A patient with apparently well-controlled epilepsy with a painful musculoskeletal condition was treated successfully with two sessions of acupuncture. However, 4 h after the first treatment and during the second, an adverse event involving impairment of consciousness occurred. The patient subsequently experienced an increased frequency of complex partial seizures resulting in the loss of his driving licence. A detailed retrospective review of the past medical history indicated that the patient probably had comorbidities in the form of rapid eye movement sleep behaviour disorder and dysfunctional somatosensory/vestibular processing. Acupuncture may have triggered the adverse event via shared neurosubstrates. This adverse event raises possible implications regarding safe clinical acupuncture practice.
- SLEEP MEDICINE
Statistics from Altmetric.com
According to national and international treatment guidelines, epilepsy is not a recognised contraindication to acupuncture therapy. In Traditional Chinese Medicine, acupuncture is a widely used treatment for epilepsy although the evidence for effectiveness is not strong.1
In September 2010 a well-built (body mass index 27.5) 38-year-old right-handed man was referred by his GP to a primary care acupuncture clinic for treatment of post-traumatic left knee pain associated with muscle imbalance and patellofemoral joint stiffness; he had associated pain in the right popliteal area. The pretreatment verbal numerical rating score range was 4–8 (where 0=no pain, 10=worst pain imaginable). The injury had been sustained in a road traffic accident 1 year before the consultation; the patient had been the driver and the accident had been attributed to mental preoccupation. His condition had not responded to physiotherapy and non-steroidal anti-inflammatory drugs.
He had developed idiopathic complex partial seizures (CPS) with occasional secondary generalisation at the age of 17. At the time of referral his epilepsy was regarded as being well-controlled by lamotrigine 150 mg and sodium valproate 2000 mg daily. He was also experiencing sporadic vague staring episodes with associated facial pallor which were attributed to presyncope and gave a history of an unusual event (UE) which had occurred during reflexology (UE1, table 1). In view of the latter episode, it was suspected that he was a strong responder. He had experienced another UE while undergoing an EEG (UE2, table 1) which had been attributed to hyperventilation.
All treatments were applied with the patient in the recumbent or semirecumbent position using S/needle J type 0.25×30 mm single-use sterile needles. In view of the patient's probable strong responder status, mild brief treatment avoiding the use of manual or electrical stimulation was employed.
At the first treatment myofascial tender points were found in the superior and inferior patellar ligaments, the proximal and distal lateral ligament insertions, GB34 and ST36 of the left knee. Mild stimulation of 1 min duration per point was applied, six needles in total. Ligamentous points were needled to a superficial depth of <1 cm and GB34 and ST36 were needled to an intermediate depth of about 1.5 cm. Stimulation of GB34 reproduced the left popliteal pain; the other points reproduced local pain. The level of pain evoked was moderate. There were no adverse events (AEs) during or immediately after treatment. The left knee pain decreased for 4–5 days after treatment, then recurred at the same intensity. The pain in the right knee resolved.
At his second appointment 3 weeks later the patient reported that he had had one of his vague staring episodes about 4 h after the first treatment. Given the time interval it was concluded that this was a coincidental event. The patient wanted to continue treatment.
On examination, myofascial tender points were located at GB34, the lateral proximal and distal joint margins and the proximal and distal lateral ligaments of the left knee. ST34, ST36, SP9 and SP10 were selected on an empirical basis. Mild stimulation to the same depth as before was applied to this group of nine needles for 5–10 min in total. The first group of needles was removed and the patient was turned onto his right side. A myofascial trigger point was located in the lateral head of the left gastrocnemius muscle; mild stimulation of this point to an intermediate depth was performed. Soon after needle insertion the patient complained of feeling strange, so the needle was removed and he was returned to a supine semirecumbent position. He complained of marked heaviness affecting both legs, then developed severe muscle spasm in both thighs, maximal on the left. As this resolved he reported a tingly ice-cold sensation initially affecting the toes of the right foot, then both feet. Next he lost muscle power in both legs, initially on the right, then both arms and flaccid paralysis with areflexia were noted. The plantar responses were neutral. There was no facial pallor, sweating or hyperventilation but he did experience transient nausea. Blood pressure and pulse rate were normal and remained so throughout the event. He then complained of being unable to move his eyes, which was the only similarity with his usual vague staring episodes, and coordinated rapid eye movements (REMs) around a central vertical axis were observed. Next he appeared to fall asleep with his eyes closed and a somnolent breathing pattern; his head slumped forward onto his chest. After a few minutes the patient was rousable. On ‘waking’ he was able to speak normally and gave an accurate account of what had been said and done during that period, but that he had been unable to speak or communicate in any way. His sensory symptoms subsequently recurred followed by moderate muscle spasm affecting both thighs, then pain affecting the left ankle. He regained power in his arms after 30 min and his legs after 45 min from the start of the AE. Rapid restoration of power along with normal tendon reflexes and bilateral flexor plantar responses occurred and the patient felt back to normal. The episode lasted for about 50 min in total.
