Introduction

Patients with diabetes are more likely to have wounds and sores that do not heal and may become infected, leading to sepsis [1]. Over 90% of patients with diabetes have type 2 diabetes mellitus (T2DM), which affects hundreds of millions of individuals worldwide [2]. T2DM is characterized by hyperglycemia, insulin resistance, impaired insulin secretion, and dyslipidemia (high triglyceride levels and low high-density lipoprotein cholesterol level) [3,4,5,6]. Moreover, diabetes alters the immune system, resulting in an increased risk of sepsis [1]. T2DM is associated with increased risks of recurrent, nosocomial, and secondary infections that lead to sepsis [1, 7]. Patients with T2DM have a higher risk of community-acquired pneumonia, biliary disease, cutaneous infections, and aspiration pneumonia during hospitalization [1, 8]. Patients with T2DM undergoing surgery may have a high risk of infectious complications that lead to sepsis, ventilator-associated pneumonia, and central venous catheter–related infections [1, 8,9,10].

Many studies evaluating the association between statin use and sepsis in different populations and at various endpoints for statin use have reported controversial findings [11,12,13,14,15,16,17,18,19,20]. In terms of sepsis prevention, statin users had superior outcomes than did nonstatin users [19, 20]. However, in severely ill hospitalized patients with diseases such as pneumonia, statin use did not prevent mortality or sepsis-related mortality [11,12,13,14,15,16,17,18]. Statins might prevent diseases through various mechanisms, such as by reducing the cholesterol level and exhibiting anti-inflammatory, immunomodulatory, antioxidant, antithrombotic, and endothelium-stabilizing properties [19,20,21,22,23]. The inconsistency in the aforementioned findings may be attributable to the slow effects of statins. Thus, statin use might prevent the progression of diseases, such as cardiovascular disease [24, 25], stroke [26, 27], and mortality [28], only in relatively healthy individuals instead of severely ill hospitalized patients with several diseases. Moreover, the inconsistent findings may be attributable to the inclusion of different populations and various endpoints for statin use [11,12,13,14,15,16,17,18,19,20]. The use of statins as a preventive medication can be beneficial in specific populations, especially patients with T2DM with a high prevalence of inflammatory diseases, immune disorders, oxidative stress conditions, and thrombotic and endothelial diseases, which lead to a high risk of sepsis [29,30,31]. A protective, safe, and long-term medication for the prevention of sepsis in the susceptible T2DM population is not yet available.

By using a real-world database, in this study, we investigated the dose-dependent protective effects of statins, specific classes of statin, and different intensities of statin use on sepsis risk in T2DM. In addition, we determined the optimal daily statin dose to prevent sepsis in patients with T2DM.

Patients and methods

Study population

We conducted a population-based cohort study by using data from Taiwan’s National Health Insurance (NHI) Research Database (NHIRD). The NHIRD contains all medical claims data regarding the disease diagnoses, procedures, drug prescriptions, demographics, and enrollment profiles of all NHI beneficiaries [32]. The NHIRD is linked by encrypted patient identifiers. In addition, the NHIRD data are linked to the Death Registry to ascertain the vital status and cause of death of each patient.

Our cohort included patients who were diagnosed as having T2DM between 2008 and 2020 and were aged  ≥ 40 years. Patients with missing information on age were excluded. To investigate the protective effects of different classes of statins on sepsis, we excluded patients who used different classes of statins during the follow-up period. Statin use was defined as using statin on most days for  > 1 months within 1 year, with a mean statin dose of  ≥ 28 cumulative defined daily doses (cDDDs) per year (cDDD-year). The index date was the date of statin use (≥ 28 cDDD-year). The observation period for each patient began from the index date and continued until death, hospital admission for sepsis, or the end of the study period (December 31, 2021). For patients with more than one episode of sepsis, we analyzed their first episode. Patients who developed sepsis before the index date were excluded. Patients with T2DM who were prescribed  ≥ 28 cDDD-year of statins with a prescription duration of  > 1 months were included in the case group, and those who were prescribed 0 cDDD of statins during the follow-up period were included in the control group.

