Acute non-A non-B Hepatitis Research Articles

Hepatitis E Virus: Epidemiology, Diagnosis, and Management

The HEV is a known cause of water-borne outbreaks of acute non-A non-B hepatitis in developing countries, which affects young people and may result in high mortality in pregnant women. In recent decades, however, HEV genotypes 3 and 4 have been known as a cause of sporadic zoonotic infections in older males from swine HEV worldwide. Most acute HEV infections are self-limited. On the other hand, in immunosuppressed patients, including solid organ transplant recipients, chronic HEV infections may exist and progress to liver cirrhosis or decompensation. Therefore, physicians need to recognize HEV as a major pathogen for acute and chronic hepatitis of unknown causes and investigate this disease.

Long-term follow-up of anti-hepatitis C virus antibodies in patients with acute nonA nonB hepatitis and different outcome of liver disease.

We screened 74 patients with nonA nonB acute hepatitis (37 of post-transfusion PTH, and 37 of non-post-transfusion, NPTH, origin) for the presence of anti-HCV by tests detecting either C100-3 antibodies alone (ELISA I) or C100-3 plus C33c plus C22-3 antibodies (ELISA II). Samples were taken at the onset of disease and then serially for a period of time ranging from 12 to 60 months. An increased number of anti-HCV positive cases (86% vs 69%) and an earlier seroconversion were observed with the second compared to the first generation ELISA. Positive samples were confirmed by recombinant immunoblot assay (RIBA), which was also used to study the kinetics of the antibody response to individual HCV antigens. Anti-C33c and anti-C22-3 antibodies were the first detectable markers of HCV infection in 80% and 20% of the patients, respectively. Thirty-two percent of the patients studied showed complete and persistent biochemical recovery, whereas 68% maintained a chronic elevation of the transaminase values. Among the 20 patients who showed early and persistent normalization of the transaminase values, complete disappearance of all antibody reactivities was limited to five of them, whereas in four cases only anti-C100-3 and anti-5-1-1 became negative.

A serological re-evaluation of acute non-A non-B hepatitis from the early 1970s.

The epidemiology and natural history of recently discovered viruses, which may be responsible for cases of seronegative infectious hepatitis, are currently being investigated. Retrospective studies of stored sera can provide a historical perspective of these infections. To re-evaluate the serological, demographic and clinical characteristics of patients hospitalised in the early 1970s with acute hepatitis. The stored sera of 57 patients hospitalised between 1971 and 1974 with acute hepatitis, designated at that time as non-A non-B (NANB) hepatitis, were re-tested using commercially available enzyme-linked immunosorbent assays (ELISAs) for the presence of anti-hepatitis A virus (HAV) IgM, hepatitis B surface antigen (HBsAg), anti-hepatitis C virus (HCV) IgG, and anti-hepatitis E virus (HEV) IgG. Stored sera from a group of 57 patients concurrently hospitalised for other conditions were also tested. Detailed records of the original epidemiological interviews were examined to compare patient demographics, risk factors for infectious hepatitis and clinical data for the NANB hepatitis group and an original control group of 604 hospitalised patients. Serum from 15 of the 57 (26%) previously designated NANB hepatitis cases had elevated anti-HAV IgM and are likely to represent missed cases of hepatitis A. Thirteen (23%) of cases previously designated as NANB hepatitis had positive hepatitis C antibody tests. These patients were younger and significantly more likely to have used intravenous drugs than control patients. Three NANB hepatitis and two hospital control patients were anti-HEV IgG antibody positive. All of these individuals were born in, or had travelled to, developing countries. Serum from 27 (47%) of the NANB hepatitis patients were negative on all tests. These hepatitis non-A-E cases included children and elderly adults, but as a group were significantly more likely to have used intravenous drugs than hospitalised control patients. Both HCV and probable non-A-E virus(es) were important causes of acute NANB hepatitis during the early 1970s.

