Skip to main content
Open Forum Infectious Diseases logoLink to Open Forum Infectious Diseases
. 2022 Jun 24;9(7):ofac316. doi: 10.1093/ofid/ofac316

IgM Positivity for Both EBV and CMV: A Clinical Conundrum

Tulip A Jhaveri 1,2,, Courtney Harris 3, Paul E Sax 4
PMCID: PMC9297312  PMID: 35873288

Abstract

A review of 28 patients who tested positive for both Epstein-Barr virus and cytomegalovirus immunoglobulin M at an academic medical center revealed that dual positivity is more common than previously reported. These cases require careful review of the history and sometimes supplemental testing. This report highlights features of patients with dual positivity and provides recommendations on interpretation of the results.

Keywords: EBV and CMV IgM, dual positivity


Infectious mononucleosis (IM) is a clinical syndrome of the reticuloendothelial and lymphatic system, mostly caused by Epstein-Barr virus (EBV) but also by other infections, including cytomegalovirus (CMV) [1]. Given the significant overlap of symptoms between acute EBV and CMV, it is difficult to distinguish EBV- from CMV-related mononucleosis. Exudative pharyngitis, cervical lymphadenopathy, and splenomegaly are more commonly associated with EBV mononucleosis, whereas individuals with CMV mononucleosis tend to have prolonged fevers and systemic symptoms. While in most cases both EBV- and CMV-induced mononucleosis resolve without specific therapy, in certain hosts, determining the specific cause of mononucleosis is important. For instance, primary CMV mononucleosis in pregnant women can cause maternal–fetal CMV transmission. Antiviral therapy such as ganciclovir, valganciclovir, or foscarnet may be indicated in immunocompromised individuals with CMV end-organ disease and/or high viral loads as well as in immunocompetent patients with protracted CMV mononucleosis with severe organ-specific complications of CMV.

EBV and CMV, members of the herpesvirus family, establish lifelong latent infection and can reactivate. More than 90% of adults have acquired at least EBV or CMV. EBV rates vary widely by age, geographic location, and race/ethnicity. A US-based National Health Examination Survey conducted among 9300 individuals aged 6–19 years between 2003 and 2010 showed that overall age-adjusted EBV seropositivity declined from 72% in 2003–2004 to 65% in 2009–2010, and from 88% among Mexican Americans and 88% among non-Hispanic Blacks to 64% among non-Hispanic Whites [2]. CMV follows similar seroprevalence trends. The US-based National Health Examination Survey conducted among >15 000 individuals aged 6–49 years between 1999 and 2004 showed an overall age-adjusted CMV seroprevalence of 50.4% [3]. With high baseline seropositivity rates, it is important to note that many patients undergoing testing for acute EBV and/or CMV infection may already have latent EBV and/or CMV.

A definitive diagnosis of acute EBV or CMV infection can be made by testing for specific immunoglobulin M and G (IgM and IgG) antibodies against EBV or CMV antigens in the right clinical context. Antibodies against EBV are formed to viral capsid antigens (VCAs), nonstructural proteins expressed early in the lytic cycle, and nuclear antigens (EBNA) expressed during latent infections and appear late during infectious mononucleosis. Acute primary EBV infection demonstrates positive EBV VCA IgM and IgG with negative EBNA, while past EBV infection shows positive EBNA and positive EBV VCA IgG with negative EBV VCA IgM. Acute CMV infection demonstrates positive CMV IgM and IgG, while past CMV infection shows negative CMV IgM with positive CMV IgG. EBV and CMV viral loads are not recommended as first-line tests for diagnosing suspected primary EBV or CMV infection, respectively, but may be useful as adjunctive tests to aid in diagnosis when serologic testing in not conclusive. This is particularly important in high-risk patients to assess disease severity and monitor response to therapy. IgM assays generally have a weaker IgM binding affinity to antigen compared with IgG binding affinity secondary to the antibody maturation effect. Further, the peptide used in the EBV IgM assay is small and synthesized, whereas the CMV IgM assay uses a whole-virus extract as the target antigen, which leads to a higher likelihood of contamination with cell proteins, and thereby higher chances of false positivity.

