It is with great interest that we read the article related to the induction of epileptic seizures by intermittent light sources (generated either by TV or video games) that appeared in your journal [1]. The clinical data available at present confirm that the frequency of the flashing lights that induce epileptic seizures lie in the range 5 to 30 pulses per second [2]. What is not known widely in the medical community is that lightning flashes also emit light pulses with a frequency that lies in the above range. A lightning flash is a composite event containing many high current events called return strokes. Each of these return strokes generates a strong light pulse since it heats the lightning channel to temperatures higher than 25 000 degrees Celsius. A typical lightning flash may contain up to about five return strokes, but in some cases the number of return strokes can exceed 20 [3]. The time interval between each event is about 50 ms on average, and, therefore, the light emitted by a lightning flash would pulsate at a frequency of about 20 pulses per second. Now, in severe thunderstorms, the lightning flashing rate can exceed 60 lightning flashes per minute and in extreme cases it may increase to 500 flashes per minute [3]. With the light from each lightning flash pulsating at a frequency of 20 pulses per second and with more than sixty such flashes taking place in a minute, the light generated by such thunderstorms might be responsible for provoking a seizure, especially when the ambient light level is low (e.g. night time when the flashes are more obvious against the dark background), in a person who is sensitive to pulsating light.

Interestingly, the first reference [1] also shows that light pulses can trigger partial epileptic seizures of the occipital lobe. The work done by Panayiotopoulos et al. [4] shows that such seizures can induce visual hallucinations with the victim remaining in a conscious state throughout the seizure. The features of these visual hallucinations are described in detail in that reference. It could be of interest for the medical community, and especially for those who are involved with epilepsy, to know that one can find descriptions of a phenomenon called ball lightning in the lightning literature. Ball lightning has been seen and described since antiquity and recorded in many places around the globe. The sightings of ball lightning are usually associated with thunderstorms; however, as it has not proved possible to produce ball lightning in the laboratory, and as the authenticity of the available photographs is questionable, the properties of ball lightning have to be extracted from eyewitness records. Ball lightning is described as being spherical in shape, although other shapes such as teardrops or ovals have also been reported. On rare occasions, ball lightning phenomena in the shape of rods have also been observed. When ball-lightning sightings have been recorded, the diameter of the balls is typically 10-40 cm, although balls as large as 1 m have been reported. The life-time of ball lightning is reported to be about 10 seconds, but occasionally ball lightning with a duration of as long as one minute has been observed. Ball lightning may manifest in different colors such as red, red-yellow, yellow, white, green and purple. In some cases, the intensity of ball lightning may increase with time and becomes a dazzling white before it disappears explosively. Furthermore, structure of the ball lightning may vary from one report to another: Sometimes a solid core surrounded by a translucent envelope is reported; in other cases it takes the form of a rotating structure or a structure that emits spark-like phenomena. Most ball lightning phenomena are said to move horizontally. Lightning scientists are still struggling to find a scientific explanation for these sightings. It may be more than a mere coincidence that, these reported features of the ball lightning are very similar to the hallucinations caused by the seizures of the occipital lobe, a fact previously recognized by Cooray and Cooray [5]. It is possible that the intermittent light pulses generated by lightning flashes in strong thunderstorms could trigger epileptic seizures in those who are sensitive to pulsating light. A small fraction of them could experience partial seizures of the occipital lobe with an associated hallucination. Such victims, especially those who are aware of the lightning literature, may misinterpret their medical condition as a ball lightning observation. It is also an established fact that children are more prone to photo convulsive seizures. Interestingly, the majority of ball lightning observations are reported by children (or by adults reporting an experience from their childhood or adolescence).

[1] Ferrie, C. D., P De Marco, R. A. Grunewald, S Giannakodimos, and C P Panayiotopoulos, Video game induced seizures, J Neurol Neurosurg Psychiatry, 57, pp. 925-931, 1994.

[2] Harding, G.F.A. and Jeavons, P.M. Photosensitive Epilepsy, 2nd ed., MacKeith Press, London, 1994.

