Seizure aggravation by ampicillin/sulbactam in an elderly patient with status epilepticus

Abstract

Background

Ampicillin/sulbactam (ABPC/ SBT) is one of the most common β-lactam antibiotics for patients with status epilepticus complicated with aspiration pneumonia. It is known that β-lactam antibiotics such as penicillin aggravate epileptic seizures or status epilepticus. Here, we investigated whether ABPC/SBT aggravates seizures using electroencephalography (EEG) monitoring.

Case presentation

An 84-year-old male with status epilepticus who presented with a new onset of clonic seizures mainly of his left side and underwent continuous video EEG was analyzed. He had been suffering from severe ulcerative colitis and infectious enteritis, delirium, atrial fibrillation and deep venous thrombosis. His cerebrospinal fluid analysis was unremarkable. Four days after starting levetiracetam, he had a cluster of seizures with impaired consciousness, consistent with status epilepticus. We started fosphenytoin and phenobarbital. We also administered ABPC/SBT twice a day, ten times in total, for aspiration pneumonia while monitoring the patient. He died twelve days after the seizure onset. We analyzed the number and duration of seizures in two hours before and after starting ABPC/SBT for each administration using EEG with trendgraph. After administration of ABPC/SBT, number of seizures significantly increased from 3.2 ± 4.7 to 7.3 ± 9.7 (mean ± SD, p = 0.047, Wilcoxson’s signed-rank test) per 2 h. Duration of seizures showed a tendency of increase from 199 ± 275 to 406 ± 536 s (p = 0.079).

Conclusions

In this elderly male patient with status epilepticus, administration of ABPC/SBT aggravated his seizures. EEG monitoring using a trendgraph is useful for evaluation of seizure severity and for analysis of causative factors.

Background

In Japan, annual incidence of aspiration pneumonia is increasing from 12.4 in 2005 to 65.1 in 2019 per 100,000 population along with the aging of society [1]. Incidence of aspiration pneumonia is high both in community- and hospital-acquired pneumonia [2]. Guidelines commonly recommend initial treatment with ampicillin/sulbactam (ABPC/SBT), a combination drug of β-lactam/β-lactamase inhibitor [3,4,5]. It is known that β-lactam antibiotics such as penicillin can aggravate epileptic seizures or status epilepticus [6, 7]. However, there is little information about adverse effect of ABPC/SBT on seizures. Here, we show clear evidence of seizure aggravation provoked by ABPC/SBT during treatment of status epilepticus in an elderly patient. Part of this manuscript was presented in the 32nd International Epilepsy Congress in 2017, Barcelona, in an abstract form.

Case presentation

An 84-year-old right-handed male presented with a new onset of focal seizures, consisting of left leg clonic seizure, evolving into left face and arm clonic seizures, sometimes into bilateral convulsive seizures (Day 1). He had been suffering from severe ulcerative colitis and infectious enteritis, delirium, atrial fibrillation, and deep venous thrombosis. He did not have alcoholism and only drank on social occasions. His brain computed tomography and cerebrospinal fluid analysis were unremarkable. Four days after starting levetiracetam (LEV) (Day 5), he developed a cluster of seizures with impaired consciousness. Blood examination showed neutrophilia (white blood cell count 7280 /µL; normal range 3300–8600 /µL, segmented neutrophil 89.0%; 38.0–74.0%), hypercoagulation (D-dimer 13.8 µg/mL; <0.5 µg/mL), renal failure and dehydration (urea nitrogen 46.5 mg/dL; 8–21 mg/dL, creatinine 1.2 mg/dL; 0.65–1.09 mg/dL), hypoalbuminemia (2.2 g/dL; 3.8–5.2 g/dL), and liver failure (γ-glutamyl transpeptidase 392 U/L; <79 U/L). The blood level of electrolytes was unremarkable (sodium 145 mmol/L; 137–147 mmol/L, potassium 4.1 mmol/L; 3.5-5.0 mmol/L, corrected calcium 9.1 mg/dL; 8.4–10.4 mg/dL). Brain magnetic resonance imaging showed high intensity lesion in the cerebral cortex and subcortical white matters of the right frontal and insular lobes in diffusion weighted image and fluid attenuated inversion recovery sequence with decreased apparent diffusion coefficient, compatible with changes due to status epilepticus (Fig. 1). Magnetic resonance angiography showed no major vessel abnormality. Thus, he was diagnosed with acute symptomatic seizure which progressed to status epilepticus mainly due to dehydration and infection. He underwent video electroencephalography (EEG) monitoring for seven days (Days 6–12). EEG signals were recoded via silver/silver chloride disk electrodes, 10 mm in diameter, placed on the scalp according to the International 10–20 system. Impedance of electrodes was kept less than 10 kΩ. Recording machines was EEG-1214 (Nihon Kohden, Tokyo, Japan). Sampling frequency was 200 Hz and bandpass filter was 0.08–60 Hz.

