Electrophysiologic Assessment of Acute Polyradiculoneuropathy in Dogs: Comparison with Guillain-Barré Syndrome in People

 Accepted March 24, 1997.

 Acknowledgments: Supported in part by the University of Wisconsin School of Veterinary Medicine Companion Animal Grant. I thank Dr. Ian Duncan and Dr. John Stewart for advice on manuscript preparation.

Idiopathic acute canine polyradiculoneuropathy (ACP) is the most commonly recognized peripheral neuropathy in dogs.1 In North America, the term Coonhound paralysis (CHP) has been used to describe this condition because many affected dogs have been bitten by a raccoon prior to the onset of neurologic signs.2 Although raccoon saliva has been incriminated as the source of the possible causative antigen or agent, the exact etiology and pathogenesis of this syndrome still remain unresolved.3 This syndrome has been suggested to be an animal model for the acute polyradicu loneuropathy of people, Guillain-Barré syndrome (GBS).4 The classic clinical signs of CHP, seen 7-14 days after a raccoon bite, consist of an initial stiff, stilted gait with rapid progression to a flaccid, lower motor neuron tetraparesis or tetraplegia. Generalized hypo- to areflexia, hypotonia, and severe neurogenic muscle atrophy are usually accompanied by aphonia. Hyperesthesia commonly develops and is secondary to the variable dorsal nerve root involvements The same presentation and progression of neurologic signs also occur in dogs with ACP that have no history of an encounter with a raccoon.

 The ventral nerve roots and the ventral root components of spinal nerves develop the most severe pathology in ACP, consisting of varying degrees of axonal degeneration, paranodal and segmental demyelination, and leukocytic infiltration.1,4,5 Lumbar and sacral ventral nerve roots are more severely involved than those in the thoracic or cervical region.5 The distal peripheral nerve trunks and the dorsal nerve roots are much less severely affected than ventral nerve roots.4,5 Axonal degeneration appears to be a more consistent and prominent histopathologic finding in ACP than in GBS.4

 Electrophysiologic investigations of dogs with ACP have been previously reported.4-8However, most have been limited to brief and sometimes conflicting reports of abnormal electromyographic (EMG) findings and selective sensory and motor nerve conduction studies. In one study, a more systematic Electrophysiologic investigation was performed, although complete analyses, including the results of F-wave studies (F-wave conduction velocities), were reported for only 5 dogs.8 There has been little discussion in the literature on an association between the observed electrophysiologic changes and the stage, severity, and pathologic distribution of the disease. The present study has incorporated an in-depth investigation of motor and sensory peripheral nerve function with a detailed study of ventral nerve root function in dogs at various stages and with different severity of ACR These data were then used to compare electrophysiologic abnormalities with known pathologic changes in this disease and to assess the utility of the individual tests in detecting changes in nerve root and peripheral nerve function.
 

Materials and Methods

 Twelve dogs with history and neurologic signs strongly indicative of ACP were studied at different stages of the disease, with dog 2 and dog 8 studied on 2 separate occasions. Although all dogs were from a rural environment, only 3 of the dogs had documented attacks by a raccoon. Four dogs had known previous encounters with raccoons, with 1 dog showing numerous facial scars indicative of a recent raccoon attack. The signalment, necrologic signs, and the time after disease onset are summarized in Table 1. Also included in Table 1 is a subjective numerical grade indicating the relative severity of each dog's clinical signs at the time of examination (grade 1—ambulatory with a stiff-stilted gait and mild hyporeflexia; grade 2—ambulatory tetraparesis with hyporeflexia; grade 3—nonambulatory tetraparesis with hyporeflexia; grade 4—severe nonambulatory tetraparesis with marked hyporeflexia to areflexia; grade 5—flaccid tetraplegia with areflexia).

 All dogs were evaluated under isoflurane (AErrane®, Anaquest Inc, Liberty Corner, NJ) (dogs 6 and 12) or halothane (Halothane®, Halocarbon Laboratories, Augusta, SC) gas anesthesia after induction with intravenous thiamylal (BIO-TAL®, Boehringer Ingelheim Animal Health Inc, St. Joseph, MO) to effect (approximately 8-10 mg/kg). Body and limb temperatures were maintained above 36°C with circulating water heating pads. A DISA 1500 Digital EMG system (Dantec Electronics Inc. Allendale, NJ) was used for all EMG and nerve conduction recordings. EMGs were recorded from the supraspinatus, infraspinatus, deltoid, biceps, triceps, extensor carpi radialis, common digital extensor, superficial and deep digital flexor, palmer interosseous, gluteal, quadriceps, semitendinosus, semimembranosus, cranial tibial, gastrocnemius, planter interosseous, and paraspinal muscles with a stainless steel, 0.45- X 40-mm concentric needle electrode.

