JournalofHepatology, 1992; 14: 88-93 @ 1992Elsevier Science Publishers B.V. Au rights reserved. 016%8278/92/$03.50
88 HEPAT 00902
Frequency of antibody to he atitis C virus in asymptomatic negative chronic active hepatitis*
Albert J. Czaja, Howard I!. Taswell, Jorge Rakela and Carol Schimek HepatobiiiaryUnit, Division of Gastroenterology,and Section of Transfuron Medicine, Division ofLaborarov Medicine, Mayo Clinic and Mayo Medical School, Rochester, MN, United Statesof America
(Received 8 October 19W)
To determine the frequency of antibodies to hepatitis C virus in asymptomatic patients with HBsAg-negative chronic active hepatitis,sera from 30 consecutive patients with few or no symptoms of liver disease were tested by an enzyme immunoassay. The reactivity of antibodies detected by enzyme immup.oassay against hepatitis C virus encoded antigens was determined by recombinant immunoblot assay. Antibodies were detected in 11 of the 30 patients (37%) and eight of the seropositive SF-ra(73%) were reactive by recombinant hmUnGblGt assay. Nonreactive patients were weakly positive by enzyme immunoassay (sample/cutoff ratio , G1.9) in contrast to reactive patients (sample/cutoff ratio, 26.3). The prevalence of immunoserologic markers was similar in patients with and without antibodies (78 vs. 87%) but high titers (31:160) were more common in seronegative patients (53 vs. al%). Additionally, seronegative patients had smooth muscle antibodies (83 vs. %5%, p < 0.05) and concurrent extrahepatic immunologic diseases (37 vs. 9%) more commonly than seropositive counterparts. We conclude that asymptomatic patients with HBsAg-negative chronic active hepatitis frequently have antibodies to hepatitis C virus. These antibodies commonly react to specific viral antigens, especially if the enzyme immunoassay is strongly positive. Seropositive patients infrequently have concurrent immunologic disorders or smooth muscle antibodies. Immunoserologic markers lack diagnostic specificity except in high titer.
Prednisone alone or in combination with azathioprine is effective in amelioratir,g symptoms, improving biochemical and histological findings, and enhancing survival in patients with severe HBsAg-negative (autoimmune) chronic active hepatitis (CAH) (l-3). The majority of patients with HBsAg-negative CAH, however, do not have severe, incapacitating, or immediately life-threatening disease and many are asymptomatic (4,5). A favorable benefit/risk ratio for corticosteroid treatment has not been established in these patients (6,7) and currently, in the absence of confident treatment guidelines, they are managed expectantly (6-9). Several studies have supported this approach by demonstrating an indolent clinical course in most patients (10-13). Progression to cirrhc -
sis occurs in 19 to 64% (10,13), death from liver failurt or complications of po :?a1 hypertension is infreqlxent (lo-13), and 5- and lo-year survivals have exceeded 80% (13). The etiology of HBcAg-negative CAH in asymptsmatic patients is uncertain. Immunoserologic markers, such as smooth muscle antibodies (SMA), antinuclear antibodies (ANA), and the lupus erythematosus (LIZ) cell phenomenon, have been found in 53% of such patients and an autoimmune disorder has been considered (14). Immunologic markers, however, lack disease specificity and their detection does not preclude a viral illness (15-18). Indeed, clinical experiences have suggested that viral CAH is more likely to be mild than autoimmune disease 0.7). Patients with anti-HCV had a higher prothrombin time at accession than counterparts without anti-HCV (Table 1, Fig. 1) but other biochemical findings were similar (Table 1, Fig. l), including serum levels of immunoglobulin A (309 + 6 vs. 360 + 52 mgldl; normal, 60-400 mg/dl), immunoglobulin G (1798 f 325 vs. 1949 + 302 mg/dl; normal 700-1500 mg/dl), and immunoglobulin M (377 + 74 vs. 351 + 110, normal, 6-300 mg/dl) . Extrahepatic immunologic disorders were present more commonly in patients without anti-HCV (37 vs. 9%, p > 0.2), although the difference was not statistically significant (Table 1). Only one patient with anti-HCV had a concurrent immunologic disease (Hashimoto’s thyroidi-
ALT
TABLE 1 25G
Clinical and laboratory features at presentation in IBsAg-negative
chronicactivehepatitisof mildto moderateinflammatoryactivity
200
Mild-moderateCAH 150
Clinical features Age (years) Duration of illness (months) Sex (male/female) Blood transfusion (n) Illicit drug use (n) Immunologic disorders (n) Previous surgery (n) Laboratory features AST (nl, C27 U/dl) ALT (nl, C32 U/dl) Bilirubin (nl, Cl. 1 mg/dl) Albumin (nl, 3.1-4.3 g/dl) y-Globulin (nl, 0.7-l .6 g/dl) Prothrombin time (nl, 10.9-12.8 s) ANA al:40 (n) SMA al:40 (n) 21 Marker present (n)
Anti-HCV( +)
Anti-I-WC(-)
(n = 11)
(n = 19)
56 + 4 37 It 21 6l5 3 2 1 8
53 + 3 25 + 9 8/11 3 0 7 14
2 100
50
0
250
I
-
7
200
.
.
150 2 3 100
50
. .
*r es *
0
-
. .f.
1,
jl/
Gamma globulin
Albumin
Prothrombin time
2.5
172 + 14 90f 13 1.7 * 0.2 3.020.1 2.4 & 0.2 12.9 + 0.2* 619 218’ 719
132 f 844 1.4 f 3.2 + 2.2 +
17 16 0.2 0.1 0.2
12.1 f 0.2’ a/9 10/12* 13115
’ Significantly different at level ofp c 0.05. Anti-HCV = antibodies to hepatitis C virus; CAH = chronic active hepatitis; AST = serum aspartate aminotransferase level; ALT = serum alanine aminotransferase level; ANA = antinuclear antibodies; SMA = smooth muscle antibodies.
