Performance of eight commercial immunoassays for the detection of cytomegalovirus-specific IgM antibodies in pregnancy – no test fits all needs
ABSTRACT
Detection of cytomegalovirus (CMV)-specific immunoglobulin M (IgM) antibodies as first-line serologic diagnosis plays an important role in identifying CMV primary infection during pregnancy. The performance characteristics of eight commercially available CMV IgM assays were compared. Sensitivity and IgM antibody kinetics were assessed using 100 acute phase and follow-up sera from 39 pregnant women with a well-defined onset of CMV primary infection. Specificity was analyzed using 50 well-characterized serum samples from pregnant women not infected or latently infected with CMV and from patients with other acute infections. Until 12 weeks after the onset of primary infection, four assays showed sensitivities of 100%, whereas the others had individual gaps to detect all primary infections in this time period. All assays showed a time-dependent decrease of IgM levels. More than 12 weeks after the onset of infection, the IgM-positive rates varied considerably between tests. The specificity was between 92% and 98% in all but one assay. The observed differences in the performance characteristics must be taken into account in CMV screening and diagnosis of primary infection during pregnancy.
INTRODUCTION
Congenital cytomegalovirus (CMV) infection is the most common infectious cause of neurodevelopmental delay and hearing loss in young children. The risk for long-term sequelae is highest following maternal primary infection (PI) around conception or in the first trimester of pregnancy (1). Thus, the timing of CMV PI in pregnancy is crucial for risk evaluation and further management. The diagnosis of CMV PI in immunocompetent patients is based on serology including IgG- and IgM- and supplementary tests like IgG avidity assays. The most common scenarios for CMV serologic testing in pregnancy are as follows: (i) suspicion of CMV PI due to symptoms (e.g., with flu-like illness); (ii) ultrasound abnormalities suggestive of fetal CMV infection; and (iii) routine or targeted CMV screening in early pregnancy.
Particularly in a screening setting, a high sensitivity of IgM assays in the acute stage of CMV PIs is of utmost importance since common screening algorithms are based on IgG and IgM testing as a first step followed by IgG avidity testing as a second step only in case of positive IgM results (2–6).
The objective of the present study is to compare the performance characteristics of eight commercial IgM assays for the diagnosis of CMV PI in pregnancy.
MATERIALS AND METHODS
Samples and study population
In this study, we used leftover patient specimens from samples submitted to our laboratory between January 2010 and June 2021. Data on initial laboratory test results and patient data were retrospectively extracted from the laboratory information system. Serum samples were de-identified, and related data were anonymized before final analysis. Samples have been stored at −20°C before testing.
Specificity panel
•
A total of 40 serum samples from pregnant women sent to the laboratory for individual CMV testing (e.g., immune status and suspicion of CMV PI), who were either CMV seronegative [DiaSorin IgM and IgG antibodies negative (group 1, n = 20)] or with latent CMV infection [DiaSorin IgM antibody negative, IgG antibody positive, and VIDAS IgG avidity high (group 2, n = 20)].
•
Ten serum samples from non-pregnant patients sent to the laboratory for diagnosis of infectious diseases (potentially interfering serum samples, group 3), who had either acute Epstein–Barr virus (EBV) infection (n = 5) or acute varicella zoster virus (VZV) infection (n = 5).
Sensitivity panel
A total of 100 acute phase or follow-up serum samples from 39 pregnant women. We reviewed cases with confirmed CMV PI diagnosed in our laboratory between 2010 and 2021. Confirmed CMV PI was defined by IgG seroconversion or by detection of IgG and IgM antibodies in the presence of IgG antibodies with low avidity. For the sensitivity panel, only cases with a well-defined onset of infection were selected by the following criteria (IgM was not a selection criterion):
1.
Seroconversion of IgG antibodies (Enzygnost Anti-CMV IgG EIA; Siemens, Germany, or Serion ELISA classic Cytomegalovirus IgG; Virion/Serion, Germany). The onset of PI (week 0) was defined as the mean time interval between two consecutive samples with an interval of ≤4 weeks (n = 27 women).
2.
