = 4.03, df 1, P = .04) and CAS symptom score (z = -2.74, P < .Ol, Table 2). The LARA patients expressed significantly more fears, anxieties, and worries and more problems with mood, friends, and social activities. We found no differences in CAS scores between the continent and incontinent groups. There was also a significant difference in CGAS score (z = -2.16, P = .03) between the HD group and the LARA group (Table 2). The continent and incontinent patients differed in CGAS score (z = -2.15, P = .03). Neither sex nor
The economic, housing, and educational situations, and the physical and mental health of the family members were generally good in both the HD and LARA patients groups with no statistically differences between the groups (Table 1). There were no statistically significant differences in the family/parental variables between the two groups (Table 2) or between the continent and incontinent patients. In the HD families, chronic family difficulties were mainly minor somatic and mental health problems in the parents or the siblings and minor intrafamilial conflicts. In the LARA families, problems were mainly caused by inadequate network support, disagreement on childcare, and marital conflicts. Sixty-five percent of the LARA families and 38% of the HD families had experienced an early mother-child separation when the child was sent away to the National Hospital for further treatment. The mother often remained in the local hospital without further information about the congenital anomaly. Fifty-three percent of the LARA parents and 15% of the HD parents recalled the early mother-child separation and the hospital admissions during the first years, as traumatic. When asked what they had found most difficult when having a child with a congenital intestinal malformation, 56% of the LARA parents described how bad they felt having to inflict pain on the child by anal dilatations, and/or having to force the child to endure the intrusive procedures of dilatations, lavages, and enemas day in and day out. Thirty-one percent of the LARA parents recollected a power struggle during these procedures. They stated that this had influenced their relationship with the child in a negative way. During long periods of childhood their children shied away from physical and emotional contact. In contrast, 23% of the HD parents recalled problems related to the intrusive procedures of crnshing clamps and lavage. Fifty-nine percent of the LARA parents and only 23% of the HD parents reported that the child’s malformation had a negative influence on their marital relationship and family life. In the LARA families the anomaly often became a taboo and was difficult to talk about. Fifty-three percent of the LARA parents recalled that they became isolated from their family, friends, and social activities, whereas 85% of the HD parents experienced support from family and friends after having a child with HD. At the time of the interview, 53% of the LARA parents and 23% of the HD parents still expressed worries about the present somatic and psychosocial situation of their adolescents, and 35% of the LARA parents expressed
EFFECTS
OF ANAL
DILATATIONS
AND
INCONTINENCE
471
worries about the future of the adolescents concerning their adolescents’ possibilities for sexual relationships, marriage, or for becoming a parent.
