Mania secondary to cotrimoxazole antibiotherapy: a case report and review of literature

Abstract

Background

Antibiomania is the manifestation of manic symptoms secondary to taking an antibiotic, which is a rare side effect. In these cases, the antibiotics most often incriminated are macrolides and quinolones, but to our knowledge, there are no published cases of antibiomania secondary to cotrimoxazole. Furthermore, we also provide an update of pharmacovigilance data concerning antibiomania through a search of the World Health Organization (WHO) database.

Case presentation

We present the case of a 66-year-old Caucasian man with history of a brief psychotic episode 16 years prior. He received levofloxacin for a urinary infection with Klebsiella pneumoniae, with a switch to cotrimoxazole after 5 days of treatment. Shortly after the introduction of cotrimoxazole, the patient presented with a significant flow of speech, which was incoherent. After 2 days of treatment, levofloxacin was restarted, but given the lack of improvement in symptoms, the patient was hospitalized. Levofloxacin was quickly replaced by ceftriaxone for 3 days without improvement, thus a psychiatric examination was requested. This examination identified manic symptoms, and so psychotropic treatment combining 5 mg of haloperidol and 75 mg of loxapine per day was initiated. Subsequently, there was a gradual improvement in symptoms within a week. Moreover, the results of the various investigations that were carried out came back normal, which this led to suspicion of an iatrogenic cause (antibiotherapy).

Discussion

We discuss the latest pharmacovigilance data and several etiopathogenic hypotheses suggested in literature, with growing interest concerning the microbiota–gut–brain axis. It appears important to pay particular attention when taking antibiotics in the context of bipolar disorder because this association seems to precipitate the appearance of secondary mania. Management consists of stopping the responsible antibiotic, with the possibility of adding psychotropic treatment.

Conclusion

It is important to be aware of antibiomania to adopt appropriate and early treatment. We think that this case report could be useful as a clinical reminder and could possibly spark lines of research.

Background

According to the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5), mania is a mood disorder characterized by an abnormal and persistent elevation in mood, energy, or activity, all constituting a break with the usual and previous psychological state. We frequently find a reduction in the need for sleep, a significant flow of speech, acceleration of thought with flight of ideas, an increase in self-esteem, and risky behaviors. The diagnosis is clinical [1]. A manic episode can occur in patients suffering from bipolar disorder or outside of bipolar disorder, particularly secondary to taking substances, various medical conditions, or even taking medication. Some treatments are identified as potential causes of manic syndrome. We can cite nonexhaustively corticosteroids, antidepressants, and methylphenidate [2]. In this case, according to the DSM-5, the diagnosis is titled “substance/medication-induced bipolar or related disorder” [1].

For several years, an increasing number of clinical cases have reported manic episodes secondary to taking antibiotics. This syndrome was named “antibiomania” by Abouesh et al. [3]. This is a rare phenomenon that remains sparsely described in literature. Some antibiotics have been more often associated with cases of mania than others.

Concerning the pathophysiology mechanisms, there are several theories that vary according to the classes of antibiotics.

Although this is a rare side effect, it is still important for practitioners to be aware of this differential diagnosis of mania, because antibiotics are frequently prescribed around the world.

At the moment, there are no clear recommendations on the management of antibiomania.

We present the case of a patient who developed manic syndrome following the use of cotrimoxazole for the treatment of a male urinary tract infection. Then, we briefly discuss a review of literature concerning antibiomania and its management and etiopathogenic hypotheses.

This clinical case aims to remind us of the importance of thinking about this differential diagnosis of mania as early as possible, to optimize the management of this adverse effect, which can have a serious impact on the patient’s quality of life.