Equivocal/low positive IgG levels followed by an at least 4-fold increase of IgG antibodies in the presence of low IgG avidity (Vidas CMV IgG avidity EIA; bioMérieux, France). The onset of infection was defined as previously described (n = 12 women) (5).
Sera were categorized in four arbitrarily defined time periods of CMV PI: A: 0–7 weeks after onset of infection (n = 24 sera); B: >7–12 weeks after onset of infection (n = 21 sera); C: >12–20 weeks after onset of infection (n = 27 sera); D: >20 weeks after onset of infection (n = 28 sera). Time periods A and B were defined as acute stage (0–12 weeks) and periods C and D (>12 weeks) as the convalescent stage of CMV PI. We investigated one to six serum samples (median 3) per woman, but no more than two samples per time period. At least one serum in the acute (A,B) and the convalescent (C,D) stage was available from 24 of the 39 women.
IgM Immunoassays
All serum samples were tested in our laboratory for CMV-specific IgM using the following eight immunoassays: Medac (Hamburg, Germany): CMV-IgM-ELA test PKS; DiaSorin (Saluggia, Italy): Liaison CMV IgM II; Virion (Würzburg, Germany): Serion ELISA classic Cytomegalovirus IgM; Euroimmun (Lübeck, Germany): Anti-CMV-ELISA (IgM) and Anti-CMV-p52-ELISA (IgM); Aesku (Wendelsheim, Germany): AESKULISA Cytomegalovirus IgM; Roche (Mannheim, Germany): Elecsys CMV IgM cobas; Vircell (Granada, Spain): Cytomegalovirus ELISA IgM Capture. In 2022, the manufacturer of the CMV-IgM-ELA test PKS had ceased production.
All assays were performed according to the manufacturer’s instructions. Assay details are given in the Supplementary Table (Table S1). The DiaSorin and Roche assays were run on the automated platforms of the manufacturer. The other assays have been adapted to the BEPIII device from Siemens Healthineers (Erlangen, Germany). Equivocal IgM test results were classed positive throughout the study.
Statistics
Specificity was calculated as (number of sera with negative results/total number tested) x 100; sensitivity as (number of sera with positive + equivocal results)/total number tested x 100. GraphPad Prism 10.1.2 was used for all analyses. The clearance of CMV IgM antibodies following PI was analyzed using non-linear regression (one-phase decay).
RESULTS
The performance of the eight IgM immunoassays in the specificity panel is presented in Table 1. The specificity of the DiaSorin IgM assay was not evaluated since negative results of this assay were part of the selection criteria in groups 1 and 2. The Medac, Virion, and Roche assays showed the highest specificity of 98.0% (95% CI: 89.4–99.5) followed by EI-Lysate, EI-p52, and Aesku with 92.0% (95% CI: 80.8–97.8). The Vircell test had the lowest specificity of 72.0% (95% CI: 57.5–83.8) due to high cross-reactivity (80%) in sera with acute EBV or VZV infection (group 3). In this potentially interfering panel (n = 10), at least 10% cross-reactivity was observed in all but one assay (Medac). The highest cross-reactivity was observed in sera from patients with acute EBV infection.
TABLE 1
| 1: CMV seronegative | 2: CMV latent infection | 3: Acute EBV infection | 3: Acute VZV infection | 1–3: Diagnostic specificitya | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Manufacturer | pos | eq | neg | pos | eq | neg | pos | eq | neg | pos | eq | neg | neg/tested | % | 95% CIb |
| Medac | 1 | 19 | 20 | 5 | 5 | 49/50 | 98.0 | 89.4–99.5 | |||||||
| DiaSorinc | 20 | 20 | 1 | 4 | 1 | 4 | ND | ND | ND | ||||||
| Virion | 20 | 20 | 1 | 4 | 5 | 49/50 | 98.0 | 89.4–99.5 | |||||||
| EI-Lysate | 20 | 1 | 19 | 2 | 1 | 2 | 5 | 46/50 | 92.0 | 80.8–97.8 | |||||
| EI-p52 | 20 | 1 | 19 | 2 | 1 | 2 | 5 | 46/50 | 92.0 | 80.8–97.8 | |||||
| Aesku | 20 | 20 | 2 | 2 | 1 | 5 | 46/50 | 92.0 | 80.8–97.8 | ||||||
| Roche | 20 | 20 | 1 | 4 | 5 | 49/50 | 98.0 | 89.4–99.5 | |||||||
| Vircell | 3 | 2 | 15 | 1 | 19 | 5 | 2 | 1 | 2 | 36/50 | 72.0 | 57.5–83.8 | |||
a
Diagnostic specificity (%) = negative results/tested x 100.
b
CI, confidence interval.
c
specificity of DiaSorin was not evaluated.
d
eq, equivocal result; neg, negative result; pos, positive result; ND, not determined.
