Influence of Smoking on Basal and on Vagally and Maximally Stimulated Gastric Acid and Pepsin Secretion A . LANAS &L T 3 . T . HIRSCHOWITZ Division of Gastroenterology, University ol Alabama at Birmingham, Birmingham. Alabama, LJSA
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Auckland on 11/13/14 For personal use only.
Lanas A . Hirschowitz BI. Influence of smoking on basal and o n vagally and maximally stimulated gastric acid and pepsin secretion. Scand J Gastroenterol 1992. 27. 208-212 Published data show that smokers have greater basal or peak acid and pepsin outputs, but the mechanisms underlying thcsc effects arc unknown. To confirm this and to determine whether these findings extend to, and implicate. m y vagal overxtivity. gastric secretions collected for 1 h basally, 1 h after 15 min of modified sham feeding (MSF), and 1 h alter pentagastrin ( 6 kg/kg subcutancously) were analy7ed for acid and pepsin content in 204 subjects, I04 with duodcnal ulcer (66 smokers) and 101 without (57 smokers). Maximal acid outputs (MAO, Keq/kg/h, i* SEM) were higher in smokers than in nonsmokers in hoth duodenal ulccr ( D U ) (623 2 35 vcrsus 491 2 35, p i0.005) and non-DU (502 1 32 versus 376 f 20, p .::O.OOS). Basal and MSF accretions were generally increased in sniokcrs but, when expressed as a percentage of MAO. were not different in smokers and non-smokers (18% versus 1744 and 43% versus 39%, respectively. in DU. and 13%, vcrsus 16% and 40%' versus 36% in [ion-DU). Maximal pepsin outputs (units x 10 '/kg/h) were also higher in smokcis than in non-smokers (nu, 129 L 7.9 vcrsus 105 2 Y.5.p = 0 . 0 5 . a n d non-DU. 101 i 7.5 versus 77 ? 10, p = 0 . 0 5 ) .Basal and MSF secretions as a percentage o f maxinial pcpsin outpul wcrc not different in smokers vcrsus non-smokera. Multivariate logistic rcgrcssion shows that smoking was most strongly associated with MA0 a n d sham fccding outputs, but the duration-intensity (pach-years) ol smoking was ;issociated only with cicvatcd M A O . We conclude t h a t smoking is assuciatcd with increased maximal acid a i d pcpsin outpul in hoth DLJ and non-DU populations. Smoking doe< not SCCIII to aICcct independently vagal tim mu la ti on of gabtric ~ccretion. K r y words Acid; duodcnal ulccr; gastric secretion: gastrin: pcpsin; smoking
Smoking increases the risk for peptic ulccr proportionally to ttic aniount smoked, impairs ulccr healing, : i d promotes the rate of rccurrencc of duodenal ulccr (DU) discase, but the ~';tthoi'hysioiogic mechanisms underlying these effects iire not understood (1-6). Although there arc contradictory rcporth (7-Yj. some published dnta show that smokers with or without D U (l& discasc have higher gastric sccrction than non-s~~i(ikers 12). The mechanisms that could explain this higher gastric secretory capacity i n smokcrs have not yet heen elucidated. I t has heen reported that chronic administration of nicotine at 'smoking doses' increases the volume of acid and pepsin outputs in rats, prdiahly through ;I viigal pathway (13. 13). Long-term cholincrgic therapy in humans ( IS) and chi-onic sham feeding in dogs ( I h j increase stimulated acid secretion. To determine whether smoking increases gastric secretion i n man and whcther it acts by vagal drive. the present study in DU populations compared basal gastric acid and pepsin secretion i n smokers and rion-smokers a n d further analyzed the response t o vagal stimulation with reference to maximal secretory capacity.
