Fluoroquinolone antibiotics are commonly prescribed due to their broad spectrum of activity, excellent oral bioavailability and well tolerated side effect profile. However, a wide range of central nervous system (CNS) effects have been reported with an estimated incidence of 1-2% in patients taking fluoroquinolones ( 1 – 3 ). Neurotoxic manifestations have included psychosis, seizures, encephalopathy, myoclonus, ataxia, dysarthria, chorea and oral facial dyskinesia. We report a case of levofloxacin induced myoclonus and review the literature on fluoroquinolone induced movement disorders.

A 62-year-old male presented to the emergency department with a 1 week history of poor oral intake, nausea, vomiting, bloody diarrhea and decreased urine output. The patient gave informed consent for publication of this case report. His past medical history included chronic obstructive pulmonary disease (COPD), hypertension, stroke, gastroesophageal reflux disease, constipation, chronic back pain, anxiety and remote stomach surgery twenty-five years prior to admission. He was an ex-smoker (56 pack/year) and drank approximately 8 beers per month. Medications prior to admission included: docusate 100 mg at the bedtime, candesartan 16 mg daily, pantoprazole 40 mg daily, tiotropium 18 mcg daily, budesonide 200 mcg/formoterol 6 mcg 2 puffs twice daily, lorazepam 0.5 mg twice daily and as needed, morphine sustained release 20 mg twice daily, morphine immediate release 5 mg twice daily as needed, salbutamol 100 mcg 2 puffs every 4 hours as needed, ipratropium 500 mcg/salbutamol 2.5 mg twice daily as needed, dimenhydrinate 50 mg daily as needed, ondansetron 8 mg every 8 hours as needed, and PEG 3350 2 tablespoons daily as needed. The patient denied any regular use of other over counter medications, herbal medications, natural health products or traditional medications. There were no medication changes in the last 3 months The allergy history was later confirmed to only include doxycycline which caused throat swelling. The patient had received numerous courses of levofloxacin in the past for acute exacerbations of COPD with no history of neurological adverse events.

Vitals on admission included: blood pressure 86/49 mmHg, heart rate 96 beats per minute, temperature 37.8°C, respiratory rate 20 breaths per minute and oxygen saturation 98% on 3 liters per minute of oxygen. On exam, the patient was drowsy, frail and cachectic with dry mucous membranes. A digital rectal exam revealed hemorrhoids and bright red blood. The remainder of the physical examination was unremarkable. Laboratory tests on admission included hemoglobin 117 (normal range 137- 180 g/L), platelet count 292 × 109/L (normal range 150 × 109 /L-400 × 109/L), white blood cell count (WBC) 9.7 × 109/L (normal range 4 × 109/L-11 × 109/L), international normalized ratio (INR) 1 (normal range 0.9–1.1), serum potassium 5.6 mmol/L (normal range 3.3-5.1 mmol/L), serum magnesium 0.98 (normal range 0.65-1.05 mmol/L), serum creatinine 540 umol/L (normal range 50-120 umol/L), noting a serum creatinine of 56 umol/L 3 weeks prior to admission. Urea 44.4 (normal range 3-9 mmol/L), fractional excretion of sodium 0.05% (FENa <1% usually indicates a prerenal state such as dehydration or other forms of prerenal azotemia), corrected calcium 1.78 mmol/L (normal range 2.10-2.55 mmol/L), serum phosphate 2.66 mmol/L (normal range 0.8-1.5 mmol/L), and venous blood gas demonstrating a severe anion gap metabolic acidosis with a pH 7.05 (normal range 7.3-7.4), anion gap 23 mmol/L (normal range 4-16 mmol/L) and a normal lactate. The patients other investigations including liver enzymes, electrocardiogram, chest x-ray and abdominal plain film were largely unremarkable except for urinalysis which showed marked pyuria and hematuria in the absence of signs/symptoms of a urinary tract infection.

The patient was empirically treated with ceftriaxone 2 g IV daily for suspected sepsis as well as crystalloid for fluid resuscitation. He also received 100 mmol of IV sodium bicarbonate 8.4% and potassium was shifted with dextrose 50% and 10 units of regular insulin. Hypocalcemia has corrected with calcium gluconate 1 gram IV x 2 doses as he was symptomatic with clonus and myoclonic jerks. Candesartan, morphine, docusate, lorazepam, and PEG 3350 were held on admission.

On day 2, an ultrasound of the kidneys and bladder did not suggest hydronephrosis. Final report of the urine culture showed no growth. Calcium carbonate 800 mg elemental calcium twice daily and cholecalciferol 2000 units daily were started. Serum phosphate had decreased to 1.81 mmol/L. The patient received hydromorphone 1mg IV x 1 dose for pain.