[3] Rakov, A. A. and M. A. Uman, Lightning: Physics and Effects, Cambridge University Press, Cambridge, 2003.

[4] Panayiotopoulos, C. P., Elementary visual hallucinations, blindness, and headache in idiopathic occipital epilepsy: differentiation from migraine, J. Neurol. Neurosurg. Psychiatry, 66, pp. 536-540, 1999.

[5] Cooray, G. and V. Cooray, Could some ball lightning observations be optical hallucinations caused by epileptic seizures?, Open Atmospheric Science Journal, vol. 2,pp. 101-105, 2008.

Conflict of Interest:

None declared

We thank J. Linn for her interest in our paper.

Following acceptance of the final version of our in vivo amyloid imaging paper in December 2009 (1), 2 subsequent papers (2,3) were published that used a different methodology and design but were in full accordance with our findings. In our paper, we report the clinical history, CT, MR and in vivo amyloid imaging in 2 cases with cortical superficial siderosis. Both cases had a positive amyloid scan. The PIB pattern in our two cases with superficial siderosis was closely similar to that typically seen in Alzheimer disease (AD). We conclude that cortical superficial siderosis falls within the spectrum of cerebral amyloid angiopathy (CAA) and Alzheimer disease (AD) and constitutes an occasional complication of cerebral amyloidosis (1). Following acceptance of our paper, a series of 38 pathologically confirmed CAA patients was published (2), including 2 cases who had superficial siderosis on MRI in the absence of microbleeds or macrohemorrhages, similarly to our case 1 (Fig 1B). On that basis, Linn et al. proposed in April 2010 to modify the criteria for CAA and include superficial siderosis as one of the imaging criteria. If we apply the modified criteria, our case 1 fulfills the modified Boston criteria of possible CAA. As mentioned in the discussion (1), case 2 fulfills the criteria of probable CAA according to the Boston criteria and also according to the modified criteria (2010). In a third paper, Kumar et al. (2010) published a consecutive series of 13 cases of atraumatic localized convexal subarachnoid hemorrhage above 60 years of age (along with 16 younger cases who are not directly relevant to the current discussion) (3). In 5-7 cases, the clinical history was remarkably similar to that observed in our 2 cases, with stereotyped paroxysmal episodes of focal neurological loss, sometimes with a migratory pattern. The CT image of intrasulcal hemorrhage also closely resembled that shown in Fig 1A and C (1). On the basis of the MRI findings (superficial siderosis, cortical microbleeds or macrohemorrhage), Kumar et al. propose that convexal subarachnoid hemorrhages in patients above 60 years of age, may be caused by cerebral amyloid angiopathy. Taken together, these 3 papers provide novel and converging evidence that, clinically, cerebral amyloid angiopathy should be considered as a cause in patients above the age of 60 years with stereotyped episodes of focal neurological loss and cognitive decline and superficial siderosis on MRI, even in the absence of other MRI abnormalities, in particular if the distribution of the siderosis is mainly supratentorial.

1. Dhollander I. et al., In vivo amyloid imaging in cortical superficial siderosis. J Neurol Neurosurg Psychiatry 2010 doi 10.1136/jnnp.2009.194480

2. Linn J. et al., Prevalence of superficial siderosis in patients with cerebral amyloid angiopathy. Neurology 2010; 74; 1346-1350

3. Kumar S. et al., Atraumatic convexal subarachnoid hemorrhage: clinical presentation, imaging patterns, and etiologies. Neurology 2010; 74; 893- 899

Conflict of Interest:

PI of a GE Healthcare sponsored phase I and phase II study, as well as of therapeutic phase II and phase III trials sponsored by EliLilly, Medivation, Novartis and Pfizer.

