Buscar en
Annals of Hepatology
Toda la web
Inicio Annals of Hepatology Safety and efficacy of dyslipidemia treatment in NAFLD Patients: a Meta-analysis...
Journal Information
Share
Share
Download PDF
More article options
Visits
50
Original article
Open Access
Pre-proof, online 2 July 2022
Safety and efficacy of dyslipidemia treatment in NAFLD Patients: a Meta-analysis of randomized controlled trials
Visits
...
Mohamed Abdallaha, Landon Brownb, John Provenzab, Raseen Tariqc, Smitha Gowdaa, Ashwani K. Singala,d,
Corresponding author
ashwanisingal.com@gmail.com

Corresponding author.
a Department of Medicine, University of South Dakota Sanford School of Medicine, Sioux Falls, SD, USA
b Department of Medicine, University of Alabama at Birmingham, Birmingham, AL, USA
c Department of Medicine, University of Rochester, NY, USA
d Division of Transplant Hepatology, Avera Transplant Institute, Sioux Falls, SD, USA
Received 28 April 2022. Accepted 24 June 2022
Article information
Abstract
Full Text
Bibliography
Download PDF
Statistics
Figures (4)
Show moreShow less
Abstract
Introduction

: Practicing physicians often hesitate to use statins and/or other lipid lowering therapies in NAFLD due to concern for hepatotoxicity.

Aim

: To examine safety of lipid lowering therapies in NAFLD patients.

Methods

: Data from randomized control trials (RCT) among NAFLD patients were pooled to examine effect of lipid lowering therapies on liver chemistry, lipid profile, and liver histology. Results are reported as mean difference of the change (pretreatment-posttreatment) between treatment and control group.

Results

: A total of 21 placebo controlled RCT on 1900 patients (304 receiving statins, 520 other lipid lowering therapies, and 61 combination) were treated for 26 weeks [Interquartile range (IQR): 17.5-52 weeks]. Pooled data showed improved lipid profile without any worsening of ALT, AST, total bilirubin, or alkaline phosphatase at the end of treatment period. NAFLD activity score improved with other lipid lowering agents but not with statins. There was no change in individual components of NAFLD activity score or fibrosis stage.

Conclusion

: This meta-analysis of randomized controlled trials examining statins and/or other lipid lowering therapies in NAFLD patients showed no evidence for worsening of liver chemistry. Studies with longer use of lipid lowering therapies are suggested to examine benefit on liver histology among patients with NAFLD.

Full Text
1Introduction

Non-alcoholic fatty liver disease (NAFLD) is the most common cause of liver disease worldwide [1]. Its spectrum extends from steatosis to non-alcoholic steatohepatitis (NASH), cirrhosis, and hepatocellular carcinoma [1]. NASH cirrhosis is currently the second leading etiology for liver transplantation and the fastest growing indication for simultaneous liver kidney transplantation, hepatocellular carcinoma, and admissions with liver failure [2–5].

NAFLD is currently recognized as the most common cause of elevated liver enzymes [1]. In addition, NAFLD represents the hepatic manifestation of metabolic syndrome and insulin resistance, and includes central obesity, type 2 diabetes mellitus, dyslipidemia and hypertension. Dyslipidemia as a risk factor for the development of NAFLD is frequently observed in these patients, and is characterized by decreased serum high density lipoprotein (HDL) and increased serum triglycerides (TG) levels [1]. Hence, appropriate treatment of dyslipidemia is important for the management of patients with NAFLD. The mainstay of treatment is HMG-CoA reductase inhibitors (i.e., statins) along with lipid lowering therapies. Improved lipid profile with statins reduces cardiovascular mortality, and statins are recommended to be used for at risk population [6]. Statins also provide anti-inflammatory, anti-oxidant, and anti-thrombotic benefits by targeting pathways that are activated in the pathophysiology of NAFLD [7]. Further, the data is accumulating on the benefits of statins on liver fibrosis, HCC risk, liver disease complications, and overall survival [8].

However, the fear of hepatotoxicity with the use of statins and/or triglyceride lowering therapy for dyslipidemia in patients with NAFLD often deters the physicians from their use in clinical practice [1, 9]. The objective of this meta-analysis is to examine the effect of using statins or lipid lowering therapies among NAFLD patients on the risk of hepatotoxicity. We also examined their effects on lipid profile and liver histology.

2Methods2.1Search Strategy and Study Selection for Analysis

A comprehensive literature search was conducted in PubMed, EMBASE, PsycINFO, and Cochrane databases through May 2020 for the safety and efficacy of lipid lowering agents and omega-3 polyunsaturated fatty acid (PUFA) supplementation in patients with NAFLD and is described in the supplementary document. As per the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines, full-length articles in English reporting on lean NAFLD were identified. Data from selected studies on patients with NAFLD comparing statins and/or lipid lowering therapy against placebo or other active agents were extracted and pooled in this meta-analysis. Identification of studies for analysis was performed independently by three investigators (M.A., J.P., A.K.S) using inclusion criteria: a) randomized-controlled trials; (b) included adult subjects; c) published in English language; and (d) reported as full papers. The studies were excluded if: (1) without a control group; (2) without data on pre- and post-treatment study outcomes; (3) studies not including lipid lowering agents; (4) non-randomized studies; or 5) not meeting ≥1 inclusion criteria. We also excluded certain studies despite meeting inclusion and exclusion criteria because they reported results in a format that was not extractable for meta-analysis . Retrieved articles were independently reviewed for their eligibility by the three investigators. Any discrepancy among investigators was resolved by consensus among all the investigators. Given that one of the aims of the study is to evaluate hepatotoxicity of lipid lowering agents, we included studies that compared the combination of two lipid lowering agents vs. another agent or a placebo.

2.2Study Quality Assessment

Two reviewers (M.A and S.G) independently assessed the methodological quality of studies using the risk of bias by Cochrane for RCTs (Supplementary Fig. 1, 2) [10]. Each study was graded on the study quality as per the following criteria: sequence generation for the randomization of subjects, allocation concealment, participant blinding, outcome assessment blinding, attrition bias, and any other sources of bias. Risk of bias for one of the first three components defined a high risk of bias.

