Scholarlink

Background <p>Acute kidney injury (AKI) and delirium are common complications in critically ill patients, both associated with adverse outcomes. Evidence suggests AKI may induce neuroinflammation and impair clearance of neurotoxic metabolites, but the clinical relationship between AKI and delirium remains incompletely characterized. We hypothesized that AKI increases the risk of delirium in a biological gradient.</p> Methods <p>We conducted a retrospective cohort study using two independent databases: MIMIC-IV (<i>n</i> = 15,219 patients) and a local institutional database (Reality) from University Hospital Münster (<i>n</i> = 3,461 patients). Adult ICU patients with length of stay &gt; 12&#xa0;h and validated delirium monitoring using the Confusion Assessment Method for the ICU (CAM-ICU) were included, with a positive assessment defining delirium presence. Patients with neurological/neurosurgical primary diagnoses or pre-existing cognitive impairment, as well as patient with delirium onset before AKI were excluded. AKI was classified according to KDIGO criteria. We employed landmark analysis at 24&#xa0;h post-ICU admission where AKI-status was recorded as the exposure variable. Subsequent delirium was recorded as follow-up during the ICU stay. Propensity score matching, and a Fine-Gray Model were added to assess the time-based association between AKI and subsequent delirium, independently of illness severity.</p> Results <p>At the 24-hour landmark, AKI was present in 58.6% (MIMIC-IV) and 55.6% (Reality) of patients. Rates of subsequent delirium were significantly higher in patients with AKI compared to those without: 25.6% versus 15.7% in MIMIC-IV (OR 1.84, CI 1.68–2.04, <i>p</i> &lt; 0.001) and 12.1% versus 6.7% in Reality (OR 1.95, CI 1.44–2.64, <i>p</i> &lt; 0.001). After adjustment for confounders, AKI remained associated with delirium (adjusted-odds ratio [OR] 1.16, 95% CI 1.10–1.23 in MIMIC-IV; OR 1.20, 95% CI 1.01–1.41 in Reality). A biological gradient was observed, with delirium rates increasing progressively across AKI stages. In MIMIC-IV, adjusted-OR increased from 1.42 for Stage 1 to 4.11 for Stage 3 AKI (trend OR 1.55 per stage, <i>p</i> &lt; 0.001). The Reality cohort showed similar patterns (Stage 1 OR 1.01 and Stage 3 OR 1.92, trend OR 1.25 per stage, <i>p</i> &lt; 0.01). Propensity score matching confirmed these findings.</p> Conclusion <p>Acute kidney injury was associated with a higher risk of subsequent delirium in critically ill patients, with a graded association across AKI severity stages. Prospective studies are needed to determine whether interventions targeting AKI prevention or treatment can also reduce delirium burden.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Association between acute kidney injury and delirium in critically ill patients: a retrospective cohort study using two independent databases

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月5日

查看全文
  • Christian Porschen,
  • Christian Strauß,
  • Sean M. Bagshaw,
  • John A. Kellum,
  • Sven G. Meuth,
  • Paul Brauckmann,
  • Ngoc-Anh Thi Nguyen,
  • Michael Fujarski,
  • Keyvan Mahjoory,
  • Carla Barbara Schwienhorst,
  • Alexander Zarbock

摘要

Background

Acute kidney injury (AKI) and delirium are common complications in critically ill patients, both associated with adverse outcomes. Evidence suggests AKI may induce neuroinflammation and impair clearance of neurotoxic metabolites, but the clinical relationship between AKI and delirium remains incompletely characterized. We hypothesized that AKI increases the risk of delirium in a biological gradient.

Methods

We conducted a retrospective cohort study using two independent databases: MIMIC-IV (n = 15,219 patients) and a local institutional database (Reality) from University Hospital Münster (n = 3,461 patients). Adult ICU patients with length of stay > 12 h and validated delirium monitoring using the Confusion Assessment Method for the ICU (CAM-ICU) were included, with a positive assessment defining delirium presence. Patients with neurological/neurosurgical primary diagnoses or pre-existing cognitive impairment, as well as patient with delirium onset before AKI were excluded. AKI was classified according to KDIGO criteria. We employed landmark analysis at 24 h post-ICU admission where AKI-status was recorded as the exposure variable. Subsequent delirium was recorded as follow-up during the ICU stay. Propensity score matching, and a Fine-Gray Model were added to assess the time-based association between AKI and subsequent delirium, independently of illness severity.

Results

At the 24-hour landmark, AKI was present in 58.6% (MIMIC-IV) and 55.6% (Reality) of patients. Rates of subsequent delirium were significantly higher in patients with AKI compared to those without: 25.6% versus 15.7% in MIMIC-IV (OR 1.84, CI 1.68–2.04, p < 0.001) and 12.1% versus 6.7% in Reality (OR 1.95, CI 1.44–2.64, p < 0.001). After adjustment for confounders, AKI remained associated with delirium (adjusted-odds ratio [OR] 1.16, 95% CI 1.10–1.23 in MIMIC-IV; OR 1.20, 95% CI 1.01–1.41 in Reality). A biological gradient was observed, with delirium rates increasing progressively across AKI stages. In MIMIC-IV, adjusted-OR increased from 1.42 for Stage 1 to 4.11 for Stage 3 AKI (trend OR 1.55 per stage, p < 0.001). The Reality cohort showed similar patterns (Stage 1 OR 1.01 and Stage 3 OR 1.92, trend OR 1.25 per stage, p < 0.01). Propensity score matching confirmed these findings.

Conclusion

Acute kidney injury was associated with a higher risk of subsequent delirium in critically ill patients, with a graded association across AKI severity stages. Prospective studies are needed to determine whether interventions targeting AKI prevention or treatment can also reduce delirium burden.