 Motor and sensory nerve functions were assessed by using previously reported recording/stimulation sites and recording parameters9,10 with the one exception that radial nerve compound muscle action potentials (CMAPs) were recorded from the extensor carpi radialis muscle. Motor nerve function was studied in the sciatic/tibial nerve (14 examinations in 12 dogs), in the radial nerve (11 examinations in 10 dogs), and in the ulnar nerve (5 examinations in 4 dogs). Sensory function was examined in the tibial nerve (14 examinations in 12 dogs), in the radial nerve (12 examinations in 11 dogs), and in the ulnar nerve (11 examinations in 9 dogs). The variables measured were insertional and spontaneous EMG activity (fibrillations and positive sharp waves), CMAP peak-to-peak amplitude,10 CMAP duration,10 proximal and distal CMAP area,11 motor nerve conduction velocity (MNCV),10,12 distal CMAP latency,10,13 sensory nerve action potential (SNAP) amplitude,9 and sensory nerve conduction velocity (SNCV).9 From these data, the presence or absence of CMAP temporal dispersion and conduction block was evaluated. In this study, temporal dispersion was defined as a relative desynchronization of the components of the CMAP due to different rates of evoked impulse conduction along the stimulated axons, resulting in prolongation of CMAP duration.

 Conduction along the entire nerve length and its associated ventral nerve roots was assessed via F-wave analyses. A total of 10-12 F waves was recorded from the planter interosseous muscles after stimulation of the distal tibial nerve just proximal to the tarsus (Fig 1). The minimum F-wave latencies, defined as the time from the onset of the stimulation artifact to the first deflection of the waveform from baseline, were measured.14 A linear regression equation was used to calculate expected F-wave latency (expected minimum F-wave latency = 3 45 + 0.33 x limb length [cm]; r = 0.87).15 Pelvic limb length in each dog was measured from the greater trochanter of the femur to the tip of the 4th digit. An F-ratio (F = [lat L - lat M) - 1]/[2 X lat M], where L = F wave and M = M wave)16 was also calculated to assess F-wave latency (conduction in the proximal nerve segment) relative to the latency of the M wave (conduction in the distal nerve segment). F-wave peak-to-peak amplitudes were also recorded.16

 Statistical Analysis

 Normal means, standard deviations, and ranges were obtained from previously published electrophysiologic data for CMAP amplitude, duration, and proximal and distal area; MNCV; distal CMAP latency; SNAP amplitude; SNCV; F-ratios; and F-wave amplitude.9,10 Results from clinically normal dogs evaluated in our laboratory closely agree with those reported in the above published studies. Results from affected dogs were compared with these published values, and the number of dogs with results outside the range provided by the mean ± 2 SDs was recorded for each outcome.
Two different criteria were used to determine the presence of conduction block because there is controversy over which is more accurate: (1) a proximal: distal CMAP amplitude ratio of <2 SDs below previously established means,11 with a <15% concomitant change in CMAP duration, and (2) a reduction in proximal: distal CMAP area ratio of <0.5 if there was marked proximal temporal dispersion or <2 SDs below the means established in dogs if temporal dispersion was not evident.11,17-19

 For F-wave latency assessment, dogs were classified as abnormal when their measured minimum F-wave latency appeared to fall outside the 95% confidence interval provided with the published linear regression equation that describes the association between limb length and expected F-wave latency.
 

 Results

 EMG Abnormalities

 Mild EMG abnormalities, consisting of increased insertion activity and abnormal spontaneous discharges (fibrillation potentials and positive sharp waves) (Table 1), were seen as early as 1.5 days after the reported onset of clinical signs and were consistently found in all 4 dogs studied within 4 days after disease onset. However, these changes were invariably mild when compared with dogs with the same severity of clinical dysfunction that were studied later in the disease course. All dogs that were studied at 5 days or more after disease onset had the above EMG abnormalities, which showed a general trend toward increasing severity in dogs with greater necrologic dysfunction. However, there was variability among dogs within the same clinical severity grouping. Complex repetitive discharges of varying frequency and severity were also commonly ob served (10/12 affected dogs [83%]). Axial muscles appeared to be involved only in dogs with the most severe necrologic dysfunction (tetraplegia).

 CMAP Amplitude Changes

 Sciatic/tibial CMAP amplitudes, after stimulation at the hock, stifle, and hip, were decreased to less than 2 SDs below previously reported means in 10/12 (83%), 11/12 (92%), and 11/12 (92%) of affected dogs, respectively (Table 2). The small number of dogs with normal CMAP amplitudes primarily showed very mild to moderate signs. The presence of decreased CMAP amplitudes was largely independent of disease severity and time of examination after disease onset although increasing necrologic deficits were generally associated with decreasing sciatic/tibial CMAP amplitudes. The lowest CMAP amplitudes were seen in dogs with flaccid tetraplegia. Similar trends were observed in radial and ulnar CMAP amplitudes (9/10 [90%] with decreased proximal radial CMAP amplitudes; 8/10 [80%] with decreased distal radial CMAP amplitudes; 4/4 [100%] with decreased proximal and distal ulnar CMAP amplitudes) during the acute progressive phase and nadir of the disease (Table 2). Normal sciatic/tibial and radial CMAP amplitudes were seen in dog 1, which showed very mild signs during the peracute phase of her disease. The worsening necrologic status in dog 2 was reflected in decreasing amplitudes in all CMAPs for both the sciatic/tibia! and radial nerves (Fig 2A,C). Improvement in all CMAP amplitudes, with a return of both proximal and distal ulnar CMAP amplitudes to normal, was seen in dog 8 at the time of examination during recovery.

 CMAP Dispersion

 CMAP temporal dispersion was present in the sciatic/tibial nerve in 8 of the affected dogs (67%), although there was no apparent association with clinical severity or the time of examination after disease onset (sciatic/tibial CMAP dispersion was not seen in 2 dogs with moderate/severe signs and in 3 dogs with severe signs). Proximal CMAPs were the most severely affected. Radial CMAP temporal dispersion was seen only in dogs 6 and 10 (20%) and was associated only with the proximal stimulation site. Proximal ulnar CMAP temporal dispersion was seen only in 1 dog with moderate/severe signs (dog 5).

