Eur J Clin Microbiol Infect Dis (2000) 19 :96–100 Q Springer-Verlag 2000

A. Chocarro Martínez (Y), F. Galindo Tobal,A. González López, F. Alvarez Navia, M.I. Martín ArribasDepartment of Internal Medicine, Hospital “Virgen de laConcha”, Avenida de Requejo 31-33, 49022 Zamora, Spaine-mail: achocarrom6medynet.com

G. Ruiz-IrastorzaDepartment of Internal Medicine, Hospital de Mendaro,Mendaro, Guipúzcoa, Spain

C. Ochoa SangradorUnit of Research, Hospital “Virgen de la Concha”,Avenida de Requejo 31-33, 49022 Zamora, Spain

Article

Risk Factors for Esophageal Candidiasis

A. Chocarro Martínez, F. Galindo Tobal, G. Ruiz-Irastorza, A. González López,F. Alvarez Navia, C. Ochoa Sangrador, M.I. Martín Arribas

Abstract The role of gastric acid inhibitors as predisposing factors for Candidaesophagitis is unknown. A retrospective case-control study of esophageal candidiasiswas conducted in human immunodeficiency virus (HIV)-negative patients diagnosedfrom January 1991 to December 1997. The diagnosis of esophageal candidiasis wasalways made on the basis of endoscopic and histological criteria. Fifty-one patientswere diagnosed with esophageal candidiasis, 15 of whom had esophageal complaintsand 48 of whom suffered from another previous chronic disease (17 had cancer). Inaddition, 20 patients had previously been treated with antibiotics, 13 with steroidsand 14 with omeprazole. In the multivariate analysis, neoplasm (odds ratio, 5.50;95% confidence interval, 1.94–15.56) and therapy with antibiotics (odds ratio, 11.97;95% confidence interval, 3.82–37.45), steroids (odds ratio, 35.52; 95% confidenceinterval, 3.90–324.01) or omeprazole (odds ratio, 18.23; 95% confidence interval,4.67–71.03) were all associated with esophageal candidiasis. These data suggest thatCandida esophagitis tends to occur in patients with chronic diseases, most of whomhave been previously treated with antibiotics, steroids or omeprazole. The findingssupport the hypothesis that treatment with omeprazole favors the development ofesophageal candidiasis.

Introduction

The Candida species is the most common cause ofinfectious esophagitis [1–3]. Andrén and Theander [4]first described the roentgenographic findings ofesophageal moniliasis in 1956. A large number ofpapers have been published since then, focusing on theclinical and epidemiological features of this disease [1,5–8]. However, most of these studies have been madewithout suitable controls, a fact that limits the conclu-sions to be drawn [1].

Esophageal candidiasis can cause severe complicationssuch as digestive tract bleeding, perforation, starvationor systemic dissemination [2, 7, 9]. Systemic dissemina-tion has been reported in 15% of patients withleukemia and 11% of renal transplant patients [1, 10].Today, it is thought that Candida infection of thegastrointestinal tract frequently precedes systemicdissemination. This is one of the reasons for thegrowing interest in candidiasis of the digestive tract[11].

The main predisposing factors for esophageal candi-diasis reported thus far are immunosuppression,previous esophageal disease and certain other condi-tions that enhance the growth of Candida in the diges-tive tract [12, 13]. Cancer, therapy with corticosteroidsor cytotoxic drugs and, more recently, HIV infectionare all well known causes of cellular immunosuppres-sion and, therefore, risk factors for Candida esophagitis[14, 15]. Indeed, since the early 1980s, most studies ofesophageal candidiasis have involved HIV-infectedpatients [16–20]. Esophageal diseases, such as noninfec-tious esophagitis and achalasia, also favor the develop-ment of Candida esophagitis [21]. Finally, broad-spec-trum antibiotics may eliminate certain bacteria that

97

inhibit fungal growth, thereby enhancing Candida over-growth [22–24].

Soon after the introduction of H2-receptor antagonists,some isolated cases of digestive candidiasis werereported [25–28]. In recent years, acid-pump inhibitorshave become widely used, and some reports linkomeprazole use with the development of esophagealcandidiasis [29–32]. However, there are very few datato prove a causative effect.