PA'T'IEN'I'S A N D METHODS Bctwccn 1979 and 1990, 266 patients underwent gastric analysis involving vagal stirnulation with modified sham feed-
ing (MSF). The studies wcre performed f o r clinical indications, and where appropriate, human use approval wits in force. All antisecretory drugs were withdrawn for at least 4X h before the test was performed. In each test, after ii 10-h overnight fast, a nasoga\tric sump tube was positioncd fluoroscopically in the most dependent part of the s t o m x h . Fasting residual contents were removed, and secretion was collected in 10-min consecutive pcriods for 3 ti a [ constant Y(L120 mmHg suction. Saliva was aspirated by continuous suction. Aliquots were titrated with 40 mM NaOH to pH 7 with a Radiometer autoburette (Radiometer Americiin, West lake, O h i o , USA). Pepsin concentraticin wits measured i n each sample with an automated method (17). using hovine hernoglohin substrate (Gibco, Madison, Wisc., USA) at pH 2.0 with reference t o crystalline pepsinogen (Sigma Chemical Co., St Louis, Mo., USA) as the st;indard ( 1 wg = 1 peptic units) . After a liasal period of 60 min the patients were asked to taste and chew, but no( swallow, a cheeseburgcr for 15 min (modified sham feeding (MSF)) (18). One h o u r after starting the MSF period, 6 Fg/kg of pentagastrin was given by subcutaneous injection, and gastric secretion was collected for an additional hour.
Smoking and Gastric Secretion
209
Table I. Characteristics of duodenal ulcer and non-duodenal ulcer populations ~
Duodenal ulcer
n
Age, years Weight, kg Men Smoking Years Pack/day
Non-duodenal ulcer
Smokers
Non-smokers
Smokers
Non-smokers
66 45.1 i 1.7 66.4 i 1.7 41 (62%)
37 45.0 2 2.9 75.3 5 2.9* 22 (59.3%)
57 45.3 i 1.9 71.2 t 2.4 40 (70%)
44 43.8 2.2 70.2 f 2.8 16 (36.3%)"
19.6 % 1.8 1.3 2 .07
0
21.4 t 2.2 1.2 t .07
0 0
0
*
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Auckland on 11/13/14 For personal use only.
The non-duodenal ulcer population consisted of 20 controls, 30 patients with esophagitis, 15 with gastroduodenitis, 12 with diarrhea, 10 with gastric ulcer, 5 with hiatal hernia, 5 with hypomotility disorders, and 4 with other diagnoses.
Blood samples were drawn in 112 persons to determine fasting gastrin levels by radioimmunoassay (19).
Stdsticul unulysis Results are expressed as mean SEM. Wilcoxon rank sum tests and Student's t tests for equal o r unequal variances were used for comparison between groups. Categoric data were analyzed by means of chi-square or Fisher's exact test when appropriate. Multiple regression (general linear models) was used when variables were to be associated with gaidric secretion variables in a multivariate context.
*
appeared when these were expressed as a percentage of maximal output. When the groups were further divided by the presence and absence of DU, smokers still had high secretion, but because of smaller numbers the differences did not reach statistical significance in all subgroups (Table
Smoking arid RUSIriC secretion I n all 204 patients combined, smokers had significantly higher basal, vagally stimulated (MSF), and maximal outputs of acid and pepsin than non-smokers (Fig. l), but differences for basal and MSF-stimulated secretion dis-
-
0Non-Smokers
.
DSrnokers
Acid f
60 50 W
RISLJLTS Two hundred and four patients were included in this analysis, 103 DU and 101 non-DU patients. D U are considered separately because as a group D U patients have increased rates of secretion (20). Thirty-eight patients who had quit smoking for variable periods were excluded from the analysis because they could not be considered either non-smokers or smokers. In addition, 24 patients with previous gastric surgery or concomitant drug ingestion were also excluded for obvious reasons. Table I shows the characteristics of smokers and nonsmokers in both D U and non-DU patients. Smokers made up two-thirds of the D U populations. Body weight was signiticantly lower in D U smokers than D U non-smokers ( M . 3 t 1.7 versus 75.3 2 2.5 kg, p = 0.007). Weight and lean body mass were lower in both male and female smokers but reached statistical significance only in the women (data not shown). Among the non-DU population the male to female ratio was twice as high in smokers (70%) as in nonsmokers (30%, p < 0.001), whereas in DU patients the sex ratio was equal.
