Clearing the Smoke in Chronic Liver Diseases

In the 1930s, serious concerns about the health risks of cigarette smoking (CS) began to surface. During subsequent decades, scientific reports linking CS and specific ailments rapidly accumulated [1,2], but it was not until 1964 that the Surgeon General’s Advisory Committee on Smoking and Health finally acknowledged that CS was linked to specific diseases and to increased mortality. Today, the evidence is robust: The adverse effects of CS on several cancer outcomes as well as on cardiovascular and respiratory disease are established [3, 4]. Although in the US the prevalence of CS has been decreasing [5], the overall worldwide prevalence is steadily rising. Independent of prevalence rates, the absolute number of smokers everywhere keeps increasing due to population growth.

The case against CS in patients with chronic liver diseases (CLD) has been highlighted recently as data reporting hepatic injury due to smoking have emerged [6,7]. A role for CS in CLD was first suggested by two studies in the mid 1990s [8,9]. By now, CS has been clearly identified as a risk factor for hepatocellular carcinoma in CLD [10–11], but its effect on histological activity or fibrosis progression in CLD still needs further characterization. Published studies have been limited predominantly by cross-sectional and retrospective study designs, and lack of supportive experimental data. Nonetheless, the evidence from clinical studies consistently indicates that CS may accelerate liver disease progression in patients with chronic hepatitis C and B and those with primary biliary cirrhosis [Table 1] [8, 12–17]. CS also appears to exacerbate the liver injury in alcoholic liver disease [8,9]. Regarding nonalcoholic fatty liver disease (NAFLD), data supporting a potential role of CS are just recently surfacing.

Table 1.

Studies in patients with CLD investigating the association between smoking and histological severity of disease.

Author REF	Condition	n	Association with Smoking
			Histological Disease Activity	Severity of Liver Fibrosis
Dev et al. 13	HCV	170	-	Yes
Pessione et al. 15	HCV	310	Yes	Yes
Hezode et al. 14	HCV	244	Yes	No
Tsochatzis et al. 12	HCV	176	-	Yes
Tsochatzis et al. 12	HBV	85	-	No
Yu et al. 16	HBV	1506	-	Yes
Corrao et al. 8	Cirrhosis	115	-	Yes
Zein et al. 17	PBC	269	-	Yes

“-“ indicates not reported or not applicable

Delineating the effect of CS in NAFLD is essential because of the vast number of subjects that may benefit from risk factor modification. Over 30 million adults in the United States have NAFLD [18] and approximately 8 million may have nonalcoholic steatohepatitis (NASH) and hence a significant risk of developing cirrhosis, its complications and liver related mortality [19–20]. Unfortunately, no beneficial therapy can be recommended yet for patients with NASH. Therefore, the identification of modifiable risk factors that may impact disease progression, by itself important, is even more critical.

Although the exact mechanisms by which CS worsens CLD are unknown, knowledge on the effects of CS in other organs together with available liver related clinical and experimental data provide insight into the pathophysiology of CS induced injury in NAFLD. These include enhanced insulin resistance (IR), altered lipid metabolism, chronic hypoxia, increased oxidative stress, and enhancement of inflammatory cytokines. Because insulin resistance influences histological severity in NAFLD [21, 22], CS may worsen NAFLD through its effect on IR, glucose intolerance and diabetes development [23,24]. Changes in lipid metabolism induced by CS may also aggravate NAFLD. Experimental studies have shown that CS aggravates the hepatic steatosis elicited by high fat diet in mice [25,26] via enhanced fatty acid synthesis through inhibition of AMP-activated protein kinase phosphorilation in liver tissue [25]. Chronic hypoxia, a hallmark side effect of CS, induces steatosis, liver inflammation and fibrosis in mice [27–29]. CS also causes oxidative stress [30], a recognized mechanism of injury in NAFLD [31]. Mice on an ethanol diet develop increased hepatocellular injury when exposed to CS and have increased CYP2E1, known to play a role in oxidative injury in NAFLD [28]. Finally, CS may worsen NAFLD through enhancing pro-inflammatory cytokines, such as tumor necrosis factor alpha, that are known to play a key role in NAFLD [32].

