Diet and Gallstones in Women of a Rural Town of Sicily


To test the possible association between dietary factors and gallstones, we performed a case-control study on the dietary habits of 71 women with gallstones in a rural area of Sicily (mean age 63.2 years) and compared these with 142 women who did not have gallstones (mean age 63.6 years) selected at random from the same general population. The dietary questionnaire, based on data collected over 7 days, was carefully compiled by two dieticians specially trained for the study. The daily calorie intake was higher in the women with gallstones (p = 0.006) and was equally distributed between fats, carbohydrates and proteins. The intakes of unsaturated fats (p = 0.03), unrefined sugars (p = 0.01) and fibre (p = 0.04) were significantly higher in the subjects with gallstones. The cholesterol and calcium intakes were not significantly different, although the average consumption in both groups was lower than the recommended quantities. However, all the nutrients were equally distributed in percentage between the two groups. When three subgroups were formed according to age, we found that the calorie intake tended to be higher in the subjects with gallstones, with the three components always proportionately distributed, but the differences were not significant. We concluded that, apart from the higher calorie intake, other factors, such as age, weight, parity, lipaemic alterations and genetic factors, may help to cause the modifications in bile secretion leading to the formation of stones.

Keywords: gallstones, epidemiology, nutrients.


Gallstones are a very topical problem as they are widely present in the general population [ 1-5] and are high on the list of the most frequent surgical intervention [6]. Thanks to the wider use of mass ultrasound screening, knowledge of the pathology and natural history of this disease has greatly increased. Studies in the literature have in fact shown a close relationship with a number of risk factors: age, sex, number of pregnancies, obesity and family history [ 1-5]. However, although numerous studies have investigated the relationship of this disease with dietary habits, the results have at times been contradictory [7-11]. This may depend both on the objective difficulty of collecting dietary data in a population (as data vary according to the seasons and over the years) and on the different types of questionnaire used by the various authors.

In the present study, performed in Ventimiglia, Sicily, about which we have already published reports [ 4,12] and where we found a positive association between gallstones and age, sex, parity, obesity and low levels of high-density lipoprotein (HDL)-cholesterol, we refer to the data of a dietary survey on patients found to be carriers of gallstones and compare them with those of non-carriers in the general population, with the aim of establishing the influence of diet on gallstone formation.


Study Population
The participants in the screening for gallstones were 1081 subjects representing 75.2% (68.2% male and 81.3% female) of the 1438 inhabitants of Ventimiglia over the age of 19 years who were asked to participate. They were previously interviewed to obtain general anamnestic data and all received a dietary form containing a printed list of the most common foods. They were carefully trained in how to complete the form and the quantity of each food consumed was noted daily for 7 days by the subjects. The less common foods, which did not appear on the printed list, were also noted, together with their quantities. To reduce the variability in collecting dietary data, the subjects were visited daily by a dietician to check the data reported on the form.

The screening protocol also included a physical examination, a blood sample to determine a number of biohumoral parameters, and an ultrasound scan of the gallbladder and biliary tract. The scan was performed in the morning after at least a 12-h fast with a portable real-time machine, the Ansaldo Au 900, using a 3.5 linear transducer. The subjects were studied by an operator unaware of the results of the questionnaire in two supine and left lateral positions, and cases of gallstones were defined in accordance with the previously described criteria [ 4]. Briefly, subjects presenting one of the following findings: ( 1) an echogenic distally shadowing structure within the gallbladder, ( 2) an echogenic movable non-shadowing structure within the gallbladder and ( 3) high-density echoes and constant shadowing in the region of the gallbladder fossa, with poor or no visualization of the gallbladder itself.

At the end of the study, 123 subjects (22 male and 101 females) were found to be positive for gallstones [ 4]. From these, we excluded all the men, owing to the limited number of cases and 30 women who were already aware of their gallstone condition at the start of the investigation (20 of whom were cholecystectomized). The remaining 71 women with gallstones represent group A of the present study.

The control group (group B) was constituted of two controls for each control case and included 142 female subjects, matched for age (+ 4 years), taken from the same general population, but without gallstones and selected as the first two progressive subjects in the general list.

The final forms were analyzed by a computerized database and the nutrient intakes were calculated using the nutrient composition database of the Italian Institute for Nutrition [13].