2.3Study outcomes

The primary outcome of this study is the effect of lipid lowering drugs on liver toxicity evaluated by the change in liver biochemical profile (pre-treatment value – post treatment value) in serum levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), and total bilirubin (TB). Secondary outcomes are the effect of lipid lowering drugs on a) lipid profile (pre-treatment – post treatment) in serum levels of total cholesterol (TC), low density lipoprotein (LDL), high density lipoprotein (HDL), triglycerides (TG) and b) liver histology (pre-treatment – post treatment) in steatosis grade, lobular inflammation score (LI), ballooning severity, NAFLD activity score (NAS), and fibrosis.

2.4Data Extraction

Data from studies selected for analysis was extracted independently by three authors (J.P., M.A., and S.G.) on: a) characteristics (sample size, study population); b) patient demographics (age, gender, and BMI); c) treatment details (type of intervention, comparator group, dose in each group, duration of treatment); c) diabetes mellitus; d) serum liver chemistry before and after treatment (ALT, AST, ALP in IU/l, and total bilirubin in mg/dL); e) serum lipid profile values before and after treatment (TC, LDL, HDL and TG in mg/dL); and f) liver histology before and after treatment (steatosis, lobular inflammation, ballooning, NAS, and fibrosis stage). If studies reported continuous variables with median values, this was converted into mean using standard statistical methods [11]. Similarly, if the unit of variation was reported as standard error of mean, this was converted to standard deviation (SD). For studies reporting laboratory values in molar concentration, these were converted to mass concentration.

2.5Definitions

NAFLD: Hepatic steatosis in absence of other causes of liver disease and alcohol use <30 g/d in males and <20 g/d in females [1].

Steatosis grade: on liver histology with % of hepatocytes showing fat as grade 1 (5-33%), grade 2 (34-66%), and grade 3 (≥66%) [12].

Lobular inflammation: graded by number of inflammatory foci per 200X field as grade 1 (<2 foci) grade 2 (2-4 foci), and grade 3 (>4 foci) [12].

Hepatocyte ballooning: graded by number of ballooned hepatocytes as grade 1 (few balloon cells), and grade 2 (many ballooned cells) [12].

NAFLD activity score (NAS): sum of the scores for steatosis, lobular inflammation, and ballooning [12].

NASH: defined on liver histology with NAS ≥5 [12].

Fibrosis stage: as stage 1 with peri-sinusoidal or portal fibrosis, stage 2 with peri-portal fibrosis, stage 3 as bridging fibrosis, and stage 4 with cirrhosis [12].

2.6Data Analyses

Comprehensive statistical analysis software was used to pool the data, using random effects model. Pooled effect size represented as Forrest plots is reported as standardized mean difference (SMD) with 95% confidence interval (CI) and considered significant if the 95% CI does not cross ‘0’For each study the SMD is the difference in means between groups divided by standard deviation among the participants in that study.

Heterogeneity between the studies was assessed using the I2 statistics (I2 >50% or P <0.05). If the data were heterogeneous, reason/s of heterogeneity were examined using the subgroup and/or sensitivity analyses. Visual inspection of funnel plots Egger's regression test were used to assess the publication bias [13]. P<0.05 was considered significant for all the analyses.

3Results3.1Baseline Characteristics

The PRISMA flow diagram and results based on the search strategies and selection criteria described above are outlined in (Supplementary Figure 1)[14]. Of 4818 studies identified on initial literature search, 35 studies met the inclusion criteria. Fourteen studies, however, were excluded (four for the lack of randomization, four other studies for not having data presented in an extractable form for meta-analysis, and six for insufficient data for treatment and/or control). Therefore, twenty-one studies were included in the final meta-analysis. There were 1900 patients (304 received statins, 520 other lipid lowering agents and 61 with combination therapy). The mean age of patients treated with single agent or a combination of two lipid lowering agents was 49.3 (95% CI: 46.4-52.2) years, with 61.6 % males (range: 35.7-87.2%). The mean body mass index was 30 (95% CI: 29-31) Kg/m2. The median duration of treatment was 26 weeks [25-75% Interquartile range (IQR): 17.5-52 weeks] (Table 1). Among eleven studies with available information, 28.4% (range: 0-100) patients were included with diabetes mellitus as comorbidity. Most importantly, 18 out of the included 21 studies (85.7%) had an elevated ALT at baseline.

Table 1.

Baseline characteristics of studies included in the analysis

Author, Year  Total N (Treated)  Mean Age (SD)  % Males  % Diabetes  Mean BMI (SD)  Intervention  Dose  Duration 
Argo, 2015 [38]  34 (17)  46.4 (12.1)  35.7  29.4  33.3 (8.0)  N-3 PUFA  3000 mg/day  1 year 
Athyros, 2006 [39]  186 (61)¿  59 (13)  66  33 (9)  Atorvastatin fen fibrates  20 mg/day 200 mg/day  54 weeks 
Athyros, 2010 [32]  437 (227)  60 (14) *  79  50.8  29 (7) *  Various statins ^  Various doses ^  3 years 
Basaranoglu, 1999 [40]  46 (23)  41.6 (6.6)  82.6  N/A  28.2 (3.7)  Gemfibrozil  600 mg/day  4 weeks 
Chan, 2010 [41]  25 (15)  N/A  60  32 (1)  Ezetimibe  10 mg/day  22 weeks 
Dasarathy, 2015 [42]  37 (18)  51.5 (6.9)  50  100  34.8 (4.6)  PUFA  EPA 2160 mg+DHA 1440 mg/day  48 weeks 
Li, 2015 [43]  78 (39)  52.6 (6.6)  87.2  N/A  28.0 (1.4)  PUFA  1:1 of EHA and DHA  6 months 
Loomba, 2015 [44]  50 (25)  49.0 (14.9)  44  28  33.8 (5.2)  Ezetimibe  10mg/day  24weeks 
Malaguarnera, 2011 [45]  65 (33)  47.8 (5.9)  60.6  26.8 (3.0)  Rosuvastatin  5 mg daily  12 months 
Nelson, 2009 [46]  16 (10)  52.6 (8.6)  70  60  37.3  Simvastatin  40 mg/day  12 months 
Qin, 2015 [47]  70 (36)  46.0 (10.68)  72  NA  26.4 (3.9)  Omega 3 fatty acid  4 gm/day  3 months 
Rana, 2016 [48]  98 (34)  NA  70  NA  27.7 (2)  Rosuvastatin  NA  24 weeks 
Rezaei, 2020 [49]  68 (34)  45.5 (8.7)  47.1  30.1 (4.1)  Flaxseed oil  20 ml/day  12 weeks 
Scoreletti, 2014 [50]  103 (51)  48.6 (11.1)  49  34.3 (5.8)  DHA+EPA  4g/day  15-18 months 
Shojasaadat, 2019 [51]  114 (38)  41.77(8.92)  51.4  NA  31.5 (3.6)  DHA+EPA  1500mg/day  12 weeks 
Sofi, 2010 [52]  11 (6)  55 (30–71)  66.7  NA  29.3 (4.1)  Olive oil enriched with n-3 PUFA  0.83 g/day  12 months 
Spadaro, 2008 [53]  40 (18)  50.2 (12.9)  61  N/A  30.1 (4.7)  PUFA  1 g/daily  6 months 
Takeshita, 2014 [54]  31 (17)  50.4 (2.9)  64.7  N/A  30.5 (1.2)  Ezetimibe  10 mg/day  6 months 
Tobin, 2018 [55]  167 (81)  55.3 (13.3)  44.4  35  32.1 (4.8)  PUFA  3 g/day  24 weeks 
Yaghoubi, 2017 [56]  90 (30)  38 (8)  N/A  N/A  27 (3)  Fen fibrate  200 mg /day  12 weeks 
Zhu, 2008 [57]  134 (72)  45 (10.9)  71  N/A  26.4 (3.1)  N-3 PUFA  2 g/TID  24 weeks 
Summary  1900 (885)  49 (46-52)  62 (36-87)    30 (29-31)      26 (18-52) 

