Risk Factors for Delta Hepatitis in a North American Cohort: Who Should be Screened?

Abstract

Background:

The AASLD recommends hepatitis D virus (HDV) screening in certain high-risk groups, however, the effectiveness is unknown.

Methods:

A study of North American (NA) patients with hepatitis B (HBV) referred to the NIH was performed to identify risk factors associated with HDV infection. Active HDV was “confirmed” by serum HDV RNA or histologic HDV antigen staining.

Results:

652 were studied of which 91 were HDV “confirmed.” Independent risk factors for HDV included: IVDU, HBV-DNA <2000 IU/mL, ALT >40 U/L, and HDV endemic country of origin.

Conclusions:

NA patients with HBV and significant risk factors should be screened for HDV.

INTRODUCTION

Hepatitis D virus (HDV) is a defective ribonucleic acid (RNA) virus that requires the presence of hepatitis B virus (HBV) to be infectious in humans, specifically the hepatitis B surface antigen (HBsAg).¹ Chronic HDV infection results in rapid progression of disease with up to 80% of patients developing cirrhosis within 5–10 years and is associated with an increased risk of mortality and hepatocellular carcinoma compared to HBV monoinfection.^{2, 3} Despite the seriousness of this disease, accurate global and country specific prevalence of HDV infection remains elusive. Historical estimations of global disease burden suggest ~15–20 million, however this may be higher.⁴ In the United States (U.S.), accurate estimations are equally lacking with data mostly pertaining to intravenous drug users (IVDU).^5–7 A potential explanation may be because the lack of clarity as to which patients with chronic HBV should be screened for HDV.⁵

The American Association for the Study of Liver Diseases (AASLD) has recommended HDV antibody screening (HDAb) in patients who are HIV-positive, IVDU, MSM (men who have sex with men), at risk for sexually transmitted diseases, or are immigrants from an HDV endemic country (HEC).⁸ Testing should also be considered in patients with low HBV-DNA levels and elevated ALT levels with or without anti-nucleo(s)tide therapy. These recommendations are based loosely on HDV risk factors reported from dated epidemiological studies mostly from Europe and the known HBV suppressive effect of HDV.^{2, 6} Nonetheless, there is a paucity of evidence supporting this approach for HDV screening, which likely has resulted in poor compliance with societal guidance; thus, most HBV patients are not screened for HDV.⁵

At the National Institutes of Health Clinical Center (NIH CC), HDV patients are routinely referred for clinical management and evaluation for experimental HDV therapies. Our primary aim was to identify risk factors associated with active HDV infection among patients with chronic HBV and test if current AASLD recommendations are sufficient and constitute the correct approach.

METHODS

We conducted a retrospective, cross-sectional cohort study of HBsAg-positive adult patients seen at the NIH CC from January 2000 through April 2019. All patients underwent a comprehensive hepatologic evaluation including serological testing, imaging, and liver biopsy as clinically indicated. Status of HDV exposure was classified according to their HDAb status. Patient were considered “exposed” if they tested positive for HDAb. HDV infection was considered “confirmed” with the presence of serum HDV-RNA and/or HDAg staining in liver tissue. Those considered “exposed” to HDV but did not have detectable serum HDV-RNA were considered HDV “cleared”. For HDV “exposed” patients that did not have an HDV-RNA result on file, stored serum samples from the time of initial evaluation were tested for the presence of HDV RNA. Country of origin was dichotomized into two groups: HDV non-endemic and endemic (Table 1).^{4, 8} Information regarding data collection, cirrhosis classification, viral diagnostics, and statistical analysis can be found in the supplementary section.

Table 1.

Countries considered endemic

Endemic countries
Benin
Cameroon
Chad
Congo
Egypt
Gambia
Ghana
Guinea
India
Italy
Ivory Coast
Liberia
Mongolia
Nigeria
Pakistan
Romania
Russia
Senegal
Sierra Leone
Somalia
Togo
Turkey
Zimbabwe

RESULTS

Of 3,373 patients tested for HBsAg, 652 were HBsAg+ and 588(90%) patients were tested for HDAb. 113(19%) were HDAb+ and were considered HDV “exposed”. 91(80.5%) of those “exposed” were confirmed to have chronic HDV infection. In the HDV “exposed” cohort; 9.8% HBeAg+, 79.5% HBeAb+, 65.5% from an HEC, 10.6% IVDU, and 12.4% were on anti-nucleo(s)tide therapy (Table 2). HDV “exposed” patients were more likely from an HEC or be IVDU (p<0.0001) compared to “un-exposed” patients. “Exposed” patients had higher baseline aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), gamma-glutamyl transferase (GGT), prothrombin time, APRI and FIB-4 and lower albumin and platelet count compared to HDV “un-exposed.”

