点击显示 收起
Department of Virology, Haartman Institute, Faculty of Medicine
Division of Microbiology, Department of Basic Veterinary Sciences, Faculty of Veterinary Medicine, University of Helsinki
Department of Virology, Helsinki University Central Hospital Laboratory, Helsinki, Finland
Sindbis virus (SINV) is widespread in Europe, Africa, Australia, and Asia, but clinical infection occurs as epidemics in a few geographically restricted areas. We recently proved, by virus isolation from patients, that SINV is the causative agent of Pogosta disease, a mosquito-borne rash-arthritis occurring as larger epidemics every seventh year in Finland. Altogether, 86 patients with serologically verified SINV infection were recruited to the present study during the 2002 outbreak. We now describe in detail the duration, incidence, and characteristics of different symptoms; hematological parameters; antibody kinetics; and presence of SINV in different tissue samples. SINV RNA detection or virus isolation from blood and/or skin lesions was successful in 8 patients. Immunoglobulin (Ig) M antibodies became detectable within the first 8 days of illness, and IgG antibodies became detectable within the first 11 days of illness. During the acute phase of Pogosta disease, the typical symptoms were arthritis, itching rash, fatigue, mild fever, headache, and muscle pain. The most notable finding was that, in 50% of the patients, joint symptoms lasted for >12 months.
Sindbis virus (SINV), a member of the western equine encephalomyelitis virus complex of the genus Alphavirus in the family Togaviridae, is an enveloped virus with a genome of single-stranded, positive-polarity 11.7 kb RNA [1]. SINV infection is widespread in insects and vertebrates in Europe, Africa, Asia, and Australia; however, clinical disease caused by SINV is common only in geographically restricted areas. Serological evidence for an association between rash-arthritis epidemics and SINV was first noted in northern Europe, in the early 1980s, and this is the area in which the disease has mainly been reported. SINV was recently proven to be the causative agent of this disease, known as "Pogosta disease" in Finland, when SINV strains were isolated from whole blood and skin lesions of patients with acute serodiagnosed disease in 2002 [2]. Pogosta disease was first recognized in Finland in 1974, as an outbreak in the province of North Karelia [3]. Since then, a larger epidemic has taken place every 7 years, as in 2002.
The main vectors of SINV are ornithophilic mosquitoes of the genera Culex and Culiseta; these mosquitoes are abundant in late summer, which explains why most acute clinical cases in Finland occur in late July through early October. The incidence and prevalence of SINV infection varies considerably between different parts of the country, being highest in eastern and central Finland. The main amplifying hosts of SINV are probably different tetraonid and migratory bird species, which function as a major blood source for vector mosquitoes, and migratory birds can also transport the virus over long geographical distances [4, 5, 6]. Consistent with this, the northern European SINV strains have been recognized as being closely related to those from South Africa [2, 7].
Diseases clinically similar to Pogosta disease in nearby geographical areas include Ockelbo disease (also known as "August-September disease") in Sweden [8] and Karelian fever in Russian Karelia [9, 10]. Two previous studies have described clinical features of acute Ockelbo [11] and Pogosta [12] disease, outlining that arthritis, rash, fever, and nausea are the most typical symptoms. Some evidence of the possible association between prolonged arthritis and Pogosta disease was previously shown in 26 patients, 50% of whom had musculoskeletal symptoms 2.5 years after the onset of symptoms [13]. Similarly, 98 patients with Ockelbo disease were interviewed 34 years after infection, and 28% of them had arthralgia at the time [14, 15].
The IgM and IgG antibody kinetics of SINV infection have been examined by Vene et al. [16], by use of consecutive samples from 16 patients. Calisher et al. [17] showed that IgM antibodies can be detected as early as 1 day after the onset of symptoms; however, the exact time of IgM seroconversion has not been defined. Furthermore, hematological parameters of patients with acute SINV infection have not been characterized previously.
SINV has been isolated from human patients once in South Africa [18] and once in China [19]. SINV RNA has also been detected, by polymerase chain reaction (PCR), in skin lesions of Swedish patients with Ockelbo disease [20]. Furthermore, we recently described the first human SINV isolates from Europe, which were isolated from whole blood and skin lesions [2].
