Proportion Positive for Epstein-Barr Virus, Cytomegalovirus, Human Herpesvirus 6, Toxoplasma, and Human Immunodeficiency Virus Types 1 and 2 in Heterophile-Negative Patients With an Absolute Lymphocytosis or an Instrument-Generated Atypical Lymphocyte Flag.
Hoagland's classic morphologic criteria utilized for the diagnosis of an infectious mononucleosis-like illness include a lymphocytosis comprising more than 50% of the white blood cell (WBC) differential with atypical lymphocytes accounting for more than 10% of the total WBC count.[4-6] Patients who meet Hoagland's criteria with a negative heterophile-antibody test have shown evidence of acute infection with a variety of infectious agents, including Epstein-Barr virus (EBV), cytomegalovirus (CMV), human herpesvirus-6 (HHV-6), Toxoplasma gondii (Toxo), acute viral hepatitis, and HIV types 1 and 2 (HIV-1, HIV-2).[7-9] However, some studies have suggested a limited effectiveness for Hoagland's criteria in the consistent detection of infectious mononucleosis. For example, a study by Fleisher et al showed that one or both of Hoagland's criteria were not met in 61% of patients with infectious mononucleosis, despite serologic evidence for acute EBV infection.
This prospective study was undertaken to determine the most common etiologic agents responsible for infectious mononucleosis-like illness in heterophile antibody-negative patients with a lymphocytosis greater than 4000/ [mm.sup.3], an instrument-generated atypical lymphocyte flag, or both. The study also examined whether a variety of other parameters, including age, sex, specific values from the 5-part differential or individual atypical lymphocyte morphologies were associated with seropositivity for individual infectious agents. A third goal was to develop a cost-effective testing algorithm for the management of patients with heterophile antibody-negative infectious mononucleosis.
MATERIALS AND METHODS
The study was conducted from July 1995 to February 1996 at Metro-McNair Clinical Laboratories and Children's Hospital, both in Vancouver, British Columbia, Canada. The assumption was made that a heterophile antibody test was initially requested by the patient's physician in response to presenting clinical signs and symptoms suggestive of an infectious mononucleosis-type illness. In total, we identified 1921 samples for which there was a physician's request for heterophile antibody testing. Blood was drawn using disposable multisample needles (Becton Dickinson Vacutainer Systems, Rutherford, NJ). Testing for heterophile antibodies was performed by registered technologists using the Monosticon Dri-Dot test (Organon-Teknika, Scarborough, Ontario, Canada). Patients with positive heterophile antibody tests were subsequently excluded from the study. Blood samples were analyzed within 1 to 3 hours of collection on either a Coulter STKS (Coulter Electronics, Inc, Hialeah, Fla) or a Sysmex NE-8000 (TOA Medical Electronics Company, Ltd, Kobe, Japan) analyzer, each calibrated according to the manufacturer's specifications. The software version utilized on the NE-8000 was 6/20, while the version on the STKS was 2B. Histogram patterns were plotted for all samples, and any operator alert flags were noted. The minimum morphologic criteria for subsequent inclusion in the study included an absolute lymphocytosis greater than 4000/ [mm.sup.3], an instrument-generated atypical lymphocyte flag, or both. The analyzer manufacturers consider the algorithms that generate the atypical lymphocyte flags to be proprietary information. An alternate a priori method of entering into the study occurred when atypical lymphocytes, lymphocytosis, or both were found on the hematology analysis and a laboratory physician-ordered heterophile antibody test was negative.
For cases meeting the study criteria, the Wright-Giemsa-stained smear was subsequently examined by a single registered technologist. Each blood film had an initial 200 WBC differential performed. A separate 200-cell lymphocyte differential was performed subsequently by examining 200 consecutive lymphocytes on each slide, utilizing a microscope eyepiece equipped with a micrometer. For the purposes of the study, lymphocytes were considered atypical or reactive if they showed the characteristic morphologic changes as described by Downey and others or were greater than 15 [micro]m in diameter. The technologist performing the manual differentials was blinded to the results of the automated differential and the serology.
