[BHJ LOGO] Original/Research ArticlesContentsHomeArchivesSearchBooksFeedback

EPIDEMIOLOGY OF GALLSTONE DISEASE - TOPLINE FINDINGS

V Jayanthi*, R Prasanthi*, G Sivakaumar*, R Surendran*, Usha Srinivas*, Sunil Mathew*, S Rajakumar*, C Palanivelu**, A Ramesh***, K Prabhakar****, G Subramanian****, B Ramathilakam*****, S Vijaya*
A collaborative study with centres at *Chennai; **Coimbatore; ***Madurai***, Salem****; *****Thanjavur; Districts in Tamil Nadu State.

Pigment gallstones are the dominant type of gallstones in south India. This study for the first time in a case controlled study has addressed the risk factors predisposing a south Indian belonging to Tamil Nadu state to pigment gallstone. Gallstones were equally common in either sex. Economic status or education did not influence a south Indian resident of Tamil Nadu state to pigment stones. Obesity and diabetes in women, a sedentary, retired life style and abstinence from smoking in men emerged as risk factors. Both sexes had equal predilection for gallstones although BMI was a mild factor for women only.

INTRODUCTION

Gallstone disease (GS) is a common gastrointestinal problem in day to day practice. Recent studies from south India have highlighted pigment and mixed variety of GS to be common
[1-4] in contrast to cholesterol stones reported from north, east and western parts of India.
[5-8]The factors predisposing a north Indian to cholesterol stones is reported to be similar to that in the west. [5,6] These factors cannot be however extrapolated to the south Indian population whose race, culture, ethnicity and diet patterns are different from north Indian patients.

The aim of the study was to quantify the risk factors for age, body mass index (BMI), parity, hereditary, life style factors and other associated systemic disorders in south Indian patients with GS disease belonging to Tamil Nadu state, India.

METHODOLOGY

The study period was between January, 1995 and June, 1998. Cases were defined as adults (15 years or above) with GS as confirmed by an ultrasound examination. [9,10] All selected cases were either residents of city of Chennai or from one of the other collaborative centres within Tamil Nadu state (Coimbatore, Madurai, Salem, Thanjavur) for atleast five years. The cases were inducted by well trained gastroenterologists who were themselves standardized between each other. A healthy control (without any current illness) was selected for each case.

Children below fifteen years and individuals with clinically overt haemolysis were excluded from the study. The demographic particulars on age, sex, parity, family size, religion, caste, type of house, household amenities, social habits—smoking, alcohol, use of betel nut/tobacco/snuff, family and past history of GS disease, diabetes and chronic liver disease were collected using a standardized questionnaire. The same questionnaire was used for both case and controls.

The data was collected by well trained medical officers and social workers who were standardized amongst each other on the data.

A written consent was obtained from each case and control to participate in the study.

Statistical Analysis

Univariate and bivariate analysis designed for case-control data was used. Chi square tests for comparison of proportions, trend Chi square for comparison of trends in proportion and analysis of variance for comparison of means was computed using the standardized statistical packages. Odds ratio and its 95% confidence interval (CI) were also computed separately for some of the factors studied.

RESULTS

A total of 320 patients with GS (cases) diagnosed by ultrasound and equal number of controls (healthy subjects without GS) matched for age and sex formed the study group. There were 156 male cases (48.8%) and equal number of male controls. The rest of the cases and an equal number of controls were women.

The age range for the cases was between 15 and 85 years and the mean age was 51.1 years (+ 14.1 years). The age range for the controls was 15 to 83 years and the mean was 50.3 (+ 14.8 years). Majority of the patients (65%) were in the age range 31 to 60 years. The mean age by selection criteria was almost similar although the difference of 0.8 year in the two groups was statistically significant because of the large numbers between the two groups (F test, p < 10-5). Thus most of the cases were middle aged persons.

Using a bivariate analysis for studying the sex difference in the mean age of cases, women were slightly younger with a mean age at 48.8 (+ 14.1 years (F test p < 10-5) (Fig. 1). However a similar age variation was not observed among the controls (51.9 years compared to 49.8 years; F test; p = 0.7).

