Advertisement

Adult community-acquired bacterial meningitis requiring ICU admission: epidemiological data, prognosis factors and adherence to IDSA guidelines

  • H. Georges
  • A. Chiche
  • S. Alfandari
  • P. Devos
  • N. Boussekey
  • O. Leroy
Article

Abstract

Numerous guidelines are available to guide empirical antimicrobial therapy (EAT) in acute bacterial meningitis (ABM) patients. We analysed prognosis factors and compliance to the Infectious Diseases Society of America (IDSA) guidelines in ABM patients requiring stay in an intensive care unit (ICU). A 10-year retrospective study, using prospectively collected data, in 82 ABM patients admitted to a 16-bed university-affiliated French ICU was undertaken. Seventeen patients (20.7%) died during ICU stay. Multivariate analysis isolated four factors associated with in-ICU death: alcoholism (P = 0.007), acute kidney injury (P = 0.006), age >60 years (P = 0.006) and ICU admission for neurological failure (P = 0.01). Causative pathogens were isolated for 62 (75.6%) patients, including 29 pneumococci, 14/28 of which were non-susceptible to penicillin. No characteristics, particularly recent hospitalisation and/or antibiotic delivery, was associated with penicillin susceptibility. Compliance to IDSA guidelines was 65%. Non-compliance concerned to be essentially the non-delivery or low dosage of vancomycin. Treatment compatible with IDSA guidelines was associated with a decreased ICU mortality in univariate (61.5% survival vs. 35.3%, P = 0.05) but not in multivariate analysis. In-ICU mortality associated with ABM remains high. Prognosis factors are related to the severity of disease or underlying conditions. Penicillin non-susceptible Streptococcus pneumoniae can occur without any of the usual predisposing factors.

Keywords

Vancomycin Intensive Care Unit Admission Intensive Care Unit Stay Glasgow Outcome Scale Glasgow Coma Score 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Acute bacterial meningitis (ABM) remains a serious challenging condition where early diagnosis and appropriate delivery of antibiotic therapy are a major concern to improve prognosis. Intensive care unit (ICU) admission is required when ABM leads to septic shock or neurological or respiratory failure [1, 2, 3]. However, few studies reporting epidemiological data, prognosis factors and long-term outcome of ICU patients with ABM are available, probably because ABM is an uncommon disease with a low ICU admission incidence rate. Thus, data about adjunctive or supportive therapy (steroids, mechanical ventilation, osmotherapy, cerebral haemodynamics monitoring), empirical antibiotic regimen and its timing of administration, Streptococcus pneumoniae strains that are non-susceptible to penicillin (PNSP) and outcome remains poorly studied in this population.

Urgent delivery of antimicrobial therapy is associated with favourable outcome. However, the recommended empirical antimicrobial therapy (EAT) vary in different guidelines, depending on the year of publication and the antimicrobial resistance of leading strains causing ABM, mainly S. pneumoniae [4, 5, 6, 7, 8]. For instance, among European guidelines, some recommend, for suspected pneumococcal meningitis, monotherapy with a beta-lactam (amoxicillin or cephalosporin), while others support a vancomycin–cephalosporin combination for suspected PNSP [4, 5, 6, 8]. The recent IDSA guidelines recommended a systematic adjunction of vancomycin to treat suspected pneumococcal meningitis [7].

The aims of our study were: (1) to determine factors associated with in-ICU mortality in adult ABM patients, (2) to identify PNSP risk factors supporting the selective addition of empirical vancomycin, (3) to evaluate the impact of the IDSA guidelines on empirical antimicrobial administration in a cohort of patients with ABM requiring ICU admission.

Materials and methods

Patients

A 10-year retrospective study, using prospectively collected data from January 1998 to December 2007, was performed in our 16-bed university-affiliated ICU (University Infectious Diseases Department set in a General Hospital).

Patients were included in the study if they were older than 15 years and 3 months and if they presented community ABM. ABM was diagnosed if the following diagnostic criteria were met: compatible clinical signs and a positive culture of cerebrospinal fluid (CSF) and/or positive gram stain and/or CSF antigen test results positive for Neisseria meningitidis or S. pneumoniae or a positive blood culture with CSF pleocytosis (leucocyte count >10 cells/mm3). Patients with non-documented bacterial meningitis were included when a compatible clinical picture was associated with CSF pleocytosis (leucocyte count >10 cells/mm3 and protein concentration >0.6 g/L, or a CSF/serum glucose ratio <0.4). ABM was considered as community-acquired if patients had not been previously treated in a hospital during the previous four weeks. ICU admission depended on the clinical criteria established by the attending ICU physician. Patients with mycobacterial meningitis were excluded from the study.