We noted that some people with mononucleosis syndromes may have simultaneous positivity of both EBV VCA IgM and CMV IgM, resulting in a diagnostic dilemma, as dual positivity could represent multiple different possibilities, including (1) acute infection due to EBV with false-positive CMV serology (which itself could be due to heterophile antibodies or viral reactivation); (2) acute infection due to CMV with false-positive EBV serology (due to viral reactivation); (3) sequential acute infections by EBV/CMV in either order with prolonged IgM positivity from the first infection; (4) simultaneous acute infection by EBV/CMV; (5) IM-like syndrome due to another etiology with false-positive EBV/CMV serology (which could be due to heterophile antibodies or viral reactivation). While there are data surrounding dual positivity of EBV and CMV serologies in pediatric populations in primary care offices [4, 5], there is a paucity of information surrounding interpretation of dual positivity in adults, especially in hospitalized immunocompromised hosts. The aim of this retrospective study was to ascertain the frequency of dual positivity of EBV VCA IgM and CMV IgM among adults tested in an academic medical center in urban Boston, with the goals of describing patients’ clinical characteristics and assisting in interpreting these and subsequent diagnostic tests.

METHODS

The Clinical Microbiology Laboratory at the Brigham and Women’s Hospital performs EBV and CMV serologic testing for inpatients in an academic tertiary care hospital, and for outpatients in its affiliated clinics. A small number of specimens from pediatric patients are tested from a few community clinics that also utilize the hospital’s laboratory services.

Among patients who underwent EBV VCA IgM and CMV IgM testing using the Liaison assay (DiaSorin, Saluggia, Italy) at Brigham and Women’s Hospital between January 1, 2016, and January 1, 2021, we investigated those who were positive for both EBV VCA IgM and CMV IgM. Their electronic medical records were reviewed to collect demographic and clinical characteristics and laboratory information, which included EBV VCA IgM, EBV VCA IgG, EBNA IgG, EBV early Ag IgG, EBV viral load, CMV IgM, CMV IgG, and CMV viral load.

The Institutional Review Board at Mass General Brigham approved the study. Two physicians independently investigated medical records. Cases were stratified based on age, gender, comorbidities, and need for hospitalization. On the basis of clinical history, additional laboratory values, relevant imaging, and treating physician’s assessment on chart review, the reviewing physician interpreted dual positive serologies into 1 of 5 broad categories—acute EBV, acute CMV, coinfection/dual infection suspected, other nonherpes infection, and equivocal/unclear.

RESULTS

Between 2016 and 2021, the BWH Clinical Microbiology Laboratory performed EBV VCA IgM tests on 4962 patients, of whom 885 had duplicate testing. Of 4077 unique patients, 268 (6.6%) tested positive, 6 were indeterminate, 32 were equivocal, and the rest tested negative; 49.6% were male, and 50.4% were female. Outpatients accounted for 2840 (69.6%) of specimens, inpatients 1143 (28%), and emergency room 102 (2.5%). The average age of patients tested was 47.3 years, and 3886 (95.1%) of the specimens were from patients aged ≥18 years, consistent with the predominantly adult population served by the hospital and its clinics. Ninety-nine percent of the pediatric specimens were from outpatients.

Of 268 unique EBV IgM–positive cases, 115 cases had CMV IgM testing performed during the same encounter, and in certain exceptions, during the same week. Dual testing for both EBV and CMV was done in individuals presenting with mononucleosis-like syndrome to assist with diagnosis. Of these, 28 were also positive for CMV IgM (dual positivity), or 24.4% of those for whom both EBV VCA IgM and CMV IgM tests were ordered who had positive EBV VCA IgM results.

A summary of clinical and laboratory characteristics of these patients with dual IgM positivity is shown in Table 1. For those who had dual positive serologies, the mean age was 36.1 years: 3 aged 4–18 years, 11 aged 19–34 years, 11 aged 35–50 years, and 3 aged 51–73 years. Fifteen (54%) patients were men, and 14 (50%) required admission to the hospital. Sixteen (57%) patients had significant comorbidities, 7 of whom were immunocompromised. All patients were symptomatic at the time of testing; 13 (46%) of these patients had symptom onset <2 weeks prior. Eighteen (64%) had fever, chills, and malaise, 14 (50%) had atypical lymphocytosis, 9 (32%) had elevated liver function tests, 6 (21%) had lymphadenopathy, and 5 (18%) had splenomegaly. Six of 17 (35%) had positive heterophile antibody. Twenty-five of 27 (93%) had positive EBV IgG, 8 of 9 (89%) had positive EBV early antigen IgG, 8 of 15 (53%) had positive EBNA IgG, and 8 of 18 (44%) had detectable EBV PCR. Thirteen of 22 (59%) had positive CMV IgG, and 8 of 20 (40%) had detectable CMV PCR. Six (21%) had follow-up serologies ordered.

