Intractable hiccup and nausea (IHN) are unique symptoms of neuromyelitis optica (NMO),1 which is a devastating neurological disorder mainly characterised by optic neuritis and myelitis.2 In our previous study, IHN was found in eight of 47 cases of relapsing neuromyelitis optica (NMO) (17%), but in none of 130 cases of multiple sclerosis (MS).1 Although the mechanism of developing IHN in NMO is not fully understood, lesions involving pericanal regions in the medulla oblongata, where the putative hiccup and vomiting centres are located, are thought to cause these symptoms.1 Recent studies strongly suggested that the pathogenesis of NMO is closely associated with anti-aquaporin-4 (AQP4) antibody,3^–6 which is a serum autoantibody that is highly specific to NMO.5 7 8 AQP4 molecules are expressed particularly in the periventricular and periaqueductal areas, including in the pericanal regions of the medulla oblongata and, more specifically, are concentrated at the astrocytic foot processes that line the outside of the blood–brain barrier.9^–11 Consistent with the distribution of AQP4, the brain lesions of NMO were reported to be localised in the periventricular and periaqueductal areas.1 12 13 Thus, the development of IHN is also thought to be closely associated with anti-AQP4 antibody.

In the previous study, we included patients with NMO who were not confirmed to have anti-AQP4 antibody.1 This uncertainty may have influenced the results of the study. Therefore, in order to re-evaluate the frequencies and clinical implications of IHN in NMO, we retrospectively investigated the patients with NMO and the high-risk syndrome of NMO, all of whom were seropositive for anti-AQP4 antibody detected by our assay.5 14 We also describe a case with IHN and show the time course of anti-AQP4 antibody titres.

MATERIALS AND METHODS

Patients and sera

We reviewed the medical records of 35 Japanese patients, including 7 patients in the previous study,1 in our facility (male/female: 0/35; age at onset: median 37 (range 13–72); disease duration 8 (1–32); EDSS 6 (1–8.5)), 25 with NMO and 10 with high-risk syndrome of NMO (cases highly suggestive of converting to NMO), which were already confirmed to be seropositive for anti-AQP4 antibody by the method described in our previous study.5 14 Although the previous study also included cases from other hospitals,1 only the cases seen at Tohoku University Hospital were analysed in the present study. NMO was defined as cases fulfilling the revised criteria of NMO.15 In this diagnosing process for NMO, we indicated the seropositivity for anti-AQP4 antibody as a substitute for the NMO-IgG status because the anti-AQP4 antibody assay was expected to have higher sensitivity and equal specificity compared with NMO-IgG.5 14 High-risk syndrome was defined as cases with optic neuritis or myelitis with seropositivity for anti-AQP4 antibody. IHN was defined as bouts lasting for more than 48 hours. We evaluated the appearance pattern of IHN and classified it into one of three categories. “Preceding” means that the IHN preceded the acute exacerbations such as optic neuritis or myelitis. The preceding periods (Days, Weeks and Months) were calculated according to the descriptions of the medical records. When we could determine the precise date, we used “Days”. When only approximate dates were available, we used “Weeks” or “Months”. “Accompanied” means that the IHN and other neurological symptoms developed at the same time. “Independent” means that the IHN were the sole symptoms in the episode. In addition, we recently experienced a case with intractable hiccup and followed the anti-AQP4 antibody titres in the patient.

All of the sera were taken in the period 2001–2007 from those 35 patients and cryopreserved at −80°C until use in the anti-AQP4 antibody assay. The study was approved by the institutional review board of the Tohoku University School of Medicine, Sendai, Japan. Patients gave spoken consent for the study.

Anti-human AQP4 antibody assay

We detected and titrated anti-AQP4 antibody by the method described in our previous report.14 Briefly, human embryonic kidney (HEK-293) cells were stably transfected with the vector with or without AQP4 cDNA, and the specimens were tested by the indirect immunofluorescent assay using these two cell lines (with or without AQP4) as the substrates. We used Alexa Fluor 488 goat anti-human IgG (Invitrogen, Eugene, Oregon, USA) instead of fluorescein-conjugated antibody, which was used in the previous studies,5 14 as the second antibody in the indirect immunofluorescent assay. This modification resulted in slightly higher antibody titres in this study compared with the previous reports.5 14 The serum samples were titrated in doubling dilutions to ascertain the maximum dilution that positively stained AQP4-transfected HEK293 cells.

We have followed the anti-AQP4 antibody titres of almost all NMO or high-risk cases in our hospital every 3–6 months from April 2006. We also titrated the antibody at all exacerbations in those cases.

Statistical analysis

We compared the clinical parameters in the patients with and without hiccup and nausea. In this comparison, Mann–Whitney’s U test and Fisher’s exact probability test were used, as appropriate. All of the statistical analyses were considered significant if the p-values were less than 0.05.