Our aim in this study was to investigate the clinicalcharacteristics of esophageal candidiasis in non-HIV-infected patients and the role of antisecretory drugs inthis infection. Accordingly, the clinical and endoscopiccharacteristics of our non-HIV-infected patients withCandida esophagitis are presented first, followed by ananalysis of the putative predisposing factors, withemphasis placed on antisecretory drugs.

Materials and Methods

Patients. The study was designed as a retrospective case-controlstudy. All cases of esophageal candidiasis diagnosed in ourhospital in non-HIV-infected patients from 1 January 1991through 31 December 1997 were reviewed. Our 380-bed generalhospital serves a population of 220000 people. Oncological treat-ments are usually carried out at another center.

Candida esophagitis was diagnosed if candidiasis was visible atendoscopy; the diagnosis was always confirmed by biopsy. Casepatients were denominated as group 1. The control group, orgroup 2, was formed by those patients without a diagnosis ofesophageal candidiasis and in whom an endoscopic examinationwas performed immediately before and after every case patientwas examined endoscopically (2 controls per case). Therefore,cases and controls were evaluated by the same medical team,using the same diagnostic criteria. Patients infected with HIVwere excluded from both groups. Esophageal candidiasis wasgraded on a scale of 1 to 4, in accordance with the method ofKodsi et al. [33] (grade 1, a few raised white plaques up to 2 mmin size with hyperemia but no edema or ulceration; grade 2,multiple raised white plaques greater than 2 mm in size withhyperemia and edema but no ulceration; grade 3, confluent,linear and nodular elevated plaques with hyperemia and frankulceration; grade 4, finding of grade 3 with an increased friabilityof the mucous membranes and occasional narrowing of thelumen).

The following putative predisposing factors were analyzed: (i)esophageal diseases – noninfectious esophagitis, hiatus hernia,achalasia, esophageal varices, diverticula, Schatzki ring oresophageal cancer; (ii) other conditions – diabetes mellitus,neoplasm, chronic hepatopathy, atrophic gastritis, azotemia, alco-holism, systemic autoimmune diseases, scleroderma, neutropeniaor other immunosuppressive or chronic diseases; and (iii)previous therapy – treatment with corticosteroids, chemotherapy,antibiotics, antacids, H2-receptor antagonists, omeprazole ordigestive surgery within 30 days prior to the diagnosis of esopha-geal candidiasis.

Neutropenia was defined as a blood neutrophil count of less than1.5!109/l [34]. Alcoholism was defined as a daily intake ofethanol greater than 80 g [35]. All gastric and esophageal condi-tions were diagnosed endoscopically, and, in addition, neoplasmand atrophic gastritis were confirmed by biopsy.

Methods. All endoscopic examinations were performed by staffmembers of our hospital’s Gastroenterology Section, using aPentax EG 2901 videoscope. Biopsy samples were placed imme-diately in 10% formol; subsequent pathological examination wascarried out using hematoxylin-eosin, periodic acid-Schiff andmethenamine silver stains. In addition, a second sample wasobtained for culture in Saboraud medium.

A bibliographic search by Medline (January 1992 throughDecember 1997) was performed, using “candidiasis” and“omeprazole” as key words. Additional papers were obtainedfrom the reference sections of the selected articles.

Statistical Analysis. Data management and statistical analyseswere performed using SPSS software (SPSS for Windows, version5.0.1; SPSS, USA). Only those variables observed in at least 10%of the patients were included. Means and standard deviations(SD) were calculated for the continuous variables, and propor-tions with 95% confidence intervals (CIs) for the categoricalvariables. The latter were compared using the chi-square andFisher’s exact tests, as indicated. The risks were estimated usingodds ratios (ORs) with 95% CIs. Mantel-Haenszel weighted ORsand 95% CIs were calculated for the stratified analysis.

A logistic-regression analysis was performed in order to assess thedifferent risk factors associated with the development of esopha-geal candidiasis, controlling for confounding factors. The initialmultivariate logistic regression analysis included all covariates forwhich the P value obtained in the univariate logistic analysis was~0.25. For all of them, the presence of the putative risk factorwas coded as 1, and its absence as 0. The modeling backwardstepwise selection strategy was used, employing the likelihoodratio change as criterion. The probability of having esophagealcandidiasis was calculated from the final model, using the coeffi-cient and the values of the predictor variables in the followingequation: pp1/(1c(Exp(-(b0cb1X1cb2X2c....cbnXn)))). Thesensitivity and specificity of the final model were calculated fromthe classification table, constructed by cross-referencing thepredictions in each case with the observed values. The cases witha probability higher than 0.5 were classified as positive prediction.The CIs were calculated with the program Two by Two (R.M.Centor and J. Keighteley, USA).

