Table 2.
Comparison of studies evaluating the characteristics of newer diagnostic tools for the detection of serious bacterial infection in febrile infants ≤90 days of age
| Test | Study design | Participants ≤90 days old (n) | Prevalence of SBI (%) | Cut off value | Sensitivity (95% CI) | Specificity (95% CI) | LR+ (95% CI) | LR− (95% CI) |
|---|---|---|---|---|---|---|---|---|
| CRP | ||||||||
| Olaciregui29 | Retrospective | 347 | 24% | 30mg/L | 59% (48,70) | 89% (85,93) | 5.4 (NR) | 0.46 (NR) |
| Stein30 | Prospective | 112 | 17% | 30mg/L | 45% (NR) | 82% (NR) | NR | NR |
| Gomez31 | Retrospective | 1112 | 26% | 20mg/L | NR | NR | 2.2 (1.2,4.0) | 0.85 (0.70,1.04) |
| Bilavsky32 | Prospective | 892 | 11% | 20mg/L | 56% (46,65) | 82% (79,85) | 3.1 (2.5,3.9) | 0.5 (0.4,0.7) |
| Milcent33 | Prospective | 2047 | 7% | 20mg/L | 77% (66,86) | 75% (72,77) | 3.1 (2.6,3.6) | 0.3 (0.2,0.5) |
| PCT | ||||||||
| Maniaci38 | Prospective | 234 | 13% | 0.13ng/mL | 97% (81,100) | 30% (24,38) | NR | 0.11 (0.02,0.76) |
| Stein30 | Prospective | 112 | 17% | 0.2ng/mL | 55% (NR) | 75% (NR) | NR | NR |
| Woelker39 | Prospective | 1551 | 8% | 0.26ng/mL | 92% (62,100) | 64% (55,72) | NR | 0.01 (0.001, 0.09) |
| Milcent33 | Prospective | 2047 | 7% | 0.3ng/mL | 74% (62,84) | 78% (75,80) | 3.3 (2.8,3.9) | 0.3 (0.2,0.5) |
| Olaciregui29 | Retrospective | 347 | 24% | 0.5ng/mL | 63% (52,74) | 87% (83,91) | 4.8 (NR) | 0.43 (NR) |
| Gomez31 | Retrospective | 1112 | 26% | 0.5ng/mL | NR | NR | 3.9 (2.1,7.3) | 0.80 (0.66,0.97) |
| Lab score | ||||||||
| Bressan41 | Retrospective | 1012 | 28% | ≥3 | 52% (46,58) | 95% (93,96) | 10.2 (9.5,10.9) | 0.5 (0.5,0.5) |
| Step-by-Step | ||||||||
| Gomez42 | Prospective | 2185 | 23% | Positive screen | 98% (96,99) | 58% (56,61) | NR | NR |
1
Included infants 2–60 days of age
Abbreviations: SBI, serious bacterial infection; CI, confidence interval; LR, likelihood ratio; CRP, C-reactive protein; PCT, procalcitonin; NR, not reported