 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 106 No. 6 December 2000, pp. 1349-1354
,
From the * Institute for Research in Extramural Medicine, Vrije
Universiteit, Amsterdam, the Netherlands; Organization for Research
and Policy Advice, Grave, the Netherlands; and the § Department of
Pediatrics, University Hospital VU, Amsterdam, the Netherlands.
 |
ABSTRACT |
|---|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
Objective. To determine the effectiveness of whey hydrolysate formula in the treatment of infantile colic in a primary care setting in the Netherlands.
Study Design. Randomized, double-blind, parallel trial with a 1-week qualification period and a 1-week intervention period.
Participants. Forty-three healthy, thriving, formula-fed infants, <6 months old, crying >3 hours per day on at least 3 days per week. Infants were randomized to whey hydrolysate formula (n = 23) or standard formula (n = 20).
Main Outcome Measure. Difference in duration of crying (minutes per day) between qualification week and intervention week.
Results. Analysis according to the intention to treat principle showed a difference in the decrease of crying duration of 63 minutes per day [95% confidence interval: 1-127 minutes per day] in favor of the whey hydrolysate formula. Five infants did not complete the trial. The scope of the study was not sufficient to expect significant differences in the subgroup analyses.
Conclusions. An extensively hydrolyzed whey formula is effective in reducing the duration of crying in a primary care setting. Key words: colic, crying, hypoallergenic formulas, randomized, controlled trial.
Infantile colic can be described as follows: a healthy,
thriving infant with paroxysms of excessive, high-pitched, inconsolable crying, frequently accompanied by flushing of the face, drawing up of
the legs, passing of gas, and difficulties with the passage of stools.
The paroxysms typically occur in the evenings. They start in the first
weeks of life and generally cease at the age of 4 to 5 months.1 To discriminate excessive crying from normal
crying, a time criterion is often used: at least 3 hours per day for at
least 3 days a week for a minimum of 3 weeks.2 Infantile
colic can be thought of as a syndrome or as a symptom.3,4
When considered as a syndrome, the excessive crying is only one of the
features, because often, it is accompanied by signs of pain,
inconsolableness, abdominal distention, passage of gas, and
difficulties in passing stools. Seen as a symptom, excessive crying is
considered to be the main feature, which may or may not lead to
abdominal distention, inconsolableness, etc. It is not clear whether
the combination of this key symptom and features such as
inconsolableness and symptoms of gastrointestinal origin Infantile colic is one of the most common problems in
infancy.5 It leads very often to unsubstantiated formula
changes.6 Mothers of infants who excessively cry
frequently believe that their child is not healthy.6 The
condition may have long-term consequences for the mother-child relationship.7-9 Cow's milk allergy is commonly regarded
as a cause of infantile colic. A recent systematic
review10showed the effectiveness of a cow's milk-free
diet. The effectiveness was established in trials using a casein
hydrolysate formula11,12 as a substitute for cow's milk,
and was not restricted to highly selected infants. Randomized trials
using whey hydrolysate formulas13-15 were located in this
review, but were excluded because of serious methodologic shortcomings
(lack of control group and too short an intervention period). Thus, the
effectiveness of whey hydrolysate formulas is not yet established.
However, compared with casein hydrolysate, whey hydrolysate has several
advantages: it is more user-friendly because its taste and smell is
better, and the cost to patient and community is lower.16
Based on the foregoing data, we conducted a randomized, controlled
trial aimed at the effectiveness of whey hydrolysate in the treatment
of infantile colic. We were also interested in the effectiveness of
this formula in specific subgroups of colicky infants: those with
atopic features or a positive family history of atopy. Finally, we
wanted to examine differences in effect in children showing the
complete picture of infantile colic: excessive crying plus
inconsolableness and other gastrointestinal symptoms versus children
with excessive crying as the only symptom.
Design
The study was conducted as a randomized, double-blind,
placebo-controlled, parallel trial, with a 1-week qualification period and a 1-week intervention period. After the 1-week qualification period, infants were allocated to experimental or control group by
random numbers. Infants in the experimental group were fed a
hypoallergenic whey hydrolysate preparation (Nutrilon Pepti, Numico,
Zoetermeer, The Netherlands) whereas infants in the control group were
given a standard cow's milk formula (Nutrilon Premium, Numico,
Zoetermeer, The Netherlands). The study has been approved by the
ethical committee of the Academic Hospital of the Vrije Universiteit,
Amsterdam.
