Abstract
Heroin-related myelopathy is an uncommon but often devastating complication of heroin intake. It is usually reported in individuals exposed to intravenous heroin after a variable drug-free period, leading to acute and complete spinal cord injury with poor long-term outcome. We describe an original case of acute longitudinally extensive transverse myelopathy following single heroin and cocaine intravenous exposure after a long period of abstinence confirmed by toxicological hair and retrospective urine drug analysis. This case could provide new insights in the understanding of this rare neurological complication.
Keywords: spinal cord, neuroimaging, drug misuse (including addiction)
Background
Acute transverse myelopathy is a potentially devastating condition associated with numerous aetiological factors; demyelination, infections and inflammatory, vascular, toxic or neoplastic/paraneoplastic diseases should be considered in the differential diagnosis.
Acute transverse myelopathy has been reported as an uncommon complication of heroin or cocaine intake. In this context, it usually has an abrupt onset, catastrophic presentation and poor outcome leading to severe long-term disability.1–9
On the contrary, myelopathies due to HIV or Human T-cell lymphotropic virus (HTLV) infection develop gradually and have a subacute or chronic clinical course.
The purpose of this report is to describe a unique case of acute extensive transverse myelopathy in association with single heroin and cocaine intravenous exposure after a long period of abstinence confirmed by toxicological hair analysis in a patient with coincidental HIV and HTLV-2 infection.
Case presentation
A white Caucasian 47-year-old woman was admitted to the emergency department of our hospital after being found in a public toilet unconscious and unresponsive.
Her vital signs were: blood pressure 90/55 mm Hg, heart rate 72 p/min, respiratory rate 10 b/min, body temperature 36°C and oxygen saturation 96%.
She had a history of HIV infection clinical stage A3, chronic hepatitis C infection and intravenous heroin drug abuse, for which she was being treated with methadone.
On the assumption of drug intoxication, naloxone and flumazenil were administered with progressive recovery of consciousness followed by psychomotor agitation with abrupt four limb movements requiring mechanical restraint and the administration of 5 mg of intramuscular haloperidol.
After awakening, couple of hours later, she complained of back pain, weakness and numbness of both legs and the perineal area. She reported abstinence from heroin and cocaine abuse for more than 1 year and confirmed that she had stopped antiretroviral treatment voluntarily since 1 year.
On physical examination, there was flaccid paraplegia and areflexia, with sensory level at T10. She had a distended bladder with urinary retention and anal sphincter atony.
Investigations
Since an acute spinal cord syndrome was suspected, an urgent spinal cord MRI was performed, observing extensive transverse T2-weighted signal hyperintensity affecting the spinal cord from thoracic segment T6 to the conus terminalis, associated with gadolinium contrast enhancement (figure 1A–C).
Figure 1.
First MRI study. (A) Sagittal T2-weighted image showing extensive T2 signal hyperintensity of thoracic and lumbar spinal cord (white arrows) associated with swelling predominantly affecting the lumbar segments (black arrow). (B) Homogeneous distribution of T2 signal abnormality on transverse T2-weighted image affecting entire thickness of the spinal cord (white arrows). (C) Transverse gadolinium-enhanced T1-weighted image showing diffuse predominantly anterior and intramedullary enhancement (white arrow).
Toxicology results in urine sample at admission are detailed in table 1.
Table 1.
Qualitative and quantitative toxicology results in urine sample at admission
| Drug screen* | Result† | Drug confirmation‡ | Result, ng/mL (ng/mg creatinine) |
| Cocaine | Positive | Benzoylecgonine | 62 166 (55 704) |
| Ecgonine methyl ester | Positive | ||
| Opiates | Positive | Morphine | 18 513 (16 589) |
| 6- monoacetylmorphine | Positive | ||
| Codeine | Positive | ||
| Methadone | Positive | Methadone | Positive |
| EDDP§ | Positive | ||
| Cannabinoids | Positive | THC-COOH¶ | 248 (223) |
*Immunoassay drug screening was performed on an Architect 16 000 automated analyser (Abbott Diagnostics, Santa Clara, California, USA).
