Association of statin use with 28-day mortality in the Medical Information Mart for Intensive Care IV database: a retrospective cohort study

Introduction

The hydroxymethylglutaryl-coenzyme A reductase inhibitors (HMG-CoA reductase inhibitors), commonly referred to as statins, have traditionally been used for the purpose of reducing blood cholesterol levels or preventing cardiovascular events in patients diagnosed with cardiovascular disease (1). Approximately 200 million patients are treated with this drug globally. These agents impede the activity of the pivotal enzyme in cholesterol synthesis, thereby diminishing cholesterol production. Simultaneously, they enhance the expression of cell surface low-density lipoprotein (LDL) receptors, accelerating the catabolism of serum LDL. Consequently, statins markedly lower levels of serum total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and apolipoprotein B (ApoB); moderately decrease serum triglyceride (TG) levels; and elevate high-density lipoprotein cholesterol (HDL-C) levels (2). Beyond lipid-lowering effects, statins possess therapeutic attributes unrelated to lipid modulation. These multifaceted effects on the vascular endothelium include immunomodulatory, anti-inflammatory, and antithrombotic activity (3-5). These non-lipid-related benefits, termed the “pleiotropic effect”, may enhance the prognosis of patients across various diseases (6), even extending potential benefit to patients with noncardiovascular conditions such as cancer, chronic obstructive pulmonary disease (COPD), and rheumatoid arthritis (7). The immunomodulatory, anti-inflammatory, and antioxidant effects of statins have garnered significant attention in recent years, leading to their investigation as a host-directed therapy for the treatment of infectious diseases (8).

Concerning anti-inflammatory activity, the growing incidence of sepsis and related fatalities necessitates the further development of therapeutic interventions. In animal models of sepsis (9,10), statins demonstrated inhibitory effects on inflammatory mediators, potentially by disrupting isoprenoid intermediates in the mevalonate pathway. This interference allows for the upregulation of nuclear factor-κB (NF-κB) inhibitors, inhibiting NF-κB-dependent inflammation. Statins also mitigate the cellular damage associated with endotoxins and exotoxins, reverse renal tubular hypoxia, and directly impact the renal vascular system, potentially ameliorating the acute kidney injury (AKI) induced by sepsis (11,12).

Fungi are a near ubiquitous class of eukaryotic organisms, and the prevalence of fungal infections has exhibited a substantial rise with the administration of immunosuppressive agents, radiotherapy for neoplasms, in vivo tube placements, and extended stays in intensive care units (ICUs). Statins have demonstrated antifungal efficacy and can synergistically enhance the effectiveness of antifungal medications (13). The possible pathway mechanisms for this effect involve, first, the inhibition of farnesyl pyrophosphate (FPP) production in the mevalonate pathway by statins, which impedes ergosterol biosynthesis, a crucial component in fungal cell membranes (14). Second, statins may suppress the isoprenylation pathway, hindering the activation of vital cellular proteins involved in cellular respiration and iron metabolism (15,16). Third, statins may induce fungal mitochondrial dysfunction and apoptosis. Finally, the statin-induced blockade of HMG-CoA reductase can dampen fungal spore germination (17). These findings were not yet applied in clinical medicine.

As it relates to antithrombotic activity, venous thromboembolism, including deep vein thrombosis and pulmonary embolism, represents a frequent complication in critically ill patients, correlating with substantial morbidity and mortality. Cluster of differentiation 36 (CD36), a single-chain glycoprotein expressed on various cell surfaces, binds to platelet membrane CD36, initiating endogenous reactive oxygen species production and causing platelet activation and thrombosis (18). Statins can exert inhibitory effects on platelet activation and aggregation either through direct interaction with CD36 or by modulating the intracellular signaling pathway of CD36, among other pathways. Obviously, platelets are not the only player in pulmonary embolism as the classical pathway of hemostasis is playing a crucial role as well.

Patients in the ICU have a high incidence of complications such as cardiovascular diseases, severe infection, and thrombosis, and almost all patients in the ICU experience systemic inflammatory response syndromes, leading to a high mortality rate (19). Investigating the potentially advantageous pleiotropic effects of statins in the ICU setting could potentially yield substantial benefits for clinical practice and public health. We employed a population subset from a comprehensive public critical care database to examine the correlation between statin usage and 28-day all-cause mortality among adult patients in the ICU. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2243/rc).

