Background Studies of groundwater sources in Sri Lanka show that in the entire Northern, North Central, Uva and Eastern Provinces, and also in a large area of the North Western Province, groundwater fluoride content is within the range 1.0 mg/l and 3.0 mg/l. The WHO recommended safe upper limit of fluoride for drinking water is 1.5 mg/l.Objective Our objective was to determine the prevalence and degree of dental fluorosis in a cohort of schoolchildren in Sinnasippikulam in Vavuniya District, the fluoride content in their serum and urine, and fluoride levels in drinking water in dug wells of the area.Methods Invitations in Tamil and Sinhala were sent to all households by courier. Demographic information and oral hygiene practices of 307 responsive children were recorded by interviewing children and their mothers. Detailed clinical examinations were performed according to WHO basic methods, with some modifications. Assessment of dental fluorosis in the children was done according to Dean's Index codes and criteria. Conclusions Our results show that harmful levels of fluoride are extremely common in groundwater sources in the study area, and consequently, dental fluorosis is highly endemic (72.9%) among resident schoolchildren.

Background The WHO recommended safe upper limit for fluoride in drinking water is 1.5 mg/l. Groundwater sources in many parts of Sri Lanka often exceed this limit. The high fluoride content of groundwater and high environmental temperatures in Vavuniya District predispose to pre-skeletal fluorosis and skeletal fluorosis in adults. Objectives To identify residents of Vavuniya District with clinical features of pre-skeletal and skeletal fluorosis; to describe their clinical, biochemical and radiographic features; to determine the fluoride content of blood and urine in individuals with established diagnoses, and of their drinking water. Methods In 98 volunteers we detected 60 with clinical features of pre-skeletal and skeletal fluorosis. Clinical examination, biochemical and radiographic investigations were performed. Forty four with confounding factors were excluded. The balance 16 had radiographic investigation for fluoride bone disease, and assessment of clinical features for pre-skeletal fluorosis. The radiographic criteria of skeletal fluorosis were trabecular haziness, osteosclerosis, osteophytes, cortical thickening and ligamentous or muscle attachment ossification. All 16 had "spot" samples of 15 ml of venous blood taken for biochemical tests and fluoride estimation; and 30 ml of urine, and water from 16 dug wells for fluoride. Results The 16 selected (11 males) had BMI between 20.6 and 31.9 kg/m 2 , and were between 22 and 84 years (59.9 + 20.4). They used water from domestic dug wells for drinking. All had adequate renal function. All serum and urine samples had raised fluoride levels way above the reference ranges for serum (0.02-0.18 mg/l) and urine (0.6-2.0 mg/l). The 16 water samples showed a mean fluoride content of 2.90 + 0.93 mg/l.