A slightly lower proportion of cases were married and living with their spouses as against the controls (Table 1). Thus 81.9% of the cases and 76.3% of the controls were married. The difference in the proportion was statistically significant (p < 10-5). However, as a risk factor, marital factor was not important as the odds ratio (OR) for this factor was only 1.07 and not significant.

Parity of the 153 married women showed that a third of them (34.1%) had none or one or two children (Table 1). The proportion was similar among the 149 controls at 33.0%. The OR was 1.03 (95% CI for OR 0.64-1.66). The mean parity for the two groups were similar for all married women at 3.41 for the former and 3.30 for the latter (F test; p=0.4). Thus a high parity was not a significant factor for the presence of GS.

The mean family size for the cases and the controls was also similar at 4.6 and 4.3 respectively (F test; p=0.3) and their distribution as seen in Table 1 was also not significant. Thus family size was not an important factor predisposing individuals to GS.

Most of the study cases and controls were Hindus (87.5% and 90.9%). Over half of the cases and controls (56.6% and 57.8%) belonged to the backward class (BC) as classified by the Tamil Nadu Gazette.11 The proportion in different castes were similar among the cases and controls (Chi sq. p=0.6) (Table 1) indicating the lack of influence of ethnic or racial factor in the distribution of gallstones in this region.

Fig 1
Fig. 1 : Mean age by gender.

 

TABLE 1
Characteristics of cases with gallstones and healthy controls
Characteristics Cases Controls Significance Levels
  No. % No. %  
1. Marital status        


P=0.05
currently married 262 81.9 244 76.3
widow/separated 39 12.2 46 14.3
2. Mean Parity (Females) + SD 3.14 +1.2 3.3 + 1.5 P=0.37
3. Family size        

 

P=0.22
Small (1-3) 86 26.9 110 34.6
Medium (4-7) 207 64.7 189 59.4
Large (8 and more) 27 8.4 19 6.0
4. Religion        

 

P=0.05
Hindu 280 87.5 291 90.9
Muslim 26 8.1 21 6.6
Christians 14 4.4 7 2.2
5. Caste        

 


P=0.61
Forward 62 19.4 58 18.1
Backward 181 56.6 185 57.8
Most backward 28 8.8 26 8.1
Sheduled castes and tribes 40 12.5 38 11.9
6. Educational Status        

 


P=0.01
Illiterate 74 23.1 79 24.7
Middle school 121 37.8 128 40.1
High school 64 20.0 71 22.2
Gradutes 60 18.8 41 12.8
7. Occupation        

 


P=0.11
Unepmloyed / housewife 193 6.3 172 53.8
Low level 53 16.6 53 16.6
High level 17 5.4 8 2.5
Business 10 3.1 11 3.4
8. Economic status          
Type of house        

 

P=0.14
Kutcha 41 12.8 49 15.3
Semi 82 25.6 90 28.1
Pucca 194 60.6 179 55.9
Amenities        

 


P=0.08
Very good 156 48.8 113 35.3
Good 65 20.3 104 32.5
Medium 25 7.8 16 5.0
Poor 71 22.2 82 25.6

* Includes skilled, semi-skilled and white collar profession with monthly income less than Rs.2000/- per month and
** with monthly income equal or more than Rs.2000/- per month

Fig 2
Fig. 2 : Education status cases versus controls.

Literacy levels were fairly high in both cases and controls with 76.6% of the cases and 75.1% of the controls being literate (Table 1 and Fig. 2). Over a third of the cases (37.8%) and controls (40.1%) had completed high school and 18.8% and 12.8% respectively were graduates. Educational qualification did not differ significantly between the two groups indicating that neither the cases nor the controls clustered in any particular educational group (Chi sq. p=0.1).

193 (60.3%) of the cases were jobless or pensioners or having a low profile skilled or semiskilled profession compared to 172 (53.8%) in controls (Table 1). The proportions were similar in both the groups (p=0.09).

86.6% of women with GS were housewives as against 79.9% of the controls. Only a few were daily wage labourers (6.7% of the cases and 9.1% of controls). Thus the OR (risk) for having GS among women who were housewives was 1.6 (95% CI 0.87 to 3.06). Occupational profile did not differ significantly (Chi sq. p=0.4) although they did so in men.

