Introduction

Skeletal fluorosis is caused by chronic, excessive exposure to fluoride through ingestion or inhalation and most commonly occurs as a result of high fluoride levels in drinking water or industrial exposure from fumes or dust [1, 2]. Other potential sources of significant exposure include some forms of tea and wines, fluoride supplements and fluoridated toothpaste or mouthwashes. It is endemic in parts of the world, notably India, Africa and China [3–7] and is characterised by axial osteosclerosis, joint pain, ligamentous ossification and fractures. In advanced cases, the morbidity is high with reduced mobility of the spine, kyphosis and severe pain. The onset is usually gradual and following withdrawal of exposure to fluoride, symptomatic and radiographic improvement over a period of years has been reported [8, 9].

Although worldwide, millions of people are affected by skeletal fluorosis, it is very uncommon in parts of the world such as Europe and North America. In cases that have been reported, excessive consumption of fluoride-containing tea or wine, fluoride-containing mineral water or fluoridated toothpaste has been described [8, 10–14]. We report a woman whose major sources of fluoride exposure were tea and excessive use of fluoridated toothpaste.

History

A 58-year-old woman presented in December 2005 with a fracture of the left third metatarsal having sustained a fall whilst walking her dog. Past medical history included mild cerebral palsy from birth leading to mild right-sided weakness, right club foot, amputation of all her toes in the right foot due to deformities in 1993, and reflux oesophagitis. She was not taking any bone active medications.

Clinical examination

The patient walked with a stick due to her deformities. Examination of the spine was normal. She had poor dentition with severe dental caries but no dental mottling.

Investigations

Bone densitometry

Bone mineral density measured by dual energy X-ray absorptiometry in April 2006 revealed a T score of +11 in the lumbar spine and +7.6 in the total hip. These values had increased markedly since 1999, when she had undergone bone densitometry as a volunteer for a population-based study. Subsequent measurements in 2007 and 2008 showed smaller increases (Table 1).

Table 1 Bone mineral density T-scores
Full size table

Radiology

Plain radiographs showed marked osteosclerosis of the spine and pelvis. There was bilateral ossification of the sacrospinous ligament and of the anterior and posterior longitudinal ligaments at some locations in the spine. The skull and mandible were spared (Fig. 1). X-rays of the feet showed a stress fracture with florid callus formation involving the neck of the third metatarsal.

Fig. 1
figure 1

Lateral radiograph of the lumbosacral spine a and thoracic spine b showing ossification of the anterior and posterior spinous intervertebral ligaments (arrows)

Full size image

Biochemistry

Laboratory investigations revealed normal serum levels of calcium, inorganic phosphate, parathyroid hormone, 25-hydroxyvitamin D, alkaline phosphatase and tryptase. Serum protein electrophoresis, full blood count and erythrocyte sedimentation rate were normal. Serum total creatine kinase (CK) and isoenzymes, including CK-BB, were normal.

Fluoride analyses, including appropriate standards, were done using the ion-specific electrode. Water and urine were analysed using the “direct method”, which requires only the addition of a buffer to the standards and samples. Brewed tea, serum, fingernail clippings, bone ash and toothpaste were analysed in triplicate after overnight HMDS- facilitated diffusion [6, 15].

Serum fluoride was first measured on May 8, 2008 and was significantly elevated at 839 μg/l (normal 10–50 μg/l). Urinary fluoride excretion was also increased at 16.9 mg/24 h (normal <1.64 mg/24 h). Fluoride renal clearance was 23 ml/min (normal 30–40 ml/min). Fluoride content in her fingernails was elevated at 9.14 mg/kg (control subject 2.39 mg/kg). Subsequent measurements showed serum fluoride levels of 881 (18/9/08), 1,200 (21/7/09) and 1,120 μg/l (27/7/10).

Analysis of an iliac crest biopsy performed by an orthopaedic surgeon at the beginning of 2008 showed markedly sclerotic bone. The bone was described as compact and mixed lamellar/woven in nature. The biopsy was inadequate for histomorphometric analysis and the patient refused to undergo further bone biopsy. The fluoride content in the bone sample was greatly elevated at 15,144 mg/kg bone ash (normal 700–1,500 mg/kg).