+
+
+
Anti-H&
Anti-H&
Anti-H&
Fig. 1. Laboratory findings at presentation in mild to moderate chronic active hepatiris with (+) and without (-) antibody to hepatitis C virus (anti-HCV). AST and ALT denote serum aspartate and aianine aminotransferase levels, respectively. The shaded areas represent normal laboratory ranges.
ANTI-HCV IN MILD TO MODERATE
CAH
91
tis) and this patient was nonreactive by RIBA. Of the w’en patients without anti-HCV who had concurrent immunologic diseases, two patients had rheumatoid arthritis, two had chronic ulcerative colitis, and one each had either Hashimoto’s thyroiditis, keratoconjunctivitis sicca, or asthma. Antinuclear antibody and/or SMA were detected with similar frequency in those patients with and without antiHCV who were tested (78 vs. 87%,p > 0.9) (Table 1). Although ANA was detected commonly in both groups, SMA was found significantly more frequently in patients who were seronegative for anti-HCV (Table 1). The median serum titers for ANA (1: 160 vs. 1:40) and SMA (1:40 vs. 0) were also higher in patients who were seronegative for anti-HCV (Fig. 2). High titers (~1: 160) of ANA or SMA were more common in seronegative patients (53 vs. ll%, p > 0.05) and this difference just missed statistical significance (Fig. 2). Patients with and without anti-HCV had a similarly high frequency of previous surgery (73 vs. 74%, p > 0.9). The frequency of previous blood transfusion and past illicit drug use was higher in seropositive patients (45 vs. 16%, p > 0.1) (Table 1) but not to a statistically significant degree. Duration and specificity of seropositivity
Successive serum samples obtained from seven of the 11 ELISA-positive patients (64%) demonstrated persistence of anti-HCV in each patient after 23 f 4 months of follow-up (range, 8 to 35 months). Eight of the 11 patients (73%) who were seropositive by ELISA had antibodies that were reactive to beth ClOO-3 and 5-l-l antigens by RIBA (Table 2). Serum from one patient had an
Antinuclear antibody
TABLE 2 Results of viroiogic and serologic tests in patients with anti-HCV by ELISA Patient
1 2 3 4 5 6 7 8 9 10 11
ELISA S/C ratio 6.3 6.8 6.8 6.7 6.8 6.8 6.5 6.7 6.7 1.3 1.0
RIBA 5-i-l
ClOO-3
1+ 4-k 4+ 4+ 2+ 4+ 4+ 4-k
2+ 4+ 1+ 3+ 2+ 4+ 4+ 4+ 4-k _ -
-
ANA
SMA
y-Globulin WW
40 80 80 20 20 0 0 40 0 5120 40
0 0 80 20 40 0 0 0 0 0 0
2.25 2.4.5 2.63 1.84 2.27 1.74 2.23 2.44 2.39 1.91 2.47
* Normal range, 0.7-1.6 g/dl. Anti-HCV = antibodies to hepatitis C virus; ELISA = enzymelinked immunosorbent assay; S/C RATIO = sample/cutoff ratio; RIBA = recombintit immunoblot assay; ANA = antinuclear antibodies; SMA = smooth muscle antibodies.
indeterminate reaction by RIBA (reactivity to ClOO-3 Only) and sera from two patients were nonreactive to both antigens (Table 2). The RIBA-nonreactive patients had a hypergammaglobulinemia that was similar to that of the RIBA-reactive patients but they had lower (borderline positive) sample/cutoff ratios by ELISA (s 1.9) (Table 2). In contrast, the RIBA-reactive patients had strongly positive sample/cutoffratios by ELISA (36.3) (Table 2). Five of the eight RIBA-reactive patients had immunoserologic markers that supported the diagnosis of autoimmune CAH but serum titers were low in each ( 0.95).
Smooth muscle antibody
Discussion
Ser;pf$ivity
t
-
Anti-HCV
+
-
Anti-HCV
Fig. 2. Reciprocal serum titers of antinuclear antibody and smooth muscle antibody at presentation in mild to moderate chronic active hepatitis with (+) and without (-) antibody to hepatitis C virus (antiHCV). Seropositivity was defined as a titer 3 140.