Sepsis patients were defined as those who were diagnosed with sepsis for the first time and received antibiotic treatment during their hospitalization, based on the ICD-9-CM and ICD-10-CM code. Patients with recurrent sepsis were excluded from the study. It is important to note that all enrolled sepsis patients had no prior history of sepsis and were experiencing the condition for the first time. The ICD-9-CM and ICD-10-CM official guidelines specify the use of specific codes for sepsis, severe sepsis, and septic shock. Specifically, the codes "038.xx" and "995.91" are utilized for sepsis, while "995.92" and "758.52" are designated for septic shock. In the ICD-10-CM coding system, the codes "A40.xx" and "A41.xx" are used to identify sepsis, with the fourth digit specifying the organism causing the infection. The code "R65.20" is used for severe sepsis without septic shock, while the codes "R65.21" and "R65.22" are used for septic shock of different severity levels.

Study covariates

We included other covariates to adjust for potential confounding effects. Patients were divided into the following age groups: 40 to 50, 51 to 60, 61 to 70, and  ≥ 71 years at the index date. To reduce the effects of potential confounders when comparing sepsis between the statin user and nonuser groups, we used the inverse probability of treatment-weighted (IPTW) [33] Cox regression models with adjustment for age groups, sex, income levels, urbanization, types of antidiabetic drugs used, antidiabetic drugs, diabetic severity (adapted Diabetes Complications Severity Index [aDCSI] score), coexisting comorbidities, and the Charlson comorbidity index (CCI) score (Table 1). We used the date of statin use (≥ 28 cDDD-year) as the index date and matched nonstatin users by using variables collected at this index date. Repeat comorbidities were excluded from CCI scores to prevent repetitive adjustment in multivariate analysis. Comorbidities were determined in accordance with International Classification of Diseases, Ninth Revision, Clinical Modification and International Classification of Diseases, Tenth Revision, Clinical Modification codes in inpatient records or based on whether the number of outpatient visits was ≥ 2 within 1 year. Onset of comorbidities during 1 year prior to the index date was recorded. Continuous variables are presented as the mean ± standard deviation or median (first and third quartiles) where appropriate.

Table 1 Baseline characteristics of the overall T2DM cohort by statin use status
Full size table

Outcome variables

Development of sepsis was the primary study outcome. Septic shock was identified as the second outcome.

Exposure to statins

Prescriptions for statins were coded in accordance with the Anatomical Therapeutic Chemical (ATC) coding system of the NHIRD pharmaceutical subsidies and were used as an interface for retrieving pharmaceutical claims data. In accordance with the ATC classification system, we selected lipophilic (atorvastatin, fluvastatin, lovastatin, simvastatin, and pitavastatin) and hydrophilic (pravastatin and rosuvastatin) statins [34] as the major exposures of interest. In addition, we examined the intensity of statin use by continually estimating the average statin dose as the defined daily dose (DDD) divided by total prescription days. The intensity of statin use was divided into average daily doses below or above 1 DDD. Furthermore, we divided patients into four subgroups that were stratified by the quartiles (Q) of cDDD-year. All analyses were adjusted for age groups, sex, income levels, urbanization, types of antidiabetic drugs used, antidiabetic drugs, diabetic severity (aDCSI score), coexisting comorbidities, current smoking, alcohol liver diseases, and CCI scores.

Statistical analysis

The IPTW [33] Cox regression model was used to overcome the imbalance in baseline characteristics between statin and nonstatin users after adjustment for age groups, sex, income levels, urbanization, types of antidiabetic drugs used, antidiabetic drugs, diabetic severity (aDCSI score), coexisting comorbidities, and CCI scores. A time-dependent Cox hazard model was used to compare sepsis between statin and nonstatin users after adjustment for the aforementioned confounding factors. Data on statin prescriptions were collected every 3 months to define a user’s status and were estimated as a time-dependent variable. “Event-free” person-times of users before their first prescription and during the 3-month period without a statin prescription were classified as unexposed follow-up times to prevent bias. In addition, we estimated the effects of individual statins on the risk of sepsis. Analyses were performed in subgroups after adjustment for baseline characteristics by using stratification instead of weighting and postdiagnosis statin use, which yielded similar results. The cumulative incidence of sepsis was estimated using the Kaplan–Meier method, and differences between statin users and nonusers were determined using the stratified log-rank test to compare cumulative incidence curves. Differences between statin users and nonusers at different cDDD-years and for specific statin classes were determined using the stratified log-rank test (Figs. 1 and 2). All statistical analyses were conducted using SAS version. 9.4 (SAS Institute, Cary, NC, USA). The study protocols were reviewed and approved by the Institutional Review Board of Tzu-Chi Medical Foundation (IRB109-015-B).