Serum samples from the study cases were subsequently coded so that further testing was both anonymous and unlinked. Ethics approval was obtained from the University of British Columbia to perform anonymous unlinked viral testing without patient consent on all specimens. Samples were then analyzed for EBV, CMV, HHV-6, Toxoplasma, HIV-1, and HIV-2. In addition, a control group that consisted of a selection of 50 patients from the 1689 original patients who were heterophile antibody negative but who had a normal lymphocyte count and generated no atypical lymphocyte flag were also tested for EBV IgM antibodies.
All viral serology was performed by the same registered technologist. For the purposes of the study, any serum sample that produced an equivocal result was treated as negative. It was also assumed that a positive IgM result was equivalent to an acute infectious process. The presence of anti-CMV-specific antibodies was tested for using the IMx CMV IgM microparticle enzyme immunoassay test kit and methods (Abbott Laboratories, Abbott Park, Ill). All sera with IMx IgM indices greater than 0.500 were subsequently subjected to rheumatoid factor neutralization and reanalyzed. The Platelia Toxo IgM immunoenzymatic double-sandwich test kit and methods (Sanofi Diagnostics Pasteur, Chaska, Minn) were used for the determination of anti-Toxo IgM antibodies. Testing for EBV viral capsid antigen was performed using the Gull EBV IgM ELISA test and methods (Gull Laboratories, Salt Lake City, Utah). Patients' sera were diluted with a specimen diluent that contained an absorbent to remove IgG and to prevent IgG/rheumatoid factor complex formation. This step served to prevent interference from competing IgG during the procedure. The 50 control specimens that were heterophile antibody negative were also tested with the Gull EBV IgM ELISA test. Antibodies to HIV-1 and HIV-2 were assayed using the HIV-AB HIV-1/HIV-2 (rDNA) enzyme immunoassay (Abbott Laboratories). Human herpesvirus 6 IgM and IgG serological testing was conducted at the British Columbia Children's Hospital using an in-house immunofluorescence assay. For this assay, HHV-6 was grown in an immature T-cell line (HSB-2). Serial dilutions of cell suspension were prepared to determine optimal levels. Welled slides were prepared, air-dried, fixed in acetone, and frozen at -70 [degrees] C. Controls and samples were first neutralized for rheumatoid factor and then tested at a 1:40 dilution using 25 [micro]L. Slides were incubated overnight, washed, and overlaid with diluted IgM or IgG conjugate (rabbit anti-human IgM or IgG, respectively). After brief incubation, washing, drying, and mounting, slides were read with a standard immunofluorescence microscope.
Statistical analysis, including logistic regression analysis, was performed using SPSS 7.5.1 for Windows (SPSS Inc, Chicago, Ill). A 2-tailed Student t test and the Mann-Whitney U test were used for analytical statistics, with a decision made a priori to use nonparametric tests if variances were unequal. For logistic regression analysis, the outcome variables were EBV IgM, CMV IgM, HHV-6 IgM, Toxo IgM, HIV-1, and HIV-2. A priori hypothesis-testing independent variables were age, sex, hematology analyzer parameters (atypical lymphocytes, total WBC count, neutrophil count, lymphocyte count, the percentage of lymphocytes/WBC), and from the special differential (percentage of atypical lymphocytes/total lymphocytes, the percentage of atypical lymphocytes/total WBC). Hypothesis-generating independent variables were hematology analyzer parameters (monocyte count, eosinophil count, basophil count) and from the special WBC differential (numbers of Downey I, II, and III lymphocytes and lymphocytes with nuclei [is greater than] 15 [micro]m).
Of the 92.2% (1771/1921) of patients testing heterophile antibody negative, 4.6% (82/1771) had an automated differential or the presence of an atypical lymphocyte flag that satisfied the minimal study criteria, and 95.4% (1689/ 1771) did not. Due to problems with dated or damaged serum, 16 of these 82 patients did not undergo subsequent serologic analysis. The remaining 66 patients underwent further testing, in addition to 4 heterophile-negative patients who were identified by the alternate a priori method described, and whose instrument-generated results otherwise have met study entry criteria, making the study population 70 in total.
The age of patients ranged from 1 to 71 years with a mean age of 21.3 years. Thirty patients were male and 40 were female. Sex did not prove to be a significant factor in the analysis. Of study patients, 70% (49/70) demonstrated 1 or more serologic findings suggestive of an acute infectious process. In 34% of patients (24/70), a single infectious agent was implicated based on a positive, specific IgM result. Serology was positive for 2 or more infectious agents in 36% of cases (25/70). The remaining 30% of cases (21/70) showed negative IgM serology for all infectious agents (Figure 1).
