- Email: [email protected]
Is cryptogenic cirrhosis different from NASH cirrhosis?
Accepted Manuscript Is Cryptogenic Cirrhosis Different from NASH Cirrhosis? Paul J. Thuluvath, Sergey Kantsevoy, Avesh J. Thuluvath, Yulia Savva PII: DOI: Reference:
S0168-8278(17)32441-8 https://doi.org/10.1016/j.jhep.2017.11.018 JHEPAT 6760
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
10 August 2017 21 October 2017 2 November 2017
Please cite this article as: Thuluvath, P.J., Kantsevoy, S., Thuluvath, A.J., Savva, Y., Is Cryptogenic Cirrhosis Different from NASH Cirrhosis?, Journal of Hepatology (2017), doi: https://doi.org/10.1016/j.jhep.2017.11.018
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
JHEPAT-D-17-01587 (clean copy)
Is Cryptogenic Cirrhosis Different from NASH Cirrhosis? Paul J. Thuluvath, MD, FRCP, FAASLD (1,2), Sergey Kantsevoy, MD, PhD (1,2), Avesh J Thuluvath (3), Yulia Savva, PhD (1) (1) Institute of Digestive Health & Liver Diseases, Mercy Medical Center, Baltimore and (2) Departments of Surgery & Medicine, University of Maryland School of Medicine, Baltimore, Maryland, (3) Johns Hopkins University School of Medicine, Baltimore Words: 3884; Tables: 6; Figure: 3 Key Words: UNOS, autoimmune hepatitis, alcoholic cirrhosis, age and gender matched Address for correspondence: Paul J. Thuluvath, MD, FAASLD, FRCP, Professor of Medicine & Surgery, Institute of Digestive Health & Liver Diseases, Mercy Medical Center, Baltimore, MD 21202; Tel: 410 332 9308; Email: [email protected] Abbreviations: NASH: Non-alcoholic steatohepatitis, CC: Cryptogenic cirrhosis, AC: alcoholic cirrhosis, AIH: autoimmune hepatitis, UNOS: United Network for Organ Sharing, HCC: hepatocellular carcinoma Financial support: None; Conflicts of interest: None Author contributions: PJT, SK, AJT and YS made substantial contributions to the conception and design; or the acquisition, analysis, or interpretation of the data; PJT contributed the drafting of the article or critical revision for important intellectual content, and all authors approved the final version, and agree to be accountable for all aspects of the work
ABSTRACT Background & Aims: We hypothesized that patients currently diagnosed as CC have truly ‘cryptogenic’ liver disease, and unlikely to have evolved from NASH. Methods: To investigate this, we compared the clinical characteristics of adults with CC (n=7,999) with NASH (n=11,302), alcohol (AC, n=21,714) and autoimmune hepatitis (AIH, n=3,447) using the UNOS database from 2002-16. We performed an age, gender and year of listing matched comparison of CC and NASH (n= 7,201 in each group), and also stratified by the presence of obesity or diabetes mellitus (DM). Results: From 2002 to 2016, patients listed with a diagnosis of NASH increased from about 1% to 16% while CC decreased from 8% to 4%. Logistic regression model using the entire UNOS data (n=138,021) suggested that the strongest predictors of NASH were type 2 DM, obesity, age > 60 years, female gender and White race. Type 2 DM was more common in NASH (53%) compared to CC (29%), AC (16%) and AIH (16%), and obesity was more common in NASH (65.3%) compared to other three groups (33-42%). There were more Whites (82.3%) in NASH group with a lower prevalence of black, Hispanic and Asian race as compared to the other three groups. HCC was more commonly seen in NASH (19% versus 913% in the other groups) and this is also not influenced by obesity and type 2 DM. The differences between CC and NASH remained unchanged even when two groups were matched for age, gender and year of listing, or when stratified by the presence or absence of obesity or type 2DM. Conclusions: Based on risk perspectives, CC should not be equated with the term ‘NASH cirrhosis’.
Cryptogenic cirrhosis (CC) is a diagnosis of exclusion when there is no other known identifiable etiology (1). About two decades ago, when non-alcoholic steatohepatitis (NASH) was emerging as a common cause of chronic liver disease in the Western world, we and others had suggested that many patients with cryptogenic cirrhosis may have evolved from NASH (2-4). These observations were made based on patient characteristics and the higher prevalence of risk factors including type 2 diabetes mellitus (DM), obesity or metabolic syndrome in those with CC since histology was often not helpful once the patient with NASH had progressed to cirrhosis (5,6). Some investigators, however, had indicated that burnt-out autoimmune hepatitis (AIH) and occult alcoholism may also be considered as ‘potential causes’ of CC (7-10). Other causes such as previous hepatitis B infection, iron overload, alpha-1 antitrypsin deficiency, and thrombotic/vascular disorders are unlikely to be major contributors of cryptogenic cirrhosis (710). Over the past two decades, physicians have become more confident in making a firm diagnosis of NASH cirrhosis based on history, risk factors and the absence of other etiological factors. The listing diagnosis in the United Network for Organ Sharing (UNOS) database suggests that the number of patients with NASH cirrhosis has increased since 2000 while those with CC has decreased (11-13). Despite the increased awareness of NASH as an important cause of cirrhosis, a significant number of patients (400-600 per year) are currently being listed for liver transplantation with the UNOS with a diagnosis of CC (11-13). In the European LT registry, about 4% of patients are listed as CC (14). It is more than likely that those patients currently listed as CC have truly ‘cryptogenic’ chronic liver disease and less likely to have evolved from NASH. Nevertheless, there is a trend among many physicians to use CC and NASH cirrhosis synonymously, and some investigators have combined them as a single entity, when outcomes
are reported, especially among the sub-group of patients with CC with obesity (defined as BMI >30) (11, 15, 16). The objective of our study is to characterize the patients currently listed for liver transplantation with a diagnosis of CC cirrhosis and compare them with cirrhosis from NASH, alcohol and AIH. PATIENTS AND METHODS We analyzed the UNOS database from 2002 to November 2016 of all adult (> 18 years) patients. We selected 2002 since the UNOS introduced MELD scores for organ allocation in February 2002 and we assumed that the data collection may have improved since then, and additionally, examination of the annual data variability by diagnosis code showed that the data collection was perhaps done more systematically since 2002. Moreover, the diagnosis of NASH was utilized in the UNOS database since 2001, and it was also around that time when NASH cirrhosis was diagnosed more consistently by liver transplant centers based on publications that suggested that many patients with CC may have progressed from NASH (2,3). Since it has been suggested that some patients with CC may have evolved from occult alcoholism or AIH, we compared CC with alcoholic cirrhosis (AC) and AIH cirrhosis in addition to NASH cirrhosis. The groups were defined based on their primary diagnosis at the time of listing with the UNOS. The only exception was those with hepatocellular carcinoma (HCC); in those patients, we included secondary diagnosis for this study. For example, an individual was considered as NASH if the primary diagnosis at listing was NASH or if the primary diagnosis was hepatocellular carcinoma (HCC) with a secondary diagnosis of NASH cirrhosis. The same rule was applied to the other three groups. For a minority of patients, the diagnosis had changed from the time of listing to the time of transplant, but we did not account for this change while
defining our groups. Since the number of patients with a change in primary diagnosis was a very small, we felt that it would have no impact on the study findings. Less than 10% of patients were listed for a simultaneous liver and kidney transplant including 1,272 (11%) for NASH, 805 (10%) for CC, 1,904 (8.8%) for AC and 190 (5.5%) for AIH. We included those received a live donor [200 (1.8%) NASH, 175 (2.2%) CC, 259 (1.25) AC, and 87 (2.5%) AIH]. Although several patients had multiple listings or were relisted, we included only one listed data point per individual. Our final study group included 11,302 NASH cirrhosis, 7,999 CC, 21,714 AC and 3,447 AIH cirrhosis. We collected following potential risk factors and confounders at the time of listing including race, age, body mass index (BMI), obesity (defined as BMI>=30), initial waiting list Model for End-stage Liver Disease (MELD) score, gender and DM. The data for DM were unknown for 5.4% of patients, and these patients were excluded from the analysis of the association between the outcome and DM. We also collected data on the prevalence of complications of cirrhosis including the prevalence of HCC, and history of spontaneous bacterial peritonitis, ascites, encephalopathy, portal vein thrombosis or dialysis. STATISTICAL ANALYSIS The descriptive statistics was presented for all four groups. The differences between the groups were tested by using logistic regressions for categorical variables, and two-sample t-test for continuous variables. Although the MELD score was not normally-distributed, considering a large sample size, a t-test was considered appropriate for hypothesis testing. To identify the patient characteristics with the strongest association with NASH, we performed a multivariate logistic regression analysis using the entire UNOS data (N=138,021).
We also compared CC and NASH cirrhosis by analyzing a 1:1 matched (matched for age, gender and year of listing) data to reduce the confounding effects of age, gender and year of listing. Each case (CC) was matched to a control (NASH) within 5-years of age, with the same gender and the year of listing (+ 2 years). Matched analysis included 7,201 CC and 7,201 NASH. Additionally, we also compared CC and NASH (unmatched) data stratified separately by presence or absence of obesity (defined by World Health Organization scale with BMI > 30) and type 2 DM to examine if there are similarities between the two groups for these two important risk factors for NASH cirrhosis.
