TYPE Original Research PUBLISHED 24 August 2022 DOI 10.3389/fendo.2022.945626 Risk of infections in patients with OPEN ACCESS NAFLD and Type 2 Diabetes EDITED BY Cristiane Nogueira, Federal University of Rio de Janeiro, under treatment with SGLT2 Brazil REVIEWED BY inhibitors and relationship with Man-Fung Tsoi, The University of Manchester, United Kingdom liver outcomes: A retrospective David Nicolás, Hospital Clinic of Barcelona, Spain case-control study Luiz Guilherme Kraemer-Aguiar, Rio de Janeiro State University, Brazil Joana Nicolau, Son Llàtzer Hospital, Spain Juan Bañares 1†, Ramiro Manzano-Nuñez 1†, Alba Prió 1, *CORRESPONDENCE Jesús Rivera-Esteban 1,2, Laura Camps-Relats 1, Ana Villarejo 1, Juan M. Pericàs Lourdes Ruiz-Ortega 1, Mònica Pons 1,3, Andreea Ciudin 2,4,5*, juan.pericas@vhir.org Andreea Ciudin Marı́a Teresa Salcedo 2,6, Vı́ctor Vargas 1,2,3, Joan Genescà 1,2,3 andreea.ciudin@vallhebron.cat and Juan M. Pericàs 1,3* † 1 Liver Unit, Internal Medicine Department, Hospital Universitari Vall d’Hebron, Vall d’Hebron Institut These authors have contributed equally to this work de Recerca (VHIR), Vall d’Hebron Barcelona Campus Hospitalari, Barcelona, Spain, 2 Medicine Department, Universitat Autònoma de Barcelona, Barcelona, Spain, 3 Centro de Investigación SPECIALTY SECTION Biomédica en Red de enfermedades digestivas y hepáticas (CIBERehd), Madrid, Instituto de Salud This article was submitted to Carlos III, Madrid, Spain, 4 Endocrinology and Nutrition Department, Hospital Universitari Vall Translational Endocrinology, d’Hebron, Vall d’Hebron Institut de Recerca (VHIR), Vall d’Hebron Barcelona Campus Hospitalari, a section of the journal Barcelona, Spain, 5 CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDem), Instituto Frontiers in Endocrinology de Salud Carlos III, Madrid, Spain, 6 Pathology Department, Hospital Universitari Vall d’Hebron, Vall d’Hebron Barcelona Campus Hospitalari, Barcelona, Spain RECEIVED 16 May 2022 ACCEPTED 04 August 2022 PUBLISHED 24 August 2022 CITATION Bañares J, Manzano-Nuñez R, Prió A, Introduction: Non-alcoholic fatty liver disease (NAFLD) is the most prevalent Rivera-Esteban J, Camps-Relats L, chronic liver disease in developed countries, with its incidence growing parallel Villarejo A, Ruiz-Ortega L, Pons M, to the epidemics of obesity and type 2 diabetes mellitus (T2DM). Sodium- Ciudin A, Salcedo MT, Vargas V, Genescà J and Pericàs JM (2022) Risk glucose co-transporter-2 inhibitors (SGLT2i) are becoming a cornerstone in of infections in patients with NAFLD the management of cardiovascular health and some studies suggest the and Type 2 Diabetes under treatment with SGLT2 inhibitors and relationship potential role in NAFLD. However, patients under treatment with SGLT2i are with liver outcomes: A retrospective at risk of developing genitourinary fungal infections (GFIs). Moreover, both case-control study. NAFLD and SGLT2i have a strong influence on the immune system, and Front. Endocrinol. 13:945626. doi: 10.3389/fendo.2022.945626 therefore the risk of infections other than GFIs could be increased in NAFLD patients treated with SGLT2i. We aimed to examine the possible association of SGLT2i with infections and hepatic outcomes in NAFLD patients. Methods: We conducted a case-control study including NAFLD patients with T2DM visited at the Liver Unit outpatient clinic from 2016 to 2021 with a minimum follow-up of 6 months by selecting 65 patients receiving SGLT2i and 130 matched patients with other types of antidiabetic treatment. Results: During follow-up, GFIs were significantly higher in the SGLT2i group (15.4% vs. 3.8%; p=0.008), whereas there were no differences in the occurrence of overall infections (41.5% vs. 30%; p=0.1) nor in other types of specific Frontiers in Endocrinology 01 frontiersin.org Bañares et al. 10.3389/fendo.2022.945626 infections. In the multivariable analysis, treatment with SGLT2i was not independently associated with higher odds of overall infection. On the other hand, SGLT2i patients showed a significantly lower incidence of hepatic events (1.5% vs. 10.7%; p=0.02). There were no significant different in all-cause mortality between cases and controls. Conclusions: NAFLD patients with T2DM receiving SGLT2i more frequently presented GFIs, whereas the incidence of other types of infections was not found to be higher than in other patients with NAFLD and T2DM treated with other drugs. Moreover, SGLT2i-treated patients had a lower occurrence of hepatic events. Further studies are warranted to validate our data. KEYWORDS NAFLD, type 2 diabetes mellitus, sodium-glucose co-transporter-2 