Paroxysmal head drops with ataxia-like symptoms presenting as Sandifer syndrome in a 3-year old girl

Abstract

We present a case of Sandifer syndrome in a 3-year-old girl who initially presented with a history of recurrent paroxysmal head drops associated with ataxia-like symptoms and recurrent falls sustaining a clavicular fracture on one occasion. She was referred to and seen by the paediatric neurologist. Physical examination, electroencephalogram, MRI brain, electromyograph single fibre study and blood tests were all normal. With the history of hiccups and choking-like episodes she was referred to the speech and language therapist (SALT). SALT assessment did not reveal indications of swallowing impairment or possible aspiration. A barium swallow later showed small amount of reflux into the distal oesophagus. This prompted a trial of lansoprazole and she was referral to the gastroenterologists. Endoscopy and oesophageal manometry were essentially normal. However, the pH impedance study revealed severe gastro-oesophageal reflux disease. She continued with lansoprazole and dairy-free diet and her symptoms resolved.

Background

Sandifer syndrome (SS) is a rare condition for which the diagnosis requires a high level of clinical suspicion. In this case report, the authors have nicely described the diagnostic workup needed to recognise this condition. By presenting the whole patient’s experience to the reader, this article highlights the key ‘clinical tips’ that any clinician should know when dealing with similar cases. Since the initial manifestations of SS are neurological, patients are usually referred for lengthy, expensive and unnecessary testing, with concurrent misdiagnosis and inappropriate use of medication. Through this report, we hope that more clinicians become aware of this syndrome and improve their diagnostic skills.

Case presentation

A 3-year-old girl with a history of recurrent paroxysmal head drops associated with ataxia-like symptoms was referred to the neurologists.

In terms of background, she had an unremarkable neonatal history. There were no significant previous medical history and her development was appropriate for age. There was a family history of multiple sclerosis and hypothyroidism.

She reported a history of unexpected head tilt towards the left which would remain in that position for about 30 seconds and then suddenly return to the centre. She would then appear to have a sudden loss of muscle tone, become hypotonic, lie on the floor with arms flexed saying she feels as if she has hiccups, would complain of abdominal discomfort and on questioning she would say ‘it's in my head’. She would unusually become distressed, extremely tired and sleep even in the presence of friends around her. These symptoms were predominantly following meal times. The symptoms were affecting her quality of life to the point where her nursery hours had to be reduced. These symptoms were often associated with a loss of balance and an unsteady gait causing her to fall and fracture her clavicle on one occasion. She was initially investigated for a possible muscle disorder but physical examination, electroencephalogram (EEG), MRI brain, electromyograph (EMG) single fibre study and blood tests including full blood count, urea and electrolytes, liver function tests, thyroid function tests, erythrocyte sedimentation rate, acetylcholine receptor antibodies, autoimmune screen, creatine kinase, coeliac, thyroid function tests and metabolic workup were all normal. During EEG, she had numerous episodes of neck flexion, all of which were not associated with changes on the EEG. She appeared to slump forward and to the side, but there was no definite atonia on EMG electrodes. On one occasion she stood still and stated she had hiccoughs, no EEG changes were observed on this occasion either. There were numerous events captured but again there was no neurophysiological evidence to support an epileptic basis for these episodes. The reason for these paroxysmal head drops remained elusive.

Due to the history of choking at 6 months of age at the time of weaning and possible association of head tilts at meal times, she was referred to speech and language therapist (SALT). During the SALT assessment, she was observed eating and drinking with no indication of swallowing impairment and no clinical indications of possible aspiration. She however complained of hiccups and was repeatedly posturing upside down (in a downward dog-like yoga posture). She came increasingly distressed and could not be distracted from this behaviour. Later, a barium swallow showed small amount of oesophageal reflux into the distal oesophagus. This prompted a trial of lansoprazole by the neurologist and she was referred to the gastroenterologists. She showed a gradual improvement in her symptoms.

Endoscopy and oesophageal manometry were essentially normal with no evidence of hiatus hernia (HH). However, the pH impedance study revealed severe gastro-oesophageal reflux disease (GORD) and frequent episodes of air-swallowing. She continued with lansoprazole and her symptoms improved but did not completely resolve. The team were considering fundoplication, however, prior to that she was put on a trial of a dairy-free diet for which she was symptom-free. Following reintroduction of dairy, her symptoms became more frequent. Her symptoms completely resolved on lansoprazole and dairy free diet.

Investigations

A summary of relevant investigations is summarised in table 1 and figures 1 and 2.

Table 1.

