Impact of magnesium on the clearance of lactate in critically sick patients with sepsis: Randomized clinical trial

Abstract

BACKGROUND

Sepsis is associated with impaired tissue perfusion and elevated serum lactate, a key prognostic marker. Magnesium plays a vital role in cellular energy metabolism and may enhance lactate clearance in sepsis.

AIM

To investigate the impact of magnesium on the clearance of lactate in critically sick patients with sepsis.

METHODS

In this triple-blind, randomized trial, 138 adult patients with a diagnosis of sepsis (according to Sepsis-3 guidelines) and with serum lactate levels > 2 mmol/L were randomized into the magnesium (M) group and the placebo (P) group. Randomization was performed using an online sequence generator. The individuals in group M were administered 2 g of magnesium sulphate infusion diluted in 50 mL of normal saline over 2 hours for 3 days. The individuals in group P were administered an identical volume of normal saline for three days. The primary outcome was to compare the time to lactate clearance (20% decrease from baseline) in both groups. The secondary outcomes included changes in lactate clearance, intensive care unit (ICU) duration, hospital stay, and 28-day mortality in both groups.

RESULTS

Time to 20% lactate clearance was considerably lower in the M group (5.73 ± 2.19 hours) when compared to the P group (8.54 ± 2.72 hours), with a mean difference [95% confidence interval (CI)]: -2.81 (-3.65 to -1.99), P < 0.001. Mean rise in lactate clearance was markedly more significant in M group than P group on day 1 [18.88% ± 1.80% vs 12.61% ± 1.94% respectively, with mean difference (95%CI): 6.26 (5.63-6.89), P < 0.001] on day 2 [30.02% ± 4.30 % vs 23.34% ± 4.76% respectively, with mean difference (95%CI): 6.68 (5.15-8.20), P < 0.001] and day 3 [50.74% ± 6.06 % vs 37.94% ± 4.53% respectively, with mean difference (95%CI): 12.80 (11.00-14.60), P < 0.001]. The length of ICU stay was considerably shorter in the M group, 8 (3-15) days, compared to the P group, 12 (5-19) days (P < 0.001). The duration of hospitalization was significantly decreased in the M group, 12 (4-24) days than in the P group, 16 (9-25) days (P < 0.001); however, the 28-day mortality was similar in both groups.

CONCLUSION

Magnesium supplementation was associated with shorter lactate clearance time and reduced ICU and hospital stay duration in patients with sepsis; however, the difference in 28-day mortality between the groups was statistically nonsignificant.

Key Words: Lactate; Magnesium; Sepsis; Intensive care unit; Lactate clearance

Core Tip: Early lactate clearance is a key therapeutic target and prognostic marker in sepsis. In this randomized clinical trial, magnesium supplementation significantly accelerated lactate clearance in critically ill patients with sepsis and was associated with reduced vasopressor requirements, shorter intensive care unit stay, and decreased overall hospital stay. Although the magnesium group demonstrated a numerically lower 28-day mortality, this difference did not reach statistical significance.

INTRODUCTION

Sepsis is an aberrant host reaction to infection, causing a potentially fatal organ failure[1]. Serum lactate is a marker of tissue hypoperfusion that may help forecast the course of sepsis[2,3]. Several therapies, encompassing fluid resuscitation, antibiotic therapy, and vasopressor administration, significantly impact lactate clearance in sepsis[4-6].

Magnesium serves as an essential cofactor in a wide range of physiological processes, including cellular energy metabolism, mitochondrial function[7-9], neuromuscular transmission, regulation of vascular tone, and modulation of immune and inflammatory responses[10-14]. Magnesium can potentially mitigate cellular damage caused by endotoxins by modulating immune system function[13]. In patients suffering from sepsis, magnesium may help avoid insulin resistance, arrhythmia, respiratory failure, muscle weakness, and poor glycemic control[2,14]. Magnesium facilitates essential enzymatic processes in aerobic metabolism and mitochondrial adenosine triphosphate synthesis; a deficiency causes cells to transition toward anaerobic glycolysis, leading to elevated lactate production[14]. By optimizing mitochondrial function, improving microcirculatory perfusion, and diminishing catecholamine-induced metabolic stress, magnesium may contribute to decreased lactate production and promote its clearance[10-14]. There are only a few studies about the effects of magnesium supplementation on sepsis patients. Thus, this research aimed to assess the duration of lactate clearance among critically ill patients with sepsis who received magnesium supplementation.