We thank Zara G for her interest in our paper on "Rituximab in patients with chronic inflammatory demyelinating polyradiculoneuropathy (CIDP): a report of 13 cases and review of the literature" (1). Most of her comment reflect indeed what we reported and discussed in the manuscript. We would like just to make a few comments to clarify some aspects. We were induced to perform this study by the few anecdotal reports (2-4) on the efficacy of Rituximab in patients with chronic inflammatory demyelinating polyradiculoneuropathy (CIDP). We therefore retrospectively analysed the response to therapy in a series of patients with CIDP unresponsive to conventional therapies treated independently in different centres. In only few patients peripheral blood immunophenotype was characterized before therapy and this is why we did not include the data on the manuscript. From a diagnostic point of view all our patients had indeed a chronic demyelinating polyneuropathy that fulfilled the diagnostic criteria of CIDP proposed by European Federation of Neurological Societies/Peripheral Nerve Society (5). According to these criteria patients with IgM monoclonal gammopathy are excluded from the diagnosis of CIDP if they have anti-MAG antibodies otherwise they are included in the diagnosis of CIDP associated with other diseases. The results of our study may indeed raise some concerns about including in CIDP patients with IgM monoclonal gammopathy and no anti-MAG reactivity, as they appear, at least from our retrospective study, to have a better response to Rituximab than patients with idiopathic CIDP. It is not possible however to know if a reduction of antibodies caused the improvement since we did not find any reactivity with nerve in our patients. Other factors, including the effect of Rituximab on immunoregulatory T cell, may have also been implicated in the response to therapy (6-8). In our study we also analysed a few additional neurophysiologic parameters, besides motor conduction velocity (MCV), including compound muscle action potential (CMAP) amplitudes, distal latencies and F waves. We only included data on MCV as these showed the higher improvement. These data are however only ancillary to our study which mainly relies on clinical findings. This was also the case of the recent randomized controlled trials on CIDP (MTX e IFN) where the results of electrophysiologic studies were not included in the main study. In conclusion our retrospective study confirms that Rituximab may have some efficacy in patients with CIDP particularly when associated with haematological diseases even if the data need to be confirmed in a randomized controlled study. In the meantime, we think that Rituximab may be considered as a possible option for patients with CIDP associated with haematological diseases not responsive to conventional therapy.

References

1. Benedetti L, Briani C, Franciotta D, et al. Rituximab in patients with CIDP: A report of 13 cases and review of the literature. J Neurol Neurosurg Psychiatry 2010 Jul 16. [Epub ahead of print]

2. Knecht H, Baumberger M, Tobon A, et al. Sustained remission of CIDP associated with Evans syndrome. Neurology 2004;63:730-2.

3. Briani C, Zara G, Zambello R, et al. Rituximab-responsive CIDP. European Journal of Neurology 2004;11:788-791.

4. Benedetti L, Franciotta D, Beronio A, et al. Rituximab efficacy in CIDP associated with idiopathic thrombocytopenic purpura. Muscle & Nerve 2008;38:1076-1077.

5. Hughes RA, Bouche P, Cornblath DR, et al. European Federation of Neurological Societies/Peripheral Nerve Society guideline on management of chronic inflammatory demyelinating polyradiculoneuropathy: report of a joint task force of the European Federation of Neurological Societies and the Peripheral Nerve Society. Eur J Neurol. 2006;13:326-332.

6. Dalakas MC, Rakocevic G, Salajegheh M, et al. A placebo-controlled trial of rituximab in IgM anti-myelin-associated glycoprotein antibody demyelinating neuropathy. Ann Neurol 2009;65:286-93.

7. Eisenberg R, Looney RJ. The therapeutic potential of anti-CD20 "what do B-cells do?".Clin Immunol. 2005;117:207-13.

8. Stasi R. Rituximab in autoimmune hematologic diseases: not just a matter of B cells. Semin Hematol. 2010;47:170-9.

Conflict of Interest:

None declared

Dear Editor,

In 2005, our case report entitled 'Acute longitudinal myelitis as the initial manifestation of Sjogren's syndrome' was published in this journal (1). In that report, we described the case of a 31-year-old woman who presented with acute longitudinal myelitis extending along the entire spinal cord. She was also diagnosed with Sjogren's syndrome on the basis of positive anti-SS-A antibody test results and findings of lip biopsy. We concluded that this condition was myelitis associated with Sjogren's syndrome, considering that it was unlikely to be Devic's syndrome or neuromyelitis optica (NMO), because of the absence of optic nerve lesions, and on the basis of the conventional diagnostic criteria at that time. After this paper was published, we received an e-mail from Dr. Weinshenker who recommended that the patient be tested for the presence of NMO-IgG. Hence, we collected her serum sample on 22 July 2004, and sent it for NMO- IgG testing to the Neuroimmunology Laboratory, Mayo Clinic; however, the results of the NMO-IgG test were negative. Thereafter, the patient further experienced 4 episodes of myelitis but none of optic neuritis. Her brain magnetic resonance (MR) image obtained during the first episode was normal, but that obtained during the second episode showed periventricular lesions with gadolinium enhancement. Recently, we developed an enzyme-linked immunosorbent assay (ELISA) for detecting anti-aquaporin-4 antibody (2), and we measured this antibody in the same serum sample that was used for testing NMO-IgG; the results of the ELISA were strongly positive. Moreover, the extended disease entity 'NMO spectrum disorder' was proposed by Wingerchuk et al. in 2007 (3), including idiopathic single or recurrent episodes of longitudinally extensive myelitis. Eventually, we revised the diagnosis of 'myelitis associated with Sjogren's syndrome' as 'neuromyelitis optica spectrum disorder'. This was supported by the fact that the serum sample that had been negative for NMO-IgG in the original test at the Neuroimmunology Laboratory was found to be positive for NMO-IgG/anti-aquaporin-4 antibody in a retest using their novel and improved assays by Dr. Pittock. There are many reports concerning the involvement of the central nervous system (CNS), including myelitis associated with Sjogren's syndrome (4). However, the association between CNS involvement, especially longitudinally extensive myelitis, and Sjogren's syndrome remains controversial. Firstly, most of the reports were published before the discovery of NMO-IgG or anti-aquaporin-4 antibody. Secondly, NMO-IgG or anti-aquaporin-4 antibody test results should be carefully interpreted. Many techniques were used for the detection of these antibodies, but the sensitivities of these techniques were thought to differ. Further, sequential analysis of anti-aquaporin-4 antibody titres in the same patient revealed dynamic changes in the antibody levels related to the disease activity and treatments. A negative result in 1 test does not exclude the possibility of a positive result in another test. Lastly, it has been known that not a few NMO patients concurrently present with anti- SS-A antibody in their serum or Sjogren's syndrome but without any disease activity of Sjogren's syndrome. Moreover, it has been reported that NMO- IgG is detected only in patients with NMO spectrum disorder, but not in Sjogren's syndrome patients who do not have optic neuritis or myelitis (5). From these points, it is necessary to carefully rule out NMO in order to evaluate CNS involvement associated with Sjogren's syndrome. Further investigation is needed to clarify whether longitudinal myelitis in patients with Sjogren's syndrome is a condition caused by Sjogren's syndrome or neuromyelitis optica spectrum disorder, and whether 'myelitis associated with Sjogren's syndrome' truly exists. P.S. We thank Dr. Weinshenker for his helpful advice, and Dr. Pittock for confirmation of the positivity of NMO-IgG/anti-aquaporin-4 antibody by novel and improved assays.

References

1. Yamamoto T, Ito S, Hattori T. Acute longitudinal myelitis as the initial manifestation of Sjogren's syndrome. J Neurol Neurosurg Psychiatry 2006;77:780.

2. Hayakawa S, Mori M, Okuta A, et al. Neuromyelitis optica and anti- aquaporin-4 antibodies measured by an enzyme-linked immunosorbent assay. J Neuroimmunol 2008;196:181-7.

3. Wingerchuk DM, Lennon VA, Lucchinetti CF, et al. The spectrum of neuromyelitis optica. Lancet Neurol 2007;6:805-15.

4. Rabadi MH, Kundi S, Brett D, et al. Primary Sjogren syndrome presenting as neuromyelitis optica. J Neurol Neurosurg Psychiatry 2010;81:213-4.

5. Pittock SJ, Lennon VA, de Seze J, et al. Neuromyelitis optica and non-organ-specific autoimmunity. Arch Neurol 2008;65:78-83.

Conflict of Interest:

None declared