A n investigation into the relationship between cigarette smoking and diverticular disease o f the colon

: A retrospective study was conducted to determine the relation ,hip between cigarette smoking and diverticular disease of the colon. One hundred and two patients undergoing barium enema were assessed for diver ticular disease and smoking history. No significant positive association between smoking and diverticular disease was found. Can J Gastroenterol 1990;4(5 ): 193,195

arc generall y acLcpted .1s re~ponsihlc for the eventual herniation nf mucosa through hn:ab in 1hc rnusck at the site of penet ration hy sm,111 arteri es, wirh the formm1o n of typical d ivcrricular sacs ( 4 ).
TPhaccn smoking has hcen ,mplrcated as a factor in 1he pni hogcnes1s of several gastrointestinal Jiscascs (5,6), a lrhnugh no stud ie, have looked at Its relationshi p lO divert icular d isease.
It was hypotherned that Ligarctte smuk mg, through nicrn me, cou ld inc rease the colonic intrnl uminal pressure and thu~ contrihute to the developmen t of diverticula. Nicoti ne, a signi fica nt conslltucnt of cigarettes, st imulates con tractile activity of the dista l de ccnd ing colon (7) and can be sufficiently forcefu l to c.n,se defecation in unancsthet izcd animab (8). Smoking has abo hcL'n shown to increase in test inal motili ty in humans (9).
An alternative mechanism prnpm.ed was through vascular cnmpromise ns a result Ll l artcrillsc leros1s. Diven ic ul:ir disease and ancrioscli:rosi~ arc a~-

PATIENTS AND METHODS
All patients who were referred to a hospital rad iology department fo r outpatient barium enemas over a four month period were chosen for entry into a retro~pective study. Of the 200 patients contacted, 102 patients suppl ied complete and re liable smo king histories and were thus 111cluded.
The smo king d;:ica obta ined was organized into a 'pack-year' system. 'One pack-year' was defined as one package a day for one year. Nonsmokers were defined as patient~ who had never smoked cigarettes.
The bari um enemas were performed by do uble contrast technique and were reviewed by o ne radiologist who was not aware of the smoking histories of the patients. The radio logic c riterion

RESULTS
The prevalence o f diverticu la r disease inc reased with age ( Figure 1 ). No d iverticula were found earlier than the fourth decade of life.
The sigmo id colon was the most common anatomic site fo r dive rtic ula (seen in 85°/ti o f patients with diverticula). Other sites, in order of decreasing frequency, were descending (56%), transverse 02%). and ascending colo n (32%).
Forty-one of the 102 patients had diverticula o n barium ene ma . Sixtyseven had a positi ve smo king history. The percentage of patient~ with a smoking history was identical in patients with diverticula LO those without diverti cul a ( Table I) .
The inc idence ofd iverticular disease did not change with increasing numbers of cigarettes smoked (Tabli.' 2).

DISCUSSION
T he data in the present !>Ludy me in general agree ment with other st udics with respect to age d istrihu tion and anatomic site of diverricula (2,4, l 1 ).
The data failed to provide evidence that divert icular disease is associated with a history of smoking o r the number of c igarettes smoked in a lifetime. T here are many difficultie/> that .irisc when undertaking experimental work into the effects of c iga rette smoktng, wh ic h were well recognized tn .t study that reviewed c iga rette smoking and inflammatory bowel disease (12). for example, there a re in excess of 3900 chemicals in c iga rette smo ke, and it is perhaps naive to propose an effect Jue to o ne compo nent ( 13).
The endpoint chosen fo r this study was the absence or presence of diverticula. Barium enema 1s a widely used and sensi ti ve measu re of diverticula (3,14). No attempt was made to quantirate the number of diverticula nor the severity of symptoms. Thus it is poss1hle that the study was not sufficientl y sensi tive, a lthN1gh t his is unlikely si nce the number of pa tiems with diverticu lnr disease d id nm increase wi th the number of cigarettes smoked.
The most evident source of error in this study is selection bias, as only patients who needed a barium enema were included. Howeve r, most data on have no t been studied extens ively in man and thus it is poss ibl e that levels during smoking are not h igh enough to stimulate rhe colon s ign ificantly. There is controversy (6) about the interpretation of the tlata in unanesthetizeJ animals (8)