Frequency of 'Chronic Active Epstein-Barr Virus Infection' in a General Medical Practice

Twenty-one percent of 500 unselected patients, aged 17 to 50 years, seeking primary care for any reason were found to be suffering from a chronic fatigue syndrome consistent with "chronic active Epstein-Barr virus (EBV) infection." They had been experiencing "severe" fatigue, usually cyclic, for a median of 16 months (range, six to 458 months), associated with sore throat, myalgias, or headaches; 45% of the patients were periodically bedridden; and 25% to 73% reported recurrent cervical adenopathy, paresthesias, arthralgias, and difficulty in concentrating or sleeping. The patients had no recognized chronic "physical" illness and were not receiving psychiatric care. While antibody titers to several EBV-specific antigens were higher in patients than in age-and sex-matched control subjects, the differences generally were not statistically significant. A chronic fatigue syndrome consistent with the chronic active EBV infection syndrome was prevalent in our primary care practice. However, our data offer no evidence that EBV is causally related to the syndrome. Indeed, we feel that among unselected patients seen in a general medical practice currently available EBV serologic test results must be interpreted with great caution. (

Frequency of 'Chronic Active Epstein-Barr Virus Infection' in a General Medical Practice Dedra Buchwald, MD; John L. Sullivan, MD; Anthony L. Komaroff, MD Twenty-one percent of 500 unselected patients, aged 17 to 50 years, seeking primary care for any reason were found to be suffering from a chronic fatigue syndrome consistent with "chronic active Epstein-Barr virus (EBV) infection."They had been experiencing "severe" fatigue, usually cyclic, for a median of 16 months (range, six to 458 months), associated with sore throat, myalgias, or headaches; 45% of the patients were periodically bedridden; and 25% to 73% reported recurrent cervical adenopathy, paresthesias, arthralgias, and difficulty in concentrating or sleeping.The patients had no recognized chronic "physical" illness and were not receiving psychiatric care.While antibody titers to several EBV-specific antigens were higher in patients than in age-and sex-matched control subjects, the differences generally were not statistically significant.A chronic fatigue syndrome consistent with the chronic active EBV infection syndrome was prevalent in our primary care practice.However, our data offer no evidence that EBV is causally related to the syndrome.Indeed, we feel that among unselected patients seen in a general medical practice currently available EBV serologic test results must be interpreted with great caution.
The clinical manifestations of initial in¬ fection are variable; some patients are asymptomatic, others develop mild res¬ piratory infection symptoms, and oth¬ ers develop the clinical, sérologie, and hématologie features of infectious mononucleosis. 13When there are symp¬ toms associated with the initial infec¬ tion, these usually resolve within a mat- ter of weeks.However, the virus can be isolated from saliva for months or years thereafter, suggesting chronic infection of the nasopharyngeal epithelium and/ or salivary glands.4Also, a small frac¬ tion of B lymphocytes remain latently infected, apparently for the rest of the patient's life.5 See also pp 2297 and 2335.
On occasion, individuals with various forms of immunodeficiency will be un¬ able to contain the population of EBV- infected B cells, and they will go on to develop polyclonal B-cell proliferation and monoclonal B-cell lymphomas.6"16 Several investigators have reported another kind of chronic illness associ¬ ated with, and perhaps secondary to, persistent EBV infection.The syn¬ drome is called chronic active EBV (CEBV) infection or chronic mono¬ nucleosis, and it is characterized by varying degrees of chronic fatigue, fever, pharyngitis, myalgias, headache, arthralgias, paresthesias, depression, and cognitive deficits.While the full syndrome has been described only in recent years,1723 earlier reports that may have been reporting the same phe¬ nomenon exist in the literature.24'36The onset of the syndrome typically seems to be in late adolescence or young adult¬ hood,1723 although it may also occur in childhood. 22By definition, patients with this syndrome have been evaluated for a variety of chronic infectious, rheu- matologic, endocrinologie, and malig¬ nant diseases, and no chronic disease is apparent.17,2123The diagnosis has been made about twice as often in women as in men.17 '20"23 The illness may follow a documented episode of infectious mono¬ nucleosis, or an acute viral syndrome for which the patient has not sought medical attention, but also may appear spontaneously.172123For most patients, the illness takes the form of a chronic, recurring "fiulike" illness.21"23Virtually all patients perceive themselves to be impaired in some way.Some patients are completely disabled by the fatigue, muscular weakness, and pain.
The results of laboratory tests gener¬ ally ordered in patients with chronic fatigue are typically unremarkable.However, three recent reports all state that patients with this chronic fatigue syndrome have significant elevation of IgG antibody to the EBV viral capsid From the Division of General Medicine and Primary Care, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston (Drs Buchwald and Komaroff); and the Department of Pedi- atrics, Molecular Genetics, and Microbiology, Univer- sity of Massachusetts, Worcester (Dr Sullivan).
Reprint requests to Division of General Medicine and Primary Care, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115 (Dr Komaroff).
antigen (VCA) and to EBV early anti¬ gens (EA-Ab), in contrast with "healthy" patients.2123Also, one ofthese reports has suggested the possibility that levels of antibodies to EB nuclear antigen (EBNA-Ab) are low in a subset of patients.22Case reports also have described patients with extraordinarily high titers ofVCA-IgG and EA-Ab.  The hititers to VCA and EA typical of patients with the syndrome also can be seen in an occasional healthy patient.40"42Also, not all patients re¬ ported to have the syndrome have ele¬ vated titers.2123Because fatigue is a common present¬ ing complaint in general medical prac¬ tices, the general physician would like to know how common CEBV infection is among patients seen in a general prac¬ tice setting, lb our knowledge, previous published reports have not systemati¬ cally addressed this question but have described only patients who in some way have come to the attention of the investigators.The general physician also would like to know the utility of sérologie testing for EBV in a primary care practice setting; the patients re¬ ported in the literature have been highly selected and may have different clinical and sérologie characteristics.
To address these two questions, we conducted a prospective study in a pri¬ mary care practice setting.