Fig. 1
figure 1

Kaplan–Meier analysis of the cumulative curves of sepsis for different classes of statins in patients with T2DM

Full size image
Fig. 2
figure 2

Kaplan–Meier analysis of the cumulative curves of sepsis for different cDDD-years of statins in patients with T2DM

Full size image

Results

From 2008 to 2020, a total of 812 420 patients were diagnosed as having T2DM. The mean age at T2DM diagnosis were 56.22 and 56.92 years for the nonstatin users and statin users, respectively. Furthermore, 35.98% of the statin users received atorvastatin, which was the most prescribed statin, followed by simvastatin (19.81%) and rosuvastatin (19.61%). To ensure postmatch balance, we used the absolute standardized mean difference (ASMD) of < 0.1 after IPTW for all baseline covariates [35]. The ASMDs for all covariates were < 0.1, indicating that the covariates after IPTW were balanced between the statin users and nonusers (Table 1) [35].

Sepsis, comparison of different classes of statin use, and dose-dependent protective effects

Among the patients with T2DM, 118 765 (27.79%) nonstatin users and 50 804 (12.03%) statin users developed sepsis. Overall, statin users had a lower incidence of sepsis than did the nonusers. The adjusted hazard ratio (aHR) of statin use was 0.37 (95% CI: 0.35, 0.38) for sepsis compared with nonstatin use (Table 2), and the log-rank test yielded a P < 0.0001 (Additional file 1: Figure S1). The findings of the Cox regression model revealed that compared with the nonstatin users, those using different classes of statins exhibited a significant reduction in sepsis, with the aHRs (95% CI) of sepsis being 0.09 (0.05, 0.14), 0.32 (0.31, 0.34), 0.34 (0.32, 0.36), 0.35 (0.32, 0.37), 0.37 (0.34, 0.39), 0.42 (0.38, 0.44), and 0.54 (0.51, 0.56), respectively, for pitavastatin, pravastatin, rosuvastatin, atorvastatin, simvastatin, fluvastatin, and lovastatin use, respectively (Table 2). The results of the log-rank test indicated that the cumulative incidence of sepsis significantly differed among the patients using different classes of statins (P < 0.0001; Fig. 1). In the patients with different cDDD-years of statins, the findings of multivariate analysis revealed a significant reduction in sepsis, with aHRs of 0.53 (0.52, 0.57), 0.40 (0.39, 0.43), 0.29 (0.27, 0.30), and 0.17 (0.15, 0.19) for Q1, Q2, Q3, and Q4 cDDD-year, respectively (P for trend < 0.0001), and the log-rank test yielded a P < 0.0001 (Fig. 2).

Table 2 Sepsis risk and adjusted hazard ratios (aHRs) associated with statin use among patients with T2DM
Full size table

Intensity of statin use

The optimal intensity of statin use was 0.84 DDD, which has a lower aHR of sepsis (Additional file 1: Figure S2) than the other DDDs. The protective effects of statins on sepsis exhibited a U-shaped dose–response relationship [36]. The optimal milligram recommendations for different statins use were shown in Additional file 1: Table S1.

Sensitivity analysis

We examined the intensity of statin use and determined that the patients who received on average both ≤ 1 and > 1 DDD had a decreased risk of sepsis. In addition, we investigated the effect of statins on patients with different comorbidities (CCI ≤ 1), age groups, sex, income levels, urbanization, types of antidiabetic drugs used, antidiabetic drugs, diabetic severity (aDCSI Score), and coexisting comorbidities. Reduction in sepsis risk observed in sensitivity analysis was comparable to that noted in the main analysis (Table 3).

Table 3 Sensitivity analyses of the association between statin use and sepsis among patients with T2DM
Full size table

IRs and IRRs for sepsis

Overall, significant IRRs of sepsis risk were obtained for the statin users and nonusers (Table 4). The IRR (95% CI) of sepsis risk in the statin users compared with the nonusers was 0.41 (0.40, 0.41), and the IRs of sepsis risk in the statin users and nonusers were 106.03 and 259.09 per 10,000 person-years, respectively. The IRs of sepsis risk for pitavastatin, pravastatin, rosuvastatin, atorvastatin, simvastatin, fluvastatin, and lovastatin users were 16.80, 90.97, 94.76, 100.67, 104.42, 130.19, and 166.85 per 10,000 person-years, respectively. The IRRs (95% CI) of sepsis risk in the statin users compared with the nonusers were 0.61 (0.60, 0.62), 0.46 (0.45, 0.47), 0.32 (0.31, 0.32), and 0.19 (0.18, 0.20), respectively, for Q1, Q2, Q3, and Q4 cDDD-years.