inhibitors, infections, hepatic outcomes Introduction suggest that SGLT2i may play a role in treating NAFLD being associated with an impact in the metabolic status, including Non-alcoholic fatty liver disease (NAFLD) encompasses a reduction in liver fat content and even histologic improvement progressive clinical spectrum: from simple steatosis, through in liver steatosis and fibrosis (7–10). Such effects could have a inflammation [i.e., non-alcoholic steatohepatitis (NASH)] and clinically relevant impact on the outcomes of patients with fibrosis, to liver cirrhosis (1, 2). From an epidemiological T2DM and NAFLD, making SGLT2i an attractive therapeutic standpoint, parallel to the growing epidemic of obesity and alternative. Even though there are currently several clinical trials type 2 diabetes mellitus (T2DM), NAFLD has emerged as the underway assessing their efficacy to treat NAFLD (11, 12), there most prevalent liver disease in the US and probably will become is insufficient information regarding an impact in the leading cause of liver transplantation in the upcoming years clinical practice. (3). It is estimated that 25% of general population in Western SGLT2i inhibit glucose reabsorption in the kidney via countries to have NAFLD, but this prevalence increases up to 60- inhibition of the SGLT channels primarily located in the 80% in patients with obesity or T2DM and can reach 80-100% proximal tubules, promoting glycosuria, which has been when both risk factors are present (2, 3). In addition, it is associated with a higher incidence of urinary and genital estimated that 20-30% of patients with NAFLD will progress infections, mainly caused by fungi, with odds ratios to liver inflammation and fibrosis (2–4). ranging approximately from 3 to 5 (13–17). Moreover, NAFLD is considered the liver manifestation of metabolic patients with T2DM have increased susceptibility to a wide syndrome, both sharing multiple pathophysiological array of infections due to variable degrees of baseline mechanisms such as insulin resistance (1–3). Moreover, immunosuppression caused by complex mechanisms that improvements in metabolic factors such as weight loss, are are tightly intertwined with pathways leading to the associated with amelioration in inflammation and liver fibrosis enhanced systemic inflammation and immune system (1). Sodium-glucose co-transporter-2 inhibitors (SGLT2i) are dysfunction characteristic of advanced liver disease, while now solidly established amongst the armamentarium to improve obesity and NAFLD may also increase infection susceptibility metabolic status and cardiovascular health in T2DM, chronic (1, 18, 19). However, no studies have investigated whether kidney disease and cardiovascular diseases (5, 6). Emerging data T2DM patients with NAFLD treated with SGLT2i present increased rates of infections and particularly genitourinary Abbreviations: NAFLD, Non-alcoholic fatty liver disease; T2DM, type 2 fungal infections (GFIs). diabetes mellitus; SGLT2i, Sodium-glucose co-transporter-2 inhibitors; GFIs, The present study aimed to examine the impact of SGLT2i genitourinary fungal infections; NASH, non-alcoholic steatohepatitis; treatment in the incidence of infections in patients with NAFLD VHUH, Vall d’Hebron University Hospital; HE, hepatic encephalopathy; and T2DM. In addition, we aimed to investigate whether COVID-19, Coronavirus disease 2019; IQR, interquartile range; BMI, body NAFLD patients treated with SGLT2i presented significant mass index; OR, odds ratio; CI, confidence of interval; MRI, magnetic differences in liver outcomes compared to those not resonance imaging. receiving SGLT2i. Frontiers in Endocrinology 02 frontiersin.org Bañares et al. 10.3389/fendo.2022.945626 Methods recommendations from the American Diabetes Association (20). Cirrhosis was defined by histological analysis when available and/or Design and setting by a combination of clinical, analytical, liver stiffness measurement, endoscopic assessment and/or radiological evidence of advanced Case-control study conducted at the Vall d’Hebron chronic liver disease. Hepatic events were defined as the presence of University Hospital (VHUH), a tertiary care setting with 1,300 ascites, hepatic encephalopathy, and upper gastrointestinal bleeding. beds in Barcelona, Spain. Infections were defined as urogenital fungal infections, urogenital bacterial infections, bacteremia, upper and lower respiratory tract infections, COVID-19, intra-abdominal, skin and soft tissues infections, and spontaneous bacterial peritonitis. Infections Participants were diagnosed in a real clinical setting