Heymann G, Rahman S, Kats D, Banini BA, Gaddam S, Aslanian E, Boateng S, Israel G, Muniraj T. Pancreatic steatosis is not associated with advanced steatohepatitis or fibrosis in metabolic dysfunction-associated steatotic liver disease. World J Gastroenterol 2025; 31(47): 114651 [DOI: 10.3748/wjg.v31.i47.114651]
Corresponding Author of This Article
Thiruvengadam Muniraj, MD, Department of Medicine, Yale University School of Medicine, 333 Cedar Street, 1080 LMP, New Haven, CT 06520, United States. thiruvengadam.muniraj@yale.edu
Research Domain of This Article
Gastroenterology & Hepatology
Article-Type of This Article
Retrospective Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Dec 21, 2025 (publication date) through Dec 19, 2025
Times Cited of This Article
Times Cited (0)
Journal Information of This Article
Publication Name
World Journal of Gastroenterology
ISSN
1007-9327
Publisher of This Article
Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA
Share the Article
Heymann G, Rahman S, Kats D, Banini BA, Gaddam S, Aslanian E, Boateng S, Israel G, Muniraj T. Pancreatic steatosis is not associated with advanced steatohepatitis or fibrosis in metabolic dysfunction-associated steatotic liver disease. World J Gastroenterol 2025; 31(47): 114651 [DOI: 10.3748/wjg.v31.i47.114651]
Author contributions: Muniraj T and Heymann G had full access to all of the data in the study and takes responsibility for the integrity of the data and accuracy of the data analysis; Heymann G, Rahman S, Kats S, Banini BA, Gaddam S, Aslanian H, Boateng S, Israel G, and Muniraj T are responsible for the study concept and design, acquisition of data, analysis and interpretation of data and drafting of the manuscript.
Institutional review board statement: This study was reviewed and approved by the Ethics Committee of Yale University.
Informed consent statement: All study participants or their legal guardian provided informed written consent about personal and medical data collection prior to study enrolment.
Conflict-of-interest statement: There is no conflict of interest associated with any of the senior author or other coauthors contributed their efforts in this manuscript.
Data sharing statement: No additional data are available.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Thiruvengadam Muniraj, MD, Department of Medicine, Yale University School of Medicine, 333 Cedar Street, 1080 LMP, New Haven, CT 06520, United States. thiruvengadam.muniraj@yale.edu
Received: September 26, 2025 Revised: October 19, 2025 Accepted: November 3, 2025 Published online: December 21, 2025 Processing time: 85 Days and 15 Hours
Abstract
BACKGROUND
Visceral fat deposition in the pancreas in the absence of significant alcohol use is termed non-alcoholic fatty pancreas disease (NAFPD) and is closely associated with metabolic dysfunction-associated steatotic liver disease (MASLD). However, few studies have assessed the relationship between the severity of NAFPD and the degree of hepatic inflammation and fibrosis in patients with MASLD.
AIM
To evaluate how NAFPD correlates with degrees of hepatic steatosis, steatohepatitis, and hepatic fibrosis in patients with MASLD.
METHODS
We performed a retrospective cohort study of patients in the Yale New Haven Health System with a diagnosis of MASLD. Chart and primary imaging data were reviewed to evaluate the degree of pancreatic steatosis and its relationship to hepatic steatosis, steatohepatitis, fibrosis, and other metabolic parameters.
RESULTS
Ninety-nine participants were identified who met additional inclusion criteria (liver biopsy and non-contrast enhanced computed tomography scan of the abdomen). 76 out of the 99 patients in our cohort met the imaging criteria for NAFPD. However, there was no association between the degree of pancreatic steatosis and hepatic steatosis (either on imaging or biopsy), or the degree of pancreatic steatosis and advanced forms of MASLD, such as the degree of metabolic dysfunction-associated steatohepatitis or stage of hepatic fibrosis.
CONCLUSION
MASLD and NAFPD are co-occurring diseases resulting from and contributing to metabolic dysregulation. Our study confirms this association but does not support a strong association between pancreatic steatosis and hepatic steatohepatitis or fibrosis in this cohort; larger prospective, longitudinal studies are needed in the future to better define the complex interplay of MASLD, NAFPD, and metabolic health.
