Differentiation of focal foveolar hyperplasia from hyperplastic polyps in gastric biopsy material

Path. Res. Pract. 191, 1198 -1202 (1995)

Differentiation of Focal Foveolar Hyperplasia from Hyperplastic Polyps in Gastric Biopsy Material M. Stolte, B. Bethke, Th. Sticht and U. Burkhard 1 Institute of Pathology, Klinikum Bayreuth, Bayreuth, Germany; 11nstitute for Statistics, ANFOMED, M6hrendorf, Germany

SUMMARY

Our objective was to investigate the question as to whether focal foveolar hyperplasia and hyperplastic polyp can be differentiated in forceps biopsy material from the stomach. Morphometric determination of the height of the epithelium layer in forceps biopsy specimens was obtained from 35 hyperplastic polyps, and forceps biopsy material was obtained from 25 focal foveolar hyperplasias. The diagnosis of hyperplastic polyp was confirmed by subsequent polypectomy. The medians and scatter range of the epithelial layer height were calculated. Using the t-test for independent samples the question was examined as to whether there is any statistically significant difference between hyperplastic polyps and focal foveolar hyperplasia in terms of the parameter "height of the foveolar epithelial layer. " The measurements revealed that the average height of foveolar epithelial cells in hyperplastic polyps is 37.70 j1m ± 7.41 j1m. In the case of focal foveolar hyperplasia, the corresponding figure was only 24.26 j1m ± 5.11 j1m. This difference was statistically highly significant (p < 0.0001). In conclusion, focal foveolar hyperplasia and hyperplastic polyp of the gastric mucosa can readily be differentiated on the basis of architectural and cytological criteria, even in forceps biopsy material. Since the hyperplastic polyp very probably does not evolve from focal foveolar hyperplasia, and the latter is not a pre-neoplastic condition or lesion, it is proposed that focal foveolar hyperplasia should no longer be referred to as "gastric polyp, " which would avoid unnecessary follow-up examinations and possibly even surgery.

Introduction In the nineteen-seventies, Muto 14 and Elster4 expressed the opinion that focal foveolar hyperplasia and the most common polyp found in the stomach, which they referred to as "hyperplasiogenous" polyp, could readily be differentiated. Subsequently, the hypothesis was advanced that these so-called "hyperplasiogenous" polyps evolve from focal foveolar hyperplasia of the gastric mucosa 11. For this reason, fo0344-0338/95/0191-1198$3.50/0

cal foveolar hyperplasia and "hyperplasiogenous" polyps were classified together as "hyperplastic polyps"7, 1l,23. As far as we know, a histological differential diagnosis of focal foveolar hyperplasia and "hyperplasiogenous" polyp has not previously been attempted in biopsy - especially forceps biopsy - material. A differentiation was undertaken only in polypectomy specimens 1o . In the present study, we show that focal foveolar hyperplasia and the larger hyperplastic polyp, which in © 1995 by Gustav Fischer Verlag, Stuttgart

Focal Foveolar Hyperplasia and Hyperplastic Polyps in the Stomach . 1199

this study we term "hyperplasiogenous" polyp as proposed by Elster4, can be differentiated not merely on the basis of architectural criteria involving the complete polypectomy specimen, but also on the basis of cellular characteristics of the foveolar epithelium. Results A comparison of the mean values plus standard deviations for the height of the foveolar epithelium in "hyperplasiogenous" polyps and in focal foveolar hyperplasia is shown in Table 1. Foveolar epithelial cells in "hyperplasiogenous" polyps are taller than those in focal foveolar hyperplasia (37.70)lm ± 7.41 )lm as compared with 24.26)lm ± 5.11 )lm). The results of the t-test of the normally distributed data are shown in Table 2. The differences in the measurements between the "hyperplasiogenous" polyp group and the focal foveolar hyperplasia group are statistically highly significant (p < 0.0001). These differences can be readily determined in routine diagnostic evaluation, even without morphometry, and may serve as a basis for the differential diagnosis. Discussion Elster4 expressed the opinion that focal foveolar hyperplasia represented a small polypoid lesion with no tendency to develop into a neoplasm, and accepted the view expressed by Muto 14 that the "hyperplasiogenous" polyp comprised a mixture of hyperplastic and blastomatous elements. In a later publicationS, however, Elster emphasized that carcinomas developed only extremely rarely in "hyperplasiogenous" polyps, that these polyps were not to be considered precancerous lesions and have, therefore, no "blastomatous" component. Ming lO,l1 was of the opinion that focal foveolar hyperplasia and "hyperplasiogenous" polyp represented a single entity, and that "hyperplasiogenous" polyps evolved from focal foveolar hyperplasia. Accordingly,

