Background: Pyelonephritis is the most common and important infection among Iranian
pediatric population. Differentiation between upper and lower Urinary Tract Infection (UTI) is often
difficult based on clinical data. Therefore, definite diagnosis is helpful for choosing appropriate
antibiotic and decision for hospital admission. The main purpose of this study was todetermine the
diagnostic value of serum STREM-1 level in children suspicious to UTI and differentiation of upper
UTI and lower UTI.
Material & Methods: This prospective cross sectional study (2010-2011) was performed to evaluate
and compare the serum level of STREM- 1 (pg. /ml) in 36 diagnosed UTI patients (24 upper
and 12 lower UTI) with 25 normal children (without UTI) in Rasoul Akram hospital, Tehran, Iran.
The mean age of studied children was 3.64 years; 24 male and 37 female. Urinary analysis and
urine culture were performed for all UTI cases and only the positive cultured cases with the same
microorganism were enrolled in the study. Distinguishing the upper from lower UTI was done on
the basis of clinical manifestation and laboratory tests and confirmed by Imaging studies (ultra
sonography /or DMSA scan). Blood sampling was taken from all children and centrifuged .The
level of STREM-1 (pg /ml) in all sera was determined by Enzyme immunoassay technique (Human
TREM-1 immunoassay Sandwich test, Quantikine, R&D systems, Minneapolis; USA).
Cut-off levels for STREM-1 were illustrated by ROC curve. The p<0.05 was considered as significant
for differences between groups.
Findings: The mean of STREM -1level had significant difference between overall cases of UTI
(427.72pg/ml) and controls (124.24 pg. /ml; P =0.000) ; with cutoff point 111.5 pg./ml ; it had
83.3% sensitivity; and 60 % specificity to distinguish UTI from control. Serum STREM -1 level
had no significantly difference between the upper and lower UTI (500pg/ml vs. 283 pg. /ml, P
value=0.1) with cutoff point 132 pg./ml it had 83.3% sensitivity ; and 60 % specificity to distinguish
upper UTI from lower UTI. Conclusion: Our study demonstrates that even low amount of serum
STREM-1 (111.5 pg./ml) has 83.3% sensitivity ; and 60 % specificity to distinguish the UTI
from normal cases (P value =0.000) but higher level (132 pg./ml) was needed for definite diagnosis
(83.3% sensitivity; 60 % specificity) of upper and lower UTI. It is concluded that serum STREM-1
level test is a valuable tool for early diagnosis of the normal cases with false positive urine culture,
or in highly suspicious upper UTI cases with false negative urine culture. Indeed higher titer of this
biomarker could be helpful for discriminating the upper from lower UTI. Therefore adding this new
biologic marker (STREM-1) to previous ones (CRP, PCT) is suggested to prevent the unnecessary
hospital admission and empiric antibiotic therapy.