Here, we provide a detailed description of clinical and laboratory manifestations during the acute and prolonged stages of human SINV infection, which had not previously been available. We describe the duration, incidence, and characteristics of different symptoms; hematological parameters; antibody kinetics; and detection of the causative virus, by use of several methods, from different tissues from 86 patients with serologically verified SINV infection, representing the largest such series reported.
PATIENTS, MATERIALS, AND METHODS
Patients and samples.
This study was performed during the 2002 Pogosta disease outbreak, in 11 health care centers in North Karelia and in Kuopio University Hospital. When acute SINV infection was suspected during a physician's examination, the patient was recruited into the study. Informed consent was obtained from all patients or their guardians. The study was performed with the permission of the coordinating ethics committee of the Hospital District of Helsinki and Uusimaa.
The investigation included the following sample and data collection from patients: serum samples, whole-blood samples, skin-lesion biopsy specimens, a tripartite questionnaire, case history reports on physicians' examinations, and results of basic blood picture analysis. The acute-phase samples included serum samples, whole-blood samples in tubes containing EDTA as an anticoagulant, and skin-lesion biopsy specimens. New serum samples were collected from most patients 12 weeks and 56 months after infection. A questionnaire study was performed to collect descriptive information on the clinical picture of Pogosta disease. The patients filled out a questionnaire (handed to them at the health care center by the physician) during the acute phase and 56 months and 1213 months after infection. The questionnaires focused on the prevalence, onset, duration, and anatomical location of different symptoms, as well as on the patients' subjective experiences. Table 1 summarizes the patients, samples, and material available for this study.
Serodiagnostic methods.
All patients were screened for SINV IgM and IgG by EIA and were screened for total antibodies by hemagglutination inhibition (HI) test. The details and the evaluation of the EIAs will be published elsewhere. Briefly, for the EIA (modified from [16]), the serum samples were diluted 1 : 200 in PBS, 0.5% bovine serum albumin, and 0.05% Tween 20 and were applied to microtiter well plates coated with purified SINV (Ockelbo strain, Edsbyn 82/5 [21]). The plates were washed with PBS and 0.05% Tween 20, were incubated with rabbit antihuman IgG or IgM peroxidase conjugate, and again were washed. 3,3,5,5-Tetramethyl-benzidine (TMB) was used as substrate, and the reaction was stopped with 0.5 mol/L sulfuric acid. The result was read by a spectrophotometer (450 nm). For HI microtitration, serum samples were absorbed with kaolin and male goose erythrocytes and were tested against 3 hemagglutinating units of SINV antigen (collected from supernatants of SINV-prototype strain grown in BHK21/WI-2 cell monolayers) and a 0.2% suspension of goose erythrocytes [4]. IgG-avidity immunofluorescence assay (IFA) was performed by comparing the titers in the urea-washed IFA slides with those washed with PBS, essentially as described elsewhere for the Puumala virus IgG-avidity test [22].
The criteria for the diagnosis of acute SINV infection were (1) a 4-fold increase in HI titer between paired serum samples and, in addition, a positive result for IgM in either sample and/or IgG seroconversion; (2) a positive result for IgM, a negative/borderline result for IgG (which excludes prior immunity), and confirmation of the IgM result by a specific reaction (titer >10) in HI; or (3) a positive result for IgM and IgG with a low IgG-avidity index and confirmation of the IgM result by a specific reaction (titer >10) in HI. The gray-zone absorbance values for IgM EIA and IgG EIA were 0.30.4 and 0.250.35, respectively, the lower limits of which were calculated on the basis of the mean absorbance values of 40 SINV-negative samples plus 3 times the SD (author's unpublished data). All patients who gave a serum sample 56 months after infection (n = 50) were also screened for rheumatoid factor (RF) in serum by use of RapiTex (Dade Behring).
RNA extraction and nested reverse-transcription (RT) PCR.
