Clinical and histological determinants of nonalcoholic steatohepatitis and advanced fibrosis in elderly patients

Authors


  • Potential conflict of interest: Dr. Behling consults for Robarts and Sundise. Dr. Sanyal consults for and received grants from Gilead, Ikaria, Salix, and Takeda. He consults for Abbott, Genentech, Merck, Norgine, Roche, Nitto Denko, and Bristol-Myers Squibb. He received grants from Exhalenz, Genentech, Gore, Intercept, Astellas, Novartis, and Galectin. He received royalties from Uptodate.

Abstract

The characteristics of nonalcoholic fatty liver disease (NAFLD) in elderly patients are unknown. Therefore, we aimed to examine the differences between elderly and nonelderly patients with NAFLD and to identify determinants of nonalcoholic steatohepatitis (NASH) and advanced fibrosis (bridging fibrosis or cirrhosis) in elderly patients. This is a cross-sectional analysis of adult participants who were prospectively enrolled in the NASH Clinical Research Network studies. Participants were included based on availability of the centrally reviewed liver histology data within 1 year of enrollment, resulting in 61 elderly (age ≥65 years) and 735 nonelderly (18-64 years) participants. The main outcomes were the presence of NASH and advanced fibrosis. Compared to nonelderly patients with NAFLD, elderly patients had a higher prevalence of NASH (56% versus 72%, P = 0.02), and advanced fibrosis (25% versus 44%, P = 0.002). Compared to nonelderly patients with NASH, elderly patients with NASH had higher rates of advanced fibrosis (35% versus 52%, P = 0.03), as well as other features of severe liver disease including the presence of ballooning degeneration, acidophil bodies, megamitochondria, and Mallory-Denk bodies (P ≤ 0.05 for each). In multiple logistic regression analyses, independent determinants of NASH in elderly patients included higher aspartate aminotransferase (AST) (odds ratio [OR] = 1.12, P = 0.007) and lower platelets (OR = 0.98, P = 0.02); and independent determinants of advanced fibrosis included higher AST (OR = 1.08, P = 0.007), lower alanine aminotransferase value (OR = 0.91, P = 0.002), and an increased odds of having low high-density lipoprotein (OR = 8.35, P = 0.02). Conclusion: Elderly patients are more likely to have NASH and advanced fibrosis than nonelderly patients with NAFLD. Liver biopsy may be considered in elderly patients and treatment should be initiated in those with NASH and advanced fibrosis. (HEPATOLOGY 2013;58:1644–1654)

Abbreviations
ALT

alanine aminotransferase

AST

aspartate aminotransferase

BMI

body mass index

CHD

coronary heart disease

GGT

gamma glutamyl trans-peptidase

HbA1c

hemoglobin A1c

HBP

high blood pressure

HDL, high-density lipoprotein; HOMA-IR

homeostasis model assessment of insulin resistance

LDL

low-density lipoprotein

NAFLD

nonalcoholic fatty liver disease

NASH

nonalcoholic steatohepatitis

Nonalcoholic Steatohepatitis Clinical Research Network (NASH CRN) NIDDK

National Institute of Diabetes and Digestive and Kidney Diseases

Nonalcoholic fatty liver disease (NAFLD) afflicts one in every three adult Americans and it is the most common cause of elevated serum aminotransferases in the United States.[1-4] NAFLD is seen in individuals who consume little or no alcohol. It can range from the presence of steatosis alone, which is expected to have a nonprogressive course, to nonalcoholic steatohepatitis (NASH), the progressive form of NAFLD that can lead to advanced fibrosis, cirrhosis, and hepatocellular carcinoma in a subset of patients.[3, 5-8] Liver biopsy in NASH is typically characterized by steatosis, lobular inflammation, and ballooning degeneration with or without perisinusoidal fibrosis.[9, 10]

Children and adolescents with NAFLD may have a different pattern of liver injury than adult patients with NAFLD.[11-13] This suggests that as an individual grows or ages NAFLD phenotypes may vary. However, there are limited data examining whether we see a different pattern of liver histology in elderly patients with NAFLD. Several groups have now shown that older age is a risk factor for NASH and advanced fibrosis in patients with NAFLD.[14, 15] Recent studies have suggested that a higher prevalence of NAFLD and more advanced fibrosis may be seen in elderly patients.[16, 17] However, little is known about the characteristics and histology of NAFLD in elderly patients.

The U.S. population is aging due to the steady rise in life expectancy.[18-20] In 2010, ∼40 million Americans were older than 65 years. By the year 2030 this age group of Americans is estimated to rise to more than 70 million.[21] The aging of the American population underscores the importance of studying the characteristics of NAFLD in elderly patients. Finally, a recent study found that alanine aminotransferase (ALT) decreases with age,[22] which may cause significant disease to be overlooked in elderly patients if that is the sole determining criterion for a referral to a specialist.

The main aims of this study were to investigate the clinical and histological characteristics of NASH and fibrosis in elderly patients compared to nonelderly patients from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Nonalcoholic Steatohepatitis Clinical Research Network (NASH CRN) cohort, and to determine the characteristics associated with NASH in the elderly compared to nonelderly patients. In this study we hypothesized that elderly patients with NAFLD have more advanced disease, reflected by a higher prevalence of NASH and fibrosis, compared to younger adults.