RESULTS During the study period, the number (Figure 1A) and proportion (Figure 1B) of patients with NASH cirrhosis increased while those with CC decreased. From 2002 to 2016, patients listed with a diagnosis of NASH increased from about 1% to 16% while cryptogenic cirrhosis decreased from 8% to 4%. Yet, a significant number of patients continued to have a listing diagnosis of cryptogenic cirrhosis (572 in 2002, 536 in 2010, 455 in 2015). Table 1A shows demographic characteristics of patients with CC, NASH, AC and AIH. Those with NASH cirrhosis were older (mean age 59 years) compared to the other groups, and the mean age increased over the study period as shown in figure 2. Until 2007, the frequency of listed individuals in the age of 50-59 and 60-69 was similar, but proportion of older (60-69 age group) patients continued to increase after 2007, and this rate of increase was significantly faster for NASH cirrhosis compared to other groups perhaps reflecting the natural history of NASH cirrhosis (figure 2). The MELD score at listing was lowest (mean 17.2) for NASH cirrhosis,
followed by CC (18.4), AIH (19.4) and AC (19.8) (Table 1). There was an apparent linear trend of an increase in listing MELD score for all four groups during the study period (Figure 3). Table 1 shows that there are significant differences in clinical characteristics among the groups. There was a higher prevalence of type 2 DM in NASH cirrhosis (53%) compared to CC (29%), AC (16%) and AIH (16%). Similarly, obesity was more common in NASH cirrhosis (65.3%) compared to other three groups (33-42%). There were more Whites (82.3%) in NASH group with a lower prevalence of black, Hispanic and Asian race as compared to the other three groups. The highest prevalence of African-American and Hispanic population was for autoimmune hepatitis (about 35%) followed by cryptogenic cirrhosis (24%). There were gender differences, as expected, between the groups with predominantly women in AIH (74%) and men (75%) in AC. Since the listing of patients with cryptogenic cirrhosis has stabilized in recent years, we analyzed the characteristics of patients listed during 2011-2016 Table 1B). As shown in table 1B, the differences remained unchanged. The complications were also different between the groups (Table 1). HCC was more common in NASH (19%) followed by AC (14%), CC (13%) and AIH (10%). Other complications of cirrhosis were higher in AC including the prevalence of spontaneous bacterial peritonitis, dialysis, portal vein thrombosis and encephalopathy (stage 2 and 3) and encephalopathy. These complications were lowest for the autoimmune group. When logistic regression modeling was done on the entire cohort of patients listed with UNOS (using logistic regression model using the entire UNOS data; (n =138,021), we found that the strongest predictors of having a NASH diagnosis were type 2 DM [Odds Ratio (OR) 4.7 after
controlling for other factors], obesity, age 60 and above, female gender and White race (Table 2). This association between type 2 DM and NASH was even stronger for non-obese population (OR=5.74) compared to the obese population (OR=4.4). Table 3 shows the characteristics of CC and NASH matched by the age, gender and year of listing. The differences between the two groups remained unchanged even when two groups were matched for age and gender. We further analyzed the groups after stratifying by obesity (Table 4) and type 2 DM (Table 5). The prevalence of type 2 DM was higher in non-obese NASH (50.4% vs. 24.6%) and obese NASH (54.1% vs. 34%) when compared to non-obese and obese CC respectively. When groups were stratified by presence or absence of type 2 DM, obesity was higher for NASH group (66.8% vs. 49.2% for those with DM, and 63.4% vs. 38% for those without DM) when compared to CC. There were also differences in HCC in NASH and CC with a higher prevalence in NASH and this was also not influenced by other confounders such as obesity and type 2 DM. DISCUSSION Our study shows that there are significant differences in the clinical characteristics of patients with CC when compared to NASH, AC or AIH. Using the large data from UNOS, we confirmed the well-known association between NASH cirrhosis with type 2 DM, obesity, older age, female gender and White ethnicity. When these characteristics were compared among four groups of patients with CC, NASH, AC and AIH, these groups had significantly different clinical characteristics. The differences remained unchanged in those who were listed between 2011 and 2016. We further analyzed these characteristics in CC and NASH after stratifying the patients with and without DM or obesity. Similarly, when CC patients were matched with NASH by age, gender and year of listing, differences remained unchanged. These data clearly indicate that CC
and NASH are different based on their risk perspectives and we cannot assume that those who are currently diagnosed as CC have evolved from NASH, AC or AIH. In addition, we cannot assume that obese or diabetic patients with CC have NASH cirrhosis. When it was suggested two decades ago that most patients with CC may have evolved from NASH, our understanding of NASH cirrhosis was just evolving (2,3). Currently, most transplant physicians are confident in making a firm diagnosis of NASH cirrhosis based on either a history long standing fatty liver disease or NASH compatible demographic characteristics and risk factors. This may explain the very different clinical characteristics between NASH and CC who are currently listed with UNOS. Our study, based on the differences in demographic characteristics, also suggests that it is quite unlikely that most patients with CC may have evolved from occult alcoholic liver disease or undiagnosed/burnt out AIH. We speculate that that physicians list patients as ‘cryptogenic cirrhosis’ only when they are unsure of the etiology, and our study appears to corroborate these assumptions. Our findings have many implications. Many physicians use CC and NASH in a synonymous fashion, and many outcome studies have combined a significant proportion of CC or unknown cirrhosis with NASH if they have obesity (11, 15,16). This may not be justified based on the findings of our study, and we suggest that future studies should treat them as distinct entities as they may have a different natural history, complications and post-liver transplant outcomes. In our experience steatosis is not uncommon after liver transplantation in obese subjects or those who develop obesity, and anecdotal experience of steatosis in those had liver transplantation for CC could not be a sound justification for assuming that CC and NASH are synonymous (17-21). The findings of our study also suggest that we should not stop exploring for other hitherto unrecognized causes of cirrhosis.
In our study, the highest prevalence of HCC was in NASH, and a combination of cirrhosis, type 2 DM and obesity may explain this finding (16, 22-31). The prevalence of HCC was 49% higher in NASH compared to CC, and patients with CC had lower prevalence of HCC when compared to AC, but the lowest prevalence was in AIH. Similarly, portal vein thrombosis was seen more commonly in NASH compared to CC, AC or AIH, but bacterial peritonitis, ascites, encephalopathy and dialysis were higher in AC with similar prevalence in CC and NASH. Some of these observations have been previously published (32,33). There are few weaknesses for study of this nature. The UNOS data are not granular enough to examine the prevalence of metabolic syndrome or explant histology. Previous studies on explant pathology has reported conflicting findings; while some studies have shown a potential etiology in many patients with CC, others did not identify any potential cause in majority of patients based on histology of the explant (1, 4, 10). Another weakness of the study is our inability to examine the components of metabolic syndrome or previous history of obesity. Loss of body mass is not uncommon in those with advanced cirrhosis, and equally important is fluid overload overestimating BMI. We believe that those changes are applicable to all four groups, and therefore, unlikely to influence our observations. Despite these weaknesses, these very large data sets come from transplant centers with significant expertise in the diagnosis of liver disease, and moreover, we believe that data collection by the UNOS has become more systematic after the introduction of MELD scores for organ allocation in 2002. Hence, based on our observations, we believe that most patients currently listed as cryptogenic cirrhosis are truly ‘cryptogenic’ in nature and based on their differences in risk factors, we cannot equate cryptogenic cirrhosis with the term ‘NASH cirrhosis’. Our observations merit further systematic examination of patients with CC to unravel other unrecognized causes of chronic liver disease. In addition, we do not
believe we should combine CC and NASH for epidemiological or outcome studies even after stratifying them for the presence of obesity or type 2 diabetes mellitus.
REFERENCES: 1. Greeve M, Ferrell L, Kim M, Combs C, Roberts J, Ascher N, Wright TL. Cirrhosis of undefined pathogenesis: absence of evidence for unknown viruses or autoimmune processes. Hepatology. 1993 Apr;17(4):593-8. 2. Poonawala A, Nair SP, Thuluvath PJ. Prevalence of obesity and diabetes in patients with cryptogenic cirrhosis: a case-control study. Hepatology. 2000; 32:689–692. 3. Caldwell SH, Oelsner DH, Iezzoni JC, Hespenheide EE, Battle EH, Driscoll CJ. Cryptogenic cirrhosis: clinical characterization and risk factors for underlying disease. Hepatology. 1999; 29:664–669. 4. Ayata G, Gordon FD, Lewis WD, Pomfret E, Pomposelli JJ, Jenkins RL, Khettry U. Cryptogenic cirrhosis: clinicopathologic findings at and after liver transplantation. Hum Pathol. 2002; 33:1098–1104. 5. Powell EE, Cooksley WGE, Hanson R, Searle J, Halliday JW, Powel LW. The natural history of nonalcoholic steatohepatitis: a follow-up study of forty-two patients for up to 21 years. Hepatology. 1990; 11:74–80. 6. Hübscher SG. Histological assessment of non-alcoholic fatty liver disease. Histopathology. 2006 Nov;49(5):450-65. 7. Czaja AJ, Carpenter HA, Santrach PJ, Moore SB, Homburger HA. Gastroenterology. The nature and prognosis of severe cryptogenic chronic active hepatitis. Gastroenterology 1993;104(6):1755-61.
8. Kaymakoglu S, Cakaloglu Y, Demir K, Türkoglu S, Badur S, Gürel S, et al. Is severe cryptogenic chronic hepatitis similar to autoimmune hepatitis? J Hepatol. 1998 Jan;28(1):78-83. 9. Heringlake S, Schütte A, Flemming P, Schmiegel W, Manns MP, Tillmann HL. Presumed cryptogenic liver disease in Germany: High prevalence of autoantibodynegative autoimmune hepatitis, low prevalence of NASH, no evidence for occult viral etiology. Z Gastroenterol 2009;47(5):417-23. 10. Duclos-Vallée JC, Yilmaz F, Johanet C, Roque-Afonso AM, Gigou M, Trichet C, et al. Could post-liver transplantation course be helpful for the diagnosis of so called cryptogenic cirrhosis? Hum Pathol. 2002 Nov;33(11):1098-104. 11. Charlton MR, Burns JM, Pedersen RA, Watt KD, Heimbach JK, , Dierkhising RA. Frequency and outcomes of liver transplantation for nonalcoholic steatohepatitis in the United States. Gastroenterology. 2011; 141:1249–1253 12. Wong RJ, Aguilar M, Cheung R, Perumpail RB, Harrison SA, Younossi ZM, Ahmed A. Nonalcoholic steatohepatitis is the second leading etiology of liver disease among adults awaiting liver transplantation in the United States. Gastroenterology. 2015; 148:547–555. 13. Kim WR, Lake JR, Smith JM, Skeans MA, Schladt DP, Edwards EB, et al OPTN/SRTR 2013 Annual Data Report: liver. Am J Transplant. 2015; 15:1–28. 22. 14. Adam R, Karam V, Delvart V, O'Grady J, Mirza D, Klempnauer J, Castaing D, Neuhaus P et al. Evolution of indications and results of liver transplantation in Europe. A report from the European Liver Transplant Registry (ELTR). J Hepatol. 2012; 57:675– 688.