Microbiology

Pathogens were identified using standard microbiological methods. S. pneumoniae and N. meningitidis isolates were tested for susceptibility to penicillin, amoxicillin and cefotaxime by E-test using the 1997 National Committee for Clinical Laboratory Standards (NCCLS) guidelines for breakpoints [9]: penicillin G ≤0.06 µg/ml = susceptible, 0.12 to 1 µg/ml = intermediate, ≥2 µg/ml = resistant; amoxicillin and cefotaxime: ≤0.5 µg/ml = susceptible, 1 µg/ml = intermediate, ≥2 µg/ml = resistant (NCCLS). For the purpose of the study, any pneumococcal strains that were not fully susceptible to penicillin (minimum inhibitory concentration [MIC] <0.12 µg/ml) were considered as non-susceptible.

Recorded data

The following data were recorded for all patients.

Patient characteristics

Age, sex, presence of diabetes mellitus, alcoholism, cirrhosis, cancer, immunosuppression (defined as one of the following: HIV infection, asplenia, long-term use of steroids and/or other suppressive medication), previous episode of meningitis, hospitalisation within 6 months before admission for ABM.

Clinical and laboratory data

  • Reasons for ICU admission: purpura, respiratory failure, neurological failure, severe sepsis or septic shock. Respiratory failure was defined by arterial hypoxaemia (PaO2/FIO2 <300) and/or respiratory rate >30/min and/or the need for mechanical ventilation. Neurological failure was defined by the presence of a Glasgow Coma Score (GCS) ≤12 and/or seizures and/or focal neurologic deficit [10]. Severe sepsis and septic shock were defined following previously established criteria [11].

  • On ICU admission and in the following 24 h, the GCS, SAPS II, organ dysfunction variables (PaO2/FIO2 <300 for lung dysfunction, acute oliguria and/or serum creatinin >20 mg/l for acute kidney injury, INR >1.5 for coagulation abnormalities, platelet count <100,000 µL−1 for thrombocytopaenia and plasma total bilirubin >40 mg/l for liver dysfunction), core temperature, C-reactive protein, platelet count, serum leukocyte count, serum creatinin, CSF leukocyte count, CSF protein level, direct examination on CSF, blood culture results, pathogens susceptibility, beta-lactam MICs for S. pneumonia and the results of cerebral computed tomography (CT) scanning [11, 12].

  • Neurologic status was determined at the end of the ICU stay using the Glasgow Outcome Scale (GOS) [13].

Antimicrobial therapy

  • We recorded antibiotic use:
    • In the previous six months

    • Home administration of antibiotics by the general practitioner for the current infection, just before hospital admission

    • EAT on admission in the emergency department (ED) when patients were treated in the ED before ICU admission

    • EAT on ICU admission

  • EAT was considered to be compliant to IDSA guidelines when presumptive pathogen identification by positive gram stain, patient age, specific predisposing condition and recommended dosages [8] strictly followed the guidelines (including patients receiving IDSA-compatible EAT before the guidelines’ publication) [8].

  • EAT was considered to be appropriate when causative pathogens were susceptible to at least one prescribed antibiotic.

  • EAT was considered to be effective when the patient’s medical condition improved in the 48 h following ICU admission (increase of GCS and fever and vasopressor requirements decrease).

  • Delivery, total daily dose and the first serum levels of vancomycin were recorded, as well as the use and dosage of third-generation cephalosporins. The first serum level of vancomycin was performed three days after its administration.

Adjuvant therapy

The use of dexamethasone (DXM), just before or during the first administration of EAT, as recently proposed in the IDSA guidelines, was recorded [8].

Statistical analysis

All statistical analyses were performed using SAS Software Release 8.2. All results are expressed as mean ± standard deviation (SD) for continuous variables or as a number (with percentages) for categorical variables. A univariate analysis was performed in order to identify the characteristics associated with adverse clinical outcome and the predisposing factors for the isolation of S. pneumoniae with reduced susceptibility to penicillin. The Chi-square test or the Fisher’s exact test was used for contrasting categorical data. Continuous variables were compared using Student’s t-test or the Wilcoxon rank-sum test. Some continuous variables were categorised into classes by selecting the best cut-offs (receiver operating characteristic [ROC] analysis, maximisation of the Chi-square). The level of significance was set at P < 0.05. Those variables with a P-value of 0.15 or less were then included in a multiple logistic regression analysis for the identification of independent prognostic factors associated with unfavourable outcomes.