Case presentation

Mr. D is a 66-year-old Caucasian French patient who consulted his general practitioner for burning during urination. There was no chronic renal condition or catheterization. A cytobacteriological examination of urine (CBEU) came back positive for Klebsiella pneumoniae at 10⁵ colony-forming units (CFU)/ml with leukocyturia; the diagnosis of male urinary infection was made. Probabilistic treatment with levofloxacin 500 mg per day was prescribed according to recommendations [4]. To limit the antibiotic pressure from quinolones, and in accordance with the antibiogram, after 5 days of treatment, levofloxacin was replaced by sulfamethoxazole/trimethoprim 800/160 mg 1 tablet in the morning and evening.

On the evening of the first day of cotrimoxazole treatment, after the morning and evening dose, the patient began to have a significant flow of speech, which was incoherent. After 2 days of treatment with cotrimoxazole, given the worsening of the clinical picture, levofloxacin was restarted. Faced with the lack of improvement 48 hours later, which meant that the patient still had a significant flow of speech that was incoherent, the patient was taken to the emergency department of the University Hospital Center of Reims.

At the emergency department, the patient was hemodynamically stable, apyretic, and conscious. He presented with incoherent speech, a significant flow of speech, and burning during urination, with no other clinical reference point.

Among the patient’s history, we noted a vascular condition (arterial hypertension, dyslipidemia, hypertensive heart disease, obesity (body mass index 35), and insulin-requiring type 2 diabetes). He wears a pacemaker and suffers from sleep apnea syndrome. The patient also had an episode of urinary infection in 2020, for which a urological assessment diagnosed benign prostatic hypertrophy. There was no evidence of immunosuppression and no neurological history. Furthermore, there were no potentially confusing molecules among his treatments, and no recent therapeutic adaptation had been made apart from the initiation of antibiotherapy.

The blood test initially carried out in the emergency department detected acute renal failure with creatinine at 120 µmol/l (N: 59–104) and creatinine clearance at 54 ml/min/1.73 m² (N: > 90) according to the chronic kidney disease epidemiology collaboration (CKD-EPI) equation (functional profile on the urinary ionogram), and serum sodium at 132 mmol/l; the rest of the ionogram was normal. C-reactive protein was less than 5 mg/l. Liver function, complete blood count, hemostasis, thyroid-stimulating hormone, albumin, and blood glucose were normal. The blood ethanol level and testing for blood toxicants, benzodiazepines, and antidepressants returned negative results.

The cytobacteriological examination of urine (CBEU) done in the emergency department indicated leukocytes at 17/mm³ and red blood cells at 11/mm³, without any microorganism on direct examination, and with a monomorphic flora of coagulase negative staphylococci at 10³ CFU/ml on the final culture (suggesting contamination). Note that, at this point in time, the patient had received 9 days of antibiotic therapy.

The brain scan did not find any abnormality at the brain level likely to explain the clinical condition. The abdominopelvic scan did not find any evidence for obstructive acute renal failure and found a bladder in repletion. A urinary catheterization allowed 400 ml to be evacuated but without improvement in the neurocognitive clinical picture.

In this context, the patient was transferred the next day to the medical department for further assessment of a clinical presentation labeled as confusional syndrome and for monitoring of renal function. It was decided to replace levofloxacin with ceftriaxone 2 g per day for 3 days (Fig. 1), thus completing the 14 days of antibiotherapy required for the treatment of a male urinary tract infection [4].

Fig. 1.

Clinical case timeline. SXT trimethoprim/sulfamethoxazole, LFX levofloxacin, CRR ceftriaxone, ↕ stopping of the antibacterial agent, J6* start of symptoms, J9 arrival at emergency department, J13 last day of ceftriaxone and introduction of haloperidol and loxapine, J21! return home

The blood tests carried out  in the emergency department were repeated and remained unchanged from the first assessment. Additional laboratory tests were added and revealed the following: serum cortisol at 08:00 a.m at 603 nmol/l, normal ammonemia at 23 µmol/l, and creatine kinase at 51 IU/l. The blood cultures came back sterile. The lumbar puncture was sterile at 48 hours, acellular, and with normal biochemical analysis. Cerebrospinal fluid analysis did not show any evidence of Alzheimer’s disease, and S-100 and neurospecific enolase proteins were normal. Renal failure progressed favorably with normalization of kidney function (creatinine: 74 μmol/l (N: 59–104) and glomerular filtration > 90 ml/min/1.73 m² (N: > 90)). After normalization of the various parameters, notably the favorable evolution of the urinary infection, and owing to the persistence of the symptoms, a psychiatric opinion was requested.