Study Setting and Patient Population
The study was conducted over a six- month period in a large (30 000 visits per year) primary care general medical practice located in a large academic teaching hospital (Brigham and Women's Hospital, Boston).The prac¬ tice was staffed by 12 faculty members, four fellows, 85 house officers, and four nurse practitioners.Prior to and during the time of the study, the interest of the authors in the CEBV infection syn¬ drome was not widely known, and the practice was not being asked to see patients with CEBV infection in con¬ sultation.
The practice is divided into six areas, each with its own waiting room.Pa¬ tients are scheduled to each area during 9 four-hour sessions per week.During a four-month period, research assistants were assigned to interview patients in a particular waiting room, during a par¬ ticular session.They were asked to in¬ terview as many patients between the ages of 17 and 50 years during each session as time would allow.The waiting rooms and sessions were rotated during the four months of the study as to be representative of the entire practice.
During each interview session, pa¬ tients were asked, prior to seeing their physician, the following question: "Have you had severe loss of energy or easy fatigability either constantly or recurrently for at least the past six months?"Patients who answered af¬ firmatively were then asked: "Since you've been tired, have you had any of the following symptoms more than once: sore throat, muscle aches, or headaches?"The patients were also asked if they were pregnant, if they were taking corticosteroids or azathioprine, or if they had any of the following chronic diseases: diabetes, chronic lung disease, cancer, heart dis¬ ease, anemia, kidney disease, liver dis¬ ease, thyroid disease, rheumatoid ar¬ thritis, systemic lupus erythematosus, or other autoimmune disorders.Each patient's medical record was reviewed in detail with reference to each of the above questions, and to determine whether they were under psychiatric care at the time of the onset of their chronic fatigue syndrome.