Patient Selection
The study took place from August 1994 to October 1996. Children
were recruited in community-based, well-baby clinics in 6 regions (7500 births per year). Children were eligible for the study when parents at
scheduled visits complained spontaneously about their infant's crying
or when they answered "yes" to a question on whether the infant
cried a lot. This question was explicitly asked when there were no
spontaneous complaints about crying. Infants who met the following
criteria (to be determined by the clinic doctor) were announced to the
trial center as eligible: 1) general good medical condition; 2)
thriving; 3) <6 months; 4) at least 1 formula feeding per day; 5) good
feeding technique; 6) no previous trial of hypoallergenic feeding; 7)
no history of anaphylactic reactions to cow's milk; and 8) the
infant's caregivers exhibiting sufficient understanding of the Dutch
language to follow instructions.
Procedure
When an infant was eligible, the name and telephone number of
the family were given to the trial center, and the research nurse for
the region was contacted on the same day. The research nurse spoke to
the parents by telephone and made an appointment for the next day. She
visited the family at home, explained the study, and obtained informed
consent. The parents were given a written outline of the study, and
were then instructed to use a structured 24-hour diary and
questionnaire and asked to start the diary and fill in the
questionnaire the next day. After this 1-week qualification period, the
research nurse returned. If, based on the information in the diary, the
infant cried at least 3 hours per day during at least 3 days in this
first week, he/she was randomized to the experimental or the control
group. During the second week, an intervention week, parents had to
fill in an identical 24-hour diary. At the end of the second week, the research nurse visited the family for the final time to collect the
study forms and to measure the remaining formula powder. Finally, the
parents were told which formula their infant had been fed.
Randomization and Blinding
In the preexecutive phase of the study, random numbers were
generated by SPSS (SPSS Inc, Chicago, IL).17 Each random
number was printed on 2 labels and stuck on 2 cans containing the same
formula (experimental or control). All cans were identical. They were
packed in boxes by the manufacturer, each box containing 12 cans of
standard formula and 12 cans of hypoallergenic whey hydrolysate
formula. The trial center distributed the boxes to the research nurses,
who executed the assignment, giving 2 identically coded cans to the
parents of each qualified infant. This procedure had the advantage of
assigning approximately equal numbers of patients in each region to
each treatment. At distribution the research nurses were unaware of the
content of the cans. Just before the nurse's last visit, she contacted
the trial center to find out which formula corresponded with the number on the cans the infant had been fed. Whey hydrolysate formula differs
from standard formula in taste, smell, and appearance. To minimize
these differences, the manufacturer substituted one fourth of the
powder in each can of standard formula with hydrolysate powder.11 To test whether adequate blinding was achieved,
we added a question to the questionnaire: "Do you know which formula
your infant has been using this week?" If the question was answered
"yes," parents had to state which formula was used Measurements
Main Outcomes
Caregivers recorded the duration of their infant's behaviors
and their activities with the infant: crying, fussing, feeding, taking
care of the infant, playing with the infant, sleeping, being awake and
content. These behaviors and activities were recorded in 10-minute
periods 24 hours a day. The diary was modified after the method
developed and validated by Barr et al.18 When a first-week
diary showed at least 3 days with 18 10-minute periods (>3 hours) of
crying, the child was randomized to experimental or standard treatment.
During the second (intervention) week the same diary was used. The main
outcome is the difference in the reduction of crying duration (not
crying/fussing duration) between qualification and intervention week.
To determine the adherence to the complete time criterion for excessive
crying, we asked for the estimated duration in weeks of the crying
problem before inclusion.
Secondary Outcomes
We compared differences in the proportion of infants, no longer
satisfying inclusion criteria after the intervention period (ie, not
crying 3 hours per day for 3 days of the week) in the experimental and
standard treatment groups.
Interventions
The independent variable was the kind of formula: whey
hydrolysate or standard formula. Every infant was provided with 2 cans of formula powder, each can containing 400 g. The remaining powder was measured at the end of the intervention week.
Covariables
Data on atopic features in the infant (atopic eczema, asthma,
food allergy) and the presence of a family history of atopy were
obtained by the questionnaire. With the exception of the presence of a
food allergy, which was measured on a 5-point scale, these features
were measured dichotomously (present/absent). Accompanying (syndromal)
features were assessed in 6 questions on a 4-point scale:
inconsolableness, pain (2 questions), abdominal distention, passage of
gas, and difficulties with the passing of stools. This results in an
overall score with a range of 6 to 24. A high score indicates that the
infant had the typical syndrome; a low score indicates that the
excessive crying was not accompanied by other features. Both atopy and
accompanying features were measured with the intention to perform
subgroup analyses.