80 70
E
40
30 20 -
1600 f 1400
-
1200
400
0 0 0 I , I
n -
X
0
-
10
0
0
' I >
F4
-
o
j
-
0
:
0
-
1
0'
0
0
0
0 :
o
i
0
:
: b b
t
: , ,
.
i :
(P
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Auckland on 11/13/14 For personal use only.
Lanas A . Hirschowitz BI. Influence of smoking on basal and o n vagally and maximally stimulated gastric acid and pepsin secretion. Scand J Gastroenterol 1992. 27. 208-212 Published data show that smokers have greater basal or peak acid and pepsin outputs, but the mechanisms underlying thcsc effects arc unknown. To confirm this and to determine whether these findings extend to, and implicate. m y vagal overxtivity. gastric secretions collected for 1 h basally, 1 h after 15 min of modified sham feeding (MSF), and 1 h alter pentagastrin ( 6 kg/kg subcutancously) were analy7ed for acid and pepsin content in 204 subjects, I04 with duodcnal ulcer (66 smokers) and 101 without (57 smokers). Maximal acid outputs (MAO, Keq/kg/h, i* SEM) were higher in smokers than in nonsmokers in hoth duodenal ulccr ( D U ) (623 2 35 vcrsus 491 2 35, p i0.005) and non-DU (502 1 32 versus 376 f 20, p .::O.OOS). Basal and MSF accretions were generally increased in sniokcrs but, when expressed as a percentage of MAO. were not different in smokers and non-smokers (18% versus 1744 and 43% versus 39%, respectively. in DU. and 13%, vcrsus 16% and 40%' versus 36% in [ion-DU). Maximal pepsin outputs (units x 10 '/kg/h) were also higher in smokcis than in non-smokers (nu, 129 L 7.9 vcrsus 105 2 Y.5.p = 0 . 0 5 . a n d non-DU. 101 i 7.5 versus 77 ? 10, p = 0 . 0 5 ) .Basal and MSF secretions as a percentage o f maxinial pcpsin outpul wcrc not different in smokers vcrsus non-smokera. Multivariate logistic rcgrcssion shows that smoking was most strongly associated with MA0 a n d sham fccding outputs, but the duration-intensity (pach-years) ol smoking was ;issociated only with cicvatcd M A O . We conclude t h a t smoking is assuciatcd with increased maximal acid a i d pcpsin outpul in hoth DLJ and non-DU populations. Smoking doe< not SCCIII to aICcct independently vagal tim mu la ti on of gabtric ~ccretion. K r y words Acid; duodcnal ulccr; gastric secretion: gastrin: pcpsin; smoking
Smoking increases the risk for peptic ulccr proportionally to ttic aniount smoked, impairs ulccr healing, : i d promotes the rate of rccurrencc of duodenal ulccr (DU) discase, but the ~';tthoi'hysioiogic mechanisms underlying these effects iire not understood (1-6). Although there arc contradictory rcporth (7-Yj. some published dnta show that smokers with or without D U (l& discasc have higher gastric sccrction than non-s~~i(ikers 12). The mechanisms that could explain this higher gastric secretory capacity i n smokcrs have not yet heen elucidated. I t has heen reported that chronic administration of nicotine at 'smoking doses' increases the volume of acid and pepsin outputs in rats, prdiahly through ;I viigal pathway (13. 13). Long-term cholincrgic therapy in humans ( IS) and chi-onic sham feeding in dogs ( I h j increase stimulated acid secretion. To determine whether smoking increases gastric secretion i n man and whcther it acts by vagal drive. the present study in DU populations compared basal gastric acid and pepsin secretion i n smokers and rion-smokers a n d further analyzed the response t o vagal stimulation with reference to maximal secretory capacity.
