You are working the morning shift in the ED, and are then asked to take a “critical lab value” on a patient who was discharged yesterday. While speaking with the lab, you quickly review the medical record and see that the patient is a 5 year old child with no known medical problems who was found to have a unilobar pneumonia and was prescribed amoxicillin. According to the chart, the patient was not hypoxic, non-toxic appearing, had no recent hospitalizations, no recent antibiotic use, and was up-to-date on their vaccinations and without any recent travel. The lab goes on to say there is positive growth in 2/4 blood culture bottles, making it hard to blow off as contamination. Fortunately, the previous provider noted follow up with the pediatrician later this afternoon.
What would you do in this situation? Does the child need admission, monitoring, and IV antibiotics? Will they have a worse outcome with oral antibiotics only? Would you have the child come in for a re-evaluation, or just alert the PMD (since the child already has a scheduled appointment)?
In this Canadian study, they reviewed all 480 true positive bloodstream infections between July 1, 2002 and March 31, 2011 that were discharged from the ED. 339 out of these 480 patients were prescribed antibiotics (70%), with 87% of those prescribed an antibiotic having cultures demonstrating susceptibility to the prescribed antibiotic.
Compared to ALL patients (321,045 in total) discharged during the study period, those with positive cultures had a statistically significant increase in mortality 2 months from the initial visit. However, culture-positive patients that were prescribed an antibiotic effective against the offending organism had similar mortality 2 months after the initial visit compared to all of the discharges during the same time period.
We’ve known that pyelonephritis carries anywhere from a 15-30% rate of positive blood cultures and that they manage just fine. In the previously mentioned Canadian study, there was conscious decision that was made that a patient was healthy enough clinically to go home. If the patient has a history of IV drug abuse and that pneumonia is really septic emboli or endocarditis, clearly, that is a different story. If they have a chemo port that may be infected and needs to be replaced, different story. If there is concern over poor follow up, different story. If the patient looks well, they have a known source and are on an appropriate antibiotic, they do as well as the general population without an infection. In select cases of bacteremia, if patients appeared clinically well enough to go home, and are on appropriate antibiotics, they can likely be safely managed as outpatients. If they are reliable and have some sort of follow up to make sure they are doing well, they likely do not need to go an infusion center for “IV antibiotics,” let alone be admitted – since their mortality rate is quite frankly the same as everyone else that was discharged.