Among the patient’s psychiatric history, 16 years before the current episode, three psychiatric hospitalizations took place during 6 months, for episodes with diffluent speech with incoherent remarks, behavioral disorders, logorrhea (a significant flow of speech), and hallucinatory manifestations, rapidly resolving under psychotropic treatments (loxapine 75 mg, diazepam 30 mg, and amisulpride 600 mg). No organic etiology was found at that time. There was no need for long-term follow-up or treatment. In view of these factors, the diagnosis of brief psychotic episodes was made according to the DSM-5 [1]. These episodes followed recognized family- and work-related stress factors. No family psychiatric history was found.

During the psychiatric interview of this current episode, disorganization of thought, a significant flow of speech, diffluency of speech (jumping from topic to topic), megalomaniacal ideas (ideas of omnipotence), exaltation of mood, a tendency to persecution, and insomnia were noted. In view of the symptoms, the diagnosis of manic episode was made, according to the criteria of the DSM-5 [1].

Haloperidol was introduced, then increased to 5 mg per day with the addition of 75 mg per day of loxapine. A gradual improvement in the various symptoms was noted within a week, which allowed the patient to leave the hospital with monotherapy of 5 mg per day of haloperidol.

Finally, through various investigations, after ruling out other potential causes, it was possible to conclude the diagnosis of mania secondary to an iatrogenic cause: antibiotherapy, according to the criteria of the DSM-5 [1].

The patient benefited from psychiatric follow-up for several months, during which haloperidol was gradually stopped without resurgence of psychiatric symptoms. The psychiatric follow-up was subsequently stopped.

This case was reported to the national pharmacovigilance center.

Discussion

Pharmacovigilance data and analysis

Antibiomania is a phenomenon that is rarely described in literature. Two reviews on the subject have been published. The first review was in 2002 [3] and the second in 2017 [5], which included cases reported to the Food and Drug Administration (FDA) and World Health Organization (WHO). They identified clarithromycin, ciprofloxacin, and ofloxacin as the agents most frequently associated with a manic episode, a male predominance, and a majority of patients having no history of bipolar disorder (96%).

To identify reports of antibiomania, we searched Vigibase, the WHO database with more than 30 million individual case reports from national pharmacovigilance centers in more than 150 countries, from 1968 to July, 2023. Cases were selected using the Anatomical Therapeutic Chemical (ATC) code J01 for systemically acting antibacterials drugs. The Medical Dictionary for Regulatory Activities (MedDRA) [6] preferred terms (PT) used were “mania” and “hypomania.”

Among 35,870,984 notifications, we found 823 cases of mania/hypomania (781 manias and 43 hypomanias) associated with the use of systemic antibacterials. The medications most frequently incriminated were clarithromycin (229 cases), ciprofloxacin (116 cases), levofloxacin (103 cases), moxifloxacin (43 cases), and metronidazole (41 cases). In terms of antibiotic families, there were 298 cases for quinolones, 263 cases for macrolides, 95 cases for beta-lactams, 43 cases for imidazole derivatives, and 32 cases for sulfonamides and trimethoprim.

Among the cases identified, 411 concerned men (49.9%) and 378 women (45.9%), and 34 cases do not have information. By age group, 55 cases (0.6%) concern those under 18 years old, 227 cases (27.5%) occurred between 18 years old and 44 years old, 253 (30.7%) between 45 years old and 64 years old, and 208 cases (25.2%) among those over 65 years old. Information was missing for 80 notifications.