15. Younossi ZM, Otgonsuren M, Henry L, Venkatesan C, Mishra A, Erario M, Hunt S et al. Association of nonalcoholic fatty liver disease (NAFLD) with hepatocellular carcinoma (HCC) in the United States from 2004 to 2009. Hepatology. 2015; 62:1723–1730. 16. Wong RJ, Cheung R, Ahmed A. Nonalcoholic steatohepatitis is the most rapidly growing indication for liver transplantation in patients with hepatocellular carcinoma in the U.S. 2014 Jun;59(6):2188-95. 17. Malik SM, Devera ME, Fontes P, Shaikh O, Sasatomi E, Ahmad J. Recurrent disease following liver transplantation for nonalcoholic steatohepatitis cirrhosis. Liver Transpl. 2009; 15:1843–1851. 18. Yalamanchili K, Saadeh S, Klintmalm GB, Jennings JW, Davis GL. Nonalcoholic fatty liver disease after liver transplantation for cryptogenic cirrhosis or nonalcoholic fatty liver disease. Liver Transpl. 2010; 16:431–439. 19. Contos MJ, Cales W, Sterling RK, et al. Development of nonalcoholic fatty liver disease after orthotopic liver transplantation for cryptogenic cirrhosis. Liver Transpl. 2001; 7:363– 373. 20. Ong J, Younossi ZM, Reddy V, Price LL, Gramlich T, Mayes J, Boparai N. Cryptogenic cirrhosis and posttransplantation nonalcoholic fatty liver disease. Liver Transpl. 2001; 7:797– 801. 21. Dureja P, Mellinger J, Agni R, Chang F, Avey G, Lucey M, Said A. NAFLD recurrence in liver transplant recipients. Transplantation. 2011; 91:684–689. 22. Ratziu V, Bonyhay L, Di Martino V, Charlotte F, Cavallaro L, Sayegh-Tainturier MH et al. Survival, liver failure, and hepatocellular carcinoma in obesity-related cryptogenic cirrhosis. Hepatology. 2002; 35:1485–1493.
23. Hassan MM, Curley SA, Li D, Kaseb A, Davila M, Abdalla EK et al. Association of diabetes duration and diabetes treatment with the risk of hepatocellular carcinoma. Cancer. 2010; 116:1938–1946. 24. Bugianesi E, Leone N, Vanni E, Marchesinin G, Brunello F, Carucci P et al. Expanding the natural history of nonalcoholic steatohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology. 2002; 123:134–140. 25. Nair S, Mason A, Eason J, Loss G, Perrillo RP. Is obesity an independent risk factor for hepatocellular carcinoma in cirrhosis? Hepatology. 2002 Jul;36(1):150-5. 26. Marrero JA, Fontana RJ, Fu S, Conjeevaram HS, Su GL, Lok AS. Alcohol, tobacco and obesity are synergistic risk factors for hepatocellular carcinoma. J Hepatol. 2005; 42:218– 224. 27. Ascha MS, Hanouneh IA, Lopez R, Tamimi TA, Feldstein AF, Zein NN. The incidence and risk factors of hepatocellular carcinoma in patients with nonalcoholic steatohepatitis. Hepatology. 2010; 51:1972–1978. 28. Pais R, Lebray P, Rousseau G, Charlotte F, Esselma G, Savier E et al. Nonalcoholic fatty liver disease increases the risk of hepatocellular carcinoma in patients with alcoholassociated cirrhosis awaiting liver transplants. Clin Gastroenterol Hepatol. 2015; 13:992. 29. Dyson J, Jaques B, Chattopadyhay D, Lochan R, Grahan J, Das D et al. Hepatocellular cancer – the impact of obesity, type 2 diabetes and a multidisciplinary team. J Hepatol. 2014; 60:110–117. 30. Mittal S, Sada YH, El-Serag HB, Kanwal F, Duan Z, Temple S et al. Temporal trends of nonalcoholic fatty liver disease-related hepatocellular carcinoma in the veteran affairs population. Clin Gastroenterol Hepatol. 2015;
31. Piscaglia F, Svegliati-Baroni G, Barchetti A, Pecorelli A, Marinelli S, Tiribelli C et al. Clinical patterns of hepatocellular carcinoma (hcc) in non alcoholic fatty liver disease (NAFLD): a multicenter prospective study. Hepatology. 2016; 63:827–838. 32. Stine JG, Shah NL, Argo CK, Pelletier SJ, Caldwell SH, Northup PG. Increased risk of portal vein thrombosis in patients with cirrhosis due to Non-Alcoholic Steatohepatitis (NASH). Liver Transpl. 2015; 21:1016–1021. 33. Tripodi A, Fracanzani AL, Primignani M, Chantarangkul V, Clerici M, Mannucci PM et al. Procoagulant imbalance in patients with nonalcoholic fatty liver disease. J Hepatol. 2014; 61:148–154.
FIGURES LEGENDS: Fig. 1. Frequency (1A) and percentage from the total UNOS population (1B) for NASH and cryptogenic cirrhosis by year at listing. Fig. 2. Frequency of the listed individuals by age. Fig. 3. The average initial MELD score at listing for NASH, cryptogenic cirrhosis (CC), alcoholic cirrhosis (AC) and autoimmune hepatitis (AIH).
Figure 1: Frequency (1A) and proportion (1B) of NASH and cryptogenic cirrhosis by year at listing. A)
B)
Figure 2: Frequency of the listed individuals by age and year of listing.
Figure 3: The average initial MELD score by year of listing for NASH, cryptogenic cirrhosis (CC), alcoholic cirrhosis (AC) and autoimmune hepatitis (AIH).
Table 1A: Characteristics of the four groups; cryptogenic cirrhosis (CC), NASH, alcoholic cirrhosis (AC), autoimmune hepatitis (AIH) for the period 2002-2016.
Characteristic
CC
NASH
p-value AC
p-value
AIH
p-value
7999
11302
.
21714
.
3447
.
Diabetes type 2, N (%)
2008 (28.5 %)
5598 (52.8 %)
<.0001
3320 (16.1 %)
<.0001
527 (16.3 %)
<.0001
Obese, N (%)
3361 (42 %)
7381 (65.3 %)
<.0001
7106 (32.7 %)
<.0001
1278 (37.1 %)
<.0001
BMI, mean(SD)
29.4(5.8)
32.7 (6.1)
<.0001
28.1(5.5)
<.0001
28.6 (6.3)
<.0001
Age, mean (SD)
57.2(10.2)
59 (8)
<.0001
54.2 (9)
<.0001
48.6 (14.3)
<.0001
Female, N (%)
3631 (45.4 %)
5536 (49 %)
<.0001
5344 (24.6 %)
<.0001
2545 (73.8 %)
<.0001
Ethnicity, N (%)
<.0001
<.0001
<.0001
White
5789 (72.4 %)
9300 (82.3 %)
.
16798 (77.4 %)
.
2094 (60.7 %)
.
Black
366 (4.6 %)
167 (1.5 %)
.
697 (3.2 %)
.
598 (17.3 %)
.
Hispanic 1516 (19 %)
1491 (13.2 %)
.
3601 (16.6 %)
.
606 (17.6 %)
.
Asian
247 (3.1 %)
189 (1.7 %)
.
313 (1.4 %)
.
91 (2.6 %)
.
Other
81 (1 %)
155 (1.4 %)
.
305 (1.4 %)
.
58 (1.7 %)
.
18.4 (8.6)
17.8 (8.2)
<.0001
19.8 (9.2)
<.0001
19.4 (9.7)
<.0001
Bacterial peritonitis, N (%)
463 (6 %)
672 (6 %)
0.82
1943 (9.2 %)
<.0001
228 (6.9 %)
0.072
Dialysis twice in prior week, N (%)
521 (6.6 %)
764 (6.8 %)
0.57
1753 (8.2 %)
<.0001
163 (4.8 %)
0.0002
Portal vein thrombosis, N (%)
451 (5.8 %)
756 (6.8 %)
0.007
868 (4.1 %)
<.0001
168 (5.1 %)
0.10
HCC, N (%)
1033 (12.9 %)
2166 (19.2 %)
<.0001
3041 (14 %)
0.01
331 (9.6 %)
<.0001
MELD at listing, mean (SD)
Other characteristics
Characteristic
CC
NASH
Ascites, N (%)
p-value AC
p-value
0.0001
<.0001
AIH
p-value <.0001
Absent
1424 (17.8 %)
2212 (19.6 %)
.
3082 (14.2 %)
.
872 (25.3 %)
.
Slight
4496 (56.3 %)
6028 (53.4 %)
.
11836 (54.6 %)
.
1839 (53.5 %)
.
Moderate 2062 (25.8 %)
3056 (27.1 %)
.
6770 (31.2 %)
.
729 (21.2 %)
.
