Article Text

Download PDFPDF

The aetiology of flaccid paralysis in West Nile virus infection
  1. A A Leis1,
  2. J A Van Gerpen2,
  3. J J Sejvar3
  1. 1Methodist Rehabilitation Center, Jackson, MS, USA
  2. 2Ochsner Clinic, New Orleans, LA, USA
  3. 3Centers for Disease Control and Prevention, Atlanta, GA, USA
  1. Correspondence to:
 J J Sejvar;
  1. M Park4,
  2. R E Bartt4,
  3. J S Hui4
  1. 4Rush University Medical Center, Chicago, ILL, USA

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    We read with interest the recent article by Park et al,1 describing a syndrome of acute anterior radiculitis associated with West Nile virus (WNV) infection. Although admittedly there is still much to learn about the clinical spectrum of disease associated with WNV, we were troubled by several of the assertions raised by the article, and the conclusions drawn.

    Recent evidence has suggested that the majority of patients developing acute asymmetrical weakness in the setting of WNV infection suffer from damage to spinal anterior horn cells, resulting in a poliomyelitis-like syndrome.2–5 This has been supported by electrodiagnostic data2–5 and by pathology demonstrating the destruction of anterior spinal grey matter.6–9 Park et al assert that “the mechanism of weakness associated with WNV infection continues to be unclear”, and they subsequently “propose an alternate explanation for the associated weakness”. This alternative explanation of “acute anterior radiculitis” is based on MRI findings that showed intradural lumbosacral nerve root enhancement in a patient with unilateral leg weakness.

    However, the authors do not describe MRI findings in the anterior spinal cord, and the MRI images provided are at the L1–L2 and L2–L3 levels, which lie caudal to the cord segments giving rise to lumbar roots; MRI of the thoracic spine is needed to adequately visualise lumbar cord segments. This is an important omission, since it is unclear whether the authors visualised the anterior lumbar cord before proposing an alternative explanation for WNV associated weakness. In addition, the authors’ contention that their case displayed ventral nerve root involvement can be challenged, since it is difficult to distinguish intradural ventral nerve root enhancement from posterior root enhancement. Furthermore, nerve root enhancement in and of itself is a relatively non-specific finding which may be seen with meningeal inflammation in general, and may not have a clinical correlate. Accordingly, the authors’ argument that the MRI findings call into question the pathophysiology of weakness, and provide evidence for an anterior radiculopathy, rather than a poliomyelitis, as the aetiology of weakness, is speculative.

    It is true that definitive MRI changes have thus far been absent in many cases of WNV poliomyelitis, although clear documentation of such changes has been reported.9 However, neuroimaging data in persons with poliomyelitis due to wild-type poliovirus or other neurotropic viruses has not been consistently gathered, and the incidence of such MRI changes is unknown. It is likely that specific MRI changes may be missed because of variations in the stage of disease at the time of imaging. Finally, the transient reversible muscle weakness seen in this case differs clinically from that in individuals with WNV poliomyelitis-like syndrome, who develop chronic profound weakness.10

    It is clear that WNV infection may be associated with a myriad of clinical and pathophysiological features, and transient weakness due to anterior radiculitis certainly may be among those features. However, generalisation based on one rather atypical case, and invoked to propose an alternative explanation for the flaccid paralysis seen in WNV infection, is problematic. It is our opinion that poliomyelitis clearly has been established as the aetiology of most cases of acute, asymmetrical paralysis seen in WNV infection.

    The article by Park et al1 raises an important issue, however, about the utility of MRI in the diagnosis of WNV; such questions about MRI findings both in WNV poliomyelitis-like syndrome and in WNV encephalitis may be addressed by serial MRI assessments of patients during and following acute illness.


    Authors’ reply

    Our understanding of the West Nile Virus (WNV) infection and its neurological manifestations has rapidly expanded in recent months. The comments submitted by Dr Leis and colleagues raise concerns regarding the case we presented. The data they cite as contrary evidence to our conclusions were published after the submission of our report (accepted for publication 10 February 2003). Since that time, pathological evidence in human cases has emerged that implicates the spinal cord anterior horn cell in the pathogenesis of WNV associated flaccid paralysis. Nonetheless, the clinical, electrodiagnostic, and MRI findings presented in our case are still valid.

    In our specific case, the patient clearly exhibited signs and symptoms of an acute lower extremity motor paralysis that was supported by electrodiagnostic studies. It is important to reiterate here that such studies are incapable of distinguishing whether the pathology is located in the anterior horn, ventral root, or motor axon. As such, the conclusions and pathological data that preceded our report were potentially premature in implicating only the anterior horn cell, despite the clinical presentation of a poliomyelitis-like syndrome.

    The MRI study we presented showed enhancement of the ventral nerve roots. In the published image no signal change was seen within the adjacent spinal cord itself. Although the nerve roots may enhance with a non-specific meningeal process, the area of signal abnormality in our patient correlated well with the clinical and electrodiagnostic findings. In addition, signal change was not seen in other areas of the MRI scan. Previous MRI studies of poliovirus infection, as well as non-poliovirus infection that involved the anterior horn cell, demonstrated signal changes and enhancement in the region of the anterior horn cell within the spinal cord.1–4 As a result, we concluded that because no such enhancement was seen in the anterior horn cell region, the suspected pathogenesis in our case study may have extended from the anterior horn to include the ventral nerve root.

    It has now been demonstrated by pathological studies that the anterior horn cell is affected in cases of flaccid paralysis caused by the WNV; however, lymphocytic infiltration of the nerve root was also seen.5 At this time, we conclude that perhaps the nerve root, in addition to, or independent of, the anterior horn cell, can also be involved in acute flaccid monoparesis caused by the WNV.

    As our understanding of WNV infection and its neurological manifestations continues to evolve, it remains important to consider varied presentations of the disease, even if at times they appear contradictory. Unfortunately, MRI is an insensitive tool for assessing WNV infection. It is our hope that future prospective and pathological studies will continue to advance our understanding of the pathophysiological mechanisms that underlie the WNV infection.