Linezolid and Serotonergic Drug Interactions: A Retrospective Survey
- 1Division of Infectious Diseases, Rochester, Minnesota
- 2Department of Pharmacy, Mayo Clinic, Rochester, Minnesota
- Reprints or correspondence: Jeremy J. Taylor, Methodist University Hospital, Pharmacy Dept., 1265 Union Ave., Memphis, TN 38104 (taylorj{at}methodisthealth.org).
Abstract
Background. Linezolid is a reversible, nonselective monoamine oxidase inhibitor. There are currently 11 published case reports of serotonin syndrome being associated with linezolid and selective serotonin reuptake inhibitors (SSRIs). Controversy exists regarding whether linezolid and SSRIs can be given concomitantly. The purpose of this study was to report the incidence of serotonin syndrome in patients receiving linezolid and SSRIs.
Methods. This study was a retrospective chart review of inpatients at the Mayo Clinic (Rochester, MN) with concomitant orders or therapy within 14 days for linezolid and an SSRI from 2000 to 2004. The Sternbach criteria and Boyer criteria for diagnosis of serotonin syndrome were used to identify clinical features of serotonin syndrome.
Results. Seventy-two patients received linezolid and an SSRI or venlafaxine within 14 days of each other. Fifty-two patients (72%) received concomitant therapy with linezolid and an SSRI or venlafaxine, and 20 patients (28%) did not receive concomitant therapy but received linezolid and an SSRI within a 14-day period. Overall, only 2 patients (3%) had a high probability of serotonin syndrome. In both patients with high probability, symptoms reversed rapidly on discontinuation of serotonergic therapy. The Boyer criteria were much more specific than the Sternbach criteria for identification of serotonin syndrome.
Conclusions. On the basis of our experience, we suggest that, if the clinical situation warrants use of linezolid in a patient receiving an SSRI, linezolid may be used concomitantly with SSRIs, without a 14-day washout period and with careful monitoring for signs and symptoms of serotonin syndrome. Serotonergic agents should be promptly discontinued if serotonin syndrome is suspected.
Linezolid (Zyvox; Pfizer), the first antibiotic in the oxazolidinone class, received approval from the US Food and Drug Administration in April 2000. It is used for the treatment of infections caused by drug-resistant, gram-positive organisms, including methicillin-resistant Staphylococcus aureus, multidrug-resistant Streptococcus pneumoniae, and vancomycin-resistant Enterococcus species, as well as susceptible gram-positive infections in patients with an allergy to or intolerance for preferred agents. Because it is a reversible, relatively weak nonselective inhibitor of monoamine oxidase, it has a potential for interaction with adrenergic and serotonergic agents. With expanding indications and options for both intravenous and oral therapy, it is likely to be prescribed with increasing frequency.
The package insert for linezolid contains the following wording regarding the potential for drug-drug interactions with other serotonergic agents: “Co-administration of linezolid and serotonergic agents was not associated with serotonin syndrome in Phase 1, 2 or 3 studies. Spontaneous reports of serotonin syndrome associated with co-administration of ZYVOX and serotonergic agents, including antidepressants such as selective serotonin reuptake inhibitors (SSRIs), have been reported” [1, p. 15]. Elsewhere, the insert says, “Where administration of ZYVOX and concomitant serotonergic agents is clinically appropriate, patients should be closely observed for signs and symptoms of serotonin syndrome such as cognitive dysfunction, hyperpyrexia, hyperreflexia, and incoordination. If signs or symptoms occur physicians should consider discontinuation of either one or both agents” [1, p. 13]. Since the introduction of linezolid, 11 reported cases of linezolid and serotonergic interactions with SSRIs or venlafaxine potentially resulting in serotonin syndrome have appeared in the literature. These cases have been summarized in table 1. This has led to strong wording in Micromedex concerning the potential for serotonergic interactions. Regarding the potential interaction with venlafaxine, an antidepressant with strong serotonin reuptake inhibition properties, Micromedex suggests that this combination is a clear contraindication: “Concurrent use of linezolid and venlafaxine is contraindicated. Wait at least 14 days after discontinuing linezolid before initiating therapy with venlafaxine. Wait at least 14 days after discontinuing venlafaxine before initiating therapy with linezolid” [11].
