Characteristics and outcome of spontaneous bacterial meningitis in patients with diabetes mellitus

We collected data from 1982 through 2017 at the Hospital de la Santa Creu i Sant Pau (Barcelona, Spain), a 540-bed tertiary university hospital serving an estimated population of 450,000 in a predominantly urban area. We included all consecutive adult patients (> 14 years) with spontaneous bacterial meningitis.

The characteristics of this cohort have been described elsewhere [4, 21, 25, 26], and also how we identified the cases. Data were collected during the hospitalization and the patient was monitored after discharge by one of the authors (PD or VP). Follow-up always included a neurological exam, neuropsychological testing, complement, and immunoglobulin levels, and audiometry.

The diagnosis of culture-proven BM was made on the basis of consistent clinical findings, a positive CSF culture. If CSF culture was negative, culture-proven BM could still be diagnosed with CSF neutrophilic pleocytosis (= > 100 neutrophils/cu mm), a positive CSF antigen test, or a positive blood culture. Meningococcal meningitis was diagnosed, in addition to those culture-proven, in patients with CSF gram-negative diplococci identified together with with a petechial or purpuric rash and a fulminant course [27].

Unknown etiology BM was diagnosed in patients with a compatible clinical picture together with neutrophilic pleocytosis and decreased CSF glucose (defined as CSF/blood glucose ratio < 0.40) or elevated CSF proteins (defined as > 0.5 g/l) [3, 6, 28].

Patients with a history of neurosurgical procedures or traumatic head/spinal injuries and cases of viral, fungal, or mycobacterial meningitis were excluded.

Isolate identification and susceptibility tests performance have been described elsewhere [4, 21, 25, 26, 29‐31].

Majority of the variables have been defined in other published works [4, 21, 25, 26].

The Charlson Comorbidity Index was used to assess comorbidity [32, 33]. Comorbid conditions were considered to be present if the patient had a confirmed diagnosis of one or more.

A diagnosis of DM was based on information provided by the patient or his/her family, recorded on medical charts or due to the presence of diagnostic criteria for DM at the time of admission (glycosylated hemoglobin [HbA1c] ≥ 6.5% or a fasting plasma glucose level ≥ 126 mg/dl [7 mmol/L] or the 2-h plasma glucose value after 75 g oral glucose tolerance test > 200 mg/dl [11,1 mmol/L] or in a patient with classic symptoms of hyperglycemia [polyuria, polydipsia or polyphagia] or hyperglycemic crisis, random plasma glucose ≥200 mg/dL [11.1 mmol/L]) [34]. We included type 1 and 2.

A primary distant focus of infection was considered when a patient had clinical symptoms and signs consistent with a focal infection distant from the central nervous system, and the same pathogenic bacterium was isolated from the primary focus of infection or a blood culture. Communication of the subarachnoid space with the skin, sinuses, or mucosal surfaces, and upper respiratory tract infection (frequent in meningococcal disease) were not considered distant foci of infection [24].

The symptoms-admission interval (SAI) was the interval, in hours, between onset of signs and symptoms of BM and admission to hospital was. When the onset of symptoms could not be precisely determined, we calculated the mean interval between the last time the patient was asymptomatic, observed by a household member, and the first time the patient was seen ill [35].

Impaired mental status, seizures and focal neurological signs (motor, sensory or cranial nerve disturbances) detected on admission or subsequently were considered neurological complications. Coma was defined as a score of 6 or less on the Glasgow Coma Scale in the absence of sedation [36].

The development of septic shock, acute respiratory failure, acute renal failure, and/or consumption coagulopathy was considered a systemic complication if it was related to bacterial meningitis and was apparent on admission or shortly afterwards [24, 37].

Adequate antibiotic treatment was defined as the intravenous administration of any antimicrobial agents to which isolated bacteria were sensitive following susceptibility testing at local laboratories, which crossed the blood-brain barrier in adequate amounts, were administered in a dose recommended for acute BM treatment and commenced on the day of admission or before deterioration of neurological and systemic conditions in inpatients [24, 38, 39].

Dexamethasone therapy was only considered when its first dose of at least 10 mg/24 h was administered prior to or concomitant with the first antibiotic dose. Steroids administered after starting antibiotic therapy were not considered [40].

Nosocomial meningitis was defined as developing more than 48 h after admission or within one week of discharge [41].

