Case Report.

K. Gyl lens ten , A. S6nnerborg , C. Jorup-R6ns tr6m, M. Halvarsson, Z. Yun

Parvovirus B19 Infection in HIV-1 Infected Patients with Anemia

Summary: Serum samples were analysed for IgM and IgG antibodies to parvovirus by ELISA and for parvovirus B19 D N A by polymerase chain reaction (PCR) in 69 HIV-1 in- fected Swedish patients/vith anemia and in 37 HIV.1 infected subjects without anemia. In 5/69 anemic patients, parvovirus B19 DNA was detected despite the lack of IgM antibody activity to the virus. The detection of parvovirus BL0 D N A was significantly correlated to the degree of anemia in the anemic patients. In two patients who had a chronic anemia, a persistent parvovirus infection was detected by PCR, but not by serology, for 1 and 1.5 years, respectively. The results suggest that persistent parvovirus infection is a rare cause of anemia, but important to identify, since the infection is potentially treatable with intrave- nous immunoglobulin.

Introduction

Parvovirus B 19 infects erythroid s tem cells and causes in- hibition of the erythropoiesis. Parvovirus infection is com- mon and the pr imary infection often occurs in childhood. Thus, 40-60% of adults exhibit I g G antibodies to the virus [1, 2]. The diagnosis of parvovirus infection was earlier bas- ed on the clinical symptoms. Several diagnostic methods have now been developed. Thus, specific IgM and IgG an- tibodies can be detected by E L I S A and the viral genome by hybridization or polymerase chain reaction (PCR) [ 1 - 5 1 .

The virus may cause several different clinical pictures. In children, the most common symptoms are fever and exan- thema, a syndrome often called erythema infectiosum [6-8]. Another clinical picture is symmetrical arthritis in adults [9, 10]. Pregnant women infected with B19 may give rise to children with hydrops fetalis or stillbirth [11, 12]. In parvovirus B19 infection, reticulocytopenia, slight leu- kopenia, lymphopenia and thrombocytopenia are often seen. This cytopenia does not result in clinical symptoms in individuals with unaffected immunity. However , in pa- tients with chronic hemolyt ic anemia, aptastic crises may appear [13, 14]. It has also been shown that parv0virus B19 may cause persistent anemia in immunocompromized pa- tients, e.g. patients with leukemia on cytotoxic t reatment , organ or bone mar row transplant recipients and A I D S pa- tients [15-24]. In these patients persistent virus replication often results in severe chronic anemia. The aim of the present study was to determine the preva- lence of persistent and previous parvovirus B19 infection in human immunodeficiency virus (HIV) infected individ- uals, and to relate the findings to the presence of anemia.

Materials and Methods

Patients: All HIV-1 infected patients attending the Department of Infectious Diseases at Roslagstull Hospital, Stockholm, Swe- den, and with a hemoglobin value of < 90 g/1 were included. Pa- tients with anemia of known origin were excluded, as well as all Africans because of possible sickle-cell anemia. The study was retrospective extending from 1986 until July 1992. The patients

