Acute Liver Injury Induced by Sitagliptin Report of Two Cases and Review of Literature

Introduction

The prevalence of diabetes mellitus (DM) has continuously increased worldwide, and the number of affected patients is expected to exceed 600 million by ii,040 (Forouzanfar et al., 2016). This epidemic of DM can be attributed to population aging, dietary preferences, lifestyle alterations, and even genetic/epigenetic predisposition (Gaulton, 2017). A major complication of DM is renal dysfunction, consisting of albuminuria, chronic kidney disease (CKD), or terminate-stage renal illness (ESRD), and renal failure accounts for 10–20% of deaths amid patients with DM (Afkarian et al., 2016). For diabetic patients developing CKD, dosage adjustment is frequently needed for their anti-diabetic regimens, due to the fact that drugs and their metabolites frequently accumulate in patients with CKD, predisposing them to adverse events. Conventional oral anti-diabetic agents, such as metformin, are contraindicated in patients with eGFR levels below xxx–45 ml/min/one.73 thousand², whilst sulfonylurea carries the risk of prolonged hypoglycemia among these patients. Options of oral anti-diabetic agents (OADs) for diabetic patients with advanced CKD are thus limited; co-ordinate to the updated guideline from the American Diabetic Clan, only meglitinides, thiazolidinediones, and dipeptidyl peptidase-iv inhibitors (DPP4i) are suitable for oral glucose reduction in patients with an eGFR of <thirty ml/min/1.73 m² (American Diabetes Clan, 2018). However, each of the to a higher place OADs has its own drawbacks, including meglitinide-related hypoglycemia and myocardial infarction (Lin et al., 2016; Wu et al., 2017b), thiazolidinediones-related eye failure exacerbation and edema, and the high price associated with DPP4i (Roussel et al., 2015). Although insulin may exist the treatment of choice for these patients, studies have suggested that OAD users in combination with insulin had a significantly lower bloodshed than those using insulin lone (Hsiao et al., 2017). Consequently, information technology is imperative that our armamentarium of OADs against DM be expanded to allow safer approaches for glycemic control in diabetic CKD/ESRD patients.

Acarbose, a competitive inhibitor of intestinal α-glucosidase, reduces post-prandial carbohydrate digestion and absorption, effectively lowering glycated hemoglobin. Apart from common side effects like gastrointestinal disturbances, early trials involving acarbose reported increases in transaminases or rarely jaundice in iii.8% of users compared to 1.one% in the placebo grouping (Balfour and McTavish, 1993). This is attributed predominantly to the aggregating of acarbose and its metabolites, just little evidence exists to back up such idea. Following the wide spread use of acarbose in DM patients more than 2 decades ago, instance reports of acarbose-related liver injury started to emerge. The first case was a 65-year-quondam lady with blazon 2 DM receiving acarbose 300 mg/day for 2 months (Andrade et al., 1996). The predominant manifestations were angst and anorexia, accompanied by laboratory changes of hepatocellular damage pattern, with spontaneous recovery after discontinuing acarbose. Subsequent instance series suggested that reports of acarbose-related liver toxicity by and large came from Japan and Spain, had an boilerplate latency of 2–6 months after therapy initiation, frequently recurred afterward re-challenge, and the associated laboratory abnormalities were hepatocellular injury predominantly (Hsiao et al., 2006). In contrast to traditional hepatotoxic medications such as acetaminophen and ethanol, whose liver toxicity is mediated through hepatic microsomal enzyme activation, acarbose may not crusade hepatotoxicity through its metabolites. Putative mechanisms include the consecration of CYP2E1 enzyme and increasing reactive oxygen species product and the influence of sexual practice hormone, merely the exact mechanism withal remains unclear due to the low and unpredictable incidence of such injury (Wang et al., 1999; Hsiao et al., 2006).

Given the lack of experiences of using acarbose in patients with advanced CKD/ESRD, and the pharmacokinetic concerns in these patients (Charpentier et al., 2000), acarbose has been contraindicated in this population (International Diabetes Federation Guideline Development Grouping, 2014; American Diabetes Association, 2018). However, whether these recommendations agree truthful decades after acarbose enters the market is unclear. We hypothesized that acarbose use might not exist associated with an increased risk of liver toxicity among diabetic patients with severe renal insufficiency. Since acarbose has non been reported as being significantly removed by dialysis in the existing literature, we think that the dosage recommendation for acarbose will not differ betwixt those with not-dialysis stage 5 CKD and dialysis-dependent ESRD. Furthermore, there is currently no evidence suggesting that dialysis influences the take chances of developing hepatotoxicity. Thus, we chose these patients who traditionally accept the fewest options of OADs, for analyzing whether acarbose use was associated with an increased risk of developing liver injury, using a retrospective population-based cohort approach. To avoid the possibility of effect dilution by the presence or absence of liver disease in these patients, nosotros planned to analyse whether this association appeared in participants without or with unlike severities of liver diseases.

Methods

Ethical Approval

The electric current study has been canonical by the institutional review lath of National Taiwan University Hospital (No. 201503028W), and its protocol adheres to the proclamation of Helsinki. Informed consent was waived due to information anonymity.

