Philip D. Hansten, PharmD

Practitioners have known for almost 2 decades that oral contraceptives (OCs) consistently and substantially reduce the serum concentrations of lamotrigine and may reduce seizure control. This mechanism appears to be caused by the induction of lamotrigine glucuronidation by OCs.

Let’s look at 3 more-recent discoveries related to the interaction that provide important new insights. First, a recent case report demonstrated that the interaction may be problematic in patients with bipolar disorder as well as those with seizures. Second, there is evidence that the interaction may be mitigated by other drugs, specifically valproic acid. Finally, it appears that lamotrigine levels may also be reduced by hormone replacement therapy (HRT).

A 23-year-old woman taking an OC (1 mg norethisterone + 0.035 mg ethinyl estradiol) was started on lamotrigine 400 mg/day for bipolar disorder.

 After several weeks, her lamotrigine serum concentrations were quite low, and her bipolar disorder had not improved. She was then switched sequentially to lithium, valproic acid, and carbamazepine, but they were all ineffective, and some had adverse events. Her health care provider then substituted leuprorelin for her OC and restarted lamotrigine at the original dose (400 mg/day), leading to therapeutic lamotrigine concentrations and effective stabilization of her bipolar disorder. This case is consistent with the findings in patients taking lamotrigine for seizures, and the interaction is rated “probable” using the Drug Interaction Probability Scale.

In a prospective study in women taking lamotrigine for epilepsy, the effect of an additional antiepileptic drug on the lamotrigine—OC interaction was studied.

 In women on lamotrigine monotherapy or on lamotrigine plus carbamazepine, lamotrigine clearance fell as expected during the OC-free week. However, in patients on lamotrigine plus valproic acid, there was no change in lamotrigine clearance during the OC-free week. The study was small, and more data are needed, but it does suggest that the expected lamotrigine—OC interaction does not occur in patients also taking valproic acid.

Valproic acid is known to inhibit uridine diphosphate—glucuronyltransferase (UGT) activity, so it is possible that UGT inhibition by valproic acid interferes with the UGT induction caused by OCs. Clinically, it is important to consider that valproic acid may inhibit the lamotrigine–OC interaction because anticipatory dose adjustments of lamotrigine when starting or stopping OCs could result in lamotrigine levels that are too high, leading to adverse outcomes, or too low, leading to seizures.

In a case-control study, women taking lamotrigine with ethinyl estradiol or HRT were matched with controls taking lamotrigine without concurrent estrogen therapy. When lamotrigine serum concentration-to-dose ratios (CDRs) were calculated, both the women on ethinyl estradiol and those on HRT had significantly lower lamotrigine CDRs than women taking lamotrigine in the absence of estrogens. Given that this was a retrospective case-control study, confirmation of this effect is needed, ideally with a prospective controlled study. However, the results are consistent with a report describing 2 women on lamotrigine who developed reduced lamotrigine serum concentrations after HRT was started, one of whom had increased seizure frequency.

The effect of HRT on lamotrigine is interesting because theoretically, HRT should simply restore serum estrogen concentrations to physiologic premenopausal levels. But the available evidence does suggest that HRT may reduce lamotrigine serum concentrations and efficacy.

The ability of OCs to substantially reduce lamotrigine serum concentrations is well documented and may necessitate adjustments in lamotrigine dosege. The interaction is also likely to occur when lamotrigine is used for bipolar disorder or other indications, and some evidence suggests that it may occur with the use of HRT as well as OCs. In patients with epilepsy, the presence of valproic acid may mitigate the interaction. This means that stopping or starting valproic acid may cause the lamotrigine—OC interaction to appear or disappear, effects that are very unlikely to be accounted for by computerized clinical decision support systems.

John R. Horn, PharmD, FCCP, and Philip D. Hansten, PharmD, are professors of pharmacy at the University of Washington School of Pharmacy in Seattle.

Sawagashira R, Fujii Y, Kusumi I. Clinical pharmacokinetic interactions between lamotrigine and hormonal contraceptives in bipolar I disorder. 

Horn JR, Hansten PD, Chan LN. Proposal for a new tool to evaluate drug interaction cases. 

 2007;41(4):674-680. doi: 10.1345/aph.1H423.

Wegner I, Wilhelm AJ, Lambrechts DA, Sander JW, Lindhout D. Effect of oral contraceptives on lamotrigine levels depends on comedication. 

. 2014;129(6):393-398. doi: 10.1111/ane.12197.

Harden CL, Herzog AG, Nikolov BG, et al. Hormone replacement therapy in women with epilepsy: a randomized, double-blind, placebo-controlled study. 