In this issue of Hepatology, Azzalini et al. provide novel evidence suggesting that CS exacerbates liver injury in NAFLD. In their study, control and obese Zucker rats were divided into smoker and nonsmoker groups based on controlled exposure to CS. Exposure to CS increased ALT levels, and increased hepatocellular ballooning and lobular inflammation in the liver of obese rats, whereas significantly smaller changes were noted in control rats. The authors showed that CS increased oxidative stress, and hepatocyte apoptosis in obese rats. In addition, CS exposure induced TIMP-1 and pro-collagen alpha2 synthesis at the transcription level. The effects of CS on ALT level, histological hepatic injury, and expression of fibrogenic genes occurred only with long-term exposure (4 weeks) to CS and did not occur with a shorter exposure (5 days). This indicates that the aggravating effects of CS on NAFLD are the result of prolonged exposure to CS. The results of Azzalini et al provide some elucidation of the underlying mechanisms involved in CS related liver injury not only in NAFLD but potentially also in other types of CLD. The value of the findings of experimental studies such as this study by Azzalini et al. is further underscored by the fact that it would be impossible to conduct a prospective randomized controlled study of the effects of CS in humans with chronic liver disease. Even case-control or cohort studies attempting to isolate the effect of CS on CLD prove to be challenging in the clinical setting and are limited by multiple confounders.

Nonetheless, the study by Azzalini et al has some limitations. As the authors acknowledge, the 4 week study design may not have allowed the occurrence of some effects of long-term CS that may have been observed with longer exposure. In this regard, although this study did not show changes in IR or lipid profiles of rats exposed to CS, it is possible that longer exposures to CS may adversely affect these metabolic factors [23–26]. Interestingly, the observation of increased hepatic injury induced by CS in the absence of worsening IR, together with the knowledge that CS also worsens IR [23,24] and IR in turn worsens NAFLD [21,22] suggests that that the deleterious effect of CS in human NAFLD, and in CLD in general, may engage several pathways. The 4 week study design may also have resulted in an inability to demonstrate increased hepatic fibrosis. A second study limitation also relates to the assessment of hepatic fibrosis. Although CS up-regulated the expression of genes involved in fibrogenesis in obese rats, this was not associated with evident development of increased liver fibrosis. However, the absence of leptin receptor in the Zucker rat model may have influenced these results. Evidence for this possibility is that whereas a methionine-choline-deficient diet induces steatohepatitis and increased oxidative stress in Zucker rats, the occurrence of increased neovascularization, hepatic expression of vascular endothelial growth factor, and liver fibrosis development are restricted in this model [33]. Therefore, although conclusions cannot be made regarding the lack of increased angiogenesis and liver fibrosis development reported in the study by Azzalini et al, the CS induced worsening of histological injury and apoptosis support the concept that CS may cause fibrosis progression in NASH [34]. Additional studies in different animal models are needed to clarify and substantiate the profibrogenic effects of CS in NALFD suggested by gene up-regulation. Finally, this study demonstrated that CS increases hepatic apoptosis in the liver of obese rats. This is of great importance given the crucial role of apoptosis in NAFLD progression. However, the exact apoptotic pathways involved were not identified. A key observation was that CS decreased caspase-3-driven-apoptosis in both obese and control rats, suggesting that CS induces a caspase-3 independent pathway in NAFLD. Further studies are warranted to elucidate the exact mechanism behind CS induced apoptosis in NAFLD.

To sum up, the study by Azzalini et al. demonstrates that CS worsens liver injury in a rat model of obesity-related NAFLD. These results, together with other experimental data [25–29], provide compelling evidence that CS exacerbates NAFLD. Similarly, clinical studies in CLD have consistently indicated that CS aggravates liver injury in humans [8,9,11–17]. There are very few published studies on the effects of CS in human NAFLD. Although two studies did not find an association between CS and the presence of NAFLD in the general population [35, 36], only one published clinical study has looked at the possible effect of CS in patients already identified as having NAFLD [37]. In that study, CS exposure was associated with increased ALT.

Future studies are needed to better elucidate the mechanistic aspects of the effects of CS in NAFLD, and to better characterize the role of CS in human NAFLD. Nonetheless, this study provides one more reminder that there is already ample experimental and clinical evidence consistently pointing in the same direction: That CS aggravates liver injury in CLD. It is time to take the harmful effects of CS in CLD more seriously. As hepatologists, we need to incorporate the intake of a more thorough smoking history during our evaluations, educate our patients on the effects of this modifiable risk factor on liver injury, and strongly recommend smoking cessation in all patients with CLD.

Abbreviations

CS

Cigarette Smoking

CLD

Chronic Liver Diseases

NALFD

Nonalcoholic fatty liver disease

NASH

Nonalcoholic steatohepatitis