Background Elevated CRP is associated with many conditions other than infection. Positive predictive value for infection of random CRP determinations soon after birth in healthy term infants (HTI) is poor. However, a decision to continue antibiotics (ATB), perform LP, and/or delay discharge is sometimes based on elevated CRP, regardless of clinical course and blood culture results.
Objective To evaluate the use and duration of ATB in “HTI” when using or not using random CRP values as indicators of possible sepsis.
Methods HTI born at GMH receive a “rule out sepsis” workup based on prenatal risk assessment (maternal fever, rupture of membranes, chorioamnionitis, etc). Ampicillin and gentamicin are started pending results. We collected data prospectively in two periods to evaluate duration of ATB treatment, CRP values, number of LPs and hospital days in all HTI who received gentamicin. In period 1 (1/03- 8/03), ATB were to be continued > 48 hours if a CRP value was > 1.6 mg/dL at 2-24 hours, even with normal clinical course and negative cultures. If a repeat CRP at 36-48 hours was still elevated, ATB were continued in-hospital for ≥ 7 days and some infants underwent workup for meningitis at ≥ 48 hours. After extensive discussion of the literature, CRP measurements were discontinued from routine “rule out sepsis” workup. Period 2 is two months after implementing new guidelines, when ATB were discontinued based on culture results and clinical course. We compared utilization of resources (ATB use and duration, LPs, hospital stay, IVs), positive blood cultures, readmissions and mortality between the 2 periods.
Results In period 1, gentamicin was used in 192 HTI (7.3% of live births). Of these, 23 (12%) received ATB ≤ 2 days and 83 (43%) for ≤ 3 days (total ≤ 3 days: 105; 55%); 87 (45%) received ATB > 3 days based on CRP results; 3 were infected. After discontinuing CRP (period 2), 54 HTI (7.5% of all LB) received gentamicin; 50% of them received ATB ≤ 48 hours and 39% ≤ 3 days (total ≤ 3 days: 48; 89%; p < .05). Only 11% had ATB > 3 days (positive culture/clinical course); 1 was infected. Resource utilization was markedly less in period 2. No infant was readmitted or died of sepsis/meningitis in either group during periods 1 and 2.
Conclusion Elevated CRP values taken randomly early after birth are of no value for managing HTI with “rule out sepsis” and are associated with unnecessary longer course of ATB, add confusion to clinical management, and do not correlate with early neonatal sepsis. We speculate that the decrease in unnecessary resource utilization associated with not using a randomly obtained CRP in HTI evaluated for “rule out sepsis” may be associated with less risk for morbidity and with less cost.