| Sažetak | Cilj istraživanja: Odrediti učinkovitost (valjanost) „Narrow-band imaging international colorectal endoscopic classification“ u razlikovanju podsluzničnog invazivnog karcinoma od adenoma te učinkovitost iste u razlikovanju adenoma od hiperplastičnih polipa. Odrediti učestalost polipa u pojedinim dijelovima debelog crijeva te odrediti rasprostranjenost pojedinih patohistoloških tipova polipa u pojedinim dijelovima debelog crijeva.
Ispitanici i metode: Provedeno je prospektivno opažajno istraživanje u kojem su uključena 122 pacijenta, 77 muškaraca i 45 žena, koji su zbog raznih indikacija upućeni na kolonoskopsku pretragu na Zavod za gastroenterologiju, Klinike za unutarnje bolesti Kliničkog bolničkog centra Split, Križine. Kolonoskopije su izvedene uz korištenje videokolonoskopa Olympus CF-H185L EVIS EXERA III i videokolonoskopa Olympus CF-H185I EVIS EXERA III. Svim ispitanicima zabilježen je spol, dob, broj polipa, veličina polipa i lokalizacija polipa. Svim otkrivenim polipima napravljena je endoskopska predikcija histološke slike koristeći NICE klasifikaciju, zatim su polipi odstranjeni i uzorci su poslani na patohistološku analizu koja je korištena kao zlatni standard. U obradi podataka korišten je programski paket SPSS Statistics for Windows, verzija 23.0 (IBM, Armonk, NY). Glavne mjere ishoda bile su negativne prediktivne vrijednosti za duboki podsluznični invazivni karcinom i adenom.
Rezultati: NICE klasifikacija pokazala se kao relativno učinkovita dijagnostička metoda u razlikovanju podsluzničnog invazivnog karcinoma od adenoma (osjetljivost je iznosila 28,6%, specifičnost 98,9%, pozitivna prediktivna vrijednost 66,7%, negativna prediktivna vrijednost 94,4% i točnost 93,6%), međutim nije pokazala dovoljnu učinkovitost u razlikovanju adenoma od hiperplastičnih polipa (osjetljivost je iznosila 82,5%, specifičnost 87%, pozitivna prediktivna vrijednost 96,6%, negativna prediktivna vrijednost 52,6% i točnost 83,3%). Od ostalih rezultata treba naglasiti činjenicu da je postojala statistički značajna razlika u veličini između hiperplastičnog polipa i adenoma (T= 7,64, p<0,001) i statistički značajna razlika u veličini između hiperplastičnog polipa i adenokarcinoma (T= 4,5, p=0,003). Nije postojala statistički značajna razlika u veličini između adenoma i adenokarcinoma (T= 1,92, p=0,089). Ukupno je odstranjeno 135 polipa - 38 u rektumu (28%), 63 u sigmoidnom kolonu (47%), 8 u silaznom kolonu (6%), 6 u poprečnom kolonu (4%), 9 u uzlaznom kolonu (7%) te 11 u cekumu (8%). Patohistološkom analizom uzoraka, 23 odstranjena polipa pokazala su se kao hiperplastični polipi (17%), 104 kao adenomi (77%) te 8 kao adenokarcinomi (6%). Nije postojala statistički značajna razlika između patohistološke dijagnoze polipa i lokalizacije polipa (p=0,245).
Zaključak: Ovo je istraživanje ukazalo na određene koristi NICE klasifikacije kao dijagnostičke metode u predikciji histološke slike kolorektalnih polipa in vivo te je svakako potrebno u idućim istraživanjima uključiti veći broj ispitanika i izbjeći nedostatke koje smo primijetili u našem istraživanju. |
| Sažetak (engleski) | Objectives: To assess the effectiveness of the „Narrow-band imaging international colorectal endoscopic classification“ in differentiating between deep submucosal invasive cancer and adenomas and to assess its effectiveness in differentiating between adenomas and hyperplastic polyps. Object is also to determine the frequency of polyps in individual parts of the large intestine and to determine the distribution of the pathohistological types of polyps in individual parts of large intestine.
Study design: Prospective observational study.
Patients and methods: Total of 122 patients were prospectively enrolled in this study, 77 males and 45 females. All of these patients were referred to colonoscopy due to various indications. Colonoscopies were performed using Olympus CF-H185L EVIS EXERA III and Olympus CF-H185I EVIS EXERA III colonoscopes. Patient's gender, age, number of polyps, size of each polyp and localization of each polyp were recorded. All of the detected polyps were viewed under NBI and after the endoscopic prediction of polyp histology using the NICE classification was made, polyps were resected and retrieved individually for pathological examination. The criterion standard for validation of real-time predictions was the pathologist's report of polyp histology. All statistical analysis were performed with SPSS statistics for Windows, version 23. The primary outcome measures for the predictive validity of the NICE classification for endoscopic assessment of the histology of colorectal polyps was the negative predictive value of an adenomatous prediction and the negative predictive value of deep submucosal invasive cancer.
Results: NICE classification has proven to be relatively effective diagnostical method in differentiating between deep submucosal invasive cancer and adenomas (the sensitivity was 28.6%, specificity was 98.9%, positive predictive value was 66.7%, negative predictive value was 94.4% and accuracy was 93.6%), however it has not proven to be effective in differentiating between adenomas and hyperplastic polyps (the sensitivity was 82.5%, specificity was 87%, positive predictive value was 96.6%, negative predictive value was 52.6% and accuracy was 83.3%). There was statistically significant difference in size between hyperplastic polyp and adenoma (T= 7.64, p<0.001) and statistically significant difference in size between hyperplastic polyp and adenocarcinoma (T= 4.5, p=0.003). There was no statistically significant difference in size between adenoma and adenocarcinoma (T= 1.92, p=0.089). Total of 135 polyps were resected - 38 in rectum (28%), 63 in sigmoid colon (47%), 8 in descending colon (6%), 6 in transverse colon (4%), 9 in ascending colon (7%) and 11 in cecum (8%). According to the pathohistological diagnosis of the retrieved samples, the prevalence of hyperplastic polyps amongst patients was 17% (23 polyps). The prevalences of adenomas and adenocarcinomas were 77% (104 polyps) and 6% (8 polyps), respectively. There was no statistically significant difference between the pathohistological diagnosis of polyps and localization of polyps (p=0.245).
Conclusion: This study has pointed out certain benefits of the NICE classification as a diagnostical method for real-time prediction of the colorectal polyp histology. Further clinical studies with greater number of patients included should be conducted to assess the generalizability of this classification. It is also necessary to avoid limitations noted in our study. |