Our findings indicate that asymptomatic patients with HBsAg-negative CAH are commonly seropositive for anti-HCV. Seropositive patients had a history of ancient blood transfusion and previous illicit drug use more frequently than seronegative counterparts and they had concomitant immunologic disorders less commonly. Individual clinical features and laboratory findings, however, did not distinguish these patients. Indeed. they commonly had immunoserologic markers and hypergammaglobulinemia that confounded their diagnosis. Importantly, se-
92 rum titers of the immunoserologic markers were higher in seronegative patients and SMA was more commonly present. High titers (31:160) of immunoserologic markers were unusual in seropositive patients, occurring in only one patient who was RIBA-nonreactive, and this observation may well be useful in discriminating between patients with viral and autoimmune CAH. previous studies have indicated that immunoserologic markers are not disease-specific and that they may accompany drug-related (27) or virus-induced disease (15,16,18,28). indeed, none of the immunoserologic markers has as yet been assigned a pathogenic role or associated with a disease-specific immunogenic stimulus (16) and the diagnostic significance of the immunoserologic findings in our patients with anti-HCV remains unknown. Recent studies using a complementary DNA polymerase chain reaction have detected HCV RNA in the liver, plasma, and serum of patients with posttransfusion CAH and anti-HCV seropositivity, suggesting that in patients with CAH the presence of anti-HCV is commonly associated with HCV viremia (29). These findings imply that anti-HCV seropositivity has an important etiologic connotation and that the immunoserologic markers in patients with anti-HCV are either epiphenomena of the viraI infection, nonspecific manifestations of the hepatocellular inflammation, or features of a concurrent autoimmune disorder. Sanchez-Tapias and colleagues have shown that anti-HCV seropositivity may be associated with low titers (Cl: 100) of ANA and that weakly positive reactions in such patients may well be epiphenomena of little diagnostic importance (30). Our findings agree with their observations and extend them by adding RIBA data. Indeed, none of our patients who were RIBA-reactive had titers of ANA or SMA greater than 1:80 (Table 2). Although the enzyme immunoassay for anti-HCV can be affected by high serum levels of y-globulin, nonspecific cross-reacting antibodies, or other undefined serum components that may produce false-positive results (24,31), eight of our 11 patients (73%) who were seropositive by SOUSA and hypergammaglobulinemic were also reactive by RIBA (Table 2). These findings indicate that the majority of our ELISA-positive patients had specific antibodies to HCV antigens and that they probably had an actual HCV infection. The two RIBA-nonreactive patients had a hypergammaglobulinemia but it did not distinguish them from the RIBA-reactive patients (Table 2). These observations suggest that hypergammaglobulinemia does not consistently affect ELISA reactions in all patients with HBsAg-negative CAH and that ELISA positivity in hypergammaglobulinemic patients must not be assumed to be falsely positive even in the presence of immunoserologic markers.
A.J. CZAJA
et al.
Importantly, the RIBA-nonreactive patients had low, borderline positive, sample/cutoff ratios by ELISA that contrasted markedly with the strongly positive ELISA reactions that uniformly were present in the RIBA-reactive patients (Table 2). Earlier studies have indicated that the ELISA is frequently weakly reactive in disorders not commonly linked to HVC infection (i.e., primayr biliary cirrhosis) and in blood donors (30) and that the &ronger the ELISA-reaction the greater the frequency of’ RIBA reactivity and detection of HCV RNA in sermr by polymerase chain reaction (32). Our findings support these observations by indicating that weak ELISA reactions may be RIBA nonreactive and falsely positive and that the actual sample/cutoff ratio may be a better means of evaluating the possibility of an HVC infection than the unquantitated ELISA result. The majority of our patients with asymptomatic CAH (63%) did not have anti-HCV seropositivity and the etiology of their liver disease remains unknown. Immunoserologic markers were commonly present in these patients (87%) and they may well have had autoimmune CAH. Viral or drug-induced disease, however, cannot be excluded. Our asymptomatic patients had a high frequency of previous surgery (73%) and although only three of the seronegative patients had received blood transfusions, the majority had received anesthetic agents and other drugs and they may have had inadvertent parenteral exposure to a hepatitic virus. Since the absence of antiHCV does not exclude HCV infection (29), we cannot discount the possibility of this occurrence in our seronegative patients. Surgery in the absence of blood transfusion has not been described previously as a risk factor for asymptomatic CAH but our findings indicate that further evaluation of this possibility is warranted. Importantly, our observarions indicate that HCV infection should be considered as an etiologic possibility in all asymptomatic patients with HBsAg-negative CAH regardless of immunoserologic,status. The presence of concurrent extrahepatic immunologic disease, smooth muscle antibody seropostivity, or immunoserologic markers in high titer makes this consideration less likely but it does not eliminate the possibility. False positive results by ELISA are possible and an independent confirmatory test for HCV infection is needed to assist in therapeutic decisions. While recombinant a-interferon is a promising drug in the management of CAH (33), it may be deleterious in patients with autoimmune CAH who are misdiagnosed as having viral CAH (34). Conversely, corticosteroid therapy may benefit ELISA-positive, RIBA-nonreactive patients with severe CAH and immunoserologic lnarkers (35). Consequently, the dechion to institute antiviral or immunosuppressive therapy in ELISA-positive patients
ANTI-I-XV IN MILD TO MODERATE CAH with autoimmune features must be well founded. In such instances, the strength of the ELISA reaction may indicate the likelihood of RIBA reactivity and help in clarifying the etiologic diagaosls. Unfortunately, the correct treatment strategy for asymptomatic patients with less than severe disease is still unclear. We hope that our ongoing randomized controlled treatment trial will provide useful treatment guidelines in the future.
