Infantile hypertrophic pyloric stenosis: genes and environment
- Dr E Chung, General and Adolescent Paediatric Unit, Institute of Child Health, University College London, The Rayne Building, 5 University Street, London WC1E 6JJ, UK;
Most paediatricians take great pleasure in making a diagnosis of infantile hypertrophic pyloric stenosis (IHPS; OMIM 179010). It is a most satisfying experience to observe the dramatic gastric peristalsis and to palpate the pyloric “tumour” during a positive test feed. Over 120 years after the condition has become a clinical entity,1 its aetiology remains unclear. The condition has an interesting age-specific and tissue-specific nature. IHPS is never seen beyond the age of 3 months except in reports of premature infants in whom enteral feeding had been started late. This suggests that a period of enteral feeding is required for the condition to become clinically evident. Either the defect is only critical to the infant in the first 3 months of life and/or there are compensatory mechanisms that will circumvent the pyloric obstruction over time. The main reported pathology is restricted to the pylorus associated with smooth muscle hypertrophy, and the pylorus has been shown to make a complete recovery after surgery.2 Recent advances in understanding of the control of gastrointestinal motility have provided a firmer basis for identification of the disease pathways underlying IHPS. The control and regulation of gastric motility and pyloric sphincter function is a complex system which involves the intrinsic myogenic activity of smooth muscle cells, the interstitial cells of Cajal (pacemaker cells), gastrointestinal hormones (eg, motilin, cholecystokinin and gastrin), the autonomic nervous system and the enteric nervous system. The excitatory pathway is mediated mainly by acetylcholine, tachykinins, serotonin, gastrin-releasing peptide and motilin. The inhibitory pathway includes the non-adrenergic, non-cholinergic enteric nervous system, the main neurotransmitters of which are nitric oxide and members of the vasoactive intestinal peptide family. Many of these pathways, proteins and their encoded genes have become targets of investigations.
Epidemiologically, IHPS is the most common condition requiring surgical intervention in the first year of life. It is most common in Caucasians and is relatively uncommon in Afro-Caribbeans and Asians. The condition has a reported incidence of 1–8 per 1000 live births. A recent decline in its incidence has been reported in a number of countries. Moreover, this decline has been observed to parallel a decline in the incidence of sudden infant death syndrome (SIDS) in Denmark and Sweden since the early 1990s.3 4 This led to the hypothesis that the two conditions may share a common environmental factor. The decline in SIDS has been mainly attributed to the “back to sleep” campaign. It has therefore been suggested that a prone sleeping position in early infancy may also be a risk environmental factor for IHPS. If true, this may also in part account for the rare occurrence of IHPS in Asians, as, culturally, these babies have routinely been placed in a supine position to sleep. There has been no biological reason to explain this association, although it has been demonstrated that most of a feed was localised to the antrum in a prone position, whereas in the supine position, the meal was localised in the fundus. It is plausible that pooling of a feed in different parts of the stomach, as a result of different posture, may exert different effects on the function of the stomach and/or the pylorus via proteins sensitive to change in volume, pressure or solute concentration. This may in turn lead to dysmotility of the stomach and/or pylorus associated with work hypertrophy. However, the decline in the incidence of IHPS appears to have preceded that of SIDS and such a parallel decline in their incidences has not been observed in Scotland, as reported in this issue of Archives of Disease in Childhood.5 This possible association between posture and the development of IHPS does warrant further investigation, especially if it offers potentially a very cost-effective preventive measure for IHPS, at least in at-risk infants. The decline in the incidence of IHPS appears to be real, and a search for its explanation will provide important clue(s) to its underlying pathophysiology.