Participant Enrollment and Data Sources

All adults (≥20 years old) diagnosed with DM co-ordinate to the International classification of affliction 9th revision—clinical modification [ICD-9-CM] codes 250.x and defined as having ≥2 discrete out-patient clinics or ≥i hospitalization with a diagnosis of DM were systemically identified from the National Health Insurance Research Database between the report period, i.e., January i, 2007 to December 31, 2013 (Effigy i). Considering the National Health Insurance provides healthcare coverage for ≥99% Taiwanese citizens and ≥97% hospitals or clinics are in contract with this national insurance organisation, the database contains unequivocally the most comprehensive records of healthcare service in this state. The index date was divers as the moment when patients reached the pre-divers criteria of having DM.

Effigy 1. Menses chart of the electric current study. CKD, chronic kidney disease; DM, diabetes mellitus; ESRD, stop-stage renal illness.

Participants with DM and non-dialysis stage 5 CKD or dialysis-dependent ESRD were selected later excluding (1) those with missing data, with an index appointment before January 1, 2008 (to reassure incident DM cases) or afterward June 30, 2013 (to allow at least 6 months of follow-up); (2) those with a history of hepatotoxic injuries (the issue of involvement) before the index date; (3) or those later on assigned to not-users but started acarbose during follow-up. We used criteria validated past the existing literature to screen for those with non-dialysis phase v CKD, consisting of those having a master diagnosis of CKD (ICD-ix-CM 585.ten, except 585.six) in ≥2 detached out-patient clinics or ≥i hospitalization, and receiving erythropoiesis-stimulating agent (ESA) treatment concurrently with the CKD diagnosis (Hsau et al., 2014; Hung et al., 2015). In Taiwan, ESA can only be initiated in CKD patients with a serum creatinine higher than 6 mg/dL according to the National Health Insurance policy, and the utilize of this feature to identify not-dialysis phase 5 CKD patients in Taiwan accept been found to exist very accurate by others (Hsau et al., 2014; Shih et al., 2014; Hung et al., 2015). The identification of those with finish-stage renal disease under chronic dialysis, or stage 5D CKD patients, also followed the approach detailed in the existing literature, based on the presence of ICD-ix-CM code 585.6 and the co-existence of official certificates of having end-phase renal illness under chronic dialysis, issued by the National Health Insurance Bureau of Taiwan (Chao et al., 2017; Lin et al., 2017; Wu et al., 2017a). The official certificates are called "catastrophic affliction certificate," serving every bit a firm proof of the patients' disease status, and are regularly audited and maintained in governmental database (chosen "Registry of Catastrophic Affliction Database") (Lin et al., 2014, 2017; Wu et al., 2017a).

Participant Categorization and Grouping

Based on existing studies, nosotros later on divided diabetic participants with non-dialysis stage five CKD or ESRD into those with (1) cirrhosis, defined by the presence of codes 571.ii, 571.5, or 571.6 in their healthcare records and the accompanying catastrophic illness certificates; (2) with chronic liver disease (CLD), including hepatitis B [070.2, 070.iii, and V02.61], hepatitis C [070.7, 070.41, 070.44, 070.51, 070.54, and V02.62], or alcoholic liver disease [291, 303, 305.0, 357.5, 425.5, 571.0, 571.one, 571.2, 571.3, 980.0, and V11.3]) but without cirrhosis; and (3) those without CLD (without any of the above codes) (Figure 1; Lin et al., 2014; Setiawan et al., 2016; Fukui et al., 2017). The spectrum of CLD traditionally includes chronic hepatitis (B, C, or non-B, non-C), booze-related liver illness, not-alcoholic fat liver affliction, cirrhosis, and hepatocellular carcinoma (Byrd et al., 2015). Population-based surveys consistently show that chronic hepatitis B, C, and alcoholic liver disease account for well-nigh CLD cases in different countries, including Asian ones (Fung et al., 2007; Bell et al., 2008). Furthermore, amid those with DM, hepatitis C and chronic hepatitis have been found to exist the master CLD etiologies (Byrd et al., 2015). Consequently, we defined CLD using a combination of diagnostic codes of chronic hepatitis B, C, and alcoholic liver disease. In addition, those with cirrhosis were specifically selected for their college liver disease severity, due to their associated complications such as portosystemic shunts which significantly increases mortality than those without cirrhosis or CLD (Simón-Talero et al., 2018). Several reports besides ostend the utility of our approach, dividing participants into those without liver disease, with non-cirrhotic liver disease, and with cirrhosis, to classify liver affliction severity (Saray et al., 2012; Montomoli et al., 2013).

We examined the healthcare records of all participants during the study catamenia and extracted their medical diagnoses, comorbidities, and medication usage and calculated their Charlson comorbidity indices (Chao et al., 2012, 2017). Comorbidities were defined using the same criteria described above. Users of medications with potential hepatotoxic take chances were defined equally receiving more thirty days inside the preceding year before index date, with the medication list defined according to literature and Taiwan adverse drug report (ADR) reporting system (Lee, 2013; Chen et al., 2014).

Because the severity of DM can potentially influence the take chances of developing hepatic injury (Adams et al., 2017), it is important to remainder this feature within the current cohort. Nosotros divers diabetic severity using the adjusted diabetes complications severity index (aDCSI), a widely used framework based on claims data without laboratory parameters (Chang et al., 2012). In brief, aDCSI was derived after totaling the severity (aDCSI scores; range 0–thirteen) or the number of 7 diabetic complications (aDCSI counts; range 0–vii). The results exhibited skilful associations with patient outcomes (Chang et al., 2012).

