Article Text

Download PDFPDF

Acute myelopathy selectively involving lumbar anterior horns following intranasal insufflation of ecstasy and heroin
  1. N Riva1,
  2. P Morana1,
  3. F Cerri1,
  4. S Gerevini2,
  5. S Amadio3,
  6. F Formaglio3,
  7. G Comi3,
  8. M Comola3,
  9. U Del Carro3
  1. 1Department of Neurology, Neuro-rehabilitation and Clinical Neurophysiology, Scientific Institute and University Ospedale San Raffaele, Milan, Italy
  2. 2Department of Neuroradiology, Scientific Institute and University Ospedale San Raffaele, Milan, Italy
  3. 3Department of Neurology, Neuro-rehabilitation and Clinical Neurophysiology, Scientific Institute and University Ospedale San Raffaele, Milan, Italy
  1. Correspondence to:
 Dr Nilo Riva
 Department of Neurology and Clinical Neurophysiology, San Raffaele Scientific Institute, Via Olgettina 60, 20132 Milano, Italy; riva.nilo{at}

Statistics from

We report a patient who developed acute myelopathy after intranasal insufflation of amphetamines and heroin. The functional prognosis was very poor; after 4 months, she remained paraplegic. MRI imaging showed selective T2 hyperintensity and intense enhancement confined to the spinal anterior horns and lumbar nerve roots and plexus. This unique MRI pattern, together with neurophysiological data, suggests that the pathological process at the first primary affected spinal anterior horns (SAH), conditioning motoneuron cell death, and then nerve roots and lumbar plexus as a consequence of wallerian degeneration

Case report

A 17-year-old girl was admitted to the emergency department in a drowsy state and unable to walk after an overdose of intranasal insufflated heroin and amphetamines. After a few hours, drowsiness progressed to stupor, and progressive weakness in all four limbs, mainly involving the lower limbs, developed. At that time, laboratory data showed massive rhabdomyolysis (creatine phosphokinase 36 880 mg/dl) with acute renal failure (ARF), and hepatic failure; medical therapy was promptly started. The patient’s past medical history was unremarkable except for habitual use of amphetamines (ecstasy) and cannabinoids since the age of 12 years. The previous week she had insufflated heroin about once a day; the previous night she reported a double dose of heroin consumption, and a high dose (approximately 1 g) of intranasal insufflation of amphetamines. The next day the patient was alert and cooperative but complained about diffuse muscle pain and tenderness, prevailing in the lower limbs, where weakness rapidly worsened to flaccid, areflexic paralysis.

Urinary chromatography detected amphetamines (3-4-methylendoxymethamphetamine (MDMA) 7680 ng/ml, 3-4-methylenedioxyamphetamine 2000 ng/ml) and opiates (>2000 ng/ml, morphine 228 ng/ml). CSF protein content was slightly increased with no intrathecal Ig synthesis; glucose, chlorine and cells were within the normal range. All virological analyses, including HIV, were negative. Rhabdomyolysis related ARF completely normalised by day 10, with resumption of spontaneous diuresis; from that time on, no more myalgias were reported. Muscle weakness promptly improved in the upper limbs to normal strength while flaccid paralysis persisted in the lower limbs. The patient had no sensory symptoms or signs, and no sphincteric abnormalities were observed.

Motor evoked potentials recorded from foot muscles were absent from day 9 onwards; both sural nerve sensory action potentials (SAPs) and tibial somatosensory evoked potentials (SSEPs) were normal. On day 16, electromyography revealed complete unexcitability of the leg motor nerves, together with fibrillation potentials without any voluntary electromyographic activity in all lower limbs and paraspinal muscles innervated by L3–S1 roots. Quantitative sensory test, skin sympathetic reflex and urodynamic test were all within the normal range.

MRI performed at the beginning of symptoms was negative. One month later, MRI showed selective T2 hyperintensity of the anterior horns and signal alteration of lumbar nerve roots, associated with volume increase and intense gadolinium enhancement (fig 1).

