Inter-observer agreement on histological diagnosis of
colorectal polyps: the APACC study
Hachung YOON (1), Antoine MARTIN (2), Robert BENAMOUZIG (1), Elisabeth LONGCHAMPT (2), Jacques DEYRA (3),
Stanislas CHAUSSADE (3) for the APACC study group
(1) Service de Gastro-Entérologie, (2) Service d’Anatomie-Pathologique, Hôpital Avicenne, 125, rue de Stalingrad, 93009 Bobigny Cedex, France ; (3) Service
de Gastro-Entérologie, Hôpital Cochin, 27, rue de Fbg Saint-Jacques, 75679 Paris Cedex 14.
SUMMARY
Introduction — Adenomas are considered as precursors of colorectal adenocarcinomas. After polypectomy, surveillance depends on the
histological analysis of the excised polyps. The aim of the study was to evaluate the reproducibility of the histological diagnosis performed by
pathologists from participating centers in the APACC study. Their interpretations were compared to those of the study’s consensual diagnosis
performed by two reference pathologists.
Methods — The agreement proportion and the concordance coefficient (kappa) were estimated for 326 polyps obtained from the first 148 patients
included.
Results — A concordant diagnosis for the adenomatous nature was shown for 246 of the 255 adenomas (96%). The diagnosis of both the
histological type and the degree of dysplasia was concordant for only 92 of the 255 adenomas (36%). Kappa values were 0.67 for the
adenomatous nature, 0.46 for the histological types, and 0.26 for the degree of dysplasia.
Conclusion — A high degree of agreement was obtained for the diagnosis of the adenomatous nature. By contrast, high variations were observed
for the interpretations of the histological type and the degree of dysplasia.
To cite the present paper, use exclusively the following reference. Yoon H, Martin A, Benamouzig R, Longchampt E, Deyra J,
Chaussade S pour le groupe d’étude APACC. Reproductibilité inter-observateurs du diagnostic anatomo-pathologique des polypes
colorectaux. Données des centres participant à l’étude APACC (full text in english on www.e2med.com/gcb). Gastroenterol Clin Biol
2002;26:220-4.
C
ancer of the colon and rectum, with more than
33,500 new cases annually, is the most frequent
cancer in France, in men and women alike [1].
Adenomas are considered as precursors of colorectal adenocar-
cinomas in the majority of cases. Currently, their detection and
resection is the only way to prevent malignant transformation [2].
The risk of malignant transformation of adenomas depends on
the size, the villous component and the degree of dysplasia [3].
Conversely, the risk of developing cancer in patients with a
limited number of hyperplastic polyps is not higher than in the
general population [4]. Although the follow-up schedule after
polypectomy largely depends on pathology results, there has
been however little work concerning the reproducibility of
histological interpretation of colonic adenomas [5-10], and no
study has been conducted in in France. The purpose of this work
was to evaluate the reproducibility of histological diagnosis
initially performed by pathologists in French centers participating
in the APACC (Association pour la Prévention par l’Aspirine du
Cancer Colorectal) chemoprevention study, considered as “gen-
eral” pathologists working in hospital or private practice, and
comparing it to the diagnosis established by 2 referent patholo-
gists, who are coordinators of the study.
Material and methods
APACC study
Data used here were collected from the APACC study, a randomized
double-blind trial designed to determine the efficacy of regular aspirin
intake in reducing the rate of recurrence of colonic adenomas. The
APACC study was approved by the Saint-Germain-en-Laye Hospital
ethics committee (Comité Consultatif de Protection des Personnes pour la
Recherche Biomédicale, CCPPRB n° 96006) and the National Commis-
sion of Liberty and Information (CNIL). The complete description of the
APACC study can be found elsewhere [11]. We recall briefly the study
design and population. Eligible subjects were aged 18 to 75 years at the
time of colonoscopy which identified at least one histologically-proven
adenoma. Two hundred ninety-one selected subjects were given a daily
dose of 300-mg lysine acetylsalicylate for 4 weeks. During this pre-
inclusion period, non-compliant patients and patients exhibiting intoler-
able side effects were excluded from the study. Thereafter, 270 subjects
were included and randomly assigned to three treatment groups:
placebo, lysine acetylsalicylate 160 mg/day or lysine acetylsalicylate
300 mg/day for 4 years. The number and type of polyps detected at
colonoscopies performed at 1 year ± 2 months and 4 years ± 3 months
were the outcome criteria.
Histological diagnosis of polyps
All polyps detected at the selection colonoscopy were removed. Polyp
specimens were fixed in buffered formalin or Bouin’s fixative. These
specimens were sent to pathologists working in the participating centers
for initial interpretation. Each polyp removed was entirely embedded .
Depending on the size of the polyp, the pathologists examined one or two
Tirés à part : R. BENAMOUZIG, CREF-APACC, Pavillon Cornil-Brissaud,
Hôpital Cochin, 27, rue du Fbg St Jacques, 75014 Paris.
E-mail: robert.benamouzig@avc.ap-hop-paris.fr
© Masson, Paris, 2001. Gastroenterol Clin Biol 2001;25:220-224
220
© 2021 Elsevier Masson SAS. Tous droits réservés. - Document téléchargé le 16/12/2021 Il est interdit et illégal de diffuser ce document.