Paradoxical deterioration in seizure control due to anticonvulsant-induced hypocalcaemia

Abstract

Hepatic cytochrome P450 enzyme induction is associated with certain antiepileptic drugs (AEDs) and may result in hypocalcaemia secondary to vitamin D deficiency. We report a case of a 44-year-old man with a history of epilepsy, who presented with breakthrough seizures after having previously been seizure-free for 11 years. Investigations revealed severe hypocalcaemia with a corrected calcium of 1.7 mmol/L. His phenytoin dose was increased, and he was started on calcium supplementation. He was discharged with a corrected calcium level of 2.05 mmol/L but was readmitted 1 week later with further seizures and a corrected calcium of 1.89 mmol/L. 25-hydroxyvitamin D was low. AED-induced hypocalcaemia was suspected, which had been made paradoxically worse by the increase in phenytoin dose. Alfacalcidol was prescribed and he was switched from phenytoin to levetiracetam with resolution of hypocalcaemia and no further seizures. The authors recommend screening for calcium and vitamin D deficiency in patients on enzyme-inducing AEDs.

Background

Epilepsy is a common, chronic neurological condition in which treatment with antiepileptic drugs (AEDs) is usually necessary in order to prevent further seizures. However, AEDs have been associated with a number of adverse effects which can sometimes limit their use. Their effect on calcium and bone metabolism is a significant concern which has been previously documented in the literature. This phenomenon is particularly seen with the cytochrome P450 (CYP450) enzyme-inducing AEDs, such as phenytoin, which are associated with vitamin D deficiency which may lead to hypocalcaemia.^{1 2} AED-induced hypocalcaemia is usually mild and asymptomatic but may be severe and may rarely result in life-threatening clinical sequelae. We describe a case of paradoxical breakthrough seizures secondary to anticonvulsant-induced hypocalcaemia in a previously very well-controlled epileptic patient.

Case presentation

We report the case of a 44-year-old man known to suffer from epilepsy and learning disability, who presented to the emergency department with breakthrough seizures. He had suffered three generalised tonic–clonic seizures over the preceding 4 months, despite having previously been seizure-free for 11 years. His regular treatment included phenytoin 100 mg three times per day, phenobarbitone 30 mg three times per day, carbamazepine 200 mg three times per day, amitriptyline 10 mg three times per day and bromazepam 3 mg three times per day. He had been compliant to his medication, denied any missed doses or any change in medication. No obvious precipitating factor was evident at first. This was his first presentation to a medical professional since the recurrence of seizures over the last 4 months.

On presentation to the emergency department, he was haemodynamically stable, afebrile and had a capillary blood glucose of 5.6 mmol/L. Clinical examination was unremarkable apart from the presence of lateral tongue bites.

A CT scan of the head was performed which did not reveal any abnormality and a chest X-ray was also normal. Initial laboratory investigations revealed a white cell count of 12.95×10⁹/L (normal values: 4.3–11.4×10⁹/L), a haemoglobin of 114 g/L (normal values: 141–172 g/L) and a platelet count of 379×10⁹/L (normal values: 146−302×10⁹/L). Renal function was normal with an estimated glomerular filtration rate of 97.4 mL/min/1.73 m². His serum sodium and potassium levels were both normal at 142 mmol/L (normal values: 135–145 mmol/L) and 4.9 mmol/L (normal values: 3.5–5.1 mmol/L), respectively. CRP was normal at 3.6 mg/L.

The corrected serum calcium level was low at 1.7 mmol/L (normal values: 2.05–2.6 mmol/L) with a phosphate level of 1.2 mmol/L (normal values: 0.87–1.45 mmol/L) and a normal magnesium level of 0.93 mmol/L (normal values: 0.65–1.05 mmol/L). Chvostek’s and Trousseau’s signs were negative. No previous recent serum calcium levels were available for comparison. Serum alkaline phosphatase was raised at 176 U/L (normal values: 40–129 U/L). He was administered 10 mL of 10% calcium gluconate over 10 min followed by 100 mL of 10% calcium gluconate in 1 L of 0.9% saline infused at 50 mL/hour.

Random drug levels revealed a phenytoin level of 5.9 mg/L (therapeutic range: 7–20 mg/L), a phenobarbitone level of 17.3 mg/L (therapeutic range: 5–30 mg/L) and a carbamazepine level of 2.1 mg/L (therapeutic range: 1.5–9 mg/L). Serum ferritin, folate, vitamin B₁₂, thyroid function tests and morning cortisol were all within normal limits. Antitissue transglutaminase antibodies were negative.

