Abstract
A 31-year-old man with acute-onset of left-sided weakness following the sniffing of cocaine was admitted with rhabdomyolysis. Neurophysiological studies showed axonal degeneration in 4/10 sensory and 3/8 motor nerves, and conduction block outside the typical compression-sites in 3/8 motor nerves. The findings are consistent with a diagnosis of multiple mononeuropathy. Ischaemia due to vasoconstriction is currently believed to be the cause of muscle necrosis following cocaine abuse and we hypothesise that it also explains the neuropathy in this case.
BACKGROUND
Cocaine abuse may provoke haemorrhagic and ischaemic strokes, symptomatic seizures, transient movement disorders and acute severe headaches. Peripheral nerve damage has only been associated with compression or intravenous administration of cocaine. It was suggested that vasoconstriction was responsible for stroke, headache and rhabdomyolysis.1,2
CASE PRESENTATION
A 31-year-old, previously healthy man, sniffed cocaine (2 runs of 1000 mg) 2 days prior to admission. The patient felt that the effect was much stronger than he had experienced previously. He had occasionally sniffed cocaine but never administered intravenously. After 6–7 hours sleep he noticed weakness and paraesthesia in the left shoulder and lower limb, and sought medical help 1 day later when his symptoms were unchanged. On admission the patient was alert and cooperative. Examination of power (Medical Research Council (MRC) scale) revealed weakness of the left upper and lower limbs and of the right foot: left elbow flexion (0); wrist and finger extension (0); intrinsic hand muscles (4); and the other muscles of the left upper limb (2–3). He had weak left hip and knee flexion and extension (4), and dorsal and plantar-flexion of the foot (3). On the right side there was weakness of toe extension. Muscle atrophy was not detected. The patient had sensory loss on the lateral side of the left arm, forearm and hand, and on the left leg and both feet (table 1). Biceps, brachioradial and Achilles reflexes could not be elicited on the left side. The other deep tendon reflexes were normal. The plantar response was flexor on the right and absent on the left side. The remainder of the physical examination was unremarkable. During the 2 weeks of admission the proximal weakness in the left lower limb recovered, but the remaining signs were unchanged.
Table 1.
Motor deficit and sensory loss associated with the affected nerves
| Nerve | Motor deficit | Sensory loss |
| Left musculocutaneous | +++ | +++ |
| Left radial | +++ | +++ |
| Left median | ++ | ++ |
| Left ulnar | ++ | − |
| Left femoral | + | − |
| Left peroneal | ++ | ++ |
| Left tibial | ++ | ++ |
| Right peroneal | + | ++ |
LABORATORY INVESTIGATIONS
MR scan of the brain was normal, on admission and also at re-examination 10 months later. MR of the spinal cord from the foramen magnum to the 5th thoracic vertebra was normal on the third day after admission.
Plasma myoglobin 1049 μg/l (normal: 28–72), aspartate aminotransferase 1424 IU/l (normal: 15–45), alanine aminotransferase 902 IU/l (normal: 10–70), lactic dehydrogenase 902 IU/l (normal: 105–205) and creatinine kinase 18980 IU/l (normal: 50–400) were greatly elevated. These findings were interpreted to indicate rhabdomyolysis and forced diuresis was started. Renal function remained normal. Negative tests were found for antinuclear antibodies, antineutrophyl cellular antigens, screens for hepatitis A, B and C, HIV, cytomeglavirus and Borrelia antibodies as well as blood and cerebrospinal fluid (CSF) cultures. Tests for Epstein-Barr virus IgG was positive and IgM negative. The CSF on first day showed normal protein concentration, 1 leukocyte per mm3, and no erythrocytes. The abnormal laboratory values gradually returned to normal within two weeks.
NEUROPHYSIOLOGICAL INVESTIGATIONS
Eighteen days after the onset of the symptoms, concentric needle electrode electromyography (ECG) showed spontaneous activity (fibrillations, positive sharp waves) in the left deltoid, biceps, gastrocnemius, tibialis anterior and bilaterally in the extensor digitorum communis muscles. Voluntary activity could not be recorded in the left biceps and extensor digitorum communis. Interference pattern was reduced in the left deltoid, tibialis anterior, gastrocnemius and in the right extensor digitorum communis muscles.
Motor conduction studies (table 2) were carried out 18 days after the onset. The amplitudes of the compound muscle action potentials at distal stimulation were decreased in the left tibial nerve and bilaterally in the peroneal nerves. Partial motor conduction blocks were found in the left median, right ulnar and left tibial nerves (Figure 1). Normal motor conduction was found in the right median and tibial, and left ulnar nerves. The presence of conduction block in the left median nerve was verified 3 weeks later by a second examination. The ulnar conduction block was at that time abolished.
