Is there evidence to support the use of extended infusion beta lactams for CNS infections?

Comment by InpharmD Researcher

Available evidence supporting extended- or continuous-infusion beta-lactams for central nervous system (CNS) infections remains limited, with insufficient high-quality data demonstrating improved clinical outcomes compared with intermittent infusion. The 2017 Infectious Diseases Society of America (IDSA) guidelines suggest that prolonged meropenem infusion over 3 hours may be effective against resistant Gram-negative organisms in healthcare-associated ventriculitis and meningitis, although this recommendation is based on low-quality evidence. Two randomized controlled trials evaluating prolonged or continuous cefotaxime infusion in pediatric bacterial meningitis found no significant improvements in mortality or neurological outcomes, while a small retrospective study of continuous cefepime infusion in postoperative intracranial infections reported greater cerebrospinal fluid (CSF) exposure and shorter treatment duration, but similar clinical cure rates. Although pharmacokinetic studies and reviews suggest potential improvements in CSF antibiotic exposure with prolonged or continuous infusion, findings remain variable, and further high-quality studies are needed to determine whether these strategies improve clinical outcomes.
Background

The 2017 Infectious Diseases Society of America (IDSA) clinical practice guidelines for healthcare-associated ventriculitis and meningitis recommend vancomycin in combination with an antipseudomonal beta-lactam (cefepime, ceftazidime, or meropenem) for empiric treatment, with antibiotic selection guided by local susceptibility patterns (strong recommendation, low-quality evidence). For infections caused by Gram-negative bacilli, the guidelines recommend selecting agents based on susceptibility testing and adequate central nervous system (CNS) penetration (strong recommendation, moderate-quality evidence). Notably, the guidelines state that prolonged infusion of meropenem, administered over 3 hours per dose, may be successful in treating resistant Gram-negative organisms (weak recommendation, low-quality evidence). The supporting evidence indicates that extended (≥3-hour) or continuous infusions of carbapenems and piperacillin/tazobactam have been associated with improved clinical outcomes compared with standard short-term infusions; however, the guidelines acknowledge that no randomized controlled trials have compared clinically meaningful outcomes between different antimicrobial agents, doses, or treatment durations specifically in healthcare-associated ventriculitis and meningitis. [1]

A 2023 international consensus guideline, endorsed by seven professional organizations, including the IDSA, evaluated the pharmacokinetic/pharmacodynamic (PK/PD) rationale, clinical efficacy, and safety of prolonged-infusion (extended and continuous) beta-lactam antibiotics compared with short infusions. The guideline conditionally recommends prolonged infusions over short infusions to reduce mortality or improve clinical cure in severely ill adults, particularly those with gram-negative infections, although the certainty of evidence was very low. However, recommendations specific to CNS infections were not provided. Among children with bacterial meningitis, a randomized controlled trial comparing continuous versus short-infusion cefotaxime found no differences in mortality or neurologic sequelae, while a population PK analysis of meropenem in infants with late-onset sepsis or meningitis found that increasing infusion duration improved serum PK/PD target attainment but decreased the percentage of time above the minimum inhibitory concentration in cerebrospinal fluid. Given the limited and heterogeneous pediatric evidence, the guideline could not recommend for or against routine prolonged-infusion beta-lactams in specific pediatric clinical settings (conditional recommendation; very low certainty of evidence) and identified CNS antibiotic exposure as an area requiring further investigation. [2]