Statistical Analysis
A statistical analysis was performed using the Student's t-test and Fisher exact test to compare the mean values and the frequencies of the associated factors, respectively. For non-parametric data the Mann-Whitney U-test was applied. A multiple logistic regression was performed between parameters that were significant using the Student's t-test, to estimate the independence of the association between each variable and the presence of gallstones; a forward stepwise regression was used with gallstones (yes/no) as a dependent variable and other parameters as forced variables. Values of p < 0.05 were taken as significant.


Table 1 shows the total calorie intake, subdivision and percentage of the major nutrients in the group of subjects with gallstones compared with the controls. An analysis of the data using the Student's t-test showed a daily calorie intake significantly higher in the subjects with gallstones (p = 0.006). This higher total calorie intake was due to a higher intake of carbohydrates (p = 0.014), fats (p = 0.036) and proteins (p = 0.024). Significantly more unrefined sugars were consumed (p = 0.01) as well as unsaturated fats (p = 0.03). The daily fibre intake was higher in the gallstone carriers (p = 0.04) but this was equally distributed between soluble and insoluble fibre. The cholesterol intake was low (<170 mg/day) in a large number of subjects, particularly the controls, while the intake was high (>250 mg/day) in a small number of the subjects with or without gallstones, but the differences were not statistically different. The calcium intake varied, but not significantly, both in the subjects with gallstones and in the controls, but was within normal limits in all cases. In addition, the alcohol values were not significantly different (Mann-Whitney U-test). However, when considering the daily intake of each nutrient in percentage terms, it can be seen that the results were quite similar.

The correlations between body weight and daily calorie intake performed using Pearson's r-test were not significant in either of the two groups (r = 0.12 in the group with gallstones and r = 0.08 in the control group). Multiple logistic regression of parameters significant using the Student's t-test (total calories, proteins, vegetable proteins, total fats, unsaturated fats, total carbohydrates, unrefined carbohydrates and fibre) showed no significant association with gallstones. Table 2 presents the total daily calorie intakes of proteins, fats and carbohydrates in the subjects with gallstones and the controls, divided into three subgroups according to age. The calorie intake was significantly higher in the 60-69 age group (p = 0.046), owing to a proportionate increase in carbohydrate, fat and protein consumption. The differences were not significant in the other groups, probably owing to the small number of cases.

An analysis of smoking habits in the two groups showed no significant difference (three out of 71 versus five out of 142, respectively; Fisher exact test p = 0.53).


The population of Ventimiglia, Sicily, is predominantly rural, with a high prevalence of women with gallstones correlated with body weight, age and sex [ 4,12]. During the preliminary phase of the study and before the abdominal ultrascan, the whole population was asked to complete a questionnaire based on the collection of dietary data over 7 days, which did not differ greatly from any other available method, such as the 24-h dietary recall or the dietary diary used in other studies [8,14].

Our results confirm a number of reports in the literature supporting a higher calorie intake in women with gallstones [7,15,16], which is said to favour stone development by increasing the lithogenic index [17]; the calories were equally divided between fats, proteins and carbohydrates, but it should be emphasized that the quotas in both groups were similar in percentage and no higher than the recommended daily intake quotas for the Italian population [18]. Among the main food components, there was a higher intake of unsaturated fats and unrefined sugars. These results are in agreement with those in the literature in the case of fats [7], but not carbohydrates, as there are more frequent reports of a greater association with refined sugar consumption, probably due more to the deficiency of fibre in the refined sugars than to the sugars themselves [19,20]. In the case of fibre, however, in contrast with reports in the literature [10,21], the intake was greater in those with gallstones. This result might not, however, be of great value as the fibre quotas in our subjects with gallstones and (of greater importance) those of the controls were both below the recommended daily levels for the Italian population [18]. The same was true for cholesterol and particularly calcium, as the consumption of milk and its derivatives was low in this population [22]. However, consumption was similar in percentage in both groups, suggesting no responsibility in gallstone formation. When the subjects were divided into three subgroups according to age, there were no notable differences, probably because the population with gallstones was homogeneously distributed in the over 50 age group, and only nine of those with gallstones were below this limit. In addition, there was probably also a levelling out of certain dietary habits characterizing the younger subjects. Naturally, in the subgroups with gallstones the trend towards a higher calorie intake compared with the controls persisted, but the differences were not statistically significant owing to the low number of subjects in each age subgroup. In the case of alcohol, no conclusions as to interference in bile secretion were possible because of the low quantity consumed by both groups. The same can be said in the case of smoking habits. In fact, very few subjects were smokers in either of the two groups. This result is quite realistic as the habit is limited in women and particularly in the more elderly subjects in rural centres in Sicily.