Table 2.
Motor nerve conduction studies
| July 2005 | ||||||
| Nerve | Side | Amplitude at distal stimulation (mV) | Conduction block | Conduction velocity (m/s) | Distal latency (ms) | F-wave latency (ms) |
| Median | Left | 9.3 | 56% (elbow-wrist) | 48 | 3.1 | 29.0 |
| Ulnar | Right | 17.3 | 68% (Erb’s point – axilla) | 60 (below elbow-wrist) | 2.8 | 30.9 (2.2 SD) |
| 52 (above -below elbow) | ||||||
| 74 (axilla-above elbow) | ||||||
| 44 (−33.4 SD) (Erb’s point-axilla) | ||||||
| Peroneal | Left | 3.0 (−2.5 SD) | − | 43 (below knee-ankle) | 4.1 | 59.8 (2.7 SD) |
| 26 (−3.9SD) (across the knee) | ||||||
| Peroneal | Right | 0.3 (−8.9 SD) | − | 35 (−3.4 SD) (knee-ankle) | 8.5 (6.5 SD) | absent |
| 39 (across the knee) | ||||||
| Tibial | Left | 5.1 (−3.2 SD) | 47% (knee-ankle) | 41 | 5.2 (2.1 SD) | absent |
| March 2006 | ||||||
| Median | Left | 16.5 | − | 54 | 3 | 30.1 |
| Ulnar | Right | 11 | − | 68 (below elbow-wrist) | 2.4 | 27.8 |
| 49 (above- below elbow) | ||||||
| 64 (axilla-above elbow) | ||||||
| 64 (Erb’s point-axilla) | ||||||
| Peroneal | Left | 5.9 | − | 38 (−2.5 SD) (knee-ankle) | 4.4 | 56.3 (2.7 SD) |
| 46 (across the knee) | ||||||
| Tibial | Left | 1.1 (−6.7 SD) | − | 36.4(−3.1 SD) (knee-ankle) | 4.7 (2.1 SD) | absent |
Only nerves with pathological findings are shown. Our normal values depend on the age and height of the patient. Thus deviation from normal is given in standard deviations (SD). Amplitudes were measured peak-to-peak.
Figure 1.
Motor nerve conduction recordings at 18 days after the acute onset (A) and 8 months later (B). Stimulating the median nerve at wrist (upper trace) and elbow (lower trace). Recording from abductor pollicis brevis muscle. Observe the improvement in the amplitude at the proximal stimulation (B). Scale: 10 mV and 3 ms, respectively.
Sensory potentials (table 3) could not be recorded in the left radial and lateral cutaneous nerve of the forearm. The amplitudes of the sensory potentials were reduced in the left median and sural nerves. Normal sensory conduction was recorded in the right median, radial, lateral cutaneous nerve of the forearm, sural nerve, and bilaterally in the ulnar nerve.
Table 3.
Sensory nerve conduction studies
| July 2005 | |||
| Nerve | Side | Amplitude (μV) | Conduction velocity (m/s) |
| Median | Left | 2.0 (−5.5 SD) | 63.8 |
| Cutaneous antebrachii lateralis | Left | Absent | |
| Radial | Left | Absent | |
| Sural | Left | 1.0 (−6.6 SD) | 44 (−2.1 SD) |
| March 2006 | |||
| Median | Left | 1.3 (−6.6 SD) | 60.7 |
| Cutaneous antebrachii lateralis | Left | absent | |
| Radial | Left | absent | |
| Sural | Left | 0.4 (−8.6 SD) | 43 (−2.5 SD) |
Only nerves with pathological findings are shown. Deviation from normal is given in standard deviations (SD).
Motor evoked potentials (MEPs) recorded 1 week after the onset were normal. The median and tibial nerve sensory evoked potentials (SEPs) showed prolonged peripheral latencies from left arm and leg. However, there was normal central conduction time from the arms and right leg, while the lumbar spinal potential after stimulation of left leg could not be identified.
DIFFERENTIAL DIAGNOSIS
The overt left-sided symptoms and signs gave rise to a suspicion of central nervous system lesion. This could not be verified by MR scans, MEP or SEP. The neurophysiological studies demonstrated that the clinical deficits were associated with involvement of peripheral nerves. The asymmetrical combination of normal and abnormal neurophysiological findings suggested multiple mononeuropathy.
