Tobacco Documents Online

.6- :~ is OM12-9270/92/8710-13901(13.00/0 T11F. l1MF.RICAN JQURNAIL or GASTROENTr:R4l.(X:Y Copynght'O 1992 by Am. Coll. of Gastrocn(crology Vol. 87. No. 10. 1992 Printcd in U.S.A. ~ ~ Origirial contributions Smoking Does Not Contribute to Duodenal Ulcer Relapse after Helicobacter pylori Eradication 'I'Iwmas J. Borody, M.D., F.R.A.C.P., Laura L. George, M.D., Susan Brandl, M.B., B.S., Peter Andrews, M.B., B.S., Eva Jankiewicz, and Noela Ostapowicz Cenlrejor Digestive Diseases, Sydney, Australia Cigarette smoking is believed to be one of the major factors influencing duodenal ulcer (DU) recurrence. However, the influence of cigarette smoking on DU recurrence after tlte eradication of Helico8acler pylori has not been separately addressed. The aim of this study was to investigate DU relapse rate in smokers and nonsmokers, botlt with confirmed eradication of H. pylori. Patients with H. pylori eradication, demon- strated at endoscopy 4 wk post-Irealment, were included in the study. Smoking history was obtained with a standard questionnaire, and patients were followed en- doscopically, both yearly and at symptomatic recur- rence, to detect anatomical DU recurrence. Of the 197 (121M:76F) patients enrolled in the study and followed ; for 1-6 yr, 80 (41%) were smokers, smoking 5-40 cigareltes/day. The 117 (59%) nonsmokers included 31 (26%) patients who had ceased smoking 4-20 yr ago. Another seven (9%) smokers ceased smoking during the follow-up period. In jjtq 197 patients with eradicated 1I. pylori and cured DU, there has been no recurrence ulcer, regardless of smoking status.l'Ye conclude that in patients with DU in whom H. pj,lori infection is eradicated; ulcer disease does not recur, as observed for up to 6 yr. Furthermore, cigarette smoking is not a risk factor for DU recurrence, provided tl. pylori is eradi- catei!J INTRODUCTION Cigarette smoking has long t?cen linked to a less favorable clinical result in peptic ulcer disease. In 1949, Battcrman and Elvenfold (I) reported that patients with ulcers who had never smoked responded better to ant- acid therapy than those who smoked. Cigarette smoking is more common in patients with peptic ulcer disease Rrc•eired Alar. 9. 19y2: arrepted Aluy6. 1992. than in non-ulcer controls (2). In smokers, duodenal ulcer.healing is significantly delayed (3) and ulcer re- lapse accelerated (4, 5), even during maintenance treat- ment with H2-receptor antagonists (H2RAs) (6). With the discovery of Ilelicobaeler pylori (H. pylori), a new variable in the duodenal ulcer (DU) equation has been introduced. 11. pylori is known to be associated with over 95% of DUs in adults (7-10), and its eradi- cation results in a marked reduction in the recurrence of DU disease (I1-13), claimed by some to be a cure of the disease ( t 1-14). Our ability to reliably eradicate H. pylori with triple therapy (15) has presented us with an opportunity to study the relevance of Il. pylori to the recurrence of DU in cigarette smokers. In this study, both cigarette smokers and nonsmokers with DU were investigated before and after H. pylori eradication, with a follow-up period of up to 6 yr, to determine ulcer recurrence rates. PATIENTS AND METHODS Patients The population of subjects studied was made up of patients with symptoms of dyspepsia, acute bleeding. anemia, weight loss, or vomiting, referred by their general practitioner to the Centre for Digestive Dis- eases, a community-based endoscopy clinic. A subgroup of these patients had resistant or recurrent DUs, and results of their DU recurrence after 11. 1~•lori eradication were reported previously (10). All patients gave informed consent to take part in the study, which was conducted in accordance with the revised Decla- ration of Helsinki (16). All patients with endoscopically confirmed DUs and the presence of 11. nplori gastritis were included in tlie study. The characteristic common to all patients was the duration of follow-up (up to 6 yr) after successful H. hrlori eradication and at least yearly, or symptom- 1390

