JRSM >
Volume 96, Number 7 >
Pp. 351-352
Department of Neurology, University Hospitals of Leicester, Royal Infirmary, Leicester LE1 5WW, UK
E-mail: yusuf.rajabally{at}uhl-tr.nhs.uk
The most frequently encountered neuropathy in diabetes is slowly progressive. Sensory symptoms predominate, sometimes accompanied by autonomic involvement, and motor deficit is usually absent.
CASE HISTORY
A man of 65 was referred after rapid onset of weakness in the arms and legs. It was his second such episode. 8 years previously, when he was 57, non-insulin-dependent diabetes mellitus had been diagnosed, and a year after onset he had developed a mild axonal diabetic polyneuropathy, causing distal paraesthesiae. Insulin was required from 2 years after diagnosis. He was hypertensive, treated with a combination of a diuretic, a calcium channel blocker and an alpha adrenoreceptor blocker. At age 60 he experienced weakness affecting all four limbs, slowly progressing over four months. There was no respiratory or bulbar involvement. He reported a concomitant increase in his sensory disturbance—ascending paraesthesiae in the four extremities. The symptoms had appeared despite adequate glycaemic control with insulin and gliclazide. On examination at that time he had symmetrical weakness of the four limbs (MRC grade 2–3) equally affecting proximal and distal muscles, distal sensory loss in a stocking-and-glove distribution, distal loss of vibration sense, and generalized areflexia. Plantar responses were flexor, and cranial nerve and cerebellar functions were unimpaired. Blood investigations (including renal function, creatine kinase, liver enzymes, thyroid function tests, vitamin B12 and folate levels and protein electrophoresis) were normal. Nerve conduction studies revealed severely reduced motor conduction velocities (left median at forearm 28.7 m/s, right ulnar nerve at forearm 3.3 m/s), prolonged distal latencies (left median 5.4 ms, right ulnar 5.3 ms, right tibial 7 ms), and absent F waves, for the three motor nerves tested. Sensory nerve action potentials were either absent (both median nerves and right sural nerve) or subnormal (right radial 1 µV), and distal conduction velocities were slowed (right radial at wrist 29.4 m/s). Cerebrospinal fluid (CSF) was normal for cellularity, glucose, and protein content (0.20 g/L, normal range 0.20–0.45 g/L). The symptoms were attributed to rapid worsening of his known diabetic polyneuropathy. Without treatment except for physiotherapy, the patient reverted to his previous neurological state over about twelve months.
The current episode was similar, and likewise occurred despite adequate control of his diabetes (by insulin and metformin). The symptoms progressed on this occasion more rapidly, over 6 weeks. Motor nerve conduction velocities were again severely reduced (right median 31.7 m/s at forearm, right ulnar 31.4 m/s at forearm, left tibial 15.6 m/s), distal motor latencies were prolonged (right median 5.6 ms, right ulnar 3.3 ms), and F-wave responses were either severely delayed (right median 40.0 ms) or absent (for all other 3 motor nerves tested). Sensory nerve action potentials were mostly absent (right median, ulnar, radial and sural nerves), the left median being reduced (5.1 µV), with a slowed conduction velocity (28.5 m/s). Electromyography showed signs of active denervation in the proximal and distal muscles of the four limbs. CSF on this occasion was acellular, with normal glucose but raised protein at 0.63 g/L. A nerve biopsy was not performed. The patient was judged to have a relapsing chronic inflammatory demyelinating polyneuropathy (CIDP), superimposed on the diabetic neuropathy, in view of the history, clinical picture, compatible neurophysiology, and the raised CSF protein on the second sample, according to the criteria outlined by Barohn et al.1. He was treated with a course of intravenous immunoglobulins (400 mg/kg daily for 5 days) and made a total recovery within less than two weeks from the end of treatment. When reviewed 6 months later, the condition had not progressed or relapsed.
COMMENT
Coexistence of diabetic neuropathy and CIDP has been documented previously.2,3 Diabetic patients are at excess risk of developing CIDP by a factor of 10 or more.3 However, patients with diabetes-associated CIDP do not differ obviously in terms of clinical progression or response to treatment from those without diabetes.2 Rapid deterioration of motor function, as in our patient's relapse, seems a good indicator of underlying CIDP in diabetes, although progression can be slow, as in the first episode. For distinguishing CIDP from diabetic polyneuropathy, nerve conduction studies are of great value.4 In the present patient motor nerve conduction velocities were low in relation to the compound muscle action potentials, and the other neurophysiological criteria for CIDP were met. CSF examination findings seem less useful because of the high prevalence of raised protein in diabetes,5 and nerve biopsy results also tend to be unhelpful, the diagnostic features of CIDP being often present in diabetic polyneuropathy.4
This case illustrates the need for a close eye on motor function in patients with diabetic neuropathy. If clinical progression suggests possible CIDP, neurophysiological tests are indicated. Whereas in diabetic neuropathy the symptoms can only be controlled, in CIDP there is the possibility of more effective therapy. In cases where the diagnosis is uncertain, a therapeutic trial of intravenous immunoglobulins may be justified. Safe and easy to administer, they have given promising results in patients with CIDP associated with diabetic neuropathy, in a small uncontrolled study.6
REFERENCES
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