Table 1.

Clinical Characteristics of Patients With Positive EBV VCA IgM Results who Also Tested Positive for CMV IgM

Clinical Characteristics No. (%)
Mean age, y 36.1
Age range, y
 (i) 4–18 3 (11)
 (ii) 19–34 11 (39)
 (iii) 35–50 11 (39)
 (iv) 51–73 3 (11)
 Male 15 (54)
 Significant comorbidities 16 (57)
 Immunocompromised 7 (25)
 Hospitalized 14 (50)
Acuity of symptoms
(i) Acute (<2 wk) 13 (46)
(ii) Subacute (2–6 wk) 11 (39)
(iii) Chronic (>6 wk) 4 (14)
Clinical presentation
 Fever, chills, diaphoresis, and malaise 18 (64)
 Atypical lymphocytosis 14 (50)
 Transaminitis 9 (32)
 Lymphadenopathy 6 (21)
 Splenomegaly 5 (18)
Laboratory testing
 Heterophile antibody positive 6/17 (35)
 EBV IgG positive 25/27 (93)
 EBV early antigen IgG positive 8/9 (89)
 EBNA IgG positive 8/15 (53)
 EBV PCR detectable 8/18 (44)
 CMV IgG positive 13/22 (59)
 CMV PCR detectable 8/20 (40)
 Follow-up serologies ordered 6 (21)

Assays used: EBV-specific serologies: Liaison assay (DiaSorin, Saluggia, Italy); CMV serologies: Liaison assay (DiaSorin, Saluggia, Italy); EBV PCR: EBV m2000 RealTime System (Abbott Molecular Inc., Des Plaines, IL, USA); CMV PCR: COBAS AmpliPrep/COBAS TaqMan CMV Test (Roche Molecular Diagnostics, Branchburg, NJ, USA).

Abbreviations: CMV, cytomegalovirus; EBNA, nuclear antigen; EBV, Epstein-Barr virus; IgG, immunoglobulin G; IgM, immunoglobulin M; PCR, polymerase chain reaction; VCA, viral capsid antigen.

Table 2 presents a case summary with interpretation of dual positive serologies. Ten patients were diagnosed with acute EBV and 6 with acute CMV. For 4 patients, dual infection could not be excluded, and in 6 patients, the diagnosis remained unknown. Two patients had alternate diagnoses; anaplasmosis and systemic lupus erythematosus (SLE) were implicated to be the etiologies of the patients’ clinical and laboratory presentations, with the positive EBV and CMV serologies being interpreted as false-positives. Six patients had follow-up serologies ordered ranging from 5 days to 6 months, with 3 of these follow-up tests assisting with the diagnosis. One patient died within 3 months of dual positive IgM test results. Further data on outcomes and duration of illness were unable to be uniformly obtained as many patients recovered and did not return for further follow-up.

Table 2.

Case Summary of Patients With Positive EBV VCA IGM Results who Also Tested Positive for CMV IgM