*Results for patients on statin therapy

^ Statin drug (median statin dose per day in mg [range]): Atorvastatin 25 (10–80), Simvastatin 22 (10–40), Pravastatin 29 (20–40), Fluvastatin 40 (40–40).

¿ [atorvastatin (n=63), fen fibrate (n=62), combination of atorvastatin and fen fibrate (n=61)]

3.2Risk of Bias within Studies

Of 21 RCTs, all studies reported on the random sequence generation (Supplementary Figures 2 and 3). Four RCTs did not report on concealment of allocation. Ten studies did not blind the study personnel or participants, and thirteen studies did not blind the outcome assessor/s. Four RCTs did not report on the attrition rate, and one study failed to provide information on selective reporting of the outcomes. Overall risk of bias was estimated to be low in twelve out of the twenty-one studies included in the meta-analysis.

3.3Primary Outcome: Change in Liver Chemistry

Data on ALT, AST, ALP, and serum total bilirubin values before and after treatment for intervention and control arms (Table 2) were pooled for SMD (95% CI) on change (pre-treatment – post treatment).

Table 2.

Liver chemistry profile [values given as mean (SD), IU/liter except for TB in mg/dL] before and after treatment

Author, Year  ALT pre-treatment  ALT post-treatment  AST pre-treatment  AST post-treatment  ALP pre-treatment  ALP post-treatment  TB pre-treatment  TB post-treatment 
Argo, 2015  72.8 (30.9)  56.7 (28.3)  N/A  N/A  N/A  N/A  N/A  N/A 
Athyros, 2006  51 (24)  30 (7)  38 (13)  24 (6)  N/A  N/A  N/A  N/A 
Athryos, 2010 ^  57 (8)  37 (6)  49 (7)  26 (4)  N/A  N/A  N/A  N/A 
Basaranoglu, 1999  72 (37.0)  49.7 (28.2)  49.1 (20.4)  38.9 (20.2)  N/A  N/A  N/A  N/A 
Chan, 2010  29 (2)  30 (3)  N/A  N/A  N/A  N/A  N/A  N/A 
Dasarathy, 2015  60.1 (27.6)  56.9 (30.9  47.7 (22.8)  41.7 (17.4)  N/A  N/A  1.1 (0.3)  1.1 (0.3) 
Li, 2015  89.2 (12.4)  67.8 (5.3)  82 (9.6)  60.3 (6.8)  N/A  N/A  N/A  N/A 
Loomba, 2015  47 (29)  48 (43)  33 (23)  33 (36)  74 (34)  89 (45)  0.4 (0.4)  0.4 (0.3) 
Malaguarnera, 2011  184 (49.2)  57.8 (21.4)  145 (46.8)  38.5 (30.2)  N/A  N/A  N/A  N/A 
Nelson, 2009  70.4 (29.6)  49.5 (15.6)  43.3 (14.8)  36.5 (11.5)  86.1 (30.9)  89.7 (23.2)  N/A  N/A 
Qin, 2015*  31.5 (25, 51.2)  25.5 (21.2, 30)  28 (23.2, 34.7)  27 (21, 30.7)  NA  NA  NA  NA 
Rana, 2016  51.9 (17.5)  59.9 (30.6)  44.6 (17.8)  52.9 (27.3)  NA  NA  NA  NA 
Rezaei, 2020  27.6 (20.8)  23 (15.8)  16.6 (9.2)  14.4 (7.9)  N/A  N/A  N/A  N/A 
Scoreletti, 2014  54.0 (43)  44.0 (34)  38.0 (24)  30.0 (27)  NA  NA  NA  NA 
Shojasaadat, 2019  34.8 (24.2)  35.3 (21.1)  28.2 (11.9)  27.1 (9.7)  NA  NA  NA  NA 
Sofi, 2010  87 (16.9)  56.3 (14.4)  48.8 (7.6)  29.2 (3.4)  NA  NA  NA  NA 
Spadaro, 2008  56.6 (24.1)  39.5 (14)  31.5 (13.2)  28 (8.8)  N/A  N/A  N/A  N/A 
Takeshita, 2014  53.2 (8.6)  49.3 (6.5)  41.8 (6.7)  33.7 (4.1)  N/A  N/A  N/A  N/A 
Tobin, 2018  37.5 (39)  38.1 (37.7)  27.1 (20.3)  28.7 (24.6)  85.5 (45.3)  83.2 (40.3)  0.5 (0.3)  0.5 (0.2) 
Yaghoubi, 2017  78 (11)  53 (26)  70 (30)  41 (17)  N/A  N/A  N/A  N/A 
Zhu, 2008  62.8 (35.9)  39.3 (18.9)  38.1 (21)  30.5 (12.7)  N/A  N/A  N/A  N/A 

^ Results for patients on statin therapy

*Median, interquartile range (IQR)

Change in serum ALT levels was similar after intervention versus control arm on pooled data from 21 studies, with mean difference (95% CI) of 0.35 (-0.19-0.88, P=0.21). The data were heterogeneous with I2=96, P<0.001. (Figure 1A). There was no publication bias, Egger's regression test P=0.21.

