- Email: [email protected]
The Role of Procalcitonin for Acute Pyelonephritis and Subsequent Renal Scarring in Infants and Young Children
Pediatric Urology
The Role of Procalcitonin for Acute Pyelonephritis and Subsequent Renal Scarring in Infants and Young Children Ji-Nan Sheu,* Hung-Ming Chang, Shan-Ming Chen, Tung-Wei Hung and Ko-Huang Lue From the Departments of Pediatrics and Internal Medicine, Chung Shan Medical University Hospital, and Departments of Pediatrics and Anatomy, School of Medicine, and Institute of Medicine, Chung Shan Medical University, Taichung, Taiwan
Abbreviations and Acronyms APN ⫽ acute pyelonephritis CRP ⫽ C-reactive protein DMSA ⫽ 99mTcdimercaptosuccinic acid NPV ⫽ negative predictive value PCT ⫽ procalcitonin PPV ⫽ positive predictive value RS ⫽ renal scarring UTI ⫽ urinary tract infection VCUG ⫽ voiding cystourethrography VUR ⫽ vesicoureteral reflux WBC ⫽ white blood cell count Submitted for publication February 27, 2011. Supported by grants from the National Science Council, Taiwan (NSC93-2314-B-040-012) and Chung Shan Medical University Hospital (CSH-2010-C-009). Study received institutional review board approval. * Correspondence: Department of Pediatrics, Chung Shan Medical University Hospital, No. 110, Section 1, Jianguo North Rd., Taichung 402, Taiwan (telephone: ⫹886 4 2473 9595 ext. 34816; FAX: ⫹886 4 2471 0934; e-mail: [email protected]).
See Editorial on page 1760.
2002
www.jurology.com
Purpose: We assessed the usefulness of procalcitonin as a biological marker in diagnosing acute pyelonephritis and for predicting subsequent renal scarring in young children with a first febrile urinary tract infection. Materials and Methods: Children 2 years old or younger with a first febrile urinary tract infection were prospectively studied. Renal parenchymal involvement was assessed by 99mTc-dimercaptosuccinic acid scan within 5 days of admission and after 6 months. Serum samples from all patients were tested for procalcitonin, C-reactive protein and white blood cell count measurements. Results: The 112 enrolled patients (age range 24 days to 24 months old) were divided into acute pyelonephritis (76) and lower urinary tract infection (36) groups according to the results of 99mTc-dimercaptosuccinic acid scans. Median values of procalcitonin, C-reactive protein and white blood cell count at hospitalization were significantly higher in patients with acute pyelonephritis than in those with lower urinary tract infection. The area under receiver operating characteristic curves showed that procalcitonin was superior to C-reactive protein and white blood cell count as a marker for diagnosing acute pyelonephritis. Initial and post-antibiotic treatment procalcitonin values were significantly higher in children with renal scarring than in those without scarring (p ⬍0.001). Procalcitonin values at hospitalization and after treatment were independent predictors of later renal scarring on logistic regression analysis. Conclusions: Our results indicate the superior diagnostic accuracy of procalcitonin for predicting acute pyelonephritis in children 2 years old or younger. Higher initial and posttreatment procalcitonin values are independent risk factors for later renal scarring. Key Words: pyelonephritis, procalcitonin, infant, vesico-ureteral reflux URINARY tract infection is a common bacterial infection in pediatric patients, especially in children younger than 2 years old, with a prevalence of 6.5% and 3.3% in girls and boys younger than 1 year old, respectively.1 The nonspecific symptoms in febrile young children make the clinical differentiation between acute pyelonephritis and lower UTI difficult. Approximately 30%
to 60% of cases in small children result in renal scarring after acute febrile UTI.2– 4 A delay in the diagnosis and treatment of APN in young children can increase the risk of kidney damage,1,5 which may cause hypertension, proteinuria and end stage renal disease later in life.2,6 99m Tc-dimercaptosuccinic acid is widely accepted as the gold standard
0022-5347/11/1865-2002/0 THE JOURNAL OF UROLOGY® © 2011 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 186, 2002-2008, November 2011 Printed in U.S.A. DOI:10.1016/j.juro.2011.07.025
AND
RESEARCH, INC.
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
for detecting APN, and assessing the extent and progression of renal damage.7,8 However, factors like availability, as well as the costs and risks of exposing patients to radiation and sedation limit the widespread use of the DMSA scan. Furthermore, commonly used clinical and laboratory parameters do not accurately localize the UTI site.8,9 Therefore, finding an accurate and readily available diagnostic test to predict APN is of great value in young children with a febrile UTI. Procalcitonin is a 116 amino acid peptide and a precursor of calcitonin.10 Its release during infection is induced directly by microbial toxins and/or indirectly by humoral factors or the cell mediated host response.11,12 Most recent studies propose serum PCT as a better marker than C-reactive protein and white blood cell count in the early detection of APN in children.13–21 In contrast, 2 studies show that the PCT test is not a sensitive marker for the early diagnosis of APN.22,23 However, few studies have demonstrated a relationship between PCT and febrile UTI in children 2 years old or younger.14,19 Furthermore, the association between PCT and vesicoureteral reflux remains controversial.14,15,18,19,21,24 Therefore, we evaluated the role of serum PCT as a reliable marker for detecting APN in children 2 years old or younger with a first febrile UTI, and assessed its ability to predict the risk of subsequent RS. We also examined the association of PCT with VUR.
PATIENTS AND METHODS Study Population and Inclusion Criteria We prospectively studied children 2 years old or younger who were hospitalized with a febrile UTI in a 3-year period. The diagnosis of a first febrile UTI was based on fever (axillary temperature 38C or greater), leukocyturia (defined as 5 or more WBC per high power field), positive urine culture (defined as growth of a single microorganism 105 colony-forming units per ml or more collected from the midstream clean void urine specimen for toilet trained young children, or 104 colony-forming units per ml or more collected from a transurethral catheterized specimen), no history of UTI and kidney or bladder disease and no other coincidental infections. All patients were treated empirically with combined intravenous cefazolin (100 mg/kg daily) and gentamicin (7.5 mg/kg daily) for at least 3 days, which was later adjusted according to the results of bacterial susceptibility tests for a treatment duration of 7 to 14 days. The institutional review board of Medical University Hospital approved the study protocol and all parents of the participants provided informed consent.
Imaging Studies All patients underwent renal ultrasound for the detection of urinary tract anomalies within the first 3 days of hospital admission. DMSA scans were performed within the first 5 days of hospitalization to verify the presence of
2003
renal parenchymal lesions. APN was defined as the presence of focal or diffuse areas of decreased uptake without evidence of cortical loss.7,8 To assess the severity of acute renal involvement on PCT values, a scoring system was used based on DMSA grading and modified by Benador et al.21 Lesions were graded in 4 groups of 0 —no lesion (lower UTI), 1—mild lesions (defect covering less than 10% of surface area), 2—moderate lesions (defect covering 10% to 30%) and 3—severe lesions (defect covering more than 30%). If the initial DMSA result was abnormal, a followup examination was performed at least 6 months later to evaluate the presence of RS. The diagnosis of APN was confirmed only in children with totally or partially reversible lesions on the followup scans. The presence of renal lesions was determined by 2 nuclear medicine physicians who were blinded to the study. VCUG was performed 1 to 2 weeks after completing treatment for the infection. VUR was graded 0 to V according to the International Reflux Study.25
Laboratory Measurements Serum and urine indexes for laboratory investigations, including peripheral WBC and differential, CRP values, urinalysis, urine and blood cultures were measured in all patients at hospitalization and before the initiation of antibiotic treatment. Serum samples were also taken for PCT measurements at hospitalization and repeated 3 days later. A rapid and quantitative measurement of PCT was performed using enzyme-linked fluorescent assay in an automated VIDAS® instrument (VIDAS BRAHMS PCT, BRAHMS Diagnostica, Berlin, Germany). The detection limit was 0.05 ng/ml and PCT 0.5 ng/ml or greater was considered abnormal.
