Brain Research, 535 (1990) 43-48 Elsevier
43
BRES 16100
E n d o n e u r i a l m i c r o e n v i r o n m e n t and acute nerve crush injury in the rat sciatic nerve Douglas W. Zochodne and Lam T. Ho Department of Medicine, Queen's University, Kingston, Ont. (Canada) (Accepted 12 June 1990)
Key words: Peripheral nerve; Crush; Nerve blood flow; Oxygen tension
In severe peripheral nerve ischemia in the rat, serial nerve blood flow (NBF) measurements have identified evidence of 'no reflow', a mechanism of continued fiber damage during reperfusion. It has been postulated that 'no reflow' also occurs in nerve compression due to direct mechanical or ischemic (if compression is prolonged) injury of microvessels, resulting in continuing nerve fiber damage. To address this question, we measured endoneurial blood flow (NBF), oxygen tension and pH at the site of an acute nerve crush injury. In further sets of experiments, NBF and endoneurial oxygen tension were examined before and after prolonged epochs of crush. NBF and MR (microvascular resistance) were not appreciably different than values obtained in control animals without intervening brief nerve crush. NBF was slightly higher and MR slightly lower 2 h after injury, but the difference was not statistically significant. No evidence of significant endoneurial hypoxia or acidosis was observed. Similarly, after more prolonged crush there was no significant oligemia or hypoxia. The studies provide no evidence that 'no reflow' occurs in crush injury even if the injury is maintained for a period of time known to induce 'no reflow' with severe ischemia. We suggest that nerve damage in crush, and possibly compression, more likely arises from direct mechanical injury of fibers.
INTRODUCTION Studies of acute mechanical injury of peripheral nerve by compression or crush have postulated that an important mechanism of injury, perhaps complicating regeneration, is ischemia 6'7A3'21'3°. A c u t e endothelial injury might arise from direct mechanical disruption, ischemia or both. Endothelial swelling, granulocyte plugging 23 and microvascular thrombosis would then result in 'no reflow' and continuing fiber d a m a g e 1. A n additional factor might be e n d o n e u r i a l e d e m a with microvascular compression ~3. T h e r e is evidence that severe peripheral nerve ischemia from vascular ligation damages endothelium, resulting in swelling, luminal narrowing and 'no reflow '2"22. A l t h o u g h compression may result in t e m p o r a r y circulatory arrest it is unclear whether this insult p e r m a n e n t l y injures nerve microvessels. Compressive ischemia, if m a i n t a i n e d long enough, might induce 'no reflow', as in the ligation experiments. If both mechanical injury to microvessels and ischemic endothelial injury were to occur with compression, relatively short periods of injury might result in 'no reflow'. Nerve crush might be considered an e x t r e m e version of compressive nerve injury with an enhanced degree of vasa nervorum disruption.
In this study we m e a s u r e d b l o o d flow, oxygen content and p H in the sciatic nerve of the rat ( N B F ) using endoneurial microelectrodes. To test the hypothesis that acute injury damages vasa n e r v o r u m , we m e a s u r e d NBF, endoneurial oxygen content and p H at the injury site i m m e d i a t e l y and 2 h following segmental mid-sciatic nerve crush. To test the hypothesis that p r o l o n g e d crush is associated with ischemic 'no reflow' we m e a s u r e d N B F and endoneurial oxygen content i m m e d i a t e l y after 2 and 3 h epochs of crush injury. A control group of animals had serial m e a s u r e m e n t s without crush injury. MATERIALS AND METHODS
Animals Animals employed were male Sprague-Dawley rats weighing 200-300 g housed in wire mesh cages with ad libitum access to rat chow and water. Preparation Rats were anesthetized with sodium pentobarbital by intraperitoneal injection (65 mg/kg). A ventral midline neck incision was made for placement of a left carotid intra-arterial line (PE-50; Intramedic; Clay Adams, Parsippany, NJ) connected to a pressure transducer (P23ID Gould; Oxnard, CA) and tracheostomy. The animal was then paralyzed using tubocurarine (1.0 mg/kg intraarterial) (Sigma Chemical Co.; St. Louis, MO) and ventilated (Rodent ventilator 683; Harvard Bioscience; South Natick, MA) through flowmeters (N032-416 and Nl12-026; Cole-Panner, Chi-
Correspondence: D.W. Zochodne, Room 206, La Salle Building, 146 Stuart Street, Department of Medicine, Queen's University, Kingston, Ont., K7L 3N6, Canada. 0006-8993/90/$03.50 (~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
44 cago, IL) to permit regulation of inspired gases. The left sciatic nerve was exposed, covered in mineral oil, and microelectrodes inserted through an epineurial window into the endoneurium. A reference KCl-agar bridge electrode was inserted subcutaneously and sutured into place. The endoneurial microelectrode was polarized at +0.250 V for hydrogen clearance studies and -0.650 V for the oxygen measurements and current detected using a microsensor (Microsensor If, Diamond General; Ann Arbor, MI). The exposed nerve preparation was maintained at 37 °C throughout the experiments using a temperature probe connected to a temperature control and feedback unit with an infrared heating lamp (TH8 and TCAT-1A, Sensortek; Clifton, N J). Additional doses of pentobarbital (20 mg/kg) and D-tubocurarine (0.5 mg/kg) were administered approximately 2 hourly through the intra-arterial catheter. Continuous print out of mean arterial blood pressure (MAP) and the microsensor output was accomplished using two channels of a polygraph recorder (79E, Grass Instruments; Quincy, MA). Arterial blood gas samples were drawn approximately hourly to ensure physiologic stability (Table I).