Aim

The aim of this study was to determine the relationship between statin therapy and all-cause mortality in critically ill patients and to compare the effects of different types of statins on mortality.

Methods

Data source

This investigation employed a retrospective cohort design using data derived from the MIMIC-IV database (version 1.0) (20). MIMIC-IV is a longitudinal, single-center database encompassing comprehensive information on patients admitted to the ICU at Beth Israel Deaconess Medical Center between 2008 and 2019. The database includes demographic details, vital signs, laboratory results, diagnostic particulars, treatment records, and instances of mortality during hospitalization. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). We (C.K.; ID: 55395496) signed a data use agreement form and completed the Collaborative Institutional Training Initiative (CITI) exam to be authorized to access and use the database and perform data extraction and analysis.

Study population

Study participants comprised individuals from the database meeting the following criteria: (I) initial admission to the ICU; (II) aged ≥18 years; and (III) patients without missing data. Participants in the statin group were required to receive statin treatment for a minimum of 3 days after ICU admission. In our study, we divided patients into statin and non-statin groups based on inclusion and exclusion criteria, and incorporated selected covariates into propensity score matching (PSM) to ensure comparability between the two groups.

Outcome indicator

The primary focus was the examination of the association between statin use and 28-day mortality. Additionally, the associations of different statin types with patient mortality was ascertained. Mortality within 28 days of admission was determined using the date of death (DOS), time of admission (INTIME), and time of discharge (OUTTIME).

Study variables

All data were extracted from the relevant tables in the data management library pgAdmin 4.26 using Structured Query Language (SQL) language. The variables subjected to analysis included age, gender, race, type of insurance, aspirin usage, the severity of condition score, comorbidities (e.g., heart failure, kidney disease, severe liver disease, myocardial infarction, acute renal failure, COPD), and vital signs (systolic blood pressure, diastolic blood pressure, heart rate, respiratory rate). Variables such as height and body mass index were excluded due to the presence of missing data in over 50% of cases.

Statistical analysis

Multivariate Cox regression analysis was employed to evaluate the correlation between statin use and 28-day all-cause mortality subsequent to admission to the ICU. An extended Cox modeling approach was implemented to account for various covariates in the model. Kaplan-Meier and log-rank analyses were used to generate survival curves. Subgroup analyses were conducted with stratification based on the pertinent covariates.

The descriptive analysis involved all patients eligible for enrollment. Categorical variables are presented as percentages (%), while continuous variables are expressed as the mean and standard deviation (SD) or as the median and interquartile range (IQR), as appropriate. Variable comparisons were executed using the Chi-squared test for categorical variables, the t-test for normally distributed variables, and the Mann-Whitney test for skewed variables. Patients with a substantial number of missing covariates were systematically excluded, along with 296 patients with missing values, illogical entries, or exit times. In the context of observational studies, the adjustment for intergroup individual differences through propensity scores renders all confounding variables, excluding exposure/treatment factors and outcome variables, comparably equivalent. This is tantamount to a post hoc randomization approach, aligning observational study data with an approximation of random assignment. PSM was gauged using standardized mean differences (SMDs), with a threshold of <0.1 considered acceptable.

All statistical analyses were carried out using R version 3.3.2 (The R Foundation of Statistical Computing) and FreeStatistics software version 1.8. A two-tailed test was employed, and statistical significance was defined as P<0.05. FreeStatistics is a software package that offers an intuitive interface for common data analyses and visualizations, with R as the underlying statistical engine and its graphical user interface scripted in Python, and is designed to facilitate reproducible analyses and interactive calculations.

Results

Population

A total of 76,540 records documenting ICU hospitalizations were retrieved from the MIMIC-IV database. From this dataset, 53,150 patient records pertaining to the initial ICU admission were meticulously selected. Following the exclusion of patients due to age or incomplete data, a final cohort of 50,624 individuals was formed, comprising 15,630 statin users and 34,994 nonusers. PSM analysis was executed, as illustrated in Figure 1.