Exposures and Principal Outcomes

The main exposure of this study is acarbose use prior to enrollment, defined as receiving more than than thirty days within the preceding year earlier index date. A sensitivity analysis excluding users who discontinued acarbose later on index date, defined as those who did non receive whatsoever acarbose prescription after enrollment, was also carried out.

The main outcome of this written report was the development of drug-induced liver injuries (DILI) during follow-upwardly. According to international consensus, DILI should be diagnosed based on clinical chemistry criteria, after excluding other diseases and being implicated as plausible candidates (Aithal et al., 2011). Still, DILI events can be very rare, and using information from any unmarried institute may suffer from a low yield rate, while using the administrative database for identifying DILI cases may accept the advantage of higher yield rate. We divers the case as the presence of whatever of the following diagnosis based on the literature: bilirubin excretion disorders (277.4), toxic hepatitis (573.iii), other specified or unspecified hepatic disorder (573.viii, 573.ix), acute/subacute hepatic necrosis (570), other biliary tract disorders (576.8), and non-neonatal unspecified jaundice (782.4), using code combinations deriving from prior reports (Kao et al., 2014; Chang et al., 2015). Based on the existing reports of acarbose-related liver injury (Kao et al., 2016), bold that the incidence of whatever liver injury event in the control during follow-up was 2–three%, with a significance level of 0.05 and a power of 0.8, at least 1,859 patients would exist needed per group.

Statistical Analysis

We showtime grouped the identified diabetic participants with stage v CKD or dialysis-dependent ESRD into those without CLD, with CLD only without cirrhosis, and with cirrhosis. After selecting all acarbose users from each group, propensity score-matched non-acarbose users with a 1:2 ratio were further selected for subsequent analysis, with propensity scores synthetic based upon logistic regression incorporating age, sex, comorbidities, diabetic severity, and concurrent medications. Matching was done using the nearest neighbor approach, which was constitute to increase precision in accomplice studies (Rassen et al., 2012). Nosotros compared the baseline clinical features betwixt users and not-users using contained t-tests. The incidence density of liver injuries was estimated by dividing event counts with follow-up lengths (Chen et al., 2014). Nosotros then used Cox proportional adventure modeling to summate the hazard ratios (HRs) with 95% confidence intervals (CIs) of acarbose employ regarding liver injuries incidence in each grouping with different severities of underlying hepatic diseases. Kaplan–Meier method was used to construct issue-gratis curves in each group, with comparisons done using a log-rank test. All statistical analyses were done using SAS software (SAS institute, Cary, NC, USA), and p-values lower than 0.05 were considered significant.

Results

Among all adults with a DM diagnosis during the study period (n = ii,036,531), 47,643 patients with incident DM and phase 5 CKD or ESRD were identified and subsequently divided into those without CLD (84.ii%), with CLD simply without cirrhosis (ten.4%), and with cirrhosis (5.4%) (Effigy 1). A total of 17.5, 16.9, and 14.five% were acarbose users in the without, with CLD, and cirrhotic groups, respectively. We selected a one:ii propensity score-matched ratio of users and non-users from each grouping for subsequent analysis. No pregnant differences in age, sex, all comorbidities, Charlson comorbidity index, DM severity, and other potential hepatotoxic medication were noted betwixt acarbose users and non-users among the 3 groups, except hypertension (Tabular array 1). The prevalence of hepatic, biliary, and pancreatic cancer amidst the three groups of patients were relatively low (without, 0.1%; balmy, 5.vi%; astringent, 17.9%) and did not differ between users and not-users.

Tabular array ane. Baseline characteristics of study participants after the procedure of propensity-score matching.

After a mean two.iv years of follow-up, 661 events of liver injury occurred during 55,517.v person-yr, yielding an incidence density of 11.9 per 1,000 person-year (Tabular array 2). Amid those without CLD, the incidence of liver injury was similar among acarbose users and non-users (HR: 1.05; 95% CI: 0.88–i.25). Afterward adjusting for demographic profiles, comorbidities, potentially hepatotoxic medication use, and aDCSI, acarbose users nevertheless exhibited a similar risk of liver injury compared to not-users (HR: 1.04–1.05) (Table 2). Among those with CLD merely without cirrhosis, acarbose users likewise exhibited a like hazard of liver injury compared to not-users (Hour: 0.97–1.00). Like results were obtained in those with cirrhosis (Table 2). We farther divided the severity of cirrhosis into those with unproblematic cirrhosis (without esophageal varices [EV] or ascites), with compensated cirrhosis (with either EV or ascites), and those with decompensated cirrhosis (with both EV and ascites), co-ordinate to past studies (Ratib et al., 2015). We found that acarbose use was non associated with a higher risk of liver injury in those with elementary, compensated, and decompensated cirrhosis than non-users (for uncomplicated, HR 0.71, 95% CI 0.3–0.65; for compensated, HR 0.three, 95% CI 0.08–1.36; for decompensated, Hr 1.94, HR 0.17–22) (Table iii). The cumulative hazard curves for liver injury using the Kaplan–Meier method showed that acarbose use was associated with a similar risk of liver injury compared to non-users amongst diabetic patients with advanced CKD/ESRD without CLD (p = 0.59), with CLD without cirrhosis (p = 0.94), and with cirrhosis (p = 0.45; Effigy two).

Table 2. Incidence and hazard ratio of hepatotoxicity events amongst blazon 2 diabetic participants with not-dialysis stage 5 CKD and Dialysis-dependent ESRD after matching.

Table iii. Analyses focusing simply on those with cirrhosis.