A 2021 systematic review and a 2022 narrative review evaluated antibiotic pharmacokinetics and infusion strategies in patients with CNS infections, particularly healthcare-associated ventriculitis and meningitis. The systematic review included 58 studies reporting antibiotic concentrations or pharmacokinetic data in the cerebrospinal fluid (CSF) of critically ill patients with uninflamed meninges and identified substantial interpatient and interstudy variability in beta-lactam CSF penetration. High-dose continuous-infusion meropenem (8 to 15 g/day) achieved CSF concentrations of 31.3 to 64 mg/L in one patient with ventriculitis, whereas extended-infusion meropenem (1 to 2 g every 8 hours over 4 hours) produced CSF concentrations of less than 0.2 to 6.2 mg/L in 21 patients. Continuous-infusion cefazolin (8 to 10 g/day) achieved CSF concentrations of 6.1 to 12 mg/L and maintained concentrations above the minimum inhibitory concentration (MIC) throughout the dosing interval in one patient, with reported clinical and bacteriological cure. Additional evidence described high CSF concentrations with continuous-infusion ceftolozane/tazobactam, while continuous-infusion cefepime demonstrated improved simulated CSF pharmacokinetic/pharmacodynamic target attainment under certain dosing conditions. However, evidence supporting prolonged or continuous infusions over intermittent administration for ventriculitis and meningitis remains preliminary, with substantial heterogeneity, limited robust pharmacokinetic analyses, and insufficient clinical outcome data to establish improved efficacy. [3], [4]

Background References: [1] Tunkel AR, Hasbun R, Bhimraj A, et al. 2017 Infectious Diseases Society of America’s clinical practice guidelines for healthcare-associated ventriculitis and meningitis. Clin Infect Dis. 2017;64(6):e34-e65. doi:10.1093/cid/ciw861
[2] Hong LT, Downes KJ, FakhriRavari A, et al. International consensus recommendations for the use of prolonged-infusion beta-lactam antibiotics: endorsed by the American College of Clinical Pharmacy, British Society for Antimicrobial Chemotherapy, Cystic Fibrosis Foundation, European Society of Clinical Microbiology and Infectious Diseases, Infectious Diseases Society of America, Society of Critical Care Medicine, and Society of Infectious Diseases Pharmacists. Pharmacotherapy. 2023;43(8):740-777. doi:10.1002/phar.2842
[3] Kumta N, Roberts JA, Lipman J, Wong WT, Joynt GM, Cotta MO. A systematic review of studies reporting antibiotic pharmacokinetic data in the cerebrospinal fluid of critically ill patients with uninflamed meninges. Antimicrob Agents Chemother. 2021;65(1):e01998-20. doi:10.1128/AAC.01998-20
[4] Karvouniaris M, Brotis A, Tsiakos K, Palli E, Koulenti D. Current perspectives on the diagnosis and management of healthcare-associated ventriculitis and meningitis. Infect Drug Resist. 2022;15:697-721. doi:10.2147/IDR.S326456
Literature Review

A search of the published medical literature revealed 4 studies investigating the researchable question:

Is there evidence to support the use of extended infusion beta lactams for CNS infections?

Level of evidence

C - Multiple studies with limitations or conflicting results  Read more→


Please see Tables 1-4 for your response.

Slow Initial Β-Lactam Infusion and Oral Paracetamol to Treat Childhood Bacterial Meningitis: A Randomised, Controlled Trial
Design

Prospective, randomized, double-blind, placebo-controlled, single-center study with a two-by-two factorial design

N= 723
Objective

To investigate whether slow initial β-lactam infusion and oral paracetamol have benefits in children with bacterial meningitis
Study Groups

Cefotaxime infusion plus paracetamol (n= 183)

Cefotaxime infusion plus placebo (n= 180)

Cefotaxime bolus plus paracetamol (n= 180)

Cefotaxime bolus plus placebo (n= 180)
Inclusion Criteria

Children aged 2 months to 13 years with signs and symptoms of bacterial meningitis and a CSF sample that was cloudy, positive for Gram-staining, or contained more than 50×10⁶/L white blood cells
Exclusion Criteria

Trauma, intracranial shunt, previous neurological disorders, hearing impairment, and immunosuppression except for HIV infection
Methods