Figure 1 (Table S2, supplement) shows the IgM detection rates in pregnant women with a well-defined onset of CMV PI during different time periods (A to D, sensitivity panel) after the onset of infection. In the first 7 weeks (A), the sensitivity was 100% (95% CI: 85.8–100) in six assays (Medac, DiaSorin, Virion, EI-Lysate, EI-p52, and Vircell), whereas Aesku and Roche exhibited a sensitivity of 95.8% (95% CI: 78.9–99.9) and 91.7% (95% CI: 73.0–99.0), respectively. In period B (>7–12 weeks after onset), the EI-p52 and the Vircell showed a noticeable decrease of sensitivity to 76.2% (95% CI: 52.8–91.8) and to 85.7% (95% CI: 63.7–97.0), respectively. Four assays (Medac, DiaSorin, Virion, and EI-Lysate) were positive in all samples of the acute stage (100% sensitivity in stages A and B). With respect to IgM persistence in the convalescent stage (C and D), the Roche and the EI-p52 assay showed low IgM detection rates, whereas the DiaSorin, Virion, and EI-Lysate yielded high rates (>85%). Concordantly, positive IgM results in all eight assays were obtained in 75.6% (34/45) of acute stage sera and in 12.7% (7/55) of convalescent stage sera.
Fig 1
The kinetics of CMV-specific IgM antibodies following CMV PI is shown for each assay in Fig. 2. The time-dependent decrease of IgM levels was most prominent with the Medac, Virion, and EI-Lysate assays, being characterized by high levels IgM in the acute stage and low level IgM in the convalescent stage of infection. In those assays, high IgM values (Medac: >2.9 Index, Virion > 100 AU/mL, and EI-Lysate >2.8 index) were exclusively detected in the acute stage, whereas with the DiaSorin, EI-p52, and Roche tests, individual sera also presented with high level IgM in the convalescent stage. IgM antibody kinetics of individual cases is shown in Fig. S1 (supplement) for four women with four to six serum samples.
Fig 2
DISCUSSION
Seroconversion of CMV-specific IgG is the gold standard for diagnosis of CMV PI in pregnant women, but usually previous serologic findings before conception or early pregnancy and stored blood samples are lacking. A positive CMV-specific IgM test may indicate CMV PI but can also result from recurrent infection and persistence of IgM for several months following a past PI or unspecific reaction. Therefore, supplementary testing (by IgG avidity tests, immunoblot, or line immunoassays) is required for diagnosis and narrowing the onset of CMV PI. Detection of moderate to high levels of IgM in the presence of low-avidity IgG confirms a CMV PI usually in the last 12–16 weeks depending on the avidity test manufacturer. Generally in CMV PI, IgM antibody levels reach their maximum in the first month after the onset of infection and then decline within the following 3–6 months (7). However, antibody kinetics varies between individuals and with the sensitivity of the assay used (8, 9).
In this work, we describe the performance characteristics of eight commercial IgM assays for the diagnosis of CMV PI in pregnancy. The specificity was examined in a small panel of pre-selected sera from daily routine serologic testing. Except for one assay with lower specificity (Vircell: 72%), all others performed well with specificities of 92%–98%. The latter is in agreement with previous reports (10, 11). Unspecific false-positive IgM results may be due to, e.g., autoantibodies, serum binding proteins, or heterophile antibodies. According to a recent review, simultaneous positivity for both EBV- and CMV-IgM antibodies in a patient seems to be more common as previously reported (12). This is consistent with our observation of high IgM cross-reactivity in acute EBV infection and emphasizes that positive IgM findings require further diagnostic clarification.
