Human Cytomegalovirus in Blood of Immunocompetent Persons during Primary Infection: Prognostic Implications for Pregnancy

Transcription

1 1170 Human Cytomegalovirus in Blood of Immunocompetent Persons during Primary Infection: Prognostic Implications for Pregnancy Maria Grazia Revello, Maurizio Zavattoni, Antonella Sarasini, Elena Percivalle, Lavinia Simoncini, and Giuseppe Gerna Institute of Infectious Diseases, University of Pavia, and Viral Diagnostic Service, IRCCS Policlinico San Matteo, Pavia, Italy Human cytomegalovirus (HCMV) was investigated in peripheral blood leukocytes (PBL) of 52 immunocompetent patients (40 pregnant women) with primary HCMV infection by quantitative determination of pp65 antigenemia, viremia, and leukodnaemia. pp65 antigenemia was detected in 12 (57.1%) of 21, 4(25%) of 16, and 0of 10 patients examined 1, 2, and 3months after onset, respectively. Viremia was detected in 5(26.3%) of 19 patients during the first month only. LeukoDNAemia was detected in 20 of 20, 17 (89.5%) of 19, and 9(47.3%) of 19 patients tested 1, 2, and 3months after onset, respectively. Four (26.6%) of 15 patients were still DNAemia-positive at 4 6months, whereas none were positive at ú6 months. HCMV was not detected in PBL of 20 HCMV-immune donors or of 9seropositive subjects with recurrent infection. Virus levels were low by all assays and did not correlate with clinical course of infection, intrauterine transmission, or severity of outcome. Invasive procedures in the presence of maternal leukodnaemia did not seem to interfere with vertical transmission of HCMV infection. Primary human cytomegalovirus (HCMV) infection in preg- The aims of the present study were to assess the diagnostic nancycarriesa50%riskofintrauterinetransmission,and10% value of virus detection in blood of nonimmunocompromised 15% of congenitally infected babies will be symptomatic at patients and to define safer protocols for the management of birth or will develop sequelae early during infancy [1]. Management pregnancywhencomplicatedbyprimaryhcmvinfection,that of pregnancy complicated by primary HCMV infec- is, whether detection or quantitation of HCMV in maternal tion requires adefinite diagnosis in the mother. Since most blood could be aprognostic marker of intrauterine transmis- infectionsaresubclinical,andvirusisolationfromurine,throat, sion. or genital secretions does not discriminate between primary andrecurrentinfections,diagnosisofprimaryhcmvinfection in immunocompetent persons still relies on serology [1]. Materials and Methods HCMV-specific IgG seroconversion represents the best diagnostic approach but rarely can be demonstrated in normal subbloodsampleswereobtainedfrom52adultsubjects(1 6samples/ Patients and clinical samples. One hundred one sequential jects. On the other hand, detection of virus-specific IgM resubject; quires cautious interpretation, particularly in pregnant women, median, 2). Subjects examined included 40 pregnant given the risk of false-positive results [2] and the sustained women, 7men, and 5nonpregnant women (mean age, 29 years; range,17 44)withdefinitediagnosisofprimaryHCMVinfection. IgMantibodyresponsefollowingprimaryinfection[3,4].Very Diagnosis was made according to virus-specific IgG seroconverrecently, the determination of HCMV-specific IgG avidity has sion in 9cases, presence of clinical symptoms (fever, myalgia) been shown to help in distinguishing primary from past or and/or abnormal liver enzyme values and virus-specific IgM in 35 recurrent infection [5]. Contrary to the extensive amount of cases, and only IgM to HCMV in 6cases. data obtained from studies in immunocompromised patients Twenty-three pregnant women (one twin pregnancy) included [6 8], very few data are available on the presence of HCMV in this study underwent prenatal diagnosis for congenital HCMV inbloodofnormal,nonimmunocompromisedpatientswithpriyears infection. Mean age of this subset of pregnant women was 28.7 mary infection [9 13]. (range, 17 39), and 11 (47.8%) of 23 were primiparous. None of the pregnant women included in this study were at risk for human immunodeficiency virus (HIV) infection, and testing for specific antibodies was not routinely done. Amniotic fluid and blood samples were obtained from 24 fetuses. Results of prenatal Received 11 June 1997; revised 25 November Presented in part: 6th International Cytomegalovirus Workshop, Orange diagnosis were confirmed in 21 newborns by urine culture at birth Beach, Alabama, 5 9March 1997 (abstract 54). and in 3fetuses by virus isolation from autopsy samples obtained Grantsupport:MinisterodellaSanità,RicercaFinalizzataIRCCSPoliclinico after termination of pregnancy. San Matteo (grant 820 RFM95/01). Peripheral blood lymphocyte (PBL) samples collected from 20 Corresponding author: Maria Grazia Revello, Servizio di Virologia, IRCCS HCMV IgG positive IgM negative healthy volunteers were used Policlinico San Matteo, Pavia, Italy (virology@ipv36.unipv.it). as controls. In addition, 9blood samples were obtained from 9 The Journal of Infectious Diseases 1998;177: by The University of Chicago. All rights reserved. subjectswithongoinghcmvrecurrentinfection.ofthese,6were /98/ $02.00 breast-feeding women with HCMV isolated from milk and 3were

