- Email: [email protected]
IN-RATIO: A new test to increase diagnostic sensitivity in ulnar nerve entrapment at elbow
Clinical Neurophysiology 119 (2008) 1600–1606 www.elsevier.com/locate/clinph
IN-RATIO: A new test to increase diagnostic sensitivity in ulnar nerve entrapment at elbow P. Caliandroa,b,*, M. Foschinia, C. Pazzagliaa, G. La Torrec, I. Aprileb, G. Granataa, P. Tonalia,b, L. Paduaa,b a
Institute of Neurology, Fondazione Pro Iuventute Don Carlo Gnocchi, Universita` Cattolica, Largo Francesco Vito 1, 00168 Roma, Italy b Fondazione Don Carlo Gnocchi ONLUS, Roma, Italy c Epidemiology and Biostatistics Unit, Institute of Hygiene, Universita` Cattolica, Rome, Italy Accepted 1 March 2008 Available online 2 May 2008
Abstract Objective: Motor conduction velocity may yield false-negative results in mild ulnar nerve entrapment at elbow (UNE). There is evidence that the clinical heterogeneity of UNE may be due to the different involvement of fascicles. We hypothesized that, if fibres to FDI are more damaged than fibres to ADM, a relative slowing of motor conduction velocity (CV) at the segment across the elbow recording from FDI (FDI-CV) versus CV at the same segment recording from ADM (ADM-CV) would occur. Methods: We calculated the ratio between FDI-CV and ADM-CV (IN-RATIO) in 60 consecutive UNE patients, 40 norms, and 16 patients with lower cervical radiculopathy. The UNE sample consisted of (1) patients with neurophysiological evidence (UNE NF+), (2) patients without neurophysiological evidence (UNE NF ). We evaluated the possible usefulness of the IN-RATIO to increase sensitivity in diagnosing UNE. Results: The IN-RATIO was lower in the UNE NF than in norms (p < 0.001) and cervicobrachialgia sample (p = 0.02). We found that if the IN-RATIO is 60.97, its sensitivity in diagnosing UNE NF is 68%, specificity 91%. Conclusions: Fibres to FDI have a higher susceptibility to damage than fibres to ADM, and the IN-RATIO is a useful neurophysiological parameter to increase diagnostic sensitivity. Significance: Our results demonstrate the usefulness of a new neurophysiological parameter in improving diagnosis of UNE. Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. Keywords: Entrapment of ulnar nerve; Diagnosis; Neurography
1. Introduction Ulnar nerve entrapment at the elbow (UNE) is a common entrapment syndrome. The clinical picture ranges from intermittent paresthesias in the fourth and fifth digits to complete sensory loss in the territory of the ulnar nerve with atrophy and weakness of the ulnar muscles (Bhala, 1976; Tackmann et al., 1984; Beekman et al., 2004a). The most frequently impaired muscles are the first dorsal inter* Corresponding author. Address: Institute of Neurology, Fondazione Pro Iuventute Don Carlo Gnocchi, Universita` Cattolica, Largo Francesco Vito 1, 00168 Roma, Italy. Tel.: +39 06 30154435; fax: +39 06 35501909. E-mail address: [email protected] (P. Caliandro).
osseous (FDI) and the abductor digiti minimi (ADM); involvement of flexor digitorum profundus (FDP) or flexor carpi ulnaris (FCU) is much less frequent (Stewart, 1987). Some evidence suggests that this clinical heterogeneity is due to the different involvement of fascicles within the nerve (Neary and Eames, 1975). In the region of the elbow nerve fibres from the terminal digital sensory branch and to the muscles of the hand lie adjacent to bone, and this position may render them more susceptible to damage (Sunderland, 1978). These clinical aspects could influence the electrophysiological findings in UNE patients. The goal of electrodiagnosis is to confirm ulnar nerve dysfunction across the elbow, and to assess the severity of nerve impairment (Bhala, 1976; Tackmann et al., 1984; Padua et al.,
1388-2457/$34.00 Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2008.03.007
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
2001; Beekman et al., 2004a). Slowed motor conduction velocity (CV) is the most common localizing electrophysiological sign (Kothari et al., 1998), although it may yield false-negative results in patients with mild lesions (Padua et al., 2001). Some authors note that sensitivity for detecting motor slowing across the elbow is higher when recording from FDI than from ADM, due to differential fascicular involvement (Stewart, 1987; Kothari et al., 1998). Other authors found that the sensitivity of motor conduction studies is similar for ADM and FDI recordings although the extra yield of recording from FDI in addition to ADM is 9% (Tackmann et al., 1984; Beekman et al., 2004a). We hypothesized that, if in UNE patients fibres to FDI are more damaged than fibres to ADM, a relative slowing of motor CV at the segment across the elbow recording from FDI (FDI-CV) versus CV at the same segment recording from ADM (ADM-CV) would occur. To verify this hypothesis we calculated the ratio between FDI-CV and ADM-CV in a sample of UNE patients and in a control group composed of norms and patients with lower cervical radiculopathy. We called this ratio the intra-nerve ratio (IN-RATIO). If FDI-CV were found to be relatively slower than ADM-CV in UNE patients as compared to the control group, we would have verified the usefulness of the IN-RATIO as a new neurophysiological parameter to increase sensitivity in diagnosing UNE. 2. Materials and methods 2.1. Clinical criteria We performed a case-control study with prospective gathering of data between January 2005 and February 2007, evaluating 60 consecutive patients (mean age 53, SD: ±15.52, 36 male, 24 female) with symptoms suggesting UNE. Patients were recruited in the neurophysiology laboratory of the university hospital A. Gemelli. In all of these patients the clinical diagnosis was made when two or more of the following criteria were present: (1) history of paresthesias and/or numbness at the fourth and fifth digit related to prolonged flexion of the elbow; (2) positive provocative test performed by prolonged (1 min) passive forced flexion of the elbow (test was positive when sensory symptoms at the fourth and fifth digit appeared or increased) (Novak et al., 1994; Rosati et al., 1998); (3) reduced motor function of ulnar innervated hand and forearm muscles, quantified according to the British Medical Research Council (BMRC) scale (Medical Research Council, 1976); (4) atrophy of the hypothenar eminence; (5) reduced sensory function focused on the ulnar hand region, evaluated through cotton wool. A skilled neurologist always examined the patient before neurophysiological evaluation was performed, and established whether the patient clinically presented UNE. The clinical picture was considered the gold standard to evaluate the neurophysiological findings. Historical/objective findings were also used to divide the UNE patients into two clinical groups:
1601
(1) patients with exclusively sensory symptoms; (2) patients with sensory–motor symptoms. Exclusion criteria were (1) previous surgery for UNE; (2) acute traumatic origin; (3) clinical or electrophysiological signs of polyneuropathy, including hereditary neuropathy with liability to pressure palsies. Patients were also excluded when electrodiagnostic studies (or additional magnetic resonance imaging) gave evidence of a radiculopathy, plexopathy, or ulnar neuropathy at Guyon’s canal. We also evaluated a control group composed by 40 norms (mean age 51, SD: ±18.09, 15 male, 25 female) and 16 patients with lower cervical radiculopathy (mean age 55, SD: ±12.35, 5 male, 11 female). In these patients the diagnosis was clinical, neurophysiological and radiological. 