Clinical and Laboratory
Data Collected Clinical Data.-We asked the fol¬ lowing patients to participate in the study: patients who stated that they did have a severe loss of energy for at least the past six months; and who had either recurrent sore throat, myalgias, or headaches; and who were not pregnant, nor taking immunosuppressive drugs, nor suffering from any of the above chronic diseases.Each patient gave written informed consent.A detailed questionnaire was then completed that inquired into present and past medical history.The patient's physician was asked to record the presence or absence of various physical examination findings on a study protocol.
Laboratory Data.-From each pa¬ tient, a single serum specimen was ob¬ tained at the index visit.Titers of IgM and IgG antibodies to VCA (VCA-IgM and VCA-IgG), as well as IgG anti¬ bodies to the early antigens (EA-Ab), were determined by indirect immuno- fluorescence.43 No attempt was made to distinguish the restricted and diffuse components of EA-Ab.The EBNA-Ab were detected by anticomplement im- munofluorescence.44Titers were re¬ ported as zero when VCA-IgG was present in a dilution of less than 1:20; when VCA-IgM was present in a dilu¬ tion of less than 1:20; when EA-Ab was present in a dilution of less than 1:20; and when EBNA-Ab was present in a dilution of less than 1:5-due to unre¬ liability of measurements at these low levels of antibody.All antibody titers were performed in a single laboratory by a single technician who routinely performs about 1000 EBV antibody pro¬ files per year.Control Subjects.-Fromthe am¬ bulatory practices of the Brigham and Women's Hospital, a group of "control" subjects matched for age (within five years) and sex to each of the study patients was identified.Control sub¬ jects were selected from patients sched¬ uled to have a venipuncture performed.Typically, they were attending the prac¬ tice for routine screening examinations; each subject was explicitly asked about, and explicitly denied, the presence of severe, prolonged chronic fatigue.Also, through a detailed review of the medical record, each control patient was deemed to be not pregnant, free of the chronic diseases listed above, and not receiving immunosuppressive drugs.Serum specimens obtained from each subject were tested for antibodies to EBV as described above.The techni¬ cian tested patient and control subject serum specimens simultaneously and was blinded to the patient or control status of the serum specimens.Statistical Analysis Calculation of Geometric Mean Titers (GMTs).-TheGMTs in the pa¬ tients were compared with those in the control subjects using both parametric and nonparametric methods.The com¬ parisons were made for the entire group of patients and control subjects and for each of several patient subgroups and their matched control subjects.Using parametric methods (Student's paired t test, two tailed), we first attempted to normalize the distribution; the value of each sérologie result was log trans¬ formed (base 10).Patients with "zero" antibody detected were given an arbi¬ trary value of 1 instead of 0, so that GMTs could be determined on the en¬ tire population, not just on patients with seropositivity.We also compared antibody levels in patients and control subjects by use of the nonparametric paired Wilcoxon rank-sum test.
Percentages.-Percentages of pa¬ tients with a certain sérologie finding were compared with the percentage of control subjects with that finding, using the x2 or Fisher's exact test (two tailed).

Study Population
Altogether, 500 patients were inter¬ viewed.They were representative of all patients between the ages of 17 and 50 years seen in our practice during the period of the study, as determined from a computerized patient registration and billing system: there were no significant Downloaded From: by a Harvard University User on 06/24/2018 Can do all the things I usually do at home or work, but feel much more easily fatigued from it: no energy left for anything else 24 Description of the frequency of the fatigue Constant fatigue that does not change 6 Always some fatigue that may get better but never goes away completely 9 The fatigue alternates with periods of feeling normal The problem has caused problems or stress at home and at work 24 Patient has sometimes thought these symptoms "might just all be in my head" Of these 500 patients, 185 stated that they had been experiencing severe fa¬ tigue for at least six months; of these 185 patients, 176 complained not only of fatigue but also of associated sore throat, muscle aches, or headaches.Of these 176 patients, 103 had no other known condition or illness that could explain the chronic fatigue and associ¬ ated symptoms.Thus, 103 (21%) of 500 patients aged 17 to 50 years seeking primary care in our practice complained of symptoms suggestive of CEBV infec¬ tion, ie, severe fatigue for at least the past six months, with associated symp¬ toms of sore throat, myalgias, or head- aches, and without any known chronic disease.
Of these 103 patients, 40 consented to participate in the study, and, thereby, to answer further questions about their symptoms, to undergo a physical ex¬ amination, and to have serum obtained for sérologie testing.The following rea¬ sons were given by 63 patients for de¬ clining to participate in the study: the patient was not interested in participat¬ ing in any kind of research studies (n = 43); the patient felt that viral infec¬ tion was unlikely and that there were "other reasons" (unspecified) to account for his or her fatigue (n = ll); and the patient did not want to have a phle¬ botomy (n = 9).Those patients partici¬ pating in the study were not signifi¬ cantly different (P>.05) from those eligible patients who declined to partici¬ pate: mean age (34 vs 33 years); sex (70% vs 75% female); and severity of illness-percentage with fatigue and sore throat and myalgias and headache (29% vs 45%).