RESULTS

Intractable hiccup and nausea were frequently seen as preceding symptoms of the exacerbations

A total of 15 of the 35 patients (43%) with anti-AQP4 antibody had episodes of IHN. No obvious differences in the clinical parameters were seen between the patients with and those without IHN except for the high prevalence of medulla lesions (47%) in the patients with IHN (p = 0.0192) (table 1). There was a total of 35 episodes of IHN in these 15 patients and, of the 35 episodes, hiccup was seen in 23 episodes (66%) and nausea was seen in 28 episodes (80%) (table 2). The IHN frequently preceded (19/35, 54%) or accompanied (10/35, 29%) other neurological symptoms and the IHN were preceded by viral infection in 7 episodes (20%). The following lesions were myelitis or medulla lesions (93%, 27/29) and optic neuritis (10%, 3/29). The majority of the IHN could be successfully treated with high-dose intravenous methylprednisolone therapy (80%, 28/35). The others were treated with oral prednisolone (11%, 4/35) or were self-limiting (8%, 3/35).

A surge of anti-AQP4 antibody titre was seen during an episode of intractable hiccup in one case

A 37-year-old woman with NMO (age at onset: 35; exacerbations: 1 with left optic neuritis and 1 with medulla lesions with IHN)—whose titre of serum anti-AQP4 antibody was 64× in the remission phase without any treatment in August 2006—developed a sore throat and cough with hiccup from 30 November 2006. The sore throat and cough were diagnosed as common cold and were self-limiting, but the hiccup persisted. She visited our hospital on 11 December 2006 and the anti-AQP4 antibody titre was 1024× at that time. In addition, she developed nausea from 15 December onwards, which persisted with the hiccup. The brain MRI on 18 December revealed small enhancing lesions in the white matter of the left temporal lobe and right frontal lobe. Then, high-dose intravenous methylprednisolone therapy was initiated on 21 December (anti-AQP4 antibody titre was 512× at that time) and the hiccup and nausea disappeared just after the initiation of the treatment. The titre of anti-AQP4 antibody became 256× on 25 December and 64× on 11 January and 12 March 2007.

DISCUSSION

This study showed a high frequency of IHN in the patients with anti-AQP4 antibody and a case with remarkably increased anti-AQP4 titre during an episode of intractable hiccup. In addition, the hiccup and nausea often preceded the neurological symptoms, such as optic neuritis and myelitis, and were preceded by an episode of viral infection. These findings could be not only clinically important, but also may be related to the pathogenesis of NMO.

Although viral fever was often described as a preceding episode for exacerbations in NMO,16 the mechanisms involved are unclear. However, our findings suggest that a surge of anti-AQP4 antibody could be induced by some immunological stimulation such as that by a common cold, and recent studies tend to support that anti-AQP4 antibody may have important roles in the pathogenesis of NMO. The loss of AQP4 was observed in the active perivascular lesions of NMO,3 4 and a correlation between anti-AQP4 antibody titres and the lengths of spinal cord lesion was seen during exacerbations of NMO.5 In addition, very recently, Hinson et al. showed that NMO patients’ serum IgG had a pathological effect on cell membranes expressing aquaporin-4, such as aquaporin-4 endocytosis/degradation and complement activation.6 Considering these findings, the surge of anti-AQP4 antibody and the following reactions with AQP4 molecules may explain, in part, the mechanism by which exacerbations are initiated. Then, why did hiccup and nausea frequently precede the main neurological symptoms? The pericanal region in the medulla oblongata, the putative centre of the hiccup and nausea, includes the area postrema, where the BBB is anatomically not seen. This means that AQP4 molecules on astrocytes in the area postrema are likely to be affected by anti-AQP4 antibody, which may lead to developing hiccup and nausea preceding other neurological symptoms, although we need to clarify in further studies why lesions in the medulla are less common than optic neuritis and spinal cord lesions. Accordingly, when IHN is seen in the patients with NMO, careful observation may be needed for timely intervention because the early initiation of treatment may prevent subsequent neurological damages such as optic neuritis or myelitis, and even one exacerbation could have critical effects on the activities of daily living of patients with this devastating disease.

All the patients enrolled in the present study were women because we included only female NMO or high-risk syndrome cases. Such a female preponderance in relapsing NMO has been reported previously,2 but further studies are needed to clarify whether the findings obtained in this study are applicable to male patients with anti-AQP4 antibody.

In conclusion, IHN were frequently seen in patients with anti-AQP4 antibody and those symptoms often preceded optic neuritis and myelitis. Careful observation is needed when hiccup or nausea is seen in patients with NMO.