From the Departments of Pediatrics and Orthopedics*, Government Medical College and Hospital, Chandigarh 160 047, India.
Reprint requests: Jatinder S. Goraya, Senior Lecturer, Department of Pediatrics, Government Medical College and Hospital, Chandigarh 160 047, India.

Manuscript Received: March 22, 1999;
Initial review completed: May 12, 1999;
Revision Accepted: August 19, 1999

Epilepsy is a common neurological disorder of childhood frequently requiring prolonged use of anticonvulsants. Most of the commonly used anticonvulsants particularly phenytoin, pheno-barbitone, and carbamazepine have the propensity to interfere with vitamin D meta-bolism. Though biochemical changes are frequent, clinically overt rickets or osteomalacia is rare(1-6). Since the signs and symptoms of osteomalacia are non-specific, the diagnosis is frequently delayed resulting in considerable morbidity(1,2). We report three cases with anticonvulsant induced osteomalacia.

Case 1: A 15-year-old adolescent girl presented with pain above the left knee and limp of 1½-month duration. She had received various analgesic drugs without any relief. She was a known epileptic, having recurrent seizures despite being on carbamazepine (26 mg/kg/day) and phenobarbitone (3 mg/kg/day) for the last 2½ and 1 year, respectively. Because of poor seizure control, she was spending most of her time indoors. About one year prior to presenta-tion, she was diagnosed to have pulmonary tuberculosis and had received anti-tuberculous drugs (2 HRZE + 4HR) for 6 months. She had been complaining of non-specific body aches and pains with generalized weakness for the last 6 months. No complaints indicative of malabsorption, liver, or renal disease were present.

Examination revealed tenderness over left distal femoral shaft and an antalgic gait. Rest of the examination was unremarkable. X-ray left knee revealed a fracture line in the distal femoral diaphysis (Fig. 1). The patient was suspected to have anticonvulsants induced osteomalacia. A skeletal survey was perofrmed which showed Looser’s zones affecting ribs, and pubic ramus (Fig. 2) and generalized decrease in bone density. In addition, radiological evidence of rickets was present in the form of metaphyseal fraying and widening of epiphyseal plates. Serum biochemistry revealed serum calcium of 7.8 mg/dL (normal 8.2-11.5 mg/dL), phosporus 3.5 mg/dL (normal 2.7-4.8 mg/dL) and a raised alkaline phosphatase of 1491 U/L (normal 66-279.0 U/L). She was treated with single oral dose of 6,00,000 units of vitamin D and plaster immobilization for 6 weeks. Patient showed significant subjective improvement in her symptoms. By 8 weeks of treatment she was fully ambulatory and had minimal limp and normal serum calcium.

Fig. 1. Radiograph showing fracture line in the distal femoral diaphysis.

Case 2: A 17-year-old adolescent female presented with backache and limp of 6 months duration following a seizure related fall. X-rays done outside revealed insufficiency fractures of both femoral necks. She was treated with traction, analgesics and calcium supplements without any relief; rather her symptoms worsened and she was unable to walk. Examination revealed tenderness and limitation of movements at both hips. Serum biochemistry revealed low serum calcium of 8.1 mg/dL, phosphorus 3.5 mg/dL and raised alkaline phosphatase of 1020 U/L. X-ray pelvis revealed insufficiency fracture in both femoral necks. She was treated with an oral dose of 6,00,000 units of vitamin D and immobilization for 6 weeks with skin traction. She was a known epileptic receiving phenobarbitone (2 mg/kg/day) and carbamazepine (14 mg/kg/day) for the last 5 and 3 years, respectively. Serum calcium normalized by 6 weeks of treatment and alkaline phos-phatase started decreasing (773 U/L). The insufficiency fracture in both femoral necks healed and she became ambulatory.

Case 3: A 7-month-old boy was noticed to have widening of wrists, costochondral beading, and frontal bossing. 