Clinical Findings
Table 1 summarizes the historical and physical examination findings.The full text of the historical questions asked of the patients is presented in Table 1.The patients described a syndrome of con¬ siderable severity; 28% stated that the fatigue at its worst was sufficient to make them "bedridden: can do virtually nothing."Nearly half of the patients reported "swollen lymph glands," par¬ ticularly in the neck.Low-grade fevers were reported by 25% of patients; 14 patients recorded their temperatures in the morning and evening for one week following the index visit, and six of the 14 reported temperatures between 37.2°C and 37.5°C.
The median duration ofthe symptoms was as follows: fatigue, 16 months (six to 458 months); pharyngitis, 13 months (0.5 to 96 months); headache, 13 months (0.5 to 144 months); and myalgias, 17 months (one to 84 months).The patients were asked explicitly if they had suf¬ fered from these associated symptoms since the onset of the severe fatigue.A detailed history of past experience with pharyngitis, headache, and myalgias was not obtained.The patients reported an average of six sick days in the preced¬ ing six months (range, zero to 40 days per patient).
A majority of patients had seen a physician previously in an attempt to diagnose and treat the problem, and 25% had seen more than one physician.Sixty percent of the patients stated that the chronic fatigue syndrome had cre¬ ated substantial stress both at home and in the workplace.Physical examination findings were benign.None of the patients had a fever at the time of examination.Fifteen per¬ cent had anterior cervical adenopathy, and 13% had enlarged submandibular glands.
fP= .015.All other differences not statistically significant.
Table 2 displays the GMTs of VCA- IgG, EA-Ab, and EBNA-Ab in all pa¬ tients and all matched control subjects, and in specific patient subsets (with their matched control subjects).In all but one of the 43 comparisons reported in the above paragraphs and in Table 2, the EBV antibody levels in patients were higher than in the matched control subjects.However, in only one instance was the difference between patients and control subjects statistically significant (P = .015)(the subset of patients with a history of mononucleosis).Given the relatively small cell sizes in the various paired comparisons, the power of the study to recognize real numerical differ¬ ences of the magnitude shown in Table 2 was weak, ranging from 3% to 69% and averaging 28%.The only comparison in which the power was greater than 50%-VCA-IgG levels in patients with a history of mononucleosis, compared with their age-and sex-matched con¬ trols-was the only instance in which a significant difference was found.