Viral RNA was extracted from skin tissue samples by use of TriPure Isolation Reagent (Roche Molecular Biochemicals) and from whole-blood samples by use of PAXgene Blood RNA Kit (PreAnalytiX). For each experiment, negative (water) and positive (RNA from SINV supernatant, always handled in a separate laboratory) controls were included. The sequences of primers used for nested RT-PCR were as follows: 5-ATACGAC(C/A)AAAGCGGAGCAG-3 (outer forward [OF]), 5-AGTACGGGTCGTAACGGTTC-3 (outer reverse [OR]), 5-GATACTTTCTCCTCGCGAAATG-3 (inner forward [IF]), and 5-GTCTTGTAATCGCCGCACTTG-3 (inner reverse [IR]). For the RT reaction, RNA was incubated with 20 pmol of each outer primer, 40 U of M-MuLV reverse transcriptase (MBI Fermentas), 5 pmol (0.22 mmol/L) of dNTP mix (Finnzymes), and 80 U of ribonuclease inhibitor (MBI Fermentas) in RT buffer (MBI Fermentas), in a total volume of 22.5 L, for 90 min at 37°C.
For the outer PCR, 10 L of cDNA was incubated at 95°C for 5 min, cooled immediately on ice, and mixed with 32 pmol of OF, 24 pmol of OR, 20 pmol (0.2 mmol/L) of dNTP mix (Finnzymes), and 2.5 U of Taq DNA polymerase in PCR buffer with 200 mmol/L (NH4)2SO4 and 2.5 mmol/L MgCl2 (MBI Fermentas), in a total volume of 100 L. The PCR conditions were 39 cycles of 96°C for 45 s, 53°C for 45 s, and 72°C for 40 s; followed by a final extension for 10 min at 72°C.
For the inner PCR, 5 L of the outer PCR product was mixed with 28 pmol of each inner primer, 25 pmol (0.25 mmol/L) of dNTP mix (Finnzymes), and 2 U of Taq DNA polymerase (MBI Fermentas), as described above. The PCR conditions were 39 cycles of 95°C for 60 s, 56°C for 45 s, and 72°C for 30 s; followed by a final extension for 10 min at 72°C. The products were run in SeaKem LE agarose (BioWhittaker Molecular Applications). The sensitivity of the PCR assay was determined to be 0.1 infectious virions, by use of SINV supernatant as a control.
Immunohistochemical analysis.
The skin-lesion biopsy specimens were fixed in 10% formalin and embedded in paraffin blocks. The immunohistochemical assay was performed on 2 thin sections with Ventana Discovery (Ventana Medical Systems), by use of 3,3-diaminobenzidine (DAB) for staining and a mixture of 3 different serum samples from mice infected with SINV (2 with the strain Ockelbo Edsbyn 82/5 and 1 with the strain Ockelbo 520; provided by S. Vene, Swedish Institute for Infectious Disease Control). A skin-lesion biopsy from an SINV antibodynegative patient and SINV-infected Vero cells were used as controls.
RESULTS
Altogether, 131 patients with suspected acute Pogosta disease were recruited to the study in 11 health stations in the province of North Karelia and in Kuopio University Hospital during JulyOctober 2002. Eighty-six of the patients were shown to have an acute, serologically confirmed SINV infection. Altogether, 23 skin-lesion biopsy specimens and 76 whole-blood samples were retrieved from these patients. The serodiagnostic criteria described in Patients, Materials, and Methods were met in this series of 86 patients as follows: criterion 1 (4-fold titer increase between paired serum samples) was met by 65 of 86 patients, criterion 2 (IgM and HI positive but not IgG positive) was met by 9 of 86 patients, and criterion 3 (IgM, IgG, and HI positive with low IgG avidity) was met by 12 of 86 patients.
The joint symptoms were likely to last for >12 months (table 3). During the acute phase, 3 or more joints were usually affected, with symptoms including tenderness during movement and ache; edema was also common. Typically, the joint symptoms occurred in a fluctuating manner but did not change anatomical location. There was no considerable difference in symptoms according to the time of day. The duration of joint symptoms was significantly dependent on age; the average age of those with symptoms for >1 year was 46 years, whereas the average age of those with full recovery was 35 years (P = .014, Student's t test). Although females tended to more often have a prolonged duration of symptoms than males, the difference was not significant (P = .0729, 2 test).