Results

Patient Characteristics. During the years 1991 through1997, 17778 upper endoscopies were performed in ourhospital. Sixty-four non-HIV-infected patients werediagnosed with Candida esophagitis on the basis ofendoscopic criteria, 13 of whom were excluded fromthe study: 11 had a negative histological examination,and the records of two additional patients did notinclude enough clinical data. Thus, group 1 comprised51 people diagnosed on the basis of endoscopic plushistological criteria. In addition, all of these patientshad a biopsy sample that was culture-positive forCandida albicans. One hundred two controls wereincluded in group 2. The ages, sexes and percentages ofinpatients were similar in both groups (Table 1).

Endoscopic Findings. Esophageal candidiasis was clas-sified as grade 1 in six patients, grade 2 in 37 and grade3 in eight. Candida affected the whole esophagus in 32patients, the lower esophagus in 16 and the upperesophagus in three. Other esophageal disease was

98

Table 1 Baseline characteristics of patients in each group

Cases(np51)

Controls(np102)

P value

Mean age BSD (yrs) 68.05B16.06 68.93B16.51 0.90No. (%) of females 23 (45) 45 (44) 0.72No. (%) of inpatients 35 (69) 75 (73) 0.52

Table 3 Results of multivar-iate logistic regression analysisof factors associated withesophageal candidiasis. Thesensitivity and specificity ofthe model were 68.63% (95%CI, 54–79%) and 87.25%(95% CI, 79–92%), respec-tively, and the global validitywas 81.05%

Factor Coefficient Standard errorof coefficient

P value Odds ratio(95% CI)

Cancer 1.7051 0.5307 0.0013 5.50 (1.94–15.56)Prior use of antibiotics 2.4830 0.5818 ~0.0001 11.97 (3.82–37.45)Prior use of steroids 3.5721 1.1269 0.0015 35.52 (3.90–324.01)Proir use of omeprazole 2.4032 0.6980 ~0.0001 18.23 (4.67–71.03)

Table 2 Results of univariate analysis of potential risk factors for acquisition of esophageal candidiasis

Variable No. (%)of cases

No. (%)of controls

P value Odds ratio(95% CI)

Cancer 17 (33.33) 17 (16.66) 0.0194 2.50 (1.14–5.45)Diabetes mellitus 12 (23.52) 14 (13.72) 0.1279 1.93 (0.81–4.56)Chronic liver disease 7 (13.72) 13 (12.74) 0.8653 1.08 (0.40–2.92)Alcoholism 15 (29.41) 19 (18.62) 0.1303 1.82 (0.83–3.97)Chronic disease 47 (92.15) 75 (73.52) 0.0068 4.23 (1.39–12.85)Prior use of antibiotics 20 (39.21) 8 (7.84) 0.0000 7.54 (3.03–18.92)Prior use of steroids 13 (25.49) 1 (0.98) 0.0000 34.55 (4.36–273.24)Prior use of antacids 4 (7.84) 10 (9.80) 0.6917 0.78 (0.23–2.62)Prior use of H2–blockers 12 (23.52) 11 (10.78) 0.0375 2.54 (1.03–6.26)Prior use of omeprazole 14 (27.45) 4 (3.92) 0.0000 9.27 (2.86–29.98)

found in 19 cases and 32 controls (difference not statis-tically significant). Thirty-five cases and 66 controlswere found to have a gastric disease (difference notsignificant). Chronic atrophic gastritis was discoveredin six group 1 patients and in nine group 2 patients(difference not significant). Gastric or esophagealneoplasm was more frequent in group 1 than in group 2(7 vs. 5 patients, respectively; difference not signifi-cant).

Clinical Findings. Thrush was observed in only twopatients with Candida esophagitis. Among the group 1patients, 92% had a chronic disease versus 73% of thecontrols. Seventeen patients in each group sufferedfrom cancer (Table 2). The number of patients withother variables was too small to make any compar-ison.