Experiments The technique of hydrogen clearance for measurement of endoneurial blood flow has been published 28. Measurements of oxygen tension were conducted by endoneurial microelectrodes as previously described zs, but without saturating the bathing mineral oil with 100% N 2. Calibration of the oxygen microelectrode was obtained within 5 mih of all in vivo measurements to prevent error from microelectrode drift. A linear calibration line was obtained by bubbled gas mixtures of 0, 10 and 25% oxygen at 37 °C in a calibration bath (Diamond General; Ann Arbor, MI). In experiments with sampling of oxygen tensions at various depths, a mean tension was calculated for each experimental time point, pH studies were conducted with 80 micron tipped antimony microelectrodes (801, Diamond General; Ann Arbor, MI) connected to an electrometer (microsensor If, Diamond General; Ann Arbor, MI). Two point calibration was conducted at 37 °C after each experiment using Trizma solutions of pH 6.70 and 7.10 (Sigma Chemical Company; St. Louis, MO) - - in vivo pH measurements were obtained from the mV vs pH calibration line. Several groups of animals were studied. In group 1, 12 animals underwent serial HC measurements (9 experiments were technically satisfactory). After an initial HC measurement, the microelectrode was withdrawn and the site of measurement subjected to 30 s of crush applied by the polyethylene-covered jaws of a hemostatic clamp - - the injured area measured 3 mm of nerve length. After removal of the clamp, the microelectrode was reinserted through the original epineurial window and repeat measurements made of hydrogen clearance. A third and final clearance curve was taken at 2 h following crush, pH measurements were made before, immediately after and 2 h following nerve crush. Measurements of oxygen tension were made after the final NBF measurement. In Group 2, 6 animals (5 experiments were technically satisfactory) underwent the same protocol as in group 1 without intervening nerve crush. In group 3, 6 animals (5 experiments were technically satisfactory) the hemostatic clamp was left in place closed across the nerve for 2 h - - hydrogen clearance measurements were made before intervention as above and following removal of the clamp. In group 4, 2 animals underwent the same protocol as in group 3 but the hemostatic clamp remained in place for 3 h. Three animals (group 5) underwent the experiment of group 1 with exclusive attention to endoneurial oxygen tension - - 5 depths of nerve was sampled prior to, immediately after and 2 h after 30 s of nerve crush. Four animals (group 6) had a protocol similar to group 5 with multiple measurements before and after 3 h of clamp placement.
Data analysis Hydrogen clearance curves were fitted to a mono- or biexponential curve as previously described by least squares regression using non-linear regression analysis software (Systat Version 4.0; Evanston, IL) on an IBM 55SX computer, yielding flow values in ml/100 g/rain (NBF). Nerve blood flow (nutritive) was determined from the
TABLE I
Arterial blood gases Values are means + S.E.M. Group 1, 30 s nerve crush; group 2, control animals, no crush; group 3, 2 h nerve crush; group 4, 3 h nerve crush.
Group
n
pO 2
pCO 2
pH
1 2 3 4
9 5 5 2
143 131 123 151
38.9 + 1.7 42.8 + 2.6 44.8 _+ 1.7 44.2
7.38 7.37 7.37 7.39
+ 10 -.+ 8 + 7 _+ 5
+ 0.01 _+ 0.02 _+ 0.01 _+ 0.02
slow component of the clearance curve 2s. Percent non-nutritive flow was calculated as a/(a + b)*100 where the biexponential clearance curve is described by y = a*exp ( - K l * t ) + b*exp (-K2*t) with y = current, t = time, K1 = fast component of washout and K2 = slow component of washout. Compositive flow (ml/100 g/min) = [a/(a + b)*Kl*100 + [b/(a + b)]*K2*100. NBF (nutritive) = K2*100. Microvascular resistance was calculated as M A P / N B E Results were compared using a non-paired Student's t-test and the null hypothesis rejected when P < 0.05. RESULTS
N B F and microvascular resistance ( M R ) NBF second
declined
and MR
serial HC
rose between
measurement
the first and
irrespective
of animal
g r o u p s t a t u s ( T a b l e II, Fig. 1). T h i s is a n e x p e c t e d f e a t u r e
TABLE II
Brief (30 s) crush injury Values are means _+ S.E.M. Flow values are in ml/100 g/min. Microvascular resistance is in mm Hg/ml/100 g/min. Mean arterial pressure is in mm Hg.