 Evidence for Conduction Block

 Proximal: distal sciatic/tibial CMAP amplitude ratios were <2 SDs below the mean in 6 dogs (Table 3). Only dog 2 (2nd examination) and dog 8 (1st examination) (17%) also demonstrated a <15% decrease in proximal versus distal CMAP duration, indicating conduction block. Radial nerve proximal: distal CMAP amplitude ratios were below 0.80 in only 4 dogs, with only dog 5 and dog 10 (20%) meeting the criteria for conduction block (Table 3). Ulnar nerve CMAP amplitude ratios were <2 SDs below the normal mean in dog 8 on both examinations although there was an improvement in this ratio paralleling the dog's clinical improvement (Table 3). This dog met the criteria for conduction block only in the nadir of her disease.

 The ratios of proximal: distal sciatic/tibial CMAP area were <2 SDs below previously published means in only 3 dogs. Dog 8 (1st examination) and dog 9 (17%) had proximal: distal CMAP area ratios of <0.5, indicating conduction block. Dog 8 demonstrated an improvement in sciatic/ tibial area ratio to normal during disease recovery (Table 3). Although there are no established normal values for radial CMAP proximal: distal area ratios in the dog, 7 of the 10 dogs had ratios   0.80 (Table 3). Dogs 5, 10, and 12 (30%) had substantially decreased area ratios, indicating conduction block. Although dog 2 had a decrease in radial CMAP area ratio from 0.96 to 0.60, as she progressed from mild to severe signs, she did not meet the criteria for conduction block. Three of 4 dogs had ulnar proximal: distal CMAP area ratios at or above established means (Table 3). As with ulnar amplitude ratios, dog 8 demonstrated a substantial decrease in ulnar area ratios during both her disease nadir and recovery and met the criteria for conduction block on both occasions.

 Preservation of MNCVs

 Despite the substantial decreases in CMAP amplitudes, all dogs, independent of disease severity, had sciatic/tibial MNCVs within 2 SDs of age-matched control means (Table 4). Radial MNCVs were <2 SDs below the normal mean in the single examinations on dogs 4, 5, and 12 and on the 2nd examination of dog 2 (Table 4). Two of these dogs were tetraplegic. Dog 2 had a marked decrease in radial MNCV from 80 m/second to 35 m/second as she progressed from mild to severe signs. Ulnar MNCVs were within the established age-matched normal ranges in all dogs tested, with almost identical values in dog 8 during her disease nadir and recovery (Table 4).

 Prolongation of distal CMAP Latencies

 An increase in distal tibial CMAP latency >2 SDs over the established mean occurred in 8 dogs (67%) (Table 4). However, only 2 of these dogs (dogs 11 and 12), which were both tetraplegic, had distal tibial CMAP latencies > 1 millisecond above the upper limit of normal. The same was found with distal radial CMAP latencies, with only 1 of the 4 dogs (dog 12) with substantially increased distal latencies having a distal radial CMAP latency > 1 millisecond above the upper limit of normal (Table 4). A prolonged distal ulnar CMAP latency was observed only in dog 6, which had moderate/severe signs (Table 4).

 Maintenance of Peripheral Sensory Nerve Function

 Thoracic and pelvic limb peripheral sensory nerve function was largely unaffected, irrespective of disease severity and the time of examination after disease onset (Table 5). Most dogs had tibial, radial, and ulnar SNAP amplitudes and SNCVs within 2 SDs of the normal means (11/12 dogs [92%], 7/11 dogs [64%], and 8/9 dogs [89%] with normal SNAP amplitudes, and 11/12 dogs [92%], 10/11 dogs [91%], and 9/9 dogs [100%] with normal SNCVs, respectively). The decreased tibial and radial SNCVs occurred in the same dog (dog 4). The decrease in radial SNAP amplitude in 36% of dogs showed no correlation with the severity of the observed necrologic dysfunction.

 Ventral Nerve Root Involvement

 Both minimum F-wave latencies and F-ratios were abnormal in all dogs with severe necrologic dysfunction (Table 6). All clinically affected dogs, with the exception of the dog with very mild signs (dog 1), had abnormalities in at least 1 of the above tests of ventral nerve root function (8/12 [67%] and 11/12 [92%] for minimum F-wave latency and F-ratios, respectively). In the 8 dogs (67%) in which there was an increase in the measured minimum F-wave latency or an absence of recordable F waves (dogs 2 and 11), a normal sciatic/tibial MNCV was recorded. The remaining 4 dogs (dogs 1, 3, 5, and 7) had normal F-wave latencies and MNCVs.

 Mean F-wave amplitudes were <2 SDs below published values in all dogs with the most severe necrologic dysfunction (Table 6). Although a number of dogs with less severe signs also had decreases in mean F-wave amplitudes of <2 SDs below the mean, this was inconsistent in dogs with milder deficits. A decrease in F-wave amplitudes was seen in 8/12 (67%) of affected dogs. Time of examination after disease onset did not contribute to the above results. The very mildly affected dog (dog 1) had normal mean F-wave amplitudes. F-wave amplitudes were already well below expected values at 2 days postonset of signs in dog 2 which subsequently progressed to complete flaccid tetraplegia and an absence of any recordable F waves (Fig 2B,D).
 