Patients and Methods

Study Design and Setting

This was a cross-sectional analysis of adult patients with biopsy-proven NAFLD who were enrolled into either the NAFLD Database Study, a prospective cohort study, or the PIVENS (Pioglitazone versus Vitamin E versus Placebo for the Treatment of Nondiabetic Patients with Nonalcoholic Steatohepatitis; Clinical Trial number NCT00063622), a randomized, placebo-controlled, double-masked clinical trial, of the NIDDK sponsored NASH CRN consortium.[23, 24]

Participants were enrolled between 2004 through 2008 by one of the eight participating medical centers in the United States: University of California at San Diego (San Diego, CA); Duke University (Durham, NC); Case Western Reserve (Cleveland, OH); Indiana University (Indianapolis, IN); Saint Louis University (St. Louis, MO); University of California at San Francisco (San Francisco, CA); University of Washington (Seattle, WA); and Virginia Commonwealth University (Richmond, VA). All enrolled patients provided written informed consent before data collection. The clinical protocols, consent forms, and manual of operations were also reviewed and approved by a data safety monitoring board established by the NIDDK specifically for the NASH CRN. In addition, the protocol and informed consent were approved by the Institutional Review Board of each site. STROBE guidelines for cross-sectional studies were followed.[25]

Patient Population

Both the NAFLD Database and PIVENS treatment studies have been published.[23, 24, 26] Briefly, the inclusion criteria for the NAFLD Database required either histological diagnosis of NAFLD, imaging suggestive of NAFLD, histological diagnosis of cryptogenic cirrhosis, or clinical evidence of cryptogenic cirrhosis. Exclusion criteria included diagnosis of other chronic liver disease or suspected or proven hepatocellular carcinoma, or an average alcohol consumption >20 g daily for men, or >10 g average for women during the 2 years before entry. PIVENS inclusion additionally required patients to have histological evidence of NASH without cirrhosis and the absence of diabetes.

Inclusion Criteria for Subanalysis

To be included in the dataset for the analysis of this study, participants were required to have a biopsy within 1 year of enrollment that was evaluated through central reading by the NASH CRN Pathology Committee. Subjects were divided into the following groups: (1) patients who were 65 years or older at the time of their biopsy were defined as elderly,[27, 28] and (2) patients between 18 and 64 years of age were defined as nonelderly.

Covariates

The following characteristics were examined: demographic factors included age, sex, race (white versus other), and ethnicity (Hispanic versus not); anthropometrics included body mass index (BMI) and waist circumference; and clinical characteristics included reported diagnosis of hypertension and diabetes. Metabolic syndrome was defined as having 3 of the following 5 factors: impaired fasting glucose (≥110 mg/dL), large waist circumference (≥88 cm in women, ≥102 cm in men), hypertriglyceridemia (≥150 mg/dL), low high-density lipoprotein (HDL) cholesterol (<50 mg/dL in women, <40 mg/dL in men), high blood pressure (HBP) (systolic BP ≥130 mmHg or diastolic BP ≥85 mmHg). In addition, we also included smoking status (yes/no) and history of coronary heart disease (CHD) (yes/no) as a covariate. This analysis also included clinical laboratory tests including: aspartate aminotransferase (AST), alanine aminotransferase (ALT), the AST/ALT ratio, gamma glutamyl trans-peptidase (GGT), alkaline phosphatase (ALK), albumin, bilirubin, international normalized ratio (INR), platelet count, total cholesterol, HDL, low-density lipoprotein cholesterol (LDL), triglycerides, hemoglobin A1c (HbA1c), fasting glucose, fasting serum insulin, the homeostasis model assessment of insulin resistance (HOMA-IR) index, and ferritin. ALT, AST, and alkaline phosphatase upper limit of normal (ULN) were defined according to local reference ranges. The aspartate aminotransferase to platelet ratio index (APRI) score was defined as [(AST/ULN)/platelets] multiplied by 100.

Liver Histologic Assessment Section

The NASH CRN Pathology Committee consisted of nine liver pathologists who were blinded to all clinical and identifying data. Biopsies were scored by consensus during pathology committee meetings using the NASH CRN Histologic Scoring System.[9]

Briefly, the following variables were recorded and analyzed in this subanalysis. Steatosis evaluation included the grade of steatosis, location of steatosis, and presence (or absence) of microvesicular steatosis. The fibrosis stage was divided into four stages including stage 0: no fibrosis; stage 1: which is comprised of stage 1a: mild, zone 3, perisinusoidal fibrosis; stage 1b: moderate, zone 3, perisinusoidal fibrosis; stage 1c: portal/periportal fibrosis; stage 2: perisinusoidal and portal/periportal fibrosis; stage 3: bridging fibrosis; and stage 4: cirrhosis. The assessment of inflammation included the number of foci of lobular inflammation, the presence of microgranulomas, the presence of large lipogranulomas, and the degree of portal inflammation. The liver cell injury assessment included the presence of ballooning degeneration, acidophil bodies, pigmented macrophages, and megamitochondria. Other components were the presence of Mallory-Denk bodies (or Mallory Hyaline) and glycogenated nuclei. The histological assessment also included diagnostic classification of NASH and liver biopsies of the participants were classified into one of the three possible categories including not NASH, possible/borderline NASH, and definite NASH.

Primary Outcomes

The main outcome variables of this study were the presence of definite NASH and advanced fibrosis defined as either bridging fibrosis or cirrhosis. Secondary outcomes included other histologic variables.

Statistical Analysis

We conducted an exploratory analysis of baseline characteristics including demographic, anthropometric, clinical, laboratory measures, and histological features. Univariate analyses were performed using this set of characteristics among different study subgroup comparisons of interest: elderly to nonelderly patients with NAFLD to examine the differences in the pattern and severity of liver injury between the two groups; elderly patients with NASH to nonelderly patients with NASH to examine if features of NASH were distinct between the two groups. Finally, we developed a logistic regression model to examine the independent determinants of NASH and advanced fibrosis in elderly patients. Differences between the distributions between subgroups were assessed using Fisher's exact test for categorical and t test for continuous features. All histological features were treated as categorical. Univariate results were reported as means and standard deviations or percentages.

