Effect of smoking on serum pepsinogen I level depends on serological status of Helicobacter pylori

Serum pepsinogen (sPG) levels are used in gastric cancer screening programs . However, modification of sPG levels by smoking habit, according to the st atus of Helicobacter pylori (H. pylori) infection has been little investiga ted. This study investigated the effects of smoking on serum levels of peps inogen I (PG I), pepsinogen II (PG II), and gastrin by IgG titer of antibod y against H. pylori (Hp-IgG titer) using the data from 356 current-smokers and 262 non-smokers (133 never-smokers and 129 ex-smokers) in a cross-secti onal study of 618 men aged 40 to 49 years. PG I, PG II, PG I/PG II ratio an d gastrin were significantly associated with Hp-IgG titer in never-smokers [Spearman's correlation coefficient (95% confidence interval): 0.23 (0.07, 0.39), 0.52 (0.41, 0,63), -0.40 (-0.54, -0.27), and 0.25 (0.10, 0.41), resp ectively], However, the correlation coefficients of PG I and PG II decrease d in current-smokers, 0.02 (-0.1, 0.13) and 0.32 (0.22, 0.42), respectively , In H. pylori seronegative and low titer cases, the mean PG I level was si gnificantly (P<0.01) higher in current-smokers, compared with non-smokers. However, in high titer cases, the mean PG I level was lower in current-smok ers. Mean PG II and gastrin levels, and PG I/PG II ratio did not differ acc ording to smoking habits by Hp-IgG titer, The gastrin level was significant ly correlated with PG II, but not PG I. These data indicate that current sm oking influences the serum PG I level depending on Hp-IgG titer and the ass ociations between sPGs and Hp-IgG titer, Gastrin is not involved in the mod ification of PG I levels by smoking.