His phenytoin dose was increased from 100–100–100 mg to 100–100–150 mg during this admission. He was started on calcium carbonate 1 g three times per day in view of the hypocalcaemia. His corrected calcium level rose to 2.08 mmol/L, he remained fit free and was discharged 6 days after admission in good health.

He was readmitted to hospital 1 week later after suffering two further generalised tonic–clonic seizures. Repeat investigations taken from the emergency department revealed that his serum corrected calcium level had once again dropped to 1.89 mmol/L. He had been compliant with treatment and random serum AED levels were within normal limits (phenytoin level of 10.7 mg/L, phenobarbitone level of 18.5 mg/L and a carbamazepine level of 2.2 mg/L).

Further blood investigations which had been taken during the first admission revealed a low serum 25-hydroxyvitamin D at 11 ng/mL (30–100 ng/mL), while serum 1,25-hydroxyvitamin D was within normal limits at 39 pg/mL (normal values: 16–81 pg/mL). Serum parathyroid hormone (PTH) was at the upper limit of normal at 66 pg/mL (normal values: 10–65 pg/mL).

An impression of drug-induced hypocalcaemia secondary to AED-induced enzyme induction was made and this had been made paradoxically worse by the recent increase in phenytoin dose during the previous admission, despite calcium carbonate supplementation. The phenytoin dose was reduced back to 100 mg three times per day and the dose was gradually tailed down by 50 mg every 2 weeks until it was stopped. He was started on levetiracetam at a dose of 500 mg per day, which was increased to 500 mg two times per day after 2 weeks. Alfacalcidol (1-alpha-hydroxycholecalciferol) was added to his treatment at a dose of 0.25 µg/day. This was gradually increased to 0.5 µg/day then two times and then three times per day over the following few months. His calcium levels improved, and he was discharged with a corrected calcium level of 2.15 mmol/L with close outpatient follow-up. His calcium carbonate tablets were switched to effervescent calcium at a dose of 1000 mg three times per day. His serum corrected calcium levels are summarised in table 1 and figure 1.

Table 1.

Serum corrected calcium levels (mmol/L) over time

Date	Corrected calcium level (mmol/L)
02/02/10	1.7
03/02/10	2.04
05/02/10	1.99
06/02/10	1.96
07/02/10	2.08
16/02/10	1.89
17/02/10	1.86
18/02/10	1.88
19/02/10	2.04
20/02/10	2.15

Figure 1.

Graphical depiction of serum corrected calcium levels over time.

Outcome and follow-up

He was followed up closely as an outpatient with regular monitoring of his serum calcium levels. He was kept on levetiracetam 500 mg two times per day while he was kept off phenytoin after the latter was tailed down. Effervescent calcium supplementation was prescribed and his alfacalcidol dose was gradually uptitrated as previously described. His calcium levels gradually normalised and his seizures did not recur.

Discussion

Hypocalcaemia results in increased neuromuscular excitability which may result in the classical signs and symptoms of paraesthesia, muscle cramps, tetany, Chvostek’s sign and carpopedal spasm as well as, more seriously, altered mental status and seizures. The development of these manifestations depends both on the degree of hypocalcaemia and on the rate of decline of serum calcium levels and also depends on the individual’s seizure threshold, with seizures being reported as occurring in 20%–25% of patients presenting with acute hypocalcaemia.³ Seizures may also occur in the absence of muscular tetany. The most commonly observed type of seizures associated with hypocalcaemia are of the generalised tonic–clonic type, though focal motor, absence, akinetic and even non-convulsive status epilepticus have all been reported.^{3 4}

Drug-induced hypocalcaemia is an important cause of this electrolyte disturbance which should always be considered along with the other common aetiological factors such as hypoparathyroidism, chronic kidney disease and malabsorption. This may occur via a variety of different mechanisms depending on the offending drug. In the case of AEDs this is observed most commonly with the inducers of CYP450 such as phenytoin, phenobarbitone and carbamazepine. The induction of hepatic microsomal enzymes results in the inactivation of vitamin D and 25-hydroxyvitamin D. This is thought to be mediated via activation of the pregnane X receptor which upregulates expression of the 24-hydroxylases which result in degradation of 25-hydroxyvitamin D. Phenytoin and phenobarbital may also contribute to hypocalcaemia by decreasing calcium absorption in the gastrointestinal tract.1 2 5 6 Vitamin D plays a vital role in calcium metabolism as it acts to increase intestinal absorption of calcium, increase bone resorption and decrease excretion of calcium and phosphate by the kidneys. Vitamin D deficiency may hence lead to hypocalcaemia as occurred in this case.