Group
Before
lmmediate post
2 hours post
NB F-- endoneurial nutritive flow* 1 Crush n=9 2 Control n=5
21.0 + 2.5
13.4 + 1.4
16.0 + 1.7
22.0 + 3.9
11.5 + 0.8
12.3 + 1.1
33.2 + 2.9 (42.1+8.0) 30.6 + 6.3 (33.6_+8.7)
30.5 + 7.7 (30.1+9.6) 26.3 + 8.7 (15.8+9.6)
Composite flow (% non-nutritive)* 1 Crush n=9 2 Control n=5
28.3 + 3.0 (12.8+6.5) 28.0 + 8.2 (11.1 ± l l . l )
Microvascular resistance* 1 Crush n=9 2 Control n=5
7.27 _+ 0.98
11.48 + 1.07
9.21 _+ 0.86
6.43 _+ 0.86
12.41 + 1.30
11.3 _+ 0.83
Mean arterial pressure* 1 Crush n=9 2 Control
134 _+ 3
143 + 5
137 _+ 6
129 _+ 7
138 + 6
136 _+ 5
n=5 * Difference between control and crush groups for each parameter is not statistically significant (P > 0.05).
45 30.0
TABLE III
Prolonged nerve crush Values are mean + S.E.M. Nerve blood flow values are in ml/100 g/min. Microvascular resistance is in mm Hg/ml/100 g/rain. Mean arterial pressure is in mm Hg.
Group
Pre-crush
~..~
o 20.0
Post-crush
N B F-- endoneurial nutritiveflo w 3 2 hcrush 17.2 + 2.4 n=5 4 3 h crush 20.6 + 1.2 n=2
,,co 15.0
16.4 + 3.0*
21.3 (9.9 28.7 (27.9
Microvascular resistance 3 2 h crush 8.00 + 0.87 n=5 4 3 h crush 5.37 + 0.30 n=2
+ + + +
~crush
Z
16.5 + 7.0*
Compositeflow (% non-nutritive) 3 2 h crush 19.5 + 2.2 n=5 (9.8 + 9.8) 4 3 h crush 28.1 + 8.7 n=2 (8.9 + 8.9)
Mean arterialpressure 3 2 h crush
-~"~ ~.E' o 25.0
~
10.0
3.0* 9.9)* 5.2* 27.9)*
9.34 + 2.68*
'
" "
'
"
__.___~
contro I
'
PRE
'
POST
2 HR
Fig. 1. Serial NBF (nerve blood flow) measurements before, immediately following and 2 h after crush for 30 s (group 1 - - dotted line). NBF measurements were taken in the center of the 3 mm crushed segment. Solid line connects NBF measurements as in group 1, but without an intervening nerve crush (group 2 - controls). Postcrush values tended toward higher values than controls, but the differences were not statistically significant.
9.25 + 3.80* brief (30 s) crush did n o t differ f r o m t h o s e w i t h o u t
129 + 4
123 + 7
110
126 + 2
i n t e r v e n i n g n e r v e crush ( T a b l e II). A t 2 h f o l l o w i n g i n t e r v e n t i o n N B F t e n d e d t o w a r d h i g h e r v a l u e s and M R
n=5
4 3 h crush n=2
t o w a r d l o w e r v a l u e s in t h e a n i m a l s that h a d u n d e r g o n e crush injury, but t h e d i f f e r e n c e s w e r e n o t statistically
* Postcrush (prolonged) compared to the 2 h postcontrol group 2 (Table If) difference is not statistically significant (P > 0.05).
significant (Fig. 1). T w o o r 3 h of p r o l o n g e d crush (Table I I I ) did not result in significant c h a n g e s in N B F o r M R -
-
t h e r e was a non-significant t r e n d t o w a r d h i g h e r N B F
of serial H C c u r v e s within n e r v e and has b e e n p r e v i o u s l y
values and l o w e r M R v a l u e s t h a n t h e ' e x p e c t e d ' result
r e p o r t e d 3'3]. N B F values w e r e similar to t h o s e r e c o r d e d
(i.e. the 2 h N B F m e a s u r e m e n t in g r o u p 2). R e s u l t s of
using H C u'28 and o t h e r t e c h n i q u e s 2°. T h e initial N B F
arterial b l o o d gases a n d M A P are p r o v i d e d in T a b l e I.
and
MR
result
(before
intervention)
did not
differ
Oxygen tension and p H
significantly a m o n g t h e groups. P o s t c r u s h o l i g e m i a o r i s c h e m i a was n o t identified in
Individual oxygen tension measurements were variable
t h e s e studies. N B F and M R v a l u e s i m m e d i a t e l y f o l l o w i n g
b o t h within a n d a m o n g e x p e r i m e n t s . D e e p e r s a m p l i n g r e c o r d e d l o w e r values.