2.2. Neurophysiological studies We adopted an electrophysiological protocol similar to that proposed by the AAEM (American Association of Electrodiagnostic Medicine, 1999). During neurophysiological evaluation skin temperature was maintained >31 °C by an infrared lamp. Motor nerve conduction studies were performed with moderate flexion (70° to 90° from horizontal) of the elbow. This position was chosen because it provides the greatest correlation between surface skin measurement and true nerve length (Checkles et al., 1971; Kincaid et al., 1986; Olney et al., 1987; Kothari and Preston, 1995; American Association of Electrodiagnostic Medicine, 1999). The ulnar nerve was stimulated by surface electrodes at the proximal wrist crease, 3 cm distal to the medial epicondyle (below elbow-BE), and 6–7 cm above this level (above elbowAE); compound muscle action potentials (CMAPs) were recorded from ADM and FDI, using surface electrodes (American Association of Electrodiagnostic Medicine, 1999). The elbow position was maintained the same when recording from FDI and ADM. Anomalies of innervation, e.g. Martin-Gruber anastomosis, were always verified. When anomalies of innervation had been excluded, the technician calculated the IN-RATIO for the UNE patients, and for the control subjects. Subjects with anastomosis were excluded from the study. We also calculated the difference in motor CV between the AE-to-BE segment and the BE-to-wrist segment for all UNE and control samples. This difference was calculated when recording both from FDI and ADM. An AE-to-BE CV more than 10 m/s slower than the BE-to-wrist CV was considered pathological (American Association of Electrodiagnostic Medicine, 1999). The cutoff values for both FDI-CV and ADM-CV in the segment across the elbow were calculated at the 97th (third) percentile of the observed values in the control group. FDICV in the AE-to-BE segment was considered pathological when it was <43 m/s; ADM-CV was considered pathological when it was <40 m/s. These cutoff values are comparable to those found by Buschbacher in a large normative
1602
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
study (Buschbacher, 1999) and are confirmed by a ROC curve evaluation we performed to compare patients with clinical UNE versus the control group (for FDI-CV area under the curve is 0.830 and for ADM-CV area under the curve is 0.698). Amplitude reduction across the elbow was considered pathological when greater than 16% (conduction block) (Beekman et al., 2004a). Sensory nerve action potentials (SNAPs) were obtained orthodromically using ring electrodes placed over the fifth digit to stimulate the nerve while recording electrodes were applied to the wrist over the ulnar nerve. The cutoff value for the negative peak amplitude of the ulnar SNAP was calculated at the 97th (third) percentile of the observed values in the control group, and was considered pathological when <5 lV. Sensory and motor neurographies were performed by different technicians without knowing whether they were evaluating a clinical UNE patient, a cervicobrachialgia patient, or a normal subject. The technicians had to set the cursors for latency determination. When severe motor deficit or hypotrophy of ulnar innervated hand muscles occurred, using concentric needle electromyography (EMG) we studied ADM and FDI for fibrillation potentials, motor unit action potential (MUAP) amplitude, and recruitment pattern. FCU was evaluated in only a few cases. MUAP amplitude was manually assessed and considered pathological when greater than 2 mV (Beekman et al., 2004a). In addition to these tests, to exclude sub-clinical polyneuropathy we performed motor neurography of peroneal and median nerves, sensory neurography of radial nerve in the first digit-wrist segment, sensory neurography of median nerve in the first and third digit-wrist segments, and sensory neurography of sural nerve. We divided the UNE sample into two groups: (1) patients with clinical ulnar neuropathy at elbow and pathological motor CV across the elbow (UNE NF+); (2) patients with clinical ulnar neuropathy but without pathological motor CV across the elbow (UNE NF ). The patient was considered as having neurophysiological evidence of UNE when either FDI-CV or ADM-CV across the elbow was pathological. UNE patients were also graded according to a previously published neurophysiological classification (Padua et al., 2001). 2.3. Statistical analysis Statistical analysis was conducted using the SPSS package for Windows, release 12.0. The Kolmogorov–Smirnov probability test was used to assess distributions for normality. Among the neurophysiological parameters, only the distribution of FDI-CV in the control group sample was normal, therefore we decided to use the Mann–Whitney U test to compare the central tendency measures (medians) in the two groups. We used the ROC curve to establish the IN-RATIO cutoff value and we compared UNE NF group versus the control group (composed by norms and patients with lower cervical radiculopathy). In order to evaluate the difference among the UNE NF+ sample,
UNE NF sample and the control group concerning the frequency of positive IN-RATIO (considered as dichotomous variable), we used the standard Pearson’s chi-square test (2 2 table), which was also used to evaluate the frequency of pathological FDI-CV and pathological ADMCV (considered as dichotomous variables) in patients with exclusively sensory symptoms and in those with concomitant sensory and motor symptoms. To calculate specificity of the neurophysiological tests, we always used the control group composed by norms and patients with lower cervical radiculopathy. 3. Results 3.1. Clinical picture We examined 59 arms in 53 patients with clinical diagnosis of UNE (6 patients had bilateral symptoms) and excluded seven arms with Martin-Gruber anastomosis. Exclusively sensory symptoms at the fourth and fifth digits were present in 34 arms (58%), and concomitant sensory and motor symptoms in 25 arms (42%). In arms with sensory–motor symptoms, we found clinical weakness at both FDI and ADM in 20 arms, exclusively at ADM in 3 arms, and exclusively at FDI in 2 arms. Only in 5 arms was FCU also weak. Among arms with weakness of ulnar innervated muscles, in 7 arms paresthesia/numbness was exclusively referred in the fourth and fifth digits (none presented CTS), while in 18 arms there was paresthesia/numbness in the hand as a whole; in all of these arms neurophysiological evidence of CTS was revealed. No patient had motor deficit without sensory symptoms. We found hypoesthesia in the palmar region of the fourth and fifth digits, evaluated through cotton wool, in only 12 arms (7 cases presented weakness at both ADM and FDI, 1 exclusively at FDI and 1 exclusively at ADM; 3 arms had no weakness). 3.2. Neurophysiological picture In the 59 examined arms with UNE, in 25 both FDI-CV and ADM-CV were pathological in the AE-to-BE segment, in 5 only FDI-CV was pathological, while in 4 only ADMCV was pathological. For the cutoff value of FDI-CV <43 m/s in the segment across the elbow, sensitivity was 51% and specificity was 100%; for the cutoff value of ADM-CV <40 m/s in the same segment, sensitivity was 49% and specificity was 100%. Evaluation of both FDICV and ADM-CV increases the sensitivity of the neurophysiological evaluation to 58%. According to a previously published neurophysiological classification (Padua et al., 2001), 25 arms were UNE NF , 10 arms were UNE with mild involvement, 17 with moderate involvement, 7 with severe involvement and none with extreme involvement. Three arms presented an atypical neurophysiological pattern with normal FDI-CV and normal ADM-CV but with reduced amplitude of ulnar SNAP; these were included in
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
the UNE NF sample. Seven arms presented conduction block associated with absolute slowing across the elbow recording from both FDI and ADM. When in addition to the absolute reduction in FDI-CV and ADM-CV we use the AAEM criteria (drop in CV between AE-to-BE and BE-to-wrist segments), the overall sensitivity of the neurophysiological evaluation increases to 73% considering the entire sample of patients (UNE NF+ and UNE NF ) with a specificity of 82% considering the entire control group (norms and patients with lower cervical radiculopathy). Nine patients of the UNE NF sample presented a pathological drop in CV between AE-to-BE and BE-towrist segments. We found that FDI-CV and ADM-CV across the elbow were never normal in arms with concomitant sensory and motor symptoms (p < 0.001 for both FDI-CV and ADMCV) (Table 1). About 25 patients of the UNE NF sample, 22 subjects met criteria 1 and 2 while 3 subjects met criteria 1, 2, and 5. Both FDI-CV and ADM-CV were significantly slower in the UNE NF+ sample compared with the UNE Table 1 Distribution of UNE NF+ and UNE NF clinical criteria
patients according to the
Clinical Picture
Conventional test
P
UNE NF+
UNE NF
Sens/motor Sens Tot
25 (42%) 9 (15%) 34 (45%)
0 (0%) 25 (43%) 25 (55%)
<0.001a
Sens/motor, sensory and motor symptoms; Sens, sensory symptoms. a Chi-square test.