The patient's knee pain resolved and has not recurred during the 2.5 years since acupuncture treatment; he is pleased with this outcome. His neurologist concluded that the episode after the first treatment had been a CPS and the AE during the second acupuncture treatment was similar to the EEG event (UE2, table 1). The patient experienced an increased frequency of sporadic clusters of CPS after acupuncture treatment which responded to modification of his medication, as well as further UEs (table 1).
He has had to surrender his driving licence so would not wish to receive acupuncture treatment again. He has been seizure-free, including vague staring episodes, for the past 6 months on his current medication, which consists of lamotrigine 300 mg/day and sodium valproate 1200 mg/day, and is enjoying improved levels of health and quality of life.
Impairment of consciousness is a central feature of CPS and is defined by the inability to respond to exogenous stimuli by virtue of altered awareness (loss of contact with events during the seizure and subsequent loss of recall) or responsiveness (inability to carry out simple commands or willed movement).2
Detailed review of past history
Since the onset of epilepsy in 1989, the patient had exhibited progressively more varied and atypical seizures, especially during the past 8 years. The past history is illustrated graphically in figure 1. When CPS became more prominent in 1999, the aura and seizures were stereotypical. In 2004 the pattern of seizures began to change. He had two episodes of CPS-associated asomatognosia, one of which resulted in a road traffic accident. In 2005 the patient developed blank staring episodes accompanied by facial pallor which were attributed to presyncope. A few months prior to acupuncture his GP recorded that these episodes were subtly altering. Following acupuncture the frequency and nature of his attacks changed again with increasing negative emotion, one episode of directed physical aggression and a complex directed motor automatism on another occasion; sometimes he dropped objects he was holding. It is thought that these episodes are CPS but could be psychogenic.
This patient poses a difficult diagnostic challenge. EEGs have shown a definite right frontal lobe focus on two occasions and a right temporal lobe focus on another separate occasion; CT and MRI brain scans have been normal. Table 2 lists a summary of the recognised features of CPS arising from the frontal and temporal lobes and the salient features of the other proposed diagnoses. Symptoms exhibited by the patient are shown in bold type. This summary suggests that he has CPS originating from the frontal and temporal lobes, with frontal lobe features becoming more prominent over recent years.
The convulsive seizures occasionally witnessed during acupuncture have been attributed to reflex anoxic seizures.5 Rosted reported repeated epileptic fits in a post-stroke patient receiving rehabilitative sensory stimulation in the form of transcutaneous electrical nerve stimulation; causality was not proven.6 In a recent review of AE reports in the Chinese literature, one case of pyknolepsy (dense clustering of seizures usually applied to absence seizures) was reported; the patient recovered and causality was not established.7
The British Medical Acupuncture Society (BMAS) recommends caution when treating patients with poorly-controlled epilepsy, that patients with epilepsy should be observed throughout treatment and that electroacupuncture should be used with particular caution.8 The WHO advises that all patients with a history of convulsions should be observed carefully during treatment.9
Sleep is a powerful modulator of epilepsy and treatment of parasomnias (sleep disorders characterised by abnormal behavioural and physiological events occurring at different sleep stages) can improve control of epilepsy as well as improving quality of life. Rapid eye movement sleep behaviour disorder (RBD) is defined as motor dyscontrol due to loss of REM sleep-related muscle atonia and is characterised by complex physical dream-enacting behaviours during REM sleep.10 The patient had a history of one episode of physical dream enactment behaviour and sporadic coherent nocturnal vocalisation with a fixed time relationship (03.00 h) when REM sleep is maximal; this suggests he may have a comorbid RBD. Dream enactment behaviour can be restricted to vocalisation.11 The vocalisation associated with nocturnal frontal lobe epilepsy is not coherent and has no fixed time relationship.10
Epileptic seizures are facilitated by non-REM sleep but are inhibited by REM sleep, leading to the hypothesis that RBD increases the likelihood of seizures.12 The fact that the patient used regularly to get a cluster of three CPS the day following nocturnal vocalisation supports this hypothesis. He is still experiencing sporadic nocturnal vocalisation although his seizures are now controlled by his current medication. This supports the existence of comorbid RBD.