Figure 1
Summary of the patient chart review. SSRI, selective serotonin reuptake inhibitor.
Table 1
Review of cases of linezolid and serotonergic interactions with selective serotonin reuptake inhibitors (SSRIs) or venlafaxine that potentially resulted in serotonin syndrome.
In addition, although it is not explicitly contraindicated, Micromedex strongly cautions against concurrent use of linezolid and SSRIs: “Concurrent use of linezolid and citalopram should be approached with caution. Wait at least 14 days after discontinuing linezolid before initiating therapy with citalopram. Wait at least 14 days after discontinuing citalopram before initiating therapy with linezolid” [11].
Serotonin syndrome is a progressive syndrome consisting of a clinical triad of symptoms, including mental status changes, autonomic hyperactivity, and neuromuscular abnormalities. Diagnosis is made on the basis of clinical manifestations and a history of drug exposure to a known medication with serotonergic properties, in temporal relationship with the clinical features. Signs and symptoms of serotonin syndrome may appear anywhere from 1 h to several days after exposure to serotonergic agents. There are 6 mechanisms that may contribute to the development of serotonin syndrome. These include increased formation of serotonin, increased release of serotonin, inhibition of serotonin metabolism, decreased reuptake of serotonin, direct serotonin receptor agonism, and increased postsynaptic response. Any combination of these mechanisms may potentially increase the risk for serotonin syndrome. The serotonergic properties of linezolid inhibit serotonin metabolism, and SSRIs and venlafaxine prevent reuptake of serotonin into the presynaptic neuron. A list of medications with serotonergic properties can be found in table 2.
Table 2
Medications with serotonergic properties.
In 1991, Sternbach [16] published the first set of criteria to assist in the diagnosis of serotonin syndrome. These criteria include the following:
Presence of at least 3 clinical features coincident with the addition of or an increase in a known serotonergic agent:
Mental status changes
Agitation
Myoclonus
Hyperreflexia
Diaphoresis
Shivering
Tremor
Diarrhea
Incoordination
Fever
Other etiologies that are ruled out:
Infectious etiologies
Metabolic etiologies
Substance abuse or withdrawal
A neuroleptic agent not started or increased in dosage prior to onset of symptoms
The clinical features included in the Sternbach criteria are nonspecific and can be found in many different conditions, including infection. When using the Sternbach criteria, it is quite difficult to exclude other etiologies of these symptoms, and it makes confirming the diagnosis of serotonin syndrome in the face of infection quite difficult.
More recently, Boyer and Shannon [12] published an algorithm for the diagnosis of serotonin syndrome that includes more specific clusters of clinical features, as follows:
Tremor and hyperreflexia
Spontaneous clonus
Muscle rigidity, a body temperature >38°C, and either ocular clonus or inducible clonus
Ocular clonus and either agitation or diaphoresis
Inducible clonus and either agitation or diaphoresis
Boyer and Shannon stress that serotonin syndrome is a progressive continuum of clinical features. They suggest that myoclonus may be the single-most important clinical feature for the diagnosis of serotonin syndrome. In addition, they also state that early signs and symptoms of serotonin syndrome may be missed when waiting for the appearance of 3 clinical features to meet criteria for serotonin syndrome. Dunkley et. al. [17] published the Hunter Serotonin Toxicity Criteria in 2003, which resulted from a retrospective analysis of prospectively collected data on patients admitted to the Hunter Area Toxicology Service following an overdose of a serotonergic drug. Data was collected with the intent of improving criteria for the diagnosis of serotonin syndrome over the nonspecific clinical features used by the Sternbach criteria. The results found clonus, agitation, diaphoresis, tremor, and hyperreflexia as the clinical features necessary for a more accurate diagnosis of serotonin syndrome. Boyer and Shannon [12] referenced these clinical features to establish support for the development of their algorithm. Because of the lack of information concerning the incidence of serotonin syndrome in the linezolid package insert and from the manufacturer's premarketing studies, the contraindication listed in Micromedex, and the appearance of 11 postmarketing case reports of probable serotonin syndrome, the significance and frequency of serotonin syndrome associated with linezolid and concomitant therapy with SSRIs or other serotonergic medications is unclear. Additionally, there are concerns regarding the abrupt discontinuation of SSRI therapy in attempt to avoid this interaction and regarding the use of alternative drugs, such as daptomycin, for which few data exist regarding its use against severe vancomycin-resistant Enterococcus infections. The objective of this study is to determine the frequency, significance, and risk factors for serotonergic drug interactions between linezolid and SSRIs or venlafaxine in inpatients at the Mayo Clinic (Rochester, MN).