Table 4 IR and IRRs for sepsis
Full size table

Septic shock, comparison of different classes of statin use, and dose-dependent protective effects

To investigate the protective effects of statins against septic shock among patients with T2DM. Of the total T2DM patients, 42,755 (10.39%) did not use statins, while 16,765 (4.18%) did. Statin users exhibited a significantly lower incidence of septic shock than nonusers. After adjustment for confounding factors, the adjusted hazard ratio (aHR) for septic shock was 0.34 (95% CI 0.33, 0.35) in statin users compared to nonusers (Additional file 1: Table S2). Further analysis revealed that the use of different classes of statins was associated with a significant reduction in sepsis, as indicated by the aHRs (95% CI) of sepsis for pitavastatin, pravastatin, rosuvastatin, atorvastatin, simvastatin, fluvastatin, and lovastatin use of 0.06 (0.04, 0.10), 0.29 (0.28, 0.31), 0.31 (0.30, 0.33), 0.32 (0.31, 0.33), 0.34 (0.33, 0.36), 0.38 (0.36, 0.40), and 0.50 (0.48, 0.53), respectively (Additional file 1: Table S2). Moreover, the protective effect of statins against sepsis was dose-dependent. Patients who used statins with higher cumulative defined daily doses (cDDD-years) had a lower incidence of sepsis, with aHRs of 0.53 (0.52, 0.55), 0.39 (0.38, 0.41), 0.26 (0.24, 0.28), and 0.14 (0.13, 0.15) for the lowest to highest quartile of cDDD-years, respectively (P for trend < 0.0001), according to the findings of the multivariate analysis.

Discussion

Statins appear to possess beneficial anti-inflammatory properties; for instance, statins can suppress the endotoxin-induced upregulation of Toll-like receptor (TLR)-4 and TLR-2 [7, 37]. Some studies have indicated the preventive effect of statins on sepsis in patients with cardiovascular diseases [19, 20]. By contrast, some studies, including the meta-analyses of randomized trials, have reported no beneficial effects of statin use on mortality or sepsis-related mortality in the hospitalized population with pneumonia or active infection [11,12,13,14,15,16,17,18]. However, these studies have included heterogeneous populations, various endpoints, and different statin classes [7, 11,12,13,14,15,16,17,18,19,20]. Moreover, these studies did not indicate a clear DDD, the dose-dependent protective effects of statins on sepsis or mortality, the intensity of statin use, and cDDD-year as well as examine the effects of the continued use or discontinuation of statins. Previous studies have reported vague findings regarding the protective effects of different classes, doses, and intensity of statin use on sepsis in the susceptible population with T2DM with a high risk of sepsis [10, 29]. No study has evaluated whether statin use can prevent sepsis in the susceptible population with T2DM. Because T2DM is an independent risk factor for sepsis and patients with T2DM have a high prevalence of sepsis [9, 10, 29], a safe, effective, and long-term protective medication for sepsis is required. We investigated the dose-dependent protective effects of statins, specific classes of statin, and different intensities of statin use on sepsis risk in patients with T2DM. In addition, we determined the optimal daily statin dose to prevent sepsis in patients with T2DM. Our results demonstrated that the aHR of statin use was 0.37 (95% CI: 0.35, 0.38) for sepsis compared with no statin use. Compared with the patients not using statins, those using different classes of statins exhibited a more significant reduction in sepsis, with aHRs (95% CIs) of sepsis being 0.09 (0.05, 0.14), 0.32 (0.31, 0.34), 0.34 (0.32, 0.36), 0.35 (0.32, 0.37), 0.37 (0.34, 0.39), 0.42 (0.38, 0.44), and 0.54 (0.51, 0.56) for pitavastatin, pravastatin, rosuvastatin, atorvastatin, simvastatin, fluvastatin, and lovastatin use, respectively. In the patients with different cDDD-years of statins, multivariate analysis indicated a significant reduction in sepsis, with aHRs of 0.53 (0.52, 0.57), 0.40 (0.39, 0.43), 0.29 (0.27, 0.30), and 0.17 (0.15, 0.19) for Q1, Q2, Q3, and Q4 cDDD-years (P for trend < 0.0001). The optimal daily statin dose of 0.84 DDD was associated with the lowest aHR.The optimal intensity of statin use was 0.84 DDD, which resulted in a lower aHR than did other DDDs. Sensitivity analysis indicated that sepsis risk was significantly decreased in the statin users, regardless of age, sex, income levels, urbanization, types of antidiabetic drugs use, antidiabetic drugs, aDCSI score, coexisting comorbidities, medication use, and CCI scores.

