Dietary habits and obesity indices in patients with gastro-esophageal reflux disease: a comparative cross-sectional study

Gastro-esophageal reflux disease (GERD) is one of the most common disorders of the upper digestive tract worldwide. The prevalence of GERD is estimated within the range of 18.1% to 27.8% in North America, 8.8% to 25.9% in Europe, and 8.7% to 33.1% in the Middle East [1]. In Iran, prevalence of GERD is estimated within the range of 1.9 to 52% [2]. GERD is characterized primarily based on typical symptoms of heartburn and regurgitation. In addition to the esophageal manifestations of GERD, its effect on quality of life as well as extra-esophageal presentations such as dry cough, chest pain, sleep difficulties, and teeth decays, have been increasingly recognized among gastroenterologists [3, 4]. Up to now, a combination of lifestyle modifications and anti-reflux medications are recommended for the management of GERD symptoms [5].

There is evidence that several socio-demographic and lifestyle factors are predictors of GERD development. The roles of some risk factors such as obesity are well known in epidemiological studies [6]. However, evidence on other variables particularly dietary habits are still inconclusive. Current guidelines recommend that reflux-triggering foods such as coffee, chocolate, spicy foods, and citrus fruit should not be routinely eliminated in management of GERD [7]. However, some studies revealed that specific dietary choices were related to higher risk of GERD development or the severity of the disease [8, 9]. In recent years, increased awareness of the side effects of medication on the general population shifted attention gradually, from pharmacological therapies to dietary modifications as the first-line management for GERD [10, 11]. Besides, it is proposed that eating habits may contribute to GERD risk through their effect on body weight [12]. Unfavorable eating patterns, which are mainly characterized by frequent consumption of carbonated drinks, fast foods, and large portion-size meals are positively associated with development of obesity [13] which in turn, may contribute to increased GERD risk.

Although, there are many studies in Iran investigating predictors of GERD risk in terms of lifestyle related factors such as smoking and obesity, evidence on the association between dietary habits and GERD are scarce and even inconsistent. A study from North of Iran showed a positive association between lying down after meals, with increased frequency of GERD, however such an association was not found for consumption of beverages and spicy foods [14]. Similarly, a study in South of Iran found no significant association between consumption of tea, coffee, and fast foods, with GERD [15]. High burden of disease in the Iranian population [16], makes it necessary to identify factors associated with GERD in different regions, for implementing prevention programs. Therefore, the aim of the present study was to determine independent factors associated with GERD in patients referred to a teaching hospital in Zahedan, South-East of Iran.

A total of 505 participants consisting of 285 GERD and 220 Non-GERD cases were included. Table 1 compares socio-demographic status and obesity indices between GERD and Non-GERD participants. In comparison with the Non-GERD group, the GERD group had significantly higher percentage of married individuals (p = 0.02) as well as the presence of concomitant disease (p = 0.001). Level of education (p = 0.001) and regular physical activity ≥2 h per week (p = 0.03) inversely influenced the prevalence of GERD. In addition, the prevalence of general obesity (p = 0.01) and central obesity (p = <0.001) were significantly higher in the GERD group compared to the Non-GERD ones. Dietary habits of participants are presented in Table 2. There were no significant differences in dietary habits between the two groups.

Table 3 shows results of univariate and multiple regression analysis for detection of independent factors associated with GERD development, as the study outcome. In univariate analysis, being married (OR = 1.57, 95%CI = 1.04, 2.36), general obesity (OR = 1.77, 95%CI = 1.11, 2.81), central obesity (OR = 2.09, 95%CI = 1.46, 3.01), and consumption of citrus fruits between meals (OR = 1.69, 95%CI = 1.04, 2.73) were significantly associated with higher odds of GERD, while higher level of education (OR = 0.53, 95%CI = 0.36, 0.77) and regular physical activity ≥2 h/week (OR = 0.53, 95%CI = 0.30, 0.94) were associated with lower odds of GERD. Based on the Hosmer-Lemeshow goodness-of-fit test, the regression model showed a good calibration (Chi-square 7.52; degrees of freedom = 8; p-value = 0.48). Also, no sign of multicollinearity was observed between variables, as all variables had VIF ≤ 2 and thus, all independent variables were included in the regression model. In the multiple regression model, central obesity (OR = 1.88, 95%CI = 1.18, 3.01) and consumption of citrus fruits between meals (OR = 2.22, 95%CI = 1.30, 3.81) were associated with increased odds of GERD, while higher level of education (>12 years) (OR = 0.55, 95%CI = 0.33, 0.91) was associated with decreased odds of GERD.

The aim of the present study was to assess dietary habits, lifestyle, and socio-demographic status of GERD patients in South-East of Iran. Although central obesity and citrus fruit intake were independent factors associated with GERD, such an association was not found for other dietary habits, BMI, and socio-demographic status. Previous researches have indicated that high BMI, an index of general obesity, is a risk factor for GERD development and severity of symptoms [20, 21]. In this study, central obesity, determined by high WC was an independent predictor of GERD. Although high BMI was associated with higher odds of GERD in the univariate analysis, this association remained non-significant in the adjusted model analysis. Excess abdominal fat can increase the risk of GERD through several mechanisms including increased intra-gastric pressure, decreased lower esophageal sphincter (LES) pressure, and increased frequency of transient lower esophageal sphincter relaxation (TLSR), which results in acid reflux [22]. Similarly, a study on 2457 GERD patients reported that central obesity as determined by WC was one of risk factors inducing erosive esophagitis [23]. Another study by Crews et al. reported that central obesity based on high values of waist-to-hip ratio (WHpR), was an independent risk factor of erosive esophagitis [24]. It should be noted that differences in obesity indices in predicting GERD risk have been reported in other chronic disorders. A previous study found that central obesity determined by high WC was a strong predictor of adverse cardiometabolic profile, compared to BMI or WHpR [25]. In contrast, Gunji et al. found that BMI was a predictor of erosive esophagitis, while such an association was not found for central obesity measured by visceral adipose tissue and waist girth [26]. Moreover, some studies have shown that general obesity is a risk factor for severity of GERD symptoms alongside its development. However, it should be noted that the prevalence of general obesity was lower than that of central obesity in our study (19.4% vs. 46.1%), which may be suggestive of the differences between the BMI in our study compared with that of others.