TABLE IV
Endoneurial oxygen tension (Torr)
10
Results are means + S.E.M.
o~ w~-w
[ ] PRE [] POST
z
Group
Before
Immediate post
2 h post
1 Brief (30 s) crush (n = 9) Nerve 54.3 + 9.7 a Blood 118.4 + 4.4 2 Control (n = 4) Nerve 67.3 -+ 16.1b Blood 121.5 -+ 5.4 5 Brief (30 s) crush (n = 3) Nerve 47.9 + 21.8 c 55.0 + 11.8d 35.2 + 11.6e Blood 175.3 + 10.8 180.4 + 6.8 185.7 + 10.3 6 Prolonged (3 h) crush (n = 4) Nerve 44.4 _+ 11.7~ 47.6 + 18.4g Blood 153.9 __+12.0 130.4 _+9.8 a vs b; c vs d, e; f vs g differences not statistically significant (P > 0.05).
rl-
0.05).
not crush - - suggesting differing mechanisms of injury. Of interest, N B F also does not decline in a severed nerve trunk, but the mechanism of injury is different 25. There may be transient ischemia during compression or crush which might induce direct injury to fibers p r o v i d e d it is maintained for an a d e q u a t e length of time. We have, however, e n c o u n t e r e d no evidence of 'no reflow' to account for continued fiber damage. Studies of nerve injury have frequently identified e d e m a at 24 h, a possible mechanism of microvascular compression and ischemia 15. Rydevik et al. 21 suggested that compressive injury p e r m a n e n t l y d a m a g e d vasa n e r v o r u m because they failed to observe perfusion by Evans blue in a segment of nerve that had been compressed. Quantitative flow measurements, however, were not conducted and precompressive nerve 'mobilization' may have c o m p r o m i s e d the microcirculation. A l t h o u g h we did not measure N B F at 24 h, a late e d e m a t o u s 'constrictive' effect b e y o n d 2 h may be less likely. Nerve e d e m a develops rapidly and the mechanism of e d e m a t o u s microvessel constriction has been disputed 12. Several mechanisms, distinct from 'no reflow' could account for the later development of perfusion defects - - these will require separate investigation. N B F m e a s u r e m e n t s postcrush were far above the residual low hydrogen clearance values o b t a i n e d during epochs of severe ischemia or circulatory arrest 4'22, and were much higher than those r e c o r d e d with 'no reflow' after severe ischemia 22. N B F m e a s u r e m e n t s were taken in the middle of a 3-mm length of evenly compressed nerve - - well b e y o n d the spatial sensitivity of the hydrogen microelectrode, r e p o r t e d to be 30 a m in diameter 14'26. If only partial 'no reflow' were to occur, the r a n d o m sampling of e n d o n e u r i u m by our electrode tip (as a result of nerve distortion from crush) would have identified a subset of animals with p o o r perfusion - - this was not observed. N B F and M R postcrush did not significantly differ from serial values in animals without an intervening injury - - in fact, 2 h after crush, N B F t e n d e d to rise. This may represent postinjury reactive hyperemia, as noted in o t h e r injured tissue beds 29. Shupeck et al. 24 observed high N B F values 48 and 72 h after nerve grafting, perhaps also an expression of postinjury hyperemia. E n d o n e u r i a l oxygen m e a s u r e m e n t s p r o v i d e d separate evidence that tissue oxygenation, which is critically d e p e n d e n t on perfusion 8 was intact. In groups 5 and 6, further experiments with exclusive attention toward multiple oxygen m e a s u r e m e n t s helped to confirm our findings. Nerve was s a m p l e d at multiple depths and mean values derived because of the topographical variation we observed, as also r e p o r t e d in e n d o n e u r i a l e d e m a due to galactose l°. M e a n oxygen tensions were higher than values r e p o r t e d w']-~'28 in animals but not human sural
47 nerve 16. Unlike our studies, oxygen tension measurements in previous studies were m a d e within nerve s u r r o u n d e d by oil b u b b l e d with 100% N 2. McManis and Low 15 identified subperineurial diffusion of oxygen into (and out of) nerve - - exclusion of bathing oxygen by bubbling N 2 m a y have lowered oxygen tensions whereas atmospheric diffusion may have increased the readings in our studies. O u r arterial oxygen tensions were also higher than those of previous studies - - this has been predicted to increase endoneurial tensions, particularly if sampling is near an arteriole 8. Values were significantly above those associated with conversion to anaerobic metabolism 9. Preserved p H measurements suggested that anaerobic driven lactate generation did not occur. Antim o n y p H microelectrodes have drawbacks in size (with a possibility of inducing further nerve injury) and accuracy 27 despite their rugged suitability in the present experiments - - r e p e a t e d calibrations were required. F u r t h e r studies will be required to compare their accuracy with that using o t h e r methods. W h y is severe peripheral nerve ischemia from vascular ligation associated with 'no reflow' but not 3 h of
continuous compressive crush injury? O n e mechanism of 'no reflow' injury has been thought to be due to generation of injurious free radical oxygen species during reperfusion 5. E n d o t h e l i u m m a y be a target of free radical injury with m e m b r a n e p e r o x i d a t i o n and m i c r o t h r o m b osis. In a substantial length of nerve trunk or ischemic limb, one might expect large quantitative differences in free radical generation and effect in comparison to a localized segment of reperfused nerve. In the crush model, n e a r b y normal perfused tissue, in contiguity with the injured segment m a y 'wash o u t ' injurious products of reperfusion limiting their concentrations and duration of action. Nearby generated prostacyclin and/or other vasodilatory substances (substance P, for example), may then prevent microthromboses and 'no reflow'. These suggestions, however, require experimental substantiation.