i I October 1992 SMOKING, ULCER RECURRENCE, AND H. PYLORI 1391 rclated, endoscopic monitoring with no H2RA main- tcnancc therapy.. Inquiries aboul patients' cigarette smoking habits and alcohol and drug usage were made with a standard questionnaire that was filled out before endospopy. Somc patients underwent additional invcs- tigations, such as colonoscopy, ultrasonic cxamination, gdstrin assays, laparoscopy, etc., where clinically indi- cated. Gaslroscopy All examinations were carried out by the same en- doscopist (TJB). DU was diagnosed if a mucosal crater with visible loss of substance was detected. Red patches, erosions, or aphthoid ulcerations did not constitute a diagnosis of "ulcer," and such patients were excluded. The presence of coexisting endoscopic diagnoses was recorded. Biopsy specimens were obtained from every patient. These included gastric antrum for rnicrobiolog- icai, culture and rapid urease test, and one each from the antrum and gastric body for histological assessment. Initially, only histology and urease tests were used to detect H. pylori infection, but since 1988, microbiolog- ical assessment became available and, thus, it has been included in our methodology. Hi.stological assessment Gastric biopsy tissue, fixed in 10% formalin, was staincd with hematoxylin and eosin for grading of se- verity of histologic gastritis and Gicmsa stain to grade the density of lIL p,),lori. Grading 1-111 was assigned, depending on the density of mononuclear lcukocytes (chronic gastritis) or the extent of neutrophil infiltration (active gastritis). Density of H. pylori infiltration was also graded 1-111, as previously described (17, 18). Alic•robiological assessment The specimens were plated during gastroscopy di- rectly onto culture mediaeChocolale agar media (Media Makers, Melbourne, Australia) with added vancomycin (10 mg/L), amphotcricin (20 mg/L), and trimethoprim (5 mg/L) were used. The plates were incubated for 4-5 days at 37`C in 3% 02, 10% CO2, and 85% H2 (Oxoid, USA). H. pflori was considered to be present if the culture and chemical testing for urease, catalase, and oxidase were positive. H. pylori infection was diagnosed on the basis of either 11. pylori-positive microbiological, histological, or urcase tests. Specificity and sensitivity of these meth- ods have been previously documented (19). 11. pJ lori eradic ution Since our original description of "triple thcrapy"(TT) for HL pylori eradication (20), TT has undergone several refinemcnts. Initially, TT consisted of a 4-wk course of colloidal bismuth subcitrate (CBS, 108-mg chewable tablets) with tetracycline HCI (500 mg), each 4x/day, togcthcr with mctronidazole (200 mg, 4X/day), taken for the first 10 days. Later, the same doses were used simultaneously 4x/day, but for a total of only 14 days. Most recently, to reduce side effects (21), lower-dose (but 5x/day) TT has been used, with CBS (108 mg). tctracyclinc HCI (250 mg), and mctronidazolc (200 mg). Regardless of the TT used, the end point was always documented eradication of IfL pylori. Ulcer healing and 11. pyluri eradication was con- f rmed by follow-up gastroscopy 6-8 wk after the orig- inal examination. All patients were reendoscoped at least yearly to detect ulcer recurrence. I