Abstract
Community-acquired bacterial meningitis (CABM) morbidity and mortality remains high in those infected. Rapid diagnosis and treatment is paramount to reducing mortality and improving outcome. This retrospective cohort study aims to assess the time from presentation to diagnosis and treatment of vaccine preventable CABM as well as identify possible factors associated with delays in diagnosis and antibiotic administration. A retrospective chart review was conducted of individuals who presented to Columbia University Irving Medical Center (CUIMC), Children’s Hospital of New York (CHONY), Mount Sinai Medical Center, and Weill Cornell Medical Center with BM due to Haemophilus influenzae type B, Streptococcus pneumoniae, and Neisseria meningitidis between January 1, 2012 and December 31, 2017. Diagnosis was delayed by more than 8 hours in 13 patients (36.1%) and 5 individuals (13.9%) had a delay of 4 hours or more from presentation to the administration of antibiotics with appropriate CNS coverage. All of these patients were also initially misdiagnosed at an outpatient clinic, outside hospital, or emergency department. This retrospective study identified febrile and/or viral infections not otherwise specified and otitis media as the most common misdiagnoses underlying delays from presentation to diagnosis and to antibiotic treatment in those with BM.
Keywords: bacterial meningitis, diagnosis, central nervous system infection, neurological sequelae, vaccine preventable, misdiagnoses
Introduction
The advent of routine vaccinations for 3 etiological agents (Haemophilus influenzae type B, Streptococcus pneumoniae, and Neisseria meningitidis) of community acquired bacterial meningitis (CABM) has altered the epidemiological landscape with decreased incidence over the last 2 decades, shifting age-specific prevalence and serotype replacement.1-3 Despite the effectiveness of vaccines in the prevention of CABM, mortality rates remain high at 10-30%, with untreated cases resulting in death within 24 to 48 hours.4-6 Permanent sequelae such as epilepsy, hearing loss, and neurocognitive deficits afflict at least 10 to 25% of survivors (WHO, 2017). 6 In 1 study, delays in diagnosis and treatment were associated with increased mortality by 12.6% per hour and increased occurrence of permanent neurological sequelae including epilepsy, hearing loss, and neurocognitive deficits in 10-20% of survivors.4,7 A prospective study found that a delay in antibiotic administration of more than 3 hours after presentation with pneumococcal meningitis was independently associated with 3-month mortality rates. 8 Comparable retrospective studies found that the adjusted odds ratio for mortality was 8.4 times higher in patients experiencing delays greater than 6 hours from presentation to appropriate antibiotic administration.9,15 While previous work does focus on the implications of delays in diagnosis and treatment, it is unclear what factors may contribute to such delays. This retrospective study aims to identify possible sociodemographic, epidemiological, and clinical factors that may influence delays in the diagnosis and treatment of vaccine preventable CABM caused by H. influenzae, S. pneumoniae, and N. meningitidis at 3 tertiary medical institutions in New York City from 2010 to 2017.
Methods
Study Design
A retrospective chart review was conducted of individuals who presented to Columbia University Irving Medical Center (CUIMC), Children’s Hospital of New York (CHONY), Mount Sinai Health System, and Weill Cornell Medical Center with bacterial meningitis due to Haemophilus influenzae type B, Streptococcus pneumoniae, and Neisseria meningitidis between January 1, 2012 and December 31, 2017. Electronic medical records (EMRs) of individuals discharged with ICD-9 and ICD-10 codes A39.0 (Meningococcal Meningitis), G00.0 (Hemophilus Meningitis), G00.1 (Pneumococcal Meningitis), G00.2 (Streptococcal Meningitis), 36 (Meningococcal Meningitis), 320 (Hemophilus Meningitis), 320.1 (Pneumococcal Meningitis), and 320.2 (Streptococcal Meningitis) were reviewed. Those who were younger than 2 months of age and had a surgical implant of any kind in the brain or skull were excluded. All included cases were CSF culture and/or PCR positive within the same admission. Sociodemographic and clinical data at presentation, 3 to 6 month and 12 month follow up were collected. General demographic information, employment status, immunocompromised conditions, and substance abuse history were extracted to identify pertinent sociodemographic factors. Clinical data of interest included length of stay in the ED, ICU, and wards as well as the time difference between differential diagnoses and lumbar puncture procedures. Poor outcome was defined as a GOS of less than four and favorable outcome a score of 4 or more.