[Figure 1 ILLUSTRATION OMITTED]
The mean age of the patients with a positive EBV IgM test was 14.1 years (N = 28), versus patients who were EBV negative who had a mean of 26.1 years (N = 42). This difference was significant at P [is less than] .001, using the Mann-Whitney U test (2 tailed).
Table 1 documents the complete serologic findings for the 70 study patients. The column headed "Pure IgM+" indicates those samples testing positive for only 1 agent (N = 45). Table 1 also provides the results of multiple positive serologies. The serologic breakdown for those 24 patients showing evidence of acute infection with a single agent is displayed in Figure 2. Of the 4 patients who entered the study via the alternate a priori route, 1 was only EBV IgM positive, 1 was only CMV IgM positive, and the other 2 were negative by all markers. In the control group assayed for EBV IgM antibodies, all 50 subjects were negative. The comparison of the 0/50 positive controls with the 28/70 EBV-positive study patients was significant at P [is less than] .0001 by [chi square] test.
[Figure 2 ILLUSTRATION OMITTED]
Table 1. Proportion of Positive Viral Serologies in Study Population
Also Positive For Primary Total IgM+, % Pure IgM+, % EBV, % CMV, % Test (N = 70) (N = 45) (N = 70) (N = 70) EBV 40 (28/70) 20 (9/45) 100 (28/28) 46 (13/28) CMV 39 (27/70) 22 (10/45) 48 (13/27) 100 (27/27) HHV-6 25 (16/65) 9 (4/45) 50 (8/16) 44 (7/16) Toxo 3 (2/70) 2 (1/45) 50 (1/2) 0 HIV 0 (0/70) 0 0 0 Also Positive for Primary HHV-6, % Toxo, % HIV, % Test (N = 65) (N = 70) (N = 70) EBV 32 (8/25) 4 (1/28) 0 CMV 28 (7/25) 0 0 HHV-6 100 (16/16) 6 (1/16) 0 Toxo 50 (1/2) 100 (2/2) 0 HIV 0 0 0
(*) EBV indicates Epstein-Barr virus; CMV, cytomegalovirus; HHV-6, human herpesvirus 6; Toxo, Toxoplasma gondii; and HIV, human immunodeficiency virus.
Hypothesis-Testing and Hypothesis-Generating Models
With EBV IgM as the outcome variable (N = 70), the best logistic regression hypothesis-testing model showed a statistically significant association with age, percent atypical lymphocytes/lymphocytes, and an interaction term (age x percent atypical lymphocytes/lymphocytes) as explanatory variables. For age, P = .62 and the odds ratio = 0.97 (95% CI, 0.88-1.08); for percent atypical lymphocytes/lymphocytes, P = .007 and the odds ratio = 1.22 (95% CI, 1.1-1.4); and for the interaction term, P = .045 and the odds ratio = 0.994 (95% CI, 0.988-0.999). The regression diagnostics for this model were acceptable. There was significant interaction between age and percent atypical lymphocytes/lymphocytes, showing that these factors were not independent in the model. For the hypothesis-testing model with EBV IgM as the outcome variable utilizing the subset of 45 subjects who had only 1 or no positives, the best model had age and percent atypical lymphocytes/lymphocytes as explanatory variables. In this case the interaction term was not significant and so was not included in the model. The P values were P = .085 for age and P = .044 for percent atypical lymphocytes/lymphocytes.
Hypothesis-generating logistic regression on further explanatory variables resulted in the significant associations listed in Table 2. With EBV IgM as the outcome variable, Downey type II lymphocytes were a significant independent variable in the model, with P = .006 and odds ratio = 1.07 (95% CI, 1.02-1.12). With HHV-6 as the outcome variable, Downey type III lymphocytes constituted a significant explanatory variable with P = .016 and odds ratio = 1.6 (95% CI, 1.1-2.3). For CMV as the outcome variable, there was a trend for Downey type I lymphocytes as an explanatory variable, with P = .097 and odds ratio = 1.38 (95% CI, 0.94-2.03).