Figure 1.

Effect of treatment on A) alanine aminotransferase (ALT), B) aspartate aminotransferase (AST), C) alkaline phosphatase (ALP), and D) total bilirubin (TB).

(0.27MB).

Mean serum AST levels from baseline value were also similar between treatment and control arm on pooled data from 19 studies, with mean difference (95% CI) of 0.58 (-0.05-1.2, P<0.07). The data were heterogeneous with I2=96, P<0.001. (Figure 1B). There was no publication bias, P=0.66.

Change in serum ALP levels after intervention compared to control arm was similar on pooled data from three studies, with mean difference (95% CI) of -0.04 (-0.32 - 0.24, P=0.78). The data were homogeneous with I2=6, P=0.34 without publication bias, P=0.46. (Figure 1C).

Change in mean serum bilirubin levels after intervention compared to control arm was similar on pooled data from three studies, with mean difference (95% CI) of 0.09 (-0.16 - 0.33, P=0.48). The data were homogeneous with I2=0, P=0.89 without publication bias, P<0.08 (Figure 1D).

3.4Secondary Outcomes3.4.1Change in Lipid Profile

Data on TC, LDL, HDL, and TG values before and after treatment for the intervention as well for the control arm were extracted (Table 3). Data were pooled in the statistical software for MD (95% CI) on the change (pre-treatment – post treatment) in the two arms.

Table 3.

Lipid profile [values given as mean (SD), mg/dL] before and after treatment

Author, Year  TC pre-treatment  TC post-treatment  LDL pre-treatment  LDL post-treatment  TG pre-treatment  TG post-treatment  HDL pre-treatment  HDL post-treatment 
Argo, 2015  191.5 (41)  193.9 (45.8)  122.6 (32.6)  124.1 (36.3)  155 (77.1)  132.2 (45.4)  N/A  N/A 
Athyros, 2006  239.8 (38.7)  158.6 (27.1)  158.6 (42.5)  92.8 (15.5)  212.6 (115.2)  106.3 (35.4)  N/A  N/A 
Athyros, 2010^  245.9 (27)  162.4 (7.7)  170.2 (19.3)  96.7 (7.7)  194.9 (53.1)  132.9 (26.6)  37.1 (7.7)  39.8 (7.7) 
Basaranoglu, 1999  N/A  N/A  N/A  N/A  229 (141)  153 (90)  N/A  N/A 
Chan, 2010  224.3 (11.6)  189.5 (11.6)  147 (11.6)  119.9 (7.7)  177.2 (17.8)  132.9 (8.9)  42.5 (3.9)  42.5 (3.9) 
Dasarathy, 2015  177.9 (40.2)  182.5 (55.5)  N/A  N/A  175.2 (91.6)  211.5 (160.8)  40.6 (7.1)  43.2 (8.9) 
Li, 2015  216.6 (15.5)  181.8 (11.6)  150.8 (11.6)  119.9 (23.2)  230.3 (26.6)  159.4 (26.6)  N/A  N/A 
Loomba, 2015  182.0 (26.0)  152.0 (46.0)  99.0 (37.0)  76.0 (30)  152.0 (63)  125.0 (59)  NA  NA 
Malaguarnera, 2011  201.1 (27.1)  177.9 (30.9)  100.5 (30.9)  77.3 (30.9)  194.9 (70.9)  177.2 (35.4)  42.5 (7.7)  42.5 (30.9) 
Nelson, 2009  230.5 (72.5)  209.1 (114.7)  138.5  102.7  388.7 (507.9)  490 (890.5)  N/A  N/A 
Qin, 2015  206.1 (45.6)  187.2 (40.6)  126.5 (25.5)  114.8 (25.9)  187.8 (91.2)  135.5 (75.3)  49.1 (7.3)  51 (8.5) 
Rana, 2016  205.1 (29.3)  146.8 (19)  113.2 (38.3)  59.4 (12.5)  260.7 (21.4)  138.1 (14.2)  37.9 (5.1)  40.4 (4.6) 
Rezaei, 2020  185.6 (42.5)  179 (46.4)  111.4 (34.8)  104 (34.8)  168.3 (141.7)  150.6 (115.2)  39.8 (6.6)  40.6 (7.3) 
Scoreletti, 2014  189.5 (42.5)  181.7 (42.5)  116 (34.8)  108.3 (34.8)  159.4 (106.3)  132.9 (106.3)  38.7 (7.7)  42.5 (11.6) 
Shojasaadat, 2019  188.4 (22.7)  201.3 (23.6)  104.4 (18.6)  107.9 (22.8)  158.6 (62.8)  165.9 (80.8)  45. (6.4)  45.1 (5.8) 
Sofi, 2010  213.8 (27.9)  213 (23.1)  131.6 (24.6)  133.6 (32.3)  164.5 (85.5)  132.8 (63.7)  44.3(5.5)  60.3 (9.3) 
Spadaro, 2008  194.4 (26.8)  189.4 (15.9)  N/A  N/A  147.4 (41.1)  110 (39.1)  43.5 (14.7)  46.8 (12.9) 
Takeshita, 2014  198.8 (7.7)  179.8 (6.6)  N/A  N/A  126.7 (9.7)  129.3 (11.5)  52.6 (3.1)  52.6 (2.3) 
Tobin, 2018  N/A  N/A  N/A  N/A  192 (125.1)  157.8 (84.2)  N/A  N/A 
Yaghoubi, 2017  186 (29)  179 (17)  N/A  N/A  143 (37)  101 (20)  N/A  N/A 
Zhu, 2008  243.6 (32.1)  196.4 (29.4)  126.1 (38.7)  3.12 (32.5)  349 (238.3)  184.2 (91.2)  39.1 (9.3)  48.3 (9.3) 

^ Results for patients on statin therapy * Percentage of patient improvement or stability after 6 months (95% CI) ** Median and interquartile range (IQR).

