Three-year multicenter surveillance of community-acquired listeria monocytogenes meningitis in adults

Listeria monocytogenes (Lm) is a Gram-positive, facultative anaerobic bacterium that primarily causes sepsis and meningitis [1] in either immunocompromised or immunocompetent hosts [2, 3]. As a foodborne pathogen, it has emerged as a significant public health problem and has caused several epidemics in the United States and Europe. Community-acquired Lm meningitis is a serious and life-threatening disease. The estimated incidence is 0.2 cases per 100.000 adults per year in developed countries [1]. Lm is the third most frequent cause of community-acquired bacterial meningitis, after Streptococcus pneumoniae and Neisseria meningitidis aetiologies, due to the vaccine-related decline in Haemophilus influenzae type b meningitis [2]. This entity occurs mainly in immunocompromised patients, newborns, and elderly individuals [4], although previously healthy adults can be affected as well [5]. Even with appropriate antibiotic therapy, this entity has a high morbidity and mortality (24%-62%) [2‐4, 6‐8].

This is a prospective multicenter observational study carried out in 9 Spanish hospitals. All these hospitals are large institutions with teaching accreditation. The study was conducted over a 39-month period. All patients over 14 years old admitted to the participating hospitals with the diagnosis of acute community-acquired bacterial meningitis (Ac-ABM) were included in this study. The Ethical committees approved the study and did not require informed consent from the patient.

One investigator at each hospital prospectively recorded variables of all Ac-ABM in a previously designed database collecting data on patients' demography, clinical antecedents, symptoms and signs at admission, laboratory findings, clinical course, treatment, and outcome. Episodes of Ac-ABM were diagnosed on the basis of compatible clinical picture and at least 1 of the following cerebrospinal fluid (CSF) findings: CSF polymorphonuclear (PMN) pleocytosis [white blood cells (WBC) count >10/μl, 90% neutrophils, hipoglycorrachia <40 mg/dL or CSF/blood glucose ratio <40%, increased protein level >1 g/L, a positive CSF Gram's stain or culture, and/or positive blood cultures. When the CSF Gram's stain and blood and CSF cultures were negative, meningitis of unknown aetiology was recorded. Meningitis was considered to be community-acquired whenever a patient had not been previously treated in a hospital. If the patient had been previously admitted, Ac-ABM was diagnosed whenever the onset of the disease occurred 2 weeks after discharge or after 4 weeks if the patient had undergone surgery [9].

Lm meningitis was diagnosed in those patients with Ac-ABM in whom Lm was isolated in the CSF, blood culture, or both. Serotyping Lm was done using specific antisera, as described elsewhere [10].

Patients were considered immunocompromised if they were receiving immunosuppressive therapy and/or had haematological and solid malignancy, connective tissue disease, diabetes mellitus, alcoholism, asplenia, liver cirrhosis, end-stage renal disease, or HIV infection [11]. Patients were considered elderly when they were 60 years or older.

The classic triad was defined when fever, neck stiffness and change in mental status were present. Hyponatremia was defined as a plasma sodium concentration <135 mmol/L [12]. Focal neurological deficit was diagnosed when aphasia, monoparesis, hemiparesis or cranial nerve palsies were present at admission or during hospitalisation. Cerebellum dysfunction was diagnosed in patients with nystagmus, dysarthria, or limb and gait ataxia.

Time to admission was defined as the appearance of symptoms to admission to the hospital. Patients with impairment of consciousness, neurological deficits, or seizures, underwent cranial computed tomography (CT) before lumbar puncture.

Empiric antibiotic therapy was defined as any parenteral antibiotic capable of crossing the blood-brain barrier, administered for the purpose of treating bacterial meningitis [8]. In the case of Lm meningitis, intravenous administration of penicillin, ampicillin, amoxicillin/clavulanic acid, piperacillin/tazobactam, a carbapenem, moxifloxacin or trimethoprim-sulfamethoxazole was considered adequate empirical therapy. During the follow up performance of a brain CT or the administration of adjunctive therapy (corticosteroids and/or phenytoin) were left for the physician in charge to decide.

All patients were followed up until death or hospital discharge. At discharge, all patients underwent a neurological examination, and outcome was measured according to the Glasgow Outcome Scale [13]. A score of 1 on this scale indicates death; a score of 2, a vegetative state; a score of 3, severe disability; a score of 4, moderate disability; and a score of 5, mild or no disability. A favourable outcome was defined as a score of 5, and an unfavourable outcome was defined as a score of 1-4.

For the statistical analysis the SPSS 16.0 software package was used. Univariate analysis of variables from the entire population was performed using unpaired student's t-test for parametric continuous variables after correction for equality of variance (Levene's test), U Mann-Whitney test for non parametric continuous variables, and Pearson's chi-square test or Fisher's exact test for categorical variables. Statistical significance was considered when p <0.05. All p values were two-sided. Furthermore, a multivariate using logistic regression analysis was used to identify independent associations with Lm meningitis with results presented as Odds-Ratio (OR) and the 95% confidence intervals (95% CI). Finally, a univariate analysis comparing immunocompromised and immunocompetent patients with Lm meningitis was performed as previously indicated.