Table 2. Hypothesis-Generating Logistic Regression Models Outcome Explanatory Variables Variable (N = 70) EBV IgM Downey II Age P = 0.0057 P = 0.0011 OR = 1.07 (95% CI 1.02-1.12) OR = 0.89 (95% CI 0.82-0.95) CMV IgM Downey I Basophils P = 0.097 P = 0.028 OR = 1.38 (95% CI 0.94-2.03) OR = 199.6 (95% CI 1.8-22 732) HHV-6 IgM Downey III ... P = 0.016 OR = 1.6 (95% CI 1.1-2.3)
(*) EBV indicates Epstein-Barr virus; CMV, cytomegalovirus; HHV-6, human herpesvirus 6; and OR, odds ratio.
Figure 3 outlines the apparent simplest and most cost effective model for triaging heterophile antibody tests. In designing this algorithm, an assumption was made that EBV represented the most significant viral infection. Applying the model to the original 1921 patients, 150 were heterophile antibody positive and 1771 were heterophile antibody negative. Of the 1771 negatives, 1689 had a normal lymphocyte count without an atypical lymphocyte flag, while 82 patients had one or both of these abnormalities. Based on the fact that none of the control group of heterophile-negative patients without lymphocytosis or an atypical lymphocyte flag were EBV IgM seropositive, both the 150 patients who were heterophile antibody positive and the 1689 patients who were heterophile antibody negative without lymphocytosis or atypical lymphocyte flag should require no further antiviral testing. The most cost-effective method for testing the remaining 82 patients would be to test all patients for EBV IgM, followed by selective testing of only the EBV IgM-negative patients for HHV-6 and CMV, that is, the 2 serologies that were most commonly also positive in our study. Since only 3% and 0% were Toxo and HIV positive, respectively, in our study population, the algorithm suggests that these tests only be performed when there is a clinical indication. Since in the province of British Columbia all laboratory testing is funded by a single payment agency, it is reasonable to look at the potential cost savings of the algorithm applied to this patient population of approximately 3.5 million people.
[Figure 3 ILLUSTRATION OMITTED]
Utilizing the algorithm for the estimated total number of heterophile-negative tests for the province of British Columbia (10626) at a cost per viral serology test of Can $21.63, with all patients tested but testing restricted to the 3 most common viruses (CMV, EBV, and HHV-6), the cost would be Can $689 521.
Incorporating the algorithm hypothesis that there were no EBV IgM-seropositive subjects in the heterophile-negative group without an absolute lymphocytosis or atypical lymphocyte flag, as was the case in our control group, if only this proportion of the total population were tested (492 patients) for the 3 most common viruses, there would be a savings of Can $657 595. Making the assumption that the detection of a positive EBV IgM serology results in no further viral serology testing would eliminate an additional 40% of the remaining population from having serologic testing for CMV and HHV-6 IgM, resulting in further savings of Can $8522.
The existing literature documents significant etiologic roles for CMV and EBV in heterophile antibody-negative infectious mononucleosis. Less well established is the proportion of heterophile antibody-negative infectious mononucleosis due to acute HHV-6,[8,11] Toxo, or HIV infection.[12-16] While infectious mononucleosis is often considered a self-limiting disease, this remains true only for the immunocompetent patient. An intricate interrelationship exists between herpesvirus infection and immune status, specifically, cellular immunity. Serological testing for herpesviruses is complex and carries the possibility for dual infections, technical error, or the potential for one virus to cross-react with another, as in the case of EBV and CMV. There is also a suggestion that an acute CMV infection can activate the EBV viral carrier state. Although such difficulties in interpretation are also possible with HHV-6 and other Herpesviridae infections, Fox and colleagues discovered no evidence of any such cross-reactions. However, more recent data have described cross-reactivity between HHV-6 and HHV-7, which also show a large degree of genetic homology.[3,19-21] Difficulty may also exist in distinguishing an acute primary infection from a reactivation event. The presence of IgM responses suggests an active viral process but does not always imply an acute primary infection. This is well established in the case of CMV and rubella, and also appears to be true for HHV-6.[7,22] Therefore, when making a definitive diagnosis of a true heterophile antibody-negative infectious mononucleosis-like illness, more confidence should be placed in those serologic investigations that only demonstrate a single infectious agent.
