BACTERIAL MENINGITIS DURING SEPSIS IN DIABETIC PATIENT

Transcription

1 Rev. Med. Chir. Soc. Med. Nat., Iaşi 2013 vol. 117, no. 4 INTERNAL MEDICINE - PEDIATRICS ORIGINAL PAPERS BACTERIAL MENINGITIS DURING SEPSIS IN DIABETIC PATIENT Cristina G. Petrovici 1, Daniela Leca 1, Andra Teodor 1, Olivia Dorneanu 2. Gabriela Juganariu 1, Carmen Dorobăț 1, Miftode Egidia 1 University of Medicine and Farmacy Grigore T. Popa - Iași Faculty of Medicine 1. Discipline of Infectious Diseases 2. Discipline of Microbiology BACTERIAL MENINGITIS DURING SEPSIS IN DIABETIC PATIENT (Abstract): The aim of the study was to evaluate the etiology, clinical features and outcome in diabetic patients with bacterial meningitis, as a nervous system determination during invasive infections. Material and methods: In a retrospective study, conducted over a period of three years, we have analyzed clinical and etiological aspects of 445 patients over 18 years old, diagnosed with sepsis of known (positive cultures from normally sterile sites) or suspected etiology (positive cultures from pus), 95 of them being included in the diabetic group. Results: Bacterial meningitis was diagnosed in 16 of 95 diabetic patients (16.8%) and 43 of 350 (12.3%) non-diabetic patients (χ 2 =0.98; GL=1; p=0.322). Among the multiple co morbidities associated in diabetic patients, as suggested by a higher Charlson score (5.44 vs. 3.25) (p=0.001), the most common underlying condition was chronic liver disease (31.3% vs. 25.6%) (p=0.916). The isolation of the microorganism concurrently from cerebro-spinal fluid and other sites (blood cultures and pus) was more frequently encountered in diabetics. The clinical picture was dominated by altered consciousness (68.8% vs. 23.3%) (p=0.003), while fever was less present (37.5% vs. 88.4%) (p=0.0003). The most frequently involved microorganism in the etiology of meningitis was S. aureus (31.3 vs. 23.3%) (p=0.771) and Gram negative bacilli: E. coli (12.5% vs. 4.7%) (p=0.629) and Klebsiella spp. (12.5% vs. 9.3%) (p=0.902). Conclusions: Altered consciousness was more frequent in diabetic patients group (68.8% vs. 23.3%) (p=0.003) where the absence of fever at admission was a more common finding than in non-diabetic septic patients with meningitis (37.5% vs. 88.4%) (p=0.0003). Keywords: BACTERIAL MENINGITIS, SEPSIS, DIABETES Bacterial infections of the central nervous system during invasive bacterial disease in diabetic population represents a challenge for the diagnosis, mainly because the clinical signs are often masked by acute and chronic complications of the diabetes, which may delay early recognition and adequate treatment. The epidemiology of bacterial meningitis varies in different populations. Several factors such as age, geographic distribution, underlying medical and surgical conditions, way of acquisition and the vaccination status may influence the prevalence of the involved pathogens (1). Association with diabetes mellitus as a co morbidity was found to be around 4-5% in patients with adult community-acquired bacterial meningitis in western reports, and as high as 25% in patients with bacterial meningitis from Taiwan (2). Therefore, the aim of the study was to evaluate the etiology, clinical features and outcome in diabet- 901

2 Cristina G. Petrovici et al. ic patients with bacterial meningitis, as a nervous system determination during invasive infections. MATERIAL AND METHODS In a retrospective study, conducted over a period of three years ( ), we have analyzed the clinical and etiological aspects of 445 patients over 18 years old with sepsis admitted to the Clinical Hospital of Infectious Diseases Iași. Only cases of sepsis with known etiology were included in the study, documented by positive cultures from different normally sterile sites or suspected etiology (positive cultures from pus). According to American Diabetes Associations criteria, 95 of patients were included in the diabetic group. Patients with bacterial meningitis were identified based on the microbiological records of the cerebrospinal fluid (CSF): direct examination, positive cultures and latex agglutination assay. The associated Chronic co morbid medical conditions were assessed by Charlson co morbidity score (3). Other variables such as onset-admission period longer than 48 hours, length of stay, discharged status, adequate treatment, mortality were collected. All the information represented data available from medical charts during hospitalization and long term outcome information was not available. For statistical analysis, all information was centralized in an Excel data base and differences between patient groups (diabetics and non-diabetics) were assessed using SPSS application, version A p-value of < 0.05 was considered to be statistically significant. RESULTS AND DISCUSSION During the period of study of three years, among 445 patients diagnosed with sepsis of known or suspected etiology, bacterial meningitis was documented in 95 cases, 16 (27.1%) of them being patients with diabetes mellitus. Regarding the frequency of meningitis diagnosed during invasive disease, no statistically significant difference was noted between the studied groups: 16 of 95 diabetic patients (16.8%) and 43 of 350 (12.3%) non-diabetic patients (χ 2 =0.98; GL=1; p=0.322) (fig.1). Fig. 1. The structure of the studied groups with bacterial meningitis The majority of cases in both groups were males (68.8% respectively 74.4%). The median age of the patients with diabetes was significantly higher (63.88 vs years) (p=0.050). Among the multiple co morbidities associated in diabetic patients, as suggested by a higher Charlson score (5.44 vs. 3.25) (p=0.001), chronic liver disease was the most commonly noted (31.3% vs. 25.6%) (p=0.916). Other underlying conditions and predisposing factors are shown in table I. Clinical evaluation in the first 24 hours of hospitalization, appreciated with SAPS II (Simplified Acute Physiology Score), revealed more severe forms of sepsis in diabetic patients (54 vs ) (p<0.001) (tab. I). 902