2 JID 1998;177 (May) HCMV in Blood of Immunocompetents 1171 pregnant women excreting HCMV in urine. All 9 were IgG-posi- divided into 5 groups according to collection time after the tive and IgM-negative before pregnancy. onset of infection. Antigenemia was detected in 50% and 23.5% Virologic and serologic tests. HCMV was tested for in PBL of PBL samples examined within 1 and 2 months, respectively, by quantitative determination of antigenemia, viremia, and leukoafter onset. Viremia was detected only in 20.8% of PBL sam- DNAemia (L-DNAemia) by polymerase chain reaction (PCR). In ples examined during the first month. On the other hand, viral particular, antigenemia was quantitated by counting the number DNA was detected in 100% and 80.9% of samples collected of PBL positive for pp65 per cells examined on cytospin preparations stained by indirect immunofluorescence with a pool 1 and 2 months after onset, respectively, and in 39.1% and of three pp65-specific monoclonal antibodies [14]. Quantitation of 16.6% of specimens collected 3 and 4 6 months after onset, viremia was done by inoculation of PBL onto human respectively. The 4 samples that were DNA-positive at later embryonic lung fibroblast cell cultures grown in shell vials and times were drawn from 4 pregnant women 115, 112, 140, and by counting the number of fibroblast nuclei positive for HCMV 180 days after onset, respectively. No PBL sample collected p72 at 24 h after infection [15]. L-DNAemia relevant to exon 4 ú6 months after onset was found positive for L-DNAemia. of the major immediate early (IE1) gene was quantitated in 10 5 Antigenemia, viremia, and DNAemia results relevant to pa- PBL according to a reported PCR procedure [16] with some modi- tients were similar to those for PBL samples (table 1). fications [17]. Briefly, an internal control of amplification, con- In pregnant women with primary HCMV infection, presence sisting of a fixed amount (100 genome equivalents [GE]) of a of viral L-DNA in blood for 3 months did not seem to recombinant DNA molecule (pac2) flanked by the target seincrease the risk of intrauterine transmission of HCMV infecquences of outer primers (N1, nt ; N2, nt ) used for viral DNA amplification in clinical samples, was routinely tion. In fact, of the 8 pregnant women positive for L-DNAemia coamplified to detect PCR inhibitors. Samples negative for HCMV for 3 months (5 subjects) or more (3 subjects) and with known DNA but competent for amplification, as shown by PCR product virologic outcome of pregnancy, 4 (50%) did transmit the virus of the internal control, were submitted to a second (nested) step and 4 did not (table 2). of amplification with an inner set of primers (CM1, nt ; No PBL sample from 20 HCMV-seropositive blood donors CM2, nt ). By using the single-step PCR, samples con- was found positive for antigenemia, viremia, or L-DNAemia. taining ú10 GE were consistently amplified; samples containing Similarly, 9 PBL samples taken from 6 breast-feeding women 1 10 GE could be detected in the nested assay and were assigned excreting HCMV in milk and 3 pregnant women with HCMV an arbitrary value of 5 GE. Antigenemia and viremia were prospecviruria were negative for all virologic markers. tively determined, whereas L-DNAemia was determined retrospec- Quantitation of HCMV in PBL. Antigenemia levels in the tively on PBL aliquots stored at 080 C. 17 positive PBL samples (table 1) were low, ranging from 1 Prenatal diagnosis of congenital HCMV infection was done by virus isolation from amniotic fluid and viral DNA detection by PCR to 20 (median, 2) pp65-positive cells. Similarly, low levels of in amniotic fluid samples. In addition, antigenemia and viremia were viremia (1 2 positive cells) were observed in the 5 positive quantitated on fetal PBL, and virus-specific IgM was determined samples. Viral DNA was detected in very low amounts as well. on fetal serum samples obtained after 20 weeks of gestation [17]. Indeed, of the 25 PCR-positive samples collected during the Virus-specific IgG and IgM were determined by ELISAs developed first month after onset (table 