Methods
This study is a retrospective chart review of inpatients at the Mayo Clinic with orders for linezolid and an SSRI or venlafaxine during the same hospitalization during April 2000–November 2004. After receiving approval from the institutional review board, patients were identified using a report generated from the Mayo Clinic's Computer-Based Antimicrobial Monitoring Program. Chart histories and electronic records were reviewed for 79 patients. Patients included in the study must have received linezolid and an SSRI or venlafaxine either concomitantly or within 14 days of each other. The Sternbach criteria and Boyer algorithm for diagnosis of serotonin syndrome were used to identify clinical features of serotonin syndrome. Patients were categorized as having or not having a documented diagnosis of serotonin syndrome. Patients without a documented diagnosis were further subdivided according to their likelihood of serotonin syndrome based on documented clinical criteria. Other data collected included age, sex, allergies, indication for linezolid, receipt of an SSRI, receipt of other serotonergic medications, concurrent medications, medication history, new or active medical conditions during hospitalization, and past medical history.
Criteria were established to assess the likelihood of serotonin syndrome for patients in this study on the basis of documented clinical criteria. Patients were ruled out for the diagnosis of serotonin syndrome if they did not meet at least 3 Sternbach clinical criteria or any of the Boyer criteria for serotonin syndrome or if they met clinical criteria with clear alternative explanations. Patients were considered to have a low probability of serotonin syndrome if they met at least 3 Sternbach clinical criteria or any of the Boyer criteria for serotonin syndrome and if their symptoms did not progress with continued therapy or if concomitant therapy was stopped but symptoms did not resolve. Patients were considered to have a high probability of serotonin syndrome if they met ⩾3 Sternbach clinical criteria or any of the Boyer criteria and no clear alternative explanation existed or symptoms reversed with discontinuation of therapy.
Results
Seventy-nine patients received linezolid and an SSRI or venlafaxine during the same hospitalization. Seven patients did not receive both types of therapy within a 14-day period and were excluded from the study. Seventy-two patients met inclusion criteria. Fifty-two patients (72%) received concomitant therapy with linezolid and an SSRI or venlafaxine. Twenty patients (28%) did not receive concomitant therapy but did receive linezolid and an SSRI within 14 days. One patient (1%) had a documented diagnosis of serotonin syndrome. Results are summarized in figure 1.
When considering the likelihood of serotonin syndrome for patients who met inclusion criteria, 68 patients (94%) were ruled out for the diagnosis of serotonin syndrome. Sixty-four of these patients did not meet at 3 of the Sternbach clinical criteria or any of the Boyer criteria for diagnosis of serotonin syndrome. Four patients met the at least 3 of the Sternbach clinical criteria for serotonin syndrome but had clear alternative explanations for symptoms; thus, they were considered to be ruled out. These explanations included increased anxiety related to severe pain, multiple medical problems with multiple surgical procedures and symptoms that did not present together, severe sepsis with a significant drop of hemoglobin and multiple blood transfusions, and mental status changes attributed to increased narcotic doses of fentanyl.
Two patients (3%) were classified as having low probability of serotonin syndrome. The first of these 2 patients did not receive concomitant therapy but did receive fluoxetine within 14 days of linezolid therapy. This patient did not receive any other serotonergic medications. Three of the Sternbach clinical criteria were met, including mental status changes, agitation, and fever, but none of the Boyer criteria were met. The second patient received concomitant therapy with sertraline. This patient also received amitriptyline, which also has serotonergic properties. Three Sternbach clinical criteria were met, including mental status changes, myoclonus, and diarrhea. This patient also met Boyer criteria with spontaneous clonus. In both of these patients, therapy was continued, but the signs and symptoms did not progress, which makes serotonin syndrome unlikely.