Estimation of fluoride exposure

Analysis of her brewed tea revealed a total fluoride content of 7.6 mg/L. The patient had consumed six 8-ounce (240 ml) cups of standard breakfast tea daily, which would have provided 10.9 mg of fluoride/day, for the last 5–10 years. She denied drinking wine. She had lived in the same house for 40 years and used tap water for drinking. Analysis of the tap water showed a fluoride level of 0.3 ppm. There was no history of exposure to mining, welding or any other industrial source of fluoride.

When initially questioned about her use of toothpaste, the patient admitted to brushing her teeth three times a day and to using one large tube of toothpaste (100 g) each week. In January 2010, we analysed her toothpaste and found a fluoride content of 1,691 mg/kg, similar to the fluoride content of other fluoridated toothpastes.

In subsequent discussions, the patient admitted to brushing her teeth eight to ten times per day for the last 30 years. She had suffered from dental caries since childhood and this led to a strong fear of dental problems and obsessive brushing habits to prevent any further dental caries. Although she denied swallowing toothpaste, it is not uncommon to swallow about 25% of the toothpaste applied to the brush. If she covered the bristles with toothpaste (~1.0 g), brushed ten times daily and swallowed 25%, the daily fluoride dose would have been 4.2 mg. Daily fluoride intake from tea and toothpaste would have been 15.1 mg. Another 2–3 mg from dietary sources would have provided a chronic daily dose of 17–18 mg, an amount sufficient to cause the skeletal changes described above.

Discussion

Fluoride is rapidly absorbed from the gastrointestinal tract, following which around 50% becomes incorporated into bone mainly as hydroxyfluorapatite and the remaining 50% is excreted in the urine [16, 17]. Although fluoride stimulates osteoblastic proliferation and increases bone formation, the bone formed may be abnormal and inferior in quality [18–20]. Once in the skeleton, fluoride has a long half-life of around 7 years.

In many reported cases of fluorosis, joint pains and fracture have been prominent symptoms. Our patient presented with a fragility fracture affecting the appendicular skeleton but to date she has not suffered fractures in the osteosclerotic axial skeleton or had joint pains or stiffness. The cause of her fluorosis was difficult to establish, as initially she denied exposure to all known sources of high fluoride intake. Industrial exposure, tap water and wine were excluded but she drank significant amounts of tea. In addition, excessive use of toothpaste was suspected and she later admitted to this. A similar case was reported by Roos et al. [12] of a woman who brushed her teeth 18 times daily, consuming a large tube of toothpaste once every 2 days and swallowing the toothpaste. Kurland et al. [8] also recently described a man with fluorosis in whom excessive toothpaste use was believed to be the cause, although the patient never admitted to this.

Even though our patient did not have symptoms attributable to fluorosis, advice to change her dental regimen and drink less tea seemed appropriate in view of the known skeletal complications of the condition. The case reported by Kurland et al [8] demonstrated that a long time period is required for serum and urine fluoride levels to return to normal; although in the first 2 years of recovery, bone mineral density decreased rapidly; thereafter, it declined at a slower rate and was still markedly elevated after 9 years. Kurland et al. also observed that withdrawal of the source of fluoride was associated with an increase in urinary calcium resulting in nephrocalcinosis and an increase (within the normal range) of serum creatinine [8]. This was associated with increases in biochemical markers of bone turnover and was therefore presumably due to increased bone resorption, although the trigger for this after fluoride withdrawal is unclear.

A striking feature of our case was the very high serum, urine, nail and bone fluoride levels, to our knowledge the highest ever reported in a patient with fluorosis. These high levels were associated with a rapid increase in lumbar spine T-score, from +4.5 to +11 in a period of just over 7 years. Interestingly, her serum alkaline phosphatase level was normal at multiple time points, whereas elevated serum alkaline phosphatase levels have often been reported in other patients with skeletal fluorosis. A possible explanation for this may be the normal vitamin D status in our patient; osteomalacia has been described in other cases, particularly in association with vitamin D deficiency.

High bone mineral density is not an uncommon finding in patients undergoing screening for osteoporosis following a fracture or identification of risk factors for fracture. This case report emphasises the need to consider fluorosis as a possible diagnosis in such individuals. In pursuing this diagnosis, urinary, serum and nail fluoride concentrations are good indicators of current and prior fluoride exposure. Bone is the “gold standard” but biopsies are often not available.