Fifteen patients in group 1 and 16 in group 2 presentedesophageal symptoms (Pp0.0464). The most frequentcomplaints were dysphagia (10 vs. 4 patients, respec-tively; Pp0.0015), heartburn (6 vs. 1; Pp0.0026) andodynophagia (3 vs. 1; not significant). In group 1, theindication for endoscopy in patients who did notpresent esophageal symptoms was abdominal pain in 11

cases, nausea and vomiting in seven, suspectedneoplasm in seven, gastrointestinal bleeding in five,anemia in two, acid dyspepsia in two, portal hyperten-sion in one and esophageal trauma in one.

Previous Therapy. Both groups were significantlydifferent in terms of previous therapy with antibiotics,steroids, H2-receptor antagonists and/or omeprazole(Table 2). However, the relationship between treat-ment with H2-receptor blockers and esophageal candi-diasis was probably explained by the associationbetween treatment with these drugs and treatment withcorticosteroids (Pp0.0012). Accordingly, the relation-ship between H2-receptor antagonists and Candidaesophagitis in patients not treated with steroids couldnot be observed in the analysis stratified by level ofcorticosteroid use (ORp1.70; 95% CI, 0.57–5.07). Norelationship between antacid consumption and esopha-geal candidiasis was observed. The number of patientswith previous cytotoxic treatment or digestive surgerywas too small to make any comparison.

Multivariate Analysis. Cancer and previous treatmentwith antibiotics, steroids or omeprazole were all relatedto esophageal candidiasis (Table 3). Other variablessuch as diabetes mellitus, chronic diseases or alco-holism were not associated with Candida esophagitis.In 44 of the 51 case patients, at least one of the riskfactors identified in the multivariate analysis (neo-plasm, or antibiotic, steroid or omeprazole therapy)was present. Furthermore, in 25 patients, only one riskfactor was identified: omeprazole therapy in six, anti-biotic treatment in seven, steroid therapy in three andcancer in nine (Figure 1).

99

Figure 1 Number of patients with different risk factors forCandida esophagitis identified in the multivariate analysis

Discussion

Candida esophagitis is not a rare disease, even amongnon-HIV-infected patients. This allowed us to study anumber of patients and draw some conclusions. One ofthem is that most patients did not complain of anyesophageal symptom [1, 10]. This finding emphasizesthe limitations of esophageal complaints as a guide todiagnose Candida esophagitis and, thus, the necessityof taking into account other data, such as the presenceof risk factors, when diagnosing this infection.

In this study, we found that the previously recognizedpredisposing factors, mainly cancer or treatment withantibiotics or corticosteroids [12, 24], indeed correlatedwith the development of esophageal candidiasis.However, unlike the findings of earlier reports, only33% of our patients had neoplasms, but most of themhad been previously diagnosed as having chronicdiseases and/or had been treated with antibiotics,steroids or omeprazole.

The published papers on the relationship betweenesophageal candidiasis and treatment with omeprazoleare scarce. In 1988, Hendel et al. [21] reported that89% of patients with scleroderma who were treatedwith H2-receptor antagonists or omeprazole hadesophageal cultures positive for Candida, comparedwith 44% of nontreated patients. The first two cases ofesophageal candidiasis in omeprazole-treated patientswere reported in 1992 [29], and four additional caseswere published in subsequent years [30–32]. However,no case-control study analyzing the relationshipbetween omeprazole use and esophageal candidiasis isavailable in the literature. Our findings in this retro-spective, controlled study show a statistical relationshipbetween omeprazole treatment and esophageal candi-diasis.

Although the ultimate mechanism that causes esopha-geal candidiasis is not known, an elevated gastric pHseems to be relevant [36]. Gastric acidity is known to bean important barrier for most microorganisms, and itsinhibition can favor the development of a number ofinfections [36, 37]. Accordingly, Karmeli et al. [36] wereable to culture Candida albicans from the gastric aspi-rate of four of 12 patients after 1 month of treatmentwith omeprazole.

However, we must also point out that a similar associa-tion between other antisecretory drugs, like H2-re-ceptor antagonists, and esophageal candidiasis was notfound. This could be due to an insufficient sample size,but it must also be remembered that these drugs aremuch less efficient than omeprazole at inhibiting gastricacid secretion. For example, Karmeli et al. [36] foundthat in all patients treated with omeprazole, the pH ofgastric juice was greater than 4.5, but only 12–20% ofgastric aspirate samples taken during cimetidinetherapy had a pH64. It is likely that this lower efficacyof H2-receptor antagonists accounts for the differentrisk of esophageal candidiasis developing.