On day 3, serum creatinine and phosphate had decreased to 91 umol/L and 0.88 mmol/L respectively. The ceftriaxone was discontinued as no infectious source was identified. Serum magnesium was 0.76 mmol/L and corrected calcium had increased to 2.53 mmol/L. The patient was complaining of chronic back pain, severe abdominal pain and nausea. The patient was prescribed hydromorphone 1 mg IV every 6 hours as needed (he was receiving 3-4 doses daily) & dimenhydrinate 50 mg IV every 4 hours and ondansetron 4-8 mg IV every 6 hours. On day 4,ova and parasite,Clostridium difficile antigen test was negative but, stool culture was positive for Campylobacter jejuni. The patient has prescribed levofloxacin 500 mg orally daily x 3 doses as there was an allergy to clarithromycin with a reaction of throat swelling documented on the hospital chart. Two hours after the initial dose of levofloxacin it was noted that the patient’s neck would shake and turn to the left and his right arm would abduct at the shoulder and flex at the elbow. The movements would occur every 1 to 2 minutes. There was muscle fasciculation’s in the bicep and forearm of the right arm briefly after neck movement. A neurological exam was completed and revealed normal muscle tone in the upper and lower extremities, no promotor drift or tremor and no clonus in the ankles. The patient was given diphenhydramine 25 mg IV x 1 dose and within 20 minutes there was a reduction in the abnormal movements. The levofloxacin was discontinued and replaced with clindamycin 300 mg orally 4 times daily x 2 days. On day 5 the allergies were clarified with the patient’s family doctor and community pharmacy and there was no allergy to clarithromycin. Regard-less, the antibiotic was not changed and the patient completed the course of clindamycin. The movement disorder had completely resolved.

On day 8, the dimenhydrinate was stopped. The patient was still having diarrhea (up to 11 bowel movements daily). Day 9 patient was started back on his home dose of morphine. Computed tomography (CT) of the abdomen and pelvis was completed for ongoing abdominal and pelvic pain, nausea and vomiting, and perfuse watery diarrhea but, no abnormal pathology was identified. On day 10 the diarrhea and nausea had improved and ondansetron was discontinued. The patient was discharged and continued all previous home medications after discharge except for docusate, PEG 3350 and lorazepam.

Hypocalcemia and morphine accumulation were the most likely causes of the initial presentation of myoclonus and myoclonic jerks. Calcium plays an important role in the propagation of neuromuscular activity but, the serum calcium was corrected prior to levofloxacin administration ( 4 ). Twenty to 40% or oral morphine is absorbed with the majority of clearance by hepatic glucuronidation as morphine-3-glucuronide (50%) and morphine- 6-glucuronide (M6G) (5 to 15%). Only M6G is active. Ninety percent of metabolites and free drug undergo renal excretion( 5 ).Advanced age, high doses of opioids, rapid increases in opioid dose and renal failure are risk factors for morphine accumulation resulting in myoclonus ( 6 ). The only risk factor present in our patient was acute kidney injury with a 10 fold increase in serum creatinine from baseline. The patient had received 1 dose of hydromorphone and no morphine in the first 2 days and serum creatinine had returned to normal suggesting opioid accumulation was not a factor on the day of levofloxacin administration.

Campylobacter infection has been associated with neurological adverse effects. Guillain-Barre syndrome is an uncommon consequence of Campylobacter jejuni (estimated at 1case per 2000 infections) but usually occurs 2-3 weeks after diarrheal illness ( 7 ). Our patient did not have the classic symptoms of numbness or tingling in the fingers and toes and the development of muscle weakness in the legs and arms. All other potential causes of the movement disorder were ruled out in the case presented. The movement disorder was considered to be most likely an adverse drug reaction to levofloxacin as assessed by the Naranjo adverse reaction probability assessment tool (score of 6)( 8 ).

A literature search of PubMed, Embase & Reactions Weekly from inception to October 2016 using the search terms, “ciprofloxacin”, “gatifloxacin”, “gemifloxacin”, “grepafloxacin”, “levofloxacin”, “moxifloxacin”, “norfloxacin”, “ofloxacin”, “trovafloxacin”, “movement disorders” and “motor dysfunction” yielded 21 case reports (see table-1)( 9 – 28 ). One case report was excluded from the analysis. The case involved a 40-year-old female who had a pseudomeningocele treated with a lumbar drain and developed Enterobacter cloacae meningitis. The patient had normal liver and kidney function. Ciprofloxacin was added to the antibiotic regimen and 4 days later the patient developed generalized tremulousness. The case was confounded by the patient’s medication regimen which included; droperidol, meperidine, metoclopramide, oxycodone, amitriptyline and incomplete documentation of doses received ( 28 ).