OUTCOME AND FOLLOW-UP
A follow-up neurological examination 8 months later revealed further partial recovery. There was still weakness in left elbow-flexion (1–2), wrist and finger extension (0), dorsal and plantar-flexion of the left foot (4) and extension of the right toes (4). The sensory loss had the same distribution except for normalisation on the left arm.
The follow-up EMG 8 months after the onset of the symptoms showed spontaneous activity in the left biceps and extensor digitorum communis muscles. Quantitative EMG showed polyphasic motor unit-potentials with prolonged duration and slightly reduced amplitude (typical for reinnervation) in the left biceps muscle.
Eight months after the onset of the symptoms, the motor nerve conduction was completely normal in the left median nerve while signs of axonal loss were found in the left tibial nerve (Figure 1).
The abnormal sensory conduction was unchanged.
DISCUSSION
We found partial conduction blocks in three nerves outside the typical locations of compression neuropathies. In two of these nerves (left median and right ulnar) the amplitude of the compound muscle action potential was normal at the distal stimulation. The conduction block cannot be explained by Wallerian degeneration, since it was detected 18 days after the acute onset and it was present in the median nerve at the repeated examination 3 weeks later. Eight months after the onset of the symptoms, when power in the left abductor pollicis brevis was restored, the conduction block could not be detected. Conduction block is generally considered to be a sign of demyelination, as shown in studies with parallel neurophysiological and morphological investigations.3 Although ischaemia is more commonly associated with fascicular axon loss, it has been shown that reperfusion after ischaemia can result in demyelination and conduction block.4–7 Cocaine blocks the presynaptic reuptake of norepinephrine and dopamine, and produces vasoconstriction. However, this is transient, and there is reperfusion when the effect of the drug fades away.
Unlike polymyositis, the EMG findings of acute rhabdomyolysis are not well-defined. A previous study showed that in patients with rhabdomyolysis only 16% of the muscles had myopathic motor unit action potentials in the proximal muscle when examined within 2 weeks after the acute onset.8 We did not find myopathic action potentials. However, the first EMG was done 18 days after the acute onset.
An alternative explanation one has to take into consideration is that vasculitis or direct toxicity due to an unusual adulterant caused the multiple mononeuropathy and/or the rhabdomyolysis.9 Against this possibility speaks the administration of cocaine in this patient by sniffing.
LEARNING POINTS
A peripheral nerve disorder should be suspected in cocaine abusers even if the presenting symptoms are lateralised.
In the acute phase of the multiple mononeuropathy conduction block and axonal degeneration may coexist.
Footnotes
Competing interests: none.
Patient consent: Patient/guardian consent was obtained for publication.
REFERENCES
- 1.Neimann J, Haapaniemi HM, Hillbom M. Neurological complications of drug abuse: pathophysiological mechanisms. Eur J Neurol 2000; 7: 595–606 [DOI] [PubMed] [Google Scholar]
- 2.Nolte KB. Rhabdomyolysis associated with cocaine abuse. Hum Pathol 1991; 22: 1141–45 [DOI] [PubMed] [Google Scholar]
- 3.Fowler CJ, Gilliatt RW. Conduction velocity and conduction block after experimental ischaemic nerve injury. J Neurol Sci 1981; 52: 221–38 [DOI] [PubMed] [Google Scholar]
- 4.Cappelen-Smith C, Lin CS, Kuwabara S, et al. Conduction block during and after ischaemia in chronic inflammatory demyelinating polyneuropathy. Brain 2002; 125: 1850–8 [DOI] [PubMed] [Google Scholar]
- 5.Homberg V, Reiners K, Toyka KV. Reversible conduction block in human ischemic neuropathy after ergotamine abuse. Muscle Nerve 1992; 15: 467–70 [DOI] [PubMed] [Google Scholar]
- 6.Kaku DA, Malamut RI, Frey DJ, et al. Conduction block as an early sign of reversible injury in ischemic monomelic neuropathy. Neurology 1993; 43: 1126–30 [DOI] [PubMed] [Google Scholar]
- 7.Nukada H, McMorran PD, Shimizu J. Acute inflammatory demyelination in reperfusion nerve injury. Ann Neurol 2000; 47: 71–9 [PubMed] [Google Scholar]
- 8.Al-Shekhlee A, Hachwi R, Jaberi MM, et al. The electromyographic features of acute rhabdomyolysis. J Clin Neuromuscul Dis 2005; 6:114–8 [DOI] [PubMed] [Google Scholar]
- 9.Richards JR. Rhabdomyolysis and drugs of abuse. J Emerg Med 2000; 19: 51–6 [DOI] [PubMed] [Google Scholar]