MATERIALS AND METHODS

Study design

The current placebo-controlled, triple-blind, randomized clinical trial was conducted from November 2022 to March 2024 in patients admitted to the intensive care unit (ICU) with sepsis at a tertiary care institute. Ethical approval for the current research was obtained from the All India Institute of Medical Sciences, Jodhpur, Rajasthan Ethics Committee on September 23, 2022, with approval No. AIIMS/IEC/2022/4177. Subsequently, the study was registered prospectively at Clinical Trial Registry - India (CTRI Reg. No. CTRI/2022/11/047031).

Study variables

Adult patients aged 18 years or more, with a diagnosis of sepsis (according to Sepsis-3 guidelines) and with serum lactate levels > 2 mmol/L, were screened. Patients with chronic renal impairment, liver failure (Child-Pugh B and C), pregnancy, diabetes mellitus on metformin treatment, diabetic ketoacidosis, alcoholism, mean arterial pressure < 65 mmHg despite taking vasopressor therapy, and those refusing consent were excluded.

Methodology

After acquiring informed signed consent, patients were randomly split into two groups: Group magnesium (M) and group placebo (P), with a 1:1 allocation ratio by a randomization method available online (http://www.randomizer.org), and the allocation concealment was performed by a sealed opaque envelope method that was opened just before the enrolment of research participants. Patients were blinded to their allocated group.

Patients assigned to group M were given 2 g of intravenous magnesium sulphate (Maget MB, Martin Brown Bio-Sciences Pvt. Ltd., India), which was diluted in 50 mL normal saline and then infused over 2 hours for three days. The chosen magnesium dose and duration were based on prior clinical trials and safety data in patients admitted to the ICU with sepsis[15]. Patients in group P received 50 mL of normal saline infused over 2 hours period for three days. Both interventions had a similar volume of 50 mL, making the appearance identical. The study infusions were administered by healthcare workers who were independent of the trial team and remained blinded to the group allocation. They were instructed to administer the ‘study drug’, which was prepared in an identical 50 mL syringe without any label indicating whether it contained magnesium sulphate or normal saline. Serum lactate levels were measured for three days using arterial blood gas analysis (Cobas b 221 analyzer; Roche Diagnostics International AG, Rotkreuz, Switzerland).

The primary research outcome was to compare the time to lactate clearance (a 20% decrease from baseline) between the two groups. The secondary outcomes included changes in lactate clearance, vasopressor dose requirement, duration of hospital stay, duration of ICU stay, and 28-day mortality. Lactate clearance was evaluated by the equation [(initial lactate - final lactate)/initial lactate] × 100 (%). Apart from demographic parameters, the Sequential Organ Failure Assessment score, patients’ comorbidities, cause of ICU admission, serum creatinine, serum bilirubin, urine output, and Acute Physiology and Chronic Health Evaluation II score, as well as ventilator requirement, were also noted in both groups. All the data were collected by an independent observer, who was not part of this study. Patient safety was closely monitored throughout the study, including regular assessments of vital signs, hemodynamic parameters, and clinical status, to ensure that any adverse events could be promptly identified and managed. The statistician analysing the data was blinded to the groups.

Sample size

Noormandi et al[15] reported a time to lactate clearance of 47.28 ± 20.59 hours in the M group compared to 61.20 ± 24.31 hours in the P group with an effect size of 0.618 (File 1 in Supplementary material). Using this data for the calculation to detect lactate clearance in the M group compared to the placebo group, we estimated a sample size of 69 per group at a 95% confidence interval, 80% power, and 10% contingency for dropouts (File 2 in Supplementary material).

Statistical analysis

Patient records were evaluated by employing IBM-SPSS software, version 23. The Kolmogorov-Smirnov test was used to assess the normality of the data. For quantitative variables with a normal distribution, results were presented as mean ± SD; for non-normally distributed variables, the median (range) was shown. Categorical data were assessed as absolute n (%). t-test was used for comparing continuous variables, while the χ² test was used for categorical variables. The Mann-Whitney test was used to analyze ordinal variables and quantitative data that did not follow a normal distribution. P < 0.05 was regarded as statistically significant.

