The trusted source for
healthcare information and
Source: Kanegaye JT, et al. Lumbar puncture in pediatric bacterial meningitis: Defining the time interval for recovery of cerebrospinal fluid pathogens after parenteral antibiotic pretreatment. Pediatrics 2001;108: 1169-1174.
Despite a dearth of relevant literature, most clinicians believe there is a minimum window of several hours after the initiation of parenteral antibiotics during which cerebrospinal fluid (CSF) can be obtained with accurate culture results. This study sought to examine the truth of this commonly held belief.
This retrospective study from Children’s Hospital of San Diego examined the medical records of children discharged with final diagnoses of bacterial meningitis or suspected bacterial meningitis during a five-year period. Patients were included if they met any of the following criteria: CSF culture positive for a known bacterial pathogen; positive CSF antigen study or gram stain along with CSF white blood cell count (WBC) higher than 10/mm3; blood culture positive with CSF WBC higher than 100/mm3; or CSF WBC higher than 4000/mm3 in the absence of bacterial isolate. Exclusion criteria included positive CSF viral study, neural tube defect, CSF shunt, penetrating injury, neurosurgical procedure in the last month other than lumbar puncture (LP), or incomplete medical record.
One hundred twenty-eight patients met study criteria, with 104 (81%) due to positive CSF cultures and the remainder divided equivalently among the other inclusion criteria. Patients were classified into one of three groups: 39 (30%) of patients had delayed LP (LPd) when LP occurred after the initiation of parenteral antibiotics; 34 (27%) had LP prior to parenteral antibiotics (LPp); and 55 (43%) had LP pre- and post-parenteral antibiotic administration (LPp/p). The primary outcomes measured were the yield of CSF cultures in the LPd and the LPp/p groups by organism and time from antibiotic administration.
Bacterial organisms were identified in 125 (98%) cases through CSF culture, blood culture, or CSF antigen study. Streptococcus pneumoniae (38%), Neisseria meningitidis (29%), and group B Streptococcus (16%) were the most common pathogens. Negative CSF cultures occurred in 44% of patients in whom parenteral antibiotics were started prior to LP, compared with 8% among patients undergoing LP prior to parenteral antibiotic administration. Antibiotics used in the LPd and LPp/p groups almost always included a high- dose, third-generation cephalosporin.
Fifteen patients had LP within one hour after initiation of antibiotics, with negative CSF cultures in three of nine cases of meningococcal meningitis. The earliest sterilization of N. meningitides occurred at 15 minutes after the initiation of ceftriaxone infusion. Only one culture remained positive after 2.3 hours. Culture-negative pneumococcal meningitis occurred as early as 4.3 hours after initiation of antibiotics. In CSF cultured between four and 10 hours after antibiotic initiation, five of seven patients with pneumococcal meningitis had negative CSF cultures. Several pneumococcal isolates were resistant to penicillin and/or ceftriaxone. CSF cultures for group B Streptococcus obtained within eight hours of parenteral antibiotics remained positive.
The authors conclude that sterilization of CSF may occur more rapidly after antibiotic initiation than previously suspected, especially among patients with meningococcal meningitis. They also emphasize that the lack of adequate CSF cultures may lead to an inability to properly tailor therapy to susceptibility, or to unnecessarily prolonged treatment among patients with suspected bacterial meningitis. The authors suggest that hemodynamically stable patients with suspected meningitis and no evidence of herniation or cerebral edema should receive LP prior to the administration of antibiotics.
Commentary by Jacob W. Ufberg, MD
The results of this study came as a surprise to me, as I’m sure they did for most readers. However, the retrospective design of this study does lead to the usual limitations regarding the accuracy of medical records. It often is difficult to tell from nursing records whether a documented time represents the beginning or end of antibiotic administration, and rates of administration surely varied from case to case. Additionally, the dosage of administration was not uniform. This being said, a prospective study would be impossible to perform due to the ethical and practical concerns involved.
The authors point out that as more children receive the pneumococcal vaccine, an increased percentage of meningitis likely will be meningococcal. This likely will result in earlier sterility of CSF isolates and an increased need for early LP. We must note, however, the medical and legal risks of delaying antibiotics in children with suspected bacterial meningitis. The answer is not delayed antibiotics, but earlier LP. As in any critically ill patient, many things must happen simultaneously. We should be performing the LP while the infusion is being prepared and started, thus obtaining the necessary culture material without delaying antibiotic administration.
Dr. Ufberg, Assistant Professor of Emergency Medicine, Department of Emergency Medicine, Temple University School of Medicine, Philadelphia, is on the Editorial Board of Emergency Medicine Alert.