Table 2.

Baseline characteristics of all HBsAg positive patients

	HBV positive All HBsAg (+) (n = 652)	HDV not checked HDAb (NC) (n = 64)	HDV un-exposed HDAb (−) (n = 475)	HDV exposed HDAb (+) (n = 113)	P ¥
Age, years	40 (13.4)	39 (15.4)	40 (13.5)	41 (12.0)	0.58
Gender
Male	385 (59.1)	40 (62.5)	274 (57.7)	71 (62.8)	0.32
Female	267 (40.9)	24 (37.5)	201 (42.3)	42 (37.2)
Race
White	145 (22.2)	33 (51.56)	74 (15.6)	38 (33.6)	<0.0001
Black	178 (27.3)	4 (6.25)	152 (32)	22 (19.5)	0.009
Asian	293 (44.9)	25 (39.06)	217 (45.7)	51 (45.1)	0.92
Other	36 (5.6)	2 (3.13)	32 (6.7)	2 (1.8)	0.04
HCV (+), %	1.3	2.6 *	1.3	1.3	0.17
HIV (+), %	2.5	6.5 *	2.7	0.9	0.28
HBeAg (+), %	19.4	32.8	20.1	9.8	0.01
HBeAb (+), %	73.1	64.9	72.6	79.5	0.14
HDV endemic country of origin, %	32.8	2.0 *	28.4	65.5	<0.0001
IVDU (+), %	2.8	3.1	0.8	10.6	<0.0001
Anti-nucleo(s)tide therapy, %	8.1	14.1	6.3	12.4	0.028
Laboratory
ALP (U/L)	80.6 (43.1)	95.3 (70.8)	76.6 (37.9)	88.7 (40.6)	0.003
AST (U/L)	48.8 (65.6)	37.8 (26.2)	44.2 (65.2)	74.0 (76.4)	<0.0001
ALT (U/L)	71.0 (109.6)	52.6 (58.8)	65.4 (112.8)	104.8 (112.2)	0.0009
Total bilirubin (mg/dL)	0.8 (0.8)	0.7 (1.1)	0.7 (0.5)	0.9 (1.4)	0.123
GGT(U/L)	45.7 (67.0)	32.1 (22.1) *	42.5 (67.1)	63.8 (73.4)	0.004
Albumin (g/dL)	4.2 (0.4)	4.2 (0.5)	4.3 (0.4)	4.1 (0.4)	0.008
PT (seconds)	13.8 (1.2)	13.8 (1.9) *	13.7 (1.0)	14.3 (1.4)	<0.0001
PLT (K/μL)	207.2 (68.7)	217.1 (68.7)	213.2 (66.3)	176.5 (70.8)	<0.0001
APRI	0.9 (1.6)	0.6 (0.5)	0.7 (1.3)	1.7 (2.6)	<0.0001
FIB-4	1.5 (2.1)	1.4 (1.9)	1.3 (1.8)	2.4 (3.0)	<0.0001

^¥Comparison made between HDV un-exposed: HDV Ab (−) with HDV exposed: HDV Ab (+) patients

Values expressed as mean (standard deviation) or n (%) unless otherwise stated.

^*= >10% missing values

Abbreviations: HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; NC, not checked; HDAb, hepatitis D virus antibody; HDV, hepatitis D virus; HBeAg, hepatitis B e-antigen; HBeAb, hepatitis B e-antibody; IVDU, intravenous drug user; ALP, alkaline phosphatase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; PT, prothrombin time; PLT, platelet count; APRI, AST-to-platelet ratio index; FIB-4, fibrosis-4 score

There were no significant differences in demographics or HBV related characteristics between HDV “confirmed” and “cleared” patients (Supplemental Table 1). However, HDV “confirmed” patients had higher AST, ALT, GGT and APRI, but lower albumin and platelets reflecting more advanced liver disease. HBV genotyping was assessed in 42% of the HBsAg+ patients (Supplemental Table 2 & 3) with the predominant genotype being HBV GT-D.

Significant predictors of HDV infection on univariate analysis included: IVDU, HBV-DNA <2000 IU/mL, origins from an HEC, ALT >40 U/L, ongoing anti-nucleo(s)tide therapy, and HBeAg negativity (Table 3). On multivariate analysis, significant predictors of HDV infection included: IVDU (OR:25.2 (95% CI:4.0–161.4)), serum HBV-DNA <2000 IU/mL (OR:7.8 (95% CI:3.6–17.1)), ALT >40 U/L (OR:7.4 (95% CI:3.9–14.1)), and origins from an HEC (OR:5.8 (95% CI:3.1–10.8)).

