710

CONCISE COMMUNICATION

Continued Transmission of West Nile Virus to Humans in SoutheasternRomania, 1997–1998

Costin Cernescu,1 Nicolae-Ion Nedelcu,2 GratielaTardei,1 Simona Ruta,1 and Theodore F. Tsai3

1Institute of Virology, Romanian Academy, and 2Ministry of Health,Bucharest, Romania; 3Division of Vectorborne Infectious Diseases,

Centers for Disease Control and Prevention, Fort Collins, Colorado

After an epidemic of West Nile (WN) virus neurologic infections in southeastern Romaniain 1996, human and animal surveillance were established to monitor continued transmissionof the virus. During 1997 and 1998, neurologic infections were diagnosed serologically as WNencephalitis in 12 of 322 patients in 19 southeastern districts and in 1 of 75 Bucharest patients.In addition, amid a countrywide epidemic of measles, the etiology of the febrile exanthem in2 of 180 investigated cases was determined serologically to be WN fever; 1 case was complicatedby hepatitis. Sentinel chickens placed in Bucharest seroconverted to WN virus during thesummer months, indicating their potential value in monitoring transmission. The continuedoccurrence of sporadic WN infections in southeastern Romania in consecutive years after the1996 epidemic is consistent with local enzootic transmission of the virus.

Mosquitoborne West Nile (WN) virus usually is an acuteself-limiting febrile illness with exanthem and myalgia. How-ever, the illness may be complicated by central nervous systeminfection, especially in the elderly, and rarely by hepatitis orother organ involvement. An unprecedented epidemic of WNencephalitis occurred in southeastern Romania in 1996, pro-ducing 393 cases—principally in Bucharest and districts bor-dering the lower Danube River [1]. Although the outbreak wassuspected to have been an extension of sylvatic transmissionfrom the delta, previous field investigations had not character-ized a local viral transmission cycle, and the virus had neverbeen isolated in the country until the 1996 outbreak. The resultsof surveillance maintained in 1997 and 1998 to monitor con-tinued transmission of the disease are reported here.

Methods

In 1997, Romanian public health epidemiologists in 19 districtswere instructed to collect serum and cerebrospinal fluid (CSF) spec-imens from patients presenting with aseptic meningitis or enceph-alitis between April and November for serologic diagnosis of WNvirus infection. In addition, 160 sentinel chickens in Bucharest werebled biweekly between 26 June and 13 August and again on 16October 1997 to monitor enzootic viral transmission. Virus-specificIgG was detected by indirect ELISA [2]. In 1998, specimens frommeningoencephalitis patients hospitalized between March and Sep-

Received 3 March 1999; revised 26 October 1999; electronically published8 February 2000.

Reprints or correspondence (present address): Dr. Theodore F. Tsai, Led-erle Vaccines, 401 N. Middletown Rd., Pearl River, NY 10965 (Tsait@war.wyeth.com).

The Journal of Infectious Diseases 2000;181:710–2q 2000 by the Infectious Diseases Society of America. All rights reserved.0022-1899/2000/18102-0040$02.00

tember in Bucharest only were referred for serologic testing. At theend of the summer, acute samples from 180 patients hospitalizedwith acute measles were tested retrospectively. Human specimenswere tested at the Institute of Virology by IgM capture and indirectIgG ELISAs [1].

Results

In 1997, specimens from 322 patients, 119 from Bucharest,were evaluated for WN virus antibodies. Twelve patients withclinical meningoencephalitis and elevated IgM absorbance ra-tios in serum or CSF samples were diagnosed with cases ofacute WN encephalitis. In the only fatal case, the serologicdiagnosis was based on tests of acute serum and CSF samplesonly, but in the others, IgG conversions between acute- andconvalescent-phase serum samples also were demonstrated. Thedistrict residences of WN encephalitis patients and those inwhom the diagnosis was excluded are shown in figure 1. Lab-oratory-confirmed WN neurologic infections occurred between13 July and 25 September with onset of 2 cases in July, 6 inAugust, and 4 in September. Case-patients were 13–76 yearsold (median, 45).

Seroconversions in sentinel chickens in Bucharest were dem-onstrated in each of four consecutive biweekly bleedings from26 June to 13 August with 23%, 26%, 40%, and 16% of birdsconverting at each interval. No seroconversions were notedagain until the final bleeding on 16 October, when 13% of thesentinels seroconverted.

