Sural nerve biopsy is reported to establish or contribute substantially to the diagnosis of peripheral neuropathy in 27% to 60% of the patients in whom it is performed (Argov et al 1989; Neundörfer et al 1990; Oh 1990). Specific, otherwise unsuspected, diagnoses are obtained in a smaller proportion, however -- generally less than 30% (Neundörfer et al 1990; Oh 1990). A large proportion of cases remain undiagnosed after biopsy, from 12% to 42%, depending on the series (12% by Argov and colleagues, 36% by Neundörfer and colleagues, and 42% by Rappaport and colleagues) (Argov et al 1989; Neundörfer et al 1990; Rappaport et al 1993).
In a prospective study of 50 patients with sural nerve biopsy in whom the investigating neurologist made a clinical diagnosis before the biopsy and an independent neurologist later evaluated the contribution of the biopsy to diagnosis and management, sural nerve biopsy changed the diagnosis in 14% and was felt to have affected management in 60% (although in these additional cases, the biopsy confirmed the clinical diagnosis) (Gabriel et al 2000).
Yield of sural nerve biopsy depends on the degree of slowing of nerve conduction velocity (Wees et al 1981; Argov et al 1989), with sural nerve histology helpful in about two thirds (65%) of patients with motor conduction velocity below 30 m/s, but only about one tenth (11%) of patients with milder reductions of conduction velocity (Argov et al 1989). Thus, Argov and colleagues suggested that sural nerve biopsy is indicated for patients with marked slowing of conduction velocity (below 30 m/s) in whom the diagnosis is "not immediately apparent." Similarly, Wees and colleagues suggested that using abnormal sural nerve conduction as a guide to use of sural nerve biopsy will greatly enhance diagnostic yield (Wees et al 1981).
Not surprisingly, the utility of sural nerve biopsy depends on the diagnosis suspected and the findings of other studies. In a study of 64 patients in whom a diagnosis of chronic inflammatory demyelinating polyneuropathy was considered, the results of sural nerve biopsy did not improve the discriminating performance of a statistical model nor the diagnostic performance of an experienced neurologist when the results of the biopsy were considered with other clinical features (ie, remitting course, symmetric sensorimotor neuropathy in arms and legs, areflexia, raised CSF protein concentration, nerve conduction studies consistent with demyelination, and absence of other significant confounding comorbidity or other relevant laboratory abnormalities) (Molenaar et al 1998). Nevertheless, a recent practice parameter by the American Academy of Neurology suggested that nerve biopsy was "generally accepted" as useful with "atypical" forms of chronic inflammatory demyelinating polyneuropathy (England et al 2009).
Sural nerve biopsy has been reported to be particularly useful in establishing or confirming a diagnosis of peripheral nerve vasculitis and paraproteinemic neuropathies (Wees et al 1981; Oh 1990; Said 2002; Hilton et al 2007; England et al 2009). However, the yield from combined superficial peroneal nerve and peroneus brevis muscle biopsy is reportedly somewhat greater than that of sural nerve biopsy, with a reported diagnostic sensitivity of about 60% for combined superficial peroneal nerve and peroneus brevis muscle biopsy and less than 50% for sural nerve biopsy (Hilton et al 2007). Furthermore, some studies have reported very low rates of confirmation of suspected vasculitic neuropathies with sural nerve biopsy, eg, 6 of 29 patients with suspected vasculitic neuropathy by Rappaport and colleagues (Rappaport et al 1993). In a study of sural nerve biopsy in 13 patients with vasculitic neuropathies, the differential diagnosis and classification of the underlying neuropathies depended on the caliber of involved vessels (Ohkoshi et al 1996).
Sural nerve biopsy has also been reported to be a sensitive test for diagnosing peripheral neuropathy due to primary amyloidosis (England et al 2009), but in 1 series of 9 patents who were ultimately diagnosed as having amyloidosis, the initial sural nerve biopsy was negative in two thirds (Simmons et al 1993). Therefore, a negative sural nerve biopsy does not certainly exclude amyloidosis as the cause of a neuropathy and should not discourage or preclude further diagnostic studies, especially when the clinical suspicion of amyloidosis is high.
In many cases, the former diagnostic utility of sural nerve biopsy has been obviated by skin biopsies or development of genetic or other tests that are either more helpful or less likely to produce complications (eg, the genetic tests for CADASIL) (Schröder et al 1995; Pleasure 2007; England et al 2009). Skin biopsies have been shown to be helpful in diagnosing vasculitis neuropathy and in quantifying axonal loss (Lee et al 2005; Chao et al 2007; Pleasure 2007) and can be helpful as well for quantitative immunochemical analysis of peripheral nervous system myelin proteins in patients with hereditary predisposition to pressure palsies or Charcot-Marie-Tooth disease associated with neurofilament light chain gene (NEFL) mutations (Li et al 2005, 2007; Miltenberger-Miltenyi et al 2007; Pleasure 2007). In selected other genetic disorders, there are circumstances in which sural nerve biopsy can be helpful in false negative and false positive diagnoses, eg, in patients with suspected metachromatic leukodystrophy when interpretation of low arylsulfatase A levels is impossible due to low arylsulfatase A levels in 1 of the parents or because information on the arylsulfatase A levels of the parents cannot be obtained (Vos et al 1982).