In addition to its properties as a proton-pump inhibitor,omeprazole has been shown to inhibit the cytotoxiceffect of lymphocytes in vivo [38]. It also diminishessalivary secretion, which can facilitate the growth ofCandida in the mouth and its subsequent spreadthroughout the digestive tract [39].

Nevertheless, our study has some limitations, most ofwhich are due to the small number of patients includedand the retrospective design. In addition, we includedonly those patients diagnosed by biopsy, and somepatients in whom mild disease could have beenexcluded.

Omeprazole was the only risk factor in six patients andwas present with another factor in eight additionalpatients. Moreover, we found similar data when otherfactors such as neoplasms, antibiotic therapy or corti-costeroid therapy were analyzed. Thus, the risk factorsidentified often work as cofactors.

Further studies are necessary to establish definite clin-ical conclusions. It is unknown whether our resultscould be confirmed in other clinical settings, such as inHIV-infected patients. In the meantime, our datasupport the recommendations made by other authors.Accordingly, we believe that omeprazole should beused with caution in immunosuppressed patients, whohave an increased risk of developing both esophagealand systemic candidiasis [30].

Acknowledgement We would like to thank Mr. G.H. Jenkins forchecking the English version of the manuscript.

100

References

1. Eras P, Goldstein MJ, Sherlock P: Candida infection of thegastrointestinal tract. Medicine (1972) 51 :367–379

2. Young JA, Elias E: Gastro-oesophageal candidiasis: diagnosisby brush cytology. Journal of Clinical Pathology (1985)38 :293–296

3. Wheeler RR, Peacock JE, Cruz JM, Richter JE: Esophagitisin the immunocompromised host: role of esophagoscopy indiagnosis. Reviews of Infectious Diseases (1987) 9 :88–96

4. Andrén L, Theander G: Roentgenographic appearances ofesophageal moniliasis. Acta Radiologica (1956) 46 :571–574

5. Holt JM: Candida infection of the oesophagus. Gut (1968)9 :227–231

6. Maggiolo P, Klinger J: Candidiasis esofágica. Experiencia de25 casos. Revista Médica de Chile (1981) 109 :1163–1167

7. Sheft DJ, Shrago G: Esophageal moniliasis. The spectrum ofthe disease. JAMA (1970) 213 :1859–1862

8. Laufer I: Radiology of esophagitis. Radiologic Clinics ofNorth America (1982) 20 :687–699

9. Gaissert HA, Breuer CK, Weissburg A, Mermel L: Surgicalmanagement of necrotizing Candida esophagitis. Annals ofThoracic Surgery (1999) 67 :231–233

10. Gupta KL, Ghosh AK, Kochhar R, Jha V, Chakrabarti A,Sakhuja V: Esophageal candidiasis after renal transplanta-tion: comparative study in patients on different immunosup-pressive protocols. American Journal of Gastroenterology(1994) 89 :1062–1065

11. Cole G, Halawa A, Anaissie E: The role of gastrointestinaltract in hematogenous candidiasis: from the laboratory to thebedside. Clinical Infectious Diseases (1996) 22, Supplement2 :73–88

12. Mathieson R, Dutta SK: Candida esophagitis. DigestiveDiseases and Sciences (1983) 28 :365–370

13. Gornig M, Emmert-Buck M, Walsh TJ: Patterns of Candidaesophagitis in cancer and AIDS patients: histopathologicalstudy of 23 patients. Mycoses (1997) 40, Supplement1 :81–85

14. Harvey CF, Mills JOM, Barros D: Oesophageal moniliasis.British Journal of Clinical Practice (1986) 40 :36–40

15. Moulinier B, Lambert R, Grenier-Boley P, Bruhière J: Lesmycoses de l’oesophage. Nouvelle Presse Mèdicale (1972)39 :2629–2632

16. López-Dupla M, Mora P, Pintado V, Valencia E, Lavilla P,Khamashta MA, Gil A: Clinical, endoscopic, immunologic,and therapeutic aspects of oropharyngeal and esophagealcandidiasis in HIV-infected patients: A survey of 114 cases.American Journal of Gastroenterology (1992) 87 :1771–1776

17. Gould E, Kory WP, Raskin JB, Ibe MJ, Redlhammer DE:Esophageal biopsy findings in the acquired immunodeficiencysyndrome (AIDS): clinicopathologic correlation in 20patients. Southern Medical Journal (1988) 81 :1392–1395