Anticonvulsants induced osteomalacia is uncommonly recognized complication of long term anti-epileptic medication(1,2). The apparent association between long-term anticonvulsant therapy and biochemical and radiological changes of rickets was first described by Kruse in 1968 from Germany(7) and later by Dent et al. in 1970 from England(8).

The frequency of problem is related to the population of patients studied and the method of detection employed(1). Majority of the studies have been done in institutionalized, non-ambulatory patients receiving high dose multiple anticonvulsants for several years with sub-optimal dietary intake of vitamin D as well as limited exposure to sunlight(1-4,6-9). Biochemical evidence of osteomalacia, i.e., alteration in serum calcium, phosphorus and alkaline phosphatase has been seen in 20-70% of these pateints. Radiological changes of osteomalacia have been detected in upto 50% of cases(1-6). Clinically overt osteomalacia is, however rare compared to biochemical and radiological changes(1,2,4). Since the clinical features of osteomalacia, viz., musculoskeletal pains, weakness are nonspecific, the diagnosis is easily over-looked as happened in our patients(Cases I and II). Rarely, occurrence of fractures, both seizure and non-seizure related, lead to the correct diagnosis as happened in two of our patients (Cases I and II).

Anticonvulsants exert their deleterious effects on mineral and bone metabolism by inducing hepatic microsomal mixed function oxidase enzyme systems, resulting in increased catabolism and excretion of vitamin D and its biologically active products(1,2). In addition, drugs especially phenytoin may directly inhibit the intestinal calcium absorption(1).

Several factors enhance the risk of anticonvulsants induced bone disease(1,2). Patient related factors include decreased dietary vitamin D, limited exposure to sunlight, reduced physical activity(1,3,10-12). The latter two factors were present in one of our patient (Case I). In addition all our patients were in the age which corresponded to the period of rapid growth, i.e., adolescence (Cases I and II) and infancy (Case III). Rapid growth might have contributed to the depletion of already vulnerable vitamin D stores of the patients(3).

Though all commonly used anticonvulsants have the propensity to cause osteomalacia, phenytoin and phenobarbitone have the greatest potential(1-3). Most of the patients, who develop anticonvulsant induced osteomalacia, usually are on these drugs for several years though osteomalacia has also been described to set in within 3-6 months of therapy(13,14). Patients who receive more than one drug are at greater risk than those who are on mono-therapy(3,4,9). Apart from being on multiple durgs, i.e., carbamazepine and phenobarbitone, one of our patients (Case I) also had received rifampicin, which is another drug, which induces hepatic enzymes and interferes with vitamin D metabolism. It may be argued that Case III had nutritional rickets which are otherwise common in infancy. But in the presence of adequate nutritional status and lack of any other risk factors, phenobarbitone was thought to have contributed to a large extent in the development of rickets in this infant.

Diagnosis of anticonvulsant induced osteomalacia involves exclusion of other causes of osteomalacia particularly malabsorption, renal and hepatic disease(2) of which there was no evidence in our patients. The diagnosis should be considered in epileptic patients receiving anticonvulsant drugs, who develop bone pains, muscle weakness, fractures after minimal trauma or have worsening of seizure control. Since the worsening of seizure control may be due to hypocalcemia secondary to rickets or ostemomalacia(8), failure to recognize this fact may result in increase in the dose of anticonvulsants and/or addition of another anticonvulsant, thus initiating a vicious cycle.

Prevention of anticonvulsant induced osteomalacia lies in avoiding polytherapy, using minimum doses, ensuring adequate vitamin D intake and encouraging physical activity. Role of prophylactic vitamin D is controversial(1). Though beneficial effect of vitamin D supple-mentation on bone and mineral metabolism has been demonstrated in epileptic patients receiv-ing various anticonvulsant drugs(13-16), there are several limitations. While some authors recommend prophylactic vitamin D at the beginning of anticonvulsant therapy(13,14) others advise an interval of at least 6 months(1). There is a wide range of dose of vitamin D used for supplementation(3,14-16). Additional costs, risk of inadvertant over dosage, and need for frequent monitoring are other potential objec-tions to this approach(1). More importantly, most of the epileptics are out-patients while majority of the reports on anticonvulsant osteomalacia have concerned institutionalized patients with multiple risk factors.