by means of clinical presentation and, when needed, radiological, laboratory and An ongoing prospective cohort study on NAFLD was used microbiological analysis. to identify patients considered eligible to be included in the present case-control study. The ongoing prospective cohort study includes consecutive patients from the VHUH NAFLD Ethics outpatient clinics diagnosed with NAFLD and T2DM from 2016 to 2021. Patients with a diagnosis of both NAFLD and T2DM The study protocol was approved by the VHUH Ethics were considered eligible for inclusion. Patients with NAFLD and Committee for Clinical Research (protocol code: PR(AG)222/ T2DM receiving treatment with SGLT2i and with a minimum of 2021) and conformed to the ethical guidelines of the 1975 6 months of follow-up were identified and defined as cases. Declaration of Helsinki. NAFLD and T2DM patients receiving other antidiabetic drug different than SGLT2i were considered as potential controls (see flowchart in Figure 1). Matching variables were sex and age (+/- Statistical analyses 3 years). We excluded patients with Child B and C cirrhosis from the Data were collected into a pre-specified and de-identified analysis to avoid bias in the allocation of SGLT2i treatment, as database in electronic format. Categorical variables were we considered that patients with more severe liver disease were presented as frequencies and percentages. Continuous less likely to receive treatment with SGLT2i because these are not variables were expressed as means with standard deviation or widely recommended in patients with moderate-severe hepatic medians with interquartile ranges as appropriate according to insufficiency, and also because patients with advanced chronic data distribution. Categorical variables were compared with liver disease are at higher risk of infections and these might Chi square test or Fisher exact test where appropriate. trigger hepatic events. Continuous variables were compared using T test or nonparametric tests when necessary. A multivariate logistic regression analysis adjusted by relevant covariates was used to examine the association between SGLT2i use and the Outcomes occurrence of infections. Findings were presented as odds ratios (OR) with 95% confidence intervals. Kaplan-Meier The occurrence of any type of infection was considered as survival curves and long-rank test were calculated for the primary outcome of interest. Secondary outcomes included the occurrence of overall infections and hepatic events the occurrence of each type of infection separately, the along time. All analyses were performed in STATA occurrence of hepatic events (only the first episode accounted statistical software. for each patient), and all-cause mortality. Results Definitions Patients The diagnostic criteria for NAFLD and T2DM are described elsewhere (1, 20). Briefly, NAFLD was diagnosed either based on The flowchart of patients’ disposition is shown in Figure 1. steatosis in abdominal ultrasound or transient elastography and the After applying the inclusion and exclusion criteria, sixty-five presence of at least one feature of metabolic syndrome or by liver patients with NAFLD and T2DM received SGLT2i and were biopsy. T2DM was diagnosed following current clinical practice considered as cases, whereas 130 diabetic patients with NAFLD not Frontiers in Endocrinology 03 frontiersin.org Bañares et al. 10.3389/fendo.2022.945626 FIGURE 1 Patients’ disposition flowchart. receiving SGLT2i were matched as controls. Characteristics of (matching variables), the prevalence of comorbidities, or the values of patients are shown in Table 1. Overall median follow-up was 34.4 glycosylated hemoglobin, albumin, and liver enzymes between months (IQR 17-48.3), with no significant differences found between groups. Patients receiving SGLT2i had significantly higher BMI cases and controls. There were neither differences regarding age, sex values and were more likely to present obesity. Frontiers in Endocrinology 04 frontiersin.org Bañares et al. 10.3389/fendo.2022.945626 Infections significantly higher rate of overall infections in patients treated with SGLT2i, we described for the first time that genitourinary Outcomes’ information is outlined in Table 2. There were no fungal infections are also more frequent in T2DM patients with significant differences in the occurrence of overall infections NAFLD treated with SGLT2i. Moreover, we found that SGLT2i- between groups (41.5% vs. 30%; p=0.1). With respect to the treated patients had lower rates of hepatic events during the infection’s etiology, the proportion of patients presenting