Children presenting to Luanda Children's Hospital in Angola were randomized 1:1:1:1 to receive cefotaxime as two 12-hour IV infusions of 125 mg/kg/dose (250 mg/kg over 24 hours), without loading doses, or four intermittent IV boluses totaling 250 mg/kg over 24 hours (every 6 hours). Patients additionally received oral paracetamol (30 mg/kg initially, followed by 20 mg/kg every 6 hours for 48 hours) or placebo. After the initial 24 hours, all patients received intermittent cefotaxime boluses at 250 mg/kg/day for the remainder of treatment. All patients also received oral glycerol four times daily for 2 days. Neurological examinations and audiological assessments were performed during hospitalization, with scheduled follow-up at 1, 3, 6, and 12 months after discharge. Outcomes were analyzed by intention to treat (ITT), with additional per-protocol and predefined pathogen-specific subgroup analyses. Logistic regression compared treatment arms using cefotaxime bolus plus placebo as the reference group.
Duration

July 18, 2005, to June 26, 2008
Outcome Measures

Primary: Death or severe neurological sequelae, deafness

Secondary: Death or any audiological or neurological sequelae
Baseline Characteristics   Cefotaxime infusion with paracetamol (n= 183)

Cefotaxime infusion with placebo (n= 180)

 Cefotaxime bolus with paracetamol (n= 180) Cefotaxime bolus with placebo (n= 180)
Median age, months

17 (1–157)

 15 (2–154) 11 (1–146) 13 (2–146)
Male

107 (58%)

 94 (52%) 90 (50%) 95 (53%)
Weight for age <–2 SD of the median

55 (30%)

 55 (31%) 56 (31%) 64 (36%)
Duration of symptoms >3 days

107/181 (59%)

 129 (72%) 112 (62%) 108/179 (60%)
Previous antibiotics

73/171 (43%)

 62/171 (36%) 64/171 (37%) 72/166 (43%)
Seizures before admission

122 (67%)

 132 (73%) 126 (70%) 126 (70%)
Glasgow coma score

11 (8–15)

 12 (8–15) 12 (8–15) 10 (8–15)
Focal neurological signs

40 (22%)

 43 (24%) 50 (28%) 41 (23%)
Another focus of infection

43 (23%)

 46 (26%) 38 (21%) 55 (31%)
Dyspnoea

90 (49%)

 78 (43%) 75 (42%) 90 (50%)
Dehydration

31 (17%)

 34 (19%) 23 (13%) 28 (16%)

Blood Haemoglobin concentrations, g/L

77 (66–85)

 80 (65–90) 80 (67–90) 74 (60–85)

Blood Glucose concentration, mmol/L

 4·6 (3·3–5·7) 4·7 (3·5–5·9) 4·6 (3·3–6·2) 4·6 (3·3–5·9)

Malaria parasites

58/176 (33%)

 51/168 (30%) 55/175 (31%) 58/175 (33%)

HIV positive

11/143 (8%)

 10/126 (8%) 11/137 (8%) 9/123 (7%)

Tuberculin test ≥10 mm

17/113 (15%)

 18/100 (18%) 12/95 (12%) 8/97 (8%)

Cerebrospinal fluid Leucocyte count, ×106/L

565 (160–1800)

 873 (210–2897) 836 (230–2680) 980 (155–2945)

Cerebrospinal fluid Glucose concentration, mmol/L

0·8 (0·4–1·7)

 0·8 (0·4–1·7) 0·7 (0·3–1·4) 0·9 (0·4–1·8)

Causative agent

Haemophilus influenzae type b

Streptococcus pneumoniae

Neisseria meningitidis

Other 

Unknown

 

41 (22%)

54 (30%)

13 (7%)

8 (4%)

67 (37%)

 

46 (26%)

38 (21%)

16 (9%)

10 (5%)

70 (39%)

 

54 (30%)

39 (22%)

11 (6%)

8 (4%)

68 (38%)

 

47 (26%)

53 (30%)

9 (5%)

6 (6%)

65 (36%)
Results

Among 723 children with suspected bacterial meningitis, 183 received cefotaxime infusion plus paracetamol, 180 received infusion plus placebo, 180 received cefotaxime bolus plus paracetamol, and 180 received bolus plus placebo.

Death or severe neurological sequelae occurred in 45%, 48%, 50%, and 46% of patients, respectively, with no significant differences between groups (all p≥ 0.40); similarly, deafness occurred in 13%, 12%, 11%, and 12%, respectively, without significant differences (all p≥ 0.86).

