The Pyloric Sphincteric Cylinder in Health and Disease



Go to chapter: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39
Chapter 23 (page 109)


According to Heinisch (l967) the etiology and pathogenesis of IHPS and AHPS remained obscure. Macroscopically and microscopically these conditions could not be differentiated from each other. As the pathological process was not limited to the pyloric ring, but also encompassed the immediate prepyloric area, Heinisch suggested that it should be termed "antrumhypertrophy" instead of pyloric hypertrophy. In his experimental work on rabbits, Heinisch attached glass spheres to the interior of the gastric fundus through a gastrostomy. This caused an "impairment of the functional unity" of the stomach; after 4 weeks a definite pyloric muscular thickening, which could not be distinguished from pyloric hypertrophy, occurred. Heinisch (l967) thought that the muscular hypertrophy was a compensatory mechanism due to the functional impairment of the proximal part of the stomach.

Smith (1970) noted that hypertrophic pyloric stenosis could occur in Chagas disease, in which myenteric plexus damage occurs. The general effect of this is loss of co-ordinated muscle contractions which propel the bolus analwards, leading to local hypertrophy and hyperplasia of smooth muscle.

Dodge (l970) found that pentagastrin injections of 3 pregnant bitches produced pyloric muscle hypertrophy and duodenal ulceration in some of the offspring.

Keet and Heydenrych (l97l) showed that electrical and mechanical stimulation of the vagal nerve trunks in the oesophageal hiatus of the diaphragm in canines, produced a temporary, tubular contraction of the pyloric sphincteric cylinder, approximately 3.0cm in length and exactly resembling hypertrophic pyloric stenosis. Increasing the stimulus caused the contraction to become firmer, until it became a rubbery hard cylinder simulating the "olive" of IHPS. It lasted as long as the stimulus was applied, while the remainder of the stomach remained flaccid. A similar result was obtained with mechanical stimulation (Chap. 32).

Dodge (l973), in a wide ranging review of the genetics of the condition, concluded that IHPS had a familial distribution but that this did not obey simple Mendelian laws. A single X-linked gene whose effects were greatly modified by other genetic and/or environmental factors would explain the increased risk to male relatives of female probands, but the excess of males with IHPS could not be accounted for by X-linkage, as male to male transmission occurred with greater frequency than expected. The most likely explanation for the distribution pattern was that multiple genetic and environmental factors, including a modifying effect of sex, operated together to produce IHPS. Female patients carried a heavier genetic load and were more likely to have affected male relatives. Suggested environmental factors were the maternal blood group, primogeniture, and the occurrence of stress shortly before birth.

Karim, Morrison and Parks (l974) administered daily pentagastrin injections to 12 bitches during the second half of pregnancy. Among 59 offspring 14 developed pyloric muscle hypertrophy, and 8 others pyloroduodenal ulcers. No muscle hypertrophy was seen in the controls.

Rogers et al. (l975) found no difference in circulating plasma gastrin levels in babies with and without IHPS, and concluded that gastrin stimulation was an unlikely cause of the muscle hypertrophy. An alternative hypothesis could be an abnormal response to intermittent milk feeding. This would release cholecystokinin and secretin from the duodenal mucosa, causing contraction, and ultimately hypertrophy, of the pylorus.

Dodge and Karim (l976) again found that pentagastrin administered to pregnant bitches produced typical pyloric muscular hypertrophy in some of their offspring. This occurred in 16 out of 56 puppies; some had duodenal ulceration. The findings indicated a variation in the response, with some pups showing no pathological changes while others were markedly affected. It suggested that individual inherent susceptibilities, such as genetic and environmental factors, were important determinants of the response.

In electron microscopic studies of the pyloric parasympathetic ganglia in IHPS, Jona (l978) found no abnormalities in the neurons and interstitial cells of the nerve plexuses; the maturation process of the ganglia proceeded normally. A small number of large axons, the significance of which was unclear, was present. While there was a marked increase in the overall thickness of the circular musculature, the individual muscle cells appeared normal. In view of the essentially normal morphology, Jona suggested that a "functional" etiology should be considered.

According to Swischuk (l980) evidence was accumulating to suggest that prolonged spasm or overactivity of the "antropyloric muscle" was the primary problem in infants with IHPS. It was probable that multiple causes of the muscular spasm could be operating.

While numerous factors had been proposed as the cause of IHPS, Haller and Cohen (l986) reiterated that the precise etiology remained uncertain.

Discussion

In the radiological examination, here as elsewhere, only the barium suspension in the lumen is visible, and from the luminal appearances the alterations in the walls are inferred (Chap. 13). With ultrasonography the actual muscle mass causing narrowing of the lumen can usually be demonstrated. Consequently ultrasound has become the investigation of choice for suspected IHPS in many centers. Pilling (l983) discussed the relative sensitivity of radiology and ultrasound in the diagnosis. Of 26 cases with palpable pyloric tumors the mass was not detected by ultrasound in 2; in these cases, as well as in cases without palpable masses, an upper gastrointestinal radiological series was deemed necessary.


Previous Page | Table of Contents | Next Page
© Copyright PLiG 1998