In total, 493 cases (59.9%) were considered severe (mainly hospitalization/prolongation of hospitalization and other severe medical situations), including 3 fatal cases (0.3%), 112 nonsevere cases (13.6%), and no details regarding the others. We noted that 73.6% of cases were declared after 2007.

Concerning levofloxacin, despite a certain number of missing data (30/103), for the 103 suspected cases of manic/hypomanic episodes secondary to taking levofloxacin, the onset time ranged from 0 days to 27 days (except for 2 cases involving delays of 3 months and 6 months), with a median delay of 1 day.

For sulfamethoxazole–trimethoprim, we found 30 notifications. The onset time was reported in 15 cases and ranged from 0 days to 23 months with a median of 2 days. To our knowledge, there are no cases published in the literature concerning antibiomania secondary to cotrimoxazole.

For the main medications of interest, an analysis of disproportionality of adverse effect notifications using the case–noncase method [7] was carried out.

The reporting odds ratio (ROR) for clarithromycin was estimated at 7.1 (95% CI (confidence interval) (6.3–8.1)), effectively reflecting a pharmacovigilance signal. The ROR for ciprofloxacin was estimated at 1.5 (95% CI (1.2–1.8)), which remains significant, but an effect that remains significantly less important. For the quinolone family (ATC J01M), the ROR was estimated at 1.2 (95% CI (1.0–1.3)) and was not significant. The ROR for levofloxacin was estimated at 0.9 (95% CI (0.8–1.2)) and was not significant. For the trimethoprim–sulfamethoxazole combination, the ROR was estimated at 0.4 (95% CI (0.3–0.6)) and did not suggest an increased risk of manic episodes compared with other medications on the basis of these pharmacovigilance data. For ceftriaxone, with 11 suspected cases of mania/hypomania, the ROR was estimated at 0.1 (95% CI (0.1–0.2)). For comparison, the ROR for fluoxetine was estimated at 22.4 (95% CI (21–23.8)), and for prednisone it was estimated at 5.3 (95% CI (4.8–6.0)).

It is important to keep in mind that these calculations are based on pharmacovigilance reports and depend on other “noncase” adverse effects reported for the medications of interest. More generally, it is not possible to calculate the incidence of antibiomania owing to the lack of data on the total number of patients exposed and possible publication bias. Furthermore, as this phenomenon is not very well known, it is likely that it is underdiagnosed and therefore underreported to pharmacovigilance.

Imputability hypothesis

Concerning this clinical case, we hypothesized an attributability mainly linked to cotrimoxazole, owing to the delay in the appearance of symptoms after the introduction of this treatment.

According to Begaud’s method of assessing imputability [8], the chronological imputability score was C2: plausible, and the semiological imputability score was S2: plausible. These two scores made it possible to define the intrinsic imputability score (ranging from I0 to I6) as I3 for cotrimoxazole alone, levofloxacin alone, and the cotrimoxazole/levofloxacin combination. The extrinsic imputability score for cotrimoxazole and/or levofloxacin was B2: effect not noted in the usual documents. According to the World Health Organization–Uppsala Monitoring Center (WHO-UMC) causality assessment method [9], causality is possible.

To develop the search for arguments concerning intrinsic imputability, we were interested in the pharmacokinetics of the two medications. The patient presented with the first symptoms 24 hours after the last dose of levofloxacin, which corresponds to four half-lives [10], and after two doses of cotrimoxazole, the maximum concentration had already been reached at that time. Moreover cotrimoxazole diffuses rapidly into the cerebrospinal fluid [11]. However, we cannot rule out a participation of levofloxacin in the attributability of cotrimoxazole owing to the patient’s renal insufficiency, which may contribute to an increase in the elimination time of these two medications. This may also contribute to the fact that the observation of the reduction in symptoms after changing antibiotics could not be seen before the introduction of haloperidol. It is also important to keep in mind that complete elimination time of cotrimoxazole was not reached at the time of switching from levofloxacin to ceftriaxone. Furthermore, given the patient’s psychiatric history of brief psychotic episodes with mood elements, there seemed to be a certain vulnerability, raising questions about the presence of a bipolar disorder. There are few documented cases of antibiomania in patients with bipolar disorder, and it is possible that a rare drug cause is less likely to be considered in cases of manic episodes in patients with a previous diagnosis of bipolar disorder.

