Seasonal variability in the activity of common chronic liver diseases

Abstract

BACKGROUND

Seasonal variations in flu-like illnesses and vaccinations, vitamin D levels, alcohol intake, and sedentary lifestyles raise the possibility that seasonal variations exist in the severity of immune-mediated, alcohol, and obesity- or dyslipidemia-related chronic liver diseases, respectively.

METHODS

We documented months–seasons in which biochemical evidence of disease activity is greatest in adult patients with common liver disorders. Months–seasons associated with peak liver enzyme levels in patients with largely immune-mediated disorders (autoimmune hepatitis, primary biliary cholangitis [PBC], and primary sclerosing cholangitis), alcoholic liver disease, and non-alcoholic fatty liver disease were documented from a hospital-based, liver diseases outpatient clinic database.

RESULTS

Aside from a spike in the severity of PBC during July (p < .005), no significant associations were found between months–seasons and peak liver enzyme activities in any of these liver disorders.

CONCLUSIONS

These findings suggest that seasonal illnesses or immunizations and vitamin D depletion, alcohol intake, and sedentary lifestyle do not significantly exacerbate common underlying immune-mediated, alcohol, or metabolic liver disorders, respectively.

Introduction

Common causes of chronic liver disease in outpatient liver clinics include non-alcoholic fatty liver disease (NAFLD), alcoholic liver disease (ALD), chronic hepatitis B virus (HBV) and hepatitis C virus (HCV) infections, autoimmune hepatitis (AIH), primary biliary cholangitis (PBC), and primary sclerosing cholangitis (PSC) (1). Of these, immune-mediated liver injury is thought to play an important role in the pathogenesis of at least three disorders (AIH, PBC, and PSC). Thus, seasons in which immune system reactivity might be enhanced by factors such as influenza illness or vaccinations and vitamin D deficiency would be expected to be associated with increased disease activity. In the case of ALD, clinical experience suggests that it is more common or severe during the winter season when alcohol consumption tends to increase during the holiday period. A more sedentary lifestyle during the winter months results in increased serum cholesterol values, which could result in worsening of liver enzymes in patients with NAFLD (2). HBV and HCV, however, would not be expected to demonstrate significant seasonal variations in disease activity. Patient gender should also be considered with respect to seasonal changes in liver disease activity. Specifically, immune-mediated diseases in general tend to be more common among women (3). Thus, seasonal variations may be more apparent in female patients with these disorders.

The clinical relevance of documenting associations between specific liver disorders and certain seasons relates to diagnostic considerations and opportunities to adjust medications vis-à-vis primary prophylaxis. Such data also inform the debate as to whether immunizations should be undertaken in patients with immune-mediated liver disorders.

The purpose of this study was to document the months–seasons when peak disease activity occurs in patients with common liver disorders and determine whether gender influences the findings.

Methods

The Phil and Ellie Kives Clinical Database is a computerized database maintained by the Section of Hepatology at the University of Manitoba in Winnipeg, Manitoba. The database was accessed to identify all patients older than age 18 years referred to the outpatient liver program at the Health Sciences Centre (HSC), a tertiary care centre, for evaluation and management of the following chronic disorders: NAFLD, ALD, HBV, HCV, AIH, PBC, and PSC. Chronicity for each disorder was defined as the presence of disease or viral persistence beyond 6 months. Diagnoses were based on the following criteria:

1. NAFLD: histology or radiologic imaging in keeping with fatty infiltration of the liver and no history of excess alcohol intake or use of medications associated with fatty infiltration of the liver and negative testing for HCV;

2. ALD: self-reported weekly alcohol intake in excess of 21 units per week for men and 14 units per week for women;

3. HBV: positive serologic testing for hepatitis B surface antigen (HBsAg);

4. HCV: positive testing for antibody to HCV and HCV core antigen, or HCV-RNA;

5. AIH: compatible histologic findings or elevated serum IgG levels and positive serologic testing (titre ≥ 1:80) for anti-nuclear (ANA) and smooth muscle (SMA) antibodies;

6. PBC: at least two of the following findings: cholestatic liver enzyme abnormalities, compatible histology, and positive serologic testing (titre ≥ 1:40) for mitochondrial (AMA) antibody; and

7. PSC: diagnostic radiologic imaging or histologic features in the absence of secondary causes of sclerosing cholangitis.

Patients were excluded from analysis if follow-up was less than 6 months, more than one liver disease was diagnosed, or follow-up visits had not been scheduled for each of the four seasons.

Seasons were defined as spring (March–May), summer (June–August), fall (September–November), and winter (December–February) inclusive. In Winnipeg, influenza season is considered to be present during November–February and vaccinations are advised during October–January.

