A 69-year-old man was admitted to an emergency department after repeated falls at home. A week earlier, he attended a shamanism workshop in the Ardèche forest (France). Four days before hospitalization, marked asthenia, diarrhea, diplopia, and dizziness developed in the patient, resulting in falls without loss of consciousness. His family reported a cough for the past year and fatigue and weight loss for several months. The patient had never traveled outside metropolitan (mainland) France.

Figure 1

Figure 1. Electroencephalogram (EEG) and brain magnetic resonance imaging (MRI) testing of a patient with encephalitis caused by pigeon paramyxovirus type 1, France. A) The first EEG (day 5 after hospitalization) revealed...

At admission, the patient was confused and unable to stand or walk and had a fever of 39°C, right-ear hearing loss, right peripheral facial paralysis, left upper limb ataxia, and multidirectional nystagmus. Deep tendon and plantar reflexes were unremarkable. Routine laboratory tests revealed mild inflammation (hyperleukocytosis at 15 G/L [reference range 4–10 G/L]) and C-reactive protein of 25 mg/L (reference range <5 mg/L). Two cerebrospinal fluid (CSF) analyses revealed no pleocytosis, negative viral and bacterial PCR testing, and negative cultures. We noted isolated hyperproteinorachia of 0.72 g/L (reference range 0.15–0.45 g/L). An electroencephalogram revealed nonspecific encephalopathy (Figure 1, panel A). Brain magnetic resonance imaging revealed no hemorrhage or recent ischemia but showed mild white matter hyperintense lesions at the junction of the mesencephalon and pons on T2 or fluid-attenuated inversion recovery.

Five days after admission, the patient was transferred to the intensive care unit for coma, respiratory failure, and aspiration pneumonia. His neurologic condition deteriorated, and dysarthria, diplopia, deafness, ataxia, swallowing disorders, and quadriparesis developed. He was promptly intubated and mechanically ventilated. Repeat electroencephalogram (Figure 1, panel B) and brain magnetic resonance imaging (Figure 1, panels C–E) confirmed severe encephalitis. A third CSF analysis confirmed hyperproteinorachia of 1.5 g/L and no pleocytosis. Routine laboratory test results were unremarkable, including investigations for rare etiologies of encephalitis (Appendix). Electroneuromyography revealed severe motor and sensory axonal neuropathy of all 4 limbs. A full-body computed tomography scan revealed enlarged centimetric mediastinal and hilar lymphadenopathies and widespread colitis from the sigmoid to the cecum. Despite supportive care and intensive medical management (Appendix), the patient died 26 days after admission from progressively worsening, life-threatening encephalitis and polyradiculoneuropathy of unknown etiology.

To investigate the origin of the encephalitis, we performed metagenomic next-generation sequencing on postmortem midbrain and cervical spinal cord tissues obtained at autopsy, generating 82.5 million midbrain reads and 67.3 million cervical spinal cord reads. We assembled the complete PPMV-1/Human/France/2023 genome from 3,245 reads (0.48%, sequencing depth 28×) from the midbrain and 3,790 reads (0.68%, sequencing depth 34×) from the cervical spinal cord samples. We did not detect any additional pathogens.

Figure 2

Figure 2. Phylogenetic analysis of fusion gene sequences of PPMV-1from a human case of fatal PPMV-1encephalitis in France and representative APMV-1 strains. A) We conducted phylogenetic analysis on the basis of the...

Phylogenetic analyses revealed that PPMV-1/Human/France/2023 belongs to APMV-1 class II, genotype VI, sub-genotype 2.1.1.2.2. The predicted fusion protein contained the polybasic cleavage motif ¹¹²RRQKRF¹¹⁷ associated with virulence in birds (10,12). PPMV-1/Human/France/2023 was related to PPMV-1 strains from China, Egypt, Belgium, and Ukraine but also APMV-1/pigeon/France/172784/2017 from France (Haute-Garonne department). The sequence from France was obtained through a surveillance program in wild birds: the original sample was collected from a wild pigeon in November 2017, ≈250 km away from Ardèche (Appendix Table 1). PPMV-1/Human/France/2023 is more closely related to Pi/SH/CH/041002/2011 (GenBank accession no. PP297102.1), which was discovered in pigeons from China in 2011 (Figure 2; Appendix Table 2).

