Background Determination of the plasma concentration of free phenytoin (C_pFL) could improve seizure control and prevent adverse effects.

Purpose To evaluate the safety profile of patients treated with phenytoin using C_pFL.

Material and methods Prospective study (2013–2014) in a hospital. Collected data: demographics, doses, C_pFL, creatinine clearance (Cl_cr), serum albumin (g/dL), degree of intoxication, days of hospitalisation and concomitant medication. Phenytoin therapeutic range, CPFL: 1–2.5 μg/mL. Moderate intoxication, C_pFL 2.5–3.0 μg/mL and severe, C_pFL >3.0 μg/mL. To determine renal clearance, we used CKD-EPI. Moderate renal impairment was defined as Cl_cr 20–50 mL/min. Polymedicated patients: >5 drugs. Statistical analysis: Spearman correlation and the χ² test.

Patients 93 (cases 192; phenytoin levels/patient 1–6). Men 51.6%. Age 58 years (range 27–84). Daily dose 299 mg/day. C_pFL 1.1 μg/mL. Cl_cr 51.7 mL/min. Serum albumin 3.6 g/dL. Levels in the therapeutic range: 49.5% (95/192), 32.8% (63/192) were suboptimals and 17.7% were toxic (34/192) (C_pFL 3.8 μg/mL; range 2.6–5.7 μg/mL). Intoxication, moderate was 64.7% and severe 35.3%. Average age (Intoxicated patients) 71 years. Cl_cr 38.9 mL/min. Serum albumin 3.4 g/dL. Three patients were hospitalised. Polymedicated patients: 71% vs. 50% for the rest. Patients with drugs that bind over 70% to plasma proteins: 48%. Patients >70 years had a higher risk of intoxication (p = 0.033). We observed an inverse correlation between C_pFL and Cl_cr (Spearman rho: -0.562; p = 0.04) and with albumin (Spearman rho: -0.623; p < 0.01). In relation to moderate intoxication, the plasma concentration of phenytoin had a value 23% higher than C_pFL.

Conclusion Elderly patients, polymedicated patients and those with moderate renal insufficiency and hypoalbuminaemia presented a higher risk of phenytoin toxicity. It would be advisable to be careful with these patients because in our study efficacy/toxicity is correlated better with C_pFL.

No conflict of interest.