Pruebas de función tiroidea en embarazadas normales (tercer trimestre) y en embarazadas con colestasis gravídica o con hepatitis aguda

Intrahepatic cholestasis of pregnancy (ICP) is a disease of unknown cause characterized by pruritus and biochemical cholestasis in the 3rd trimester of pregnancy. Its pathogenesis may be due to the interaction of abnormalities in the metabolism of estrogens and progesterone, while still unknown environmental factor (s) modulate the expressivity of a genetic predisposing trait. To verify if thyroid function tests (TFT) are altered in ICP as in other hepatic diseases and whether a dietary iodine deficiency could be involved. From 1983 to 1986, 13 normal pregnancies (3rd trimester), 26 ICP patients (with 30 pregnancies) and 4 patients with acute non-A non-B hepatitis in pregnancy, were studied. Serum T3, rT3, T4, fT4 and TSH (before and after TRH) were measured by RIA; in ICP patients, measurements were repeated in puerperium. Urinary 24 h iodine excretion was measured in normal pregnancies and in 6 ICP patients. In normal pregnancies, T3 (3.00 +/- 0.22 nmol/L) and rT3 (0.40 +/- 0.03 nmol/L) were higher than the values detected in non-pregnant women; other TFT were unchanged. Urinary iodine excretion was normal in all individuals tested. Patients with acute hepatitis in pregnancy or with ICP had lower T3 than normal pregnancies (1.82 +/- 0.19 nmol/L in hepatitis; 2.24 +/- 0.12 nmol/L in ICP; p < 0.01) and higher rT3 (0.80 +/- 0.25 nmol/L in hepatitis; 0.54 +/- 0.05 nmol/L in ICP; p < 0.05), while other TFT were unchanged. None of them had clinical signs of hypo or hyperthyroidism. A "euthyroid sick syndrome" was detected in 2 ICP patients and in 2 acute hepatitis in pregnancy. In puerperium of ICP patients, T3 and rT3 returned to levels in non-pregnant women. In ICP patients, TFT show similar trends than in more severe hepatic and non-hepatic diseases. Although thyroid binding-globulin was not measured in our patients, the pattern of TFT suggests that an impaired peripheral (hepatic?) deiodination of T4 is responsible for these changes. The influence of a dietary iodine deficiency can be ruled out.

Etiology of sporadic acute viral hepatitis in Taiwan: the role of hepatitis C virus, hepatitis E virus and GB virus-C/hepatitis G virus in an endemic area of hepatitis A and B.

The etiology of sporadic acute hepatitis was studied in 334 consecutive patients from Taiwan (237 men and 97 women, aged 16-81 years), with emphasis on the role of hepatitis C virus (HCV), hepatitis E virus (HEV), and GB virus-C/hepatitis G virus (GBV-C/HGV) in acute non-A, non-B (NANB) hepatitis and in HBsAg carriers with superimposed acute hepatitis. According to the conventional diagnostic criteria, there were 12 cases (3.6%) of acute hepatitis A, 17 cases (5.1%) of acute hepatitis B, 128 cases (38.3%) of acute NANB hepatitis, and 177 cases (53.0%) of acute hepatitis in HBsAg carriers (those who were HBsAg positive but IgM anti-HBc negative). Among 128 cases of acute NANB hepatitis, 70 (54.7%) had acute hepatitis C (HCV RNA positive), 5 (3.9%) had acute hepatitis E (IgM anti-HEV positive), and the other 53 (41.4%) were presumably acute hepatitis non-A-E. The prevalence of acute hepatitis A, B, E, and non-A-E showed no significant sex difference, whereas acute hepatitis C was significantly more prevalent in females. The prevalence of acute hepatitis A and B decreased and that of acute hepatitis C increased significantly with increasing age. In contrast, acute hepatitis E and non-A-E showed no significant age predominance. Of 177 HBsAg carriers with acute hepatitis, 64 (36.1%) demonstrated non-B hepatotropic virus superinfection, with HCV being the most common (60.9%), followed by hepatitis D, E, and A viruses, and the other 55 (31.1%) and 58 (32.8%) were presumed to have acute exacerbation of chronic hepatitis B or superimposed acute hepatitis non-A-E, respectively. Serum GBV-C/HGV RNA was detected in 3-4% of acute hepatitis non-A-E cases, suggesting its limited role in these cases.

The role of hepatitis E virus in acute sporadic non-A, non-B hepatitis.

Hepatitis E virus (HEV) is the cause of enterically transmitted non-A, non-B (NANB) hepatitis. Water-borne epidemics have been reported in many developing countries primarily affecting young and middle-aged adults. To determine the role of HEV in acute sporadic NANB hepatitis, we have studied the profiles of anti-HEV IgG in the sera of patients previously diagnosed for NANB hepatitis. We tested the sera of 53 patients with acute NANB hepatitis and 100 healthy people as a control group for anti-HEV IgG. Thirty-nine of the 53 patients (73.3%) and none of the control group were positive for HEV infection, according to results shown by ELISA. This result suggests that HEV is a common cause of acute sporadic NANB hepatitis in Turkey. Further studies are needed in the other regions to determine the true prevalence of HEV infection in Turkey.

