CLINICOPATHOLOGICAL CORRELATION OF UPPER GASTROINTESTINAL TRACT ENDOSCOPIC BIOPSIES IN A TERTIARY CARE HOSPITAL IN RURAL AREA OF NORTH INDIA

1. Lecturer, Department of Pathology, Government Medical College, Jammu, Jammu and Kashmir, India. 2. Registrar, Department of Medicine, Government Medical College, Jammu, Jammu and Kashmir, India. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History Received: 25 January 2020 Final Accepted: 27 February 2020 Published: March 2020


Materials and Method:-
The present study was a prospective study and 200 patients were included in the study over a period of 2 years. The inclusion criteria consisted of patients with lesions presenting with ulcers, abnormal growths, precancerous conditions and malabsorption syndrome (MAS) and Lesions present in esophagus, stomach and up to second part of duodenum while, patients presenting with lesions in the oral cavity & oropharynx, inadequate biopsy in terms of no glands, only fibro collagenous tissue and all duodenal biopsies below the second part of duodenum were excluded out.
The endoscopic biopsy specimens so obtained were put in saline, placed on the filter paper with mucosal surface upwards and fixed in 10% formalin. All the bits were embedded together for ideal visualization. Then, sections 4-6 microns were stained routinely with Haematoxylin and Eosin. The well oriented and adequate endoscopic gastrointestinal biopsies were then correlated with endoscopic findings. The neoplastic lesions were diagnosed as per WHO classification of tumours 4 . The cases of MAS were graded as per modified MARSH criteria 5 .
The data collected was analyzed as percentage of concordance of the two. The discordant cases were analyzed to obtain reasons for discordance. SPSS (Statistical package for social sciences v21.0) was used to perform statistical analyses.

Results:-
Out of 217 biopsies obtained, 17 were excluded as they contained only fibro collagenous tissue, scanty material & non-specific findings. Among the neoplastic lesions 34 (17.0%), the most common was Squamous cell carcinoma constituting 20 cases (10.0%) followed by 14 cases of adenocarcinoma (7.0%). The most common age group was 51-70 years for both squamous cell & adenocarcinoma. The non-neoplastic lesions including gastritis, duodenitis, villous atrophy, polyp, lymphangectasia & GAVE were commonly seen in males accounting for 166 cases (83.0%). (TABLE 1) Endoscopic and histological correlation of 26 esophageal biopsies shows that out of 6 cases of esophagitis, 1 was diagnosed as carcinoma upon endoscopy whereas remaining 5 where diagnosed as esophagitis in both yielding 83.3% concordance. Similarly, out of 20 cases of carcinoma, 19 were found to be malignant on both endoscopy as well as histology whereas 1 case was documented as esophagitis upon endoscopy accounting to 95.0% concordance. (TABLE 2)      Endoscopic and histological correlation of 44 duodenal biopsies shows that the most common discordant lesion on endoscopy was polypoidal lesion wherein a case of lymphangectasia was diagnosed to be a polypoidal lesion upon endoscopy, followed by scalloping lesions which accounted for 60.7% concordance between the two diagnostic methods. Duodenitis and carcinomas yielded 100% concordance in our study. (TABLE 4) 1158

Discussion:-
Endoscopy, when combined with biopsy is an easy, minimally invasive & cost effective procedure when it comes to arriving at a specific diagnosis of a patient with non-specific symptoms. The present study included such endoscopic biopsies that were studied with respect to age & sex distribution and correlation with endoscopic findings. Out of total 217 endoscopic biopsies from upper GIT , 200 lesions were included in the study. The biopsies that were included in the study comprised of 26 (13.0%) esophageal biopsies & 130 (65%) gastric biopsies. The remaining were 44 (22%) duodenal biopsies & none (0.0%) from GEJ. These lesions were then classified non-neoplastic lesions which, comprised 166 (83.0%) cases & neoplastic lesions, which comprised 34 (17%) cases. The age & sex distribution, correlation with endoscopic findings were calculated separately for both these categories.
In the present study it was found that the majority of patients were between 31-40 years of age in contrast to study done by Froehlich et al 6 , where the age group was found to be over 60 years. The non-neoplastic lesions included cases of esophagitis, gastritis, GAVE, polyps, duodenitis, villous atrophy& lymphangectasia. The peak age group of 1159 non-neoplastic lesions was found to be 31-40 years. These findings were in contrast to the study done by Wei et al where mean age group was 56 years 7 . Gastritis was found to be common in 31-40 years age group in the present study similarly, it was found to be common in 31-40 years in other studies & the incidence was also found to increase with age 8,9 . In the present study, gastric polyps were found more in females over 31-  16 .
In the present study the number of males undergoing upper GI endoscopy were more than the number of females. Similar findings were found in the previous study by David 17 , Froehlich et al 6 and Shennak et al 18 . In all the three studies the men out numbered women due to more prevalence of smoking, alcoholism and stressful life. The nonneoplastic lesions were also commonly seen in males (69.3%), which was similar to Lee et al 19 . Gastritis was also more common in males in the present study similar to Afzal et al but in contrast to Adisa et al where it was more common in females 20 . This could be due to more number of males attending the hospital and also due to higher consumption of alcohol in males. In the present study, gastric polyps were found to occur more in frequency in females than males while Ljubicic et al showed a slight male preponderance 10  The esophageal lesions showed a better correlation between the two diagnostic modalities with positive endoscopy findings. Out of 20 esophageal carcinomas, 19 diagnosed as carcinomas on endoscopy were confirmed by histology whereas 1 case came out benign. Esophagitis seen as inflammatory change could only be found in 5 out of 6 histologically confirmed cases while 1 was diagnosed as carcinoma endoscopically. According to Pope 21 endoscopic finding did not rule out possibility of esophagitis, confirming our study that few lesions are likely to be missed on endoscopic examination alone.
In the gastric lesions there was a modest concordance between the two diagnostic modalities specially in cases of diagnosing polypoidal lesions wherein out of 5 histologically confirmed cases of polyps, a case of GAVE and another of carcinoma was misjudged to be a polyp on endoscopy accounting for 71.4% concordance. Thus because of this discordance the endoscopist should convey the findings as erythema, exudate to the pathologist along with the biopsy rather than diagnosing it alone. Morson 22 has also documented similar observations. The carcinomas on the other hand yielded 100% concordance in our study.
In the duodenal lesions too, the correlation studies showed similar trends as in gastric lesions with polypoidal lesions yielding total discordance, as single case diagnosed as polyp on endoscopy came out as lymphangectasia upon histology. This was followed by scalloping lesions with loss of folds seen in 28 cases on endoscopy with only 17 confirmed cases on histology accounting for 60.7% concordance between the two methods. Our findings were in agreement to the study by Paoluzi et al 23 . Similar to the gastric findings, the carcinomas yielded 100% concordance in our study.

Conclusion:-
The conclusion of the study was that endoscopic examination alone might miss out in diagnosing majority of the lesions. So, histological examination in adjunct with endoscopy should be considered as much more valuable diagnostic tool rather than endoscopy alone.