RESULTS

Among 280 patients screened for eligibility, 142 were excluded. The relatives of 32 patients denied consent, and 110 patients did not meet the inclusion criteria [alcoholism (n = 39), diabetes mellitus on metformin (n = 28), chronic renal failure (n = 22), liver failure (n = 11), diabetic ketoacidosis (n = 8), and pregnancy (n = 2)]. The remaining 138 patients who fulfilled the inclusion criteria were recruited for randomization and were equally distributed into two groups: Group M (n = 69) and Group P (n = 69) (Figure 1). All randomized patients were included in the primary analysis according to the intention-to-treat principle.

Figure 1  Consort flow diagram.

The demographic, baseline clinical, and laboratory variables were comparable in both groups (Tables 1 and 2). There was no difference in baseline hemodynamic parameters between the two groups. Between the two groups, initial serum lactate values were comparable [2.78 (2.11-6.35) in group M and 2.67 (2.11-7.71) in group P, P = 0.988]. The time to 20% lactate clearance was significantly less in the M group (5.73 ± 2.19 hours) compared to the P group (8.54 ± 2.72 hours), P < 0.001.

Table 1 Comparison of demographic parameters, n (%)/median (range).

Parameter	Group M (n = 69)	Group P (n = 69)	P value
Age (years)	54.00 (18-93)	42.00 (19-84)	0.1221
Gender
Male	44 (63.77)	43 (62.31)	0.8602
Female	25 (36.23)	26 (37.68)
Weight (kg)	68.00 (48-92)	68.00 (48-90)	0.9521
Comorbidities
Hypertension	25 (36.23)	24 (34.78)	0.8592
Diabetes	24 (34.78)	18 (26.09)	0.2672
Chronic lung disease	04 (5.79)	05 (7.24)	0.7302
Cerebrovascular disease	08 (11.59)	05 (7.24)	0.3822
Cause of ICU admission
Surgical	34 (49.28)	32 (46.38)	0.1162
Medical	35 (50.72)	37 (53.62)
Source of sepsis
Lung
CAP	11 (15.94)	13 (18.84)
VAP	10 (14.49)	11 (15.94)
ARDS	5 (7.24)	3 (4.34)
Lung abscess	1 (1.44)	2 (2.89)
Abdomen
Peritonitis	13 (18.84)	11 (15.94)
Urinary tract infection	9 (13.04)	11 (15.94)
Pancreatitis	3 (4.34)	2 (2.89)
Liver abscess	2 (2.89)	1 (1.44)
Blood
Candidemia	2 (2.89)	3 (4.34)
Brain/CNS
Meningitis	3 (4.34)	4 (5.79)
Pots spine	2 (2.89)	1 (1.44)
Encephalitis	2 (2.89)	1 (1.44)
Skin and soft tissue
Soft tissue	2 (2.89)	1 (1.44)
Gluteal abscess	1 (1.44)	2 (2.89)
Necrotizing fasciitis	2 (2.89)	1 (1.44)
Pectoral abscess	1 (1.44)	2 (2.89)

¹Mann-Whitney U test.

²χ² test.

M: Magnesium group; P: Placebo group; ICU: Intensive care unit; CAP: Community-acquired pneumonia; VAP: Ventilator-associated pneumonia; ARDS: Acute respiratory distress syndrome; CNS: Central nervous system.

Table 2 Comparison of clinical and laboratory parameters, n (%)/median (range).

