Hyperphosphatemia and Outcomes in Critically Ill Patients: A Systematic Review and Meta-Analysis

Abstract

Introduction: Serum phosphate level is often deranged during critical illness. Hyperphosphatemia, as a marker of disease severity, attracts more and more attention. This study aimed to evaluate the impact of hyperphosphatemia on clinical outcomes in critically ill patients. Methods: We searched for relevant studies in PubMed, EMBASE, and the Cochrane database up to Jan 10, 2022. Two authors independently screened studies, extracted data, and assessed the study quality. Meta-analyses were performed to determine hyperphosphatemia prevalence and evaluate its relationship with prognosis and important clinical outcomes. We also conducted subgroup analysis and sensitivity analyses to explore the sources of heterogeneity. Results: Ten studies with 60,358 patients met the inclusion criteria. These studies were moderate to high quality. The median prevalence of hyperphosphatemia was 30% (range from 5.6 to 45%). Patients with hyperphosphatemia had a significantly higher risk of all-cause mortality than those without (OR 2.85; 95% CI, 2.35 to 3.38, P < 0.0001). Subgroup analyses, sensitivity analyses, and regression analyses further confirmed these results. In addition, patients with hyperphosphatemia required more CRRT (OR 4.96; 95% CI, 2.43 to 10.2, P < 0.0001) but not significantly increased duration of mechanical ventilation (mean difference, MD 0.13, 95% CI −0.04 to 0.30; P = 0.138), length of stay in intensive care unit (ICU) (SMD 0.164 day, 95% CI −0.007 to 0.335; P = 0.06), and length of stay in hospital (SMD 0.005 day, 95% CI −0.74 to 0.75; P = 0.99). Conclusions: Our results indicated that hyperphosphatemia was associated with all-cause mortality in critically ill patients. However, due to the retrospective design of the included studies, more prospective, well-designed research is required in the future. Systematic Review Registration: [https://doi.org/10.37766/inplasy2021.12.0130], identifier [INPLASY2021120130].