Figure 2. Kaplan–Meier curves for cumulative hepatotoxicity hazard during follow-upward amongst patients without CLD (A), with CLD but without cirrhosis (B), and with cirrhosis (C). CLD, chronic liver disease.

We used defined daily dose (DDD) between 1 yr prior to index date and the cease of follow-up, to gauge the dose-response relationship. Categorizing the daily dose into three tertiles, we found that none of the acarbose users from each DDD tertile had an increased hazard of liver injury event compared to non-users (Table 4). We also subdivided participants into those using acarbose as monotherapy and those using acarbose combined with other OADs for analyses. We found that amidst those using acarbose as monotherapy, users of the non-CLD, CLD without cirrhosis, and the cirrhosis groups were not at an increased risk of liver injury compared to non-users (for not-CLD, HR 0.88, 95% CI 0.56–1.38; for CLD without cirrhosis, HR 0.38, 95% CI 0.ane–1.46; for cirrhosis, HR could not be calculated due to no event in the user group) (Tabular array five). Among those who used acarbose with other OADs, the results were unaltered (for non-CLD, Hour one.08, 95% CI 0.9–i.31; for CLD without cirrhosis, 60 minutes ane.two, 95% CI 0.72–two.01; for cirrhosis, 60 minutes 0.8, 95% CI 0.36–1.75; Table five).

Table four. Using defined daily dose (DDD) for evaluating the dose-outcome relationship.

Table 5. Results from dividing users into acarbose monotherapy or those using acarbose combined with other OADs.

We performed several sensitivity analyses to verify our findings. Nosotros first excluded acarbose users who discontinued acarbose afterwards enrollment, and the results were essentially the same (Table 6). After further polishing the definition of liver injury by excluding those with alternative owing diagnoses including autoimmune hepatitis (571.42) and shock (785.5x), we found that amongst patients without CLD, with CLD but without cirrhosis, or with cirrhosis, acarbose use was not associated with an increased chance of liver injury (for non-CLD, Hr 1.09, 95% CI 0.93–1.28; for CLD without cirrhosis, 60 minutes 0.84, 95% CI 0.56–1.28; for cirrhosis, HR 0.61, 95% CI 0.29–one.27) (Table 7). We also narrowed downwards outcome definition by including simply those with DILI who discontinued acarbose immediately afterwards DILI occurred, and derived like findings (for non-CLD, Hr 0.88, 95% CI 0.73–1.06; for CLD without cirrhosis, HR 0.88, 95% CI 0.54–1.44; for cirrhosis, Hr 0.58, 95% CI 0.25–1.38; Table 8). Finally, nosotros modified our definition of acarbose exposure past using stricter definitions of acarbose exposure, focusing on those receiving more than 30 days of acarbose within 6 or 3 months preceding the index date. We found that if we used the definition of half-dozen-month, acarbose users were not at an increased risk of liver injury than not-users amongst the 3 groups of patients (for not-CLD, 60 minutes 0.97, 95% CI 0.74–1.28; for CLD without cirrhosis, 60 minutes ane.90, 95% CI 0.88–4.10; for cirrhosis, HR 0.89, 95% CI 0.22–3.75; Table 9). Using 3-month definition, acarbose users were all the same not at an increased risk of liver injury than not-users among the three groups of patients (for not-CLD, Hour 1.04, 95% CI 0.81–1.32; for CLD without cirrhosis, Hr 1.15, 95% CI 0.58–2.30; for cirrhosis, HR 0.61, 95% CI 0.17–two.26; Table 9).

Tabular array half dozen. Sensitivity analyses excluding acarbose users who discontinued acarbose after enrollment.

Table seven. Sensitivity analyses excluding participants with culling attributable diagnoses (propensity-score re-matched).

Tabular array 8. Sensitivity analyses in which merely those who discontinued acarbose immediately after developing liver injury were selected in the acarbose group.

Tabular array 9. Sensitivity analyses using different definitions of acarbose exposure.

Discussion

Based on a nation-wide cohort followed up prospectively, nosotros found that among diabetic patients with phase five CKD or ESRD, the utilise of acarbose was non associated with an increased risk of liver injury. This finding clearly illustrates that the theoretical pharmacokinetic concerns of acarbose utilise in these patients practise not occur in the real-world setting, and it is probable that nosotros might be able to administer acarbose safely even in diabetic patients with both renal and hepatic dysfunction, a subgroup in need of more therapeutic options confronting hyperglycemia.

Co-ordinate to past reports, a latency menstruum from 2 to half-dozen months can be between the initiation of acarbose and liver toxicity events among affected diabetic patients (Hsiao et al., 2006). Taken this proposed latent flow into account, our findings lend support to this hypothesis by showing that the cumulative incidence curves of users and non-users nearly converged during the initial half dozen months of follow-upward (Figure 2). Furthermore, we extended the spectrum of their case summary past demonstrating the absence of differences in liver toxicity incidence up to 4–5 years subsequently the latent period. In addition, indigenous deviation has been suggested to play a role in affecting the gamble of metabolic idiosyncrasy related acarbose hepatotoxicity (Andrade et al., 1996), and CYP2E1 could be a plausible candidate (Wang et al., 1999). Multiple studies discovered that in patients of Chinese origin, single nucleotide polymorphism of CYP2E1 influenced the run a risk of lupus erythematosus and different cancers (Liao et al., 2011; Wang et al., 2014). Others also revealed that distinct polymorphisms of CYP2E1 might increment the risk of anti-tuberculosis drug-induced hepatotoxicity (Wang et al., 2016). Consequently, We believe that ethnic differences in CYP2E1 activity tin can influence the chance of developing acarbose-related hepatotoxicity in our studies, and farther pharmacogenetic assay are needed to elucidate the exact mechanisms of our findings.