We also investigated the impact of SINV infection on the patients' everyday lives (table 4). Most patients considered the severity of handicap and the need for daily adjustments in their everyday lives to be mild or moderate. However, more than one-third of patients graded the subjective handicap caused by the disease as considerable or major.
Table 5 presents the serological and virological findings. Forty-eight percent of the first serum samples collected from the 86 patients had positive IgM results, suggestive of acute infection. Sixteen percent of the patients had a borderline IgM result in the first serum sample. SINV RNA could be detected by RT-PCR in 4 of 23 skin-lesion biopsy specimens and in 5 of 73 whole-blood samples. Virus isolation was successful from 4 of 23 skin-lesion samples and from 1 of 73 whole-blood samples, as we have reported elsewhere [2]. Altogether, SINV could be either isolated or detected by RT-PCR (or both) from 8 of 86 patients. These samples were collected during the first 29 days of illness (data available from 6/8 cases). Of 23 skin-lesion biopsy specimens, 21 were analyzed twice by immunohistochemical assay, and no SINV antigen was detected by staining with mouse antisera raised against SINV.
To define the antibody kinetics in SINV infection, we evaluated the serum samples from 50 patients from whom a serum sample was available 56 months after infection. The patients were evaluated for IgM and IgG antibodies in a series of 24 serum samples per patient (figure 4). IgM antibodies became detectable within the first 8 days of illness. Thirty-six percent of the patients had IgM antibodies detectable 56 months after infection; however, 2 of them were RF positive. IgG antibodies became detectable within the first 11 days of illness. In each case, the IgG absorbance value was higher at 56 months after infection than during the acute phase (average absorbance value difference, 2.0).
Clinical hematological and infection laboratory parameters were analyzed during the acute phase, including white and red blood cell count, hemoglobin level, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, platelet count, lymphocyte percentage, neutrophil percentage, C-reactive protein level, and erythrocyte sedimentation rate. All parameters studied were within the normal range in almost all cases.
The symptoms and clinical hematological parameters of the seronegative patients and patients with prior immunity were also determined (n = 34), when available. In these patients, arthritis (80%) and rash (72%) were somewhat less common than in seropositive patients. The incidence of fatigue, fever, muscle pain, headache, nausea, and dizziness were approximately the same in both groups. The clinical hematological parameters were usually within the normal range, as in seropositive patients.
DISCUSSION
This study describes in detail the clinical picture of acute and prolonged SINV infection, serological and hematological findings, and results of viral detection attempts in 86 patients. The hallmarks of the typical clinical manifestation of Pogosta disease are joint symptoms and rash. In addition, fatigue, muscle pain, headache, and mild fever were seen in approximately one-half of the patients in our study. The arthritic symptoms most commonly affected the ankle, wrist, finger, and knee joints. These findings are in agreement with those of some previous, smaller studies [11, 12].
Questionnaires were used for data collection to characterize typical symptoms of the disease. This methodology is susceptible to errors of recall or other mistakes, and there are, therefore, limitations in the interpretation of the results. Despite the confounding issues, the questionnaires were found to contain more accurate and comparable information than did the review of the usually brief medical records (also available from almost all patients). The patients filled out the questionnaires during the acute phase, in most cases immediately after seeing their physicians.
The acute symptoms of Pogosta disease sometimes resemble those of human parvovirus B19 infection [23], which is the primary differential diagnostic alternative. Secondary alternatives include varicella, measles, rubella, borreliosis, reactive arthritis, and rheumatic diseases without infectious etiology. Since our seronegative patients and patients with prior immunity had, in most cases, the same symptoms as did the confirmed cases, it is recommendable to confirm the diagnosis with serological assessment. Specific diagnosis reduces unnecessary treatment attempts and provides a prediction of the clinical manifestation of the disease. Our results showed that the clinical hematological and infection parameters in SINV infection are typically within the normal range.
We screened for IgM antibodies in 50 patients, by use of paired serum samples collected during the acute phase and a third sample collected 56 months after infection. The results showed that IgM seroconversion occurs within the first 8 days of illness, and IgG seroconversion occurs within the first 11 days. Thus, a negative result of SINV antibody testing during the first week of illness does not rule out Pogosta disease. However, 16 (38%) of 42 of the patients had IgM antibodies detectable in serum samples collected within the first 6 days of illness.
