The Journal of Nutrition Vol. 129 No. 1 January 1999, pp. 207-213

Fish Oil Supplementation Prevents Diabetes-Induced Nerve Conduction Velocity and Neuroanatomical Changes in Rats^1,2

Alain Gerbi^{*, , 3}, Jean-Michel Maixent, Jean-Luc Ansaldi^**, Michele Pierlovisi^*, Thierry Coste^*, Jean-Francois Pelissier, Philippe Vague^*, and Denis Raccah^*

^* Diabetology Department, CHU Timone, 13385 Marseille Cedex 5, France,  Cardiology Department, School of Medicine, I.F.R. Jean Roche, Bd Pierre Dramard 13015 Marseille, France, ^** Electron Microscopic Department, CHU Timone, 13385 Marseille Cedex 5, France and  Neuropathology Department, School of Medicine, 13385 Marseille Cedex 5, France

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Diabetic neuropathy has been associated with a decrease in nerve conduction velocity, Na,K-ATPase activity and characteristic histological damage of the sciatic nerve. The aim of this study was to evaluate the potential effect of a dietary supplementation with fish oil [(n-3) fatty acids] on the sciatic nerve of diabetic rats. Diabetes was induced by intravenous streptozotocin injection. Diabetic animals (n = 20) were fed a nonpurified diet supplemented with either olive oil (DO) or fish oil (DM), and control animals (n = 10) were fed a nonpurified diet supplemented with olive oil at a daily dose of 0.5 g/kg by gavage for 8 wk. Nerves were characterized by their conduction velocity, morphometric analysis and membrane Na,K-ATPase activity. Nerve conduction velocity, as well as Na,K-ATPase activity, was improved by fish oil treatment. A correlation was found between these two variables (R = 0.999, P < 0.05). Moreover, a preventive effect of fish oil was observed on nerve histological damage [endoneurial edema, axonal degeneration (by 10-15%) with demyelination]. Moreover, the normal bimodal distribution of the internal diameter of myelinated fibers was absent in the DO group and was restored in the DM group. These data suggest that fish oil therapy may be effective in the prevention of diabetic neuropathy.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

The development of experimental diabetic neuropathy is accompanied by the appearance of a variety of metabolic abnormalities in nerve that are generally believed to contribute significantly to the pathogenesis of this disorder. Prominent among these are changes in essential fatty acid metabolism which may be important in the aetiology of peripheral nerve dysfunction (Dines et al. 1993). Diabetic neuropathy is characterized by neuroanatomical changes and by decreased nerve conduction velocity (Clark and Lee 1995). A relationship has been demonstrated between structural lesions and slowed nerve conduction velocity in chronically diabetic animals (Sima et al. 1990 and 1993). The reduced nerve conduction velocity has been associated with a diminished activity of Na,K-ATPase (Das et al. 1976, Greene and Lattimer 1983 and 1984, Greene et al. 1984 and 1985, Raccah et al. 1994). This membrane-bound enzyme is very sensitive to altered membrane environment (Gerbi et al. 1993a, 1994 and 1997a). In a previous study, we demonstrated that fish oil supplementation, rich in (n-3) fatty acids, improved sciatic nerve acitivity of Na,K-ATPase isoenzymes and expression in diabetic rats (Gerbi et al. 1997b). The aim of this study was to extend our biochemical observation with functional and anatomical nerve variables. We have studied the effect of fish oil supplementation on nerve conduction velocity and neuroanatomical lesions in diabetic rats.

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Animals.  Five-week-old male Sprague-Dawley (n = 30) rats weighing ~200 g were randomly divided into three groups of 10. In two groups, diabetes was induced by intravenous injection of streptozotocin (STZ)⁴ at a dose of 60 mg/kg (STZ, Sigma, L'Isle d'Abeau, Chesne, France) diluted immediately before injection in citric acid buffer (0.01 mol/L pH 5.5). In the control group only citric acid buffer was injected. One group of diabetic animals (DM) was fed a nonpurified diet, 3% of lipids, (A04, UAR, Villemoisson-sur-orge, France) supplemented with (n-3) fatty acid-enriched fish oil concentrate (MaxEPA, Pierre Fabre Santé, Castres, France) administered for 8 wk at a daily dose of 0.5 g/kg by gavage. This supplement is rich in eicosapentaenoic acid [20:5 (n-3)] and docosahexenoic acid [22:6 (n-3)]. The other group of diabetic rats (DO) was fed a nonpurified diet supplemented with olive oil at the same dose (Table 1). Diabetic rats were not treated with insulin. The nondiabetic control group (CO) was also fed a nonpurified diet supplemented with olive oil. Olive oil was chosen as the placebo because it does not contain (n-3) fatty acids. Water was given with free access to all groups. All animal treatments adhered strictly to all institutional and national ethical guidelines.

