When Mononucleosis Does Not Get Better

Abstract & Commentary

By Joseph F. John, Jr., MD, FACP, FIDSA, FSHEA, Chief of Staff for Education, Medical Subspecialty Services, Ralph H. Johnson Veterans Administration Medical Center, Professor of Medicine, Medical University of South Carolina, Charleston, is Associate Editor for Infectious Disease Alert.

Dr. John does research for Merck, is a consultant for Cubist, Roche, and bioMerieux, and is on the speaker’s bureau for Pharmacia, GSK, Merck, Bayer, and Wyeth.

Synopsis: The resolution of symptomatic IM is not determined by control of viremia, nor is it easily explained by altered host responses to EBV infection. The detailed determinants of delayed recovery remain to be elucidated.

Source: Cameron B, et al. Prolonged Illness After Infectious Mononucleosis is Associated with Altered Immunity But Not with Increased Viral Load. J Infect Dis. 2006;193:664-671.

When Primary Ebstein-Barr Infection occurs in early childhood, it is usually asymptomatic. Infection later, during adolescence or in adulthood, is associated with a classic syndrome that we term infectious mononucleosis (IM), mono in the vernacular. There are patients whose mono does not resolve over a month or 6 weeks are considered to have chronic active Ebstein-Barr Virus infection (CAEBV). That type of prolonged illness—one in which fatigue and other symptoms persists more than 6 months—was the subject of an Australian study. The study evolved from a larger cohort being studied for acquisition of various infections over a long period time, known as the Dubbo Infection Outcome Study.

Participants in the Dubbo study were enrolled if their IgM antibody to EBV viral capsid antigen (VCA) was positive and if they had symptoms of IM. A health questionnaire known as SOMA-6 was used to follow the general symptoms of the cohort. A SOMA score of > 3 indicates severe disability. A control cohort was a group of IM patients who returned to health within 2 months. Assessments were performed on patients at 2-3 weeks, 4-6 weeks, and at 3 months. For those patients with symptoms > 3 months, further evaluation was performed, including detailed testing of EBV viral load, tests of humeral and cellular immunity, flow cytometry, and interferon-gamma determination (IFN-g). Controls were HLA matched.

The results were very interesting. Only 8 patients had prolonged illness. They were compared in the study to 31 controls. At the time of enrollment, the prolonged cases compared to controls had been in bed for a mean of 6 days and out of their normal roles for an average of 14 days. The mean time of illness for cases was 34 weeks compared to 8 weeks for controls. The mean time in bed for cases vs controls was 21 vs 5 days respectively.

IM is a disease that triggers a brisk cytotoxic lymphocytic response primarily with EBV-specific CD8+ cells, a humoral response to EBV nuclear antigens, and reduction of a measurable circulating viral load. For case patients as compared to controls, there was no abrogation of the CD8+ cell response, and the pattern of decline to normal was the same in both groups. The cases were more likely to show a blunted CTL response to latent antigens. Also, cases were more likely to have higher anti-EBNA at 3-6 weeks and at 3 months than the controls. Controls were more likely to develop quicker responses to specific lytic antigens. There was no significant difference in viral load between the 2 groups.


This study helps resolve the issue of why some young adults (mean age of 23 years in this study) resolve their primary EB virus infection very slowly. Common observation is that a youngster with mononucleosis gets better in about 2 months. Controls without prolonged illness health returned in about 8 weeks compared to 4 times as long in cases. Repeated psychological testing showed no differences between the 2 groups that would ascribe the chronic symptoms to alterations in mood.

IM also produces a peak gamma-interferon response in about 4-8 weeks, as measured for controls also in this study. Cases were more likely to take on the average of 30 weeks to reach a peak interferon response. In contrast, cases were more likely to make a slower antibody response to EBNA. Cameron and colleagues emphasize that this finding contradicts classic dogma (Henle, et al Proc Natl Acad Sci. USA 1987) that linked convalescence to a rising anti-EBNA, whereas that rising or persisting response to EBNA represents a delay in the humoral response.

EBV viral loads are now available at many tertiary care centers. It will be tempting to measure these levels in patients with unresolving IM. Nevertheless, this current study by Cameron et al found that their viral loads mitigated as early in cases as in controls, suggesting that it is not the virus per se that produces prolongation of the illness. It is more likely the differences in cytokine responses explain more of the differences in symptom recovery from IM.

It is interesting to me that many patients with chronic fatigue who may eventually be diagnosed as having chronic fatigue syndrome present at some point in their illness with elevated humoral responses to EBV. Studies have measured elevated interferon levels in patients with CFS. In Japan, CFS patients are referred to as having cytokine disease. The exact role of EBV in CFS remains elusive. At the very least with the development of numerous biologics that influence cytokine production, such agents may prove worthwhile to test in patients with prolonged EBV and possibly in patients with CFS. It is unlikely that chronic symptoms of lassitude, fatigue, and the need to remain in bed will be due to the effects of persisting whole virus.