Cephalosporin-Related Neurotoxicity in a Ceftriaxone-Treated Critically Ill Child: A Case Report

Ceftriaxone is a first-choice drug in empirical sepsis therapy in children [1] and recommended for children with community-acquired sepsis in many guidelines [2, 3]. High-dosing regiments are well tolerated and effective for the treatment of serious infections, including meningitis [4, 5]. However, serious side effects of ceftriaxone, such as encephalopathy, related to excessively high serum levels have recently been reported in adult and elderly patients on normal dosing regiments [6, 7].

Ceftriaxone clearance is dependent on biliary and renal excretion [8] and has a long half-life due to its high albumin-binding capacity [9]. These pharmacokinetic parameters have been known to change in critically ill patients [10], resulting in abnormal plasma concentrations of ceftriaxone [11].

These factors combined should prompt clinicians to consider monitoring ceftriaxone exposure in critically ill children, especially after unexpected changes in neurological status and in case of developing renal failure or severe hypoalbuminemia. Potentially due to the relative novelty and low incidence of some of these findings, recommendations regarding therapeutic drug monitoring (TDM) for ceftriaxone are lacking. The following case advocates for TDM in critically ill children. Awareness of ceftriaxone toxicity is especially relevant considering its widespread use.

A four-year-old girl with an unremarkable medical history was referred to the pediatric intensive care unit for meningococcal sepsis. She had been ill for half a day prior to admission with symptoms of fever, vomiting, and rapid development of petechiae and purpura.

The patient was tachypneic, tachycardic, and pale colored with cold extremities. Her heart rate was 170 beats per minute, she had a mean arterial blood pressure of 60 mm Hg, and she had a respiratory rate of 40 breaths per minute. Laboratory results showed a respiratory-compensated metabolic acidosis, thrombocytopenia, and prolonged clotting tests (Table 1).

She was given fluid resuscitation (40 mL/kg), ceftriaxone 100 mg/kg once daily, 2 L/min nasal oxygen, dobutamine (5 μg/kg/min), and low-dose noradrenaline (0.05 μg/kg/min). Fresh frozen plasma was administered for consumption coagulopathy. Her circulatory condition stabilized shortly thereafter, allowing inotropic and respiratory support to be discontinued after 48 h. She had become oliguric after admission, her blood urea and creatinine rose, and the renal failure was attributed to the preceding shock. Renal replacement therapy was not initiated because her overall clinical condition had improved and absolute indications (metabolic or fluid overload) were lacking. Blood culture confirmed a ceftriaxone-susceptible Neisseria meningitidis.

From the fourth day of admission onward, she showed an altered state of consciousness with marked drowsiness and incoherent speech. This was in clear contrast to her mental state in the previous days when she was able to communicate with her parents and caregivers. During formal neurological examination, she opened her eyes in response to pain, actively resisted passive movement of all the limbs, and uttered sounds. Neither myoclonus nor focal abnormalities were observed.

To rule out a cerebrovascular cause or hydrocephalus for the altered mental state, a CT scan of the brain was performed, which revealed no signs of abnormalities. After psychiatric evaluation, delirium was considered unlikely due to the lack of characteristic symptoms such as fluctuation in consciousness or hallucinatory behavior.

Taking the preceding period of critical illness and renal failure into account, “metabolic encephalopathy” became the working diagnosis. Whether renal failure was the predominant cause for the encephalopathy was doubtful, given the moderately elevated urea of 22.8 mmol/L at the onset of symptoms. Therefore, other causes for an encephalopathy were considered. Apart from the known elevated urea and creatinine, laboratory testing revealed no other significant abnormalities (shown in Table 1).

Potential side effects of administered drugs were reviewed. Since sedatives and analgesics other than paracetamol had scarcely been administered, and none of them continuously, we considered cephalosporin-related neurotoxicity (CRN). CRN was particularly suspected because of the renal failure and hypoalbuminemia, and thus potentially increased risk of high unbound ceftriaxone exposure. Measurement of serum levels of ceftriaxone on day five revealed elevated total and unbound trough concentrations (130 and 33.9 mg/L, respectively), well above the cut-off value for dose reduction (>10x MIC). The free fraction was also elevated 26.1% (reference 4–17%) [12]. On day 6 of ceftriaxone therapy, the free fraction increased even further to 42.5%. Hence, CRN was considered the likely cause of her altered mental state.