REFERENCES
injury and increased endoneurial fluid pressure: a 'miniature compartment syndrome', J. Neurol. Neurosurg. Psychiat., 46 (1983) 1119-1124. 14 McManis, P.G., Low, P.A. and Yao, J.K., Relationship between nerve blood flow and intercapillary distance in peripheral nerve edema, Am. J. Physiol., 251 (1986) E92-E97. 15 McManis, P.G. and Low, P.A., Factors affecting the relative viability of centrifascicular and subperineurial axons in acute peripheral nerve ischemia, Exp. Neurol., 99 (1988) 84-95. 16 Newrick, P.G., Wilson, A.J., Jakubowski, J., Boulton, A.J.M. and Ward, J.D., Sural nerve oxygen tension in diabetes, Br. Med. J., 293 (1986) 1053-1054. 17 Nukada, H., Post-traumatic endoneurial neovascularization and nerve regeneration: a morphometric study, Brain Research, 449 (1988) 89-96. 18 Ochoa, J., Fowler, T.J. and Gilliat, R.W., Anatomical changes in peripheral nerves compressed by a pneumatic tourniquet, J. Anat., 113 (1972) 433-455. 19 Powell, H.C., Myers, R.R., Costello, M.L. and Lampert, P.W., Endoneurial fluid pressure in wallerian degeneration, Ann. Neurol., 5 (1979) 550-557. 20 Rundquist, I., Smith, Q.R., Michel, M.E., Ask, P., Oberg, P.A. and Rapoport, S.I., Sciatic nerve blood flow measured by laser doppler flowmetry and (C14) iodoantipyrine, Am. J. Physiol., 248 (1985) H311-H317. 21 Rydevik, B., Lundborg, G. and Bagge, U., Effects of graded compression on intraneural blood flow. An in vivo study on rabbit tibial nerve, J. Hand. Surg., 6 (1981) 3-12. 22 Schmelzer, J.D., Zochodne, D.W. and Low, P.A., Ischemic and reperfusion injury of rat peripheral nerve, Proc. Natl. Acad. Sci., 86 (1989) 1639-1642. 23 Schmid-Schonbein, G.W., Capillary plugging by granulocytes and the no-reflow phenomenon in the microcirculation, Fed. Proc., 46 (1987) 2397-2401. 24 Shupeck, M., Ward, K.K., Schmelzer, J.D. and Low, P.A., Comparison of nerve regeneration in vascularized and conventional grafts: nerve electrophysiology, norepinephrine, prostacyclin, malondialdehyde and the blood-nerve barrier, Brain Research, 493 (1989) 225-230.
1 Ames, A., Wright, R.L., Kowada, M., Thurston, J.M. and Majno, G., Cerebral ischemia. I1. The no-reflow phenomenon, Am. J. Pathol., 52 (1968) 437-453. 2 Benstead, T.J., Dyck, P.J., Sangalang, V. and Steele, A., Ischemia-induced endothelial swelling of endoneurial microvessels, Neurology, Suppl. 1, 38 (1988) 402. 3 Day, T.J., Schmelzer, J.D. and Low, P.A., Aortic occlusion and reperfusion and conduction, blood flow and the blood-nerve barrier of rat sciatic nerve, Exp. Neurol., 103 (1989) 173-178. 4 Day, T.J., Lagerlund, T.D. and Low, P.A., Analysis of H 2 clearance curves used to measure blood flow in the rat sciatic nerve, J. Physiol., 414 (1989) 35-54. 5 Del Maestro, R.E, An approach to free radicals in medicine and biology, Acta Physiol. Scand., Suppl. 492 (1980) 153-168. 6 Denny-Brown, D. and Brenner, C., Paralysis of nerve induced by direct pressure and by tourniquet, Arch. Neurol. Psychiat., 51 (1944) 1-26. 7 Denny-Brown, D. and Brenner, C., Lesion in peripheral nerve resulting from compression by spring clip, Arch. Neurol. Psychiat., 52 (1944) 1-19. 8 Lagerlund, T.D. and Low, P.A., A mathematical simulation of oxygen delivery in rat peripheral nerve, Microvasc. Res., 34 (1987) 211-222. 9 Low, P.A., Schmelzer, J.D., Ward, K.K. and Yao, J.K., Experimental chronic hypoxic neuropathy: relevance to diabetic neuropathy, Am. J. Physiol., 250 (1986) E94-E99. 10 Low, P.A., Nukada, H., Schmelzer, J.D., Tuck, R.R. and Dyck, P.J., Endoneuriai oxygen tension and radial topography in nerve edema, Brain Research, 341 (1985) 147-154. I1 Low, P.A. and Tuck, R.R., Effects of changes of blood pressure, respiratory acidosis and hypoxia on blood flow in the sciatic nerve of the rat, J. Physiol., 347 (1984) 513-524. 12 Low, P.A., Endoneurial fluid pressure and microenvironment of nerve. In P.J. Dyck, P.K. Thomas, E.H. Lambert and R. Bunge (Eds.), Peripheral Neuropathy, Vol. 1, Saunders, Amsterdam, 1984, pp. 599-618. 13 Lundborg, G., Myers, R. and Powell, H., Nerve compression
Acknowledgements. Lisa Fregeau provided excellent secretarial assistance. Funding support was provided by the Medical Research Council of Canada, Canadian Diabetes Association, the Muscular Dystrophy Association of Canada, and Botterell, Principal's Development Fund Category B, and Research Initiation Grants from Queen's University. Ethicon Canada provided suture material.
