Clinical analysis of high-risk factors for steroid resistance in children with primary nephrotic syndrome

@article{gongping2026clinical,
  title = {{Clinical analysis of high-risk factors for steroid resistance in children with primary nephrotic syndrome}},
  author = {Zhao Gong-ping and Li Li-jie and Liu Yu-jie and Zhang Ya-xin and Jia Dong-hua and Li Ji-tong and Zhang Shu-feng and Liu Cui-hua},
  journal = {Linchuang shenzangbing zazhi},
  year = {2026},
  month = oct,
  url = {http://www.lcszb.com/thesisDetails#10.3969/j.issn.1671-2390.2026.10.004},
  note = {CC BY-NC-ND},
}

TY  - JOUR
TI  - Clinical analysis of high-risk factors for steroid resistance in children with primary nephrotic syndrome
AU  - Zhao Gong-ping
AU  - Li Li-jie
AU  - Liu Yu-jie
AU  - Zhang Ya-xin
AU  - Jia Dong-hua
AU  - Li Ji-tong
AU  - Zhang Shu-feng
AU  - Liu Cui-hua
PY  - 2026
DA  - 2026/10/24
JO  - Linchuang shenzangbing zazhi
UR  - http://www.lcszb.com/thesisDetails#10.3969/j.issn.1671-2390.2026.10.004
AB  - ObjectiveTo investigate the clinically relevant risk factors for steroid resistance in newly diagnosed and treatment-naïve children with primary nephrotic syndrome （PNS）.MethodsA retrospective analysis was performed on the clinical data of 406 children with PNS admitted to the Department of Nephrology， Rheumatology， and Immunology of Children’s Hospital Affiliated to Zhengzhou University from January 2020 to December 2024. According to whether the children were sensitive to steroid therapy， they were divided into a steroid-resistant nephrotic syndrome group （n=36） and a steroid-sensitive nephrotic syndrome group （n=370）. Logistic regression analysis was used to identify risk factors for steroid resistance in children with PNS. To assess the robustness of the multivariable regression model， receiver operating characteristic curve analysis was performed to evaluate its predictive value， and five-fold cross-validation and Bootstrap methods were used for internal validation. The concordance index （C-index） was used to assess discrimination.ResultsThere were significant differences between the two groups in age （5.89 ［2.96， 9.17］ years vs 3.87 ［2.62， 6.24］ years）， microscopic hematuria （39% vs 11%）， lymphocyte count （3.50 ［2.44， 5.00］×10⁹/L vs 4.26 ［3.04， 5.92］×10⁹/L）， estimated glomerular filtration rate （120.00 ［93.93， 120.00］ mL·min⁻¹·［1.73 m²］⁻¹ vs 120.00 ［120.00， 120.00］ mL·min⁻¹·［1.73 m²］⁻¹）， IgG （2.83 ［1.14， 3.84］ g/L vs 1.89 ［1.01， 2.85］ g/L）， IgA （1.15 ［0.97， 1.44］ g/L vs 0.94 ［0.64， 1.37］ g/L）， blood urea nitrogen （5.45 ［3.73， 8.30］ mmol/L vs 4.30 ［3.40， 5.70］ mmol/L）， urine protein-to-creatinine ratio （5.19 ［3.92， 9.98］ vs 4.35 ［3.28， 6.78］）， urinary β₂-microglobulin （447.90 ［259.10， 1264.40］ μg/L vs 339.80 ［221.80， 652.70］ μg/L）， and 24-hour urinary protein excretion （137.04 ［90.00， 198.02］ mg/kg vs 108.70 ［76.95， 157.39］ mg/kg） （P<0.05）. Multivariable logistic regression analysis showed that increasing age （OR=1.205， 95%CI： 1.025-1.416）， microscopic hematuria （OR=4.058， 95%CI： 1.623-10.141）， elevated IgG level （OR=1.385， 95%CI： 1.060-1.810）， and increased 24-hour urinary protein excretion （OR=1.011， 95%CI： 1.005-1.018） were risk factors for steroid resistance in children with PNS （P<0.05）. Receiver operating characteristic curve analysis showed that the combined prediction of steroid resistance by these four factors had a sensitivity of 91.4%， a specificity of 56.7%， and an area under the curve of 0.770 （95%CI： 0.687-0.853， P<0.001）. Bootstrap validation showed that the C-index of this regression model was 0.755 （95%CI： 0.645-0.865）， and the mean C-index from five-fold cross-validation was 0.760.ConclusionIncreasing age， microscopic hematuria， elevated IgG level， and increased 24-hour urinary protein excretion are independent risk factors for steroid resistance in children with PNS.
SN  - 1671-2390
ER  -