There was a wide variation in the range of age from 25 to 87 years in the cases reviewed (see table-1). A history of neurological disease was present in 3 patients: transient ischemic attack ( 9 ), stroke ( 10 ), and Alzheimer dementia ( 11 ). Ciprofloxacin was implicated in 50% of the cases reported. The most frequently reported neurological manifestation was oral facial dyskinesia. No pathologic cause was identified on CT of the head ( 9 – 14 ) or magnetic resonance imaging of the brain in these cases ( 10 , 15 – 19 ). Three patients had a normal lumbar puncture ( 16 , 18 , 20 ) and 5 patients a normal electroencephalogram (EEG) ( 12 , 13 , 18 , 19 , 21 ). Two patients had abnormal electromyography (EMG). The case by Striano showed irregular, continuous, and multifocal myoclonus, more evident in the upper limbs and multifocal paroxysmal activity on EEG ( 22 ). The case by Post ( 19 ) noted increased muscle activity in the biceps brachii and paraspinal muscles with EMG bursts varying from 80 msec to 400 msec and symptoms lasted for 1 year. The authors proposed that ciprofloxacin triggered a spinal myoclonic generator that continued to be active even after discontinuation. In one case a nerve conduction study suggested a demyelinating polyneuropathy ( 27 ). Review of the cases suggested the onset of symptoms generally ranged from hours after administration after the first dose up to day 5 of therapy with complete resolution of symptoms within 5 days in the majority of cases. Anticholinergics, neuroleptics, and benzodiazepines were the most common antidotes used. In all cases, the antibiotic was discontinued and the patient was not rechallenged. The literature review suggests renal ( 9 , 10 , 12 , 14 , 21 , 22 , 24 ) or hepatic dysfunction ( 17 , 18 , 25 ) are risk factors associated with fluoroquinolone induced movement disorders. In our case, the onset of symptoms occurred within 2 hours of administration which has been reported with both ciprofloxacin ( 14 ) and gatifloxacin ( 23 ). The ciprofloxacin case reported by MacLeod involved a 69-year-old woman with a history of metastatic cervical cancer, ischemic heart disease and diabetes mellitus ( 14 ). The patient presented to the emergency with anuria and fever. A uretic stent was replaced and following the procedure she was placed on cefazolin 1 gram IV every 6 hours and norfloxacin 400 mg PO daily (no other concomitant medications were reported). Results of blood and urine cultures revealed Pseudomonas sensitive to ciprofloxacin. Norfloxacin was discontinued and the patient was started on ciprofloxacin 400 mg IV every 12 h with a serum creatinine of 271 umol/L. During the first infusion, the patient became unresponsive, demonstrating decorticate posturing with a rightward deviation of her gaze. A clinical diagnosis of the cerebral vascular accident was made and the results of a CT of her head was normal. Over the next 2 hours, she developed facial tics, consisting of grimacing and protruding of the tongue, an automatism of her right hand, echolalia and echopraxia. Ciprofloxacin was discontinued and her neurological symptoms resolved within several hours with supportive care. There were no residual effects and the patient was discharged 7 days later ( 14 ). In a case of Mohan, a 25-year-old patient received gatifloxacin for acute bronchitis but, was otherwise young and healthy and did not receive anyother concomitant medications ( 23 ). After receiving gatifloxacin he experienced an unsteady stance and gait but, the remainder of his neurological examination was normal. No further investigations were reported. The ataxia resolved within 6 hours of gatifloxacin discontinuation and the patient was discharged from thehospital( 23 ). This parallels the patient’s rapid resolution of symptoms in our case after the cessation of levofloxacin. The neuromuscular symptoms improved within 20 minutes of IV diphenhydramine and had completely resolved with 24 hours of levofloxacin discontinuation. The mechanism by which fluoroquinolones cause neurotoxicity is not fully understood. Postulated mechanisms for fluoroquinolonemediated CNS toxicity include inhibition of GABA-A-receptors as well as activation of the excitatory NMDA receptors ( 2 ). This is thought to induce a hyperexcitable neuronal state resulting in the development of a movement disorder ( 10 , 15 , 19 , 20 , 25 ). There is a similarity between the structure of certain substituents at position 7 of the fluoroquinolone nucleus and the neurotransmitter GABA ( 29 ). It is hypothesized that fluoroquinolones may compete with and displace GABA from its receptor sites in the CNS leading to stimulation of the CNS ( 29 ). Other mechanisms that have been suggested include the interaction of fluoroquinolones with opioid, dopamine and glutamate receptors ( 29 ).