were categorized according to stage of infection, and into asymp- tomatic (early, CDC II) and symptomatic (late, CDC IV a-d) HIV infection. Measurements of CD4+ cell counts and use of zidovudine treat- ment were also recorded. A group of consecutive patients with HIV-1 infection but without anemia or other HIV-related symp- toms served as a control group. Polymerase chain reaction: Serum samples had been stored for between 0.5-6 years. Five ~zl serum was used without previous extraction. PCR was run in a "nested" configuration. The pri- mers were obtained from the VP1 region of parvovirus B19 (ou- ter: ATAAATCCATATACTCATT and CTAAAGTATCCI'- GACCTTG; inner: TGGGTTTCAAGCACAAGTAG and GCCCCCTCACTCCACATG), corresponding to nucleotide positions 2936-2954, 3635-3617, 3001-3020, 3560-3541. The first PCR was run 35 cycles and the second PCR was run 20 cycles [25]. The total incubation volume was 50 ~1, containing 100 mM Tris-C1 (pH 8.3), 500 mM KCI, 2 mM MgC12 (optimized), 0.1% w/v gelatin, 50 p~M of each dNTP and 2 units of Taq-polymerase. The second PCR was performed containing 5 pJ product from the first reaction and the same buffer conditions. The PCR pro- ducts were assayed by electrophoresis of 4/xl amplified DNA on a 3% agarose gel (NuSieve; FMC, Rockland, ME, USA) con- taining 0.8 pJ/ml ethidium bromide and visualized with an UV light source. Assay of anti-parvovirus B19 antibodies: IgG.and IgM antibodies to parvovirus B19 were determined by ELISA using recom- binant parvovirus B19 VP-2 antigen (kindly provided by Dako- patts, Denmark), as earlier described [26]. One hundred ~1 of a 1:100 dilution of 10 pJ of serum was incubated overnight at room temperature, whereafter the plate was washed four times. Bound IgG or IgM antibodies were detected with a 1:500 dilution of al- kaline-phosphatase-conjugated rabbit anti-human IgG or IgM using diethanolamine as substrate. The absorbance was read at 405 nm. This assay has earlier been shown repeatedly and specif-

Received: 26 August 1993/Revision accepted: 11 July 1994 K. Gyllensten, M. D., Christina Jorup-ROnstrOm, M. D., Ph.D., M. Hal- varsson, R. N., Dept. of Infectious Diseases, Huddinge Hospital; A. SOnnerborg, Asst. Prof. Z. Yun, M. D., Division of Clinical Virology, Karolinska Institutet, Huddinge Hospital, Sweden. Correspondence to: Dr. Christina Jorup-ROnstrOm, Infectious Diseases Unit, S6dersjukhuset, S-11883 Stockholm, Sweden.

5 2 / 3 5 6 Infection 22 (1994) No. 5 © MMV Medizin Verlag GmbH Miinchen, Mtinchen 1994

ically to detect parvovirus B19 antibodies in immunocompetent individuals [27]. Statistical methods: Mann-Whitney's test and Spearman-Rank correlation test were used in comparative calculations.

Results

Altogether 76 patients fulfilled the inclusion criteria. Sev- en of these subjects were excluded, because the anemia was of known origin or stored sera were not found. Sixty- six patients had late stage disease and three early stage. Thirty-seven asymptomatic patients served as controls (Table 1). Anemic patients who were PCR positive had significantly lower hemoglobin values than anemic subjects who were PCR negative (mean 65 and 79 g/l, respectively, p=0.002). There was no significant correlation between CD4+ cell counts and PCR positivity (data not shown). No difference was found concerning the prevalence of IgG antibodies between the patients in the early and late stages of infec- tion. In two AIDS patients, parvovirus B19 DNA was detected repeatedly when analysing six and eight consecutive sam- ples obtained over 1 and 1.5 years, respectively. Neither of the two patients had IgM or IgG antibodies to parvovirus. In both patients, parvovirus DNA became undetectable after initiation of treatment with foscarnet and Zido- vudine, respectively. After cessation, however, parvovirus DNA became detectable again. Both patients had low hemoglobin levels (mean 83 and 73 g/l, respectively). One of them had received 150 units of erythrocytes over 1.5 years. A further 28 of the 66 late stage patients were treat- ed with zidovudine, and all of them were PCR negative.

Discussion

The usefulness of PCR and antibody analyses for diagnos- ing acute and previous parvovirus B19 infection in im- munocompetent individuals has earlier been proven [2, 4, 8, 25]. Analysis of IgM antibodies is known to be useful for the diagnosis of acute parvovirus infection in immuno- competent individuals. Furthermore, in one study IgM an- tibodies were found in 5/25 AIDS patients with severe anemia [23] and a high rate has also been reported in chil- dren with AIDS despite lack of anemia [28]. However, in the present study, IgG and IgM antibodies could not be found in two parvovirus DNA positive AIDS patients with chronic anemia. Thus, these and other data [19] suggest that the diagnosis of parvovirus B 19 infection in HIV-1 in- fected patients with advanced immunodeficiency cannot rely only on antibody immunoassay, but should be based on the detection of the virus. The prevalence of I'gG antibodies to parvovirus was high (64%) in the present study, i.e. slightly higher than in the general population [1, 2], despite the inclusion of severely immunosuppressed patients, who might have decreased capacity to synthesize antibodies. Hemophiliac patients with HIV infection have been reported to exhibit the high-

K. GyUensten et al.: Parvovirus B19 Infection and Anemia

Table 1: PCR and antibody assays of parvovirus in 106 HIV-1 infected patients.

