Phlebovirus Meningoencephalis Complicated by Pseudomonas aeruginosa Pneumonia: A Case Report

Introduction

T he genus Phlebovirus (family Bunyaviridae) consists currently of 53 antigenically distinct serotypes, including the sandfly fever viruses, mostly transmitted by phlebotomine sandflies (Nichol et al. 2005). There are three phleboviruses circulating around the Mediterranean region: Toscana virus (TOSV), Sandfly fever Naples virus (SFNV), and Sandfly fever Sicilian virus (SFSV). SFSV and SFNV cause a transient febrile illness, whereas TOSV often causes aseptic meningitis and meningoencephalitis (Charrel et al. 2005). In central Italy, where TOSV was first isolated in 1971, it is considered responsible for >80% of aseptic meningitis cases occurring during summer, as the disease coincides with the seasonal activity of its vectors (Valassina et al. 2000). Laboratory diagnosis of clinically suspected phlebovirus infections is set by serology, by virus isolation, or by RNA detection, although the last two methods are successful only if applied on samples taken during the first days of the disease, due to the short-term viremia.

In 1997, Dobler et al. reported a human case of acute TOSV infection in a 73-year-old German woman who had returned from Lavrion, a city 60 km south of Athens, Greece (Dobler et al. 1997). In 2002 a major outbreak of febrile syndrome among the Greek Army forces in Cyprus was caused by a Sicilian-like phlebovirus (Cyprus strain) (Papa et al. 2006). Recent studies revealed high seroprevelance among the apparently healthy Greek population; the highest prevalence is observed in Ionian islands and the western mainland of Greece (60% and 35%, respectively) (Charrel et al. 2005). However, phlebovirus infections are very seldom included in differential diagnosis of central nervous system (CNS) infections in Greece.

Case Report

In June 2004 an 8-year-old boy was admitted to the Chidren's Intensive Care Unit of Hippokration Hospital of Thessaloniki because of 3-day fever (39.1°C), mild tachypnea, hypotonia, diarrhea, and tonoclonic convulsions in body and extremities. Twenty-four hours later the patient was intubated because of deterioration of his neurological condition. Laboratory tests on the first day of admission showed hematocrit 34.2%, hemoglobin 11.7 g/dL, and white blood cells count of 3000/μL, with 50.7% neutrophils and 42.7% lymphocytes, and platelet count of 212,000/μL. Analysis of cerebrospinal fluid (CSF) taken on admission revealed 0–1 cells/mm³, with normal glucose (64 mg/dL) and protein (26 mg/dL) levels. No brain abnormalities have been evidenced by computed tomography or magnetic resonance imaging.

Blood and CSF cultures for common bacteria were negative. Three serum samples collected on 4th, 17th, and 19th day after admission were tested negative for IgM and IgG antibodies to tick-borne encephalitis virus, West Nile virus, herpes simplex virus, Parvo B19 virus, Ricketsia typhi, and Ricketsia conorii. Negative results were also taken in polymerase chain reactions (PCRs) for herpes viruses, enteroviruses, coronaviruses, flaviviruses, and orthobunyaviruses applied on the serum sample taken the fourth day of the disease. CSF sample taken upon admission was not available.

The three serum samples were tested for the presence of IgG and IgM antibodies to TOSV using enzyme-linked immunosorbent assay (ELISA; Diesse, Siena, Italy). The first sample was IgG negative, whereas the second and third samples were positive (index 1.6, positive when >1.1). No IgM antibodies were detected. An RT-nested PCR with generic primers amplifying partial L gene of phleboviruses was applied on the first serum sample and found negative (Sanchez-Seco et al. 2003).

The patient was treated for 17 days with acyclovir and ceftriaxone. His condition was complicated with hospital-acquired pneumonia caused by Pseudomonas aeruginosa, for which he received colistin. He was discharged from the hospital after 2 months with a slight ataxia.

Discussion

Laboratory diagnosis of acute CNS infections is achieved only in limited number of cases, even when molecular methods for various pathogens are promptly applied (Frantzidou et al. 2008). This is due to a great variety of pathogens causing CNS infections, while many factors (quality and quantity of samples, time of sampling, duration of viremia, etc.) play a role in the success of detection of the causative agent. Many febrile cases were being observed in the summer time in Cyprus; however, the virus was detected only when samples were tested for many pathogens, including phleboviruses, whereas isolation was achieved only when first-day samples were used for isolation procedures (Papa et al. 2006, Konstantinou et al. 2007).

Cross reactions are often observed among phleboviruses, especially between TOSV and other serotypes of SFNV (Tesh et al. 1982, Schwarz et al. 1996, Charrel et al. 2005). In the present case the IgG seroconvertion was suggestive of a phlebovirus infection. This fact, together with the failure to detect TOSV IgM antibodies in ELISA (in which a TOSV recombinant nucleoprotein antigen is used), suggests that the causative agent was a phlebovirus distinct from TOSV. Similar results were observed in samples taken from patients during the Cyprus outbreak, where all samples were IgM negative and IgG positive (using the same ELISA). It was explained when sequencing and phylogenetic analysis showed that Cyprus virus is genetically related to SFSV rather than to SFNV and TOSV.

Records of our laboratory studied retrospectively gave evidence that 2004 was an epidemic year; it was found that in summer of 2004, 14 of 27 (51.9%) patients who were hospitalized with acute CNS infection had high titers of TOSV IgG antibodies, percentage that differs significantly from that observed in 2003 (4/17, 23.5%) and 2005 (4/20, 20%) (p < 0.05). Although a past phlebovirus infection cannot be excluded, the high IgG titers together with the clinical symptoms and the seasonality give evidence of acute phlebovirus infection. For none of these cases a very acute sample was available; as expected, RT-nested-PCR applied on samples taken 1 week after the onset of illness was negative.

The current study is the first report of human phlebovirus infection in Greece. Most probably, cases occur every year in Greece, and physicians have to include them in differential diagnosis of febrile cases observed in summer, especially when neurological symptoms are also present. Further studies on human samples taken the first days of the disease, as well as on sandflies, will enable the genetic detection and characterization of the phlebovirus strain(s) circulating in Greece.