- Split View
-
Views
-
Cite
Cite
José Antonio Morales-Molina, Javier Mateu-de Antonio, Mónica Marín-Casino, Santiago Grau, Linezolid-associated serotonin syndrome: what we can learn from cases reported so far, Journal of Antimicrobial Chemotherapy, Volume 56, Issue 6, December 2005, Pages 1176–1178, https://doi.org/10.1093/jac/dki368
- Share Icon Share
Abstract
Objectives: To study the characteristics of the linezolid-associated serotonin syndrome cases.
Methods: Database search for linezolid-associated serotonin syndrome.
Results: Twelve cases were found. The mean age of patients was 52.8 years. All patients received linezolid concomitantly with selective serotonin re-uptake inhibitor drugs (SSRID). The onset of syndrome was 9.5 days after linezolid introduction and was directly correlated to patients' age (P = 0.024). The symptoms resolved in 2.9 days. Citalopram was associated with a delayed resolution (P = 0.018). A trend was observed towards longer resolution time the longer the half-life of the interacting drug (P = 0.096).
Conclusions: Patients are at risk when receiving SSRID concomitant with linezolid. The syndrome onset could be delayed in older patients. The resolution could be delayed when citalopram is involved in the syndrome.
Introduction
Linezolid is an oxazolidinone antibiotic with activity against Gram-positive microorganisms, including vancomycin-resistant Enterococcus (VRE), methicillin-resistant Staphylococcus aureus (MRSA) and penicillin-resistant Streptococcus pneumoniae.1 It is indicated for the treatment of nosocomial and community-acquired pneumonia and complicated skin and soft tissue infections without concomitant osteomyelitis. It has additionally a weak, reversible, competitive and non-specific monoamine oxidase (MAO) inhibitor activity.1 Serotonin is implicated in psychiatric and non-psychiatric events and it has been associated with different pathophysiological symptoms. Serotonin syndrome (SS) is a potentially life-threatening adverse drug reaction that results from therapeutic drug use, intentional self-poisoning or inadvertent interaction between serotonergic agents and MAO inhibitors.2 It leads to an increase in serotonin neurotransmission in the CNS and peripheral serotonergic receptors. SS can manifest as a wide spectrum of clinical symptoms, from mild to fatal cases. Often, this syndrome is underdiagnosed as physicians are unaware of its clinical diagnosis. A detailed review of this syndrome has been published recently elsewhere.2 To date, several individual cases of linezolid-associated SS have been reported, but no case series review has been published.
We studied the relationship between patient demographics, onset and resolution of linezolid-associated SS, and involved drug characteristics.
Methods
We reviewed MEDLINE, EMBASE and IDIS databases for reported cases of linezolid-associated SS. We employed two strategies for database search: a Medical Subject Headings (MeSH) search and a free search without restrictions except for papers related to humans. The last search was performed on 29 July 2005. Initially no papers were excluded. In the second instance, case reports where diagnosis of linezolid-associated SS was not clearly stated were discarded. In the remaining reports, we recorded demographics, comorbidity factors, parameters related to the pharmacological treatment, such as concomitant drugs, time of symptom onset and time to resolution of SS. Statistical tests performed were the Spearman test for correlating quantitative non-parametric variables and the χ2 test or Fisher's exact test, when appropriate, for studying dichotomous non-parametric variables. Values of P < 0.05 were considered significant.
