Abstract

Methods

Selection criteria

The group's consensus diagnosis of ALS was based on the following criteria¹:

• Clinical and electromyographic evidence of progressive LMN degeneration

• Clinical evidence of UMN degeneration

• LMN and UMN signs in at least two of the four central nervous system regions: brain stem, cervical, thoracic or lumbosacral spinal cord

• Clear progression of the disease.

Enrolment in this study further required:

• No evidence of diabetes mellitus, malignancy, paraproteinaemia, nutritional deficiency or other known potential causes of polyneuropathy

• No family history of ALS

• Being an ambulatory patient

• Confirmation of ALS diagnosis by follow‐up.

Open in a separate window

Figure 1Outline of the group of patients with ALS. Of the 159 patients with anterior horn cell disorder (AHCD) sampled, 61 were excluded: 11 at the data validation level because of incomplete data submission (n=3) or because of a consensus diagnosis other than anterior horn cell disorder (n=8); 43 at the peer review level as they did not meet the established criteria for ALS (23 primarily due to absence of upper motor neurone signs, 17 primarily due to lower motor neurone signs in one central nervous system region only or due to unconvincing electromyographic findings, 2 with Kennedy's disease and 1 with spinal muscular atrophy), 6 with diabetes mellitus and 1 with other systemic disorder. At follow‐up, 10 patients with an ALS consensus diagnosis were excluded: 9 patients were lost to follow‐up and 1 patient was excluded as there had been no disease progression over 5 years.

Results

Of the 88 patients with ALS, 68 (77.3%) patients were entered into the ALS group, 11 (12.5%) patients into the ALS+GSSA group and 9 (10.3%) patients into the ALS+possible GSSA group. The incidence of patients with additional consensus diagnosis of polyneuropathy was significantly higher in the ALS group than in the control group (12.5% v 2.0%; p<0.01).

Comparison of sex, disease duration and region of onset among the three subgroups of patients with ALS did not show any differences. The mean age in the three groups was ALS: 62.7 years (range 36–85); ALS GSSA: 69.3 years (range 50–81); and ALS+possible GSSA: 62.8 years (range 52–77). The mean duration to examination in the three groups was ALS: 10.4 months (range 2–48); ALS+GSSA: 10 months (range 2–24); and ALS+possible GSSA: 8.9 months (range 3–30). The frequency of patients with bulbar onset was ALS: 26%; ALS+GSSA: 27%; and ALS+possible GSSA: 22%.

When comparing the ALS+GSSA and ALS groups separately, the average number of sensory nerves examined per patient was higher in the ALS+GSSA group (3.7 v 2.8; p=0.04). Furthermore, there was a tendency (p=0.09) towards older age of the patients in the ALS+GSSA group.

No sensory deficits on neurological examination were reported in any of the patients in the ALS+GSSA or ALS+possible GSSA groups.

Electrophysiological findings

Of 41 sensory nerves examined in the ALS+GSSA group, 32 (78%) nerves had abnormal SNC parameters: 6 had decreased SNCV and decreased SNAP amplitude, 10 had decreased SNCV, 14 had decreased SNAP amplitude and 2 nerves did not elicit any response (table 1​1).). Of the 41 sensory nerves examined in the ALS+possible GSSA group, 15 (37%) nerves had electrophysiological deficits: 5 had decreased SNCV and SNAP amplitude, 7 had decreased SNAP amplitude, 2 had decreased SNCV and 1 nerve did not yield any sensory response. None of the abnormalities in any of the patients in the two groups was due to nerve entrapment.

Table 1Electrophysiological findings in the ALS+GSSA and ALS+possible GSSA groups