Pt ID Age/Sex Immunocompromised Presenting Symptoms and Signs Heterophile Ab EBV VCA IgM EBV VCA IgG EBV Early Ag, IgG EBV EBNA, IgG EBV DNA, IU/mL CMV IgM CMV IgG CMV DNA, IU/mL Follow-up Serology Final Interpretation Management/Outcomes/Other Coexisting Illness
1 48/M No Abdominal and chest paina Pos Pos Pos Detected; 752 Pos Pos Not detected No Acute EBV infectionn Also diagnosed and treated for Lyme disease and babesiosis
2 26/F No Night sweats, cervical lymphadenopathy Pos Pos Pos Neg Pos No Acute EBV infectionb Two weeks later, symptoms resolved except minimal fatigue
3 47/F Heart transplant recipient Nasal congestion, cough, diarrheaa Pos Neg Pos Pos Detected; <137 No Unclearb Cardiogenic shock thought to be precipitated by nonspecific viral infection, patient died
4 73/M Heart transplant recipient Volume overload, dyspneaa Pos Pos Not detected Pos Pos Not detected No Unclearb CHF exacerbation; EBV/CMV tests done to workup pulmonary nodules in transplant patient
5 43/M No Petechiae, fatigue, headachesa Pos Pos Not detected Pos Detected; <137 No Unclear ITP thought to be precipitated by a possible viral infection; platelet count improved with steroids
6 46/M No Nasal congestion, eye irritation, headache, cough, neck rash Neg Pos Pos Pos Pos No Unclear Treated for ehrlichiosis 1 mo prior; self-limiting symptoms, improved without follow-up
7 18/M No Sore throat, abdominal pain, fatigue, cervical lymphadenopathy Pos Pos Pos Pos Neg No Acute EBV infection Saw multiple providers over 3- wk period with persistent symptoms that eventually improved
8 60/F GCA on steroids Worsening from baseline GCA headache Pos Pos Pos Pos No Unclear Viral studies sent by patient’s rheumatologist for persistent acute on chronic headaches
9 47/F No Myalgias, headaches, fatigue, poor appetite, diarrhea, cervical lymphadenopathy Pos Pos Pos Neg Pos Neg No Dual infection Recently treated for Lyme; viral studies sent for atypical lymphocytosis and transaminitis; self-limiting symptoms, improved
10 29/F No Nausea, poor appetite, fatigue, hot flashes Pos Pos Pos Not detected Pos Pos Not detected No Dual infection Recently treated for Lyme; clinical improvement after 3 mo
11 69/F No Fevers and malaisea Neg Pos Pos Pos Pos Pos Not detected No Alternate diagnosis (false+ in the setting of anaplasmosis)b Treated with doxycycline for anaplasmosis; self-limiting symptoms, improved
12 21/F SLE on belimumab None Neg Pos Pos Neg Detected; 949 Pos Pos Not detected CMV IgM in 1 wk (+)
EBV DNA in 1 wk: undetectable
Acute EBV infectionb Viral studies sent for incidentally discovered atypical lymphocytosis after starting belimumab; patient later developed sore throat, eventually improved
13 30/F No Left upper quadrant abdominal pain, fevers, night sweatsa Neg Pos Pos Detected; 127 Pos Neg Detected;1210 copies CMV IgG in 4 wk (+)
EBV DNA in 4 wk: 285
Acute CMV infectionb Viral studies sent for atypical lymphocytosis; course complicated by infected splenic infarct requiring drainage and IV antibiotics
14 24/F No Headaches, severe fatigue, feversa Neg Pos Pos Pos Pos Not detected Pos Pos Detected; <137 No Acute EBV infectionb Viral studies sent for atypical lymphocytosis; symptoms improved during hospital stay
15 29/F Crohn’s disease on infliximab Sore throat, fatigue, gastroenteritis, cervical lymphadenopathya Pos Pos Pos Pos Detected; <100 Neg No Acute EBV infectionb Diagnosed initially as recent EBV infection with necrotizing encephalitis (thalamic infarcts with negative LP) concerning for postinfectious inflammatory syndrome; long, complicated course, but recovered well