Results
Initially, we found 11 references and an additional article in press (which has now been published). Two references were discarded. In the first of these cases, authors did not relate the SS to linezolid3 and in the second case, authors stated that patient's symptoms did not meet the criteria for SS.4 The rest of the selected papers included 12 cases of linezolid-associated SS.5–14 The mean age of patients was 52.8 years (range 4–85). Studied patients and involved drugs characteristics are summarized in Table 1. Linezolid was prescribed for treatment of pulmonary (4 patients, 33%), skin–soft tissue (3 patients, 25%), osteomyelitis (2 patients, 17%), prosthesis (1 patient, 8%), cholecystopancreatitis (1 patient, 8%) and not-reported infections (1 patient, 8%). Microbiological isolates were MRSA (6), VRE (3), methicillin-susceptible Staphylococcus aureus (2), methicillin-resistant Staphylococcus epidermidis (1) and a non-reported microorganism (1). In three cases (25%), linezolid prescription was for osteomyelitis and prosthesis infections that are not labelled indications. Patients received linezolid mainly by intravenous route, although three patients (25%) received it also by oral route. Selective serotonin re-uptake inhibitor drugs (SSRID) were prescribed for the following psychiatric diagnoses; depression (12 patients, 100%); bipolar disease (2 patients, 17%), schizoaffective disorder (1 patient, 8%) and essential tremor (1 patient, 8%). Patients also had several comorbidity factors, mainly diabetes, high blood pressure, leukaemia and mechanical ventilation. The mean number of comorbidity factors per patient was 2.0 (range 0–3). All patients received SSRID before linezolid introduction. Ten patients (83%) continued receiving the SSRID during linezolid therapy. Two patients (17%) discontinued SSRID before linezolid introduction, one who received the last dose of SSRID 3 days before linezolid6 and the other patient, 18 days before linezolid.13 Duration of SSRID use was reported in five cases (42%) and ranged from 11 days to 1 year, prior to the beginning of linezolid treatment. The onset of SS was 9.5 days after linezolid prescription (range 0.5–21). This was not correlated to the number of comorbidities or concomitant serotonergic drugs, but was directly correlated with increasing age (Spearman's rho = 0.671; P = 0.024). When the paediatric patient was excluded, the correlation remained without substantial changes (Spearman's rho = 0.689; P = 0.028). After SS onset, three patients (25%) discontinued linezolid, four patients (33%) discontinued linezolid and also the SSRID, three patients (25%) discontinued only the SSRID, one patient (8%) reduced the SSRID dose and in one case (8%) no drug discontinuation was reported.8 The symptoms resolved in 2.9 days (range 1–9). The resolution time was not correlated with any variable. However, a trend was observed towards longer resolution time the longer the half-life of the interacting drug (Spearman's rho = 0.555; P = 0.096). Considering the drugs involved separately, citalopram was associated with a delayed resolution of more than 2 days, of the SS (Fisher's exact test; P = 0.018). Three patients (25%) received cyproheptadine as part of SS treatment. Cyproheptadine use was not associated with changes in duration of SS resolution (χ2 test; P = 0.632). Gender did not influence any variable. Two patients died, however, authors did not directly relate these deaths to the SS.7,8
Age (years)/gender . | Infection . | Selective serotonin re-uptake inhibitor drug, dose (half-life)15 . | Other serotonergic drugs, dose (half-life)15 . | Symptoms onset after introducing linezolid . | Resolution time . |
---|---|---|---|---|---|
45/M5 | skin–soft tissue | sertraline, 200 mg/day (23 h) | trazodone, 50 mg/day (6 h) | 10 days | 2 days |
56/F6 | skin–soft tissue | paroxetine, NR (17 h) | — | 1 day | 2 days |
81/M7 | osteomyelitis | citalopram, 40 mg/day (33 h) | — | 21 days | NR died |
56/F8 | pulmonary | citalopram, 20 mg/day (33 h) | — | 13 days | 9 days died |
36/M8 | pulmonary | sertraline, 50 mg/day (23 h) | — | 5 days | 1 day |
85/M9 | prosthesis | venlafaxine, 150 mg/day (5 h) | — | 20 days | 2 days |
4/F10 | NR | fluoxetine, 5 mg/day (NR in children) | fentanyl, 200 μg/day (3.7 h) | 2 days | 2 days |
85/F11 | osteomyelitis | citalopram, NR (33 h) | — | NR | 3 days |
38/F12 | pulmonary | venlafaxine, 300 mg/day (5 h) | — | 8 days | 1.5 days |
37/M12 | skin–soft tissue | citalopram, 40 mg/day (33 h) | — | 3 days | 5 days |
39/F13 | pulmonary | fluoxetine, NR (53 h) | — | 0.5 days | 2 days |
72/F14 | cholecystopancreatitis | paroxetine, 20 mg/day (17 h) | amitriptyline, 10 mg/day (21 h) | 21 days | 2 days |
Age (years)/gender . | Infection . | Selective serotonin re-uptake inhibitor drug, dose (half-life)15 . | Other serotonergic drugs, dose (half-life)15 . | Symptoms onset after introducing linezolid . | Resolution time . |
---|---|---|---|---|---|
45/M5 | skin–soft tissue | sertraline, 200 mg/day (23 h) | trazodone, 50 mg/day (6 h) | 10 days | 2 days |
56/F6 | skin–soft tissue | paroxetine, NR (17 h) | — | 1 day | 2 days |
81/M7 | osteomyelitis | citalopram, 40 mg/day (33 h) | — | 21 days | NR died |
56/F8 | pulmonary | citalopram, 20 mg/day (33 h) | — | 13 days | 9 days died |
36/M8 | pulmonary | sertraline, 50 mg/day (23 h) | — | 5 days | 1 day |
85/M9 | prosthesis | venlafaxine, 150 mg/day (5 h) | — | 20 days | 2 days |
4/F10 | NR | fluoxetine, 5 mg/day (NR in children) | fentanyl, 200 μg/day (3.7 h) | 2 days | 2 days |
85/F11 | osteomyelitis | citalopram, NR (33 h) | — | NR | 3 days |
38/F12 | pulmonary | venlafaxine, 300 mg/day (5 h) | — | 8 days | 1.5 days |
37/M12 | skin–soft tissue | citalopram, 40 mg/day (33 h) | — | 3 days | 5 days |
39/F13 | pulmonary | fluoxetine, NR (53 h) | — | 0.5 days | 2 days |
72/F14 | cholecystopancreatitis | paroxetine, 20 mg/day (17 h) | amitriptyline, 10 mg/day (21 h) | 21 days | 2 days |
NR, not reported.
