Clinical manifestations and endoscopic presentations of gastric lymphoma: a multicenter seven year retrospective survey

Transcription

1 /2017/109/8/ Revista Esañola de Enfermedades Digestivas Coyright SEPD y ARÁN EDICIONES, S.L. Rev Es Enferm Dig 2017, Vol. 109, N.º 8, ORIGINAL PAPERS Clinical manifestations and endoscoic resentations of gastric lymhoma: a multicenter seven year retrosective survey Xianghua Cui 1,2, Tao Zhou 2,3, Dalei Jiang 1, Huiya Liu 4, Jian Wang 5, Shengan Yuan 6, Hongyun Li 7, Peng Yan 8 and Yanjing Gao 2 1 Deartment of Gastroenterology. Qingdao Municial Hosital. Qingdao, China. 2 Deartment of Gastroenterology. Qilu Hosital of Shandong University. Jinan, China. 3 Deartment of Gastroenterology. Linyi Mental Health Center. Linyi, China. 4 Deartment of Gastroenterology. Heze Municial Hosital. Heze, China. 5 Deartment of Gastroenterology. Linyi Peole s Hosital. Linyi, China. 6 Deartment of Gastroenterology. Zibo Central Peole s Hosital. Zibo, China. 7 Deartment of Gastroenterology. Jining Peole s Hosital. Jining, China. 8 Deartment of Gastroenterology. Liaocheng Second Peole s Hosital. Liaocheng, China ABSTRACT Background and aim: To imrove the diagnostic rate of gastric lymhoma by analyzing clinical and endoscoic features of atients with gastric lymhoma and susected gastric lymhoma. Methods: Clinical and endoscoic records of atients with gastric lymhoma (ositive ) and 133 atients with susected gastric lymhoma but subsequent non-malignant athology (negative ) were analyzed retrosectively. Data from another 99 gastric lymhoma atients with malignant athology but nonsecific endoscoy (endoscoy non-susect ) were analyzed. Results: Abdominal ain was the redominant symtom reorted in both the ositive and negative lymhoma s, reresenting 60.0 and 52.5%, resectively. No significant differences in age, sex and clinical manifestations in subjects from the two s were found. In the ositive, 54.3% were ulcerative; 34.3%, infiltrative; 8.5%, olyoid; and 2.9%, granulonodular. In the negative, 52.6% were infiltrative; 42.1%, ulcerative; 4.5%, granulonodular; and 0.75%, olyoid. The endoscoic results varied between the two s ( < 0.05). In the non-susect, 66.7% were ulcerative; 17.2%, infiltrative; 14.1%, olyoid; and 2.0%, granulonodular. With regards to histology, diffuse large B cell lymhoma was the most common subtye. The sensitivity of endoscoy was 60% for detecting malignancy and 21% for gastric lymhoma. Conclusion: The resent study suggests that gastric lymhoma and susected gastric lymhoma have similar clinical features. Gastric lymhoma resented mainly as macroscoic ulcerative lesions, whereas susected gastric lymhoma aeared mainly as infiltrative lesions. Although the diagnostic rate of gastric lymhoma was relatively low (21%), it can be identified by endoscoy (60%). To imrove diagnosis, reetitive endoscoic biosies should be erformed and novel endoscoic techniques develoed in the future. Key words: Gastric lymhoma. Clinical features. Endoscoy. Pathology. Diagnosis. reresenting ~50% of all PGI NHL (1-3) and 3-5% of gastric malignancies (4,5). In recent years, the incidence of gastric lymhoma has been increasing worldwide, notably in westernized countries (6,7). A recent study involving a large adult North American oulation demonstrated