Core Tip: Non-alcoholic fatty pancreas disease (NAFPD) and metabolic dysfunction-associated steatotic liver disease (MASLD) are commonly co-occurring disorders associated with metabolic dysregulation with important clinical implications. Few studies have examined the relationship between these conditions in terms of disease severity. We perform a retrospective analysis of patients with biopsy-proven MASLD, and non-contrast enhanced abdominal computed tomography to assess how NAFPD correlates with degrees of hepatic steatosis, steatohepatitis, and hepatic fibrosis. We find that most patients with MASLD met criteria for NAFPD, but the degree of pancreatic steatosis did not portend a greater degree of hepatic steatosis, steatohepatitis, or hepatic fibrosis.
Citation: Heymann G, Rahman S, Kats D, Banini BA, Gaddam S, Aslanian E, Boateng S, Israel G, Muniraj T. Pancreatic steatosis is not associated with advanced steatohepatitis or fibrosis in metabolic dysfunction-associated steatotic liver disease. World J Gastroenterol 2025; 31(47): 114651
Metabolic syndrome and central obesity can lead to visceral fat deposition. This process in the liver results in hepatic steatosis or metabolic dysfunction-associated steatotic liver disease (MASLD, previously known as non-alcoholic fatty liver disease or NAFLD), an increasingly common etiology of chronic liver disease that can progress to liver inflammation [steatohepatitis or metabolic dysfunction-associated steatohepatitis (MASH)] and fibrosis[1]. An emerging clinical entity is non-alcoholic fatty pancreas disease (NAFPD), characterized by fat infiltration of the pancreatic parenchyma with further progression of insulin resistance and metabolic dysregulation[2,3]. Many studies have shown a significant association between MASLD and NAFPD, independent of co-occurring metabolic parameters[4-9]. However, the few studies that examined the relationship between NAFPD and advanced forms of MASLD, such as steatohepatitis (MASH) and advanced liver fibrosis, have demonstrated conflicting results. Rosenblatt et al[10] found that pancreatic steatosis measured by ultrasonography was positively correlated with advancing liver fibrosis and that ‘extensive’ fatty pancreas was associated with MASH. Measuring pancreatic steatosis with proton-density fat fraction on magnetic resonance imaging (MRI-PDFF) in a small cohort of patients with biopsy and biochemically evidenced MASH, Patel et al[11] found that pancreatic fat inversely correlated with advancing fibrosis stage. Two subsequent studies using the same MRI-PDFF technique in patients with biopsy-proven MASLD found no association between pancreatic fat content and hepatic steatosis, steatohepatitis, or liver fibrosis[12,13]. Whether these varying results are related to different imaging techniques (ultrasound vs MRI-PDFF), patient selection, or stage of disease progression remains unclear. We use computed tomography (CT), a more commonly available modality of abdominal imaging in patients with biopsy-proven MASLD to assess the relationship between NAFPD and MASLD.
Steatohepatitis resulting from lipotoxicity is thought to drive fibrotic remodeling of the liver[14,15], and advanced fibrosis portends poorer outcomes in MASLD[16,17], highlighting the importance of understanding the predictors and contributors of MASH and fibrosis progression. Given the differing results of prior studies examining the association between NAFPD and hepatic inflammation and fibrosis in patients with MASLD, we aimed to clarify this relationship further. In this study, we hypothesize that pancreatic and hepatic steatosis are co-occurring conditions, but that pancreatic steatosis does not predict advanced hepatic fibrosis or steatohepatitis. Understanding the interaction of these disorders in the context of major clinical outcomes may help guide clinical surveillance and management.
MATERIALS AND METHODS
Study design
This is an Institutional Review Board approved, retrospective cohort study using the Yale New Haven Health System Epic EMR database. A query of the Yale New Haven Health System database revealed 1614 patients with a chart diagnosis of NAFLD or MASLD, a history of liver biopsy, and a history of abdominal imaging. These patient charts were reviewed to ensure a historical and biopsy-confirmed diagnosis of MASLD without other etiologies of chronic liver disease as well as a non-contrast CT scan of the abdomen within 1 year of biopsy, resulting in 99 patients for study inclusion (Figure 1). Liver biopsy reports were then reviewed for the grade of hepatic steatosis and steatohepatitis, and the stage of fibrosis. CT scans were reviewed by radiologists (Rahman S and Israel G) for quantification of pancreatic and hepatic steatosis.
