Cefepime: a novel cause of thrombocytopaenia

Abstract

Thrombocytopaenia is a commonly encountered finding in hospitalised patients. Many antibiotics, especially beta-lactams, are well known to cause thrombocytopaenia by an immune-mediated mechanism. We present a 55-year-old woman who was admitted to the hospital with a complicated urinary tract infection resulting in right-sided pyonephrosis and pararenal abscess. She was observed to develop thrombocytopaenia after initiation of cefepime therapy. Following an extensive work-up for her new-onset thrombocytopaenia, she was diagnosed as a case of a drug (cefepime)-induced thrombocytopaenia. Her platelet count recovered back to normal levels after cessation of cefepime therapy. Based on our PubMed search, there are only a few cases of cefepime-induced thrombocytopaenia published in the literature. This paper illustrates that physicians should include cefepime among the possible aetiologies of thrombocytopaenia. Additionally, this article outlines the currently available approaches to the diagnosis and management of drug-induced thrombocytopaenia.

Background

Cefepime is a beta-lactam antibiotic and a fourth-generation cephalosporin that is used to treat a broad category of infections caused by both Gram-positive and Gram-negative bacteria. Just like other commonly used antibiotics, cefepime has many known adverse effects, with fever, skin rash and diarrhoea the most common ones. Thrombocytopaenia is one of the rare side effects of cefepime use, with a reported incidence of less than 1%.¹ Although it is not very well studied, the proposed mechanism behind cefepime-induced thrombocytopaenia is an idiosyncratic immune-mediated reaction.

Classically, platelet count starts to decline a few days after starting the drug and is usually asymptomatic. Thrombocytopaenia typically resolves once the medication is stopped. Drug-induced thrombocytopaenia is suspected once other more common causes of thrombocytopaenia have been excluded, and the platelet count starts trending up after the culprit drug is stopped. The Narango Drug Adrverse Reaction Probability Scale can also be useful in the diagnosis of drug-induced thrombocytopaenia.² To the best of our knowledge, only one case of thrombocytopaenia explicitly associated with cefepime has been reported to date.¹ We report here a case of cefepime-induced thrombocytopaenia in a middle-aged woman.

Case presentation

A 55-year-old female patient was admitted to our hospital with intermittent right flank pain, dysuria and fever for the last 2 weeks. The patient reported that her symptoms first started around 2 years ago and gradually worsened. She received multiple antibiotic treatments for presumed urinary tract infections over 2 years, which intermittently alleviated some of her symptoms. However, she never returned to her normal state of health. She denied any history of nephrolithiasis or urological procedures.

On examination, the patient’s vital signs were as follows: blood pressure 127/71 mm Hg, heart rate 82 beats per minute, respiratory rate 18 per minute, oxygen saturation 98% on room air and temperature 38.1°C. She had right costovertebral angle tenderness on palpation. Her cardiopulmonary and neurological examinations were unremarkable. Laboratory investigations at admission revealed marked leucocytosis (18.2×10⁹/L), anaemia (haemoglobin of 9.8 g/dL), normal platelet count (180×10⁹/L) and elevated serum creatinine (1.4 mg/dL) with an unknown baseline. Urinalysis showed bacteriuria and pyuria and therefore urine culture was ordered. The rest of the laboratory investigations, including metabolic panel, liver function test and coagulation studies, were unremarkable. Renal ultrasound demonstrated increased parenchymal echogenicity and perinephric free fluid. CT of the kidneys revealed additional findings of right-sided pyonephrosis along with right perinephric abscess (figure 1).

Figure 1.

Contrast-enhanced CT scan. (A) Level of the interpole of the right kidney demonstrating a non-enhancing area (yellow arrow) with surrounding perinephric fat-stranding. (B) Similar lesion was seen at the lower pole (yellow arrow). Other multiple foci were also seen (not shown).

