Acute subarachnoid hemorrhage, intraventricular bleeding, tumors of the brain stem,and other acute intracranial pathologies may require insertion of external ventricular drainsto manage hydrocephalus and monitor intracranial pressure, which holds the potential for intracerebral infections such as ventriculitis and/or meningitis^[1].  The spectrum of pathogens in ventriculitis is characterized mainly by Gram-positive, the Infectious Diseases Societyof America therefore recommends vancomycin in combination with meropenem (or ceftazidime) for the initial treatment of infections after neurosurgery ^[2]. However, due to the hinder of blood-CSF-barrier, achieving and maintaining appropriate concentrations at the target site of infection is a significant challenge for physicians.

In addition, it has been documented that there are also large individual vaiations in the cerebrospinal fluid permeability of both vancomycin and meropenem in patients with ventriculitis. To maintain effective antimicrobial concentrations in the CSF, continuous infusion of high concentrations of the drugs (daily dose of up to 20 g meropenem and 8 g vancomycin) is required. This strategy was implemented in the local hospital to ensure the curative effect.

This study retrospectively analyzed the concentrations of serum and cerebrospinal fluid in 9 patients with ventriculitis who were receiving consecutive infusion of meropenem and vancomycin.

The present study enrolled 11 patients treated for ventriculitis. However, two of these patients were ultimately excluded due to the retrieve of other sources of infections, leading to frequent change of antibiotics, and only 9 patients were retained. In the trial, the first serum and cerebrospinal fluid concentration tests were performed 24 hours after initiation of dosing, and dose adjustments were made at the 48-hour time point. This was followed by a second drug concentration test at 72 hours, followed by dose adjustment at the 96-hour time point. As shown in Figure 1.

Figure 1  Treatment regimen. Serum and CSF concentrations were analyzed after 24 and 72 h with consecutive dose adjustment after 48 or 96 h from the start of infusion. Further analysis and dose adjustment were only performed where necessary.

The median values of the initial dose for 9 patients were 8.8 g of meropenem and 4.25 g of vancomycin, which equals 114.3 mg/kg bodyweight/24h meropenem and 57.2 mg/kg of bodyweight/24h vancomycin respectively. It is noteworthy that no renal impairment could be detected despite high-dose vancomycin therapy. Dose adjustments after 48h were made in four of the patients (three reduced the dose of both drugs and one reduced the therapeutic dose of vancomycin only) according to the results of serum TDM. After the second TDM analysis, another 4 patients underwent dose adjustments, of which the dose was enhanced for 1 patient. After dose adjustment, final cerebrospinal fluid blood levels reached the target values in all 9 patients.

As seen in the comparison of serum and cerebrospinal fluid data based on TDM, respectively, the median value of the C_CSF/C_serum ratio for meropenem was 0.15 (0.06; 0.33), and the median value of the C_CSF/C_serum ratio for vancomycin was 0.07 (0.02; 0.50). As shown in Table 1, vancomycin cerebrospinal fluid concentrations were more variable than meropenem. However, two of the patients (#6, #7) had low C_CSF/C_serum ratios and resulted in insufficient CSF concentrations, whereas vancomycin had low CSF concentrations in only one case (patient #6).

Table 1. Serum and CSF concentration in each patient 24h after infusion start.

In the present study, all but one patient achieved CSF concentrations of meropenem and vancomycin above 1 mg/L with initial dosing. However, there was considerable variation in both serum and cerebrospinal fluid concentrations as well as CSF/serum ratios. This is consistent with previous studies and suggesting the routine TDM to prevent underexposure in meropenem and vancomycin brain therapy.

In the present study, the starting daily dose used to achieve a serum effective therapeutic concentration of 20-30 mg/L exceeded the recommended dose (meropenem 6 g/day and vancomycin 2 g/day), which may be related to the stronger renal function of the patients in the present study compared with the normal creatinine clearance of the patients, which is more common in neurologically critically ill patients.

The evaluation of efficacy was not performed in this study, and it is unclear which pharmacokinetics - pharmacodynamics targets should be used to  optimize therapeutic outcome for intracranial infections. Therefore, this paper cannot conclude whether TDM is potential to improve efficacy. However, in this study, all patients achieved effective MIC concentrations under the guidance of TDM, and it is reasonable to believe that the use of TDM will be helpful to improve the efficacy of antimicrobial therapy in patients with ventribulitis.