In 1998, clinical samples from 75 patients in Bucharest hos-pitalized with encephalitis (25), aseptic meningitis (36), and me-ningoencephalitis (14) were submitted for serologic testing. Onepatient was diagnosed as having acute WN encephalitis.

From May 1997 to September 1998, the country experienced

by guest on October 21, 2012

http://jid.oxfordjournals.org/D

ownloaded from

JID 2000;181 (February) West Nile Virus Transmission in Romania 711

Figure 1. Serologically confirmed cases of West Nile virus and other neurologic infections by district of residence and etiology, Romania, 1997.

its greatest measles epidemic in the postvaccine era: 120,000cases were reported, principally among high school childrenand young adults (14–25 years old). Serum samples from 180hospitalized patients were tested for WN IgG and IgM anti-bodies in parallel with confirmatory tests for measles. Two pa-tients with a clinical diagnosis of measles without serologicevidence of measles virus infection had elevations of WN virus-specific IgM, indicating recent infection with that virus. Thefirst case patient, a 24-year-old woman whose onset of illnesswas on 24 May, made an uneventful recovery. The second pa-tient, a 19-year-old woman, developed an acute febrile illnesswith exanthem on 2 June and was initially diagnosed with acase of acute measles, but her illness was complicated by hep-atitis and sustained elevated hepatic transaminases for 2 weeksbefore clinical recovery.

Discussion

The sporadic cases of WN virus infections in Bucharest andthe lower Danube delta in 1997 and in Bucharest in 1998 arepersuasive evidence of ongoing virus transmission in south-

eastern Romania. Compared with the 1996 epidemic, when 352(80%) of 441 hospitalized encephalitis patients with adequateclinical specimens were diagnosed with cases of WN neurologicinfection, the 12 patients with laboratory-confirmed cases in1997 comprised only 4% of the patients referred for evaluation,and in 1998 only 1% [1]. The epidemiologic and clinical criteriafor specimen referral were similar in all years, underscoring theunusual pattern of viral transmission in 1996. Systematicallycollected data from other years are unavailable; however, theexperience from the aftermath of the epidemic, when clinicalsuspicion of the diagnosis might still have been high, suggeststhat WN encephalitis occurs sporadically and with a relativelylow frequency in Bucharest and in the southeastern quadrantof Romania. Sentinel chickens were posted in Bucharest ratherlate, but by the first bleeding on 26 June, nearly one quarterof the sentinels had seroconverted, preceding the onset of thefirst case in a human in the city by 6 weeks; this indicates thepotential value of sentinel bird surveillance for forecasting fu-ture epidemics.

During the course of the 1996 investigation, hospital-basedcase finding focused on identifying neurologic infections and

by guest on October 21, 2012

http://jid.oxfordjournals.org/D

ownloaded from

712 Cernescu et al. JID 2000;181 (February)

attempts to measure the full extent of the epidemic by identi-fying patients with milder illnesses were unsuccessful becauseof the relatively low infection rate and the nonspecific symp-toms associated with uncomplicated WN fever. In previous re-ports, WN fever has been characterized as an undifferentiatedself-limited febrile illness, often with exanthem, lymphadeno-pathy, and polyarthralgias [3]. But illness complicated by fatalhepatitis, pancreatitis, myocarditis, and neurologic infectionhave been described [4–7]. The 2 patients hospitalized with aclinical diagnosis of measles appear to have had cases of WNfever with exanthem, complicated in the second patient by acutehepatitis. The frequency of the latter complication is uncertain,although in the 1998 Georgia outbreak, in which 624 febrileexanthem case patients were investigated, 3 had fatal hepatitis[7]. Hepatitis complicating measles appears to be even rarer. Arecent review found only 27 reported cases with hepatobiliaryinvolvement [8]. On the basis of the negative serologic findingsof acute measles infection and the presence of WN virus-specificIgM antibodies, it seems likely that this was a case of WNfever-associated hepatitis.