an overall increase in the yearly incidence of PGI NHL from 0.13 to 2.10 er 100,000 from 1999 to Gastric lymhoma also dislayed a general increase in incidence during the study eriod (7). The clinical manifestations of gastric lymhoma are often vague and indistinguishable from other benign or malignant gastric tumors (8,9). The endoscoic atterns of gastric lymhoma are also varied and non-secific, and may range from simle mucosal changes to a resemblance to adenocarcinoma, thus leading to missed diagnosis or a misdiagnosis (10,11). Doglioni et al. (12) found that the endoscoic study of gastric lymhoma is indicative in only one third of cases. Therefore, the diagnosis of gastric lymhoma remains a challenge, esecially in the early stages of the disease, and thus decreases the likelihood of a successful management (12). Although several recent studies (2,13,14) have addressed the clinical and endoscoic features of gastric lymhoma, limited data are available in order to make an accurate clinical diagnosis of gastric lymhoma by endoscoic or other techniques in susect cases. Thus, we erformed a retrosective multicenter survey to comare the clinical manifestations and endoscoic features of atients with gastric lymhoma and susected gastric lymhoma in eight hositals in different cities in the Shandong rovince in China from June 2008 to December The aim of the study was to develo imroved criteria for the diagnosis of gastric lymhoma. INTRODUCTION Gastric lymhoma is one of the most common tyes of rimary gastrointestinal non-hodgkin lymhoma (PGI NHL), METHODS This was a retrosective, multicenter study involving eight hositals in eight cities in the Shandong rovince in China. The study Received: Acceted: Corresondence: Yanjing Gao. Deartment of Gastroenterology. Qilu Hosital of Shandong University. Wenhua Xi Road, Jinan, P.R. China yanjinggao@medmail.com.cn Cui X, Zhou T, Jiang D, Liu H, Wang J, Yuan S, Li H, Yan P, Gao Y. Clinical manifestations and endoscoic resentations of gastric lymhoma: a multicenter seven year retrosective survey. Rev Es Enferm Dig 2017;109(8): DOI: /reed /2017

2 2017, Vol. 109, N.º 8 CLINICAL MANIFESTATIONS AND ENDOSCOPIC PRESENTATIONS OF GASTRIC LYMPHOMA: 567 A MULTICENTER SEVEN YEAR RETROSPECTIVE SURVEY was aroved by the Ethics Committee of the Shandong University Qi-Lu Hosital and followed standard ethical guidelines. A total of 267 atients were identified between 2008 and 2015, and their relevant clinical, endoscoic and athologic data were analyzed. Patients with an uncertain diagnosis or < 18 years of age at diagnosis were excluded from the study. Subjects included those with gastric lymhoma diagnosed by endoscoy (endoscoy susected, n = 168) and those who had athologically confirmed gastric lymhoma which was not susected by endoscoy (endoscoy non-susected, n = 99). For endoscoic assessment, the redominant endoscoic attern, texture and tumor site of each subject were retrieved and carefully evaluated. The endoscoic attern was classified as ulcerative (single, multile, and diffuse), olyoid, granulonodular or infiltrative (distortion, thickening mucosal folds and oor eristalsis) (15) (Fig. 1). For histological assessment, each biosy was immunohistochemically stained for CD20 and CD3 and diagnosed by an exerienced athologist