he classified these two lesions together under the term "hyperplastic polyps." This lumping together of focal foveolar hyperplasia and "hyperplasiogenous" polyps in the single group of hyperplastic polyps, as is still done in the latest edition of the WHO classification of tumours of the stomach24, leads to a confusion of different statistics on the frequency of "hyperplasiogenous" polyps, their location and their size. While in our own material we have diagnosed "hyperplasiogenous" polyps in 29% of a total of 5,515 gastric polyps22, their reported fre~uency in other statistics varies between 50% and 90% ,13. Only some of the publications dealing with polyps of the stomach list focal foveolar hyperplasia separately from "hyperplasiogenous" polyps. In these reports, the frequency of focal foveolar hyperplasia among all stomach polyps varies between 5.0% and 95.1 %, that of "hyperplasiogenous" polyps between 2.6% and 68.1 %1,2,8,9,15,16,17,18,19. We have been unable to observe the evolution of a "hyperplasiogenous" polyp from focal foveolar hyperplasia, nor have we found any convincing evidence for this in the literature. If the "hyperplasiogenous" polyp actually does evolve from focal foveolar hyperplasia, it ought to be detected mainly in the antrum, for this is the predilection site for focal foveolar hyperplasia which develops during the course of healing of the mucosal lesions occurring in Helicobacter pylori gastritis 21 or chemically induced gastritis, and which may persist as a regenerative remnant. Although focal foveolar hyperplasia is also seen in association with type A gastritis, this latter, which accounts for about 3% to 6% of all cases of gastritis, is much less common than Helicobacter pylori gastritis21. The most common site of "hyperplasiogenous" polyps, however, is not the antrum but the corpus, where 49% of the 2,000 "hyperplasiogenous" polyps we evaluated were located9, only 38% being found in the antrum. This distribution of site also shows that the earlier classification of "hyperplasiogenous" polyps as "regenerative polyps"12 cannot be correct, since most gastric mucosal lesions, and thus also focal regenerative tissue, are found in the antrum. On the basis of these observations, we believe that "hy-

Table 1. Comparison of mean value, median value, standard deviation (SD), standard error (SE) and range of the data of measurements of heights of foveolar epithelia (in /lm) of hyperplasiogenous polyps (HPP) and focal foveolar hyperplasia (FFH)

HPP FFH

Median

Minimum

Maximum

1.25

37.19

25.69

60.69

1.02

22.87

16.74

34.77

N

Mean

SD

SE

35

37.70

7.41

24.26

5.11

25

Table 2. Comparison of mean values of the heights of foveolar epithelia: results of the t-test for independent samples (two tailed) Variable

F-test F-value

Prob.

t-test t-value

Prob.

DF

Value

2.10

0.0607

8.32

0.0001

57

1200 . M. Stolte et al.

Figs. 1-2. Even in the routine diagnostic evaluation without morphometry, it can be noted that the foveolar epithelium in hyperplasiogenous polyps (left) is taller than that in foval foveolar hyperplasia (right). Haematoxylin & eosin same magnification. perplasiogenous" polyps and focal foveolar hyperplasia should no longer be classified together as a single histological entity. Since neither benign nor malignant neoplasias develop from focal foveolar hyperplasia - that is, this condition represents a harmless regenerative process - and since the gastric polyp statistics are falsified by the high rate of focal foveolar hyperplasia, we are in agreement with Elster et a1. 5 that this lesion should no longer be recorded in the register of gastric polyps, thus avoiding unnecessary follow-up examination, and possibly even unnecessary surgery. This proposal, however, presupposes that focal foveolar hyperplasia and "hyperplasiogenous" polyp can be differentiated by means of endoscopy and biopsy. Endoscopically, focal foveolar hyperplasia can be distinguished from the "hyperplasiogenous" polyp on the basis of site, size and superficial structure. Focal foveolar hyperplasia is usually located within the antrum, is often multiple, lentil-, pea-, or bean-sized, and has a non-eroded surface. The "hyperplasiogen-

ous" polyp is mainly solitary, more commonly located in the corpus than in the antrum, is larger than the focal foveolar hyperplastic lesion and almost always shows focal surface erosion 22 • The histological differential diagnosis on the basis of endoscopic polypectomy specimens is simple: while focal foveolar hyperplasia reveals only lengthening of the foveolae, but normal architecture, the "hyperplasiogenous" polyp is characterized by the following changes in tissue architecture: - lengthening, tortuosity and variable cystic dilatation of the foveolae, - widening of the stroma and oedema, with increased and hyperplastic bundling of smooth muscle cells arranged vertically in the lamina propria, - increased numbers, and dilation, of the capillaries in the lamina propria, and - apical erosions. Most of these architectural changes, however, are not identifiable in forceps biopsy material. In our daily routine diagnostic evaluation work, we noted that focal foveolar hyperplasia and "hyperplasiogenous" polyp

