Table 1.
Demographics of infants with late-onset sepsis episodes and characteristics of empiric antimicrobials. Categorical variables are reported as number and percentage. Continuous variables are reported as median and IQR.
| Variables | All Episodes (n = 83) | Non-Fatal Episodes (n = 66) |
Fatal Episodes (n = 17) |
p |
|---|---|---|---|---|
| Male sex | 42 (50.6) | 32 (48.5) | 10 (58.8) | 0.59 |
| Gestatational age, weeks | 26.0 (25.0–27.0) | 27.0 (25.0–28.0) | 25.0 (24.0–26.0) | 0.019 |
| Birth weight, g | 780.0 (654.0–993.8) | 800.0 (697.0–1050.0) | 690.0 (624.8–895.5) | 0.050 |
| Age at sepsis onset, days | 18.0 (10.3–32.8) | 22.0 (10.0–33.0) | 14.0 (10.5–21.8) | 0.281 |
| Body weight at sepsis onset, days | 1056.0 (766.8–1301.8) | 1100.0 (885.0–1342.0) | 760.0 (594.8–1078.3) | 0.003 |
| Gram-negative pathogen | 50 (60.2) | 36 (54.5) | 14 (82.4) | 0.051 |
| Time to first antibiotics, hours | 2 (0–6) | 3 (0.5–10) | 1 (0–2) | 0.035 |
| In vitro active empirical antimicrobials * | 59 (71.1) | 50 (75.8) | 9 (52.9) | 0.121 |
| Appropriate empirical antimicrobials § | 43 (51.8) | 39 (59.1) | 4 (23.5) | 0.019 |
| Appropriate empirical antimicrobials (cases with undetermined appropriateness excluded) ¶ | 43/70 (61.4) | 39/58 (67.0) | 4/12 (33.3) | 0.048 |
* Active empirical antimicrobials: the pathogen was susceptible (in vitro) to at least one empirical antimicrobial (or each pathogen was susceptible to at least one antimicrobial in the case of polymicrobial infections). Inactive empirical antimicrobials: the pathogen was resistant to all empirical antimicrobials. § Appropriate empirical antimicrobials: the empirical antimicrobial was active (in vitro) against the pathogen and (i) meningitis was ruled out (through a lumbar puncture) or (ii) the antimicrobial ideally penetrated at high levels the blood-brain barrier. Inappropriate empirical antimicrobials: (i) inactive empirical antimicrobials or (ii) active empirical antimicrobials ideally did not pass the blood-brain barrier at high levels (e.g., an aminoglycoside) and lumbar puncture findings were consistent with meningitis. Undetermined appropriateness of empirical antimicrobials: active empirical antimicrobials ideally did not pass the blood-brain barrier at high levels (e.g., an aminoglycoside) and meningitis was not ruled out (no lumbar puncture was performed). ¶ Percentages are calculated by excluding cases for whom empirical antimicrobial effectiveness was undetermined (non-fatal episodes n = 8, fatal episodes n = 5).