Misdiagnosis was defined as an initial differential at presentation that did not include meningitis and/or the diagnosis of an illness other than meningitis leading up to admission at an outside clinic or transferring hospital. Data used to calculate time variables were collected from timestamps derived from clinical and procedural notes. Time of presentation is defined as the time of admission to the emergency room for those who presented to the ED or time of admission for direct admits. Time of LP is defined as the timestamp recorded on the procedural note for the initial LP. Time of diagnosis was defined as the read-back time of confirmatory laboratory results to the clinical team. Timestamp data for outside hospital (OSH) transfers was unavailable and such cases were excluded from time calculations. Delay from presentation to lumbar puncture (LP) was defined as more than 6 hours, delayed presentation to administration of antibiotics was defined as more than 4 hours, and delay of diagnosis was defined as more than 8 hours.
Statistical Analysis
Descriptive analyses including mean, median, and interquartile range were calculated for each continuous variable including time to from presentation to LP, diagnosis, administration of appropriate antibiotics and length of hospital stay. Frequencies of each categorical variable (diagnosis, path to admission, outcome, sociodemographic, clinical and follow-up data) were calculated to compare differences between groups based on diagnosis and outcome.10,11
Results
Thirty-six patients presented with BM during the study period and 27 patients (75%) had favorable outcomes at discharge, while 4 patients (11.1%) expired. Sociodemographic characteristics of the cohort are detailed below in Table 1 and follow-up data is presented in Table 2. S. pneumoniae was identified in 32 (88.9%) cases, N. meningitidis in 2 (5.6%), and H. influenzae in 2 (5.6%) patients. Twelve (33.3%) individuals presented to an outpatient clinic prior to admission, with 2 individuals (5.6%) admitted directly from a primary care clinic. Twenty-four (66.7%) initially presented to the ED of study sites with a median ED stay of 8 hours (IQR 6), and 11 individuals (30.6%) were transferred from an outside hospital (OSH). Ten (27.8%) of these transfer patients had been diagnosed with BM and had begun treatment at the OSH before being transferred for further management.
Table 1.
Sociodemographic and Epidemiological Factors.
| All | S. pneumoniae | H. influenzae | N. meningitidis | |||||
|---|---|---|---|---|---|---|---|---|
| Total | 36 | % | 32 | % | 2 | % | 2 | % |
| Sociodemographic factors | ||||||||
| Age (Years) | 42 | 42 | 2 | 1.25 | ||||
| Gender | ||||||||
| Female | 12 | 33.3% | 11 | 34.4% | 1 | 50.0% | 0 | 0.0% |
| Male | 24 | 66.7% | 21 | 65.6% | 1 | 50.0% | 2 | 100.0% |
| Race | ||||||||
| Caucasian | 17 | 47.2% | 16 | 50.0% | 1 | 50.0% | 0 | 0.0% |
| Black/African-American | 10 | 27.8% | 9 | 28.1% | 1 | 50.0% | 0 | 0.0% |
| Asian | 1 | 3.6% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Other/unknown | 8 | 22.2% | 6 | 18.8% | 0 | 0.0% | 2 | 100.0% |
| Ethnicity | ||||||||
| Hispanic/Latino | 10 | 27.8% | 10 | 31.3% | 0 | 0.0% | 0 | 0.0% |
| Non-Hispanic/Latino | 16 | 44.4% | 15 | 46.9% | 1 | 50.0% | 0 | 0.0% |
| Unknown | 10 | 27.8% | 7 | 21.9% | 1 | 50.0% | 2 | 100.0% |
| Limited English proficiency | 8 | 22.2% | 8 | 25.0% | 0 | 0.0% | 0 | 0.0% |
| Employment | ||||||||
| Retired | 4 | 11.1% | 4 | 12.5% | 0 | 0.0% | 0 | 0.0% |
| Unemployed | 6 | 16.7% | 6 | 18.8% | 0 | 0.0% | 0 | 0.0% |
| On disability | 0 | 0.0% | 0 | 0.0% | 0 | 0.0% | 0 | 0.0% |
| In school | 1 | 2.8% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Unknown | 6 | 16.7% | 6 | 18.8% | 0 | 0.0% | 0 | 0.0% |
| <5 years of age | 11 | 30.6% | 7 | 21.9% | 2 | 100% | 2 | 100.0% |