Figure 1 Flow chart of study design and inclusion criteria.
NAFPD: Non-alcoholic fatty pancreas disease; MASLD: Metabolic dysfunction-associated steatotic liver disease; MASH: Metabolic dysfunction-associated steatohepatitis; CT: Computed tomography.
Demographic data
Charts were reviewed to gather baseline demographic, metabolic, and laboratory data. Attempts were made to obtain data from a time point approximately midway between the CT scan and liver biopsy. A diagnosis of diabetes was based on the presence of an elevated A1c or the use of an anti-glycemic agent at the time of CT scan or liver biopsy. A diagnosis of hypertension or dyslipidemia was made based on the use of blood pressure or lipid-lowering therapy, respectively, at the time of CT scan or liver biopsy. Alanine transaminase, aspartate transaminase, platelet count, and albumin level were used to calculate Fibrosis-4 index (FIB-4) and NAFLD fibrosis scores.
Histopathologic liver grading
Liver Biopsy reports were reviewed for steatosis, steatohepatitis, and fibrosis. While many reports explicitly stated the degrees and stage, some required interpretation of the description and classification using the Brunt et al[18] scoring system.
Imaging analyses
Non-contrast-enhanced CT images of the abdomen were evaluated by Dr. Rahman under the guidance of Dr. Israel. To quantify pancreatic steatosis, density measurements were performed in the pancreatic head, body, and tail, with the average value subtracted from the density measurement of the spleen for internal normalization as previously described[19-22]. To quantify hepatic steatosis, density measurements were performed in the left and right hepatic lobes, with the average value included in the analysis as previously described[21,23,24].
Statistical analyses
The Shapiro-Wilk test was applied to variables to determine normality and to guide the use of the appropriate correlation and comparison test. The analyses were performed in SPSS and GraphPad Prism. An adjusted P value using Bonferroni correction was applied to the multiple comparisons in Tables 1 and 2. Several statistical analyses were performed to assess whether our sample was adequately powered. For the correlation analysis between pancreatic steatosis and hepatic steatosis, with an effect size of r = 0.3, a sample size of 85 participants was required to achieve 80% power at a significance level of 0.05 (Figure 2A and B). For the following non-parametric comparisons, a 15% increase in sample size was added to account for the lower statistical power of these tests[25]. For comparisons across multiple groups in relation to pancreatic steatosis, we assumed a moderate effect size of Cohen’s f = 0.25 with a power of 80% at a significance level of 0.05, resulting in sample size requirements of 52 participants per group for comparisons across hepatic steatosis grades (Figure 2C), 60 participants per group for comparisons across MASH grades (Figure 3A), and 45 participants per group for comparisons across fibrosis stages (Figure 3B). For the comparison between minimal and advanced fibrosis categories in relation to pancreatic steatosis, we assumed a moderate size of Cohen’s d = 0.5 with a power of 80% at a significance level of 0.05, resulting in a sample size requirement of 74 participants per group (Figure 3C).
Figure 2 Relationship between pancreatic and hepatic steatosis.
A: Scatterplot for pancreatic steatosis by hepatic steatosis on computed tomography (n = 99), Pearson correlation test, r = 0.04, P = 0.71; B: Scatterplot for pancreatic steatosis by percent hepatic steatosis on biopsy (n = 99), Spearman correlation test, r = -0.12, P = 0.29; C: Box and whiskers plot with all values included for pancreatic steatosis by steatosis grade on biopsy (n = 11 for grade 0, n = 33 for grade 1, n = 35 for grade 2, n = 20 for grade 3), Kruskal-Wallis test, P = 0.54.
Figure 3 Relationship between pancreatic steatosis and metabolic dysfunction-associated steatohepatitis grade and fibrosis stage.