Mean (95% CI) serum TC level was lower by 0.58 (0.03-1.31), P<0.04 after intervention compared to control arm on pooled data from 18 studies. The data were heterogeneous with I2=95, P<0.001. (Figure 2A). There was no publication bias, P<0.07.

Figure 2.

Effect of treatment on A) serum total cholesterol (TC), B) low density lipoprotein (LDL), C) high density lipoprotein (HDL), and D) triglycerides (TG).

(0.26MB).

Mean (95% CI) serum LDL level was lower by 0.83 (0.20-1.5, P=0.01), after intervention compared to control arm on pooled data from 16 studies. The data were heterogeneous with I2=95, P<0.001. (Figure 2B). There was no publication bias, P=0.90.

Mean (95% CI) serum HDL level changed by -0.04 (-0.39 - 0.32), P=0.84 (Figure 2C) after intervention compared to control arm on pooled data from eight studies. The data were homogeneous with I2=67, P=0.003, with a publication bias on Egger's test, P<0.05.

Mean (95% CI) serum TG level was lower by 0.46 (0.05-0.88), P<0.03 after intervention compared to control arm on pooled data from 21 studies. The data were heterogeneous with I2=93, P<0.001 (Figure 2D). There was no publication bias, P=0.67.

3.4.2Change in Liver Histology

Data on steatosis grade, lobular inflammation score, hepatocyte ballooning score, NAS and fibrosis before and after treatment for the intervention as well for the control arm were extracted (Supplementary Table 1). Data were pooled in the statistical software for MD (95% CI) on the change (pre-treatment – post treatment) in the two arms.

Change in steatosis grade was similar after intervention compared to control arm on pooled data from six studies, with mean difference (95% CI) of 0.63 (-0.23-1.49, P=0.15). The data were heterogeneous with I2=90, P<0.001 (Figure 3A). There was no publication bias, P=0.69.

Figure 3.

Effect of treatment on A) steatosis grade, B) lobular inflammation score (LI), C) hepatocyte ballooning score, D) Non-alcoholic fatty liver disease (NAFLD) activity score (NAS), and E) fibrosis stage.

(0.29MB).

Change in lobular inflammation score was similar after intervention compared to control arm on pooled data from six studies, with mean difference (95% CI) of 0.21 (-0.93-1.34, P=0.80). The data were heterogeneous with I2=94, P<0.001 (Figure 3B). There was no publication bias, P=0.79.

Change in hepatocyte ballooning severity was similar after intervention compared to control arm on pooled data from four studies, with mean difference (95% CI) of 0.02 (-0.27-0.31, P=0.89). The data were homogeneous with I2=0, P=0.53 (Figure 3C). There was no publication bias, P=0.87.

Mean (95% CI) NAS was lower by 0.49 (0.09-0.89), P<0.02 (Figure 3D) after intervention compared to control arm on pooled data from three studies. The data were homogeneous with I2=3, P=0.36. There was no publication bias, P=0.16.

Mean (95% CI) fibrosis stage was lower by 0.51 (-0.25 - 1.27), P=0.19 (Figure 3E) after intervention compared to control arm on pooled data from six studies. The data were heterogeneous with I2=88, P<0.001. There was no publication bias, P=0.36.

Figure 4.

Figure 4.

xxx

(0.36MB).
3.5Subgroup analyses

Results were similar when pooled data were analyzed only for 12 studies with low risk of study bias (Supplementary Table 2). Similarly, results did not change on separately pooling 10 studies using lipid lowering therapies and five using statins. The study using combination of statins and lipid lowering drugs was counted in the subgroup of statins. Both subgroups improved LDL levels by 1.8 on 8 non-statin studies and 1.64 in 5 studies using statins. Similarly, TG decreased by 1.1 in 10 non-statin studies and 2.6 in 5 studies using statins. However, the decrease TC levels decreased by 1.1 in 9 studies using non-statin based medications, but there was no change in 5 studies using statins (Supplementary Table 3). Although, no publication bias was detected, heterogeneity between studies was noted in both subgroups for most analyses.

4Discussion

The main finding of this meta-analysis is that drugs used for dyslipidemia among patients with NAFLD are not associated with worsening of liver enzymes, including patients with elevated liver chemistry at baseline. Further, lipid lowering therapies are associated with improvement in lipid profile, especially LDL levels and TG. Although there was improvement in NAFLD activity score, there was no benefit on its components or on fibrosis.

Asymptomatic ALT elevation with statins is dose‐related, and occurs in less than 3% within the first 12 weeks of therapy, which improves spontaneously in the majority of patients [15]. In another meta‐analysis (n=49,275), statin use versus placebo was not associated with liver enzymes elevation of > 3 times upper limit normal (1.1% vs. 1.1%, OR 1.3, 95% CI: 0.99–1.62) [16]. Further, the incidence of acute liver failure with statins has been shown to be similar to that of the general population (1:130,000 vs. 1:114,000) [17]. Our NAFLD focused analysis also showed no increased risk of elevation of liver enzymes with use of statins or lipid lowering drugs. Our study findings strengthen the recently published American Gastroenterology Association Clinical Practice Update based on expert review advocating that statins and other lipid lowering therapies are safe and effective in NAFLD patients and should not be avoided because of fears of hepatotoxicity [18]. Although drug induced hepatotoxicity are linked to fibrates, compelling evidence for fibrate-induced hepatotoxicity is not available [19]. In a meta-analysis of twenty-six studies on 9494 individuals, the use of fibrates combined with statins was associated with increased risk of hepatotoxicity [20]. However, in the current meta-analysis, use of fibrates and omega-3 fatty acids among NAFLD patients did not increase this risk.

The benefit of statins or lipid lowering therapies on histological improvement of NASH remains controversial. For example, use of eicosapentanoic acid among patients with NASH or NAFLD did not show any benefit on the liver histology [21, 22]. In another study on 100 subjects, use of statins protected from steatosis, NASH, and fibrosis stage F2-4. This effect was dose dependent and was stronger in subjects not carrying the PNPLA3 at risk variant [23]. Also, resolution of NASH was shown in 19 out of 20 patients after treatment with rosuvastatin in some reports [24]. Our subgroup analysis, however, did not show improvement of NAS or any of its components with statin or non-statin therapy. This finding is similar to recently published American Gastroenterology Association Clinical Practice Update based on expert review concluding that no firm recommendation could be derived on the benefit of lipid lowering therapies on liver histology [18]. The lack of benefit could be due to short period of treatment and observation of about 26 weeks in this pooled data. In a recent report, the observation period on the benefit of statins to reduce the risk of hepatocellular carcinoma was 4-19 years [25]. Studies with long-term treatment are needed to establish the role of these drugs among patients with NASH on improvement of liver disease.

