First-trimester aggregate index of systemic inflammation and adverse pregnancy outcomes in hospitalized women with threatened miscarriage: a single-center retrospective cohort study

Abstract

Abstract Background Complete blood count-derived inflammatory indices have been associated with miscarriage, but the incremental value of the aggregate index of systemic inflammation (AISI) beyond established indices such as the systemic immune-inflammation index (SII) remains uncertain. We examined the association and prognostic performance of first-trimester AISI in hospitalized women with threatened miscarriage. Methods This single-center retrospective cohort study included 312 women hospitalized with threatened miscarriage between January 2020 and December 2022. The primary outcome was a composite of miscarriage before 14 weeks, preterm birth before 37 weeks with or without preterm prelabor rupture of membranes (PPROM), or fetal demise from 14 weeks onward. AISI was analyzed continuously on the log2 scale. Multivariable logistic regression, restricted cubic splines, receiver operating characteristic analysis with DeLong comparisons, calibration, 500-resample bootstrap internal validation, and decision curve analysis were performed. Results Adverse outcomes occurred in 134 women (42.9%). Median AISI was higher in the adverse-outcome group than in women with term birth (339.9 [IQR 250.2-432.1] vs. 236.6 [167.0-313.9]; p < 0.001). Each doubling of AISI was associated with higher adjusted odds of an adverse outcome (adjusted odds ratio 4.46, 95% CI 2.59–7.67; p < 0.001), with no evidence of nonlinearity ( p = 0.499). AISI showed moderate discrimination (AUC 0.717, 95% CI 0.661–0.774), which was not significantly different from SII (AUC 0.696; DeLong p = 0.299). The exploratory AISI cut-off of 287.3 had 66.4% sensitivity (95% CI 58.1–73.9%) and 69.1% specificity (95% CI 62.0-75.4%). Adding AISI to the clinical plus C-reactive protein model increased the bootstrap-corrected AUC from 0.811 to 0.853 and reduced the corrected Brier score from 0.176 to 0.153; the incremental decision-curve benefit was modest. Conclusions First-trimester AISI was independently associated with adverse pregnancy outcomes in this selected hospitalized cohort and added prognostic information to clinical variables and C-reactive protein. However, AISI alone had only moderate discrimination and was not superior to SII. These findings do not support a stand-alone clinical cut-off or a specific management intervention; prospective multicenter validation is required.