Focal Foveolar Hyperplasia and Hyperplastic Polyps in the Stomach . 1201 can also be distinguished on the basis of characteristic cytological changes of the epithelium of the foveolae. We observed an increase in width and height of foveolar epithelial cells in "hyperplasiogenous" polyps, together with an increase in the production of mucus. In contrast, the foveolar epithelium in focal foveolar hyperplasia cannot be distinguished from normal foveolar epithelium. And this equally applies to foveolar epithelium in chemical gastritis3,2o, which is also not taller and shows no increase in mucus production. In contrast to focal foveolar hyperplasia, however, in chemical gastritis there are no endoscopically detectable tiny polypoid elevations, the foveolar tortuosity is diffusely represented in all antral biopsies and the lengthening of the foveolae is not so pronounced. In comparison with the morphology of the "hyperplasiogenous" polyp, the features characteristic of chemical gastritis like oedema of the lamina propria, fibrosis and increased smooth muscle fibres in the lamina propria are much less marked and more homogeneous than in "hyperplasiogenous" polyp. In order to investigate our impression that an assessment of the height of the foveolar epithelium can be helpful for the differential diagnosis between focal foveolar hyperplasia and "hyperplasiogenous" polyp in forceps biopsy material obtained from these polypous lesions, we undertook the present morphometric study on randomly mixed forceps biopsy material obtained from "hyperplasiogenous" polyps and focal foveolar hyperplasia. The diagnosis of "hyperplasiogenous" polyp was confirmed by the subsequent histological assessment of the endoscopic polypectomy specimen. The diagnosis was established and morphometry done, respectively, by two different investigators. A comparative statistical evaluation of the two groups revealed that the cytological criteria are indeed most suitable for establishing a differential diagnosis between focal foveolar hyperplasia and "hyperplasiogenous" polyp, for the differences in the numerical figures for the two groups were statistically highly significant. This means that our suggestion to stop classifying focal foveolar hyperplasia as a gastric polyp, can be realized on the basis not only of endoscopic criteria and the histological work-up of the polypectomy specimen, but also of the histological differential diagnostic examination of forceps biopsy material - without the need for morphometry. Since the lesion termed "hyperplasiogenous" polyp by both Muto 14 and Elster 4 contains no "blastomatous" elements as originally thought, the designation "hyperplasiogenous," still used in many parts of the world, should be replaced by "hyperplastic." The term hyperplastic polyp, however, should, in the future, be used only to describe the former "hyperplasiogenous" polyp, but not focal foveolar hyperplasia. This also has practical consequences for patient care: focal foveolar hyperplasia is a harmless regenerative process, and further endoscopy/biopsy follow-up examinations are not required. A stomach containing hyperplastic polyps, however, is a precancerous condition 18 , so that

appropriate follow-up with endoscopy/biopsy commended.

IS

re-

Material and Methods Two series of forceps biopsy specimens obtained from the stomach via the endoscope were investigated: Group 1: Forceps biopsy material obtained from 35 "hyperplasiogenous" polyps which, after primary diagnostic evaluation of the biopsy material, were removed by polypectomy, thus permitting the diagnosis of "hyperplasiogenous" polyp to be confirmed. Group 2: Forceps biopsy material from endoscopically detected pea- to bean-sized elevations of the gastric mucosa, histologically diagnosed as focal foveolar hyperplasia, was obtained in 25 patients. A total of 60 histological sections (fixation of the material in 4% buffered formalin solution, paraffin-embedding, staining of the 3 to 5 Il-thick sections with haematoxylin & eosin) were randomly mixed and submitted to a morphometric examination by another investigator blinded to the original diagnostic classification or the subsequent classification performed on the polypectomy specimens, as applicable.

Morphometry For the measurement of the height of the foveolar epithelium, optimally oriented parts of the histological sections containing longitudinally cut foveolae (see Figures 1 and 2) were selected. Within the various parts of the foveolae, epithelial cells were selected for measurement from the lower, middle and upper thirds. Measurement was performed only on those cells that had a cylindrical configuration and basally located nuclei, and thus very probably represented vertically cut cells. Using these selection criteria, 25 to 30 foveolar cells were identified for morphometry in each case. Measurement of the height of the epithelium was effected with the aid of a Leitz CBA Image Analysis System, and median values and standard deviations calculated.