Results

Patients and microbiological data

Eighty-two patients were studied. Among them, 26 (31.7%) had been admitted after the IDSA guidelines’ publication. Twelve patients (14.6%) had been directly admitted to the ICU. The characteristics of the patients on ICU admission are reported in Table 1. Causative pathogens were retrieved for 62 (75.6%) patients: S. pneumoniae n = 29, N. meningitidis n = 21, Listeria monocytogenes n = 6, Staphylococcus aureus n = 3, Escherichia coli n = 1, Haemophilus influenzae n = 1, S. bovis n = 1. Blood cultures were positive for 30 (36.5%) patients: S. pneumoniae n = 21, N. meningitidis n = 3, S. aureus n = 3, L. monocytogenes n = 1, H. influenzae n = 1, S. bovis n = 1. Susceptibility to antibiotics was determined for 61 pathogens. The mean duration of ICU stay was 8.3 ± 8.9 days. For the 34 mechanically ventilated patients, mean duration of mechanical ventilation was 10.9 ± 10.3 days.
Table 1

Clinical and biological characteristics on intensive care unit (ICU) admission in 82 patients with acute bacterial meningitis (ABM)

Characteristic

No. of meningitis episodes

Age in years, mean (range)

49.1 (15–85)

Male gender, no. (%)

46 (56.1)

Comorbidity, no. (%)

 Diabetes

18 (21.9)

 Alcoholism

16 (19.5)

 Cirrhosis

6 (7.3)

 Cancer

8 (9.7)

 Immunodepression

12 (14.6)

Previous meningitis, no. (%)

10 (12.2)

Hospital admission in the previous six months, no. (%)

14 (17.1)

SAPS II, mean (SD)

33.7 (19.1)

Reason(s) for ICU admission*

 Purpura

20 (24.4)

 Respiratory failure

14 (17.1)

 Neurological failure

51 (62.2)

 Severe sepsis

64 (78.1)

 Septic shock

14 (17.1)

Seizures, no. (%)

8 (9.7)

Mean core temperature (°C), mean (SD)

38.6 (1.3)

Glasgow Coma Score, mean (SD)

11.3 (2.9)

Mechanical ventilation, no. (%)

37 (45.1)

Organ dysfunction**, no. (%)

60 (73.2)

 Respiratory, no. (%)

20 (24.4)

 Coagulation, no. (%)

7 (8.5%)

 Kidney, no. (%)

10 (12.2)

 Liver, no. (%)

2 (2.4)

 Platelets, no. (%)

12 (14.6)

Cerebral CT scan, no. (%)

61 (74.4)

Laboratory data

 CRP (mg/l), mean (SD)

205.4 (125.4)

 Leukocytes count in CSF (cells/mm3), mean (SD)

4,340 (6413)

 Protein in CSF (g/l), mean (SD)

3.06 (2.5)

 Serum creatinin (mg/l), mean (SD)

12.6 (7.6)

 Positive direct examination on CSF, no. (%)

49 (60.4)

 Positive blood culture, no. (%)

30 (36.5)

SAPS = Simplified Acute Physiological Score; CSF = cerebrospinal fluid

*Patients could be admitted for many reasons

**Some patients could have multiple organ dysfunctions

Outcome and prognosis factors

Seventeen patients (20.7%) died during ICU stay. Vegetative state (GOS = 2) and severe disability (GOS = 3) at ICU discharge were present for one and two patients respectively. Univariate and multivariate analysis of the relationship between baseline characteristics and mortality is reported in Table 2. Multivariate analysis identified four variables associated with in-ICU death: alcoholism, acute kidney injury, age >60 years and ICU admission for neurologic failure.
Table 2

Univariate and multivariate analysis of baseline factors associated with in-ICU mortality in ABM

 

Survivors (n = 65)

Non-survivors (n = 17)

P-value

Odds ratio

95% confidence interval

P-value

Age in years, mean (SD)

46.7 (20.2)

58.2 (18)

0.05

   

Age >60 years, no. (%)

16 (24.6)

11 (64.7)

0.001

8.3

1.8–38.9

0.006

Male gender, no. (%)

36 (55.3)

10 (58.8)

0.79

   

Diabetes, no. (%)

12 (18.4)

6 (35.3)

0.13

   

Alcoholism, no. (%)

9 (13.8)

7 (41.1)

0.01

9.4

1.8

48.1

Cirrhosis, no. (%)

2 (3)

4 (23.5)

0.003

   

Cancer, no. (%)

4 (6.1)

4 (23.5)

0.03

   

Immunodepression, no. (%)

8 (12.3)

4 (23.5)

0.24

   

Previous meningitis, no. (%)

9 (13.8)

1 (5.9)

0.37

   

Hospital admission in the previous six months, no. (%)

8 (12.3)

6 (35.3)

0.02

   

SAPS II, mean (SD)

28.8 (14.6)

52.8 (22.4)

0.001

   

SAPS II >40, no. (%)

16 (24.6)

11 (64.7)

0.001

   

Reason for ICU admission*, no. (%)

 Purpura

18 (27.7)

2 (11.7)

0.17

   

 Respiratory failure

8 (12.3)

6 (35.3)

0.02

   

 Neurological failure

36 (55.3)

15 (88.2)

0.01

11.9

1.6–89.1

0.01

 Severe sepsis

49 (75.4)

15 (88.2)

0.45

   

Septic shock

8 (12.3)

6 (35.3)

0.02

   

Seizures, no. (%)

5 (76.9)

3 (17.6)

0.21

   

Glasgow Coma Score, mean (SD)

11.8 (2.6)

9.3 (3.2)

0.004

   

