Article Text


Perifocal metabolism in a patient with brain abscess: insights from cerebral microdialysis
  1. Verena Rass1,
  2. Mario Kofler1,
  3. Alois J Schiefecker1,
  4. Max Gaasch1,
  5. Claudia Unterhofer2,
  6. Claudius Thomé2,
  7. Paul Rhomberg3,
  8. Bettina Pfausler1,
  9. Ronny Beer1,
  10. Erich Schmutzhard1,
  11. Raimund Helbok1
  1. 1 Neurological Intensive Care Unit, Department of Neurology, Medical University of Innsbruck, Innsbruck, Austria
  2. 2 Department of Neurosurgery, Medical University of Innsbruck, Innsbruck, Austria
  3. 3 Department of Neuroradiology, Medical University of Innsbruck, Innsbruck, Austria
  1. Correspondence to Dr Raimund Helbok, Department of Neurology, Medical University of Innsbruck, Innsbruck 6020, Austria; raimund.helbok{at}

Statistics from


Current management of bacterial brain abscess consists of parenteral antimicrobial therapy, abscess drainage and symptomatic treatment.1 Strict normoglycaemia is recommended in neurocritical care patients2; However, the optimal target for systemic glucose in patients with brain abscess is unknown.

Cerebral microdialysis (CMD) allows the assessment of cerebral energy metabolism in patients with severe brain injury. Tight glycaemic control is known to be associated with a higher rate of metabolic distress, lower cerebral glucose levels and poor outcome in neurocritical care patients supporting the idea of a more liberal glucose regimen in these patients.3

Here, we present a patient with bacterial brain abscess in whom CMD was used for brain metabolic monitoring in the perilesional area.

Case presentation

A 43-year-old previously healthy man presented with sudden onset of Broca’s aphasia and headache for 3 weeks. Neurological examination revealed a mild right facial palsy without other focal deficit. After admission, the patient developed a generalised tonic-clonic seizure. Anticonvulsive treatment with levetiracetam (1000 mg twice daily) was initiated. MRI showed a left frontotemporal lesion (maximum diameter 36 mm) with a ring-like contrast enhancement, diffusion restriction and prominent perilesional oedema suggestive of brain abscess (online Supplementary figure 1, panels A and B). Information on clinical presentation is given in the online Supplementary file 2. Stereotactic aspiration of the abscess and implantation of a CMD catheter (71 High Cut-Off, M Dialysis AB, Stockholm, Sweden; online Supplementary file 2) and brain tissue oxygen catheter (Licox Integra LifeSciences, Saint Priest, France) was performed on day 1 (first full calendar day) after admission. Positioning was confirmed by brain computed tomography (online Supplementary figure 1, panels D and G). After surgery, the patient was awake and was on mild adequate analgesia without sedatives. Based on the positive culture for Streptococcus intermedius from abscess fluid, antimicrobial therapy was modified from empirical intravenous fosfomycin (8 g three times a day) and piperacillin/tazobactam (4/0.5 g three times a day) to intravenous moxifloxacin (400 mg once daily) and clindamycin (900 mg three times a day). Two days after abscess aspiration, the patient developed another generalised tonic-clonic seizure and antiepileptic drug therapy was intensified (levetiracetam 1000 mg twice daily). Neuromonitoring values revealed a drop in brain tissue oxygen tension (PbtO2) and a CMD-glutamate peak associated with the seizure.

Supplementary data

Supplementary data

Repeated brain imaging showed progression of the brain abscess despite adequate antimicrobial treatment to a maximum diameter of 38 mm with progressive perifocal oedema and a midline shift of 8 mm (online Supplementary figure 1, panel E) requiring repuncture and drainage on day 22. Culture results remained negative at that time. Antimicrobial therapy with moxifloxacin (400 mg once daily) was continued and the patient could be discharged home on day 31 with a modified Rankin Scale score of 1. Five-month follow-up MRI scan indicated almost complete remission of brain tissue damage, without any signs of complications following CMD. The patient remained seizure free with anticonvulsive therapy (levetiracetam 1000 mg twice daily).