patients was correlated to several mental health (CAS, YSR) scores (CAS areas scores for fear [rs = 57, P = ,021 and reality testing [rs = .51, P = .04], YSR total score [rs = .59, p = ,011, internalizing and externalizing scores, and several YSR subscales; withdrawn, anxiousldepressed, thought problems, delinquent behavior, aggressive behavior). The duration of weekly lavages showed no such correlations. Number and length of the hospital stays were not correlated to any variables of mental health and psychosocial functioning. To further investigate the relative importance of the independent treatment, somatic and family/parental variables in predicting psychosocial (CGAS) and mental (CA& YSR, CBCL) outcome, stepwise (forward) multiple regression analyses were conducted. With CGAS (psychosocial functioning) as the dependent variable, the following variables were entered: somatic correlates; diagnostic group, fecal continence function, flatus continence function, and family/parental correlates; chronic family difficulties (CFD), parental warmth, and parental criticism, as explanatory variables. Because duration of anal dilatation was not significantly associated to CGAS either in the total group or in the diagnostic groups, we did not enter this variable into the model. Chronic family difficulties was the strongest predictor of CGA& explaining 56% of the variance (B = -6.9, P < .OOOl). With parental warmth added, the explained variance increased to 67% (l3 = -03, P = .OOZ).Flatus continence function added further 7% units (B = -1.53, P = .006), and fecal continence function added at last 3% units (l3 = -4.57, P = ‘04) to the explained variance of CGA,?. The diagnostic group and the other parental variables did not significantly contribute to the explained variance of psychosocial functioning of the adolescents. With CAS symptom score (mental health) as the
Associations Between Somatic Variables, Parental/Family Variables, and Mental and Psychosocial Outcome The initial bivariate analyses indicated significant associations between diagnostic groups, duration of anal dilatation, bowel function, parental/family variables, and mental health and psychosocial adjustment (Table 3). When splitting the patient group into the two diagnostic groups, fecal incontinence was significantly correlated with poorer psychosocial functioning (+rs = -58, P = .Ol) and more parental criticism (rs = .63. P = .03) in the HD patients, whereas in the LARA group a number of significant correlations were found between the flatus continence function and several variables of mental health (CAS, YSR) and psychosocial functioning (CAS symptom score [rs = .73, P < .OOl], several CAS content areas scores; worries/anxieties, family, mood, expression of anger, YSR total score [rs = .61, P = .Ol], extemalizing and internalizing scores, several YSR subscales; anxious/depressed, social problems, aggressive behavior, and psychosocial functioning [ +rs = - .53, P = ,031). Examination of the relation between treatment procedures and outcome variables showed that the anal invasive procedures of extra crushing clamps and weekly lavage in the 10 HD patients were not significantly correlated to CAS or CGAS scores. But there was a significant difference in CBCL externalizing scores (z = -2.46, P = .Ol) and a trend toward difference with regard to YSR externalizing scores (z = - 1.65, P = .09) with higher externalizing scores in the group with extra invasive procedures. In contrast, the duration of anal dilatation in the LARA Table 3. Spearman
Rank-Sum Treatment
Diagnostic group: HD:LARA Duration of dilatationt
Correlation
Coefficients
Bowel Function
Describing
DUtmOtl of Dilatation
Fecal cant
Flatus ccmt.
CAS Symptom
.72ll
.31 21
5711 .35*
.37* .62§
.33*
29
Fecal incontinence Flatus incontinence
.67ll
YSR total CBCL total CGAS
§P< .Ol. IIPC ,001.
With
CGAS
.cl4
.30 .33
-.36$ -.35
.I2 .I6
.21
-.39* ~.48§
.I1 .I7
p.7411
.35 .26
.83ll .25 .44§
.32 .44*
HD and LARA
Farmiy Psychosocnl
CBCL Total
34
warmth critvzism
*A total set of family variables were only available Korrelation coefficients only for those who were tP < .05.
in Adolescents Outcome
YSR Total
w
CAS symptom
Parental Parental
Correlations
Mental and Psychosocv.4
-.5711 -.6Ol\
Parental Warmth
.53§ - .76\j
Parental Critusm
CFD
.34
.24
.06 .30
.05 19
.38* .31
.29 .45§
.31 .44* -.58/j
.58ll
.57§
for 30 adolescents. anal dliatated (n = 161.