18. Bonacini M, Laine L, Gal AA, Lee MH, Martin SE, Strigle S:Prospective evaluation of blind brushing of the esophagus forCandida esophagitis in patients with human immunodefi-ciency virus infection. American Journal of Gastroenterology(1990) 85 :385–389

19. Porro GB, Parente F, Cernuschi M: The diagnosis of esopha-geal candidiasis in patients with acquired immune deficiencysyndrome: is endoscopy always necessary? American Journalof Gastroenterology (1989) 84 :143–146

20. Chaisson RE, Gallant JE, Keruly JC, Moore RD: Impact ofopportunistic disease on survival in patients with HIV infec-tion. AIDS (1998) 12 :29–33

21. Hendel L, Svejgaard E, Walsoe I, Kieffer M, Stenderup A:Esophageal candidosis in progressive systemic sclerosis:occurrence, significance, and treatment with fluconazole.Scandinavian Journal of Gastroenterology (1988)23 :1182–1186

22. Drouhet E, Dupont B: Biologie des Candida. Physiopatho-logie et épidémiologie des candidoses digestives. Annales deGastroentérologie et D’Hépatologie (1984) 20 :41–46

23. Nichols PE, Henry K: Gastritis and cimetidine: a possibleexplanation. Lancet (1978) i:1905–1906

24. Ortuno JA, Tovar A, Ruiz J, Garcia A: Candidiasis esofágicaen pacientes VIH negativos. Revista Espanola de Enferme-dades Digestivas (1997) 89 :503–510

25. Kochhar R, Talwar P, Singh S, Mehta SK: Invasive candi-diasis following cimetidine therapy. American Journal ofGastroenterology (1988) 83 :102–103

26. Trigger DR, Goepel JR, Slater DN, Underwood JCE:Systemic candidiasis complicating acute hepatic failure inpatients treated with cimetidine. Lancet (1981) ii:837–838

27. Cipollini F, Altilia F: Candidiasis complicating cimetidinetreatment. Lancet (1981) ii:1047

28. Goenka MK, Kochar R, Chakrabarti A, Kumar A, Gupta O,Talwar P, Mehta SK: Candida overgrowth after treatment ofduodenal ulcer. A comparison of cimetidine, famotidine, andomeprazole. Journal of Clinical Gastroenterology (1996)23 :7–10

29. Larner AJ, Lendrum R: Oesophageal candidiasis afteromeprazole therapy. Gut (1992) 33 :860–861

30. Mosimann F: Esophageal candidiasis, omeprazole therapy,and organ transplantation – a word of caution. Transplanta-tion (1993) 56 :492–493

31. Sood A, Sharma M, Jain NP, Chawla LS, Kumar R: Esopha-geal candidiasis following omeprazole therapy: a report oftwo cases. Indian Journal of Gastroenterology (1995)14 :71–72

32. González P, Toro H, Balanzó J, Villardel F: Candidiasisesofágica durante el tratamiento con omeprazol. MedicinaClínica (1993) 100 :639

33. Kodsi BE, Wickrermesinghe PC, Kozinn PJ, Iswara K, Gold-berg PK: Candida esophagitis: a prospective study of 27 cases.Gastroenterology (1976) 71 :715–719

34. Bagby G: Disorders of neutrophil production. In: Bennet JC,Plum F (eds): Cecil textbook of medicine. Saunders, Philadel-phia (1996) pp 908–915

35. Savage D, Lindenbaum J: Anemia in alcoholics. Medicine(1986) 65 :322–338

36. Karmeli Y, Stalnikowitz R, Eliakim R, Rahav G: Conven-tional dose of omeprazole alters gastric flora. DigestiveDiseases and Sciences (1995) 40 :2070–2073

37. Gianella RA, Broitman SA, Zamchek N: Influence of gastricacidity on bacterial and parasitic enteric infections. Aperspective. Annals of Internal Medicine (1973) 78 :271–276

38. Scaringi L, Cornacchione P, Fettucciari K, Rosati E, Rossi R,Marconi P, Capodicasa E: Activity inhibition of cytolyticlymphocytes by omeprazole. Scandinavian Journal of Immu-nology (1996) 44 :204–214

39. Teare JP, Spedding C, Whitehead MW, Greenfield SM, Chal-lacombe SJ, Thompson RP: Omeprazole and dry mouth.Scandinavian Journal of Gastroenterology (1993)30 :216–218

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.