Impact of Hypophosphatemia on Outcome of Patients in Intensive Care Unit

A Retrospective Cohort Study

Lichun Wang; Chaoxing Xiao; Lei Chen; Xiaofei Zhang; Qiuye Kou


BMC Anesthesiol. 2019;19(86) 

In This Article

Abstract and Introduction


Background: Hypophosphatemia generally occurs in Intensive Care Units (ICUs), but its impact is often ignored. The aim of this study was to investigate whether hypophosphatemia can be a risk factor for ICU 28-day mortality.

Methods: A single-center retrospective cohort study was conducted by collecting data from 1073 patients admitted to general ICU and then presented to the Sixth Affiliated Hospital, Sun Yat-sen University (Guangzhou City, Guangdong Province, China) from 1 January 2016 to 31 December 2017. The patients were divided into a normal control group (serum phosphate levels 0.80–1.60 mmol/L) and a hypophosphatemia group (serum phosphate levels < 0.80 mmol/L), based on the concentration of phosphorus at the time of ICU admission. The association between phosphate levels and ICU 28-day mortality was evaluated by binary logistic regression analysis. Multivariate logistic regression was employed to predict the ICU 28-day mortality.

Results: The cohort included 946 patients with a median phosphate concentration of 0.77 mmol/L (interquartile range 0.55–1.03 mmol/L). Patients with hypophosphatemia had a higher ICU 28-day mortality than the normal control group (33.3% vs 24.0%, P < 0.05). Patients with hypophosphatemia had a longer ICU and hospital stays, and prolonged duration of mechanical ventilation (all P < 0.05). Hypophosphatemia was an independent risk factor for ICU 28-day mortality (adjusted OR = 1.5, 95% CI = 1.1–2.1, P = 0.01) in the multivariate logistic regression analysis.

Conclusions: Hypophosphatemia at admission is an independent risk factor for 28-day mortality in general ICU patients.

Trial registration: The medical study was approved by the Institutional Ethics Committee of the Six Affiliated Hospital, Sun Yat-sen University (Approval number: 2017ZSLYEC-110). No consent was given as the data were analyzed anonymously.