| Working | 8 | 22.2% | 8 | 25.0% | 0 | 0.0% | 0 | 0.0% |
| Epidemiological factors | ||||||||
| Total immunocompromised | 16 | 44.4% | 16 | 50.0% | 0 | 0.0% | 0 | 0.0% |
| Diabetes mellitus | 3 | 8.3% | 3 | 9.4% | 0 | 0.0% | 0 | 0.0% |
| Cancer | 4 | 11.1% | 4 | 12.5% | 0 | 0.0% | 0 | 0.0% |
| HIV | 1 | 2.8% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Asplenia | 1 | 2.8% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Alcoholism/drug abuse disorder | 4 | 11.1% | 4 | 12.5% | 0 | 0.0% | 0 | 0.0% |
| Transplant | 1 | 2.8% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Other | 2 | 5.6% | 2 | 6.3% | 0 | 0.0% | 0 | 0.0% |
| Chronic infections | 6 | 16.7% | 6 | 18.8% | 0 | 0.0% | 0 | 0.0% |
| Sinus | 3 | 8.3% | 3 | 9.4% | 0 | 0.0% | 0 | 0.0% |
| Upper respiratory | 1 | 2.8% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Ear | 1 | 2.8% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Meningeal | 1 | 2.8% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Recent illness | 24 | 66.7% | 4 | 11.1% | 1 | 50.0% | 2 | 100% |
| Upper respiratory | 17 | 47.2% | 2 | 6.3% | 1 | 50.0% | 2 | 100% |
| Sinus | 7 | 19.4% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Ear | 8 | 22.2% | 1 | 3.1% | 0 | 0.0% | 0 | 0.0% |
| Gastrointestinal | 3 | 8.3% | 0 | 0.0% | 0 | 0.0% | 0 | 0.0% |
| Alcohol use in past year | 13 | 36.1% | 13 | 40.6% | 0 | 0.0% | 0 | 0.0% |
| Tobacco use in past year | 8 | 22.2% | 8 | 25.0% | 0 | 0.0% | 0 | 0.0% |
| Drug use in past year | 7 | 19.4% | 7 | 21.9% | 0 | 0.0% | 0 | 0.0% |
Table 2.
Clinical and Outcome Factors.
| Clinical information | ||
|---|---|---|
| Measures | Median | IQR |
| Length of hospital stay | 12 days | 15 days |
| Length of ICU stay | 6 days | 6 days |
| Initial neurological symptom to presentation at any clinical setting (OSH, clinic, ED) | 1 day | 2 days |
| Presentation to LP | 8 hours | 7 hours |
| Presentation at hospital to diagnosis (First positive CSF culture or PCR result) | 12 hours | 9 hours |
| LP to diagnosis | 3 hours | 5 hours |
| Presentation to antibiotics | 4 hours | 5 hours |
| Misdiagnoses | N | % |
| Total | 13 | 36.1% |
| Febrile and/or viral infections not otherwise specified | 5 | 13.9% |
| Otitis media | 2 | 5.6% |
| Stroke | 1 | 2.8% |
| Drug overdose | 1 | 2.8% |
| Alcohol withdrawal | 1 | 2.8% |
| Mechanical problems relating to a recent injury/fall | 1 | 2.8% |
| Post-epidural headache | 1 | 2.8% |
| Neuroleptic malignant syndrome | 1 | 2.8% |
| Complications | N | % |
| Patients in ICU | 31 | 86.1% |
| Patients intubated | 19 | 52.7% |
| Patients with seizures | 11 | 30.6% |
| Status epilepticus | 3 | 8.3% |
| Patients with cerebral edema | 8 | 22.2% |
| Outcome | ||
| Favorable outcome at discharge (GOS ≥ 4) | 27 | 75% |
| Deceased patients | 4 | 11.1% |
| Symptoms at discharge | 36 | 100% |
| Hearing loss | 11 | 30.6% |
| Seizures | 3 | 8.3% |
| Behavioral/cognitive deficits | 16 | 44.4% |
| Unable to complete ADLs | 11 | 30.6% |
| Symptoms at 3-6 month follow up | 23 | 63.9% |
| Hearing loss | 8 | 22.2% |
| Seizures | 2 | 5.6% |
| Behavioral/cognitive deficits | 12 | 33.3% |
| Unable to complete ADLs | 5 | 13.8% |
| Symptoms at 1 year follow up | 17 | 47.2% |
| Hearing loss | 6 | 16.7% |
| Seizures | 1 | 2.8% |
| Behavioral/cognitive deficits | 6 | 16.7% |
| Unable to complete ADLs | 3 | 8.3% |
Diagnosis was delayed by more than 8 hours in 13 patients (36.1%), all of whom were initially misdiagnosed at either an outpatient clinic, OSH, or ED. Seventeen patients (47.2%) had a delay of more than 6 hours from presentation to LP. Of those delayed LP, 11 patients (64.7%) had initial misdiagnoses. Additionally, 2 (11.8%) due to the healthcare proxy’s hesitancy to consent to LP, and 1 (5.9%) was deferred out of concern for cerebral edema. Two (5.6%) patients, who were initially misdiagnosed, were non-English speaking. Five individuals (13.9%) who had a delay of 4 hours or more from presentation to the administration of antibiotics with appropriate CNS coverage also had initial misdiagnoses.