In the context of a manic episode secondary to an antidepressant, stopping the antidepressant alone is rarely sufficient for the disappearance of the manic symptoms, and the management recommendations are to stop the antidepressant and offer an antipsychotic or a mood-regulating treatment [12]. These different elements may explain at least in part the fact that the symptoms did not spontaneously decrease when cotrimoxazole was stopped.

Pathophysiological hypotheses

The mechanisms involved in antibiomania are not well known. Different etiopathogenic hypotheses for antibiomania are mentioned in literature, such as a neurotoxic effect of an increase in the concentration of the antibiotic in the central nervous system, which could be secondary to drug interactions and/or inhibition of the hepatic cytochromes P450 [5] or a possible epileptic effect that could mimic a psychiatric presentation [3]. Another hypothesis for certain antibiotics could be that they could interfere with the neurotransmission of gamma-aminobutyric acid (GABA), which plays a role in reducing neuronal excitability [5]. Other hypotheses suggest an increased concentration of cortisol, prostaglandin E1, proinflammatory cytokines or C-reactive protein, and mitochondrial dysfunction induced by the antibiotic [5]. Indeed, antibiotics that inhibit protein synthesis target bacterial ribosomal ribonucleic acid (RNA) similar to mitochondrial ribosomal RNA. Mutations in mitochondrial ribosomal RNA can mimic the structure of bacterial ribosomal RNA and bind to antibiotics, causing mitochondrial dysfunction [5]. It is also interesting to note that bipolar disorder has been associated with abnormalities in mitochondrial function [13].

We may also be interested in the fact that cotrimoxazole interferes with folate metabolism, in particular by inhibiting dihydrofolate reductase (DHFR) [11], which occurs upstream of methylene tetrahydrofolate reductase (MTHFR) [14]. Folate (vitamin B9) is a water-soluble vitamin that is vital for cellular activity and functions of the immune and nervous systems. It is a key element in the synthesis of neurotransmitters and many other metabolic pathways and has been found to be abnormally synthesized in many psychiatric diseases. There may be an association between low folate levels and bipolar disorder. Among the hypotheses that could explain this low rate in this context, the authors mention in particular the dysregulation of the MTHFR enzyme, which is a product of the MTHFR gene, which is itself a source of interest in the pathophysiology of bipolar disorder [15]. Moreover, a systematic review of randomized controlled trials concluded that folate as an adjuvant is effective and safe for the treatment of bipolar manic episodes and major depressive disorder [16].

Furthermore, more and more recent studies are linking intestinal microbiota and psychiatric disorders. The gut microbiota can influence the brain and behaviors through direct and indirect pathways. There is a bidirectional communication pathway called the microbiota–gut–brain axis [17]. A 2019 study [18] looked at the effect of psychotropic drugs on the microbiota and showed that lithium, valproate, and aripiprazole increase microbial diversity and, above all, that psychotropic drugs can have an impact on the microbiota and gastrointestinal permeability. A 2021 meta-analysis suggested a transdiagnostic community of microbial disruptions in major depressive disorder, bipolar disorder, anxiety, psychosis, and schizophrenia, characterized by depleted butyrate-producing antiinflammatory bacteria and proinflammatory enriched bacteria [19]. A 2022 systematic review found differences in the overall composition of the microbiota. Specific bacterial taxa commonly associated with mental illness were identified. Here, too, a lower level of bacterial genera that produce short-chain fatty acids (for example, butyrate) was observed, but higher levels of lactic acid-producing bacteria and bacteria associated with glutamate and GABA metabolism were also observed [20]. Data from the study by Yolken et al. lead to the conclusion that people hospitalized for a manic episode have a significantly increased rate of exposure to antibiotics owing to the presence of bacterial infections [21].