genital follow-up time included in the study. mycotic infections was significantly higher in patients receiving Despite there were no significant differences in the SGLT2i compared to those not receiving SGLT2i (15.4% vs. occurrence of overall infectious episodes among SGLT2i 3.8%; p=0.008), with incidence rates of 5.12 and 1.28 cases per patients, these were more likely to present genitourinary 100 person-year in SGLT2i and non-SGLT2i groups, fungal infections (GFIs). In the general population, the respectively (p<0.001). There were no significant differences in incidence of GFIs has been estimated to be approximately 1- the proportion of the other types of infections between the two 2% (13). As previously reported across a large body of literature, group of patients, including urinary tract infections of bacterial through their glycosuric effect there is a higher risk of these types etiology (10.7% vs. 7.7%, p=0.4). on infections in SGLT2i users, with rates between 3 to 8% (13, As shown in Table 3, multivariate logistic regression analysis 15, 16). Several randomized controlled trials revealed a 3- to 4- showed no significant risk-adjusted odds of infections with the fold increase for SGLT2i compared with placebo (14, 17), which use of SGLT2i (OR 1.68, 95% CI 0.93-3.05; p=0.084). led manufacturers to include GFIs as common adverse reactions Kaplan-Meier survival curve on the occurrence of overall in the prescribing information of SGLT2i. Furthermore, infections is shown in Figure 2A (log-rank test= 0.02). observational real-world studies analyzing claims data revealed similar results. For example, a large population-based observational study, analyzing data of more than 40,000 patients, found a two-fold increase in the risk of GFIs with the Hepatic events use of SGLT2i in older diabetic patients (16). Our study is the first to corroborate these findings in a NAFLD cohort. Overall, patient who received SGLT2i showed a significantly Furthermore, the 15% rate of GFIs in NAFLD patients found lower incidence of hepatic events (1.5% vs. 10.7%; p=0.02) in our cohort is higher than those previously reported in although the proportion of cirrhosis was similar between SGLT2i-treated diabetic patients without NAFLD (17), which groups. Each individual decompensating event was less could point to a synergistic effect of NAFLD-related underlying frequent in SGLT2i treated patients (Table 2) although the immune dysfunction (1) added to the SGLT2i intrinsic increased difference did not reach statistical significance. Figure 2B risk of these infections. NAFLD indeed has been linked to shows the Kaplan-Meier curve for hepatic events (log-rank immune system malfunctioning (18), including test= 0.1). microcirculation disarrangements leading to diminished microbial clearance, undermined function of neutrophils and natural killer cells, and deficiency of vitamin-D levels, which in Mortality turn further impair innate immunity (19). Notwithstanding the increased risk of GFIs infection, No significant differences in overall mortality [1.5% vs 6.9%; treatment with SGLT2i was not associated with and overall p=0.1] were observed between groups. Out-of-hospital cardiac increase of infections. In contrast, HbA1c levels were arrest was the cause of death of the only patient who died in the independently associated with the risk of infections. This has SGLT2i group. In the non-SGLT2i group 5 patients died because been previously described in patients with T2DM (21–23). To of an infection (two of them of COVID-19) and one as a direct our knowledge, this is the first study that corroborates this complication of hepatocellular carcinoma. The other three died finding in NAFLD patients. As infections can precipitate acute of cardiac arrest, extrahepatic cancer and a motor decompensation and acute-on-chronic liver failure in cirrhotic vehicle collision. patients (24), improving glycemic control could reduce the risk of infection. Further studies are required to better understand the pathologic pathways and clinical implications of Discussion these findings. We found that patients with T2DM and comorbid NAFLD In this study we aimed at examining the frequency of receiving treatment with SGLT2i were less likely to present infections and hepatic events in patients with NAFLD and hepatic events. Though preliminary, these data suggest the T2DM treated with SGLT2i as compared to patients treated potential beneficial effects of SGLT2i in avoiding liver disease- with other antidiabetic agents. Although we did not find a related complications amongst patients with T2DM and Frontiers in Endocrinology 