West Nile virus-associated flaccid paralysis

Abstract

The causes and frequency of acute paralysis and respiratory failure with West Nile virus (WNV) infection are incompletely understood. During the summer and fall of 2003, we conducted a prospective, population-based study among residents of a 3-county area in Colorado, United States, with developing WNV-associated paralysis. Thirty-two patients with developing paralysis and acute WNV infection were identified. Causes included a poliomyelitislike syndrome in 27 (84%) patients and a Guillain-Barré-like syndrome in 4 (13%); 1 had brachial plexus involvement alone. The incidence of poliomyelitislike syndrome was 3.7/100,000. Twelve patients (38%), including 1 with Guillain-Barré-like syndrome, had acute respiratory failure that required endotracheal intubation. At 4 months, 3 patients with respiratory failure died, 2 remained intubated, 25 showed various degrees of improvement, and 2 were lost to followup. A poliomyelitislike syndrome likely involving spinal anterior horn cells is the most common mechanism of WNV-associated paralysis and is associated with significant short- and long-term illness and death.

Figure 1

Age distribution of 32 patients with West Nile virus–associated paralysis.

Figure 2

Saggital (A) and axial (B) T2-weighted magnetic resonance images of the cervical spinal cord in a patient with acute asymmetric upper extremity weakness and subjective dyspnea. A shows a diffuse cervical cord signal abnormality, and B shows an abnormal signal in the anterior horn region.

Figure A1

Patterns of weakness at strength nadir and 4 months later in patients with acute paralysis and West Nile virus infection. Shadings indicate strength scoring on manual muscle testing and represent the average strength score of all tested muscles proximally (upper extremity: shoulder adduction/abduction, arm internal/external rotation; lower extremity: hip flexion/extension, thigh adduction/abduction), medially (upper extremity: elbow flexion/extension, pronation/supination; lower extremity: knee flexion/extension), and distally (upper extremity: wrist flexion/extension, finger flexion/extension; lower extremity: ankle plantarflexion/dorsiflexion, foot inversion/eversion, toe flexion/extension) in each limb and the facial muscles. Scores were rounded down to the lowest whole number. Patients with respiratory weakness are indicated by circles (intubated) or triangles (not intubated). Pt, patient.

Figure A2

Sample correlation matrices of the 14 strength score measurements at nadir (a) and follow-up (b) for 21 patients with poliomyelitislike syndrome and excluding those who died or were lost to follow-up. Each cell is shaded to represent the correlation between strength scores at the various sites, indicated along the axes, as described and recorded in Figure A1. Darker shading represents lower correlation and white represents perfect (1.0) correlation. The heavy black lines splitting the figure into quadrants separate right and left, and the lighter-shaded lines separate the quadrants into face, arm, and leg measurements. At nadir, consistent correlation is displayed within limbs across the body, with lower correlation between limbs on the same body side; lowest correlation is observed between face and leg. At follow-up, correlations were observed at nadir (e.g., stronger correlation within limbs across body) persist, but associations were weakened somewhat, suggesting greater heterogeneity of strength scores. Weakening of association between distal and proximal limbs suggests distal to proximal improvement. L, left; R, right.