Glasgow Coma Score < 11, no. (%)

30 (46.1)

11 (64.7)

0.01

   

Mechanical ventilation, no. (%)

21 (32.3)

13 (76.4)

0.001

   

Respiratory dysfunction, no. (%)

15 (23)

5 (29.4)

0.01

   

Coagulation dysfunction, no. (%)

6 (9.2)

1 (5.8)

0.66

   

Kidney dysfunction, no. (%)

5 (7.7)

5 (29.4)

0.01

   

Liver dysfunction, no. (%)

0

2 (11.7)

0.005

   

Platelets dysfunction, no. (%)

6 (9.2)

6 (35.3)

0.006

   

Abnormal cerebral CT scan (n = 61), no. (%)

13 (27.6)

4 (28.6)

0.75

   

Core temperature (°C), mean (SD)

38.7 (1.2)

38.6 (1.8)

0.71

   

CRP (mg/l), mean (SD)

217.8 (127.3)

145.8 (98.6)

0.05

   

CRP > 150 mg/, no. (%)

19 (29.2)

9 (52.9)

0.008

   

Platelets counts (×109 cells/l), mean (SD)

207.9 (104.6)

157.2 (124)

0.06

   

Serum creatinin (mg/l), mean (SD)

11.7 (6.6)

16.0 (9.9)

0.11

   

Serum creatinin > 20 mg/l, no. (%)

5 (7.7)

5 (29.4)

0.01

23.07

2.4–221.5

0.006

Leucocytes count in CSF (cells/mm3), mean (SD)

4,855 (6,755)

2,278 (4,391)

0.03

   

Protein in CSF (g/l), mean (SD)

2.98 (2.51)

3.4 (2.3)

0.4

   

Positive direct examination on CSF, no. (%)

37 (56.9)

12 (70.5)

0.18

   

Positive blood culture, no. (%)

25 (38.4)

5 (29.4)

0.46

   

Micro-organism in CSF, no. (%)

0.17

   

 S. pneumoniae

23

6

    

 N. meningitidis

19

2

    

 L. monocytogenes

3

3

    

Other

4

2

    

S. pneumoniae meningitis

Among the 29 isolated S. pneumoniae strains, 14 were PNSP and one unknown (positive CSF pneumococcal antigen alone). Among the 14 PNSP, the MICs of penicillin were intermediate for eight strains and resistant for six strains. The MICs of amoxicillin were ≤0.5 and 1 mg/l for five and nine strains respectively. The MICs of cefotaxime were ≤0.5 and 1 mg/l for nine and five strains, respectively. No strains had an MIC >1 mg/l for cefotaxime. Research of factors associated with decreased penicillin susceptibility yielded non-significant results (Table 3). Mortality was similar in the two groups.
Table 3

Predisposing factors for penicillin non-susceptible pneumococci in 28 patients with meningitis

 

Penicillin-susceptible S. pneumoniae (n = 14)

Penicillin non-susceptible S. pneumoniae (n = 14)

P-value

Age in years, mean (range)

59.5 ± 13.9

56.1 ± 16.8

0.2

Sex, male/female, no.

7/7

9/5

0.44

Comorbidity, no.

 MacCabe RF, UF, NF

1/1/12

0/5/9

 

 Diabetes

6

2

0.09

 COPD

3

0

0.07

 Alcoholism

5

1

0.06

 Cirrhosis

2

1

0.54

 Cancer

1

0

0.3

 Immunodepression

1

3

0.28

Previous meningitis, no.

4

2

0.35

Hospitalisation in the previous six months, no.

1

3

0.28

SAPS II, mean (range)

37.9 ± 15.1

35.1 ± 17.4

0.25

Positive blood culture

11

10

0.66

Severe sepsis

10

9

0.89

Septic shock

3

1

0.28

Mechanical ventilation

7

6

0.7

Antibiotics started at home for meningitis

3

7

0.11

Antibiotics in the previous six months

1

3

0.28

Deaths

4

1

0.13

RF, UF, NF = rapidly fatal, ultimately fatal, non-fatal; COPD = chronic obstructive pulmonary disease

Antimicrobial therapy

Before hospital admission

Ten patients (12.2%) had received antimicrobial therapy in the previous six months. Thirty-four patients (41.5%) received antimicrobial therapy at a mean of 3 ± 2.6 days before ICU admission, when meningitis was not yet present or not diagnosed by the patient’s physician.

On emergency department admission

EAT was started in the ED for 70 patients. Treatment was similar to the IDSA recommendations for 44/70 (62.8%) patients. EAT was appropriate in 51/56 (91.1%) patients with definitive bacterial documentation.