To the best of our knowledge, no study has evaluated the protective effects of different classes of statins on sepsis in patients with T2DM. This is the first study to demonstrate the protective effect of specific statins on sepsis in patients with T2DM. Pitavastatin exerted the strongest protective effect on sepsis, followed by pravastatin, rosuvastatin, atorvastatin, simvastatin, fluvastatin, and lovastatin. Statins exert protective effects possibly by reducing low-density lipoprotein (LDL) and triglyceride levels and increasing the high-density lipoprotein (HDL) level [38,39,40]. For example, rosuvastatin is more potent than atorvastatin [38, 39], and rosuvastatin is significantly more potent than simvastatin, atorvastatin, fluvastatin, and lovastatin [39, 40]. At the maximal prescribed doses, LDL cholesterol reduction is greater with rosuvastatin than with the aforementioned three statins [39, 40]. The efficacy of the aforementioned four statins in reducing the LDL level is similar to their protective effects on sepsis in patients with T2DM (Table 2 and Fig. 1). Statins alter the HDL cholesterol level (also known as the good cholesterol), typically increasing them, but these effects vary by the class and dose of statins [41]. For example, an increase in the HDL cholesterol level is noted with the increasing doses of simvastatin and rosuvastatin, whereas the increase in the HDL cholesterol level caused by atorvastatin is attenuated at its higher doses [41]. Moreover, rosuvastatin was more effective in reducing the triglyceride level than other statins in patients with hypercholesterolemia [39]. However, the association of the effects of specific statins on LDL, HDL, and triglycerides with sepsis remains unclear. In our current study, the effects of specific statins on LDL, HDL, and triglycerides appeared to be proportional to the protective effect of statins on sepsis in the patients with T2DM (Table 2 and Fig. 1). In addition, pitavastatin, pravastatin, and fluvastatin are less likely to have drug interactions or cause muscle toxicity than some other statins [42, 43]. Fewer pharmacokinetic drug interactions are likely to occur with pravastatin, rosuvastatin, pitavastatin, and fluvastatin because they are not metabolized through CYP3A4 [42, 43]. Patients with T2DM receive many types of medication (Table 1); thus, statins with fewer drug–drug interactions, including pitavastatin and pravastatin, might lead to a balance between effects and toxicities [42, 43]. Although the detailed mechanisms of specific classes of statins and their preventive effects on sepsis remain unclear, statins that result in fewer pharmacokinetic drug interactions, including pitavastatin and pravastatin [42, 43], and exert stronger effects on lowering LDL and triglycerides and increasing HDL, including rosuvastatin [38,39,40], might be better choices. However, because the sample size of pitavastatin users in our study was small, our findings might be biased. Therefore, future studies should investigate the detailed effects of specific statins on sepsis and their underlying mechanisms.

The effects of statins on LDL, HDL, and triglycerides might differ on the basis of their intensity and daily dose because we observed a U-shaped dose–response relationship for the effect of the daily dose of statins on LDL, HDL, and triglycerides [41, 44]. Thus, the U-shaped dose–response relationship was observed for not only the pharmacological but also toxicological effects of statins (Additional file 1: Figure S2) [36]. Thus, in our study, we observed that an increased daily dose of statins did not result in a better protective effect [45]. This might be the reason for the inconsistency in the findings of previous studies on the association of statin use with sepsis risk [7, 11,12,13,14,15,16,17,18,19,20]. This is the first study to demonstrate that the optimal intensity of DDD for statin users was 0.84 DDD, which was associated with a lower risk of sepsis in the T2DM population. The U-shaped dose–response relationship observed for the protective effects of statins on sepsis is in agreement with the findings of previous biological, toxicological, and pharmacological studies [36]. Part of the variability in the response to and side effects of statins may be related to genetic differences in the rate of drug metabolism [46,47,48]. CYP2D6 is a member of the cytochrome P450 superfamily of drug-oxidizing enzymes. CYP2D6 is functionally absent in 7% of White and African American individuals, and its deficiency is rare among Asian individuals. Asian individuals (mostly those from China, Japan, and Korea) may exhibit greater responses to low doses of statins than do European American individuals [47]. Thus, statin therapy should be started with a lower initial daily dose in Asian individuals than in other groups considering the observed differences in pharmacokinetics [47, 49]. Therefore, our study demonstrated that the optimal intensity of statin daily dose was 0.84 DDD, and this value would be valuable for Asian patients and explain the previous inconsistent findings [7, 11,12,13,14,15,16,17,18,19,20]. The optimal milligram recommendations for different statins use were shown in Additional file 1: Table S1.