Statistical Comparison of the Measured Results On the basis of the measured values of the two groups, the t-test for independent samples was employed to determine whether the measured values for the height of the foveolar epithelium differed statistically significantly between focal foveolar hyperplasia and "hyperplasiogenous" polyps. References 1 Bosseckert H, Raabe G (1983) Multiple polyps in the stomach - how many should be ectomized. Endoscopy 15: 150-151 2 Bosseckert H, Kratzsch KH, Machnik G, Raabe G, Waller H, Winkelvoss H, Stibenz J, Eichhorn K, Lichteblau A (1988) Hyperplasiogene Polypen und Magenkarzinomrisiko. Dtsch Z Verdau- und Stoffwechs. Krankh. 48: 1-9 3 Dixon MF, O'Connor HJ, Axon ART, King RFJG, Johnston D (1986) Reflux gastritis: distinct histopathological entity. J Clin Pathol 39: 524-530 4 Elster K (1974) A new approach to the classification of gastric polyps. Endoscopy 6: 44-47 5 Elster K, Carson W, Eidt H, Thomasko A (1983) Significance of gastric polypectomy (histological aspect) Endoscopy 15: 148-149

1202 . M. Stolte et al. 6 Fenoglio-Preiser CM, Lantz P, Listrom M, Davis M, Rilke F (1988) Gastrointestinal Pathology. Raven Press, New York 178-189 7 Johansen A, Rubio CA (1989) Other tumours of the stomach. In: Whitehead R (Ed) Gastrointestinal and Oesophageal Pathology. Churchill Livingstone, Edingburgh 711-732 8 Koch H, Lesch R, Cremer M, Oehlert W (1979) Polyps and polypoid foveolar hyperplasia in gastric biopsy specimens and their precancerous prevalence. Front gastrointest Res 4: 183-191 9 Koch H, Viebahn B (1983) The significance of polypectomy in the stomach. Endoscopy 15: 144-147 10 Ming SC (1977) The classification and significance of gastric polyps. Int Acad pathol Monogr 18: 149-175 11 Ming SC (1977) The classification and significance of gastric polyps in the gastrointestinal tract. In: Yardley JH, Morson BC, Abell MR (Eds) Williams and Wilkins, Baltimore 12 Morson BC, Dawson IMP (1979) Gastrointestinal Pathology. Blackwell, Oxford 140-147 13 Morson BC, Dawson IMP (1990) Morson & Dawson's Gastrointestinal Pathology. 3rd Edition Blackwell Scientific Publications, Oxford 134-142 14 Muto T, Oata K (1970) Polypogenesis of gastric mucosa. Gann (now Jpn J Cancer Res) 61: 435 -442 15 Nakamuro T (1970) Pathohistologische Einteilung der Magenpolypen. Chirurg 41: 122-127

16 Remmele W, Pfannkuche (1979) Epitheliale Polypen und Driisenkorperzysten der Magenschleimhaut. Pathologe 1: 25-39 17 Rosch W (1980) Epidemiology, pathogenesis, diagnosis and treatment of benign gastric tumours. Front gastrointest Res 6: 167-194 18 Seifert E, Gail K, Weismiiller J (1983) Gastric polypectomy - long-term results (survey of 23 centers in Germany) Endoscopy 15: 8-11 19 Snover DC (1985) Benign epithelial polyps of the stomach. Pathology Annual 20: 228-230 20 Sobala GM, King RFG, Axon ATR, Dixon MF (1990) Reflux gastritis in the intact stomach. J Clin Pathol43: 303306 21 Stolte M, Bethke B, Ritter M, Lauer E, Eidt H (1990) Praxis der Gastritis-Klassifikation. Endoskopie heute 4: 228-230 22 Stolte M, StichtT, Eidt S, Ebert D, Finkenzeller G (1994) Frequency, location, and age and sex distribution of various types of gastric polyp. Endoscopy 26: 659-665 23 Tomasulo J (1971) Gastric polyps. Histologic types and their relationship to gastric carcinoma. Cancer 27: 13461355 24 Watanabe H,JassJR, Sobin LH (1990) Histological typing of oesophageal and gastric tumours. Springer, Berlin Heidelberg

Received June 1, 1994 . Accepted in revised form September 12, 1995

Key words: Gastric polyps - Focal foveolar hyperplasia - Hyperplastic polyp Prof. Dr. M. Stolte, Institut fiir Pathologie, Klinikum Bayreuth, Preuschwitzer StraBe 101, D-95445 Bayreuth, Germany