Two hundred and seventy-eight patients with Ac-ABM were included in this prospective study. The most common pathogen was Streptococcus pneumoniae (Table 1). Forty-six episodes of Lm meningitis were identified in 46 patients. Three patients were excluded from the analysis because they were prematurely transferred to another hospital. Lm grew from the CSF in 40/43 (93%) cases and from blood culture in 27/43 (63%).

Table 2 shows the comparison between Lm meningitis and episodes with other aetiologies. In this univariate analysis those risk factors and clinical characteristics showing differences at admission were: age, immunosupression, previous otitis, focal neurological deficit, cerebellum dysfunction, WBC count CSF and percentage of neutrophils, CSF/blood glucose ratio, and CSF protein levels. However in the multivariate analysis only age (OR 1.026; 95% CI 1.00-1.05; p = 0.042), immunosupression (OR 2.520; 95% CI 1.05-6.00; p = 0.037), and CSF/blood glucose ratio (OR 39.42; 95% CI 4.01-387.50; p = 0.002) were independently associated with Lm meningitis.

All patients with acute community-acquired Lm meningitis were immunocompromised or over 50 years old. The baseline characteristics of these patients and the differences between patients with and without immunocompromise are shown in Table 3. The median age was 69 ± 30 years and 24 of the 43 cases (56%) were male. Twenty-nine (68%) patients were immunocompromised and fourteen (32%) were immunocompetent patients. The nature of immunocompromise were: Twelve (41%) patients were receiving immunosuppressive therapy, 8 (27.5%) had haematological and solid malignancy, 2 (6.8%) with connective tissue disease, 2 (6.8%) with end-stage renal disease and 5 (17.2%) with HIV infection.

In our series, in patients with Lm meningitis, fever was the most common symptom (91%). The majority of patients also had altered mental status at presentation, while headache was present in nearly half the patients. The classic triad of fever, neck stiffness and altered mental status was present in 21 (49%) patients; however, almost all patients (95%) had at least 1 or more of these symptoms. Thirty-two percent of the patients had focal neurological findings at presentation, 12% presented cerebellum dysfunction, and 9% had seizures. The most frequent focal neurological finding was cranial nerve palsies. The median duration of symptoms before admission to the hospital was two days. Immunocompromised patients had a lower incidence of the classic triad (45% vs. 71%; p = 0.039), headache (59% vs 85%; p = 0.076), and focal neurological deficit at presentation (28% vs 43%; p = 0.317) as compared with immunocompetent patients with acute community-acquired Lm meningitis.

Thirty-nine (91%) patients underwent CT in the emergency department. In 30 of 39 (77%), there were no abnormalities associated with the listerial infection. In the remaining 9 (23%) patients, the most common findings were focal lesions and/or hydrocephalus. All patients with hydrocephalus were treated neurosurgically.

Lumbar puncture was performed in all patients. CSF findings are outlined in Table 3. There was no relation between low CSF leukocyte count and age or immunocompromise. The most frequently found serotype was the 4b in 23 (82%) of 28 cases.

Empirical antimicrobial therapy was intravenous ampicillin (2 g every four hours) for 34 (79%) of 43 patients, in 11 (32%) of them combined with aminoglycosides. The rest of six patients received trimethoprim-sulfamethoxazole therapy in two cases and meropenem in the other four cases as empirical antibiotic treatment. Three patients that did not received appropiated empirical treatment receiving ceftriaxone plus vancomycin. Twenty-one patients (49%) received adjunctive therapy with dexamethasone; in 14 cases, the first dose was given previously or concomitantly to the first antibiotic dose, and eleven patients received phenytoin. There were no major side-effects of dexamethasone therapy (gastro-intestinal bleeding, hyperglycemia).

The overall mortality was 28% (12 of 43 patients) and five of thirty-one (16.1%) surviving patients had a neurological deficit at discharge. Neurological abnormalities identified were cranial nerve palsies in 2 (6.4%) and hemiparesis in 3 (9.6%). No patients had hearing impairment. Definitive ampicillin plus gentamicin therapy was significantly associated with unfavourable outcome (67% vs 28%; p = 0.024) and a higher mortality (67% vs 32%; p = 0.040) due to Lm infection, Table 4.

Early administration of adequate antimicrobial therapy is the cornerstone of treatment in patients with meningitis. Despite developments in antimicrobial agents, Lm meningitis is still a serious disease that carries high morbidity and mortality rates. A relevant aspect in this study is the prospective inclusion of adult patients with acute community-acquired Lm meningitis, to evaluate risk factors associated with Lm aetiology which should contribute to early appropriate antimicrobial therapy of this serious and life-threatening disease. Thus, in a large cohort of patients with Ac-ABM, our study reveals that elderly, immunocompromised patients, and a higher CSF/blood glucose ratio were independently associated with isolation of Lm as the responsible pathogen.