IgM and IgA antibodies generated against hepatitis C virus core antigen in patients with acute and chronic HCV infection.

Antibody subclasses directed against the core protein (HCc) of hepatitis C virus (HCV) were measured in 27 patients with acute non-A, non-B (NANB) hepatitis, and 99 patients with chronic HCV-associated liver disease. IgM, IgA, and IgG anti-HCc responses were observed in 11 (40.7%), 7 (25.9%), and 18 (67%) patients with acute NANB hepatitis, respectively. Twenty-four (24.2%) and 40 (40.4%) patients with chronic HCV infection also had detectable IgM and IgA, respectively. IgM anti-HCc inconsistently detected acute infection, and HCV ribonucleic acid (RNA) could be detected preceding the rise in anti-HCc antibodies in five consecutive patients with acute hepatitis. IgM anti-HCc also could not distinguish acute from chronic infection and did not correlate with histologic progression. However, the form of IgA present (polymeric vs monomeric) did discriminate acute from chronic infection and the IgA anti-HCc titer correlated with histologic evidence of liver disease in patients with chronic HCV infection.

ＨＰＳ（Ｈｅｍｏｐｈａｇｏｃｙｔｉｃ Ｓｙｎｄｒｏｍｅ）を呈したＥｐｓｔｅｉｎ‐Ｂａｒｒ ｖｉｒｕｓ感染性Ｔ細胞性悪性リンパ腫の２剖検例

Here we report two cases of Epstein-Barr virus related T-cell lymphomas manifested primarily as hemophagocytic syndrome. In case 1, although acute non-A non-B hepatitis with DIC was suspected, bone marrow aspiration was interpreted to be compatible with T cell malignancy. Autopsy revealed a widespread T cell lymphoma with prominent hemophagocytosis by macrophages. The second case was diagnosed as malignant histiocytosis on clinical symptoms and bone marrow findings. Lymph node biopsy was reported as diffuse lage cell lymphoma of T cell origin with reactive macrophages, which was confirmed at autopsy.In both cases, approximately one in each twenty neoplastic cells contained EB virus genomes as revealed by in situ hybridization. HPS in these cases might be explained by reactivation of lymphoma related EB virus. Macrophages should be activated by both viruses and lymphokines which were produced by tumor cells.

Seroepidemiology of hepatitis C virus in Beijing, China

To investigate the seroprevalence of hepatitis C virus (HCV) in China we tested sera from healthy individuals without hepatitis and no history of parenteral blood exposure and from patients admitted to a hepatitis hospital in Beijing. Sera were tested for anti-HCV by first-generation enzyme immunoassay; selected positives were tested with two second-generation EIAs, one utilizing recombinant antigens and the other synthetic peptides. We found anti-HCV with the following frequencies: 10 of 164 (6%) individuals with no disease; 2 of 36 (5.5%) patients with acute non-A non-B hepatitis (NANBH); 26 of 39 (67%) patients with post-transfusion NANBH; 10 of 34 (29%) patients with chronic hepatitis negative for hepatitis B surface antigen (HBsAg); 3 of 30 (10%) patients with chronic HBsAg-positive hepatitis; 0 of 19 patients with acute HBsAg-positive hepatitis. Of 24 repeat-positive sera, 19 were positive by both and 4 by one second-generation tests. We conclude that hepatitis C infection is common in China, that it contributes substantially to the incidence of post-transfusion hepatitis, and that HCV plays a significant role in both acute and chronic hepatitis. Further studies are needed to extend these observations and to define the predominant routes of transmission of HCV in China.