Dyslipidemia among NAFLD patients confers increased risk of cardiovascular disease, the most common cause of death in these patients [26–28]. The association of effective treatment of dyslipidemia on the cardiovascular outcomes is variable with benefit in many studies [29–33], however no benefit was shown in a recently reported analysis of subjects enrolled in the National Health and Nutrition Examination Survey [34]. In addition, statins use has been shown to protect from NASH and fibrosis among at risk individuals with improvement in liver histology in NASH patients. Statins also reduce portal pressure, liver disease severity and decompensation, hepatocellular carcinoma risk, and patient mortality in patients with cirrhosis [23, 24, 35, 36].

Pooled data from 21 RCT with large sample size and satisfactory follow up duration is a strength of our study. However, our study is limited by heterogeneous data due to variations across studies on active ingredient, dose, treatment regimen, duration of treatment, and follow up period. Despite heterogeneous data, this pooled data confirms safety of treatment of dyslipidemia among NAFLD patients, with no increased risk for liver enzyme elevations and hepatotoxicity. Although, there are previous studies highlighting safety of drugs for dyslipidemia in NAFLD patients including a recently published practice guidance document by American Gastroenterological Association, our meta-analysis of all randomized controlled studies on different lipid lowering therapies has not been performed earlier and provides a high level of evidence to further reassure physicians and providers on the safety of these drugs for treatment of dyslipidemia in NAFLD patients. With effective improvement of lipid profile and reduced risk for cardiovascular event, and added potential benefit on liver disease, the findings of this meta-analysis on safety on treatment of dyslipidemia in NAFLD patients will be reassuring to the practicing physicians. However, it is recommended to obtain baseline liver chemistry profile before initiating the treatment, and practice repeated testing on follow up during treatment of dyslipidemia among patients with NAFLD [37].

Conclusion

This meta-analysis of randomized controlled trials examining statins and/or other lipid lowering therapies in NAFLD patients showed no evidence for worsening of liver chemistry. Studies with longer use of lipid lowering therapies are suggested to examine benefit on liver histology among patients with NAFLD.

Formatting of funding sources

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authorship and Contributions

MA drafted the article and interpreted the data; LB conceived the idea and extracted the study data; JP extracted the study data; RT contributed drafting the article; SG extracted the study data; AKS conceived the study design, performed analyses, provided study supervision, and drafted the article. All the authors approved the final version of the manuscript.