3 Bacterial meningitis during sepsis in diabetic patient TABLE I Characteristics of diabetic and non-diabetic patients with bacterial meningitis Characteristic Patients with bacterial meningitis (no/% of the whole group) Diabetic patients (n=95) (%) Non-diabetic patients (n=350) (%) 16 (16.8%) 43 (12.3%) Median age (years) 63.88± ± Gender Male 11 (68.8%) 32 (74.4%) Female 5 (31.2%) 11 (25.6%) Predisposing conditions p Chronic liver disease 5 (31.2%) 11 (25.6%) Alcoholism 2 (12.5%) 6 (13.9%) Chronic kidney disease 1 (6.3%) 1 (2.3%) Tumors 1 (6.3%) 2 (4.7%) AIDS - 2 (4.7%) Charlson Score 5.44± ± SAPS II 54.0± ± Clinical features Fever 6 (37.5%) 38 (88.4%) Altered consciousness 11 (68.8%) 10 (23.3%) Seizures 2 (12.5%) 7 (16.3%) Shock 3 (18.8%) 7 (16.3%) MODS a 4 (25.0%) 5 (11.6%) Brain abscess - 1 (2.3%) Epiduritis 1 (6.3%) 1 (2.3%) Positive cultures from CSF and other pathological products 13 (81.3%) 14 (32.6%) Duration onset-admittance >48 ore 6 (37.5%) 10 (23.3%) Hospital stay 15.75± ± Adequate initial therapy 11 (68.8%) 30 (69.8%) Mortality rate 2 (12.5%) 11 (25.6%) Transfer 4 (25%) 4 (9.3%) a MODS multiple organ dysfunction syndrome, # CSF cerebrospinal fluid 903

4 Cristina G. Petrovici et al. A period of more than 48 hours between the onset of clinical signs and admission in hospital was noticed in diabetic patients (37.5% vs. 23.3%) (p=0.444). This is probably due to the fact that classical clinical signs of meningitis were less obvious, metabolic imbalance being also a cause of altered consciousness, which may be misleading, delaying thus the diagnosis of meningitis. Before admission, the clinical picture was dominated by altered consciousness (68.8% vs. 23.3%) (p=0.003), while fever was less present (37.5% vs. 88.4%) (p=0.0003). Seizures were more often encountered in non-diabetic patients 16.3% vs. 12.5%, but without statistical significance (p=0.961). Patients with diabetes were more likely to have other foci of infection than patients without diabetes, multiple organ dysfunction syndrome being documented in 25% of this immunocompromised group vs. 11.6% in the non-diabetic group (p=0.388). It is noteworthy that the isolation of the same microorganism concurrently from cerebrospinal fluid and other different pathologic collections (blood cultures and pus) was also more frequently encountered in diabetic group (81.3% vs. 32.6%) (p=0.0003). Complications of bacterial meningitis were epiduritis (one case in each group) and brain abscess (one case in the non-diabetic group). TABLE II The etiology of bacterial meningitis in diabetic and non-diabetic patients Etiology Diabetic patients (n=16) Non-diabetic patients (n=43) S. aureus 5 (31.3%) 10 (23.3%) Coagulase-negative staphylococci 2 (12.5%) 3 (7.0%) Streptococcus grup B 1 (6.3%) Streptococcus pneumoniae 2 (12.5%) 12 (27.9%) E. coli 2 (12.5%) 2 (4.7%) Klebsiella spp. 2 (12.5%) 4 (9.3%) Neisseria spp 1 (6.3%) 2 (4.7%) Aerococcus viridans 1 (6.3%) 1 (2.3%) Enterococcus spp - 3 (7.0%) Acinetobacter spp - 3 (7.0%) Pseudomonas aeruginosa - 2 (4.7%) Citrobacter freundii - 1 (2.3%) P The mean hospital period of stay varied between 12 and 16 days in both groups (p>0.05). Unfavorable outcome was not signifi- 904