 2006;47(9):1447-1551. doi: 10.1111/j.1528-1167.2006.00507.x.

Articles

Practitioners have known for almost 2 decades that oral contraceptives consistently and substantially reduce the serum concentrations of lamotrigine and may reduce seizure control.

Estrogens and Lamotrigine: New Developments

The combined use of valproic acid (VPA) and a carbapenem is likely to occur in patients with central nervous system infections resulting from head trauma or surgical procedures; serious systemic infections, including pneumonia; and generalized seizure disorders.

The data describing this interaction have been based primarily on case reports and retrospective studies. Nevertheless, this potential drug interaction presents a risk to patients.

VPA is primarily metabolized by glucuronidation to VPA-glucuronide, which is eliminated from the body via bile and urine. However, VPA-glucuronide can be deconjugated by acylpeptide hydrolase to the parent compound VPA.

 The conversion of VPA-glucuronide back to VPA provides for a longer half-life and duration of action for VPA. Although the mechanism responsible for the reported interaction between VPA and carbapenem has been debated, it appears that carbapenems act by inhibiting the activity of the acylpeptide hydrolase enzyme that converts VPA-glucuronide to VPA.

 This results in greater elimination, a shortened half-life, and subtherapeutic plasma concentrations of VPA, all of which represent an uncommon drug interaction mechanism that can produce clinically significant patient outcomes.

summarizes several of the larger studies of the effect of carbapenems on VPA. In addition, there have been several case reports of subtherapeutic VPA concentrations in patients treated with carbapenems. The magnitude of this interaction ranges from about a 50% to an 80% reduction in VPA plasma concentrations. This reduction has been reported to occur in as few as 24 hours after initiation of the carbapenem. The half-life of VPA is reduced from 15 to 4 hours with concurrent carbapenem administration. It appears that all of the available carbapenems can cause VPA concentrations to decline. Imipenem appears to demonstrate the least effect on VPA, although it clearly can cause a loss of VPA efficacy.

In several studies, patients were noted to have developed increased seizure frequency during carbapenem administration. Although carbapenems have been noted to cause seizures, an acute reduction in the VPA concentration has been associated with status epilepticus and may precipitate seizures in these patients. The carbapenems are eliminated by the kidneys, and this interaction may be more likely to occur in patients with renal dysfunction. Following the discontinuation of the carbapenem, the effect on VPA has been noted to continue for 7 to 14 days. The likely explanation of this prolonged duration of the interaction despite the short half-life of the carbapenems is because of an irreversible inhibition of acylpeptide hydrolase by the carbapenem, resulting in a half-life for recovery of deconjugation estimated to be almost 2 days.

 In nearly all of the reports of this interaction, increasing the dose of VPA did not result in therapeutic VPA concentrations while the carbapenems were coadministered.

The effect of a carbapenem on VPA is significant and may result in reduced efficacy and increased seizure activity. Because of the mechanism of this interaction, increasing the dose of the VPA will unlikely achieve therapeutic plasma concentrations. If possible, an alternative antibiotic should be chosen for patients stabilized on VPA. When a carbapenem is required, additional antiepileptic therapy would be appropriate during carbapenem coadministration and for 1 to 2 weeks after the antibiotic is discontinued.

Suzuki E, Nakai D, Ikenaga H, et al. In vivo inhibition of acylpeptide hydrolase by carbapenem antibiotics causes the decrease of plasma concentration of valproic acid in dogs. 

. 2016;46(2):126-131. doi: 10.3109/00498254.2015.1054002.

Ishikawa T, Otaki H, Mizuta S, et al. Computational study of the competitive binding of valproic acid glucuronide and carbapenem antibiotics to acylpeptide hydrolase. 

. 2017;32(4):201-207. doi: 10.1016/j.dmpk.2017.04.002.

Wu CC, Pai TT, Hsiao FY, Shen LJ, Wu FL. The effect of different carbapenem antibiotics (ertapenem, imipenem/cilastatin, and meropenem) on serum valproic acid concentrations. 

. 2016;38(5):587-592. doi: 10.1097/FTD.0000000000000316.

Wen Z-P, Fan SS, Du C, et al. Drug-drug interaction between valproic acid and meropenem: a retrospective analysis of electronic medical records from neurosurgery inpatients. 

. 2017;42(2):221-227. doi: 10.1111/jcpt.12501.