05 frontiersin.org Bañares et al. 10.3389/fendo.2022.945626 TABLE 1 Demographics and clinical characteristics of cases (SGLT2i users) and controls (non-SGLT2i users) in a cohort of patients with NAFLD and type 2 diabetes. Total (n = 195) SGLT2i users (n = 65) Non-SGLT2i users (n = 130) p-value Follow-up, median months (IQR) 34.4 (17-48.3) 34.4 (16.5-48.2) 34.4 (17.1-49) 0.5 Age, median (IQR) 62 (57-68) 61 (57-66) 63 (57-68) 0.3 ≥70 years, n (%) 40 (20.5) 10 (15.4) 30 (23) 0.2 Male sex, n (%) 104 (53.3) 35 (53.8) 69 (53.1) 0.9 Comorbidities Body Mass Index, median (IQR) 31.5 (28-34) 33.5 (29.7-36.2) 30.4 (27.3-34.2) 0.001 Obesity, n (%) 116 (61.4) 47 (74.6) 69 (54.7) 0.008* Hypertension, n (%) 145 (74.3) 51 (78.5) 94 (72.3) 0.3 Dyslipidemia, n (%) 152 (78.0) 55 (84.6) 97 (74.6) 0.1 Ischemic Heart Disease, n (%) 18 (9.2%) 5 (7.8%) 13 (10) 0.7 Heart Failure, n (%) 17 (8.7) 7 (10.7) 10 (7.7) 0.4 COPD, n (%) 24 (12.3) 12 (18.4) 12 (9.2) 0.06 CKD, n (%) 20 (10.2) 6 (9.2) 14 (10.7) 0.7 Child A cirrhosis, n (%) 65 (33.7) 21 (33.3) 44 (33.8) 0.9 SGLT2i Canagliflozin, n (%) – 9 (13.9) – Dapagliflozin, n (%) – 31 (47.7) – Empagliflozin, n (%) – 35 (35.4) – Other/No data, n (%) – 2 (3.1) – Concomitant T2DM treatment Insulin, n (%) 76 (39.0) 32 (49.2) 44 (33.9) 0.03 Metformin, n (%) 161 (82.6) 53 (81.5) 108 (83.1) 0.7 DDP-4 inhibitors, n (%) 53 (27.2) 18 (27.7) 35 (26.9) 0.9 GLP-1 receptor agonists, n (%) 50 (25.6) 26 (40.0) 24 (18.5) 0.001 Pioglitazone, n (%) 35 (18.0) 12 (18.5) 23 (17.7) 0.9 Sulfonylureas, n (%) 4 (2.1) 2 (3.1) 2 (1.5) 0.5 Metaglinides, n (%) 5 (2.6) 2 (3.1) 3 (2.3) 0.8 Number of antidiabetic drugs Monotherapy 49 (25.1) 0 (0.0) 49 (37.7) <0.001 2-drugs regime 68 (34.9) 13 (20.0) 55 (42.3) 0.002 3-drugs regime 52 (26.7) 30 (46.2) 22 (16.9) <0.001 4-drugs regime 19 (9.7) 16 (24.6) 3 (2.3) <0.001 5-drugs regime 7 (3.6) 6 (9.2) 1 (0.8) 0.002 Lipid-lowering agents Statins† 110 (56.4) 43 (66.2) 67 (51.5) 0.05 Fibrates† 35 (17.9) 18 (27.7) 17 (13.1) 0.1 Ezetimibe† 12 (6.2) 2 (3.1) 10 (7.7) 0.2 Other/No data 4 (2.1) 0 (0.0) 4 (3.1) 0.146 No treatment 58 (29.7) 13 (20.0) 45 (34.6) 0.03 Baseline Blood Tests HbA1c, median (QIR) 7.25 (6.7-8) 7.5 (6.9-8.5) 7 (6.5-8) 0.01 Glucose, median (IQR) 137 (110-167) 141 (114-167) 133 (108-165) 0.1 Platelets, median (IQR) 224 (161-273) 232 (183-271) 215 (152-274) 0.1 Creatinine, median (IQR) 0.79 (0.63-0.93) 0.8 (0.63-0.95) 0.77 (0.62-0.92) 0.5 Albumin, median (IQR) 4.3 (4.1-4.5) 4.3 (4.1-4.5) 4.3 (4.1-4.5) 0.5 INR, median (IQR) 0.97 (0.92-1.04) 0.96 (0.93-1.02) 0.98 (0.92-1.04) 0.6 AST (GOT), median (IQR) 36 (24-52) 35 (24-52) 36 (24-51) 0.8 ALT (GPT), median (IQR) 40 (26-56) 39 (25-56) 40 (27-60) 0.6 (Continued) Frontiers in Endocrinology 06 frontiersin.org Bañares et al. 10.3389/fendo.2022.945626 TABLE 1 Continued Total (n = 195) SGLT2i users (n = 65) Non-SGLT2i users (n = 130) p-value GGT, median (IQR) 72 (37-137) 67 (34-132) 72 (38-147) 0.2 Alkaline phosphatase, median (IQR) 85 (69-114) 81 (62-100) 89 (72-123) 0.01* FIB-4, median (IQR) 1.66 (1.16-2.9) 1.56 (1.16-2.6) 1.68 (1.16-3.17) 0.3 LSM, median kPa (IQR) Overall 10 (6.8-18.3) 10.1 (7.7-16.6) 10 (6.4-18.6) 0.6 Cirrhotics 19.8 (15.6-34.9) 19.8 (15-27) 20.4 (17-43.2) 0.4 † Lipid-lowering agents alone or in combination with other drug classes. ALT, alanine transferase; AST, aspartate transferase; GGT, gamma glutamyl transferase; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; HbA1c, glycosylated hemoglobin; LSM, liver stiffness measurement; SGLT2i, sodium-glucose co-transporter-2 inhibitors. * means statistical significance at, p < 0.05. NAFLD. Translational research in animal models and clinical while SLGT2i undoubtedly increase GFIs, by improving the studies has shown that SLGT2i use is related to improved metabolic control the overall risk of infection could decrease. hepatic metabolism (9). Its use has been associated with This balancing situation presents a trade-off for physicians and reductions in hepatic steatosis measured by controlled patients between metabolic control and adverse effects. attenuation parameter, MRI-proton density fat fraction or Cirrhotic patients, especially those that have already magnetic resonance spectroscopy (7), as well as liver stiffness presented decompensations, are of special interest regarding ameliorations in patients with previous significant fibrosis (8). the role of SGLT2i because their combined glycosuric and In a single-arm pilot study with paired biopsies demonstrated natriuric effects might be beneficial in patients with that SGLT2i might induce histological improvements in NASH hypervolemia and renal