In our own series, the incidence of acute community-acquired Lm meningitis in adults was 16.5% (46/278). Previously, in a multicenter study [14] conducted in the United States, the authors reported that Lm accounted for 8% of the cases of bacterial meningitis. Other studies from Europe and North America have reported incidences of 5%-10% or more among all episodes of Ac-ABM among adults [15‐18]. Over a 36-year period, Durand et al. described that Lm accounted for 11% of the episodes of Ac-ABM in adults [7]. The probable increase in the incidence may be attributed to longer life expectancy, changes in diet or food processing, and the higher number and longer survival of immunocompromised people [2].

On the other hand, symptoms and signs of patients presenting with Lm meningitis were not different from those found in the general population of patients with Ac-ABM, according to previous studies [19].

The CSF profile in patients with Listeria infection revealed significantly fewer WBC and lower protein concentrations than patients with infection due to other pathogens, and a trend towards less hypoglycorrhachia and a lower percentage of PMNs. However, these findings do not apply to all cases [20]. In the appropriate clinical situation, clinicians should not exclude Lm based on the degree of pleocytosis, the percentage of PMNs, or the concentration of glucose or protein in the CSF.

The experience in the present series suggests that the Gram stain is negative in two-thirds of the cases of Lm meningitis, and may be misleading in many of the remaining cases. Clinicians should be aware of these difficulties, inform the microbiology department when they suspect this pathogen, and not rule out Lm based on Gram stain alone. Furthermore, despite the diversity of serotypes of Lm, only three serotypes are responsible for >90% of human disease: 1/2a, 1/2b, and 4b. The majority of strains were serotype 4b [21] (23 of 28 cases; 82%), suggesting that serotype 4b of Lm is more virulent than others. Otherwise, blood cultures were positive in more than half of the cases in the present series. Some reports advocate the use of the polymerase chain reaction or immunoassays for the early detection of Lm [22, 23], but these techniques need further evaluation.

Most patients in our case series received appropriated empirical therapy, 40/43 (93%), in 11 of them associated to aminoglycosides. Our study showed that a definitive combined therapy with aminoglycosides was significantly associated with an increasing trend for an unfavourable outcome and mortality, which corresponds with a previous study [24]. These results could be explained by Lm meningitis patients have an associated co-morbid condition, in which addition of an aminoglycoside may be harmful. For this reason, actually this combined therapy is been questioned, moreover of the drug's associated nephrotoxicity and inability to cross the blood-brain barrier. In this way, every case should be approached independently, although certain subsets of patients probably require empiric antibiotic treatment for Lm, such as patients over 45-50 years of age, patients with immunosuppression, a higher CSF/blood glucose ratio, or patients with a Gram stain of CSF revealing Gram-positive bacilli [19], and the associated with aminoglycosides should be avoid in order to improve patient's outcome.

A significant number of patients, 5 of 31 (16.1%) of the survivors in the present series had some degree of residual neurological deficit at hospital discharge. Although an even higher incidence of residual neurological deficits (32%) was noted among survivors enrolled in 3 previous series and deficits persisted for more than a year in 4 of 10 of those cases with adequate follow up [25‐27].

Moreover, mortality due to Lm is among the highest of all causes of acute bacterial meningitis [28]. Part of the high mortality of Listeria infection may relate to the increased number of immunocompromised and older patients affected by this pathogen [29, 30]. Although evidence for use of anticonvulsant prophylaxis in Lm meningitis is lacking, it may be considered as seizures have previously been related to increases mortality [31], as did we in the present series, although this was not statistically significant. Further studies are needed in order to clarify this point.

Finally, the role of adjunctive therapy with corticosteroids has been widely debated. A clinical trial showed a beneficial effect of early dexamethasone treatment in adults with bacterial meningitis [32]. Nevertheless, there are no studies on the effect of dexamethasone in adult Listeria meningitis. We observed a higher survival rate among those patients that received dexamethasone concomitant to antibiotic, although this association did not reach statistical significance. Our results suggest that further studies are needed to evaluate the effect of corticosteroid therapy in adult patients with Lm meningitis.

The present study has several limitations. First, given the small number of patients, we were unable to develop a statistical model to identify risk factors independently associated with a poor outcome or death in patients with acute community-acquired Lm meningitis. Secondly, only patients who underwent lumbar puncture and who had a positive cerebrospinal fluid culture were included. Negative cerebrospinal fluid cultures occur in 20 percent of patients with acute community acquired bacterial meningitis. Third, delay of antimicrobial therapy was not recorded. Despite all these limitations, we consider that this study contributes to our knowledge of risk factors associated with development of acute community-acquired Lm meningitis, and to evaluating epidemiology, clinical features, management, and outcome in this homogenous population of adult patients with acute community-acquired Lm meningitis.

To sum up, our study showed elderly or immunocompromised patients and a higher CSF/blood glucose ratio in patients with Ac-ABM must alert clinicians about Lm aetiology. Furthermore, we observed a high incidence of acute community-acquired Lm meningitis in adults and the addition of aminoglycosides to treatment should be avoid in order to improve the patients' outcome. Nevertheless, despite developments in intensive care and antimicrobial therapy, this entity is still a serious disease that carries high morbidity and mortality rates. Further studies should focus on specific interventions (e.g. adjuvant therapies) that could help to improve the poor prognosis of these patients.