RBD, narcolepsy (disturbed nocturnal and abnormal daytime sleep pattern with rapid onset REM sleep) and cataplexy (sudden episode of muscle weakness usually triggered by emotions during wakefulness and compatible with REM sleep atonia, rarely stiffening of the muscles can occur) tend to coexist.13 It is possible that the AE described was due to narcolepsy/cataplexy. The rapid coordinated eye movements observed before the onset of apparent sleep are suggestive of narcolepsy. Although the maximal duration of atonia associated with cataplexy is normally a few minutes, in ‘status catalepticus’ continual cataplectic episodes can last for several hours.14 The rapid restoration of flexor plantar responses after the AE suggests a non-epileptiform aetiology. Generalised atonic seizures associated with partial epilepsy are uncommon but have been reported in association with frontal and parietal lobe partial epilepsy, although muscle atonia normally lasts for only a few seconds.15 The prodromal somatosensory symptoms are more consistent with an epileptiform event.
Functional brain scanning has shown that acupuncture deactivates multiple limbic areas and activates the somatosensory cortex.16 A hypothetical neuronal network underlying RBD/narcolepsy/cataplexy involves neural substrates shared with acupuncture.11 A positron emission tomographic study has demonstrated shared neural substrates in subjects with narcolepsy/cataplexy.17 Intravenous procaine activates limbic system structures and produces sensations similar to CPS.18 Clinical experience suggests that acupuncture can increase as well as decrease activity in the anterior limbic area, especially in the presence of brain pathology.16 Patients with CPS often have damage to limbic structures.2
‘Out of body’ experiences are thought to reflect a failure by the brain to integrate complex somatosensory and vestibular information and have been evoked by direct electrical stimulation of the cortex in the region of the right angular gyrus, but distant from the epileptic focus, in a patient with CPS.19 Another patient with CPS experienced an ‘out of body’ experience while receiving acupuncture.16 The symptoms the patient experienced in UE4 (table 1) indicate that he experienced difficulty integrating somatosensory and vestibular information.
This patient has a complex combination of CPS, probable RBD, possible narcolepsy/cataplexy and possible dysfunctional vestibular/somatosensory processing. The AE might have resulted from activation by acupuncture of the neural substrates shared by the patient's comorbid conditions.
Similar mechanisms could underlie the UEs occasionally associated with acupuncture therapy: altered consciousness during and persisting after treatment (Adrian White, personal communication, 2010); transient aphonia7 ,20; and a tonic-clonic seizure sustained by a patient with post-traumatic stress disorder (PTSD) whose original diagnosis had been myoclonic epilepsy in whom the PTSD improved after acupuncture (Mark Andrews, personal communication, 2011).
Acupuncture treatment compliant with BMAS and WHO guidelines and tailored to the patient's anticipated strong responder status was provided. Given the information that was available at the time, there were no grounds for declining treatment which indeed proved to be very effective. Causality between the delayed first AE and acupuncture treatment cannot be definitely established. This AE was described as being like the vague staring episodes which had a non-epileptic diagnosis. Continuation of therapy was therefore consistent with safe acupuncture practice. Clinical observation supports causality in the case of the second AE. The temporal relationship between acupuncture treatment and the increased frequency of CPS supports causality and indicates similarities between this case and the one reported in the Chinese literature.
As a result of the additional information and knowledge gained from researching this case, the acupuncturist would decline to treat a similar patient with acupuncture unless the possession of a driving licence and employment considerations were not of concern to the patient. Extreme cataplectic attacks can involve all striated muscles excluding the diaphragm,14 so respiratory compromise is possible. The acupuncturist would therefore only wish to treat a similar patient in a clinical environment with access to basic cardiopulmonary resuscitation (CPR) equipment and other CPR-trained clinical staff.
Although this case was alarming, it needs to be put in context with other medical procedures. In a prospective study of AEs associated with acupuncture consultations, a rate of convulsive reflex anoxic seizures of 1/32 000 was reported.5 A review of AEs associated with blood donation reported a rate of ‘syncopations convulsivantes’ of 1/11 873.21
This case study was completed in part fulfilment of the MSc in Western Medical Acupuncture at the University of Hertfordshire. The excellent online literature searching facilities provided for distance-learning students enabled me to gain unrestricted access to high-quality international literature. My thanks go to the patient and his wife for their altruistic stance and their ready cooperation with the production of this case report. My thanks also go to all the members of Trevithick Surgery who facilitated detailed assessment and reporting of this adverse event.
Competing interests None.
Patient consent Obtained.
Provenance and peer review Not commissioned; internally peer reviewed.
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.