Independent predictors of either definite NASH or advanced fibrosis among elderly patients were determined using unadjusted and adjusted multiple logistic regression.[29] Odds ratios (OR), 95% confidence intervals (95% CI), and P-values were used to report the results. The candidate set for the multivariable-adjusted models was limited to features that have been linked to NAFLD and based on biological plausibility and included key demographics (age, race, ethnicity), smoking, h/o CHD, diabetes, components of metabolic syndrome (BMI, hypertriglycerides, low HDL, high blood pressure, insulin resistance [HOMA-IR]), and liver disease biomarkers (ALT, AST, GGT, platelets, ferritin). Ethnicity was not included in the candidate set for the advanced fibrosis model due to multicollinearity with metabolic traits. The adjusted model was determined from backward stepwise regression using a 0.05 level of significance of definite NASH and advanced fibrosis on the candidate set forcing age, gender, and race into the model. Final models were assessed using Hosmer-Lemeshow goodness of fit and the Akaike Information Criterion (AIC).[30-33]

All analyses were performed using STATA (v. 12) and SAS statistical software (v. 9.3).[34, 35] Nominal, two-sided P values were used and were considered statistically significant if P ≤ 0.05, a priori.

Results

Demographic, Clinical, and Biochemical Characteristics in Elderly Compared to Nonelderly Patients With NAFLD

Among the 796 patients with biopsy-proven NAFLD who met the inclusion criteria for this study, 61 patients age ≥65 years were classified into the elderly patients group, and the remaining 735 patients age 18-65 years were classified into the nonelderly patients group.

A detailed description of the cohort categorized into elderly versus nonelderly patients with NAFLD is shown in Table 1. Compared to nonelderly patients, the elderly patients group with NAFLD had more females and subjects were more likely to be hypertensive. The elderly patients group had a lower mean BMI and smaller waist circumference. Although the elderly patients group had a higher average AST and a lower average ALT, this difference was not statistically significant. The elderly patients group had a higher mean AST/ALT ratio, lower mean platelet count, and higher mean APRI score, all of which are suggestive of advanced liver disease.

Table 1. Demographic, Anthropometric, and Clinical Characteristics of Patients With NAFLD Enrolled in NASH CRN Studies by Age Group
 Age Group 
CharacteristicsNonelderly (18-64 years old) (N = 735)Elderly (≥ 65 years old) (N = 61)Pa
  1. a

    P values (two-sided) determined from either a Fisher's exact test for categorical variables or t test for continuous variables.

  2. b

    Ever smoked regularly defined as smoking at least one cigarette per day for a year; four nonelderly patients were missing values. Cardiovascular disease defined as ever diagnosed with cerebrovascular or coronary heart disease.

  3. c

    Abbreviations: AST: aspartate aminotransferase, ALT: alanine aminotransferase, GGT: gamma glutamyl trans-peptidase, HDL: high density lipoprotein, LDL: low density lipoprotein, HbA1c: hemoglobin A1c, HOMA-IR: the homeostasis model assessment of insulin resistance index.

Demographics & lifestyle:   
Male, n (%)295 (40%)14 (23%)0.01
Age (mean years ± SD)47 ± 1168 ± 3<0.001
White, n (%)593 (84%)46 (78%)0.20
Hispanic, n (%)98 (13%)7 (12%)0.84
Ever smoked regularly, n (%)b263 (36%)31 (51%)0.03
Clinical   
Hypertension, n (%)337 (46%)42 (69%)0.001
Cardiovascular disease (CVD), n (%)b33 (4%)11 (18%)<0.001
Type 2 diabetes, n (%)174 (24%)15 (25%)0.88
Metabolic syndrome, n (%)451 (61%) 34 (56%)0.41
Anthropometric (mean ± SD)   
Body mass index (kg/m2)35 ± 632 ± 5<0.001
Waist circumference (cm)109 ± 14103 ± 12<0.001
Hepatology panelc (mean ± SD)   
AST (U/L)54 ± 3661 ± 440.20
ALT (U/L)76 ± 5267 ± 440.16
AST/ALT ratio0.8 ± 0.41.0 ± 0.4<0.001
Alkaline phosphatase (ALK) (U/L)87 ± 3394 ± 370.13
GGT (U/L)70 ± 7877 ± 750.53
Albumin (g/dL)4.2 ± 0.44.2 ± 0.40.20
Bilirubin, total (mg/dL)0.8 ± 0.40.8 ± 0.40.26
Bilirubin, direct (mg/dL)0.1 ± 0.10.2 ± 0.20.03
International normalized ratio1.0 ± 0.21.0 ± 0.10.18
Other laboratory studiesc (mean ±SD)
Platelet count (1,000/mm3)245 ± 71218 ± 660.004
Total cholesterol (mg/dL)195 ± 42199 ± 440.48
HDL cholesterol (mg/dL)43 ± 1247 ± 110.01
LDL cholesterol (mg/dL)119 ± 36122 ± 380.65
Triglycerides (mg/dL)178 ± 131159 ± 790.08
HbA1c (%)6.0 ± 1.16.0 ± 1.00.70
Fasting serum glucose (mg/dL)103 ± 34107 ± 270.31
Fasting serum insulin (lU/mL)23 ± 1919 ± 130.07
HOMA-IR (mg/dL x lU/mL/405)6.0 ± 6.05.5 ± 4.40.34
Ferritin (ng/mL)235 ± 264326 ± 4540.13
APRI score0.5 ± 0.40.7 ± 0.60.04

Histologic Characteristics in Elderly Compared to Nonelderly Patients With NAFLD

Table 2 presents the comparison of the detailed histological features in the elderly and nonelderly patients with NAFLD. Compared to nonelderly patients with NAFLD, the elderly had a higher prevalence of NASH (72% versus 56%, P = 0.02) (Fig. 1), advanced fibrosis (44% versus 25%, P = 0.002) (Fig. 2) and azonal-distribution of steatosis (43% versus 27%, P = 0.01) (Table 2).