1 Soloway RG, Summerskill WHJ, Baggenstoss AH, et al. Ciinicai, biochemical, and histological remission of severe chronic active lives disease. a cnstrolled study of treatments and early progZU. iiasiioenterology 1972; 63: 820-33. 2 Czaja AJ, Davis GL, Ludwig J, Baggenstoss AH, TasweU HF. Autoimmune features as determinants of prognosis i? steroidtreated chronic active hepatitis of uncertain etiology. Gastroenteroloev 1983: 85: 713-7. 3 Czaja AJ. Natural history, clinical features, and treatment of autoimmune hepatitis. Semin Liver Dis 1984; 4: l-12. 4 Koretz RL, Lewin KJ, Higgins J, Fagen ND, Gitnick GL. Chronic active hepatitis. Who meets treatment criteria? Dig Dis Sci 1980; 25: 695-9. 5 Hodges JR, Millward-Sadler GH, Wright R. Chronic active hepatitis: the spectrum of disease. Lancet 1982; i: 550-2. 6 Czaja AJ, Summer&iii WHJ. Chronic hepatitis. To treat or not to treat? Med Clin N Am 1978; 62: 71-85. 7 Czaja AJ, Current problems in the diagnosis and management of chronic active hepatitis. Mayo Ciin Proc 1981; 56: 31 l-23. 8 Czaja AJ, Strategies in the management of chronic active hepatitis. Survey Dig Dis 1984; 2: 233-43. 9 Czaja AJ. Treatment strategies in chronic active hepatitis. In: Czaja AJ, Dickson ER, eds. Chronic Active Hepatitis. The Mayo Clinic Experience. pew York: Marcel Dekker, i98ti 247-67. 10 Thaler H. The natural history of chronic hepatitis. In: Schaffncr F, Sherlock S, Leevy CM, eds. The Liver and its Diseases. New York: Stratton Intercontinental, 1974: 207-U. 11 DeGroote J, Fevery J, Lepoutre L. Long-term follow-up of chronic active hepatitis of moderate severity. Gut 1978; 19: 510-3. 12 Fevery J, Desmet VJ, DeGroote J. Long-term follow-up and management of asymptomatic chronic active hepatitis. In: Cohen S, Soloway RD, eds. Chronic Active Liver Disease. New York: Churchill Livingstone, 1983: 51-64. 13 Kemeny MJ, Q’Harlton G, Gregory PB, Asymptomatic chronic active hepatitis. Prognosis and treatment (Absu.). Gastroenterology 1984; 86: 1325. 14 Hay JE, Czaja AJ1 Rakela J, Ludwig J. The nature of unexplained chronic aminotransferase elevations of a mild to moderate degree in asymptomatic patients. Hepatoiogy 1989; 9: 193-7. 15 Sol~?y RD, Summerskill WHJ, Baggenstoss AH, Schoenfield LJ. ‘Lupoid’ hepatitis, a nonentity in the spectrum of chronic active liver disease. Gastroenteroiogy 1972; 63: 458-65. 16 Czaja AJ. Autoimmune chronic active hepatitis. A specific entity? The negative argument. J Gastroent Hepatol 1990; 5: 343-51. 17 Esteban JI, Estebau R, Viiadomiu L, et al. Hepatitis C virus antibodies among risk groups in Spain. Lancet 1989; ii: 294-7. 18 Lenzi M, BaMardini G, Fusconi M, et al. Type 2 autoimmune hepatitis and hepatitis C virus infection. Lancet 1990; 335: 258-97.
93
Linda Grande the preparation
provided
expert secretarial
of the submitted
ses was performed
manuscript.
assistance
in
Data analy-
in part using the CLINFO Data Anz!y-
sis System.
19 Berman M, Alter HJ, Ishak KG, Purcell RH, Jones EA. The ChKmk! sequeiae of non-A, non-B hepatitis. Ann Intern Med 1979; 91: l-6. 20 Koretz RL, Stone 0, Gitnick GL. The icng-term course of nonA, non-B post-transfusion hepatitis. Gastroenteroiogy 1980; 79: 893-8. 21 Czaja Ad, Davis GL. Hepatitis non-A, non-B. Manifestations and implications of acute and chronic disease. Mayo Clin Proc 1982; 57: 639-52. 22 Choo Q-L. Kuo G, Weiner AJ, Overby LR, Brad!ey DW, Houghton M. Isolation of a cDNA clone derived from a bloodborne non-A, non-B viral hepatitis genome. Science 1989; 244: 359-61. 23 KUO G, Choo Q-L, Alter HJ, et al. An assay for circulating antibodies to a major etiologic virus of human non-A, non-B hepatitis. Science 1989; 244: 362-4. 24 McFariane IG, Smith HM, Johnson PJ, Bray GP, Vergani D, Williams R. Hepatitis C virus antibodies in chronic active hepatitis: pathogenetic factor or false-positive result? Lancet 1990; 335: 754-7. 25 DeGroote J, Desmet VJ, Gedigk P, et al. A classification of chronic hepatitis. Lancet 1968; ii: 626-8. 26 Czaja AJ, Davis GL, Ludwig J, Tasweil HF. Complete resoiution of inflammatory activity following corticosteroid treatment of HBsAg-negative chronic active hepatitis. Hepatology 1984; 4: 622-7. 27 Seeff LB. Drug-induced chronic liver disease with emphasis on chronic active hepatitis. Semin Liver Dis 1981; 1: 104-15. 28 Wood JR, Czaja AJ, Beaver SJ, et al. Frequency and significance of antibody to double-stranded DNA in chronic active hepatitis. Hepatoiogy 1986; 6: 976-80. 29 Weiner AJ, Kuo G, Bradley DW, et al. Detection of hepatitis C viral sequences in non-A, non-B hepatitis. Lancet 1990: 335; l-3. 30 Sanchez-Tapias JM, Barrera JM, Costa J, et al. Hepatitis C virus infection in patients with nonalcoholic chronic liver disease. Ann Intern Med 1990; 112: 921-4. 31 McFarlane IG, Johnson PJ, Williams R. Implications of anti-hepatitis C reactivity in ‘autoimmune’ chronic active hepatitis. Gastroenterology 1990; 99: 1532-3. 32 Weiner. AJ, Truett MA, Rosenbiatt J, et al. HCV testing in lowrisk population. Lancet 1990; 336: 695. 33 DiBisceglie AM, Martin P, Kassianides C, et al. Recombinant interferon alpha therapy for chronic hepatitis C. A randomized, double-blind, placebo-controlled trial. N Engl 3 Med 1989; 321: 1506-10. 34 Vento S, DiPerri G, Garofano T, et al. Hazards of interferon therapy for HBV-seronegative chronic hepatitis. Lancet 1989; ii: 926. 35 Czaja AJ, TasweU HF, Rakeia J, Schimek C. Frequency and significance of antibody to hepatitis C virus in severe corticosteroidtreated autoimmune chronic active hepatitis. Mayo Ciin PrOc 1991; in press.