Statistical analysis Student's I test and xZ with Yates correction have been used to detect demographic difference's between smoking and nonsmoking groups. A p value of 0.05 or less was considered to be statistically significant. ,' RESULTS . c Over a 6-yr period, 220 patients who presented to the Centre with an endoscopically confirmed ulcer were included in the study. Thirteen (6%) of these patients were excluded from the study, because they were found to be H. pflori negative during the initial endoscopy; 207 patients satisfied entry criteria, and after complet- ing their treatment, were asked to return for annual endoscopic reexaminations. During the course of the study, 10 (5%) of those patients failed to return for various reasons (e.g., moved away, felt well and saw no need for further follow-up, or declined to participate in the study). One hundred ninety-seven (121M:76F) pa- tients were finally included in this study. Six of those patients did not initially clear If. pflori infection, and were re-treated. All six became H. pylori negative and were eventually included in the study. Seven (4%) patients required re-endoscopy because of ulcer-like° symptoms (2- to 4-wk duration). On examination, either no abnormality, apthoid erosions, or esophagitis were detected. All of these patients remained H. pylori negative. Demographic data for these 197 patients can be found in Table 1. Of the 197 patients, 80 (41 %) were smokers, smoking TABLF. I Drmographic Characlcri.clics of Smokers and Nonsmokcrs x•idt Duodcna! Ulcer Charactcristic Smokcrs Nonsmokers No. 80 117 Malc:fcmale 52:28 69:48 Age range (yr) 31-76 28-76 Age (yr), mean (±SD) 50.16 (t 11.66) 54.61 (t i 3.49) Mean duration of follow- 35.71 (t 17.83) months up(tSD)

~ .~ ,.. .r. I 1392 BORODY el al. o.... < r, wr. .rt ri ,+w, DA-0.4wr P..... . ..q, a,+Aft~d ,~..,+..... ~~.. { ~n~w Ft<a. I. DU rccurroncc in smokors and nonsmokers: comparison of published data with prescnt results. between 5 and 40 cigarettes/day. Nonsmokers, I 17/ 197 (59%), included 31 (26%) people who had stopped smoking, 4-20 yr earlier; 7/80 (9%) patients who ini- tially reported smoking between 5 and 40 cigarettes/ day had ceased smoking during the follow-up period. Statistical analysis showed that there was no difference between "smokers" and "nonsmoker" groups (P = 0.186). There also was no difference in the distribution of sexes between these two groups, with p = 0.482 (Table 1). After 11. l~hlru-i eradication, patients included in the study have been followed for I yr (n = 53), 2 yr (n = 48),3yr(n=40),4yr(n=22),5yr(n=21),and6 yr (n = 13). The minimum number of follow-up months was 12 and maximum was 73, with a mean (± SD) of 35.71 (t 17.83) months. In all 197 patients with eradicated 11. pylor•i infection and cured DU, we have not seen an ulcer recur, regard- less of whether the patient was a smoker or not. In Figure 1, we compare our findings with published results. Relapse data ,f,9r smokers vc~rsus nonsmokers who have been lrcatcd and maintained on 142RA was 91 % versus 51 % at I yr, 97 % versus 82% at 2 yr, 98% vcrstrs 86% at 3 yr, and 98% I'ersus 88% at 4 yr (22). These figures do not take into account the patient's /1. 