The present study focused on the prevalence of EBV, CMV, HHV-6, Toxo, and HIV in 70 patients meeting our chosen criteria for a heterophile antibody-negative infectious mononucleosis-like illness. These patients had either an absolute lymphocytosis or an atypical lymphocyte flag as defined by the hematology analyzers utilized. We believe that these selection criteria are more realistic for application in a high-volume clinical setting than Hoagland's traditional criteria, since the modern hematology analyzers can flag all of the patients subsequently requiring specific serologic testing in an objective and reproducible fashion. There were 2 a priori routes for entry into the study, and 4 patients entered via the alternate route with the heterophile antibody test being ordered by the laboratory physician. We recognize that our assumption that these 2 groups were clinically the same could represent a bias in this study, but it would be difficult to determine the direction and magnitude of this potential bias. Our data contained no evidence indicating that these groups represented different patient populations, and we do not think there was any effect on our conclusions.
We were particularly interested in the role of HHV-6 in this study population. Nine percent (4/45) of study patients had evidence for acute HHV-6 infection, which is comparable with the findings of Horwitz et al who showed a prevalence rate of 6.5% for HHV-6. Human herpesvirus 6-induced heterophile antibody-negative infectious mononucleosis-like illness is believed to arise from a late primary HHV-6 infection, not a reactivation event. There is a restricted pool of patients susceptible to late HHV-6 infection, in that more than 90% of children aged 1 to 4 years are HHV-6 IgG positive.[23-26] It is possible, however, that a proportion of the patients investigated in this study were undergoing a viral reactivation event.
The majority of study patients in whom a single positive IgM titer was observed displayed evidence of an acute CMV or EBV infection (22% and 20%, respectively). Horwitz et al have presented data in 2 studies[2,7] showing significantly different results for the incidence of CMV and EBV (42% and 33%, respectively, for CMV and 70% and 16%, respectively, for EBV). However, there is a significant difference in the percentage of patients for whom a diagnosis was not determined between our study and that of Horwitz et al. In our investigation, a diagnosis could not be obtained for 30% of patients (21/70) versus 16% in the investigation of Horwitz et al.
There are several explanations for the large percentage of unknown patients in our study. Anti--HHV-6 IgM antibodies may not be detectable in serum until 4 to 7 days after the onset of clinical symptoms. Furthermore, CMV can produce clinical illness without an associated rise in IgM titer.[28,29] While the sensitivity has been reported at greater than 90% for all of the test kits employed in this study, false-negative results are still a possibility. Also, there are many other infectious agents that may have been responsible for the observed quantitative and morphologic changes in blood lymphocytes, including hepatitis B virus, adenovirus, rubella, or various drugs. Finally, other as of yet unidentified agents may have been responsible. Thus, although a positive IgM titer does not definitively confirm an etiologic diagnosis, this methodology appears most clinically appropriate for a high-volume commercial laboratory setting. In clinical practice, the evaluation of a positive IgM serology in individual patients in whom multiple viruses are detected can be further aided by examining the absorbance and cutoff values for each virus, the clinical picture, and by other diagnostic tests for the specific viruses.
This study also examined both hypothesis-testing and hypothesis-generating logistic regression models. The association demonstrated between age and acute EBV infection simply confirms earlier data on the age-specific incidence of infectious mononucleosis. While laboratories no longer quantitate the various morphologic types of atypical lymphocytes, the association between the individual viral infections and Downey types as determined by the hypothesis-generating model was interesting (Table 2). There is a well-known correlation between acute EBV infection and the presence of Downey type II lymphocytes observable on the peripheral blood film, as confirmed by our results.[5,16,17,30] Our findings of a significant association between HHV-6 titers and the presence of Downey type III lymphocytes were notable, as was the trend between the presence of Downey type I lymphocytes and CMV infection. While probably only of clinical interest, the data do provide further information regarding an association between different herpesvirus infections and the presence of various morphologic types of atypical lymphocytes.