Number 3 66 37 Hemoglobin g/1

Mean 69 84 143 Range 62-75 56-90 120-166

CD 4+ cells x 109/1 Mean 0.383 0.085 0.480 Range 0.040-0.840 0.001-0.320 0.230-1.430

Zidovudine therapy 0 29 0 Parvovirus

PCR+ 2 3 0 IgM+ 0 0 0 IgG+ 0 46 22

est prevalence (92%), suggesting that blood products may transmit the virus [29]. The rate of IgG positivity to parvovirus seems otherwise not to be increased in HIV in- fected subjects. A high frequency of parvovirus B19 infection has been re- ported in anemic HIV-1 infected patients [22]. In contrast, we found parvovirus B19 DNA in only 5/69 (7%) anemic HIV-1 infected patients. In these subjects, the presence of parvovirus correlated with the severity of the anemia. Par- vovirus DNA was not found in HIV-1 infected control per- sons without anemia. It seems unlikely that technical fac- tors contributed to these differences. Thus, the sensitivity and the specificity of the PCR assay is high. It can be ar- gued that a substantial portion of our anemic patients was treated with zidovudine. It seems very unlikely, though, that the presence of zidovudine or foscarnet inhibited the PCR resulting in false negative reactions. Thus, several other viruses, e.g. H I V and CMV, have been amplified with success from serum of HIV-1 infected patients, who have been treated with these drugs. It can- not be excluded with certainty that a potential improve- ment of the immune deficiency due to the zidovudine treatment might have led to the eradication of parvovirus from the blood in these patients. This seems less likely, how- ever, since many patients had been treated with zidovu- dine for a long period of time and the CD4+ cells had con- tinued to decrease in these patients. Although parvovirus infection seems to be a rare cause of anemia in Swedish HIV-1 infected patients, the diagnosis should be considered in severely anemic patients. In these subjects, intravenous immunoglobulin might be an alter- native therapeutic method to repeated blood transfusions [19, 30321.

Acknowledgements

This work was supported by the Swedish Medical Research Council (no. 9179) and Swedish Physicians against AIDS.

Infection 22 (1994) No. 5 © MMV Medizin Verlag GmbH Mt~nchen, Mtinchen 1994 357 / 53

K. Gyllensten et al.: Parvovirus B19 Infection and Anemia

Zusammenfassung: Parvovirus B19 Infektion bei HIV-infi. zierten Patienten mit Aniimie. Serumproben von 69 HIV-infi- zierten schwedischen Patienten mit An~imie und von 37 HIV- Infizierten ohne Anfimie wurden mittels ELISA auf IgM- und IgG-Antik6rper gegen Parvovirus B19 untersucht, auch wur- de mittels Polymerasekettenreaktion (PCR) das Vorliegen von Parvovirus B19 DNA geprtift. Bei 5/69 an~imischen Pati- enten war Parvovirus B19 DNA nachzuweisen, obwohl keine IgM-Antik6rperreaktion gegen das Virus vorlag. Zwischen dem Nachweis yon Parvovirus B19 DNA und dem Schwere-

grad der An~imie bestand bei den an~imischen Patienten eine signifikante Korrelation. Bei zwei Patienten mit seit 1 bezie- hungsweise 11/2 Jahren bestehender chronischer Angmie konnte unter Anwendung der PCR eine persistierende Parvo- virus-Infektion festgestellt werde n, filr die serologisch kein Hinweis vorlag. Die Ergebnisse lassen annehmen, dab die persistierende Parvovirus-Infektion eine seltene Ursache for eine An~imie darstellt, ihr Nachweis jedoch wegen der m6gli- then Behandlung mit intraven6sem Immunglobulin von klini- scher Relevanz ist.