48 25 Smith, D.R., Kobrine, A.I. and Rizzoli, H.V., Blood flow in peripheral nerves. Normal and post severance flow rates, J. Neurol. Sci., 33 (1977) 341-346. 26 Stosseck, K. and Lubbers, D.W., Local generation and measurement of hydrogen in tissue by means of bare platinum electrodes, In M. Kessler, D.E Bruley, L.C. Clark, D.W. Lubbers, I.A. Silver and J. Strauss, Oxygen Supply, University Park, Baltimore, 1971, pp. 113-114. 27 Thomas, R.C., Ion-sensitive Intracellular Microelectrodes: How To Make And Use Them, Academic, London, 1978. 28 Tuck, R.R., Schmelzer, J.D. and Low, P.A., Endoneurial blood flow and oxygen tension in the sciatic nerves of rats with
experimental diabetic neuropathy, Brain, 107 (1984) 935-950. 29 Westerman, R.A., Magerl, EW., Szolcsanyi, L, Handwerker, H.O., Low, A., Pratt, A., Widdop, R. and Kozak, W., Vasodilator axon reflexes. In P.M. Vanhoutte (Ed.), Vasodilatation: Vascular Smooth Muscle, Peptides, Autonomic Nerves and Endothelium, Raven, New York, 1988, pp. 107-112. 30 Woodhall, B. and Davis, C., Changes in the arteriae nervorum in peripheral nerve injuries in man, J. Neuropath. Exp. Neurol.. 9 (1950) 335-343. 31 Zochodne, D.W. and Ho, L., Activity-induced endoneurial hyperemia, Neurology, Suppl. 1, 40 (1990) 160.
43
BRES 16100
E n d o n e u r i a l m i c r o e n v i r o n m e n t and acute nerve crush injury in the rat sciatic nerve Douglas W. Zochodne and Lam T. Ho Department of Medicine, Queen's University, Kingston, Ont. (Canada) (Accepted 12 June 1990)
Key words: Peripheral nerve; Crush; Nerve blood flow; Oxygen tension
In severe peripheral nerve ischemia in the rat, serial nerve blood flow (NBF) measurements have identified evidence of 'no reflow', a mechanism of continued fiber damage during reperfusion. It has been postulated that 'no reflow' also occurs in nerve compression due to direct mechanical or ischemic (if compression is prolonged) injury of microvessels, resulting in continuing nerve fiber damage. To address this question, we measured endoneurial blood flow (NBF), oxygen tension and pH at the site of an acute nerve crush injury. In further sets of experiments, NBF and endoneurial oxygen tension were examined before and after prolonged epochs of crush. NBF and MR (microvascular resistance) were not appreciably different than values obtained in control animals without intervening brief nerve crush. NBF was slightly higher and MR slightly lower 2 h after injury, but the difference was not statistically significant. No evidence of significant endoneurial hypoxia or acidosis was observed. Similarly, after more prolonged crush there was no significant oligemia or hypoxia. The studies provide no evidence that 'no reflow' occurs in crush injury even if the injury is maintained for a period of time known to induce 'no reflow' with severe ischemia. We suggest that nerve damage in crush, and possibly compression, more likely arises from direct mechanical injury of fibers.
INTRODUCTION Studies of acute mechanical injury of peripheral nerve by compression or crush have postulated that an important mechanism of injury, perhaps complicating regeneration, is ischemia 6'7A3'21'3°. A c u t e endothelial injury might arise from direct mechanical disruption, ischemia or both. Endothelial swelling, granulocyte plugging 23 and microvascular thrombosis would then result in 'no reflow' and continuing fiber d a m a g e 1. A n additional factor might be e n d o n e u r i a l e d e m a with microvascular compression ~3. T h e r e is evidence that severe peripheral nerve ischemia from vascular ligation damages endothelium, resulting in swelling, luminal narrowing and 'no reflow '2"22. A l t h o u g h compression may result in t e m p o r a r y circulatory arrest it is unclear whether this insult p e r m a n e n t l y injures nerve microvessels. Compressive ischemia, if m a i n t a i n e d long enough, might induce 'no reflow', as in the ligation experiments. If both mechanical injury to microvessels and ischemic endothelial injury were to occur with compression, relatively short periods of injury might result in 'no reflow'. Nerve crush might be considered an e x t r e m e version of compressive nerve injury with an enhanced degree of vasa nervorum disruption.