Velez-Diaz-Pallares M, Delgado SE, Alvarez Diaz AM, Perez Menendez-Conde C, Vicente ON, Bermejo VT. Analysis of the valproic acid-meropenem interaction in hospitalized patients. 

. 2012;27(1):34-38. doi: 10.1016/j.nrl.2011.03.008.

Haroutiunian S, Ratz Y, Rabinovich B, Adam M, Hoffman A. Valproic acid plasma concentration decreases in a dose-independent manner following administration of meropenem: a retrospective study. 

. 2009;49(11):1363-1369. doi: 10.1177/0091270009334377.

Huang CR, Lin CH, Hsiao SC, et al. Drug interaction between valproic acid and carbapenems in patients with epileptic seizures. 

. 2017;33(3):130-136. doi: 10.1016/j.kjms.2016.12.001.

Miranda Herrero MC, Alcaraz Romero AJ, Escudero Vilaplana V, et al. Pharmacological interaction between valproic acid and carbapenem: what about levels in pediatrics? 

. 2015;19(2):155-161. doi: 10.1016/j.ejpn.2014.12.010.

Doribax [prescribing information]. Osaka, Japan; Shinogi & Co Ltd, 2007. www.accessdata.fda.gov/drugsatfda_docs/label/2009/022106s002s005lbl.pdf. Accessed September 4, 2018.

Masuo Y, Ito K, Yamamoto T, Hisaka A, Honma M, Suzuki H. Characterization of inhibitory effect of carbapenem antibiotics on the deconjugation of valproic acid glucuronide. 

. 2010;38(10):1828-1835. doi: 10.1124/dmd.110.034231.

The combined use of valproic acid and a carbapenem is likely to occur in patients with central nervous system infections resulting from head trauma or surgical procedures; serious systemic infections, including pneumonia; and generalized seizure disorders.
 

Valproic Acid and Carbapenem Antibiotics Interaction

The antibacterial agent linezolid is a nonselective inhibitor of monoamine oxidase (MAO), and when it is combined with serotonergic agents, it has been associated with serotonin syndrome. A recent case of serotonin syndrome was reported with concurrent use of linezolid and methadone. 

A 39-year-old man on darunavir, lamivudine, methadone, ritonavir, and tenofovir was admitted with a high fever and septicemia.

 After initial antibiotic therapy with amoxicillin and ceftriaxone, he was switched to 600 mg of clindamycin 4 times a day and 600 mg of linezolid twice a day. Within 3 days he became agitated, confused, and disoriented, and he developed increased tone in both legs, hyperreflexia, and clonus. These symptoms clearly warrant a diagnosis of serotonin syndrome using the more specific Hunter criteria.

 Assessing the causality of the serotonin syndrome because of linezolid plus methadone in this case using the Drug Interaction Probability Scale gives a rating of “probable.”

Methadone is one of the opioids that is known to have serotonergic effects, along with fentanyl, meperidine, and tapentadol. In the case described above, the methadone plasma concentrations may have been higher than expected because 2 of methadone’s metabolic pathways are inhibited by ritonavir (CYP2D6 and CYP3A4) and darunavir (CYP3A4).

Linezolid is a relatively weak nonselective MAO inhibitor, but a number of cases of serotonin syndrome have been reported when it has been combined with serotonergic agents. Most of the case reports have involved the use of linezolid with serotonin-norepinephrine reuptake inhibitors (SNRIs) or selective serotonin reuptake inhibitors (SSRIs), though isolated cases have been reported when linezolid was used with other serotonergic drugs.

Results from retrospective studies suggest that there is a low overall risk of serotonin syndrome in large populations of people receiving linezolid with SNRIs or SSRIs,

 and in many of the reported cases, the patients had a variety of predisposing factors that probably increased their risk.

In one case, a 98-year-old man developed serotonin syndrome after the concurrent use of linezolid with both amiodarone and citalopram.

 The citalopram serum concentrations were probably higher than normal because amiodarone inhibits several of the metabolic pathways and transporters for citalopram such as CYP2D6, CYP3A4, and P-glycoprotein.

In another case, a 68-year-old woman who developed well-documented serotonin syndrome while on linezolid had received 2 serotonergic drugs, fentanyl and amitriptyline.

 Although amitriptyline does not have strong serotonergic effects, it probably added to the serotonergic effects of fentanyl in this case. Other cases of linezolid-associated serotonin syndrome have involved the combined use of fentanyl and fluoxetine in a 4-year-old girl

 and duloxetine and fentanyl in a 55-year-old woman.