Overall mortality was 33%, 41%, 38%, and 38%, respectively, with no significant differences between groups (all p≥ 0.32). However, in a predefined subgroup analysis of children with pneumococcal meningitis, cefotaxime infusion plus placebo was associated with lower odds of death or any sequelae compared with bolus plus placebo (odds ratio [OR] 0.18; 95% confidence interval [CI] 0.03–0.90; p= 0.04), whereas infusion plus paracetamol showed a nonsignificant reduction (OR 0.22; 95% CI 0.04–1.09; p= 0.06).

Additionally, a post-hoc analysis demonstrated lower mortality with infusion plus paracetamol versus bolus plus placebo at 48 hours (OR 0.45; 95% CI 0.26–0.77; p= 0.003) and 72 hours (OR 0.52; 95% CI 0.32–0.85; p= 0.009), although no significant improvement in final clinical outcomes was observed.
Adverse Events

No significant adverse events attributable to the study interventions were noted.
Study Author Conclusions

Although no regimen tested in this study proved significantly beneficial in terms of the predefined endpoints, an exploratory, post-hoc analysis showed that a regimen of cefotaxime infusion for 24 h plus paracetamol for 48 h was associated with lowered mortality for at least 3 days, irrespective of the causative agent.
Critique

This randomized, double-blind trial provides direct clinical evidence comparing prolonged versus intermittent cefotaxime administration in pediatric bacterial meningitis, but the 24-hour infusion intervention did not significantly improve the predefined primary outcomes. Although exploratory analyses suggested lower early mortality and fewer deaths or sequelae in selected subgroups, the short infusion duration, concurrent paracetamol intervention, and severely ill pediatric population limit conclusions regarding the clinical benefit of prolonged β-lactam infusion for CNS infections more broadly.
Table 1 References:
[5] Pelkonen T, Roine I, Cruzeiro ML, Pitkranta A, Kataja M, Peltola H. Slow initial -lactam infusion and oral paracetamol to treat childhood bacterial meningitis: a randomised, controlled trial. Lancet Infect Dis. 2011;11(8):613-621. doi:10.1016/S1473-3099(11)70055-X

Extended Continuous β-Lactam Infusion With Oral Acetaminophen in Childhood Bacterial Meningitis: A Randomized, Double-blind Clinical Trial
Design

Randomized, double-blind, parallel-group study

N= 375
Objective

To investigate whether extending the treatment regimen of continuous β-lactam infusion with oral acetaminophen to 4 days would improve outcomes in childhood bacterial meningitis
Study Groups

Intervention group (n= 187)

Control group (n= 186)
Inclusion Criteria

Children aged 2 months to 15 years presenting with symptoms and signs of bacterial meningitis, with cloudy CSF, positive Gram stain, or ≥50 leukocytes/mm3
Exclusion Criteria

Previous known neurological abnormalities or hearing impairment, immunosuppression except for HIV, active tuberculosis, known hepatic disease, or pretreatment with >1 dose of parenteral antibiotic
Methods

Children received cefotaxime 250 mg/kg/day as continuous infusion with oral acetaminophen (first dose 30 mg/kg, then 20 mg/kg every 6 hours) or cefotaxime as boluses (62.5 mg/kg, 4 times per day) with placebo orally. All received oral glycerol
Duration

January 2012 to January 2017
Outcome Measures

Primary: Mortality by day 7

Secondary: In-hospital mortality, death or severe neurological sequelae, death or any neurological sequelae
Baseline Characteristics  

Bolus + Placebo (n= 186)

 Infusion + Acetaminophen (n= 187)
Age, months (range)

31 (1–161)

 28 (2–177)
Male

115 (62%)

 101 (54%)
Weight for age < −2 SD

48/185 (26%)

 48/186 (26%)
Duration of symptoms > 3 d

116/181 (64%)

 136/184 (74%)
Previous antibiotics

73/165 (44%)