1603
NF sample, normal subjects, and the cervicobrachialgia sample (p < 0.001 for both FDI-CV and ADM-CV); no other difference was found (Fig. 1). EMG evaluation was performed in 10 arms. In 8 we found fibrillation potentials at FDI and in 7 at ADM. Pathological recruitment was found at both FDI and ADM in all 10 arms. In two arms we found pathological recruitment from FDI and ADM without increased MUAP amplitude; in these two arms we also evaluated FCU, and found fibrillation potentials and pathological recruitment with increased MUAP amplitude. 3.3. IN-RATIO evaluation In the UNE NF sample IN-RATIO was lower than that we found in norms (p < 0.001) and in the cervicobrachialgia sample (p = 0.02). No other statistical difference relative to IN-RATIO was found among the groups (Fig. 2). We calculated the cutoff value for IN-RATIO by comparing the UNE NF and the control group (norms and cervicobrachialgia sample), and found that if IN-RATIO is 60.97, then its sensitivity in diagnosing UNE NF is 68%, specificity 91%. Fig. 3 shows the ROC curve evaluation for IN-RATIO; the area under the curve is 0.77. Table 2 shows the distribution of subjects with IN-RATIO 60.97 and INRATIO >0.97. IN-RATIO 60.97 was more frequent in the UNE NF sample than in normal subjects, and cervicobrachialgia sample (p < 0.001), IN-RATIO 60.97 was more frequent in the UNE NF+ sample than in normal subjects (p < 0.001) and in cervicobrachialgia sample (p = 0.002).
Fig. 1. FDI-CV and ADM-CV across the elbow in the UNE NF+ sample, the UNE NF cervical radiculopathy (Cerv).
sample, normal subjects, and the sample of patients with lower
1604
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
Fig. 2. IN-RATIO in the UNE NF+ sample, the UNE NF
sample, normal subjects, and the sample of patients with lower cervical radiculopathy (Cerv).
When in addition to the absolute reduction in FDI-CV and ADM-CV we use the IN-RATIO, the overall sensitivity is 86%, specificity is 91%. 4. Discussion
Fig. 3. The ROC curve evaluation for IN-RATIO.
Table 2 Distribution of arms with positive and negative IN-RATIO in the four groups studied Samples
IN-RATIO Positive (60.97)
Negative (>0.97)
UNE NF+ UNE NF Norms Cerv
15 (44%) 17 (68%) 4 (10%) 1 (6%)
19 8 36 15
a
Chi-square test.
P
(56%) (32%) (90%) (94%)
<0.001a
In UNE diagnosis electrophysiological evaluation is crucial for localizing nerve damage at the elbow and for assessing the degree of nerve involvement. Nowadays motor nerve conduction studies are considered the most useful tests for localizing nerve injury (Kothari et al., 1998; American Association of Electrodiagnostic Medicine, 1999); however, these tests may not reveal milder nerve lesions. There is no internationally accepted agreement concerning the sensitivity and specificity of electrodiagnostic studies in diagnosing UNE. Reported sensitivity ranges from 37% to 86%, and specificity ranges from 95% upwards (American Association of Electrodiagnostic Medicine, 1999). Some authors have attempted to find the single most sensitive and specific electrodiagnostic test for UNE (Raynor et al., 1994; Merlevede et al., 2000; Lo et al., 2001); however, to our knowledge no international agreement has been reached. In our study sensitivity is 58% and specificity is 100% when both FDI-CV and ADM-CV are evaluated. Our sensitivity is lower than that reported elsewhere (Bhala, 1976; Tackmann et al., 1984; Beekman et al., 2004a), but the specificity is the best possible. We think that this difference is mainly due to the clinical characteristics of the patients enrolled. In fact, we included also patients without sensory and motor deficits, while in the studies just mentioned (Bhala, 1976; Tackmann et al., 1984; Beekman et al., 2004a) patients without abnormalities at physical
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
examination were excluded, and this may have increased the neurophysiological sensitivity. Our data provide further evidence that the AAEM criteria are useful and allow the neurophysiological diagnosis of UNE when an absolute reduction of FDI-CV and ADMCV across the elbow is absent. We considered the clinical picture as the gold standard to evaluate the neurophysiological findings. Although that approach may cause some difficulties in localizing the exact site of damage, we believe that the combination of inclusion and exclusion criteria may adequately prevent misdiagnosis and avoid inclusion of patients with pathologies that may mimic UNE (hereditary neuropathy with liability to pressure palsies, radiculopathy, plexopathy, or ulnar neuropathy at Guyon’s canal). We did not perform a follow-up evaluation to confirm UNE clinical diagnosis because we used to give instructions to the patient concerning the anatomical condition that causes the damage and the position in which the nerve is more compressed. We give information also when UNE diagnosis is not neurophysiologially confirmed and other causes of damage are clinically and instrumentally excluded (Padua et al., 2002). This approach may determine remission/improvement of symptoms and therefore the exclusion of actual UNE patients (Padua et al., 2002; Beekman et al., 2004b). Much evidence is reported concerning the fascicular organization of peripheral nerves. Sunderland described, in the distal part of a nerve, a clear somatotopic organization of nerve fibres clustered together into individual fascicles, while in the proximal region there is a kind of plexiform organization, with apparent intermingling of nerve fibres and fascicles (Sunderland, 1978). Other authors have pointed out that in spite of the plexiform structure, groups of functionally related nerve fibres can be identified at the proximal site of a nerve (Chow et al., 1985; Watchmaker et al., 1991). In other words, functionally related nerve fibres remain together even if they crisscross from one fascicle to another. We hypothesized that, if in UNE patients the fibres to FDI are more damaged than the fibres to ADM, the IN-RATIO can reveal a relative slowing of FDI-CV compared to ADM-CV in the segment across the elbow. Our neurophysiological findings show that in normal subjects FDI-CV is faster than ADM-CV. Since conduction velocity of each normal axon depends on the external nerve fibre diameter (Gasser and Grundfest, 1939; Hursh, 1939; McLeod and Wray, 1967; Waxman and Bennett, 1972; Ritchie, 1982), a possible explanation is that the fastest fibres to FDI have a larger diameter than the fastest fibres to ADM. In patients group, fibres to FDI are more impaired than fibres to ADM and this finding is mainly manifest when FDI-CV and ADM-CV across the elbow are within the normal range. In the UNE NF+ sample impairment of fibres to FDI and fibres to ADM is more uniform although a prevalent susceptibility of FDI fibres persists. A possible explanation of these findings is that for somatotopic characteristics of ulnar nerve, fibres to