Statistical Analysis
Basic characteristics are presented as means or absolute numbers
with 95% confidence intervals.19 For the analysis of the
main outcome, we compared the mean difference in duration of crying
between the qualification and intervention week in the experimental
group with the mean difference in the qualification and intervention
week in the control group. Individual differences were used in the
analysis. These differences were checked for normality and analyzed
with the F test from SPSS' multivariate analysis of
variance (MANOVA).20 The analysis was performed both
according to the intention-to-treat principle and applied to those
participants who completed the trial. To test whether unblinding
influenced the outcome, we repeated the analysis without those infants
whose parents were correctly aware of the formula. Numbers of infants
in both groups, no longer satisfying the inclusion criteria after the
intervention period, were compared using the The following variables were considered as possible confounding
factors: age, sex, level of crying in the qualification week, atopic
family history, atopic features in the infant, food allergy in the
infant, and the presence of the syndromal features of infantile colic.
The mutual relation between the confounding variables was examined with
Pearson correlation coefficients.19 From those variables,
which were strongly correlated, only one was used in the further
analysis. Thus, the remaining variables were entered in a multivariate
model (MANOVA) for the continuous outcome measure, and in a logistic
regression equation19 for the dichotomous outcome measure.
Subgroup analyses were planned for atopic versus nonatopic infants, and
for infants with the complete (syndromal) picture of infantile colic
(excessive crying including inconsolableness and gastrointestinal
symptoms) versus infants with an incomplete picture. Because of
insufficient scope, these subgroup analyses are uninformative and will
not be presented.
We defined a difference in the decrease of crying 60 minutes per 24 hours as clinically relevant. At the planning stage of this trial, we
were not aware of any data on the standard deviation (SD) of the
reduction of the duration of crying. Our study consisted of 150 infants.
The progress of the patients through the trial is shown in Fig
1. Ninety-eight infants satisfying the
eligibility criteria were referred to the trial center. Of the 98, 14 did not start and another 8 did not complete the qualification period for various reasons (Table 1). Of the
remaining 76 infants who completed the qualification period, 33 infants
did not meet the inclusion criteria of crying for 3 or more hours per
day, at least 3 days per week. The length of the duration of the crying problem before the qualification period of these 33 infants (mean: 4.8 weeks, SD 1.6 weeks, P = .96) did not differ from those
who did qualify. Most of the parents of the nonqualified children stated that their infant was crying considerably less during the qualification week. Forty-three infants were randomized in total. From
the 5 infants who were withdrawn during the intervention week, data
were available for <3 days. Thirty-eight infants completed the trial,
of whom 20 were fed whey hydrolysate formula, and 18 standard cow's
milk formula. The trial had to be terminated before the estimated
number of patients were included because of the time already spent on
sampling participants (recruitment more than fivefold less than
expected) and the expiration date of the formula powder.
TABLE 1
as described
aboveconstitute a clinical entity distinct from excessive crying
alone.
METHODS
Top
Abstract
Methods
Results
Discussion
References
whey
hydrolysate or standard formulaand why they thought they knew which
formula they had fed their child .
2
statistic with Yates' correction.19 The difference in
duration of crying between trial infants (n = 38) and
nonqualified infants (n = 33) was analyzed with the 2-sample t tests.19
RESULTS
Top
Abstract
Methods
Results
Discussion
References
View larger version (32K):
[in a new window]
Fig. 1.
Flowchart describing progress of patients participating in a
randomized, clinical trial21 on the effectiveness of whey
hydrolysate formula in infantile colic.
Selection of Trial Population: Reasons and Numbers for Not Entering the
Trial
The baseline characteristics of the 38 infants completing the trial are given in Table 2. The experimental and control group differed in some aspects: age, gender, crying and crying/fussing levels and the presence of a food allergy. For the entire population, overall mean crying time was 284 minutes per day (SD: 93 minutes/day); mean crying plus fussing time was 368 minutes per day (SD: 114 minutes/day); mean duration of the problem was 4.8 weeks (SD: 1.4 weeks); mean decrease in crying time was 75 minutes per day (SD: 97 minutes/day).
