To the Editor: Pain is a common complain in elderly people, often linked to neuropathic syndromes that affect old people particularly, such as trigeminal neuralgia, painful diabetic neuropathy, and postherpetic neuralgia.1 The effectiveness of anticonvulsant drugs in the management of neuropathic pain has been proven; for example, carbamazepine has been extensively employed and evaluated for pain relief in trigeminal neuralgia and has received Food and Drug Administration approval for this use.2–4 Nevertheless, carbamazepine has a narrow therapeutic window and has caused, albeit rarely, severe side effects, including hepatic failure, agranulocytosis, aplastic anemia, Stevens-Johnson syndrome, and toxic epidermal necrolysis. In addition, in polymedicated elderly patients, there is a high potential for adverse effects from drug-to-drug interactions, carbamazepine being a powerful inducer of the hepatic P450 cytochrome system. This could lead to reduced effectiveness of the drugs co-prescribed with carbamazepine. A 90-year-old woman was admitted to the hospital with exacerbation of long-standing arterial hypertension. She was also known to have symptomatic facial neuralgia, which had been treated with carbamazepine over many years. During her hospital stay, secondary causes of hypertension, such as pheochromocytoma and renal artery stenosis, were excluded. High systolic blood pressure measurements (range 240–260 mmHg) were noted several times daily despite pentatherapy (calcium channel blocker, loop diuretic, thiazide diuretic, angiotensin II receptor antagonist, and beta-blocker). Left ventricular hypertrophy was diagnosed according to echocardiogram. Other treatments with drugs that are not metabolized by the liver, in this case glyceryl trinitrate and moxonidine, appeared to be more effective. This observation suggested that the treatment resistance of the hypertension might be due to an effect of carbamazepine at the level of the antihypertensive drugs' hepatic metabolism. After discontinuing carbamazepine (replaced by gabapentin) and allowing sufficient time for its elimination, significant improvement in the blood pressure profile was observed; mean arterial blood pressures decreased from 177/64 mmHg (38 measures/24 h) to 157/65 mmHg (17 measures/24 h). Blood pressure measurements greater than 200 mmHg disappeared completely (>10 peaks/24 h previously), allowing the antihypertensive treatment regime to be reduced from five to three drugs. Some months after discharge home, carbamazepine was reintroduced for neurological symptoms, leading again to an uncontrolled elevation in blood pressure and hospital readmission. Carbamazepine has been reported to be associated with arterial hypertension and (more frequently) with exacerbation of preexisting hypertension through its effect on hepatic drug metabolism.5–7 Carbamazepine induces the cytochrome P450 enzymes 3A4, 2C9, and 1A2, which catalyze many antihypertensive drug types, for example, the calcium channel blockers, angiotensin II receptor antagonists, some loop diuretics (torsemide), and some alpha- and beta-blockers. Consequently, the pharmacological effect of these antihypertensive drugs is diminished. The hepatic cytochrome system does not metabolize other blood pressure–lowering molecules (such as angiotensin-converting enzyme (ACE) inhibitors, thiazide diuretics, potassium-sparing diuretics, agonists of alpha-2 central receptors, nitrates, and some beta-blockers (atenolol)), and hence the concomitant use of carbamazepine or other cytochrome P450–inducing drugs (e.g., rifampicin, dexamethasone, phenytoin, phenobarbital, and St. John's wort (Hypericum perforatum) does not affect their pharmacological effect. In addition to antihypertensive agents, cytochrome P450 metabolizes many other drugs, whose effect may also be expected to be compromised in the presence of carbamazepine. In our patient's treatment, cytochrome P450–metabolized drugs, including paracetamol, clopidogrel, omeprazole, zolpidem, and mianserin, were prescribed. The clinical case presented here illustrates the importance of considering potential drug-to-drug interactions whenever an anticipated therapeutic effect is not realized. The phenomenon of induced or exacerbated hypertension by carbamazepine is well known but is rarely mentioned in the geriatric literature, probably because the drug's primary indication is in the treatment of epilepsy, a condition that usually presents in younger patients, who rarely have concomitant hypertension, but carbamazepine has many other indications, such as for bipolar disorder, withdrawal procedures, and neuropathic pain, and its use has become more frequent, including in geriatric populations, in spite of the risk of interactions, or the risk of hepatic, cutaneus, and hematological toxicities. Carbamazepine is consequently not recommended for use in elderly patients, and an alternative treatment should be considered. Gabapentin, alone or in combination with clonazepam, should be tried for neuropathic pain,8,9 and antidepressants (tricyclics), topical anaesthetic agents, opiates, and analgesics could also be considered.10,11 Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this letter. Author Contributions: Emilia Frangos Lordos and Véronique Trombert: preparation of manuscript, acquisition of subject, interpretation of data. Nicole Vogt: revision of manuscript. Jean-Jacques Perrenoud: Revision of manuscript, supervision. Sponsor's Role: None.