Sporadic cases and clusters of encephalitis cases occurringduring the summer and attributed to WN virus were previouslyreported in areas of Romania, but some cases had epidemio-logic features more consistent with tick-borne encephalitis, andWN virus was not isolated, so the diagnosis remains unconfir-med [9, 10]. Although reports and serosurveys reporting WNHI antibodies in birds, mammals, and humans in the Danubedelta strongly suggested that the virus was transmitted locally,a mechanism of enzootic transmission was never defined, andperiodic reintroductions of the virus from Africa by migratorybirds also seemed possible. Perennial transmission of the virusin the Danube delta is plausible in view of the plentiful pop-ulation of birds in that expanse of wetlands, and the continuedhuman infections in 2 consecutive years after the 1996 outbreakfurther supports that likelihood. The chance downing of south-ward migrating storks (Ciconia ciconia) as they transited Israelin August and September 1998 and the isolation of WN virusfrom moribund and dead animals provide additional evidenceof viral transmission in Europe, at least in areas where thisspecies summers or along its southern flyway [11]. This obser-vation underscores the possibility of intercontinental virustransfer between Europe and Africa in both directions.

WN virus has been isolated sporadically since 1962 from abroad area of southern and central Europe, including Ukraine,Belarus, and western Russia, and outbreaks affecting humansand horses have been documented in France, Ukraine, and the

Volga delta of Russia [12]. In addition, there is serologic evi-dence of human and animal infection elsewhere on the conti-nent. In this decade, WN outbreaks have been reported in Tus-cany, Italy (14 cases in horses), in Georgia (624 cases inhumans), in the Czech Republic (5 cases in humans), and inAlgeria and Morroco (16 cases in humans); in addition to theepidemic and postepidemic cases in Romania, these outbreaksindicate increased transmission or recognition of the disease insouthern Europe and neighboring areas [7, 12–15].

References

1. Tsai TF, Popovici F, Cernescu C, Campbell GC, Nedelcu NI, investigativeteam. West Nile encephalitis epidemic in southeastern Romania. Lancet1998;352:767–71.

2. Olson JG, Scott TW, Lorenz LH, Hubbard JL. Enzyme immunoassay fordetection of antibodies against eastern equine encephalomyelitis virus insentinel chickens. J Clin Microbiol 1991;29:1457–61.

3. Marberg K, Goldblum N, Stork V, Jasinska-Klingberg W, Klingberg M. Thenatural history of West Nile fever. I. Clinical observations during an ep-idemic in Israel. Am J Hyg 1956;69:259–69.

4. George AJ, Lesbordes JL, Georges-Courbot MC, Meunier DMY, GonzalezJP. Fatal hepatitis from West Nile virus. Ann Inst Pasteur 1987;138:237–44.

5. Perelman A, Stern J. Acute pancreatitis in West Nile fever. Am J Trop MedHyg 1974;23:1150–2.

6. Albagli C, Chaimoff R. A case of West Nile virus myocarditis. Harefuah1959;57:274.

7. Tsereteli DG, Tsiklauri RA, Ivanidze EA. West Nile fever in West Georgia[abstract 60.006]. Program and abstracts: 8th International Congress onInfectious Diseases, Boston, 1998:206.

8. Khatib R, Siddique M, Abbass M. Measles associated hepatobiliary disease:an overview. Infection 1993;21:112–4.

9. Draganescu N, Garjabu E, Iftimovici R, Klebleev E. Etudes serologiques surla presence de quelques arbovirus dans la region de sylvo-steppe de Tulcea.Rev Roum Virol 1991;42:47.

10. Draganescu N, Girjabu E, Iftimovici R. Investigations serologiques chezl’homme et chez differentes especes d’animaux domestiques de la regionde sylvo-steppe de Roumanie, pour le depistage de quelques Togaviridaeet Bunyaviridae. Rev Roum Virol 1993;44:207–10.

11. Malkinson M, Banet C, Weisman J, Pokamonski S, King R. West Nile fever;recent evidence for intercontinental dispersion [abstract VW28,03]. Pro-gram and abstracts: XI International Congress of Virology, Sydney, 1999:56.

12. Hubalek Z, Halouzka J. West Nile fever, a reemerging mosquito-borne viraldisease in Europe. Emerg Infect Dis 1999;5:643–50.

13. Le Guenno B, Bougermouh A, Azzam T, Bouakaz R. West Nile: a deadlyvirus. Lancet 1996;349:1315.

14. El Harrack M, LeGuenno B, Gouno P. Isolement du virus West Nile auMaroc. Virologie 1997;1:248–9.

15. Mitchell CJ. Geographic spread of Aedes albopictus and potential for in-volvement in arbovirus cycles in the Mediterranean basin. J Vector Ecol1995;20:44–58.

by guest on October 21, 2012

http://jid.oxfordjournals.org/D

ownloaded from