The association between diet and GERD has been an issue of interest in epidemiological studies. In the present study, citrus fruit intake was the only dietary factor that might be associated with higher odds of GERD. Up to now, results on diet-GERD relationship are inconsistent among various studies. Several studies reported that consumption of caffeinated beverages (coffee and tea), carbonated drinks, citrus fruit, chocolate, spicy foods, and large-volume meals, were related to higher risk of GERD or triggering symptoms [8, 9, 27‐29]. It is proposed that these foods can decrease LES pressure and/or slow gastric emptying which may result in development of GERD symptoms [27]. In addition, the association between coffee and GERD is still unclear. Shimamoto et al. found no association between coffee intake and GERD among 8103 healthy Japanese individuals [30]. Moreover, results of a meta-analysis reported no association between coffee intake and GERD [31]. In our study, about 90% of participants did not consume coffee. Therefore, it seems that any possible relationship between coffee and GERD may not be obvious in this study. In terms of tea consumption and GERD, several epidemiological surveys had regarded tea as a risk factor of GERD [27, 32], although, this relationship varied depending on the type of tea consumed, such as green tea, strong tea or peppermint tea [27, 33, 34], as well as quantity of tea intake [32]. In this study, no significant association was found between the timing of tea intake and risk of GERD, which is in line with the finding of previous epidemiological reports [35, 36]. Also, we found no association between carbonated beverages and GERD. Similarly, a systematic review demonstrated that carbonated beverages were neither related to GERD development, nor exacerbation of GERD complication [37]. It should be noted that in this study, the amount of the liquid intake was not recorded. Some studies have shown that intake of large volumes of liquid may be associated with GERD development or disease severity [8, 28]. Moreover, it is proposed that this association may vary, depending on whether or not the liquid is consumed with meals. Accordingly, a study reported that intake of coffee alone had no significant changes on LES pressure, while consumption of coffee after a lunch meal significantly reduced LES pressure [38]. However, this topic seems to be still inconclusive and more researches with an emphasis on both quantity and timing of liquid intake are needed.

In terms of socio-demographic status and GERD, being married was associated with higher odds of GERD in univariate analysis. While, higher educational level was associated with lower odds of GERD in both univariate and adjusted analysis. Similarly, previous studies reported lower educational level or being married as risk factors of GERD [39, 40]. It has been suggested that the above-mentioned factors may be indirectly related to GERD through their effects on lifestyle conditions including eating habits, psychosocial stress, and hygiene practices [41]. However, other studies have demonstrated otherwise [42, 43]. It should be noted that this association could not be simply explained due to this cross-sectional design, warranting further researches in this field.

Another concern is the lack of a clear association between smoking and GERD. While previous studies have indicated smoking as an independent predictor of GERD development [21, 26, 29, 44], others could not confirm such an association [24, 35, 40], in the present study, smoking was not associated with GERD development. This is inconsistent with findings of the study conducted by Esmaillzadeh et al., which showed that in Iranian adult men, smoking was not a significant predictor for GERD [45]. A similar population-based cohort study in Norwegian population clarified that cessation of smoking was associated with improvement of GERD from severe to no or minor complaints, only among individuals within normal range of BMI but not among overweight individuals [46]. Although, smoking has a stimulatory effect on acid exposure, only 4% of the variations in GERD symptoms are attributed to acid exposure [47, 48]. Besides, the inconsistency among studies might be due to differences in the socio-demographic status of study populations, sample sizes, criteria for diagnosis of GERD, and definition of smoking status. Of note is that in this study, similar to that of Esmaillzadeh et al. (7.8% of the study participants were smokers) [45], prevalence of smoking was low, making it difficult to draw a significant relationship between smoking with GERD.

There are some limitations of our study, worth mentioning. First, we cannot ignore the effect of recall bias on our investigation, although trained staff were responsible for data collection, to reduce any possible errors which might be influenced in the results. Second, dietary intake from other food groups such as dairy products, meats, grains, and fats, as well as total energy and nutrient intake were not investigated. A study by Wu et al. found that intake of meat, oil, salt, and calcium were related to an increased risk of GERD, while intake of carbohydrate, protein, vitamin C, starches, egg, and fruits were inversely associated with GERD development [49]. However, the present study focused on those specific dietary sources which have mostly known for their refluxogenic effect. We suggest that future studies should focus on the diet-GERD relationship in the context of dietary patterns, rather than of specific nutrients, food items, or food groups. This approach therefore, will assess the combined effect of various dietary components, and as a result, provide a more comprehensive and precise information on the diet-GERD relationship.

To our knowledge, this study provides the first evidence on predictors of GERD in the Sistan and Baluchistan province. Central obesity as determined by WC and citrus fruit intake were independent factors associated with GERD development in the study population. Due to increasing prevalence of GERD in Iran, detection of risk factors for GERD can assist health professional to promote preventive programs in this region.