5 Bacterial meningitis during sepsis in diabetic patient cantly different in diabetic and non-diabetic patients. Transfer to other medical or surgical clinics was necessary in 25% of diabetic cases (neurosurgery - 1 case of epiduritis, and in intensive care service - 3 cases, for acute respiratory distress syndrome). S. aureus was the most common pathogen responsible for meningitis in diabetic patients with sepsis (31.3 vs. 23.3%) (p=0.771) followed by Gram negative bacilli: E. coli (12.5% vs. 4.7%) (p=0.629) and Klebsiella spp. (12.5% vs. 9.3%) (p=0.902). We did not find significant differences between the two groups of patients regarding the etiology (tab. II). DISCUSSION A report from Koreea (4) analyzing cases of community-acquired meningitis from 1998 to 2008 showed that 31.3% of patients were diabetic compared with 27.1% in our cohort of patients with meningitis during sepsis. It is worth noting that we included in our study only cases with known etiology (positive cultures of CSF or blood), therefore, true prevalence of bacterial meningitis may be underestimated. In a previous Romanian study, conducted on 844 cases of meningitis, diabetes was associated in only 1.5% of cases. These important differences may be due to the fact that in the study patients younger than 18 years were also included, children under 15 years representing 43.5% and only 7.7% being patients over 65 years (5). Similar studies from Taiwan showed a high incidence of diabetes (34-36%), among the adult patients with communityacquired spontaneous bacterial meningitis (1, 6). In a prospective study from the Netherlands on adult community-acquired bacterial meningitis, diabetes was reported in 7% of cases (7). A lot of reports support the concept that changes in mental status are the strongest clinical indicators of bacterial meningitis in elderly, the same findings being noted in our study in diabetic patients, whose median of age was higher than in non-diabetics (8. 9). In this category of patients, altered consciousness was noted in almost 70% of cases. Beside diabetes, alcohol consumption and/or liver cirrhosis were frequent additional underlying conditions in our study, being present in more than 80%. The etiology was related to the most prevalent pathogen causing nosocomial infections (Staphylococcus aureus/ coagulase negative staphyloccocci) in diabetics group, whereas in non-diabetic patients, S. pneumonia was more frequently isolated. In community-acquired meningitis in the same region, staphylococcus accounted for only 5% of cases with known etiology (5). This figure is in contrast with the findings of the Taiwanese report that mentioned Klebsiella species as the most frequent etiological agent of meningitis in diabetic patients (2). An investigation concerning community-acquired meningitis in the Hospital of Infectious Diseases Iași during reported that the overall mortality rate was 13.1%, lower than that found in the Korean study (19%) (4, 5). Kragsbjerg et al. in Sweden, revealed a mortality rate of 16% in patients over 14 years of age with pneumococcal meningitis, while, the mortality rate was 33% for those over 70 years of age (10). This value contrasts with the mortality rate of 30% reported for patients with pneumococcal meningitis in USA (11). The rate of death in our study was not 905