References

1. Cohen, B., Mortimer, P., Pereira, M.: Diagnostic assays with mono- clonal antibodies for the human serum parvovirus-like virus. J. Hy- giene 91 (1983) 113-130.

2. Anderson, L., Tsou, C., Parker, R., Chorba, T., Wulff, H., Tattersaii, P., Mortimer, P.: Detection of antibodies and antigens of human par- vovirus B 19 by enzyme-linked immunosorbent assay. J. Clin. Micro- biol. 24 (1986) 522-526.

3. Anderson, M., Jones, S., Minson, A: Diagnosis of human parvovirus infection by dot-blot hybridization using cloned viral DNA. J. Med. Virol. i5 (1985) 163-172.

4. Clewley, J.: Polymerase chain reaction assay of parvovirus B19 DNA in clinical specimens. J. Clin. Microbiol. 27 (1989) 2647-2651.

5. Mnsiani, M., Azzi, A., Zerbini, M., Gibellini, D., Venturoli, S., Za- krzewska, K., Re, M., Gentiiomi, G., Gallineila, G., La Placa, M.: Neste d polymerase chain reaction assay for the detection of Bt9 par- vovirus DNA in human immunodeficiency virus patients. J. Med. Vi- rok 40 (1993) 157-160.

6. Anderson, M., Lewis, E., Kidd, L, Hall, S., Cohen, B.: An outbreak oferythema infectiosum associated with human parvovirus infection. J. Hygiene 93 (1984) 85-93.

7. Nnnoue, T., Okochi, K., Mortimer, Po, Cohen, P.: Human parvovirus (B19) and erythema infectiosum. J. Pediatrics 107 (1985) 38-40.

8. Erdman, D., Usher, J., Caui, O., Gary, W., Kajigaya, S., Young, N., Anderson, L: Human parvovirus B19 specific IgG, IgA, and IgM an- tibodiesand DNA in serum specimens from persons with erythema infectiosum. J. Med. Virol. 35 (1991) t10-115.

9. Reid, D., Reid, T., Brown, T., Rennie, J., Eastmond, C.: Human par- vovirus-associated arthritis: a clinical and laboratory description. Lancef~i (t985) 422-425.

10. White, D., Mortimer, P., Blake, D., Woolf, A., Cohen, B., Bacon, P.: Humafiparvovirus arthropathy. Lancet i (1985) 419-421.

11. Brown~IT., Anand, A., Ritchie, L', Clewley, J., Reid, T,: Intrauterine parvovirus infection associated with hydrops fetalis. Lancet ii (1984) 1033-I'034.

12. Knott, P., Welply, G., Anderson, M.: Serologically proved intraute- rine infeqtion with parvovirus. Br. Med. J. 289 (1984) 1660.

13, Serjeant, G.i Mason, K., Topley, J., Serjeant, B., Pattison, J., Jones, S., Mohamed, R.: Outbreak of aplastic crises in sickle cell anaemia associated With parvovirus-like agent. Lancet ii (i981) 595-597.

14. Lefrere, J., Courouc6, A. M., Bertrand, Y., Girot, R., Soulier, J. P.: Human parvovirus and aplastic crisis in chronic hemolytic anemias. Am. J. Hematok 23 (1986) 271-275.

15. Kurzman, G., Ozawa, K., Cohen, B., Hanson, G., Oseas, R., Young, N.: Chronic bone marrow failure due to persistent B19 parvovirus in- fection. N. Engl. J. Med. 317 (1987) 287-294.

16. Kurzman, G., Meyers, P., Cohen, B., Amnullah, A., Young, N.: Per- sistent B19 parvovirus infection as a cause of severe chronic anaemia in children with acute lymphocytic leukemia. Lancet ii (1988) 1159-1162.

i7. Couiombei, L., Morinet, F, Mielot, F., Tchernia, G.: Parvovirus in- fection, leukemia and immunodeficiency. Lancet i (1989) 101-102.