Fig. 1

Brain MRI, axial diffusion weighted image. High intensity lesion resides in the cerebral cortex and subcortical white matters of the right frontal and insular lobes

EEG showed continuous evolving EEG seizure patterns arising from the right hemisphere, consistent with electroclinical status epilepticus (Fig. 2A) [8]. We added fosphenytoin (fosPHT) and phenobarbital (PB) from Day 6. Trough blood concentration of antiseizure medications drawn early in the morning before administration on Day 8 and Day 11 was PHT 10.3 and 14.5 µg/mL (corrected for albumin level 18.1 and 27.4 µg/mL), PB 17.9 and 27.0 µg/mL, respectively, and LEV 28.7 µg/mL on Day 8. We also administered ABPC/SBT, 3 g, intravenous drip over 2 h, every 12 h, for seven days for aspiration pneumonia (Days 7–13). The number of seizures decreased from Day 9. He died twelve days after the seizure onset (Day 13).

Fig. 2

(A) Electroclinical status epilepticus on Day 7, approximately 3 h before the first ampicillin/sulbactam administration shown in (B). Similar electroencephalographic (EEG) pattern continued from start of video EEG monitoring on Day 6. An evolving EEG seizure pattern for 82.5 s, arising from the right hemisphere. The patient showed focal impaired awareness seizures with clonic movement on the left side of his body. Electromyographic artifacts in frontal-central-frontotemporal areas and 0.3–0.5 Hz movement artifact in O2 are seen correlating with the convulsion. (B) A density spectral array display of the right hemisphere two hours before and after ampicillin/sulbactam (ABPC/SBT) administration (black arrow). Abrupt change in wide band spectral color indicates epileptic seizures (blue arrows). Szs; seizures. (C) Number of seizures in two hours before (Pre) and after (Post) ampicillin/sulbactam administration. The number of seizures significantly increased from 3.2 ± 4.7 to 7.3 ± 9.7 times per two hours (mean ± SD, p = 0.047, Wilcoxon’s signed-rank test). (D) Duration of electrographic seizures in two hours before (Pre) and after (Post) ampicillin/sulbactam administration. The duration of seizures showed a tendency of increase from 199 ± 275 to 406 ± 536 s (p = 0.079). n.s.; not significant

To evaluate whether the administration of ABPC/SBT was responsible to the undesired outcome, we analyzed the number and duration of seizures in two hours before and after ABPC/SBT for each intravenous drip infusion, ten times in total. His wife gave informed consent on the study. Based on the retrospective and non-invasive study design with assured anonymity, the current study was exempt from the local ethics committee approval.

Figure 2B shows a density spectral array trendgraph which clearly demonstrates change in seizures before and after a single dose of ABPC/SBT administration. The number of seizures per two hours doubled from 3.2 ± 4.7 to 7.3 ± 9.7 (mean ± SD) (Fig. 2C) and duration of each EEG seizure showed twofold increase from 199 ± 275 to 406 ± 536 s (Fig. 2D) (p = 0.047 and p = 0.079, respectively, Wilcoxson’s signed-rank test), irrelevant to administration of intravenous antiseizure drugs (Supplementary Fig. 1).

Discussion and conclusions

To the best of our knowledge, this is the first study to evaluate seizure aggravation by ABPC/SBT using EEG monitoring with a density spectral array. Previous studies describe that antibiotic-associated encephalopathy (AAE) may be underrecognized considering a 15% incidence of cefepime neurotoxicity among critically ill patients and difficulty to identify association between antibiotic agents and neurological symptoms [9, 10]. Bhattacharyya et al. divided AAE into three clinical phenotypes and our patient can be classified as type I, which is characterized by onset within days of antibiotic administration, myoclonus or seizures, and abnormal EEG [10]. One of the most plausible pathophysiology of type I AAE is noncompetitive binding of β-lactams to γ-aminobutyric acid class A receptor that causes disruption of inhibitory synaptic neurotransmission and excitotoxicity [10]. These information strongly suggest that seizure aggravation of our patient is caused by β-lactam of ABPC. In contrast, SBT is known to have neuroprotective effect against excitotoxicity of glutamate by upregulating glutamate transporter 1 in brain ischemia, Alzheimer’s disease, and amyotrophic lateral sclerosis [11, 12].

Since status epilepticus is often treated with multiple antiseizure medications and other concomitant drugs for associated conditions, possible risk for drug-drug interactions should be considered. PHT and PB are strong inducers of cytochrome P450 and uridine-glucuronyl transferase enzymes which decrease concentration of other drugs [13]. On the contrary, many co-administrated drugs can increase or decrease concentration of PHT and PB [13,14,15]. ABPC/SBT does not affect blood concentration of parenteral PHT and PB, thus it is unlikely that pharmacological interaction caused seizure worsening of our case.

Dose adjustment is needed in elderly patients with renal dysfunction which can cause decreased drug clearance. We adjusted ABPC/SBT dosing regimen according to the Sanford Guide to Antimicrobial Therapy which employs an extended interval of 12 h. In addition, blood concentration of PHT and PB was monitored because renal dysfunction and hypoalbuminemia can lead to increased free fraction and adverse effects.

In conclusion, antibiotics including ABPC/SBT can cause AAE and aggravate seizures in patients with status epilepticus. EEG monitoring using a density spectral array trendgraph is useful for detection and evaluation of seizure severity and for analysis of causative factors.