In infancy, the symptoms may take special forms
M. Bellaiche, M.A. Maspoli, A.L. Lapeyraque, N. Boige, P. Foucaud (Versailles)
There is considerable variation in the reported incidence of gastroesophageal reflux (GER) in infants. One of the reasons for this may be that the clinical picture is polymorphic. Although upper gastrointestinal symptoms are an essential part of the diagnostic picture for GER, there are also more atypical manifestations. Certain pulmonary and ENT disorders, certain serious infantile diseases and even sudden deaths may be attributable to GER that has eluded diagnosis. GER can be identified under these clinical situations by 24-h pH monitoring, although nothing can be said about the cause-effect relation. The etiopathogenesis of these disorders is often multifactorial and only the correction of these extra-alimentary manifestations after antireflux treatment can give an indication of the true role of GER.
GER and respiratory disorders in the infant
Physiopathology
a. Induction of respiratory diseases by GER. Two mechanisms have been schematically proposed to account for the pulmonary complications of GER: bronchial aspiration of gastrointestinal fluid and reflex bronchoconstriction [1].
Bronchial inhalation of gastric juice causes pulmonary disease. Its toxic effects can cause bronchoconstriction [2], surfactant destruction with atelectasis and an increase
in total pulmonary resistance [3]. In some cases the clinical picture is very suggestive of an aspiratory mechanism. An acid taste in the mouth, confirmed by pH measurement, prior to the onset of cough or an asthma attack [4] is one example here. The combination of difficulty in swallowing and global loss of sphincter tone, in children with encephalopathy presenting lung disease associated with repeated swallowing, is another example.
Direct diagnostic methods based on tracheal intubation or bronchoscopy have been envisaged: detection in the airways of macrophages infiltrated with lipid droplets [5], lactose [6] or intragastrically administered dyes. However, these procedures are too invasive and not sufficiently specific. In addition, the intubation itself can influence the results. Radiological investigations can also assist the diagnosis, as when a bronchographic image is obtained after esophagogastroduodenal examination after administration of barium or pulmonary contamination after esophageal technetium scintigraphy. Unfortunately the sensitivity of these investigations is still insufficient, especially for the detection of repeated microaspiration [3]. Therefore, objective demonstration is not always possible and the diagnosis must be supported by the history and compatible radiological data [7,8].
Reflex bronchospasm was first evoked in adults with a combination of asthma and GER [9]. It is in fact known that intraesophageal infusion of acid causes broncho-spasm and potentiates the effect of the cholinergics [10]. A pediatric population with this same disease combination was investigated by Davis et al. [11]. The authors found that intraesophageal instillation of decinormal hydrochloric acid only causes bronchospasm in children with esophagitis and not in asthmatic subjects with uncomplicated GER. The damaging effects of esophagitis on the afferent nerve impulses may account for the increased sensitivity to variations of pH [12]. Animal studies indicate that the vagus nerve plays a part in this reflex bronchoconstriction [9] and this would explain the efficacy of atropine here. This mechanism should therefore be suspected, especially in asthmatics with nocturnal symptoms that are resistant to treatment with the usual bronchodilators.
b. GER induced by respiratory disease. Chronic respiratory disease causes symptoms and requires treatment which themselves can render the patient susceptible to GER. The excessive abdominal pressure created by forced expiration during wheezing and cough potentiate GER. This has been observed especially in patients with cystic fibrosis, asthma, bronchopulmonary dysplasia and infective lung disease. Negative intrathoracic pressure secondary to stridor or hiccup may have the same consequences on GER. Breathing exercises involving forced expiratory and postural maneuvers also facilitate the disorder [13].
The pressure of the lower esophageal sphincter (LES) may be reduced in subjects with respiratory disease, either because of the thoracic distension, or as a result of treatment with drugs such as the [3 adrenoceptor stimulants, the anticholinergics, or theophylline (which also increases gastric acidity) [14]. By combined evaluation of GER by pH monitoring and determination of respiratory function, Berquist et al. [15] confirmed that episodes of GER accompany bronchodilator therapy, but also showed that there are no spirometric repercussions.