†The cut-offs (ng/mL) for the qualitative applications were: cocaine: 150; opiates: 300, methadone: 300; cannabinoids: 50.
‡Gas chromatography–mass spectrometry analyses were carried out on Agilent HP 7890A GC coupled with an Agilent MSD 5975 C MS (Agilent Technologies, Santa Clara, California, USA).
§2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine.
¶11-nor-9-carboxy-Δ 9 -tetrahydrocannnabinol (THC-COOH).
Lumbar puncture was carried out. The cerebrospinal fluid (CSF) was transparent, and contained 50.9 mg/dL proteins, 59 mg/dL glucose and 37 leucocytes (85% lymphocytes). Routine CSF cultures were negative and pathological analysis did not reveal cellular atypia. Prognostic markers of HIV were as follows: viral load 22 580 copies/ml and CD4 +47; therefore, antiretroviral therapy was started again. Anti HTLV-2 antibodies were detected by Chemiluminescence immunoassay (CLIA) test and confirmed later by western blot testing.
The rest of the ancillary tests performed during hospitalisation, including blood analysis for liver and kidney function, autoimmunity test, chest radiography and ECG, were unremarkable.
A 20 cm long segment of her hair was sent for toxicological analysis. Segmental analysis of four hair segments, 3 cm each, was carried out. Given the length of the patient’s hair, abstinence to cocaine and heroin was confirmed for at least the previous 12 months.
Control MRI was performed 9 months later, showing atrophy of the distal thoracic and lumbar segments of the spinal cord, where signal abnormality selectively affected both anterior horns (figure 2A–C).
Figure 2.
Control MRI study. (A) Sagittal T2-weighted image showing resolution of the signal alteration in upper thoracic segments appreciating decreased volume distally from T10 to the conus terminalis suggestive of atrophy (white arrows). (B) Signal abnormality on control transverse T2-weighted image affecting both anterior horns symmetrically (white arrows). (C) Transverse T2-weighted image. Note the absence of signal abnormality in the upper thoracic segments.
Differential diagnosis
Spinal cord infarction: suggested as an underlying mechanism of spinal cord injury in patients with heroin myelopathy. Improbable for several reasons: the spinal cord is relatively resistant to ischaemia in comparison with more susceptible brain watershed regions, which were unaffected in our case; lymphocytic pleocytosis, early spinal cord signal abnormality and contrast enhancement in MRI did not support this diagnosis.
HIV or HTLV myelopathy: usually occurs in the late stages of AIDS, showing a slowly progressive clinical course with spastic paraparesis as a dominant manifestation. MRI is commonly normal.
Neuromyelitis optica spectrum disorders: should be considered in differential diagnosis of longitudinally extensive myelopathies especially in the case of negative toxicological analysis. It usually presents with bilateral optic neuritis with poor visual outcome or transverse myelitis with a relapsing course. The onset is acute but usually develops more insidiously over days.
Treatment
During hospitalisation, 1000 mg of intravenous methylprednisolone for five consecutive days followed by 1 mg/kg/day oral corticosteroids with tapering and early rehabilitation therapy were initiated.
Outcome and follow-up
At the time of discharge, 1 month after admission, there was no substantial improvement in the patient’s neurological status. Six months later she was unable to walk and presented with sphincter incontinence.
Discussion
It has been a half century since Richter and Rosenberg published an original paper reporting four cases of acute heroin-related myelopathy; since then, this condition has been recognised and reported by several other authors.1–10
Interestingly, most of the patient reports resemble each other: all except one6 used intravenous heroin; almost all consumed the drug after a variable period of abstinence;1 3–8 the neurological picture was established within hours, manifesting as complete transverse myelitis or spinal shock with a predilection for the thoracic segments of the spinal cord;1–3 5–7 and most of them had poor outcome with long-term residual disability, or even death.1
Most reports describe normal CSF findings on routine analysis,1–3 6 8 although pleocytosis and increased total protein may occur.1 5 7
Since the initial reports, little progress has been made in understanding the pathophysiological process, so this still remains a rare and poorly understood condition. Direct toxic effects, prolonged periods of hypotension, hypoxia, compression or immunological effects of the drug or its contaminant have been suggested as potential injury mechanisms.