The patient was started on intravenous hydration. The previous culture and sensitivity report showed resistance to third-generation cephalosporin including ceftriaxone and therefore intravenous cefepime was started. Two days after initiation of cefepime treatment, a declining trend in the patient’s platelet count was observed, with levels approaching as low as 102×10⁹/L. Intravenous cefepime therapy was continued while searching for other possible causes of thrombocytopaenia. The patient did not have any previous exposure to antiplatelet or anticoagulation therapy. Her 4Ts (thrombocytopenia, timing of platelet count fall, thrombosis or other sequelae, other causes for thrombocytopenia) score was 2, excluding heparin-induced thrombocytopaenia. Pseudothrombocytopaenia was ruled out by repeating the complete blood count in a citrated tube. Peripheral smear examination did not reveal any abnormal platelet clumping. Hepatitis C and HIV tested negative. Disseminated intravascular coagulation and other microangiopathies were ruled out in the setting of normal serum haptoglobin, coagulation studies and normal fibrinogen degradation product levels.

The patient’s platelet count continued a downward trend until it reached a nadir level of 6×10⁹/L on day 7 of cefepime therapy. After a careful review of the medication list, cefepime was stopped as a possible cause of her thrombocytopaenia. Luckily she did not develop any symptoms of bleeding diathesis. During the entire period of low platelet count, her creatinine level remained close to the value at the time of admission. Her urine culture sent at the time of admission grew Escherichia coli and she was started on intravenous meropenem based on the antibiotic sensitivity report. The next day, her platelet count started to gradually improve until it returned to baseline after 7 days of stopping cefepime.

Her medication history included valsartan 80 mg for the last 2 years. For analgesia, she took tramadol 50 mg two times per day, which was continued during her hospital stay and as her discharge regimen. Both valsartan and tramadol were continued during the period of thrombocytopaenia and recovery. She had also taken oral ciprofloxacin multiple times before presenting to our department. None of the drugs had any temporal relationship with the thrombocytopaenia except for cefepime.

The patient ultimately underwent right nephrectomy in the setting of long-standing pyonephrosis and poor residual right renal function (<10%) on diethylene triamine penta-acetic acid scan.

Outcome and follow-up

Overall, the patient had an uneventful postoperative recovery. Her thrombocytopaenia resolved and her platelet count returned to baseline after stopping cefepime. She was discharged home on the seventh postoperative day in a stable condition with a normal platelet level. She was followed regularly in the outpatient clinic with interval testing of blood counts. Her platelet counts remained within normal limits and she continues to do well to date.

Discussion

Thrombocytopaenia is a common finding seen in majority of hospitalised patients and is usually defined as a platelet count below 150×10⁹/L or a 50% decrease in platelet from the baseline level.³ Thrombocytopaenia is asymptomatic in most cases but can result in a myriad of clinical manifestations, ranging from minor petechiae to life-threatening spontaneous intracranial haemorrhage. Thrombocytopaenia can arise from various aetiologies, including decreased production, increased platelet consumption or sequestration. Drug-induced thrombocytopaenia is considered a diagnosis of exclusion. The most crucial factor linking thrombocytopaenia with a particular drug is the temporal association between initiation of the drug treatment and the onset of thrombocytopaenia. The reported incidence of drug-induced thrombocytopaenia is between 19% and 25% in acutely ill patients.⁴

Various pathways have been implicated in the development of drug-induced thrombocytopaenia. One of the mechanisms is decreased platelet production through direct myelosuppression, as seen with most chemotherapeutic agents.⁵ Beta-lactam antibiotics, mainly first-generation, second-generation and third-generation cephalosporins, are well known for causing thrombocytopaenia by an immune-mediated reaction cascade.^6–8 Cephalosporins carry a beta-lactam ring in their chemical structure that binds the free hapten in blood and results in hapten-antibody carrier complexes that act as an antigenic source for antibody production. This hapten-related antibody production is thought to be the cause of immune-mediated platelet consumption and resultant thrombocytopaenia.⁸