GER and chronic respiratory disease present the risk of mutual exacerbation. However, Hampton et al. [16] found no significant correlation between the severity of GER and the change in respiratory function in 38 children suffering from chronic respiratory disease. The authors suggested that there may be individual susceptibility.
Epidemiology
In large-scale, multicenter pediatric studies, the incidence of respiratory disorders seen in conjunction with GER fluctuates between 15 and 45% [3,17-20]. It is very important to take into account the general condition of the patients when interpreting these figures. As was already seen, chronic respiratory disease can cause secondary GER. Analysis of six cumulative pediatric populations showed a 56% incidence of GER confirmed by 24-h pH monitoring. Cystic fibrosis [21] and asthma [22] are at the forefront here and are diseases in which antireflex treatment is often an integral part of the therapeutic scheme.
Patients with fixed or progressive encephalopathy are also very prone to the pulmonary complications of GER because of global hypotonia, disorders of swallowing, restraining devices, scoliosis and the need for physiotherapy. Deforming diseases can also be associated with reflux: operated atresia of the esophagus where there is major impairment of motility, and to a lesser degree, gastroschisis or omphalocele, cleft lip/palate and Pierre-Robin syndrome [3,8,20]. Short bowel syndrome, because of the impaired motility and the frequent dependence on gastrostomy feeding, is also a predisposing condition [23]. Age seems to play a part as witnessed by the variability of the percentage of respiratory disease in GER as a function of the mean age of the population considered. Ghisolfi et al. [19] found only 1% respiratory complications in babies under 3 months, compared with 20% in infants and 33% in children.
Clinical characteristics
The respiratory symptoms secondary to GER are very heterogeneous: bronchitis, with or without an asthma-like component, cough, wheezing, lung disease. A number of characteristics of these episodes suggest that GER might be responsible. GER should be checked for if the respiratory symptoms are recurrent, exacerbated in the supine position with a predominance of cough at night, especially 1 to 3 hours after going to bed [24,25], if the disorders are not seasonal in character [20] and if they affect the middle lobe in particular [3,17,20], The existence of other potential complications of GER such as apnea, faintness or repeated episodes of ENT disease tends to support the diagnosis, as does the existence of patent gastrointestinal symptoms, especially if these predominate at night [20].
Paraclinical characteristics
It is much easier to demonstrate the existence of GER in a patient exhibiting respiratory symptoms than to confirm the cause-effect relation between these
disorders. The supplementary investigations carried out for these respiratory conditions can give useful pointers here, however.
A high level of reflux of barium or disorders of swallowing during radiological examination of the esophagus, stomach and duodenum following administration of barium, can be indicative of aspiration. However, the observation period is too short and the main value of these examinations is that it allows visualization of the anatomy of the upper gastrointestinal tract [3]. Twenty-four hour pH monitoring is the investigation of choice for confirmation of GER [3,19,20,26]. Respiratory symptoms coinciding with a fall in pH or increased esophageal acidity at night are useful indicators. However, in the experience of the French-speaking pediatric nutrition and gastroenterology group, it would seem that the traces examined show a greater tendency to reveal very short-lived and repetitive acid refluxes predominating during the day without GER at night (unpublished data). Although esophageal scintigraphy, by revealing pulmonary contamination, is of theoretical interest, in practice its sensitivity is not greater than 25% [3,19,20,27]. Manometric analysis has shown a reduction in the pressure of the LES in several studies [28-30]. The results concerning the pressure of the upper esophageal sphincter are more questionable, however: Cargill et al. [28] and Gerhardt et al. [31] observed hypotonia of the upper esophageal sphincter whereas Sondheimer et al. [8] did not. The value of esophagogastric endoscopy resides in its ability to detect esophagitis which is recognized as a potential inducer of bronchospasm [12].