Variables*
.34 -.781/ .76ll .52§
DISETH,
472
dependent variable, diagnostic group, duration of dilatation, fecal continence function, flatus continence function, CFD, parental warmth, and parental criticism were entered as explanatory variables. Duration of anal dilatation was the only significant predictor of CAS, explaining 28% of the variance of CAS in the whole group (l3 = 12.97, P = .008). Because anal dilatation was a treatment procedure primarly used in the LARA group. further stepwise multiple regression analyses were conducted to examine to what extent treatment and somatic variables contributed significantly to mental outcome in the LARA group. The mental health scores CAS symptom score, YSR total score, and CBCL total score were dependent variables entering the somatic correlates duration of anal dilatation, fecal continence, and flatus continence, into the model. The most significant independent predictor of the adolescents’ mental outcome as measured by the CAS symptom score was duration of anal dilatation, which explained 41% of the variance (B = 15.27, p = .005). With flatus continence function added, the explained variance increased to 58% (B = 2.10, P = .03). Entering the family/parental variables into the model did not significantly increase the explained variance of CAS . Duration of anal dilation was also the strongest independent predictor of mental health as measured by a youth self report; YSR total problem score (R2 = .47,l3 = 27.02, P = .002). Also for YSR total score, with flatus continence function added, the explained variance increased (R2 = .61, l3 = 3.14, P = .04). Duration of anal dilatation was the only significant predictor of mental health as measured by the questionnaire of another informant, the parents; CBCL total problem score (R2 = .24,B = 25.29, P = .04). For all dependent variables, the same significant predictors were selected regardless of whether log transformations were performed. However, the P values refer to the original scale. DISCUSSION
When analyzing possible reasons for the diverse outcome of mental health and psychosocial outcome in HD and LARA patients, we found that although both the HD and LARA groups experienced persistent impairment of fecal continence, the groups differed with regard to treatment procedures. such as type and length of invasive anal treatment regimes; in parent-child experiences, eg, age at diagnosis, postpartum mother-child separation, length of hospital stays, and painful experiences associated with treatment procedures; and in flatus continence function at the follow-up. There were no statistically significant differences in family or parental factors between the two groups. The duration of anal invasive treatment procedure and current bowel function were associated with mental and
EGELAND,
AND
EMBLEM
psychosocial outcome. The early treatment variable, duration of anal dilatation, was the most significant predictor of the adolescents’ mental health, whereas chronic family difficulties and parental warmth together with the current bowel function variables, fecal and flatus continence function, best explained the variance in psychosocial outcome at follow-up. The retrospective design makes it difficult to draw conclusions about cause and effect, and the relatively small number of patients in each group limits the generalizability of findings. However, statistical power considerations before the implementation of the study showed that the study was worthwhile, ie, chances were that interesting clinical differences would be detected with sufficient power. Furthermore, because of the centralization of treatment of these low-frequency congenital anomalies, the present sample is representive of the Norwegian patient population. The two patient groups had been treated at the same hospital, were assessed by the same methods, and were comparable with regard to gender, age, and sociodemographic characteristics. Our findings suggest some risk factors that may explain the difference in mental and psychosocial outcome for patients with HD and LARA: 1. The diagnosis of HD was generally made days to weeks after birth, and most parents experienced a healthy child in the neonatal period and had time to establish the important first relationship. When hospitalized, the mother could accompany the HD child. In the LARA group, the anomaly was usually noted soon after birth and immediately created parental worries and anxieties. This was often followed by a postpartum mother-child separation when the newborns were transferred to the National Hospital and were subjected operations and invasive anal treatment procedures. Parental reports based on the interviews indicate that this may have influenced the long-term parent-child relationship. 2. The two groups also differed in duration of hospital stays but not in number. Repeated hospitalizations in early childhood have been found to be associated with high rates of both short-term and long-term behavioral and emotional prob1ems.l“-16 Patients in our sample were born in the 1970s and early 1980s and at that time rooming-in facilities and round-theclock visiting for parents and siblings were not routine. However, neither number or duration of hospital stays nor age at diagnosis were correlated to long-term effects on mental health or psychosocial functioning of these adolescents. 3. Type and duration of invasive anorectal operation techniques and type of after-treatment regimes also differed between the two groups. Children with HD operated on at the median age of 7 months were