Detection of hepatitis C viral RNA in sporadic acute non-A, non-B hepatitis by polymerase chain reaction: Its usefulness for the early diagnosis of seronegative infection

To determine the prevalence of hepatitis C viral infection in patients with sporadic non-A, non-B (NANB) acute hepatitis, hepatitis C viral RNA was studied in the plasma of 15 patients by reverse transcription-polymerase chain reaction assay. Plasma samples were sequentially obtained from 15 patients, and polymerase chain reaction was performed with two nested pairs of primers deduced from the 5'-non-coding region of hepatitis C viral sequences. Anti-C100 and anti-GOR antibodies were also measured with an enzyme-linked immunosorbent assay system. Plasma hepatitis C viral RNA was detected transiently in 7 of 15 patients (47%) at an early phase of the clinical course, while anti-C100 antibodies were detectable in only 2 (29%) of hepatitis C viral RNA-positive patients, and in 1 (13%) of the negative patients. Of 7 patients that were positive for plasma hepatitis C viral RNA, 4 (57%) had relapsing or protracted courses. In contrast, in all patients with undetectable hepatitis C viral RNA, hepatitis C viral RNA recovered and remained normal for at least 1 year. Thus, hepatitis C viral infection represents almost half the patients with acute sporadic NANB hepatitis, and detection of hepatitis C viral RNA in an early clinical phase is superior to anti-C100 measurement for diagnosing acute sporadic hepatitis C viral infection.

Hepatitis C virus antibodies in a long-term follow-up of beta-thalassaemic children with acute and chronic non-A non-B hepatitis.

The presence of antibodies toward hepatitis C virus (HCV) was examined in 78 polytransfused beta-thalassaemic children. The anti-HCV status was correlated with acute and chronic non-A non-B (NANB) hepatitis that developed during a follow up of about 13 years. Anti-HCV was present in 83.3% of children with acute NANB hepatitis and in 82.9% of those with chronic NANB hepatitis. The percentage of chronic evolution was 56.7% for acute anti-HCV positive NANB hepatitis and 50.0% for anti-HCV negative NANB hepatitis. The long-term persistence of anti-HCV antibodies did not correlate with chronic evolution of liver infection in thalassaemic patients. Histological features of chronic hepatitis showed little or no difference between HCV associated or non-associated liver disease. The multifactorial liver injury in beta-thalassaemic children explains the high prevalence of cirrhosis (about 30%) observed in these patients with NANB hepatitis. On the other hand, independent of liver disease, some patients never seroconverted during the follow up in spite of the high number of transfusions suggesting the existence of "non-responders".

The development of hepatitis C antibody shortly after acute icteric non-A non-B hepatitis.

To determine the relationship between the development of hepatitis C antibody (anti-HCV) and the clinical symptoms in acute hepatitis C. Retrospective analysis of sera from patients with acute non-A non-B hepatitis. Patients admitted to Fairfield Hospital with the diagnosis of acute non-A non-B hepatitis between 1979 and 1989. Inclusion criteria included a typical clinical illness, accompanied by an alanine aminotransferase level of more than 2.5 times the upper limit of normal (normal, less than or equal to 40 U/L) and negative serological test results for acute hepatitis A and B. Time to develop anti-HCV after the onset of symptoms in patients with acute hepatitis C. Seroconversion was demonstrated in 26 of the 128 patients who fulfilled the inclusion criteria. In these patients, antibody was detected between one week and 32 weeks after the onset of dark urine; more than half the patients (54%) had seroconverted by four weeks and a third (34%) developed antibodies within two weeks. Of 20 patients who had sera collected within four weeks of the onset of dark urine, 14 (70%) had developed antibody. These results suggest that in patients with community-acquired hepatitis C, seroconversion occurs significantly earlier than is observed in patients who have been infected by blood transfusion. Sera taken shortly after the onset of symptomatic hepatitis C may be useful in the diagnosis of this condition.

Serodiagnosis of hepatitis C virus infection by ELISA for antibodies against the putative core protein (p20 c) expressed in Escherichia coli

The putative core gene of hepatitis C virus (HCV) was incorporated into a plasmid vector (pCC5-J4), and expressed in Escherichia coli. The product of 180 amino acids (p20 c) was purified by gel electrophoresis in the presence of sodium dodecyl sulfate, and used in enzyme-linked immunosorbent assay for antibodies against the putative core protein of HCV (anti-p20 c). Anti-p20 c was detected in 13 (1.5%) of 873 apparently healthy blood donors. It was detected in 205 (86.5%) of 237 patients with acute or chronic non-A, non-B (NANB) hepatic disease, significantly more frequently ( p < 0.01) than antibodies against the C100-3 protein encoded by nonstructural regions of HCV (anti-C100-3) that was found in 178 (75.1%). Anti-p20 c developed in the circulation of a patient with acute NANB hepatitis much earlier than anti-C100-3. HCV RNA was detected by polymerase chain reaction in serum samples from blood donors positive for anti-p20 c in high titers, one of which was negative for anti-C100-3. These results indicated that anti-p20 c would be useful in complementing anti-C100-3 for the diagnosis of NANB hepatitis and further decreasing the incidence of posttransfusion NANB hepatitis.