On ICU admission

On ICU admission (12 direct and 70 for the ED), EAT was similar to the IDSA recommendations for 46/82 (56.1%) patients. Reasons for divergence were:
  • Antibiotics excess in 14 patients (vancomycin n = 3, amoxicillin n = 7, gentamicin n = 5, levofloxacin n = 2)

  • Lack of one or more recommended antibiotics in 13 patients (vancomycin n = 13, third-generation cephalosporin n = 3, amoxicillin n = 2, gentamicin n = 2)

  • Insufficient vancomycin dosage in nine patients

However, EAT was appropriate for 60 of 61 patients with adequate bacterial documentation.

Vancomycin

EAT in the ICU includes vancomycin in 49 patients (59.7%). The total daily dose varied from 20 to 70 mg/kg and was inferior to the recommended dose of 30 mg/kg in 15 patients (30.6%). The serum level on day 3 was performed in 34 patients. The mean serum level was 25.2 ± 9.5 mg/l. A level ≥15 mg/l and ≥20 mg/l was obtained for 29 (85.2%) and 25 patients (73.5%), respectively. Among patients with pneumococcal meningitis, 25 patients (86.2%) received vancomycin. A serum level has been performed for 19 patients (12 with PNSP), with a rate ≥15 mg/l for 17 patients.

Third-generation cephalosporin

Third-generation cephalosporins were delivered in 76 patients (92.7%). The antibiotics used were cefotaxime for 54 patients at a minimum dosage of 70 mg/kg/d and ceftriaxone for 22 patients at a minimum dosage of 200 mg/kg/d.

Prognosis factors

Among antibiotic therapy administration or strategy, only two items were significant in the univariate analysis (Table 4): ICU started with EAT-compatible with IDSA guidelines (P = 0.05) and effective EAT (P < 0.0001). Neither were significant in the multivariate analysis.
Table 4

Univariate analysis studying the role of antimicrobial therapy on prognosis

 

Survivors (n = 65)

Non-survivors (n = 17)

P-value

Antibiotics in the previous six months, no. (%)

6 (9.2)

4 (23.5)

0.1

Antibiotics delivered at home for the current infectious process, no. (%)

30 (46.1)

4 (23.5)

0.09

Compliance to IDSA guidelines for EAT started in the ED (n = 70), no. (%)

34 (60.7)

10 (71.4)

0.45

Appropriate administration of EAT started in the ED (n = 56), no. (%)

39 (92.8)

12 (85.7)

0.85

Compliance to IDSA guidelines for EAT started in the ICU, no. (%)

40 (61.5)

6 (35.3)

0.05

Appropriate administration of EAT started in the ICU (n = 61), no. (%)

48 (97.9)

12 (100)

0.78

Effective EAT

57 (87.7)

5 (29.4)

<0.0001

EAT: empirical antibiotic therapy

ED: emergency department

ICU: intensive care unit

When stratifying before and after the publication of the IDSA guidelines, compatibility to guidelines increased (reaching statistical significance for treatment started in the ED) but without a significant improvement in outcome (Table 5).
Table 5

Influence of the publication of IDSA guidelines on the delivery of EAT

 

Before IDSA guidelines publication* (n = 56)

After IDSA guidelines publication, (n = 26)

P-value

Antibiotics prescribed in the ED** (n = 70)

46

24

 

Compliance to IDSA guidelines, no. (%)

25 (55.5%)

19 (79.2%)

0.05

Antibiotics prescribed in the ICU admission (n = 82)

56

26

 

Compliance to IDSA guidelines, no. (%)

29 (51.8%)

17 (65.4%)

0.43

Deaths (n = 17), no. (%)

12 (21.4%)

5 (29.4%)

0.49

*IDSA guidelines published in November 2004

**For these patients, antibiotics were delivered in the ED, in the Infectious Diseases Unit or by the family physician

Steroid therapy

Only three patients received DXM therapy. They were admitted after the publication of the IDSA guidelines. DXM was delivered for two patients with pneumococcal meningitis and one patient with meningococcal meningitis. None died during their ICU stay.

Discussion

The main results of our study are as follows: (1) we identified four prognosis factors associated with in-ICU mortality: age, alcoholism, acute kidney injury and cerebral damage; (2) we did not identify any factor predictive of penicillin non-susceptibility in pneumococcal meningitis; (3) compliance to IDSA guidelines is accurate with two-thirds of adherence.

About 20% of our patients died. This is similar to other studies reporting a relatively constant ABM mortality, ranging from 15 to 30% [1, 2, 14, 15, 16, 17, 18, 19, 20]. Many factors associated with poor outcome have already been described: low GCS, seizures, advanced age, delayed antimicrobial therapy and CSF leukocyte count <1,000 cells per mm3 [15, 16, 17, 18, 21]. In the ICU, disease severity and altered mental status are the two factors associated with a poor outcome [1, 2]. In our study, age >60 years and neurological failure were also poor prognosis factors. However, we identified two other prognosis factors rarely reported in other studies: alcoholism and acute renal failure.

Alcoholism worsens the host’s defences, decreasing humoral and cellular immune responses [22]. Moreover, chronic alcohol intake has been associated, in septic patients, with persistent fever, delayed resolution of symptoms, increased rates of bacteraemia, increased use of intensive care, prolonged duration of hospital stay and increased cost of hospitalisation [23].