Iigin et al studied the effects of ciprofloxacin on neurotransmitters ( 30 ). Rats were fed ciprofloxacin 20 mg/kg or 50 mg/kg day or saline solution as a placebo for 14 days. There were no significant alterations in the dopamine or noradrenaline levels measured in brain homogenates. Adrenaline levels were decreased and glutamate levels increased in both treatment groups. Ciprofloxacin 50 mg/kg significantly reduced serotonin and GABA.

Malondialdehyde, brain glutathione, superoxide dismutase, catalase were measured to determine the contribution of oxidative stress to ciprofloxacin neurotoxicity. In the rats fed ciprofloxacin decreased levels of brain glutathione and catalase activity were noted. There was no difference in superoxide dismutase activity. Brain malondialdehyde levels were higher in rats fed ciprofloxacin. The results suggest an enhanced oxidative stress/damage with a weakened oxidative defense system ( 30 ).

Penetration of drugs into the cerebral spinal fluid and brain tissue are dependent on meningeal inflammation, molecular size, lipophilicity, plasma protein binding, susceptibility to active transport systems & the degree of ionization of the medication ( 31 ). Most fluoroquinolones are moderately lipophilic with a molecular mass of approximately 300 daltons and low protein binding to plasma proteins (approximately 20-40%)( 31 ). At physiologic pH, most of the agents are uncharged which favors their CNS penetration ( 31 ).

CNS reactions to fluoroquinolones are dose or concentration dependent but, neurotoxic reactions to fluoroquinolones are not strictly mildly inflamed meninges ranges from 0.3-0.7 and with strong meningeal inflammation 0.7- 0.9 ( 31 ). Due to their potential to cross the blood-brain barrier, it is logical that fluoroquinolones may be indicated in the treatment of certain CNS infections. They have not been extensively studied in the treatment of bacterial meningitis and are not part of standard therapy but may be used in patients with multiple severe drug allergies or in the treatment of highly resistant organisms ( 32 , 33 ). In these instances, ciprofloxacin or moxifloxacin may be considered as an alternative therapy ( 32 , 33 ). Moxifloxacin or levofloxacin may also be used in the treatment of CNS tuberculosis ( 34 ). Low serum concentrations of norfloxacin are attained after oral administration of usual dosages. Use of norfloxacin is generally limited to genitourinary or gastrointestinal tract infections ( 35 ). Gatifloxacin (dysglycemia), gemifloxacin, grepafloxacin (QT prolongation), ofloxacin, trovafloxacin (hepatotoxicity) have been withdrawn from the market.

There is very little data on the concentrations of fluoroquinolones in brain tissue with only data on ciprofloxacin being reported. Five patients undergoing surgery for removal of brain tumor, aneurysms or other vascular malformations received ciprofloxacin 750 mg (route not specified) 3 to 5 hours prior to surgery. Three to 5 hours post administration the mean concentration of ciprofloxacin in serum was 2.82 mcg/mL (range 0.62-7.54 mcg/mL) and at 3 hours in brain tissue was 2.18 (0.69-5.05 mcg/mL). This study suggested ciprofloxacin brain tissue concentrations were correlated with high plasma concentrations ( 1 ). approximately 25% lower than serum however, The AUCCSF/AUCserum for ciprofloxacin, levofloxacin, and moxifloxacin in uninflamed or there was a wide variation in ciprofloxacin brain tissue concentrations ( 36 ). In another study, 14 patients undergoing brain tumor excision received a single dose of ciprofloxacin 200 mg IV administered over 20 min. Blood samples were drawn 10 minutes after the infusion. Samples from subcutaneous fat, skull bone, dura mater and brain tissue were collected 60 min after ciprofloxacin administration. The concentrations showed considerable variability and were approximately 2-20 times higher than the corresponding concentrations in CSF ( 37 ). No clear relationship has been established between CNS penetration and adverse events for fluoroquinolones ( 2 , 3 ).

Movement disorders are a rare adverse effect associated with fluoroquinolones. The proposed mechanism involves neuronal excitability through GABA inhibition and enhanced oxidative stress. The adverse reaction appears to be a class effect. Oral facial dyskinesia has been most commonly reported. Half of the cases reviewed occurred in patients with renal or hepatic dysfunction necessitating the need for appropriate dosing. The onset of symptoms is usually within 5 days after fluoroquinolone administration with complete resolution of symptoms within 7 days after fluoroquinolone discontinuation. The case presented and literature review summarizes the data on fluoro-quinolone induced movement disorders.