Of the 50 patients, 18 (36%) were IgM positive 56 months after infection (2 were RF positive); 10 (56%) of these patients had joint symptoms at the time. Of the 32 IgM-negative patients, 14 (44%) had joint symptoms at the time 56 months after infection. Although IgM antibodies have previously been shown to persist even for several years after infection, no correlation between persisting IgM and prolonged joint symptoms has been stated [15]. No such tendency was shown in our study, either, neither were there indications that early total antibody levels would correlate to the duration of arthritis symptoms. The persistence of IgM antibodies could suggest that viral replication takes place somewhere in the body, possibly in synovia. In mouse experiments, it has been shown that replication of SINV can take place in the periosteum or endosteum [24]; whether this is true for humans has remained unclear, despite attempts to detect virus in synovial fluid [25].
Altogether, SINV RNA could be detected in 4 of 23 skin-lesion biopsy specimens and in 5 of 73 whole-blood samples by means of RT-PCR. In 2 of the skin-lesion specimens from which virus isolation was successful, RT-PCR results remained negative. This could be the result of a varying viral load in different sections of the skin, possibly related to the amount of blood in the skin tissue.
Patients with mild symptoms might have been excluded from the study, since they are less likely to seek health care services or their condition might remain unrecognized by physicians. Brummer-Korvenkontio et al. have estimated that, during the 1981 outbreak in Finland, there were 17 times more subclinical and mild (or undiagnosed) SINV infections than serodiagnosed cases [4]. It remains to be studied whether subclinical acute infections could predispose patients to later occurring joint or other manifestations. Laine et al. studied 142 Finnish patients with acute arthritis and showed that 4.2% of the patients had acute SINV infection at the time of the onset of symptoms [26].
Although previous studies [13, 14, 15] have indicated that there is a considerable risk of prolonged joint symptoms after SINV infection, the results of our study show an unexpectedly high prolonged disease burden; 50% of the patients had joint symptoms >12 months after infection. Seventy-seven percent of the 12-month follow-up questionnaires were returned. It is possible that the healthier individuals were not as eager to answer, and the proportion of arthritic patients could be overrepresented. In any case, a clear conclusion from these results is that prolonged arthritis after infection with Finnish strains of SINV is more a typical manifestation than a rarity. Those with joint symptoms 1 year after the acute infection were significantly older than those without joint symptoms, showing that advanced age is a predisposing factor for prolonged manifestations. Females were overrepresentedalthough not significantlyin the group with prolonged joint symptoms. Further investigations are required to define the possible etiologic role of SINV infection in acute and chronic rheumatic diseases in general.
The present study suggests that an etiologic role of SINV infection should be considered for patients with prolonged, undefined joint symptoms. Combining serodiagnostic data with the clinical time frame, as well as with epidemiological and anamnestic data, might link SINV infection with patients' current symptoms. Because of the unexpectedly high incidence of prolonged symptoms in this study, a separate study with matched SINV antibodynegative controls will be conducted in the future.
Predisposing factors that might lead to clinical human disease caused by SINV infection could be related to viral genetics, host factors, or ecological circumstances. Other alphavirusessuch as the Mayaro [27], Chikungunya [28], O'nyong-nyong [29], and, especially, Ross River virus (RRV) [30]are known to cause similar symptoms. RRV has also been reported to cause prolonged joint symptoms in a substantial proportion of patients [31]. SINV is the most commonly isolated arbovirus in Australia, yet reported clinical cases have been rare [31, 32, 33]. It seems unlikely that the genetic susceptibility of Australian and Fennoscandian people would be considerably different. Previous studies have shown that there could be small antigenic differences, between prototype SINV isolates and the Fennoscandian strains, that may have epidemiologic significance [34, 35]. Could the Fennoscandian SINV strains and RRV have crucial similarities in their antigenic properties that affect the manifestation of infection Or do ecological circumstances and vector competence favor outbreaks and spillover to humans only under certain circumstances that exist for SINV in Finland during late summer The latter is, in any case, evident, because even within Finland, similar SINV prevalences are found in tetraonid birds throughout most of the country (authors' unpublished data), whereas human cases cluster in certain regions.
