The EMT squad has just left your office, headed for the local community hospital with one of your patients in an ambulance. A short time ago, the 4-year and 11-month-old boy experienced a left-sided focal seizure in an examining room of the practice that progressed to a generalized tonic-clonic seizure. You were able to control the seizure with intravenous lorazepam, 0.1 mg/kg of body weight, while awaiting EMS. Now, you pause in the middle of an otherwise routine day of appointments to ask: What chain of events led to these tense moments?
Just over three weeks ago, this Caucasian boy, who has a notable history of cognitive developmental delay and reported penicillin allergy, was seen by one of your partners for right-sided acute otitis media (AOM). Despite the boy's speech delay, he is known in the practice to be able to communicate appropriately with his mother, and had complained of right-ear pain. Your partner, his regular pediatrician, started the patient on cefdinir (Omnicef).
An uneventful three weeks passed. Then, today, the boy was brought in by his mother with a four-day history of fever (maximum recorded axillary temperature, 102°F) and discharge from the right ear, along with an upper respiratory tract infection. In your partner's absence, you saw the patient. He wasn't in discomfort and hadn't complained of earache at home. Other than the discharge, fever, and respiratory complaint, your exam was unremarkable. You prescribed ofloxacin otic drops (Floxin Otic) for possible perforated AOM and advised his mother to bring the boy back in 72 hours for a recheck.
But the patient's mother returned just 10 minutes later, her son in her arms, agitated because he "wasn't making eye contact and was stiff with his jaws locked up". As she explains this, the seizure begins; you and your staff take action, and EMS is called.
Deterioration on admission
The story picks up on the pediatric ward of the community hospital, where the boy has been admitted. Blood, cerebrospinal fluid (CSF), and urine specimens are immediately sent for culture. The patient undergoes noncontrast cranial computed tomography; the radiologist interprets the scan as right-sided mastoiditis—"opacity of the right mastoid without gross evidence of bone destruction."
Spinal tap reveals the following CSF values: cell count, 7/mm3 (with a differential count of 92% neutrophils and 8% lymphocytes); glucose, 65 mg/dL; and protein, 83 mg/dL.
Complete blood count reveals a white blood cell count of 5.6 X 103 /μL (with a differential count of 23% segmented neutrophils, 47% band forms, 14% lymphocytes, and 16% monocytes); hemoglobin, 11.4 g/dL; and a platelet count of 214 X 103 /μL. Urinalysis parameters are within normal ranges.
A blood chemistry panel shows a serum sodium level of 137 mmol/L; potassium, 2.9 mmol/L; chloride, 103 mmol/L; bicarbonate, 22 mmol/L; glucose, 172 mg/dL; blood urea nitrogen, 2 mg/dL; serum creatinine, 0.8 mg/dL; aspartate aminotransferase, 304 IU/L; alanine aminotransferase, 149 IU/L; serum alkaline phosphatase, 154 IU/L; total bilirubin, 2.5 mg/dl; total protein, 3.6 g/L; serum albumin, <2.0 g/L; serum ammonia, 43 μmol/L (reference rage, 11 to 35 μmol/L); and serum lactate dehydrogenase, 575 IU/L (reference range, 76 to 182 IU/L).
The low serum albumin level is judged to likely reflect increased vascular permeability secondary to incipient septic shock. Because the boy is deteriorating clinically, he is transferred to the pediatric intensive care unit. Intravenous cefotaxime is started. With a prothrombin time of 22.2 sec (reference range, 10.2 to 12.3 sec) and an INR of 3.7, as well as an activated partial thromboplastin time of 40 sec (range, 24 to 35 sec), he is also given human activated protein C (drotrecogin [Xigris]) for disseminated intravascular coagulation, and vasopressors for hypotension associated with the septic shock.
When preliminary results of blood and CSF cultures showed gram-positive cocci that evening, IV vancomycin is added to cover the usual suspect, Streptococcus pneumoniae.
Based on the dictum that not every patient reads the book, the real "shock" comes afterward—but it's for you and the medical team. Within four and 14 hours, cultures of CSF and blood, respectively, grow β-hemolytic group A streptococcus (GAS)!
