Association Between Novel Indicators for Inflammation and Metabolism and Kidney Stones: A Cross-Sectional Analysis of NHANES 2007-2018

Four composite blood ratios tied to kidney stone prevalence in 27K Americans — but one of the four didn't hold up under scrutiny

Journal: BMC Urology | Published: 2026-03-27 | Type: Cross-sectional | PMID: 41896861 Authors: Wang Jingxian et al., Department of Urology, The Affiliated Hospital of Qingdao University, China Funding/COI: Shandong Province Medical Health Science and Technology Project. No competing interests declared.

Summary

Using six cycles of NHANES data (2007–2018), this study tested whether four composite biomarker ratios — each bundling an immune cell count with a metabolic marker — are associated with kidney stone history in a nationally representative US sample. Three of the four held up across multiple analytical approaches; the fourth (NAR) fell apart once the researchers moved beyond basic logistic regression. The framing as "novel indicators" is generous — these are ratios of already well-known risk factors dressed up in new acronyms.

Claims

• NPAR (neutrophil percentage ÷ albumin) associated with kidney stones: OR 1.63 (95% CI 1.22–2.18, p=0.001) in fully adjusted models; nonlinear association not significant on RCS
• NAR (neutrophil count ÷ albumin): OR 1.20 (95% CI 1.04–1.40, p=0.015) in logistic regression, but association disappeared in quartile analysis and trend testing — the authors acknowledge this directly
• UHR (uric acid × 100 ÷ HDL-C): OR 1.43 (95% CI 1.15–1.79, p=0.002); strongest and most consistent across analyses; most pronounced in adults aged ≥60
• LHR (lymphocyte count ÷ HDL-C): OR 1.15 (95% CI 1.02–1.30, p=0.026); weakest association; sex was a significant effect modifier, with stronger association in females
• 2,636 of 27,420 enrolled participants had kidney stones (9.6% prevalence)
• Participants with kidney stones were older (mean 53.01 vs. 46.86 years), more often male (54.9% vs. 48.0%), had higher BMI, and higher rates of hypertension, diabetes, and hyperlipidemia

Study Quality

The NHANES platform is a genuine strength — stratified multistage probability sampling, well-documented protocols, and decades of methodological refinement make it a reliable substrate. The authors appropriately applied survey-weighted analyses and adjusted for a broad covariate set including age, sex, race, BMI, smoking, alcohol, physical activity, hypertension, diabetes, hyperlipidemia, and serum uric acid. RCS modeling to detect nonlinearity and formal interaction testing for subgroup effects are methodologically sound choices.

Cross-sectional design is the ceiling here, and it's a low ceiling. Kidney stone history was captured entirely by self-report — "Have you ever had kidney stones?" — with no imaging, clinical records, or laboratory confirmation. There's no way to determine whether elevated inflammatory or metabolic markers preceded stone formation or followed from it, and recall bias in self-reported stone history is well-documented. The composite ratios also contain components (uric acid, HDL-C, albumin) that are themselves established stone-risk variables, which makes independent contribution murky.

Red Flags

• Causation is impossible to establish: cross-sectional design with self-reported outcome means directionality is entirely unknown
• NAR failed its own validation: the authors report OR 1.20 in logistic regression, then acknowledge it was non-significant in quartile and trend analyses — yet it remains in the headline findings
• Self-reported kidney stone diagnosis: no imaging or medical record confirmation; misclassification is likely
• "Novel" is oversold: NPAR, NAR, UHR, and LHR are algebraic combinations of existing biomarkers. UHR is essentially a uric acid/HDL ratio — uric acid and HDL are each independently linked to kidney stones
• No clinical thresholds proposed: the paper identifies associations but provides no cutoff values, limiting any practical interpretation even for researchers
• Heterogeneous indicator performance: the four indicators behave differently by sex and age, which the authors frame as nuance but which also suggests these composite ratios aren't capturing a single coherent biological signal

Strengths

• Large, nationally representative US sample (n=27,420 across 11 years of NHANES cycles)
• Appropriate survey-weighted statistical methods throughout
• Multiple analytical approaches (continuous regression, quartile analysis, trend testing, RCS) — and the authors reported where results diverged, rather than hiding it
• Subgroup analyses with interaction tests are reported transparently
• Adjustment for a comprehensive confounder set

Verdict

This is a competent secondary analysis of NHANES data that demonstrates associations between composite inflammatory-metabolic ratios and self-reported kidney stone history. The UHR finding is the most credible — it's consistent across all analytical approaches and aligns with prior literature on uric acid metabolism and stone formation. The decision to include NAR as a positive finding despite it failing quartile and trend testing is the most telling weakness: when your own robustness checks don't replicate, that's a signal, not a footnote. The paper is adequate as hypothesis-generating material for prospective work, but the "novel indicators" framing overstates what are ultimately repackaged versions of already-known risk factors.