Table 2. Histological Features of Patients With NAFLD Comparing Elderly to Nonelderly Patients
 Age Group 
Histological FeatureNonelderly (N = 735) No. (%)Elderly (N = 61) No. (%)Pa
  1. a

    P values determined from Fisher's exact test or Cuzick nonparametric test for trend across ordered categories, except for NAS, a t test was used.

Steatosis:   
Grade  0.20
0-1 (0%-33%)309 (42.0%)31 (50.8%) 
2 (>33%-66%)252 (34.3%)18 (29.5%) 
3 (>66%)174 (23.7%)12 (19.7%) 
Location (predominant)  0.02
Zone 3 (central)307 (41.8%)15 (24.6%) 
Zone 1(periportal)11 (1.5%)0 (0.0%) 
Azonal195 (26.6%)26 (42.6%) 
Panacinar221 (30.1%)20 (32.8%) 
Microvesicular steatosis: present74 (10.1%)9 (14.8%)0.27
Fibrosis:   
Stage:   
None (0)200 (27.4%)5 (8.2%)0.001
Mild/moderate (zone 3), portal/periportal (1A, 1B, 1C)209 (28.3%)13 (21.3%) 
Zone 3 & periportal (2)141 (19.3%)16 (26.2%) 
Bridging (3)122 (16.7%)19 (31.0%) 
Cirrhosis (4)61 (8.3%)8 (13.1%) 
Advanced fibrosis:  0.002
None (0), mild (1), moderate (2)548 (75.0%)34 (55.7%) 
Bridging (3) or cirrhosis (4)183 (25.0%)27 (44.3%) 
Inflammation:   
Lobular inflammation (score) (no. foci per 200X field)  0.007
0 to <2 foci (0-1)392 (53.3%)22 (36.1%) 
2 to 4 foci (2)267 (36.3%)26 (42.6%) 
>4 foci (3)76 (10.3%)13 (21.3%) 
Microgranulomas: present603 (82.0%)52 (82.3%)0.60
Large lipogranulomas: present279 (38.0%)33 (54.1%)0.02
Portal Inflammation (score)  0.23
None (0)111 (15.1%)5 (8.2%) 
Mild (1)466 (63.4%)39 (63.9%) 
More than mild (2)158 (21.5%)17 (27.9%) 
Liver cell injury:   
Ballooning (score):  0.004
None (0)251 (34.2%)15 (24.6%) 
Few (1)199 (27.1%)9 (14.8%) 
Many (2)285 (38.8%)37 (60.7%) 
Acidophil bodies: many215 (29.3%)26 (42.6%)0.04
Pigmented macrophages: many644 (87.6%)54 (88.5%)1.00
Megamitochondria: many106 (14.4%)15 (24.6%)0.04
Other findings   
Mallory Denk bodies: many194 (26.4%)32 (52.5%)<0.001
Glycogenated nuclei: many392 (53.3%)29 (47.5%)0.42
NAFLD Activity Score (NAS)  0.23
0 - 4 384 (52.2%)27 (44.3%) 
5 - 8351 (47.8%)34 (55.7%) 
Mean ± SD4.38 ± 1.674.82 ± 1.690.05
Diagnostic classification:   
Steatohepatitis (diagnosis):  0.12
Not steatohepatitis (0)166 (22.6%)10 (16.4%) 
Possible/borderline:   
Zone 3 pattern (1A)148 (20.1%)7 (11.5%) 
Zone 1, periportal (1B)9 (1.2%)0 (0.0%) 
Definite steatohepatitis (2)412 (56.1%)44 (72.1%) 
Definite NASH (yes)412 (56.1%)44 (72.1%)0.02
Figure 1.

Prevalence of definite NASH between nonelderly and elderly patients with NAFLD. Compared to nonelderly (green dotted bar) patients with NAFLD, elderly patients (red bar) had a higher prevalence of NASH (72.1% versus 56.1%, P = 0.02).

Figure 2.

Distribution of fibrosis stage between nonelderly and elderly patients with NAFLD. The distribution of fibrosis between nonelderly (green dotted bar) versus elderly patients (red bar) for various stage of fibrosis was as follows: stage 0: 27.4% versus 8.2%, stage 1: 28.3% versus 21.3%, stage 2: 19.3% versus 26.2%, stage 3: 16.7% versus 31%, and stage 4: 8.3% versus 13.1%.

Furthermore, elderly patients had other features consistent with progressive liver disease, including a higher degree of lobular inflammation and a higher prevalence of acidophil bodies, megamitochondria, Mallory-Denk bodies, as well as more prominent ballooning (Table 2). As expected, elderly patients had a higher prevalence of lipogranulomas.

Histological Comparison Between Elderly and Nonelderly Patients With Definite NASH

In order to examine whether the advanced histologic features in elderly patients with NAFLD were due to the increased prevalence of NASH or whether these features were seen across the spectrum of NAFLD irrespective of presence or absence of NASH, we compared the detailed liver histologic features between elderly versus nonelderly patients who had biopsy-proven NASH. There were 44 patients with biopsy-proven NASH in the elderly patients group and 412 patients with biopsy-proven NASH in the nonelderly patients group (Table 3). Compared to nonelderly patients with NASH, elderly patients with NASH had higher rates of advanced fibrosis (52% versus 35%, P = 0.03), as well as other features suggestive of severe liver disease including ballooning degeneration, acidophil bodies, megamitochondria, and Mallory-Denk bodies (P ≤ 0.05 for each) (Table 3). In contrast, compared to nonelderly patients with NASH, elderly patients had lesser degrees of steatosis (48% versus 67% >33% steatosis, P = 0.01).

