Hepatitis A epidemic in the elderly.

ABSTRACT

Background. The purpose of this study was to characterize acute hepatitis A in patients more than 40 years old.

Methods. Retrospective chart reviews and physician questionnaires were used. Hospitalization, laboratory tests, comorbid conditions, complications, and mortality were assessed.

Results. A total of 180 cases of acute hepatitis A occurred in patients more than age 40. Hospitalization rates increased from 3% for the group aged 40 to 49 to 12% in the group aged 50 to 69 and 42% in those [greater than or equal to] 70 years old. The mean peak total bilirubin value was higher in hospitalized patients than in nonhospitalized patients (12.4 mg/dL vs 7.7 mg/dL). Comorbid conditions included significant alcohol consumption (8%), diabetes (16.6%), cardiovascular disease (54%), and chronic obstructive pulmonary disease (COPD) (21%). Complications included pancreatitis (4.1%), ascites (8.3%), prolonged cholestatic features (7.6%), and death (0.8%).

Conclusion. Hospitalization rates and rare complications associated with acute hepatitis A increase with age.

**********

ACUTE HEPATITIS A viral infection is generally thought to be a benign, self-limited disease that affects children and young adults in underdeveloped countries, with occasional foodborne or waterborne outbreaks in the United States. (1) However, hepatitis A can be associated with severe extrahepatic manifestations as well as fulminant hepatitis. (24) Although the severity of acute hepatitis A increases with age, there is a paucity of data on clinical manifestations and mortality rates of acute hepatitis A in the elderly.

British investigators examined the incidence of infectious hepatitis as well as the mortality due to hepatitis A from 1979 through 1985. (5) Cumulative deaths from the entire period were summed for individuals in each of the age groups. During this period, the mortality rate was 0.03% to 0.06% in the group aged 25 to 35, increasing 100-fold to 3% to 6% for patients more than age 55 and approaching 15% in patients older than 75. (5) In another study, the mortality rate for patients more than 40 years of age was 5.7% during a 2-year urban epidemic. (3) Further-more, in a liver transplantation referral center, 57% of fulminant hepatitis A cases in patients more than 40 years of age were fatal (13/23 cases). (4) Acute hepatitis A case fatality rates in the United States from 1983 to 1987 were 0.004% in patients 5 to 14 years old and 2.7% in patients more than 49 years old, (6) confirming that hepatitis A fatality rates increase with age.

In addition, various clinical complications have been associated with hepatitis A including immune complex disease, (7) prolonged intrahepatic cholestasis, (8) cholecystitis, (9) and pancreatitis. (10-16) Hepatitis A-associated cholecystitis was first described in an elderly man in whom the hepatitis A viral antigen and hepatitis A virus (HAV) 1gM were observed in epithelial cells of the gallbladder wall. (9) Although cholecystitis might theoretically be associated with prolonged cholestasis, the latter entity has not been reported in an elderly population. In a 1984 series, 5 of 6 patients were younger than 45 years of age. (8) This apparent lack of association with advanced age could be spurious or could be the result of an immunologic phenomenon associated with age.

The association of hepatitis and pancreatitis was initially suspected in 1944 (10) and was confirmed in later years. (11-17) The severity of pancreatic involvement may range from an abnormal appearance on ultrasonography without clinical symptoms to frank pancreatitis. In cases of fulminant hepatic failure, the majority being of viral etiology, 33% to 35% of patients were found to have pancreatitis by biopsy in retrospective autopsy analysis. (12, 13) In a prospective study, investigators found hyperamylasemia in 14 of 160 patients with viral hepatitis, and in 10 of these patients, pancreatitis was confirmed at autopsy. (13) In 5 case series, only 3 of 20 patients were more than 45 years old. (12-17) Pancreatitis appears to be an unusual occurrence in the elderly, with a reported rate of 14% to 15% in patients more than 45 years old.

The purpose of this study was to explore not only the mortality rate in the elderly, but also the clinical features of hepatitis A during a single source outbreak in Dallas County. The estimates were based on the number of cases and deaths reported to the Dallas County Health Department during the epidemic that lasted approximately 20 days and was associated with a food worker at the restaurant. The study also compared the clinical features of patients who were hospitalized with those of patients who were not hospitalized during the epidemic. In this epidemic, we found that elderly patients were more likely to have more severe disease, as confirmed by increased hospitalization rates; higher bilirubin levels and prothrombin times; and more clinical complications, such as prolonged cholestasis, pancreatitis, and ascites.

