Introduction

Phocomelia is a congenital malformation in which the proximal portions of the extremities are poorly developed or absent and the hands and feet are attached directly to the trunk or by means of a poorly formed bone [1]. It is extremely rare but emerged as a devastating side effect and one of a range of limb defects in newborn children [2–5] exposed to the drug thalidomide in utero. In contrast to many teratogenic drug side effects there was no dose response relationship but its impact markedly time dependent acting between 20 and 36 gestational days in utero [6–7], however even a single tablet at the appropriate time could result in the full blown condition. This led to its withdrawal from use in humans in 1962. Thalidomide is now used only in particular conditions where its side effects are a distal, axonal, symmetrical, predominantly sensory [8], dose-dependent [9–10] peripheral neuropathy [11–12].

Individuals born with thalidomide embryopathy (TE) have reported symptoms that may be attributable to peripheral nervous system (PNS) dysfunction in the maldeveloped limbs in later life [13, 14] and it has been hypothesized that “late onset neuropathy in thalidomiders will be recognized as a sensory equivalent of post-polio syndrome” [15]. We set out to determine whether there was any evidence of peripheral nerve dysfunction in subjects with TE limb anomalies and neurological symptoms in the upper limbs. In contrast to a previous study [14] we systematically investigated large and small somatic and sympathetic nerve function not only in the symptomatic maldeveloped upper limbs, but also in the largely asymptomatic lower limbs to ascertain whether there was any electrophysiological evidence of a more generalized neuropathic dysfunction.

Materials and Methods

Results

Large fibre neuropathic abnormalities are present in the affected upper limbs and sensory changes are present in the asymptomatic, clinically unaffected lower limbs

The CMAP of the median nerve showed no evokable response in six TE subjects and in some a response could only be obtained by stimulating at the axilla. The mean median CMAP amplitude was significantly reduced in subjects compared with controls. The ulnar CMAP amplitude and the amplitude of the median and ulnar sensory nerve action potential (SNAP) were not significantly different in subjects compared with controls (Table 1). Entrapment of median nerve at the distal segment (at the wrist) was seen in five subjects (bilaterally = 1, right = 2 and left = 2). The superficial radial SNAP could not be assessed in twelve TE subjects (due to radial aplasia) but it was not significantly different compared with controls. Sural nerve SNAPs were recordable in the lower limb and the mean amplitude was significantly reduced in subjects, compared with controls (Table 1). No entrapment mono-neuropathies were found in the lower limbs and no needle EMG abnormalities were found.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Nerve conduction studies of the upper and lower limbs in controls (C) and thalidomide embryopathy (TE) subjects.

https://doi.org/10.1371/journal.pone.0152902.t001

Sympathetic skin responses were abnormal in the affected, symptomatic upper limbs whereas thermal thresholds were abnormal in asymptomatic lower limbs

The mean perception threshold for warm sensation was significantly increased in the clinically normal lower limbs in subjects compared with controls. Perception threshold for cool sensation and cold and heat pain thresholds were similar between TE subjects and controls in the upper and lower limbs. SSRs were present in the upper and lower limbs of TE subjects but the mean SSR amplitude in the palms was significantly reduced in the TE subject group, when compared with the controls. For the SSR amplitudes on the plantar surfaces, however, no differences in TE subjects compared with the controls could be detected (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Thermal thresholds and sympathetic skin responses in the upper and lower limbs in controls (C) and thalidomide embryopathy (TE) subjects.

https://doi.org/10.1371/journal.pone.0152902.t002

Discussion

To the best of our knowledge, this is the first study systematically assessing the peripheral nervous system in both the clinically affected upper limb but also the lower limbs in middle-aged subjects born with TE. All had confirmed thalidomide exposure in utero, limb reduction defects ranging from thumb hypoplasia to phocomelia of the upper limbs with sensory symptoms, and most had normal appearing lower limbs.

In this group we found a range of problems that raised important clinical considerations for individuals with TE [14]. Three-quarters had features suggesting upper limb compressive neuropathy but, electrophysiological abnormalities consistent with compressive CTS were only found in 5 subjects. This discrepancy arises from the complex anatomy, making the typical electrophysiological findings of CTS unhelpful. In three subjects, cervical radiculopathy was suspected clinically, despite atypical dermatomal and myotomal anatomy, and subsequently confirmed on cervical magnetic resonance imaging. These findings support the need for a detailed examination of subjects with TE and neurological symptoms to exclude other treatable pathologies.

In the abnormal upper limbs, evidence of a motor large fibre dysfunction was documented. SSR was also reduced in the palms, consistent with additional sudomotor (efferent cholinergic fibre) dysfunction. Longitudinal shortening of the radial aspect in the upper limb and tibial aspect in the lower limb have been seen in the abnormal limb of phocomelic subjects [15]. Consistent with these observations our subjects had radial aspect abnormalities in the clinically abnormal upper limb but they did not have tibial aspect abnormalities. However, in the predominantly, grossly anatomically normal lower limbs, the mean amplitude of the sural SNAP was reduced. Furthermore, TTT showed that warm perception was significantly different in the lower limb in TE subjects. These findings together suggest axonal nerve dysfunction affecting not only sensory large fibers but also the small fibres. The exclusion of other potential causes of peripheral neuropathy increases the possibility that the electrophysiological abnormalities could have resulted from thalidomide exposure in utero. At the present time, the majority of TE subjects do not have clinical evidence (signs or symptoms) of a peripheral neuropathy affecting the lower limbs. As such, it is currently unclear whether the electrophysiological abnormalities found in the lower limbs have been present long term, whether these are progressive in nature, and whether they will prove to represent a clinically important part of the TE syndrome in future. Thalidomide exposure in adults in some work but not all has been found to cause a dose-dependent peripheral neuropathy [10, 16]. In TE the damage is markedly time dependent as opposed to being dose dependent [6–7]. As Thalidomide exposure in the TE group was in utero, it is possible that the neuropathic changes demonstrated here may have arisen from the embryonic pathobiological defects causing TE and are not likely to be a direct effect of thalidomide exposure. However, whether this is clinically relevant can only be determined with a follow-up study to determine whether further neuropathic symptoms develop and to look for evolution of the electrophysiological abnormalities.

Supporting Information

Acknowledgments

The Thalidomide Trust commissioned a pilot study to evaluate the presence of peripheral neuropathy. The authors gratefully acknowledge the help provided by Anne Horton at HealthLink, Dr Richard Sills and Dr Dee Morrison, Medical Advisors at The Thalidomide Trust.