As soon as CRN was considered a possible cause for the encephalopathy, ceftriaxone administration was switched from once daily dosing to continuous infusion to avoid high peak exposure. When the laboratory results of the ceftriaxone blood levels returned on day 6, its administration was discontinued completely, and she had completed the protocolled treatment course for meningococcal sepsis at this point [2]. Concurrent with these changes, continuous venovenous hemofiltration was initiated as renal recovery was considered too slow with metabolic deterioration (potassium 5.8 mmol/L, urea 36.3 mmol/L) and the development of fluid overload.

After discontinuation of ceftriaxone and initiation of continuous venovenous hemofiltration, the girl’s mental status improved markedly in the following 2 days (Fig. 1). Renal function recovered during the following weeks, initially with increasing urine output, followed by improved clearance.

The patient was transferred from the pediatric intensive care unit to the pediatric ward after 14 days, where treatment for the sequelae of meningococcal sepsis (skin infection) and acute renal failure (hypertension, hyperphosphatemia) was continued. She was discharged after 30 days of admission and remains in long-term follow-up. At 1 year after the admission, her estimated glomerular filtration rate was 109 mL/min/1.73 m² with mild proteinuria. Due to fatigue, she was able to attend 4 out of 5 school days. The CARE Checklist for this case report is attached as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000534890).

CRN has been described in both adults and [13, 14]. CRN can be caused by different cephalosporins, including ceftriaxone [6, 14]. In a pharmacovigilance study investigating CRN, ceftriaxone was the second most frequent drug associated with CRN, cefepime being the most frequent [6]. CRN symptoms vary from confusion, agitation, hallucinations, myoclonus, seizures, tremor to coma [7]. Usually, CRN resolves within days after cephalosporin withdrawal [6, 7].

In patients with suspected CRN, cephalosporin serum levels are often elevated. In the pharmacovigilance study, 90% of the measured cases were above the therapeutic range [6]. A total ceftriaxone trough plasma concentration above 100 mg/L is associated with neurological adverse drug reactions in high-dose ceftriaxone regiments [15]. In most studies, solely the total concentration of ceftriaxone is reported. However, the plasma unbound concentration might correlate better with a risk for the development of adverse drug reaction, as this might better represent tissue levels of ceftriaxone concentration [16].

Ceftriaxone clearance can be markedly reduced in critically ill patients and has been shown to correlate with renal clearance [11, 17]. Interestingly, ceftriaxone to albumin binding in serum from patients with impaired renal function was shown to be decreased [11]. Accordingly, in critically ill patients the free fraction of ceftriaxone is elevated compared to healthy volunteers [11]. In the pharmacovigilance study investigating CRN, ceftriaxone-induced CRN was accompanied by renal impairment in half of the cases [6]. Interestingly, the aforementioned reported pediatric cases of CRN involved children on chronic peritoneal dialysis [13, 14].

Renal impairment thus appears to result in an increased total and – to an even stronger extent – unbound concentration of ceftriaxone. Although renal impairment is frequently present in cases of ceftriaxone CRN, serum albumin levels and levels of unbound ceftriaxone are infrequently reported. However, these might play an important yet under-recognized role in ceftriaxone toxicity and warrant further investigation.

Despite the wide use of ceftriaxone, there are a few reports on CRN in children. This raises the question of whether children are less frail than adults, noting that increased age is a risk factor for neurological side effects of high-dose ceftriaxone regiments [15]. On the other hand, CRN might go unrecognized as critically ill children usually suffer from multiple organ dysfunction and are often intubated and sedated. The patient presented here did not require respiratory support beyond low-flow oxygen and thus may have been the exception to the rule. We argue that when ceftriaxone TDM services are available, determining serum concentration may be helpful in preventing CRN in children with renal failure.

Ethical approval is not required for this study in accordance with local or national guidelines. Parents have given written informed consent for publication of the details of their child’s medical case and the accompanying images.

The authors have no conflicts of interest to declare.

The authors have no funding to declare.

M. Sylva was involved in writing the case report. Alan Abdulla, Birgit Koch, and Jaap Mulder were involved in writing and redirecting the case report.

All data generated or analyzed during this study are included in this article and its online supplementary material. Further inquiries can be directed to the corresponding author. A protocol for the serum ceftriaxone measurement is available upon request.