according to the 4 th edition of the World Health Organization (WHO) criteria (14). Furthermore, additional staining for CD10, CD30, CD43, CD79a, CK19, bcl-2, and Ki-67 (12) were erformed in selected cases. The articiating centers and the number of cases for each center were as follows: Qilu Hosital of Shandong University, Jinan city (87 and 58 cases); Linyi Peole s Hosital, Linyi city (38 and 12 cases); Qingdao Municial Hosital, Qingdao city (17 and nine cases); Zibo Central Peole s Hosital, Zibo city (12 and five cases); Jining Peole s Hosital, Jining city (five and seven cases); Binzhou Peole s Hosital, Binzhou city (five and two cases); Heze Municial Hosital, Heze city (three and zero cases); and Liaocheng Peole s Hosital, Liaocheng city (one and six cases). Statistical analysis The Student s t-test was alied for aired observations or the Chi-squared test as aroriate. All statistical analyses were erformed using the SPSS statistical software (version 18.0; SPSS Inc., Chicago, IL, USA) with a statistical significance level set at < Fig. 1. Endoscoic atterns. A. Ulcerative tye. B. Polyoid tye. C. Granulonodular tye. D. Infiltrative tye. RESULTS Clinical characteristics In total, 168 atients had susected gastric lymhoma by endoscoy, the athology was conformed in atients (ositive sub) and 133 atients had a non-confirmed athology (negative sub). The endoscoy susected was comrised of 101 males and 67 females and the male to female ratio was 1.51 to 1. The mean age at diagnosis was 52.1 (range, 19 to 80), and 90% of cases were over 40 years of age. The ositive sub was comrised of 20 males and 15 females with a mean age of 52.5 (range, 23 to 80). The negative sub consisted of 81 males and 52 females and the male to female ratio was 1.56 to 1. The average age at diagnosis was 52.0 (range, 19 to 80) (Table 1). There was no significant difference between the age and sex distribution in the two subs ( > 0.05) (Table 1). Abdominal ain or discomfort was the redominant symtom in both subs, reresenting 60.0% (15/25) and 52.5% (53/101) of cases, resectively. Dysesia or bloating were less frequently reorted and were resent in 16.0% (4/25) and 17.8% (18/101) of cases, resectively. Weight loss and heartburn had a similar occurrence in both s, whereas gastrointestinal bleeding was rarely Table 1. Clinical characteristics of atients in the endoscoy susected Endoscoy susected Clinical characteristic No. Gender (n, %) Male < Female < Age (years) Mean Range Clinical symtoms/signs (n, %) Abdominal ain/discomfort Dysesia/bloating Weight loss Heartburn Gastrointestinal bleeding Positive 20 4 (20.0) 3 (15.0) 8 (40.0) 4 (20.0) 1 (5.0) 15 3 (20.0) 4 (26.7) 2 (13.3) 2 (13.3) 4 (26.7) (60.0) 4 (16.0) 3 (12.0) 2 (8.0) 1 (4.0) Negative (60.9) 12 (9.0) 17 (12.8) 28 (21.1) 19 (14.3) 5 (3.8) 52 (39.1) 14 (26.9) 10 (19.2) 7 (13.5) 12 (23.1) 9 (17.3) (52.5) 18 (17.8) 12 (11.9) 10 (9.9) 8 (7.9)

