Dear Editor

We would like to thank Dr Chandrasoma for his attentive reading and kind comments on our work published in Gut.[1] He has also provided the readers with an admirable synthesis of the most recent research on the development of the different mucosal types in the gastro-oesophageal junction region. By means of this letter, we want to reflect on some of his comments.

The quintessence of Dr Chandrasoma’s vision on cardiac mucosa (CM) is, that it is not a normal structure but develops through metaplasia in a context of gastro-oesophageal reflux disease. The presence of a small length of CM in many “normal” adults could be the result of asymptomatic, low-level reflux. According to his view, the “non-ciliated, non-glandular late fetal foregut epithelium” (which we call CM in our study) will develop into either oesophageal squamous epithelium or gastric mucosa with parietal cell-containing glands. The necessary corrolary of his theory is that there can be no such thing as a normal CM. He also puts forward that the presence of CM in some infants might be due to deviant differentiation of the uncommitted epithelium in a context of reflux or other trauma such as nasogastric intubation. Even if this hypothesis is correct, we think that other possibilities should be considered. One possible situation could be the persistence of the uncommitted epithelium with development of a sort of heterotopic CM (analogous to the heterotopic fundic-type mucosa described in the upper third of the oesophagus). Clearly, much more research is needed for a final answer.

Obviously, our work is not completely representative for the development of the gastro-oesophageal junction region throughout gestation. Notably, we need extra specimens from third trimester fetuses. At this moment we are gathering this material for future research.

As Dr Chandrasoma himself says, the most important reason for the divergent conclusions of his work and ours are the terminology and the interpretation of the data. What we call CM is, in Dr Chandrasoma’s opinion, an uncommitted epithelium devoid of glands. He specifically warns for applying the designation “gland” to the tangentially cut tortuous ends of the foveolar pits (our Figure 2 and Figure 4). We believe glands are present in these illustrations. We formed this conclusion both on a purely morphological basis (the gland cells are cuboidal to triangular and contain a centrally located round nucleus, as opposed to the tall columnar foveolar and pit cells with basically located nuclei) and after histochemical evaluation (the foveolar and pit cells contain a large amount of mostly neutral mucins, whereas the gland cells for a long time contain only a small amount of mostly acidic mucins). We used the term CM for this zone interposed between squamous and fundic mucosa because of its morphologic analogy with adult CM (whether normal or abnormal). Its principal characteristic is the presence of mucus producing glands devoid of parietal cells. We stated that CM develops during gestation and is present at birth. We don’t know what happens with this CM in infants and children. We cannot comment on the identity of adult CM – has it always been there, or did it develop through metaplasia ? To prove or disprove Dr Chandrasoma’s theory, evidently much further research has to be performed.

Reference

(1) Chandrasoma PT. Fetal "Cardiac Mucosa" is not adult cardiac mucosa [electronic response to De Hertogh G et al. On the existence and location of cardiac mucosa: an autopsy study in embryos, fetuses, and infants] gutjnl.com 2003 http://gut.bmjjournals.com/cgi/eletters/52/6/791#89

Editor,

Although it is valuable in some respects, the study from Nakamura et al (Gut 2001;48:454-460) highlights and potentially exacerbates a critical stumbling block in the diagnosis and treatment of gastric lymphomas.

The authors divide gastric "MALT" lymphomas into low grade and high grade, based on the absence or presence of, "... clusters or sheets of large cells [comprising] at least 1% of the neoplastic lymphoid population." Although the 1% cut off was used in a previous study, one would be hard pressed to find a hematopathologist who would call such a lesion "high grade" in clinical practice. Because marginal zone lymphomas tend to remain localized and often respond to antibiotic therapy, while diffuse large B cell lymphomas have aggressive behavior and do not respond to antibiotics, muddying the waters by the promulgation of the ambiguous term "high grade MALT lymphoma" is a practice that should be discouraged.

Indeed, in order to avoid confusion between marginal zone lymphoma and diffuse large B cell lymphoma, the authors of the upcoming WHO lymphoma classification have urged that marginal zone lymphomas be simply designated as such, and that they not be graded (Harris, et al, Mod Pathol 13;193-207:2000). Part of the reason for this is to underscore the very important distinction of marginal zone lymphoma ("MALT"), an indolent entity that may respond to antibiotics, from diffuse large B cell lymphoma, an aggressive entity that requires surgery and/or multi-agent chemotherapy.