A slightly higher but significant proportion of men with GS had retired or were without a job —32.7% versus 26.3% in controls (Chi sq. P < 0.05). Thus the OR for developing GS among retired/jobless men was slightly higher at 1.3 (95% CI 0.77 to 2.2).

Thus in our series, patients of either sex, who were jobless or had retired had a higher risk for GS formation.

Economic status for the two groups was assessed by the type of the house, which they occupied and the amenities present in the house, rather than the reported income since the reliability of the same is considered poor. As far as the amenities in the house was concerned, the cases had slightly better amenities compared to the controls (Table 1). Thus, nearly half of the cases (48.8%) had a television and a vehicle in their house compared to just a third of the controls (35.3%) (p < 0.001). However, this criteria was not sensitive enough to detect differences in the type of house due to definition problems. Using the standard definition of Government of India, [12] of kutcha house for a hut, a pucca for a permanent structure and semi-pucca for a temporary structure, with either the wall, flooring or ceiling being temporary, 60.6% of the study cases and 55.9% of the controls lived in a pucca house (Table 1). Induction of a large proportion of the study group from those living in a pucca house could be due to the urban nature of the study, wherein pucca houses are dominant. The size and facility in each house, however, could vary enormously. But the two groups again did not differ significantly (p=0.14 and 0.08 respectively) and we could not capture the socio-economic variability among the cases and controls using this criterion.

A few cases (3.1%) had associated cirrhosis and 7.2% had history of diabetes. The latter was significantly lower in the controls (1.9%) (p=0.002). Thus these patients who had associated diabetes had an OR of 4.5 indicating a 4.5 times higher risk for GS than those without. The 95% CI for this OR was also significant with OR ranging from 4.1 to 4.9. Diabetes emerged as a significant factor for GS in this series.

A positive family history of GS (with another family member having suffered from GS) was present in equal proportions among the case and controls ruling out the familial predilection possibility (1.6% versus 1.3%). Similarly, chronic illness such as acid peptic disease, bronchitis, hypertension was present in equal proportions in both the groups (28.8% versus 27.2% respectively).

Life Style Variables

The life style variables considered in this study were social habits such as smoking, alcohol intake, chewing of tobacco and betel nuts and use of snuff. Analysis of the latter three variables for GS has been attempted for the first time in India.

Smoking and alcoholism were not reported among any female patients in this study and hence women were excluded from the analysis of these two variables.

Among men, while 44.2% of the men were either current smokers or had ever smoked in the control group, the same was significantly lower in the cases at 32% (p < 0.05) (Table 2). Sex standardized OR for smoking was 0.69 indicating a mild protective effect of smoking on GS disease.

TABLE 2
Life style variables
Characteristics Cases Controls Odds Ratio 95% CI For (OR)
  No. % No. %   (OR)
1. Smoking            
never 156 67.9 87 55.8 0.65 0.32-0.45
current and ever 50 32.1 69 44.2    
2. Alcohol in men            
never 104 66.7 105 105    
current and ever 50 32.1 50 32.1    
3. Betelnut 53 16.6 31 6.7    
4. Tobacco used 20 6.3 22 6.9    
5. Snuff used 6 1.9 4 1.3 0.1  

Alcoholism in men however was not associated with presence with GS in that the proportion of alcoholics was similar among the two groups. Thus 25.6% of the male cases and 17.9% of the controls were current alcoholics and 6.4% and 14.1% were so, in the past (Table 2). The proportion of teetotalers or the current alcoholics did not differ significantly between the two groups, indicating an absence of any association (prospective or otherwise) between alcoholism and presence of GS disease.

A very small proportion of the cases and controls used tobacco, betelnut or snuff. Thus, a significant proportion i.e. 16.6% of the cases as against 6.7% of the controls reported regular consumption of betelnut (p < 0.5) (Table 2). Tobacco was used by 6.3% of the cases and 6.9% of the controls and snuff by 1.9% and 1.3% respectively.

Clinical Features

Using the standard definition of BMI, the patients were classified as lean, normal build, obese when the BMI values were less than 18, 18-25 and above 26 respectively. Accordingly, only 12.6% of cases were lean, and nearly for fifths (80.8%) had normal BMI and 6.6% were obese. This was in contrast to 15.5%, 83.0% and 1.6% among the controls. Although a large number of the controls were normal or lean, these were not significantly different from the cases (p=0.7) the mean BMI for the two groups was similar at 22.9 + 4.8 for the cases and 21.6 + 3.9 for the controls. In general, although obesity was slightly more prevalent among the cases than among the controls, overall, it wasot a predominant risk factor in this study, when both sexes were considered together.