1kiplori status Published data on DU recurrence in 11. />.t/ori-ncgativc patients varies from source to source, with a reported ralc of 0-48% at I yr (13, 23-27). There appears to be no difference between recurrence rates in smokers and nonsmokers in those reports. Our data show clearly that when 11. pt'lori eradication is achieved, there is no ulcer recurrence, regardless of the patient's smoking status. DISCUSSION It is known that eradication of 11. pylvri profoundly reduces the rate of, or even prevents the recurrence of, DU (11, 13, 14, 26). However, the role of cigarette smoking in DU recurrence after 11. pylori eradication 1'rrl. 87, No. /(J, 1992 has not.;iiccn separately addressed. This study shows that smoking docs- not influence DU recurrence in patients who remain free of 11. p)'lur•i long-term. Even with a prolonged follow-ttp period of up to 6 yr, rcgard- Icss of smoking status, there was no observed duodenal ulcer recurrence. Although continuing smoking does not appear to influence the natural history of duodenal ttlcer after IIL pi'Iru•i eradication, we stress that our study is not aimed to promote cigarette smoking. Previous studies have shown that smokers are more likely to develop DU (28, 29) and have retarded DU healing (30), but those studies report only patients with ongoing 11. py'lori infection. Smokers are more likely to be 11. pylori positive (28), and those who continue to be 11. 1~I'hrri positive, even on H2RAs, are.more likely to develop recurrent DU (22, 26, 27, 31). However, the mechanism of increased aggressiveness of a DU in smokers with H. hvlori is unclear. Possible factors may include higher acid secretion, reduced pancreatic bicar- bonate secretion, reduced neutralization by alkaline' saliva, slower gastric emptying, reduced mucus produc- tion, tion, decreased mucosal blood flow, and reduced pros-~' taglandin synthesis (3, 32, 33). The virtual absence of DU recurrence, particularly in patients who continue to smoke cigarettes, suggests strongly that the contribution of !1. 1>_rlori as a risk factor for DU is much greater than ihat of cigarette smoking. Removal of this infection and healing of the associated gastritis allows the stomach and duodenum to withstand the destructive effects of smoking. We conclude, therefore, that smoking is not a risk factor for DU recurrence, provided 11. Irrluri is eradicated. ACKNOWLEDGMENT Prcliminary results were presented in abstract form at the Digestive Disease Week meeting. May 12-18. 1990, San Antonio. Texas. Reprint requests and correspondence: Dr. Thomas J. 13orody. Ccntic fnr Digestive Discases, 144 Great North Road. Five Dock- 2Q46 N.S.W.. Australia. REFERENCES 1. Rattcrman RC. l:hrcnfold 1. lnflucncc of smoking upon the management of thc pcptic ulccr patient. Gastrocntcrology 1949; I 2:57-85. 2. Kikcndall JW. Evaul J. Johnson LF. EfTcct nf cigarette smoking iin gastrointestinal physiology and non-neoplastic digestive dis- cace. J Clin Gaorocntcrol 1984:6:65-79. 3. Katschinski nl). Gochcll I t, Arnold R, ct al. Smoking as a risk factor for slnw <iuodcnal ulccn healing. Eur J Gastrocnlcrol I Icpatill 1991:3:443-7. ._...-. 4. Sontag S. Graham I)t', nclsilo A, ct al. Cimctidinc. cigarcttc ~ smoking and recurrence of duodenal ulcer. N Engl J Mcd 1984. 331:6R9-93. ~ 5. Korman MG. ilansky J. Eavcs ER. ci al. Intlucncc of cigarcttc smoking on healing and relapse in <lucx)cnal ulccr discasc. Gas- trncnlcrology 1983:85:871-4. 6. l.cc FI, 1lardman M, Jadcrixrg ME. Maintenance treatment of ri~ W

Oc'luhc'r 1992 SMOKING, ULCER RECURRENCE, AND N. PYLORI 1393 duodcnal ulccration: Ranilidine 300 mg at nighl is bettcr than i50 mg in cigarcttc smokers. Gut 1991;32:t51-3. 7. Ho J, Lui 1, Hui WM, ct al. A study on thc corrclation of duodcnal ulcer healing with campylobaclcr-likc organisms. Gas- trocnlcrol FlcpaloJ 1986:1:69-74. 8. Rauws EAJ, Langcnbcrg W. Houthoff HJ, ct al. CijmpJlnbac•rcr 1!rloridi.c=associated active chronic gastrilis: A prospective study of its prevalence and the effects of antibacterial and anti-ulccr slicatment. Gastrocntcrology 1988;94:33-40. 