Lastly, I would like to add that I was prompted to write this letter by a real life manifestation of the type of confusion I attempt to point out. On the day that the Nakamura article was published, I diagnosed a gastric diffuse large B cell lymphoma via an endoscopic biopsy. When I explained to the submitting gastroenterologist that this lymphoma would not respond to antibiotic therapy, he countered with, "There's a paper in today's Gut that says it will. It says that MALT lymphomas will respond, as long as they're superficial, regardless of grade."

Sadly, a week later I received the patient's bone marrow biopsy, which was completely overtaken by an aggressive diffuse large B cell lymphoma.

Dear Editor

Hepatitis C virus (HCV) infection is widespread among patients on long-term hemodialysis (HD) and the intravenous drug abusers (IVDAs). However, there seem to be striking similarities in the mode of transmission between the two groups since both form high-risk groups for the parenterally transmitted HCV infection.

The indispensable requirement of having vascular access site possibly adds to the risk of acquiring HCV infection among patients on long-term HD through nosocomial transmission especially in the high HCV prevalence units. Preliminary data suggests that among various types of vascular accesses used for HD, arteriovenous fistula (AVF) and polytetrafluoroethylene grafts (PTFE) which require extra skillful handing perhaps play a more significant role in the transmission of the HCV than the permanent or temporary central venous catheters.[1] Sharing of contaminated dialysis equipments, dialyzer-reuse and the physical proximity of an infected patient during the HD are the additional important factors incriminated in the transmission of HCV in the busy HD units.[2] Gilli et al. reported an outbreak of HCV in an Italian HD unit due to the sharing of multidose heparin vials.[3] Another recent study from USA reported an outbreak of HCV occurring when a multidose saline vial was contaminated with blood from HCV-infected patient in a Florida hospital.[4] Above all, the breakdowns in the implementation of the standard infection control safety measures recommended by CDC, is essentially responsible for the rapid rise of HCV infection among HD patients, worldwide.

On the other hand, sharing of contaminated equipment (needles and syringes) among IVDAs as well is the primary concern attributed to the continuous increase in the HCV infection. However, in a most recent report from Kolkata, India [5] the dissemination of HCV accelerated paradoxically, from the baseline prevalence of 17% in 1996 to 66% in 2002 and further to 80 % during the next year regardless of the supply of fresh needles and syringes on daily basis, under the supervision of trained field workers with the equipment being taken back from IVDAs, on next day after use. Most of the IVDAs did not share their syringes and needles; nonetheless they shared the multidose vials of the drugs. Indirectly, the sharing of the drug ampoules suggested contaminated body fluids/ blood being the means of transmission of HCV through direct access to the blood circulation. Transmission of virus was also suspected to occur from sharing of a small pot containing water that some IVDAs use to clean the syringes and needles before using them again.

With strict implementation of standard infection control precautions and probably the isolation of the anti-HCV positive patients it might be possible to effectively control the spread of HCV infection among patients on long-term HD.[6] However, the promiscuous sexual behavior, lack of personal and community hygiene and absolutely disregarded attitude towards life, prevalent among IVDAs, are the major determinants posing practical problems for interventional measures to put into practice, for the control of the spread of HCV in this high-risk group.

References

(1) Saxena AK, Panhotra BR, Sundaram DS. The role, the vascular access plays in the transmission of hepatitis C virus in a high prevalence hemodialysis unit. J Vasc Access 2002; 3:158-63.

(2) Dentico P, Boungiorno R, Volpe A. Prevalence and Incidence of Hepatitis C virus in Hemodialysis patients: Study of risk factors. Clin Nephrol 1992;38:49-52.

(3) Gilli P, Moretti M, Soffritti S. Non-A, non-B hepatitis and anti- HCV antibodies in dialysis patients. Int J Artif Organs 1990; 13:737-41.

(4) Krause G, Trepka MJ, Whisenhunt RS et al. Nosocomial transmission of hepatitis C virus associated with the use of multidose saline vials. Infect Control Hosp Epidemiol 2003; 24:122-27.

(5) Saxena AK, Panhotra BR, Naguib M et al. The impact of dedicated space, dialysis equipment and nursing staff on the transmission of hepatitis C virus in a hemodialysis unit of Middle East. Am J Infect Control 2003; 31:26-33.

(6) Sarkar K, Mitra S, Bal B, Chakraborty S, Bhattacharya SK. Rapid spread of hepatitis C virus and needle exchange programme in Kolkata, India. Lancet 2003;361:9365.