Table 3. Histological Features of Patients With Definite NASH Comparing Elderly to Nonelderly Patients
 Age Group 
Histological FeatureaNonelderly (N = 412) No. (%)Elderly (N = 44) No. (%)Pb
  1. a

    Determination of histological features from centrally reviewed biopsies using the NASH CRN Scoring System.9

  2. b

    P values determined from Fisher's exact test or Cuzick nonparametric test for trend across ordered categories, except for NAS, a t test was used.

  3. c

    One nonelderly patient and zero elderly patients had no ballooning.

Steatosis:   
Grade  0.01
0-1 (0%-33%)136 (33.0%)23 (52.3%) 
2-3 (>33%)276 (67.0%)21 (47.7%) 
Location (predominant)  0.07
Zone 3 (central)152 (36.9%)9 (20.5%) 
Zone 1(periportal)2 (0.5%)0 (0.0%) 
Azonal119 (28.9%)20 (45.5%) 
Panacinar139 (33.7%)15 (34.1%) 
Microvesicular steatosis  0.82
Not present353 (85.7%)37 (84.1%) 
Present59 (14.3%)7 (15.9%) 
Fibrosis:   
Stage:  0.03
None (0)32 (7.8%)1 (2.3%) 
Mild/moderate (zone 3), portal/periportal (1A, 1B, 1C)133 (32.4%)7 (15.9%) 
Zone 3 & periportal (2)101 (24.6%)13 (29.6%) 
Bridging (3)107 (26.1%)15 (34.1%) 
Cirrhosis (4)37 (9.0%)8 (18.2%) 
Advanced fibrosis:  0.03
None (0), mild (1), moderate (2)266 (64.9%)21 (47.7%) 
Bridging (3) or cirrhosis (4)144 (35.1%)23 (52.3%) 
Inflammation:   
Lobular inflammation (score) (no. foci per 200X field)  0.08
0 to <2 foci (0-1)162 (39.3%)13 (29.6%) 
2 to 4 foci (2)183 (44.4%)18 (40.9%) 
>4 foci (3)67 (16.3%)13 (29.6%) 
Microgranulomas: present357 (86.7%)40 (90.9%)0.64
Large lipogranulomas: present172 (41.8%)29 (65.9%)0.002
Portal Inflammation (score)  0.69
None (0)43 (10.4%)3 (6.8%) 
Mild (1)260 (63.1%)27 (61.4%) 
More than mild (2)109 (26.5%)14 (31.8%) 
Liver cell injury:   
Ballooning (score)c: many (2)276 (67.0%)37 (84.1%)0.03
Acidophil bodies: many160 (38.8%)25 (56.8%)0.02
Pigmented macrophages: many372 (90.3%)41 (93.2%)0.79
Megamitochondria: many76 (18.5%)14 (31.8%)0.05
Other findings:   
Mallory Denk bodies: present186 (45.2%)32 (72.7%)<0.001
Glycogenated nuclei: many171 (41.5%)23 (52.3%)0.20
NAFLD Activity Score (NAS)  0.72
0 – 4111 (26.9%)13 (29.5%) 
5 – 8301 (73.1%)31 (70.5%) 
Mean ± SD5.36 ± 1.255.43 ± 1.370.73

Characteristics of Elderly Patients With Definite NASH

We then investigated the characteristics of the presence of NASH in elderly patients by comparing how it differs from those without NASH in this age group (Supporting Table 1). Elderly patients with NASH had significantly higher average values for AST (70 ± 48 versus 38 ± 12 U/L, P < 0.001), ALT (75 ± 49 versus 49 ± 21 U/L, P = 0.006), and GGT (88 ± 82 versus 49 ± 44 U/L, P = 0.02). In addition, the average platelet count was lower (204 ± 59 versus 254 ± 71 ×1,000/mm3, P = 0.02) (Supporting Table 1). The mean APRI score was significantly higher in elderly patients with NASH compared to elderly patients without NASH (0.8 ± 0.7 versus 0.4 ± 0.3, P < 0.001). There was no significant difference in steatosis and degree of lobular inflammation between those with and without NASH. However, as would be expected, NASH patients were more likely to have ballooning degeneration. In addition, Mallory-Denk bodies were present in 72% of NASH patients, while these were absent in those who did not have NASH (P < 0.001) (Supporting Table 1). The NAFLD activity score (NAS) as indicated by the percentage of patients with NAS ≥5 was higher in elderly patients with NASH compared to elderly patients without NASH (70% versus 18%, P < 0.001). Elderly patients with NASH were more likely to have advanced fibrosis compared to elderly patients who did not have NASH (52% versus 24%, P = 0.05) (Supporting Table 1).

Independent predictors of NASH among elderly patients determined from multivariable-adjusted logistic regression analyses were: younger age among this cohort with age ≥65 (OR = 0.65, 95% CI: 0.46-0.91, P = 0.01); higher AST value (OR = 1.12, 95% CI: 1.03-1.22, P = 0.007), and lower platelet count (OR = 0.98, 95% CI: 0.96-1.00, P = 0.02) (Table 4).

Table 4. Multiple Logistic Regression Analysis of Demographic, Clinical, and Histological Characteristics in Elderly Patients With NAFLD: Characteristics Independently Associated With Definite NASH and Advanced Fibrosis
 UnadjustedbAdjustedc
CharacteristicsOR (95%CI)POR (95%CI)P
  1. a

    Total number = 59; no. with definite NASH = 42, not definite NASH = 17; no. with advanced fibrosis = 25, not advanced = 34.

