Hypomagnesemia as a Risk Factor for the Non-recovery of the Renal Function in Critically Ill Patients With Acute Kidney Injury

Sarah Cascaes Alves; Cristiane Damiani Tomasi; Larissa Constantino; Vinícius Giombelli; Roberta Candal; Maria de Lourdes Bristot; Maria Fernanda Topanotti; Emmanuel A. Burdmann; Felipe Dal-Pizzol; Cassiana Mazon Fraga; Cristiane Ritter

Disclosures

Nephrol Dial Transplant. 2013;28(4):910-916. 

In This Article

Results

A total of 355 patients admitted into the ICU were screened, and 123 of these patients were excluded from the study for the following reasons: 83 patients remained in the ICU for <8 h, 32 patients had ICU admission SCr values ≥3.5 mg/dL and 8 patients were <18 years of age. Thus, 232 patients were included in the study (Figure 1).

Figure 1.

Flowchart of patients in the study.

Table 1 summarizes the general characteristics in patients with and without AKI. Patients with AKI presented more frequently with sepsis (P = 0.007), had a higher APACHE II score (P = 0.007), were admitted due to medical reasons (P = 0.005), and needed vasopressors (P = 0.03). There was a significant difference among the mortality rates of patients with and without AKI (32.2 versus 10.5%, respectively, P < 0.0001). Renal replacement therapy (RRT) was needed in 35 (30%) of AKI patients, and 8 remained on RRT at the end of study. All patients received daily intermittent hemodialysis or extended RRT.

The prevalence of hypomagnesemia was 63% (146 of 232). Ninety-seven of the patients who were admitted into the ICU already presented with hypomagnesemia, with a mean hypomagnesemia duration during ICU stay of 3.1 ± 2.9 days. Among the patients with hypomagnesemia, 75 developed AKI, but 20 of them developed AKI before the occurrence of hypomagnesemia, and were thus excluded in the analyses related to AKI. The prevalence of hypomagnesemia was similar in the groups with and without AKI (47 versus 62%, P = 0.36), and the presence or absence of hypomagnesemia was not associated with AKI (Figure 2A). There was no significant association between serum magnesium levels or the presence of hypomagnesemia and RIFLE stages (P = 0.10 and P = 0.15, data not shown).

Figure 2.

(A) Kaplan–Meier curves for the development of AKI in patients with and without hypomagnesemia. AKI, acute kidney injury. P = 0.15. (B) Kaplan–Meier curves for the recovery of the renal function in patients with and without hypomagnesemia. AKI, acute kidney injury. P = 0.008. (C) Kaplan–Meier curves for the intensive care mortality in patients with and without hypomagnesemia. P = 0.63.

Table 2 summarizes the general characteristics of the patients who recovered and whose did not recover renal function. The prevalence of hypomagnesemia was significantly higher in the patients who did not recover renal function compared with those who recovered the renal function (70 versus 31%, respectively, P = 0.003), and the presence or absence of hypomagnesemia was associated with recovery of renal function (Figure 2B). Neither the lower magnesium level or the mean magnesium levels during ICU stay were significantly different in the patients that recovered and whose did not recover the renal function (Table 2). Using a ROC curve analyses the discriminatory power of the lower magnesium levels and the mean magnesium levels was weak (AUROC = 0.542 and 0.544). Even when hypomagnesemia was stratified as mild, moderate or severe we did not find any significant difference in the renal function recovery between the two magnesium levels groups (Table 2). The RIFLE score correlated significantly with the renal function recovery. Patients in the R category recovered renal function more frequently than those in the I or F categories (P = 0.001). In the multivariate analyses, hypomagnesemia was a significant risk factor for non-recovery of renal function (Table 3). ICU mortality was similar in patients with and without hypomagnesemia (Figure 2C).

processing....