Statistical Analyses All statistical analyses were performed using SPSS® for Windows (version 15.0). Nonparametric data were assessed by the Mann-Whitney U test or Kruskal-Wallis 1-way ANOVA, and expressed as medians and interquartile ranges (Q1-Q3). Logistic regression analysis was used to evaluate the impact of potential risk on the development of RS in patients. The receiver operating characteristic curve analysis was performed to assess quantitative variables at hospitalization for diagnosing APN and later RS. The diagnostic values of each cutoff point, including sensitivity, specificity, PPV, NPV and the likelihood ratio for a positive result, were all calculated with p ⬍0.05 considered statistically significant.
RESULTS Patient and Clinical Characteristics There were 66 boys and 46 girls 24 days to 24 months old (median age 5.0 months), with a diagnosis of a first febrile UTI (table 1). Escherichia coli was isolated as a single pathogen in the urine culture of 100 (89.3%) patients. The number of urine collections by catheterized vs clean void specimens was 103 vs 9. Renal ultrasound revealed findings of normal (89), mild to moderate hydronephrosis (renal pelvic
2004
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
Table 1. Clinical characteristics and laboratory data of patients with APN and lower UTI APN Median mos age (IQR) No. male/female No./total no. VUR grade (%): III or Greater I–II No./total no. E. coli (%) Median ng/ml PCT (IQR) Median mg/dl CRP (IQR) Median WBC/mm3 (IQR)
Lower UTI
5.1 (3.0–8.0) 40/36 30/70 (42.9) 22/30 (73.3) 8/30 (26.7) 65/76 (85.5) 2.95 (1.37–8.37) 9.61 (5.39–14.60) 16,125 (12,278–19,673)
p Value
4.0 (2.0–10.9) 26/10 6/31 (19.4) 2/6 (33.3) 4/6 (66.70) 35/36 (97.2) 0.35 (0.25–0.44) 3.09 (1.49–5.68) 13,595 (9,460–16,368)
0.312* 0.078† 0.040† 0.150† 0.150† 0.769† ⬍0.001* ⬍0.001* 0.013*
* Mann-Whitney U test for continuous variables. † Chi-square or Fisher’s exact test for categorical variables.
diameter 5 mm or greater in 20), duplex kidney (2) and unilateral renal agenesis (1). Of the 101 (90.2%) patients who underwent VCUG studies (70 with APN and 31 with lower UTI) 36 (35.6%) had VUR with grade II (12), III (11), IV (9) and V (4) (maximum degree of reflux given if bilateral). Laboratory Markers for APN Diagnosis at Hospitalization PCT, CRP and WBC values of the severity of acute renal involvement are shown in table 2. Median PCT tended to increase significantly with increasing DMSA score (p ⬍0.001, fig. 1). Median PCT at hospitalization was significantly higher in patients with APN than in those with lower UTI (p ⬍0.001, table 1). Median PCT values initially and after 3 days of antibiotic treatment were 2.95 ng/ml (IQR 1.37– 8.37) vs 0.38 ng/ml (IQR 0.08 –1.10) (p ⬍0.001). To differentiate between APN and lower UTI at hospitalization, the ROC curves were used by plotting the sensitivity vs 1-specificity for different cutoff values of PCT, CRP and WBC. The area under the curve of the ROC was 0.910 for PCT (p ⬍0.001), 0.770 for CRP (p ⫽ 0.015) and 0.594 for WBC (p ⫽ 0.529). The ROC analysis demonstrated that PCT values had a greater AUC than CRP and WBC values for differentiating APN from lower UTI (fig. 2). The cutoff point for maximum diagnostic efficiency of PCT was 1.0 ng/ml, with 81.6% sensitivity and 91.7% specificity (table 3). Combined analysis of PCT and CRP showed that a concomitant PCT value of 1 ng/ml or greater and CRP 3.5 mg/dl or greater had a sensitivity of 77.6%, specificity 91.7%, PPV
95.2% and NPV 66%. However, this diagnostic efficiency was not superior to PCT alone for predicting APN. Laboratory Markers and RS Followup DMSA scan was performed 6 months after the infection in all 76 cases. Four patients who experienced a second UTI between the 2 DMSA scans were excluded from analysis. Renal scars were seen in 34 (47.2%) of the remaining 72 patients whereas 38 (52.8%) exhibited total regression of the initial lesions (table 4). Of the laboratory markers at hospitalization only median PCT was significantly higher in patients with RS (p ⬍0.001). All patients with RS had a PCT value greater than 2 ng/ml. The effect of age on PCT values was analyzed and no correlation was found (r ⫽ 0.060, p ⫽ 0.608). Of the 34 patients with RS 4 of 25 (16.0%), 13 of 26 (50.0%) and 17 of 21 (81.0%) had scores of 1, 2 and 3, respectively, with an overall median PCT of 8.57 ng/ml (IQR 5.37–15.24). The incidence of RS tended to increase significantly as the initial DMSA score increased (p ⬍0.001). In the 38 patients with normal followup DMSA results the overall median PCT was 1.45 ng/ml (IQR 0.57–2.27). The initial median PCT was significantly higher in patients with RS (p ⬍0.001, fig. 3). Moreover median posttreatment PCT in patients with RS was 1.13 ng/ml (IQR 0.84 – 1.99) vs 0.08 ng/ml (IQR 0.06 – 0.21) (p ⬍0.001). VUR was found in 11 (32.4%) of 34 patients with RS (table 4). No association was noted between RS and VUR grade. There was no significant difference between PCT in patients with and in those without VUR (p ⫽
Table 2. PCT, CRP and WBC values indicating acute renal involvement severity
No. pts Median ng/ml PCT (IQR) Median mg/dl CRP (IQR)† Median WBC/mm3 (IQR)†
DMSA Score 0
DMSA Score 1
DMSA Score 2
DMSA Score 3
p Value*
36 0.35 (0.25–0.44) 3.09 (1.49–5.68) 13,595 (9,460–16,368)
27 1.35 (0.48–2.15) 10.10 (7.55–13.60) 18,250 (11,300–20,640)
28 3.45 (1.87–8.99) 8.08 (4.45–12.01) 16,530 (12,308–21,158)
21 8.57 (4.79–15.46) 14.50 (4.85–18.50) 14,750 (2,750–16,640)
⬍0.001 ⬍0.001 0.034
* Kruskal-Wallis 1-way ANOVA for continuous variables. † There were no significant differences among DMSA scores 1, 2 and 3 in CRP (p ⫽ 0.082) and WBC (p ⫽ 0.245).
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
2005
Table 3. Sensitivity, specificity, PPV, NPV and likelihood ratio for a positive result of PCT, CRP and WBC for the diagnosis of APN
Figure 1. PCT values of patients with APN at hospitalization in terms of acute renal lesion severity by DMSA score. Horizontal line indicates median and box indicates quartiles (Kruskal-Wallis test p ⬍0.001).