1), only 15 (60%) were positive in the laboratory and previously described [4, 18]. by single-step PCR (median no. of GE, 20; range, ). Of the 30 PCR-positive samples collected afterwards, only 1 Results (3.3%) was positive by single-step PCR (10 GE) 40 days after onset of illness, whereas all of the remaining samples were Determination of HCMV antigenemia, viremia, and L- detected by nested PCR only, thus containing õ10 GE. DNAemia. Of 101 sequential PBL samples taken from 52 No correlation was observed between virus levels in blood adult subjects, 57 were examined for antigenemia, 56 for vire- and severity or duration of symptoms in the patients examined mia, and 97 for L-DNAemia (table 1). PBL samples were in this study. Four representative patients are reported in table Table 1. Results of HCMV detection in blood of 52 immunocompetent subjects (40 pregnant women) with primary HCMV infection. Antigenemia Viremia LeukoDNAemia Days after onset (median) PBL samples Patients PBL samples Patients PBL samples Patients 0 30 (21) 13/26 (50) 12/21 (57.1) 5/24 (20.8) 5/19 (26.3) 25/25 (100) 20/20 (100) (45) 4/17 (23.5) 4/16 (25) 0/18 0/12 17/21 (80.9) 17/19 (89.5) (80) 0/14 0/10 0/14 0/8 9/23 (39.1) 9/19 (47.3) (145) 0/0 0/0 0/0 0/0 4/24 (16.6) 4/15 (26.6) ú181 (212.5) 0/0 0/0 0/0 0/0 0/4 0/4 NOTE. PBL, peripheral blood leukocytes. Data are no. positive/no. examined (%).

3 1172 Revello et al. JID 1998;177 (May) Table 2. Duration of maternal DNAemia and intrauterine transmis- Within the limits of the few cases studied, no correlation sion of HCMV infection in 8 women with known virologic outcome was observed between levels of HCMV in blood and likelihood of pregnancy. of intrauterine transmission. In fact, of the 7 women with viral Day of first Virologic DNA levels 50 GE, 3 did transmit the virus and 4 did not, Duration of DNA-negative outcome of including the 2 pregnant women with the highest levels of Patient no. DNAemia (days) PBL sample pregnancy antigenemia (20 positive cells; table 2, patient 1) or L-DNAemia (398 GE; table 4, patient 10) No CI HCMV L-DNA in maternal blood and invasive procedures 2 Ç180 Ç200 CI NA No CI for prenatal diagnosis. We then investigated whether the No CI presence of HCMV in maternal blood at the time of procedures 5 77 NA No CI for prenatal diagnosis had any impact on the virologic outcome CI of pregnancy. Therefore, maternal L-DNAemia, results of pre CI natal diagnosis, and virologic outcome at birth were compara NA CI tively examined. Data were available from 23 mother-child NOTE. CI, congenital infection; NA, not available; PBL, peripheral blood pairs (table 4). Ten women were positive for L-DNAemia when leukocytes. prenatal diagnosis was done (table 4, patients 1 10), and 2 women (patients 2 and 10) had two sequential results positive for L-DNAemia. Antenatal testing was negative in 5 fetuses 3. Patients 1 (a pregnant woman) and 3 (a man) both presented (nos. 6 10). Of these, 4 were born without HCMV infection, with comparable levels of antigenemia and viremia, although whereas 1 (no. 6) was found to excrete HCMV in urine at patient 3 required hospitalization because of high-grade fever birth. On the other hand, 13 women were found to be negative and persistent headache whereas patient 1 had very mild clini- for L-DNAemia at the time of amniocentesis (table 4, patients cal symptoms. It must be noted that the antigenemia levels 11 23). In this group, prenatal diagnosis was negative in 7 detected in these 2 patients were the highest observed through- fetuses (nos ), and absence of infection was confirmed out the study. On the other hand, patient 2 (a pregnant woman) at birth. and 4 (a man) experienced mild to moderate HCMV-related Positive prenatal diagnosis results indicated that intrauterine clinical symptoms, and both had a low number of pp65-positive transmission of the infection had already occurred in 6 (60%) PBL in the earliest PBL sample examined. Patient 2 fully recov- of 10 fetuses sampled in the presence (table 4, nos. 1 5) and ered in a very short time, whereas in patient 4, asthenia per- in 6 (46.1%) of 13 fetuses sampled in the absence (table 4, sisted for at least 2 months. In this patient, however, neither nos ) of maternal L-DNAemia. All positive prenatal antigenemia nor viremia were detected by day 19, and viral results were confirmed either at birth or after termination of DNA was detectable only by nested PCR up to 60 days after pregnancy. In these cases, the possibility of an iatrogenic transmission the onset. of the virus to the fetus was obviously excluded. Table 3. Virologic and clinical findings in 4 patients with primary HCMV infection. HCMV in blood Day after Patient no., symptoms Clinical course onset Antigenemia* Viremia LeukoDNAemia IgM 1, asthenia, AST/ALT alteration Very mild , cough, asthenia, AST/ALT alteration Mild , headache, fever, AST/ALT alteration, thrombocytopenia Severe ú10 4, fever, asthenia, AST/ALT alteration Moderate (prolonged) ú ú10 60 ND ND NOTE. AST, aspartate aminotransferase; ALT, alanine aminotransferase; ND, not determined; PBL, peripheral blood leukocytes. * No. of pp65-positive PBL/ cells examined. No. of pp72-positive human embryonic lung fibroblasts inoculated with PBL. No. of genome equivalents. Absorbance at 492 nm of test serum/absorbance at 492 nm cutoff serum value.