3 568 X. CUI ET AL. Rev Es Enferm Dig resent in both s. There was no significant difference in the clinical manifestations in the two s ( > 0.05) (Table 1). Endoscoic resentations The endoscoic characteristics, including gross endoscoic atterns, texture and redominant involvement sites, of the endoscoy susected and the endoscoy non-susected are listed in tables 2 and 3. Endoscoy susected In the ositive sub, the most common endoscoic attern was the ulcerative tye in 19 (54.3%) cases (single: eleven; multile: seven; and diffuse: one) infiltrative in 12 (34.3%) cases, olyoid in three (8.5%) cases and granulonodular in one (2.9%) case. In the negative sub, the main endoscoic attern was infiltrative in 52.6% (70/133), ulcerative (single: 32; multile: 23; and diffuse: one) in 42.1% (56/133), granulonodular in 4.5% (6/133) and olyoid in 0.75% (1/133) of cases. With regard to the texture of biosy secimens on endoscoic examination, the main texture was stiff in 15 cases (48.4%), cris in eight (25.8%) cases, tenacious in Table 3. Endoscoic features of atients with gastric lymhoma Endoscoic features Endoscoic attern (n, %) Ulcerative Single Multile Diffuse Polyoid Granulonodular Infiltrative Texture* Soft Cris Tenacious Stiff Predominant site Solitary site Corus Antrum Fundus Cardia Multile sites Pathology confirmed (ositive ) 19 (54.3) 11 (31.4) 7 (20.0) 3 (8.5) 12 (34.3) 31 4 (12.9) 8 (25.8) 4 (12.9)\ 15 (48.4) 25 (71.4) 14 (40.0) 9 (25.6) 10 (28.6) Endoscoy non-susected (66.7) 45 (45.5) 13 (13.1) 8 (8.1) 14 (14.1) 2 (2.0) 17 (17.2) 81 6 (7.5) 29 (.8) 14 (17.3) 32 (39.4) (76.7) 38 (38.4) 34 (34.3) 3 (3.0) 1 (1.0) 23 (23.3) *Data were unavailable in four cases of the athology confirmed and 17 cases of the endoscoy non-susected. Table 2. Endoscoic features of atients in the endoscoy susected Endoscoic features Endoscoic attern (n, %) Ulcerative Single Multile Diffuse Polyoid Granulonodular Infiltrative Texture* Soft Cris Tenacious Stiff Predominant site Solitary site Corus Antrum Fundus Cardia Multile sites Endoscoy susected Positive 19 (54.3) 11 (31.4) 7 (20.0) 3 (8.5) 12 (34.3) 31 4 (12.9) 8 (25.8) 4 (12.9)\ 15 (48.4) 25 (71.4) 14 (40.0) 9 (25.6) 10 (28.6) Negative (42.1) 32 (24.1) 23 (17.3) 1 (0.75) 1 (0.75) 6 (4.5) 70 (52.6) (18.5) 27 (26.2) 44 (42.7) 13 (12.6) (64.7) 45 (33.9) 26 (19.5) 13 (9.8) 2 (1.5) 47 (.3) *The data were unavailable in four cases of the ositive and 30 cases of the negative. four (12.9%) and soft in four (12.9%) cases in the ositive sub. In the negative sub, the main texture was tenacious in 44 (42.7%), cris in 27 (26.2%), soft in 19 (18.5%) and stiff in the remaining 13 (12.6%) cases. There were significant differences in the endoscoic tyes and texture of the two subs ( < 0.05) (Table 3). The redominant site of lesions was frequently the corus of the stomach in both subs, with 14 (40.0%) cases and 45 (33.9%) cases, resectively. Diffuse involvement was observed in ten (28.6%) and 47 (.3%) cases, resectively. Antral involvement was noted in nine (25.6%) and 26 (19.5%) cases resectively and lesions in the cardia/ fundus were found in two (5.8%) and 15 (11.3%) cases, resectively. The distribution of sites was similar between the two subs ( > 0.05) (Table 3). Endoscoy non-susected The features of the endoscoy non-susected are detailed in table 3. With regard to the texture of biosy secimens, the main texture was stiff in 32 (39.4%) cases, cris in 29 (.8%) cases, tenacious in 14 (17.3%) cases and soft in six (7.5%) cases. The main endoscoic attern was ulcerative in 66.7% (66/99) (single: 45; multile: 13; diffuse: eight), infiltrative in 17.2% (17/99), olyoid in 14.1% (14/99) and granulonodular in 2.0% (2/99) of cases.