A 58-year-old man with obsessive-compulsive disorder was taking 60 mg of fluoxetine a day for 10 years, 100 mg of sertraline for a day for 7 weeks, and 100 mg of fluvoxamine a day for 5 weeks without evidence of serotonin syndrome.

 Fentanyl and linezolid were added, and his SSRIs were stopped. But he developed signs of serotonin syndrome with spontaneous clonus, hyperreflexia, rigidity, and diaphoresis. He responded to discontinuation of the fentanyl and linezolid, along with the serotonin antagonist cyproheptadine, with a 12-mg loading dose, followed by 2 mg every 2 hours. Cyproheptadine has been recommended by others for the prophylaxis or treatment of serotonin syndrome.

Many cases of serotonin syndrome have been reported in patients receiving linezolid, along with serotonergic drugs. Many of the patients had predisposing factors, such as more than 1 serotonergic drug or the presence of other drugs that increase the plasma concentration of the serotonergic drug. Although results from retrospective studies suggest a low prevalence of serotonin syndrome in patients receiving linezolid with serotonergic drugs, those on such therapy should be monitored for evidence of serotonin syndrome. Prompt discontinuation of the serotonergic drugs and the linezolid (if possible) is indicated. Cyproheptadine treatment has apparently been useful in treating serotonin syndrome in several cases, but no controlled studies are available. 

John R. Horn, PharmD, FCCP, and Philip D. Hansten, PharmD, are professors of pharmacy at the University of Washington School of Pharmacy in Seattle

The antibacterial agent linezolid is a nonselective inhibitor of monoamine oxidase (MAO), and when it is combined with serotonergic agents, it has been associated with serotonin syndrome. A recent case of serotonin syndrome was reported with concur- rent use of linezolid and methadone.

Linezolid and Serotonin Syndrome

Myopathy is a well-established adverse outcome of statin use, and the muscle damage can range from minor muscle pain and weakness to life-threatening rhabdomyolysis. Statin-induced myopathy often occurs and is the result of increased statin plasma concentrations caused by interacting drugs, usually inhibitors of CYP3A4. For some statin drug interactions, other mechanisms may also be involved.

Colchicine can also produce myopathy, and rhabdomyolysis has been reported in severe cases. As with statins, muscle damage often occurs when CYP3A4 inhibitors cause large increases in colchicine plasma concentrations. Given that both colchicine and statins can cause myopathy individually, health care professionals might expect that the combination of statins and colchicine would increase the risk. Numerous credible case reports published over the past 2 decades do suggest that the combination of colchicine and statins may increase the risk of myopathy compared with either drug alone.

In cases of myopathy in patients receiving colchicine and a statin, muscle weakness was the most common finding, occurring in virtually every case. In some cases, the weakness was so severe that the patient could not even stand up. In about one-third of the cases, muscle weakness was accompanied by muscle pain. Darkening of the urine to a tea-colored hue also sometimes accompanied the physical symptoms. 

The onset of the symptoms usually occurred after 2 to 3 weeks of combined therapy, but in some cases, the patient noticed changes after about a week. Occasionally, there was a latent period of 2 to 3 months or more. This is important to note because clinicians may not have a high index of suspicion about something that occurred months previously. 

A study of statin-induced myopathy found that 40% of the patients received a drug that may have predisposed them to the muscle damage, and colchicine was the second-most-common interacting drug.

A more recent report focusing only on simvastatin drug—drug interactions (DDIs) reported that the combination with colchicine was the second-most-common DDI exposure, after gemfibrozil. 

One remarkable case of fatal rhabdomyolysis involving colchicine and simvastatin is instructive.

A 66-year-old man presented with muscle weakness so severe that he had trouble walking and getting up from a chair. His medications were colchicine 1.2 mg per day, cyclosporine 150 mg per day, mycophenolate 500 mg per day, prednisone 7.5 mg per day, and simvastatin 60 mg per day. His creatine kinase (CK) was 2538, suggesting myopathy, so the colchicine and simvastatin were stopped, but he was then given intravenous methylprednisolone and propofol. He developed severe rhabdomyolysis with a CK of 33,580 and died 5 days after admission. 

shows that he received 5 drugs that can individually cause myopathy (represented by the single arrows) and 5 combinations of drugs that have been associated with myopathy and/or pharmacokinetic DDIs (represented by the double arrows). Although it is impossible to determine which drugs were primarily responsible in this tragic case, the ability of cyclosporine to increase plasma concentrations of colchicine, methylprednisolone, and simvastatin may have played a significant role. 