Concerning sensitivity and antibody kinetics, considerable differences between the investigated tests were observed. From the four assays with 100% sensitivity in the acute phase of PI, the Medac test showed the lowest detection rate >20 weeks after onset of PI (D: 42.9%), whereas three assays (DiaSorin, Virion, and EI-Lysate) still recorded more than 85% of sera with positive/equivocal IgM results in period D. Depending on the test scenario in pregnancy, CMV-specific IgM tests have to meet different needs. Serologic screening in early pregnancy requires IgM assays with high sensitivity in the acute stage and low IgM detection rates in the convalescent stage. One goal of screening in early pregnancy is to identify women who are seronegative and might benefit from hygiene counseling (13). A further aim is the timely detection of CMV PI to offer treatment options for secondary prevention of fetal infection (14, 15). Recurrent CMV infections are not the target of serologic screening, as they can hardly be recognized by diagnostic means. As shown in a recent meta-analysis by Chatzakis et al., the risk for long-term sequelae in children with congenital CMV infection is greatest following maternal PI in the periconceptional period (broadly defined as 4 weeks before until 6 weeks after the last menstrual period) and in the first trimester (1). Therefore, screening should ideally be performed before 8 gestational weeks using an assay that is sensitive enough to detect CMV-specific IgM for at least 12 weeks after the onset of PI. High sensitivity in the acute stage is especially important, if the screening algorithm includes IgG and IgM antibody detection as the first step and (in case of a positive IgM) IgG avidity testing as the second step.
In our study, four IgM assays (Medac, DiaSorin, Virion, and EI-Lysate) fulfilled the requirements of excellent sensitivity up to 12 weeks after onset of CMV PI, whereas the others displayed individual gaps. While we were able to demonstrate a sensitivity of 100% for the DiaSorin test during the acute stage of infection, others reported early clearance of IgM antibodies in about 5% of patients within 12 weeks after the onset of infection (8). The recombinant EI-p52 (anti-p52 CMV IgM) showed the lowest sensitivity (76%) of all immunoassays in period B (>7 to 12 weeks after onset of infection). A similar lack in sensitivity has also been observed by Zelini et al. The EI-p52 assay was designed to serve as a second-line assay together with the Euroimmun recombinant anti-gB IgG EIA (16). However, the reduced sensitivity of a second-line IgM test can lead to a CMV PI being misclassified as not acute, in particular when other assays used to date the onset of infection (e.g., IgG avidity and immunoblot) give inconclusive results. The use of both recombinant Euroimmun assays (anti-p52 CMV IgM and anti-gB CMV IgG) as first-line screening in comparison to the conventional approach with the lysate-based IgM-, IgG-, and IgG avidity Euroimmun assays was recently investigated (17). In daily routine, the recombinant testing approach resulted in a lower number of inconclusive results and reduced laboratory workload. However, diagnostic accuracy of the applied IgM tests was not an objective of this study. The Roche test (ECLIA) investigated in our series also uses recombinant antigens and was initially designed to identify the IgM response during the acute stage (10). However, a relevant number of negative IgM test results were found during the acute stage (A: 8%; B: 9%). These findings are in concordance with those of a previous study reporting a relatively high rate of negative IgM results (15%) up to 90 days after maternal infection (10).
Three assays with 100% sensitivity in the acute phase in our study (DiaSorin, Virion, and EI-Lysate) showed still a high IgM detection rate (>85%) in the convalescent stage >12–20 weeks and even >20 weeks after the onset of CMV PI, which is not desirable in a screening setting in early pregnancy. Long-persisting IgM detected by those assays results from CMV PI that occurred long before pregnancy and is assumed not to be associated with fetal risk. Positive or equivocal IgM results always require further diagnostic workup and may lead to unnecessary anxiety in women and in the worst case, if pregnancies are incorrectly managed, to termination of pregnancy.
In contrast, if CMV serology is performed for the first time in the presence of abnormal ultrasound findings in the second or third trimester, IgM assays with a high detection rate more than 20 weeks after primary infection might be beneficial. They may give hints that fetal abnormalities can be caused by a PI CMV infection in early pregnancy or around conception.