References
[1]
N Chalasani, Z Younossi, JE Lavine, M Charlton, K Cusi, M Rinella, et al.
The diagnosis and management of nonalcoholic fatty liver disease: Practice guidance from the American Association for the Study of Liver Diseases.
Hepatology, 67 (2018), pp. 328-357
[2]
RJ Wong, M Aguilar, R Cheung, RB Perumpail, SA Harrison, ZM Younossi, et al.
Nonalcoholic steatohepatitis is the second leading etiology of liver disease among adults awaiting liver transplantation in the United States.
Gastroenterology, 148 (2015), pp. 547-555
[3]
AK Singal, H Salameh, YF Kuo, RH. Wiesner.
Evolving Frequency and Outcomes of Simultaneous Liver Kidney Transplants Based on Liver Disease Etiology.
[4]
P Axley, Z Ahmed, S Arora, A Haas, YF Kuo, PS Kamath, et al.
NASH Is the Most Rapidly Growing Etiology for Acute-on-Chronic Liver Failure-Related Hospitalization and Disease Burden in the United States: A Population-Based Study.
Liver Transpl, 25 (2019), pp. 695-705
[5]
Z Younossi, M Stepanova, JP Ong, IM Jacobson, E Bugianesi, A Duseja, et al.
Nonalcoholic Steatohepatitis Is the Fastest Growing Cause of Hepatocellular Carcinoma in Liver Transplant Candidates.
Clin Gastroenterol Hepatol, 17 (2019), pp. 748-755
[6]
BB Adhyaru, TA. Jacobson.
New Cholesterol Guidelines for the Management of Atherosclerotic Cardiovascular Disease Risk: A Comparison of the 2013 American College of Cardiology/American Heart Association Cholesterol Guidelines with the 2014 National Lipid Association Recommendations for Patient-Centered Management of Dyslipidemia.
Endocrinol Metab Clin North Am, 45 (2016), pp. 17-37
[7]
AS Antonopoulos, M Margaritis, R Lee, K Channon, C. Antoniades.
Statins as Anti-Inflammatory Agents in Atherogenesis: Molecular Mechanisms and Lessons from the Recent Clinical Trials.
Current pharmaceutical design, 18 (2012), pp. 1519-1530
[8]
J Bosch, J Gracia-Sancho, JG. Abraldes.
Cirrhosis as new indication for statins.
[9]
N. Chalasani.
Statins and hepatotoxicity: focus on patients with fatty liver.
Hepatology, 41 (2005), pp. 690-695
[10]
JP Higgins, DG Altman, PC Gøtzsche, P Jüni, D Moher, AD Oxman, et al.
The Cochrane Collaboration's tool for assessing risk of bias in randomised trials.
Bmj, 343 (2011), pp. d5928
[11]
S McGrath, X Zhao, R Steele, BD Thombs, A. Benedetti.
Estimating the sample mean and standard deviation from commonly reported quantiles in meta-analysis.
Stat Methods Med Res, (2020),
[12]
DE Kleiner, EM Brunt, M Van Natta, C Behling, MJ Contos, OW Cummings, et al.
Design and validation of a histological scoring system for nonalcoholic fatty liver disease.
Hepatology (Baltimore, Md), 41 (2005), pp. 1313-1321
[13]
JJ AD Deeks, MJ Bradburn.
Statistical methods for examining heterogeneity and combining results from several studies in meta-analysis.
2nd ed., BMJ Books, (2005),
[14]
D Moher, A Liberati, J Tetzlaff, DG. Altman.
Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.
[15]
CY Chang, TD. Schiano.
Review article: drug hepatotoxicity.
Alimentary pharmacology & therapeutics, 25 (2007), pp. 1135-1151
[16]
S de Denus, SA Spinler, K Miller, AM. Peterson.
Statins and liver toxicity: a meta-analysis.
Pharmacotherapy, 24 (2004), pp. 584-591
[17]
MD Onofrei, KL Butler, DC Fuke, HB. Miller.
Safety of statin therapy in patients with preexisting liver disease.
Pharmacotherapy, 28 (2008), pp. 522-529
[18]
EK Speliotes, M Balakrishnan, LS Friedman, KE. Corey.
Treatment of Dyslipidemia in Common Liver Diseases.
Clin Gastroenterol Hepatol, 16 (2018), pp. 1189-1196
[19]
MS Hedrington, SN. Davis.
Peroxisome proliferator-activated receptor alpha-mediated drug toxicity in the liver.
Expert opinion on drug metabolism & toxicology, 14 (2018), pp. 671-677
[20]
Q Geng, JY Ren, SF Li, H. Chen.
[A meta-analysis on the safety of combination therapy with fenofibrate and statins].
Zhonghua xin xue guan bing za zhi, 41 (2013), pp. 1063-1068
[21]
AJ Sanyal, MF Abdelmalek, A Suzuki, OW Cummings, M Chojkier, E-AS. Group.
No significant effects of ethyl-eicosapentanoic acid on histologic features of nonalcoholic steatohepatitis in a phase 2 trial.
Gastroenterology, 147 (2014), pp. 377-384
[22]
E Scorletti, L Bhatia, KG McCormick, GF Clough, K Nash, L Hodson, et al.
Effects of purified eicosapentaenoic and docosahexaenoic acids in nonalcoholic fatty liver disease: results from the Welcome* study.
Hepatology (Baltimore, Md), 60 (2014), pp. 1211-1221
[23]
P Dongiovanni, S Petta, V Mannisto, RM Mancina, R Pipitone, V Karja, et al.
Statin use and non-alcoholic steatohepatitis in at risk individuals.
J Hepatol, 63 (2015), pp. 705-712
[24]
K Kargiotis, VG Athyros, O Giouleme, N Katsiki, E Katsiki, P Anagnostis, et al.
Resolution of non-alcoholic steatohepatitis by rosuvastatin monotherapy in patients with metabolic syndrome.
World J Gastroenterol, 21 (2015), pp. 7860-7868
[25]
MM Islam, TN Poly, BA Walther, HC Yang, YC. Jack Li.
Statin Use and the Risk of Hepatocellular Carcinoma: A Meta-Analysis of Observational Studies.
Cancers (Basel), 12 (2020),
[26]
P Angulo, DE Kleiner, S Dam-Larsen, LA Adams, ES Bjornsson, P Charatcharoenwitthaya, et al.
Liver Fibrosis, but No Other Histologic Features, Is Associated With Long-term Outcomes of Patients With Nonalcoholic Fatty Liver Disease.
Gastroenterology, 149 (2015), pp. 389-397
[27]
R Tariq, P Axley, AK. Singal.
Extra-Hepatic Manifestations of Nonalcoholic Fatty Liver Disease: A Review.
J Clin Exp Hepatol, 10 (2020), pp. 81-87
[28]
G Musso, R Gambino, M Cassader, G. Pagano.
Meta-analysis: natural history of non-alcoholic fatty liver disease (NAFLD) and diagnostic accuracy of non-invasive tests for liver disease severity.
[29]
I Graham, MT Cooney, D Bradley, A Dudina, Z. Reiner.
Dyslipidemias in the prevention of cardiovascular disease: risks and causality.
Current cardiology reports, 14 (2012), pp. 709-720
[30]
N Katsiki, DP Mikhailidis, CS. Mantzoros.
Non-alcoholic fatty liver disease and dyslipidemia: An update.
Metabolism, 65 (2016), pp. 1109-1123
[31]
KE Corey, J Misdraji, L Gelrud, H Zheng, RT Chung, RM. Krauss.
Nonalcoholic steatohepatitis is associated with an atherogenic lipoprotein subfraction profile.
Lipids in health and disease, 13 (2014), pp. 100
[32]
VG Athyros, K Tziomalos, TD Gossios, T Griva, P Anagnostis, K Kargiotis, et al.
Safety and efficacy of long-term statin treatment for cardiovascular events in patients with coronary heart disease and abnormal liver tests in the Greek Atorvastatin and Coronary Heart Disease Evaluation (GREACE) Study: a post-hoc analysis.