METHODS

From February 12, 1999, to March 10, 1999, 232 cases of acute hepatitis were reported to the Dallas County Health Department. Of these, 180 cases occurred among people more than 40 years old. The outbreak was traced to a single source, a restaurant in Dallas County.

A positive test for IgM antibody to hepatitis A was required to diagnose an acute hepatitis A infection in conjunction with compatible symptoms and history of eating at the single source eating establishment. Data were obtained on the need for hospitalization, underlying comorbid conditions (chronic liver disease, alcohol consumption >50 g/day for 5 years, cardiovascular disease, cancer, immunosuppression, and diabetes), complications (acalculous cholecystitis, hemolysis, autoimmune hepatitis, pancreatitis, and prolonged cholestatic features), and mortality. The height and duration of elevated aminotransferase levels, total bilirubin values, and prothrombin times were evaluated. Twenty-four medical charts of patients who required hospitalization were reviewed, and 22 telephone interviews with physicians of patients who did not require hospitalization were conducted. The 24 medical charts were specifically examined for evidence of rare extrahepatic manifestations of hepatitis A. All records for the patients ho spitalized in the epidemic were located and reviewed.

These data were entered into a database and then summed. The results were analyzed by UTSTAT program using Ngroup analysis, which uses the Student's t test to determine significance.

RESULTS

We identified 232 patients with acute hepatitis A during this single source outbreak reported to the Dallas County Health Department and documented by tests for HAV IgM antibody. Hepatitis A was noted in one of the food handlers at the single source (restaurant). Of the 180 patients who were >40 years of age, most were white (175 whites versus 5 African Americans); 95 were men and 85 were women. Almost equal percentages of patients were aged 40 to 49 (64 patients), 50 to 59 (57 patients) and >60 (59 patients). There were 26 patients more than 70 years old.

Higher rates of hospitalization occurred with each decade of life (Table 1). Of the 24 hospitalized patients over age 40, 13 were female and 11 were male. Most of the hospitalized men were more than 70 years old (Table 1). There was a significant difference in rates of hospitalization before and after the age of 60 years (9/121 [7%] versus 15/59 [25%]); (P = .001). In the age group >70 years, 42% (11/26) of the patients required hospitalization, and the 2 deaths occurred in men from this group. These results indicate that the elderly had an increased rate of hospitalization and death during this single source outbreak of hepatitis A.

The hospitalized patients had significantly higher levels of serum bilirubin than the nonhospitalized patients (P = .005, Fig 1). The bilirubin values ranged from 3.8 to 51.0 mg/dL in the hospitalized patients (n = 16) and 0.8 to 14.3 mg/dL in the nonhospitalized patients (n = 19). In contrast to the bilirubin levels, no significant difference was noted in the mean aspartate aminotransferase (AST) values for the hospitalized and nonhospitalized patients (2,200 versus 1,700 IU/L, Fig 1). The AST values ranged from 169 to 7,409 IU/L in the hospitalized patients (n = 14) and from 46 to 4,450 IU/L in the nonhospitalized patients (n = 19).

Some elderly patients had prolonged elevation of both bilirubin and aminotransferases. In one 81-year-old male patient, the AST and alanine aminotransferase (ALT) levels remained abnormal for more than 100 days (Fig 2). The total bilirubin did not return to normal for nearly 2 months after the initial presentation (Fig 2). An elderly female patient had elevated liver test results for >4 weeks (initial values--AST 2,200 IU/mL, ALT 1,500 IU/mL and total bilirubin 28 mg/dL; values 4 weeks later-- AST 210 IU/mL, ALT 217 IU/mL, and total bilirubin 4.7 mg/dL). These data suggest that elderly patients may have prolonged illness associated with hepatitis A. Of the 26 patients >70 years old, 2 (7.6%) had prolonged elevation of the bilirubin level, whereas none of the 64 patients <50 years of age exhibited this finding.