In this study we m e a s u r e d b l o o d flow, oxygen content and p H in the sciatic nerve of the rat ( N B F ) using endoneurial microelectrodes. To test the hypothesis that acute injury damages vasa n e r v o r u m , we m e a s u r e d NBF, endoneurial oxygen content and p H at the injury site i m m e d i a t e l y and 2 h following segmental mid-sciatic nerve crush. To test the hypothesis that p r o l o n g e d crush is associated with ischemic 'no reflow' we m e a s u r e d N B F and endoneurial oxygen content i m m e d i a t e l y after 2 and 3 h epochs of crush injury. A control group of animals had serial m e a s u r e m e n t s without crush injury. MATERIALS AND METHODS
Animals Animals employed were male Sprague-Dawley rats weighing 200-300 g housed in wire mesh cages with ad libitum access to rat chow and water. Preparation Rats were anesthetized with sodium pentobarbital by intraperitoneal injection (65 mg/kg). A ventral midline neck incision was made for placement of a left carotid intra-arterial line (PE-50; Intramedic; Clay Adams, Parsippany, NJ) connected to a pressure transducer (P23ID Gould; Oxnard, CA) and tracheostomy. The animal was then paralyzed using tubocurarine (1.0 mg/kg intraarterial) (Sigma Chemical Co.; St. Louis, MO) and ventilated (Rodent ventilator 683; Harvard Bioscience; South Natick, MA) through flowmeters (N032-416 and Nl12-026; Cole-Panner, Chi-
Correspondence: D.W. Zochodne, Room 206, La Salle Building, 146 Stuart Street, Department of Medicine, Queen's University, Kingston, Ont., K7L 3N6, Canada. 0006-8993/90/$03.50 (~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
44 cago, IL) to permit regulation of inspired gases. The left sciatic nerve was exposed, covered in mineral oil, and microelectrodes inserted through an epineurial window into the endoneurium. A reference KCl-agar bridge electrode was inserted subcutaneously and sutured into place. The endoneurial microelectrode was polarized at +0.250 V for hydrogen clearance studies and -0.650 V for the oxygen measurements and current detected using a microsensor (Microsensor If, Diamond General; Ann Arbor, MI). The exposed nerve preparation was maintained at 37 °C throughout the experiments using a temperature probe connected to a temperature control and feedback unit with an infrared heating lamp (TH8 and TCAT-1A, Sensortek; Clifton, N J). Additional doses of pentobarbital (20 mg/kg) and D-tubocurarine (0.5 mg/kg) were administered approximately 2 hourly through the intra-arterial catheter. Continuous print out of mean arterial blood pressure (MAP) and the microsensor output was accomplished using two channels of a polygraph recorder (79E, Grass Instruments; Quincy, MA). Arterial blood gas samples were drawn approximately hourly to ensure physiologic stability (Table I).
Experiments The technique of hydrogen clearance for measurement of endoneurial blood flow has been published 28. Measurements of oxygen tension were conducted by endoneurial microelectrodes as previously described zs, but without saturating the bathing mineral oil with 100% N 2. Calibration of the oxygen microelectrode was obtained within 5 mih of all in vivo measurements to prevent error from microelectrode drift. A linear calibration line was obtained by bubbled gas mixtures of 0, 10 and 25% oxygen at 37 °C in a calibration bath (Diamond General; Ann Arbor, MI). In experiments with sampling of oxygen tensions at various depths, a mean tension was calculated for each experimental time point, pH studies were conducted with 80 micron tipped antimony microelectrodes (801, Diamond General; Ann Arbor, MI) connected to an electrometer (microsensor If, Diamond General; Ann Arbor, MI). Two point calibration was conducted at 37 °C after each experiment using Trizma solutions of pH 6.70 and 7.10 (Sigma Chemical Company; St. Louis, MO) - - in vivo pH measurements were obtained from the mV vs pH calibration line. Several groups of animals were studied. In group 1, 12 animals underwent serial HC measurements (9 experiments were technically satisfactory). After an initial HC measurement, the microelectrode was withdrawn and the site of measurement subjected to 30 s of crush applied by the polyethylene-covered jaws of a hemostatic clamp - - the injured area measured 3 mm of nerve length. After removal of the clamp, the microelectrode was reinserted through the original epineurial window and repeat measurements made of hydrogen clearance. A third and final clearance curve was taken at 2 h following crush, pH measurements were made before, immediately after and 2 h following nerve crush. Measurements of oxygen tension were made after the final NBF measurement. In Group 2, 6 animals (5 experiments were technically satisfactory) underwent the same protocol as in group 1 without intervening nerve crush. In group 3, 6 animals (5 experiments were technically satisfactory) the hemostatic clamp was left in place closed across the nerve for 2 h - - hydrogen clearance measurements were made before intervention as above and following removal of the clamp. In group 4, 2 animals underwent the same protocol as in group 3 but the hemostatic clamp remained in place for 3 h. Three animals (group 5) underwent the experiment of group 1 with exclusive attention to endoneurial oxygen tension - - 5 depths of nerve was sampled prior to, immediately after and 2 h after 30 s of nerve crush. Four animals (group 6) had a protocol similar to group 5 with multiple measurements before and after 3 h of clamp placement.