16 22/M No Headache, fatigue, nausea, sore throat, scleral icterusa Pos Pos Pos Pos Neg Detected; 844 Pos Neg No Acute EBV infection Course complicated by respiratory distress and hemolytic anemia requiring steroids
17 26/M No Nausea, fevers, abdominal pain, pelvic lymphadenopathya Pos Pos Pos Detected; 1050 Pos Neg Neg No Acute EBV infection Hospitalized with initial concern for heme malignancy with atypical lymphocytosis, thought to be acute EBV due to elevated viral load
18 57/M No Fever, chills, night sweats, and myalgiasa Neg Pos Pos Not detected Pos Pos Detected;33 035 CMV IgM in 6 mo (−)
CMV IgG in 6 mos (+)
Acute CMV infectionb Initial workup concerning for hematologic malignancy, but negative flow and viral testing with positive findings
19 50/M No Dyspnea, fatigue, weight loss, migratory arthralgias with new pericardial effusiona Pos Pos Not detected Pos Pos Neg No Alternate diagnosis (false+ in the setting of SLE) Diagnosed with SLE by rheumatology during current admission
20 38/M No Cough, nightsweats, sore throat, myalgias, diarrhea, arthralgias Neg Pos Pos Pos Pos Not detected Pos Pos Detected; 477 No Acute CMV infectionb ID thought positive CMV IgM and viremia (even low level) consistent with primary CMV
21 26/M Type 1 insulin-dependent diabetes Fever, malaise, mild cougha Neg Pos Pos Neg Not detected Pos Pos Detected; 1354 No Acute CMV infection Was diagnosed outpatient with strep (positive testing) before hospital stay
22 19/M No No symptoms at time of lab test, elevated LFTs and atypical lymphocytosis on lab screen Pos Pos Pos Pos Neg Detected; 7440 Pos Pos Neg No Acute EBV infection No further case details, follow-up closer to home with PCP
23 47/M Psoriatric arthritis on secukinumab Fever, cough, dyspnea, and diarrheaa Pos Pos Detected; 4390 Pos Neg Detected; 55 572 CMV IgM in 5 d (+)
CMV IgG in 5 d (+)
Acute CMV infectionb Admitted to the hospital with F&N; found to have CMV pneumonitis, started on ganciclovir, transitioned to valganciclovir, completing over a month of therapy once 2 viral loads were undetectable
24 40/F No Cough, fevers, fatigue, myalgias, cervical lymphadenopathy, trunk rash Neg Pos Pos Pos Not detected Pos Neg CMV IgM in 6 wk (−)
CMV IgG in 6 wk (−)
Acute CMV infectionb Was being treated empirically with a 2-wk course of doxycycyline for Lyme disease (Lyme testing was eventually negative)
25 7/M No Headache, cervical lymphadenopathy, nasal congestion, cough, gastroenteritis Neg Pos Pos Neg Pos Neg No Dual infection Clinically improved within a week with only some residual swollen lymph nodes
26 4/M No Fever, headache, poor appetite, cervical lymphadenopathy, abdominal pain Pos Pos Neg Pos Neg No Dual infection PCP initially concerned about acute otitis media; patient failed to improve with a course of augmentin; re-presented with persistent fevers, found to have mono; recovered within a week
27 42/F No Headache, bilateral arthralgias, cervical lymphadenopathy, fatigue, low appetite Pos Neg Not detected Pos Neg Not detected CMV IgM in 1 wk (−)
CMV IgG in 4 wk (+)
EBV IgM in 1 wk (−)
EBV IgG in 4 wk (−)
Unclearb Persistent symptoms for 3-4 wk; no therapy given, did end up resolving; ID thought this was an acute mononucleosis syndrome but was suspicious of Bartonella or toxoplasmosis with cat exposure (both serologies neg)
28 23/F No Sore throat, emesis, fever, jaundice, lymphadenopathy, abdominal pain Neg Pos Pos Pos Pos Pos Not detected No Acute EBV infectionb Did not return after 2 wk of symptoms except to get labs, which showed improving LFTs and resolution of atypical lymphocytosis