1605
FDI lie adjacent to bone and therefore are more susceptible to damage. When the nerve damage is greater, a larger number of fibres are of interest and from these also fibres to ADM. IN-RATIO allows to compare the function of two different portions of a nerve within the same segment. This approach allows to eliminate all the variability typical of segmental and comparative techniques in which two different segments of one nerve or two different nerves are compared. In addition, although the accuracy of latency measurement is very important in the evaluation of INRATIO, this is independent of the absolute values of FDI-CV and ADM-CV, and therefore each laboratory may use its own normal values for these CVs to diagnose UNE. Our data demonstrate that IN-RATIO allows to increase the sensibility of electrodiagnosis in UNE although its specificity is lower than that we found for absolute reduction of motor CV across the elbow. However, considering that in UNE the global reliability of neurophysiological studies is not so high as in carpal tunnel, for example, we think that IN-RATIO may be a useful instrument in association with the clinical picture. In conclusion, our findings demonstrate that fibres to FDI have a higher susceptibility to damage than fibres to ADM and that IN-RATIO is a useful neurophysiological parameter to increase sensitivity in patients with clinical UNE. We think that when IN-RATIO is positive and FDI-CV and ADM-CV across the elbow are within the normal range, the patient may be classified as having a ‘‘minimal UNE”. Acknowledgments Thanks to Mr. Kim H. Wessling for language editing. The authors report no conflicts of interest. References American Association of Electrodiagnostic Medicine. Practice parameter for electrodiagnostic studies in ulnar neuropathy at the elbow: summary statement. American Association of Electrodiagnostic Medicine, American Academy of Neurology, American Academy of Physical Medicine and Rehabilitation. Muscle Nerve 1999;22:408–11. Beekman R, Van Der Plas JP, Uitdehaag BM, Schellens RL, Visser LH. Clinical, electrodiagnostic, and sonographic studies in ulnar neuropathy at the elbow. Muscle Nerve 2004a;30:202–8. Beekman R, Wokke JH, Schoemaker MC, Lee ML, Visser LH. Ulnar neuropathy at the elbow: follow-up and prognostic factors determining outcome. Neurology 2004b;63:1675–80. Bhala RP. Electrodiagnosis of ulnar nerve lesions at the elbow. Arch Phys Med Rehabil 1976;57:206–12. Buschbacher RM. Ulnar nerve motor conduction to the abductor digiti minimi. Am J Phys Med Rehabil 1999;78:S9–S14. Checkles NS, Russakov AD, Piero DL. Ulnar nerve conduction velocity effect of elbow position on measurement. Arch Phys Med Rehabil 1971;52:362–5. Chow JA, Van Beek AL, Meyer DL, Johnson MC. Surgical significance of the motor fascicular group of the ulnar nerve in the forearm. J Hand Surg [Am] 1985;10A:867–72. Gasser HS, Grundfest H. Axon diameters in relation to the spike dimensions and the conduction velocity in mammalian A fibers. Am J Physiol 1939;127:393–414.
1606
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
Hursh JB. Conduction velocity and diameter of nerve fibres. Am J Physiol 1939;127:131–9. Kincaid JC, Phillips LH, Daube JR. The evaluation of suspected ulnar neuropathy at the elbow: normal conduction study values. Arch Neurol 1986;43:44–7. Kothari MJ, Heistand M, Rutkove SB. Three ulnar nerve conduction studies in patients with ulnar neuropathy at the elbow. Arch Phys Med Rehabil 1998;79:87–9. Kothari MJ, Preston DC. Comparison of the flexed and extended elbow positions in localizing ulnar neuropathy at the elbow. Muscle Nerve 1995;18:336–40. Lo YL, Dan YF, Lee MP, Ratnagopal P. Segmental mixed nerve conduction studies in ulnar neuropathy at the elbow. J Clin Neurophysiol 2001;18:456–9. McLeod JG, Wray SH. Conduction velocity and fibre diameter of the median and ulnar nerves of the baboon. J Neurol Neurosurg Psychiatry 1967;30(3):240–7. Medical Research Council. Aids to examination of the peripheral nervous system. London: HMSO; 1976. Merlevede K, Theys P, Van Hees J. Diagnosis of ulnar neuropathy: a new approach. Muscle Nerve 2000;23:478–81. Neary D, Eames RA. The pathology of ulnar nerve compression in man. Neuropathol Appl Neurobiol 1975;1:69–88. Novak CB, Lee GW, Mackinnon SE, Lay L. Provocative testing for cubital tunnel syndrome. J Hand Surg [Am] 1994;19(5):817–20. Olney RK, Budingen HJ, Miller RG. The effect of temporal dispersion on compound muscle action potential area in human peripheral nerve. Muscle Nerve 1987;10:728–33.
Padua L, Aprile I, Caliandro P, Foschini M, Mazza S, Tonali P. Natural history of ulnar entrapment at elbow. Clin Neurophysiol 2002;113:1980–4. Padua L, Aprile I, Mazza O, Padua R, Pietracci E, Caliandro P, et al. Neurophysiological classification of ulnar entrapment across the elbow. Neurol Sci 2001;22:11–6. Raynor EM, Shefner JM, Preston DC, Logigian EL. Sensory and mixed nerve conduction studies in the evaluation of ulnar neuropathy at the elbow. Muscle Nerve 1994;17:785–92. Ritchie JM. On the relation between fibre diameter and conduction velocity in myelinated nerve fibres. Proc R Soc Lond B Biol Sci 1982;217(1206):29–35. Rosati M, Martignoni R, Spagnolli G, Nesti C, Lisanti M. Clinical validity of the elbow flexion test for the diagnosis of ulnar nerve compression at the cubital tunnel. Acta Orthop Belg 1998;64(4):366–70. Stewart JD. The variable clinical manifestations of ulnar neuropathies at the elbow. J Neurol Neurosurg Psychiatry 1987;50:252–8. Sunderland S. Nerves and nerve injuries. Edinburgh: Churchill Livingstone; 1978. Tackmann W, Vogel P, Kaeser HE, Ettlin T. Sensitivity and localizing significance of motor and sensory electroneurographic parameters in the diagnosis of ulnar nerve lesions at the elbow. J Neurol 1984;231:204–11. Watchmaker GP, Gumucio CA, Crandall E, Vannier MA, Weeks PM. Fascicular topography of the median nerve: a computer based study to identify branching patterns. J Hand Surg [Am] 1991;16A:53–9. Waxman SG, Bennett MV. Relative conduction velocities of small myelinated and non-myelinated fibres in the central nervous system. Nat New Biol 1972;238(85):217–9.
IN-RATIO: A new test to increase diagnostic sensitivity in ulnar nerve entrapment at elbow P. Caliandroa,b,*, M. Foschinia, C. Pazzagliaa, G. La Torrec, I. Aprileb, G. Granataa, P. Tonalia,b, L. Paduaa,b a
Institute of Neurology, Fondazione Pro Iuventute Don Carlo Gnocchi, Universita` Cattolica, Largo Francesco Vito 1, 00168 Roma, Italy b Fondazione Don Carlo Gnocchi ONLUS, Roma, Italy c Epidemiology and Biostatistics Unit, Institute of Hygiene, Universita` Cattolica, Rome, Italy Accepted 1 March 2008 Available online 2 May 2008
Abstract Objective: Motor conduction velocity may yield false-negative results in mild ulnar nerve entrapment at elbow (UNE). There is evidence that the clinical heterogeneity of UNE may be due to the different involvement of fascicles. We hypothesized that, if fibres to FDI are more damaged than fibres to ADM, a relative slowing of motor conduction velocity (CV) at the segment across the elbow recording from FDI (FDI-CV) versus CV at the same segment recording from ADM (ADM-CV) would occur. Methods: We calculated the ratio between FDI-CV and ADM-CV (IN-RATIO) in 60 consecutive UNE patients, 40 norms, and 16 patients with lower cervical radiculopathy. The UNE sample consisted of (1) patients with neurophysiological evidence (UNE NF+), (2) patients without neurophysiological evidence (UNE NF ). We evaluated the possible usefulness of the IN-RATIO to increase sensitivity in diagnosing UNE. Results: The IN-RATIO was lower in the UNE NF than in norms (p < 0.001) and cervicobrachialgia sample (p = 0.02). We found that if the IN-RATIO is 60.97, its sensitivity in diagnosing UNE NF is 68%, specificity 91%. Conclusions: Fibres to FDI have a higher susceptibility to damage than fibres to ADM, and the IN-RATIO is a useful neurophysiological parameter to increase diagnostic sensitivity. Significance: Our results demonstrate the usefulness of a new neurophysiological parameter in improving diagnosis of UNE. Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. Keywords: Entrapment of ulnar nerve; Diagnosis; Neurography
1. Introduction Ulnar nerve entrapment at the elbow (UNE) is a common entrapment syndrome. The clinical picture ranges from intermittent paresthesias in the fourth and fifth digits to complete sensory loss in the territory of the ulnar nerve with atrophy and weakness of the ulnar muscles (Bhala, 1976; Tackmann et al., 1984; Beekman et al., 2004a). The most frequently impaired muscles are the first dorsal inter* Corresponding author. Address: Institute of Neurology, Fondazione Pro Iuventute Don Carlo Gnocchi, Universita` Cattolica, Largo Francesco Vito 1, 00168 Roma, Italy. Tel.: +39 06 30154435; fax: +39 06 35501909. E-mail address: [email protected] (P. Caliandro).
osseous (FDI) and the abductor digiti minimi (ADM); involvement of flexor digitorum profundus (FDP) or flexor carpi ulnaris (FCU) is much less frequent (Stewart, 1987). Some evidence suggests that this clinical heterogeneity is due to the different involvement of fascicles within the nerve (Neary and Eames, 1975). In the region of the elbow nerve fibres from the terminal digital sensory branch and to the muscles of the hand lie adjacent to bone, and this position may render them more susceptible to damage (Sunderland, 1978). These clinical aspects could influence the electrophysiological findings in UNE patients. The goal of electrodiagnosis is to confirm ulnar nerve dysfunction across the elbow, and to assess the severity of nerve impairment (Bhala, 1976; Tackmann et al., 1984; Padua et al.,
1388-2457/$34.00 Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2008.03.007
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
2001; Beekman et al., 2004a). Slowed motor conduction velocity (CV) is the most common localizing electrophysiological sign (Kothari et al., 1998), although it may yield false-negative results in patients with mild lesions (Padua et al., 2001). Some authors note that sensitivity for detecting motor slowing across the elbow is higher when recording from FDI than from ADM, due to differential fascicular involvement (Stewart, 1987; Kothari et al., 1998). Other authors found that the sensitivity of motor conduction studies is similar for ADM and FDI recordings although the extra yield of recording from FDI in addition to ADM is 9% (Tackmann et al., 1984; Beekman et al., 2004a). We hypothesized that, if in UNE patients fibres to FDI are more damaged than fibres to ADM, a relative slowing of motor CV at the segment across the elbow recording from FDI (FDI-CV) versus CV at the same segment recording from ADM (ADM-CV) would occur. To verify this hypothesis we calculated the ratio between FDI-CV and ADM-CV in a sample of UNE patients and in a control group composed of norms and patients with lower cervical radiculopathy. We called this ratio the intra-nerve ratio (IN-RATIO). If FDI-CV were found to be relatively slower than ADM-CV in UNE patients as compared to the control group, we would have verified the usefulness of the IN-RATIO as a new neurophysiological parameter to increase sensitivity in diagnosing UNE. 2. Materials and methods 2.1. Clinical criteria We performed a case-control study with prospective gathering of data between January 2005 and February 2007, evaluating 60 consecutive patients (mean age 53, SD: ±15.52, 36 male, 24 female) with symptoms suggesting UNE. Patients were recruited in the neurophysiology laboratory of the university hospital A. Gemelli. In all of these patients the clinical diagnosis was made when two or more of the following criteria were present: (1) history of paresthesias and/or numbness at the fourth and fifth digit related to prolonged flexion of the elbow; (2) positive provocative test performed by prolonged (1 min) passive forced flexion of the elbow (test was positive when sensory symptoms at the fourth and fifth digit appeared or increased) (Novak et al., 1994; Rosati et al., 1998); (3) reduced motor function of ulnar innervated hand and forearm muscles, quantified according to the British Medical Research Council (BMRC) scale (Medical Research Council, 1976); (4) atrophy of the hypothenar eminence; (5) reduced sensory function focused on the ulnar hand region, evaluated through cotton wool. A skilled neurologist always examined the patient before neurophysiological evaluation was performed, and established whether the patient clinically presented UNE. The clinical picture was considered the gold standard to evaluate the neurophysiological findings. Historical/objective findings were also used to divide the UNE patients into two clinical groups:
1601
(1) patients with exclusively sensory symptoms; (2) patients with sensory–motor symptoms. Exclusion criteria were (1) previous surgery for UNE; (2) acute traumatic origin; (3) clinical or electrophysiological signs of polyneuropathy, including hereditary neuropathy with liability to pressure palsies. Patients were also excluded when electrodiagnostic studies (or additional magnetic resonance imaging) gave evidence of a radiculopathy, plexopathy, or ulnar neuropathy at Guyon’s canal. We also evaluated a control group composed by 40 norms (mean age 51, SD: ±18.09, 15 male, 25 female) and 16 patients with lower cervical radiculopathy (mean age 55, SD: ±12.35, 5 male, 11 female). In these patients the diagnosis was clinical, neurophysiological and radiological. 2.2. Neurophysiological studies We adopted an electrophysiological protocol similar to that proposed by the AAEM (American Association of Electrodiagnostic Medicine, 1999). During neurophysiological evaluation skin temperature was maintained >31 °C by an infrared lamp. Motor nerve conduction studies were performed with moderate flexion (70° to 90° from horizontal) of the elbow. This position was chosen because it provides the greatest correlation between surface skin measurement and true nerve length (Checkles et al., 1971; Kincaid et al., 1986; Olney et al., 1987; Kothari and Preston, 1995; American Association of Electrodiagnostic Medicine, 1999). The ulnar nerve was stimulated by surface electrodes at the proximal wrist crease, 3 cm distal to the medial epicondyle (below elbow-BE), and 6–7 cm above this level (above elbowAE); compound muscle action potentials (CMAPs) were recorded from ADM and FDI, using surface electrodes (American Association of Electrodiagnostic Medicine, 1999). The elbow position was maintained the same when recording from FDI and ADM. Anomalies of innervation, e.g. Martin-Gruber anastomosis, were always verified. When anomalies of innervation had been excluded, the technician calculated the IN-RATIO for the UNE patients, and for the control subjects. Subjects with anastomosis were excluded from the study. We also calculated the difference in motor CV between the AE-to-BE segment and the BE-to-wrist segment for all UNE and control samples. This difference was calculated when recording both from FDI and ADM. An AE-to-BE CV more than 10 m/s slower than the BE-to-wrist CV was considered pathological (American Association of Electrodiagnostic Medicine, 1999). The cutoff values for both FDI-CV and ADM-CV in the segment across the elbow were calculated at the 97th (third) percentile of the observed values in the control group. FDICV in the AE-to-BE segment was considered pathological when it was <43 m/s; ADM-CV was considered pathological when it was <40 m/s. These cutoff values are comparable to those found by Buschbacher in a large normative
1602
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
study (Buschbacher, 1999) and are confirmed by a ROC curve evaluation we performed to compare patients with clinical UNE versus the control group (for FDI-CV area under the curve is 0.830 and for ADM-CV area under the curve is 0.698). Amplitude reduction across the elbow was considered pathological when greater than 16% (conduction block) (Beekman et al., 2004a). Sensory nerve action potentials (SNAPs) were obtained orthodromically using ring electrodes placed over the fifth digit to stimulate the nerve while recording electrodes were applied to the wrist over the ulnar nerve. The cutoff value for the negative peak amplitude of the ulnar SNAP was calculated at the 97th (third) percentile of the observed values in the control group, and was considered pathological when <5 lV. Sensory and motor neurographies were performed by different technicians without knowing whether they were evaluating a clinical UNE patient, a cervicobrachialgia patient, or a normal subject. The technicians had to set the cursors for latency determination. When severe motor deficit or hypotrophy of ulnar innervated hand muscles occurred, using concentric needle electromyography (EMG) we studied ADM and FDI for fibrillation potentials, motor unit action potential (MUAP) amplitude, and recruitment pattern. FCU was evaluated in only a few cases. MUAP amplitude was manually assessed and considered pathological when greater than 2 mV (Beekman et al., 2004a). In addition to these tests, to exclude sub-clinical polyneuropathy we performed motor neurography of peroneal and median nerves, sensory neurography of radial nerve in the first digit-wrist segment, sensory neurography of median nerve in the first and third digit-wrist segments, and sensory neurography of sural nerve. We divided the UNE sample into two groups: (1) patients with clinical ulnar neuropathy at elbow and pathological motor CV across the elbow (UNE NF+); (2) patients with clinical ulnar neuropathy but without pathological motor CV across the elbow (UNE NF ). The patient was considered as having neurophysiological evidence of UNE when either FDI-CV or ADM-CV across the elbow was pathological. UNE patients were also graded according to a previously published neurophysiological classification (Padua et al., 2001). 2.3. Statistical analysis Statistical analysis was conducted using the SPSS package for Windows, release 12.0. The Kolmogorov–Smirnov probability test was used to assess distributions for normality. Among the neurophysiological parameters, only the distribution of FDI-CV in the control group sample was normal, therefore we decided to use the Mann–Whitney U test to compare the central tendency measures (medians) in the two groups. We used the ROC curve to establish the IN-RATIO cutoff value and we compared UNE NF group versus the control group (composed by norms and patients with lower cervical radiculopathy). In order to evaluate the difference among the UNE NF+ sample,
UNE NF sample and the control group concerning the frequency of positive IN-RATIO (considered as dichotomous variable), we used the standard Pearson’s chi-square test (2 2 table), which was also used to evaluate the frequency of pathological FDI-CV and pathological ADMCV (considered as dichotomous variables) in patients with exclusively sensory symptoms and in those with concomitant sensory and motor symptoms. To calculate specificity of the neurophysiological tests, we always used the control group composed by norms and patients with lower cervical radiculopathy. 3. Results 3.1. Clinical picture We examined 59 arms in 53 patients with clinical diagnosis of UNE (6 patients had bilateral symptoms) and excluded seven arms with Martin-Gruber anastomosis. Exclusively sensory symptoms at the fourth and fifth digits were present in 34 arms (58%), and concomitant sensory and motor symptoms in 25 arms (42%). In arms with sensory–motor symptoms, we found clinical weakness at both FDI and ADM in 20 arms, exclusively at ADM in 3 arms, and exclusively at FDI in 2 arms. Only in 5 arms was FCU also weak. Among arms with weakness of ulnar innervated muscles, in 7 arms paresthesia/numbness was exclusively referred in the fourth and fifth digits (none presented CTS), while in 18 arms there was paresthesia/numbness in the hand as a whole; in all of these arms neurophysiological evidence of CTS was revealed. No patient had motor deficit without sensory symptoms. We found hypoesthesia in the palmar region of the fourth and fifth digits, evaluated through cotton wool, in only 12 arms (7 cases presented weakness at both ADM and FDI, 1 exclusively at FDI and 1 exclusively at ADM; 3 arms had no weakness). 3.2. Neurophysiological picture In the 59 examined arms with UNE, in 25 both FDI-CV and ADM-CV were pathological in the AE-to-BE segment, in 5 only FDI-CV was pathological, while in 4 only ADMCV was pathological. For the cutoff value of FDI-CV <43 m/s in the segment across the elbow, sensitivity was 51% and specificity was 100%; for the cutoff value of ADM-CV <40 m/s in the same segment, sensitivity was 49% and specificity was 100%. Evaluation of both FDICV and ADM-CV increases the sensitivity of the neurophysiological evaluation to 58%. According to a previously published neurophysiological classification (Padua et al., 2001), 25 arms were UNE NF , 10 arms were UNE with mild involvement, 17 with moderate involvement, 7 with severe involvement and none with extreme involvement. Three arms presented an atypical neurophysiological pattern with normal FDI-CV and normal ADM-CV but with reduced amplitude of ulnar SNAP; these were included in
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
the UNE NF sample. Seven arms presented conduction block associated with absolute slowing across the elbow recording from both FDI and ADM. When in addition to the absolute reduction in FDI-CV and ADM-CV we use the AAEM criteria (drop in CV between AE-to-BE and BE-to-wrist segments), the overall sensitivity of the neurophysiological evaluation increases to 73% considering the entire sample of patients (UNE NF+ and UNE NF ) with a specificity of 82% considering the entire control group (norms and patients with lower cervical radiculopathy). Nine patients of the UNE NF sample presented a pathological drop in CV between AE-to-BE and BE-towrist segments. We found that FDI-CV and ADM-CV across the elbow were never normal in arms with concomitant sensory and motor symptoms (p < 0.001 for both FDI-CV and ADMCV) (Table 1). About 25 patients of the UNE NF sample, 22 subjects met criteria 1 and 2 while 3 subjects met criteria 1, 2, and 5. Both FDI-CV and ADM-CV were significantly slower in the UNE NF+ sample compared with the UNE Table 1 Distribution of UNE NF+ and UNE NF clinical criteria
patients according to the
Clinical Picture
Conventional test
P
UNE NF+
UNE NF
Sens/motor Sens Tot
25 (42%) 9 (15%) 34 (45%)
0 (0%) 25 (43%) 25 (55%)
<0.001a
Sens/motor, sensory and motor symptoms; Sens, sensory symptoms. a Chi-square test.
1603
NF sample, normal subjects, and the cervicobrachialgia sample (p < 0.001 for both FDI-CV and ADM-CV); no other difference was found (Fig. 1). EMG evaluation was performed in 10 arms. In 8 we found fibrillation potentials at FDI and in 7 at ADM. Pathological recruitment was found at both FDI and ADM in all 10 arms. In two arms we found pathological recruitment from FDI and ADM without increased MUAP amplitude; in these two arms we also evaluated FCU, and found fibrillation potentials and pathological recruitment with increased MUAP amplitude. 3.3. IN-RATIO evaluation In the UNE NF sample IN-RATIO was lower than that we found in norms (p < 0.001) and in the cervicobrachialgia sample (p = 0.02). No other statistical difference relative to IN-RATIO was found among the groups (Fig. 2). We calculated the cutoff value for IN-RATIO by comparing the UNE NF and the control group (norms and cervicobrachialgia sample), and found that if IN-RATIO is 60.97, then its sensitivity in diagnosing UNE NF is 68%, specificity 91%. Fig. 3 shows the ROC curve evaluation for IN-RATIO; the area under the curve is 0.77. Table 2 shows the distribution of subjects with IN-RATIO 60.97 and INRATIO >0.97. IN-RATIO 60.97 was more frequent in the UNE NF sample than in normal subjects, and cervicobrachialgia sample (p < 0.001), IN-RATIO 60.97 was more frequent in the UNE NF+ sample than in normal subjects (p < 0.001) and in cervicobrachialgia sample (p = 0.002).
Fig. 1. FDI-CV and ADM-CV across the elbow in the UNE NF+ sample, the UNE NF cervical radiculopathy (Cerv).
sample, normal subjects, and the sample of patients with lower
1604
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
Fig. 2. IN-RATIO in the UNE NF+ sample, the UNE NF
sample, normal subjects, and the sample of patients with lower cervical radiculopathy (Cerv).
When in addition to the absolute reduction in FDI-CV and ADM-CV we use the IN-RATIO, the overall sensitivity is 86%, specificity is 91%. 4. Discussion
Fig. 3. The ROC curve evaluation for IN-RATIO.
Table 2 Distribution of arms with positive and negative IN-RATIO in the four groups studied Samples
IN-RATIO Positive (60.97)
Negative (>0.97)
UNE NF+ UNE NF Norms Cerv
15 (44%) 17 (68%) 4 (10%) 1 (6%)
19 8 36 15
a
Chi-square test.
P
(56%) (32%) (90%) (94%)
<0.001a
In UNE diagnosis electrophysiological evaluation is crucial for localizing nerve damage at the elbow and for assessing the degree of nerve involvement. Nowadays motor nerve conduction studies are considered the most useful tests for localizing nerve injury (Kothari et al., 1998; American Association of Electrodiagnostic Medicine, 1999); however, these tests may not reveal milder nerve lesions. There is no internationally accepted agreement concerning the sensitivity and specificity of electrodiagnostic studies in diagnosing UNE. Reported sensitivity ranges from 37% to 86%, and specificity ranges from 95% upwards (American Association of Electrodiagnostic Medicine, 1999). Some authors have attempted to find the single most sensitive and specific electrodiagnostic test for UNE (Raynor et al., 1994; Merlevede et al., 2000; Lo et al., 2001); however, to our knowledge no international agreement has been reached. In our study sensitivity is 58% and specificity is 100% when both FDI-CV and ADM-CV are evaluated. Our sensitivity is lower than that reported elsewhere (Bhala, 1976; Tackmann et al., 1984; Beekman et al., 2004a), but the specificity is the best possible. We think that this difference is mainly due to the clinical characteristics of the patients enrolled. In fact, we included also patients without sensory and motor deficits, while in the studies just mentioned (Bhala, 1976; Tackmann et al., 1984; Beekman et al., 2004a) patients without abnormalities at physical
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
examination were excluded, and this may have increased the neurophysiological sensitivity. Our data provide further evidence that the AAEM criteria are useful and allow the neurophysiological diagnosis of UNE when an absolute reduction of FDI-CV and ADMCV across the elbow is absent. We considered the clinical picture as the gold standard to evaluate the neurophysiological findings. Although that approach may cause some difficulties in localizing the exact site of damage, we believe that the combination of inclusion and exclusion criteria may adequately prevent misdiagnosis and avoid inclusion of patients with pathologies that may mimic UNE (hereditary neuropathy with liability to pressure palsies, radiculopathy, plexopathy, or ulnar neuropathy at Guyon’s canal). We did not perform a follow-up evaluation to confirm UNE clinical diagnosis because we used to give instructions to the patient concerning the anatomical condition that causes the damage and the position in which the nerve is more compressed. We give information also when UNE diagnosis is not neurophysiologially confirmed and other causes of damage are clinically and instrumentally excluded (Padua et al., 2002). This approach may determine remission/improvement of symptoms and therefore the exclusion of actual UNE patients (Padua et al., 2002; Beekman et al., 2004b). Much evidence is reported concerning the fascicular organization of peripheral nerves. Sunderland described, in the distal part of a nerve, a clear somatotopic organization of nerve fibres clustered together into individual fascicles, while in the proximal region there is a kind of plexiform organization, with apparent intermingling of nerve fibres and fascicles (Sunderland, 1978). Other authors have pointed out that in spite of the plexiform structure, groups of functionally related nerve fibres can be identified at the proximal site of a nerve (Chow et al., 1985; Watchmaker et al., 1991). In other words, functionally related nerve fibres remain together even if they crisscross from one fascicle to another. We hypothesized that, if in UNE patients the fibres to FDI are more damaged than the fibres to ADM, the IN-RATIO can reveal a relative slowing of FDI-CV compared to ADM-CV in the segment across the elbow. Our neurophysiological findings show that in normal subjects FDI-CV is faster than ADM-CV. Since conduction velocity of each normal axon depends on the external nerve fibre diameter (Gasser and Grundfest, 1939; Hursh, 1939; McLeod and Wray, 1967; Waxman and Bennett, 1972; Ritchie, 1982), a possible explanation is that the fastest fibres to FDI have a larger diameter than the fastest fibres to ADM. In patients group, fibres to FDI are more impaired than fibres to ADM and this finding is mainly manifest when FDI-CV and ADM-CV across the elbow are within the normal range. In the UNE NF+ sample impairment of fibres to FDI and fibres to ADM is more uniform although a prevalent susceptibility of FDI fibres persists. A possible explanation of these findings is that for somatotopic characteristics of ulnar nerve, fibres to
1605
FDI lie adjacent to bone and therefore are more susceptible to damage. When the nerve damage is greater, a larger number of fibres are of interest and from these also fibres to ADM. IN-RATIO allows to compare the function of two different portions of a nerve within the same segment. This approach allows to eliminate all the variability typical of segmental and comparative techniques in which two different segments of one nerve or two different nerves are compared. In addition, although the accuracy of latency measurement is very important in the evaluation of INRATIO, this is independent of the absolute values of FDI-CV and ADM-CV, and therefore each laboratory may use its own normal values for these CVs to diagnose UNE. Our data demonstrate that IN-RATIO allows to increase the sensibility of electrodiagnosis in UNE although its specificity is lower than that we found for absolute reduction of motor CV across the elbow. However, considering that in UNE the global reliability of neurophysiological studies is not so high as in carpal tunnel, for example, we think that IN-RATIO may be a useful instrument in association with the clinical picture. In conclusion, our findings demonstrate that fibres to FDI have a higher susceptibility to damage than fibres to ADM and that IN-RATIO is a useful neurophysiological parameter to increase sensitivity in patients with clinical UNE. We think that when IN-RATIO is positive and FDI-CV and ADM-CV across the elbow are within the normal range, the patient may be classified as having a ‘‘minimal UNE”. Acknowledgments Thanks to Mr. Kim H. Wessling for language editing. The authors report no conflicts of interest. References American Association of Electrodiagnostic Medicine. Practice parameter for electrodiagnostic studies in ulnar neuropathy at the elbow: summary statement. American Association of Electrodiagnostic Medicine, American Academy of Neurology, American Academy of Physical Medicine and Rehabilitation. Muscle Nerve 1999;22:408–11. Beekman R, Van Der Plas JP, Uitdehaag BM, Schellens RL, Visser LH. Clinical, electrodiagnostic, and sonographic studies in ulnar neuropathy at the elbow. Muscle Nerve 2004a;30:202–8. Beekman R, Wokke JH, Schoemaker MC, Lee ML, Visser LH. Ulnar neuropathy at the elbow: follow-up and prognostic factors determining outcome. Neurology 2004b;63:1675–80. Bhala RP. Electrodiagnosis of ulnar nerve lesions at the elbow. Arch Phys Med Rehabil 1976;57:206–12. Buschbacher RM. Ulnar nerve motor conduction to the abductor digiti minimi. Am J Phys Med Rehabil 1999;78:S9–S14. Checkles NS, Russakov AD, Piero DL. Ulnar nerve conduction velocity effect of elbow position on measurement. Arch Phys Med Rehabil 1971;52:362–5. Chow JA, Van Beek AL, Meyer DL, Johnson MC. Surgical significance of the motor fascicular group of the ulnar nerve in the forearm. J Hand Surg [Am] 1985;10A:867–72. Gasser HS, Grundfest H. Axon diameters in relation to the spike dimensions and the conduction velocity in mammalian A fibers. Am J Physiol 1939;127:393–414.
1606
P. Caliandro et al. / Clinical Neurophysiology 119 (2008) 1600–1606
Hursh JB. Conduction velocity and diameter of nerve fibres. Am J Physiol 1939;127:131–9. Kincaid JC, Phillips LH, Daube JR. The evaluation of suspected ulnar neuropathy at the elbow: normal conduction study values. Arch Neurol 1986;43:44–7. Kothari MJ, Heistand M, Rutkove SB. Three ulnar nerve conduction studies in patients with ulnar neuropathy at the elbow. Arch Phys Med Rehabil 1998;79:87–9. Kothari MJ, Preston DC. Comparison of the flexed and extended elbow positions in localizing ulnar neuropathy at the elbow. Muscle Nerve 1995;18:336–40. Lo YL, Dan YF, Lee MP, Ratnagopal P. Segmental mixed nerve conduction studies in ulnar neuropathy at the elbow. J Clin Neurophysiol 2001;18:456–9. McLeod JG, Wray SH. Conduction velocity and fibre diameter of the median and ulnar nerves of the baboon. J Neurol Neurosurg Psychiatry 1967;30(3):240–7. Medical Research Council. Aids to examination of the peripheral nervous system. London: HMSO; 1976. Merlevede K, Theys P, Van Hees J. Diagnosis of ulnar neuropathy: a new approach. Muscle Nerve 2000;23:478–81. Neary D, Eames RA. The pathology of ulnar nerve compression in man. Neuropathol Appl Neurobiol 1975;1:69–88. Novak CB, Lee GW, Mackinnon SE, Lay L. Provocative testing for cubital tunnel syndrome. J Hand Surg [Am] 1994;19(5):817–20. Olney RK, Budingen HJ, Miller RG. The effect of temporal dispersion on compound muscle action potential area in human peripheral nerve. Muscle Nerve 1987;10:728–33.
Padua L, Aprile I, Caliandro P, Foschini M, Mazza S, Tonali P. Natural history of ulnar entrapment at elbow. Clin Neurophysiol 2002;113:1980–4. Padua L, Aprile I, Mazza O, Padua R, Pietracci E, Caliandro P, et al. Neurophysiological classification of ulnar entrapment across the elbow. Neurol Sci 2001;22:11–6. Raynor EM, Shefner JM, Preston DC, Logigian EL. Sensory and mixed nerve conduction studies in the evaluation of ulnar neuropathy at the elbow. Muscle Nerve 1994;17:785–92. Ritchie JM. On the relation between fibre diameter and conduction velocity in myelinated nerve fibres. Proc R Soc Lond B Biol Sci 1982;217(1206):29–35. Rosati M, Martignoni R, Spagnolli G, Nesti C, Lisanti M. Clinical validity of the elbow flexion test for the diagnosis of ulnar nerve compression at the cubital tunnel. Acta Orthop Belg 1998;64(4):366–70. Stewart JD. The variable clinical manifestations of ulnar neuropathies at the elbow. J Neurol Neurosurg Psychiatry 1987;50:252–8. Sunderland S. Nerves and nerve injuries. Edinburgh: Churchill Livingstone; 1978. Tackmann W, Vogel P, Kaeser HE, Ettlin T. Sensitivity and localizing significance of motor and sensory electroneurographic parameters in the diagnosis of ulnar nerve lesions at the elbow. J Neurol 1984;231:204–11. Watchmaker GP, Gumucio CA, Crandall E, Vannier MA, Weeks PM. Fascicular topography of the median nerve: a computer based study to identify branching patterns. J Hand Surg [Am] 1991;16A:53–9. Waxman SG, Bennett MV. Relative conduction velocities of small myelinated and non-myelinated fibres in the central nervous system. Nat New Biol 1972;238(85):217–9.