|
The crying of infants on hypoallergenic formula decreased by 77 minutes
per day more than the crying in children on standard cow's milk
formula (Table 3). Unadjusted for
confounding factors, this difference is statistically significant
(P = .01). Weight and length were strongly correlated
with age and thus not entered in the multivariate model. The remaining
covariables were analyzed in a simultaneous model and showed 2 confounding factors: the infant's gender and the duration of crying
during the qualification period. Adjusted for these confounders, the
difference of the decrease in crying time between whey hydrolysate and
standard formula group was 47 minutes per day (n = 38, P = .04). Analysis, including the 5 patients who did
not complete the trial (Fig 1), revealed a net effect of 63 minutes per
day (n = 43, P = .05). The number of
infants no longer satisfying the inclusion criteria after the
intervention week (ie, crying <3 hours per day on at least 3 days) was
8 of 20 (40%) in the whey hydrolysate group versus 5 of 18 (28%) in
the standard formula group (2 = 0.20, P = .65). This result was not confounded by the
infant's gender and the duration of crying in the qualification
period.
|
Concerning the question on blinding, 6 parents indicated that they were possibly aware of the formula their infant had been fed. Of these 6 parents, 2 falsely identified their formula as whey hydrolysate, whereas 4 correctly identified the formula as whey hydrolysate. Reasons for unblinding were: 1) aspect of the stools and 2) color and smell of the formula. Sensitivity analysis removing these 4 infants from the analysis revealed an unadjusted effect of 75 minutes per day and an adjusted effect of 58 minutes per day (P = .03).
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
Substitution of cow's milk formula by an extensively hydrolyzed whey formula is effective in the treatment of infantile colic. The intervention diminishes crying ~1 hour per day, which is, as we predetermined, a clinically relevant reduction. It is possible that this result is an underestimation of the real effect because the strength of the intervention was reduced by substituting one fourth of the standard formula with hydrolysate powder. Consequently, the effect of cow's milk substitution is demonstrable in a primary care setting, and is applicable to healthy, thriving infants, <6 months, crying for >3 hours a day, at least 3 days per week. The subgroup analyses, aimed at the identification of atopy or the presence of additional features as modifying factors, had insufficient scope because of small numbers.
The baseline characteristics showed some differences between the intervention and control group. In the multivariate analysis, 2 modifying variables were identified: gender and the duration of crying in the qualification week. Both factors reduced the net effect. It is most likely that these confounders were caused by unequal distribution by chance because of the small sample, and not to the randomization procedure itself. Failure of blinding parents adequately is a possible source of information bias, because the parents had to assess the primary outcome (amount of crying) in the diary. Knowledge of using the experimental treatment could result in a psychological benefit, leading to more favorable registration of the outcome.22 During this trial, 4 parents seemed to know that their child was provided with the experimental formula. Despite this, the effect of the intervention was not clearly influenced.
In accordance with most diet studies in infantile colic, the duration of the intervention period was 1 week. In this period an effect is to be expected, assuming an association between infantile colic and cow's milk allergy. However, some doubt of the persistence of the effect remains as Forsyth11 demonstrated a decreasing effect of cow's milk elimination after repeated elimination/provocation procedures.
An interesting finding is that approximately one third of the infants, whose parents initially complained of excessive crying, did not meet the inclusion criteria after the qualification period . Most parents of the nonqualifying infants reported a decrease in duration of crying during the qualification week. This might be because of the natural course of infantile colic or to the Hawthorne effect.23
The primary difficulty in performing this trial was the recruitment of cases. Data on the prevalence of infantile colic vary widely, depending on how it is defined and measured; rates of 10% to 40% are reported.24 We conservatively assumed a prevalence of 10% and expected to find 750 cases within 1 year. Posthoc, the following reasons can be considered as an explanation for the low actual recruitment numbers. First, the way we measured the presence of infantile colic was perhaps more accurate than in other studies: parents had to keep up a 7-day diary with 10-minute periods. Most likely, this reduces the number of cases as opposed to measures relying on the parents' memory. For example, a recent study showed an infantile colic occurrence rate of 5.3% when registered with a diary, as opposed to 12.1% relying on parents' memory.25 Second, our cases fulfilled both of Wessel et al's criteria2: they cry for >3 hours on at least 3 days for >3 weeks. Other trials used less stringent criteria.26-34 The use of an accurate measurement instrument and a strict definition in this trial led to nonparticipation of 33 infants that might have been included when solely the parents' report of problem crying was used. Third, the motivation of the clinic doctors and nurses may not have been sufficient. We have no data to investigate this hypothesis. However, we tried every means possible to keep motivation high: explanation of the study to all participants personally, extra visits of the participants after several months and keeping them informed by newsletters 4 times a year. Although it is generally known that recruitment numbers often lag behind expected numbers,22 the final attrition rate in this study was not anticipated. Associated with the low recruitment numbers was the final sample size. When we started the trial, the estimated sample size was 150 infants to be included in 1 year. This estimation was arbitrary because of the lack of data on the standard deviation of the decrease of crying. The imminent expiration date of the formula powder (October 1996) led us to stop the trial at attrition of 43 participants in September 1996.