Two patients (3%) met criteria for having high probability of serotonin syndrome. The first of these patients was a 30-year-old woman who received linezolid for a pancreatic pseudocyst infected with vancomycin-resistant Enterococcus species. She had a diagnosis of serotonin syndrome documented in her chart. This patient received concomitant therapy with linezolid and sertraline. The patient also received trazodone and fentanyl, which also have serotonergic properties. Four Sternbach clinical criteria were met, including agitation (anxiety), myoclonus, diaphoresis, and shivering. In addition, 2 Boyer criteria were met, including spontaneous clonus and inducible clonus with either agitation or diaphoresis. When symptoms were first recognized, this patient had complaints of restlessness, anxiety, and alternating hot and cold sweats. The patient had a body temperature of 33.9°C and was in mild distress, with blood pressure of 173/105 mm Hg. In a progress note from later that same day, the patient had complaints of anxiousness and myoclonic jerks. Sertraline was discontinued, but linezolid was continued, and the patient was given lorazepam. The diagnosis was questionable mild serotonin syndrome versus β-blocker withdrawal. By the next day, the symptoms resolved, and there was no further documentation of any signs and symptoms of serotonin syndrome. The second patient with high probability of serotonin syndrome did not have a diagnosis of serotonin syndrome documented in her chart. The patient was an 81-year-old woman who received linezolid for a urinary tract infection caused by vancomycin-resistant Enterococcus species. She received concomitant therapy with linezolid and venlafaxine, followed by citalopram. Six Sternbach clinical criteria were met, including mental status changes, agitation, myoclonus, hyperreflexia, tremor, and incoordination. In addition, 4 Boyer criteria were met, including tremor and hyperreflexia, spontaneous clonus, muscle rigidity with a body temperature >38°C and ocular or inducible clonus, and ocular clonus with either agitation or diaphoresis. Initially, the patient refused to eat, became confused with respect to time and place, and started shouting. The next day, the patient became unarousable, only mumbling a few words. A head CT scan was obtained, which demonstrated no significant changes. A nursing note documented the patient as being lethargic, with leg and arm twitching and jerking, eyes rolling back in her head, and labored respirations. Vital signs included a blood pressure of 180/110 mm Hg, a heart rate of 120 beats/min, and respiratory a rate of 50 breaths/min. The patient developed respiratory distress, and naloxone was given for potential fentanyl-induced respiratory depression. Mental status improved, but the patient's respiratory status did not. The patient was then intubated and sedated. An electroencephalogram was obtained, which showed no significant changes from baseline. Linezolid was discontinued at this time to simplify the patient's drug regimen. However, no mention was made of possible serotonin syndrome. According to a neurology note from the following day, the patient had open eyes with no spontaneous movements but withdrew from stimulus and nodded to questions. Five days after the onset of symptoms and 2 days after discontinuing linezolid, the patient was extubated, had a baseline mental status, and was able to communicate and follow commands.
Discussion
Four patients (6%) met clinical criteria for serotonin syndrome. However, only 2 (3%) of these patients had high-probability cases, whereas the other 2 (3%) had low-probability cases. Three of 4 patients who met the clinical criteria for serotonin syndrome, including both high-probability patients, received concomitant therapy with an SSRI (1 received citalopram, and 2 received sertraline). The fourth low-probability patient meeting the clinical criteria did not receive concomitant therapy but did receive fluoxetine within 14 days of discontinuation of linezolid therapy. Sixty-five (90%) of the patients included in the study also received at least 1 other serotonergic agent, which provides the potential for enhanced serotonergic effects. This included 3 of 4 patients who met clinical criteria for serotonin syndrome.
In both patients who were classified as having low probability of serotonin syndrome, concomitant therapy with linezolid and an SSRI was continued, despite the presence of clinical features. However, the signs and symptoms of serotonin syndrome did not progress, even with continued therapy, making the diagnosis of serotonin syndrome unlikely. Both patients with high probability met at least 3 of the Sternbach clinical criteria and at least 1 of the Boyer criteria. It is important to note that once concomitant therapy was discontinued, the signs and symptoms of serotonin syndrome resolved within 24–48 h.