Abbreviations: CHF, congestive heart failure; CMV, cytomegalovirus; EBNA, nuclear antigen; EBV, Epstein-Barr virus; F&N, febrile neutropenia; GCA, giant cell arteritis; ID, infectious diseases; IgG, immunoglobulin G; IgM, immunoglobulin M; ITP, idiopathic thrombocytopenic purpura; PCP, primary care physician; SLE, systemic lupus erythematosus; VCA, viral capsid antigen.

a

Hospitalized.

b

ID consulted.

DISCUSSION

In this review of patients testing positive for EBV IgM, simultaneously having a positive CMV IgM (dual positivity) was more common (∼25%) than previously reported in the literature. True antigen cross-reactivity triggering these results is unlikely as the amino acid sequences of these orthologues are quite different. A more plausible explanation is that interfering substances (such as autoantibodies, heterophile antibodies, and serum binding proteins) can lead to false-positives [6]. The diversity of clinical characteristics of cases makes individual interpretation of positive and negative test results challenging.

There are no clear guidelines on how to interpret dual positivity for these 2 viruses. Hence, diagnosis should be established after taking the overall clinical context into consideration via case-by-case discussion. In more than half the cases, CMV and EBV viral load testing helped resolve the uncertainty when significantly elevated for one diagnosis compared with the other, suggesting that elevated viral loads may also reflect reactivation. EBV antinuclear IgG positivity was useful in context with other testing to demonstrate remote EBV infection. Repeating these serologies in 4–6 weeks can, in some cases, help ascertain the etiological agent of mononucleosis. In a minority of our cases, dual positivity was thought to represent false-positive results for both viruses when a separate infection occurred. In our population in the Northeast United States, we saw several cases where a recent tick-borne infection was the culprit, and with transaminitis in the setting of fever and other nonspecific symptoms, EBV/CMV was challenging to diagnose. Ultimately, in a sizeable proportion of our cases, the final diagnosis remained unknown. As obtaining the final diagnosis in patients with dual positivity can represent a significant challenge, clinicians must acknowledge the need for diagnostic stewardship and the potential for unneeded therapies at the time of ordering serologic testing.

Dual seropositivity for EBV and CMV IgM may be more common than previously reported. One possible explanation is that most of the patients in this predominantly hospital-based cohort were sicker and more complicated than those typically seen in outpatient clinical care. As there is no definitive diagnosis when both EBV and CMV IgM results are positive, clinicians should strongly consider supplemental testing (especially additional serologies and viral load assays) to resolve the diagnostic dilemma.

Acknowledgments

We thank Fred Wang, MD, for his contributions to the design of the study during the process of drafting this article.

Financial support. The authors received no financial support for the research.

Potential conflicts of interest. All authors: no reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

Author contributions. T.A.J. and P.E.S. contributed to designing the study. T.A.J. wrote the first draft. T.A.J. and C.H. participated in data collection. All authors revised the manuscript and provided final approval.

Patient consent. The study did not include factors necessitating patient consent. The Institutional Review Board at Mass General Brigham approved the study.

Contributor Information

Tulip A Jhaveri, Division of Infectious Diseases, University of Mississippi Medical Center, Jackson, Mississippi, USA; Division of Medical Microbiology, Brigham and Women’s Hospital, and Harvard Medical School, Boston, Massachusetts, USA.

Courtney Harris, Division of Infectious Diseases, Brigham and Women’s Hospital, and Harvard Medical School, Boston, Massachusetts, USA.

Paul E Sax, Division of Infectious Diseases, Brigham and Women’s Hospital, and Harvard Medical School, Boston, Massachusetts, USA.

References

  • 1. Ishii  T, Sasaki  Y, Maeda  T, Komatsu  F, Suzuki  T, Urita  Y. Clinical differentiation of infectious mononucleosis that is caused by Epstein-Barr virus or cytomegalovirus: a single-center case-control study in Japan. J Infect Chemother  2019; 25:431–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Balfour  HH, Sifakis  F, Sliman  JA, Knight  JA, Schmeling  DO, Thomas  W. Age-specific prevalence of Epstein-Barr virus infection among individuals aged 6-19 years in the United States and factors affecting its acquisition. J Infect Dis  2013; 208:1286–93. [DOI] [PubMed] [Google Scholar]
  • 3. Bate  SL, Dollard  SC, Cannon  MJ. Cytomegalovirus seroprevalence in the United States: the National Health and Nutrition Examination Surveys, 1988-2004. Clin Infect Dis  2010; 50:1439–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Sohn  MJ, Cho  JM, Moon  JS, Ko  JS, Yang  HR. EBV VCA IgM and cytomegalovirus IgM dual positivity is a false positive finding related to age and hepatic involvement of primary Epstein-Barr virus infection in children. Medicine (Baltimore)  2018; 97:e12380. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Medović  R, Igrutinović  Z, Radojević-Marjanović  R, et al.  Clinical and laboratory differences between Epstein-Barr and cytomegalovirus infectious mononucleosis in children. Srp Arh Celok Lek  2016; 144:56–62. [DOI] [PubMed] [Google Scholar]
  • 6. De Carolis  S, Santucci  S, Botta  A, et al.  False-positive IgM for CMV in pregnant women with autoimmune disease: a novel prognostic factor for poor pregnancy outcome. Lupus  2010; 19:844–9. [DOI] [PubMed] [Google Scholar]

Articles from Open Forum Infectious Diseases are provided here courtesy of Oxford University Press

RESOURCES