In our report, we used Taiwan National Health Insurance Research database, an administrative healthcare database covering near all citizens in Taiwan, suitable for identifying rare cases or events such as DILI. Prior attempts to test for the utility of ICD-9-CM codes in identifying DILI cases based upon ICD-9-CM codes combination suggesting acute liver injury with or without codes suggesting drug poisoning/overdose (Jinjuvadia et al., 2007); although liver injury code-based approach exhibited an increased sensitivity, the results were still of business organization with regard to their validity. Nonetheless, due to the limitation of the merits database we used, including the lack of biochemical data to satisfy the diagnostic criteria for DILI, and the disability to ascertain temporal causal relationship betwixt offending drug and DILI, we could only identify DILI events based on diagnostic codes combinations which have been utilized before. Kao et al., in a study addressing anti-fungal agent hepatotoxicity, defined DILI events based on ICD-ix-CM codes only (Kao et al., 2014). Others also used a very like approach to uncover statin-related liver injury from the same database we used (Chang et al., 2015). Although our approach may suffer from low specificity, the large example number of participants is expected to partially offset this methodological insufficiency. We further increased the accuracy of our approach past excluding those with autoimmune hepatitis or daze, as shown in Table 7, with the aforementioned results. Consequently, our findings might still be useful, although prospective studies are still needed for confirmation.

In this study, we found that acarbose use in diabetic individuals with advanced CKD/ESRD and liver cirrhosis was not associated with an increased incidence of liver injury (Tabular array ii). This is particularly important, since the options of OADs for patients with DM and hepatic impairment are limited, and the option of handling choice is more challenging due to their tendency of developing hypoglycemia during treatment (Scheen, 2014). However, this upshot is rarely addressed in the literature (Tolman et al., 2007). An effective while safe OAD in this increasing population is urgently needed. Metformin, nigh sulfonylureas, meglitinide, acarbose, thiazolidinedione, and DPP4i are mostly metabolized past the liver, and pharmacokinetic changes tin can be observed for meglitinide and DPP4i amongst those with different severities of hepatic damage (Scheen, 2014). Because the metabolites of acarbose are excreted in the urine, the combination of renal and liver dysfunction casts dubiousness on the appropriateness of acarbose usage amongst these patients. Our findings may provide a temporary relief for physicians caring for diabetic patients with liver dysfunction, among whom acarbose can be a useful offshoot for glycemic control even in those with CKD and cirrhosis.

The incidence of liver injury in this study was comparable to those in the literature (Chen et al., 2014). A prior written report tested the relationship between acarbose utilise and the incidence of liver injury among diabetic patients with severe CKD using a more than simplistic approach (Kao et al., 2016). They compared diabetic acarbose users vs. not-users with astringent CKD only, without those carrying ESRD, enrolled fewer cases in their written report, and did not adjust for other hepatotoxic medications or account for background hepatic diseases explicitly (Kao et al., 2016). Our findings further strengthen their conclusions past showing that acarbose tin can be used safely in diabetic CKD/ESRD patients regardless of the presence or absence of CLD. We believe that the findings of this study serves to support the renaissance of acarbose as a useful adjunct in diabetic patients with phase 5 and 5D CKD. Although a randomized controlled trial is definitely needed to validate our finding, results gained from clinical practice in real-world settings are still noteworthy and can aid us in making a reasonable selection when selecting OADs for diabetic patients with multiple morbidities.

Our study has several of import limitations. First, laboratory data were unavailable in the electric current inquiry database, which constitutes significant barrier for fully characterizing the nature of DILI events identified. DILI is a diagnosis of exclusion, and other alternative diagnoses should be ruled out, which can be difficult to exist accomplished using this database due to its retrospective nature and the possibility of missing information. Using ICD-9-CM codes to identify DILI cases has been accounted problematic by some researchers (Jinjuvadia et al., 2007). However, the main criteria we used for recognizing avant-garde CKD, ESRD, CLD, and cirrhosis accept been cross-validated with proofs of catastrophic illness certificates, prominently lowering the uncertainty in diagnosis coding. Although the incidence of liver injury in this study was comparable to others, it is still likely that delayed hepatotoxic events occur subsequently prolonged acarbose use. Finally, our definition of acarbose exposure may non be strict enough (more than 30 days inside ane year), but our sensitivity analyses confirmed that changing this definition to more strict ones did non change our findings. Nonetheless, the medication adherence of the participants could not exist determined accurately in this claim database study, and the cumulative acarbose dose exposed for each participant was non direct assessed. Studies incorporating longer follow-up menstruum in these patients are needed to confirm our findings.

Author Contributions

C-TC, G-LC: Report design; C-TC, JW, J-WH, and K-LC: Data assay; C-TC, JW, J-WH, and M-LC: Article drafting. All authors approved the last version of the manuscript.

Funding

The study is financially sponsored by National Taiwan University Hospital BeiHu branch and the Ministry of Scientific discipline and Technology, Taiwan (Virtually 106-2314-B-002-132-MY2).

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of whatsoever commercial or financial relationships that could be construed equally a potential conflict of interest.

