Fatal Coxsackievirus A-16 Pneumonitis in Adult

Coxsackievirus A-16 (CVA-16) is the agent of hand, foot, and mouth disease in children. We report a case of fatal pneumonitis in an adult due to a CVA-16 strain with a low (78.6%) rate of sequence homology with the reference strain. A modified, more virulent, strain of CVA-16 could be emerging.


The Case
In April 2006, a 76-year-old man was admitted to the emergency department of Pontchaillou University Hospital, Rennes, France, with acute onset of fever, lumbar pain, and dyspnea. Examination found a temperature of 37.9°C and bilateral pulmonary crackles. Laboratory results were the following: leukocytes 9,600 cells ×10 6 /L (90.6% neutrophils), C-reactive protein 216 mg/L (normal value <5 mg/ L), and arterial oxygen partial pressure 67 mm Hg (room air). Chest radiograph was unremarkable, and Legionella pneumophila urinary antigen was not found. The patient was treated with amoxicillin-clavulanate and ofl oxacin. On day 3, acute respiratory distress syndrome developed, and the patient required mechanical ventilation. Computed tomographic scan of the thorax showed bilateral alveolointerstitial infi ltrates (Figure). Transthoracic echocardiograph and pulmonary artery catheterization showed normal left ventricular function. Serum troponin levels were within normal limits.
A bronchoalveolar lavage (BAL) was performed; no pyogenic bacteria, L. pneumophila, Mycobacterium tuberculosis, Pneumocystis jiroveci, or Aspergillus sp. were isolated despite appropriate staining for direct examination and cultures on appropriate media. Negative results were obtained in M. tuberculosis PCR and immunofl uorescence assays (IFAs) for P. jiroveci and L. pneumophila. The result of a PCR for human herpesviruses, which used herpes consensus identifi cation, was negative. Results of testing for respiratory syncytial virus, infl uenza virus, parainfl uenza virus, and adenovirus by IFA and ELISA, which used specifi c monoclonal antibody, were all negative. In addition, serologic test results for L. pneumophila, Mycoplasma pneumoniae, Chlamydophila pneumoniae, C. psittaci, Coxiella burnetti, and HIV were negative. Three sets of blood cultures taken at the time of hospital admission remained sterile. Test results for antinuclear factors, antiglomerular basement-membrane antibodies, and antineutrophil cytoplasmic antibodies were negative.
BAL was positive in 2 enteroviral PCR assays that used EV1 primer from Rotbart et al. (3) and real-time PCR with primers and probe adapted from Verstrepen et al. (4). In addition, on day 3, MRC-5 cell culture showed a specifi c cytopathic effect, which was confi rmed as enterovirus by indirect IFA that used enterovirus mouse monoclonal antibody (Novocastra, Newcastle, UK) and fl uorescein isothiocyanate-conjugated Affi niPure Goat Anti-Mouse (Jackson Immuno-Research, West Grove, PA, USA). Enterovirus was also detected in serum and pharyngeal samples by real-time PCR. Results of real-time PCR and cell culture (MRC-5 and LLC-MK2) were also positive for enterovirus on BAL performed on days 8 and 14. Kinetics analysis in real-time PCR showed a 100-fold decrease in viral load by comparison of cycle thresholds between day 8 and day 14 BAL. Serologic testing for enterovirus showed an 8-fold increase in enterovirus antibody titration by complement fi xation between days 3 and 11. Immunoglobulin M to echovirus/coxsackievirus was detected by ELISA in serum on day 11 (Genzyme, Virotech, Chilly Mazarin, France).
No drug is currently approved for the treatment of enterovirus infection. Pleconaril may be of value in severe enteroviral infections (5) but is no longer available because licensure was not pursued. The patient did not improve and died on day 28 of intractable hypoxemia. Histologic examination of postmortem pulmonary biopsy specimen showed diffuse alveolar damage and fi brosis, real-time PCR detected enterovirus, and viral cultures were negative. The enterovirus isolated on the day-3 BAL was identifi ed as CVA-16 by partial sequencing of the VP1 region that encompasses the BC loop. This region was amplifi ed with primers 292 as sense primer and 222 as antisense primer 1 Current affi liation: Centre Hospitalier Saint-Brieuc, Saint-Brieuc, France *Centre Hospitalier Universitaire Pontchaillou, Rennes, France; and †Centre Hospitalier Universitaire Edouard Herriot, Lyon, France (6). The sequences obtained (reference no. bankit 845096-DQ993173 until defi nitive number assigned) were aligned with the corresponding region in GenBank. Comparison of a 338-nt sequence with that of the CVA-16 reference strain (prototype BrCr) and the EV 71 reference strain (prototype G-10) showed nucleotide identity rates of 78.6% and 64.6%, respectively. On subsequent questioning, the patient's wife reported that he had had close contact with his granddaughter, who had active HFMD, while in the United Kingdom during the week before he was admitted. Although a large HFMD outbreak occurred in the United Kingdom in 1994 (7), no enterovirus outbreaks had been reported at the time of the patient's admission. During the 1994 UK outbreak, CVA was isolated in 28 of 40 patients (type CVA-16 in 21 patients and CVA-10 in 7 patients), but secondary cases in family members were rare and no case of pneumonitis or death was reported.

Conclusions
Fatal CVA-16 infection has been described infrequently in children who had HFMD associated with myocarditis (8). We report a fatal CVA-16 infection associated with pneumonitis in an adult; to our knowledge, this is the fi rst such report. Our patient had neither myocarditis nor left ventricular dysfunction, as demonstrated by pulmonary artery catheterization results, echocardiograph results, and serum troponin levels. In 2003, 7 fatal cases of HFMD in children were reported in Singapore (9). These children had interstitial pneumonitis, either alone or associated with myocarditis or encephalitis. EV 71 was isolated in 4 cases and echovirus 16 in 1. The CVA-16 strain isolated from our patient had a low percentage of nucleotide identity with the reference strain (78.6%); a threshold of 90% is usually required to defi ne strain homology. This may be a sign that this virus is evolving. A strain similar to that from our patient was circulating in China from 1999 through 2004 (98% nucleotide identity; GenBank accession no. AY821798) and was isolated from fecal samples of children with HFMD or suspected enterovirus infection (10). This strain was associated with local yearly outbreaks in which only a few cases of neurologic disease and no deaths were reported. According to phylogenetic analysis based on VP4 207-bp nucleotide sequence, the authors concluded that 3 genetic lineages were circulating in Asia at that time and suggested that the same tendency may apply in other continents (10).
We report what we believe to be the fi rst case of CVA-16 pneumonitis in an adult, with fatal outcome. Preliminary sequence analysis revealed a low rate of homology between the CVA-16 strain we isolated and those previously published, which suggests that a new, more virulent, strain of CVA-16 could be emerging. To compare the sequences to those published in GenBank, we are sequencing the complete part of the genome encoding the VP1 capsid protein.