The prevalence of antibody to hepatitis C virus (anti-HCV) in patients with acute and chronic liver diseases in Jakarta, Indonesia

The seroepidemiology of HBV and HCV infections in the patients with acute and chronic liver diseases in Jakarta was investigated. The sera from 141 cases with acute hepatitis, 176 liver cirrhosis and 70 hepatocellular carcinoma (HCC) were examined. Anti-HA IgM, HBsAg, anti-HBc IgM and anti HCV (Ortho) were detected by Elisa method. In acute hepatitis, 83 cases (58.9%) out of 141 cases were hepatitis A and 9 cases (6.4%) hepatitis B. The others were diagnosed non-A, non-B (NANB) hepatitis and anti-HCV in 4 cases (11.8%) out of 34 cases with NANB hepatitis was positive. The low prevalence of anti-HCV in acute NANB hepatitis seems to be due to inadequate date of serum sampling. HBsAg and anti-HCV in liver cirrhosis were positive 36.5% and 73.9% respectively, including 22.7% of double infection. HBsAg and anti-HCV in HCC were 58.6% and 34.2%, including 17.1% of double infection. In 16.7% fo chronic liver disease (liver cirrhosis and HCC), neither HBsAg nor anti-HCV were detected.

Epidemiology of Acute Non-A, Non-B Hepatitis in a Large County Hospital Outpatient Population

The epidemiology of acute non-A, non-B (NANB) hepatitis in a large county hospital was determined by reviewing all acute hepatitis serology profiles over a 26-month period. Acute NANB hepatitis was identified in 77 patients from their clinical history, negative serologic findings, and elevated hepatic enzyme levels. The age and sex distributions for hepatitis A, B, and NANB are in agreement with previous epidemiologic studies. However, the frequency of various risk factors for NANB obtained from this epidemiologic investigation vary from those previously reported. Information regarding risk factors was available for 69 of the 77 patients identified. A history of recent intravenous drug abuse (IVDA) was obtained in 45 patients (65.2%). In the 6 months prior to presentation, blood transfusion was documented in five patients (7.2%). Eighteen patients (26.1%) denied having had a previous transfusion or IVDA. In our population a history of transfusion is obtained less frequently from patients with acute NANB hepatitis than previously reported. IVDA is the predominant risk factor for acute NANB hepatitis.

Detection of hepatitis C viral RNA by the polymerase chain reaction

These studies used the polymerase chain reaction (PCR) to identify hepatitis C virus (HCV) RNA in clinical samples collected from patients with chronic liver disease or from blood donors. The relative role of this assay compared with detection of HCV antibodies in serum was evaluated. Weiner and colleagues found that 9 of 15 patients with non-A, non-B (NANB) chronic hepatitis and the only patient with cryptogenic cirrhosis had persistent antibodies to a nonstructural protein of HCV (C100-3). Seven of the 10 antibody-positive patients had HCV RNA detected by a highly sensitive PCR assay; two of the antibody-negative patients also had HCV RNA. The pattern of viral RNA and antibody was evaluated in one patient with acute posttransfusion NANB hepatitis (PTNANB) and in three chimpanzees with acute infection. Viral RNA was detected early after infection and, in each case, before the appearance of antibody. In one chimpanzee, viral RNA disappeared when antibody became detectable. The authors conclude that most patients with chronic NANB liver disease have HCV infection. Garson and colleagues found that 6 of 1,100 donor blood units were repeatedly positive for antibodies to the C100-3 antigen (anti-C100-3). Only one of these six donor units was found to contain HCV RNA by a “nested PCR” assay. Nested PCR refers to the use of two successive rounds of amplification, with the second round using primers internal to, or nested within, the original two primers. The positive donor unit was the only one of the six antibody-positive units to cause posttransfusion hepatitis in a recipient. Sera from three other confirmed cases of posttransfusion hepatitis were evaluated and in each case one donor unit of blood was found to contain HCV RNA and anti-C100-3. The authors concluded that anti-C100-3 may not be a good predictor of infectivity in blood donors and that direct identification of viral RNA may be required.