Figure 1

 MRI imaging. (A) Post-gadolinium (standard dose) T1 weighted MRI image, coronal view: linear enhancement along the nerve roots is evident. (B) T2 weighted MRI image, coronal plane: widespread T2 hyperintensity along nerve roots is more evident on the right side. (C) Post-gadolinium (standard dose) T1 weighted MRI image, axial view: selective anterior horn intense enhancement is seen. (D) T2 weighted MRI image, axial plane: selective anterior horn T2 hyperintensity is noted.

Intravenous high dose steroid therapy was started with no clinical benefit. MRI signal alteration and enhancement of the nerve roots almost completely disappeared while signal alteration of the anterior horns remained, although it was reduced in intensity. Four months later, the patient was still paraplegic.


Rhabdomyolysis, ARF, acute hepatotoxicity and transverse myelitis (TM) are well known complications of intravenous heroin assumption.1,2 A single case of acute myelopathy as a consequence of heroin inhalation has been reported following, in common with other cases, a period of abstinence preceding the event; moreover, recovery was almost complete 1 month later.1 No cases of acute MDMA myelopathy are known; rhabdomyolysis is known to complicate MDMA consumption.3

This case particularly highlights selective involvement of the lumbar motoneurons, as supported by neurophysiological and neuroradiological evidence. Unexcitability of motor fibres and complete denervation in the lower limbs and paraspinal muscles, with spared SAPs and SSEPs, suggest damage of SAH and/or ventral root axons. MRI confirmed selective T2 hyperintensity confined to the SAH and nerve roots, with intense enhancement in both, spreading from D10 up to the conus (fig 1). This unique MRI pattern has never been reported previously to our knowledge, and differs from MRI patterns observed in TM and radiculitis. In TM, MRI can sometimes show an inflammatory lesion localised within the cord, usually not involving more than 3–4 vertebral segments (more often at the dorsal level), hyperintense in T2 weighted images and with nerve root sparing. In the acute phase, enhancement is generally present, but not involving the nerve roots.4 In radiculitis (eg, Guillain–Barré syndrome), MRI signal alterations are commonly localised to spinal anterior roots, without any involvement of the spine.5 Moreover, the MRI pattern did not suggest a spinal infarction. The anterior spinal artery derives from a single anterior radiculo-medullary artery (Adamkiewicz artery) that arises from L3 upwards; spinal infarction is rare and occurs mostly in the cervical spine. MRI abnormalities on sagittal T2 weighted images in our patient are not consistent with typical “pencil-like” hyperintensities and cord enlargement.

Therefore, taken together, the clinical, neuroradiological and neurophysiological data suggest that the pathological process primary affected SAH conditioning motoneuron cell death, and then nerve roots and lumbar plexus as a consequence of wallerian degeneration.

Suggested pathogenetic mechanisms of heroin associated acute myelopathy include hypotension watershed zones ischaemia, vasculitis and hypersensitivity reaction. Among the most credited hypothesis is the possibility of a direct toxic effect of heroin.1,2 In our case, we speculate that a pathogenetic role played by MDMA cannot be excluded and may contribute towards an explanation of some of the atypical features such as the lack of a clear period of abstinence from heroin, as described in most cases of opiate toxic myelopathy, and the unusually high dose of MDMA inhaled. Moreover, it has been proved that the lumbar motoneurons express high levels of serotonin receptors (mediating MDMA effects) whereas opiates receptors are three times more represented in the dorsal horns compared with the ventral horns of the spinal cord.6

This case increases our knowledge of the possible severe acute toxic effects of intranasal insufflation of heroin and ecstasy, often considered by the public as “safe” drugs. The selective involvement of the anterior horns of the spinal cord indicates a poor functional prognosis and suggests a specific, but still unknown, pathogenetic effect on the spinal motoneurons, possibly toxic or immunopathological in nature.


View Abstract


  • Competing interests: None.

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.