If it is necessary to give patients colchicine and a sta- tin, use conservative doses of both drugs, if possible. The patient should be advised to immediately report any muscle pain or weakness as well as brownish or darkened urine. The patient is likely to fare better if the myopathy is diagnosed quickly and the drugs are discontinued

John R. Horn, PharmD, FCCP, and Philip D. Hansten, PharmD, are both professors of pharmacy at the University of Washington School of Pharmacy in Seattle. For an electronic version of this article, including references, visit hanstenandhorn.com.

1. Hsu W-C, Chen WH, Chang MT, Chiu HC. Colchicine-induced acute myopathy in a patient with concomitant use of simvastatin. 

2. Baker SK, Goodwin S, Sur M, Tarnopolsky MA. Cytoskeletal myotoxicity from simvastatin and colchicine. 

2004;30(6):799-802. doi: 10.1002/mus.20135.

3. Justiniano M, Dold S, Espinoza SR. Rapid onset of muscle weakness (rhabdomyolysis) associated with the combined use of simvastatin and colchicine. 

2007;13(5):266-268. doi: 10.1097/RHU.0b013e318156d977.

4. Sahin G, Korkmaz C, Yalcin AU. Which statin should be used together with colchicine? Clinical experience in three patients with nephrotic syndrome due to AA type amyloidosis. 

2008;28(3):289-291. doi: 10.1007/s00296-007-0435-1.

5. Francis L, Bonilla E, Soforo E, et al. Fatal toxic myopathy attributed to propofol, methylprednisolone, and cyclosporine after prior exposure to colchicine and simvastatin. 

2008;27(1):129-131. doi: 10.1007/s10067-007-0696-9.

6. Tufan A, Dede DS, Cavus S, Altintas ND, Iskit AB, Topeli A. Rhabdomyolysis in a patient treated with colchicine and atorvastatin. 

. 2006;40(7-8):1466-1469. doi: 10.1345/aph.1H064.

7. Phanish MK, Krishnamurthy S, Bloodworth LL. Colchicine-induced rhabdomyolysis. 

8. Vasudevan AR, Uthamalingam S, Kumar S, Tamarin F, Brensilver JM. Colchicine-induced rhabdomyolysis: the whole is greater than the sum of its parts! 

9. Torgovnick J, Sethi N, Arsura E. Colchicine and HMG Co-A reductase inhibitors induced myopathy—a case report. 

2006;27(6):1126-1127. doi: 10.1016/j.neuro.2006.09.003.

10. Atasoyu EM, Evrenkaya TR, Solmazgul E. Possible colchicine rhabdomyolysis in a fluvastatin-treated patient. 

2005;39(7-8):1368-1369. doi: 10.1345/aph.1E653.

11. Sarullo FM, Americo L, Di Franco A, Di Pasquale P. Rhabdomyolysis induced by co-administration of fluvastatin and colchicine. 

2010;74(3):147-149. doi: 10.4081/monaldi.2010.264.

12. Alayi G, Cengiz K, Canturk F, Durmus D, Akyol Y, Menekse EB. Acute myopathy in a patient with concomitant use of pravastatin and colchicine. 

. 2005;39(7-8):1358-1361. doi: 10.1345/aph.1E593.

13. Bouquié R, Deslandes G, Renaud C, Dailly E, Haloun A, Jolliet P. Colchicine-induced rhabdomyolysis in a heart/lung transplant patient with concurrent use of cyclosporin, pravastatin, and azithromycin. 

2011;17(1):28-30. doi: 10.1097/RHU.0b013e3182056042. 

14. Boonmuang P, Nathisuwan S, Chaiyakunapruk N, et al. Characterization of statin-associated myopathy case reports in Thailand using the health product vigilance center database. 

. 2013;36(9):779-787. doi: 10.1007/s40264-013-0055-5.

15. Siriangkhawut M, Tansakul P, Uchaipichat V. Prevalence of potential drug interactions in Thai patients receiving simvastatin: the causality assessment of musculoskeletal adverse events induced by statin interaction. 

. 2017;25(6):823-829. doi: 10.1016/j.jsps.2016.12.006.

 Myopathy From Colchicine and Statins

Endothelin antagonists—ambrisentan (Letairis), bosentan (Tracleer), and macitentan (Opsumit)&mdash;are indicated to treat pulmonary arterial hypertension. Although these drugs have similar pharmacologic activity, they differ in their potential for their plasma concentrations to be altered by other medications (Table 1).

 Taking other medications with endothelin antagonists can have various effects (Table 2).