Encephalopathy, N (%)
0.26
<.0001
<.0001
None
2757 (34.5 %)
3775 (33.4 %)
.
6233 (28.7 %)
.
1491 (43.3 %)
.
1-2
4728 (59.2 %)
6808 (60.3 %)
.
13741 (63.4 %)
.
1683 (48.9 %)
.
3-4
497 (6.2 %)
713 (6.3 %)
.
1714 (7.9 %)
.
266 (7.7 %)
.
The differences between the groups were tested by using logistic regressions for categorical variables, and two-sample t-test for continuous variables. P-values are for hypotheses testing the differences from cryptogenic cirrhosis.
Table 1B. Characteristics of the four groups; cryptogenic cirrhosis (CC), NASH, alcoholic cirrhosis (AC), autoimmune hepatitis (AIH) for the period 2011-2016.
CC
NASH
Characteristic
AC p-value
N
2728
7308
Diabetes type 2, N (%)
850 (31.9 %)
3933 (56 %)
Obese, N (%)
AIH p-value
10856
p-value 1541
<0.01
1928 (18 %)
<0.01
306 (20.2 %)
<0.01
1057 (38.7 %) 4689 (64.2 %)
<0.01
3671 (33.8 %)
<0.01
587 (38.1 %)
0.67
BMI, mean (SD)
29 (5.7)
32.6 (6.2)
<0.01
28.3 (5.5)
<0.01
28.8 (6.4)
0.44
Age, mean (SD)
57.5 (10.8)
59.4 (8)
<0.01
54.2 (9.4)
<0.01
48.7 (14.4)
<0.01
Female, N (%)
1218 (44.6 %) 3498 (47.9 %)
<0.01
2764 (25.5 %)
<0.01
1105 (71.7 %)
<0.01
Ethnicity, N (%)
<0.01 White
1853 (67.9 %) 5910 (80.9 %)
8249 (76 %)
883 (57.3 %)
Black
130 (4.8 %)
110 (1.5 %)
390 (3.6 %)
292 (18.9 %)
Hispanic 611 (22.4 %)
1046 (14.3 %)
1844 (17 %)
296 (19.2 %)
Asian
99 (3.6 %)
139 (1.9 %)
193 (1.8 %)
44 (2.9 %)
Other
35 (1.3 %)
103 (1.4 %)
180 (1.7 %)
26 (1.7 %)
19.5 (8.8)
18.3 (8.4)
<0.01
20.9 (9.6)
<0.01
20.7 (9.9)
<0.01
Bacterial peritonitis, N (%)
194 (7.2 %)
486 (6.7 %)
0.42
1097 (10.3 %)
<0.01
124 (8.1 %)
0.26
Dialysis twice in prior week N (%)
244 (8.9 %)
576 (7.9 %)
0.08
1103 (10.2 %)
0.06
95 (6.2 %)
<0.01
Portal vein thrombosis, N (%)
229 (8.5 %)
577 (8 %)
0.41
537 (5 %)
<0.01
107 (7 %)
0.09
HCC, N (%)
439 (16.1 %)
1577 (21.6 %)
<0.01
1636 (15.1 %)
0.18
164 (10.6 %)
<0.01
MELD at listing, mean (SD)
Other characteristics
Ascites, N (%)
<0.01 Absent
556 (20.4 %)
1541 (21.1 %)
1756 (16.2 %)
416 (27 %)
CC
NASH
Characteristic
AC p-value
Slight
1423 (52.2 %) 3685 (50.4 %)
Moderate 748 (27.4 %)
2079 (28.5 %)
AIH p-value
p-value
5445 (50.2 %)
769 (50 %)
3650 (33.6 %)
354 (23 %)
Encephalopathy, N (%)
<0.01 None
1015 (37.2 %) 2500 (34.2 %)
3338 (30.8 %)
679 (44.1 %)
1-2
1534 (56.3 %) 4324 (59.2 %)
6612 (60.9 %)
734 (47.7 %)
3-4
178 (6.5 %)
901 (8.3 %)
126 (8.2 %)
481 (6.6 %)
The differences between the groups were tested by using logistic regressions for categorical variables, and two-sample t-test for continuous variables. P-values are for hypotheses testing the differences from cryptogenic cirrhosis.
Table 2: Multivariate logistic regression modeling for probability of having NASH diagnosis.
Effect
Lower Upper 95% 95% Odds Confidence Confidence Ratio Limit for Limit for Estimate Odds Ratio Odds Ratio
p-value
Diabetes type 2
4.706
4.505
4.917
<.0001
Obesity
3.476
3.326
3.632
<.0001
Age 60+
2.43
2.327
2.538
<.0001
Female
2.006
1.921
2.095
<.0001
Caucasian
2.315
2.192
2.445
<.0001
The probability of NASH (N=11302) was compared to the rest of the UNOS population (N=126,719) by multivariate logistic regression modeling.
25
Table 3: Characteristics of NASH and cryptogenic cirrhosis matched by age, gender and year of listing.
Characteristics
Level
Cryptogenic cirrhosis
NASH
p-value
N
7201
7201
Diabetes type 2, N (%)
1947 (30.2 %)
3369 (50.7 %)
<.0001
Obese, N (%)
3045 (42.3 %)
4718 (65.5 %)
<.0001
BMI, mean (SD)
29.4 (5.8%)
32.7 (6%)
<.0001
Ethnicity, N (%)
<.0001 White
5196 (72.2 %)
5955 (82.7 %)
Black
311 (4.3 %)
118 (1.6 %)
Hispanic
1393 (19.3 %)
916 (12.7 %)
Asian
228 (3.2 %)
116 (1.6 %)
Other
73 (1 %)
96 (1.3 %)
Portal vein thrombosis, N (%)
427 (6.1 %)
432 (%)
0.67
HCC, N (%)
968 (13.4 %)
1296 (18 %)
<.0001
Other characteristics
The differences between the groups were tested by using a conditional logistic regression for categorical variables, and paired t-test for continuous variables.
Table 4. Characteristics of NASH and cryptogenic cirrhosis stratified by BMI below and above 30.
26
CC(BMI<30)
NASH (BMI<30)
p-value CC(BMI>=30) NASH(BMI>=30) p-value
Number
4638
3921
.
Diabetes type 2, N (%)
1020 (24.6 %)
1861 (50.4 %)
<.0001 988 (34 %)
Age, N (%)
7381
.
3737 (54.1 %)
<.0001
<.0001
<.0001
19-30
157 (3.4 %)
16 (0.4 %)
34 (1 %)
25 (0.3 %)
30-59
2180 (47 %)
1498 (38.2 %)
1766 (52.5 %)
3682 (49.9 %)
60+
2301 49.6 %)
2407 (61.4 %)
1561 (46.4 %)
3674 (49.8 %)
2071 (44.7 %)
1967 (50.2 %)
<.0001 1560 (46.4 %)
3569 (48.4 %)
Female, N (%) Ethnicity, N (%)
HCC, N (%)
3361
<.0001
0.06 <.0001
White
3236 (69.8 %)
3161 (80.6 %)
2553 (76 %)
6139 (83.2 %)
Black
263 (5.7 %)
62 (1.6 %)
103 (3.1 %)
105 (1.4 %)
Hispanic
889 (19.2 %)
552 (14.1 %)
627 (18.7 %)
939 (12.7 %)
Asian
208 (4.5 %)
111 (2.8 %)
39 (1.2 %)
78 (1.1 %)
Other
42 (0.9 %)
35 (0.9 %)
39 (1.2 %)
120 (1.6 %)
584 (12.6 %)
715 (18.2 %)
<.0001 449 (13.4 %)
1451 (19.7 %)
<.0001
The differences between the groups were tested by using logistic regression
Table 5: Characteristics of NASH and cryptogenic cirrhosis stratified presence or absence of diabetes type 2 (individuals with missing values for diabetes were excluded).
27
CC (with type NASH (with type CC (no Type 2 NASH (no p-value DM) 2 DM) 2 DM) Type 2 DM)
p-value
Number
2008
5598
.
Obese, N (%)
988 (49.2 %)
3737 (66.8 %)
Female, N (%)
927 (46.2 %)
2653 (47.4 %)
Age, N (%)
4999
<.0001
1919 (38 %)
3167 (63.4 %) <.0001
0.34
2260 (44.8 %)
2524 (50.5 %) <.0001
0.34
<.0001
19-30
4 (0.2 %)
6 (0.1 %)
182 (3.6 %)
33 (0.7 %)
30-59
862 (42.9 %)
2326 (41.6 %)
2606 (51.6 %)
2534 (50.7 %)
60+
1142 (56.9 %)
3266 (58.3 %)
2260 (44.8 %)
2432 (48.6 %)
Ethnicity, N (%)
HCC, N (%)
5048
<.0001
<.0001
White
1382 (68.8 %)
4525 (80.8 %)
3709 (73.5 %)
4192 (83.9 %)
Black
70 (3.5 %)
50 (0.9 %)
254 (5 %)
105 (2.1 %)
Hispanic
465 (23.2 %)
814 (14.5 %)
880 (17.4 %)
584 (11.7 %)
Asian
64 (3.2 %)
118 (2.1 %)
160 (3.2 %)
61 (1.2 %)
Other
27 (1.3 %)
91 (1.6 %)
45 (0.9 %)
57 (1.1 %)
354 (17.6 %)
1297 (23.2 %)
587 (11.6 %)
740 (14.8 %)
<.0001
The differences between the groups were tested by using logistic regression
<.0001
28
•
About 4% of patients are listed for liver transplant with a diagnosis of cryptogenic cirrhosis (CC)
•
Patients with CC have different clinical characteristics when compared to non-alcoholic steatohepatitis (NASH), alcoholic (AC) or autoimmune hepatitis (AIH) cirrhosis
•
Differences between CC and NASH remained unchanged when matched for age, sex and year of listing, or when stratified by the presence of diabetes mellitus or obesity
•
Based on the significant differences of risk factors, CC should not be equated with the term ‘NASH cirrhosis’.
