Neurocritical Care

27. Drug Interactions

A 71-year-old female with a prior history of a glioblastoma, treated with radiation and temozolomide, was admitted with new-onset, generalized, tonic-clonic seizures. The patient had developed a recent deep venous thrombosis that required warfarin. Because she failed to maintain an adequately opened airway she was emergently intubated and–because of an allergy to phenytoin–she was treated with intravenous levetiracetam before transfer to the neurological intensive care unit. After admission, the patient received an additional dose of 2 grams of IV levetiracetam because of recurrent seizures, and after another seizure she also received valproic acid (1100 mg as a loading dose followed by 500 mg IV b.i.d.). Seizures were adequately controlled with this combination. The next day, the international normalized ratio (INR), which was measured basically as a routine follow-up, is 5.5 (increased from admission INR of 3.0), and a repeat INR hours later was 7.6. Brain MRI scan shows a glioma with large amount of vasogenic edema, but no evidence of a recent hemorrhage. Intratumoral blood products could be seen in the gradient recalled echo (GRE) sequence.

What do you do now?

It has been known for years that a single memorable experience with a drug interaction is needed to make physicians aware of that interaction. Without such experience—usually leading to a potential complication— physicians are not typically aware of major drug interactions and certainly not the less frequent ones. Fortunately, most of the drug interactions are clinically inconsequential and do not require adjustments.

Drug interactions are expected in critically ill patients—as a result of polypharmacy. Drug interactions can lead to medical complications, and best known are QTc prolongation, hypokalemia, hypotension, hypertension, and cardiac arrhythmias.

Drug interactions in a neurosciences intensive care unit (NICU) are often different from those caused by commonly used drugs in medical and surgical intensive care units. Although theoretically there are many, the 5 most common interactions in NICU are shown in Table 27.1. Any physician rotating through or attending in the NICU should be cognizant of these interactions, but in our experience they are often unrecognized. ICU pharmacists have been helpful in identifying these drug interactions early and in pointing out their pharmacokinetic basis.

One class of drugs that are mostly used in NICU are antiepileptic drugs, and that is where the unfamiliarity among physicians is the greatest. One such significant effect is that most antiepileptic drugs increase metabolism of warfarin and therefore decrease its effect. This has important repercussions if these antiepileptic drugs are suddenly discontinued, which could lead to sudden increase in warfarin effect. The only antiepileptic agent that increases the action of warfarin and therefore causes a significant increase in INR is valproic acid. Both drugs—warfarin and valproic acid—are acidic compounds with a small volume of distribution and they are highly protein bound. There is a high affinity for the same binding site on human albumin. Therefore, competition for albumin-binding sites between these drugs results in displacement of warfarin from albumin-binding sites and a transient increase in INR. In anticoagulated patients, displacement of less than 1% of total plasma warfarin can result in significant change of warfarin action.

Our patient is a typical example of such an interaction. This drug interaction is more profound if valproic acid is given in a large (loading) dose, and thus anticoagulated patients receiving a loading dose of valproic acid should have their INR closely monitored. INR may rise substantially and may potentially lead to bleeding either spontaneously or in areas prone to bleed (e.g., tracheostomy sites, prior wound beds, recent ischemic or hemorrhagic stroke, recent craniotomy, recent brain biopsy or ventriculostomy.) Our patient received 10 mg of vitamin K and fresh frozen plasma and did not develop any bleeding complication.

TABLE 27.1 The 5 Drug Interactions You Need to Know About in the NICU

Warfarin and Valproic acid

Mode of action: Drug displacement in protein binding site; a high loading dose reaching a higher serum level may displace warfarin from of valproic acid binding site.

Phenytoin and Fluconazole

Mode of action: Fluconazole inhibits phenytoin metabolism and may increase phenytoin level up to 4 times. Serum concentration monitoring with a reduction in phenytoin dosage is warranted.

Valproic acid and Carbapenems

Mode of action: The exact mechanism is unknown. Carbapenems, especially meropenem, may inhibit valproic acid absorption. Meropenem may accelerate the renal excretion and may result in low valproic acid serum level and increase risk of seizures. Additionally, carbapenems lower seizure threshold.

Statin and Levofloxacin or Amiodarone

Mode of action: The exact mechanism is unknown, but severe rhabdomyolysis may occur.

Clopidogrel and Omeprazole

Mode of action: Omeprazole inhibits CYP2C19, which is responsible for the conversion of clopidogrel into its active form. The effect of clopidogrel is reduced up to 47%.

In situations where acute decisions will have to be made—such as with our patient—drug interactions may not be on the radar, but it is important to consider the effects antiepileptic drugs have on anticoagulation. The antiepileptic drug that seems to have the safest profile is levetiracetam. Most importantly the frequent use of antiepileptic drugs in the NICU requires knowledge of their effects on other drugs and on other antiepileptics. These are summarized in Table 27.2.

TABLE 27.2 Drug Interactions with Antiepileptic Drugs


Decreases effect
























Increases effect

Valproic acid



· Most antiepileptic drugs decrease INR, and valproate increases INR, in patients receiving warfarin.

· Sudden discontinuation of antiepileptic drugs can result in marked increase in INR and lead to bleeding complications.

· Many antiepileptic drugs decrease the effect of commonly used drugs including corticosteroids and tricyclic antidepressants.

Further Reading

Bertsche T, Pfaff J, Schiller P, et al. Prevention of adverse drug reactions in intensive care patients by personal intervention based on an electronic clinical decision support system. Intensive Care Med 2010; 36:655-672.

Guthrie SK, Stoysich AM, Bader G, Hilleman DE. Hypothesized interaction between valproic acid and warfarin. J Clin Psychopharm 1995; 15:138-139.

Juurlink DN, Mamdani M, Kopp A, Laupacis A, Redelmeier DA. Drug-drug interactions among elderly patients hospitalized for drug toxicity. JAMA 2003; 289:1652-1658.

Ko Y, Malone DC, D’Agostino JV, Skrepnek GH, Armstrong EP, Brown M, Woosley RL. Potential determinants of prescribers’ drug-drug interaction knowledge. Research in Social and Administrative Pharmacy 2008; 4:355-366.

Moura C, Prado N, Acurcio F. Potential drug-drug interactions associated with prolonged stays in the intensive care unit: a retrospective study. Clin Drug Investig 2011;31:309-316.

Reimche L, Forster AJ, van Walraven C. Incidence and contributors to potential drug-drug interactions in hospitalized patients. J Clin Pharmacol 2011; 51:1043-1050.

Seller EM, Koch-Weser J. Kinetics and clinical importance of displacement of warfarin from albumin by acidic drugs. Annals of the New York Academy of Sciences 1971; 179:213-225.