 95/170 (56%)

Seizures before admission

122/175 (70%)

 116/178 (65%)

Glasgow Coma Scale score, median (IQR)

11 (8–15)

 12 (9–15)
Focal neurological signs

53/181 (29%)

 60/176 (34%)
Another focus of infection

46/169 (27%)

 62/178 (35%)
Dyspnea

111/177 (63%)

 100/181 (55%)
Dehydration

43/183 (23%)

 46/182 (25%)
Hb, g/L, median (IQR)

73 (60–92)

 74 (62–89)
Leukocyte count, ×109/L, median (IQR)

15.1 (9.0–21.0)

 13.7 (8.8–19.7)
CRP, mg/L, median (IQR)

160 (96–160)

 160 (91 to > 160)
Malaria test positive

51/186 (27%)

 40/187 (21%)
HIV test positive

4/98 (4%)

 11/93 (12%)
CSF Leukocyte count, ×106/L

495 (130–2770)

 800 (140–2289)
CSF Protein level, mg/dL

197 (123–265)

 187 (110–250)
CSF Glucose level, mmol/L

0.9 (0.5–2.7)

 1.1 (0.5–2.4)
Streptococcus pneumoniae

38 (20%)

 35 (19%)
Neisseria meningitidis

21 (11%)

 16 (9%)
Haemophilus influenzae

4 (2%)

 3 (2%)
Other bacteria

16 (9%)

 23 (12%)
Unknown

107 (58%)

 110 (59%)

Abbreviations: IQR, interquartile range; CRP, C-reactive protein; Hb, hemoglobin
Results  

Infusion + Acetaminophen, % (n= 187)

 Bolus + Placebo, % (n= 186) Risk Ratio (95% CI) Absolute Risk Difference, % (95% CI)

Day 7 death

32.6 (61/187)

 34.4 (64/186) 0.95 (.71–1.26) 1.8 (−7.8 to 11.4)

Death at discharge

38.0 (71/187)

 40.3 (75/186) 0.94 (.73–1.21) 2.3 (−7.6 to 12.2)

Severe neurological sequelae or death, day 7

46.5 (80/172)

 43.4 (75/173) 1.07 (.85–1.36) 3.1 (−7.4 to 13.6)

Severe neurological sequelae or death, at discharge

48.4 (90/186)

 46.4 (85/183) 1.04 (.84–1.29) 2.0 (−8.2 to 12.2)

Any neurological sequelae or death, day 7

57.8 (96/166)

 51.2 (86/168) 1.13 (.93–1.38) 6.6 (−4.1 to 17.3)

Any neurological sequelae or death, at discharge

57.5 (104/181)

 49.7 (89/179) 1.16 (.95–1.40) 7.8 (−2.5 to 18.1)
Adverse Events

No significant differences in adverse events between groups. Potential adverse events were registered as part of normal follow-up.
Study Author Conclusions

Prolonged continuous β-lactam infusion combined with oral acetaminophen did not improve the outcomes of childhood bacterial meningitis in Angola.
Critique

This randomized, double-blind trial provides direct clinical evidence comparing continuous versus intermittent cefotaxime administration in pediatric bacterial meningitis, with no significant improvements in mortality or neurological outcomes. However, concurrent acetaminophen administration only in the continuous-infusion group prevents isolation of the infusion strategy's independent effect, while the high proportion of unconfirmed meningitis cases, smaller per-protocol population, and limited audiological assessments further constrain interpretation.
Table 2 References:
[6] Savonius O, Rugemalira E, Roine I, Cruzeiro ML, Peltola H, Pelkonen T. Extended Continuous -Lactam Infusion With Oral Acetaminophen in Childhood Bacterial Meningitis: A Randomized, Double-blind Clinical Trial. Clin Infect Dis. 2021;72(10):1738-1744. doi:10.1093/cid/ciaa341

Continuous Versus Intermittent Infusion of Cefepime in Neurosurgical Patients with Post-Operative Intracranial Infections
Design