 Discussion

 The results of the above studies strongly suggest that ACP represents a severe motor axonopathy affecting the entire length of the peripheral nerve, which tends to overshadow any signs of demyelination in most dogs. The electrophysiologic evidence for this axonopathy is largely in dependent of the clinical severity of the disease and the time of examination after disease onset, although worsening severity of signs tends to be associated with more severe axonal involvement. Severe axonopathy as well as demyelination, however, also occur at the most proximal portion of the motor nerves, ventral nerve roots, or both and also have a strong tendency to worsen with increasing disease severity, although less severely affected dogs may have normal F-wave characteristics. These observations are in agreement with the previously reported findings of de creased F-wave conduction velocities or the absence of recordable F waves in a small number of dogs with suspected ACP.8 Evidence of demyelination along the peripheral portions of motor nerves is present in some dogs, with a higher percentage of abnormalities seen in the more clinically af fected dogs. These changes, however, are somewhat inconsistent among nerves of the same dog. This finding does not support an earlier report that demonstrated fairly consistent MNCV slowing or no measurable CMAPs in many dogs with suspected ACP.8 The reason for this discrepancy is not known. Electrophysiologic evidence of sensory nerve involvement in this disease is minimal, with the most notable abnormality being a decrease in SNAP amplitude without temporal dispersion. This is primarily related to the radial nerve in this study. This finding would suggest mild sensory axonal loss in some patients, although this interpretation is tenuous due to the wide variation of SNAP amplitudes in normal dogs and the proximity of many of the "abnormal" amplitudes to the lower limits of normal.

 The electrophysiologic findings that appear to be the earliest and most reliable indicators of ACP are EMG changes in multiple muscles, markedly decreased CMAP amplitudes in multiple peripheral nerves, absent F waves, and an increased sciatic/tibial nerve F-ratio. Increased minimum F-wave latencies and, in dogs with the most severe signs, significantly decreased F-wave amplitudes would also strongly support this diagnosis. Although evidence of ventral nerve root axonal changes (decreased F-wave amplitudes) does not appear to occur independent of CMAP amplitude changes, many dogs had a preferential distribution of demyelination in the ventral nerve roots and most prox imal regions of the peripheral nerves. This is supported by the fact that 75% of dogs with ACP in this study had an increase in sciatic/tibial minimum F-wave latencies, despite having normal MNCVs. This percentage is much higher than that reported in GBS and chronic inflammatory de myelinating polyneuropathy in people, where only 20.1% of 403 tested nerves showed preferential proximal nerve and ventral nerve root conduction changes.20

 The electrophysiologic abnormalities seen in ACP are not as consistent with the classic form of GBS (87% of cases) as with the acute axonal or intermediate forms of the disease (3% and 10% of patients, respectively).21 Electrophysiologic studies of classic GBS reveal an evolving picture of multifocal demyelinating polyneuropathy with secondary axonal degeneration. In contrast, the acute axonal form of GBS is characterized by absent or severely decreased proximal and distal CMAP amplitudes with later development of extensive EMG changes, despite relatively preserved MNCVs and distal motor latencies.17,22-24 Evidence of widespread segmental demyelination (conduction block, CMAP temporal dispersion, and prolonged distal latencies), which occurs in 50% of classic GBS patients in the first 14 days and in at least 85% of patients between 14 and 21 days, was not a prominent feature in the majority of the dogs with ACP in this study.17,25-27 The degree of clinical impairment in ACP showed a much greater association with the prominent, generalized CMAP amplitude decreases and ventral nerve root abnormalities than with the presence or degree of motor conduction block, as is reported in classic GBS in people.25,28-31

 The time course for development of clinical and electrophysiologic abnormalities in dogs with ACP is much faster than in classic GBS patients. Dogs with ACP reach maximum disease severity within 10 days (and commonly within 4-5 days) of disease onset1 and often demonstrate EMG abnormalities, CMAP amplitude reduction, and F-wave latency prolongation within 4 days. In contrast, peak neurologic dysfunction occurs in only 50% of classic GBS patients even after 14 days,26 with inconsistent development of fibrillations and positive sharp waves only after the first 14-21 days.17,25 F-wave latency prolongation also may not occur in GBS patients until the 4th week after disease onset (46% within the first 14 days).14,17,32-34

 Despite published data indicating that the onset of spontaneous muscle activity secondary to denervation occurs after 5-6 days in the dog,1,35 all dogs in this study presenting between 1.5 and 4 days postonset of signs had sparse mild EMG activity in appendicular muscles. Analysis of the results in this study, therefore, suggests that EMG examination of dogs with suspected ACP, presenting prior to day 5 of disease, can provide valuable information on the pathologic process occurring.