Treatment
A number of factors must be taken into consideration when deciding on the therapeutic strategy for this type of reflux. The ventral tilting position can render respiratory physiotherapy difficult. A compromise is therefore necessary. Most authors contraindicate the use of bethanechol because of its potential bronchospastic properties [32,33]. Treatment of the esophagitis must be optimal as this condition tends to maintain the disease. In patients receiving theophylline, it is advisable to use H2-receptor antagonists other than cimetidine. The course is usually favorable except in certain cases (bronchodysplasia, encephalopathy, esophageal atresia and gastrostomy feeding) where surgical treatment is often required [3,18,20].
GER and "near miss" sudden infant death syndrome
Episodes of suffocation, apnea or fainting can be life-threatening or may be premonitory signs of sudden death (main cause of death in children under 1 year of age). Investigations for etiological factors have emphasized the important role of GER [3,20,34,35].
Physiopathology
Several studies have shown that there is a correlation between the fall in intra-
esophageal pH and apnea [9,36,37]. Bradycardiac apnea was found to be induced by instillation of decinormal hydrochloric acid in five infants admitted to hospital after a "near miss" sudden death syndrome [3]. This was not caused by instillation of milk. The monitoring of pH visualized this phenomenon during acid reflux occurring at the same time as apnea in 40 to 100% of cases investigated for serious fainting and GER [38]. Furthermore, this sudden fall in esophageal pH was found to precede a fall in transcutaneous oxygen pressure in half of the premature babies investigated by Pasquis et al. [39]. Menon et al. [40] observed brief or even prolonged apnea associated with postprandial reflux in infants. However, this chronological relationship between GER and apnea was not seen in polygraphic studies [3,27,41] performed with infants who had been saved from sudden death.
The apnea can be accompanied by laryngospasm and bradycardia. The mechanism appears to be a vagal reflex triggered by esophageal and/or laryngeal chemoreceptors. This vagal hyperreflexia has been demonstrated by decelerations in ECG traces obtained with Holter ECG monitoring equipment and studies of the oculocardiac reflex in certain "near miss" infants [42,43]. In this population, GER was found in 18 to 30% of cases [27,34,44], This vagal dystonia is very probably related to a process of deficient maturation [34,35]. This is supported by the fact that anatomo-pathological and functional studies of the brain stem [45,46] have revealed immaturity of the autonomic nervous system. A common embryological origin for the triggering of fainting, apnea and GER has even been suggested in the Pierre-Robin syndrome. At the esophageal level, the vagal hyperreflexia also manifests by incomplete achalasia of the LES [34,35]. This has been observed in 75% of "near miss" infants with vagal hyperreflexia compared with 30% of infants without this condition [47].
Anatomopathological study
Systematic postmortem examinations have allowed considerable advances to be made in the understanding of the etiopathogenesis of sudden infant death, which is nowadays less and less inexplicable. The presence of esophagitis in 20 to 40% of cases [48,49] demonstrates the importance of the reflux factor in these infants.
Clinical studies
The initial demonstrations of GER in infants who had suffered respiratory arrest [36,37] were very encouraging. However, although the link between these two disorders is not in doubt, subsequent studies failed to confirm that all cases of "near miss" sudden infant death are the same. Near misses tend to exhibit polymorphic symptoms of necessarily fairly short duration. The episodes described comprise cyanosis and/or pallor, hypotonia, loss of contact, suffocation, apnea or cardiorespi-ratory arrest which responds to resuscitation. These episodes can be prolonged and life-threatening. A history of regurgitation or vomiting argues in favor of the link with GER. However, this parameter is lacking in sensitivity [3,27]. In practice, a history of this type was only found in one-third of cases on average in the various published studies [35,49-52]. Symptoms of esophagitis, the triggering of the disorder
on a change of posture and repeated respiratory symptoms are also factors to be checked for.
The incidence of GER among near miss infants ranges from 10 to 100% according to the study considered [35,49-52]. However, these figures are invalidated by recruitment biases and the methods of investigation used, and the true incidence is probably between 10 and 20% [3,20,27].
The diurnal character of near miss sudden infant death can perhaps assist in determining its etiology. The GER diagnosed in near miss infants by polygraphic investigations was found to occur in the waking period in 60 to 75% of cases [52-54]. In the cases where GER developed at night, however, Jeffrey et al. [54] have found that it occurs most frequently during REM sleep.