Retrospective, non-randomised, comparative study

N= 68
Objective

To evaluate the clinical efficacy of continuous infusion therapy of cefepime compared to intermittent infusion in neurosurgical patients with post-operative intracranial infections
Study Groups

Continuous infusion (n= 34)

Intermittent infusion (n= 34)
Inclusion Criteria

Patients with Gram-negative or Gram-positive intracranial infections following neurosurgical operations who received initial empirical antibiotic therapy with cefepime over a 2-year period (October 2010–November 2012)
Exclusion Criteria

Patients aged <18 years, patients with hypersensitivity/allergy to β-lactam antibiotics, and patients with renal dysfunction (dialysis or creatinine clearance <40 mL/min)
Methods

Medical records of eligible neurosurgical patients treated at Fujian Provincial Hospital, China, were retrospectively reviewed. Patients received either CI cefepime (0.5 g IV loading dose over 30 minutes, followed by 4 g IV over 24 hours) or II cefepime (2 g IV over 30 minutes every 12 hours), as determined by the treating physician. Plasma and CSF samples were collected at 0.5, 1, 2, 4, 6, 8, 12, 16, 20, and 24 hours after infusion, and cefepime concentrations were measured using high-performance liquid chromatography. Minimum inhibitory concentrations (MICs) were determined by microdilution, while area under the concentration–time curve (AUC) and percentage of time cefepime concentrations exceeded the MIC (%T>MIC) were calculated. Clinical outcomes were independently assessed by 3 neurosurgeons blinded to treatment allocation. Clinical cure required normalization of temperature and meningeal signs, improvement in CSF parameters, and 3 consecutive negative CSF cultures; failure was defined as persistent or progressive meningeal irritation or positive CSF cultures. Cefepime was discontinued upon clinical cure.
Duration

October 2010 to November 2012
Outcome Measures

Primary: Clinical cure of intracranial infection

Secondary: Plasma and CSF cefepime concentrations, %T>MIC, AUC
Baseline Characteristics

 

 CI (n= 34) II (n= 34) P-value

Age, years

 47.2 ± 12.1 48.4 ± 13.5 0.685

Female

 20/14 14/20 0.146

Tumour location (supratentorial/infratentorial), n

 19/15 15/19 0.332

Cefepime treatment duration, days

 6.6 ± 1.9 7.8 ± 2.6 0.036
Results

 

 CI (n= 34) II (n= 34) P-value

Cefepime treatment duration, days

 6.6 ± 1.9 7.8 ± 2.6 0.036

%T>MIC for plasma, MIC 8 µg/mL

 100% 75%  

AUC for plasma, µg h/mL

 1197.99 ± 72.15 890.84 ± 140.78 0.655

%T>MIC for CSF, MIC 4 µg/mL

 83.3% 25%  

%T>MIC for CSF, MIC 8 µg/mL

 75% 0%  

AUC for CSF, µg h/mL

 220.56 ± 13.59 86.34 ± 5.69 0.003
All 68 patients achieved clinical cure with continuous infusion (CI) or intermittent infusion (II) cefepime; however, treatment duration was significantly shorter with CI (6.6 ± 1.9 vs 7.8 ± 2.6 days; p= 0.036).

Mean plasma cefepime exposure (AUC₀.₅–₂₄) was 1197.99 ± 72.15 µg·h/mL with CI versus 890.84 ± 140.78 µg·h/mL with II (p= 0.655), while plasma concentrations exceeded an MIC of 8 µg/mL for 100% versus 75% of the dosing interval, respectively.

Mean CSF exposure was significantly higher with CI than II (220.56 ± 13.59 vs 86.34 ± 5.69 µg·h/mL; p=0.003), with CSF concentrations exceeding MICs of 4 and 8 µg/mL for 83.3% and 75% of the dosing interval with CI versus 25% and 0% with II, respectively.

CSF penetration (AUCCSF/AUCplasma) was also greater with CI (18.4% vs 9.7%).