0.409). There was also no significant correlation of PCT and VUR grade (p ⫽ 0.166). Variables Predicting APN and Late RS The ROC analyses demonstrated that initial and posttreatment PCT values were useful indexes for predicting later RS (AUC 0.942, p ⬍0.001 and AUC 0.963, p ⬍0.001, respectively). The optimal cutoff value of 3.5 ng/ml at hospitalization and 1.0 ng/ml for posttreatment had maximum diagnostic effi-
Figure 2. Comparison of ROC curves for PCT, CRP and WBC values to distinguish APN from lower UTI. PCT (AUC 0.910, p ⬍0.001) was determined to be best index compared to CRP (AUC 0.770, p ⫽ 0.015) and WBC (AUC 0.594, p ⫽ 0.529).
PCT (ng/ml): 0.5 or Greater 1 or Greater* 1.5 or Greater 2 or Greater CRP (mg/dl): 2 or Greater 3.5 or Greater 6 or Greater 10 or Greater WBC (/mm3): 12,000 or Greater 15,000 or Greater 20,000 or Greater
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
Pos Likelihood Ratio
89.5 81.6 72.4 64.5
83.3 91.7 94.4 97.2
91.9 95.4 96.5 98.0
78.9 70.2 61.8 56.5
5.4 9.8 12.9 23.0
100 90.8 73.7 47.4
38.9 58.3 80.6 94.4
77.6 82.1 88.9 94.7
100 75.0 59.2 45.9
1.6 2.2 3.8 8.5
78.9 57.9 21.1
38.9 66.7 88.9
73.2 78.6 80.0
46.7 42.9 34.8
1.3 1.7 1.9
* Maximum diagnostic efficiency.
ciency for RS (table 5). Multivariate analyses revealed that the independent predictors of APN were initial PCT and CRP values as well as the presence of VUR (table 6). Initial and posttreatment PCT values were independent predictors of later RS.
DISCUSSION This study focused on young children 2 years old or younger because this age population has the most frequent cases of febrile UTI. To date to our knowledge, our study enrolled the largest case number of a pediatric population age 2 years or younger to explore the diagnostic value of PCT for predicting APN in children with a febrile UTI. Our results revealed that PCT values only were influenced by the severity of renal lesions and were not subject to an age related effect in this age population. This study shows that initial PCT is a reliable marker in distinguishing APN from lower UTI in young children, consistent with the published data.13–21 Previous studies have shown that PCT was increased in children with APN, with a sensitivity of 70% to 100% and a specificity of 51% to 97%.13–21 Our data show that using a PCT cutoff value of 1 ng/ml has the best performance, with a high sensitivity (81.6%) and specificity (91.7%). The AUCs of ROC show that PCT has a much greater area than those of CRP and WBC for predicting APN. In this study 81.6% of patients with APN had PCT values greater than 1 ng/ml compared to only 8.3% of those with a lower UTI. Moreover all patients with RS presented with an initial PCT greater than 1 ng/ml. Our findings suggest that when PCT values at hospitalization are less than 1 ng/ml, the possibility of renal involvement is low in young chil-
2006
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
Table 4. Comparisons of clinical characteristics and laboratory markers at hospitalization in patients with and those without RS DMSA Scan
No. male/female Median mos age (IQR) Median ng/ml PCT (IQR) Median mg/dl CRP (IQR) Median WBC/mm3 (IQR) No. VUR/no VUR No. VUR grade I–II/III–V
RS⫹
RS⫺
p Value
21/13 5.6 (3.9–8.0) 8.57 (5.37–15.24) 10.83 (5.05–16.85) 15,190 (12,720–20,868) 11/23 1/10
18/20 5.8 (2.9–10.3) 1.45 (0.57–2.27) 9.52 (5.44–13.43) 16,240 (11,460–19,045) 17/21 3/14
0.324* 0.973† ⬍0.001† 0.417† 0.446† 0.404* 1.000*
* Chi-square or Fisher’s exact test for categorical variables. † Mann-Whitney U test for continuous variables.
dren with febrile UTI, and DMSA studies during the acute phase or after 6 months can be omitted for the selection of cases. The results of followup DMSA scans show that RS developed in 47.2% of patients after APN. These 34 cases of RS were characterized by higher initial PCT values compared to other cases of APN. The results also demonstrate significant associations of increased PCT at hospitalization with severity of acute renal involvement and later RS, consistent with few studies on the correlation between PCT and RS in younger children14 and children.15,17,18,21 Moreover our results also demonstrate a positive association of posttreatment PCT values with RS. This finding has not yet been reported in other studies. Our results indicate that using a PCT cutoff greater than 3.5 ng/ml at hospitalization and greater than 1.0 ng/ml after treatment can provide 17.8 and 12.8 times, respectively, higher likelihood of finding patients with RS. Therefore, higher initial and posttreatment PCT values are associated with a higher risk of RS.
The association of PCT with VUR in children with febrile UTI remains controversial. This study revealed no significant differences in PCT values in children with and in those without VUR and VUR grade, similar to the published data.15,18,19,21 In contrast, Bressan et al found significantly higher PCT values in children with VUR, but revealed no correlation between PCT levels and VUR grade.14 However, only 13 patients with VUR were enrolled in their study and the researchers declared that this small number of patients prevented drawing definite conclusions on the subject. Although VUR may predispose to APN, to our knowledge no association has been established between the presence of VUR and persistent renal damage in children with febrile UTI.2,26 Previous studies indicate that increased PCT concentrations during the early phase of the disease are closely related to the onset of bacterial infection and correlated to disease severity.27,28 Our findings support this suggestion. This area needs further investigation in a large number of children to clarify the role of PCT as a predictor of VUR. This study has some limitations. Most patients with a febrile UTI were treated as outpatients. Therefore, a selection population bias might have Table 5. Indices of diagnostic quality and efficiency of PCT for predicting subsequent RS
Figure 3. Median PCT at hospitalization in patients with lower UTI, APN, and APN with and without RS. Horizontal line indicates median and box indicates quartiles (Kruskal-Wallis test p ⬍0.001).
Cutoff (ng/ml) at hospitalization: 1.0 or Greater 2.0 or Greater 3.5 or Greater* 5.0 or Greater 6.0 or Greater Cutoff (ng/ml) after treatment: 1.0 or Greater* 1.5 or Greater 2.0 or Greater
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
Pos Likelihood Ratio
100 97.1 94.1 76.5 70.6
34.2 65.8 94.7 94.7 97.4
57.6 71.7 94.7 92.9 96.0
100 96.2 94.1 81.8 78.7
1.5 2.9 17.8 14.4 27.2
67.6 44.1 23.5
94.7 97.4 97.4
93.8 92.0 88.9
66.1 76.6 58.7
12.8 17.0 9.0
* Maximum diagnostic efficiency.