4 2017, Vol. 109, N.º 8 CLINICAL MANIFESTATIONS AND ENDOSCOPIC PRESENTATIONS OF GASTRIC LYMPHOMA: 569 A MULTICENTER SEVEN YEAR RETROSPECTIVE SURVEY The main sites involved were the corus of the stomach in 38 (38.4%) cases, the antrum in 34 (34.3%), multile involved sites in 23 (23.3%) and cardia/fundus in four (4.0%) cases. There were no significant differences in the endoscoic features between the endoscoy non-susected and athology confirmed (ositive ) ( > 0.05). Imortantly, we found that the lesions with irregularly ulifted mucosa and a central ulceration or with multile suerficial ulcerations were often suggestive of gastric lymhoma in both samles. Histological subtye The histological subtyes of atients with gastric lymhoma are detailed in table 4. B cell gastric lymhoma comrised 97.1% (34/) and 97.0% (96/99) of cases in the athology confirmed sub and endoscoy non-susect, resectively. Among the B cell lineage, the main histological subtye was diffuse large B cell lymhoma (DLBCL) in both s (65.8 vs 63.6%), followed by mucosal-associated lymhoid tissue (MALT) lymhoma (25.8 vs 28.3%) and other tye B lymhoma (5.6 vs 5.1%). Furthermore, there was no significant difference in the histological subtye of gastric lymhoma atients between the two s ( > 0.05) (Table 4). Furthermore, the histoathology of the non-gastric lymhoma cases was evaluated in the endoscoy susected (Fig. 2). Chronic inflammation (63%) was the most common athological lesion, followed by oorly differentiated adenocarcinoma (23%), signet-ring cell carcinoma (10%), well-differentiated adenocarcinoma (1%), adenoma (1%) and others (2%). The diagnostic sensitivity of endoscoy A total of atients (20.8%, /168) in the endoscoy susected had a gastric lymhoma that was verified by athological analysis. Therefore, the diagnostic sensitivity of endoscoy was only 20.8% in the study. According to the gross morhologic aearance of the lesion, in 59 of 99 atients (60%) it was characterized as Table 4. Histological subtye of atients with gastric lymhoma Histological subtye B lymhoma (n %) DLBCL MALT lymhoma MCL Pathology confirmed (Positive ) 34 (97.1) 23 (65.8) 9 (25.8) 1 (2.8) Endoscoy non-susected 96 (97.0) 63 (63.6) 28 (28.3) 0 (0.0) Others 1 (2.8) 5 (5.1) T lymhoma (n, %) 1 (2.8) 3 (3.0) Fig. 2. Histological characteristics of atients in the athology non-confirmed (negative ). malignant in the endoscoy non-susect and 32 of 99 atients (32%) were diagnosed with a gastric ulcer (with an unidentified nature). Thus, this method exhibited a sensitivity of 60% for the detection of malignancy. DISCUSSION In this retrosective multicenter study, we comared the endoscoic resentations and clinical manifestations of atients with gastric lymhoma and susected gastric lymhoma. We found that age, sex distribution and clinical manifestations were similar in both s. With regard to the macroscoic attern, subjects with gastric lymhoma had ulcerative lesions in both the athology confirmed sub (ositive sub) and the endoscoy non-susected. On the other hand, in the susected gastric lymhoma the main attern was the infiltrative tye in the athology non-confirmed sub (negative sub). The clinical manifestations of atients with gastric lymhoma were often obscure and non-secific. In this study, the most common symtoms were abdominal ain or discomfort (51.2%) and dysesia (18.4%) in both the endoscoy susected and the endoscoy non-susected. These data are consistent with other studies (13,16). In addition, most of the atients were over 40 years of age and were routinely evaluated by endoscoy. With regard to endoscoic findings, the most common attern of the gastric lymhoma in the athology confirmed sub (ositive ) and endoscoy non-susected was the ulcerative tye. This is also consistent with several recent studies (13,16,17). Unfortunately, the main macroscoic attern of susected gastric lymhoma in the athology non-susected sub was the infiltrative tye (49.4%) and the ulcerative tye was less frequent (42.1%). The differences between the macroscoic atterns of the two subs may be exlained by the high morbidity of gastric carcinoma in China, and thus endoscoy may be erformed to identify gastric carcinoma when gastric ulcerative lesions are found. Therefore, it is imortant to imrove the understanding of the use of endoscoic techniques for evaluating gastric lymhoma.