No adverse effects related to cefepime or its route of administration were observed in either group during treatment.
Adverse Events

No adverse effects related to the drug or route of administration were observed in any patient during the study period.
Study Author Conclusions

We conclude that continuously infusing cefepime consistently generates concentrations in plasma and CSF above 4 μg/mL and may produce more reliable plasma and CSF drug concentrations in neurosurgical patients with post-operative intracranial infections than intermittent infusion.
Critique

This study provides direct clinical and pharmacokinetic evidence supporting continuous cefepime infusion for postoperative intracranial infections, demonstrating significantly greater CSF exposure and shorter treatment duration compared with intermittent infusion, although all patients achieved clinical cure. However, the retrospective, nonrandomized design, small sample size, physician-directed treatment allocation, and exclusion of patients with significant renal dysfunction limit the strength and generalizability of the findings.
Table 3 References:
[7] Huang H, Huang S, Zhu P, Xi X. Continuous versus intermittent infusion of cefepime in neurosurgical patients with post-operative intracranial infections. Int J Antimicrob Agents. 2014;43(1):68-72. doi:10.1016/j.ijantimicag.2013.08.019

Meropenem Pharmacokinetics in Cerebrospinal Fluid: Comparing Intermittent and Continuous Infusion Strategies in Critically Ill Patients—A Prospective Cohort Study
Design

Prospective, observational study

N= 16
Objective

To evaluate whether the choice of administration type (continuous infusion vs intermittent infusion) influences the concentrations and target attainment of meropenem in cerebrospinal fluid
Study Groups

Continuous infusion (n= 8)

Intermittent infusion (n= 8)
Inclusion Criteria

Adult patients with an external ventricular drain recruited from August 2020 to September 2021
Exclusion Criteria

Not specified
Methods

Patients received meropenem either as a standard short infusion (2 g every 8 h over 30 min) or as a continuous infusion (250 mg/h). Dosing regimens were adapted based on therapeutic drug monitoring results. Plasma and CSF samples were collected and analyzed using LC-MS/MS. Population pharmacokinetic modeling was performed using NONMEM 7.5.
Duration

August 2020 to September 2021
Outcome Measures

Primary: Concentrations and probability of target attainment of meropenem in CSF

Secondary: Correlation of interleukin-6 in CSF with penetration rate
Baseline Characteristics  

Overall (N= 16)

 Continuous infusion (n= 8) Intermittent infusion (n= 8)
Primary CNS infection

9 (56%)

 5 (63%) 4 (50%) >0.9

Infection

Brain abscess

Meningitis

Mycotic aneurysm

Pneumonia

Ventriculitis

 

1 (6.3%)

1 (6.3%)

1 (6.3%)

7 (44%)

6 (38%)

 

0

0

1 (13%)

3 (38%)

4 (50%)

 

1 (13%)

1 (13%)

0

4 (50%)

2 (25%)

 

0.7

0.7

0.7

0.7

0.7

Female

3 (19%)

 2 (25%) 1 (13%) >0.9

Meropenem daily dose, g

6 (3–13)

 7 (5–13) 6 (3–10) 0.2

Meropenem plasma concentration, mg/L

12.2 (0.1–66.0)

 14.7 (1.0–50.7) 5.1 (0.1–66.0) <0.001

Meropenem CSF concentration, mg/L

0.6 (0.0–9.0)

 0.6 (0.0–6.9) 0.5 (0.0–9.0) 0.5

Age, years

59 (22–77)

 46 (22–77) 65 (37–76) 0.2

Height, cm

175 (161–190)

 175 (165–190) 177 (161–183) 0.7

Weight, kg

78 (62–153)

 78 (65–153) 79 (62–107) >0.9

Serum creatinine, µmol/L

61.9 (26.5–185.6)

 57.5 (26.5–123.8) 61.9 (44.2–185.6) 0.7

Urea, mmol/L

7.3 (2.5–11.7)

 7.9 (2.5–11.7) 6.8 (3.0–10.9) 0.7

eGFR (CG formula), mL/min

120 (29–248)

 193 (55–248) 111 (29–219) 0.3

Bilirubin, µmol/L

5.1 (3.4–59.9)