Acknowledgments
We thank the staff at North Karelian health stations involved in material collection; Suvi-Sirkku Kaukoranta (North Karelia Central Hospital) and Camilla Jokinen (Kuopio University Hospital), for organizing support; Leena Kostamovaara (University of Helsinki), Irina Suomalainen (University of Helsinki), and Sirkka Vene (the Swedish Institute for Infectious Disease Control), for assistance with immunohistochemical experiments; and Kati Vapalahti for help with statistics.
References
1. Strauss EG, Rice CM, Strauss JH. Complete nucleotide sequence of the genomic RNA of Sindbis virus. Virology 1984; 133:92110. First citation in article
2. Kurkela S, Manni T, Vaheri A, Vapalahti O. Causative agent of Pogosta disease isolated from blood and skin lesions. Emerg Infect Dis 2004; 10:88994. First citation in article
3. Brummer-Korvenkontio M, Kuusisto P. Onko Suomen Lnsiosa sstynyt "Pogostalta" (Has western Finland been spared the "Pogosta"). Suom Lkril 1981; 32:26067. First citation in article
4. Brummer-Korvenkontio M, Vapalahti O, Kuusisto P, et al. Epidemiology of Sindbis virus infections in Finland 198196: possible factors explaining a peculiar disease pattern. Epidemiol Infect 2002; 129:33545. First citation in article
5. Lundstrm JO, Lindstrm KM, Olsen B, Dufva R, Krakower DS. Prevalence of Sindbis virus neutralizing antibodies among Swedish passerines indicates that thrushes are the main amplifying hosts. J Med Entomol 2001; 38:28997. First citation in article
6. Lundstrm JO, Turell MJ, Niklasson B. Viremia in three orders of birds (Anseriformes, Galliformes and Passeriformes) inoculated with Ockelbo virus. J Wildl Dis 1993; 29:18995. First citation in article
7. Norder H, Lundstrm JO, Kozuch O, Magnius LO. Genetic relatedness of Sindbis virus strains from Europe, Middle East, and Africa. Virology 1996; 222:4405. First citation in article
8. Skogh M, Espmark A. Ockelbo disease: epidemic arthritis-exanthema syndrome in Sweden caused by Sindbis-virus like agent. Lancet 1982; 1:7956. First citation in article
9. Lvov DK, Skvortsova TM, Berezina LK, et al. Isolation of Karelian fever agent from Aedes communis mosquitoes. Lancet 1984; 2:399400. First citation in article
10. Lvov DK, Vladimirtseva EA, Butenko AM, Karabatsos N, Trent DW, Calisher CH. Identity of Karelian fever and Ockelbo viruses determined by serum dilution-plaque reduction neutralization tests and oligonucleotide mapping. Am J Trop Med Hyg 1988; 39:60710. First citation in article
11. Espmark A, Niklasson B. Ockelbo disease in Sweden: epidemiological, clinical, and virological data from the 1982 outbreak. Am J Trop Med Hyg 1984; 33:120311. First citation in article
12. Turunen M, Kuusisto P, Uggeldahl PE, Toivanen A. Pogosta disease: clinical observations during an outbreak in the province of North Karelia, Finland. Br J Rheumatol 1998; 37:117780. First citation in article
13. Laine M, Luukkainen R, Jalava J, Ilonen J, Kuusisto P, Toivanen A. Prolonged arthritis associated with Sindbis-related (Pogosta) virus infection. Rheumatology (Oxford) 2000; 39:12724. First citation in article
14. Niklasson B, Espmark A. Ockelbo disease: arthralgia 3-4 years after infection with a Sindbis virus related agent. Lancet 1986; 1:103940. First citation in article
15. Niklasson B, Espmark A, Lundstrm J. Occurrence of arthralgia and specific IgM antibodies three to four years after Ockelbo disease. J Infect Dis 1988; 157:8325. First citation in article
16. Vene S, Franzen C, Niklasson B. Development of specific antibody patterns and clinical symptoms following Ockelbo virus infection. Arch Virol 1994; 134:6171. First citation in article