In conclusion, epileptic children on long term anticonvulsants should be closely super-vised for the development of osteomalacia. In view of non-specific symptomatology, there is always a danger of the diagnosis beign over-looked. A high index of suspicion is required to make the diagnosis. Since the treatment is easy and effective, the need for early diagnosis cannot be over emphasized.

1. Hahn TJ. Bone complications of anticonvulsants. Drugs 1976; 12: 201-211.

2. Reynolds EH. Chronic antiepileptic toxicity. A Review. Epilepsia 1975; 16: 319-352.

3. Richens A, Rowe DJF. Disturbance of calcium metabolism by anticonvulsant drugs. BMJ 1970; 3: 73-76.

4. Lifshitz F, Maclaren NK. Vitamin D dependent rickets in institutionalized mentally retarded children receiving long term anticonvulsant therapy: A survey of 288 patients. J Pediatr 1973; 83: 612-620.

5. Tolman KG, Jubiz W, Sannella JJ, Madsen JA, Belsey RE, Gold Smith RS, et al. Osteomalacia associated with anticonvulsant drug therapy in mentally retarded children. Pediatrics 1975; 56: 45-50.

6. Hoikke V, Savolainen K, Alhava EM, Silvenius J, Karjalainen P, Repo A. Osteomalacia in institutionalized epileptic patients on long term anticonvulsant therapy. Acta Neurol Scand 1981; 64: 122-131.

7. Kruse R. Osteopathien bie antiepilaptischer Langeihterapie. Monatsselve Kinderheilkd 1968; 116: 378-381.

8. Dent CE, Richens A, Rowe DJF, Stamp TCB. Osteomalacia with-long term anticonvulsant therapy in epilapsy. BMJ 1970; 4: 69-72.

9. Borgstedt AD, Bryson MF, Young LW, Forbes GB. Long term administration of anti epileptic drugs and the development of rickets. J Pediatr 1972; 81: 9-15.

10. Baer MT, Kozlowski BW, Blyler EM, Taham CM, Taylor ML, Hogan MP. Vitamin D, calcium and bone status in children with development delay in relation to anticonvulsant use and ambulatory status. Am J Cllin Nutr 1997; 65: 1042-1051.

11. Mahapatro AK, Behari M, Prasad K, Ahuja GK. Phenytoin induced osteomalacia and fracture in a young epileptic. J Assoc Phys India. 1990; 38: 675-676.

12. Williams C, Netzolff M, Foklerts L, Vargas A, Garmica A, Frias J. Vitamin D, metabolism and anticonvulsant therapy: Effect of sunshine on incidence of osteomalacia. South Med J. 1984; 77: 834-836.

13. Rodbro P, Christiansen C, Lund M. Development of anticonvulsant osteomalacia in an epileptic patient on phenytoin treatment. Acta Neurol Scand, 1974; 50: 527-552.

14. Liakakos D, Papadonlos S, Vlashos P, Boviatsi E, Varonos DD. Serum alkaline phosphatase and urinary hydroxyproline values in children receiving phenobarbital with and without vitamin D. J Pediatr 1975; 87: 291-296.

15. Collins N, Maher J, Cole M, Baker N, Callaghan N. A prospective study to evaluate the dose of vitamin D required to correct low 25 hydroxy vitamin D, calcium and alkaline phosphatase in patients at risk of developing antiepileptic drug-inducted osteomalacia. QJ Med 1991, 78: 113-122.

16. Christiansen C, Rodbro P, Nielson CT. Iatrogenic osteomalacia in epileptic children. A controlled therapeutic trial. Acta Pediatr Scand 1975; 64: 219-224.