dysfunction such as those with features without a worsening of fibrosis (10, 25). Moreover, a hepatorrenal syndrome (26). Nonetheless, SGLT2i are seldom small randomized controlled trial found that SGLT2i treatment prescribed in patients with advanced liver disease. Preliminary resulted in reduced liver fat content by MRI-derived proton findings by Saffo, Garcia-Tsao and Tadei in a cohort of seventy- density fat fraction and improvements in liver enzymes, eight patients with cirrhosis treated with SGLT2i, including 39 compared to placebo (12). Several ongoing clinical trials are (50%) with non-alcoholic steatohepatitis and 63 (81%) with currently assessing the effects of SGLT2i in patients with compensated disease, rates of hepatic decompensation and NAFLD and NASH, which may provide more definitive mortality were not increased compared to most published answers. Parallel to their well-known beneficial cohorts of cirrhotics, and SGLT2i use was not identified as a cardiovascular and renal effects (5), SLGT2i may exert cause of decompensation (27). Whether administering SGLT2i additional hepatic protection, which could be a cornerstone before first decompensation or as earlier as possible in both in the treatment of T2DM patients with NAFLD. Moreover, diabetic and non-diabetic patients might be not only safe TABLE 2 Outcomes in cases (SGLT2i users) and controls (non- SGLT2i users) in a cohort of patients with NAFLD and type 2 diabetes. Total (n = 195) SGLT2i users (n = 65) Non-SGLT2i (n = 130) p-value Overall infection (total), n (%) 66 (39) 27 (41.5) 39 (30) 0.1 Respiratory tract infection, n (%) 16 (8.2) 5 (7.7) 11 (8.4) 0.9 Abdominal infection, n (%) 4 (2) 2 (3.1) 2 (1.5) 0.6 Bacterial UT infection, n (%) 17 (8.7) 7 (10.7) 10 (7.7) 0.4 Mycotic UT/genital infection, n (%) 15 (7.7) 10 (15.4) 5 (3.8) 0.008 Skin and Soft tissue infection, n (%) 4 (2) 1 (1.5) 3 (2.3) 1 Bacteremia, n (%) 4 (2) 1 (1.5) 3 (2.3) 1 COVID-19, n (%) 2 (1) 1 (1.5) 1 (1) 1 SBP, n (%) 1 (1.9) 0 (0) 1 (2.5) 1 Hepatic Events, n (%) 15 (7.7) 1 (1.5) 14 (10.7) 0.02 Ascites, n (%) 10 (5.1) 1 (1.5) 9 (6.2) 0.1 HE, n (%) 6 (3.1) 0 (0) 6 (4.6) 0.1 UGB, n (%) 7 (13.2) 0 (0) 5 (17.5) 0.1 Mortality, n (%) 10 (5.1) 1 (1.5) 9 (6.9) 0.1 SGLT2i, Sodium-glucose co-transporter-2 inhibitors; UT, urinary tract; COVID-19, Coronavirus disease 2019; SBP, Spontaneous bacterial peritonitis; HE, hepatic encephalopathy UGB, upper gastrointestinal tract bleeding. Frontiers in Endocrinology 07 frontiersin.org Bañares et al. 10.3389/fendo.2022.945626 TABLE 3 Multivariate logistic regression analysis for risk of infection. Variable Adjusted HR (95% CI) p-value SGLT2i 1.68 (0.93-3.05) 0.084 HbA1c 1.09 (0.87-1.36) 0.43 Obesity† 1.35 (0.70-2.59) 0.36 Platelets 1.0 (0.99-1.004) 0.39 Age 1.02 (0.98-1.07) 0.17 Sex (males) 0.95 (0.52.1.74) 0.89 HbA1c, glycosylated hemoglobin; SGLT2i, Sodium-glucose co-transporter-2 inhibitors †(BMI >30 kg/m2). A B FIGURE 2 Kaplan-Meier survival curves. (A) Overall infections; (B) Hepatic events. but beneficial as suggested by our findings warrants Conclusion further investigation. When compared with patients with NAFLD and T2DM without SGLT2i treatment, patients receiving SGLT2i were more likely to present genitourinary fungal infections but not other Limitations types of infections. In addition, patients receiving SGLT2i had a lower occurrence of hepatic events. Further studies are required This study is not without limitations and results should be to confirm these results. interpreted in the context of the study design. First, we performed a case-control study, which is prone to selection bias thereby limiting the validity of the results presented. In the case of our study, cases had higher rates of obesity, were using Data availability statement more drugs for T2DM, had more insulin use and also statin, and had also higher glycated hemoglobin, which might contribute to The raw data supporting the conclusions of this article will both infections and hepatic events. Second, the collection of be made available by the authors, without undue reservation. relevant data was retrospective. Third, the relatively small sample size constrains the robustness of multivariate analyses. Fourth, the potential effects of medications other than SGLT2i (e.g., GLP1 receptor agonists) were not addressed in our analysis. Ethics statement Despite our limitations, we provide data about infectious and metabolic outcomes that are relevant to clinical practice. The studies involving human participants were reviewed and Nevertheless, these results should be interpreted in the context approved by Vall d’Hebron