Summary of investigations

Investigations	Results (with normal references in parenthesis)
Haemoglobin	111 g/L (115–145)
MCV	73.5 fL (75–87)
Ferritin	5L μg/L (8.6–74)
U&E, LFTs, TFTs, CK, ESR, CRP, AChR antibodies, ANA, ANCA, anti-dsDNA and metabolic work	Normal
Anti-TTG	0.2 U/mL (0.0–6.9)
Total IgA	0.89 g/L (0.4–2.0)
Iron	5.6 μmol/L (5–25)
TIBC	79.5 μmol/L (29–63)
Total IgE	3.2 kU/L (0–52)
EEG, MRI, EMG	Normal
Barium swallow	Unremarkable appearances of the oesophagus, no hold up of contrast column. Normal GOJ and stomach. The DJ flexure is normally sited. There is a small amount of oesophageal reflux into the distal oesophagus.
Endoscopy	Oesophageal squamous epithelium with no significant histological abnormalities. Gastric antral and small bowel biopsy showing no abnormalities. No helicobacter-like organisms seen.
pH metry	Total acid exposure time 1% (normal <5.4%). Conclusion: Normal amount of reflux.

AChR, acetylcholine receptor antibodies; ANA, antinuclear antibody; ANCA, antineutrophil cytoplasmic antibody; anti-dsDNA, antidouble strand DNA; anti-TTG, tissue transglutaminase autoantibodies; CK, creatine kinase; CRP, C-reactive protein; DJ, duodenojejunal; EEG, electroencephalogram; EMG, electromyograph; ESR, erythrocyte sedimentation rate; GOJ, gastro-oesphageal junction; LFTs, liver function tests; MCV, mean corpuscular volume; TFTs, thyroid function tests; TIBC, total iron binding capacity; U&E, urea and electrolytes.

Figure 1.

Multichannel intraluminal impedance combined with pH (MII‐pH).

Figure 2.

Oesophageal manometry.

Figure 1 illustrating the MII-pH findings. The report as shown in the summary below showed severe GORD and frequent episodes of air swallowing. The symptom association probability (SAP) score was 100%.

• Study duration 18.39 hours.

• Increased total acid exposure time 18.4% (normal <5.4%), may be underestimated, due to very low impedance baseline and increased air swallowing.

• Increased total number or reflux episodes 104 (normal <70), in which 85 were acidic and 19 weakly acidic.

• The longest acid exposure lasted 33.5 min.

• 39/104 reflux episodes reached proximal impedance channels.

• 34 episodes of V positioning positively associated with reflux (SAP=100%).

• Four episodes of fatigue reported during the study, positively associated with reflux (SAP 100%).

• Frequent episodes of air-swallowing.

Conclusion: Severe GORD and frequent episodes of air-swallowing.

• SAP, a parameter that expresses the likelihood that the patient’s symptoms are related to reflux.¹

Figure 2 showing the oesophageal manometry study. The report as shown below was essentially normal.

• Normal resting pressures of upper and lower oesophageal sphincters (LES).

• Normal relaxation of LES with each swallow.

• Normal oesophageal body peristalsis.

• No episodes of LES relaxation or reflux seen.

Differential diagnosis

Paroxysmal abnormal movement or posturing are commonly seen in infants, young children and adolescents. Given the spectrum of clinical conditions associated with these symptoms, it is vital that there is a rational diagnostic workup as the management differs in each case. The diagnosis often remains a challenge for paediatricians. Differential diagnoses can include paroxysmal dyskinesias which are rare involuntary movement disorders that are typically episodic in nature. Neurological examination may be entirely normal between the attacks.²

It is important to rule out a structural brain lesion for the underlying symptoms. Other important differentials to consider are Tic disorders including Tourette syndrome, infantile spasms, anomalies of cervical vertebrae and congenital muscular torticollis. It is worth mentioning that although rare in developed countries, Tetanus can manifest with the same symptomatology. In addition, another rare but potentially fatal condition is Nodding syndrome. This is a poorly understood neurological disorder of unknown aetiology that affects children and adolescents in Africa, it is commonly seen in United republic of Tanzania, South Sudan and Uganda. Typically, there is involuntary drops of the head to the chest which can be associated with cognitive and growth decline as well as other epileptic activity.³

Paroxysmal abnormal movement or posturing occurring during or after meals can be suggestive of Sandifer syndrome (SS) and this should alert the clinician to consider this differential if there is a clear history and relationship between food intake and these paroxysmal events. Furthermore, symptoms such as vomiting, choking, hiccoughs and burping should be taken into account when considering SS.

Treatment

Treatment with lansoprazole and dairy free diet allowed for complete resolution of her symptoms.

Outcome and follow-up

The patient’s symptoms had fully resolved. To date, the patient has remained asymptomatic at follow-up appointments. She remains closely followed-up with her general paediatrician.