Inpatient mortality was considered to be due to BM when meningitis was the underlying and immediate cause of death. BM-related death was defined and recorded if there were no other cause and if there were agreement between the reviewing physician and the underlying cause of death recorded on the medical records. We classified the causes of death in two categories: neurological causes (intractable seizures, brain herniation, cerebrovascular complications or coma) and systemic causes (septic shock, respiratory failure, disseminated intravascular coagulation or multiple organ failure) [37, 42]. Sepsis resulted from a host’s systemic inflammatory response syndrome (SIRS) to infection with 2 or more of temperature > 38 °C or < 36 °C, heart rate > 90, respiratory rate > 20/min or pCO2 < 32 mmHg or white blood cell count > 12,000/mm3 or < 4000/mm3 or > 10% immature bands. Septic shock was defined as sepsis induced hypotension persisting despite adequate fluid resuscitation. Disseminated intravascular coagulation is primarily a laboratory diagnosis, based on the combination of elevated fibrin-related markers (FRM), with decreased procoagulant factors and platelets.

Sequelae were defined as any disability, disorder, or injury demonstrated during hospital stay or upon discharge from hospital that was not present before the episode of BM and persisted at 6 months after discharge [42].

Qualitative variables were expressed as absolute numbers and percentages, whereas quantitative variables were expressed as means and standard deviation (SD).

The comparison of continuous variables was performed through Student’s t-test or Mann-Whitney U test, whereas categorical data were compared by c2 test or Fisher’s exact test.

Independent predictive factors for mortality were explored through logistic regression analysis with adjustment for clinically relevant covariates was used, and the outcome shown sed as adjusted odds ratios. The proportion of the total variation of the outcome explained by a model was assessed through Nagelkerke’s R2.

The demographic characteristics and clinical features of the population are summarized in Table 1. Patients with DM were older and have a Charlson comorbidity score of ≥3 more often than non-diabetics. When the Charlson score was recalculated excluding DM without and with end-organ failure, the comorbidity score for the two groups was similar (Charlson comorbidity index ≥3 was 18/106 [17%] in diabetic patients versus 89/609 [15%], p = 0.555).

Hypertension and cancer were the most common underlying conditions in these patients.

A distant focus of infection was present in 41 of the 106 episodes (39%), with otitis or sinusitis (12 episodes, 11%), upper respiratory tract infection (7 episodes, 7%) and pneumonia (6 episodes, 6%) being the most frequently diagnosed distant foci.

Altered mental status was more frequent among patients with DM than among non-diabetics, whereas neck stiffness, headache, nausea, and rash were less frequent.

Lumbar puncture was performed in all episodes, and the CSF showed at least one CSF finding suggestive of acute bacterial meningitis: increased protein levels in 103 cases (97%), a decreased CSF glucose/blood glucose ratio in 83 cases (78%), and pleocytosis with an elevated neutrophil count in 100 cases (94%). CSF cyto-biochemical findings did not differ significantly between the two groups of patients (Table 2). On the contrary, blood cultures were more frequently positive in diabetic patients (63% vs. 45%, p < 0.001) (Table 2).

The most common organisms were S. pneumoniae (24% in people with DM vs. 25% in non-diabetic patients, p = 0.632) followed by L. monocytogenes (18% vs. 10%, respectively, p < 0.033). There were only 11 (10%) cases of meningococcal meningitis, which was significantly more common in non-diabetic patients (194 episodes, 32%, p < 0.001). S. agalactiae was detected in 9 diabetic patients (8.5%) and in 6 non-diabetic patients (0.98%), with OR = 9.32 (2.88–32.42), p < 0.001. (Table 2). The etiologic spectrum of BM did not vary significantly between the two periods of the study except for listerial meningitis, which was more frequent in the second period: 3 (7%) vs. 16 (26%) cases (p = 0.01).

There was a high number of patients with DM having received out-of-hospital antibiotic therapy, although not statistically different from the rest of the population (Table 1).

Initial empirical antibiotic treatment was appropriate in 93 patients with diabetes (88%), being less frequent, although not statistically significant, than for non-diabetic patients (93%, p = 0.109) (Table 3). There was no difference between the 2 periods (37 [40%] vs. 56 [60%], p = 0.230). The most organisms not adequately covered by initial empiric therapy were L.monocytogenes, Pseudomonas aeruginosa, and Staphylococcus aureus.

The median duration of antibiotic treatment was 15 days (IQR 11).

Neurological complications were not different between patients with and without DM. However, the presence of focal neurological deficits after discharge (between survivors) was more frequent in diabetics (7.5% vs. 3.1%, p = 0.046) (Table 3).

The overall mortality rate was significantly higher in patients with DM (25% vs. 16%, p = 0.025) (Table 3), being higher in the first study period (18 between 1982 and 1999 (66.7%) vs. 9 between 2000 and 2017 (33.3%), p = 0.006). The BM-attributable death in diabetic patients was 52%, and among non-diabetics was 63% (p = 0.72). The mean number of days until death did not differ between diabetic and non-diabetic patients (14 vs. 15, p = 0.857). Independent factors associated with an unfavorable outcome by multivariate analysis included: septic shock, coagulation disorder, age 65 or older, acute renal failure, the periods of the study, and gram-negative bacillary etiology. Diabetes was not associated with higher mortality (Table 4).