Table 3.

Univariate and multivariate analysis of the risk factors associated with HDV infection among HBsAg positive patients who were checked for HDV Ab (n = 588)

Factor	Crude OR	(95% CI)	P	Adjusted OR*	(95% CI)	P
Age >= 40 (years)	0.9	(0.6 – 1.4)	0.78	0.6	(0.4 – 1.2)	0.14
Gender - Male	1.2	(0.8 – 1.9)	0.32	0.6	(0.3 – 1.1)	0.08
HIV	0.3	(0.04 – 2.6)	0.3	0.6	(0.04 – 9.3)	0.74
HCV	0.7	(0.09 – 6.0)	0.76	0.2	(0.01 – 4.3)	0.33
HBeAg positivity	0.4	(0.2 – 0.8)	0.013	0.5	(0.1 – 1.8)	0.29
HBeAb positivity	1.5	(0.9 – 2.4)	0.14	0.6	(0.2 – 1.6)	0.26
HBV-DNA below 2000 IU/mL	6.1	(3.3 – 11.2)	<0.0001	7.8	(3.6 – 17.1)	<0.0001
ALT > 40 U/L	4.3	(2.7 – 6.9)	<0.0001	7.4	(3.9 – 14.1)	<0.0001
IVDU	14.0	(4.4 – 44.2)	<0.0001	25.2	(4.0 – 161.4)	0.0007
HDV endemic country of origin	4.8	(3.1 – 7.4)	<0.0001	5.8	(3.1 – 10.8)	<0.0001
Anti-nucleo(s)tide therapy	2.1	(1.1 – 4.1)	0.03	2.6	(0.8 – 8.4)	0.12

Values expressed as mean (standard deviation) or n (%); Patients in whom HDAb was not checked were excluded.

^*Adjusted for ALT > 40 IU/L.

Abbreviations: HIV, human immunodeficiency virus; HCV, hepatitis C virus; HBeAg, hepatitis B e-antigen; HBeAb, hepatitis B e-antibody; HBV, hepatitis B virus; ALT, alanine aminotransferase; IVDU, intravenous drug user, HDV, hepatitis D virus

DISCUSSION

Given the minimal description of HDV disease characteristics in the U.S., we examined our referral population in this study and highlight risk factors that can aid in the diagnosis of active HDV infection. We found that risk factors for active HDV infection include a history of IVDU, baseline serum HBV-DNA <2000 IU/mL, ALT >40 U/L, and origination from an HEC. Our results support the current AASLD recommendations on whom to screen. Therefore, HBV patients who have at least one of these risk factors should be screened for HDV infection with serum HDAb testing.

One of the major strengths of this study is that we explored a large cohort of HBV patients in which most patients (90.2%) were tested for HDAb and then underwent confirmation of HDV infection with serum HDV-RNA PCR testing or HDAg staining on histology. Prior studies have relied solely on HDAb testing for diagnosing HDV infection which lacks specificity for active HDV infection and only provides evidence of prior exposure to HDV which may have cleared.⁹ The main limitation of our study is its retrospective nature and that our cohort is subjected to referral bias, due to the nature of the NIH. The prevalence rate of HDV in our study is higher than most studies (14%) investigating primarily non-IVDU HBsAg positive population.^{5, 6, 10–13} Thus, the true prevalence of HDV in the U.S. should not be interpreted from this study and still requires further investigation. However, our main goal was to identify risk factors for active HDV infection to provide guidance on which HBV patients need to be screened for HDV. Widespread adoption of our findings will hopefully lead to better HDV epidemiologic studies.

In conclusion, this study shows that HBsAg positive patients in the U.S. should be screened for HDV with serum HDAb testing especially if they have one of the following risk factors: a history of IVDU, origination from an HDV endemic country, or laboratory findings of an ALT >40 U/L or HBV-DNA <2000 IU/mL. Widespread awareness and adoption of this practice is paramount as the diagnosis of HDV carries a significant alteration in prognosis and management strategies.

Abbreviations:

AASLD

American Association for the Study of Liver Diseases

AST

aspartate aminotransferase

ALT

alanine aminotransferase

ALP

alkaline phosphatase

HBV

hepatitis B virus

HDAb

hepatitis D antibody

HDV

hepatitis D virus

GGT

gamma-glutamyl transferase

HEC

HDV endemic country

IVDU

intravenous drug users

MSM

men who have sex with men

NIH CC

National Institutes of Health Clinical Center

RNA

ribonucleic acid

U.S

United States