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
2007
Table 6. Multivariate logistic regression models predicting APN and later RS in patients after febrile UTI
Age Gender PCT CRP WBC Posttreatment PCT Reflux grade: 0 I–II III–V DMSA lesion score: 1 2 3
OR (95% CI) APN
p Value
0.898 (0.750–1.076) 3.127 (0.521–18.777) 12.189 (2.310–64.300) 1.434 (1.103–1.864) 1.000 (1.000–1.000) —
0.243 0.213 0.003 0.007 0.075
1 [Reference] 22.857 (1.379–378.967) 43.818 (2.179–881.245) — — — —
0.029 0.014
OR (95% CI) RS 1.085 0.261 3.133 0.903 1.000 2.451
p Value
(0.851–1.583) (0.024–2.790) (1.549–6.336) (0.769–1.060) (1.000–1.000) (1.368–4.389)
0.510 0.267 0.001 0.211 0.230 0.003
1 [Reference] 0.051 (0.000–27.294) 0.354 (0.045–2.780)
0.353 0.324
1 [Reference] 0.568 (0.050–6.445) 0.897 (0.067–11.99)
0.648 0.935
For multivariate analysis a multivariate logistic regression model was used.
been present. In addition, VCUG examination was not performed in nearly 10% of our cohort, which might have affected our results in terms of the analysis of PCT in patients with VUR. Another limitation is that patients with UTI without fever or with a temperature less than 38C and the absence of leukocyturia were excluded from study. Thus, some diagnostic bias in the study might have affected the data. How PCT can replace DMSA is yet to be determined. The price of DMSA examination is approximately $150, which is 5 times that of PCT measurement (around $30). From a cost-effective point of view of diagnostic tests a PCT based diagnostic strategy vs DMSA based study can reduce costs by 80% per test. From a patient safety standpoint, the most important point, a PCT based strategy can avoid the risks of radiation and sedation exposure in small children, risks which are associated with DMSA examinations. Our results suggest that chil-
dren 2 years old or younger with a low PCT value (less than 1 ng/ml) at hospitalization may not require DMSA study after a first UTI. For pediatricians, knowing the limitations of DMSA, these findings are of great clinical importance because they may reserve DMSA examinations for those with a PCT value greater than 1 ng/ml.
CONCLUSIONS Our study indicates the superior diagnostic accuracy of PCT in APN in children 2 years old or younger. Higher initial and posttreatment PCT values indicate a high risk of later RS. These results may lead to more personalized diagnostic and therapeutic strategies for children 2 years old or younger with a first febrile UTI and may contribute to the avoidance of further unnecessary DMSA studies in children with low PCT values.
REFERENCES 1. American Academy of Pediatrics, Committee on Quality Improvement, Subcommittee on Urinary Tract Infection: Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and younger children. Pediatrics 1999; 103: 843. 2. Stokland E, Hellström M, Jacobsson B et al: Early 99mTc dimercaptosuccinic acid (DMSA) scintigraphy in symptomatic first-time urinary tract infection. Acta Paediatr 1996; 85: 430.
5. Hiraoka M, Hashimoto G, Tsuchida S et al: Early treatment of urinary infection prevents renal damage on cortical scintigraphy. Pediatr Nephrol 2003; 18: 115.
9. Garin EH, Olavarria F, Araya C et al: Diagnostic significance of clinical and laboratory findings to localize site of urinary infection. Pediatr Nephrol 2007; 22: 1002.
6. Jacobson SH, Eklöf O, Lins LE et al: Long-term prognosis of post-infectious renal scarring in relation to radiological findings in childhood–a 27year follow-up. Pediatr Nephrol 1992; 6: 19.
10. Russwurm S, Wiederhold M, Oberhoffer M et al: Molecular aspects and natural source of procalcitonin. Clin Chem Lab Med 1999; 37: 789.
3. Benador D, Benador N, Slosman D et al: Are younger children at highest risk of renal sequelae after pyelonephritis? Lancet 1997; 349: 17.
7. Rushton HG: The evaluation of acute pyelonephritis and renal scarring with technetium 99m-dimercaptosuccinic acid renal scaintigraphy: evolving concepts and future directions. Pediatr Nephrol 1997; 11: 108.
4. Sheu JN, Chen SM, Meng MH et al: The role of serum and urine interleukin-8 on acute pyelonephritis and subsequent renal scarring in children. Pediatr Infect Dis J 2009; 28: 885.
8. Biggi A, Dardanelli L, Pomero G et al: Acute renal cortical scintigraphy in children with a first urinary tract infection. Pediatr Nephrol 2001; 16: 733.
11. Müller B, White JC, Nylén ES et al: Ubiquitous expression of the calcitonin-I gene in multiple tissues in response to sepsis. J Clin Endocrinol Metab 2001; 86: 396. 12. Dandona P, Nix D, Wilson MF et al: Procalcitonin increase after endotoxin injection in normal subjects. J Clin Endocrinol Metab 1994; 79: 1605. 13. Nikfar R, Khotaee G, Ataee N et al: Usefulness of procalcitonin rapid test for the diagnosis of acute
2008
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
pyelonephritis in children in the emergency department. Pediatr Int 2010; 52: 196.
scarring in children with urinary tract infection. Pediatr Infect Dis J 2003; 22: 438.
14. Bressan S, Andreola B, Zucchetta P et al: Procalcitonin as a predictor of renal scarring in infants and young children. Pediatr Nephrol 2009; 24: 1199.
19. Smolkin V, Koren A, Raz R et al: Procalcitonin as a marker of acute pyelonephritis in infants and children. Pediatr Nephrol 2002; 17: 409.
15. Kotoula A, Gardikis S, Tsalkidis A et al: Comparative efficacies of procalcitonin and conventional inflammatory markers for prediction of renal parenchymal inflammation in pediatric first urinary tract infection. Urology 2009; 73: 782. 16. Gürgöze MK, Akarsu S, Yimaz E et al: Proinflammatory cytokines and procalcitonin in children with acute pyelonephritis. Pediatr Nephrol 2005; 20: 1445. 17. Pecile P, Miorin E, Romanello C el al: Procalcitonin: a marker of severity of acute pyelonephritis among children. Pediatrics 2004; 114: e249. 18. Prat C, Domínguez J, Rodrigo C et al: Elevated serum procalcitonin values correlate with renal
20. Gervaix A, Galetto-Lacour A, Gueron T et al: Usefulness of procalcitonin and C-reactive protein rapid tests for the management of children with urinary tract infection. Pediatr Infect Dis J 2001; 20: 507. 21. Benador N, Siegrist CN, Gendrel D et al: Procalcitonin is a marker of severity of renal lesions in pyelonephritis. Pediatrics 1998; 102: 1422.
24. Leroy S, Adamsbaum C, Marc E et al: Procalcitonin as a predictor of vesicoureteral reflux in children with a first febrile urinary tract infection. Pediatrics 2005; 115: e706. 25. Lebowitz RL, Olbing H, Parkkulainen KV et al: International system of radiographic grading of vesicoureteric reflux. Pediatr Radiol 1985; 15: 105. 26. Ditchfield MR, Grimwood K, Cook DJ et al: Persistent renal cortical scintigram defects in children 2 years after urinary tract infection. Pediatr Radiol 2004; 34: 465.
22. Güven AG, Kazdal HZ, Koyun M et al: Accurate diagnosis of acute pyelonephritis: how helpful is procalcitonin? Nucl Med Commun 2006; 27: 715.
27. van Rossum AM, Wulkan RW and OudesluysMurphy AM: Procalcitonin as an early marker of infection in neonates and children. Lancet Infect Dis 2004; 4: 620.
23. Tuerlinckx D, Vander Borght T, Glupczynski Y et al: Is procalcitonin a good marker of renal lesion in febrile urinary tract infection? Eur J Pediatr 2005; 164: 651.
28. Müller B, Becker KL, Schächinger H et al: Calcitonin precursors are reliable markers of sepsis in a medical intensive care unit. Crit Care Med 2000; 28: 977.