Group	Patient	No of examined sensory nerves	Percentage of abnormal sensory nerves	Abnormal nerve(s)	SNCV	SNAP amplitude
ALS+GSSA	1	5	100	Radial	↓	↓
				Median	↓	↓
				Ulnar	↓	↓
				Sural		No response
				Peroneal		No response
	2	3	67	Ulnar		↓
				Sural	↓	↓
	3	2	100	Ulnar	↓
				Peroneal	↓
	4	4	50	Median	↓
				Median	↓
	5	3	100	Median	↓
				Ulnar		↓
				Sural		↓
	6	4	75	Radial	↓
				Median		↓
				Sural	↓
	7	3	67	Median	↓
				Ulnar	↓
	8	4	50	Median	↓
				Sural		↓
	9	2	100	Median		↓
				Ulnar		↓
	10	8	75	Median	↓	↓
				Median	↓	↓
				Ulnar		↓
				Sural		↓
				Peroneal		↓
				Peroneal		↓
	11	3	100	Median		↓
				Sural		↓
				Sural		↓
ALS+possible GSSA	12	4	25	Sural		↓
	13	5	20	Median		↓
	14	5	20	Ulnar		↓
	15	3	67	Ulnar		↓
				Ulnar		↓
	16	5	40	Median	↓	↓
				Ulnar		No response
	17	3	33	Median	↓
	18	2	50	Median	↓	↓
	19	8	25	Peroneal Peroneal	↓	↓
					↓	↓
	20	6	33	Peroneal Peroneal	↓

Open in a separate window

ALS, amyotrophic lateral sclerosis; GSSA, generalised sensory system abnormalities; SNAP, sensory nerve action potential; SNCV, sensory nerve conduction velocity; ↓, decrease.

Discussion

In this study, 20 (22.7%) patients from a series of 88 patients with ALS sampled by an international group of seven experienced neurophysiologists showed abnormalities in sensory NCS parameters. The doctors, serving as a peer review group, furthermore diagnosed 11 (12.5%) patients with additional electrophysiological polyneuropathy. This proportion clearly exceeds the reported frequencies of polyneuropathy in the background population by 0.04–0.1%, and the 1.6% prevalence of polyneuropathy in patients >55 years with no recognised exposure to diseases or neurotoxic agents found in an Italian multicentre study.^19,20,21 These rates may be underestimated, as they were based on clinical assessment; higher frequencies are expected when electrodiagnostic tests are applied.²² However, the frequency of additional polyneuropathy in our age‐matched control group comprising patients with various disorders examined in the involved laboratories was 2%.

The significantly greater proportion of patients with normal sensory NCS findings in this study indicates that sensory nerves are not generally affected in patients with ALS. Rather, patients with sensory nerve dysfunction may represent a variant of ALS. The similar disease duration of <11 months in the three ALS subgroups indicates that the sensory nerve involvement is part of the primary disease, which is supported by the non‐progressive subclinical findings of the sensory system in patients with ALS by Theys et al,⁷ and the recent quantitative demonstration of a marked degeneration of sensory axons in the dorsal roots of the SOD1^G93A mouse model of ALS, which appeared to be non‐progressive with advancing disease.²³

Although the patients in this study were diagnosed with electrophysiological peripheral polyneuropathy, the sensory nerve abnormalities can be generated in the dorsal root ganglion, as a peripheral sensory axonal neuropathy and a ganglionopathy cannot be distinguished using electrodiagnostic methods. Morphological findings of sensory nerve biopsies from patients with ALS have shown axonal degeneration, segmental demyelination and a reduced number of large neurones of the L5 dorsal root ganglion.^8,9,10,12 These findings and the findings of disparate electrophysiological abnormalities in this study, not consistently indicative of either axonal loss or demyelination, are in keeping with the suggestion by Heads et al¹⁰ that the primary pathology in the sensory peripheral nerve in ALS is dorsal root ganglion neuronopathy, resulting initially in progressive axonal atrophy followed by secondary de‐remyelination and ultimately by axonal loss.

The revised El Escorial (World Federation of Neurology) criteria for the diagnosis of ALS allow abnormal sensory NCS only in the presence of entrapment syndrome or coexisting peripheral nerve disease.¹ Normal sensory NCS have been shown in a series of patients with ALS.^24,25 Other neurophysiological studies have reported frequencies of patients with ALS with abnormal findings of the peripheral nervous system ranging between 13% and 22%.^3,5,6,7 In this study, in a few patients only two sensory nerves were studied or reduced SNCV was the sole finding. Clearly, caution should be taken in these cases; however, broad criteria that include the presence of some electrophysiological sensory abnormality may be appropriate. Further prospective studies are needed to determine how much abnormality is allowable in the diagnosis of ALS.

Abbreviations

Footnotes

References