  2. b

    Unadjusted odds ratios and P values determined from logistic regression of either definite NASH or advanced fibrosis on each characteristic.

  3. c

    The adjusted model determined from backward stepwise regression of either definite NASH or advanced fibrosis on the candidate set forcing age, gender, and race into the model. If the characteristic was not in the final model, n/s was used to indicate to indicate this. Smoking and CVD were excluded from the candidate set for definite NASH due to noncollinearity (noted with —); ethnicity was excluded from the candidate set for the advanced fibrosis model due to no non-Hispanic males with fibrosis.

Definite NASHa   
Demographics    
Female (vs. male)1.62 (0.45, 5.80)0.461.15 (0.18, 7.50)0.79
Age (years)0.86 (0.72,1.04)0.110.65 (0.46, 0.91)0.01
White (vs. non-white)0.69 (0.16, 2.88)0.610.30 (0.04, 2.27)0.25
Hispanic (vs. non-Hispanic)2.53 (0.28, 22.71)0.41n/s 
Ever smoked regular (vs. not)0.89 (0.29, 2.73)0.84 
Clinical    
Cardiovascular disease (yes vs. no)1.04 (0.24, 4.48)0.91 
Type 2 diabetes (yes vs. no)1.75 (0.43, 7.19)0.44n/s 
BMI (kg/m2)1.04 (0.93, 1.18)0.47n/s 
Laboratory markers    
AST (U/L)1.08 (1.02, 1.13)0.011.12 (1.03, 1.22)0.007
ALT (U/L)1.02 (1.00, 1.05)0.06n/s 
GGT (U/L)1.01 (1.00, 1.03)0.10n/s 
Platelets (1000/mm3)0.99 (0.98, 1.00)0.020.98 (0.96, 1.00)0.02
Hypertriglyceridemia (yes vs. no)1.09 (0.35, 3.38)0.89n/s 
Low HDL (yes vs. no)1.41 (0.46, 4.37)0.55n/s 
High blood pressure (yes vs. no)0.60 (0.18, 2.01)0.41n/s 
HOMA-IR (mg/dL x lU/mL/405)1.03 (0.89, 1.18)0.70n/s 
Ferritin (ng/mL)1.00 (1.00, 1.00)0.48n/s 
Advanced fibrosis (yes vs. no)3.56 (1.00, 12.64)0.05n/s 
     
Advanced Fibrosisa
Demographics    
Female (vs. males)1.08 (0.32, 3.59)0.900.44 (0.06, 2.79)0.38
Age (years)1.13 (0.5, 1.36)0.171.34 (0.97, 1.87)0.08
Ever smoked regular (vs. not)1.41 (0.51, 3.88)0.51n/s 
White (vs. non-white)0.82 (0.24, 2.83)0.762.92 (0.40, 21.19)0.29
Clinical    
Cardiovascular disease (yes vs. not)4.35 (1.03, 18.4)0.05n/s 
Type 2 diabetes (yes vs. no)1.14 (0.35, 3.66)0.83n/s 
BMI (kg/m2)1.18 (1.03, 135)0.02n/s 
Laboratory markers    
AST (U/L)1.00 (0.99, 1.01)0.641.08 (1.02, 1.14)0.007
ALT (U/L)0.99 (0.98, 1.00)0.120.91 (0.86, 0.97)0.002
GGT (U/L)1.01 (1.00, 1.02)0.04n/s 
Platelets (1,000/mm3)0.98 (0.97, 1.00)0.01n/s 
Hypertriglyceridemia (yes vs. no)1.50 (0.54, 4.18)0.44n/s 
Low HDL (yes vs. no)3.21 (1.08, 9.59)0.048.35 (1.50, 46.5)0.02
HOMA-IR (mg/dL x lU/mL/405)1.15 (1.00, 1.32)0.05n/s 
High blood pressure (yes vs. no)2.54 (0.85, 7.58)0.10n/s 
Ferritin (ng/mL)1.00 (1.00, 1.00)0.58n/s 
Definite NASH (yes vs. no)3.56 (1.00, 12.64)0.0510.37 (1.22, 87.94)0.03

Characteristics of Elderly Patients With Advanced Fibrosis

 Characteristics of elderly patients with advanced fibrosis compared to those with stage 0-2 fibrosis are shown in Supporting Table 2. Patients with advanced fibrosis were more likely to have metabolic syndrome, higher average BMI, and increased fasting serum insulin, HOMA-IR, INR, and AST/ALT ratio. In addition, patients with advanced fibrosis had lower mean platelet count, total cholesterol, and LDL cholesterol levels. Patients with advanced fibrosis had significantly less steatosis but more portal inflammation, ballooning, Mallory-Denk bodies, and higher prevalence of NASH.

In multivariable-logistic regression analysis, the independent predictors of advanced fibrosis included a higher AST level (OR = 1.08, 95% CI: 1.02-1.14, P = 0.007), a lower ALT value (OR = 0.91, 95% CI: 0.86-0.97, P = 0.002), and increased odds of having a low HDL cholesterol (OR = 8.35, 95% CI: 1.50-46.50, P = 0.02) (Table 4).