S0168-8278(17)32441-8 https://doi.org/10.1016/j.jhep.2017.11.018 JHEPAT 6760
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
10 August 2017 21 October 2017 2 November 2017
Please cite this article as: Thuluvath, P.J., Kantsevoy, S., Thuluvath, A.J., Savva, Y., Is Cryptogenic Cirrhosis Different from NASH Cirrhosis?, Journal of Hepatology (2017), doi: https://doi.org/10.1016/j.jhep.2017.11.018
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
JHEPAT-D-17-01587 (clean copy)
Is Cryptogenic Cirrhosis Different from NASH Cirrhosis? Paul J. Thuluvath, MD, FRCP, FAASLD (1,2), Sergey Kantsevoy, MD, PhD (1,2), Avesh J Thuluvath (3), Yulia Savva, PhD (1) (1) Institute of Digestive Health & Liver Diseases, Mercy Medical Center, Baltimore and (2) Departments of Surgery & Medicine, University of Maryland School of Medicine, Baltimore, Maryland, (3) Johns Hopkins University School of Medicine, Baltimore Words: 3884; Tables: 6; Figure: 3 Key Words: UNOS, autoimmune hepatitis, alcoholic cirrhosis, age and gender matched Address for correspondence: Paul J. Thuluvath, MD, FAASLD, FRCP, Professor of Medicine & Surgery, Institute of Digestive Health & Liver Diseases, Mercy Medical Center, Baltimore, MD 21202; Tel: 410 332 9308; Email: [email protected] Abbreviations: NASH: Non-alcoholic steatohepatitis, CC: Cryptogenic cirrhosis, AC: alcoholic cirrhosis, AIH: autoimmune hepatitis, UNOS: United Network for Organ Sharing, HCC: hepatocellular carcinoma Financial support: None; Conflicts of interest: None Author contributions: PJT, SK, AJT and YS made substantial contributions to the conception and design; or the acquisition, analysis, or interpretation of the data; PJT contributed the drafting of the article or critical revision for important intellectual content, and all authors approved the final version, and agree to be accountable for all aspects of the work
ABSTRACT Background & Aims: We hypothesized that patients currently diagnosed as CC have truly ‘cryptogenic’ liver disease, and unlikely to have evolved from NASH. Methods: To investigate this, we compared the clinical characteristics of adults with CC (n=7,999) with NASH (n=11,302), alcohol (AC, n=21,714) and autoimmune hepatitis (AIH, n=3,447) using the UNOS database from 2002-16. We performed an age, gender and year of listing matched comparison of CC and NASH (n= 7,201 in each group), and also stratified by the presence of obesity or diabetes mellitus (DM). Results: From 2002 to 2016, patients listed with a diagnosis of NASH increased from about 1% to 16% while CC decreased from 8% to 4%. Logistic regression model using the entire UNOS data (n=138,021) suggested that the strongest predictors of NASH were type 2 DM, obesity, age > 60 years, female gender and White race. Type 2 DM was more common in NASH (53%) compared to CC (29%), AC (16%) and AIH (16%), and obesity was more common in NASH (65.3%) compared to other three groups (33-42%). There were more Whites (82.3%) in NASH group with a lower prevalence of black, Hispanic and Asian race as compared to the other three groups. HCC was more commonly seen in NASH (19% versus 913% in the other groups) and this is also not influenced by obesity and type 2 DM. The differences between CC and NASH remained unchanged even when two groups were matched for age, gender and year of listing, or when stratified by the presence or absence of obesity or type 2DM. Conclusions: Based on risk perspectives, CC should not be equated with the term ‘NASH cirrhosis’.
Cryptogenic cirrhosis (CC) is a diagnosis of exclusion when there is no other known identifiable etiology (1). About two decades ago, when non-alcoholic steatohepatitis (NASH) was emerging as a common cause of chronic liver disease in the Western world, we and others had suggested that many patients with cryptogenic cirrhosis may have evolved from NASH (2-4). These observations were made based on patient characteristics and the higher prevalence of risk factors including type 2 diabetes mellitus (DM), obesity or metabolic syndrome in those with CC since histology was often not helpful once the patient with NASH had progressed to cirrhosis (5,6). Some investigators, however, had indicated that burnt-out autoimmune hepatitis (AIH) and occult alcoholism may also be considered as ‘potential causes’ of CC (7-10). Other causes such as previous hepatitis B infection, iron overload, alpha-1 antitrypsin deficiency, and thrombotic/vascular disorders are unlikely to be major contributors of cryptogenic cirrhosis (710). Over the past two decades, physicians have become more confident in making a firm diagnosis of NASH cirrhosis based on history, risk factors and the absence of other etiological factors. The listing diagnosis in the United Network for Organ Sharing (UNOS) database suggests that the number of patients with NASH cirrhosis has increased since 2000 while those with CC has decreased (11-13). Despite the increased awareness of NASH as an important cause of cirrhosis, a significant number of patients (400-600 per year) are currently being listed for liver transplantation with the UNOS with a diagnosis of CC (11-13). In the European LT registry, about 4% of patients are listed as CC (14). It is more than likely that those patients currently listed as CC have truly ‘cryptogenic’ chronic liver disease and less likely to have evolved from NASH. Nevertheless, there is a trend among many physicians to use CC and NASH cirrhosis synonymously, and some investigators have combined them as a single entity, when outcomes
are reported, especially among the sub-group of patients with CC with obesity (defined as BMI >30) (11, 15, 16). The objective of our study is to characterize the patients currently listed for liver transplantation with a diagnosis of CC cirrhosis and compare them with cirrhosis from NASH, alcohol and AIH. PATIENTS AND METHODS We analyzed the UNOS database from 2002 to November 2016 of all adult (> 18 years) patients. We selected 2002 since the UNOS introduced MELD scores for organ allocation in February 2002 and we assumed that the data collection may have improved since then, and additionally, examination of the annual data variability by diagnosis code showed that the data collection was perhaps done more systematically since 2002. Moreover, the diagnosis of NASH was utilized in the UNOS database since 2001, and it was also around that time when NASH cirrhosis was diagnosed more consistently by liver transplant centers based on publications that suggested that many patients with CC may have progressed from NASH (2,3). Since it has been suggested that some patients with CC may have evolved from occult alcoholism or AIH, we compared CC with alcoholic cirrhosis (AC) and AIH cirrhosis in addition to NASH cirrhosis. The groups were defined based on their primary diagnosis at the time of listing with the UNOS. The only exception was those with hepatocellular carcinoma (HCC); in those patients, we included secondary diagnosis for this study. For example, an individual was considered as NASH if the primary diagnosis at listing was NASH or if the primary diagnosis was hepatocellular carcinoma (HCC) with a secondary diagnosis of NASH cirrhosis. The same rule was applied to the other three groups. For a minority of patients, the diagnosis had changed from the time of listing to the time of transplant, but we did not account for this change while
defining our groups. Since the number of patients with a change in primary diagnosis was a very small, we felt that it would have no impact on the study findings. Less than 10% of patients were listed for a simultaneous liver and kidney transplant including 1,272 (11%) for NASH, 805 (10%) for CC, 1,904 (8.8%) for AC and 190 (5.5%) for AIH. We included those received a live donor [200 (1.8%) NASH, 175 (2.2%) CC, 259 (1.25) AC, and 87 (2.5%) AIH]. Although several patients had multiple listings or were relisted, we included only one listed data point per individual. Our final study group included 11,302 NASH cirrhosis, 7,999 CC, 21,714 AC and 3,447 AIH cirrhosis. We collected following potential risk factors and confounders at the time of listing including race, age, body mass index (BMI), obesity (defined as BMI>=30), initial waiting list Model for End-stage Liver Disease (MELD) score, gender and DM. The data for DM were unknown for 5.4% of patients, and these patients were excluded from the analysis of the association between the outcome and DM. We also collected data on the prevalence of complications of cirrhosis including the prevalence of HCC, and history of spontaneous bacterial peritonitis, ascites, encephalopathy, portal vein thrombosis or dialysis. STATISTICAL ANALYSIS The descriptive statistics was presented for all four groups. The differences between the groups were tested by using logistic regressions for categorical variables, and two-sample t-test for continuous variables. Although the MELD score was not normally-distributed, considering a large sample size, a t-test was considered appropriate for hypothesis testing. To identify the patient characteristics with the strongest association with NASH, we performed a multivariate logistic regression analysis using the entire UNOS data (N=138,021).
We also compared CC and NASH cirrhosis by analyzing a 1:1 matched (matched for age, gender and year of listing) data to reduce the confounding effects of age, gender and year of listing. Each case (CC) was matched to a control (NASH) within 5-years of age, with the same gender and the year of listing (+ 2 years). Matched analysis included 7,201 CC and 7,201 NASH. Additionally, we also compared CC and NASH (unmatched) data stratified separately by presence or absence of obesity (defined by World Health Organization scale with BMI > 30) and type 2 DM to examine if there are similarities between the two groups for these two important risk factors for NASH cirrhosis.