Septic renal failure has already been reported as a poor prognosis factor in the ICU [24, 25, 26]. Measures such as adequate hydration and the maintenance of sufficient circulating blood volume and mean arterial pressure must be rapidly started in these patients. Rapid admission to the ICU associated with fluid challenge is necessary for meningitis patients with acute renal impairment.

Decreased susceptibility to penicillin remains an important issue. In France, PNSP incidence in ABM nears 40% of pneumococci [27]. The main factors reported as associated with PNSP are: antimicrobial use within the previous three months, older age, institutionalisation or recent hospital admission [28, 29, 30]. To our knowledge, only one study investigated the risk factors for PNSP in pneumococcal meningitis [17]. The authors concluded that neither age nor the absence of identified risk factors can accurately predict penicillin susceptibility. In our study, we too did not find evidence of any factor associated with or allowing us to rule out PNSP. Our number of patients with PNSP meningitis is low (n = 14) and this result must be confirmed by other studies. However, we show that, in some areas, PNSP can be present, with no predisposing factors, although recent European guidelines recommend determining whether predisposing factors associated with PNSP are present so as to add vancomycin when pneumococcal meningitis is suspected [4, 5, 31].

Many guidelines are available for the management of ABM but their impact on ABM outcome has not been evaluated. It is, furthermore, unclear if strict adherence to the recommended treatment is the best therapeutic option in ICU patients. Studies evaluating meningitis guidelines compliance are limited and concern only North European guidelines [6, 32]. In our study, we chose to study compliance to the IDSA guidelines, as it was one of the most recently published. IDSA guidelines differ from European guidelines by the recommendation of the systematic administration of vancomycin for suspected S. pneumoniae ABM. In our unit, there are no written recommendations and EAT depends on the attending physician. Vancomycin has been widely prescribed, even before the publication of IDSA guidelines, as our area (northern France) had a high PNSP prevalence for many years. During our studied period, publication of the IDSA guidelines improved the empirical delivery of recommended antibiotics after November 2004, with 65% of adherence. Compliance with this guideline improved ICU outcome in univariate analysis. This could be explained by the systematic use of vancomycin in suspected pneumococcal meningitis, the use of a third-generation cephalosporin rather than a penicillin in suspected pneumococcal meningitis, the systematic use of amoxicillin in patients over 50 years of age with non-documented meningitis or more appropriate dosages for prescribed antibiotics.

Appropriate EAT is the cornerstone of ABM outcome. The impact of IDSA guidelines in the appropriateness of our EAT is difficult to appreciate, as appropriate coverage of isolated strains was already high before its publication. In our study, EAT appropriateness was 98%, despite a moderate compliance (55%) to recommended IDSA EAT. However, among 36 non-compliance cases, 23 included patients with under-dosed or unnecessary antibiotics (n = 23). We might have chosen too stringent criteria; however, we considered these conditions to be deleterious, as they increase the risk of treatment failure or favour the emergence of multidrug-resistant bacteria.

We used DXM only rarely (3.6%). The use of adjuvant steroid therapy showed a beneficial impact in the outcome of patients with pneumococcal meningitis [33]. Recent guidelines recommend the delivery of DXM, just before or during the first administration of antibiotics, for patients with suspected pneumococcal meningitis [7, 8]. Our study underlines that adjuvant steroids therapy is still not a current practice for emergency and intensive care physicians.

Our study has some limitations. First, during the 10 studied years, the management of ICU septic patients has changed. Independently of antimicrobial therapy, prognosis has been improved with hydrocortisone use in septic shock, intensive insulin therapy and protective mechanical ventilation.

Second, our report is retrospective and we could not study a recently described important prognosis factor, the time between hospital admission and EAT delivery [3, 34].

Third, the ABM guidelines at our institution target medical wards and the ED, but not the ICU. Thus, even if most physicians followed the IDSA guidelines after their publication, the lack of homogeneity in our antimicrobial therapy limits the relevance of our results.

Fourth, our studied population is heterogeneous. We have chosen to include all ABM patients and our studied patients reflect probably the population of patients admitted in other ICUs. The severity of disease and the outcome of pneumococcal and meningococcal meningitis are not the same (20% mortality for S. pneumoniae and 10% mortality for N. meningitidis in our report), and analysis for each of these two pathogens could have given other results. However, causative pathogens, as a prognosis factor, was non-significant on outcome.

In summary, in our unit, compliance to IDSA guidelines is accurate, with about two-thirds of adherence. Non-compliance essentially concerns low dosage or the lack of delivery of vancomycin. We have displayed PNSP with no predisposing factors. ICU-admitted ABM remains a serious therapeutic challenge, despite frequently updated epidemiological data and therapeutic guidelines.