The Role of Procalcitonin for Acute Pyelonephritis and Subsequent Renal Scarring in Infants and Young Children Ji-Nan Sheu,* Hung-Ming Chang, Shan-Ming Chen, Tung-Wei Hung and Ko-Huang Lue From the Departments of Pediatrics and Internal Medicine, Chung Shan Medical University Hospital, and Departments of Pediatrics and Anatomy, School of Medicine, and Institute of Medicine, Chung Shan Medical University, Taichung, Taiwan
Abbreviations and Acronyms APN ⫽ acute pyelonephritis CRP ⫽ C-reactive protein DMSA ⫽ 99mTcdimercaptosuccinic acid NPV ⫽ negative predictive value PCT ⫽ procalcitonin PPV ⫽ positive predictive value RS ⫽ renal scarring UTI ⫽ urinary tract infection VCUG ⫽ voiding cystourethrography VUR ⫽ vesicoureteral reflux WBC ⫽ white blood cell count Submitted for publication February 27, 2011. Supported by grants from the National Science Council, Taiwan (NSC93-2314-B-040-012) and Chung Shan Medical University Hospital (CSH-2010-C-009). Study received institutional review board approval. * Correspondence: Department of Pediatrics, Chung Shan Medical University Hospital, No. 110, Section 1, Jianguo North Rd., Taichung 402, Taiwan (telephone: ⫹886 4 2473 9595 ext. 34816; FAX: ⫹886 4 2471 0934; e-mail: [email protected]).
See Editorial on page 1760.
2002
www.jurology.com
Purpose: We assessed the usefulness of procalcitonin as a biological marker in diagnosing acute pyelonephritis and for predicting subsequent renal scarring in young children with a first febrile urinary tract infection. Materials and Methods: Children 2 years old or younger with a first febrile urinary tract infection were prospectively studied. Renal parenchymal involvement was assessed by 99mTc-dimercaptosuccinic acid scan within 5 days of admission and after 6 months. Serum samples from all patients were tested for procalcitonin, C-reactive protein and white blood cell count measurements. Results: The 112 enrolled patients (age range 24 days to 24 months old) were divided into acute pyelonephritis (76) and lower urinary tract infection (36) groups according to the results of 99mTc-dimercaptosuccinic acid scans. Median values of procalcitonin, C-reactive protein and white blood cell count at hospitalization were significantly higher in patients with acute pyelonephritis than in those with lower urinary tract infection. The area under receiver operating characteristic curves showed that procalcitonin was superior to C-reactive protein and white blood cell count as a marker for diagnosing acute pyelonephritis. Initial and post-antibiotic treatment procalcitonin values were significantly higher in children with renal scarring than in those without scarring (p ⬍0.001). Procalcitonin values at hospitalization and after treatment were independent predictors of later renal scarring on logistic regression analysis. Conclusions: Our results indicate the superior diagnostic accuracy of procalcitonin for predicting acute pyelonephritis in children 2 years old or younger. Higher initial and posttreatment procalcitonin values are independent risk factors for later renal scarring. Key Words: pyelonephritis, procalcitonin, infant, vesico-ureteral reflux URINARY tract infection is a common bacterial infection in pediatric patients, especially in children younger than 2 years old, with a prevalence of 6.5% and 3.3% in girls and boys younger than 1 year old, respectively.1 The nonspecific symptoms in febrile young children make the clinical differentiation between acute pyelonephritis and lower UTI difficult. Approximately 30%
to 60% of cases in small children result in renal scarring after acute febrile UTI.2– 4 A delay in the diagnosis and treatment of APN in young children can increase the risk of kidney damage,1,5 which may cause hypertension, proteinuria and end stage renal disease later in life.2,6 99m Tc-dimercaptosuccinic acid is widely accepted as the gold standard
0022-5347/11/1865-2002/0 THE JOURNAL OF UROLOGY® © 2011 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 186, 2002-2008, November 2011 Printed in U.S.A. DOI:10.1016/j.juro.2011.07.025
AND
RESEARCH, INC.
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
for detecting APN, and assessing the extent and progression of renal damage.7,8 However, factors like availability, as well as the costs and risks of exposing patients to radiation and sedation limit the widespread use of the DMSA scan. Furthermore, commonly used clinical and laboratory parameters do not accurately localize the UTI site.8,9 Therefore, finding an accurate and readily available diagnostic test to predict APN is of great value in young children with a febrile UTI. Procalcitonin is a 116 amino acid peptide and a precursor of calcitonin.10 Its release during infection is induced directly by microbial toxins and/or indirectly by humoral factors or the cell mediated host response.11,12 Most recent studies propose serum PCT as a better marker than C-reactive protein and white blood cell count in the early detection of APN in children.13–21 In contrast, 2 studies show that the PCT test is not a sensitive marker for the early diagnosis of APN.22,23 However, few studies have demonstrated a relationship between PCT and febrile UTI in children 2 years old or younger.14,19 Furthermore, the association between PCT and vesicoureteral reflux remains controversial.14,15,18,19,21,24 Therefore, we evaluated the role of serum PCT as a reliable marker for detecting APN in children 2 years old or younger with a first febrile UTI, and assessed its ability to predict the risk of subsequent RS. We also examined the association of PCT with VUR.
PATIENTS AND METHODS Study Population and Inclusion Criteria We prospectively studied children 2 years old or younger who were hospitalized with a febrile UTI in a 3-year period. The diagnosis of a first febrile UTI was based on fever (axillary temperature 38C or greater), leukocyturia (defined as 5 or more WBC per high power field), positive urine culture (defined as growth of a single microorganism 105 colony-forming units per ml or more collected from the midstream clean void urine specimen for toilet trained young children, or 104 colony-forming units per ml or more collected from a transurethral catheterized specimen), no history of UTI and kidney or bladder disease and no other coincidental infections. All patients were treated empirically with combined intravenous cefazolin (100 mg/kg daily) and gentamicin (7.5 mg/kg daily) for at least 3 days, which was later adjusted according to the results of bacterial susceptibility tests for a treatment duration of 7 to 14 days. The institutional review board of Medical University Hospital approved the study protocol and all parents of the participants provided informed consent.
Imaging Studies All patients underwent renal ultrasound for the detection of urinary tract anomalies within the first 3 days of hospital admission. DMSA scans were performed within the first 5 days of hospitalization to verify the presence of
2003
renal parenchymal lesions. APN was defined as the presence of focal or diffuse areas of decreased uptake without evidence of cortical loss.7,8 To assess the severity of acute renal involvement on PCT values, a scoring system was used based on DMSA grading and modified by Benador et al.21 Lesions were graded in 4 groups of 0 —no lesion (lower UTI), 1—mild lesions (defect covering less than 10% of surface area), 2—moderate lesions (defect covering 10% to 30%) and 3—severe lesions (defect covering more than 30%). If the initial DMSA result was abnormal, a followup examination was performed at least 6 months later to evaluate the presence of RS. The diagnosis of APN was confirmed only in children with totally or partially reversible lesions on the followup scans. The presence of renal lesions was determined by 2 nuclear medicine physicians who were blinded to the study. VCUG was performed 1 to 2 weeks after completing treatment for the infection. VUR was graded 0 to V according to the International Reflux Study.25
Laboratory Measurements Serum and urine indexes for laboratory investigations, including peripheral WBC and differential, CRP values, urinalysis, urine and blood cultures were measured in all patients at hospitalization and before the initiation of antibiotic treatment. Serum samples were also taken for PCT measurements at hospitalization and repeated 3 days later. A rapid and quantitative measurement of PCT was performed using enzyme-linked fluorescent assay in an automated VIDAS® instrument (VIDAS BRAHMS PCT, BRAHMS Diagnostica, Berlin, Germany). The detection limit was 0.05 ng/ml and PCT 0.5 ng/ml or greater was considered abnormal.