Age (years)/gender . | Infection . | Selective serotonin re-uptake inhibitor drug, dose (half-life)15 . | Other serotonergic drugs, dose (half-life)15 . | Symptoms onset after introducing linezolid . | Resolution time . |
---|---|---|---|---|---|
45/M5 | skin–soft tissue | sertraline, 200 mg/day (23 h) | trazodone, 50 mg/day (6 h) | 10 days | 2 days |
56/F6 | skin–soft tissue | paroxetine, NR (17 h) | — | 1 day | 2 days |
81/M7 | osteomyelitis | citalopram, 40 mg/day (33 h) | — | 21 days | NR died |
56/F8 | pulmonary | citalopram, 20 mg/day (33 h) | — | 13 days | 9 days died |
36/M8 | pulmonary | sertraline, 50 mg/day (23 h) | — | 5 days | 1 day |
85/M9 | prosthesis | venlafaxine, 150 mg/day (5 h) | — | 20 days | 2 days |
4/F10 | NR | fluoxetine, 5 mg/day (NR in children) | fentanyl, 200 μg/day (3.7 h) | 2 days | 2 days |
85/F11 | osteomyelitis | citalopram, NR (33 h) | — | NR | 3 days |
38/F12 | pulmonary | venlafaxine, 300 mg/day (5 h) | — | 8 days | 1.5 days |
37/M12 | skin–soft tissue | citalopram, 40 mg/day (33 h) | — | 3 days | 5 days |
39/F13 | pulmonary | fluoxetine, NR (53 h) | — | 0.5 days | 2 days |
72/F14 | cholecystopancreatitis | paroxetine, 20 mg/day (17 h) | amitriptyline, 10 mg/day (21 h) | 21 days | 2 days |
Age (years)/gender . | Infection . | Selective serotonin re-uptake inhibitor drug, dose (half-life)15 . | Other serotonergic drugs, dose (half-life)15 . | Symptoms onset after introducing linezolid . | Resolution time . |
---|---|---|---|---|---|
45/M5 | skin–soft tissue | sertraline, 200 mg/day (23 h) | trazodone, 50 mg/day (6 h) | 10 days | 2 days |
56/F6 | skin–soft tissue | paroxetine, NR (17 h) | — | 1 day | 2 days |
81/M7 | osteomyelitis | citalopram, 40 mg/day (33 h) | — | 21 days | NR died |
56/F8 | pulmonary | citalopram, 20 mg/day (33 h) | — | 13 days | 9 days died |
36/M8 | pulmonary | sertraline, 50 mg/day (23 h) | — | 5 days | 1 day |
85/M9 | prosthesis | venlafaxine, 150 mg/day (5 h) | — | 20 days | 2 days |
4/F10 | NR | fluoxetine, 5 mg/day (NR in children) | fentanyl, 200 μg/day (3.7 h) | 2 days | 2 days |
85/F11 | osteomyelitis | citalopram, NR (33 h) | — | NR | 3 days |
38/F12 | pulmonary | venlafaxine, 300 mg/day (5 h) | — | 8 days | 1.5 days |
37/M12 | skin–soft tissue | citalopram, 40 mg/day (33 h) | — | 3 days | 5 days |
39/F13 | pulmonary | fluoxetine, NR (53 h) | — | 0.5 days | 2 days |
72/F14 | cholecystopancreatitis | paroxetine, 20 mg/day (17 h) | amitriptyline, 10 mg/day (21 h) | 21 days | 2 days |
NR, not reported.