Abbreviations

aDCSI, adapted diabetes complications severity index; CI, confidence interval; CKD, chronic kidney disease; DILI, drug-induced liver injury; DKD, diabetic kidney affliction; DM, diabetes mellitus; DPP4i, dipeptidyl peptidase-4 inhibitors; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; EV, esophageal varices; HR, run a risk ratio; ICD-nine-CM, international classification of affliction ninth revision—clinical modification; OAD, oral anti-diabetic amanuensis.

References

Adams, L. A., Anstee, Q. K., and Tilg, H. (2017). Non-alcoholic fatty liver disease and its human relationship with cardiovascular disease and other extrahepatic diseases. Gut 66:1138. doi: 10.1136/gutjnl-2017-313884

PubMed Abstract | CrossRef Full Text | Google Scholar

Afkarian, One thousand., Zelnick, Fifty. R., Hall, Y. N., Heagerty, P. J., Tuttle, Chiliad., Weiss, N. S., et al. (2016). Clinical manifestations of kidney disease among us adults with diabetes, 1988–2014. JAMA 316, 602–610. doi: 10.1001/jama.2016.10924

PubMed Abstract | CrossRef Full Text | Google Scholar

Aithal, G. P., Watkins, P. B., Andrade, R. J., Larrey, D., Molokhia, M., Takikawa, H., et al. (2011). Example definition and phenotype standardization in drug-induced liver injury. Clin. Pharmacol. Ther. 89, 806–815. doi: ten.1038/clpt.2011.58

PubMed Abstract | CrossRef Total Text | Google Scholar

American Diabetes Association (2018). eight. Pharmacologic approaches to glycemic treatment: standards of medical intendance in diabetes-−2018. Diabetes Care 41, S73–S85. doi: x.2337/dc18-S008

PubMed Abstruse | CrossRef Full Text

Balfour, J. A., and McTavish, D. (1993). Acarbose: an update of its pharmacology and therapeutic use in diabetes mellitus. Drugs 46, 1025–1054.

PubMed Abstruse | Google Scholar

Bell, B. P., Manos, M. M., Zaman, A., Terrault, Due north., Thomas, A., Navarro, V. J., et al. (2008). Epidemiology of newly diagnosed chronic liver disease in gastroenterology practices in the United states: results from population-based surveillance. Am. J. Gastroenterol. 103, 2727–2736. doi: 10.1111/j.1572-0241.2008.02071.x

PubMed Abstract | CrossRef Full Text | Google Scholar

Byrd, K. K., Mehal, J. M., Schillie, S. F., Holman, R. C., Haberling, D., and Murphy, T. (2015). Chronic liver disease-associated hospitalizations among adults with diabetes, National inpatient sample, 2001–2012. Public Health Rep. 130, 693–703. doi: 10.1177/003335491513000619

PubMed Abstruse | CrossRef Full Text | Google Scholar

Chang, C.-H., Chang, Y.-C., Lee, Y.-C., Liu, Y.-C., Chuang, L.-M., and Lin, J.-W. (2015). Severe hepatic injury associated with different statins in patients with chronic liver affliction: a nationwide population-based accomplice study. J. Gastroenterol. Hepatol. 30, 155–162. doi: 10.1111/jgh.12657

PubMed Abstract | CrossRef Total Text | Google Scholar

Chang, H., Weiner, J., Richards, T., Bleich, Due south. N., and Segal, J. B. (2012). Validating the adapted diabetes complications severity alphabetize in claims data. Am. J. Manag. Care 18, 721–726.

PubMed Abstract | Google Scholar

Chao, C.-T., Hou, C.-C., Wu, V.-C., Lu, H.-K., Wang, C.-Y., Chen, L., et al. (2012). The impact of dialysis-requiring acute kidney injury on long-term prognosis of patients requiring prolonged mechanical ventilation: nationwide population-based study. PLoS Ane vii:e50675. doi: 10.1371/journal.pone.0050675

PubMed Abstruse | CrossRef Full Text | Google Scholar

Chao, C.-T., Tang, C.-H., Cheng, R. West.-Y., Wang, Thou.-West., and Hung, K.-Y. (2017). Poly peptide-energy wasting significantly increases healthcare utilization and costs amid patients with chronic kidney affliction: a propensity-score matched cohort written report. Curr. Med. Res. Opin. 33, 1705–1713. doi: ten.1080/03007995.2017.1354823

PubMed Abstract | CrossRef Full Text | Google Scholar

Charpentier, G., Riveline, J., and Varroud-Vial, One thousand. (2000). Direction of drugs affecting blood glucose in diabetic patients with renal failure. Diabetes Metab. 26, 73–85.

