Skip to main content

Table 1 Summary of findings

From: Diagnostic value of biomarkers for paediatric urinary tract infections in primary care: systematic review and meta-analysis

Which biomarkers and predictions rules are useful for ruling out UTI in children?

Population acutely ill children < 18 years of age

Index test biomarkers or clinical prediction rules

Reference standard urine culture (cystitis), DMSA scan (pyelonephritis)

Target condition urinary tract infection (cystitis, pyelonephritis)

Setting primary care, outpatient

Positive likelihood ratio (LR+)

Negative likelihood ratio (LR-)

> 10 = very large increase in probability

0–0.1 = very large decrease in probability

5–10 = large increase in probability

0.1–0.2 = large decrease in probability

2–5 = moderate increase in probability

0.2–0.5 = moderate decrease in probability

1–2 = slight increase in probability

0.5–1 = slight decrease in probability

Outcome

Biomarkers

Number of patients

(Number of studies)

LR+ (95%CI)

LR- (95%CI)

AUC (95%CI)

Quality of the evidence (GRADE)

Comments

Cystitis (positive urine culture)

N

55,402 (26a)

39 (21–73)

0.61 (0.55–0.68)

0.79 (0.63–0.91)

⨁⨁⨁◯ due to indirectness

d

LE

65,204 (26b)

  

0.94 (0.80–0.98)

⨁⨁⨁◯ due to indirectness and risk of bias

d

 trace

 

8 (6–14)

0.13 (0.09–0.25)

   

 +

 

13 (9–19)

0.16 (0.11–0.31)

   

 ++

 

28 (15–114)

0.24 (0.14–0.76)

   

 +++

 

61 (26–994)

0.35 (0.15–0.96)

   

N or LE

25,238 (22a)

9 (6–13)

0.13 (0.10–0.18)

0.93 (0.91–0.96)

⨁⨁⨁◯ due to indirectness

d

N and LE

38,070 (10a)

115 (33–394)

0.65 (0.62–0.68)

0.51 (0.38–0.83)

⨁⨁⨁◯ due to indirectness

d

Protein ≥trace

665 (3a)

3 (2–3)

0.69 (0.56–0.83)

0.70 (0.47–0.91)

⨁◯◯◯ due to inconsistency, indirectness and imprecision

d

WBC ≥5/hpf (manual)

21,763 (18b)

6 (6–25)

0.27 (0.22–0.69)

0.90 (0.77–0.98)

⨁⨁⨁◯ due to indirectness

d

WBC ≥10/μL (automatic)

56,286 (9b)

4 (2–6)

0.13 (0.06–0.23)

0.91 (0.58–0.98)

⨁⨁⨁◯ due to indirectness

d

B ≥ 1/hpf

2979 (6b)

3 (2–8)

0.11 (0.06–0.24)

0.93 (0.25–1.00)

⨁◯◯◯ due to risk of bias, indirectness and imprecision

d

Gram stain ≥1/hpf

13,945 (12b)

20 (14–34)

0.10 (0.06–0.15)

0.98 (0.92–0.99)

⨁⨁⨁⨁

d

CRP ≥10 mg/L

13,729 (9b)

2 (1–2)

0.38 (0.24–0.77)

0.75 (0.62–0.83)

⨁⨁◯◯ due to inconsistency and indirectness

 

PCT ≥0.25 ng/mL

6585 (4b)

2 (2–8)

0.56 (0.24–0.69)

0.71 (0.62–0.80)

⨁⨁◯◯ due to risk of bias and indirectness

e

DUTY (+LE, N, Hb) ≥5 points

2277 (1c)

5 (5–6)

0.22 (0.13–0.37)

/

⨁⨁⨁◯ due to imprecision

 

UTIcalc (+LE, N)

229 (1c)

18 (10–33)

0.05 (0.01–0.26)

/

⨁◯◯◯ due to indirectness and imprecision

f

Pyelonephritis (positive DMSA scan)

WBCc 12 to 20 × 103/μL

209 (2c)

1.62 to 3.57

0.35 to 0.83

/

⨁◯◯◯ due to serious indirectness and imprecision

g

CRP ≥20 mg/L

7179 (5c)

1.32 to 2.74

0.10 to 0.37

/

⨁⨁◯◯ due to imprecision and indirectness

 

PCT ≥2 ng/mL

436 (5c)

2.93 to 14.22

0.32 to 0.74

/

⨁◯◯◯ due to inconsistency, indirectness and imprecision

g

  1. GRADE Working group grades of evidence
  2. High quality: further research is very unlikely to change our confidence in the estimate of effect. ⨁⨁⨁⨁
  3. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. ⨁⨁⨁◯
  4. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. ⨁⨁◯◯
  5. Very low quality: We are very uncertain about the estimate. ⨁◯◯◯
  6. DMSA Dimercaptosuccinic acid scan, 95%CI 95% confidence intervals, AUC Area Under the Receiver Operating Characteristic (ROC) Curve Analysis, Assessment, Development and Evaluations, N Nitrite (urine), LE Leukocyte esterase (urine), Hb Hemoglobin (urine), WBC White blood cell (urine), B Unstained bacteria (urine), CRP C-reactive protein (blood), PCT Procalcitonin (blood), DUTY Diagnosis of Urinary Tract Infections in Young children, UTIcalc UTI calculator, hpf High-power field, μL Microliter, mg/L Milligram per liter, ng/mL Nanogram per milliliter, mm3 Cubic milliliter
  7. aBivariate random effects model
  8. bHSROC model
  9. cDescriptive statistics
  10. dThe majority of studies included patients with suspicion of UTI or with UTI features, and therefore the results might not be applicable for patients without suspicion of UTI or UTI features
  11. eMajority of children had fever without a source for infection
  12. fDerived in setting where there was high circumcision rate in boys
  13. gIn children with high pre-test probability of UTI
Back to article page

BMC Primary Care

ISSN: 2731-4553

Contact us