The Sternbach criteria appear to be overly sensitive in patients being treated for infection. Eight patients had at least 3 of the clinical features, according to the Sternbach criteria, but 4 of these patients had clear alternative explanations for their signs and symptoms; thus, they were ruled out for serotonin syndrome. Additionally, 2 patients meeting the clinical features had a low probability of serotonin syndrome, and only 2 of the 8 patients (25%) meeting the Sternbach criteria had a high probability of serotonin syndrome. Conversely, only 1 patient in the low-probability group met any of the Boyer criteria, whereas both patients in the high-probability group met Boyer criteria for serotonin syndrome. The low-probability serotonin syndrome patient meeting the Boyer criteria did not have progression of symptoms with continued concomitant therapy. These findings suggest that applying the Boyer algorithm, which focuses on myoclonus, may be more specific for the diagnosis of serotonin syndrome, compared with the Sternbach criteria.
Because of the low overall incidence of serotonin syndrome in our patient population, it was not feasible to compare the incidence of serotonin syndrome by specific SSRI or venlafaxine. Additionally, risk factors for the development of serotonin syndrome could not be assessed with a small number of patients. It should be noted, however, that both patients with high probability of serotonin syndrome were receiving concomitant therapy with linezolid and an SSRI. Both patients were receiving at least 1 other serotonergic agent, which may increase the potential for signs and symptoms of serotonin syndrome.
Limitations of this study include its retrospective nature. Only documented signs and symptoms could be used to determine the likelihood of serotonin syndrome in these patients. If the physician was unaware of the signs and symptoms of serotonin syndrome, these clinical features may not have been documented in the chart. There was also considerable variability in documentation among physicians with respect to patient progress notes. Another limitation to the study includes concurrent medications that could mask, mimic, or enhance the signs and symptoms of serotonin syndrome. Sixty-five of the 72 patients in this study received a serotonergic medication other than linezolid or an SSRI. This may have increased the risk for the signs and symptoms of serotonin syndrome. Other medications may have mimicked clinical features without actually contributing to serotonin syndrome. In addition, patients who were sedated and in an intensive care unit may not have demonstrated recognizable signs and symptoms of serotonin syndrome.
Conclusions
In our study, only 2 patients with high probability of serotonin syndrome were identified among 52 patients (4%) receiving concomitant therapy with venlafaxine or an SSRI, and no patients with high probability were identified among those receiving therapy with linezolid and an SSRI plus venlafaxine within 14 days after hospitalization. The total incidence of high-probability serotonin syndrome among patients receiving either concomitant SSRI-venlafaxine therapy or therapy within 14 days was 2.8%. Resolution promptly occurred within 24–48 h on discontinuation of concomitant therapy in patients who developed signs and symptoms of serotonin syndrome.
When using the Boyer criteria, the onset of serotonin syndrome can be clearly recognized if physicians are familiar with the signs and symptoms (i.e., the presence of myoclonus, inducible clonus, tremor, and hyperreflexia, all of which are not typical of infection or sepsis). Serotonin syndrome with a linezolid plus either an SSRI or venlafaxine is a progressive process, which is another recognizable feature. When recognized, however, serotonin syndrome is readily reversible on discontinuation or a decrease of the dosage of serotonergic agents. On the basis of our experience, we suggest that, if the clinical situation warrants use of linezolid in a patient receiving an SSRI, linezolid may be used concomitantly with SSRIs, without a 14-day washout period, with careful monitoring for signs and symptoms of serotonin syndrome. Use caution when administering linezolid and SSRIs concomitantly in situations in which they may mask (e.g., paralytics) or enhance (e.g., other serotonergic agents) the signs and symptoms of serotonin syndrome. Other serotonergic agents should be avoided, if doing so is feasible.
Acknowledgments
Potential conflicts of interest. All authors: no conflicts.
- Received December 9, 2005.
- Accepted April 13, 2006.
- © 2006 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
-
- Articles
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