88 HEPAT 00902
Frequency of antibody to he atitis C virus in asymptomatic negative chronic active hepatitis*
Albert J. Czaja, Howard I!. Taswell, Jorge Rakela and Carol Schimek HepatobiiiaryUnit, Division of Gastroenterology,and Section of Transfuron Medicine, Division ofLaborarov Medicine, Mayo Clinic and Mayo Medical School, Rochester, MN, United Statesof America
(Received 8 October 19W)
To determine the frequency of antibodies to hepatitis C virus in asymptomatic patients with HBsAg-negative chronic active hepatitis,sera from 30 consecutive patients with few or no symptoms of liver disease were tested by an enzyme immunoassay. The reactivity of antibodies detected by enzyme immup.oassay against hepatitis C virus encoded antigens was determined by recombinant immunoblot assay. Antibodies were detected in 11 of the 30 patients (37%) and eight of the seropositive SF-ra(73%) were reactive by recombinant hmUnGblGt assay. Nonreactive patients were weakly positive by enzyme immunoassay (sample/cutoff ratio , G1.9) in contrast to reactive patients (sample/cutoff ratio, 26.3). The prevalence of immunoserologic markers was similar in patients with and without antibodies (78 vs. 87%) but high titers (31:160) were more common in seronegative patients (53 vs. al%). Additionally, seronegative patients had smooth muscle antibodies (83 vs. %5%, p < 0.05) and concurrent extrahepatic immunologic diseases (37 vs. 9%) more commonly than seropositive counterparts. We conclude that asymptomatic patients with HBsAg-negative chronic active hepatitis frequently have antibodies to hepatitis C virus. These antibodies commonly react to specific viral antigens, especially if the enzyme immunoassay is strongly positive. Seropositive patients infrequently have concurrent immunologic disorders or smooth muscle antibodies. Immunoserologic markers lack diagnostic specificity except in high titer.
Prednisone alone or in combination with azathioprine is effective in amelioratir,g symptoms, improving biochemical and histological findings, and enhancing survival in patients with severe HBsAg-negative (autoimmune) chronic active hepatitis (CAH) (l-3). The majority of patients with HBsAg-negative CAH, however, do not have severe, incapacitating, or immediately life-threatening disease and many are asymptomatic (4,5). A favorable benefit/risk ratio for corticosteroid treatment has not been established in these patients (6,7) and currently, in the absence of confident treatment guidelines, they are managed expectantly (6-9). Several studies have supported this approach by demonstrating an indolent clinical course in most patients (10-13). Progression to cirrhc -
sis occurs in 19 to 64% (10,13), death from liver failurt or complications of po :?a1 hypertension is infreqlxent (lo-13), and 5- and lo-year survivals have exceeded 80% (13). The etiology of HBcAg-negative CAH in asymptsmatic patients is uncertain. Immunoserologic markers, such as smooth muscle antibodies (SMA), antinuclear antibodies (ANA), and the lupus erythematosus (LIZ) cell phenomenon, have been found in 53% of such patients and an autoimmune disorder has been considered (14). Immunologic markers, however, lack disease specificity and their detection does not preclude a viral illness (15-18). Indeed, clinical experiences have suggested that viral CAH is more likely to be mild than autoimmune disease 0.7). Patients with anti-HCV had a higher prothrombin time at accession than counterparts without anti-HCV (Table 1, Fig. 1) but other biochemical findings were similar (Table 1, Fig. l), including serum levels of immunoglobulin A (309 + 6 vs. 360 + 52 mgldl; normal, 60-400 mg/dl), immunoglobulin G (1798 f 325 vs. 1949 + 302 mg/dl; normal 700-1500 mg/dl), and immunoglobulin M (377 + 74 vs. 351 + 110, normal, 6-300 mg/dl) . Extrahepatic immunologic disorders were present more commonly in patients without anti-HCV (37 vs. 9%, p > 0.2), although the difference was not statistically significant (Table 1). Only one patient with anti-HCV had a concurrent immunologic disease (Hashimoto’s thyroidi-
ALT
TABLE 1 25G
Clinical and laboratory features at presentation in IBsAg-negative
chronicactivehepatitisof mildto moderateinflammatoryactivity
200
Mild-moderateCAH 150
Clinical features Age (years) Duration of illness (months) Sex (male/female) Blood transfusion (n) Illicit drug use (n) Immunologic disorders (n) Previous surgery (n) Laboratory features AST (nl, C27 U/dl) ALT (nl, C32 U/dl) Bilirubin (nl, Cl. 1 mg/dl) Albumin (nl, 3.1-4.3 g/dl) y-Globulin (nl, 0.7-l .6 g/dl) Prothrombin time (nl, 10.9-12.8 s) ANA al:40 (n) SMA al:40 (n) 21 Marker present (n)
Anti-HCV( +)
Anti-I-WC(-)
(n = 11)
(n = 19)
56 + 4 37 It 21 6l5 3 2 1 8
53 + 3 25 + 9 8/11 3 0 7 14
2 100
50
0
250
I
-
7
200
.
.
150 2 3 100
50
. .
*r es *
0
-
. .f.
1,
jl/
Gamma globulin
Albumin
Prothrombin time
2.5
172 + 14 90f 13 1.7 * 0.2 3.020.1 2.4 & 0.2 12.9 + 0.2* 619 218’ 719
132 f 844 1.4 f 3.2 + 2.2 +
17 16 0.2 0.1 0.2
12.1 f 0.2’ a/9 10/12* 13115
’ Significantly different at level ofp c 0.05. Anti-HCV = antibodies to hepatitis C virus; CAH = chronic active hepatitis; AST = serum aspartate aminotransferase level; ALT = serum alanine aminotransferase level; ANA = antinuclear antibodies; SMA = smooth muscle antibodies.
