Fever Pitch

The three scariest words uttered in emergency medicine are undoubtedly, “Remember the patient…?” Those three words can make your heart skip a beat, can cause your stomach to churn, or can bring tears to your eyes. In this case, here’s how the sentence ends: “Remember the patient with the fever who came in during the middle of the night?”

This month, we’re discussing a classic late night blunder that led to a monster claim. A concerned mother brought her 75-day-old daughter to the Emergency Department and said that her infant had a 103-degree fever. There was minimal charting, but the triage note stated that the child had a fever, no cough, and one episode of vomiting. The intake nurse recorded a lot of basic history, including normal delivery (NSVD), that the family had a car seat and a pediatrician, and that the infant didn’t smoke. Interestingly, there was no nursing documentation of the infant’s vaccination status. The vital signs annotations registered a temperature of 102.6, a pulse of 140, and respiration of 30. The baby’s blood pressure wasn’t recorded.

The physician notes were similarly sparse. The emergency physician noted that the mother said that the baby had a fever of 103 at home, but there was no coughing, urinary symptoms, or diarrhea. The physical exam revealed that the infant was alert and easily consolable. The chart was template-driven and indicated a reddened tympanic membrane, but failed to identify which ear was infected. The baby’s mucus membranes were moist and her lungs were clear. Her abdomen was soft and nontender, and her skin was warm and dry. The emergency physician documented the baby’s neurological exam as appropriate for her age.

No laboratory tests or radiographs were ordered, and no repeat examination or trial of oral fluids was documented. The infant was given Tylenol and discharged with a diagnosis of a middle ear infection. The physician prescribed a 10-day supply of amoxicillin, and the mother was advised to administer Tylenol and follow up with their pediatrician the next day.

Later the next day, the mother took the infant to the pediatrician’s office. The pediatrician noted that the baby was pale, cool to the touch, and lethargic. The office called an ambulance and the infant was taken to the Emergency Department, where she had a septic workup, was intubated, and was ultimately diagnosed as having meningitis. The infant was administered antibiotics and admitted. After nearly a month of hospitalization, the child was discharged with hypoxic brain injury.

The plaintiffs claimed that the emergency physician should have ordered blood cultures and a urinalysis to exclude bacteremia and meningitis. They also said that the physician should have scheduled a follow-up visit within 24 to 48 hours of the first visit (even though the patient saw the pediatrician the next day). The defendant claimed that the bacteremia and meningitis developed after the patient left the hospital, and that the strain of pneumococcus responsible for the problems was resistant to amoxicillin.

A verdict of just under $2 million was returned.

Meningitis is the most common diagnosis involved in pediatric malpractice lawsuits, as well as one of the top two diagnoses seen in cases in which a child dies. It is one of the most serious of the common pediatric infections, with a fatality rate of five to ten percent. Thirty percent of neonates will additionally have long-term neurological sequelae.

There are now vaccines for three of the most common causes of meningitis: Neisseria meningitides, Streptococcus pneumonia, and Haemophilus influenzae type b (Hib). Introduced in the late 1980s, the Hib vaccine alone reduced the incidence of invasive Hib disease in children aged <5 years by 99%, to less than one case per 100,000 children. Despite the anti-vaccination movement, the average annual incidence rate of invasive Hib disease in children aged <5 years in the U.S. remains below the Healthy People 2020 goal of 0.27/100,000, with the CDC 2014 rate being 0.19/100,000. Similar results have arisen thanks to other vaccines. Nevertheless, it’s important to remember that no vaccine is 100% protective.

This case raises the question of how to determine which patients should be identified as potentially having meningitis and sepsis, and which patients you can safely send home. It’s important to note that there are no criteria that can safely identify low-risk patients in the neonatal period (age 1-28 days). Any fever during this time period requires a septic evaluation and should involve antibiotics. Similarly, if the decision is made not to perform invasive studies on a febrile, well-appearing infant, patients must have reliable follow-up.

Various criteria have been developed in an attempt to identify the 28- to 90-day-old infant who is at low risk for a serious infection. The Rochester, Boston, and Philadelphia criteria are protocols that strive to identify low-risk patients who do not require a septic workup, admission, and IV antibiotics. The criteria differ slightly, but the chance of occult bacteremia drops to less then one percent when all of the following are met:

  1. Infant appears well
  2. Infant has been previously healthy
  3. Infant has no evidence of skin, soft-tissue, bone, joint, or ear infection
  4. WBC count of 5,000-15,000 cells/μL and absolute band count is 1,500 bands/μL or less
  5. UA has less than 10 WBCs/HPF on microscopic examination of the urine
  6. If the patient has diarrhea, stool has less than 5 WBCs/HPF on microscopic examination
  7. No immunizations or antimicrobials were given in the preceding 48 hours
  8. Caretaker is available by telephone
  9. There is no infiltrate on chest radiograph (if one was obtained)

Another significant issue is antibiotic selection in relation to the increase in resistance to Streptococcus pneumoniae. If you are going to discharge the patient, I recommend administering an injection of a long-acting third-generation cephalosporin, such as ceftriaxone.

If IV antibiotics are required and the organism is unknown, the following is a recommended regimen based on the infant’s age:

  • < 30 days: ampicillin and an aminoglycoside or a cephalosporin
  • 30-60 days: ampicillin and a cephalosporin; because of resistance, consider vancomycin instead of ampicillin
  • >61 days: a cephalosporin or ampicillin plus chloramphenicol with vancomycin
  • Antivirals (i.e., acyclovir) should also be considered as well as steroids (i.e., dexamethasone)

The incidence of a serious bacterial infection in infants who are categorized as low risk after a full evaluation is small; unfortunately, only those that you miss will follow the three words, “Remember the patient…?”

The Monster Claim is the nightmare, $1,000,000 plus settlement or judgment (not including defense costs) that could happen to any unsuspecting physician or medical care provider. As in most cases in emergency medicine, it is a case of failure to suspect the diagnosis and involves a series of unfortunate events. We believe that it is important that all emergency providers become aware of these claims so they won’t repeat the same mistakes in the future.

Pin It on Pinterest