A specific, positive viral serology was identified in 70% of heterophile-negative patients with an absolute lymphocytosis, an atypical lymphocyte flag generated by an automated hematology analyzer, or both. Cytomegalovirus represented the most common viral serology detected, followed by EBV and HHV-6. No HIV-1- or HIV-2-positive serologies were found, but these agents should always be considered in the differential diagnosis of heterophile antibody-negative infectious mononucleosis, based on the individual clinical scenario. While a specific etiologic diagnosis is not always possible owing to negative results or patients with multiple positive tests, the methodology utilized in this study should prove beneficial in restricting the differential diagnosis. The testing algorithm developed is based on objective criteria provided by automated hematology analyzers and sequential testing for viruses and has the potential to generate significant cost savings depending on individual clinical practice. Where viral testing is currently restricted for economic reasons, utilizing the testing algorithm should allow for a more focused approach to the heterophile-negative patient.
The authors thank the following institutions and people for their contributions to this study: Metro-McNair Laboratory, Vancouver, British Columbia, Canada, in particular Caroline Huen, RT; Children's Hospital Virology Laboratory, Vancouver, British Columbia, Canada; Jeff Quon, MHSc; Riyad Abu-Laban, MD, MHSc; and Gull Laboratories, Salt Lake City, Utah; Abbott Laboratories, Abbott Park, Ill; and Sanofi Diagnostics Pasteur, Chaska, Minn, for providing reagents.
[1.] Aronson MD, Komaroff AL, Pass TM, Ervin CT, Branch WT. Heterophil antibody in adults with sore throat. Ann Intern Med. 1982;94:505-508.
[2.] Horwitz CA, Henle W, Henle G, et al. Heterophil-negative infectious mononucleosis and mononocleosis-like illnesses. Am J Med. 1977;63:947-957.
[3.] Frenkel N, Schimer EC, Wyatt LS, et al. Isolation of a new herpesvirus from CD4+ T cells. Proc Natl Acad Sci U S A. 1990;87:748-752.
[4.] Hoagland RJ. Infectious mononucleosis. Primary Care. 1975;2:295-307.
[5.] Wood TA, Frenkel EP. The atypical lymphocyte. Am J. Med. 1967;42:923-936.
[6.] Pejme J. Infectious mononucleosis: a clinical and haematological study of patients and contact, and a comparison with healthy subjects. Acta Med Scand Suppl. 1964;413:1-83.
[7.] Horwitz CA, Krueger GRE Steeper TA, Bertram G. HHV-6-induced mononucleosis-like illnesses. In: Ablashi DV, Krueger GRF, Salahuddin SZ, eds. Human Herpesvirus-6: Epidemiology, Molecular Biology, and Clinical Pathology, Amsterdam, Holland: Elsevier Science Publishers; 1992:159-174.
. Steeper TA, Horwitz CA, Ablashi DV, et al. The spectrum of clinical and laboratory findings resulting from human herpesvirus-6 (hhv-6) in patients with mononucleosis-like illnesses not resulting from Epstein-Barr virus or cytomegalovirus. Am J Clin Pathol. 1990;93:776-783.
[9.] Finch SC. Laboratory findings in infectious mononucleosis, In: Carter RL, Penman HG, eds. Infectious Mononucleosis. Oxford, England: Blackwell; 1969:47.
[10.] Fleisher GR, Collins M, Fager S. Limitations of available tests for diagnosis of infectious mononucleosis. J Clin Microbiol. 1983; 17:619-624.
[11.] Irving WL, Cunningham AL. Serological diagnosis of infection with human herpesvirus-6. BMJ. 1990;300:156-159.
[12.] Lusso P, Ablashi DY, Luka J. Interactions between HHV-6 and other viruses. In: Ablashi D, Krueger GRF, Salahuddin SZ, eds. Human Herpesvirus-6: Epidemiology, Molecular Biology, and Clinical Pathology: Perspectives in Medical Virology Vol 4. Amsterdam, Holland: Elsevier Science Publishers; 1992:121-133.
[13.] Spira TJ, Bozeman LH, Sanderlin KC, et al. Lack of correlation between human herpesvirus-6 infection and the course of human immunodeficiency virus infection. J Infect Dis. 1990; 161:567-570.
[14.] Carrigan DR, Knox KK, Tapper MA. Suppression of human immunodeficiency virus type I replication by human herpesvirus-6. J Infect Dis. 1990;162: 844-851.
[15.] Levy J, Landay A, Lennette E. Human herpesvirus 6 inhibits human immunodeficiency virus type 1 replication in cell culture. J Clin Microbiol. 1990; 28:2362-2364.