17. Calisher CH, Meurman O, Brummer-Korvenkontio M, Halonen PE, Muth DJ. Sensitive enzyme immunoassay for detecting immunoglobulin M antibodies to Sindbis virus and further evidence that Pogosta disease is caused by a western equine encephalitis complex virus. J Clin Microbiol 1985; 22:56671. First citation in article
18. Malherbe H, Strickland-Cholmley M, Jackson AL. Sindbis virus infection in man: report of a case with recovery of virus from skin lesions. S Afr Med J 1963; 37:54752. First citation in article
19. Zhou G, Liang G, Li L. Complete nucleotide sequence of the nonstructural gene of alphavirus YN87448 strain isolated in China and its relationship to other Sindbis viruses. Zhonghua Shi Yan He Lin Chuang Bing Du Xue Za Zhi 1999; 13:31420. First citation in article
20. Hrling J, Vene S, Franzen C, Niklasson B. Detection of Ockelbo virus RNA in skin biopsies by polymerase chain reaction. J Clin Microbiol 1993; 31:20049. First citation in article
21. Niklasson B, Espmark A, LeDuc JW, et al. Association of a Sindbis-like virus with Ockelbo disease in Sweden. Am J Trop Med Hyg 1984; 33:12127. First citation in article
22. Hedman K, Vaheri A, Brummer-Korvenkontio M. Rapid diagnosis of hantavirus disease with an IgG-avidity assay. Lancet 1991; 338:13536. First citation in article
23. Reid DM, Reid TM, Brown T, Rennie JA, Eastmond CJ. Human parvovirus-associated arthritis: a clinical and laboratory description. Lancet 1985; 1:4225. First citation in article
24. Heise MT, Simpson DA, Johnston RE. Sindbis-group alphavirus replication in periosteum and endosteum of long bones in adult mice. J Virol 2000; 74:92949. First citation in article
25. Luukkainen R, Laine M, Nirhamo J. Chronic arthritis after Sindbis-related (Pogosta) virus infection. Scand J Rheumatol 2000; 29:399400. First citation in article
26. Laine M, Vainionp R, Uksila J, et al. Prevalence of Sindbis-related (Pogosta) virus infections in patients with arthritis. Clin Exp Rheumatol 2003; 21:2136. First citation in article
27. Pinheiro FP, Freitas RB, Travassos da Rosa JF, Gabbay YB, Mello WA, LeDuc JW. An outbreak of Mayaro virus disease in Belterra, Brazil. I. Clinical and virological findings. Am J Trop Med Hyg 1981; 30:67481. First citation in article
28. Fourie ED, Morrison JG. Rheumatoid arthritic syndrome after chikungunya fever. S Afr Med J 1979; 56:1302. First citation in article
29. Shore H. O'nyong-nyong fever: an epidemic virus disease in East Africa. III. Some clinical and epidemiological observations in the northern province of Uganda. Trans R Soc Trop Med Hyg 1961; 55:36173. First citation in article
30. Proll S, Dobler G, Pfeffer M, Jelinek T, Nothdurft HD, Loscher T. Persistent arthralgias in Ross-River-Virus disease after travel to the South Pacific. Dtsch Med Wochenschr 1999; 124:75962. First citation in article
31. Mackenzie JS, Lindsay MD, Coelen RJ, Broom AK, Hall RA, Smith DW. Arboviruses causing human disease in the Australasian zoogeographic region. Arch Virol 1994; 136:44767. First citation in article
32. Doherty RL, Bodey AS, Carew JS. Sindbis virus infection in Australia. Med J Aust 1969; 2:10167. First citation in article
33. Guard RW, McAuliffe MJ, Stallman ND, Bramston BA. Haemorrhagic manifestations with Sindbis infection. Case report. Pathology 1982; 14:8990. First citation in article
34. L'vov DK, Skvortsova TM, Kondrashina NG, Vershinskii BV, Lesnikov AL. Etiology of Karelia fevera new arbovirus infection. Vopr Virusol 1982; 27:6902. First citation in article
35. Calisher CH, el-Kafrawi AO, Al-Deen Mahmud MI, et al. Complex-specific immunoglobulin M antibody patterns in humans infected with alphaviruses. J Clin Microbiol 1986; 23:1559. First citation in article