Comparative evaluation of seven commercial tests for detection of heterophile antibody in infectious mononucleosis

M SKULNICK, DE Low, AE SIMOR, M PATEL, P GEORGE, R CHUA. Comparative evaluation of seven commercial tests for detection of heterophile antibody in infectious mononucleosis . Can J Infect Dis 1991 ;3(1):23-26. Detection of heterophile antibodies in infectious mononucleosis is the most rapid and cost-effective m ethod for confirming the clin ical diagnosis of the disease. This study compared seven commercial test kits (the Oxoid Infectious Mononucleosis Kit [Oxoid Ltd]. Immunoscan 1m-Latex [Baxter Travenol]. Mon o-Latex [Wampole Laboratories]. Monospot and lm Screen Test [Ortho Diagnostics[. lmmunoscan Im-RBC Test [BaA1:er Travenol[ . and Infectious Mononucleosis Test [NCS DiagnosticsJ) to the Davidsohn differen tial test. All of the kits were shown to be acceptable for use. with specificities and sensitivities greater than 96.5% and 95.5%. respectively.

self-limiting disease caused by the Epstein-Barr virus .The disease is most commonly seen in adolescents and young a dul ts.The diagnosis of infectio u s mononucleosis is usually made on the basis of clinical symptoms (such as pharyngitis, atypical lymphocytosis.rash, fever.lymphadenopathy an d splenomegaly) and confirmed by the detection of infectious mononucleosis heterophile antibodies and/ or antibodies to Epstein-Barr virus (1.2).However.certain adult patients may not produce infectious mononucleosis heterophile antibodies.and some children with acute infectious mononucleosis have no measurable heterophile antibodies with current rapid methods (3,4).Other heterophile antibodies such as the Forssman and serum sickness antibodies are produced in a variety of diseases such as rheumatoid arthritis, lymphoma and hepatitis.and must be differentiated from the infectious mononucleosis heterophile antibody (5).In response to this concern, Davidsohn (5) developed a confirmatory differential test which removes Forssman and sheep heterophile antibodies by absorption wiU1 beef erythrocytes and guinea pig kidney cells.

MATERIALS AND METHODS
A total of 318 consecutive sera obtained from nonhospitalized adult and adole cent patients seen in fan1ily physicians' offices and suspected of having infectious mononucleosis were tested with each of lhe comm ercia l kits and compared to the results of the Davidsohn differential test (6).All testing was done in a blinded manner and comparative results were unknown until ta bula tion of all data was complete.
All sera with discrepant test kil result compared to wilh the Davidsohn differential tesl were repealed wilh the test kit or kits.In addition.specific Epstein -Barr serology using immunol1uorescenl lechniques was performed.These tests were viral capsid antigen IgG (Granbio Inc.California).viral cap id antigen IgM (Gull Laboratories , Utah).early antigen IgG and Epstein-Barr nuclear antigen IgM.Acute infection was defined as U1e TABlE 1 presence of either viral capsid antigen IgM antibodies or antibody to early antigen (7).The two latter tests were performed at the Ministry of Health Laboratories using fluorescent antibody technique.
The kits tested in this study were a cross-section of those available on the market.) .The latex agglutination kits all used latex polystyrene beads sensitized with purified bovine red blood cell heterophile antigen.This glycoprotein has been shown to be specific for the infectious mononucleosis heterophile antibody and, therefore, removed the need for procedures to absorb Forssman or serum sickness antibodies as with the monospot kit (8 ,9).The other four kits all used fresh or modified horse red blood cells.

RESULTS
Of the 318 sera tested, 89 (28%) were positive for heterophile antibodies using the Davidsohn differential lest.A comparison of the test kit results is shown in Table 1.The results show that all kits h ad acceptable sensitivities and specificities greater than 95.5% and 96.5%, respectively.
The sera giving discordant results are shown in Tab le 2. The majority of sera were discordant against two or fewer kits.However, patient sera 153.156.200 and 259 gave discordant results against four test kits each.One serum.213, positive for the Davidsohn test with a titre of 1:40, was negative against all test kits.However. the viral capsid antigen IgM test was less than l : 10 and the viral capsid a ntigen IgG and Epstein-Barr nuclear antigen were positive at greater than or equal to l :640 and greater than or equal to l :40 , respectively.specimen was a false positive.In most cases the Epstein -Barr virus serology was in agreement with the •gold standard•.Analysis of discrepant results showed that in all but two patients Epstein-Barr serology was consistent with no acute infection.The two other sera showed evidence of acute infection and could.therefore.represent false negative results of the Davidsohn differential test.

DISCUSSION
There are many commercial test kits available on the market for detection of infectious mononucleosis heterophile antibody.Several authors have compared many of these kits and have shown that there is no significant statistical advantage con cerning sensitivity and specificity (10)(11)(12).The present results agree wiU1 published reports.In the present study, the erythrocytebased kits.although equal in sensitivity to the latex kits.appeared to be less specific.However, the difference was not statistically significant.This result is also comparable to the study by Hoiby et al (11) which showed an increased number of false positives in erythrocyte-based tests.The Ortho monospot test, which is widely used.has been shown to give false positive reactions occasionally with leukemia, rubella, malaria and several other diseases (13,14).In the present study the monospot test demonstrated a specificity comparable to that of other erythrocyte-based tests.However, the monospot had the disadvantage of an absorption step that was not required in any other test.
In general the latex kits were technically simpler to perform and the endpoint agglutination was easier to CAN J INFECT DIS V OL 3 No 1 J ANUARY / FEBRUARY 1992 read.All tests were rapid, ranging from 2 mins for the NCS test to 4 mins for the Immunoscan latex test.Th e costs for kits are comparable and would vary from institution to institution depending on the volume of tests.
Evaluation comparative de sept tests commerciaux servant a Ia detection des anticoi'ps heterophiles dans Ia mononucleose infectieuse RESUME: La detection des anticorps heterophiles dans Ia monon u cleose infectieuse est Ia fac;:on Ia p lus rapide et Ia p lus eco n omique de confirmer Je diagnostic cli n ique.La presente etude a compare sept tests commerciaux (Oxoid Infectious Mononucleosis Kit [Oxoid Ltd[.lmmunoscan 1m-Latex [Baxter Travenoll.Mono-Latex [Wam pole Laboratories[.Monospot et Im Screen Te t [OrU1o DiagnosticsJ) au test d ifferentiel de Davidsohn .Taus les tests se sonl averes acceptables.avec des specificites et sensibili tes superieures a 96.5 eta 95,5 %. respectivemenl.SKULNICK et Of I NFECTIOUS MONONUCLEOSIS IS A COMMON AND USUALLY These results indicate that the Davidsohn test with thisHeterophile antibody results for seven test kits compared with the Davidsohn differential test •n=89. 1 n=229.' The glycoprotein is derived from bovine red blood cells.See text for complete names of test kits