Allergic or hypersensitivity reactions have been among the dominant theories.1 3 5–8 This is supported by the fact that the majority of patients develop acute myelopathy in relation to single intravenous exposure to heroin after a variable drug-free period.
Some authors hypothesise a sensitisation of spinal cord parenchyma via haptenation of heroin with an unknown protein that leads to neurotoxic immune-mediated injury after heroin exposure.6
The vascular process is another commonly suggested mechanism.2 3 5 The spinal cord has an extensive collateral blood supply that makes it relatively resistant to ischaemia, but heroin overdose frequently provokes respiratory depression followed by hypotension, and vascular injury can be further promoted due to the prolonged hyperextended position of the neck seen in some cases.
Furthermore, our patient consumed cocaine simultaneously, which can lead to vasospasm or cardioembolism as the underling mechanisms for spinal cord infarction.
Although not completely disposable, the vascular theory does not explain why these patients do not usually develop watershed brain infarctions.
The predisposition for thoracic segments being affected seen in most reported cases can be explained by less adequate blood supply of this area of the spinal cord.
Our patient had extensive spinal cord involvement that vastly exceeded the thoracic segment, affecting the spinal cord in a continuous and transverse manner from T6 all the way up to the conus medullaris.
Selective anterior horn involvement observed in control MRI has been reported previously in one case following intranasal insufflation of heroin and ecstasy.10 This pattern, known as ‘snake eye’ or ‘owl’s eye’, is not specific for spinal cord ischaemic lesions and can be seen in other diseases which selectively or dominantly affect the motoneurons in the anterior horns, such as spinal muscle atrophy or viral ‘polio-like’ infection.11
Additional neuroimaging observations do not support this possibility: unlike our patient, MRI findings are usually normal within the first hours after onset in cases of spinal cord infarction and enhancement is often absent at this stage.12 13 Moreover, ischaemia does not usually produce lymphocytic pleocytosis, as seen in our case.13
For these reasons, we consider the vascular process and consequent cocaine toxicity unlikely to be the pathophysiological mechanism of spinal cord injury in our patient.
The hyperacute onset, poor evolution and the extensive affection of the spinal cord also excluded demyelinating disease, vitamin deficiency, HIV-associated vacuolar myelopathy or HTLV-associated myelopathy, as the principal cause of the myelopathy. We consider the HIV and HTLV-2 detected in our patient to be incidental given the fact that co-infection by these viruses in Europe and USA are mainly reported among parenteral drug users.
It is of important clinical interest to confirm or exclude drug intake and define its temporal relation with the onset of the neurological disorder, in order to establish appropriate diagnosis and avoid unnecessary testing. For this purpose, toxicological testing for drugs of abuse in hair has become a helpful diagnostic tool in determining recent past drug use or long-term drug history and has many potential clinical and forensic toxicological applications. The examination is carried out by multisectional analysis of the hair, and the identification and quantification of drugs in different hair segments which correspond to different periods of time.14
Hair samples are useful for drug testing since drugs can be identified for longer periods than in blood, urine or other biological samples.
Retrospectively, we reviewed urine drug results that were performed in a drug treatment centre during the last 20 months. The results confirmed refrainment from illicit drugs and adherence to methadone maintenance treatment. In our case, toxicological hair and urine analyses were complementary to her testimony of drug abstinence in the last year, therefore excluding the possibility of a cumulative toxic effect as a potential cause of the neurological disorder.
We would like to highlight the originality of this paper, which, according to our knowledge, is the first case of acute heroin-related myelopathy in which toxicological hair analysis was used to corroborate long-term drug abstinence, making the hypothesis of spinal neurotoxic immune-mediated injury a credible mechanism. Further reports and studies are necessary to establish a more consistent causal relation and to improve our understanding of this rare neurological complication.
Learning points.
Heroin myelopathy is an uncommon but potentially catastrophic complication of heroin abuse with a poor outcome.
It usually occurs following single heroin exposure after a period of abstinence.
The pathophysiology is unknown, although many authors suggest hypersensitive immune-mediated spinal injury as the main underlying pathophysiological mechanism.