Blood samples were collected regularly from the tip of the tail, and blood glucose was measured with a reagent strip (Reflolux, Boehringer Mannheim, Mannheim Germany).

Fatty acid composition of the oil supplementation.  Fatty acids were analyzed as methyl esters on a Varian Model 3300 gas chromatograph (Varian, les Ulis, France) equpped with a flame ionization detector using a spiravax capillary column (25 m × 0.2 mm internal diameter). The temperature program was 150-210°C at 1.5°C/min. Peak areas from the resulting chromatogram were measured with a Merk model D 2000 integrator (Merk, Nogent, France). Fatty acid methyl esters were prepared according to Hagenfeldt (1966). Nonadecanoic acid (C19:0) was added to the mixture before methylation as internal standard.

Measurement of nerve conduction velocity.  After 8 wk motor nerve conduction velocity was recorded in 10 animals of each group from the left sciatic tibial nerve in temperature-controlled environment under ether anaesthesia. The rectal temperature was maintained at 36-37°C with heating lamp and pad. The left sciatic nerve was stimulated proximally at the sciatic noch and distally at the ankle via bipolar electrodes with supramaximal stimuli (6 mA) done by rectangular pusle, 0.3 ms in duration, from a stimulator at 10 Hz on Neuropack 2 EMG (Nihon Kohden). The muscle action potential was recorded from the first interosseous muscle of the left hindlimb by unipolar pin electrodes. The latencies were measured from the stimulus artifact to the onset of the negative M-wave deflection. The nerve conduction velocity (NCV) was calculated by the ratio distance between the two sites of stimulation in mm divided by the difference between proximal and distal latencies in ms, giving a value for MNCV in meters per second (m/s).

Tissue preparation.  Immediately after decapitation, sciatic nerves were removed, rapidly rinsed with ice-cold saline in <30 s, frozen in liquid nitrogen and stored at 80°C until use. Preparation consisted of isolating a highly Na,K-ATPase-enriched membrane fraction according to a previously described procedure (Maixent et al. 1991). Frozen sciatic nerves (left and right) (200 mg) were homogenized directly in 10 volumes of ice cold buffer containing 20 mmol/L sodium pyophosphate, 0.1 mmol/L phenylmethane sulfonyl fluoride, 1 mmol/L EDTA, 250 mmol/L sucrose, 80 mmol/L KCl and 20 mmol/L imidazol/HCl (pH 7.4 at 25°C) with a polytron PT20 (25 s; setting 7). The homogenate was subfractionated by three sequential differential centrifugations at 120 × g for 5 min, 7000 × g for 15 min and 48,000 × g for 30 min using JA. 20 rotor (Beckman Instruments France, Gagny, France) in the Beckman J2.21 (Beckman Instruments France, Gagny, France). The pellet was resuspended in 100 mmol/L NaCl, 250 mmol/L sucrose and 30 mmol/L imidazol/HCl (pH 7.4 at 25°C) and stored frozen in liquid nitrogen. Protein yield was similar in the CO, DO and DM groups.

Enzymological study.  Na,K-ATPase activity was determined by coupled assays at 37°C with or without ouabain as previously described defined in µmol inorganic phosphate (Pi)·h¹·mg protein¹ (Gerbi et al. 1993b). Assays were carried out with native vesicles.

Morphometric and quantitative analysis of sciatic nerve fibers.  Sciatic nerve specimens were removed after decapitation in nine rats (three of each group: CO, DO, DM). A first part of the specimens was immediatly fixed in 0.05 mol/L phosphate-buffered (30 g/L) glutaradehyde solution for 6 h, at pH 7.4, and then divided in three portions.

For light microscopy, transverse semi-thin sections (1 µm thick) were performed on the first portion and were stained by hematein-phloxin-safroon. The second portion was postfixed in 1% osmium tetroxide, dehydrated in graded concentrations of ethanol and embedded in araldite; semi-thin sections were cut and stained with paraphenylediamine (PPD). Ultrathin sections were mounted in copper grids, stained with lead citrate and uranyl acetate, and examined with JEOL 1220 electron microscope. The third section, after osmification for 24 h, was infiltrated with glycerol and single nerve fibers were teased out.