Different cDDD-years for statins might exert different effects on LDL, HDL, and triglycerides and thus different effects on sepsis risk in patients with T2DM. Therefore, we determined the effects of the cumulative doses of Q1, Q2, Q3, and Q4 cDDD-years on sepsis risk in the patients with T2DM. Our results revealed that the aHRs (95% CIs) of the cDDD-year of Q1, Q2, Q3, and Q4 were 0.53 (0.52, 0.57), 0.40 (0.39, 0.43), 0.29 (0.27, 0.30), and 0.17 (0.15, 0.19; P for trend < 0.0001). A higher cDDD-year of statins was associated with an increased reduction of sepsis risk in the patients with T2DM. Our results demonstrated the dose-dependent protective effect of statin use on sepsis in the patients with T2DM.

The strengths of our study is that it included the largest sample size of statin users and examined the effects of the intensity and dose-dependent protective effects of statins on sepsis in the patients with T2DM (Figs. 1 and 2 and Additional file 1: Figures S1 and S2). Compared with the findings of previous studies examining the association of statin use with sepsis in different populations, our study provided more reliable and long-term follow-up real-world evidence to indicate that the persistent use of statins can reduce sepsis risk in patients with T2DM (Tables 2–4). In addition, in terms of the intensity of statin use, the optimal daily statin dose of 0.84 DDD was associated with the lowest sepsis risk (Additional file 1: Figure S2). Moreover, pitavastatin exerted the strongest protective effect on sepsis, followed by pravastatin, rosuvastatin, atorvastatin, simvastatin, fluvastatin, and lovastatin (Table 2 and Fig. 1). This is the first study to investigate the dose-dependent protective effects of statins, specific classes of statins, and different intensities of statin use on sepsis risk in T2DM.

This study has some limitations. First, this study was conducted using a claims database. Laboratory values or lipid profiles were not available. Therefore, we could not evaluate whether changes in lipid profiles following the initiation of statin use were associated with sepsis. Second, we could not completely avoid the possibility that statin users might be a different population compared with nonusers, which might have been an unmeasured confounding factor in our study. We used IPTW to balance the difference in covariates. Several subgroup analyses were conducted to examine potential bias resulting from unmeasured confounders. We examined the effects of statins for different age groups, sex, income levels, urbanization, types of antidiabetic drugs use, antidiabetic drugs, aDCSI Score, coexisting comorbidities, medication use, and CCI scores. The reduction in sepsis with statin use was similar in patients with T2DM in sensitivity analysis. Third, we did not have information on the body mass index and other lifestyle factors at the time of T2DM diagnosis. Therefore, we were unable to evaluate the impact of those factors on sepsis. Fourth, event numbers were small in some of the subgroups of specific statin classes, which limited our statistical power. Finally, our study population was 95% Han Chinese [50], which limits the generalizability of our results to other ethnic groups. The prevalence of statin use was approximately 76.5% in North Americans, 69.9% in Western Europeans, and 60.5% in Asians [51]. Therefore, other ethnicities with higher rates of statin use might have slightly different results. However, some previous studies conducted in different ethnic populations also demonstrated a reduction in sepsis risk associated with statin use.

Conclusion

Our real-world evidence demonstrated that the persistent use of statins reduced sepsis risk in the patients with T2DM and a higher cDDD-year of statins was associated with more reduction in sepsis risk in these patients. The optimal daily statin dose of 0.84 DDD was associated with the lowest mortality. Moreover, pitavastatin exerted the strongest protective effect on sepsis, followed by pravastatin, rosuvastatin, atorvastatin, simvastatin, fluvastatin, and lovastatin.