 The consistency of observed electrophysiologic changes among nerves in the same dog was greatest for CMAP amplitude reduction. The parameters used for detection of peripheral demyelination were much less consistent. When all of the criteria to assess demyelination that were evaluated in this study are taken into consideration, the sciatic/tibial nerve appeared to be more severely involved than the thoracic limb nerves. This more severe involvement may have been partly due to the greater length of the sciatic/tibial nerve under investigation, especially in relation to the higher frequency of CMAP temporal dispersion and increased distal CMAP latency reported in this nerve. Agreement among techniques to assess different aspects of demyelination in the sciatic/tibia! nerve, however, was poor. Temporal dispersion of proximal and, to a lesser extent, middle sciatic/tibial CMAPs (stimulation at the stifle) suggested demyelination in a high percentage of affected dogs although its presence did not relate to disease severity. The presence or absence of temporal dispersion was also the least consistent parameter in suggesting demyelination among nerves of the same dog (no dog showed consistent temporal dispersion in all three nerves examined). The most consistent assessor of the presence or absence of demyelination among nerves of the same dog was MNCV. Normal MNCV was found in all nerves tested (at least 2) in 67% (8/12) of the evaluations. In the remaining 4 studies, decreased MNCV was seen only in the radial nerve. One possible explanation for this finding may be related to technical error in the measurement of the distance between radial nerve stimulation sites, which was <10 cm in all dogs tested. In addition, reported "normal" radial nerve conduction velocities have been established with young adult dogs, with no adjustment for decreased MNCVs with increased age,9 in contrast to the sciatic/tibial and ulnar nerves.12 Realistically, however, this would have affected the interpretation of our results only for dog 4, a 10-year-old Labrador cross, whose radial MNCV was just less than 2 SDs below the normal mean.

 When the 2 methods used for the detection of conduction block (proximal: distal CMAP amplitude and area ratios) were compared, there was good agreement in the results from the 3 nerves tested. There was, however, a marked inconsistency in the finding of conduction block among nerves of the same dog. Conduction block was observed in more than 1 nerve in only 1 dog (dog 8), which differs from the finding of multifocal conduction block in many patients with classic GBS.26,31

 The severity of the pathologic changes in the ventral nerve roots in ACP is well documented1,4,5 and has been strongly substantiated in this study by the abnormalities seen in minimum F-wave latency, F-ratio, and F-wave amplitude. The addition of F-wave evaluation has distinct advantages over M-wave analysis alone, including the detection of purely proximal motor nerve segment and nerve root pathology, amplification of borderline abnormalities due to a longer pathway being tested, and an increased testing sensitivity because F-wave latencies have a much narrower normal range than MNCVs.14 Although minimum F-wave latency measurements and F- ratios both assess proximal motor nerve and ventral nerve root conduction abnormalities, they use very different measurement criteria—a regression equation based on limb length and a comparison of F-wave and distal M-wave latencies, respectively. Agreement between the two techniques on the presence of nerve root demyelination was relatively strong (9/12 dogs [75%]). With the exception of dog 1, all dogs showed ventral nerve root conduction abnormalities in at least 1 of the techniques used. These findings are similar to those in people, where 92% of nerves from patients with GBS in 1 study showed F-wave abnormalities.20 Because neither of the techniques was 100% predictive when used alone, however, both should be employed to obtain the maximum amount of information on ventral nerve root function.

 Interpretation of the results of F-ratios strongly indicates that conduction abnormalities are selectively much more severe in the most proximal motor nerve segments and ventral nerve roots in ACP when compared with the distal nerve segments. Only dog 8 had more severe electrophys iologic changes in the distal tibial nerve than in the ventral nerve roots, despite increased F-wave latencies. Despite the fact that 2 dogs with severe signs (dogs 11 and 12) also had substantially increased distal tibial residual latencies, the increased F-ratio in dog 12 and the absence of F waves in dog 11 indicated more severe demyelination in the ventral nerve roots. This finding was confirmed by the minimum F-wave latency regression equation. These findings are in sharp contrast to those reported in GBS, where the F-ratio was normal in 51% of 126 nerves studied in 45 GBS patients, decreased in 25%, and increased in only 24%.36 Therefore, the conduction abnormality in GBS affects both proximal and distal nerve segments equally in the majority of patients and, if selective, is distributed at random between these two areas.36

 Despite the fact that not all dogs in this study had a history of a raccoon attack, the classic antecedent event associated with ACP, all consistently had a similar electrophysiologic pattern of an acute primary motor axonopathy with severe ventral nerve root demyelination and axonal involvement. Although ACP should still be regarded as a natural animal model for human GBS, the much more prominent axonal involvement in ACP seen on electrophysiology in this study, and on histopathology in the pasty suggests a much closer affinity to the rarer axonal and intermediate forms of the disease in people.

 1. Cummings. JE Canine inflammatory neuropathies. In Kirk RW, Bonagura JD, eds. Current Veterinary Therapy XI. Philadelphia, PA: WB Saunders; 1992:1034-1037.

 2. Duncan ID. Peripheral neuropathy in the dog and cat. Prog Vet Neurol 1991;2:111-128.

 3. Cuddon PA. Electrophysiological and immunological evaluation in Coonhound paralysis. Proc ACVIM 8th Annual Veterinary Medical Forum, Washington, DC, 1990:1009-1012.

 4. Cummings JF, de Lahunta A, Holmes DF, et al. Coonhound paralysis. Further clinical studies and electron microscopic observations. Acta Neuropathol (Berl) 1982;56:167-178.

 5. Cummings JF, Hass DC. Coonhound paralysis. An acute idiopathic polyradiculoneuritis in dogs resembling the Landry Guillain-Barré syndrome. J Neurol Sci 1967;4:51-81.