RESULTS During the study period, the number (Figure 1A) and proportion (Figure 1B) of patients with NASH cirrhosis increased while those with CC decreased. From 2002 to 2016, patients listed with a diagnosis of NASH increased from about 1% to 16% while cryptogenic cirrhosis decreased from 8% to 4%. Yet, a significant number of patients continued to have a listing diagnosis of cryptogenic cirrhosis (572 in 2002, 536 in 2010, 455 in 2015). Table 1A shows demographic characteristics of patients with CC, NASH, AC and AIH. Those with NASH cirrhosis were older (mean age 59 years) compared to the other groups, and the mean age increased over the study period as shown in figure 2. Until 2007, the frequency of listed individuals in the age of 50-59 and 60-69 was similar, but proportion of older (60-69 age group) patients continued to increase after 2007, and this rate of increase was significantly faster for NASH cirrhosis compared to other groups perhaps reflecting the natural history of NASH cirrhosis (figure 2). The MELD score at listing was lowest (mean 17.2) for NASH cirrhosis,
followed by CC (18.4), AIH (19.4) and AC (19.8) (Table 1). There was an apparent linear trend of an increase in listing MELD score for all four groups during the study period (Figure 3). Table 1 shows that there are significant differences in clinical characteristics among the groups. There was a higher prevalence of type 2 DM in NASH cirrhosis (53%) compared to CC (29%), AC (16%) and AIH (16%). Similarly, obesity was more common in NASH cirrhosis (65.3%) compared to other three groups (33-42%). There were more Whites (82.3%) in NASH group with a lower prevalence of black, Hispanic and Asian race as compared to the other three groups. The highest prevalence of African-American and Hispanic population was for autoimmune hepatitis (about 35%) followed by cryptogenic cirrhosis (24%). There were gender differences, as expected, between the groups with predominantly women in AIH (74%) and men (75%) in AC. Since the listing of patients with cryptogenic cirrhosis has stabilized in recent years, we analyzed the characteristics of patients listed during 2011-2016 Table 1B). As shown in table 1B, the differences remained unchanged. The complications were also different between the groups (Table 1). HCC was more common in NASH (19%) followed by AC (14%), CC (13%) and AIH (10%). Other complications of cirrhosis were higher in AC including the prevalence of spontaneous bacterial peritonitis, dialysis, portal vein thrombosis and encephalopathy (stage 2 and 3) and encephalopathy. These complications were lowest for the autoimmune group. When logistic regression modeling was done on the entire cohort of patients listed with UNOS (using logistic regression model using the entire UNOS data; (n =138,021), we found that the strongest predictors of having a NASH diagnosis were type 2 DM [Odds Ratio (OR) 4.7 after
controlling for other factors], obesity, age 60 and above, female gender and White race (Table 2). This association between type 2 DM and NASH was even stronger for non-obese population (OR=5.74) compared to the obese population (OR=4.4). Table 3 shows the characteristics of CC and NASH matched by the age, gender and year of listing. The differences between the two groups remained unchanged even when two groups were matched for age and gender. We further analyzed the groups after stratifying by obesity (Table 4) and type 2 DM (Table 5). The prevalence of type 2 DM was higher in non-obese NASH (50.4% vs. 24.6%) and obese NASH (54.1% vs. 34%) when compared to non-obese and obese CC respectively. When groups were stratified by presence or absence of type 2 DM, obesity was higher for NASH group (66.8% vs. 49.2% for those with DM, and 63.4% vs. 38% for those without DM) when compared to CC. There were also differences in HCC in NASH and CC with a higher prevalence in NASH and this was also not influenced by other confounders such as obesity and type 2 DM. DISCUSSION Our study shows that there are significant differences in the clinical characteristics of patients with CC when compared to NASH, AC or AIH. Using the large data from UNOS, we confirmed the well-known association between NASH cirrhosis with type 2 DM, obesity, older age, female gender and White ethnicity. When these characteristics were compared among four groups of patients with CC, NASH, AC and AIH, these groups had significantly different clinical characteristics. The differences remained unchanged in those who were listed between 2011 and 2016. We further analyzed these characteristics in CC and NASH after stratifying the patients with and without DM or obesity. Similarly, when CC patients were matched with NASH by age, gender and year of listing, differences remained unchanged. These data clearly indicate that CC
and NASH are different based on their risk perspectives and we cannot assume that those who are currently diagnosed as CC have evolved from NASH, AC or AIH. In addition, we cannot assume that obese or diabetic patients with CC have NASH cirrhosis. When it was suggested two decades ago that most patients with CC may have evolved from NASH, our understanding of NASH cirrhosis was just evolving (2,3). Currently, most transplant physicians are confident in making a firm diagnosis of NASH cirrhosis based on either a history long standing fatty liver disease or NASH compatible demographic characteristics and risk factors. This may explain the very different clinical characteristics between NASH and CC who are currently listed with UNOS. Our study, based on the differences in demographic characteristics, also suggests that it is quite unlikely that most patients with CC may have evolved from occult alcoholic liver disease or undiagnosed/burnt out AIH. We speculate that that physicians list patients as ‘cryptogenic cirrhosis’ only when they are unsure of the etiology, and our study appears to corroborate these assumptions. Our findings have many implications. Many physicians use CC and NASH in a synonymous fashion, and many outcome studies have combined a significant proportion of CC or unknown cirrhosis with NASH if they have obesity (11, 15,16). This may not be justified based on the findings of our study, and we suggest that future studies should treat them as distinct entities as they may have a different natural history, complications and post-liver transplant outcomes. In our experience steatosis is not uncommon after liver transplantation in obese subjects or those who develop obesity, and anecdotal experience of steatosis in those had liver transplantation for CC could not be a sound justification for assuming that CC and NASH are synonymous (17-21). The findings of our study also suggest that we should not stop exploring for other hitherto unrecognized causes of cirrhosis.
In our study, the highest prevalence of HCC was in NASH, and a combination of cirrhosis, type 2 DM and obesity may explain this finding (16, 22-31). The prevalence of HCC was 49% higher in NASH compared to CC, and patients with CC had lower prevalence of HCC when compared to AC, but the lowest prevalence was in AIH. Similarly, portal vein thrombosis was seen more commonly in NASH compared to CC, AC or AIH, but bacterial peritonitis, ascites, encephalopathy and dialysis were higher in AC with similar prevalence in CC and NASH. Some of these observations have been previously published (32,33). There are few weaknesses for study of this nature. The UNOS data are not granular enough to examine the prevalence of metabolic syndrome or explant histology. Previous studies on explant pathology has reported conflicting findings; while some studies have shown a potential etiology in many patients with CC, others did not identify any potential cause in majority of patients based on histology of the explant (1, 4, 10). Another weakness of the study is our inability to examine the components of metabolic syndrome or previous history of obesity. Loss of body mass is not uncommon in those with advanced cirrhosis, and equally important is fluid overload overestimating BMI. We believe that those changes are applicable to all four groups, and therefore, unlikely to influence our observations. Despite these weaknesses, these very large data sets come from transplant centers with significant expertise in the diagnosis of liver disease, and moreover, we believe that data collection by the UNOS has become more systematic after the introduction of MELD scores for organ allocation in 2002. Hence, based on our observations, we believe that most patients currently listed as cryptogenic cirrhosis are truly ‘cryptogenic’ in nature and based on their differences in risk factors, we cannot equate cryptogenic cirrhosis with the term ‘NASH cirrhosis’. Our observations merit further systematic examination of patients with CC to unravel other unrecognized causes of chronic liver disease. In addition, we do not
believe we should combine CC and NASH for epidemiological or outcome studies even after stratifying them for the presence of obesity or type 2 diabetes mellitus.
REFERENCES: 1. Greeve M, Ferrell L, Kim M, Combs C, Roberts J, Ascher N, Wright TL. Cirrhosis of undefined pathogenesis: absence of evidence for unknown viruses or autoimmune processes. Hepatology. 1993 Apr;17(4):593-8. 2. Poonawala A, Nair SP, Thuluvath PJ. Prevalence of obesity and diabetes in patients with cryptogenic cirrhosis: a case-control study. Hepatology. 2000; 32:689–692. 3. Caldwell SH, Oelsner DH, Iezzoni JC, Hespenheide EE, Battle EH, Driscoll CJ. Cryptogenic cirrhosis: clinical characterization and risk factors for underlying disease. Hepatology. 1999; 29:664–669. 4. Ayata G, Gordon FD, Lewis WD, Pomfret E, Pomposelli JJ, Jenkins RL, Khettry U. Cryptogenic cirrhosis: clinicopathologic findings at and after liver transplantation. Hum Pathol. 2002; 33:1098–1104. 5. Powell EE, Cooksley WGE, Hanson R, Searle J, Halliday JW, Powel LW. The natural history of nonalcoholic steatohepatitis: a follow-up study of forty-two patients for up to 21 years. Hepatology. 1990; 11:74–80. 6. Hübscher SG. Histological assessment of non-alcoholic fatty liver disease. Histopathology. 2006 Nov;49(5):450-65. 7. Czaja AJ, Carpenter HA, Santrach PJ, Moore SB, Homburger HA. Gastroenterology. The nature and prognosis of severe cryptogenic chronic active hepatitis. Gastroenterology 1993;104(6):1755-61.
8. Kaymakoglu S, Cakaloglu Y, Demir K, Türkoglu S, Badur S, Gürel S, et al. Is severe cryptogenic chronic hepatitis similar to autoimmune hepatitis? J Hepatol. 1998 Jan;28(1):78-83. 9. Heringlake S, Schütte A, Flemming P, Schmiegel W, Manns MP, Tillmann HL. Presumed cryptogenic liver disease in Germany: High prevalence of autoantibodynegative autoimmune hepatitis, low prevalence of NASH, no evidence for occult viral etiology. Z Gastroenterol 2009;47(5):417-23. 10. Duclos-Vallée JC, Yilmaz F, Johanet C, Roque-Afonso AM, Gigou M, Trichet C, et al. Could post-liver transplantation course be helpful for the diagnosis of so called cryptogenic cirrhosis? Hum Pathol. 2002 Nov;33(11):1098-104. 11. Charlton MR, Burns JM, Pedersen RA, Watt KD, Heimbach JK, , Dierkhising RA. Frequency and outcomes of liver transplantation for nonalcoholic steatohepatitis in the United States. Gastroenterology. 2011; 141:1249–1253 12. Wong RJ, Aguilar M, Cheung R, Perumpail RB, Harrison SA, Younossi ZM, Ahmed A. Nonalcoholic steatohepatitis is the second leading etiology of liver disease among adults awaiting liver transplantation in the United States. Gastroenterology. 2015; 148:547–555. 13. Kim WR, Lake JR, Smith JM, Skeans MA, Schladt DP, Edwards EB, et al OPTN/SRTR 2013 Annual Data Report: liver. Am J Transplant. 2015; 15:1–28. 22. 14. Adam R, Karam V, Delvart V, O'Grady J, Mirza D, Klempnauer J, Castaing D, Neuhaus P et al. Evolution of indications and results of liver transplantation in Europe. A report from the European Liver Transplant Registry (ELTR). J Hepatol. 2012; 57:675– 688.