Our results were in general agreement with some data in the literature [7,15,17], but also either partially or totally in disagreement with others [8,10,11,16]. In the light of the present study, we believe that investigations of the diet-gallstones association are difficult owing to the complexity of evaluating one dietary component isolated from all the others, which also varies according to the season, while gallstones generally require months or years to develop. It is also difficult to compare dietary components in populations with different characteristics and associate them with gallstones, while it is easier to determine associations with more objective parameters such as age, sex or obesity. The difficulties of a dietary study in this sector depend not so much on the collection of data as on the lack of correspondence between the actual quantity and quality of foods compared with the actual presence or absence of gallstones. In conclusion, the dietary data we recorded at Ventimiglia, Sicily, in female subjects with gallstones, compared with controls, show an increased daily calorie intake equally divided between the three components of carbohydrates, fats and proteins, with no difference in percentage between the two groups. We conclude that, apart from dietary habits, other factors of a different nature, such as age, sex, weight, lipaemic alterations and parity, are more important parameters of the risk of developing gallstones in the population of Ventimiglia.


This study was in part supported by grant 60% from MURST, 1993.

TABLE 1. Mean +/- SD daily consumption of major nutrients in women gallstones and matched controls
Women with
gallstones Controls
Nutrient n = 71 n = 142 p

Total calories
(kcal) 932.3 +/- 624.3 1718.1 +/- 535.6 0.006

Protein (g) 63.6 +/- 20.4 57.7 +/- 16.6 0.024
(% of total
calories) (13.6) (14)

proteins (g) 42.4 +/- 18.6 38.4 +/- 13.6 0.076
(% of total
calories) (9.1) (9.4)

proteins (g) 21.2 +/- 6.3 19.3 +/- 5.8 0.030
(% of total
calories) (4.5) (4.7)

Total fats (g) 67.2 +/- 23.8 58.8 +/- 28.6 0.036
(% of total
calories) (32.4) (32.3)

Saturated fats (g) 20.1 +/- 8.6 17.9 +/- 9.3 0.115
(% of total
calories) (9.7) (9.8)

fats (g) 47.1 +/- 17.2 40.9 +/- 21.1 0.031
(% of total
calories) (22.8) (22.4)

carbohydrates (g) 51.1 +/- 107.8 219.4 +/- 78.4 0.014
(% of total
calories) (53.9) (53.6)

carbohydrates (g) 46.8 +/- 19.5 44.6 +/- 19.5 0.434
(% of total
calories) (10.0) (10.9)

carbohydrates (g) 04.3 +/- 100.4 174.8 +/- 69.8 0.019
(% of total
calories) (43.8) (42.7)

Fibre (g) 16.7 +/- 6.8 14.9 +/- 5.2 0.040
(g 1000
kcal-1) (8.64) (8.67)

Cholesterol (mg) 71.2 +/- 18.7 157.7 +/- 115.7 0.894
(mg 1000
kcal-1) (88) (91)

Calcium (mg) 85.4 +/- 208.1 509.6 +/- 212.2 0.426
(mg 1000
kcal-1) (251) (296)

Alcohol (g) 9.6 +/- 9.3 11.4 +/- 12.5 0.284(a)

(a) Mann-Whitney U-test.
TABLE 2. Mean +/- SD daily consumption of major nutrients in the subjects with gallstones and controls, divided into three age groups
Subjects with gallstones

Nutrient (n = 25) (n = 25)

Calories 1919.1 +/- 418.7 1936.8 +/- 855.1(*)
Proteins 64.2 +/- 14.3 66.2 +/- 26.2
Fats 75.9 +/- 20.2 63.7 +/- 24.6
Carbohydrates 244.6 +/- 70.1 274.6 +/- 157.5

gallstones Controls

>/= 70 years Nutrient (n = 21) (n = 50)

Calories 1705.7 +/- 467.1 1799.7 +/- 548.2
Proteins 59.4 +/- 60.1 60.1 +/- 15.3
Fats 60.6 +/- 72.4 72.4 +/- 31.1
Carbohydrates 230.6 +/- 229.8 229.8 +/- 80.3


60-69 years >/= 70 years
Nutrient (n = 50) (n = 42)

Calories 1610.9 +/- 544.6(*) 1494.1 +/- 475.1
Proteins 59.9 +/- 15.3 52.9 +/- 24.1
Fats 54.4 +/- 31.1 49.0 +/- 24.1
Carbohydrates 220.4 +/- 80.3 210.3 +/- 68.1

(*) p < 0.046.
[1] GRAEPCO. Prevalence of gallstone disease in an Italian adult female population. Am J Epidemiol 1984; 119: 786-805.