9.' Wyatt J). Relationship of C nvlnri to duodcnal ulcer disease. In: Blascr MJ, cd. Camlrylnbacier pylnri in gastritis and peptic ulcer disease. New York: lgaku-Shoin, 1989:99-104. 10. Borody TJ, George Lly Brandl S, et al. 1ielicobacler p1/nri- ncgativc duodcnal ulcer. Am 1 Gastroentcrol 1991;86:1154-7. i I. George LI, Borody TJ, Andrews P, ct al. Cure of duodenal ulcer a(ler eradication of Helicobacier pylori. Med J Aust 1990;153t' 145-9. 12. Rauws EAJ,'I'ytgat GNJ. Curt of duodenal ulcer associated with eradication of Helicabacler py%ri. Lancet 1990;335:1233-5. 13. Borody TJ, Cole P, Noonan S, ei al. Recurrence of duodenal ulcer and Campylnbacter pylori infeciion after eradication. Mcd J Aust 1989;151:431-5. 14. Graham DY, Lew GM, Klein PD, et al. Effect of treatment of llelicobacier pylorl on the long-term recurrence of gastric or duodenal ulcer: A randomized controlled study. Ann intern Med 1992;116:705-8. 15. Axon ATR. Ilelicobacter pylori therapy: Effect on peptic ulcer disease.l Gastroenterol Hepatol 1991;6:131-7. 16. World Medical Assembly. The declaration of Helsinki. World Mcd 1 1982;29:86-8. 17. Marshall Bi, Warren JR. Unidentified curved bacilli in the stomachs of patients with gastritis and peptic ulceration. Lancet 1984;I:131I-5. - 18. Whitehead R, Truelove SC, Gear MWC. The histological diag- nosis of gastritis in fibreoptic gastroscope biopsy specimens. J Clin Pathol 1972:25:1-11. 19. Borody TJ, Andrews P, Brandl S, et al. Endoscopic diagnosis of llelicabrielcrl~tlnri (HP)-approaching the °gold standard." In: Abstracts of the World Congress of Gastrocnlcrology. Sydney 1990. Abingdon: The Medicine Group (UK) Ltd, 1990:PD91. 20. Borody TJ, Carrick 1, Hazel SL. Symptoms improve after cradi- cation of gastric Campylabac•rer pylaridis. Mcd J Aust 1987; 146:450-I. 21. Borody TJ, Brandl S. Andrews P. Side cffccts of triple therapy. )7th International Congress of Chemotherapy (proceedings). June 23-28. 1991. Munchen: Futuramcd Verlagsgcselischafl mbH Munchen. Abstract 561. 22. Lane MR, Lee SP. Recurrence of duodenal ulcer after medical treatment. Lancet 1988;1:1147-9. 23. Porro GB, Lazzaroni M. Campylobacrer pylori and ulcer rccur- rence. Lancet 1988;1:593. 24. Bardhan •KD, Singh S. Thompson M. Preventing relapse of duodenal ulcer (DU): Maintenance treatment (MT) with colloi- dal bismuth. Gastrocntcrology 1990:98(S):A 18. 25. Patchett S, O'Riordan T,•Leen E, et al. A prospective study of Ilelicobacler pylnri eradication in duodenal ulcer. Gastroenler- ology 1990;98:A 104. 26. Coghlan JG, Humphries H, Dooley C, el al. Campylobacrerpylori and recurrence of duodenal ulcers: A 12-month follow-up study. Lancet 1987;2:1109-11. 27. Fiocca R, Solcia E, Santoro B. Duodenal ulcer relapse aftcr eradication of lielicobacler pylnri. Lancet 1991;337:1614. 28. McNulty CAM, Dent JC, Uff JS, ei at. The prevalence of C. pylnri in 1447 patients al endoscopy. Am J Gastroenterot 1988: 83:1035. , 29. Martin DF, Montgomery E, Dobek AS, ei al. CampJ•lobacrer pJlori, NSAIDS, and' smoking: Risk factors for pcptie ulcer disease. Am J Gaslrocnterol 1989;84:1268-72. 30. Lauritsen K, Rune S1, Bytzcr P, el al. Effect of omeprazolc and cimetidinc on duodenal ulcer. N Engl 1 Med 1985;312:958-61. 31. Lee Fl, Samloff IM, Hardman M. Comparison of lri-potassium di-citrato bismuthale tablets with ranitidinc in healing and relapse of duodenal ulcers. Lancet 1985:1:1299-301. 32. Kaufmann D, Wilder-Smith CH. Kcmpf M, el al. Cigarette smoking gastric acidity and peptic ulceration. Dig Dis Sci 1990; 35:1482-7. 33. Lindell G, Brudin L Ohlin P. et al. Does smoking influence intragastric acidity? Eur J Gastrocnlcrol I lcpatol 1990:2:305-8. N C .: ,