Discussion
SS has not been described in clinical trials of linezolid, but several cases have been reported after commercialization of this antibiotic. SS is thought to be induced by activation mainly of 5-HT2A serotonin receptors along with other receptor subtypes such as 5-HT1A in combination with increased levels of norepinephrine in the CNS. Stimulation or modification of these receptors in the dorsal and median raphe nuclei of the brain stem and spinal cord play an important role in the pathophysiology of the syndrome. High doses of SSRID or a combination of serotonergic agents could induce SS. Most cases resolve quickly after discontinuation of one of the interacting drugs, but in some patients it could take longer. The mortality rate was 1.27% in patients with SSRID exposure who developed significant toxic effects.2
With respect to linezolid, we could not correlate the onset of SS with any of the parameters studied except for age. Older patients developed SS later than younger patients did. This finding agrees with changes in serotonergic activity related to ageing. The number of 5-HT2A receptors in the human brain declines by 18% per decade of age resulting in a decrease in serotonergic activity with age.16 Additionally, a decrease in neurotransmitter synthesis, especially that of serotonin, catecholamines and opioids, and post-synaptic receptor sensibility has also been described.17 Both mechanisms contribute to delayed hyperstimulation of the serotonergic transmission in older patients.
Several patients received SSRID for a long time before linezolid prescription. This could have increased the risk of developing SS by increasing basal serotonergic activity. The interaction of MAO inhibitors and SSRID are well known. A washout period is generally recommended prior to the introduction of the new drug when alternating drugs between these two pharmacological groups, but this washout period is not well-established and controversial times have been recommended when linezolid has been involved.7,8,11–13 However, it is not always possible to delay treatment of infection to avoid possible drug interaction, as recommended in some cases.8,13 Another alternative is to consider SSRID dose reduction. In one case, the dose of SSRID was reduced empirically to half with resolution of the SS.12 Considering that three cases of SS have been reported in off-label indications of linezolid, linezolid prescription should also be carefully considered when patients are receiving concomitant SSRID or drugs with patent serotonergic activity and alternative antibiotics should be used when possible.
In cases where citalopram was involved in the SS, there was delayed resolution of symptoms as a result of the long half-life prolonging the serotonergic effect despite discontinuation. In the studied cases, SSRID half-life presented a correlation trend with resolution time.
Cyproheptadine is the recommended drug for treating SS, although its efficacy has not been sufficiently established.2 Cyproheptadine had no effect on the resolution time and its use seemed not to have any advantage in the studied patients.
In conclusion, clinicians must be aware that patients with SSRID receiving concomitant linezolid could be at risk of developing SS. In young patients, the SS could develop during the first days of the antibiotic treatment, but in older patients, the onset could be delayed. A delayed recovery could be expected when citalopram, a long half-life SSRID, is involved in the SS.
Transparency declarations
None to declare.
References
Rubinstein E, Isturiz R, Standiford HC et al. Worldwide assessment of linezolid's clinical safety and tolerability: comparator-controlled phase III studies.
Aga VM, Barklage NE, Jefferson JW. Linezolid, a monoamine oxidase inhibiting antibiotic, and antidepressants.
Ferry T, Ponceau B, Simon M et al. Possibly linezolid-induced peripheral and central neurotoxicity: report of four cases.
Lavery S, Ravi H, McDaniel WW et al. Linezolid and serotonin syndrome.
Hachem RY, Hicks K, Huen A et al. Myelosuppression and serotonin syndrome associated with concurrent use of linezolid and selective serotonin reuptake inhibitors in bone marrow transplant recipients.
Bernard L, Stern R, Lew D et al. Serotonin syndrome after concomitant treatment with linezolid and citalopram.
Jones SL, Athan E, O'Brien D. Serotonin syndrome due to co-administration of linezolid and venlafaxine.
Thomas CR, Rosenberg M, Blythe V et al. Serotonin syndrome and linezolid.
Tahir N. Serotonin Syndrome as a consequence of drug-resistant infections: an interaction between linezolid and citalopram.
Bergeron L, Boule M, Perrault S. Serotonin toxicity associated with concomitant use of linezolid.
Morales N, Vermette H. Serotonin syndrome associated with linezolid treatment after discontinuation of fluoxetine.
Morales-Molina JA, Mateu-de Antonio J, Grau Cerrato S et al. Probable serotonin syndrome by interaction of amitriptiline, paroxetine and linezolid.
Thummel KE, Shen DD. Design and optimization of dosage regimens: pharmacokinetic data. In: Hardman JG, Limbird LE, eds. Goodman and Gilman's The Pharmacological Basis of Therapeutics, 10th edn. NY: McGraw-Hill,
Sheline Y, Mintun MA, Moerlein SM et al. Greater loss of 5-HT2A receptors in midlife than in late life.