Among both hospitalized and nonhospitalized patients, the older patients had higher bilirubin levels than the younger ones (Fig 3). The peak bilirubin in the group aged >70 was 51 mg/dL in the hospitalized patients versus 14.3 mg/dL in the nonhospitalized patients. There was significant difference between the bilirubin values in the 17 hospitalized patients >60 years old (range, 4.9 to 51 mg/dL; mean 16.5 [+ or -] 2.84 mg/dL) and the 9 hospitalized patients <60 years old (range, 6.4 to 14.1 mg/dL; mean 8.77 [+ or -] 1.3 mg/dL, P = .032). Furthermore, there was a significant difference in the bilirubin values in the 8 nonhospitalized patients aged [greater than or equal to] 60 years (range, 0.8 to 16.5 mg/dL) and the 15 nonhospitalized patients aged <60 (range, 0.8 to 13.8 mg/dL; mean 5.52 [+ or -] 0.8927 mg/dL, P = .01). These data suggest that in both hospitalized and nonhospitalized patients, the bilirubin level trended upward with age. These data also suggest that even without severe hepatitis necessitating hospitalization, the bilirubin level is higher in the elderly.

There was also an upward trend in the prothrombin time in 19 hospitalized patients. The prothrombin time ranged from 10.5 to 47.0 seconds in the patients aged [greater than or equal to]60 (mean 19.97 [+ or -] 3.19 seconds) and from 11.8 to 22.0 seconds in the patients aged <60 (mean 14.3 [+ or -] 1.19 seconds). These data suggest that prothrombin time in the hospitalized patients also increases with age in patients with hepatitis A (Fig 4).

Preexisting underlying medical problems in the patients requiring hospitalization because of hepatitis A included cardiovascular disease in 13 patients, diabetes in 4, COPD in 5, cancer in 3, and significant alcohol consumption in 2. Furthermore, most of the patients with cardiovascular disease and diabetes were >70 years old. Of the 2 patients with cardiovascular disease, 1 had hemochromatosis and 1 was HIV positive (Table 2).

The complications noted in the hospitalized patients included ascites (2 patients, 73 and 80 years old), prolonged cholestatic features (2 patients, 80 years old), and pancreatitis (1 patient, 73 years old). The pancreatitis was diagnosed by hyperamylasemia. Of the patients >60 years old, 2 (3.4%) died of fulminant hepatic failure (71 and 80 years old). None of the patients <60 years of age had any of these clinical manifestations.

DISCUSSION

This is the first report of a single source outbreak of hepatitis A in which the large majority (180/232, 78%) of the patients were aged 40 years and older. This observational study of outcome of acute hepatitis A in patients >40 years old revealed a number of important trends in the disease process in an aged population. As in other recent epidemics, there was no female or male predominance. (3) Unlike other recent epidemics in the United States however, this one affected predominantly white patients. (2,3,18-20) The population in this outbreak was older, with a median age of 55 years, compared with 26 to 32 years in other recent epidemics. (3) This age difference may be due to the source of the outbreak (a food establishment popular among the area's elderly people) or to the shift in hepatitis A prevalence.

Similar to epidemics in other developed countries, this epidemic suggests that many elderly individuals in this country did not acquire immunity by childhood infection with HAV. The proportion of aged patients affected by foodborne hepatitis A may continue to expand after the elimination of person-to-person epidemics among children. (21) Furthermore, the prevalence of the hepatitis A antibody has been examined in 7,196 individuals in England. (22) In 1994, the prevalence of antibody positivity was 89.5% among patients >60 years of age, and 11% of the population >60 years of age were susceptible to hepatitis A infection. (22)

With the shift of the hepatitis A outbreaks toward the elderly population, examining the clinical responses of the elderly to HAV will be important. As early as 1965, Fenster (23) reported that the mortality rate from viral hepatitis in the elderly was approximately 24%. He suggested that the poor prognosis correlated with subacute hepatic necrosis, poor nutrition, and associated diseases such as cancer, cardiovascular, or chronic pulmonary disease. (23) Others have reported that the mortality rate in the aged has decreased from 52% in 1947, the post-World War II era, to 20% in 1965, (10) and 3% or 4% between 1960 and 1970. (24) However, as recently as 1999, hepatitis A was observed to be an agent of fulminant hepatitis in developed countries. In 1999, 15 patients with hepatitis A were admitted to the Kings College Liver Failure Unit in England. The 7 who died had a higher median age (47 years) than the survivors (33.5 years). (5)

Death from cerebral edema during acute liver failure appears to be more frequent among the young. (25) Fulminant hepatic failure in the elderly is often associated with renal failure, sepsis, and multiple system organ failure. (25) Importantly, in the epidemic we are reporting, the 2 patients with fulminant and subfulminant hepatic failure evidenced no cerebral edema, posturing, or uncal herniation. Mortality associated with acute liver failure due to acetaminophen and other drug toxicities has decreased over the past three decades. (26) Mortality due to acute liver failure from hepatitis A appears to have remained unchanged since the 1960s and 1970s. Specifically in the elderly, these rates may be associated with their coexisting illnesses.