A further epidemiological observation has implicated the motilin pathway in the aetiology of IHPS. A report of a sevenfold increase in the incidence of IHPS among infants who had received erythromycin,6 a motilin agonist, for post-exposure pertussis prophylaxis suggests that abnormalities in the motilin pathway may predispose some infants to developing IHPS. The gastrokinetic effects of erythromycin are variable and complex and include effects on the timing, duration, amplitude and distribution of the migrating motor complex. In at-risk infants, it has been speculated that the motilin-like effects of erythromycin on antral smooth muscle function may lead to abnormal or excessive pyloric/gastric motility, resulting in pyloric muscular hypertrophy. A recent study found no mutations in the gene for motilin in 57 patients with IHPS. The number of patients is small, and they have no documented history of exposure to erythromycin, which makes interpretation of the results more difficult. Obviously other genes in the pathway may be the culprit, such as the gene encoding the motilin receptor.
Perspective on the paper by Sommerfield et al (see p 1007)
Many studies have been attempted to identify the aetiological factors for IHPS. A large number of abnormalities have been reported that mainly involve the pyloric musculature and its surrounding neuronal network. Other reported associations include Helicobacter pylori infection and hyperacidity in the duodenum. Most of these findings are likely to be secondary changes arising from as yet unidentified primary pathophysiological pathway(s). A better understanding of the pathophysiology of IHPS will no doubt come from the elucidation of its molecular genetics. A genetic predisposition to the condition has been well established, with a male preponderance of about 4 to 1. It is often inherited as a complex multifactorial trait, which is a result of interaction between genetic and environmental factors. IHPS has been associated with several genetic syndromes, such as Cornelia de Lange and Smith–Lemli-Opitz syndromes, and chromosomal abnormalities, including translocation of chromosome 8 and 17 and a partial trisomy of chromosome 9. Autosomal-dominant monogenic forms of IHPS have also been reported in several extended pedigrees.
Five IHPS genetic loci have so far been identified: IHPS1 (NOS1; 12q24.2–q24.31)7–9; IHPS2 (16p13–p12)10; IHPS3 (11q14–q22)11; IHPS4 (Xq23)11; IHPS5 (16q24).12 Accumulating evidence has strongly implicated NOS1 (IHPS1; 12q24.2–q24.31), which encodes the gene for neuronal nitric oxide synthase (nNOS) in the aetiology of IHPS. nNOS is the critical enzyme for the production of nitric oxide, which mediates relaxation of the pyloric smooth muscle. It is proposed that defects in nNOS may lead to failure of pyloric sphincter relaxation with subsequent hypertrophy. NOS1 is the most complex human gene yet described in its promoter diversity, which offers great potential for developmental regulation. A polymorphism in the regulatory region of NOS1 has been reported to be associated with IHPS but in only a small number of people.13 This result has not been replicated (unpublished data), although the risk allele was shown to be associated with a reduction in the NOS1 transcription. The role of NOS1 in IHPS awaits further investigation.
IHPS2 (16p13–p12) and IHPS5 (16q24) were identified using single large multiplex pedigrees and represent monogenic variants of IHPS. The existence of the monogenic form in a complex disease is not uncommon and may yet reveal a relevant disease pathway in the more common sporadic form. IHPS3 (11q14–q22) and IHPS4 (Xq23) were identified using a cohort of over 80 pedigrees, each of which has two or more affected individuals. It is of particular interest that one of these loci is on the X chromosome. A predisposing locus on the X chromosome will provide some of the explanation for the male preponderance of the condition. Each of these two loci harbours a gene of the same canonical transient receptor potential cation (TRPC) family of calcium ion channels. They have a well-recognised role in smooth muscle hypertrophy and are strong candidate genes for IHPS. These loci merely represent the first step in identifying the disease genes and/or the molecular changes that cause IHPS. Further work will involve large-scale sequencing, further association study and ultimately functional evaluation.
IHPS is indeed a good example of a complex disease in its pathophysiology and genetics. Although the aetiology of IHPS remains unclear, an increasing amount of data from epidemiological and genetic studies are beginning to provide glimpses of hope in unravelling some of the genetic and environmental factors.