Figure 3

Figure 3. PPMV-1 dissemination in the central and peripheral nervous systems and in axillary lymph nodes in a patient with fatal PPMV-1 encephalitis in France. A) Schematic representation of PPMV-1 viral load...

To investigate the dissemination of PPMV-1 in the patient, we conducted semiquantitative real-time PCR on all clinical samples (Appendix Table 3). PPMV-1 was detected in the CNS, thoracic spinal ganglia, and axillary lymph nodes. We identified the highest viral burdens in the brain, specifically localized to the forebrain. Posthoc tests revealed an increasing trend in PPMV-1 cycle threshold values from the brain to the peripheral samples (Figure 3).

Figure 4

Figure 4. Histologic examination of the nervous tissue in fatal human encephalitis case caused by pigeon paramyxovirus type 1, France. A) Brown discoloration of the anterior horns (gray arrows) of the spinal...

Postmortem examination of the CNS revealed brain edema, friable midbrain tissue, and gray matter discoloration of the spinal cord (Appendix). Histopathologic testing (Appendix Table 4) revealed a diffuse inflammatory infiltrate of activated macrophages, microglia, and CD3+ and CD8+ T lymphocytes, with severe neuronal loss in the spinal cord, brainstem, and cerebellum. Neuronophagia and microglial nodules were frequent. We did not observe cytopathic effect. The choroid plexus and ependymal cells appeared unremarkable. We found perivascular inflammatory infiltrates in peripheral nerves (Figure 4). Newcastle disease virus immunostaining was positive in the spinal cord, midbrain, cerebellum, hypothalamus, and lymph nodes (Figure 4).

We report a case of fatal encephalitis associated with PPMV-1 in France that belonged to virulent class II, genotype VI, subgenotype 2.1.1.2.2, with polybasic F cleavage motif RRQKRF, in a patient with no known immunosuppression. That motif is associated with high intracerebral pathogenicity in birds (13) and was reported in another fatal encephalitis case (10). PPMV-1 belongs to the velogenic pathotype of AOAV-1 (Appendix).

The patient had never traveled outside France, indicating autochthonous infection. Although the source of infection is undetermined, the temporal association with a shamanism workshop in the Ardèche forest suggests environmental exposure to avian pathogens, potentially through contact with avian feces. PPMV-1 is shed orally and cloacally, remains stable in pigeon feces, and can spread by windborne dust, extending risk beyond localized environments (8).

Two previously reported human APMV-1 encephalitis cases occurred in severely immunosuppressed patients, suggesting that immune status might modulate APMV-1 neuropathogenicity (10,11). However, the patient we report had no known underlying disease or immunosuppressive treatment. His family reported a deterioration in his general condition, including severe weight loss, in the year before death, without medical evaluation. No laboratory markers of immunosuppression were found. We quantified blood torque teno virus loads to further assess immunosuppression, but results were inconclusive regarding immune status at the time of PPMV-1 infection (14). Detecting unusual pathogens in seemingly immunocompetent patients should prompt immune status investigation.

Real-time PCR of postmortem and peripheral samples exhibited strong neurotropism of PPMV-1 without meningitis or systemic infection at the time of neurologic manifestations. PPMV-1 was not detected in the CSF by PCR or metagenomics. PCR of bronchoalveolar lavage was negative 10 days after symptom onset, suggesting weak and transient respiratory replication despite the respiratory tract being the usual entry route. Negative blood PCR results on days 16 and 24 might reflect a brief early viremia.

The patient’s clinical manifestations included severe polyradiculoneuropathy with inflammatory infiltrates in the spinal cord and peripheral nerves, which were not described previously (10,11). The patient also exhibited enlarged lymphadenopathies, and PPMV-1 was detected at low burden in the right and left axillary lymph nodes; immunostaining confirmed viral tropism in those nodes. Lymph node aspiration could provide a minimally invasive alternative to brain biopsy for diagnosing PPMV-1 encephalitis.

In summary, in the case we report and 2 other reported encephalitis cases, class II APMV-1 with a virulence-associated F-protein cleavage site was identified. APMV-1 infection might be underdiagnosed and should be considered in neurologic disease of unknown etiology, particularly among patients who report exposure to Columbidae fauna.