In an observational investigation of patients diagnosed with sepsis, researchers evaluated the prognostic significance of circulating antibody classes and discovered that higher serum immunoglobulin A concentrations were associated with increased mortality among individuals who maintained preserved circulating immunoglobulin G levels. The multi-center investigation, conducted across two hospital intensive care units in China alongside an independent cohort derived from the public MIMIC-IV critical care database, specifically examined sepsis patients presenting with serum immunoglobulin G concentrations at or above 670 milligrams per deciliter. While aggregate analyses of unstratified critical care cohorts revealed no crude baseline differences in total immunoglobulin concentrations between surviving patients and nonsurvivors, targeted survival modeling uncovered a distinct association between elevated circulating immunoglobulin A and reduced survival probability within the subset of patients with preserved baseline immunoglobulin G.
Sepsis represents a complex and life-threatening medical emergency defined by a profoundly dysregulated host systemic inflammatory response to invasive microbial infection, which frequently culminates in progressive multi-organ injury, vascular collapse, and clinical mortality. In intensive care medicine, evaluating underlying humoral immune competence and adaptive immune status has traditionally centered on quantifying circulating serum immunoglobulins, which comprise five structurally distinct functional isotypes including immunoglobulin G, immunoglobulin A, and immunoglobulin M. Prior critical care investigations evaluating whether baseline serum immunoglobulin concentrations can reliably predict clinical trajectories and overall survival among septic patients have yielded highly conflicting results across different patient populations, a discrepancy that investigators attribute to the extensive clinical heterogeneity inherent to severe systemic infection.
To investigate whether baseline humoral stratification could clarify these conflicting prognostic patterns, the study investigators structured their experimental framework around an established literature-derived threshold of 670 milligrams per deciliter of immunoglobulin G. This stratification boundary was adopted from earlier clinical findings indicating that profound hypogammaglobulinemia reflects a distinct immunological phenotype characterized by heightened 28-day mortality risk. Building upon that observation, the researchers hypothesized that the clinical relevance of other circulating antibody classes, particularly immunoglobulin A, might differ markedly depending on whether a patient has depleted or relatively preserved background levels of circulating immunoglobulin G. Immunoglobulin A serves as the predominant antibody class protecting mucosal surfaces throughout the respiratory and gastrointestinal tracts, yet its quantitative relationship with systemic organ dysfunction and sepsis outcomes had remained uncertain.
The study protocol drew upon two distinct patient populations to test this hypothesis under real-world clinical conditions. The primary clinical cohort comprised 233 consecutively evaluated adult patients admitted to the intensive care units of Fujian Medical University Union Hospital and Nan’an City Hospital between May 1, 2024, and December 31, 2025. All included individuals met standardized Sepsis-3 diagnostic criteria with infection-related organ dysfunction indicated by Sequential Organ Failure Assessment scores of two or greater, an age between 18 and 85 years, and hospital stays exceeding 24 hours. Patients presenting with active connective tissue disorders, acquired immunodeficiency conditions, active post-transplant antirejection regimens, pregnancy, or intensive care stays shorter than 48 hours were excluded. For cross-cohort comparison, the team extracted data from the MIMIC-IV critical care database, evaluating 1,121 eligible sepsis patients who had documented serum immunoglobulin measurements available from their critical care admissions.
Initial baseline evaluations conducted across the full cohorts confirmed that circulating immunoglobulin concentrations did not systematically differ when evaluated broadly between survivors and nonsurvivors. In the primary clinical cohort of 233 patients, including 176 survivors and 57 nonsurvivors, median baseline concentrations of serum globulin, immunoglobulin G, immunoglobulin A, and immunoglobulin M exhibited no statistically significant variations between outcome groups. Similar findings emerged within the overall MIMIC-IV cohort of 1,121 patients, where 906 individuals survived hospitalization and 215 died. Across both cohorts, nonsurvivors were distinguished primarily by standard physiologic and hematologic indices of critical illness, displaying markedly higher acute illness scores on the Sequential Organ Failure Assessment and Acute Physiology and Chronic Health Evaluation II systems, accompanied by significantly lower peripheral platelet counts.
Pronounced statistical divergences appeared, however, when researchers implemented multivariable Cox proportional hazards regression models that tested interaction terms between circulating immunoglobulin A and dichotomized immunoglobulin G status. Within the primary clinical cohort, testing revealed a statistically significant interaction between continuous serum immunoglobulin A and the immunoglobulin G stratification threshold of 670 milligrams per deciliter. Among the 153 clinical patients categorized into the subgroup with immunoglobulin G at or above 670 milligrams per deciliter, who experienced 34 observed deaths, each 100 milligram per deciliter increment in serum immunoglobulin A was associated with a multivariable-adjusted hazard ratio for mortality of 1.269 after adjusting for patient age, sex, and Sequential Organ Failure Assessment score. Replacing organ failure scores with Acute Physiology and Chronic Health Evaluation II scores yielded a comparable adjusted hazard ratio of 1.284.