Discussion
This retrospective study identifies the most common misdiagnoses that may contribute to management delays, and poor outcome. The rates of residual hearing loss, seizure, behavioral and cognitive deficits, and ability to complete ADLs in this cohort remained consistent with those found in past studies.1,3,7,12-14 Overall survival rates, as well as the rates of most common sequelae of BM, were also comparable to those in previous studies.
This study reports that initial misdiagnoses accounted for all delays of 4 hours or more from presentation to LP, and to targeted antibiotic administration.8,9 All individuals who were not initially misdiagnosed were administered antibiotics in 2 hours or less. Contributing factors to misdiagnosis in this study include initial presentation to an outpatient clinic and a preceding or coinciding illness veiling BM. Factors influencing misdiagnosis such as clinical presentation at outpatient care centers, symptomology shared by BM and other common illnesses, language barriers and other sociodemographic factors should be explored in further studies. Learning how to better distinguish BM cases from other illnesses earlier in disease course and facilitate faster diagnosis and treatment would be immeasurably useful in improving mortality and morbidity rates in such cases.
This study has several important limitations. Firstly, the ICD discharge codes used to identify potential cases likely do not include all cases and may not fully capture the diagnostic and clinical landscape of BM. As we were reliant on EMR records, this study also lacks detailed symptomology, which could possibly identify clinical presentations associated with misdiagnoses and diagnostic delays. Additionally, the accuracy of measured time intervals is reliant on the accuracy of the timestamps registered in the medical records. Lastly, the study has a small sample size and focuses only on 3 sites in NYC, limiting generalizability.
Future multicenter large scale studies are needed to investigate contributing factors of diagnostic and treatment delays such as detailed clinical symptomology, language barriers, and geographic proximity/physical accessibility to care facilities that may contribute to poor prognosis. Identifying such factors will help inform possible solutions to these barriers and provide more comprehensive, high-quality healthcare to communities. As the epidemiological profile of BM continues to evolve by virtue of antibiotic resistant strains and serotype replacement, it is essential to optimize preventative, diagnostic and therapeutic measures to reduce mortality and increase quality of life for survivors.
Acknowledgments
The authors thank Yifei Sun for her contributions to statistical data analysis.
Authors’ Note: Sarah D. Torres and Mitashee Das: Data collection, initial analyses and writing, review and revision of manuscript. Carla Kim: Data collection, analyses and writing, review and revision of manuscript. Jyoti V. Ankam: Initial analyses and writing, review and revision of manuscript. Nicole Luche and Brittany Glassberg: Data collection, review and revision of manuscript. Michael Harmon: Data collection, revision of manuscript. Emily M. Schorr, Don Weiss, Jacqueline S. Gofshteyn, and Anusha K. Yeshokumar: Reviewed and revised the manuscript. Stephen S. Morse: Conceptualized and designed the study, data collection, drafted the initial manuscript, and reviewed and revised the manuscript. Kiran T. Thakur: Conceptualized and designed the study, drafted the initial manuscript, and reviewed and revised the manuscript.