University Hospital Ethics of hypothesis generating research and should be tested in Committee. The patients/participants provided their written subsequent, larger, studies. informed consent to participate in this study. Frontiers in Endocrinology 08 frontiersin.org Bañares et al. 10.3389/fendo.2022.945626 Author contributions European Commission/EFPIA IMI2 853966-2, IMI2 777377, H2020 847989, and ISCIII PI19/01898. Conceptualization and design: JB, JP; Data collection: JB, AP, The remaining authors declare that the research was JR-E, LC-R, AV, LR-O, MS; Data analysis: RM-N, MP; conducted in the absence of any commercial or financial Interpretation: JR-E, AC, MS, VV, JG, JP; First manuscript relationships that could be construed as a potential conflict drafting: JB, RM-N; Revision and acceptance of last version: of interest All authors; Supervision: JP. All authors contributed to the article and approved the submitted version. Publisher’s note Acknowledgments All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated Preliminary analyses related to the present study were organizations, or those of the publisher, the editors and the presented during the 46 th Spanish Conference of the reviewers. Any product that may be evaluated in this article, or Association for the Study of the Liver (AEEH, Abstract 01213, claim that may be made by its manufacturer, is not guaranteed June 14-16, 2021, Madrid, Spain). or endorsed by the publisher. Conflict of interest JP reports having received consulting fees from Boehringer Ingelheim and Novo Nordisk. He has received speaking fees Supplementary material from Gilead, and travel expenses from Gilead, Rubió , Pfizer, Astellas, MSD, CUBICIN, and Novo Nordisk. He has received The Supplementary Material for this article can be found educational and research support from Gilead, Pfizer, Astellas, online at: https://www.frontiersin.org/articles/10.3389/ Accelerate, Novartis, Abbvie, ViiV, and MSD. Funds from fendo.2022.945626/full#supplementary-material References 1. Eslam M, Sanyal AJ, George JInternational Consensus Panel. MAFLD: A type 2 diabetes mellitus. Dig Dis Sci (2020) 65(2):623–31. doi: 10.1007/s10620-019- consensus-driven proposed nomenclature for metabolic associated fatty liver disease. 5477-1 Gastroenterology (2020) 158(7):1999–2014. doi: 10.1053/j.gastro.2019.11.312 11. Mantovani A, Petracca G, Csermely A, Beatrice G, Targher G. Sodium-glucose 2. Powell EE, Wong VW, Rinella M. Non-alcoholic fatty liver disease. Lancet. cotransporter-2 inhibitors for treatment of nonalcoholic fatty liver disease: A meta-analysis (2021) 397(10290):2212–24. doi: 10.1016/S0140-6736(20)32511-3 of randomized controlled trials. Metabolites. (2020) 11(1):22. doi: 10.3390/metabo11010022 3. Younossi ZM. Non-alcoholic fatty liver disease - A global public health 12. Kuchay MS, Krishan S, Mishra SK, Farooqui KJ, Singh MK, Wasir JS , et al. perspective. J Hepatol (2019) 70(3):531–44. doi: 10.1016/j.jhep.2018.10.033 Effect of empagliflozin on liver fat in patients with type 2 diabetes and nonalcoholic fatty liver disease: A randomized controlled trial (E-LIFT trial). Diabetes Care 4. Harrison SA, Gawrieh S, Roberts K, Lisanti CJ, Schwope RB, Cebe KM, et al. (2018) 41(8):1801–8. doi: 10.2337/dc18-0165 Prospective evaluation of the prevalence of non-alcoholic fatty liver disease and steatohepatitis in a large middle-aged US cohort. J Hepatol (2021) 75(2):284–91. 13. Nichols GA, Brodovicz KG, Kimes TM, Dé ruaz-Luyet A, Bartels DB. doi: 10.1016/j.jhep.2021.02.034 Prevalence and incidence of urinary tract and genital infections among patients with and without type 2 diabetes. J Diabetes Complications. (2017) 31(11):1587–91. 5. Kalyani RR. Glucose-lowering drugs to reduce cardiovascular risk in type 2 doi: 10.1016/j.jdiacomp.2017.07.018 diabetes. N Engl J Med (2021) 384(13):1248–60. doi: 10.1056/NEJMcp2000280 14. Li D, Wang T, Shen S, Fang Z, Dong Y, Tang H. Urinary tract and genital 6. Storgaard H, Gluud LL, Bennett C, Grøndahl MF, Christensen MB, Knop K, infections in patients with type 2 diabetes treated with sodium-glucose co- et al. Benefits and harms of sodium-glucose Co-transporter 2 inhibitors in patients transporter 2 inhibitors: A meta-analysis of randomized controlled trials. with type 2 diabetes: A systematic review and meta-analysis. PloS One (2016) 11 Diabetes Obes Metab (2017) 19(3):348–55. doi: 10.1111/dom.12825 (11):e0166125. doi: 10.1371/journal.pone.0166125 15. McGovern AP, Hogg M, Shields BM, Sattar NA, Holman RR, Pearson ER, 7. Dougherty JA, Guirguis E, Thornby KA. A systematic review of newer et al. Risk factors for genital