The half-life of ambrisentan is 9 to 15 hours, with its primary metabolism via uridine 5’-diphosphate glucuronosyltransferases (UGTs); CYP450 (CYP) 3A4, CYP2C19, P-glycoprotein (P-gp), and organic anion transporting polypeptides (OATPs) also contribute to its elimination.

 The mean area under the curve (AUC) of ambrisentan increases about 2-fold with the addition of cyclosporine but is not markedly affected by ketoconazole.

 Both cyclosporine and ketoconazole inhibit CYP3A4 and P-gp, but cyclosporine also inhibits OATPs, perhaps accounting for the greater effect seen with cyclosporine. The importance of OATP in ambrisentan elimination is key in rifampin’s effect on ambrisentan pharmacokinetics. Although the steady-state AUC of ambrisentan was minimally changed, during the first 2 days following rifampin plus ambrisentan, the AUC rose nearly 90%. This initial effect is likely because of rifampin inhibition of OATPs that is offset by the enzyme induction of rifampin following several days of coadministration.

 Clarithromycin has been reported to increase the AUC of ambrisentan by 41%, similar to the effect of ketoconazole.

 Other inhibitors of CYP3A4 would be expected to produce modest increases in ambrisentan plasma concentrations.

Bosentan is metabolized by CYP3A4, and CYP2C9, and, to a lesser extent, CYP2C19. It is a potent inducer of CYP enzyme activity, including CYP3A4, CYP2C9, and CYP2C19. Thus, bosentan causes autoinduction of its metabolism. With chronic dosing, the AUC of bosentan will be reduced by up to 65% of the value observed following a single dose. The CYP3A4 inhibitors cyclosporine and ketoconazole produce about a doubling of bosentan AUC.

 Rifampin is known to induce CYP3A4 and CYP2C9 but acutely can inhibit OATPs. As was observed with ambrisentan, rifampin produced an 6-fold increase in the AUC of bosentan, but with steady-state rifampin dosing, the net effect was nearly a 60% reduction in bosentan AUC.

 When a combination of lopinavir and ritonavir was coadministered with bosentan, the steady-state AUC of bosentan increased more than 400%.

 Bosentan will likely reduce the plasma concentration of other drugs that are metabolized by CYP2C9 or CYP3A4.

Inhibitors of CYP3A4 ketoconazole and cyclosporine increased macitentan’s AUC by 2.3-fold and 10%, respectively.

 The administration of the CYP3A4 inducer rifampin led to an 80% decrease in macitentan concentrations.

 Macitentan may be affected similarly by other CYP3A4 inhibitors and inducers.

Drug interaction data are limited for these agents. However, based on the pharmacology of the endothelin receptor inhibitors, it is possible to predict how these agents will interact with other drugs. The relatively greater dependence on CYP3A4 for elimination by bosentan and macitentan compared with ambrisentan appears to make ambrisentan less likely to result in clinically significant interactions with CYP3A4 inhibitors or inducers. UGT inhibitors, such as gemfibrozil (Lopid), may increase ambrisentan concentrations. Pending data, monitor patients for an alerted ambrisentan response. Similarly, patients receiving inhibitors of CYP2C9 (eg, amiodarone, cimetidine, fluconazole, sulfamethoxazole, and voriconazole) should have their response to bosentan monitored. The enzyme induction response to bosentan will likely reduce the response to many other drugs.

John R. Horn, PharmD, FCCP, and Philip D. Hansten, PharmD, are both professors of pharmacy at the University of Washington School of Pharmacy in Seattle.  

Letairis [prescribing information]. Foster City, CA: Gilead Sciences, Inc; 2015. gilead.com/~/media/Files/pdfs/medicines/cardiovascular/letairis/letairis_pi.pdf. Accessed May 10, 2018.

Tracleer [prescribing information]. San Francisco, CA: Actelion Pharmaceuticals US, Inc; 2017. tracleer.com/assets/PDFs/Tracleer_Full_Prescribing_Information.pdf. Accessed May 10, 2018.

Opsumit [prescribing information]. San Francisco, CA: Actelion Pharmaceuticals US, Inc; 2017. opsumit.com/opsumit-prescribing-information.pdf. Accessed May 10, 2018.

Spence R, Mandagere A, Richards DB, Magee MH, Dufton C, Boinpally R. Potential for pharmacokinetic interaction between ambrisentan and cyclosporine. 

 2010;88(4):513-520. doi: 10.1038/clpt.2010.120.

Richards DB, Walker GA, Mandagere A, Magee MH, Henderson LS. Effect of ketoconazole on the pharmacokinetic profile of ambrisentan. 