Furthermore, the hospitalization rate and the total bilirubin level increased with advancing age. Since we have the admitting diagnosis for all hospitalized patients, we know that these patients were not admitted for other comorbid conditions, but for acute hepatitis. Bilirubin was also lower in the nonhospitalized patients than in the hospitalized patients. Less frequent laboratory testing may account for some of the differences, but it is unlikely to account for all of the differences.

Complications also occurred more frequently in the elderly. One of our elderly patients had evidence of pancreatitis. In one study, 3.7% of patients found to have pancreatitis at autopsy were noted to have evidence of viral hepatitis, (27) as compared with overall autopsy incidence of viral hepatitis of 0.9%. (12) Pancreatitis in patients with hepatitis A-related fulminant hepatic failure (FHF) is often observed, but this extrahepatic manifestation is a rare entity in hepatitis A or hepatitis B without FHF. (12-16,28-32)

In another report on FHF, 21 patients with FHF were admitted with acute or subacute hepatic necrosis. The patients were highly selected with grade II to IV encephalopathy at some stage of their illness, and ranged in age from 40 to 94 years. Eleven of the 21 patients had autopsy evidence of pancreatitis, ranging from severe hemorrhagic to mild pancreatitis, suggesting that pancreatitis may be seen in more than 50% of the patients more than 40 years old with FHF. (12) Similarly, in another case series, 8 of 18 patients with fulminant viral hepatitis had changes characteristic of acute pancreatitis. (29) Conversely, only 2 patients with acute pancreatitis were observed in another series of 33 patients with fatal liver necrosis attributed to halothane. (10) In our series, the 73-year-old white woman who exhibited hyperamylasemia had hepatic encephalopathy, suggesting that in this nonfatal case of hepatitis A, there was evidence of FHF. The associated pancreatitis in this individual as in others may be a marker o f severe disease, suggesting that elderly patients with severe disease may have pancreatitis.

Cholestasis was observed in 2 of 26 patients (7.6%) >70 years of age in this epidemic. Previously reported data showed that 84% of the patients with prolonged cholestasis were younger than 45 years of age. (8) Although hepatitis A generally has a brief clinical course, with only 8.5% of patients reportedly having an elevated aminotransferase level after 14 weeks, jaundice has been reported to last from 80 to 125 days. (8) In one of the elderly patients in our epidemic, the duration of hyperbilirubinemia was approximately 120 days. HAV-associated RNA and antigen has been observed in hepatocytes, Kupffer cells, gallbladder, and bile duct epithelium. The latter two tissues were examined because of a patient with HAV-associated cholecystitis and the recurrence of hepatitis A after liver transplantation suggesting that HAV might infect cells other than hepatocytes. Thus, direct infection of the biliary epithelium may have a direct role in the pathogenesis of HAV-associated cholangiopathic infection. In patients >7 0 years of age infected with hepatitis A, a similar percentage as younger patients may have prolonged cholestasis, possibly due to direct invasion of HAV into the gallbladder or bile duct epithelium.

Hepatitis A may be more severe in the elderly because of specific immunologically mediated processes. Aging of the T cells that clear viruses may in fact be important in this epidemic of hepatitis A. Investigators have described similar aging of cytotoxic T cells that clear the influenza virus and other infectious diseases. (33-35) Furthermore, since the inflammatory infiltrate around and within the biliary epithelium consisted almost exclusively of lymphocytes, cell-mediated immunity may also have a role in the pathogenesis of hepatitis A--associated cholangiopathies.