Data regarding ethnicity, rural versus urban residence, treatment with immunomodulants, antivirals, or vitamin D were not available, nor were data on whether patients had been infected or vaccinated against influenza or other non-hepatotropic pathogens.

Peak disease activity was defined as the highest serum alanine aminotransferase (ALT) value recorded during the duration of follow-up in patients with NAFLD, HBV, HCV, and AIH. In patients with ALD, serum aspartate aminotransferase (AST) values were used. In patients with PBC or PSC, serum alkaline phosphatase (ALP) levels served as the indicator of disease activity.

All testing was performed by accredited clinical laboratories at the HSC or, in the case of viral testing, the Cadham Provincial Laboratory.

The study protocol was approved by the University of Manitoba’s conjoint human ethics committee.

Statistics

Continuous variables are reported as means and standard deviations, and categorical variables are reported as percentages. To assess for quantitative differences in the monthly and seasonal distribution of visits with peak disease activity, a Poisson regression analysis was performed using NCSS Statistical Software (version 11; NCSS LLC, Kaysville, UT).

Results

Table 1 provides the demographic findings, number of clinic visits per season, total number of clinic visits, and duration of follow-up for each liver disease. Overall, patient visits were equally distributed throughout the year, regardless of the underlying liver disease (approximately 25% per season). Mean follow-up was in excess of 4 years for each liver condition (range 4.4–10.3 years).

Table 1:

Demographics and distribution of clinic visits in patients with various liver diseases

Clinic visits, n (%)
	N	Age, y, mean (SD)	Men, n (%)	Spring	Summer	Fall	Winter	Total visits	Follow-up, y, mean (SD)
NAFLD	1,247	50 (13)	600 (48)	3,436 (27)	2,836 (22)	3,305 (26)	3,042 (24)	12,619	4.4 (3.8)
ALD	162	53 (11)	106 (65)	625 (28)	502 (22)	591 (26)	554 (24)	2,272	4.5 (4.1)
HBV	1,256	39 (14)	699 (56)	5,077 (27)	4,363 (23)	5,278 (28)	4,385 (23)	19,103	6.9 (5.9)
HCV	2,299	44 (12)	1,400 (61)	13,780 (26)	12,443 (24)	13,617 (26)	12,801 (24)	52,641	7.7 (5.6)
AIH	203	47 (16)	42 (21)	1,896 (26)	1,749 (25)	1,801 (26)	1,656 (23)	7,102	9.5 (7.1)
PBC	146	55 (11)	10 (6.8)	985 (27)	832 (23)	990 (27)	796 (22)	3,603	10.3 (6.2)
PSC	106	46 (16)	74 (70)	1,007 (25)	987 (24)	1,057 (26)	1,014 (25)	4,065	9.8 (7.1)

NAFLD = non-alcoholic fatty liver disease; ALD = alcoholic liver disease; HBV = hepatitis B virus;

HCV = hepatitis C virus; AIH = autoimmune hepatitis; PBC = primary biliary cholangitis; PSC = primary sclerosing cholangitis

Figure 1 indicates when (month and season) ALT levels or, in the case of ALD and PBC and PSC, respectively, AST levels and ALP levels, peaked during the entire follow-up period. No overt monthly or seasonal increases in peak ALT values were found among patients with AIH (Figure 1a). Patients with PBC had an abrupt increase in the prevalence of peak ALP levels during July that was significant (p < .005) compared with the aggregated prevalence rate for the remaining months (Figure 1b). There was also a tendency toward bimodal increases in disease activity during the summer and winter seasons that did not reach statistical significance. This pattern was not observed in the other possibly immune-mediated disorders, AIH and PSC (Figures 1a and 1c, respectively).

Figure 1:

Distribution of the appropriate peak liver enzyme levels (ALT, AST, or ALP) during the months and seasons of clinic visits. The increase in number of peak ALP levels during July in PBC patients was significant (p < 0.005) when compared with the aggregated results of the remaining months.

ALT = alanine aminotransferase; AST = aspartate aminotransferase; ALP = alkaline phosphatase; AIH = autoimmune hepatitis; PBC = primary biliary cholangitis; PSC = primary sclerosing cholangitis; HBV = hepatitis B virus; ALD = alcoholic liver disease; NAFLD = non-alcoholic fatty liver disease; HCV = hepatitis C virus

Contrary to clinical impressions, there was no striking increase in the prevalence of peak AST values during the winter months in patients with ALD (Figure 1e). Also contrary to predictions, peak ALT values in patients with NAFLD did not occur more frequently during the winter season (Figure 1f). As predicted, peak ALT values in patients with HBV and HCV were evenly distributed throughout the year (Figure 1d and 1g).