Quantitatively, IgM antibodies decreased from high levels in the acute stage to lower levels more than 12 weeks after onset of infection by all assays. We observed that at least three assays had IgM antibody levels above a certain value only in the acute phase but not thereafter. In contrast, two assays based on recombinant antigens (EI-p52, Roche) and the DiaSorin assay detected high-level IgM in several sera of the convalescent stage. This may lead to diagnostic pitfalls: high-level IgM in combination with a delayed avidity maturation, which has been described in 12% of pregnant women with CMV PI (8), might mistakenly be interpreted as acute PI.
A limitation of our study is the relatively small sample size, especially in the specificity panel. In the sensitivity panel, we included only acute and follow-up specimens from pregnant women with a very well-defined onset of infection. Appropriate data sets to evaluate the reliability of assays for detection of acute PI are difficult to obtain. To extrapolate our results to a screening setting, further large-scale studies with concomitant IgG avidity testing are desirable.
In conclusion, commercial CMV IgM assays show significant differences in IgM detection rates in the acute and convalescent stage of CMV PI. As no test fits all needs, the laboratory must be aware of the performance characteristics of the current test in use.
ACKNOWLEDGMENTS
We thank all manufacturers for contributing the test kits free of charge.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
SUPPLEMENTAL MATERIAL
Fig. S1 - jcm.01407-23-s0001.tif
Kinetics of CMV-specific IgM antibody in eight commercial immunoassays in 4 pregnant women with primary CMV infection with 4-6 serum samples (time periods A: 0-7 weeks, B: >7-12 weeks, C: >12-20 weeks, D: >20 weeks after onset of infection). Assay-specific symbols are unfilled with thin border lines in case of negative IgM according to the manufacturers' cut-offs. In order to improve the visibility the y-axes were enlarged in the range of negative to low positive results.
- Download
- 2.08 MB
Legend to Figure S1 - jcm.01407-23-s0002.docx
- Download
- 13.50 KB
Table S1 - jcm.01407-23-s0003.docx
Characteristics of eight commercial immunoassays for detection of CMV-specific IgM antibodies.
- Download
- 16.94 KB
Table S2 - jcm.01407-23-s0004.docx
Detection rate of CMV-specific IgM antibodies in eight commercial IgM immunoassays in 100 sera of 39 pregnant women with well-defined onset of CMV primary infection.
- Download
- 16.09 KB
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.
REFERENCES
1.
Chatzakis C, Ville Y, Makrydimas G, Dinas K, Zavlanos A, Sotiriadis A. 2020. Timing of primary maternal cytomegalovirus infection and rates of vertical transmission and fetal consequences. Am J Obstet Gynecol 223:870–883.
2.
Delforge M-L, Eykmans J, Steensels D, Costa E, Donner C, Montesinos I. 2019. Combination of line immunoassays Mikrogen recomLine CMV IgG and recomLine CMV IgG avidity helps to date the onset of CMV primary infection. Diagn Microbiol Infect Dis 93:208–212.
3.
Leruez-Ville M, Ville Y. 2020. Is it time for routine prenatal serological screening for congenital cytomegalovirus?. Prenat Diagn 40:1671–1680.
4.
Rawlinson WD, Boppana SB, Fowler KB, Kimberlin DW, Lazzarotto T, Alain S, Daly K, Doutré S, Gibson L, Giles ML, Greenlee J, Hamilton ST, Harrison GJ, Hui L, Jones CA, Palasanthiran P, Schleiss MR, Shand AW, van Zuylen WJ. 2017. Congenital cytomegalovirus infection in pregnancy and the neonate: consensus recommendations for prevention, diagnosis, and therapy. Lancet Infect Dis 17:e177–e188.
5.
Enders G, Daiminger A, Bäder U, Exler S, Schimpf Y, Enders M. 2013. The value of CMV IgG avidity and immunoblot for timing the onset of primary CMV infection in pregnancy. J Clin Virol 56:102–107.
6.
Fourgeaud J, Nguyen C-Â, Guilleminot T, Ville Y, Leruez-Ville M. 2023. Comparison of two serological screening strategies for cytomegalovirus primary infection in the first trimester of pregnancy. J Clin Virol 169:105614.
7.
