Does hypoxemia aggravate sepsis-associated acute kidney injury? Integrated clinical and experimental evidence.

BACKGROUND: Sepsis-associated acute kidney injury (SA-AKI) is a common and severe complication in critically ill patients, but the association between hypoxemia and renal dysfunction remains uncertain. METHOD: We retrospectively analyzed 2292 patients with SA-AKI from the MIMIC-IV database and stratified them into four groups based on their highest arterial partial pressure of oxygen (PO₂) within 24 h of admission: < 60 mmHg, ≥ 60 to < 80 mmHg, ≥ 80 to  < 100 mmHg, and ≥ 100 mmHg. Associations between PO₂ and renal injury markers (serum creatinine [SCr] and blood urea nitrogen [BUN]) were evaluated using multivariable regression analyses, and survival outcomes were compared with Kaplan-Meier methods. To explore mechanistic pathways, a murine model was established with four experimental conditions: normoxia, hypoxia (10% O₂), lipopolysaccharide (LPS)-induced sepsis, and combined sepsis plus hypoxia. Serum biochemical parameters, histological injury, and protein expression of hypoxia-inducible factor-1α (HIF-1α) were measured at 6, 24, and 48 h. Mitochondrial autophagy was assessed by LC3 and TOMM20 immunofluorescence colocalization. RESULT: Patients with lower PO₂ had higher illness severity and unadjusted BUN and SCr levels, multivariable analyses revealed no independent association between PO₂ and renal injury markers. Survival differed significantly across groups, with the ≥ 100 mmHg group showing the best outcomes (log-rank P < 0.001). In animal experiments, sepsis groups developed increased SCr and BUN at 24 and 48 h, but combined hypoxia did not exacerbate these parameters compared to sepsis alone. Histological analysis revealed severe tubular injury with no significant aggravation in the sepsis-plus hypoxia group. HIF-1α expression was lowest in sepsis-only kidneys but markedly upregulated in the sepsis-plus-hypoxia group at 6 h. Immunofluorescence demonstrated less colocalization of LC3 and TOMM20 in the sepsis-only group than in sepsis-plus-hypoxia mice, suggesting more efficient mitophagy with hypoxemia. CONCLUSIONS: These clinical and experimental findings indicate that hypoxemia was not independently associated with aggravated renal injury in SA-AKI and may activate HIF-1α and promote adaptive mitophagy. This challenges the conventionally held belief that hypoxemia is uniformly detrimental to renal function during sepsis.