A: Box and whiskers plot with all values included for pancreatic steatosis by metabolic dysfunction-associated steatohepatitis grade on biopsy (n = 22 for grade 0, n = 66 for grade 1, n = 11 for grade 2 and 3), Kruskal-Wallis test, P = 0.99; B: Box and whiskers plot with all values included for pancreatic steatosis by fibrosis stage on biopsy (n = 15 for stage 0, n = 20 for stage 1, n = 16 for stage 2, n = 10 for stage 3, n = 38 for stage 4), Kruskal-Wallis test, P = 0.98; C: Box and whiskers plot with all values included for pancreatic steatosis by fibrosis category [minimal 0 to 2 (n = 51) vs advanced 3 to 4 (n = 48)] on biopsy, Wilcoxon test, P > 0.99. MASH: Metabolic dysfunction-associated steatohepatitis.
Table 1 Baseline characteristics of study participants compared by stage of fibrosis.
A comparison of baseline characteristics by fibrosis stage revealed a significant increase in age (individual stage 0-4, one-way ANOVA, P = 0.003; minimal vs advanced, t-test, P = 0.002), hemoglobin A1c by individual fibrosis stage (stage 0-4, Kruskal-Wallis test, P = 0.005) but not by fibrosis category (minimal vs advanced, Wilcoxon test, P = 0.03), history of diabetes (individual stage 0-4, χ2 test, P = 0.001; minimal vs advanced, χ2 test, P = 0.009), and history of hypertension by individual fibrosis stage (stage 0-4, χ2 test, P = 0.006) but not by fibrosis category (minimal vs advanced, χ2 test, P = 0.03). There was no difference in male-female distribution, average body mass index, or prevalence of dyslipidemia. FIB-4 and NAFLD fibrosis scores significantly increased with advancing fibrosis [FIB-4: Individual stage 0-4, Kruskal-Wallis test, P < 0.0001; minimal vs advanced, Wilcoxon test, P < 0.0001; NAFLD fibrosis score: Individual stage 0-4, one-way ANOVA, P < 0.0001; minimal vs advanced, t-test, P < 0.0001 (Tables 1 and 2)]. A comparison of baseline characteristics and FIB-4 and NAFLD fibrosis scores by MASH grade revealed no significant differences (Table 3). Pancreatic steatosis was significantly correlated with older age (Pearson correlation test, r = -0.21, P = 0.03) and elevated A1c (Spearman correlation test, r = -0.25, P = 0.02) but not BMI, FIB-4, or NAFLD fibrosis scores.
Table 3 Baseline characteristics of study participants compared by metabolic dysfunction-associated steatohepatitis grade.
The accepted threshold for pancreatic steatosis on CT scan using the spleen for internal normalization is -5 Hounsfield Units (HU), with a lower value indicating a greater degree of steatosis[21,26]. In our sample, 76 out 99 patients (77%) of the patients in our cohort met this criterion for pancreatic steatosis and the average pancreatic steatosis measurement was -17.3 ± 1.85 HU, consistent with an overall association between MASLD and NAFPD. However, within our study population, there is no association between the degree of pancreatic steatosis and hepatic steatosis, either when quantified on CT scan (Figure 2A, Pearson correlation test, r = 0.04, P = 0.71), percent steatosis on biopsy (Figure 2B, Spearman correlation test, r = -0.12, P = 0.29), or when stratified by steatosis grade on biopsy (Figure 2C, Kruskal-Wallis test, P = 0.54).
Pancreatic steatosis and MASH grade
We do not see a significant association when comparing the degree of pancreatic steatosis across MASH grades (Figure 3A, Kruskal-Wallis test, P = 0.99) with average pancreatic density measurements of -16.6 ± 3.7 HU for grade 0, -16.6 ± 2.1 HU for grade 1, and -23 ± 8.3 HU for grades 2 and 3 (Figure 3A).