+
+
+
Anti-H&
Anti-H&
Anti-H&
Fig. 1. Laboratory findings at presentation in mild to moderate chronic active hepatiris with (+) and without (-) antibody to hepatitis C virus (anti-HCV). AST and ALT denote serum aspartate and aianine aminotransferase levels, respectively. The shaded areas represent normal laboratory ranges.
ANTI-HCV IN MILD TO MODERATE
CAH
91
tis) and this patient was nonreactive by RIBA. Of the w’en patients without anti-HCV who had concurrent immunologic diseases, two patients had rheumatoid arthritis, two had chronic ulcerative colitis, and one each had either Hashimoto’s thyroiditis, keratoconjunctivitis sicca, or asthma. Antinuclear antibody and/or SMA were detected with similar frequency in those patients with and without antiHCV who were tested (78 vs. 87%,p > 0.9) (Table 1). Although ANA was detected commonly in both groups, SMA was found significantly more frequently in patients who were seronegative for anti-HCV (Table 1). The median serum titers for ANA (1: 160 vs. 1:40) and SMA (1:40 vs. 0) were also higher in patients who were seronegative for anti-HCV (Fig. 2). High titers (~1: 160) of ANA or SMA were more common in seronegative patients (53 vs. ll%, p > 0.05) and this difference just missed statistical significance (Fig. 2). Patients with and without anti-HCV had a similarly high frequency of previous surgery (73 vs. 74%, p > 0.9). The frequency of previous blood transfusion and past illicit drug use was higher in seropositive patients (45 vs. 16%, p > 0.1) (Table 1) but not to a statistically significant degree. Duration and specificity of seropositivity
Successive serum samples obtained from seven of the 11 ELISA-positive patients (64%) demonstrated persistence of anti-HCV in each patient after 23 f 4 months of follow-up (range, 8 to 35 months). Eight of the 11 patients (73%) who were seropositive by ELISA had antibodies that were reactive to beth ClOO-3 and 5-l-l antigens by RIBA (Table 2). Serum from one patient had an
Antinuclear antibody
TABLE 2 Results of viroiogic and serologic tests in patients with anti-HCV by ELISA Patient
1 2 3 4 5 6 7 8 9 10 11
ELISA S/C ratio 6.3 6.8 6.8 6.7 6.8 6.8 6.5 6.7 6.7 1.3 1.0
RIBA 5-i-l
ClOO-3
1+ 4-k 4+ 4+ 2+ 4+ 4+ 4-k
2+ 4+ 1+ 3+ 2+ 4+ 4+ 4+ 4-k _ -
-
ANA
SMA
y-Globulin WW
40 80 80 20 20 0 0 40 0 5120 40
0 0 80 20 40 0 0 0 0 0 0
2.25 2.4.5 2.63 1.84 2.27 1.74 2.23 2.44 2.39 1.91 2.47
* Normal range, 0.7-1.6 g/dl. Anti-HCV = antibodies to hepatitis C virus; ELISA = enzymelinked immunosorbent assay; S/C RATIO = sample/cutoff ratio; RIBA = recombintit immunoblot assay; ANA = antinuclear antibodies; SMA = smooth muscle antibodies.
indeterminate reaction by RIBA (reactivity to ClOO-3 Only) and sera from two patients were nonreactive to both antigens (Table 2). The RIBA-nonreactive patients had a hypergammaglobulinemia that was similar to that of the RIBA-reactive patients but they had lower (borderline positive) sample/cutoff ratios by ELISA (s 1.9) (Table 2). In contrast, the RIBA-reactive patients had strongly positive sample/cutoffratios by ELISA (36.3) (Table 2). Five of the eight RIBA-reactive patients had immunoserologic markers that supported the diagnosis of autoimmune CAH but serum titers were low in each ( 0.95).
Smooth muscle antibody
Discussion
Ser;pf$ivity
t
-
Anti-HCV
+
-
Anti-HCV
Fig. 2. Reciprocal serum titers of antinuclear antibody and smooth muscle antibody at presentation in mild to moderate chronic active hepatitis with (+) and without (-) antibody to hepatitis C virus (antiHCV). Seropositivity was defined as a titer 3 140.