4 JID 1998;177 (May) HCMV in Blood of Immunocompetents 1173 Table 4. Maternal, fetal, and neonatal findings in 23 mother-child pairs. Weeks gestation Prenatal Mother-child HCMV Prenatal Maternal HCMV DNAemia diagnosis pair no. infection diagnosis (no. of genome equivalents) result Outcome Pos (5) Pos CI Pos (5) Pos 23 Pos (5) Pos CI* Pos (5) Pos CI Pos (5) Pos CI Pos (5) Pos TOP / HCMV Pos (30) Neg CI Pos (5) Neg No CI 8? 23 Pos (5) Neg No CI 9? 23 Pos (5) Neg No CI Pos (398) Neg 22 Pos (5) Neg No CI Neg (0) Pos CI Neg (0) Neg No CI NOTE. Diagnostic criteria for prenatal diagnosis are reported in Methods. CI, congenital infection; TOP, termination of pregnancy; Pos, positive; Neg, negative. * 1 set of twins. Two pregnancies were terminated and HCMV was isolated from fetal autopsy samples. In conclusion, among the 12 women with negative prenatal failed to detect viral DNA by PCR in the monocyte fraction diagnosis, 1 (20%) of 5 positive for L-DNAemia could have [24] or in PBL [6, 24 27] or reported low (4% 6%) positivity transmitted the virus during invasive procedures. On the other rates [20, 28]. hand, none of the remaining 7 women negative for L-DNAemia Results of the present study support the concept that viral did transmit the virus. DNA is not detectable in PBL of HCMV-seropositive immunocompetent subjects. However, our data indicate that sporadic Discussion PCR positivity in blood donors (or healthy controls ) could be indeed observed (and properly interpreted) when considering The diagnostic impact of antigenemia, viremia, and L- that viral DNA can persist in PBL of 25% of persons for DNAemia assays in immunocompetent subjects has never been as long as 6 months after primary HCMV infection. On the investigated. This study shows that a rapid and specific diagnoú6 other hand, our data also indicate that 100% of subjects tested sis of primary HCMV infection can be done whenever any of months after the infection are negative for viral L-DNA- these virologic markers is detected in blood. This conclusion emia, and, moreover, that viral DNA is not detectable in PBL relies on the observation that in no other instance, including of HCMV-immune persons even during recurrent HCMV in- active recurrent HCMV infection, can virus be demonstrated fections. The possibility remains that mononuclear cells rather in blood of immunocompetent persons by the techniques herein than PBL may be privileged sites of HCMV persistence or used. Although the number of controls examined in this study latency [22]. is limited, additional data supporting this statement come from Determination of L-DNAemia led to diagnosis of primary the extensive experience we gained with immunosuppressed HCMV infection in all subjects examined within 1 month after patients (solid organ transplant recipients) in whom neither onset and in Ç90% of those tested within 2 months. Antigen- antigenemia nor viremia or DNAemia could be detected in emia and viremia showed lower sensitivity, detecting Ç50% PBL before transplantation. So far, only one study reported and 25% of positive patients, respectively, in the first month HCMV isolation from the blood of 2 of 35 normal donors [19], after the infection, whereas at 2 months, only 25% of patients but a more recent study failed to confirm these data with 86 remained antigenemia-positive and none were positive for vireblood donors [20]. mia. Thus, antigenemia and viremia appear to be unique mark- Presence of viral DNA in leukocytes is a more controversial ers of acute primary HCMV infection. The same seems to issue. In 3 reports investigating HCMV in healthy adult volun- apply to positive single-step PCR results, considering that 15 teers, viral DNA was found in virtually all seropositive persons (93.7%) of 16 PBL samples positive by single-step PCR and and in most seronegative persons when monocytes [21, 22] or containing ú10 GE were collected during the first month after PBL [23] were examined. Conversely, other investigators either the onset of infection.