Discussion

Paul Sandifer was a paediatric neurologist who worked at the Hospital for Sick Children, Great Ormond Street. It was here that his former student, Marcel Kinsbourne, described in 1964, the association of HH with abnormal posturing and movements of the head, neck and body in five children,⁴ two of these patients were under the care of Paul Sandifer.⁵

This association was further described by Sutcliffe and Gellis in 1969 and 1971, respectively.^{6 7} The syndrome was named ‘Sandifer syndrome’ (SS) by Sutcliffe in 1969 and is sometimes known as Sandifer-Sutcliffe syndrome.⁶

Murphy and Bray suggested that the abnormal movements maybe associated with GORD and that HH is not compulsory for the diagnosis as previously stated by Kinsbourne.^{8 9}

SS manifests as sudden onset dystonia affecting the neck, trunk or limbs due to gastro-oesophageal reflux (GOR).¹⁰ Patients with SS are typically seen initially by the neurologists and are evaluated with unnecessary investigations.⁹ In SS, various abnormal movements and postures can occur that affect the different parts of the body, most commonly the neck, the torso and the upper limbs.¹⁰ The attacks can be sudden. They usually present with cervical dystonia and due to the involuntary contracting of the head and neck muscles, can lead to torticollis (rotation), tilting of the neck, or extension of the neck.⁹ Duration of the abnormal position is usually short, about 1–3 min, but can last longer and there have been reports for these posturing lasting 30 min.¹¹

There may also be opisthotonic posturing and eye movements.¹² SS is more common in children with underlying neurological conditions.¹³ These abnormal movements are typically absent during sleep^{6 14 15} and have been shown to be associated with food intake.¹⁶

Patients may show signs and symptoms of dysphagia, vomiting, abdominal pain, anaemia or haematamesis.¹⁷ Babies can present with apnoeas, cyanosis, choking episodes, vomiting and jerks.¹⁰ The clinical picture of GORD can have a wide range of symptoms. The classical symptoms of reflux may be absent. In one particular study, 50% of patients diagnosed with SS had no symptoms of GOR.¹⁸ Vomiting which is usually a common symptom of GOR, may be absent.¹³

There are studies suggesting that cow’s milk protein allergy can be associated with SS.¹⁹ Bamji et al reported two cases of suspected SS where there was no improvement despite treatment with proton pump inhibitors. On changing the milk to an amino-acid-based formula they showed resolutions of symptoms. Furthermore, there is a known association between food allergies and GORD.^{16 20} The symptoms are thought to be due to allergic oesophagitis.²¹

SS is more common in children who have GORD or HH.¹² The incidence of SS is unknown. It is believed that less than 1% of children with GORD also have SS.²² This figure could be underestimated as SS is under reported.²² Kotagal et al performed a study on 883 patients who were monitored for paroxysmal non-epileptic events between 1989 and 1995. They found that 16% of the children from the age group of 2 months to 5 years had GORD as a cause for their abnormal neurological movement.²³

SS is usually a disease of infancy or childhood. However there have been reports of cases in the adolescent and adult age group.²⁴ SS has also been reported in neonates,^{10 11} the youngest being a 9 day old.¹⁰ There have been to the best of our knowledge, only six case reports that have been described in the adult population.^{22 25–29}

Most studies have highlighted that SS is more common in boys, ^{6 30} however, Gorrotxategi et al reports it is more common in girls than boys in their study.¹⁸ Some authors advise investigating all patients with torticollis for GORD.⁸ Generally, the older the patient, the more varied the symptomatology might be and the more difficult the diagnosis becomes.³¹ Patients are usually diagnosed with epilepsy and are placed on multiple anticonvulsants with no improvement and are then found to have SS as the underlying cause for their symptoms.³²

What is still unclear is why some patients with GORD present with extraintestinal signs while others do not. The actual pathophysiology of SS is still not understood and there are various mechanisms described in the literature.

Most authors including ourselves consider the abnormal movements associated with SS to be a learned behaviour. With reference to our patient, the abnormal posturing (downward dog-like yoga posture) most probably provided relief from her discomfort. The findings of Puntis et al seem also to support this hypothesis.¹⁴ It is thought that children by chance perform these movements and discover that there is a relief of the discomfort they experience and then they continue performing this.^{4 6 16}

Kinsbourne’s results were conflicting in that he stated that his patients reported an improvement in reflux symptoms on posturing, however, radiological studies at the same time suggested a worsening of reflux.⁴ Puntis showed that these posturing movements cause an increase in oesophageal motility leading to a clearance of acid thus improving their symptoms.¹⁴

Webb proposed that the diaphragmatic muscles would become irritated secondary to the HH and coexisting acid reflux and thus causing referred cervical dystonia.³³ The authors explained that since the diaphragm and neck muscles share common nerve roots, the dystonic movements in the neck occur when the diaphragmatic nerves are stimulated by reflux. However, this fails to explain why during sleep patients with reflux do not display any dystonic movements.