The platelet count typically begins to drop within 2–7 days of initiating a drug, like in our case,⁹ whereas the platelet downtrend in the previously reported case of cefepime-induced thrombocytopaenia by Lim et al¹ began after only 1 day of intravenous cefepime.¹ Although it can be asymptomatic, the most common symptoms associated with cephalosporin-induced thrombocytopaenia are petechiae, purpuric lesions, deep tissue haematoma and gastrointestinal bleeding.^{7 8 10} However, our patient remained asymptomatic from thrombocytopaenia despite an extremely low level of 6×10⁹/L, just like the patient in comparison (table 1 and figure 2).¹

Table 1.

Comparison between previously published case and the current case

	Lim et al1	Current case
Age	45 years	50 years
Gender	Male	Female
Admission diagnosis	Chest injury	Pyonephrosis
Culprit drug	Cefepime	Cefepime
Baseline platelet count	102×109/L	180×109/L
Platelet count after 2 days of cefepime	62×109/L	102×109/L
Platelet count after 4 days of cefepime	23×109/L	40×109/L
Platelet nadir level	15×109/L	6×109/L
Platelet at the time of stopping cefepime	15×109/L	6×109/L
Platelet on the second day of stopping cefepime	42×109/L	38×109/L
Platelet on the third day of stopping cefepime	80×109/L	64×109/L
Platelet on the eighth day of stopping cefepime	123×109/L	184×109/L
Symptoms from thrombocytopaenia	None	None
Treatment offered	Stopping the drug	Stopping the drug

Figure 2.

Graph showing the comparison between the current case and Lim et al’s¹ case (platelet count trend plotted against the day of cefepime therapy). This figure/graphical work is originally made by the authors.

A work-up for new-onset thrombocytopaenia should start by ruling out pseudothrombocytopaenia, which can be done by simply repeating complete blood count in a non-EDTA anticoagulant and by the absence of platelet clumping on peripheral smear. The presence of other cytopaenias should lead to evaluation for possible bone marrow suppression, sepsis, hepatitis C, HIV infection and microangiopathies, including disseminated intravascular coagulation. Testing for heparin-induced thrombocytopaenia should be considered, especially in hospitalised patients with thrombosis, based on their pretest probability, which is calculated using the 4Ts score. A detailed medication review should be performed to look for any possible causative agents. Our patient underwent extensive work-up for her thrombocytopaenia and was diagnosed with cefepime-induced thrombocytopaenia, given the temporal relationship with the initiation of cefepime and quick recovery after stopping the drug. We also used the Naranjo Drug Adverse Reaction Probability Scale to calculate the possible association of cefepime with decreasing number of platelets. It yielded a positive result with a score of 6, falling in the category of ‘Probable’.²

The treatment for drug-induced thrombocytopaenia primarily includes stopping the causative agent and close monitoring for resolution of thrombocytopaenia. It is also important to monitor renal function as cefepime is renally cleared and decreased renal function can worsen the side effects associated with cefepime. Most of the patients do well with this strategy alone, just like our patient. Some patients with severe, life-threatening bleeding may require plasmapheresis and intravenous immunoglobulins.¹¹ Platelet transfusion can also be considered in refractory cases.¹²

Learning points.

• Thrombocytopaenia is a rarely reported side effect of cefepime that can have potentially life-threatening consequences if missed.

• We suggest checking a baseline level of platelet count prior to starting cefepime therapy, daily monitoring of platelet count and stopping the drug promptly if the patient develops new-onset thrombocytopaenia, especially within 2–7 days of drug therapy initiation.

• Patients with severe thrombocytopaenia (<10–20×10⁹/L) and significant bleeding may require intravenous immunoglobulins, plasmapheresis and platelet transfusion, in addition to stopping the causative agent.

Ethics statements

Patient consent for publication

Obtained.