Figure 1 The flowchart of the cohort selection process. ICU, intensive care unit; MIMIC-IV, Medical Information Mart for Intensive Care IV.

Baseline characteristics

Comparative analysis of baseline characteristics revealed that individuals in the statin group exhibited an advanced age and a higher proportion of males compared to their non-statin-treated counterparts (62.0% vs. 53.1%); the statin group also had a higher percentage of Medicare recipients (50.2% vs. 39.5%) and elevated rates of concurrent aspirin usage (70.8% vs. 18.2%). In terms of comorbidities, the statin group had higher incidences of congestive heart failure (CHF), AKI, myocardial infarction, and COPD. Notably, the prevalence of severe liver disease was lower in the statin group than in the nonstatin (0.9% vs. 5.9%), underscoring hepatotoxicity as a principal adverse effect of statins (see Table 1). After PSM, a significant reduction in SMDs between the two groups was observed. Prior to PSM adjustment, most covariates exhibited imbalances between the statin group and nonstatin group; after PSM, the SMD for all covariates was <0.10, indicative of balanced inclusion of covariates between the two groups (Figure S1).

Univariate analysis

Following PSM, univariate analysis was conducted on patients to clarify the association between covariates and 28-day mortality in the ICU. The statin group exhibited lower mortality compared to the nonstatin group, with reductions of 29% for atorvastatin, 48% for rosuvastatin, 44% for simvastatin, and 28% for other statins. Patients aged 65 years or older had a mortality rate 2.06 times higher than their counterparts younger than 65 years. Individuals using aspirin demonstrated a 31% lower risk of death compared to those not using aspirin. Additionally, an increase in mortality of 4% per 1-point increase in Acute Physiology Score III (APSIII) score, a 25% increase per 1-point rise in the Charlson comorbidity index (CCI) score, and elevated mortality in patients with comorbidities compared to those without comorbidities were observed (Table 2).

Multivariate analysis

After PSM, 6,213 (58.36%) in the atorvastatin group, 730 (6.86%) in the rosuvastatin group, 2,817 (26.46%) in the simvastatin group, and 886 (8.32%) in the other statin group. After PSM and adjustment for all covariates to mitigate confounding effects, patients in the statin group exhibited an overall 31% lower risk of death compared to the nonstatin group [hazard ratio (HR) =0.69; 95% confidence interval (CI): 0.63–0.75; P<0.001]. As it relates to different statin types, compared to the nonstatin group, atorvastatin demonstrated a 26% lower risk (HR =0.74; 95% CI: 0.67–0.82; P<0.001), rosuvastatin a 34% lower risk (HR =0.66; 95% CI: 0.5–0.88; P=0.004), simvastatin a 44% lower risk (HR =0.56; 95% CI: 0.48–0.65; P<0.001), and other statins a 26% lower risk (HR =0.74; 95% CI: 0.59–0.92; P=0.008) (Table 3). Please refer to Table S1 for the pre-PSM results.

Survival analysis

The Kaplan-Meier survival curves indicated that the 28-day survival probability of the statin group surpassed that of the nonstatin group. Before PSM, the cohort comprised 50,624 patients, with 15,630 in the statin group and 34,994 in the nonstatin group, and the respective survival probabilities were 92.09% and 85.05%. Stratification by specific statins indicated survival rates of 94.08% for rosuvastatin, 93.32% for simvastatin, 91.66% for other statins, and 91.43% for atorvastatin. Following PSM, a total of 21,292 patients were included, who were evenly distributed between the statin and nonstatin groups (10,646 each). The corresponding survival probabilities were 91.51% and 87.42%. Stratified analysis based on individual statins indicated survival rates of 93.15% for rosuvastatin, 92.76% for simvastatin, 90.74% for other statins, and 90.86% for atorvastatin (Figure 2).

Figure 2 Kaplan-Meier survival curves for patients with and without statin administration. PSM, propensity score matching.

Subgroup analysis

Within the various subgroups, including patients diagnosed with COPD, AKI, sepsis, and myocardial infarction, the use of statins exhibited an association with a diminished 28-day mortality rate. Conversely, in the context of severe liver disease, statin use did not demonstrate a reduction in 28-day mortality and may notably pose a risk factor for mortality in patients with severe liver disease. There was no difference between the sexes regarding the association between statin use and 28-day mortality (P>0.05) (Figure 3).