Pancreatic steatosis and fibrosis stage
Similarly, we do not see a significant association when comparing the degree of pancreatic steatosis by individual fibrosis stage with average pancreatic density measurements of -16.6 ± 3.7 HU for stage 0, -16.6 ± 2.1 HU for stage 1, -23 ± 8.3 HU for stage 2, -23 ± 8.3 HU for stage 3, and -23 ± 8.3 HU for stage 4 (Figure 3B, Kruskal-Wallis test, P = 0.98). We also stratified our sample into minimal (individual fibrosis stages 0, 1, and 2) and advanced fibrosis (individual fibrosis stages 3 and 4) groups and again did not see a significant relationship between fibrosis category and pancreatic steatosis measurements (Figure 3C, Wilcoxon rank sum test, P > 0.99).
DISCUSSION
With the increasing prevalence of obesity and components of metabolic dysregulation, understanding the complex interplay between pancreas and liver pathophysiology is a principal public health concern. Several studies have shown a dynamic relationship between MASLD, NAFPD, and other parameters of metabolic dysregulation[4,5,7,26-29]. While the association between MASLD and NAFPD is well established[26,30,31], only four retrospective studies in the adult literature evaluate how the severity of these diseases correspond with each other, with contradictory findings[10-13]. In the present study, we aimed to further elucidate this relationship.
Based on the existing literature, we hypothesized that pancreatic steatosis leads to increasing hepatic steatosis, but that increasing degrees of pancreatic steatosis would not be associated with advanced hepatic inflammation and fibrosis. While a majority of the MASLD patients in this study met the imaging criterion for NAFPD (76 out of 99 patients exceeded the -5 HU threshold with an average pancreatic steatosis measurement of -17.3 ± 1.85 HU), contrary to our hypothesis, we do not see a correlation between the degree of pancreatic steatosis and hepatic steatosis (Figure 2). These results are consistent with three of the four published studies: Rosenblatt et al[10] study in which they found a weak, non-significant correlation between pancreatic and hepatic steatosis using ultrasonography; Mak et al[12] and Idilman et al[13] did not find a relationship between pancreatic steatosis on MRI-PDFF and hepatic steatosis on MRI-PDFF or biopsy. Interestingly, Patel et al[11] found a significant increase in pancreatic fat measurements with histologic hepatic steatosis grade while using the same MRI-PDFF imaging modality. One possible explanation for these findings is the phenomenon of burnt-out steatohepatitis, where over time, the previously steatotic hepatic parenchyma is replaced with fibrotic tissue, thus leading to reduced measurements of hepatic steatosis on biopsy or imaging. To test this hypothesis, we reanalyzed our data set excluding all patients with stage 4 fibrosis noted to have biopsy findings consistent with burnt-out steatohepatitis (n = 12). This cohort demonstrated a weaker correlation between pancreatic and hepatic steatosis (CT scan: Pearson correlation test, r = 0.05, P = 0.62; Biopsy: Spearman correlation test, r = -0.06, P = 0.29) that remains non-significant (Figure 4A and B). The same applies to this subpopulation by steatosis grade on biopsy (Figure 4C: Kruskal-Wallis test, P = 0.79).
Figure 4 Relationship between pancreatic and hepatic steatosis after exclusion of patients with ‘burnt-out steatohepatitis’.
A: Scatterplot for pancreatic steatosis by hepatic steatosis on computed tomography (n = 87), Pearson correlation test, r = 0.05, P = 0.62; B: Scatterplot for pancreatic steatosis by percent hepatic steatosis on biopsy (n = 87), Spearman correlation test, r = -0.06, P = 0.29; C: Box and whiskers plot with all values included for pancreatic steatosis by steatosis grade on biopsy (n = 3 for grade 0, n = 29 for grade 1, n = 35 for grade 2, n = 20 for grade 3), Kruskal-Wallis test, P = 0.79.