5 1174 Revello et al. JID 1998;177 (May) The only previous studies in which HCMV isolation from study reported a significant association of needling procedures blood of immunocompetent patients with mononucleosis was during pregnancy (such as amniocentesis) and mother-to-child attempted reported either negative results [9] or isolation rates transmission of HIV-1 infection [32]. Therefore, at present we ranging from 16.6% [10] to 83% 100% during the acute phase check maternal blood for HCMV DNA before prenatal testing, [11 13] and from 14% to 16% during the convalescent phase and whenever possible, procedures are done only when HCMV [11, 12]. Multiple factors may account for the higher rate of DNA is no longer detected in blood. positivity for viremia reported in the latter studies: The majority Our findings also provide a new important diagnostic tool of patients had posttransfusion mononucleosis, which may have for the interpretation of positive IgM results in pregnant women resulted in higher virus load; PBL were used for virus by facilitating unequivocal diagnosis of acute or recent primary isolation, which included an initial observation period of 4 5 HCMV infection and a more precise definition of the onset of weeks followed by two blind passages of 2 3 weeks each; and subclinical infections. Dating of the infection is important both conventional virus isolation has been shown to be more sensitive for planning of prenatal diagnostic procedures [17, 33] and for than the shell-vial technique for viremia determination in prognosis (primary infections occurring before conception can the presence of very low amounts of infectious virus [15]. be conceivably considered at a lower risk of transmission). Contrary to what has been reported for immunocompromised However, gestational age does not seem to have any influence patients [6, 29], a clear correlation did not seem to exist between on the risk of intrauterine transmission in primary HCMV in- the amount of virus in blood and the clinical course of fection occurring during pregnancy. In agreement with a previ- the infection in the immunocompetent subjects examined in ous study [34], our own experience with 54 infants exposed to this study. Peak virus levels occurring very early after onset HCMV in utero indicated that transmission occurred in 50%, and/or the very low amounts of virus detected in blood of the 40%, and 71% of fetuses after maternal infection in the first, immunocompetent subjects may account for such a difference. second, and third trimester of pregnancy, respectively (unpublished As far as maternal virus load and the risk of transmission data). of HCMV from mother to child is concerned, this study does The major findings described herein are as follows. HCMV not show any correlation between virus levels in the mother can be detected in blood of immunocompetent persons only and intrauterine transmission of the infection. However, the during acute or recent primary infections. Neither infectious limited data examined do not lead to any final conclusion. virus nor viral components (pp65, DNA) are detectable in PBL Similarly, transmission of HIV-1 has also been reported to of HCMV-immune persons with or without concomitant recurrent occur at a wide range of maternal plasma HIV-1 viremia levels HCMV infections. Also, virus levels determined by speoccur [30]. Furthermore, the persistence of HCMV DNA in PBL for cific assays are low and do not correlate with the clinical course 3 months or ú3 months did not seem to be a factor associated of infection or with the likelihood of intrauterine transmission with a higher risk of fetal infection. of HCMV following primary infection during pregnancy. Finally, since L-DNAemia was detectable in Ç50% of patients ú60 days after the onset (an interval during which a procedure for prenatal diagnosis, when required, is most likely Acknowledgments to be done), we investigated whether procedures done in the We thank Giuliana Casali and Maria Torsellini