Figure 3 Subgroup analysis of statin use and 28-day mortality. HR, hazard ratio; CI, confidence interval; Ref, reference; CHF, congestive heart failure; AKI, acute kidney injury; MI, myocardial infarction; COPD, chronic obstructive pulmonary disease.

Sensitivity analysis

We generate a PSM among the statin and non-statin group using a multivariable regression model. A 1:1 nearest neighbor matching algorithm was applied using a caliper width of 0.01. An SMD was used to examine the PSM degree. Furthermore, the variables mentioned above as covariates were selected to generate the propensity score. The estimated propensity scores were used as weights. Pairwise algorithmic (PA), standardized mortality ratio weight (SMRW) models were used to generate a weighted cohort to adjust the baseline confounders, thus reflecting more truly the independent association between statin use and mortality.

The E value (21) of this study was 2.26, signifying that in the presence of an unmeasured covariate, the association with 28-day mortality was required to surpass a threshold of at least 2.26 to exert an impact on the robustness of the findings.

Discussion

Statement of key findings

Using the MIMC-IV database, we sought to clarify the potential link between statin usage and 28-day mortality among patients in the ICU. The comprehensive retrospective cohort analysis revealed a significant association between statin therapy and a reduced risk of 28-day all-cause mortality in patients in the ICU. The robustness of these results persisted even after PSM, multivariate analysis, and subgroup analyses. Notably, the observed association was most pronounced in the simvastatin group.

Interpretation

Studies have presented conflicting findings regarding the impact of statins on mortality in critically ill patients. Christensen et al. (22) conducted a retrospective analysis involving over 10,000 critically ill patients and found that preadmission statin use was associated with a reduced risk of death after intensive care. Similarly, Kyu Oh et al. (23) examined more than 20,000 critically ill adults, revealing that preadmission statin use, particularly with rosuvastatin, was linked to lower 90-day mortality, even in cases without cardiovascular disease. The study conducted by Yao et al. used the MIMIC-IV database to examine the correlation between statin usage and the risk of in-hospital mortality among patients with sepsis-induced coagulation disorder. The findings suggested a significant association between statin administration and a decreased likelihood of in-hospital death in individuals with sepsis-induced coagulopathy (SIC) (24). The administration of statins prior to admission in patients with acute respiratory distress syndrome (ARDS) has been associated with potential clinical benefits. For instance, pre-ICU administration of statins was found to produce a significant reduction in both 30- and 90-day mortality rates, as well as a decrease in ICU length of stay among patients with ARDS; these findings support the efficacy of statin therapy for the treatment of ARDS (25). Recent cell culture experiments have found that simvastatin can inhibit the growth of pancreatic neuroendocrine tumors both in vivo and in vitro. These findings highlight simvastatin as a promising candidate drug for the treatment of neuroendocrine tumors (26). Wang et al. provided evidence for the potential protective effect of simvastatin against gastric cancer through Mendelian randomization studies. Considering the use of simvastatin as an adjuvant to traditional cancer treatments may propose a new strategy to improve patient outcomes (27). In contrast, Papazian et al. (28) completed a multicenter randomized controlled trial (RCT) in 26 French ICUs and reported that adjunctive simvastatin therapy did not enhance 28-day survival in adults with suspected ventilator-associated pneumonia (VAP). The authors speculated that this discrepancy could be due to variations in statin toxicity and renal elimination between critically ill patients and healthy individuals and emphasized the notably high peak plasma statin levels in the former. Our study diverges from previous meta-analyses that reported no mortality benefit from statin therapy in patients with sepsis or acute lung injury/ARDS (29,30). Notably, we focused on a diverse ICU population using a large public database, including patients admitted with at least 3 days of statin use irrespective of their preadmission statin history. This approach allowed us to investigate the potential pleiotropic effect of statins on all-cause mortality. It is important to note that previous randomized clinical trials often targeted specific conditions such as acute lung injury/ARDS, VAP, or sepsis. Mixed ICU populations, as in our study, may generally have healthier individuals or encounter more self-limiting diseases. This divergence in patient characteristics could explain the differences in outcomes between our study and previous research.