 6. Chrisman CL. Differentiation of tick paralysis and acute polyradiculoneuritis in the dog using electromyography. J Am Anim Hosp Assoc 1975;11:455-458.

 7. Northington JW, Brown MJ, Farnbach GC, et al. Acute idiopathic polyneuropathy in the dog. J Am Vet Med Assoc 1981;179: 375-379.

 8. Northington JW, Brown MJ. Acute canine idiopathic polyneuropathy. A Guillain-Barré-like syndrome in dogs. J Neurol Sci 1982; 56:259-273.

 9. Redding RW, Ingram JT, Colter SB. Sensory nerve conduction velocity of cutaneous afferents of the radial, ulnar, peroneal, and tibial nerves of the dog: Reference values. Am J Vet Res 1982;43:517-521.

10. Walker TL, Redding RW, Braund KG. Motor nerve conduction velocity and latency in the dog. Am J Vet Res 1979;40:1433-1439.

11. Malik R, Ho S, Church DB. A new method for recording and analysing evoked motor potentials from dogs. J Small Anim Pract 1989;30:13-19.

12. Swallow JS, Griffiths IR. Age related changes in the motor nerve conduction velocity in dogs. Res Vet Sci 1977;23:29-32.

13. Bowen JM. Electromyographic analysis of evoked potentials of canine muscle motor points. J Am Vet Med Assoc 1974;164:509-512.

14. Lachman T, Shahani BT, Young RR. Late responses as aids to diagnosis in peripheral neuropathy. J Neurol Neurosurg Psychiatry 1980:43:156-162.

15. Steiss JE. Linear regression to determine the relationship between F-wave latency and limb length in control dogs. Am J Vet Res 1984;45:2649-2650.

16. Poncelet L, Balligand M. Nature of the late potentials and F-ratio values in dogs. Res Vet Sci 1991;51:1-5.

17. Asbury AK, Cornblath DR. Assessment of current diagnostic criteria for Guillain-Barré syndrome. Ann Neurol 1990;27(Suppl):S21-S24.

18. Rhee EK, England JD, Sumner AJ. A computer simulation of conduction block: Effects produced by actual block versus interphase cancellation. Ann Neurol 1990;28:146-156.

19. Hausmanowa Petrusewicz I. Conduction bloc in peripheral nerves. Facts and hypotheses. Neurol Neurochir Pol 1994;28:157-166.

20. Kiers L, Clouston P, Zuniga G, et al. Quantitative studies of F responses in Guillain-Barré syndrome and chronic inflammatory demyelinating polyneuropathy. Electroencephalogr Clin Neurophysiol 1994;93:255-264.

21. Albers JW, Donofrio PD, McGonagle TK. Sequential electrodiagnostic abnormalities in acute inflammatory demyelinating polyradiculoneuropathy. Muscle Nerve 1985;9:528-539.

22. Meulstee J, van der Meche FG. Electrodiagnostic criteria for polyneuropathy and demyelination: Application in 135 patients with Guillain-Barré syndrome. Dutch Guillain-Barré Study Group. J Neurol Neurosurg Psychiatry 1995;59:482-486.

23. Feasby TE, Gilbert JJ, Brown WF, et al. An acute axonal form of Guillain-Barré polyneuropathy. Brain 1986;109:1115-1126.

24. Yuki N, Yoshino H, Sato S, et al. Severe acute axonal form of Guillain-Barré syndrome associated with IgG anti-GD14 antibodies. Muscle Nerve 1992;15:899-903.

25. Albers JW, Kelly JJ. Acquired inflammatory demyelinating polyneuropathies: Clinical and electrodiagnostic features. Muscle Nerve 1989;12:435-451

26. Arnason BG, Soliven B. Acute inflammatory demyelinating polyradiculoneuropathy. In: Dyck PJ, Thomas PK, eds. Peripheral Neuropathy. Philadelphia, PA: WB Saunders; 1993:1437- 1497.

27. Kinoshita A, Kawamura J, Hashimoto S, et al. Sequential studies of motor nerve conduction in relation to the disease stages of Guillain-Barré syndrome. Rinsho Shinkeigaku 1993;33:600- 605.

28. Brown WF, Feasby TE. Conduction block and denervation in Guillain-Barré polyneuropathy. Brain 1984;107:219-239.

29. Feasby TE, Brown WF, Gilbert JJ, Hahn AE The pathological basis of conduction block in human neuropathies. J Neurol Neurosurg Psychiatry 1985;43:239-244.

30. Ropper AH. The Guillain-Barré syndrome. N Engl J Med 1992;326:1130-1136.

31. Cros D, Triggs WJ. There are no neurophysiologic features characteristic of "axonal" Guillain-Barré syndrome. Muscle Nerve 1994;17:675-677.

32. Cornblath DR, Mellits ED, Griffin JW, et al. Motor conduction studies in Guillain-Barré syndrome: Description and prognostic value. Ann Neurol 1988;23:354-359.

33. Gilmore RL, Nelson KR. SSEP and F-wave studies in acute inflammatory demyelinating polyradiculoneuropathy. Muscle Nerve 1989;12:538-543.

34. Cornblath DR. Electrophysiology in Guillain-Barré syndrome. Ann Neurol 1990;27(Suppl):S17-S20.

35. Bowen JM. Electromyography. In: Oliver JE, Hoerlein BF, Mayhew IG, eds. Veterinary Neurology. Philadelphia, PA: WB Saunders; 1987:145-168.