PubMed Abstract | Google Scholar

Chen, G.-L., Hsiao, F.-Y., Dong, Y.-H., Shen, L.-J., and Wu, F.-Fifty. (2014). Statins and the risk of liver injury: a population-based instance-control study. Pharmacoepidemiol. Drug Saf. 23, 719–725. doi: 10.1002/pds.3646

PubMed Abstract | CrossRef Total Text | Google Scholar

Forouzanfar, M. H., Afshin, A., and Alexander, 50. T. (2016). Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet 388, 1659–1724. doi: 10.1016/S0140-6736(16)31679-8

PubMed Abstract | CrossRef Total Text | Google Scholar

Fukui, North., Golabi, P., Otgonsuren, 1000., Mishra, A., Venkatesan, C., and Younossi, Z. M. (2017). Demographics, resource utilization, and outcomes of elderly patients with chronic liver illness receiving hospice care in the Us. Am. J. Gastroenterol. 112, 1700–1708. doi: ten.1038/ajg.2017.290

PubMed Abstract | CrossRef Full Text | Google Scholar

Fung, M. T., Fung, J., Lai, C. L., and Yuen, M. F. (2007). Etiologies of chronic liver diseases in Hong Kong. Eur. J. Gastroenterol. Hepatol. xix, 659–664. doi: 10.1097/MEG.0b013e3281ace0b7

PubMed Abstract | CrossRef Full Text | Google Scholar

Hsau, T.-W., Liu, J.-South., Hung, S.-C., Kuo, 1000.-L., Chang, Y.-G., Chen, Y.-C., et al. (2014). Renoprotective upshot of renin-angiotensin-aldosterone organisation blockade in patients with predialysis advanced chronic kidney disease, hypertension, and anemia. JAMA Intern. Med. 174, 347–354. doi: x.1001/jamainternmed.2013.12700

PubMed Abstract | CrossRef Full Text | Google Scholar

Hsiao, P.-J., Wu, Thousand.-L., Chiu, S.-H., Chan, J.-S., Lin, Y.-F., Wu, C.-Z., et al. (2017). Impact of the use of anti-diabetic drugs on survival of diabetic dialysis patients: a 5-yr retrospective cohort study in Taiwan. Clin. Exp. Nephrol. 21, 694–704. doi: 10.1007/s10157-016-1330-4

PubMed Abstruse | CrossRef Full Text | Google Scholar

Hung, S.-C., Chang, Y.-1000., Liu, J.-S., Kuo, K.-L., Chen, Y.-H., Hsu, C.-C., et al. (2015). Metformin use and bloodshed in patients with avant-garde chronic kidney disease: national retrospective, observational, cohort study. Lancet Diabetes Endocrinol. 3, 605–614. doi: 10.1016/S2213-8587(15)00123-0

PubMed Abstract | CrossRef Full Text | Google Scholar

International Diabetes Federation Guideline Evolution Group (2014). Global guideline for type 2 diabetes. Diabetes Res. Clin. Pract. 104, one–52. doi: 10.1016/j.diabres.2012.10.001

PubMed Abstract | CrossRef Total Text

Jinjuvadia, One thousand., Kwan, W., and Fontana, R. J. (2007). Searching for a needle in a haystack: use of ICD-9-CM codes in drug-induced liver injury. Am. J. Gastroenterol. 102, 2437–2443. doi: 10.1111/j.1572-0241.2007.01456.x

PubMed Abstruse | CrossRef Full Text | Google Scholar

Kao, C.-C., Wu, P.-C., Wu, C.-H., Chen, Fifty.-Thousand., Chen, H.-H., Wu, M.-Southward., et al. (2016). Hazard of liver injury after α-glucosidase inhibitor therapy in advanced chronic kidney affliction patients. Sci. Rep. 6:18996. doi: 10.1038/srep18996

PubMed Abstruse | CrossRef Total Text | Google Scholar

Kao, West.-Y., Su, C.-Due west., Huang, Y.-S., Chou, Y.-C., Chen, Y.-C., Chung, Westward.-H., et al. (2014). Risk of oral antifungal agent-induced liver injury in Taiwanese. Br. J. Clin. Pharmacol. 77, 180–189. doi: x.1111/bcp.12178

PubMed Abstract | CrossRef Full Text | Google Scholar

Liao, L.-H., Zhang, H., Lai, M.-P., Chen, S.-50., Wu, M., and Shen, N. (2011). Single-nucleotide polymorphisms and haplotype of CYP2E1 factor associated with systemic lupus erythematosus in Chinese population. Arthritis Res. Ther. 13:R11. doi: x.1186/ar3232

PubMed Abstract | CrossRef Full Text | Google Scholar

Lin, C.-H., Hung, P.-H., Hu, H.-Y., Chen, Y.-J., Guo, H.-R., and Hung, K.-Y. (2017). Infection-related hospitalization and hazard of end-stage renal illness in patients with systemic lupus erythematosus: a nationwide population-based report. Nephrol. Dial. Transplant. 32, 1683–1690. doi: ten.1093/ndt/gfw407

PubMed Abstract | CrossRef Full Text | Google Scholar

Lin, T.-T., Wu, C.-C., Yang, Y.-H., Lin, L.-Y., Lin, J.-L., Chen, P.-C., et al. (2016). Anti-hyperglycemic agents and new-onset acute myocardial infarction in diabetic patients with stop-stage renal illness undergoing dialysis. PLoS I xi:e0160436. doi: 10.1371/periodical.pone.0160436

PubMed Abstract | CrossRef Full Text | Google Scholar

Lin, Y.-T., Wu, P.-H., Lin, C.-Y., Lin, M.-Y., Chuang, H.-Y., Huang, J.-F., et al. (2014). Cirrhosis as a risk factor for tuberculosis infection—a nationwide longitudinal study in Taiwan. Am. J. Epidemiol. 180, 103–110. doi: 10.1093/aje/kwu095

PubMed Abstract | CrossRef Full Text | Google Scholar

Montomoli, J., Erichsen, R., Christiansen, C. F., Ulrichsen, S. P., Pedersen, 50., Nilsson, T., et al. (2013). Liver illness and 30-day bloodshed later on colorectal cancer surgery: a Danish population-based accomplice report. BMC Gastroenterol. 13:66. doi: 10.1186/1471-230X-13-66