[2] Barbara L, Sama C, Morselli Labata AM, et al. A population study on the prevalence of gallstone disease. The Sirmione study. Hepatology 1987; 7: 913-19.

[3] Glambek I, Kvaale G, Arnesso B, et al. Prevalence of gallstones in an Norwegian population. Scand J Gastroenterol 1987; 22: 1089-94.

[4] Montalto G, Muratori R, Soresi M, et al. Prevalence of gallstone disease in a small town of Sicily. Eur J Int Med 1995; 6: 221-6.

[5] Jorgensen T. Prevalence of gallstones in a Danish population. Am J Epidemiol 1987; 126: 912-21.

[61 Soper NJ, Barteau JA, Clayman RV, et al. Comparison of early postoperative results for laparoscopic versus standard open cholecystectomy. Surg Gynecol Obstet 1992; 174: 114-18.

[7] Scragg RKR, McMichael AJ, Baghurst PA. Diet alcohol and relative weight in gallstone disease: a case control study. BMJ 1984; 288: 1113-19.

[8] Pixley F, Mann J. Dietary factors in the aetiology of gallstones: a case control study. Gut 1988; 29: 1511-15.

[9] Wheeler M, Hills LL, Laby B. Cholelithiasis: a clinical and dietary survey. Gut 1970; 1]: 430-7.

[10] Smith DA, Gee MI. A dietary survey to determine the relationship between diet and cholelithiasis. Am Clin Nutr 1979; 32: 1519-26.

[11] Jorgensen T, Morch Jorgensen L. Gallstones and diet in a Danish population. Scand J Gastroenterol 1989; 24: 821-6.

[12] Montalto G, Soresi M, Carroccio A, et al. Prevalence of biliary lithiasis in elderly people of a small town of Sicily. Age Ageing 1992; 21: 338-42.

[13] Istituto Nazionale della Nutrizione (Italian Institute for Nutrition). Food composition tables. Rome Istituto Nazionale della Nutrizione, 1986.

[14] Morgan RW, Jain M, Miller AB, et al. A, comparison of dietary methods in epidemiologic studies. Am J Epidemiol 1978; 107: 488-98.

[15] Sarles H, Charbert C, Pommeau Y, et al. Diet and cholesterol gallstones. A study of 101 patients with cholelithiasis compared to 101 matched controls. Am J Dig Dis 1969, 14: 531-7.

[16] Williams LN, Johnston JL. Prevalence of gallstones and risk factors in Caucasian women in a rural Canadian Community. Can Med Assoc J 1980; 120: 664-8.

[17] Lefkof IR, Marks JW, Schoenfield LJ. Diet in the formation and treatment of cholesterol gallstones. In: Cohen S, Soloway RD, eds. Gallstones. Contemporary issues in gastroenterology. New York: Churchill Livingstone, 1985; 4; 113-27.

[18] Livelli di assunzione giornaliera raccomandati di energia e nutrient) per la popolazione italiana (LARN), a cure della Societa Italiana di Nutrizione Umana. Revisione, 1986-87, INN, Roma.

[19] Thornton JR, Emmett PM, Heaton KW. Diet and gallstones: effects of refined and unrefined carboydrate diets on bile cholesterol saturation and bile acid metabolism. Gut 1983, 24: 2-6.

[20] Werner D, Emmet PM, Heaton KW. Effects of dietary sucrose on factors influencing cholesterol gallstone formation. Gut 1984, 25: 269-74.

[21] Marcus SN, Heaton KW. Effect of a new, concentrated wheat fibre preparation on intestinal transit, deoxycholic acid metabolism and the composition of bile. Gut 1986; 27: 893-900.

[22] Montalto G, Carroccio A, Cavera G, et al. Malassorbimento di lattosio in una popolazione rurale siciliana. Atti XCVI Congresso Nazionale SIMI, Roma 1995, p 127.


By G. MONTALTO MD, A. CARROCCIO MD, M. SORESI MD, A. CARTABELLOTTA MD, D. LORELLO MD, G. CAVERA MD, M. R. AVERNA MD, C. M. BARBAGALLO MD, G. ANASTASI MD, F. BASCONE MD AND A. NOTARBARTOLO MD, Cattedra di Medicina Interna, Universita di Palermo, via del Vespro 143, 1-90127 Palermo, Italy

Share this with your friends