 2009;49(6):719-724. doi: 10.1177/0091270009335870.

Harrison B, Magee MH, Mandagere A, et al. Effects of rifampicin (rifampin) on the pharmacokinetics and safety of ambrisentan in healthy subjects: a single-sequence, open-label study. 

 2010;30(12):875-885. doi: 10.2165/11539110- 000000000-00000.

Market C, Hellwig R, Burhenne J, et al. Interaction of ambrisentan with clarithromycin and its modulation by polymorphic SLCO1B1. 

 2013;69(10):1785- 1793. doi: 10.1007/s00228-013-1529-1.

Binet I, Wallnofer A, Weber C, Jones R, Thiel G. Renal hemodynamics and pharmacokinetics of bosentan with and without cyclosporine A. 

 2000;57(1):224-231. doi: 10.1046/j.1523-1755.2000.00838.x.

van Giersbergen PL, Halabi A, Dingemanse J. Single- and multiple-dose pharmacokinetics of bosentan and its interaction with ketoconazole. 

van Giersbergen PL, Treiber A, Schneiter R, Dietrich H, Dingemanse J. Inhibitory and inductive effects of rifampin on the pharmacokinetics of bosentan in healthy subjects [published erratum appears in Clin Pharmacol Thera. 2007;81(3):460]. 

 2007;81(3):414-419. doi: 10.1038/sj.clpt.6100075.

Dingemanse J, Van Giersbergen PL, Patat A, Nilsson PN. Mutual pharmacokinetic interactions between bosentan and lopinavir/ritonavir in healthy participants. 

 2010;15(2):157-163. doi: 10.3851/IMP1506.

Atsmon J, Dingemanse J, Shaikevich D, Volokhov I, Sidharta PN. Investigation of the effects of ketoconazole on the pharmacokinetics of macitentan, a novel dual endothelin receptor antagonist, in healthy subjects. 

 2013;52(8):685- 692. doi: 10.1007/s40262-013-0063-8.

Bruderer S, Aänismaa P, Homery MC, et al. Effect of cyclosporine and rifampin on the pharmacokinetics of macitentan, a tissue-targeting dual endothelin receptor antagonist. 

 2012;14(1):68-78. doi: 10.1208/s12248-011-9316-3.

Endothelin antagonists&mdash;ambrisentan (Letairis), bosentan (Tracleer), and macitentan (Opsumit)&mdash;are indicated to treat pulmonary arterial hypertension. Although these drugs have similar pharmacologic activity, they differ in their potential for their plasma concentrations to be altered by other medications.

Endothelin Receptor Antagonist Drug Interactions

In 2006, the FDA issued a warning of possible serotonin syndrome in patients receiving triptans together with either selective serotonin/norepinephrine reuptake inhibitors (SNRIs) or selective serotonin reuptake inhibitors (SSRIs).

 Despite periodic reports providing evidence that the combination is not dangerous, the warning persists in the triptan labeling. Given that about 1/4 of patients on triptans receive SNRIs or SSRIs, it probably contributes substantially to “alert fatigue” for pharmacists. New evidence has appeared regarding the clinical importance of this interaction.

Isolated cases of serotonin syndrome have been reported in patients using triptans and SNRIs or SSRIs,

 and the FDA apparently has received additional reports. It was not clear, however, that all the patients actually had serotonin syndrome, and it is possible that the reactions were caused by the SNRI or SSRI alone or by some other serotonergic drug given concurrently with it. The evidence in support of serotonin syndrome from triptans plus SNRIs or SSRIs, therefore, is meager.

A 14-year electronic health records study looked at data from 19,017 patients who had concurrently received a triptan and an SNRI or SSRI.

 Just 2 patients were classified as having definite serotonin syndrome. These data are consistent with previous clinical and pharmacokinetic studies that failed to observe serotonin syndrome in more than 2000 patients receiving triptans and SSRIs.

 Although we cannot rule out the possibility that isolated patients on the combination may develop serotonin syndrome, the risk appears to be extraordinarily low. Indeed, the risk may be similar to the extremely small risk of sero- tonin syndrome from the use of SSRIs alone.

In addition, from a theoretical standpoint, triptans should not interact with SNRIs or SSRIs to produce serotonin syndrome. The condition is thought to be mediated by 5HT2A receptors and possibly 5HT1A, yet the affinity of triptans is primarily for 5HT1B and 5HT1D receptors. Pharmacokinetic interactions are possible because frovatriptan and zolmitriptan are metabolized by CYP1A2, and fluvoxamine is a potent inhibitor of CYP1A2. Although this would not result in serotonin toxicity, it would be prudent to avoid these triptans in patients receiving fluvoxamine.