Statistical Analyses All statistical analyses were performed using SPSS® for Windows (version 15.0). Nonparametric data were assessed by the Mann-Whitney U test or Kruskal-Wallis 1-way ANOVA, and expressed as medians and interquartile ranges (Q1-Q3). Logistic regression analysis was used to evaluate the impact of potential risk on the development of RS in patients. The receiver operating characteristic curve analysis was performed to assess quantitative variables at hospitalization for diagnosing APN and later RS. The diagnostic values of each cutoff point, including sensitivity, specificity, PPV, NPV and the likelihood ratio for a positive result, were all calculated with p ⬍0.05 considered statistically significant.
RESULTS Patient and Clinical Characteristics There were 66 boys and 46 girls 24 days to 24 months old (median age 5.0 months), with a diagnosis of a first febrile UTI (table 1). Escherichia coli was isolated as a single pathogen in the urine culture of 100 (89.3%) patients. The number of urine collections by catheterized vs clean void specimens was 103 vs 9. Renal ultrasound revealed findings of normal (89), mild to moderate hydronephrosis (renal pelvic
2004
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
Table 1. Clinical characteristics and laboratory data of patients with APN and lower UTI APN Median mos age (IQR) No. male/female No./total no. VUR grade (%): III or Greater I–II No./total no. E. coli (%) Median ng/ml PCT (IQR) Median mg/dl CRP (IQR) Median WBC/mm3 (IQR)
Lower UTI
5.1 (3.0–8.0) 40/36 30/70 (42.9) 22/30 (73.3) 8/30 (26.7) 65/76 (85.5) 2.95 (1.37–8.37) 9.61 (5.39–14.60) 16,125 (12,278–19,673)
p Value
4.0 (2.0–10.9) 26/10 6/31 (19.4) 2/6 (33.3) 4/6 (66.70) 35/36 (97.2) 0.35 (0.25–0.44) 3.09 (1.49–5.68) 13,595 (9,460–16,368)
0.312* 0.078† 0.040† 0.150† 0.150† 0.769† ⬍0.001* ⬍0.001* 0.013*
* Mann-Whitney U test for continuous variables. † Chi-square or Fisher’s exact test for categorical variables.
diameter 5 mm or greater in 20), duplex kidney (2) and unilateral renal agenesis (1). Of the 101 (90.2%) patients who underwent VCUG studies (70 with APN and 31 with lower UTI) 36 (35.6%) had VUR with grade II (12), III (11), IV (9) and V (4) (maximum degree of reflux given if bilateral). Laboratory Markers for APN Diagnosis at Hospitalization PCT, CRP and WBC values of the severity of acute renal involvement are shown in table 2. Median PCT tended to increase significantly with increasing DMSA score (p ⬍0.001, fig. 1). Median PCT at hospitalization was significantly higher in patients with APN than in those with lower UTI (p ⬍0.001, table 1). Median PCT values initially and after 3 days of antibiotic treatment were 2.95 ng/ml (IQR 1.37– 8.37) vs 0.38 ng/ml (IQR 0.08 –1.10) (p ⬍0.001). To differentiate between APN and lower UTI at hospitalization, the ROC curves were used by plotting the sensitivity vs 1-specificity for different cutoff values of PCT, CRP and WBC. The area under the curve of the ROC was 0.910 for PCT (p ⬍0.001), 0.770 for CRP (p ⫽ 0.015) and 0.594 for WBC (p ⫽ 0.529). The ROC analysis demonstrated that PCT values had a greater AUC than CRP and WBC values for differentiating APN from lower UTI (fig. 2). The cutoff point for maximum diagnostic efficiency of PCT was 1.0 ng/ml, with 81.6% sensitivity and 91.7% specificity (table 3). Combined analysis of PCT and CRP showed that a concomitant PCT value of 1 ng/ml or greater and CRP 3.5 mg/dl or greater had a sensitivity of 77.6%, specificity 91.7%, PPV
95.2% and NPV 66%. However, this diagnostic efficiency was not superior to PCT alone for predicting APN. Laboratory Markers and RS Followup DMSA scan was performed 6 months after the infection in all 76 cases. Four patients who experienced a second UTI between the 2 DMSA scans were excluded from analysis. Renal scars were seen in 34 (47.2%) of the remaining 72 patients whereas 38 (52.8%) exhibited total regression of the initial lesions (table 4). Of the laboratory markers at hospitalization only median PCT was significantly higher in patients with RS (p ⬍0.001). All patients with RS had a PCT value greater than 2 ng/ml. The effect of age on PCT values was analyzed and no correlation was found (r ⫽ 0.060, p ⫽ 0.608). Of the 34 patients with RS 4 of 25 (16.0%), 13 of 26 (50.0%) and 17 of 21 (81.0%) had scores of 1, 2 and 3, respectively, with an overall median PCT of 8.57 ng/ml (IQR 5.37–15.24). The incidence of RS tended to increase significantly as the initial DMSA score increased (p ⬍0.001). In the 38 patients with normal followup DMSA results the overall median PCT was 1.45 ng/ml (IQR 0.57–2.27). The initial median PCT was significantly higher in patients with RS (p ⬍0.001, fig. 3). Moreover median posttreatment PCT in patients with RS was 1.13 ng/ml (IQR 0.84 – 1.99) vs 0.08 ng/ml (IQR 0.06 – 0.21) (p ⬍0.001). VUR was found in 11 (32.4%) of 34 patients with RS (table 4). No association was noted between RS and VUR grade. There was no significant difference between PCT in patients with and in those without VUR (p ⫽
Table 2. PCT, CRP and WBC values indicating acute renal involvement severity
No. pts Median ng/ml PCT (IQR) Median mg/dl CRP (IQR)† Median WBC/mm3 (IQR)†
DMSA Score 0
DMSA Score 1
DMSA Score 2
DMSA Score 3
p Value*
36 0.35 (0.25–0.44) 3.09 (1.49–5.68) 13,595 (9,460–16,368)
27 1.35 (0.48–2.15) 10.10 (7.55–13.60) 18,250 (11,300–20,640)
28 3.45 (1.87–8.99) 8.08 (4.45–12.01) 16,530 (12,308–21,158)
21 8.57 (4.79–15.46) 14.50 (4.85–18.50) 14,750 (2,750–16,640)
⬍0.001 ⬍0.001 0.034
* Kruskal-Wallis 1-way ANOVA for continuous variables. † There were no significant differences among DMSA scores 1, 2 and 3 in CRP (p ⫽ 0.082) and WBC (p ⫽ 0.245).
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
2005
Table 3. Sensitivity, specificity, PPV, NPV and likelihood ratio for a positive result of PCT, CRP and WBC for the diagnosis of APN
Figure 1. PCT values of patients with APN at hospitalization in terms of acute renal lesion severity by DMSA score. Horizontal line indicates median and box indicates quartiles (Kruskal-Wallis test p ⬍0.001).