PubMed Abstract | CrossRef Full Text | Google Scholar

Rassen, J. A., Shelat, A. A., Myers, J., Glynn, R. J., Rothman, 1000. J., and Schneeweiss, S. (2012). One-to-many propensity score matching in cohort studies. Pharmacoepidemiol. Drug Saf. 21, 69–eighty. doi: 10.1002/pds.3263

PubMed Abstract | CrossRef Total Text | Google Scholar

Ratib, S., Fleming, K. M., Crooks, C. J., Walker, A. J., and West, J. (2015). Causes of decease in people with liver cirrhosis in England cmpared with the full general population: a population-based cohort report. Am. J. Gastroenterol. 110, 1149–1158. doi: 10.1038/ajg.2015.191

PubMed Abstract | CrossRef Full Text | Google Scholar

Roussel, R., Lorraine, J., Rodriguez, A., and Salaun-Martin, C. (2015). Overview of data concerning the safe use of antihyperglycemic medications in type two diabetes mellitus and chronic kidney disease. Adv. Ther. 32, 1029–1064. doi: ten.1007/s12325-015-0261-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Saray, A., Mesihovic, R., Vanis, Due north., Gornjakovic, S., and Prohic, D. (2012). Clinical significance of haemostatic tests in chronic liver disease. Med. Arch. 66, 231–235. doi: 10.5455/medarh.2012.66.231-235

PubMed Abstract | CrossRef Full Text | Google Scholar

Scheen, A. J. (2014). Pharmacokinetic and toxicological considerations for the handling of diabetes in patients with liver affliction. Proficient Opin. Drug Metab. Toxicol. x, 839–857. doi: 10.1517/17425255.2014.902444

PubMed Abstract | CrossRef Full Text | Google Scholar

Setiawan, Five. W., Stram, D. O., Porcel, J., Lu, South. C., Le Marchand, L., and Noureddin, M. (2016). Prevalence of chronic liver disease and cirrhosis by underlying cause in understudied ethnic groups: the multiethnic cohort. Hepatology 64, 1969–1977. doi: 10.1002/hep.28677

PubMed Abstract | CrossRef Full Text | Google Scholar

Shih, C.-J., Chen, Y.-T., Ou, S.-G., Yang, W.-C., Kuo, Due south.-C., and Tarng, D.-C. (2014). The touch on of dialysis therapy on older patients with advanced chronic kidney disease: a nationwide population-based study. BMC Med. 12:169. doi: 10.1186/s12916-014-0169-3

PubMed Abstract | CrossRef Full Text | Google Scholar

Simón-Talero, M., Roccarina, D., Martinez, J., Lampichler, K., Baiges, A., Low, G., et al. (2018). Association between portosystemic shunts and increased complications and mortality in patients with cirrhosis. Gastroenterology 154, 1694–1705. doi: 10.1053/j.gastro.2018.01.028

PubMed Abstruse | CrossRef Full Text | Google Scholar

Tolman, K. G., Fonseca, 5., Dalpiaz, A., and Tan, M. H. (2007). Spectrum of liver disease in blazon 2 diabetes and management of patients with diabetes and liver disease. Diabetes Intendance 30:734. doi: 10.2337/dc06-1539

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, F.-J., Wang, Y., Niu, T., Lu, W.-10., Sandford, A. J., and He, J.-Q. (2016). Update meta-analysis of the CYP2E1 Rsal/Pstl and Dral polymorphisms and risk of antituberculosis drug-induced hepatotoxicity: testify from 26 studies. J. Clin. Pharm. Ther. 41, 334–340. doi: ten.1111/jcpt.12388

PubMed Abstruse | CrossRef Full Text | Google Scholar

Wang, P.-Y., Kaneko, T., Wang, Y., and Sato, A. (1999). Acarbose lone or in combination with ethanol potentiates the hepatotoxicity of carbon tetrachloride and acetaminophen in rats. Hepatology 29, 161–165. doi: ten.1002/hep.510290109

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, Y.-D., Yang, H.-Y., Liu, J., and Wang, H.-Y. (2014). Updated meta-analysis of the clan between CYP2E1 Rsal/Pstl polymorphisms and lung cancer risk in Chinese population. Asian. Pac. J. Cancer Prev. fifteen, 5411–5416. doi: 10.7314/APJCP.2014.15.13.5411

PubMed Abstract | CrossRef Full Text | Google Scholar

Wu, H.-H., Kuo, C.-F., Li, I.-J., Weng, C.-H., Lee, C.-C., Tu, Yard.-H., et al. (2017a). Family aggregation and heritability of ESRD in Taiwan: a population-based written report. Am. J. Kidney Dis. lxx, 619–626. doi: 10.1053/j.ajkd.2017.05.007

PubMed Abstract | CrossRef Full Text | Google Scholar

Wu, P.-C., Wu, Five.-C., Lin, C.-J., Pan, C.-F., Chen, C.-Y., Huang, T.-M., et al. (2017b). Meglitinides increment the risk of hypoglycemia in diabetic patients with advanced chronic kidney disease: a nationwide, population-based study. Oncotarget eight, 78086–78095. doi: x.18632/oncotarget.17475

PubMed Abstract | CrossRef Full Text | Google Scholar

whitebrour1997.blogspot.com

Source: https://www.frontiersin.org/articles/10.3389/fphar.2018.00860/full

Share this post