infections in people initiating SGLT2 inhibitors and antidiabetic agents in the treatment of nonalcoholic fatty liver disease. Ann their impact on discontinuation. BMJ Open Diabetes Res Care (2020) 8(1): Pharmacother. (2021) 55(1):65–79. doi: 10.1177/1060028020935105 e001238. doi: 10.1136/bmjdrc-2020-001238 8. Shimizu M, Suzuki K, Kato K, Jojima T, Iijima T, Murohisa T, et al. 16. Lega IC, Bronskill SE, Campitelli MA, Guan J, Stall NM, Lam K, et al. Evaluation of the effects of dapagliflozin, a sodium-glucose co-transporter-2 Sodium glucose cotransporter 2 inhibitors and risk of genital mycotic and urinary inhibitor, on hepatic steatosis and fibrosis using transient elastography in tract infection: A population-based study of older women and men with diabetes. patients with type 2 diabetes and non-alcoholic fatty liver disease. Diabetes Obes Diabetes Obes Metab (2019) 21(11):2394–404. doi: 10.1111/dom.13820 Metab (2019) 21(2):285–92. doi: 10.1111/dom.13520 17. Engelhardt K, Ferguson M, Rosselli JL. Prevention and management of genital 9. Sumida Y, Yoneda M, Tokushige K, Kawanaka M, Fujii H, Yoneda M, et al. mycotic infections in the setting of sodium-glucose cotransporter 2 inhibitors. Ann Antidiabetic therapy in the treatment of nonalcoholic steatohepatitis. Int J Mol Sci Pharmacother. (2021) 55(4):543–8. doi: 10.1177/1060028020951928 (2020) 21(6):1907. doi: 10.3390/ijms21061907 18. Nseir W, Taha H, Khateeb J, Grosovski M, Assy N. Fatty liver is associated 10. Lai LL, Vethakkan SR, Nik Mustapha NR, Mahadeva S, Chan WK. with recurrent bacterial infections independent of metabolic syndrome. Dig Dis Sci Empagliflozin for the treatment of nonalcoholic steatohepatitis in patients with (2011) 56(11):3328–34. doi: 10.1007/s10620-011-1736-5 Frontiers in Endocrinology 09 frontiersin.org Bañares et al. 10.3389/fendo.2022.945626 19. Adenote A, Dumic I, Madrid C, Barusya C, Nordstrom CW, Rueda Prada L. glucose metabolism of non-alcoholic fatty liver disease complicated by diabetes NAFLD and infection, a nuanced relationship. Can J Gastroenterol Hepatol (2021) mellitus. Hepatol Res (2019) 49:531–9. doi: 10.1111/hepr.13304 2021:5556354. doi: 10.1155/2021/5556354 26. Saffo S, Taddei T. SGLT2 inhibitors and cirrhosis: A unique perspective on 20. Association AD. 2. classification and diagnosis of diabetes: Standards of the comanagement of diabetes mellitus and ascites. Clin Liver Dis (Hoboken) (2018) medical care in diabetes–2021. Diabetes Care (2021) 44(Suppl 1):S15. doi: 10.2337/ 11(6):141–4. doi: 10.1002/cld.714 dc21-S002 27. Saffo S, Garcia-Tsao G, Taddei T. SGLT2 inhibitors in patients with 21. Mor A, Dekkers OM, Nielsen JS, Beck-Nielsen H, Sørensen HT, Thomsen cirrhosis and diabetes mellitus: A tertiary center cohort study and insights about RW. Impact of glycemic control on risk of infections in patients with type 2 a potential therapeutic target for portal hypertension. J Diabetes. (2021) 13(3):265– diabetes: A population-based cohort study. Am J Epidemiol. (2017) 186(2):227–36. 9. doi: 10.1111/1753-0407.13136 doi: 10.1093/aje/kwx049 22. Critchley JA, Carey IM, Harris T, DeWilde S, Hosking FJ, Cook DG. Glycemic control and risk of infections among people with type 1 or type 2 diabetes in a Large COPYRIGHT primary care cohort study. Diabetes Care (2018) 41(10):2127–35. doi: 10.2337/dc18-0287 © 2022 Bañares, Manzano-Nuñez, Prio,́ Rivera-Esteban, Camps-Relats, 23. Carey IM, Critchley JA, DeWilde S, Harris T, Hosking FJ, Cook DG. Risk of Villarejo, Ruiz-Ortega, Pons, Ciudin, Salcedo, Vargas, Genescà and Pericàs. infection in type 1 and type 2 diabetes compared with the general population: A This is an open-access article distributed under the terms of the Creative matched cohort study. Diabetes Care (2018) 41(3):513–21. doi: 10.2337/dc17-2131 Commons Attribution License (CC BY). The use, distribution or 24. Villanueva C, Albillos A, Genescà J, Garcia-Pagan JC, Brujats A, Calleja JL, et al. reproduction in other forums is permitted, provided the original author Bacterial infections adversely influence the risk of decompensation and survival in (s) and the copyright owner(s) are credited and that the original compensated cirrhosis. J Hepatol (2021) 75(3):589–99. doi: 10.1016/j.jhep.2021.04.022 publication in this journal is cited, in accordance with accepted 25. Akuta N, Kawamura Y, Watanabe C, Nishimura A, Okubo M, Mori Y, et al. academic practice. No use, distribution or reproduction is permitted Impact of sodium glucose cotransporter 2 inhibitor on histological features and which does not comply with these terms. Frontiers in Endocrinology 10 frontiersin.org