Our findings indicate that asymptomatic patients with HBsAg-negative CAH are commonly seropositive for anti-HCV. Seropositive patients had a history of ancient blood transfusion and previous illicit drug use more frequently than seronegative counterparts and they had concomitant immunologic disorders less commonly. Individual clinical features and laboratory findings, however, did not distinguish these patients. Indeed. they commonly had immunoserologic markers and hypergammaglobulinemia that confounded their diagnosis. Importantly, se-
92 rum titers of the immunoserologic markers were higher in seronegative patients and SMA was more commonly present. High titers (31:160) of immunoserologic markers were unusual in seropositive patients, occurring in only one patient who was RIBA-nonreactive, and this observation may well be useful in discriminating between patients with viral and autoimmune CAH. previous studies have indicated that immunoserologic markers are not disease-specific and that they may accompany drug-related (27) or virus-induced disease (15,16,18,28). indeed, none of the immunoserologic markers has as yet been assigned a pathogenic role or associated with a disease-specific immunogenic stimulus (16) and the diagnostic significance of the immunoserologic findings in our patients with anti-HCV remains unknown. Recent studies using a complementary DNA polymerase chain reaction have detected HCV RNA in the liver, plasma, and serum of patients with posttransfusion CAH and anti-HCV seropositivity, suggesting that in patients with CAH the presence of anti-HCV is commonly associated with HCV viremia (29). These findings imply that anti-HCV seropositivity has an important etiologic connotation and that the immunoserologic markers in patients with anti-HCV are either epiphenomena of the viraI infection, nonspecific manifestations of the hepatocellular inflammation, or features of a concurrent autoimmune disorder. Sanchez-Tapias and colleagues have shown that anti-HCV seropositivity may be associated with low titers (Cl: 100) of ANA and that weakly positive reactions in such patients may well be epiphenomena of little diagnostic importance (30). Our findings agree with their observations and extend them by adding RIBA data. Indeed, none of our patients who were RIBA-reactive had titers of ANA or SMA greater than 1:80 (Table 2). Although the enzyme immunoassay for anti-HCV can be affected by high serum levels of y-globulin, nonspecific cross-reacting antibodies, or other undefined serum components that may produce false-positive results (24,31), eight of our 11 patients (73%) who were seropositive by SOUSA and hypergammaglobulinemic were also reactive by RIBA (Table 2). These findings indicate that the majority of our ELISA-positive patients had specific antibodies to HCV antigens and that they probably had an actual HCV infection. The two RIBA-nonreactive patients had a hypergammaglobulinemia but it did not distinguish them from the RIBA-reactive patients (Table 2). These observations suggest that hypergammaglobulinemia does not consistently affect ELISA reactions in all patients with HBsAg-negative CAH and that ELISA positivity in hypergammaglobulinemic patients must not be assumed to be falsely positive even in the presence of immunoserologic markers.
A.J. CZAJA
et al.
Importantly, the RIBA-nonreactive patients had low, borderline positive, sample/cutoff ratios by ELISA that contrasted markedly with the strongly positive ELISA reactions that uniformly were present in the RIBA-reactive patients (Table 2). Earlier studies have indicated that the ELISA is frequently weakly reactive in disorders not commonly linked to HVC infection (i.e., primayr biliary cirrhosis) and in blood donors (30) and that the &ronger the ELISA-reaction the greater the frequency of’ RIBA reactivity and detection of HCV RNA in sermr by polymerase chain reaction (32). Our findings support these observations by indicating that weak ELISA reactions may be RIBA nonreactive and falsely positive and that the actual sample/cutoff ratio may be a better means of evaluating the possibility of an HVC infection than the unquantitated ELISA result. The majority of our patients with asymptomatic CAH (63%) did not have anti-HCV seropositivity and the etiology of their liver disease remains unknown. Immunoserologic markers were commonly present in these patients (87%) and they may well have had autoimmune CAH. Viral or drug-induced disease, however, cannot be excluded. Our asymptomatic patients had a high frequency of previous surgery (73%) and although only three of the seronegative patients had received blood transfusions, the majority had received anesthetic agents and other drugs and they may have had inadvertent parenteral exposure to a hepatitic virus. Since the absence of antiHCV does not exclude HCV infection (29), we cannot discount the possibility of this occurrence in our seronegative patients. Surgery in the absence of blood transfusion has not been described previously as a risk factor for asymptomatic CAH but our findings indicate that further evaluation of this possibility is warranted. Importantly, our observarions indicate that HCV infection should be considered as an etiologic possibility in all asymptomatic patients with HBsAg-negative CAH regardless of immunoserologic,status. The presence of concurrent extrahepatic immunologic disease, smooth muscle antibody seropostivity, or immunoserologic markers in high titer makes this consideration less likely but it does not eliminate the possibility. False positive results by ELISA are possible and an independent confirmatory test for HCV infection is needed to assist in therapeutic decisions. While recombinant a-interferon is a promising drug in the management of CAH (33), it may be deleterious in patients with autoimmune CAH who are misdiagnosed as having viral CAH (34). Conversely, corticosteroid therapy may benefit ELISA-positive, RIBA-nonreactive patients with severe CAH and immunoserologic lnarkers (35). Consequently, the dechion to institute antiviral or immunosuppressive therapy in ELISA-positive patients
ANTI-I-XV IN MILD TO MODERATE CAH with autoimmune features must be well founded. In such instances, the strength of the ELISA reaction may indicate the likelihood of RIBA reactivity and help in clarifying the etiologic diagaosls. Unfortunately, the correct treatment strategy for asymptomatic patients with less than severe disease is still unclear. We hope that our ongoing randomized controlled treatment trial will provide useful treatment guidelines in the future.