Clinically, it manifests as an acute spinal shock syndrome and radiologically as longitudinally extensive transverse myelopathy.
Toxicological urine and hair analysis for drugs of abuse are helpful diagnostic tools in determining recent or past drug use or long-term drug history.
Footnotes
Contributors: TI was responsible for patient management during hospitalisation, performed literature search, most part of editing and coordination. AMEI was responsible for critical review and correction of the work in progress and the final manuscript. BBM and IGM conducted hair toxicological analysis, reviewed previous available drug analysis and wrote the toxicological part of the manuscript.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Obtained.
References
- 1. Richter RW, Rosenberg RN. Transverse myelitis associated with heroin addiction. JAMA 1968;206:1255–7. 10.1001/jama.1968.03150060029005 [DOI] [PubMed] [Google Scholar]
- 2. Ell JJ, Uttley D, Silver JR. Acute myelopathy in association with heroin addiction. J Neurol Neurosurg Psychiatry 1981;44:448–50. 10.1136/jnnp.44.5.448 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Goodhart LC, Loizou LA, Anderson M. Heroin myelopathy. J Neurol Neurosurg Psychiatry 1982;45:562–3. 10.1136/jnnp.45.6.562 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Grassa C, Montanari E, Scaglioni A, et al. Acute heroin myelopathy-case report. Ital J Neurol Sci 1984;5:63–6. 10.1007/BF02043972 [DOI] [PubMed] [Google Scholar]
- 5. Malik MM, Woolsey RM. Acute myelopathy following intravenous heroin: a case report. J Am Paraplegia Soc 1991;14:182–3. 10.1080/01952307.1991.11735851 [DOI] [PubMed] [Google Scholar]
- 6. McCreary M, Emerman C, Hanna J, et al. Acute myelopathy following intranasal insufflation of heroin: a case report. Neurology 2000;55:316–7. 10.1212/WNL.55.2.316-a [DOI] [PubMed] [Google Scholar]
- 7. Sveinsson O, Herrman L, Hietala MA. Heroin-induced acute myelopathy with extreme high levels of CSF glial fibrillar acidic protein indicating a toxic effect on astrocytes. BMJ Case Rep 2017;2017:bcr-2017-219903 10.1136/bcr-2017-219903 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Dujmovic I, Nikolic I, Martinovic V, et al. Heroin-induced acute longitudinally extensive transverse myelopathy. Neurol Sci 2018;39:791–2. 10.1007/s10072-017-3197-x [DOI] [PubMed] [Google Scholar]
- 9. Di Lazzaro V, Restuccia D, Oliviero A, et al. Ischaemic myelopathy associated with cocaine: clinical, neurophysiological, and neuroradiological features. J Neurol Neurosurg Psychiatry 1997;63:531–3. 10.1136/jnnp.63.4.531 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Riva N, Morana P, Cerri F, et al. Acute myelopathy selectively involving lumbar anterior horns following intranasal insufflation of ecstasy and heroin. J Neurol Neurosurg Psychiatry 2007;78:908–9. 10.1136/jnnp.2006.109207 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Weidauer S, Wagner M, Nichtweiß M. Magnetic resonance imaging and clinical features in acute and subacute myelopathies. Clin Neuroradiol 2017;27:417–33. 10.1007/s00062-017-0604-x [DOI] [PubMed] [Google Scholar]
- 12. Vargas MI, Gariani J, Sztajzel R, et al. Spinal cord ischemia: practical imaging tips, pearls, and pitfalls. AJNR Am J Neuroradiol 2015;36:825–30. 10.3174/ajnr.A4118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Mariano R, Flanagan EP, Weinshenker BG, et al. A practical approach to the diagnosis of spinal cord lesions. Pract Neurol 2018;18:187–200. 10.1136/practneurol-2017-001845 [DOI] [PubMed] [Google Scholar]
- 14. Cooper GA, Kronstrand R, Kintz P. Society of Hair Testing. Society of hair testing guidelines for drug testing in hair. Forensic Sci Int 2012;218:20–4. 10.1016/j.forsciint.2011.10.024 [DOI] [PubMed] [Google Scholar]