5 570 X. CUI ET AL. Rev Es Enferm Dig Some studies indicate that large suerficial ulcers with long stellate branches or irregular, rotrusive lesions with central ulcerations are characteristic of lymhoma (11,14). However, several other tyes of ulcers such as small or enetrating ulcers may be indistinguishable from benign and carcinomatous ulcers. Furthermore, the infiltrative attern may also be observed in benign conditions such as Menetrier s disease and in linitis lastica carcinoma (11). In addition, both tyes of lesions mainly occurred in the middle and distal arts of the stomach, as reorted in revious studies (18). Thus the distribution of nonsecific lesions may contribute to misdiagnosis and missed diagnosis. Therefore, it is extremely difficult to obtain a correct diagnosis by endoscoy. Endoscoy in our study showed a sensitivity of 60% for the detection of malignancy, whereas the endoscoic sensitivity declined to 20.8% for gastric lymhoma. Kelessis et al. (17) reorted similar results, with a sensitivity of 61% for malignancy and only 27% for non-hodgkin s lymhoma. This henomenon may be caused by the submucosal growth of the lesion, necrotic material taken during biosy, inadequate amounts of secimen or inadvertent samling resulting from a crush (17,20). In order to enhance the sensitivity of endoscoy, reeat biosies are recommended. Xu et al. (20) found that obtaining reeat biosies could imrove the overall accuracy rate by 40.8%. In addition, endosonograhy (EUS) layed a vital role in identifying gastric lymhoma and other gastric tumors. For examle, infiltrative carcinoma tended to show vertical growth in the gastric wall. However, lymhoma tended to show mainly a horizontal extension (21). However, EUS is not a histological technique, and thus should be combined with endoscoic biosy in order to avoid false negative results. Confocal laser endomicroscoy, a technique that examines tissue histology in vivo, has shown very romising results for diagnosis and differential diagnosis of gastric lymhoma and may be used as an alternative to conventional biosy (22,23). Moreover, histoathological and immunohistochemical examination is crucial for imroving the diagnosis of gastric lymhoma. With regard to the histoathological subtye of lesions, our results verified the fact that DLBCL was the most common tye in atients with gastric lymhoma. The limitation of the study was its retrosective design, which can comromise data integrity and lead bias in the conclusions. In conclusion, atients with gastric lymhoma and susected gastric lymhoma in our study had a similar clinical manifestation. Gastric lymhoma mainly resented as macroscoic ulcerative lesions, whereas susected gastric lymhoma mainly resented as the infiltrative tye. Even though the diagnostic rate of gastric lymhoma was relatively low (21%), gastric lymhoma can usually be diagnosed by endoscoy (60%). In order to imrove diagnostic sensitivity, endoscoic examinations should be combined with reetitive biosies, and novel endoscoic techniques should be develoed in the future. ACKNOWLEDGEMENTS The authors would like to thank the Qilu Hosital of the Shandong University and the Qingdao municial hosital (China). REFERENCES 1. Koch P, Del Valle F, Berdel WE, et al. Primary gastrointestinal non- Hodgkin s lymhoma: I. Anatomic and histologic distribution, clinical features, and survival data of 371 atients registered in the German Multicenter Study GIT NHL 01/92. J Clin Oncol 2001;19: Ding D, Pei W, Chen W, et al. Analysis of clinical characteristics, diagnosis, treatment and rognosis of 46 atients with rimary gastrointestinal non-hodgkin lymhoma. Mol Clin Oncol 2014;2: Ding W, Zhao S, Wang J, et