Not a customary culprit
GAS is the cause of a variety of common clinical illnesses, including tonsillopharyngitis, scarlet fever, cellulitis, cervical lymphadenitis, erysipelas, and otitis media. Although necrotizing fasciitis, osteomyelitis, sepsis and toxic shock syndrome are well recognized among invasive GAS infections, GAS-associated meningitis occurs but rarely, accounting, in various studies, for none to 3.2% of all bacterial meningitides and for 0.8% of all invasive GAS infections reported in the US between 1995 and 1999.
GAS meningitis has been reported in all age groups, with disproportionately higher incidence in children. The prevalence of comorbid conditions is high among all age groups: In one report, 67% of patients had a concomitant or earlier neurosurgical condition. That report also noted the growth of variable M-type GAS with genotypic variability of speC and speA in strains isolated from different patients.
Among adults, GAS meningitis is often preceded by otitis media or sinusitis. In one retrospective study, 80% of patients had an extrameningeal focus, predominantly in the ear, nose, throat, and upper respiratory tract.Similar to what was seen in your patient, an association between GAS meningitis and both OM and mastoiditis has been documented in several pediatric case reports and case series. Newborns, patients with a CSF leak, and those who have an immunosuppressive condition are the groups at high risk of GAS meningitis. Your patient was in none of these categories.
GAS meningitis is a virulent disease, with a mortality of 10% to 12% and a high incidence of neurologic sequelae in children and adults. Mortality from GAS meningitis is comparable to mortality associated with Haemophilus influenzae and Neisseria meningitidis meningitis. Overall, the incidence of neurologic sequelae among children who have had GAS meningitis is 36% to 46%.(In fact, your patient had residual left-sided hemiparesis at discharge.)
Timeline of subsequent events
Clinical case definition of streptococcal TSS in a child
The boy had a protracted hospital course, exhibiting clinical features of meningitis and septic shock. In addition, the clinical findings fulfill the definition of streptococcal toxic shock syndrome (TSS) . Notwithstanding what was seen in this case, TSS is rarely associated with meningitis in invasive GAS infection. The other unusual finding here was a low WBC count in CSF; a recent review of the international literature on GAS meningitis reported otherwise.
Once the pathogen was confirmed, and in light of the TSS, the antimicrobial regimen was switched to a four-week course of ampicillin and clindamycin to treat the mastoiditis. (Note: The boy tolerated ampicillin well; the rash reported to your partner by his parents when the boy was an infant, and assumed to be penicillin allergy because it was coincidental to a course of amoxicillin was apparently of a different, unknown cause.) He was also given 2 g/kg of IV immune globulin for the TSS.
After recovery from the acute illness, the patient was transferred to a children's hospital for neurorehabilitation. Before transfer, he underwent brainstem auditory evoked response testing which confirmed bilaterally normal hearing. A magnetic resonance imaging scan of the brain showed right-sided cerebral infarcts in the parieto-occipital areas.
Drug of the hour: Clindamycin
The efficacy of clindamycin in GAS infection is independent of the size of the inoculum or the stage of bacterial growth. Additionally, the drug suppresses both synthesis of bacterial toxin and release of penicillin-binding proteins that are involved in cell-wall synthesis and degradation. Clindamycin also has a longer post-antibiotic effect compared to that of β-lactam antibiotics. It enhances phagocytosis of GAS by inhibiting M-protein synthesis—observations that are based on animal and in vitro studies. It is also postulated that the effectiveness of clindamycin may be related to its ability to modulate host immune response.
Because the minimum inhibitory concentration of penicillin has remained relatively unchanged for GAS isolates in the past eight decades, penicillin, 250,000 U/kg/day to a maximum of 24 million U every 24 hours, is recommended to attain a CSF level of 1 μg/mL. The usual empiric choice of a third-generation cephalosporin to cover more common bacteria that cause meningitis does suffice to treat GAS, although it is advisable to tailor therapy with antibiotics specific to GAS.3 (For a discussion of selecting antibiotics empirically, read What bug, which drug? Optimizing empiric antimicrobial therapy).
A minute of your time, please!
Except for residual, but improving, left hemiparesis, the patient's clinical course has been unremarkable since he was discharged from the children's hospital and returned to your practice. In accordance with recommendations made by the Centers for Disease Control and Prevention for invasive GAS disease among household contacts of index patients, no other of the boy's family members were screened, or offered chemoprophylaxis, for GAS.
To sum up: GAS is rare in the era of antibiotics. Always take a few minutes, however—10, say?—to consider it as a cause of meningitis that is associated with recurrent otitis media and mastoiditis.