0.409). There was also no significant correlation of PCT and VUR grade (p ⫽ 0.166). Variables Predicting APN and Late RS The ROC analyses demonstrated that initial and posttreatment PCT values were useful indexes for predicting later RS (AUC 0.942, p ⬍0.001 and AUC 0.963, p ⬍0.001, respectively). The optimal cutoff value of 3.5 ng/ml at hospitalization and 1.0 ng/ml for posttreatment had maximum diagnostic effi-
Figure 2. Comparison of ROC curves for PCT, CRP and WBC values to distinguish APN from lower UTI. PCT (AUC 0.910, p ⬍0.001) was determined to be best index compared to CRP (AUC 0.770, p ⫽ 0.015) and WBC (AUC 0.594, p ⫽ 0.529).
PCT (ng/ml): 0.5 or Greater 1 or Greater* 1.5 or Greater 2 or Greater CRP (mg/dl): 2 or Greater 3.5 or Greater 6 or Greater 10 or Greater WBC (/mm3): 12,000 or Greater 15,000 or Greater 20,000 or Greater
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
Pos Likelihood Ratio
89.5 81.6 72.4 64.5
83.3 91.7 94.4 97.2
91.9 95.4 96.5 98.0
78.9 70.2 61.8 56.5
5.4 9.8 12.9 23.0
100 90.8 73.7 47.4
38.9 58.3 80.6 94.4
77.6 82.1 88.9 94.7
100 75.0 59.2 45.9
1.6 2.2 3.8 8.5
78.9 57.9 21.1
38.9 66.7 88.9
73.2 78.6 80.0
46.7 42.9 34.8
1.3 1.7 1.9
* Maximum diagnostic efficiency.
ciency for RS (table 5). Multivariate analyses revealed that the independent predictors of APN were initial PCT and CRP values as well as the presence of VUR (table 6). Initial and posttreatment PCT values were independent predictors of later RS.
DISCUSSION This study focused on young children 2 years old or younger because this age population has the most frequent cases of febrile UTI. To date to our knowledge, our study enrolled the largest case number of a pediatric population age 2 years or younger to explore the diagnostic value of PCT for predicting APN in children with a febrile UTI. Our results revealed that PCT values only were influenced by the severity of renal lesions and were not subject to an age related effect in this age population. This study shows that initial PCT is a reliable marker in distinguishing APN from lower UTI in young children, consistent with the published data.13–21 Previous studies have shown that PCT was increased in children with APN, with a sensitivity of 70% to 100% and a specificity of 51% to 97%.13–21 Our data show that using a PCT cutoff value of 1 ng/ml has the best performance, with a high sensitivity (81.6%) and specificity (91.7%). The AUCs of ROC show that PCT has a much greater area than those of CRP and WBC for predicting APN. In this study 81.6% of patients with APN had PCT values greater than 1 ng/ml compared to only 8.3% of those with a lower UTI. Moreover all patients with RS presented with an initial PCT greater than 1 ng/ml. Our findings suggest that when PCT values at hospitalization are less than 1 ng/ml, the possibility of renal involvement is low in young chil-
2006
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
Table 4. Comparisons of clinical characteristics and laboratory markers at hospitalization in patients with and those without RS DMSA Scan
No. male/female Median mos age (IQR) Median ng/ml PCT (IQR) Median mg/dl CRP (IQR) Median WBC/mm3 (IQR) No. VUR/no VUR No. VUR grade I–II/III–V
RS⫹
RS⫺
p Value
21/13 5.6 (3.9–8.0) 8.57 (5.37–15.24) 10.83 (5.05–16.85) 15,190 (12,720–20,868) 11/23 1/10
18/20 5.8 (2.9–10.3) 1.45 (0.57–2.27) 9.52 (5.44–13.43) 16,240 (11,460–19,045) 17/21 3/14
0.324* 0.973† ⬍0.001† 0.417† 0.446† 0.404* 1.000*
* Chi-square or Fisher’s exact test for categorical variables. † Mann-Whitney U test for continuous variables.
dren with febrile UTI, and DMSA studies during the acute phase or after 6 months can be omitted for the selection of cases. The results of followup DMSA scans show that RS developed in 47.2% of patients after APN. These 34 cases of RS were characterized by higher initial PCT values compared to other cases of APN. The results also demonstrate significant associations of increased PCT at hospitalization with severity of acute renal involvement and later RS, consistent with few studies on the correlation between PCT and RS in younger children14 and children.15,17,18,21 Moreover our results also demonstrate a positive association of posttreatment PCT values with RS. This finding has not yet been reported in other studies. Our results indicate that using a PCT cutoff greater than 3.5 ng/ml at hospitalization and greater than 1.0 ng/ml after treatment can provide 17.8 and 12.8 times, respectively, higher likelihood of finding patients with RS. Therefore, higher initial and posttreatment PCT values are associated with a higher risk of RS.
The association of PCT with VUR in children with febrile UTI remains controversial. This study revealed no significant differences in PCT values in children with and in those without VUR and VUR grade, similar to the published data.15,18,19,21 In contrast, Bressan et al found significantly higher PCT values in children with VUR, but revealed no correlation between PCT levels and VUR grade.14 However, only 13 patients with VUR were enrolled in their study and the researchers declared that this small number of patients prevented drawing definite conclusions on the subject. Although VUR may predispose to APN, to our knowledge no association has been established between the presence of VUR and persistent renal damage in children with febrile UTI.2,26 Previous studies indicate that increased PCT concentrations during the early phase of the disease are closely related to the onset of bacterial infection and correlated to disease severity.27,28 Our findings support this suggestion. This area needs further investigation in a large number of children to clarify the role of PCT as a predictor of VUR. This study has some limitations. Most patients with a febrile UTI were treated as outpatients. Therefore, a selection population bias might have Table 5. Indices of diagnostic quality and efficiency of PCT for predicting subsequent RS
Figure 3. Median PCT at hospitalization in patients with lower UTI, APN, and APN with and without RS. Horizontal line indicates median and box indicates quartiles (Kruskal-Wallis test p ⬍0.001).
Cutoff (ng/ml) at hospitalization: 1.0 or Greater 2.0 or Greater 3.5 or Greater* 5.0 or Greater 6.0 or Greater Cutoff (ng/ml) after treatment: 1.0 or Greater* 1.5 or Greater 2.0 or Greater
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
Pos Likelihood Ratio
100 97.1 94.1 76.5 70.6
34.2 65.8 94.7 94.7 97.4
57.6 71.7 94.7 92.9 96.0
100 96.2 94.1 81.8 78.7
1.5 2.9 17.8 14.4 27.2
67.6 44.1 23.5
94.7 97.4 97.4
93.8 92.0 88.9
66.1 76.6 58.7
12.8 17.0 9.0
* Maximum diagnostic efficiency.