In hypocalcaemia secondary to vitamin D deficiency, biochemical analysis would be expected to reveal a low 25-hydroxyvitamin D, a normal to high serum PTH levels due to compensatory increase in PTH secretion by the parathyroid glands in response to low calcium levels, and a raised alkaline phosphatase which rises in response to the effect of PTH on calcium absorption from bone.⁷ These biochemical findings were all present in this case which fits with the diagnosis of hypocalcaemia due to vitamin D deficiency.

Vitamin D is synthetised by the skin in exposure to ultraviolet B radiation from sunlight which converts 7-dihydroxycholesterol to vitamin D₃ and is also obtained from the diet, particularly a diet rich in oily fish. Although subclinical vitamin D deficiency is common, hypocalcaemia secondary to vitamin D deficiency is rare in developed countries with an adequate diet particularly in sunny countries as in this case. Apart from drug-induced hepatic enzyme induction, vitamin D deficiency might also occur due to decreased synthesis by the skin (eg, burns or sunscreen), malabsorption (eg, coeliac disease, Crohn’s disease, cystic fibrosis or after gastric bypass surgery), liver failure (due to decreased synthesis of 25-hydroxyvitamin D from vitamin D₃), chronic kidney disease (due to decreased conversion of 25-hydroxyvitamin D to 1,25-dihydroxyvitamin D) as well as other rarer causes such as genetic conditions like vitamin D-resistant rickets.^{8 9} The clinical and biochemical picture in this case was most consistent with hepatic enzyme induction induced by his AEDs with no convincing features to suggest any of these other causes.

AED-induced enzyme induction is a well-known phenomenon and subsequent hypocalcaemia has been documented in the literature. Seizures secondary to this phenomenon, however, are probably rare or possibly under-recognised and reports of this complication have scarcely been previously documented in the literature. Nseir et al described a case of hypocalcaemia thought to be precipitated by phenytoin which resulted in a seizure in a patient with tuberous sclerosis and chronic kidney disease who was on prophylactic phenytoin due to cortical tubers.¹⁰ A similar case was reported by Ali et al who similarly described loss of seizure control in an epileptic patient on phenytoin and phenobarbital also thought to be secondary to phenytoin-induced hypocalcaemia.¹¹ In our case, the patient was on three different enzyme-inducing AEDs namely phenytoin, phenobarbitone and carbamazepine all of which likely played a role in the development of hypocalcaemia.

In this case, despite the apparent improvement in the calcium levels during the first admission once oral calcium carbonate was initiated, on increasing the phenytoin dose a further drop in the serum calcium which precipitated further seizures was observed 2 weeks later. This occurred despite continued calcium supplementation. This is keeping with phenytoin’s pharmacokinetics which has an average half-life of 22 hours and commonly reaches a steady state in 7–10 days.^{12 13} This lends further evidence that underlying AED-induced hepatic enzyme induction resulting in vitamin D deficiency and subsequent hypocalcaemia was the underlying mechanism for the breakthrough seizures in this case. Since tailing down the phenytoin and switching to levetiracetam (a newer AED which along with lamotrigine and topiramate is not associated with vitamin D deficiency),¹⁴ he has since been able to maintain his serum calcium levels within the normal range with the aid of supplementation and has now been seizure-free for many years.

In conclusion, we describe a case of paradoxical loss of seizure control due to AED-induced enzyme induction resulting in vitamin D deficiency and hypocalcaemia which worsened on increasing the dose of phenytoin. This case highlights the importance of considering this phenomenon in all patients on enzyme-inducing AEDs such as phenytoin who present with breakthrough seizures, particularly if these occur despite an increase in dose. In view of the potential serious sequelae of this phenomenon as outlined in this case report, we recommend that routine screening for calcium and vitamin D deficiency at neurology outpatient or general practitioner (GP) visits should be considered for all patients on phenytoin and other drugs known to cause enzyme induction even if well-controlled and asymptomatic.

Learning points.

• Certain antiepileptic drugs (AEDs), such as phenytoin, may cause hypocalcaemia primarily via cytochrome P450 enzyme induction which results in increased degradation of vitamin D.

• AED-induced hypocalcaemia may result in breakthrough seizures which might worsen on increasing the AED dose.

• AED-induced hypocalcaemia should be considered in all patients on enzyme-inducing AEDs presenting with loss of seizure control, particularly if seizures recur on increasing the AED dose and one might expect to find a low serum calcium, a low serum 25-hydroxyvitamin D, a raised ALP and a normal to high parathyroid hormone level.

• Routine screening of serum calcium and vitamin D levels is recommended for patients on enzyme-inducing AEDs.

• Should seizures secondary to AED-induced hypocalcaemia be suspected, one should consider switching to newer AEDs which are not associated with enzyme induction, in addition to calcium and vitamin D supplementation.