Hepatitis C Virus Antibody in Poland

To assess the epidemiology of hepatitis C virus (HCV) in Poland, anti-HCV was studied in patients with acute and chronic non-A, non-B (NANB) hepatitis, in healthy adults, and in subjects at risk. Anti-HCV prevalence was 2% in 152 blood donors, 78% in 95 parenteral drug addicts, 21% in 112 alcoholics, and 86% in 42 patients with chronic NANB hepatitis. Among 34 prospectively followed patients with acute NANB hepatitis 17 (50%) developed anti-HCV. It seems that HCV infection is responsible for the majority of NANB hepatitis in Poland and is common in parenteral drug abusers and alcoholics.

Anti-hepatitis C virus antibody prevails in fulminant hepatic failure

Serial serum samples obtained from 27 patients with fulminant hepatic failure (FHF) in a variety of etiology were tested for anti-hepatitis C virus antibody (anti-HCV). Seven out of 10 patients (70%) with FHF due to hepatitis B virus (HBV) infection were positive for anti-HCV, showing a significantly higher rate than that in acute HBV hepatitis (0/17, 0%): In particular, all 3 post-transfusion HBV-FHF cases were found to be positive for the antibody. The incidence of anti-HCV in sporadic non-A non-B (NANB)-FHF patients (7/11, 64%) tended to be greater than that in acute sporadic NANB hepatitis as recently surveyed in this country. In addition, anti-HCV was also detected in a patient with hepatitis A virus (HAV)-FHF and in 2 out of 4 drug-induced FHF patients. Moreover, anti-HCV appeared earlier in FHF (median; 27.5 days, n = 9) regardless of etiology, when compared to acute NANB hepatitis (3 to 6 months). Hence, co-infection and/or superinfection of HCV with enhanced antibody response may play an important role in the development of this fatal disease.

Serological diagnosis of infection with human herpesvirus type 6.

To identify clinical consequences of acute human herpesvirus type 6 infection by hypothesising that the virus will induce similar clinical syndromes to cytomegalovirus. Examination of consecutive serum samples from patients with illnesses compatible with acute cytomegalovirus infection or exanthem subitum by indirect immunofluorescence for the presence of antibodies to human herpesvirus type 6. An IgG absorption step was included to avoid false positive and negative results for IgM. The criterion standard for diagnosis of human herpesvirus type 6 infection was the presence of IgM human herpesvirus type 6 antibody (titre greater than 20) and a rising titre of IgG human herpesvirus type 6 antibody without serological evidence of alternative infection. Routine viral diagnostic and reference laboratory in the largest teaching hospital in Sydney. 341 Consecutive serum samples were analysed from patients with hepatitis (147 samples); infectious mononucleosis-like illness (106); screens for toxoplasma, other viruses, rubella, cytomegalovirus, and herpesvirus (38); fever in an immunocompromised patient (eight); unusual neurological (nine) or haematological syndromes (14); splenomegaly (six); and rash in a child (13). Three cases of acute human herpesvirus type 6 infection were identified: in one patient aged 65 with a previous diagnosis of acute non-A non-B hepatitis, one aged 25 with a glandular fever-like illness, and one aged 6 with a glandular fever-like illness. All three illnesses resolved completely. 15 Further serum samples were positive for human herpesvirus type 6 antibody but were also diagnostic for acute infection with other viruses (cytomegalovirus (nine), Epstein-Barr virus (three), and HIV (one] or had a titre of IgM human herpesvirus type 6 antibody less than 20 (two). Acute human herpesvirus type 6 infection in immunocompetent patients may result in a mononucleosis-like illness or an acute but self limiting hepatitis.

Successful treatment of a patient with posthepatitic severe aplastic anemia with bolus methylprednisolone.

A 31-year-old female developed severe aplastic anemia following acute non-A non-B hepatitis. Because of the impossibility of bone marrow transplantation, bolus methylprednisolone therapy was done. Soon after the therapy, pancytopenia improved. She has since taken mepitiostane and no longer needs blood transfusions. Good response to the immunosuppressive therapy suggests that the pathogenesis of this anemia is related to the immune system. In vitro culture demonstrated that the patient's serum contained a factor which suppressed hematopoietic activity, possibly gamma-interferon.