Declaration of Conflicting Interests: The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Jacqueline Gofshteyn is supported by the NIH/NINDS NSADA-K12 Career Development Award (NS5250799523). Kiran Thakur is supported by the National Institute of Health, NINDS K23 NS105935-01 and NIH/NICHD 1R01HD074944-01A1.
ORCID iDs: Carla Y. Kim, BA 
https://orcid.org/0000-0001-5714-7212
Kiran T. Thakur, MD
https://orcid.org/0000-0003-0050-0323
References
- 1.Thigpen MC, Whitney CG, Messonier NE, et al. Bacterial meningitis in the United States, 1998-2007. N Engl J Med. 2011;364(21):2016–2025. [DOI] [PubMed] [Google Scholar]
- 2.Hsu HE, Shutt KA, Moore MR, et al. Effect of pneumococcal conjugate vaccine on pneumococcal meningitis. N Engl J Med. 2009;360(3):244–256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Brouwer MC, Tunkel AR, van de Beek D. Epidemiology, diagnosis, and antimicrobial treatment of acute bacterial meningitis. Clin Microbiol Rev. 2010;23(3):467–492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Glimaker M, Johansson B, Grindborg O, Bottai M, Lindquist L, Sjolin J.Adult bacterial meningitis: earlier treatment and improved outcome following guideline revision promoting prompt lumbar puncture. Clin Infect Dis. 2015;60(8):1162–1169. [DOI] [PubMed] [Google Scholar]
- 5.Cohn AC, MacNeil JR, Harrison LH, et al. Changes in Neisseria meningitidis disease epidemiology in the United States, 1998-2007: implications for prevention of meningococcal disease. Clin Infect Dis. 2010;50(2):184–191. [DOI] [PubMed] [Google Scholar]
- 6.World Health Organization (WHO). Meningococcal meningitis; fact sheet 2017. 2017. Updated December 2017. Accessed April 20, 2019. http://www.who.int/mediacentre/factsheets/fs141/en/
- 7.Oordt-Speets AM, Bolijn R, van Hoorn RC, Bhavsar A, Kyaw MH. Global etiology of bacterial meningitis: a systematic review and meta-analysis. PLoS One. 2018;13(6):e0198772. doi:10.1371/journal.pone.0198772 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Auburtin M, Wolff M, Charpentier J, Varon E, et al. 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. 2006;34(11):2758–2765. [DOI] [PubMed] [Google Scholar]
- 9.Proulx N, Fréchette D, Toye B, Chan J, Kravcik S. Delays in the administration of antibiotics are associated with mortality from adult acute bacterial meningitis. QJM. 2005;98(4):291–298. [DOI] [PubMed] [Google Scholar]
- 10.RStudio Team. RStudio: Integrated Development for R 2015. RStudio Inc.; 2015. [Google Scholar]
- 11.SAS Institute Inc. SAS. 2013. [Google Scholar]
- 12.van de Beek D, de Gans J, Tunkel AR, Wijdicks EF. Community-acquired bacterial meningitis in adults. N Engl J Med. 2006;354(1):44–53. [DOI] [PubMed] [Google Scholar]
- 13.Zoons E, Weisfelt M, de Gans J, et al. Seizures in adults with bacterial meningitis. Neurology. 2008;70(22 Pt 2):2109–2115. [DOI] [PubMed] [Google Scholar]
- 14.Worsoe L, Caye-Thomasen P, Brandt CT, Thomsen J, Ostergaard C. Factors associated with the occurrence of hearing loss after pneumococcal meningitis. Clin Infect Dis. 2010;51(8):917–924. [DOI] [PubMed] [Google Scholar]
- 15.Bodilsen J, Dalager-Pedersen M, Schonheyder HC, Nielsen H. Time to antibiotic therapy and outcome in bacterial meningitis; a Danish population-based cohort study. BMC Infect Dis. 2016;16(1):392. doi:10.1186/s12879-016-1711-z [DOI] [PMC free article] [PubMed] [Google Scholar]