Parameter	Group M (n = 69)	Group P (n = 69)	P value
Serum lactate at diagnosis (mmol/L)	2.78 (2.11-6.35)	2.67 (2.11-7.71)	0.9881
APACHE-II score at ICU admission	19.00 (5 -32)	20.00 (10-32)	0.1921
SOFA score
Day 1	6.00 (2-13)	6.00 (2-13)	0.2651
Day 2	6.00 (2-13)	6.00 (2-13)	0.1361
Day 3	7.00 (2-13)	6.00 (2-12)	0.1311
Serum bilirubin (mg/dL)
Day 1	0.5 (0.3-4.7)	0.5 (0.2-4.7)
Day 2	0.5 (0.2-3.8)	0.5 (0.2-3.8)	0.6381
Day 3	0.5 (0.3-4.7)	0.5 (0.2-4.7)	0.8491
Serum creatinine (mg/dL)
Day 1	0.80 (0.3-1.2)	0.72 (0.3-1.2)	0.5701
Day 2	0.7 (0.4-1.0)	0.7 (0.4-1.0)	0.6841
Day 3	0.8 (0.4-1.2)	0.7 (0.3-1.1)	0.2371
Urine output (mL/day)
Day 1	1430 (820-1980)	1430 (800-2000)	0.3791
Day 2	1350 (820-2000)	1450 (800-1990)	0.0901
Day 3	1430 (930-2000)	1350 (800-1980)	0.8811
Ventilator requirement
Day 1	34 (49.28)	34 (49.28)	1.0002
Day 2	38 (55.07)	42 (60.87)	0.4902
Day 3	30 (43.48)	36 (52.17)	0.3072

¹Mann-Whitney U test.

²χ² test.

M: Magnesium group; P: Placebo group; APACHE: Acute Physiology and Chronic Health Evaluation; ICU: Intensive care unit; SOFA: Sequential Organ Failure Assessment.

Lactate clearance significantly increased in the M group from day 1 to day 3 (Table 3). The mean lactate clearance was 18.88 ± 1.80 vs 12.61 ± 1.94 on day 1, 30.02 ± 4.30 vs 23.34 ± 4.76 on day 2, and 50.74 ± 6.06 vs 37.94 ± 4.53 on day 3 in M and P groups, respectively (P < 0.001). Median vasopressor dosage was much lower in group M than in P group [1.2 (0.6-1.6) vs 1.5 (1.2-1.7) in group M and group P, respectively, on day 2, P = 0.022 and 1.5 (0.9-2.3) vs 1.9 (1.5-2.3) in group M and group P, respectively, on day 3, P = 0.033]. Median length of ICU stay was considerably shorter in the M group, 8 (3-15) days, than in the P group, 12 (5-19) days (P < 0.001). The median duration of hospitalization was significantly decreased in the M group, 12 (4-24) days, compared to the P group, 16 (9-25) days (P < 0.001). During the 28-day follow-up, 39 (28.26%) patients died. Fifteen out of 69 (21.73%) patients in the M group died by day 28, compared to 24 out of 69 patients in the P group, indicating an insignificant difference in 28-day mortality between the two groups (P = 0.12) (Table 3). We did not observe any adverse event in this trial.

Table 3 Comparison of various outcomes, mean ± SD/median (range).

Parameter	Group M (n = 69)	Group P (n = 69)	Mean difference (95%CI)	P value
Time to lactate clearance (hours)	5.73 ± 2.19	8.54 ± 2.72	-2.81 (-3.65 to -1.99)	0.0011
Lactate clearance (%)
Day 1	18.88 ± 1.80	12.61 ± 1.94	6.26 (5.63-6.89)	0.0011
Day 2	30.02 ± 4.30	23.34 ± 4.76	6.68 (5.15-8.20)	0.0011
Day 3	50.74 ± 6.06	37.94 ± 4.53	12.80 (11.00-14.60)	0.0011
Vasopressor dose (mcg/kg/minute)
Day 1	0.6 (0.1-1.2)	0.7 (0.3-0.9)		0.0742
Day 2	1.2 (0.6-1.6)	1.5 (1.2-1.7)		0.0222
Day 3	1.5 (0.9-2.3)	1.9 (1.5-2.3)		0.0332
Duration of ICU stay (days)	8 (3-15)	12 (5-19)		0.0012
Duration of hospital stay (days)	12 (4-24)	16 (9-25)		0.0012
28-day mortality; n (%)	15 (21.73)	24 (34.78)		0.123

¹Independent sample t-test.

²Mann-Whitney U test.

³χ² test.

M: Magnesium group; P: Placebo group; CI: Confidence interval; ICU: Intensive care unit.

DISCUSSION

In the current investigation, we observed that critically ill patients with sepsis who received magnesium supplementation exhibited improved lactate clearance, reduced vasopressor dose requirements, and shorter ICU and hospital stays. The M group showed a lower 28-day mortality rate than the placebo group, although this difference was not statistically significant.