15. Younossi ZM, Otgonsuren M, Henry L, Venkatesan C, Mishra A, Erario M, Hunt S et al. Association of nonalcoholic fatty liver disease (NAFLD) with hepatocellular carcinoma (HCC) in the United States from 2004 to 2009. Hepatology. 2015; 62:1723–1730. 16. Wong RJ, Cheung R, Ahmed A. Nonalcoholic steatohepatitis is the most rapidly growing indication for liver transplantation in patients with hepatocellular carcinoma in the U.S. 2014 Jun;59(6):2188-95. 17. Malik SM, Devera ME, Fontes P, Shaikh O, Sasatomi E, Ahmad J. Recurrent disease following liver transplantation for nonalcoholic steatohepatitis cirrhosis. Liver Transpl. 2009; 15:1843–1851. 18. Yalamanchili K, Saadeh S, Klintmalm GB, Jennings JW, Davis GL. Nonalcoholic fatty liver disease after liver transplantation for cryptogenic cirrhosis or nonalcoholic fatty liver disease. Liver Transpl. 2010; 16:431–439. 19. Contos MJ, Cales W, Sterling RK, et al. Development of nonalcoholic fatty liver disease after orthotopic liver transplantation for cryptogenic cirrhosis. Liver Transpl. 2001; 7:363– 373. 20. Ong J, Younossi ZM, Reddy V, Price LL, Gramlich T, Mayes J, Boparai N. Cryptogenic cirrhosis and posttransplantation nonalcoholic fatty liver disease. Liver Transpl. 2001; 7:797– 801. 21. Dureja P, Mellinger J, Agni R, Chang F, Avey G, Lucey M, Said A. NAFLD recurrence in liver transplant recipients. Transplantation. 2011; 91:684–689. 22. Ratziu V, Bonyhay L, Di Martino V, Charlotte F, Cavallaro L, Sayegh-Tainturier MH et al. Survival, liver failure, and hepatocellular carcinoma in obesity-related cryptogenic cirrhosis. Hepatology. 2002; 35:1485–1493.
23. Hassan MM, Curley SA, Li D, Kaseb A, Davila M, Abdalla EK et al. Association of diabetes duration and diabetes treatment with the risk of hepatocellular carcinoma. Cancer. 2010; 116:1938–1946. 24. Bugianesi E, Leone N, Vanni E, Marchesinin G, Brunello F, Carucci P et al. Expanding the natural history of nonalcoholic steatohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology. 2002; 123:134–140. 25. Nair S, Mason A, Eason J, Loss G, Perrillo RP. Is obesity an independent risk factor for hepatocellular carcinoma in cirrhosis? Hepatology. 2002 Jul;36(1):150-5. 26. Marrero JA, Fontana RJ, Fu S, Conjeevaram HS, Su GL, Lok AS. Alcohol, tobacco and obesity are synergistic risk factors for hepatocellular carcinoma. J Hepatol. 2005; 42:218– 224. 27. Ascha MS, Hanouneh IA, Lopez R, Tamimi TA, Feldstein AF, Zein NN. The incidence and risk factors of hepatocellular carcinoma in patients with nonalcoholic steatohepatitis. Hepatology. 2010; 51:1972–1978. 28. Pais R, Lebray P, Rousseau G, Charlotte F, Esselma G, Savier E et al. Nonalcoholic fatty liver disease increases the risk of hepatocellular carcinoma in patients with alcoholassociated cirrhosis awaiting liver transplants. Clin Gastroenterol Hepatol. 2015; 13:992. 29. Dyson J, Jaques B, Chattopadyhay D, Lochan R, Grahan J, Das D et al. Hepatocellular cancer – the impact of obesity, type 2 diabetes and a multidisciplinary team. J Hepatol. 2014; 60:110–117. 30. Mittal S, Sada YH, El-Serag HB, Kanwal F, Duan Z, Temple S et al. Temporal trends of nonalcoholic fatty liver disease-related hepatocellular carcinoma in the veteran affairs population. Clin Gastroenterol Hepatol. 2015;
31. Piscaglia F, Svegliati-Baroni G, Barchetti A, Pecorelli A, Marinelli S, Tiribelli C et al. Clinical patterns of hepatocellular carcinoma (hcc) in non alcoholic fatty liver disease (NAFLD): a multicenter prospective study. Hepatology. 2016; 63:827–838. 32. Stine JG, Shah NL, Argo CK, Pelletier SJ, Caldwell SH, Northup PG. Increased risk of portal vein thrombosis in patients with cirrhosis due to Non-Alcoholic Steatohepatitis (NASH). Liver Transpl. 2015; 21:1016–1021. 33. Tripodi A, Fracanzani AL, Primignani M, Chantarangkul V, Clerici M, Mannucci PM et al. Procoagulant imbalance in patients with nonalcoholic fatty liver disease. J Hepatol. 2014; 61:148–154.
FIGURES LEGENDS: Fig. 1. Frequency (1A) and percentage from the total UNOS population (1B) for NASH and cryptogenic cirrhosis by year at listing. Fig. 2. Frequency of the listed individuals by age. Fig. 3. The average initial MELD score at listing for NASH, cryptogenic cirrhosis (CC), alcoholic cirrhosis (AC) and autoimmune hepatitis (AIH).
Figure 1: Frequency (1A) and proportion (1B) of NASH and cryptogenic cirrhosis by year at listing. A)
B)
Figure 2: Frequency of the listed individuals by age and year of listing.
Figure 3: The average initial MELD score by year of listing for NASH, cryptogenic cirrhosis (CC), alcoholic cirrhosis (AC) and autoimmune hepatitis (AIH).
Table 1A: Characteristics of the four groups; cryptogenic cirrhosis (CC), NASH, alcoholic cirrhosis (AC), autoimmune hepatitis (AIH) for the period 2002-2016.
Characteristic
CC
NASH
p-value AC
p-value
AIH
p-value
7999
11302
.
21714
.
3447
.
Diabetes type 2, N (%)
2008 (28.5 %)
5598 (52.8 %)
<.0001
3320 (16.1 %)
<.0001
527 (16.3 %)
<.0001
Obese, N (%)
3361 (42 %)
7381 (65.3 %)
<.0001
7106 (32.7 %)
<.0001
1278 (37.1 %)
<.0001
BMI, mean(SD)
29.4(5.8)
32.7 (6.1)
<.0001
28.1(5.5)
<.0001
28.6 (6.3)
<.0001
Age, mean (SD)
57.2(10.2)
59 (8)
<.0001
54.2 (9)
<.0001
48.6 (14.3)
<.0001
Female, N (%)
3631 (45.4 %)
5536 (49 %)
<.0001
5344 (24.6 %)
<.0001
2545 (73.8 %)
<.0001
Ethnicity, N (%)
<.0001
<.0001
<.0001
White
5789 (72.4 %)
9300 (82.3 %)
.
16798 (77.4 %)
.
2094 (60.7 %)
.
Black
366 (4.6 %)
167 (1.5 %)
.
697 (3.2 %)
.
598 (17.3 %)
.
Hispanic 1516 (19 %)
1491 (13.2 %)
.
3601 (16.6 %)
.
606 (17.6 %)
.
Asian
247 (3.1 %)
189 (1.7 %)
.
313 (1.4 %)
.
91 (2.6 %)
.
Other
81 (1 %)
155 (1.4 %)
.
305 (1.4 %)
.
58 (1.7 %)
.
18.4 (8.6)
17.8 (8.2)
<.0001
19.8 (9.2)
<.0001
19.4 (9.7)
<.0001
Bacterial peritonitis, N (%)
463 (6 %)
672 (6 %)
0.82
1943 (9.2 %)
<.0001
228 (6.9 %)
0.072
Dialysis twice in prior week, N (%)
521 (6.6 %)
764 (6.8 %)
0.57
1753 (8.2 %)
<.0001
163 (4.8 %)
0.0002
Portal vein thrombosis, N (%)
451 (5.8 %)
756 (6.8 %)
0.007
868 (4.1 %)
<.0001
168 (5.1 %)
0.10
HCC, N (%)
1033 (12.9 %)
2166 (19.2 %)
<.0001
3041 (14 %)
0.01
331 (9.6 %)
<.0001
MELD at listing, mean (SD)
Other characteristics
Characteristic
CC
NASH
Ascites, N (%)
p-value AC
p-value
0.0001
<.0001
AIH
p-value <.0001
Absent
1424 (17.8 %)
2212 (19.6 %)
.
3082 (14.2 %)
.
872 (25.3 %)
.
Slight
4496 (56.3 %)
6028 (53.4 %)
.
11836 (54.6 %)
.
1839 (53.5 %)
.
Moderate 2062 (25.8 %)
3056 (27.1 %)
.
6770 (31.2 %)
.
729 (21.2 %)
.
Encephalopathy, N (%)
0.26
<.0001
<.0001
None
2757 (34.5 %)
3775 (33.4 %)
.
6233 (28.7 %)
.
1491 (43.3 %)
.