The efficacy of antireflux treatment is the most convincing argument here. The disappearance of the near miss episodes during treatment demonstrates the responsibility of the GER [34,35,37,49]. If medical treatment is not sufficient to improve the clinical picture and to normalize esophageal pH, the infant may require surgical treatment [3,20,27].
Paraclinical studies
In all cases of near miss sudden infant death syndrome, an exhaustive investigation is required to check for pre-existing neurological, metabolic, cardiac or ENT disease. A systematic investigation for GER must also be carried out. Barium meal, scintigraphic and endoscopic investigations have the same value as in patients with respiratory disorders associated with GER. The monitoring of pH can be combined with permanent ECG monitoring using a Holter apparatus or, better still, integrated into a single polygraphic trace. In addition to its diagnostic value, it helps to guide patient management. In practice, optimal antireflux treatment is based on the results obtained for control pH measurements. Esophageal manometry designed to check for incomplete achalasia will give useful information concerning the state of esophageal immaturity [47]. In addition, a greatly diminished lower esophageal sphincter pressure (LESP) is the best indication for the use of urecholine.
Hygienic-dietetic treatment of reflux (ventral tilting at 30°, thickening and fractionating of meals, early diversification of diet, large cot, absence of passive smoking) is initially sufficient. However, recent studies [55] have reported a reduced incidence of sudden infant death syndrome in the Netherlands following a government advisory circular recommending that all infants should always be placed in lateral decubitus in their cots. Initially pronate ventral decubitus appears to be a risk factor, especially because of the stifling and hyperthermia that it can generate.
Drug therapy is often required, mainly with cisapride and bethanechol. The presence of vagal hypertonia in addition to the GER complicates management. Prescription of an atropine-like drug such as diphemanil in theory reduces the lower esophageal sphincter pressure (LESP) and antagonizes the parasympathomimetic
motility stimulants used. However, in clinical practice the anticholinergics do not appear to aggravate GER [56] or to reduce LESP in infants receiving antireflux treatment [57].
Where medical treatment fails, surgical treatment should be considered. Cardiorespiratory monitoring at home is sometimes combined with antireflux treatment until the esophagus has matured.
GER and ENT disorders
The physiopathological mechanisms that engender ENT symptoms in GER are the same as those described for the respiratory symptoms. GER also acts to maintain lesions in cases of laryngeal surgery or acquired stenosis in intubated children. In these cases antireflux treatment can reduce the morbidity caused by inflammation and edema that obstructs the upper airways.
ENT symptoms can develop at several levels. GER should be checked for in subjects with a chronic, dry, barking cough occurring mainly at night, once the possibility of malformation, allergy or a foreign body have been eliminated. Recurrent acute laryngitis, as observed in the adult [58], occurring without fever or nasopharyn-gitis and without any seasonal factor, is more frequently a sign of GER than of allergy or adenoiditis. This diagnosis is supported if laryngoscopy reveals erythema of the posterior rim [59].
Pharyngeal symptoms resulting from GER often take the form of acute or repeated pharyngitis or pharyngeal paraesthesia in older children [60]. Posterior pharyngeal or retrocricoid granulations surrounded by erythema are often visible. By extrapolation, the GER could favor the development of otitis or nasopharyngitis. However, this is very difficult to demonstrate, especially in these multifactorial disorders. In order to corroborate these pediatric clinical findings, a 2-channel esophageal and oro/hypo-pharyngeal pH monitoring technique was used in a recent study [61] to demonstrate the pathogenic role of GER in a population of 21 children suffering from recurrent otitis and pharyngitis, by comparison with a control group. A significant difference was observed for both the esophageal trace and the pharyngeal trace, thus confirming the presence of gastric fluid in the ENT sphere in these children. The role of GER in this type of condition will be best established, however, by analysis of the symptoms, 24-h pH monitoring and the success of antireflux therapy.
References
14. Christensen J. Effects of drugs on esophageal motility. Arch Intern Med 1976:136:532-537.
49. Herbst JJ, Myers WF. Gastroesophageal reflux in children. Adv Pediatr 1981;28:159-186.