LAUSR

I read with interest “Lymphocytic esophagitis: an update on histologic diagnosis, endoscopic findings, and natural history,” by Patil et al.1 The authors stated that lymphocytic esophagitis (LyE) is characterized by increased intraepithelial lymphocytes (IELs) (>20/high-power field) with rare, or absent, granulocytes. Furthermore, more than a decade following its original description, LyE remains an enigmatic entity, and most of the confusion regarding the clinical significance of this disorder stems from its diagnostic criteria.1 The concept of LyE in humans was first reported 12 years ago.2 During its “childhood,” the histologic diagnosis of LyE was based on hematoxylin and eosin (H&E)–stained sections exhibiting a high number of IELs/high power field (HPF). The results in 20 LyE cases showed that when 10 × oculars and 40× objective (lens aperture: 0.95) were used, a total of 1508 IELs (mean 75.4 IEL/case) were found.2 Among 1508 IELs, 1102 and 406 were peripapillar (mean 55.1) and interpapillar IELs (mean 20.3), respectively. Regrettably, the lower limit of intraepithelial lymphocytosis used to diagnose LyE was not given.2 In a subsequent study of LyE, also using H&E-stained sections, the lower limit was set at 20 IELs.3 It has been axiomatic that all IELs can be detected with H&E. However, the microtome cuts lymphocytes at various levels, and histological sections of 4 m in diameter contain not only the entire IEL nuclei but also small nuclear and small cytoplasm fragments. Those nuclear and cytoplasmic IEL fragments are not counted as lymphocytes in H&Estained sections. Moreover, when the CD3 antibody reacts with its cognate antigen present in the cytoplasm,3 the entire IEL and cytoplasmic fragments included in the sections become clearly visible. With that method, a significantly higher number of CD3+ lymphocytes are highlighted in the area encompassed by the 4 m thick section than when counting IEL in H&E stain exclusively.2 Thus, accruing entire IEL and its fragments can be detected in routine tissue sections immunostained with CD3, but not when stained with H&E.2 Twelve years after the first description2 (i.e., at the time of LyE “puberty”), we took an advantage of the aforementioned knowledge and stained tissue sections with H&E and with CD3 independently, one at a time.3 The result in 311 patients consulting for dysphagia, food impaction, dysphagia with fluids and/or solids, and sense of foreign body revealed four histological/immunohistological esophagitis phenotypes: (1) lymphocytic esophagitis (LyE): 40 CD3+ lymphocytes/HPF using CD3 immunostaining; (2) eosinophilic esophagitis (EoE): 15 eosinophils/HPF with lymphocytic infiltration ( 39 CD3+/HPF) in H&E stain; (3) compound lymphocytic esophagitis–eosinophilic esophagitis (Co LyE–EoE), that is, LyE, 40 CD3+ lymphocytes/HPF in CD3 immunostaining and 15 eosinophils/HPF in H&E stain; and (4) lymphocytic infiltration ( 39 CD3+/HPF). These four esophagitis phenotypes were easily recognized, provided that sections were stained with H&E and CD3 independently, one at a time.3 The serendipitous finding that 11% (33/311) of the cases had compound LyE and EoE2 may explain some of the difficulties experienced by endoscopists while trying to discriminate between LyE and EoE. But, there is a caveat. Recent literature showed that group 2 innate lymphocytes (ILC2s) enriched in EoE4 remain undetected in CD3 immunostaining. If ILC2s also participate in EoE with lymphocytic infiltration, then the frequency of cases with compound LyE–EoE might have been much higher.