In contrast, parallel survival analyses conducted within the larger MIMIC-IV critical care subgroup yielded substantially more attenuated and less consistent statistical associations. Among the 771 patients in the MIMIC-IV database with baseline immunoglobulin G at or above 670 milligrams per deciliter, who experienced 130 in-hospital deaths, an unadjusted Cox proportional hazards analysis showed an initial hazard ratio of 1.053 per 100 milligram per deciliter increase in immunoglobulin A. Following multivariable adjustment for age, sex, and Sequential Organ Failure Assessment score, the corresponding hazard ratio was 1.043 with a 95 percent confidence interval spanning from 0.999 to 1.089, which did not cross the threshold of statistical significance. When Acute Physiology and Chronic Health Evaluation II scores were incorporated into the MIMIC-IV model instead, the adjusted hazard ratio reached 1.045 with a confidence interval that narrowly excluded unity, reflecting marginal statistical significance.
To explore potential physiological mechanisms underlying these statistical associations, the research team conducted a focused sub-investigation examining mucosal compartment dynamics and pulmonary tissue alterations in 45 consecutive patients presenting with nonpulmonary sepsis and preserved immunoglobulin G. Nonpulmonary sepsis patients often develop acute indirect pulmonary complications and exudative radiographic abnormalities despite their primary infections originating at distant anatomical sites such as the abdomen, urinary tract, or soft tissues. Serial chest computed tomography imaging was performed to quantify structural lung changes using a standardized 25-point radiographic scoring scale, measuring the numerical difference between admission scans and peak follow-up scans. Both circulating serum immunoglobulin A levels and bronchoalveolar lavage fluid immunoglobulin A concentrations normalized via urea nitrogen calibration displayed positive linear correlations with progressive pulmonary computed tomography lesion scores.
The researchers further utilized high-parameter flow cytometric analysis to examine bronchoalveolar lavage cellular suspensions recovered from a subset of 36 mechanically ventilated patients with nonpulmonary sepsis. Bronchoalveolar lavage fluid was harvested during clinically required therapeutic bronchoscopic suctioning procedures rather than elective sampling, adhering to standardized double-lavage protocols using sterile saline instilled into dependent lower lung lobes. Flow cytometric gating identified singlets, viable CD45-positive leukocytes, CD19-positive B-cell lineages, and specific surface immunoglobulin A1 expression. Intriguingly, flow cytometry revealed no statistically significant differences between surviving patients and nonsurvivors regarding either the relative proportion of immunoglobulin A1-positive B cells among recovered airway leukocytes or the mean fluorescence intensity of immunoglobulin A1 expression within the B-cell subset.
The study authors highlighted several critical methodological constraints that prevent translating these observational findings into clinical decision-making or definitive causal models. A central limitation involves the profound selection pressure present within retrospective critical care databases; in MIMIC-IV, over 97 percent of potentially eligible sepsis patients lacked documented immunoglobulin testing and were excluded, indicating that clinical laboratory ordering may have been driven by specific diagnostic suspicions. Furthermore, while circulating immunoglobulin A correlated with progressive lung imaging changes, the absence of differences in local airway B-cell populations indicates that current evidence cannot discern whether mucosal immunoglobulin A elevations stem from local hypersecretion, passive systemic microvascular leakage across damaged alveolar-capillary barriers, or non-causal reflection of severe underlying systemic tissue damage.
Subject of Research: Serum immunoglobulin A concentrations and mortality risk in sepsis patients with preserved immunoglobulin G levels
Article Title: Association between serum IgA levels and mortality in patients with sepsis and serum IgG ≥670 mg/dL: an observational study of clinical and MIMIC-IV cohorts
Article References: Tu, F., Huang, Y., Li, W., Lin, K., Wu, Q., Lin, Y., Yu, R., Chen, J., Zhou, L., Ye, R., & Wu, W. (2026). Association between serum IgA levels and mortality in patients with sepsis and serum IgG ≥670 mg/dL: an observational study of clinical and MIMIC-IV cohorts. Frontiers in Medicine, 13, Article 1961362. https://doi.org/10.3389/fmed.2026.1961362
Image Credits: AI Generated
DOI: 10.3389/fmed.2026.1961362
Keywords: Sepsis, Immunoglobulin A, Immunoglobulin G, Intensive Care Unit, Mortality Risk, MIMIC-IV, Biomarkers, Immune Dysregulation, Sequential Organ Failure Assessment, APACHE II, Bronchoalveolar Lavage Fluid, Flow Cytometry, Pulmonary Exudation, Computed Tomography, Nonpulmonary Sepsis, Observational Cohort Study
News Source: Ophelia Keating. (November 13, 2026). Study Links Higher Serum IgA to Mortality in Specific Sepsis Subgroup. Scienmag.