References

  1. 1.
    Auburtin M, Porcher R, Bruneel F, Scanvic A, Trouillet JL, Bédos JP, Régnier B, Wolff M (2002) Pneumococcal meningitis in the intensive care unit: prognostic factors of clinical outcome in a series of 80 cases. Am J Respir Crit Care Med 165:713–717PubMedGoogle Scholar
  2. 2.
    Flores-Cordero JM, Amaya-Villar R, Rincón-Ferrari MD, Leal-Noval SR, Garnacho-Montero J, Llanos-Rodríguez AC, Murillo-Cabezas F (2003) Acute community-acquired bacterial meningitis in adults admitted to the intensive care unit: clinical manifestations, management and prognostic factors. Intensive Care Med 29:1967–1973CrossRefPubMedGoogle Scholar
  3. 3.
    Auburtin M, Wolff M, Charpentier J, Varon E, Le Tulzo Y, Girault C, Mohammedi I, Renard B, Mourvillier B, Bruneel F, Ricard JD, Timsit JF (2006) Detrimental role of delayed antibiotic administration and penicillin-nonsusceptible strains in adult intensive care unit patients with pneumococcal meningitis: the PNEUMOREA prospective multicenter study. Crit Care Med 34:2758–2765CrossRefPubMedGoogle Scholar
  4. 4.
    Les méningites purulentes communautaires. 9ème conférence de consensus en thérapeutique anti-infectieuse. Med Mal Infect 1996;26:952–973Google Scholar
  5. 5.
    Begg N, Cartwright KAV, Cohen J, Kaczmarski EB, Innes JA, Leen CLS, Nathwani D, Singer M, Southgate L, Todd WTA, Welsby PD, Wood MJ (1999) Consensus statement on diagnosis, investigation, treatment and prevention of acute bacterial meningitis in immunocompetent adults. British Infection Society Working Party. J Infection 39:1–15CrossRefGoogle Scholar
  6. 6.
    van de Beek D, de Gans J, Spanjaard L, Vermeulen M, Dankert J (2002) Antibiotic guidelines and antibiotic use in adult bacterial meningitis in the Netherlands. J Antimicrob Chemother 49:661–666CrossRefPubMedGoogle Scholar
  7. 7.
    Tunkel AR, Hartman BJ, Kaplan SL, Kaufman BA, Roos KL, Scheld WM, Whitley RJ (2004) Practice guidelines for the management of bacterial meningitis. Clin Infec Dis 39:1267–1284CrossRefGoogle Scholar
  8. 8.
    Chaudhuri A, Martinez-Martin P, Kennedy PGE, Andrew Seaton R, Portegies P, Bojar M, Steiner I; EFNS Task Force (2008) EFNS guideline on the management of community-acquired bacterial meningitis: report of an EFNS Task Force on acute bacterial meningitis in older children and adults. Eur J Neurol 15:649–659CrossRefPubMedGoogle Scholar
  9. 9.
    National Committee for Clinical Laboratory Standards (NCCLS) (1997) Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 4th edn. Approved standard. NCCLS document M7-A4. Villanova, PA: NCCLSGoogle Scholar
  10. 10.
    Teasdale G, Jennett B (1974) Assessment of coma and impaired consciousness. A practical scale. Lancet 2:81–84CrossRefPubMedGoogle Scholar
  11. 11.
    Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, Cohen J, Opal SM, Vincent JL, Ramsay G (2003) 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 31:1250–1256CrossRefPubMedGoogle Scholar
  12. 12.
    Le Gall JR, Lemeshow S, Saulnier F (1993) A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA 270:2957–2963CrossRefPubMedGoogle Scholar
  13. 13.
    Jennet B, Teasdale G (1981) Glasgow outcome scale. Management of head injuries, 2nd edn. FA Davis, PhiladelphiaGoogle Scholar
  14. 14.
    Durand ML, Calderwood SB, Weber DJ, Miller SI, Southwick FS, Caviness VS Jr, Swartz MN (1993) Acute bacterial meningitis in adults. a review of 493 episodes. N Engl J Med 328:21–28CrossRefPubMedGoogle Scholar
  15. 15.
    Proulx N, Fréchette D, Toye B, Chan J, Kravcik S (2005) Delays in the administration of antibiotics are associated with mortality from adult acute bacterial meningitis. Q J Med 98:291–298Google Scholar
  16. 16.
    Weisfelt M, van de Beek D, Spanjaard L, Reitsma JB, de Gans J (2006) Clinical features, complications, and outcome in adults with pneumococcal meningitis: a prospective case series. Lancet Neurol 5:123–129CrossRefPubMedGoogle Scholar
  17. 17.
    Meyer CN, Samuelsson IS, Galle M, Bangsborg JM (2004) Adult bacterial meningitis: aetiology, penicillin susceptibility, risk factors, prognostic factors and guidelines for empirical antibiotic treatment. Clin Microb Infect 10:709–717CrossRefGoogle Scholar