However obesity was a risk factor for women: in that the mean BMI in women with GS was slightly but significantly higher than their corresponding controls. Thus the mean BMI for the cases was 23.6 + 5.0 compared to 22.0 + 4.0 in controls (F test; p=0.002). On the contrary among men, BMI was similar to that of the controls at 22.2 + 4.5 in the former compared to 21.2 + 3.8 in the later (F test; p=0.3). Two-thirds of the patients were symptomatic (212 patients). 41% underwent cholecystectomy. Predominant symptoms were abdominal pain, dyspepsia and cholangitis. The symptoms decreased with advancing age i.e. among those less than 30 years, 77.4% were symptomatic, compared to 68.8% in 31-60 years and 57% in those above 60 years (trend Chi sq. P=0.02) (Fig. 3).

Based on visual inspection of the stones, 80% of the cases had pigment GS, 6.9% each had cholesterol or mixed type of GS and in three cases, the stones could not be classified and no data on the type of stones were available in five.

Fig 3
Fig. 3 : Symptomatic patients in different age groups

DISCUSSION

In India till date, there are several reports describing north-south differences in the type of gallstones. [1-3] , [5,6], [13] The north being predominantly of the cholesterol type and south of the pigment variety. While the characteristics of the individuals with gallstones from north have been identified, similar reports are not available for the south. The present case-control study on GS has for the first time from south India emphasized on a lack of association between various socio-economic factors and life style variables for GS especially of the pigment variety. The study is also unique in using for the first time using matched controls to compare the characteristics.

Eighty per cent of the patients with GS in the present series and another 7% had mixed/intermediate type of GS. This confirms our earlier hypothesis that south Indian GS are predominantly of the pigment/mixed type. These being the common type of GS, information on the study group hitherto referred to would obviously pertain to pigment stones only.

An equi-distribution of GS in both sexes was one of the main features in our series. This is in line with the observation for pigment GS reported elsewhere from India [3,5] and other parts of the world. [14] A similar finding was made in an earlier hospital study from the same centre. [1] The average age for pigment GS reported in a Dutch series was 50 years. [15] In our series the mean age was similar at 51 years and two-thirds of the patients were young adults (age 31-60 years) Sarin et al [5] had made similar observations even for cholesterol GS in north India. However, an increase in prevalence with advance in age observed by several other workers was not observed in this study. [16-21]

Parity is reported to be an important variable for cholesterol GS. [22-25] Sarin et al [5] report that 94% of patients with GS are multiparous with a 1.5 times greater risk compared to those women with pigment GS. Diehl et al [23] could not demonstrate such an association. In our series with predominant pigment GS, parity did not emerge as an important risk factor since the parity distribution for women with and without stones (controls) remained the same.

Socio economic variables studied were indirect ones and based on the amenities available at home - TV, radio, cycle etc., the structure of the house itself vis a vis the number of members in the family. Although data on per capita income was available for all our patients and controls, its reliability was considered poor largely due to poor reporting of the true income by the individuals. Overall it was observed that a moderately rich patient had a higher risk for pigment GS compared to controls; nearly 50% of patients had good amenities in their house as compared to their controls. However family size was not an associated factor in this series. Thus socio-economic factor did not prove to be a risk factor in our series. In Japan, [26-27] individuals residing in rural areas were at a high risk for black pigment GS compared to the urban population. Sarin et al, [5] from north India did not observe a similar association for either cholesterol or pigment GS. Khuroo et al [6] did not find any relationship between prevalence of cholesterol GS and socio-economic status of patients.

The predominance of the backward ethic group among the cases and controls confirmed that GS were equally prevalent in all ethnic groups. Trotman et al [14] found no significant difference in stone types between the blacks and the whites in Philadelphia. Kaufman et al [28] found blacks have a significantly higher proportion of pigment stones than the whites. Diehl et al [23] found a slightly higher proportion of black pigment stones in the blacks, although the difference was not statistically significant.