The findings for immune-mediated and other causes of chronic liver disease remained unchanged when female patients were considered separately (data not shown).

Discussion

The results of this study do not support the hypothesis that immune-mediated liver diseases are likely to undergo exacerbations during seasons when the immune system is boosted by influenza infections or vaccinations or when vitamin D levels are lowest. They also do not support the clinical impression that ALD and NAFLD disease activity is greatest during the December holiday season and winter, respectively. As predicted, HBV and HCV disease activity do not demonstrate seasonal variability.

There are a paucity of reports describing seasonal variations in immune-mediated liver diseases. Indeed, in the only article relevant to this topic published to date, McNally et al did not identify increases in the rates of diagnosing PBC during the fall or winter months (4). Of interest, they did document a marked peak in the diagnosis of PBC during June, a finding similar to that shown by our July PBC disease activity data. The proposed explanation for this finding relates to the increase in Escherichia coli infections, a pathogen implicated in the pathogenesis of PBC, that occurs during the summer months (5).

Although the absence of AIH and PBC flares during late fall and winter does not support the hypothesis that exogenous immune stimulation enhances immune-mediated liver disease activity, it also does not eliminate that possibility. Indeed, the majority of these patients were receiving immunosuppressive and bile salt therapies, which effectively decrease the likelihood of exacerbations occurring. Thus, more definitive data would have been derived from a natural history study in which untreated patients with these conditions were followed for prolonged periods of time, a study that would not be ethically acceptable. An alternative approach, limiting analyses to the date of patients’ initial visits when disease activity would presumably be highest and treatment not yet instituted, was also not feasible as referral wait times were too variable. Another argument that the findings do not refute the hypothesis relates to the database used for the study, which does not capture which patients were infected or immunized against influenza (or other pathogens) and who had low serum vitamin D levels. Thus, it is conceivable that these patients constituted only a small percentage of the total immune-mediated disease study populations.

It is of interest to note the one immune-mediated disorder not routinely treated is PSC, which was also not associated with a spike in disease activity during later fall or winter seasons. However, it has been suggested that although PSC is initiated by immune hyperreactivity, the subsequent course of the disease is determined by the extent of retained toxic bile acids (6).

According to national alcohol sales data, alcoholic beverage consumption significantly increases during the December holiday season, yet peak ALD activity (as reflected by AST values) was not higher in December or January, the months when consumption would be expected to be highest (7). Similar results (albeit describing no festive season increases in alcohol-induced pancreatitis rather than hepatitis) were described by Bertilsson et al in Sweden (8). There are two probable explanations for this finding. First, because acute alcoholic hepatitis is a life-threatening illness that often requires hospitalization, patient presentation to an outpatient liver clinic during an ALD flare would be less common. Second, the drinking pattern of ALD patients attending outpatient liver clinics may differ from that of seasonal drinkers in that the former cohort tend to drink alcohol to excess throughout the year with no significant seasonal variability (authors’ unpublished observations).

Ockene et al have reported the estimated prevalence of hypercholesterolemia increases by 22% during the winter months, largely as a result of a more sedentary lifestyle (2). Because dyslipidemia and a sedentary lifestyle are well-recognized risk factors for NAFLD, one might expect peak disease activity to occur during the winter season. However, that was not the case in this study, in which peak ALT values in patients with NAFLD were evenly distributed throughout the year. A likely explanation for this finding is the distinction between simple steatosis and the necro-inflammatory disease associated with NAFLD (steatohepatitis). Dyslipidemia and a sedentary lifestyle contribute to hepatic steatosis but not necessarily the severity of steatohepatitis, which can be captured by serum aminotransferase levels (9,10).

The clinical implications of this study are largely limited to concerns regarding the safety of vaccinations in patients with immune-mediated liver disorders (11,12). The data provided do not describe increases in AIH, PBC, or PSC peak disease activity during the months when influenza vaccinations would be offered, but there are too many confounding variables to categorically state that such vaccinations are not associated with an increased risk of disease exacerbation. Thus, in our opinion, when indicated, influenza vaccinations should be provided to patients with chronic liver disease, but patient follow-up within 1–3 months after vaccination is warranted to detect evidence of disease activation.

In conclusion, none of the common liver diseases studied demonstrated seasonal variations in maximal disease activity. These findings do not support the hypothesis that exogenous, seasonal factors influence the course of these disorders.

Ethics Approval:

The study protocol was approved by an ethics committee and the ethics certificate information is available from the authors upon request.

Informed Consent:

Informed consent was obtained from the patient(s).

Registry and Registration No. of The Study/Trial:

N/A

Funding:

None to declare

Disclosures:

None to declare