Introduction

Gastroesophageal reflux disease (GERD) is becoming increasingly common, as is obesity. Up to now, a clear association between the two conditions has not been established. However, a new study from the Karolynska Institute in Stockholm provides evidence of a relationship. The investigators also wanted to see if the relationship was influenced by female sex hormones.

Method

This was a large, population-based, cross-sectional, case-control study. Data were obtained from two surveys conducted in Norway with a 9-year interval between them. Each participant completed an extensive questionnaire at a local health center, where they also had a physical exam, and measurement of body weight and height (to provide the body mass index, or BMI).

Determination of GERD status was based on self-assessment of severe heartburn or regurgitation symptoms (i.e. reflux) during the previous 12 months. The analyses were done on the 58,596 second survey subjects who answered this question. Sixty-nine percent said they had had no reflux symptoms, 26% had had minor symptoms, and 5% reported severe symptoms. After exclusion of pregnant women and those with incomplete data, there were 3,113 subjects with severe symptoms who represented the 'case' group, and 39,872 subjects with no symptoms who were the control group.

The net changes in BMI between the first and the second survey were used to establish a possible effect of weight gain or weight loss on the occurrence of GERD.

Results

Over half the participants in the analysis were women (57%). The mean age of the case subjects was 52 years, while that for the controls was 48 years. The mean BMI for cases was 28.1, compared with 25.8 for controls. Among the female cases 14.8% had used hormone therapy, compared with 9.6% of the female controls.

There was a dose-dependent relationship between BMI and reflux symptoms in both men and women. Compared with normal-weight persons (BMI <25), the odds ratio (OR) for developing GERD symptoms was 3.3 (95% CI; 2.4-3.7) for severely obese men (BMI >35). In women, the OR was 6.3 ((5% CI; 4.9-8.0) for the severely obese.

Premenopausal women had a significantly higher risk than postmenopausal women (ORs of 6.8 and 4.2, respectively). However, normal-weight postmenopausal women who had used estrogen-only replacement therapy at anytime had more than twice the risk of developing GERD than women who had never used hormones. (Women who had used combination hormone replacement therapy had a non-significant tendency to increased risk - OR 1.3; 95% CI; 0.9-1.8.)

Emphasizing the roles of obesity and estrogen in association with GERD, the OR for women treated with estrogen-only hormone therapy (i.e. previous hysterectomy subjects) and with a BMI >35 was 33.3 (95% CI; 11.8-94.4).

Weight changes during the previous 9 years appeared relevant. A net gain of 3.5 BMI units was associated with an increase of 2.7 in the OR for GERD, while a loss of 3.5 BMI units was linked to a 0.6 decrease in the OR.

Comment

The investigators have shown that there is a clear association between overweight and GERD, as well as between estrogen use by women and GERD. They suggest that this may be because obese women have increased synthesis of estrone in fatty tissues, as well as raised circulating levels of unbound estradiol.

Whether this explanation for the findings is correct or not, the important finding from this study is that an increase in weight increases the likelihood of developing GERD, while decreasing weight can reduce the risk of this. So, there's another reason to attack overweight first!