Cholesterol GS were slightly higher but not significantly high amongst Mexican-American patients [29] residing in Novo Scotia.

The educational status for the case and controls were fairly high at almost 75% in our series. Using univariate analysis, Diehl et al [22] found no relationship between educational status and the type of gallstones.

The type of occupation between case and controls did not differ in our study for women but they did differ in men. A significantly higher but significant proportion of men among the cases were retired or without a job when compared to the controls. The mean age for the jobless men were consequently higher than those with a job. Sarin et al [5] found cholesterol GS to be more prevalent amongst sedentary workers compared to those with moderate activity. Physical activity did not influence the occurrence of pigment GS in their series.

Family history of GS disease was found in equal proportion of cases and controls. Likewise other systemic disorders such as hypertension, diabetes and gastrointestinal disorders such as acid peptic disease, liver disease were present in equal proportions of men and women. In a study from Israel [30] the relative risk for GS formation was 2 fold high amongst first degree relatives for patients with GS. Cholesterol saturation was higher in relatives of GS patient. [31]

Smoking appeared to be protective for GS formation. 44.2% with GS disease were smokers compared to 32.0% in controls. In a study by Diehl et al [23] and Murray et al [32] smoking enhanced the risk for GS formation. Nicotine probably alters the lithogeneity of the bile.

Alcohol did not influence the occurrence of GS disease. Thus almost a quarter of male cases and 18% of controls were alcoholics. The proportion of teetotalers or the current alcoholics did not differ in the 2 groups. This is similar to the reports by Schweisenger et al [33] where alcohol was not associated with greater risk to GS disease. Several other workers however found that use of alcohol in small to moderate amounts may be protective against GS formation. The exact mechanism is not clear but probably it lowers the biliary cholesterol saturation index. [34-37]

Other life style variables such as use of tobacco, chewing of betelnut and use of snuff has not been looked into by other workers, even though these are common social habits in the Indian community. How nicotine could influence risk of GS disease is not clear. In our study a very small proportion of the cases and controls used tobacco, betelnut and snuff. A significantly greater proportion of cases used betelnut compared to the controls. Tobacco chewing and snuff usage was not different between the case and controls.

80% of our patients had a normal BMI and was similar to the controls; their means were also similar. For women, however obesity was a factor, but not heavily so, in that fat women had a slightly higher risk of suffering from GS compared to the lean and normal ones. This is similar to the female patients with cholesterol GS5, [6], [22,23], [38,39] especially for women. This is in tune with the belief that females who are fat have a higher probability for GS. Trotman observed that the BMI in patients with cholesterol GS and pigment GS were similar. [16]

Only 2 patients had hypothyroidism, 3.1% had an associated cirrhosis and all were asymptomatic for GS disease. Diehl et al [23] and several other workers [14], [40-42] have found a strong association for pigment GS in cirrhotics. The risk appears to be independent of the aetiology of cirrhosis. Associated diabetes appeared to increase the risk for GS in our series by about 4.5 times. Trotman et al [16] reported diabetes in 8% of patients with GS and hypothyroidism in 3%. These figures were similar in patients with pigment and cholesterol GS. Whether diabetes is a risk factor for GS formation is uncertain. Most authorities do not consider it as a risk factor even for cholesterol GS. Several studies have failed to show an association between diabetes and GS patients in age-matched controls. [43] The GREPCO and Sirmione studies as also the Dutch study failed to show any association. [44]

In conclusion, a south Indian patient with pigment GS would be a middle aged person (30-50 years)-if a female, a fat housewife and if a male, a diabetic, slightly obese, retired, and a nonsmoker. Diabetes enhances the risk of GS formation while smoking appears to be protective. Genetic factors predisposing to GS formation may not be simply related to an autosomal dominant or recessive traits but is likely to be more complex.

ACKNOWLEDGEMENT

The authors thank the Director of Medical Education, Tamil Nadu for permitting us to undertake this study and the RD Birla Smarak Kosh for providing the grant to undertake this study.