6 Cristina G. Petrovici et al. statistically different among patients with and without diabetes (12% vs. 25%). There are some controversies among the studies concerning the analysis of variables associated with prognosis. In van de Beek study, the strongest risk factors for an unfavorable outcome are those that are indicative of systemic compromise, a low level of consciousness and S. pneumoniae etiology (11). Factors associated with poor prognosis in a previous local analysis were: pneumococcal etiology, age over 60 and the presence of seizures or coma at admission (5). Other risk factors associated with fatality rate were the absence of headache (finding probably associated with the delay in diagnosis), immunocompromised status, presence of coma and pneumonia in the study of Papaiordanou (12). A high index of suspicion for meningitis must be maintained during the healthcare of diabetic patients with changes in mental status and/or fever, because a delay in diagnosis and therapy will result in a devastating illness. Accurate knowledge of epidemiology of sepsis and meningitis in diabetic patients and local pattern of antibiotic susceptibility may help us in filling the gaps related to the outcome of this particular category of patients. CONCLUSIONS This study shows a relatively high frequency of meningitis during sepsis in diabetic patients. Bacterial meningitis was more frequently microbiologically documented (S. aureus, Coagulase-negative staphylococci, Streptococcus pneumoniae, E. coli, Klebsiella spp.) from the CSF in diabetic population. The risk of late recognition was higher, mainly due to chronic complications of diabetes. Altered consciousness was more frequently in diabetic patients (68.8%) in whom the absence of fever at admission was a more common finding (88.4%) than in non-diabetic septic patients with meningitis. 906 REFERENCES 1. Lai WA, Chen SF, Tsai VW, Chang CC, Lu CH, Chuang YC, Chien CC, Huang CR et al Clinical Characteristics and Prognosis of Acute Bacterial Meningitis in Elderly Patients Over 65: A Hospitalbased Study. BMC Geriatr ; 11(1): Huang CR, Lu CH, Chang HW et al. Community-Acquired Spontaneous Bacterial Meningitis in Adult Diabetic Patients: An Analysis of Clinical Characteristics and Prognostic Factors. Infection 2002; 30(6): Charlson M, Szatrowski TP, et al. Validation of a combined co morbidity index. J Clin Epidemiol, 1994; 47 (11): Moon SY, Chung DR, Kim SW et al. Changing etiology of community-acquired bacterial meningitis in adults: a nationwide multicenter study in Korea. Eur J Clin Microbiol Infect Dis 2010; 29: Miftode E, Vâță A, Leca D, Hurmuzache M, Dorneanu O, Manciuc C, Luca V, Dorobăţ C. Community-acquired meningitis: a 10-year review. Rev Med Chir Soc Med Nat Iasi. 2009; 113(2): Chang WN, L, Tsai NW, Chuang YC, Chang CC, Chen SF, Chien CC. Changing epidemiology of adult bacterial meningitis in southern Taiwan: a hospital-based study. Infection, 2008; 36: Schut ES, Willeke FW et al. Hiperglicemia in bacterial meningitis: a prospective cohort study. BMC Infect Dis 2009; 9:

7 Bacterial meningitis during sepsis in diabetic patient 8. Erdem H, Kilic S, Coskun O, Ersoy Y, Cagatay A, Onguru P, Alp S. Community-acquired acute bacterial meningitis in the elderly in Turkey. Clin Microbiol Infect 2010; 16(8): Domingo P, Pomar V, Benito N, Coll P. The spectrum of acute bacterial meningitis in elderly patients. BMC Infect Dis 2013; 13: Kragsbjerg P, Kallman J, Olcen P. Pneumococcal Meningitis in Adults, Scand J Infect Dis 1994; 26(6): Van de Beek D, Gans J, Spanjaard L, Weisfelt M,. Reitsma JB. Clinical Features and Prognostic Factors in Adults with Bacterial Meningitis. N Engl J Med 2004; 351: Papaiordanou PM, Cadogan SM, Ribeiro Rosende M, Oliveira Campos E, Teixeira Garcia M, Moretti Branchini ML. Bacterial Meningitis in the Elderly: An 8-Year Review of Cases in a University Hospital. Braz J Infect Dis 1999; 3(3): NEWS METASTATIC CROHN DISEASE - A RARE CUTANEOUS ENTITY Metastatic Crohn disease is a rare cutaneous disorder characterized by noncaseating, granulomatous skin lesions present at sites anatomically separate from the gastrointestinal tract.first described by Crohn et al, Crohn disease is a chronic granulomatous inflammatory bowel disorder that may involve any segment of the gastrointestinal tract. Along with the characteristic gastrointestinal findings of this disease, patients with Crohn disease may also present with extraintestinal manifestations including ocular findings, musculoskeletal pathology, and mucocutaneous manifestations. Mucocutaneous findings are the most frequent extraintestinal manifestation of Crohn disease; 22% to 44% of patients present with mucocutaneous changes, which may be categorized as granulomatous skin disease. Included within the granulomatous cutaneous category are perianal, peristomal, and perifistular inflammatory lesions, which are contiguous with the gastrointestinal tract and are the most common cutaneous manifestations of Crohn disease. Parks et al were the first to describe the presence of sterile, noncaseating, granulomatous lesions of the skin at sites noncontiguous with the gastrointestinal tract in patients with Crohn disease. This entity is known as met a- static Crohn disease (MCD), the name first coined in 1970 by Mountain. It is characterized by cutaneous, noncaseating granulomas at sites anatomically separate from the gastrointestinal tract. It is the least common dermatologic manifestation of Crohn disease, and the differential diagnosis includes numerous similarly appearing, granulomatous skin entities. Males and females appear to be equally affected, and children tend to present with clinical lesions different from those of adults. An integration of clinical information, microscopic findings, and ancillary studies is necessary to accurately diagnose this rare cutaneous disease. Our objective is to review the clinical features, histopathologic characteristics, suggested pathogenesis, differential diagnosis, and current therapeutic options of metastatic Crohn disease. (Alan Siroy, Jay Wasman. Metastatic Crohn Disease - A Rare Cutaneous Entity. Arch Pathol Lab Med. 2012;136: ) Doina Butcovan 907