for serologic presence of maternal HCMV L-DNAemia had any impact on assays, Umberto Nicolini and Alessandra Kusterman (Istituto Oste- the virologic outcome of pregnancy. In fact, our results indicate trico Ginecologico L. Mangiagalli, University of Milan) for that maternal DNAemia at the time of amniocentesis does not performing prenatal diagnosis procedures, and Linda D Arrigo for represent a major risk of iatrogenic transmission. On the other revision of the English. hand, such a risk cannot be ruled out. In fact, the only falsenegative prenatal diagnosis was obtained in a fetus sampled in References the presence of maternal L-DNAemia 4 weeks after maternal 1. Britt WJ, Alford CA. Cytomegalovirus. In: Fields BN, Knipe DM, Howley infection, whereas no congenital infection was detected in any PM, eds. Virology. New York: Raven Press, 1996: of the 7 newborns with negative prenatal diagnosis whose 2. Naot Y, Barnett EV, Remington JS. Method for avoiding false-positive mothers were negative for L-DNAemia. A recent study [31] results occurring in immunoglobulin M enzyme-linked immunosorbent assay due to presence of both rheumatoid factor and antinuclear antibodreported as many as 10 false-negative prenatal diagnoses when ies. J Clin Microbiol 1981;14:73 8. amniocentesis was done early (3.8 weeks) after the onset of 3. Stagno S, Tinker MK, Elrod C, Fuccillo DA, Cloud G, O Beirne AJ. maternal primary infection. These discrepant results may be Immunoglobulin M antibodies detected by enzyme-linked immunosorbent due to procedures done too close to maternal infection, as well assay and radioimmunoassay in the diagnosis of cytomegalovirus as to the erratic behavior of HCMV transmission or suboptimal infections in pregnant women and newborn children. J Clin Microbiol 1985;21: sensitivity of the technique used for HCMV detection (amniotic 4. Revello MG, Percivalle E, Zannino M, Rossi V, Gerna G. Development fluid culture). However, the possibility of accidental transmisand evaluation of a capture ELISA for IgM antibody to the human sion must be taken into consideration until properly designed cytomegalovirus major DNA binding protein. J Virol Methods 1991; studies provide conclusive results. Moreover, a very recent 35:

6 JID 1998;177 (May) HCMV in Blood of Immunocompetents Grangeot-Keros L, Mayaux MJ, Lebon P, et al. Value of cytomegalovirus 20. Smith KL, Kulski JK, Cobain T, Dunstan RA. Detection of cytomegalovi- (CMV) IgG avidity index for the diagnosis of primary CMV infection rus in blood donors by the polymerase chain reaction. Transfusion 1993; in pregnant women. J Infect Dis 1997;175: : Gerna G, Zipeto D, Parea M, et al. Monitoring of human cytomegalovirus 21. Stanier P, Taylor DL, Kitchen AD, Wales N, Tryhorn Y, Tyms AS. Persisinfections and ganciclovir treatment in heart transplant recipients by tence of cytomegalovirus in mononuclear cells in peripheral blood from determination of viremia, antigenemia, and DNAemia. J Infect Dis blood donors. Br Med J 1989;299: ;164: Taylor-Wiedeman J, Sissons JG, Borysiewicz LK, Sinclair JH. Monocytes 7. The TH, van der Ploeg M, van der Berg AP, Vlieger AM, van der Giessen are a major site of persistence of human cytomegalovirus in peripheral M, van Son WJ. Direct detection of cytomegalovirus in peripheral blood blood mononuclear cells. J Gen Virol 1991;72: leukocytes: a review of the antigenemia assay and polymerase chain 23. Bevan IS, Daw RA, Day PJ, Ala FA, Walker MR. Polymerase chain reaction. Transplantation 1992;54: reaction for detection of human cytomegalovirus infection in a blood 8. Locatelli F, Percivalle E, Comoli P, et al. Human cytomegalovirus infec- donor population. Br J Haematol 1991;78:94 9. tion in pediatric patients given allogeneic bone marrow transplantation: 24. Bitsch A, Kirkner H, Dupke R, Bein G. Failure to detect human cytomegarole of early antiviral treatment for HCMV antigenemia on patients lovirus DNA in peripheral blood leukocytes of healthy blood donors by outcome. Br J Hematol 