EFFECTS
OF ANAL
DILATATIONS
AND
INCONTINENCE
kept tied in the bed for 5 days after the operation with crushing clamps in the anus. In 37% of the HD children, an extra procedure with crushing clamps was performed at an average of 1.5 years of age. This “tying up” of the child was reported by their parents to be unpleasant for the child. In children with LARA, anal dilatation has been advocated to avoid anal stricture.17 Generally, no analgesia is used in this procedure. Anal dilatations were performed once to twice daily by the parents up to median 2 years of age (range, O-3 years). The parents reported that this had a traumatic impact on the child and affected the parent-child relationship creating a power struggle and parental guilt by inflicting pain on the child. Both the procedure and the child’s reaction to it, may have influenced the relationship between the child and the parents. This may explain the significant associations found in the present study between duration of anal dilatation and the adolescents’ present mental outcome (CAS, YSR, CBCL), and parental factors (CID, marital relationship) as reported in both low and high anorectal malformations.3 In the present study, the anal dilatation was correlated to fear, reality testing, withdrawal, anxious/depressed mood, and delinquent or aggressive behavior. The duration of anal dilatation in childhood was the most important predictor of the adolescents’ mental health at the follow-up. The difference in mental outcome in patients with HD and LARA suggests long-term negative implications when anal dilatations are performed for years. 4. Chronic family difficulties was the most important predictor of the adolescents’ psychosocial functioning. The LARA families experienced less network support and reported more disagreement in childcare and/or marital conflicts than the HD families. The LARA families reported problems with handling the child’s fecal soiling, and the anorectal anomaly had become a subject to be avoided. At follow-up we also found more worries about the present and future situation in the LARA group. The cross-sectional design of this part of the study does not allow us to draw conclusions. However, the parents’ experiences of long-lasting emotional and practical strain related to the child’s congenital disease may create risk for vicious circles with early strong ambivialent feelings in the mother creating tensions in the marital relationship. Such interactive effects are indicated by Ludman’s 3-year prospective study of full-term neonates undergoing surgery.l* The parents of sick babies were under greater general stress and had significantly more psychological problems than controls. This indi-
473
cates that a combination of stress factors rather than just the anxiety associated with the birth of a sick baby may contribute to the increased incidence of maternal depression. 5. The associations between less parental warmth/ more criticism and poorer psychosocial outcome of the children, are in line with previous findings.‘9-z1 We found high parental warmth to be one important predictor of the adolescents’ psychosocial functioning. This finding is in accordance with the documented protective nature of parental warmth. In this sample, many factors may have contributed to the parent’s Ievel of warmth and criticism. Chronic illness may lead to heightened maternal empathy toward the child,2z and strongly protective and emotionally involved parenting may be psychologically adaptive for children under stress. In the HD group, findings suggest that this may have contributed to counteract risk factors such as early separation. hospital admissions. and invasive treatment procedures. 6. The association between severity of physical impairment (fecal and flatus incontinence) and psychosocial outcome also corresponds to findings in other studieszz We found flatus incontinence to be one of four important predictors of CGAS and the two diagnostic groups differed in flatus continence function. Uncertainty about flatus control created many additional worries and preoccupations for the adolescents. To hide the unpleasant smell, they reported use of extensive amounts of perfume and/or they smoked cigarettes. Flatus incontinence was significantly correlated to poorer resting pressure, thus emphazising the importance of applying surgical techniques, preserving the internal anal sphincter. Our findings suggest that invasive anal treatment procedures and continence dysfunction may have negative impact on mental health and psychosocial functioning for patients with anorectal malformations. This exploratory study cannot document the mechanisms by which anal dilatation or incontinence affect mental and psychosocial outcome, but the findings raise interesting considerations. Possible consequences of daily dilatations performed for prolonged periods and of chronic fecal continence dysfunction may be addressed on the basis of developmental theories. The 2- to 3-years-of-age period is a critical age for the important developmental tasks of self control and autonomy. The ability of a child to control defecation at the end of this anal period represents psychosocial maturation and is associated with the building of autonomy and pride. Intrusive procedures in the anal region, a sensitive zone in this developmental phase, may be experienced as scary; they evoke protest and also
474
DISETH,
muscle contractions. Thus, both the anal dilatation and the persistent problems with continence may influence the child’s natural development of self-esteem, autonomy, and social activity, factors so important for mental health and psychosocial functioning. Concerning anal dilatation, one of the authorities in the field, Albert0 Pefia statesl’: “The parents are encouraged to perform the dilatations themselves twice a day at home. But if they have difficulties, it can be done at the hospital. The process of dilatation continues until the desired size is reached, then the colostomy may be closed, but the process of dilatation continues but gradually tapering the frequency over months.” The LARA parents in our study had performed daily anal dilatations for 2 to 3 years at home, and the dilatation had become a chronic stressor for both the child and the parents. One may question whether the recommended procedure of anal dilatation is based on empirical or documented effects. Penal’ noted that “the rationale behind dilatations consists of appreciation of the fact that the anus and the rectum are surrounded by muscle structures and therefore remain closed at rest, particularly in cases with good muscles. Thus, if the patient is not dilatated, the anus will tend to heal closed or very narrowly, provoking an anal stricture.” However, the effect of anal dilatation or the significance of the length of dilatations on outcome for patients with anorectal anomalies have not been documented. Our findings indicate that the use of anal dilatation needs to be reevaluated. Less embarrassing and painful procedures and more supportive interventions (eg, use of anesthetics, parent substitutes to perform the procedures, teaching relaxation techniques, and/or to have psychosocial experts work with the child and the parents) should be considered. To achieve optimal quality of life overall, the challenge is to find a compromise between the physically optimal procedures and procedures that are not psychosocially detrimental. In our study, a significant number of the HD and LARA patients had fecal incontinence problems (32% v 59%) and flatus incontinence problems (16% v 65%). In another study,z all adolescents with high anorectal anomalies had persistent fecal dysfunction, but only 19% had a permanent colostomy. One may ask whether too few of the patients with incontinence problems have been presented with the option of alternative procedures at an early age, eg, rectal irrigation procedures, antegrade colonic irrigation through appendixstomy, physiotherapeu-
EGELANDV
AND
EMBLEM
tic or biofeedback training, or permanent colostomy. Offering a permanent colostomy may have been delayed or avoided for fear of potentially serious psychosocial consequences. A colostomy may represent a chronic disability and loss of the hope for a perfect cure, but it may also reduce the practical and emotional strain of soiling and invasive treatment procedures, both for the child and the parents. In a controlled study of psychosocial adjustment to stoma surgery for childhood inflammatory bowel disease, Lask et alZ3 found no evidence that stoma surgery in childhood is psychologically harmful for psychosocial adjustment, self esteem, or quality of life. They concluded that as long as children are well prepared and followed-up carefully and sensitively, then stoma surgery should not be deferred for fear of adverse psychological consequences. However, although imperious not controlled defecation is an important complaint also in inflammatory bowel disease, the two patient groups are not quite comparable, and this may limit the generalizability of the findings in Lask’s study. This study of adolescents with congenital intestinal malformations demonstrates that the anatomic defect and its operative treatment is only one determinant for the long-term outcome. The duration of invasive procedures, the continence dysfunction. and the family/parental functioning, all seem to be associated with mental and psychosocial outcome of the adolescents. Performing invasive treatment procedures may have physical and mental implications, and findings suggest that the indications for and ways of performing these procedures, need to be reassessed, addressing both the medical and psychosocial aspects and outcome. Persistent incontinence into adolescence represents an outcome with negative psychosocial impact, and a colostomy may represent an alternative procedure to attain fecal control. The findings that chronic family difficulties and parental strain and emotional problems expressed as less warmth and more criticism, have an adverse effect on the adolescents’ psychosocial outcome, should be taken into consideration early in treatment, as well as the inclusion of psychosocial experts in the health care teams3 that treat patients with congenital intestinal malformations and their families. ACKNOWLEDGMENTS The authors thank Professor Inger Helene Vandvik for scientific advice, SonJa Heyerdahl, MD. PhD, for statistical support. and Medinnova for financial support.