Lancet, 376 (2010), pp. 1916-1922
[33]
MJ Tikkanen, R Fayyad, O Faergeman, AG Olsson, CC Wun, R Laskey, et al.
Effect of intensive lipid lowering with atorvastatin on cardiovascular outcomes in coronary heart disease patients with mild-to-moderate baseline elevations in alanine aminotransferase levels.
Int J Cardiol, 168 (2013), pp. 3846-3852
[34]
O Shahab, R Biswas, J Paik, H Bush, P Golabi, ZM. Younossi.
Among Patients With NAFLD, Treatment of Dyslipidemia Does Not Reduce Cardiovascular Mortality.
Hepatol Commun, 2 (2018), pp. 1227-1234
[35]
E Pose, L Napoleone, A Amin, D Campion, C Jimenez, S Piano, et al.
Safety of two different doses of simvastatin plus rifaximin in decompensated cirrhosis (LIVERHOPE-SAFETY): a randomised, double-blind, placebo-controlled, phase 2 trial.
Lancet Gastroenterol Hepatol, 5 (2020), pp. 31-41
[36]
E Pose, J Trebicka, RP Mookerjee, P Angeli, P. Gines.
Statins: Old drugs as new therapy for liver diseases?.
J Hepatol, 70 (2019), pp. 194-202
[37]
H Bays, DE Cohen, N Chalasani, SA Harrison.
The National Lipid Association's Statin Safety Task F. An assessment by the Statin Liver Safety Task Force: 2014 update.
Journal of clinical lipidology, 8 (2014), pp. S47-S57
[38]
CK Argo, JT Patrie, C Lackner, TD Henry, EE de Lange, AL Weltman, et al.
Effects of n-3 fish oil on metabolic and histological parameters in NASH: a double-blind, randomized, placebo-controlled trial.
J Hepatol, 62 (2015), pp. 190-197
[39]
VG Athyros, DP Mikhailidis, TP Didangelos, OI Giouleme, EN Liberopoulos, A Karagiannis, et al.
Effect of multifactorial treatment on non-alcoholic fatty liver disease in metabolic syndrome: a randomised study.
Curr Med Res Opin, 22 (2006), pp. 873-883
[40]
M Basaranoglu, O Acbay, A. Sonsuz.
A controlled trial of gemfibrozil in the treatment of patients with nonalcoholic steatohepatitis.
Journal of hepatology, 31 (1999), pp. 384
[41]
DC Chan, GF Watts, SK Gan, EM Ooi, PH. Barrett.
Effect of ezetimibe on hepatic fat, inflammatory markers, and apolipoprotein B-100 kinetics in insulin-resistant obese subjects on a weight loss diet.
Diabetes care, 33 (2010), pp. 1134-1139
[42]
S Dasarathy, J Dasarathy, A Khiyami, L Yerian, C Hawkins, R Sargent, et al.
Double-blind randomized placebo-controlled clinical trial of omega 3 fatty acids for the treatment of diabetic patients with nonalcoholic steatohepatitis.
Journal of clinical gastroenterology, 49 (2015), pp. 137-144
[43]
YH Li, LH Yang, KH Sha, TG Liu, LG Zhang, XX. Liu.
Efficacy of poly-unsaturated fatty acid therapy on patients with nonalcoholic steatohepatitis.
World journal of gastroenterology, 21 (2015), pp. 7008-7013
[44]
R Loomba, CB Sirlin, B Ang, R Bettencourt, R Jain, J Salotti, et al.
Ezetimibe for the treatment of nonalcoholic steatohepatitis: assessment by novel magnetic resonance imaging and magnetic resonance elastography in a randomized trial (MOZART trial).
Hepatology (Baltimore, Md), 61 (2015), pp. 1239-1250
[45]
M Malaguarnera, M Vacante, C Russo, MP Gargante, M Giordano, G Bertino, et al.
Rosuvastatin reduces nonalcoholic fatty liver disease in patients with chronic hepatitis C treated with α-interferon and ribavirin: Rosuvastatin reduces NAFLD in HCV patients.
Hepatitis monthly, 11 (2011), pp. 92-98
[46]
A Nelson, DM Torres, AE Morgan, C Fincke, SA. Harrison.
A pilot study using simvastatin in the treatment of nonalcoholic steatohepatitis: A randomized placebo-controlled trial.
Journal of clinical gastroenterology, 43 (2009), pp. 990-994
[47]
Y Qin, Y Zhou, SH Chen, XL Zhao, L Ran, XL Zeng, et al.
Fish Oil Supplements Lower Serum Lipids and Glucose in Correlation with a Reduction in Plasma Fibroblast Growth Factor 21 and Prostaglandin E2 in Nonalcoholic Fatty Liver Disease Associated with Hyperlipidemia: A Randomized Clinical Trial.
[48]
H Rana, SS Yadav, HD Reddy, S Singhal, DK Singh, K. Usman.
Comparative Effect of Insulin Sensitizers and Statin on Metabolic Profile and Ultrasonographical Score in Non Alcoholic Fatty Liver Disease.
J Clin Diagn Res, 10 (2016), pp. OC19-OC23
[49]
S Rezaei, MR Sasani, M Akhlaghi, A. Kohanmoo.
Flaxseed oil in the context of a weight loss program ameliorates fatty liver grade in patients with non-alcoholic fatty liver disease: a randomized double-blind controlled trial.
[50]
E Scorletti, L Bhatia, KG McCormick, GF Clough, K Nash, L Hodson, et al.
Effects of purified eicosapentaenoic and docosahexaenoic acids in nonalcoholic fatty liver disease: results from the Welcome* study.
Hepatology (Baltimore, Md), 60 (2014), pp. 1211-1221
[51]
F Shojasaadat, P Ayremlou, A Hashemi, A Mehdizadeh, R. Zarrin.
A randomized controlled trial comparing effects of a low-energy diet with n-3 polyunsaturated fatty acid supplementation in patients with non-alcoholic fatty liver disease.
J Res Med Sci, 24 (2019), pp. 21
[52]
F Sofi, I Giangrandi, F Cesari, I Corsani, R Abbate, GF Gensini, et al.
Effects of a 1-year dietary intervention with n-3 polyunsaturated fatty acid-enriched olive oil on non-alcoholic fatty liver disease patients: a preliminary study.
Int J Food Sci Nutr, 61 (2010), pp. 792-802
[53]
L Spadaro, O Magliocco, D Spampinato, S Piro, C Oliveri, C Alagona, et al.
Effects of n-3 polyunsaturated fatty acids in subjects with nonalcoholic fatty liver disease.
Dig Liver Dis, 40 (2008), pp. 194-199
[54]
Y Takeshita, T Takamura, M Honda, Y Kita, Y Zen, K Kato, et al.
The effects of ezetimibe on non-alcoholic fatty liver disease and glucose metabolism: a randomised controlled trial.
Diabetologia, 57 (2014), pp. 878-890
[55]
D Tobin, M Brevik-Andersen, Y Qin, JK Innes, PC. Calder.
Evaluation of a High Concentrate Omega-3 for Correcting the Omega-3 Fatty Acid Nutritional Deficiency in Non-Alcoholic Fatty Liver Disease (CONDIN).
[56]
M Yaghoubi, S Jafari, B Sajedi, S Gohari, S Akbarieh, AH Heydari, et al.
Comparison of fenofibrate and pioglitazone effects on patients with nonalcoholic fatty liver disease.
Eur J Gastroenterol Hepatol, 29 (2017), pp. 1385-1388
[57]
FS Zhu, S Liu, XM Chen, ZG Huang, DW. Zhang.
Effects of n-3 polyunsaturated fatty acids from seal oils on nonalcoholic fatty liver disease associated with hyperlipidemia.
Article options
Tools
es en pt

¿Es usted profesional sanitario apto para prescribir o dispensar medicamentos?

Are you a health professional able to prescribe or dispense drugs?

Você é um profissional de saúde habilitado a prescrever ou dispensar medicamentos