Quantitative studies were performed in each case. The density and diameter of myelinated nerve fibers were determined by SAMBA (Systeme d'analyse microphotometrique à balayage automatique) analysis imaging system with a specific program from transverse semi-thin sections stained with PPD. Selected endoneurial areas were ranged from 0.40 to 0.50 mm². The study of unmyelinated fibers was made using an analyzing system for electron microscopy at magnification of 6000. The selected areas were of 0.16-0.23 mm².

Statistical analysis.  Differences among the three groups (CO, DO and DM) were evaluated by one-way analysis of variance followed by multiple comparisons with Scheffé tests using Statview software. Correlations were evaluated by a single linear regression analysis. Values of P < 0.05 were considered significant.

	RESULTS

Abstract Introduction Methods Results Discussion References

Changes in body weight and blood glucose.  Rats given STZ exhibited a marked hyperglycemia and no gain of body weight at the end of 8 wk (Table 2). Fish oil supplementation had no effect on blood glucose levels and body weights in diabetic rats.

Effects of STZ-induced diabetes and fish oil supplementation on the NCV.  Diabetic rats had 28.2% lower NCV than controls (Fig. 1). Fish oil supplementation (group DM) resulted in 13% greater NCV compared to diabetic rats fed olive oil. However, the NCV in DM groups was still significantly lower than the control group (Fig. 1).

View larger version (22K): [in this window] [in a new window]   Fig 1. Nerve conduction velocity (A) and Na,K-ATPase activity (B) in nondiabetic (C) and diabetic (D) rats fed olive oil (O) or fish oil (M). Data are means ± SEM, n = 10. Bars in the histogram not bearing a common superscript letter were significantly different, P < 0.05.

Effects of STZ-induced diabetes and fish oil supplementation on Na,K-ATPase activity.  The Na,K-ATPase activity in the DO group was 68% lower than in CO (P < 0.05, Fig. 1). The fish oil supplementation in the DM group partially prevented the loss of the enzymatic activity (P < 0.05).

A significant correlation (r = 0.999) was found between diabetes and fish oil related changes in NCV and Na,K-ATPase (Fig. 2).

View larger version (15K): [in this window] [in a new window]   Fig 2. Relationship between Na,K-ATPase activity and NCV in data of Fig. 1. R = 0.999, P < 0.05

Morphometric and quantitative analysis of sciatic nerve fibers.  In sciatic nerve from the DO group, an endoneurial edema with dissocation of nerve fibers was observed. From 10 to 15% of myelinated fibers exibited axonal degeneration, and some of them were demyelinated (Fig. 3A). This demyelination was segmental in the teasing study compared to control group (Fig. 4A, B). There was no microvascular change compared to CO. Fish oil treatment (DM) prevented all these abnormalities in diabetic rats (Fig. 3B).

View larger version (160K): [in this window] [in a new window]   Fig 3. Photomicrograph (×930) of cross section through sciatic nerve of diabetic (A) and DM (B) rats. Axonal degeneration is evident in the untreated diabetics, while the profile of the DM group is similar to that in nondiabetic COs.

View larger version (67K): [in this window] [in a new window]   Fig 4. Photomicrograph of teasing (×190) of segmental demyelination of a diabetic myelinated nerve fiber (A) and CO group (B).

Electron microscopic study of DO nerve specimens showed degenerative axonal changes (Fig. 5A) and axonal atrophy (Fig. 5B) in abnormal myelinated nerve fibers compared to control (Fig. 5C). Unmyelinated fibers were normal. Fish oil treatment partially prevented these effects.

View larger version (125K): [in this window] [in a new window]   View larger version (155K): [in this window] [in a new window]   Fig 5. Photomicrograph (×10,000) of cross section of sciatic nerve of diabetic (A) degenerative axonal changes in diabetic myelinated nerve fiber by electron microscopic study. (B) Axonal atrophy in diabetic nerve fiber by electron microscopic study. (C) Normal nerve fibers in CO group.