PA'T'IEN'I'S A N D METHODS Bctwccn 1979 and 1990, 266 patients underwent gastric analysis involving vagal stirnulation with modified sham feed-
ing (MSF). The studies wcre performed f o r clinical indications, and where appropriate, human use approval wits in force. All antisecretory drugs were withdrawn for at least 4X h before the test was performed. In each test, after ii 10-h overnight fast, a nasoga\tric sump tube was positioncd fluoroscopically in the most dependent part of the s t o m x h . Fasting residual contents were removed, and secretion was collected in 10-min consecutive pcriods for 3 ti a [ constant Y(L120 mmHg suction. Saliva was aspirated by continuous suction. Aliquots were titrated with 40 mM NaOH to pH 7 with a Radiometer autoburette (Radiometer Americiin, West lake, O h i o , USA). Pepsin concentraticin wits measured i n each sample with an automated method (17). using hovine hernoglohin substrate (Gibco, Madison, Wisc., USA) at pH 2.0 with reference t o crystalline pepsinogen (Sigma Chemical Co., St Louis, Mo., USA) as the st;indard ( 1 wg = 1 peptic units) . After a liasal period of 60 min the patients were asked to taste and chew, but no( swallow, a cheeseburgcr for 15 min (modified sham feeding (MSF)) (18). One h o u r after starting the MSF period, 6 Fg/kg of pentagastrin was given by subcutaneous injection, and gastric secretion was collected for an additional hour.
Smoking and Gastric Secretion
209
Table I. Characteristics of duodenal ulcer and non-duodenal ulcer populations ~
Duodenal ulcer
n
Age, years Weight, kg Men Smoking Years Pack/day
Non-duodenal ulcer
Smokers
Non-smokers
Smokers
Non-smokers
66 45.1 i 1.7 66.4 i 1.7 41 (62%)
37 45.0 2 2.9 75.3 5 2.9* 22 (59.3%)
57 45.3 i 1.9 71.2 t 2.4 40 (70%)
44 43.8 2.2 70.2 f 2.8 16 (36.3%)"
19.6 % 1.8 1.3 2 .07
0
21.4 t 2.2 1.2 t .07
0 0
0
*
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Auckland on 11/13/14 For personal use only.
The non-duodenal ulcer population consisted of 20 controls, 30 patients with esophagitis, 15 with gastroduodenitis, 12 with diarrhea, 10 with gastric ulcer, 5 with hiatal hernia, 5 with hypomotility disorders, and 4 with other diagnoses.
Blood samples were drawn in 112 persons to determine fasting gastrin levels by radioimmunoassay (19).
Stdsticul unulysis Results are expressed as mean SEM. Wilcoxon rank sum tests and Student's t tests for equal o r unequal variances were used for comparison between groups. Categoric data were analyzed by means of chi-square or Fisher's exact test when appropriate. Multiple regression (general linear models) was used when variables were to be associated with gaidric secretion variables in a multivariate context.
*
appeared when these were expressed as a percentage of maximal output. When the groups were further divided by the presence and absence of DU, smokers still had high secretion, but because of smaller numbers the differences did not reach statistical significance in all subgroups (Table
Smoking arid RUSIriC secretion I n all 204 patients combined, smokers had significantly higher basal, vagally stimulated (MSF), and maximal outputs of acid and pepsin than non-smokers (Fig. l), but differences for basal and MSF-stimulated secretion dis-
-
0Non-Smokers
.
DSrnokers
Acid f
60 50 W
RISLJLTS Two hundred and four patients were included in this analysis, 103 DU and 101 non-DU patients. D U are considered separately because as a group D U patients have increased rates of secretion (20). Thirty-eight patients who had quit smoking for variable periods were excluded from the analysis because they could not be considered either non-smokers or smokers. In addition, 24 patients with previous gastric surgery or concomitant drug ingestion were also excluded for obvious reasons. Table I shows the characteristics of smokers and nonsmokers in both D U and non-DU patients. Smokers made up two-thirds of the D U populations. Body weight was signiticantly lower in D U smokers than D U non-smokers ( M . 3 t 1.7 versus 75.3 2 2.5 kg, p = 0.007). Weight and lean body mass were lower in both male and female smokers but reached statistical significance only in the women (data not shown). Among the non-DU population the male to female ratio was twice as high in smokers (70%) as in nonsmokers (30%, p < 0.001), whereas in DU patients the sex ratio was equal.
80 70
E
40
30 20 -
1600 f 1400
-
1200
400
0 0 0 I , I
n -
X
0
-
10
0
0
' I >
F4
-
o
j
-
0
:
0
-
1
0'
0
0
0
0 :
o
i
0
:
: b b
t
: , ,
.
i :
(P