REFERENCES 1, Diseth TH, BJcWnland K. Novik T. et al: Bowel function. mental health and psychosocial function in adolescents with Hirschsprung’s disease. Arch Dis Child 76.100-106, 1997 2. Diseth TH. Emblem R: Somatic function, mental health and psychosocial adjustment of adolescents with anorectal anomalies. J Pediatr Surg 31:638-643. 1996
3. Diseth TH, Emblem R, Vandvik IH: Adolescents with anorectal malformations and their famihes. Family Systems Medicine 13:215231,1995 4. Duhamel B: A new operation for the treatment of Hirschsprung’s disease. Arch Dis Child 35:38, 1960 5. Stephens FD, Smith ED: Classification, identification, and assess-
EFFECTS
OF ANAL
DILATATIONS
AND
INCONTINENCE
ment of surgical treatment of anorectal anomalies. Pedratr Surg Int 1:200-2051986 6. Achenbach TM~ Edelbrock CS: Manual for the Youth Self Report and Profile. Burlington. VT, University of Vermont. Department of Psychiatry. 1987 7. Achenbach TM: Manual for the Child Behavior Checklist and 1991 Profile. Burlington, VT, Umversity of Vermont, Department of Psychiatry, 1991 8. Shaffer D, Gould US, Brasic J, et al: Children’s Global Assessment Scale (CGAS). Psychopharm Bull 21:747-748, 1985 9. Taylor EA (ed): The overactive child, in Clinics in Developmental Medtcine 97. Oxford, England, Blackwell Scientific, 1986 10. Vandvik IH, Hoyeraal HM, Fagerhm H: Chronic family drfficulties and stressful life events in recent onset juvenile arthritis. J Rheumatol 16:1088-1092, 1989 11 Bjomstad P, Lindberg H, Spurkland I: Unge hjerter i faresonen. (Young hearts at risk). Oslo. Norway, Tano AS, 1990 12. Altman DG: Practical Statistics for Medical Research. London. England, Chapman and Hall, 1991 13. Kleinbaum D. Kupper L, Muller K: Applied regression analysis and other multivariable methods (2nd ed). Boston, MA, PWS-KENT Publishing Company, 1988 14. Douglas JWB: Early hospital admission and later disturbances of behavior and learning. Dev Med Child Neural 17:456-480. 1975 15. Quinton D, Rutter M: Early hospita1 adnusston and later
475
disturbances of behaviour: An attempted rephcation of Douglas’ findings. Dev Med Child Neurol 18:447-459, 1976 16. Graham P: Prevention. m Rutter M. Taylor E, Hersov L: Child and Adolescent Psychiatry: Modem Approaches. Oxford, England, Blackwell Scientific Publications. 1994 17. Peria A: Atlas of Surgical Management of Anorectal Malformations. New York, NY, Springer-Verlag, 1990 18. Ludman L, Lansdown R, Spitz L: Factors associated with developmental progress of full term neonates who required intensive care. Arch Dis Child 64:333-337, 1989 19. Rutter M. Yule B, Quinton D, et al: Attainment and adjustment in two geographical areas: III. Some factors accounting for area differences. Br J Psychiatr 126:520-533, 1975 20. Quinton D, Rutter M: Family pathology and child psychiatric disorder: A four-year prospective study. in Nicol AR (ed): Longitudinal Studies in Chtld Psychology and Psychtatry. Chichester, Wiley, 1985 21. Richman N, Stevenson J. Graham P: Preschool to school-A behavioural study. London. England, Academic Press. 1982 22. Garralda ME: Chronic physical illness and emotional disorder in childhood Where the brain’s not involved. there may still be problems. Br J Psychiatr 164:8-10, 1994 23. Lask B, Jenkins J. Nabarro L. et al: Psychosocial sequelae of stoma surgery for inflammatory bowel disease m childhood. Gut X1257-1260, 1987