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
2007
Table 6. Multivariate logistic regression models predicting APN and later RS in patients after febrile UTI
Age Gender PCT CRP WBC Posttreatment PCT Reflux grade: 0 I–II III–V DMSA lesion score: 1 2 3
OR (95% CI) APN
p Value
0.898 (0.750–1.076) 3.127 (0.521–18.777) 12.189 (2.310–64.300) 1.434 (1.103–1.864) 1.000 (1.000–1.000) —
0.243 0.213 0.003 0.007 0.075
1 [Reference] 22.857 (1.379–378.967) 43.818 (2.179–881.245) — — — —
0.029 0.014
OR (95% CI) RS 1.085 0.261 3.133 0.903 1.000 2.451
p Value
(0.851–1.583) (0.024–2.790) (1.549–6.336) (0.769–1.060) (1.000–1.000) (1.368–4.389)
0.510 0.267 0.001 0.211 0.230 0.003
1 [Reference] 0.051 (0.000–27.294) 0.354 (0.045–2.780)
0.353 0.324
1 [Reference] 0.568 (0.050–6.445) 0.897 (0.067–11.99)
0.648 0.935
For multivariate analysis a multivariate logistic regression model was used.
been present. In addition, VCUG examination was not performed in nearly 10% of our cohort, which might have affected our results in terms of the analysis of PCT in patients with VUR. Another limitation is that patients with UTI without fever or with a temperature less than 38C and the absence of leukocyturia were excluded from study. Thus, some diagnostic bias in the study might have affected the data. How PCT can replace DMSA is yet to be determined. The price of DMSA examination is approximately $150, which is 5 times that of PCT measurement (around $30). From a cost-effective point of view of diagnostic tests a PCT based diagnostic strategy vs DMSA based study can reduce costs by 80% per test. From a patient safety standpoint, the most important point, a PCT based strategy can avoid the risks of radiation and sedation exposure in small children, risks which are associated with DMSA examinations. Our results suggest that chil-
dren 2 years old or younger with a low PCT value (less than 1 ng/ml) at hospitalization may not require DMSA study after a first UTI. For pediatricians, knowing the limitations of DMSA, these findings are of great clinical importance because they may reserve DMSA examinations for those with a PCT value greater than 1 ng/ml.
CONCLUSIONS Our study indicates the superior diagnostic accuracy of PCT in APN in children 2 years old or younger. Higher initial and posttreatment PCT values indicate a high risk of later RS. These results may lead to more personalized diagnostic and therapeutic strategies for children 2 years old or younger with a first febrile UTI and may contribute to the avoidance of further unnecessary DMSA studies in children with low PCT values.
REFERENCES 1. American Academy of Pediatrics, Committee on Quality Improvement, Subcommittee on Urinary Tract Infection: Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and younger children. Pediatrics 1999; 103: 843. 2. Stokland E, Hellström M, Jacobsson B et al: Early 99mTc dimercaptosuccinic acid (DMSA) scintigraphy in symptomatic first-time urinary tract infection. Acta Paediatr 1996; 85: 430.
5. Hiraoka M, Hashimoto G, Tsuchida S et al: Early treatment of urinary infection prevents renal damage on cortical scintigraphy. Pediatr Nephrol 2003; 18: 115.
9. Garin EH, Olavarria F, Araya C et al: Diagnostic significance of clinical and laboratory findings to localize site of urinary infection. Pediatr Nephrol 2007; 22: 1002.
6. Jacobson SH, Eklöf O, Lins LE et al: Long-term prognosis of post-infectious renal scarring in relation to radiological findings in childhood–a 27year follow-up. Pediatr Nephrol 1992; 6: 19.
10. Russwurm S, Wiederhold M, Oberhoffer M et al: Molecular aspects and natural source of procalcitonin. Clin Chem Lab Med 1999; 37: 789.
3. Benador D, Benador N, Slosman D et al: Are younger children at highest risk of renal sequelae after pyelonephritis? Lancet 1997; 349: 17.
7. Rushton HG: The evaluation of acute pyelonephritis and renal scarring with technetium 99m-dimercaptosuccinic acid renal scaintigraphy: evolving concepts and future directions. Pediatr Nephrol 1997; 11: 108.
4. Sheu JN, Chen SM, Meng MH et al: The role of serum and urine interleukin-8 on acute pyelonephritis and subsequent renal scarring in children. Pediatr Infect Dis J 2009; 28: 885.
8. Biggi A, Dardanelli L, Pomero G et al: Acute renal cortical scintigraphy in children with a first urinary tract infection. Pediatr Nephrol 2001; 16: 733.
11. Müller B, White JC, Nylén ES et al: Ubiquitous expression of the calcitonin-I gene in multiple tissues in response to sepsis. J Clin Endocrinol Metab 2001; 86: 396. 12. Dandona P, Nix D, Wilson MF et al: Procalcitonin increase after endotoxin injection in normal subjects. J Clin Endocrinol Metab 1994; 79: 1605. 13. Nikfar R, Khotaee G, Ataee N et al: Usefulness of procalcitonin rapid test for the diagnosis of acute
2008
PROCALCITONIN FOR ACUTE PYELONEPHRITIS AND RENAL SCARRING
pyelonephritis in children in the emergency department. Pediatr Int 2010; 52: 196.
scarring in children with urinary tract infection. Pediatr Infect Dis J 2003; 22: 438.
14. Bressan S, Andreola B, Zucchetta P et al: Procalcitonin as a predictor of renal scarring in infants and young children. Pediatr Nephrol 2009; 24: 1199.
19. Smolkin V, Koren A, Raz R et al: Procalcitonin as a marker of acute pyelonephritis in infants and children. Pediatr Nephrol 2002; 17: 409.
15. Kotoula A, Gardikis S, Tsalkidis A et al: Comparative efficacies of procalcitonin and conventional inflammatory markers for prediction of renal parenchymal inflammation in pediatric first urinary tract infection. Urology 2009; 73: 782. 16. Gürgöze MK, Akarsu S, Yimaz E et al: Proinflammatory cytokines and procalcitonin in children with acute pyelonephritis. Pediatr Nephrol 2005; 20: 1445. 17. Pecile P, Miorin E, Romanello C el al: Procalcitonin: a marker of severity of acute pyelonephritis among children. Pediatrics 2004; 114: e249. 18. Prat C, Domínguez J, Rodrigo C et al: Elevated serum procalcitonin values correlate with renal
20. Gervaix A, Galetto-Lacour A, Gueron T et al: Usefulness of procalcitonin and C-reactive protein rapid tests for the management of children with urinary tract infection. Pediatr Infect Dis J 2001; 20: 507. 21. Benador N, Siegrist CN, Gendrel D et al: Procalcitonin is a marker of severity of renal lesions in pyelonephritis. Pediatrics 1998; 102: 1422.
24. Leroy S, Adamsbaum C, Marc E et al: Procalcitonin as a predictor of vesicoureteral reflux in children with a first febrile urinary tract infection. Pediatrics 2005; 115: e706. 25. Lebowitz RL, Olbing H, Parkkulainen KV et al: International system of radiographic grading of vesicoureteric reflux. Pediatr Radiol 1985; 15: 105. 26. Ditchfield MR, Grimwood K, Cook DJ et al: Persistent renal cortical scintigram defects in children 2 years after urinary tract infection. Pediatr Radiol 2004; 34: 465.
22. Güven AG, Kazdal HZ, Koyun M et al: Accurate diagnosis of acute pyelonephritis: how helpful is procalcitonin? Nucl Med Commun 2006; 27: 715.
27. van Rossum AM, Wulkan RW and OudesluysMurphy AM: Procalcitonin as an early marker of infection in neonates and children. Lancet Infect Dis 2004; 4: 620.
23. Tuerlinckx D, Vander Borght T, Glupczynski Y et al: Is procalcitonin a good marker of renal lesion in febrile urinary tract infection? Eur J Pediatr 2005; 164: 651.
28. Müller B, Becker KL, Schächinger H et al: Calcitonin precursors are reliable markers of sepsis in a medical intensive care unit. Crit Care Med 2000; 28: 977.