Considering the severity of the disease in the elderly and the ease of vaccination, we believe that elderly individuals who are not immune should be vaccinated. An inactivated vaccine against hepatitis A as a single dose at month 0 with a booster at month 6 has been administered to 200 subjects (134 patients aged 20 to 39 and 66 patients aged 40 to 62). At month 7, 100% in both groups had become seropositive. The vaccine was well tolerated and did not cause any severe reactions. This hepatitis A vaccine offers protection for at least 6 months in the majority of cases and may be used in an older population. (36)

CONCLUSION

Hepatitis A is usually a self-limiting disease of children and young adults, but fatal cases can occur in the elderly. During this epidemic of acute hepatitis A, nearly one third of those infected were more than 60 years old. Similar to estimated fatality rates among the elderly in the 1960s, the case fatality rate in patients more than the age of 60 was 3.0%. Thus, mortality after acute hepatitis A infection in the elderly appears unchanged over the past 30 years. Furthermore, complications such as pancreatitis and prolonged intrahepatic cholestasis may also be observed in the elderly and may be markers of severity of disease.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

TABLE 1

Epidemiology of the Single Source Outbreak of Hepatitis A


Sex
 Male         95
 Female       85

Race
 White       175
 Black         5

Age (years)
 40-49        64
 50-59        57
 60-69        33
 >70          26
Hospitalization (by age)  Women  Men     Total

     40-49                  2     0    (2/64, 3%)
     50-59                  3     4    (7/57, 12%)
     60-69                  4     0    (4/33, 12%)
     70-79                  4     7    (11/26, 42%)
TABLE 2

Underlying Medical Conditions

                                        Age (years)
                      40-49     50-59     60-69      >70      Total
Condition            (n = 2)   (n = 7)   (n = 4)   (n = 11)  (n = 24)

Cardiovascular          2         2 *       2 +       7        13
  disease
Diabetes                0         1         0         3         4
COPD                    0         2         2         1         5
Cancer                  0         2         0         1         3
Significant alcohol     0         1         0         1         2
  consumption

* HIV positive.

+ Hemochromatosis.

References

(1.) Hutlin Y, Pool V, Cramer E, et al: A multistate, foodborne outbreak of hepatitis A. N Engl J Med 1998; 340:595-602

(2.) Gocke D: Hepatitis A revisited. Ann Intern Med 1996; 105:960-961

(3.) Wilner I, Uhl M, Rely C, et al: Serious hepatitis A: an analysis of patients hospitalized during an urban epidemic in the United States. Ann Intern Med 1998; 128:111-114

(4.) Gimson A, White Y, Eddleston A, et al: Clinical and prognostic differences in fulminant hepatitis type A, B and non A, non-B. Gut 1983; 24:1194-1198

(5.) Forbes A, Williams R: Increasing age. an important adverse prognostic factor in hepatitis A virus infection. J R Coll Physicians Lond 1988; 22:237-239

(6.) Hadler SC: Global impact of hepatitis A virus infection changing patterns. Viral Hepatitis and Liver Disease. Hollinger FB, Lemon SM, Margolis H (eds). Baltimore, Williams & Wilkins Co, 1990, pp 14-20

(7.) Inman RD, Hodge M. Johnston ME, et al: Arthritis, vasculitis and cryoglobulinemia associated with relapsing hepatitis A virus infection. Ann Intern Med 1986; 105:700-703

(8.) Gordon SC, Reddy KR, Schiff L, et al: Prolonged intrahepatic cholestasis secondary to acute hepatitis A. Ann Intern Med 1984; 101:635-637

(9.) Mourani S, Dobbs SM, Genta RM, et al: Hepatitis A virus-associated cholecystitis. Ann Intern Med 1994; 120:398-400

(10.) Lisney AA: Infective hepatitis in Leicestershire: a survey of 1062 cases. Proc R Soc Med 1944; 37:165

(11.) Geokas MC, Olsen H, Swanson V et al: The association of viral hepatitis and acute pancreatitis. Calif Med J 1972; 117:1-7

(12.) Ham JM, Fitzpatrick P: Acute pancreatitis in patients with acute hepatic failure. Am J Dig Dis 1973; 18:1079-1083

(13.) Mishra A, Saigal S, Gupta R, et al: Acute pancreatitis associated with viral hepatitis: a report of six cases with review of literature, Am J Gastroenterol 1999; 94:2292-2295

(14.) Davis TV, Keeffe EB: Acute pancreatitis associated with acute hepatitis A. Am J Gastroenterol 1992; 87:1648-1650

(15.) Klar A, Branski D, Nadjari M, et al: Gallbladder and pancreatic involvement in hepatitis A. J Clin Gastroenterol 1998; 27:143-145

(16.) Shrier LA, Karpen SJ, McEvoy C: Acute pancreatitis associated with acute hepatitis A in a young child. J Pediatr 1995; 126:57-59