REFERENCES

  1. Jayanthi V. Pattern of gallstone disease in Chennai city, South India - a hospital based survey. Journ Assoc Physicians India 1996; 44 : 461-4.
  2. Tandon RK, Thakur US, Basak AK, Lal K, Jayanthi V, Nijahawan S. Pigment gallstone predominate in south India (abs.) Indian J Gastroenterol 1994; 13 : A18.
  3. Ananthakrishnan N. Current concepts in the pathogenesis of gallstones. Indian Journ Surg 1998; 60 : 85-9.
  4. Rathnaswami A, Vijayan J, Omprakash R, Balasubramanian S, Rangabashyam N. Gallstone diseases - our experience. South Indian Journal Clinics 1989; 3 : 89-93.
  5. Sarin SK, Kapur BML, Tandon RK. Cholesterol and pigment gallstones in northern India. A prospective analysis. Dig Dis Sci 1986; 31 : 1041-5.
  6. Khuroo MS, Mahajan R, Zargar SA, Javid G, Sapru S. Prevalence of biliary tract disease in India: a sonographic study in adult population in Kashmir. Gut 1989; 30 : 201-5.
  7. Tandon RK. Pigment gallstone. Indian J Gastroenterol 1988; 7 : 5-6.
  8. Kotwal MR, Rinchen CZ. Gallstone disease in the Himalayas (Sikkim and North Bengal): causation and stone analysis. Indian J Gastroenterol 1998; 17 : 87-9.
  9. Zeman RK. Gallbladder imaging. The state of art. Gastroenterol Clinics North America 1991; 20 : 127-56.
  10. Coopberg P, B urhenne HJ. Real time ultrasonography. Diagnostic technique of choice in calculous gallbladder disease. New Engl J Med 1980; 302 : 1277-79.
  11. Appendix 1994-1995 sessions. Gazette of Tamil Nadu.
  12. National environment hygiene committee. Report of the environmental hygienic committee. New Delhi : Ministry of health, Govt. of India. 1949.
  13. Agarwal DK, Choudhari G, Rao KV, Saxena R, Kapoor VK. Gallstone composition in North India population changing pattern (abs). Indian J Gastroenterol 1991; 10 : A14.
  14. Trotman BW, Soloway RD. Pigment vs cholesterol cholelithiasis : clinical and epidemiologial aspects. Dig Dis 1975; 20 : 735-9.
  15. Van Erpecum KJ, van Bege Henegouwen GP, Stoelwindern B, et al. Cholesterol and pigment gallstone disease : comparison of the reliability of three bile acid tests for differentiation between the two stone types. Scand J Gastroenterol 1988; 23 : 948-54.
  16. Trotman BW, Ostrow JD, Soloway RD. Pigment vs cholesterol cholelithiasis : comparison of stone and bile composition. Am J Dig Dis 1974; 19 : 585-90.
  17. The Rome Group for epidemiology and prevention of cholelithiasis (GREPCO). The epidemiology of gallstone disease in Rome, Italy. Part 1. Prevalence data in men. Hepatology 1988; 9 : 904-6.
  18. The Rome Group for epidemiology and prevention of cholelithiasis (GREPCO). The epidemiology of gallstone disease in Rome, Italy. Part 1. Prevalence data in an Italian adult female population. Am J Epidemiol 1984; 119 : 796-805.
  19. Barbara L, Sama C, Labate AMM, et al. Population study on the prevalence of gallstone disease : the Sirmione study. Hepatology 1987; 7 : 913-7.
  20. Barabara L, Festi D, Frabboni R, et al. Incidence and risk factors for gallstone disease : the Sirmione study. Hepatology 1988; 8 : 1256.
  21. Festi D, Barbara L, Frabboni R, et al. Prevalence and incidence of gallstone disease : the Sirimione study. In capocaccia L, Ricci G, Angelico F, et al; eds. Recent advances in the epidemiology and prevalence of gallstone disease; Dec. 4-5, 1989, Rome Dordrecnt, Netherlands. Kluwer Academic Publishers 1991; 7-11.
  22. Sama C, Labate AMM, Taromi F, et al. Epidemiology and natural history of gallstone disease. Semin Liv Dis 1990; 10 : 149-58.
  23. Diehl AK, Schwesinger WH, Holleman DR Jr, Chapman JB, Kurtin WE. Clinical correlates of gallstone composition: Distinguishing pigment from cholesterol stones. Am J Gastroenterol 1995; 90 : 967-72.