Discussion

This is a secondary analysis of prospectively collected clinical, biochemical, and histologic data of a large number (n = 796) of adult patients with biopsy-proven NAFLD that allowed the detailed characterization of NAFLD in elderly versus nonelderly patients. The main findings of this study are that elderly patients with NAFLD have significantly higher rates of NASH and advanced fibrosis than nonelderly patients with NAFLD. Furthermore, among those NAFLD patients who already have NASH on liver biopsy, elderly patients are more likely to have advanced fibrosis as well as other features suggestive of severe liver disease including the presence of ballooning degeneration, acidophil bodies, and Mallory-Denk bodies compared to nonelderly patients with NASH. These findings suggest that the severity of liver histology is shifted to a more aggressive phenotype in elderly patients across the spectrum of NAFLD, including those with or without NASH. This is further supported by the surprising finding that the prevalence of advanced fibrosis was 24% in elderly patients who did not have evidence of NASH on biopsy, and the advanced fibrosis prevalence rate increased to 52% in elderly patients who had evidence of NASH on liver histology (P = 0.05). Further studies with a larger sample size are needed to confirm this phenomenon of presence of advanced fibrosis in non-NASH elderly patients with NAFLD.

A higher AST, a lower platelet count, and a lower (not higher) ALT and a low HDL cholesterol are independent predictors of NASH and advanced fibrosis; these commonly available factors can be used to aid clinical decision-making regarding when to consider a liver biopsy in elderly patients with NAFLD. These data are in agreement with previously published studies by the NASH CRN and other cohorts but highlight that ALT may be lower in elderly patients and low HDL was the most significant predictor of advanced disease in elderly patients.[26, 36, 40]

NAFLD is reported to be more prevalent in men than in women, and its prevalence may increase with age.[16, 37-39] Other studies have also suggested that the prevalence of NAFLD may be influenced by menopause, and it may be more prevalent in women after menopause.[39-41] Our results are consistent with previous studies that elderly patients with NAFLD are more likely to be women. Frith et al.[17] have shown that fibrosis and cirrhosis are higher in elderly patients with NAFLD. Using this well-characterized clinicopathologic cohort, we confirm their findings by demonstrating that elderly patients had laboratory findings that are suggestive of more advanced liver disease including, higher AST/ALT ratio, higher APRI score, and lower platelet count.[42, 43] Furthermore, we showed that elderly patients have more definite NASH, advanced fibrosis, and cirrhosis compared to nonelderly patients. Given that this a cross-sectional study, one can argue that the higher prevalence of advanced liver disease found in elderly patients can be due to the fact that they have more metabolic risk factors.[44] However, in our cohort the elderly patients did not have more risk factors such as diabetes or insulin resistance.[42] Indeed, elderly patients had lower BMI and waist circumference.

The novelty of the study is the detailed histological description of NAFLD and NASH by a panel of expert pathologists, and the availability of a clinical, demographic, and biochemical dataset that allowed the comparison between elderly and nonelderly patients with biopsy-proven NAFLD.

Our findings in the context of the previous studies may suggest that early in the natural history of NAFLD, steatosis starts in zone 3 and with progressive aging (as well as with disease progression because they are collinear with each other), steatosis spreads to other zones and the pattern of steatosis distribution becomes pan-acinar with more cellular injury. Then, perhaps due to progressive fibrosis and regeneration/remodeling, the pattern is further modified, and steatosis distribution becomes azonal as patients develop more advanced fibrosis. In addition, steatosis paradoxically decreases in elderly patients despite having more severe disease. Frith et al. and Permutt et al. have previously shown that steatosis grade on histology and liver fat content estimated by magnetic resonance imaging (MRI), respectively, are significantly lower in patients with cirrhosis compared to those with less degree of fibrosis.[10, 45, 46] One plausible explanation of this paradoxical reduction in steatosis may be related to reduced ability of the stiffened fibrotic liver to store and accumulate fat in the hepatocytes. The collagen deposition in the liver tissue replaces fat in the liver and restricts further accumulation of fat in hepatocytes. Prospective studies are needed to confirm this hypothesis. Moreover, the mechanisms underlying these alterations in steatosis distribution by age need to be studied further.

Strengths and Limitations

 The strengths of the study include the prospective design of the NASH CRN studies and availability of well-characterized liver histology data. The study utilized the well-accepted and previously validated NASH CRN Histologic Scoring System.[9, 47] Liver biopsy assessment was performed by a panel of expert liver pathologists during central review by consensus of the members of the pathology committee. This study included comparisons between elderly and nonelderly patients with NAFLD as well as NASH. Although our cohort is large, the number of elderly patients was relatively small but provided sufficient power to detect clinically significant differences. This also illustrates the challenges in the recruitment and enrollment of elderly patients in cohort studies, especially with the requirement of liver biopsy. It took several years for this large multicenter study to recruit patients with biopsy-proven NAFLD. Age correlates with duration of disease and the association with more advanced disease may not be attributable to age alone, but also to duration of disease. However, it is not possible to control for duration of disease. Despite this limitation, it is elderly patients who have higher rates of NASH and advanced fibrosis whether it is due to aging or due to duration of disease. Longitudinal studies with serial liver biopsies will be required to investigate the natural progression of the disease in younger and older adults and to examine the evolution of fat distribution.

In conclusion, elderly patients with NAFLD are more likely to have features of advanced fibrosis as well as aggressive NASH. NAFLD cannot be considered a benign disease in elderly patients. Elderly patients are at increased risk of NASH and advanced fibrosis but are underrepresented in cohort studies. Advanced fibrosis can also occur in elderly patients with NAFLD without specific histologic features of NASH. This observation may reflect the previous observation that key features of NASH such as steatosis, ballooning, and Mallory-Denk bodies may be lost as the disease progresses towards cirrhosis. Thus, liver biopsy evaluation can be helpful in this age group to guide the implementation of treatment recommendations such as weight reduction and increased physical activity. Due to the aging of American society, further research is needed in NAFLD in elderly patients. It is important to identify elderly NAFLD patients who are at risk of progressive liver disease, especially because newer treatment modalities are emerging.[23, 48-54] Furthermore, clinical trials should be conducted to test the efficacy and safety of the available treatment modalities, such as vitamin E, in this subpopulation and every effort should be made to avoid excluding patients older than 65 years in future trials and cohort studies.