(17.) Buntain WL, Wood JB, Woolley MM: Pancreatitis in childhood. J Pediatr Surg 1978; 13:143-149

(18.) Osmon DR, Melton J III, Keys TF, et al: Viral hepatitis. a population-based study in Rochester, Minn, 1971-1980. Arch Intern Med 1987; 147:1235-1240

(19.) Shaw FE Jr, Sudman JH, Smith SM, et al: A community-wide epidemic of hepatitis A in Ohio. Am J Epidemiol 1986; 123:1057-1065

(20.) Bergeisen GH, Hinds MW, Skaggs JW: A waterborne outbreak of hepatitis A in Meade County, Kentucky. Am J Public Health 1985; 75:161-164

(21.) Hayashi H, Yagi A, Ichimiya H, et al: An outbreak of foodborne hepatitis A in a factory: a possible shift in age of patients in Japan. Int J Epidemiol 1988; 17:870-873

(22.) Gay NJ, Morgan-Capner P. Wright j, et al: Age-specific antibody prevalence to hepatitis A in England: implications for disease control. Epidemiol Infect 1994; 113:113-120

(23.) Fenster LF: Viral hepatitis in the elderly. Gastroenterology 1965; 49:262

(24.) Gibinski K, Fojt E, Suchan L: Hepatitis in the aged. Digestion 1973; 8:254-260

(25.) Schiodt FV, Atillasoy E, Shakil AO, et al: Etiology and outcome for 295 patients with acute liver failure in the United States. Liver Transplant Surg 1999; 5:86-89

(26.) Devlin J, Williams R: Prognosis and consideration of transplantation. Acute Liver Failure. Lee W, Williams R (eds). Cambridge, England, Cambridge University Press, 1997, pp 173-185

(27.) Joshi RA, Probsterin JG, Blumenthal HT: A survey of experiences with three hundred clinical and one hundred and eight autopsy cases of acute pancreatitis. Am Surg 1957; 23:24

(28.) Parbhoo SP, Welch J, Sherlock S: Acute pancreatitis in patients with fulminant hepatic failure. Gut 1973; 14:428

(29.) Wands JR, Salyer DC, Boitnott JD et al: Fulminant hepatitis complicated by pancreatitis. Johns Hopkins Med J 1973; 133:156-160

(30.) Lopez Morante A, Rodriques de Lope C, San Miguel G, et al: Acute pancreatitis in hepatitis A infection. Postgrad Med J 1986; 62:407-408

(31.) Amarapurkar DN, Begani MM, Mirchangani K: Acute pancreatitis A infection. Trop Gastroenterol 1996; 17:30-31

(32.) Eugene C, Cadranel JF, Bergue A, et al: Acute pancreatitis associated with non-A, non-B hepatitis. J Clin Gastroenterol 1990; 12:195-197

(33.) Powers DC: Influenza A virus-specific cytotoxic T lymphocyte activity declines with advancing age. J Am Geriatr Soc 1993; 41:1-5

(34.) Schwab R, Szabo P, Manavalan JS, et al: Expanded CD4+ and CD8+ T cell clones in elderly humans. J Immunol 1997; 158:4493-4499

(35.) Fantin B, Joly V, Elbim C, et al: Lymphocyte subset counts during the course of community-acquired pneumonia: evolution according to age, human immunodeficiency virus status, and etiologic microorganisms. Clin Infect Dis 1996; 22:1096-1098

(36.) Briem H, Safary A Immunogenicity and safety in adults of hepatitis A virus vaccine administered as a single dose with a booster 6 months later. J Med Viral 1994; 44:443-445

RELATED ARTICLE: KEY POINTS

* Higher rates of hospitalization occurred with each decade of life.

* Some elderly patients were noted to have prolonged elevation of both bilirubin and aminotransferases.

* Older patients were observed to have higher bilirubin levels compared with the younger patients, both in hospitalized and nonhospitalized cases.

* The complications noted in the elderly hospitalized patients included ascites, prolonged cholestatic features, and pancreatitis.

From the Division of Digestive and Liver Diseases, Department of Internal Medicine, University of Texas Southwestern Medical Center at Dallas.

Reprint requests to Geri R. Brown, MD, University of Texas Southwestern Medical Center at Dallas, Division of Digestive and Liver Diseases, Department of Internal Medicine, 5323 Harry Hines Blvd, Dallas, TX 75390-9151.