36. Kimura J. Proximal versus distal slowing of motor nerve conduction velocity in the Guillain- Barré syndrome. Ann Neurol 1978;3:344-350.

Fig 1. Normal F waves recorded from a control dog, after stimulation at the hock, to demonstrate typical waveform shape and amplitudes. Electrodes are positioned such that the negative polarity is in the upward direction.

Fig 2. Sciatic/tibial CMAPs (after stimulation at the hock, stifle, and hip) and F waves (after stimulation at the hock) recorded from dog 2 at 2 days (A,B, respectively) and 11 days (C,D, respectively) after the onset of acute polyradiculoneuropathy after a raccoon bite. On day 2, there was a decrease in CMAP amplitudes (3.9, 1.85, and 1.63 mV) and proximal CMAP temporal dispersion despite a normal MNCV of 61.6 m/ second. F waves at this time showed a substantial decrease in amplitude (45.5-90  V), apparent polyphasia, and an increase in minimum latency (expected F-wave latency = 19.32 millisecond). Paralleling this dog's marked disease progression, there was a further decrease in CMAP amplitudes on day 11 (0.08, 0.055, and 0.04 mV), without temporal dispersion, despite a continued normal MNCV of 54.6 m/second. F waves were unobtainable at this second examination, indicating the severity of ventral nerve root and proximal peripheral nerve pathology. In all tracings, electrodes are positioned such that the negative polarity is in the upward direction.
 

^aTime of electrophysiologic assessment after reported onset of clinical signs.

^bNumerical assessment of the degree of EMG changes (increased insertion activity and spontaneous activity consisting of fibrillation potentials and positive sharp waves): 0 = normal; 1 = increased insertion activity followed by 5 or less abnormal spontaneous discharges on the oscilloscope screen at any one time in  25% of thoracic and pelvic limb muscles; no abnormalities in axial muscles; 2 = increased insertion activity and between 1 and 5 abnormal spontaneous discharges on the oscilloscope screen at any one time in >50% of thoracic and pelvic limb muscles; no abnormalities in axial muscles; 3 = increased insertion activity and between 5 and 10 abnormal spontaneous discharges on the oscilloscope screen at any one time in 75-100% of thoracic and pelvic limb muscles; no abnormalities in axial muscles; 4 = increased insertion activity and 10 abnormal spontaneous discharges on the oscilloscope screen at any one time in all appendicular muscles; no abnormalities in axial muscles; 5 = complete obliteration of the baseline on the oscilloscope screen with increased insertion activity and abnormal spontaneous discharges in all appendicular muscles: variable abnormal spontaneous activity (ranging from grade 1 to grade 5) in axial muscles.

Table 2. Compound muscle action potential (CMAP) amplitudes after stimulations of the sciatic/tibial, radial, and ulnar nerves in 12 dogs with polyradiculoneuropathy.
 

Dog	Numerical Severity of Clinical Signs	Sciatic/Tibial Nerve CMAP Amplitude (mV)			Radial CMAP Amplitude (mV)		Ulnar CMAP Amplitude (mV)
		Proximal	Middle	Distal	Proximal	Distal	Proximal	Distal
1	1	8.80	10.80	10.40b	12.50	7.80	--	--
2 (2 days)	2	1.63b	1.85b	3.90b	0.40b	0.40b	--	--
2 (11 days)	5	0.04b	0.06b	0.08b	0.03b	0.06b	--	--
3	3	4.00b	7.60b	11.60	10.20b	8.00	--	--
4	3	6.25b	6.50b	11.50	3.50b	4.00b	--	--
5	3	1.90b	1.67b	4.90b	2.10b	3.45b	5.30b	7.50b
6	4	2.50b	2.60b	3.80b	2.40b	2.25b	0.96b	1.40b
7	4	6.60b	6.20b	8.00b	4.40b	3.60b	5.00b	6.00b
8 (15 days)	4	1.20b	1.80b	3.25b	5.10b	6.10b	4.10b	11.00b
8 (120 days)	2	2.05b	2.90b	5.20b	--	--	9.60	18.00
9	5	0.35b	0.42b	1.12b	--	--	--	--
10	5	0.05b	0.04b	0.07b	0.30b	0.80b	--	--
11	5	0.94b	0.75b	1.26b	--	--	--	--
12	5	0.04b	0.04b	0.03b	2.00b	5.50b	--	--
Normals10 (mean ± SD)		19.6 ± 5.59	22.2 ± 5.81	22.7 ± 6.04	24.5 ± 6.04	21.0 ± 6.71	20.7 ± 6.0	24.0 ± 6.48

^aStimulation sites: proximal sciatic/tibial nerve--deep within the groove formed between the greater trochanter of the femur and the ischiatic tuberosity; middle sciatic/tibial nerve deep in the popliteal fossa caudal to the distal end of the femur; distal sciatic/tibial nerve deep to the gastrocnemius tendon just proximal to the tuber calcaneus; proximal radial nerve--two-thirds the distance down the humeral diaphysis where the nerve spirals laterally; distal radial nerve--at the proximocranial border of the extensor carpi radialis muscle; proximal ulnar nerve--at the elbow in the groove formed by the olecranon and the medial epicondyle of the humerus; distal ulnar nerve--at the carpus just proximal and deep to the accessory carpal bone.

bCMAP amplitudes <2 SDs below the normal means.