Our primary focus in this study was related to the relationship between pancreatic steatosis and steatohepatitis and fibrosis, as inflammation resulting from hepatic steatosis is known to drive fibrotic remodeling of the liver, and fibrosis is the greatest independent predictor of mortality in patients with MASLD[14-17]. Consistent with our hypothesis, we found that increasing pancreatic steatosis does not correlate with increased liver inflammation or fibrosis. These results are consistent with the studies by Mak et al[12], which also found no association between pancreatic steatosis and steatohepatitis or liver fibrosis on biopsy, as well as Idilman et al[13], which found no association between pancreatic steatosis and liver fibrosis, although they do not comment on steatohepatitis. Patel et al[11], however, found increased pancreatic fat measurements in patients without fibrosis (stage 0) compared to those with fibrosis (stage 1 to 4), although there was no difference when comparing pancreatic fat among fibrosis stages 1 through 4. Rosenblatt et al[10] is the only study to demonstrate a positive association between extensive pancreatic steatosis and hepatic inflammation as well as advanced hepatic fibrosis. However, the study utilizes ultrasound for pancreatic fat quantification and notes significant age differences between those with extensive vs minimal or absent fatty pancreas. Pancreatic echogenicity on ultrasound is less sensitive compared to other imaging modalities[32,33] and is known to increase with age[34], which is also associated with advanced liver fibrosis in patients with MASLD[15], thus presenting a possible confounder for their results.
Our results have several possible explanations. Most simply, it is possible that pancreatic steatosis impacts the development of MASLD and possibly the degree of hepatic steatosis but has no effect on subsequent progression to MASH or fibrosis. Alternatively, a positive correlation may exist, but fibrotic remodeling of the pancreas also occurs, similar to that of the liver, with a resultant decrease in pancreatic steatosis, hindering the ability to ascertain a significant association in this cross-sectional and retrospective analysis.
From a methodological perspective, there are several possible limitations in our study to acknowledge. First, our sample is relatively small and limited to patients with a liver biopsy and abdominal CT scan, likely selecting a non-generalizable cohort of MASLD patients, specifically those with milder disease or possibly with consideration of additional etiologies to their chronic liver disease. Based on our power analysis, our sample sizes per group do not meet the threshold to detect small to moderate differences in some of our comparisons. Second, we used non-contrast-enhanced CT scan as our imaging modality to increase our sample size and mitigate the variable of contrast bolus timing affecting our measurements, but it is not the gold standard for measuring pancreatic and hepatic steatosis and tends to underestimate hepatic steatosis in advanced liver fibrosis.
CONCLUSION
Given the increasing burden of metabolic dysregulation and its association with MASLD and NAFPD, further research is needed to understand the dynamic and complex interactions between these disorders. Our findings do not support a strong association between pancreatic steatosis and hepatic fibrosis in this cohort. Larger prospective studies are needed in the future to examine the longitudinal progression of metabolic parameters, MASLD progression to MASH and liver fibrosis, and overall health outcomes in relation to pancreatic steatosis to better establish a causal relationship warranting clinical consideration.
Footnotes
Provenance and peer review: Invited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Gastroenterology and hepatology
Country of origin: United States
Peer-review report’s classification
Scientific Quality: Grade B, Grade B, Grade C
Novelty: Grade A, Grade B, Grade B
Creativity or Innovation: Grade B, Grade B, Grade B
Scientific Significance: Grade B, Grade B, Grade C
P-Reviewer: Li X, Academic Fellow, Associate Chief Physician, China; Liu DF, MD, Professor, China S-Editor: Qu XL L-Editor: A P-Editor: Wang WB
Mak AL, Wassenaar N, van Dijk AM, Troelstra M, Houttu V, van Son K, Driessen S, Zwirs D, van den Berg-Faay S, Shumbayawonda E, Runge J, Doukas M, Verheij J, Beuers U, Nieuwdorp M, Cahen DL, Nederveen A, Gurney-Champion O, Holleboom A. Intrapancreatic fat deposition is unrelated to liver steatosis in metabolic dysfunction-associated steatotic liver disease.JHEP Rep. 2024;6:100998.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 7][Reference Citation Analysis (0)]
Dulai PS, Singh S, Patel J, Soni M, Prokop LJ, Younossi Z, Sebastiani G, Ekstedt M, Hagstrom H, Nasr P, Stal P, Wong VW, Kechagias S, Hultcrantz R, Loomba R. Increased risk of mortality by fibrosis stage in nonalcoholic fatty liver disease: Systematic review and meta-analysis.Hepatology. 2017;65:1557-1565.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 984][Cited by in RCA: 1473][Article Influence: 184.1][Reference Citation Analysis (0)]