All these elements seem to suggest a link between the bacterial infection leading to the taking of an antibiotic and mood, in particular, in this case study, the appearance of manic symptoms. Research is needed to clarify all the mechanisms involved.

Antibiomania management

There are no clear recommendations on the management of antibiomania, but stopping the causative agent appears to be the essential intervention, allowing remission of manic symptoms without antimanic treatment in 39% of cases [5]. There is the possibility of adding psychotropic antimanic treatment. Here, the two antibiotics prescribed before the onset of symptoms were stopped and antipsychotic treatments with loxapine and haloperidol were introduced.

Concerning the different psychotropic antimanic treatments used, antipsychotics are the most prescribed (55% of cases, representing 26 cases). Next, in order of frequency, there are lithium (4%, two cases) and valproic acid (4%, two cases) and finally carbamazepine (2%: one case) [5].

Strengths and limitations

This article has limitations, notably owing to the fact that the patient received several antibiotics. Furthermore, relying on pharmacokinetic data remains difficult in the context of a rare adverse effect whose mechanism of action is not clearly elucidated and where unknown parameters come into play. Depending on the mechanism of action, this adverse effect could depend on the exposure time or peak exposure.

Despite these limitations, the patient’s management followed the recommendations, the differential diagnoses were explored and eliminated, no etiological factor other than the use of antibiotics was found, and the diagnosis of mania was made on the basis of the DSM-5 criteria. In addition, we have provided new, updated global pharmacovigilance data, with computer search methods allowing direct access to all pharmacovigilance reports.

Conclusion

Antibiotimania is a rare adverse effect increasingly reported to pharmacovigilance around the world, although probably underreported owing to a lack of awareness. It is important to remember the importance of reporting adverse drug reactions to pharmacovigilance.

Case reports allow the provision of important clinical elements, particularly concerning the diagnostic criteria or the evolution of symptoms depending on the treatment. More research would provide more information on antibiomania and more etiopathogenic assumptions, whose mechanisms remain not well known. This could also help develop a track of research into the mechanisms involved in mood disorders and open up the field for therapeutic means.

We would like to encourage particular attention to be given to patients with bipolar disorder taking antibiotics that can cause (hypo)manic episodes. It is interesting to note that a combination of antibiotics could have a synergistic effect on the emergence of symptoms. This case report also illustrates the importance of being aware of the phenomenon of antibiomania, especially in bipolar patients taking antibiotics, to initiate appropriate care as early as possible. Its management consists of stopping the responsible antibiotic with the possibility of adding psychotropic treatment.

Abbreviations

ATC

Anatomical therapeutical chemical

CBEU

Cytobacteriological examination of urine

CI

Confidence interval

CKD-EPI

Chronic kidney disease epidemiology collaboration

DHFR

Dihydrofolate reductase

DNA

Desoxyribonucleic acid

DSM-5

Diagnostic and Statistical Manual of Mental Disorders, 5th Edition

GABA

Gamma-aminobutyric acid

MedDRA

Medical Dictionary for Regulatory Activities

MTHFR

Methylene tetrahydrofolate reductase

PT

Preferred terms

RNA

Ribonucleic acid

ROR

Reporting odds ratio

WHO

World Health Organization

WHO-UMC

World Health Organization-Uppsala Monitoring Centre

Author contributions

TR participated in the care of the patient and contributed to the conception of this case report. MJ, with the help of TR, wrote the case report and conducted the literature search. JH conducted the pharmacovigilance research and wrote the section regarding this data. MB and AK provided a review of the manuscript before submission. All authors read and approved the final manuscript.

Funding

The present work did not receive any funding.

Availability of data and materials

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

The identity of the patient reported herein remained fully anonymous.

Consent for publication

Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.

Competing interests

The authors declare that they have no competing interests.