1-2
4728 (59.2 %)
6808 (60.3 %)
.
13741 (63.4 %)
.
1683 (48.9 %)
.
3-4
497 (6.2 %)
713 (6.3 %)
.
1714 (7.9 %)
.
266 (7.7 %)
.
The differences between the groups were tested by using logistic regressions for categorical variables, and two-sample t-test for continuous variables. P-values are for hypotheses testing the differences from cryptogenic cirrhosis.
Table 1B. Characteristics of the four groups; cryptogenic cirrhosis (CC), NASH, alcoholic cirrhosis (AC), autoimmune hepatitis (AIH) for the period 2011-2016.
CC
NASH
Characteristic
AC p-value
N
2728
7308
Diabetes type 2, N (%)
850 (31.9 %)
3933 (56 %)
Obese, N (%)
AIH p-value
10856
p-value 1541
<0.01
1928 (18 %)
<0.01
306 (20.2 %)
<0.01
1057 (38.7 %) 4689 (64.2 %)
<0.01
3671 (33.8 %)
<0.01
587 (38.1 %)
0.67
BMI, mean (SD)
29 (5.7)
32.6 (6.2)
<0.01
28.3 (5.5)
<0.01
28.8 (6.4)
0.44
Age, mean (SD)
57.5 (10.8)
59.4 (8)
<0.01
54.2 (9.4)
<0.01
48.7 (14.4)
<0.01
Female, N (%)
1218 (44.6 %) 3498 (47.9 %)
<0.01
2764 (25.5 %)
<0.01
1105 (71.7 %)
<0.01
Ethnicity, N (%)
<0.01 White
1853 (67.9 %) 5910 (80.9 %)
8249 (76 %)
883 (57.3 %)
Black
130 (4.8 %)
110 (1.5 %)
390 (3.6 %)
292 (18.9 %)
Hispanic 611 (22.4 %)
1046 (14.3 %)
1844 (17 %)
296 (19.2 %)
Asian
99 (3.6 %)
139 (1.9 %)
193 (1.8 %)
44 (2.9 %)
Other
35 (1.3 %)
103 (1.4 %)
180 (1.7 %)
26 (1.7 %)
19.5 (8.8)
18.3 (8.4)
<0.01
20.9 (9.6)
<0.01
20.7 (9.9)
<0.01
Bacterial peritonitis, N (%)
194 (7.2 %)
486 (6.7 %)
0.42
1097 (10.3 %)
<0.01
124 (8.1 %)
0.26
Dialysis twice in prior week N (%)
244 (8.9 %)
576 (7.9 %)
0.08
1103 (10.2 %)
0.06
95 (6.2 %)
<0.01
Portal vein thrombosis, N (%)
229 (8.5 %)
577 (8 %)
0.41
537 (5 %)
<0.01
107 (7 %)
0.09
HCC, N (%)
439 (16.1 %)
1577 (21.6 %)
<0.01
1636 (15.1 %)
0.18
164 (10.6 %)
<0.01
MELD at listing, mean (SD)
Other characteristics
Ascites, N (%)
<0.01 Absent
556 (20.4 %)
1541 (21.1 %)
1756 (16.2 %)
416 (27 %)
CC
NASH
Characteristic
AC p-value
Slight
1423 (52.2 %) 3685 (50.4 %)
Moderate 748 (27.4 %)
2079 (28.5 %)
AIH p-value
p-value
5445 (50.2 %)
769 (50 %)
3650 (33.6 %)
354 (23 %)
Encephalopathy, N (%)
<0.01 None
1015 (37.2 %) 2500 (34.2 %)
3338 (30.8 %)
679 (44.1 %)
1-2
1534 (56.3 %) 4324 (59.2 %)
6612 (60.9 %)
734 (47.7 %)
3-4
178 (6.5 %)
901 (8.3 %)
126 (8.2 %)
481 (6.6 %)
The differences between the groups were tested by using logistic regressions for categorical variables, and two-sample t-test for continuous variables. P-values are for hypotheses testing the differences from cryptogenic cirrhosis.
Table 2: Multivariate logistic regression modeling for probability of having NASH diagnosis.
Effect
Lower Upper 95% 95% Odds Confidence Confidence Ratio Limit for Limit for Estimate Odds Ratio Odds Ratio
p-value
Diabetes type 2
4.706
4.505
4.917
<.0001
Obesity
3.476
3.326
3.632
<.0001
Age 60+
2.43
2.327
2.538
<.0001
Female
2.006
1.921
2.095
<.0001
Caucasian
2.315
2.192
2.445
<.0001
The probability of NASH (N=11302) was compared to the rest of the UNOS population (N=126,719) by multivariate logistic regression modeling.
25
Table 3: Characteristics of NASH and cryptogenic cirrhosis matched by age, gender and year of listing.
Characteristics
Level
Cryptogenic cirrhosis
NASH
p-value
N
7201
7201
Diabetes type 2, N (%)
1947 (30.2 %)
3369 (50.7 %)
<.0001
Obese, N (%)
3045 (42.3 %)
4718 (65.5 %)
<.0001
BMI, mean (SD)
29.4 (5.8%)
32.7 (6%)
<.0001
Ethnicity, N (%)
<.0001 White
5196 (72.2 %)
5955 (82.7 %)
Black
311 (4.3 %)
118 (1.6 %)
Hispanic
1393 (19.3 %)
916 (12.7 %)
Asian
228 (3.2 %)
116 (1.6 %)
Other
73 (1 %)
96 (1.3 %)
Portal vein thrombosis, N (%)
427 (6.1 %)
432 (%)
0.67
HCC, N (%)
968 (13.4 %)
1296 (18 %)
<.0001
Other characteristics
The differences between the groups were tested by using a conditional logistic regression for categorical variables, and paired t-test for continuous variables.
Table 4. Characteristics of NASH and cryptogenic cirrhosis stratified by BMI below and above 30.
26
CC(BMI<30)
NASH (BMI<30)
p-value CC(BMI>=30) NASH(BMI>=30) p-value
Number
4638
3921
.
Diabetes type 2, N (%)
1020 (24.6 %)
1861 (50.4 %)
<.0001 988 (34 %)
Age, N (%)
7381
.
3737 (54.1 %)
<.0001
<.0001
<.0001
19-30
157 (3.4 %)
16 (0.4 %)
34 (1 %)
25 (0.3 %)
30-59
2180 (47 %)
1498 (38.2 %)
1766 (52.5 %)
3682 (49.9 %)
60+
2301 49.6 %)
2407 (61.4 %)
1561 (46.4 %)
3674 (49.8 %)
2071 (44.7 %)
1967 (50.2 %)
<.0001 1560 (46.4 %)
3569 (48.4 %)
Female, N (%) Ethnicity, N (%)
HCC, N (%)
3361
<.0001
0.06 <.0001
White
3236 (69.8 %)
3161 (80.6 %)
2553 (76 %)
6139 (83.2 %)
Black
263 (5.7 %)
62 (1.6 %)
103 (3.1 %)
105 (1.4 %)
Hispanic
889 (19.2 %)
552 (14.1 %)
627 (18.7 %)
939 (12.7 %)
Asian
208 (4.5 %)
111 (2.8 %)
39 (1.2 %)
78 (1.1 %)
Other
42 (0.9 %)
35 (0.9 %)
39 (1.2 %)
120 (1.6 %)
584 (12.6 %)
715 (18.2 %)
<.0001 449 (13.4 %)
1451 (19.7 %)
<.0001
The differences between the groups were tested by using logistic regression
Table 5: Characteristics of NASH and cryptogenic cirrhosis stratified presence or absence of diabetes type 2 (individuals with missing values for diabetes were excluded).
27
CC (with type NASH (with type CC (no Type 2 NASH (no p-value DM) 2 DM) 2 DM) Type 2 DM)
p-value
Number
2008
5598
.
Obese, N (%)
988 (49.2 %)
3737 (66.8 %)
Female, N (%)
927 (46.2 %)
2653 (47.4 %)
Age, N (%)
4999
<.0001
1919 (38 %)
3167 (63.4 %) <.0001
0.34
2260 (44.8 %)
2524 (50.5 %) <.0001
0.34
<.0001
19-30
4 (0.2 %)
6 (0.1 %)
182 (3.6 %)
33 (0.7 %)
30-59
862 (42.9 %)
2326 (41.6 %)
2606 (51.6 %)
2534 (50.7 %)
60+
1142 (56.9 %)
3266 (58.3 %)
2260 (44.8 %)
2432 (48.6 %)
Ethnicity, N (%)
HCC, N (%)
5048
<.0001
<.0001
White
1382 (68.8 %)
4525 (80.8 %)
3709 (73.5 %)
4192 (83.9 %)
Black
70 (3.5 %)
50 (0.9 %)
254 (5 %)
105 (2.1 %)
Hispanic
465 (23.2 %)
814 (14.5 %)
880 (17.4 %)
584 (11.7 %)
Asian
64 (3.2 %)
118 (2.1 %)
160 (3.2 %)
61 (1.2 %)
Other
27 (1.3 %)
91 (1.6 %)
45 (0.9 %)
57 (1.1 %)
354 (17.6 %)
1297 (23.2 %)
587 (11.6 %)
740 (14.8 %)
<.0001
The differences between the groups were tested by using logistic regression
<.0001
28
•
About 4% of patients are listed for liver transplant with a diagnosis of cryptogenic cirrhosis (CC)
•
Patients with CC have different clinical characteristics when compared to non-alcoholic steatohepatitis (NASH), alcoholic (AC) or autoimmune hepatitis (AIH) cirrhosis
•
Differences between CC and NASH remained unchanged when matched for age, sex and year of listing, or when stratified by the presence of diabetes mellitus or obesity
•
Based on the significant differences of risk factors, CC should not be equated with the term ‘NASH cirrhosis’.