Kawahara et al proposed that in patients with HH, straining or coughing or any activity that would cause contraction of the abdominal musculature would induce reflux.³⁴

Frankel et al suggested stimulation of the diaphragm or abdominal wall contractions would trigger reflux and lead to head tilt.³⁰ The authors here mention that their study further adds to the evidence that acid reflux is linked with posturing. They hypothesise that the head tilting is a learned behaviour or tic. The tic in turn causes contraction of the neck and abdominal wall muscles leading to strain-induced reflux. They showed an association between abdominal wall contractions and reflux episodes.

Studies describing the relationship between pH, GOR and abnormal movements are few in number and based on single patients and yielded conflicting results.^{11 34 35}

However, all these proposed mechanisms fails to explain why these abnormal movements are absent during sleep. The theories currently described in the literature fail to explain why some patients with GORD do not have these neurological manifestations. The pathophysiology still remains elusive.

The diagnosis of GORD is based on clinical history, symptomatology and physical examination. The diagnosis can be difficult due to its range of clinical manifestations and due to the various diagnostic modalities.³⁶

Joint recommendations from the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition and the European Society for Paediatric Gastroenterology, Hepatology and Nutrition 2018 ahead of print advise in the context of investigating GORD, barium contrast studies and ultrasound may be used to exclude anatomical abnormalities and not merely for the diagnosis of GORD. Furthermore, endoscopy can be used to assess complications of GORD and manometry when a motility disorder is suspected. They suggest scintigraphy should not be used for the diagnosis of GORD.³⁷

Multichannel intraluminal impedance combined with pH (MII‐pH) detects acid, weak acid and non-acid reflux episodes and is superior to pH metry.³⁸ MII-pH monitoring is now becoming the investigation of choice for diagnosing GOR.³⁹

Treatment of the underlying GOR/GORD will eliminate the symptoms.⁹ However, the abnormal movements improve gradually over several weeks¹⁶ and this can be due to the improvement of the GOR or the fact that it is a learned behaviour which gradually settles.

Surgery for GORD is usually the last resort after other options have failed or is usually offered to manage GORD-related complications.³⁸ In patients with HH, surgical repair of the HH improves symptoms.⁴ Laparoscopic Nissen fundoplication is considered the gold standard of severe GORD and has replaced Nissen’s fundoplication.⁴⁰

In our case, our patient presented with abnormal neurology with no classical reflux symptoms. Her background was unremarkable for reflux apart from the brief self-resolving choking episodes at the age of 6 months. Although she was complaining of hiccups and abdominal discomfort, the classical reflux symptoms were absent, there was no vomiting.

As mentioned earlier, one study found 50% of patients had no symptoms of GOR.¹⁸ The barium swallow showed very small amount of reflux, however, barium swallow might well be normal.^{10 18} One particular study showed pH metry was normal (% of time with intraoesophageal pH less than 4) in 62.5% of patients with SS.¹⁸ The pH impedance study revealed severe GORD with frequent episodes of aerophagia. Aerophagia is rarely discussed in the paediatric literature. It is quite possible that her abnormal neurology were exacerbated by her frequent air-swallowing. The definition of aerophagia is included in the Rome II and III criteria for functional gastrointestinal disorders.⁴¹ Despite her frequent air-swallowing, manometry did not show any signs of motility disorder.

Fundoplication was considered as although she made a general improvement on lansoprazole her symptoms were still present. However, prior to that she was put on a trial of a dairy-free diet for which she was symptom-free. Following re-introduction of dairy, her symptoms became more frequent. Her symptomatology completely resolved on lansoprazole and a dairy-free diet.

Our case report provides further information about SS, which may occur in the absence of reflux symptoms, absence of oesophagitis on endoscopy and absence of HH. It further adds to the evidence regarding the implications of dairy foods on SS and highlights the role and superiority that MII-pH monitoring has over standard pH monitoring in diagnosing GOR.

Learning points.

• Patients with paroxysmal abnormal movement or posturing, refractory to medication for epilepsy and movement disorders should be evaluated for Sandifer syndrome (SS) even in children beyond the infantile age group.

• We recommend that a dairy-free diet should be trialled in conjunction with medication prior to consideration of surgery.

• SS can present with no reflux symptoms.

• SS should be considered in cases of refractory seizures resistant to anticonvulsant therapy.

• Early diagnosis allows prompt treatment and relief of symptoms.