Strengths and weaknesses

Our study involved several limitations that warrant further consideration. First, we employed a retrospective, observational design using pre-existing data and did not conduct randomization. Second, due to an incomplete long-term follow-up in the database, a comprehensive long-term effect analysis was not feasible, and thus our focus was on 28-day mortality. Third, there were limitations related to the population selection, specifically for previously healthy patients admitted to the hospital for accidental injury. In cases where there were no comorbidities, such as infection during hospitalization, and no indications for statin use, the mortality rate of patients in the nonstatin group might have been inflated solely due to trauma-related deaths, introducing bias into the results. Nevertheless, we implemented rigorous methods to address both known and unknown confounding factors. Techniques such as PSM, multifactor analysis, and calculation of the E value were employed, yielding relatively robust research outcomes. The results of our subgroup analysis point to the potential benefits of statin use in clinical practice for patients with sepsis, AKI, myocardial infarction, or COPD. This information can serve as a valuable reference for clinicians in optimizing treatment strategies.

Further research

Future large-scale, multicenter, RCTs are needed to verify the potential pleiotropic effects of statins.

Conclusions

Statin use was potentially associated with a reduction in the 28-day all-cause mortality among patients in the ICU, and this association was more pronounced in the simvastatin group.

Acknowledgments

We would like to thank the staff of the Department of Pharmacy, The Second Hospital of Hebei Medical University, for their guidance and help. We would also like to express our gratitude to the team of clinical scientists for their helpful review and comments regarding the manuscript.