In the quantitative study there was no loss of myelinated fibers in specimens from DO-fed rats, but the bimodal repartition was modified [CO vs. DO (Fig. 6A, B)] and normal in DM (Fig. 6C), exhibiting a loss of the fibers larger than 11 µm, compared to CO. With fish oil treatment the histogram of myelinated nerve fibers showed a bimodal distribution with similar mean internal diameters of 4.77 and 9.39 (CO) vs. 4.34 and 8.49 µm (DM). Histograms of myelin sheath widths revealed similar results in the three groups. The histograms of nerve fiber repartition for unmyelinated fibers were comparable for each group (Fig. 7A-C).

View larger version (20K): [in this window] [in a new window]   Fig 6. Quantitative study of myelinated fibers in CO (A), DO, (B) and DM groups. A bimodal distribution was defined by Gaussian analysis in the CO and DM groups, whereas the distribution was normal in the DO group.

View larger version (31K): [in this window] [in a new window]   Fig 7. Quantitative study of myelinated and unmyelinated fibers in CO (A), DO (B) and DM (C) groups. Fibers larger than 11 µm in both DO and DM groups decreased compared to CO.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Data obtained in this study indicate that (n-3) fatty acid supplementation at a dosage corresponding to currently applied treatment protocols used in humans partially ameliorates diabetes-associated changes in Na,K-ATPase activity and NCV. These effects were associated with attenuated anatomical damage and preventive effect on the bimodal distribution of myelinated fiber diameters. The improved NCV was strongly correlated with Na,K-ATPase activity.

These results support the hypothesis that treatment with (n-3) fatty acids is beneficial for diabetic patients (Malasanos and Stacpool 1991). This hypothesis resulted from experimental and clinical data showing that (n-3) fatty acids could restore hyperglycemia-induced alterations, such as platelet aggregation (Colwell et al. 1983, Szirtes 1970), vasoconstriction (Boulanger et al. 1990, Shimokawa and Vanhoutte 1988), endothelial permeability chemotaxis erythrocyte deformability, blood viscosity and granulocyte/endothelium interactions and elevated plasma triglyceride level (review Malasano and Stacpool 1991, Simopoulos 1991).

Moreover, evidence suggests that dietary fish oils rich in eicosapentaenoic (20:5) and docosahexaenoic (22:6) fatty acids may prevent or delay diabetic complications attributable to microvascular diseases (Jensen et al. 1989). While (n-3) fatty acids are recognized for their antiplatelet actions and effects on lipoprotein metabolism, there is increasing evidence that their vascular protective actions are expressed at the level of the vessel wall after incorporation into the phospholipids of cell membranes (Leaf 1990). This preventive action of fish oil has been observed in cardiomyopathy (Gerbi et al. 1997c) and neuropathy (Gerbi et al. 1997b). Indeed, it was documented that this treatement prevents the loss of activity and expression of Na,K-ATPase isoenzymes. Nevertheless, deleterious effects of fish oil supplementation have been documented in retinopathy (Hammes et al. 1996). In this study, we extend the biochemical study of the treatment of neuropathy by fish oil to functional and anatomical nerve parameters. This is the first time that beneficial effects of fish oil supplementation on the neuroanatomical lesions have been found, although amelioration of altered NCV has been previously suggested (Dines et al. 1993).

Our morphological study confirmed the histological damage induced by diabetes on the sciatic nerve fibers (Dick et al. 1994, Sima et al. 1988 and 1993): endoneurial edema and axonal degeneration with occasional secondary segmental demyelination. The quantitative study of nerve fibers showed clearly that myelin sheath was unaffected by diabetes. In this study, in spite of the lack of microvascular abnormality, the hypoxic hypothesis remains plausible. However among these histological damages, endoneurial edema could result from altered sodium cell gradient related to impairment of Na,K-ATPase activity. In contrast, although we hypothesize that the loss of activity of Na,K-ATPase was associated with the altered membrane environment, the decrease of protein expression (Gerbi et al. 1997b) could result from demyelination. Then the decrease in Na,K-ATPase activity could be early as the result of metabolic abnormality and late as the result of histological damage. Moreover, the decrease in NCV could result from alterations in Na,K-ATPase activity, the membrane environment and histological damage, particularly the loss of myelinated large fibers. Dietary fish oil supplementation for 8 wk almost completely prevented or delayed the histological damage induced by diabetes. These effects were independent of hyperglycemia, as this variable was not affected by the dietary treatment. We have previously shown that dietary fish oil supplementation prevented the membrane alteration of the (n-6) fatty acid incorporation, and the protein expression and activity of the Na,K-ATPase 1 and 3 isoenzymes (Gerbi et al. 1997b). These preventive effects on Na,K-ATPase and histological nerve damage could be associated with the improvment of NCV. The relationships between biochemical, functional and neuroanatomical variables remain unresolved, as is the mechanism by which fish oil treatment prevents or delays the histological damage.

We speculate that fish oil prevents the membrane alteration, and by this mechanism prevents the changes in Na,K-ATPase activity, histological damage and prevents partially the loss of NCV. A second hypothesis is that fish oil acts by a transcriptional effect on Na,K-ATPase gene which induces a prevention in functional and histological parameters.

In conclusion, this nutritional study shows the beneficial effect of fish oil supplementation on diabetes-induced abnormalities of the sciatic nerve and suggests a treatment for diabetic neuropathy.

	FOOTNOTES

Manuscript received 20 February 1998. Initial reviews completed 7 April 1998. Revision accepted 11 August 1998.

	ACKNOWLEDGMENT

We thank G. Pieroni (Inserm U476, Marseille France) for the determination of the fatty acid composition of the different oils used in this study.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Boulanger C., Schini V. B., Hendrickson H., Vanhoutte P. M. Chronic exposure of cultured endothelial cells to eicosapentaenoic acid potentiates the release of endothelium-derived relaxing factor(s). Br. J. Pharmacol. 1990; 99:176-180[Medline]
• Clark C. M., Lee D. A. Prevention and treatment of the complications of diabetes mellitus. New Engl. J. Med. 1995; 332:1210-1217[Free Full Text]
• Colwell J. A., Winocour P. D., Lopes-Virella M., Halushka P. V. New concepts about the pathogenesis of atherosclerosis in diabetes mellitus. Am. J. Med. 1983; 75:67-80[Medline]
• Das P. K., Bray G. M., Aguayo A. J., Rasminsky M. Diminished ouabain sensitive, sodium potassium ATPase activity in sciatic nerves of rats with streptozotocin induced diabetes. Exp. Neurol. 1976; 53:285-288[Medline]
• Dick, P. J., Karnes, J. L., Lais, A., Lofgren, E. P. & Stevens, J. C. (1984) Pathologic alterations of the peripheral nervous system of human. In: Peripheral Neuropathy (Dick, P. J., Thomas, P. K., Lambert, E. H. & Bunge, R., eds.), pp. 760-870. W.B. Saunders, Philadelphia.
• Dines K. C., Cotter M. A., Cameron N. E. Contrasting effects of treatment with -3 and -6 essential fatty acids on peripheral nerve function and capillarization in streptozotocin-diabetic rats. Diabetolgia 1993; 36:1132-1138[Medline]
• Gerbi A., Barbey O., Raccah D., Coste T., Jamme I., Nouvelot A., Ouafik L., Levy S., Vague P., Maixent J-M. Alteration of Na,K-ATPase isoenzymes in diabetic cardiomyoaphy: effect of dietary supplementation with fish oil (n-3 fatty acids) in rats. Diabetologia 1997c; 40:496-505[Medline]
• Gerbi A., Debray M., Maixent J-M., Chanez C., Bourre J-M. Heterogeneous Na sensitivity of Na,K-ATPase isoenzymes in whole brain membranes. J. Neurochem. 1993b; 60:246-252[Medline]
• Gerbi, A., Maixent, J-M., Barbey, O., Coste, T., Pierlovisi, M., Nouvelot, A., Vague, P. & Raccah, D. (1997b) Beneficial effect of omega 3 fatty acids on Na,K-ATPase activity and nerve function in diabetic neuropathy. J. Peripheral Nervous System 2, 261(abs).
• Gerbi A., Maixent J-M., Zérouga M., Berrebi-Bertrand I., Debray M., Chanez C., Bourre J-M. Specific modulation of two neuronal digitalis receptors by anaesthesia. J. Rec. Transd. Sign. 1997a; 17:447-448
• Gerbi A., Zérouga M., Debray M., Durand G., Chanez C., Bourre J-M. Effect of fish oil diet on fatty acid composition of phospholipids of brain membranes and on kinetic properties of Na,K-ATPase isoenzymes of weaned and adult rats. J. Neurochem. 1994; 62:1560-1569[Medline]
• Gerbi. A., Zérouga M., Debray M., Durand G., Chanez C., Bourre J-M. Effect of dietary -linolenic acid on functional characteristic of Na,K-ATPase isoenzymes in whole brain membrane of weaned rats. Biochim. Biophys. Acta 1993a; 1165:291-298[Medline]
• Greene D. A., Lattimer S. A. Impaired rat sciatic nerve sodium-potassium adenosine triphosphatase in acute streptozotocin diabetes and its correction by dietary myo-inositol supplementation. J. Clin. Invest. 1983; 72:1058-1063
• Greene D. A., Lattimer S. A. Impaired energy utilization and Na,K-ATPase in diabetic peripheral nerve. Am. J. Physiol. 1984; 246:E311-E318[Abstract/Free Full Text]
• Greene D. A., Lattimer S. A., Ulbrecht J., Carroll P. Glucose-induced alterations in nerve metabolism: current perspective on the pathogenesis of diabetic neuropathy and future directions for research and therapy. Diabetes Care 1985; 8:290-299[Abstract]
• Greene D. A., Yagihashi S., Lattimer S. A., Sima A.A.F. Nerve Na,K-ATPase, conduction and myo-inositol in the insulin-deficient BB rat. Am. J. Physiol. 1984; 247:E534-E539[Abstract/Free Full Text]
• Hagenfeldt L. A gas chromatographic method for the determination of fatty acids in plasma. Clin. Chim. Acta. 1966; 13:266-268[Medline]
• Hammes H. P., Weiss A., Fuhrer D., Kamer H. J., Papavassilis C., Grimminger F. Acceleration of experimental diabetic retinopathy in the rat by omega 3 fatty acids. Diabetologia 1996; 39:251-255[Medline]
• Jensen T., Stender S., Goldstein K., Holmer G., Deckert T. Partial normalisation by dietary cod liver oil of increased microsvascular albumin leakage in patients insulin-dependent diabetes and albuminuria. N. Engl. J Med. 1989; 321:1572-1577[Abstract]
• Leaf A. Cardiovascular effects of fish oils. Beyond the platelet. Circulation 1990; 82:624-628[Free Full Text]
• Maixent J-M., Fenard S., Kawamoto R. M. Tissue localisation of active Na,K-ATPase isoenzymes by determination of their profil of inhibition with ouabain, digoxin, digitoxigenin and cordil, a new aminosteroid cardiotonic. J. Rec. Res. 1991; 11:687-698
• Malasanos T. H., Stacpool P. W. Biological effects of -3 fatty acids in diabetes mellitus. Diabetes Care 1991; 14:1160-1179[Abstract]
• Raccah, D., Jannot, M. F., Issautier, T. & Vague, P. (1994) Effect of experimental diabetes on Na,K-ATPase activity in red blood cells, peripheral nerve and kidney. Diabète et Metabolisme 20: 271-274.
• Shimokawa H., Vanhoutte P. M. Dietary cod liver oil improve endothelium-dependent responses in hypercholesterolemic and atherosclerotic porcine coronary arteries. Circulation 1988; 78:1421-1430[Abstract/Free Full Text]
• Sima A.A.F., Nathaniel V., McEwen T. A. J., Greene D. A. Histopathological heterogeneity of neuropathy in insulin-dependent and non-insulin dependent diabetes, and demonstration of axo-glial dysjunction in human diabetic neuropathy. J. Clin. Invest. 1988; 81:349-364
• Sima A.A.F., Prashar A., Nathaniel V., Bril V., Werb M. R., Greene D. A. Overt diabetic neuropathy: Repair of axo-glial dysjunction and axonal atrophy by aldose reductase inhibiton and its correlation to improvment in nerve conducton velocity. Diabetic Med. 1993; 10:115-121[Medline]
• Sima A.A.F., Prashar A., Zhang W. X., Chakrabarti S., Greene D. A. The preventive effect of a long term adose reductase inhibition (Ponalrestat) on nerve conduction and sural nerve structure in the spontaneously diabetic BB rat. J. Clin. Invest. 1990; 85:1410-1420
• Simopoulos A. P. Omega-3 fattyacids in health and dsease and in growth and development. Am. J. Clin. Nutr. 1991; 54:438-463[Abstract/Free Full Text]
• Szirtes M. Platelet aggregation in diabetes mellitus. Adv. Cardiol. 1970; 4:179-186[Medline]

0022-3166/99 $3.00 ©1999 American Society for Nutritional Sciences