al. Gastrointestinal lymhoma in Southwest China: Subtye distribution of 1,010 cases using the WHO (2008) classification in a single institution. Acta Haematol 2016;1:21-8. DOI: / Halme L, Mecklin JP, Juhola M, et al. Primary gastrointestinal non-hodgkin s lymhoma. A oulation based study in central Finland in Acta Oncol 1997;36: DOI: / Hahn JS, Lee S, Chong SY, et al. Eight-year exerience of malignant lymhoma - Survival and rognostic factors. Yonsei Med J 1997;38: DOI: /ymj Gurney KA, Cartwright RA. Increasing incidence and descritive eidemiology of extranodal non-hodgkin lymhoma in arts of England and Wales. Hematol J 2002;3: DOI: /sj.thj Howell JM, Auer-Grzesiak I, Zhang J, et al. Increasing incidence rates, distribution and histological characteristics of rimary gastrointestinal non-hodgkin lymhoma in a North American oulation. Can J Gastroenterol 2012;26: DOI: /2012/ Fischbach W, Kestel W, Kirchner T, et al. Malignant lymhomas of the uer gastrointestinal tract. Results of a rosective study in 103 atients. Cancer 1992;70: DOI: / ( )70:5<1075:: AID-CNCR >3.0.CO; Taal BG, Burgers JM, Van Heerde P, et al. The clinical sectrum and treatment of rimary non-hodgkin s lymhoma of the stomach. Ann Oncol 1993;4: Tursi A, Paa A, Cammarota G, et al. The role of endoscoy in the diagnosis and follow-u of low-grade gastric mucosa-associated lymhoid tissue lymhoma. J Clin Gastroenterol 1997;25: DOI: / Taal BG, Boot H, Van Heerde P, et al. Primary non-hodgkin lymhoma of the stomach: Endoscoic attern and rognosis in low versus high grade malignancy in relation to the MALT concet. Gut 1996;39: DOI: /gut Doglioni C, Ponzoni M, Ferreri AJ, et al. Gastric lymhoma: The histology reort. Dig Liver Dis 2011;43(Sul 4):S310-S8. DOI: / S (11) Andriani A, Zullo A, Di Raimondo F, et al. Clinical and endoscoic resentation of rimary gastric lymhoma: A multicentre study. Aliment Pharmacol Ther 2006;23: DOI: /j x 14. Rotaru I, Ciurea T, Foarfa C, et al. The diagnostic characteristics of a of atients with rimary gastric lymhoma: Macroscoic, histoathological and immunohistochemical asects. Rom J Morhol Embryol 2012;53: Al MI. Endoscoic features of rimary uer gastrointestinal lymhoma. J Clin Gastroenterol 1994;19:69-73, Radic-Kristo D, Planinc-Peraica A, Ostojic S, et al. Primary gastrointestinal non-hodgkin lymhoma in adults: Clinicoathologic and survival characteristics. Coll Antrool 2010;34: Kelessis NG, Vassilooulos PP, Tsamakidis KG, et al. Is gastroscoy still a valid diagnostic tool in detecting gastric MALT lymhomas? A dilemma beyond the eye. Mucosa-associated lymhoid tissue. Surg Endosc 2003;17:

6 2017, Vol. 109, N.º 8 CLINICAL MANIFESTATIONS AND ENDOSCOPIC PRESENTATIONS OF GASTRIC LYMPHOMA: 571 A MULTICENTER SEVEN YEAR RETROSPECTIVE SURVEY 18. Aoun JP, Moukarbel N, Khoury S. Endoscoic atterns of rimary gastric MALT lymhoma. J Med Liban 1998;46: Arista-Nasr J, Jiménez A, Keirns C, et al. The role of the endoscoic biosy in the diagnosis of gastric lymhoma: A morhologic and immunohistochemical rearaisal. Hum Pathol 1991;22: DOI: / (91)90082-Z 20. Xu W, Zhou C, Zhang G, et al. Reeating gastric biosy for accuracy of gastric lymhoma diagnosis. Gastroenterol Nurs 2010;33: DOI: /SGA.0b013e3181ea Caletti G, Fusaroli P, Togliani T, et al. Endosonograhy in gastric lymhoma and large gastric folds. Eur J Ultrasound 2000;11: DOI: /S (99) Dolak W, Kiesewetter B, Müllauer L, et al. A ilot study of confocal laser endomicroscoy for diagnosing gastrointestinal mucosa-associated lymhoid tissue (MALT) lymhoma. Surg Endosc 2016;30: DOI: /s Kav T, Ozen M, Uner A, et al. How confocal laser endomicroscoy can hel us in diagnosing gastric lymhomas? Bratisl Lek Listy 2012;113: DOI: /BLL_2012_155