Brain metabolic profile in the periabscess area

The brain metabolic profile indicated perifocal neuroglucopenia (CMD-glucose ≤0.7 mmol/L) during the first 2 days of neuromonitoring which improved over time. As long as PbtO2 was monitored (3 days), 12% of neuroglucopenia occurred during brain tissue hypoxia (PbtO2 <20 mm Hg). Daily caloric needs were accomplished by oral food intake and parenteral amino acid supplementation (Aminoven 3.5%). No insulin was administered. Brain glucose concentrations increased following oral food intake and correlated with blood glucose levels (figure 1). Episodes of neuroglucopenia only occurred during serum glucose levels below 6.1 mmol/L (<110 mg/dL) and were not observed at higher systemic glucose levels. Furthermore, brain metabolic profile suggested mitochondrial dysfunction (CMD-lactate/pyruvate ratio >30 and CMD-pyruvate >70 µmol/L)3 in 47% of measurements. CMD-lactate levels were constantly increased (>4 mmol/L in 98.6%) and decreased over monitoring time.

Figure 1

Monitoring and metabolic profile. Continuous values of brain tissue oxygen tension (PbtO2), hourly values of brain metabolism assessed by cerebral microdialysis (CMD) and values of serum (S)-glucose over the monitoring period. Horizontal bars and vertical lines represent nutrition management. The height of vertical lines indicates the amount of food intake. The arrow indicates an epileptic seizure. Dashed lines represent published thresholds for pathological values of the respective parameters; PbtO2 <20 mm Hg, CMD-glutamate >16 µmol/L, lactate-to-pyruvate ratio (LPR) >30, CMD-lactate >4 mmol/L, CMD-pyruvate <70 µmol/L, CMD-glucose ≤0.7 mmol/L, S-glucose <6.1 mmol/L. Grey shaded areas represent pathological ranges. Day 0, admission; day 1, first full calendar day after admission to our neurological intensive care unit.


This is the first report of continuous brain metabolic monitoring in a patient with bacterial brain abscess. Cerebral metabolic profile was suggestive of mitochondrial dysfunction, and episodes of neuroglucopenia were common in the initial phase despite normal systemic glucose levels.

Little is known about the metabolic pattern in the brain tissue surrounding a brain abscess. The observation of low brain glucose levels may simply reflect direct glucose consumption by bacteria in the brain abscess. Another explanation may be increased metabolism and high glucose turnover in the periabscess brain tissue. Brain glucose metabolism in 18-fludeoxy-glucose-positron emission tomography (18-FDG-PET) studies showed increased FDG uptake in brain inflammatory lesions including brain abscess4 suggestive of a higher metabolic rate. Interestingly, PbtO2 levels were mostly within normal range, despite low cerebral blood volume (online Supplementary figure 1 , panel C). One can hypothesise that oxygen consumption within the periabscess region was reduced due to decreased oxygen demand.

Neuroglucopenia is associated with poor outcome in patients with traumatic brain injury and subarachnoid haemorrhage3 and may contribute to compromised brain function in the periabscess area. So far, little is known how to influence neuroglucopenia in patients with severe brain injury. In this patient, CMD-glucose levels highly correlated with serum glucose levels. Episodes of neuroglucopenia were only observed at glucose levels within the range of tight glycaemic control (4.4–6.1 mmol/L), whereas accepting liberal systemic glucose concentrations (6.1–8.3 mmol/L) resulted in improved brain extracellular glucose metabolism. Brain glucose levels increased after oral food intake, which supports the idea of early enteral feeding in critically ill patients.

The brain metabolic profile was suggestive of mitochondrial dysfunction. In bacterial meningitis, impaired cerebral energy metabolism based on mitochondrial dysfunction is more common than an ischaemic pattern.5 Ineffective mitochondrial substrate utilisation may contribute to neuroglucopenia and secondary brain injury.

The interpretation of our results has several limitations. Findings are associations, and a causative relationship cannot be concluded. CMD represents only a small brain area; therefore, exclusive cerebral metabolism of the periabscess region was demonstrated. Our results are hypothesis generating, suggesting that a more liberal glucose management may contribute to normal brain glucose levels in the brain tissue surrounding the brain abscess. Importantly, no therapeutic decisions can be concluded based on a single case presentation.

In conclusion, invasive neuromonitoring indicated metabolic derangement and neuroglucopenia in the periabscess brain tissue. Further studies are needed to confirm this finding and interpret the results for their clinical relevance.


We like to thank our nursing staff and the whole team of the neurocritical care unit.


View Abstract


  • Contributors VR, MK, MG, BP, RB and RH were involved in the acquiring, analysing and interpreting of the patient data and drafted the manuscript. VR was involved in the writing of the manuscript. RH was a major contributor in drafting and writing the manuscript. CU and CT performed the neurosurgical procedure. PR analysed the brain imaging. All authors read and approved the final manuscript.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval Ethics committee of the Medical University of Innsbruck, Austria (AN3898-285/4.8).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The data set used and analysed during the current study is available from the corresponding author on reasonable request.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.