In research conducted by Noormandi et al[15] in patients with severe sepsis, the time to lactate clearance (a 50% reduction from baseline) was found to be 47.28 ± 20.59 hours in the M group and 61.20 ± 24.31 hours in the P group. The findings from our study are comparable to those of Noormandi et al[15], demonstrating the effectiveness of magnesium supplementation in improving time to lactate clearance in individuals diagnosed with sepsis. In our research, we also found a rise in lactate clearance in the M group compared to the P group on days 1, 2, and 3. Similarly, Noormandi et al[15], in their study, found that the increase in lactate clearance on day 2 was (27.53 vs 23.79) and on day three was (49.83 vs 37.02) on the M and P groups, respectively.

In the current research, the vasopressor dosage was significantly lower in the M group on days 2 and 3. Similarly, in a retrospective cohort study, Lv et al[16] found that patients with sepsis who received magnesium had a median vasopressor-free time [72.4 (40.8, 171.6) hours] in contrast to individuals who did not receive magnesium [51.6 (30.8, 94) hours] Magnesium’s potential to influence vascular tone and improve microcirculatory perfusion may contribute to a reduced need for vasopressors in these patients.

Current research indicates that the median duration of ICU stay was shorter for the M group compared to the placebo group. Similarly, Noormandi et al[15] observed that the duration of ICU stay was shorter in the M group than in the P group. Assarian et al[17] discovered that the length of ICU admissions in the hypomagnesemia group compared to the normomagnesemia group was 14 (12.75-17.25) vs 10 (8.25-14.25) days, respectively. The median hours of ICU stay in research performed by Lv et al[16] was 145 (71, 296) hours vs 106 (57.5, 228) hours in the hypomagnesemia group and normomagnesemia group. Patil and Aslam[18] observed that the mean length of ICU stay was longer in patients with hypomagnesemia, 7.21 ± 1.74 days, compared to 5.24 ± 1.38 days in normomagnesemic patients. Comparing the findings of our study with the above studies, it was found that hypomagnesemia patients had a longer ICU stay duration, and hence, magnesium supplementation may significantly reduce the duration of ICU stay in patients having sepsis.

Our research revealed that the length of stay in the hospital for the M group was less than that for the placebo group. Islam et al[19] noticed that the hospital stay duration was more prolonged, at 14.94 ± 7.78 days in the hypomagnesemia group, compared with 10.47 ± 7.60 days in the normomagnesemia group, hence, magnesium supplementation in patients with sepsis may significantly reduce the length of hospital stay.

In our study, the M group exhibited a lower 28-day mortality rate compared to the placebo group, although this difference did not reach statistical significance. Similarly, Noormandi et al[15] in their investigation of patients with severe sepsis found that sepsis-associated 28-day mortality in the M group was lower than in the placebo group; however, the difference was not statistically significant (10.71% vs 21.42%, respectively, P = 0.65) In a study of the association of magnesium sulphate use with mortality, Gu et al[20] found that the 28-day all-cause mortality was 20.2% in the M group and 25.0% in the non- M group. Lv et al[16] analyzed the effect of magnesium supplementation on 14-, 28-, and 60-day mortality and showed a significant reduction in mortality in the M group (30.8%, 33.8%, 35.5%, respectively) compared to the P group (48.9%, 48.9%, 48.9%, respectively). In the present study, the difference in mortality was not statistically significant, likely because, as a single-centre study, the sample size may be insufficient to demonstrate a substantial survival effect.

Limitations

This research is the first to evaluate the time to 20% lactate clearance in individuals diagnosed with sepsis after magnesium supplementation. No other study compared the vasopressor dose requirements in individuals diagnosed with sepsis with or without magnesium supplementation. However, this study had certain limitations. First, it was conducted at a single centre, which may limit the external validity and restrict the applicability of the findings to other healthcare settings. Second, serum magnesium levels were not measured, which prevented an assessment of baseline magnesium status and limited the interpretation of treatment effects. Finally, data on renal replacement therapy were not collected, which may influence lactate kinetics and could affect the accuracy of outcome comparisons.

CONCLUSION

The present study focuses on the influence of the addition of magnesium on lactate clearance among critically ill patients with sepsis. Magnesium supplementation accelerated lactate clearance time and reduced vasopressor dose requirements, as well as the duration of ICU and hospital stays. However, the 28-day mortality was not statistically significant in either group.