Table 3. Proximal: distal compound muscle action potential (CMAP) amplitude ratios (in relation to the percentage change in CMAP duration) and proximal: distal CMAP area ratios for the sciatic/tibia!, radial, and ulnar nerves in 12 dogs with acute polyradiculoneuropathy. Normal proximal: distal CMAP amplitude and area ratios are not available for the radial nerve.
 

Dog	Numerical Severity of Clinical Signs	CMAP Amplitude Radio  (% Change in Duration)			Proximal : Distal CMAP Area Ratio
		Sciatic/Tibial Nerve	Radial Nerve	Ulnar Nerve	Sciatic/Tibial Nerve	Radial Nerve	Ulnar Nerve
1	1	0.84  (9.7)	1.60  (9.6)	--  --	1.27	1.80	--
2 (2 days)	2	0.41  (80.0)	1.00  (1.5)	--  --	0.81	0.96	--
2 (11 days)	5	0.50a  (9.8)	0.50  (35.5)	--  --	0.91	0.60	--
3	3	0.35  (28.4)	1.28  (1.7)	--  --	0.53	1.22	--
4	3	0.54  (32.6)	0.88  (-1.5)	--  --	0.92	0.86	--
5	3	0.39  (34.5)	0.61a  (-9.9)	0.70  (97.0)	0.90	0.47a	1.17
6	4	0.66  (57.1)	1.07  (-5.2)	0.69  (24.4)	0.91	1.00	1.65
7	4	0.83  (10.6)	1.22  (-1.6)	0.83  (10.0)	0.77	1.05	0.88
8 (15 days)	4	0.37a  (-5.4)	0.84  (1.6)	0.34a  (21.1)	0.49a	0.89	0.53a
8 (120 days)	2	0.39  (75.4)	--  --	0.53  (52.8)	0.64	--	0.33a
9	5	0.31  (62.0)	--  --	--  --	0.42a	--	--
10	5	0.72  (19.5)	0.38a  (7.9)	--  --	1.16	0.34a	--
11	5	0.75  (0.0)	--  --	--  --	1.17	--	--
12	5	1.30  (11.4)	0.33a  (18.0)	--  --	1.44	0.43a	--
Normals11  (mean ± SD)		0.68 ± 0.12  (<15%)	--  (<15%)	0.79 ± 0.1  (<15%)	0.79 ± 0.08	NA	0.89 ± 0.08

NA, not available.

^aMeets the criteria for conduction block.

Table 4. Sciatic/tibial, radial, and ulnar motor nerve conduction velocities (MNCVs) and distal compound muscle action potential (CMAP) latencies in 12 dogs with acute polyradiculoneuropathy. Each sciatic/tibial and ulnar MNCV is compared with those of previously published age-matched control dogs. Age-matched controls are not available for the radial nerve. Stimulation sites are the same as in Table 2.
 

Dog	Numerical Severity of Clinical Signs	Sciatic/Tibial Nerve		Radial Nerve		Ulnar Nerve
		MNCV  (m/seconds)  (normal age-matched MNCV12 [m/second ± SD])	Distal CMAP Latency  (m/second)	MNCV  (m/second)	Distal CMAP Latency  (m/second)	MNCV  (m/second)  (normal age-matched MNCV12 [m/second ± SD])	Distal CMAP Latency  (m/second)
1	1	57.6  (60.5 ± 6.9)	3.9a	61.5	2.3a	--	--
2 (2 days)	2	61.6  (64.7 ± 5.7)	4.2	80.0	1.8	--	--
2 (11 days)	5	54.6  (64.7 ± 5.7)	3.9a	35.0a	1.5	--	--
3	3	71.5  (65.5 ± 3.2)	3.4	77.3	1.7	--	--
4	3	53.2  (58.1 ± 4.2)	3.6	55.0a	1.4	--	--
5	3	51.2  (56.6 ± 4.2)	3.0	33.3a	1.4	48.4  (5.19 ± 6.2)	1.7
6	4	68.8  (64.2 ± 3.4)	4.2a	61.0	1.8	59.7  (62.1 ± 1.3)	3.1a
7	4	67.8  (64.2 ± 3.4)	2.8	76.9	1.9	57.1  (62.0 ± 4.5)	1.9
8 (15 days)	4	49.6  (56.6 ± 4.2)	4.3a	66.7	2.2a	51.5  (51.9 ± 6.2)	2.0
8 (120 days)	2	50.2  (56.6 ± 4.2)	3.5	--	--	53.6  (51.9 ± 6.2)	1.8
9	5	55.7  (62.2 ± 4.1)	4.2a	--	--	--	--
10	5	55.6  (56.6 ± 4.2)	4.1a	68.8	2.1a	--	--
11	5	52.0  (60.5 ± 6.9)	6.2a	--	--	--	--
12	5	39.6  (48.2 ± 6.5)	5.1a	22.8a	3.0a	--	--
Normals10  (mean ± SD)			3.4 ± 0.22	73.1 ± 6.04	1.5 ± 0.22	NA	2.1 ± 0.22

^aMNCVs and distal CMAP latencies <2 SDs below the normal means.

^aSNCVs and SNAP amplitudes <2 SDs below the normal means.

^bF-ratios and F-wave amplitudes that are <2 SDs below the normal means.

  Hop Back to the Neurology Journal Club Page!