1994;88: the polymerase chain reaction. Transfusion 1992;32: Klemola E, Kääriäinen L. Cytomegalovirus as a possible cause of disease 25. Shibata D, Martin WJ, Appleman MD, Causey DM, Leedom JM, Arnheim resembling infectious mononucleosis. Br Med J 1965;2: N. Detection of cytomegalovirus in peripheral blood of patients infected 10. Jordan MC, Rousseau WE, Stewart JA, Noble GR, Chin TDY. Spontane- with human immunodeficiency virus. J Infect Dis 1988;158: ous cytomegalovirus mononucleosis. Ann Intern Med 1973;79: Jiwa NW, Van Gemert GW, Raap AK, et al. Rapid detection of human 11. Rinaldo CR, Black PH, Hirsch MS. Interaction of cytomegalovirus with cytomegalovirus DNA in peripheral blood leukocytes of viremic transleukocytes from patients with mononucleosis due to cytomegalovirus. plant recipients by the polymerase chain reaction. Transplantation 1989; J Infect Dis 1977;136: : Rinaldo CR, Carney WP, Richter BS, Black PH, Hirsch MS. Mechanisms 27. Rowley AH, Wolinsky SM, Sambol SP, Barkholt L, Ehrnst A, Andersson of immunosuppression in cytomegaloviral mononucleosis. J Infect Dis JP. Rapid detection of cytomegalovirus DNA and RNA in blood of renal 1980;141: transplant patients by in vitro enzymatic amplification. Transplantation 13. Carney WP, Hirsch MS. Mechanism of immunosuppression in cytomega- 1991;51: lovirus mononucleosis. II. Virus-monocyte interactions. J Infect Dis 28. Cassol SA, Poon MC, Pal R, et al. Primer-mediated enzymatic amplifica- 1981;144: tion of cytomegalovirus (CMV) DNA. Application to the early diagnosis 14. Gerna G, Revello MG, Percivalle E, Morini F. Comparison of different of CMV infections in marrow transplant recipients. J Clin Invest 1989; immunostaining techniques and monoclonal antibodies to the lower 83: matrix phosphoprotein (pp65) for optimal quantitation of human cyto- 29. Gerna G, Percivalle E, Revello MG, Morini F. Correlation of quantitative megalovirus antigenemia. J Clin Microbiol 1992;30: human cytomegalovirus pp65-, p72-, and p150-antigenemia, viremia 15. Gerna G, Revello MG, Percivalle E, Zavattoni M, Parea M, Battaglia M. Quantification of human cytomegalovirus viremia by using monoclonal and circulating endothelial giant cells with clinical symptoms and antiviantibodies to different viral proteins. J Clin Microbiol 1990;28: ral treatment in immunocompromised patients. Clin Diagn Virol 1993; : Gerna G, Baldanti F, Sarasini A, et al. Effect of foscarnet induction treatine treatment, and the risk of transmission of human immunodeficiency 30. Sperling RS, Shapiro DE, Coombs RW, et al. Maternal viral load, zidovudment on quantitation of human cytomegalovirus (HCMV) DNA in pevirus ripheral blood polymorphonuclear leukocytes and aqueous humor of type 1 from mother to infant. N Engl J Med 1996;335: AIDS patients with HCMV retinitis. Antimicrob Agents Chemother 31. Mulongo KN, Lamy ME, Van Lierde M. Requirements for diagnosis of 1994;38: prenatal cytomegalovirus infection by amniotic fluid culture. Clin Diagn 17. Revello MG, Baldanti F, Furione M, et al. Polymerase chain reaction for Virol 1995;4: prenatal diagnosis of congenital human cytomegalovirus infection. J 32. Mandelbrot L, Mayaux MJ, Bongain A, et al. Obstetric factors and mother- Med Virol 1995;47: to-child transmission of human immunodeficiency virus type 1: the 18. Gerna G, Revello MG, Palla M, Percivalle E, Torsellini M. Microneutrali- French perinatal cohorts. Am J Obstet Gynecol 1996;175: zation as a reference method for selection of the cut-off of an enzymehuman 33. Nicolini U, Kustermann A, Tassis B, et al. Prenatal diagnosis of congenital linked immunosorbent assay for detection of IgG antibody to human cytomegalovirus infection. Prenat Diagn 1994;14: cytomegalovirus. Microbiologica 1992;15: Stagno S, Pass RF, Cloud G, et al. Primary cytomegalovirus infection in 19. Diosi P, Moldovan E, Tomescu N. Latent cytomegalovirus infection in pregnancy. Incidence, transmission to fetus, and clinical outcome. blood donors. Br Med J 1965;4: JAMA 1986;256: