Med Pregl 2011; LXIV (1-2): 11-14. Novi Sad: januar-februar. 11

ORIGINALNI NAUČNI RADOVI ORIGINAL STUDIES

Clinical Centre of Vojvodina Novi Sad Originalni naučni rad Department of Neurology Original study UDK 616.379-008.64:616.833-073.7 DOI: 10.2298/MPNS1102011C THE SENSITIVITY OF ELECTROMYONEUROGRAPHY IN THE DIAGNOSIS OF DIABETIC SENZITIVNOST ELEKTROMIONEUROGRAFIJE U RANOJ DIJAGNOSTICI DIJABETESNE POLINEUROPATIJE Milan CVIJANOVIĆ, Miroslav ILIN, Petar SLANKAMENAC, Sofija BANIĆ HORVAT and Zita JOVIN Summary – Diabetic polyneuropathy is a complex set of clinical syndromes, which deplete various regions of the nervous system. The process leading to diabetic neuropathy is multi-factorial. Its symptoms are paresthesia, dysesthesia and pain. The signs of damage to the peripheral neurons are hypoesthesia, hypoalgesia, hyperesthesia and hyperalgesia, decreased tendon reflexes, and, possibly, weakness and muscle atrophy. There is no universal classification. Electromyoneurography is indispensable in the diagnosis of dia- betic polyneuropathy. However, there is no agreement on the most sensitive parameter for an early diagnosis. One hundred patients with mellitus were examined in order to investigate the sensitivity of different electromyographic parameters. Electromyo- graphic techniques proved to be entirely sensitive for the early diagnosis of diabetic polyneuropathy. Some of the parameters are more suitable for an early detection of peripheral damage, and others, which are not so sensitive but easy to use and stable, are suitable to follow up the course of diabetic polyneuropathy. Key words: Diabetic Neuropathies; Signs and Symptoms; ; Early Diagnosis; Sensitivity and Specificity Introduction Pathophysiology of diabetic polyneuropathy Diabetic polyneuropathy is a complex set of clini- In agreement with the results of retrospective and cal syndromes, which deplete various regions of the prospective studies suggesting a strong association be- nervous system, either individually or in a combina- tween hyperglycemia and the development and severi- tion [1]. It is the most common complication of diabe- ty of diabetic polyneuropathy (DPN), its prevalence is tes mellitus, leading to major morbidity and mortality highest among people with poorly controlled diabetes. and producing huge costs [2]. A simple and widely Until recently, there were two theories explaining the used clinical definition is that diabetic polyneuropathy mechanism of diabetic neuropathy –metabolic and is the presence of symptoms and/or signs of peripheral vascular. It seems most probable, however, that an in- nervous dysfunction in diabetic patients when other teraction between the two mechanisms leads to neu- possible causes have been excluded. The prevalence of ropathy [8]. More recently, it has become generally neuropathy in diabetes varies widely in literature, be- recognized that the process leading to diabetic neu- tween 0.1% and 100% [3], and most frequently the es- ropathy is multi-factorial [9], although dominated by timates are 10-100% [4]. The wide prevalence rates re- metabolic factors involving the polyol pathway and ported are due to different diagnostic criteria used, vascular disorders. Other factors that are related to different methodological approaches adopted [5] and those previously mentioned, such as accumulation of populations studied [6]. Approximately 10% of pa- end-products of non-enzymatic glycosylation in the tients have polyneuropathy at the time they are diag- nerve and blood vessel proteins have also been recog- nosed with diabetes. The data on such a high preva- nized [10]. Furthermore, decreased n-6-essential fatty lence of polyneuropathy in newly-diagnosed diabetics acids and prostaglandin result in damage to the nerve and data estimating the prevalence rate at 100% result sheath and microvascular and blood abnormalities. In from electrophysiological studies [7]. Since a timely addition, a significant role is played by nitric oxides diagnosis is a prerequisite for the adequate treatment (oxidative stress), decreased neurotrophic factors, in and prevention of more severe damage to peripheral particular nerve growth factor but also neurotrophin–3 neurons, it is of utmost importance to identify the most and insulin-like growth factor, which affects axonal sensitive electromyographic technique for the early transport, decreased protein kinase C activity and im- detection of diabetic polyneuropathy. munological factors. Laminin has also been frequently

Corresponding Author: Doc. dr Milan Cvijanović, Clinic of Neurology, Clinical Centre of Vojvodina, 21000 Novi Sad, Hajduk Veljkova 1-11, E-mail: [email protected] 12 Cvijanović M, et al. Pathophysiology of diabetic polyneuropathy

Abbreviations Subclinical neuropathy DPN – diabetic polyneuropathy Rapid reversible phenomena: CHRD – chronodispersion 1) Distal sensory symptoms SCV – sensory conduction velocity 2) Reduced nerve conduction velocity implicated lately. Laminin, a large heterotrimeric pro- 3) Resistance to ischemia-associated conduction di- tein of basement membranes, appears to have a signif- sturbances icant role in the nerve regulation and manifestations of neuropathy [11]. It promotes neurite extension in cul- Clinically manifest neuropathy tured neurons. An abnormal expression of laminin Focal and multifocal neuropathy: beta 2 gene (laminin is composed of one large alpha 1) Cranial neuropathy and two small beta chains – beta 1 and beta 2) may 2) Thoracoabdominal neuropathy contribute to the pathogenesis of diabetic neuropathy. 3) Focal neuropathy of the extremities 4) Motor asymmetric neuropathy of the proximal Clinical picture of DPN part of lower extremities (diabetic amyotrophy) The most common symptoms of DPN are pare- Symmetric neuropathy sthesias, and dysesthesias, which are described as 1) Sensory/(sensory-motor)/autonomous neuropathy feelings of walking on water, or cotton or wool stro- 2) Proximal motor neuropathy of the lower extremities kes over the soles, or similar. Cramps are also com- mon and usually occur in the lower leg, in particular Electromyoneurography in the diagnosis the calf and toe flexors. Pain is usually described as of diabetic neuropathy sudden and poignant pain resembling piercing by a sharp tool, but also as dull and tormenting, as if The foundations of clinical electromyography were holding legs in the freezing cold water [12]. It may laid by Buchthal and Clemmesen in 1941 [13]. However, occur at any stage of disease and is difficult to ob- this electrophysiological technique, including peripheral jectify. By its characteristics, this small fibre pain nerve studies in diabetic patients, has become more has all characteristics of neuropathic pain. Due to widely used only in the last decades. Early studies fo- its unpleasant manifestations with the paresthetic cused on the analysis of electromyographic findings, modalities, it severely impairs the quality of life and i.e. firing patterns. In diabetic neuropathy, the pattern is frequently resistant to treatment. is usually thin and there is an increased percentage of Impaired vibration sensitivity is frequently report- polyphasic potentials. However, a significantly more ed as a convincing sign of DPN. Hypoesthesia and sensitive parameter is the nerve conduction velocity hypoalgesia also occur and are frequently preceded study, including both sensory and motor nerve fibres, by hyperesthesia and hyperalgesia. Decreased tendon which belongs to the domain of conventional elec- reflexes, almost invariably the Achilles reflex, occur troneurography. Compared to diabetics with no clini- subsequently or simultaneously. Muscle paresis and cal signs of neuropathy, in diabetics with signs of neu- atrophies, more often in the lower extremities, occur ropathy, motor conduction velocities are more de- in advanced disease. These signs are followed by the creased and reduction of the evoked potential ampli- signs of vegetative neuropathy, but the latter is not tude occurs earlier [14]. Tibial and peroneal discussed in the present paper. show more abnormality than ulnar and medial nerves, i.e., the lower extremities are more sensitive to diabetic Classification of diabetic neuropathy damage than the upper extremities. The involvement of peripheral nerves is diffuse, more distal than proxi- Recognition of various clinical syndromes of pe- mal parts. However, some authors have long consid- ripheral neuron damage and their likely overlapping ered that distal latency in axonal neuropathies, includ- prevents us from adopting a universal classification. ing diabetic neuropathy, does not reflect only dysfunc- There have been many attempts at systematizing tion of the distal segment. Namely, the loss of alpha different manifestations, using different criteria, motor neurons leads firstly to decreased motor electric both clinical and topographic, as well as subclinical, conduction velocity in the distal part of the peripheral morphometric and neurophysiological signs of pe- nerve. Sensory conduction velocity is considered a ripheral nerve damage. This has produced a great more sensitive study than the motor conduction velo- number of classifications of diabetic neuropathy. city [15]. Burke et al. consider that sensory conduction However, there are classifications that are widely ac- velocity study of the sural nerve is the most sensitive cepted, such as Thomas’ classification [10], which electrodiagnostic method for the detection of polyneu- differentiates between diffuse symmetrical polyneu- ropathy. More recently, sophisticated techniques for ropathy on one hand, and focal and multifocal neu- studying late responses, including and H re- ropathy on the other. This classification includes flex, have been used. The diagnostic value of these also the criterion of intensity (subclinical and clini- electrophysiological parameters, especially of F wave, cal). This practical classification with minor adapta- has been reported by many authors [16-18]. F wave is tions comprises the following: a late response widely used in clinical practice. It was first described in the annals of Johns Hopkins hospital Med Pregl 2011; LXIV (1-2): 11-14. Novi Sad: januar-februar. 13

Table 1. Results of individual EMNG techniques in 100 studied parts of peripheral motor neurons, in addition to sen- diabetic patients sory conduction velocity and the late responses H re- Tabela 1. Rezultati pojedinačnih EMNG tehnika kod 100 ispiti- flex and F wave in two modalities - minimal latency vanih pacijenata sa dijabetesom (F min) and chronodispersion of F waves. The results FP MCV D.L. SCV F min CHRD H ref were compared with the reference values and classi- I:U MBP D:L: SBP F min CHRD H ref fied into normal or abnormal ones accordingly. A Normal/Normalno 27% 60% 44% 6% 41% 7% 18% higher percentage of abnormal findings of an electro- Abnormal/Abnormalno 73% 40% 56% 94% 59% 93% 82% myographic parameter found in the patient population Total/Ukupno 100%100% 100% 100% 100% 100% 100% indicated a greater sensitivity of the parameter in the early diagnosis of diabetic polyneuropathy. EMNG - electromyoneurography by Magladery and McDougal in 1951. At first, it was Results and Discussion considered to be a polysynaptic reflex, but soon it was proved to be centripetal discharge of alpha motor neu- We found around 73% of abnormal firing pat- rons following antidromic stimulation of the motor fi- terns in our study sample. Among these, 56% of pa- bres. This response reflects the interaction between tients had the abnormal motor conduction velocity the anterior horn cells and the possibility of afferent in the distal part and 40% in the proximal part of the and efferent conduction through the motor fibres, pro- peripheral motor neuron. The abnormal sensory con- viding information on the function of the whole path- duction velocities were found in 94% of subjects. way of the peripheral motor neuron. The fact that the The abnormal CHRD was found in 93% and abnor- pathway is used for inward and outward conduction mal minimal F in 59%. H reflex was abnormal in indicates a range of parameters in both healthy indi- 82% of our subjects. viduals and patients with diseases of the peripheral The most sensitive electromyographic parameter nervous system. Although the minimal F wave latency for the early detection of polyneuropathy was clearly is most frequently used for clinical applications, there sensory conduction velocity (SCV), which can be ex- are authors who consider chronodispersion (CHRD), plained by a smaller diameter of the sensory fibres i.e. the difference between the minimal and maximal and their higher sensitivity to various endogenous and latencies, the most sensitive parameter in neurophysi- exogenous influences. On the other hand, considering ological studies of diabetic neuropathy [19]. The same the early damage to these fibres, which is often an ir- author (Tojokura, 1998.) termed F wave chronodisper- reversible process, this method is not suitable for the sion F duration. He found abnormal values of F wave follow up of the changes in neurogenic lesions. A sur- chronodispersion in patients with normal F wave la- prising sensitivity, i.e. a high percentage of abnormal tencies. Conversely, in 1989 [17] Bernbaum et al. pre- findings, was recorded with CHRD, which was shown ferred a study of H reflex in the electrophysiological a reliable method for the early detection of diabetic diagnosis of diabetic polyneuropathy. The latter meth- polyneuropathy. However, the pronounced inconsist- od enables a better insight into the condition of periph- ency of this F wave modality may be a factor limiting eral neurons, providing information on the reflex ac- the precise detection. On the other hand, although not tivity and thus enabling the evaluation of the function nearly so sensitive in the early detection, F min is a not only of the peripheral nerve but also of superior reliable parameter for following up the course of spinal areas. As early as 1918, Paul Hoffmann de- polyneuropathy. H reflex showed a significant sensi- scribed a reaction that was named after him H reflex. tivity and is an important indicator of the function of Much later, this reaction was analyzed and applied in the whole population of the nerve fibres involved in the practice. Today it is well known that H reflex is a the reflex arc, from sensitive, via transmitters in the result of the sensory fibre stimulation by low-voltage spinal cord, to the motor nerve fibres. It could also be impulses which are transmitted by fibres from motor a suitable parameter to follow up the course of diabet- spindles and which, after monosynaptic delay, lead to ic polyneuropathy. generation of action motor potential in the cells of the anterior horn and their propagation into the muscle tis- Conclusion sue within certain motor units. Although accepted as a highly sensitive sign of S1 , it has recent- The extremely high percentage of subjects with ly been used also in the study of demyelinating dis- diabetic polyneuropathy found in our study indicates eases as well as diabetic neuropathy. the necessity of applying electromyographic tech- niques in the evaluation of diabetic patients before Material and methods the onset of clinical symptoms and signs of neuropa- thy. Whereas some of the electromyographic param- We studied a hundred patients with diabetes melli- eters are more suitable for the early detection of pe- tus lasting for over a year. The firing pattern, motor ripheral nerve damage, others, which are not so sen- conduction velocity and distal latency were analyzed sitive but easy to use and stable, are suitable to fol- as parameters of the function of proximal and distal low up the course of diabetic polyneuropathy. 14 Cvijanović M, et al. Pathophysiology of diabetic polyneuropathy

References 1. Vinik AI. Managment of neuropathy and foot problems in 10. Thomas PK. Metabolic neuropathy. J R Coll Physicians diabetic patients. Clin Cornerst 2003;5(2):38-55. Lond 1973;7:154. 2. Vinik AI, Mitchell BD, Leichter SB, et al. Epidemiology of 11. Rivas RJ. Burneister DWA, Goldberg DJ Rapid effects of the complications of diabetes. In: Leslie RDG, Robbins DC, eds. laminin on the growth cone. Neuron 1992;8:107-15. Diabetes clinical science in practice. Cambridge: Cambridge Uni- 12. Tešić D, Cvijanović M, Pantelinac P. Akutna bolna neuro- versity Press; 1995. p. 221-87. patija. Med Pregl 1995;48(11-12):425-8. 3. Holyer SE, Camerota A, Martens L, et al. Cost and durati- 13. Shields WR. Single fiber electromyography is a sensitive on of care for lower extremity ulcers in patents with diabetes. indicator of axonal degeneration in diabetes. Neurology 1987;37: Clin Ther 1998;20:169-81. 1394-7. 4. Melton LI, Dyck PI. Epidemiology. In: Dyck PI, Thomas 14. Horowitz SH, Ginsberg-Fellner F. Ischemia and sensory PK, Asbury AK, Inegrad AI, Porte ID, eds. Diabetic neuropathy. nerve conduction in diabetes mellitus. Neurology 1979;29:695- Philadelphia: W. B. Sannders; 1987. p. 27-35. 704. 5. Perić Z. Diabetesna neuropatija. Niš: Prosveta; 1993. 15. Jusić A, Tomić M, Fronjek N. Secondary potentials of tri- 6. Young MJ, Boulton AJM, MacLeod AF, et al. A multicen- ceps sural muscles (F, FH and H potentials) in healthy subjects. tre study of the prevalence of diabetic in the Electomyogr Clin Neurophysiol 1986;26:33-9. United Kingdom hospital clinic population. Diabetologia. 1993; 16. Dyck JP, Karnes LJ, O’Brien P, Service J. Clinical and 36:1-5. neuropathological criteria for the diagnosis and staging of diabe- 7. Cvijanović M, Ilin M, Cvijanović SM. Prevalencija dijabete- tic polyneuropathy. Brain 1985;108:861-80. sne polineuropatije u kliničkoj populaciji: elektrofiziološki parame- 17. Bernbaum M, Alberts GS, McGarry DJ. Diabetic neuro- tri. Aktuelnosti iz neurologije, psihijatrije i graničnih područja. pathy and braille ability. Arch Neurol 1989;46:1179-81. 1996;4(2):24-6. 18. Ilin M, Cvijanović M, Cvijanović SM. Hronodisperzija F 8. Cameron NE, Cotter MA. Metabolic and vascular factors in talasa: prilog standardizaciji parametara. Aktuelnosti iz neurolo- the pathogenesis of diabetic neuropathy. Diabetes 1997;46(Suppl): gije, psihijatrije i graničnih područja. 1996;4(1-2)39-41. S31-7. 19. Toyokura M. F wave duration in diabetic polyneuropathy. 9. Ceriello A. New insights on oxidative stres and diabetic Muscle Nerve 1998;21(2):246-9. complications may lead to a «Causal Antioxidant Therapy». Diab Care 2003;26:1589-96.

Sažetak Uvod Materijal i metode Dijabetesna polineuropatija kompleksan je skup kliničkih sindro- U istraživanju je, pomoću raznih elektromiografskih metoda, ispi- ma koji oštećuju različite regione nervnog sistema, bilo pojedinač- tivano stotinu bolesnika s dijabetesom melitusom u trajanja od no ili kombinovano. Ona je najčešća i najneprijatnija komplikacija preko godinu dana. dijabetesa, vodi velikom morbiditetu i mortalitetu i rezultira veli- Rezultati i diskusija kim ekonomskim troškovima. U poslednje vreme je opšteprihvaćen Ispitivanje je izvedeno s više elektromioneurografskih tehnika koje stav da je proces koji dovodi do dijabetesne neuropatije multifak- su se pokazale senzitivnim, a neke, zbog osetljivosti i postojanosti torski, iako u njemu dominiraju metabolički faktori s poremećajem odgovara, i neophodnim u ranoj detekciji i praćenju toka dijabete- poliolskog puta i vaskularni poremećaji. Simptomi su parestezije, sne polineuropatije. dizestezije i bol. Znaci oštećenja perifernog nerva su hipoestezija, Zaključak hipoalgezija, hiperestezija i hiperalgezija, kao i sniženje tetivnih Elektromioneurografija se u celini pokazala neophodnom u ranoj refleksa, te eventualno slabost i atrofija mišića. detekciji i praćenju dijabetesne polineuropatije. Ključne reči: Dijabetesna polineuropatija; Znaci i simptomi; Elektromiografija; Rana Dijagnoza; Senzitivnost i specifičnost Rad je primljen 2. IX 2010. Prihvaćen za štampu 8. III 2010. BIBLID.0025-8105:(2011):LXIV:1-2:11-14.