Both casein hydrolysate and extensively hydrolyzed whey formula are effective in the treatment of infantile colic. Therefore, because it smells and tastes better and costs less, whey hydrolysate should be the first choice. The results of this trial are applicable to infants receiving at least 1 formula feeding per day. Research with casein hydrolysates in exclusively breastfed infants has shown a reduction in the duration of crying.10 Therefore, we assume that these results are also applicable to breastfed infants.
Interventions, proved to be effective in the treatment of infantile colic, include substitution of cow's milk with hypoallergenic formula, medication (dicyclomine) and behavioral interventions (ie, the advice to reduce stimulation).10 Which intervention should be used first? Or, should a combination of interventions be used? As there are serious side effects, dicyclomine should not be used at all.10 Although our trial lacked the scope to support this recommendation, the choice between hypoallergenic formula and behavioral intervention could depend on the presence of atopic features in the infant, a positive family history of atopy or the presence of the more typical clinical picture of infantile colic. So, one could choose to start with whey hydrolysate in infants with atopic features on pathophysiological grounds. In all other cases, the treatment of first choice depends on the preferences of the physician.
Additional research is needed to establish the effectiveness of hydrolysate formulas in infants with atopic features and in infants showing the complete picture of infantile colic, as opposed to infants with excessive crying as the single symptom.
| |
ACKNOWLEDGMENTS |
|---|
This project was funded by the Praeventie Fonds, Grant No. 002824560. Nutricia kindly provided the formulas.
We thank E. Buiting, M. Boersma, R. Rip, C de Jongh-v.d. Velde, F. Bijleveld, M. Oey-Spauwen and the doctors and nurses of the well-baby clinics for including crying infants: Th. Reynders-Jenniskens, E. Royackers-v.Rixtel, R. Veldkamp-de Wit, A. Schreurs-Crienen, K. Könings-Sanders, A. Roodhart, C.v. Eijk-Driessen, D. Kemp-Verhoeven, D.v. Deenen-Rutgers, and R. Lucassen-Hermans for visiting families with excessively crying infants; P. Jansen-Ypma, Y. Peters-Moors, S. Hoornveld for managing the trial center; Nutricia for providing the experimental and standard formula-especially J. Bindels, F v.d. Wees and L. Thörig; the members of the Scientific Committee of the Dutch College of General Practitioners for their help in the development of the trial protocol; and Rosemarie Tomes for linguistic help.
| |
FOOTNOTES |
|---|
Received for publication Jun 1 1999; accepted Mar 21 2000.
Address correspondence to Peter L. B. J. Lucassen, MD, Akkerroosstraat 18, 5761 EX, Bakel, the Netherlands.
| |
ABBREVIATIONS |
|---|
MANOVA, multivariate analysis of variance; SD, standard deviation.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
less common than previously estimated? Acta Paediatr. 196;85:454-458This article has been cited by other articles:
 |
|
 |
|
  F. Savino, E. Pelle, E. Palumeri, R. Oggero, and R. Miniero Lactobacillus reuteri (American Type Culture Collection Strain 55730) Versus Simethicone in the Treatment of Infantile Colic: A Prospective Randomized Study Pediatrics, January 1, 2007; 119(1): e124 - e130. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Hill, N. Roy, R. G. Heine, C. S. Hosking, D. E. Francis, J. Brown, B. Speirs, J. Sadowsky, and J. B. Carlin Effect of a Low-Allergen Maternal Diet on Colic Among Breastfed Infants: A Randomized, Controlled Trial Pediatrics, November 1, 2005; 116(5): e709 - e715. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. K C Leung and J. F. Lemay Infantile colic: a review The Journal of the Royal Society for the Promotion of Health, July 1, 2004; 124(4): 162 - 166. [Abstract] [PDF]
|
|
|
|
 |
|
 S. Wade and T. Kilgour Extracts from ""Clinical Evidence"": Infantile colic BMJ, August 25, 2001; 323(7310): 437 - 440. [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Walker Whey hydrolysate formula reduced crying time in infantile colic Evid. Based Nurs., July 1, 2001; 4(3): 76 - 76. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2000 American Academy of Pediatrics. All rights reserved. |
||
|
||