Acknowledgment:

The study was sponsored by the NIDDK, NIH. As per the policy of the network, the article was reviewed by the NIDDK prior to publication. The authors take full responsibility for the data analyses and credibility of findings.

Author contributions:

All authors approved the final submission. Mazen Noureddin: Drafting of the article, interpretation of data, critical revision of the article. Katherine P. Yates: Analysis and interpretation of data, statistical analysis, critical revision of the article. Ivana A. Vaughn: Analysis and interpretation of data, statistical analysis, critical revision of the article. Brent A. Neuschwander-Tetri, Arun J. Sanyal, Arthur McCullough, Raphael Merriman, Bilal Hameed, Edward Doo, David E. Kleiner, Cynthia Behling: Critical revision of the article. Rohit Loomba: Study concept and design, analysis and interpretation of data, drafting of the article, critical revision of the article, obtained funding, study supervision.

Appendix

Members of the Nonalcoholic Steatohepatitis Clinical Research Network: Adult Clinical Centers: Case Western Reserve University Clinical Centers: MetroHealth Medical Center, Cleveland, OH: Arthur J. McCullough, MD; Patricia Brandt; Diane Bringman, RN (2004-2008); Srinivasan Dasarathy, MD; Jaividhya Dasarathy, MD; Carol Hawkins, RN; Yao-Chang Liu, MD (2004-2009); Nicholette Rogers, PhD, PAC (2004-2008); Cleveland Clinic Foundation, Cleveland, OH: Arthur J. McCullough, MD; Srinivasan Dasarathy, MD; Mangesh Pagadala, MD; Ruth Sargent, LPN; Lisa Yerian, MD; Claudia Zein, MD; California Pacific Medical Center, San Francisco, CA: Raphael Merriman, MD; Anthony Nguyen; Columbia University, New York, NY: Joel E. Lavine, MD, PhD; Duke University Medical Center, Durham, NC: Manal F. Abdelmalek, MD; Stephanie Buie; AnnaMae Diehl, MD; Marcia Gottfried, MD (2004-2008); Cynthia Guy, MD; Meryt Hanna (2010);Christopher Kigongo; Paul Killenberg, MD (2004-2008); Samantha Kwan, MS (2006-2009); Yi-Ping Pan; Dawn Piercy, FNP; Melissa Smith (2007-2010); Savita Srivastava, MD; Indiana University School of Medicine, Indianapolis, IN: Naga Chalasani, MD; Oscar W. Cummings, MD; Marwan Ghabril, MD; Ann Klipsch, RN; Linda Ragozzino, RN; Girish Subbarao, MD; Sweta Tandra, MD; Raj Vuppalanchi, MD; Saint Louis University, St Louis, MO: Brent A. Neuschwander-Tetri, MD; Joan Siegner, RN; Susan Stewart, RN; Debra King, RN; Judy Thompson, RN; University of California San Diego, San Diego, CA: Cynthia Behling, MD, PhD; Jennifer Collins; Janis Durelle; Tarek Hassanein, MD (2004-2009); Joel E. Lavine, MD, PhD (2002-2010); Rohit Loomba, MD; Anya Morgan (2009-2010); Thu Nguyen; Heather Patton, MD; Claude Sirlin, MD; University of California San Francisco, San Francisco, CA: Bradley Aouizerat, PhD; Kiran Bambha, MD (2006-2010); Marissa Bass; Nathan M. Bass, MD, PhD; Linda D. Ferrell, MD; Bo Gu (2009-2010); Bilal Hameed, MD; Mark Pabst; Monique Rosenthal (2005-2010); Tessa Steel (2006-2008); University of Washington Medical Center, Seattle, WA: Matthew Yeh, MD, PhD; Virginia Commonwealth University, Richmond, VA: Sherry Boyett, RN, BSN; Melissa J. Contos, MD; Michael Fuchs, MD; Amy Jones; Velimir A.C. Luketic, MD; Puneet Puri, MD; BimalijitSandhu, MD (2007-2009); Arun J. Sanyal, MD; Carol Sargeant, RN, BSN, MPH; KimberlyNoble; Melanie White, RN, BSN (2006-2009); Virginia Mason Medical Center, Seattle, WA: Sarah Ackermann; Kris V. Kowdley, MD; Jane Park; Tracey Pierce; Jody Mooney, MS; James Nelson, PhD; Cheryl Shaw, MPH; Alice Stead; Chia Wang, MD; Washington University, St. Louis, MO: Elizabeth M. Brunt, MD. Resource Centers: National Cancer Institute, Bethesda, MD: David E. Kleiner, MD, PhD; National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD: Edward C. Doo, MD; Jay H. Hoofnagle, MD; Patricia R. Robuck, PhD, MPH (2002-2011); Averell Sherker, MD; Johns Hopkins University, Bloomberg School of Public Health (Data Coordinating Center), Baltimore, MD: Patricia Belt, BS; Frederick L. Brancati, MD, MHS (2003-2009); Jeanne M. Clark, MD, MPH; Erin Corless, MHS; Ryan Colvin, MPH (2004-2010); Michele Donithan, MHS; Mika Green, MA (2007-2012); Rosemary Hollick (2003-2005); Milana Isaacson, BS; Wana K. Jin, BS (2008-2011); Alison Lydecker, MPH (2006-2008); Pamela Mann, MPH (2008-2009); Kevin P. May, MS; Laura Miriel, BS; Alice Sternberg, ScM; James Tonascia, PhD; Aynur Ünalp-Arida, MD, PhD; Mark Van Natta, MHS; Ivana Vaughn, MPH; Laura Wilson, ScM; Katherine Yates, ScM.

Ancillary