18. Weiland, H., Salimans, M., Fibbe, W., Kluin, P., Cohen, B.: Pro-

longed parvovirus B19 infection with severe anaemia in a bone mar- row transplant recipient. Br. J. Hematol. 71 (1989) 300.

19. Frickhofen, N., Abkowitz, J., Safford, M., Berry, M., Mayolo, J., Astrow, A., Cohen, R., Haiperin, I., King, L., Mintzer, D~, Cohen, B., Young, N,: Persistent B19 parvovirus infection in patients infected with human immunodeficiency virus type 1. Ann. Intern. Med. 113 (1990) 926-933.

20. Mayolo, J., Temple, D.: Pure red cell aplasia due to parvovirus B19 infection in a man with HIV infection. South. Med. J. 83 (1990) 1480-1481.

21. Chrystie, L, Almeida, J., Welch, J~: Case report: electron microscop- ic detection of human parvovirus (B 19) in a patient with HIV infec- tion. J. Med. Virol. 30 (1990) 249-252.

22. Naides, S., Howard, J. Swack, N., Stapleton, J.: Parvovirus B19 in- fection in HIV-1 infected persons failing or intolerant to zidovudine therapy. J. Infect. Dis. 168 (1993) 101-105.

23. Cinque, P., Monforte, A., Barbi, M., Boldorini, R., Luraschi, C., Ruggieri, A., Angeli, L., Costanzi, G., Lazzarin, A., Moroni, M.: Hu- man parvovirus B19: a possible causative agent of chronic anaemia in AIDS patients. World AIDS conference, Florence 1991, MB 2331.

24. Griffin, T., Squires, J., Timmons, C., Buchanan, G.: Chronic human parvovirus B19-induced erythroid hypoplasia as the initial manifes- tation of human immunodeficiency virus infection. J. Pediatrics 118 (1991) 899-901.

25. Koch-Weser, W., Adler, S.: Detection of human parvovirus B19 DNA by Using the polymerase chain reaction. J. Clin. Microbiol. 28 (1990) 65-69.

26. Brown, Ci, van Lent, J., Vlak, J., Spaan, W.: Assembly of empty cap- sids by using bacutovirus recombinants expressing human parvovirus B19 structural proteins?J. Virol. 65 (1991) 2702-2706.

27. Siinnerborg, A, Forsgren, M., Ahlander, L, Yun, Z. B.: Detection of acute parvovirus infection in immunocompetent patients. Ann. Meeting of Swed. Med. Soc. 1992, abstract MM 9GP.

28. Nigro, G., Luzi, G., Frideli, E., Ferrara, M., Pisano, P., Gattinara, G., Mezzaroma, I., Siiderlund, l~I., Rasnoveanu, D., Aiuti, F.: Parvovirus infection in children with AIDS. AIDS 6 (1992) 679-684.

29. Schwarz, T., Hottentriiger, B., Roggendorf, M.: Prevalence of anti- bodies to parvovirus B19 in selected groups of patients and healthy individuals. Int. J. Med. Microbiol. Virol. Parasitol. Infect. Dis. 276 (1992) 437-442.

30. Kurzman, G., Frickhofen, N., Kimball, J., Jenkins, D., Nienhuis, A., Young, N.: Pure red celi aplasia of 10 years' duration due to persist- ent parvovirus B 19 infection and its cure with immunoglobulin thera- py. N. Engl. J. Med. 321 (1989) 519-523.

31. Mitchell, S., Welch, J., Weston-Smith, S., Nicholson, F., Bradbeer, C.: Parvovirus infection and anaemia in a 15atient with AIDS. Geni- tourin. Med. 66 (1990) 95-96.

32. Gottlieb, F., Deutsch, J.: Red cell aplasia responsive to immunoglob- ulin therapy as initial manifestation of human immunodeficiency vi- rus infection. Am. J. Med. 92 (1992) 331-333.

54/358 Infection 22 (1994) No. 5 © MMV Medizin Verlag GmbH MiJnchen, Miinchen 1994