COMMENT
Chronic active EBV infection is a controversial clinical entity.There seems to be little doubt that some pa- tients with extraordinarily high titers of EBV antibodies (particularly VCA-IgG and EA-Ab) really are suffering from a syndrome associated with reactivation of EBV.19 '37"39   Several recent reports have described patients with less dramatic elevations of EBV antibody titers.21"23These reports involved highly selected patients who were referred to investigators known to have an interest in CEBV infection.No attempt was made by these investiga¬ tors to ascertain the prevalence of the condition.Levels of EBV antibodies (VCA-IgG and EA-Ab) were signifi¬ cantly higher in these patients than in control subjects, suggesting that mea¬ surement of antibodies to EBV has diagnostic value in highly selected pa¬ tients.
These recent reports2123 provide fur¬ ther evidence of a chronic fatigue syn¬ drome associated with sérologie evi¬ dence of reactivated, latent EBV infection.However, as the authors ac¬ knowledge, the demonstration of such an association does not prove that reac¬ tivation of EBV is the cause of the fatigue and associated symptoms seen in these patients.Indeed, the elevated EBV antibody titers may represent only an epiphenomenon; some other pri¬ mary pathogenetic event (such as infec¬ tion with another lymphotropic virus) could have caused the patients' illness and also reactivated a latent EBV infec¬ tion.
In this study, we attempted to assess the prevalence of a syndrome sugges¬ tive of the CEBV infection syndrome in an unselected population of patients seeking primary medical care.We found that a clinical syndrome suggestive of CEBV infection or chronic mononucleo¬ sis was present in a surprisingly high fraction (21%) of patients aged 17 to 50 years, coming to our primary care prac¬ tice; none of these patients were felt by their physicians to have evidence of other preexisting "organic" or psychi¬ atric illnesses that could account for their symptoms.On the other hand, while these patients tended to have higher levels of antibody to EBV (par¬ ticularly VCA-IgG and EA-Ab) than did age-and sex-matched control sub¬ jects who did not have a chronic fatigue syndrome, the differences between pa¬ tients and control subjects almost never were statistically significant.
In contrast with several previous studies, we obtained control subjects from among patients (not laboratory technicians or employees) attending the same practice.The control subjects were matched for both age and sex, were asked specifically whether they had been experiencing chronic fatigue, and explicitly denied having chronic fa¬ tigue; the "normality" of the control subjects was not assumed.Finally, the antibody determinations in control sub¬ jects were performed blindly by the same technician who performed the determinations on patients.
It seems likely that our patients were less sick than patients in the several other recently reported studies.2123None of the 40 patients in our study had been fully disabled by his or her illness, in contrast with some patients in the previous studies, and the patients had relatively few days when they could not work.Also, the VCA-IgG and EA-Ab levels in our patients were lower than in these previous studies (although the fact that our sérologie studies were per¬ formed in a different laboratory makes such comparisons hazardous).We ex¬ pected any patients we detected with a syndrome suggestive of CEBV infec¬ tion to be less sick than previously described patients, since we were see¬ ing patients seeking primary care for any reason; we were not seeing patients referred for evaluation of chronic fa¬ tigue.
At the same time, the patients re¬ ported in this study had been experienc¬ ing considerable morbidity from their syndrome (Table 1), for a median dura¬ tion of 16 months (range, six to 458 months).Sixty percent reported that the syndrome had caused substantial stress at home and at work.The major¬ ity stated that the fatigue became so severe that they were periodically bed¬ ridden; the associated myalgias and headaches became so severe at times that the patients had to stop all normal activities and rest.Twenty-five percent of the patients stated that they had sought help for the problem from multi¬ ple physicians, to no avail.
Many of the symptoms associated with the CEBV infection syndromeparticularly the chronic fatigue, my¬ algias, and headache-are also man¬ ifestations of chronic depression and anxiety.Therefore, it is possible that some of these patients were suffering from a psychoneurosis, even though none of them were being treated for a psychoneurosis at the time that their chronic fatigue syndrome started.At the same time, depression and anxiety do not so easily explain the recurrent pharyngitis, adenopathy, paresthesias, arthralgias, or fevers reported by many patients.
The results of this study indicate that a chronic fatigue syndrome suggestive of a mild form of the CEBV infection syndrome may be surprisingly common in a primary care practice.However, our study offers no proof that EBV plays a causal role in the pathogenesis of this syndrome.Furthermore, our study emphasizes the difficulties of diagnos¬ ing this syndrome using currently avail- able sérologie techniques, at least with patients seen in a general medicine pri¬ mary care practice.
500-patient sample and the entire population with regard to age, sex, or medical insurance coverage.

Table 1 .
-Clinical Findings Findings HistoryChief complaint is the fatigue syndrome 12 (30) Severity of the fatigue at its worst Bedridden: can do virtually nothing 11 Shut-in: cannot do even light housework or its equivalent 5

Table 2 .
-Comparisons of Serologie Results, Different Patient Groups, and Their Matched Control Subjects*