Given the compelling evidence for an extremely low risk of serotonin syndrome in patients using triptans with SNRIs or SSRIs, it does not appear clinically necessary to warn prescribers about these interactions. Nonetheless, as long as the warnings remain on the triptan labels, it presents a dilemma for pharmacists. If, for example, the patient on a triptan and SSRI took some third drug with serotonergic effects, such as tramadol, the interaction between the SSRI and the tramadol might produce actual serotonin syndrome, and the triptan-SSRI combination might be falsely blamed for the adverse effect. An alternative to ignoring the interaction might be to advise the patient that the interaction is mentioned on the label but not generally considered to be clinically important. The patient also could be advised to be alert for symptoms of serotonin syndrome, especially the characteristic neuromuscular effects such as myoclonus, rigidity, and tremors. That would be useful even for patients on SNRIs or SSRIs because the syndrome occasionally occurs with the addition of other serotonergic agents.

Although isolated cases of serotonin syndrome have been reported with combined use of triptans with SNRIs or SSRIs, it is not clear that these cases represent an actual drug interaction.

Epidemiologic and pharmacokinetic studies suggest that the risk of serotonin syndrome from combining triptans with SNRIs or SSRIs is extraordinarily low and may be no higher than the use of SNRIs or SSRIs alone.

Given that the triptan labels warn of possible serotonin syndrome if SNRIs or SSRIs are used concurrently, there may be medicolegal considerations if the patient develops serotonin syndrome for whatever reason. Warning the patient about the signs and symptoms of serotonin syndrome is useful.

John R. Horn, PharmD, FCCP, and Philip D. Hansten, PharmD, are both professors of pharmacy at the University of Washington School of Pharmacy in Seattle. For an electronic version of this article, including the online table, visit hanstenandhorn.com.

FDA. Public health advisory - combined use of 5-hydroxytryptamine receptor agonists (triptans), selective serotonin reuptake inhibitors (SSRIs) or selective serotonin/norepinephrine reuptake inhibitors (SNRIs) may result in life-threatening serotonin syndrome. FDA website. wayback.archive- it.org/7993/20170406044820/https://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSaf etyInformationforPatientsandProviders/ucm124349.htm Updated August 16, 2013. Accessed April 11, 2018.

Hendrix Y, van Zagten MS. Serotonin syndrome as a result of concomitant use of paroxetine and sumatriptan. Ned Tijdschr Geneeskd. 2005;149(16):888-890.

Mathew NT, Tietjen GE, Lucker C. Serotonin syndrome complicating migraine pharmacotherapy. Cephalalgia. 1996;16(5):323-327. doi: 10.1046/j.1468- 2982.1996.1605323.x.

Gardner DM, Lynd LD. Sumatriptan contraindications and the serotonin syndrome. Ann Pharmacother. 1998;32(1):33-38. doi: 10.1345/aph.17204.

Orlova Y, Rizzoli P, Loder E. Association of coprescription of triptan antimigraine drugs and selective reuptake inhibitor or selective norepinephrine reuptake inhibitor antidepressants with serotonin syndrome [published online February 26, 2018]. JAMA Neurol. 2018. doi:10.1001/jamaneurol.2017.5144. 4

Tepper SJ, Millson D. Safety profile of the triptans. Expert Opin Drug Saf. 2003;2(2):123-132. doi: 10.1517/14740338.2.2.123.

Putnam GP, O’Quinn S, Bolden-Watson CP, Davis RL, Gutterman DL, Fox AW. Migraine polypharmacy and the tolerability of sumatriptan: a large-scale prospective study. Cephalalgia. 1999;19(7):668-675. doi: 10.1046/j.1468-2982.1999.019007668.x.

Mathew NT, Hettiarachchi J, Alderman J. Tolerability and safety of eletriptan in the treatment of migraine: a comprehensive review. Headache. 2003;43(9):962-974.

Smith DA, Cleary EW, Watkins S, Huffman CS, Polvino WJ. Zolmitriptan (311C90) does not interact with fluoxetine in healthy volunteers. Int J Clin Pharmacol Ther. 1998;36(6):301-305.

Goldberg MR, Lowry RC, Musson DG, et al. Lack of pharmacokinetic and pharmacodynamic interaction between rizatriptan and paroxetine. J Clin Pharmacol. 1999;39(2):192-199.

Isolated cases of serotonin syndrome have been reported in patients using triptans and SNRIs or SSRIs.