 5.1 (3.4–59.9) 5.1 (3.4–6.8) 0.4

C-reactive protein, mg/L

75.0 (2.0–155.0)

 80.5 (2.0–155.0) 60.5 (31.0–114.0) 0.5
Leukocyte count, G/L

9.6 (5.6–15.9)

 9.8 (5.6–15.9) 9.2 (7.6–11.7) 0.5

Interleukin-6, pg/mL

18.7 (3.5–84.7)

 24.7 (3.5–84.7) 17.8 (6.2–35.5) 0.6

Parameters in CSF

Neuron-specific enolase (ng/mL)

Protein S100 (ng/mL)

Ferritin (µg/L)

Erythrocyte count (G/L)

Protein (g/L)

Total cell count (G/L)

Glucose (mmol/L)

Interleukin-6 (pg/mL)

Lactate (mmol/L)

 

18.6 (9.6–600.0)

6.8 (0.6–30.0)

515.0 (76.1–655.0)

6.5 (0.0–111.0)

0.7 (0.0–2.3)

0.3 (0.0–2.4)

3.4 (0.8–5.0)

1,515 (7–13,799)

0.2 (0.2–0.6)

 

15.7 (9.6–134.0)

6.8 (0.6–30.0)

411.0 (76.1–607.0)

4.0 (0.0–22.0)

0.7 (0.1–1.3)

0.1 (0.0–0.8)

3.3 (1.6–4.6)

1,000 (27–13,799)

0.5 (0.2–0.6)

 

21.5 (15.7–600.0)

6.8 (3.2–30.0)

515.0 (307.0–655.0)

13.0 (0.0–111.0)

0.7 (0.0–2.3)

0.3 (0.0–2.4)

4.0 (0.8–5.0)

2,969 (7–9,723)

0.2 (0.2–0.6)

 

0.3

0.9

0.3

0.6

0.8

0.2

0.6

0.8

0.12
Results
All 68 patients achieved clinical cure with continuous infusion (CI) or intermittent infusion (II) cefepime, although treatment duration was significantly shorter with CI (6.6 ± 1.9 vs 7.8 ± 2.6 days; p= 0.036).
 
Mean plasma cefepime exposure (AUC₀.₅–₂₄) was higher with CI than II (1197.99 ± 72.15 vs 890.84 ± 140.78 µg·h/mL; p= 0.655), with plasma concentrations exceeding an MIC of 8 µg/mL for 100% versus 75% of the dosing interval, respectively.
 
Mean CSF exposure was significantly higher with CI (220.56 ± 13.59 vs 86.34 ± 5.69 µg·h/mL; p= 0.003), with CSF concentrations exceeding MICs of 4 and 8 µg/mL for 83.3% and 75% of the dosing interval, respectively, compared with 25% and 0% with II.
 
CSF penetration was also greater with CI (18.4% vs 9.7%).
 
No drug- or administration-related adverse effects were observed in either group during treatment.
Adverse Events

Not specified
Study Author Conclusions

We conclude that continuously infusing cefepime consistently generates concentrations in plasma and CSF above 4 μg/mL and may produce more reliable plasma and CSF drug concentrations in neurosurgical patients with post-operative intracranial infections than intermittent infusion.
Critique

This study provides direct clinical and pharmacokinetic evidence supporting continuous cefepime infusion for postoperative intracranial infections, demonstrating significantly greater CSF exposure and shorter treatment duration compared with intermittent infusion, although all patients achieved clinical cure. However, the retrospective, nonrandomized design, small sample size, physician-directed treatment allocation, and exclusion of patients with significant renal dysfunction limit the strength and generalizability of the findings.
Table 4 References:
[8] Wicha SG, Kinast C, Mnchow M, et al. Meropenem pharmacokinetics in cerebrospinal fluid: comparing intermittent and continuous infusion strategies in critically ill patients-a prospective cohort study. Antimicrob Agents Chemother. 2024;68(9):e0045124. doi:10.1128/aac.00451-24