  24. Strom BL, West SL. The epidemiology of gallstone disease. In: Cohen S, Soloway RD, eds. Contemporary issues in Gastroenterology, Gallstones. New York: Churchill Livingstone, 1985; 4 : 1-26.
  25. Everson GJ. Pregnancy and gallstones. Hepatology 1993; 17 : 156-61.
  26. Hikasa Y, Nagase M, Soloway RD et al. Gallstones in western Japan. Epidemiological factors affecting the type and location of gallstones. Arch Jpn Chir 1981; 50 : 277-88.
  27. Nagase M, Tanimura H, Setoyama M, et al. Present feature of gallstones in Japan. A collective review of 2144 cases. Am J Surg 1978; 135 : 788-90.
  28. Kaufman HS, Magnusoi TH, Pitt HA, et al. The distribution of calcium salt precipitates in the core, periphery and shell of cholesterol, black pigment and brown pigment gallstones. Hepatology 1994; 19 : 1124-32.
  29. Williams CN, Johnston JL, Weldon KLM. Prevalence of gallstone and gallbladder disease in Canadian Mirunac women. Can Med Assoc J 1977; 117 : 758-60.
  30. Gilat T, Feldman C, Halpern Z, et al. An increased familial frequency of gallstone. Gastroenterology 1983; 84 : 242-6.
  31. Danzinger RG, Gordon H, Schoenfield LJ, et al. Lithogenic bile in siblings of young women with cholelithiasis. Mayo Clin Proc 1972; 47 : 762-6.
  32. Murray FE, Logan RFA, Hannaford PC, Kay CR. Cigarette smoking and parity as risk factors for the development of symptomatic gallbladder disease in women : results of the Royal College of General practitioners oral contraceptives study. Gut 1994; 35 : 107-11.
  33. Schwesinger WH, Kurtin WE, Levine BA, et al. Cirrhosis and alcohol as pathogenetic factors in pigment gallstone formation. Ann Surg 1985; 210 : 319-22.
  34. Scragg RKR, McMichael AJ, Baghurst PA. Diet, alcohol and relative weight in gallstone disease : a case control study. Br J Med 1984; 288 : 1113-9.
  35. MacClure KM, Hayes KC, Colditz GA, et al. Weight, diet and the risk of symptomatic gallstones in middle aged women. New Engl J Med 1989; 321 : 563-9.
  36. Kong S, Ikea Yanai F, et al. Dietary habits and gallstones : a study of male self defense officials in Japan. In : Capoccia L, Ricci G, Angelico F, et al. Eds. Recent advances in the epidemiology and prevalence of gallstone disease. Proceedings of the second international workshop on epidemiology and prevalence of gallstone disease. Dec. 4-5, 1989; Rome. Dordrecht, Netherlands : Kluwer Academic Publishers 1991; 129-137.
  37. Diehl AK. Epidemiology and natural history of gallstone disease. Gastroenterol Clin North Amer 1991; 20 : 1-19.
  38. Bennion LJ, Grundy SM. Risk factors for the development of cholelithiasis in man. Part 2. New Engl J Med 1978; 299 : 1221-7.
  39. Bennion LJ, Grundy SM. Risk factors for the development of cholelithiasis in man. Part 1. New Engl J Med 1978; 299 : 1161-7.
  40. Trotman BW, Moris TA III, Cheney HM, et al. Pigment gallstone composition in cirrhotic and noncirrhotic subjects. Am J Dig Dis 1978; 23 : 872-6.
  41. Acalovschi M, Badea R, Diemitrascu D, et al. Prevalence of gallstone in liver cirrhosis : a sonographic study. Am J Gastroenterol 1988; 83 : 954-6.
  42. Bouchier IA. Post mortem study of the frequency of gallstones in patients with cirrhosis of the liver. Gut 1969; 10 : 705-10.
  43. Honore LH. The lack of a positive association between symptomatic cholesterol cholelithiasis and clinical diabetes mellitus : a retrospective study. J Chr Dis 1980; 33 : 465-9.
  44. Jorgensen T. Gallstone in a Danish population. Relation to weight, physical activity, smoking, coffee consumption and diabetes mellitus. Gut 1989; 30 : 528-34.
To Section TOC
Sponsor-Dr.Reddy's Lab