Funding: None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2243/rc

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2243/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2024-2243/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Kulkarni S, Watts MM, Kostapanos M. Statins. BMJ 2024;384:e072584. [Crossref] [PubMed]
2. Joint Committee on the Chinese Guidelines for Lipid Management. Chinese guidelines for lipid management (2023). Zhonghua Xin Xue Guan Bing Za Zhi 2023;51:221-55. [Crossref] [PubMed]
3. Mach F. Statins as immunomodulatory agents. Circulation 2004;109:II15-7. [Crossref] [PubMed]
4. Viasus D, Garcia-Vidal C, Simonetti AF, et al. The effect of simvastatin on inflammatory cytokines in community-acquired pneumonia: a randomised, double-blind, placebo-controlled trial. BMJ Open 2015;5:e006251. [Crossref] [PubMed]
5. Glynn RJ, Danielson E, Fonseca FA, et al. A randomized trial of rosuvastatin in the prevention of venous thromboembolism. N Engl J Med 2009;360:1851-61. [Crossref] [PubMed]
6. Mihos CG, Salas MJ, Santana O. The pleiotropic effects of the hydroxy-methyl-glutaryl-CoA reductase inhibitors in cardiovascular disease: a comprehensive review. Cardiol Rev 2010;18:298-304. [Crossref] [PubMed]
7. He Y, Li X, Gasevic D, et al. Statins and Multiple Noncardiovascular Outcomes: Umbrella Review of Meta-analyses of Observational Studies and Randomized Controlled Trials. Ann Intern Med 2018;169:543-53. [Crossref] [PubMed]
8. Zhang K, Liu W, Liang H. Effect of statins on sepsis and inflammatory factors: A Mendelian randomization study. Eur J Clin Invest 2024;54:e14164. [Crossref] [PubMed]
9. Ando H, Takamura T, Ota T, et al. Cerivastatin improves survival of mice with lipopolysaccharide-induced sepsis. J Pharmacol Exp Ther 2000;294:1043-6.
10. Zhang S, Rahman M, Zhang S, et al. Simvastatin antagonizes CD40L secretion, CXC chemokine formation, and pulmonary infiltration of neutrophils in abdominal sepsis. J Leukoc Biol 2011;89:735-42. [Crossref] [PubMed]
11. Spitzer AL, Harris HW. Statins attenuate sepsis. Surgery 2006;139:283-7. [Crossref] [PubMed]
12. Yasuda H, Yuen PS, Hu X, et al. Simvastatin improves sepsis-induced mortality and acute kidney injury via renal vascular effects. Kidney Int 2006;69:1535-42. [Crossref] [PubMed]
13. Tavakkoli A, Johnston TP, Sahebkar A. Antifungal effects of statins. Pharmacol Ther 2020;208:107483. [Crossref] [PubMed]
14. Bhattacharya S, Esquivel BD, White TC. Overexpression or Deletion of Ergosterol Biosynthesis Genes Alters Doubling Time, Response to Stress Agents, and Drug Susceptibility in Saccharomyces cerevisiae. mBio 2018;9:e01291-18. [Crossref] [PubMed]
15. Oesterle A, Laufs U, Liao JK. Pleiotropic Effects of Statins on the Cardiovascular System. Circ Res 2017;120:229-43. [Crossref] [PubMed]
16. Haas H. Fungal siderophore metabolism with a focus on Aspergillus fumigatus. Nat Prod Rep 2014;31:1266-76. [Crossref] [PubMed]
17. Nagy G, Vaz AG, Szebenyi C, et al. CRISPR-Cas9-mediated disruption of the HMG-CoA reductase genes of Mucor circinelloides and subcellular localization of the encoded enzymes. Fungal Genet Biol 2019;129:30-9. [Crossref] [PubMed]
18. Magwenzi S, Woodward C, Wraith KS, et al. Oxidized LDL activates blood platelets through CD36/NOX2-mediated inhibition of the cGMP/protein kinase G signaling cascade. Blood 2015;125:2693-703. [Crossref] [PubMed]
19. Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med 2003;348:138-50. [Crossref] [PubMed]
20. Johnson A, Bulgarelli L, Pollard T, et al. Mimic-IV (Version1.0). PhysioNet. 2021. Available online: https://doi.org/10.13026/s6n6-xd98
21. Haneuse S, VanderWeele TJ, Arterburn D. Using the E-Value to Assess the Potential Effect of Unmeasured Confounding in Observational Studies. JAMA 2019;321:602-3. [Crossref] [PubMed]
22. Christensen S, Thomsen RW, Johansen MB, et al. Preadmission statin use and one-year mortality among patients in intensive care - a cohort study. Crit Care 2010;14:R29. [Crossref] [PubMed]
23. Kyu Oh T, Song IA, Lee JH, et al. Preadmission Statin Use and 90-day Mortality in the Critically Ill: A Retrospective Association Study. Anesthesiology 2019;131:315-27. [Crossref] [PubMed]
24. Yao Y, Zhao X, Wang M, et al. Association between the use of statins and in-hospital mortality risk in patients with sepsis-induced coagulopathy during ICU stays: a study based on medical information mart for intensive care database. BMC Infect Dis 2024;24:738. [Crossref] [PubMed]
25. Mao H, Yu Y, Wang Q, et al. Association between pre-ICU statin use and ARDS mortality in the MIMIC-IV database: a cohort study. Front Med (Lausanne) 2023;10:1328636. [Crossref] [PubMed]
26. Shi XT, Yan LJ, Lu FY, et al. Exploring the therapeutic potential of simvastatin in pancreatic neuroendocrine neoplasms: insights into cell cycle regulation and apoptosis. Transl Cancer Res 2024;13:4315-23. [Crossref] [PubMed]
27. Wang J, Cao G, Liu Y, et al. Mendelian randomization study on simvastatin and gastric cancer: exploring the therapeutic potential of statins in oncology. Transl Cancer Res 2024;13:4671-7. [Crossref] [PubMed]
28. Papazian L, Roch A, Charles PE, et al. Effect of statin therapy on mortality in patients with ventilator-associated pneumonia: a randomized clinical trial. JAMA 2013;310:1692-700. [Crossref] [PubMed]
29. Chen M, Ji M, Si X. The effects of statin therapy on mortality in patients with sepsis: A meta-analysis of randomized trials. Medicine (Baltimore) 2018;97:e11578. [Crossref] [PubMed]
30. Feng Y. Efficacy of statin therapy in patients with acute respiratory distress syndrome/acute lung injury: a systematic review and meta-analysis. Eur Rev Med Pharmacol Sci 2018;22:3190-8. [Crossref] [PubMed]

(English Language Editor: J. Gray)