  18. 18.
    van de beek D, de Gans J, Spanjaard L, Weisfelt M, Reitsma JB, Vermeulen M (2004) Clinical features and prognostic factors in adults with bacterial meningitis. N Engl J Med 351:1849–1859CrossRefPubMedGoogle Scholar
  19. 19.
    Dauchy FA, Gruson D, Chêne G, Viot J, Bebear C, Maugein J, Bézian MC, Dutronc H, Dupon M (2007) Prognostic factors in adult community-acquired bacterial meningitis: a 4-year retrospective study. Eur J Clin Microbiol Infect Dis 26:743–746CrossRefPubMedGoogle Scholar
  20. 20.
    Aronin SI, Peduzzi P, Quagliarello VJ (1998) Community-acquired bacterial meningitis: risk stratification for adverse clinical outcome and effect of antibiotic timing. Ann Intern Med 129:862–869PubMedGoogle Scholar
  21. 21.
    Lepur D, Barsić B (2007) Community-acquired bacterial meningitis in adults: antibiotic timing in disease course and outcome. Infection 35:225–231CrossRefPubMedGoogle Scholar
  22. 22.
    Pavia CS, La Mothe M, Kavanagh M (2004) Influence of alcohol on antimicrobial immunity. Biomed Pharmacother 58:84–89CrossRefPubMedGoogle Scholar
  23. 23.
    Moss M (2005) Epidemiology of sepsis: race, sex, and chronic alcohol abuse. Clin Infec Dis 41:S490–S497CrossRefGoogle Scholar
  24. 24.
    Brivet FG, Kleinknecht DJ, Loirat P, Landais PJ (1996) Acute renal failure in intensive care units—causes, outcome, and prognostic factors of hospital mortality; a prospective, multicenter study. French Study Group on Acute Renal Failure. Crit Care Med 24:192–198CrossRefPubMedGoogle Scholar
  25. 25.
    Levy EM, Viscoli CM, Horwitz RI (1996) The effect of acute renal failure on mortality. A cohort analysis. JAMA 275:1489–1494CrossRefPubMedGoogle Scholar
  26. 26.
    Neveu H, Kleinknecht D, Brivet F, Loirat P, Landais P (1996) Prognostic factors in acute renal failure due to sepsis. Results of a prospective multicentre study. The French Study Group on Acute Renal Failure. Nephrol Dial Transplant 11:293–299PubMedGoogle Scholar
  27. 27.
    Varon E, Gutmann L (2006) Centre National de Reference des Pneumocoques. Rapport d’activité 2007. Epidemiologie. Home page at: http://www.invs.sante.fr/surveillance
  28. 28.
    Vanderkooi OG, Low DE, Green K, Powis JE, McGeer A; Toronto Invasive Bacterial Disease Network (2005) Predicting antimicrobial resistance in invasive pneumococcal infections. Clin Infect Dis 40:1288–1297CrossRefPubMedGoogle Scholar
  29. 29.
    Clavo-Sánchez AJ, Girón-González JA, López-Prieto D, Canueto-Quintero J, Sánchez-Porto A, Vergara-Campos A, Marín-Casanova P, Córdoba-Doña JA (1997) Multivariate analysis of risk factors for infection due to penicillin-resistant and multidrug-resistant streptococcus pneumoniae: a multicenter study. Clin Infect Dis 24:1052–1059CrossRefPubMedGoogle Scholar
  30. 30.
    Bédos JP, Chevret S, Chastang C, Geslin P, Régnier B (1996) Epidemiological features of and risk factors for infection by Streptococcus pneumoniae strains with diminished susceptibility to penicillin: findings of a French survey. Clin Infect Dis 22:63–72PubMedGoogle Scholar
  31. 31.
    Roord JJ, de Gans J, Spanjaard L. Consensus-Bijeenkomst Bacteriele Meningitis. CBO/MWR, Utrecht, the NetherlandsGoogle Scholar
  32. 32.
    Cullen MM (2005) An audit of the investigation and initial management of adults presenting with possible bacterial meningitis. J infect 50:120–124CrossRefPubMedGoogle Scholar
  33. 33.
    de Gans J, van de Beek D; European Dexamethasone in Adulthood Bacterial Meningitis Study Investigators (2002) Dexamethasone in adults with bacterial meningitis. N Engl J Med 347:1549–1556CrossRefPubMedGoogle Scholar
  34. 34.
    Køster-Rasmussen R, Korshin A, Meyer CN (2008) Antibiotic treatment delay and outcome in acute bacterial meningitis. J Infec 57:449–454CrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  • H. Georges
    • 1
  • A. Chiche
    • 1
  • S. Alfandari
    • 1
  • P. Devos
    • 2
  • N. Boussekey
    • 1
  • O. Leroy
    • 1
  1. 1.Intensive Care and Infectious Disease UnitTourcoing HospitalTourcoing cedexFrance
  2. 2.Department of BiostatisticsCentre Hospitalier Régional Universitaire de LilleLilleFrance

Personalised recommendations