When evaluating children with CSF pleocytosis, bacterial meningitis is the greatest concern, although most children end up having aseptic meningitis. Therefore, these children are routinely admitted and treated with broad spectrum antibiotics while awaiting final culture results. The Bacterial Meningitis Score is a clinical prediction rule that was developed to help classify patients into low risk for bacterial meningitis. Patients that lack all of the 5 criteria are classified into low risk. The original study (Pediatrics. 2002;110:712-719) looked at previously healthy children 29 days to 19 years old with CSF pleocytosis (CSF WBC ≥7cells/μL) and showed a negative predictive value of 100%, meaning that none of the 144 patients classified as low risk had bacterial meningitis. The validation study (JAMA. 2007;297:52-60) showed similar results (sensitivity 98.3%; NPV 99.9%). There were only 2 children out of the 1714 categorized as low risk that had bacterial meningitis. Both of these children were under 2 months of age and had E.coli meningitis from a urinary tract infection. Therefore, the authors caution against the use of Bacterial Meningitis Score in children less than 2 months where the sensitivity drops to 92%. Also, the scoring system should not be used inn children pre-treated with antibiotics because this may affect the CSF profiles. The authors conclude that using the Bacterial Meningitis Score prediction rule to assist with clinical decision making could substantially reduce unnecessary hospital admissions for children with CSF pleocytosis at very low risk of bacterial meningitis.
Source: Nate Kuppermann, MD
(ed. note: Dr. Kupperman served as an investigator on this paper.)
2. The Seat Belt Sign: a Harbinger of Internal Injury!
There are three types of seat belt signs, the most serious of these being the lower abdominal bruising caused by the compression by the lap belt. Sokolove et al (Acad Emerg Med 2005;12) showed that children with the abdominal seat belt sign were 3x more likely to suffer intra-abdominal injury with higher rates of intestinal, pancreatic and mesenteric injury. All of these children with injury had abdominal pain and tenderness. CT scans are inadequate at detecting these injuries (hollow viscus injury, mesenteric tears, blunt pancreatic injury) in the acute setting. FAST scan has also been shown to be inadequate as a screening test with a sensitivity of 78%. So what should our approach be for diagnosing these rare injuries that are radiographically invisible? We should start with CT for those that have abdominal pain or tenderness to rule-out solid organ injury or obvious abnormalities. They may then need to be observed with serial abdominal exams for the development of peritonitis and surgical consultation. The neck seat belt sign is much more rare and is also associated with serious injury. Of the 131 patients studied at USC (DiPerna et. al Am Surg 2002) with the neck seat belt sign, only 1 patient had a serious injury (intimal flap with carotid artery stenosis). The presence of the cervicothoracic belt sign plus and abnormal physical exam provides an adequate screening test. These abnormalities include:
LOC, GCS deficits
Clavicle and 1st rib fractures
Anterior neck triangle involvement
Hoarseness, dysphagia, odynophagia
Any neurologic findings or complaints-even transient
If any of these are present, then further imaging using CT or MR angiography is warranted to look for pseudoaneurysms or carotid flaps.
Source: Jan Shoenberger MD, USC Essentials 2007
Source: Jan Shoenberger MD, USC Essentials 2007
3. Etomidate OK in Sepsis
Dr. Ron Walls weighs in on his perspective of the safety of etomidate in sepsis and presents a counterpoint to the September interview with Dr. Al Sacchetti. It is true that a single dose of etomidate can suppress cortisol production and blunt the ACTH response, as measured by the cosyntropin stimulation test. This is a transient effect. Although the cortisol levels are low, they are still in the normal range. Early studies show increased mortality in septic patients that received etomidate, but theses studies did not separate out those patients that were intubated vs. not intubated. Therefore, they were only able to show an association and not a cause and effect. This makes sense- the patients that received etomidate were sicker because they needed to be intubated and therefore had higher mortality. Dr. Walls returns to the den Brinker study (Intensive Care Med 2008;34:163) that retrospectively analyzed children with meningococcal sepsis and found that etomidate use was one of the factors predictive of a decrease in cortisol. No mortality effect was actually observed but the authors argue that etomidate “might have” increased the risk for mortality. Dr. Walls concludes that “there has been no credible study that goes against using etomidate in septic shock and…the greatest service we can do to our patients is to conduct the large, high quality trials needed to base our clinical practice on truly robust evidence.”
Source: Ron Walls MD; Annals of Emergency Medicine 2008;52(1):13-14