1 Soloway RG, Summerskill WHJ, Baggenstoss AH, et al. Ciinicai, biochemical, and histological remission of severe chronic active lives disease. a cnstrolled study of treatments and early progZU. iiasiioenterology 1972; 63: 820-33. 2 Czaja AJ, Davis GL, Ludwig J, Baggenstoss AH, TasweU HF. Autoimmune features as determinants of prognosis i? steroidtreated chronic active hepatitis of uncertain etiology. Gastroenteroloev 1983: 85: 713-7. 3 Czaja AJ. Natural history, clinical features, and treatment of autoimmune hepatitis. Semin Liver Dis 1984; 4: l-12. 4 Koretz RL, Lewin KJ, Higgins J, Fagen ND, Gitnick GL. Chronic active hepatitis. Who meets treatment criteria? Dig Dis Sci 1980; 25: 695-9. 5 Hodges JR, Millward-Sadler GH, Wright R. Chronic active hepatitis: the spectrum of disease. Lancet 1982; i: 550-2. 6 Czaja AJ, Summer&iii WHJ. Chronic hepatitis. To treat or not to treat? Med Clin N Am 1978; 62: 71-85. 7 Czaja AJ, Current problems in the diagnosis and management of chronic active hepatitis. Mayo Ciin Proc 1981; 56: 31 l-23. 8 Czaja AJ, Strategies in the management of chronic active hepatitis. Survey Dig Dis 1984; 2: 233-43. 9 Czaja AJ. Treatment strategies in chronic active hepatitis. In: Czaja AJ, Dickson ER, eds. Chronic Active Hepatitis. The Mayo Clinic Experience. pew York: Marcel Dekker, i98ti 247-67. 10 Thaler H. The natural history of chronic hepatitis. In: Schaffncr F, Sherlock S, Leevy CM, eds. The Liver and its Diseases. New York: Stratton Intercontinental, 1974: 207-U. 11 DeGroote J, Fevery J, Lepoutre L. Long-term follow-up of chronic active hepatitis of moderate severity. Gut 1978; 19: 510-3. 12 Fevery J, Desmet VJ, DeGroote J. Long-term follow-up and management of asymptomatic chronic active hepatitis. In: Cohen S, Soloway RD, eds. Chronic Active Liver Disease. New York: Churchill Livingstone, 1983: 51-64. 13 Kemeny MJ, Q’Harlton G, Gregory PB, Asymptomatic chronic active hepatitis. Prognosis and treatment (Absu.). Gastroenterology 1984; 86: 1325. 14 Hay JE, Czaja AJ1 Rakela J, Ludwig J. The nature of unexplained chronic aminotransferase elevations of a mild to moderate degree in asymptomatic patients. Hepatoiogy 1989; 9: 193-7. 15 Sol~?y RD, Summerskill WHJ, Baggenstoss AH, Schoenfield LJ. ‘Lupoid’ hepatitis, a nonentity in the spectrum of chronic active liver disease. Gastroenteroiogy 1972; 63: 458-65. 16 Czaja AJ. Autoimmune chronic active hepatitis. A specific entity? The negative argument. J Gastroent Hepatol 1990; 5: 343-51. 17 Esteban JI, Estebau R, Viiadomiu L, et al. Hepatitis C virus antibodies among risk groups in Spain. Lancet 1989; ii: 294-7. 18 Lenzi M, BaMardini G, Fusconi M, et al. Type 2 autoimmune hepatitis and hepatitis C virus infection. Lancet 1990; 335: 258-97.
93
Linda Grande the preparation
provided
expert secretarial
of the submitted
ses was performed
manuscript.
assistance
in
Data analy-
in part using the CLINFO Data Anz!y-
sis System.
19 Berman M, Alter HJ, Ishak KG, Purcell RH, Jones EA. The ChKmk! sequeiae of non-A, non-B hepatitis. Ann Intern Med 1979; 91: l-6. 20 Koretz RL, Stone 0, Gitnick GL. The icng-term course of nonA, non-B post-transfusion hepatitis. Gastroenteroiogy 1980; 79: 893-8. 21 Czaja Ad, Davis GL. Hepatitis non-A, non-B. Manifestations and implications of acute and chronic disease. Mayo Clin Proc 1982; 57: 639-52. 22 Choo Q-L. Kuo G, Weiner AJ, Overby LR, Brad!ey DW, Houghton M. Isolation of a cDNA clone derived from a bloodborne non-A, non-B viral hepatitis genome. Science 1989; 244: 359-61. 23 KUO G, Choo Q-L, Alter HJ, et al. An assay for circulating antibodies to a major etiologic virus of human non-A, non-B hepatitis. Science 1989; 244: 362-4. 24 McFariane IG, Smith HM, Johnson PJ, Bray GP, Vergani D, Williams R. Hepatitis C virus antibodies in chronic active hepatitis: pathogenetic factor or false-positive result? Lancet 1990; 335: 754-7. 25 DeGroote J, Desmet VJ, Gedigk P, et al. A classification of chronic hepatitis. Lancet 1968; ii: 626-8. 26 Czaja AJ, Davis GL, Ludwig J, Tasweil HF. Complete resoiution of inflammatory activity following corticosteroid treatment of HBsAg-negative chronic active hepatitis. Hepatology 1984; 4: 622-7. 27 Seeff LB. Drug-induced chronic liver disease with emphasis on chronic active hepatitis. Semin Liver Dis 1981; 1: 104-15. 28 Wood JR, Czaja AJ, Beaver SJ, et al. Frequency and significance of antibody to double-stranded DNA in chronic active hepatitis. Hepatoiogy 1986; 6: 976-80. 29 Weiner AJ, Kuo G, Bradley DW, et al. Detection of hepatitis C viral sequences in non-A, non-B hepatitis. Lancet 1990: 335; l-3. 30 Sanchez-Tapias JM, Barrera JM, Costa J, et al. Hepatitis C virus infection in patients with nonalcoholic chronic liver disease. Ann Intern Med 1990; 112: 921-4. 31 McFarlane IG, Johnson PJ, Williams R. Implications of anti-hepatitis C reactivity in ‘autoimmune’ chronic active hepatitis. Gastroenterology 1990; 99: 1532-3. 32 Weiner. AJ, Truett MA, Rosenbiatt J, et al. HCV testing in lowrisk population. Lancet 1990; 336: 695. 33 DiBisceglie AM, Martin P, Kassianides C, et al. Recombinant interferon alpha therapy for chronic hepatitis C. A randomized, double-blind, placebo-controlled trial. N Engl 3 Med 1989; 321: 1506-10. 34 Vento S, DiPerri G, Garofano T, et al. Hazards of interferon therapy for HBV-seronegative chronic hepatitis. Lancet 1989; ii: 926. 35 Czaja AJ, TasweU HF, Rakeia J, Schimek C. Frequency and significance of antibody to hepatitis C virus in severe corticosteroidtreated autoimmune chronic active hepatitis. Mayo Ciin PrOc 1991; in press.