[16.] Steeper TA, Horwitz CA, Hanson M, et al. Heterophil-negative mononucleosis-like illnesses with atypical lymphocytosis in patients undergoing seroconversions to the human immunodeficiency virus. Am J Clin Pathol. 1988;90:169-174.
[17.] Fiala M, Heiner DC, Turner JA, Rosenbloom B, Guze LB. Infectious mononucleosis and mononucleosis syndromes. West J Med. 1977;126:445-459.
[18.] Fox JD, Ward P, Briggs M, Irving W, Stammers TG. Production of IgM antibody to HHV6 in reactivation and primary infection. Epidemiol Infect. 1990; 104:289-296.
[19.] Nicholas J. Determination and analysis of the complete nucleotide sequence of human herpesvirus-7. J Virol. 1996;70:5975-5989.
[20.] Berneman ZN, Ablashi DY, Li G, et al. Human herpesvirus-7 is a T-lymphotropic virus and is related to, but significantly different from, human herpesvirus 6 and human cytomegalovirus. Proc Natl Acad Sci U S A. 1992;89:10552-10556.
[21.] Pellet PE, Black JB, Yamamoto M. Human herpesvirus 6: the virus and the search for its role as a human pathogen. Adv Virus Res. 1992;41:1-52.
[22.] Straus SE. Epstein-Barr virus infections: biology, pathogenesis, and management. Ann Int Med. 1993;118:45-58.
[23.] Tagawa S, Mizuki M, Onoi U, et al. Transformation of large granular lymphocytic leukemia during the course of a reactivated human herpesvirus-6 infection. Leukemia. 1992;6:465-469.
[24.] Levine PH, Jarrett R, Clark DA. The epidemiology of human herpesvirus-6. In: Ablashi DV, Krueger GRF, Salahuddin SZ, eds. Human Herpesvirus-6: Epidemiology, Molecular Biology, and Clinical Pathology. Amsterdam, Holland: Elsevier Science Publishers; 1992:9-23.
[25.] Axelrod P, Finestone AJ. Infectious mononucleosis in older adults. Am Fam Physician. 1990;42:1599-1606.
[26.] Secchiero P, Cleghorn FR. HHV-6 infection in immunocompromised patients. Infect Med. 1998;15:192-198.
[27.] Briggs M, Fox J, Tedder RS. Age prevalence of antibody to human herpesvirus-6. Lancet. 1988;1:1058-1059.
[28.] Klemola E, Von Essen R, Henle G, Henle W. Infectious-mononucleosis-like disease with negative heterophil agglutination test: clinical features in relation to Epstein-Barr virus and cytomegalovirus antibodies. J Infect Dis. 1970; 121:608-614.
[29.] Weber JM, Wilson G. Disease association and diagnosis of human herpesvirus 6. Can J Infect Dis. 1994;5:185-185.
[30.] Klemola E, Kaariainen L. Cytomegalovirus as a possible cause of a disease resembling infectious mononucleosis. Br Med J. 1965;2:1099-1102.
Accepted for publication March 31, 2000.
From the University of Calgary Medical School, Calgary, Alberta, Canada (Mr Tsaparas); Department of Medical Oncology, BC Cancer Agency-Center for the Southern interior, Kelowna, British Columbia, Canada (Dr Brigden); and the Departments of Health Care and Epidemiology (Drs Mathias and Raboud) and Pathology and Laboratory Medicine (Drs Thomas and Doyle), University of British Columbia, Vancouver, British Columbia, Canada.
Presented in part at the Canadian Society of Transfusion Medicine meeting, Ottawa, Ontario, Canada, May 9, 1998, and at the Interscience Conference on Antimicrobial Agents and Chemotherapy, San Diego, Calif, September 26, 1998.
Reprints: Patrick W. Doyle, MD, Canadian Blood Services, 4750 Oak St, Vancouver, BC, Canada V6H 2N9.
|Printer friendly Cite/link Email Feedback|
|Author:||Tsaparas, Yotis F.; Brigden, Malcolm L.; Mathias, Richard; Thomas, Eva; Raboud, Janet; Doyle, Patric|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Sep 1, 2000|
|Previous Article:||An Algorithm to Aid in the Investigation of Thalassemia Trait in Multicultural Populations.|
|Next Article:||Evaluation of the Extent of Nonlinearity in Reportable Range Studies.|