Revello MG, Gerna G. 2013. State of the art and trends in cytomegalovirus diagnostics, p 382–401. In Reddehase M (ed), Cytomegaloviruses: from molecular pathogenesis to intervention. Vol. 2. Caister Academic Press, Norfolk, UK.
8.
Sarasini A, Arossa A, Zavattoni M, Fornara C, Lilleri D, Spinillo A, Baldanti F, Furione M. 2021. Pitfalls in the serological diagnosis of primary human cytomegalovirus infection in pregnancy due to different kinetics of IgM clearance and IgG avidity index maturation. Diagnostics (Basel) 11:396.
9.
Maine GT, Stricker R, Stricker R. 2012. Kinetics of CMV seroconversion in a Swiss pregnant women population. Diagn Microbiol Infect Dis 73:275–277.
10.
Revello MG, Vauloup-Fellous C, Grangeot-Keros L, van Helden J, Dickstein Y, Lipkin I, Mühlbacher A, Lazzarotto T. 2012. Clinical evaluation of new automated cytomegalovirus IgM and IgG assays for the Elecsys analyser platform. Eur J Clin Microbiol Infect Dis 31:3331–3339.
11.
BaAlawi F, Robertson PW, Lahra M, Rawlinson WD. 2012. Comparison of five CMV IgM immunoassays with CMV IgG avidity for diagnosis of primary CMV infection. Pathology 44:381–383.
12.
Jhaveri TA, Harris C, Sax PE. 2022. IgM positivity for both EBV and CMV: a clinical conundrum. Open Forum Infect Dis 9:ofac316.
13.
Revello MG, Tibaldi C, Masuelli G, Frisina V, Sacchi A, Furione M, Arossa A, Spinillo A, Klersy C, Ceccarelli M, Gerna G, Todros T, CCPE Study Group. 2015. Prevention of primary cytomegalovirus infection in pregnancy. EBioMedicine 2:1205–1210.
14.
Amir J, Chodick G, Pardo J. 2023. Revised protocol for secondary prevention of congenital cytomegalovirus infection with valaciclovir following infection in early pregnancy. Clin Infect Dis 77:467–471.
15.
Chatzakis C, Shahar-Nissan K, Faure-Bardon V, Picone O, Hadar E, Amir J, Egloff C, Vivanti A, Sotiriadis A, Leruez-Ville M, Ville Y. 2023. The effect of valacyclovir on secondary prevention of congenital cytomegalovirus infection, following primary maternal infection acquired periconceptionally or in the first trimester of pregnancy. an individual patient data meta-analysis. Am J Obstet Gynecol 230:109–117.
16.
Zelini P, Fornara C, Furione M, Sarasini A, Klemens J, Arossa A, Spinillo A, Gerna G, Lilleri D. 2019. Determination of anti-p52 IgM and anti-gB IgG by ELISA as a novel diagnostic tool for detection of early and late phase of primary human cytomegalovirus infections during pregnancy. J Clin Virol 120:38–43.
17.
Müller J, Flindt J, Pollmann M, Saschenbrecker S, Borchardt-Lohölter V, Warnecke JM. 2023. Efficiency of CMV serodiagnosis during pregnancy in daily laboratory routine. J Virol Methods 314:114685.
Information & Contributors
Information
Published In
Journal of Clinical Microbiology
Volume 62 • Number 4 • 10 April 2024
eLocator: e01407-23
Editor: Elitza S. Theel, Mayo Clinic Minnesota, USA
PubMed: 38426762
Copyright
© 2024 American Society for Microbiology. All Rights Reserved.
History
Received: 27 October 2023
Accepted: 4 February 2024
Published online: 1 March 2024
Keywords
Contributors
Editor
Elitza S. Theel
Editor
Mayo Clinic Minnesota, USA
Metrics & Citations
Metrics
Note:
- For recently published articles, the TOTAL download count will appear as zero until a new month starts.
- There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
- Citation counts come from the Crossref Cited by service.
Citations
Citation
2024.
Performance of eight commercial immunoassays for the detection of cytomegalovirus-specific IgM antibodies in pregnancy – no test fits all needs. J Clin Microbiol
62:e01407-23.
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.