Data analysis Hydrogen clearance curves were fitted to a mono- or biexponential curve as previously described by least squares regression using non-linear regression analysis software (Systat Version 4.0; Evanston, IL) on an IBM 55SX computer, yielding flow values in ml/100 g/rain (NBF). Nerve blood flow (nutritive) was determined from the
TABLE I
Arterial blood gases Values are means + S.E.M. Group 1, 30 s nerve crush; group 2, control animals, no crush; group 3, 2 h nerve crush; group 4, 3 h nerve crush.
Group
n
pO 2
pCO 2
pH
1 2 3 4
9 5 5 2
143 131 123 151
38.9 + 1.7 42.8 + 2.6 44.8 _+ 1.7 44.2
7.38 7.37 7.37 7.39
+ 10 -.+ 8 + 7 _+ 5
+ 0.01 _+ 0.02 _+ 0.01 _+ 0.02
slow component of the clearance curve 2s. Percent non-nutritive flow was calculated as a/(a + b)*100 where the biexponential clearance curve is described by y = a*exp ( - K l * t ) + b*exp (-K2*t) with y = current, t = time, K1 = fast component of washout and K2 = slow component of washout. Compositive flow (ml/100 g/min) = [a/(a + b)*Kl*100 + [b/(a + b)]*K2*100. NBF (nutritive) = K2*100. Microvascular resistance was calculated as M A P / N B E Results were compared using a non-paired Student's t-test and the null hypothesis rejected when P < 0.05. RESULTS
N B F and microvascular resistance ( M R ) NBF second
declined
and MR
serial HC
rose between
measurement
the first and
irrespective
of animal
g r o u p s t a t u s ( T a b l e II, Fig. 1). T h i s is a n e x p e c t e d f e a t u r e
TABLE II
Brief (30 s) crush injury Values are means _+ S.E.M. Flow values are in ml/100 g/min. Microvascular resistance is in mm Hg/ml/100 g/min. Mean arterial pressure is in mm Hg.
Group
Before
lmmediate post
2 hours post
NB F-- endoneurial nutritive flow* 1 Crush n=9 2 Control n=5
21.0 + 2.5
13.4 + 1.4
16.0 + 1.7
22.0 + 3.9
11.5 + 0.8
12.3 + 1.1
33.2 + 2.9 (42.1+8.0) 30.6 + 6.3 (33.6_+8.7)
30.5 + 7.7 (30.1+9.6) 26.3 + 8.7 (15.8+9.6)
Composite flow (% non-nutritive)* 1 Crush n=9 2 Control n=5
28.3 + 3.0 (12.8+6.5) 28.0 + 8.2 (11.1 ± l l . l )
Microvascular resistance* 1 Crush n=9 2 Control n=5
7.27 _+ 0.98
11.48 + 1.07
9.21 _+ 0.86
6.43 _+ 0.86
12.41 + 1.30
11.3 _+ 0.83
Mean arterial pressure* 1 Crush n=9 2 Control
134 _+ 3
143 + 5
137 _+ 6
129 _+ 7
138 + 6
136 _+ 5
n=5 * Difference between control and crush groups for each parameter is not statistically significant (P > 0.05).
45 30.0
TABLE III
Prolonged nerve crush Values are mean + S.E.M. Nerve blood flow values are in ml/100 g/min. Microvascular resistance is in mm Hg/ml/100 g/rain. Mean arterial pressure is in mm Hg.
Group
Pre-crush
~..~
o 20.0
Post-crush
N B F-- endoneurial nutritiveflo w 3 2 hcrush 17.2 + 2.4 n=5 4 3 h crush 20.6 + 1.2 n=2
,,co 15.0
16.4 + 3.0*
21.3 (9.9 28.7 (27.9
Microvascular resistance 3 2 h crush 8.00 + 0.87 n=5 4 3 h crush 5.37 + 0.30 n=2
+ + + +
~crush
Z
16.5 + 7.0*
Compositeflow (% non-nutritive) 3 2 h crush 19.5 + 2.2 n=5 (9.8 + 9.8) 4 3 h crush 28.1 + 8.7 n=2 (8.9 + 8.9)
Mean arterialpressure 3 2 h crush
-~"~ ~.E' o 25.0
~
10.0
3.0* 9.9)* 5.2* 27.9)*
9.34 + 2.68*
'
" "
'
"
__.___~
contro I
'
PRE
'
POST
2 HR
Fig. 1. Serial NBF (nerve blood flow) measurements before, immediately following and 2 h after crush for 30 s (group 1 - - dotted line). NBF measurements were taken in the center of the 3 mm crushed segment. Solid line connects NBF measurements as in group 1, but without an intervening nerve crush (group 2 - controls). Postcrush values tended toward higher values than controls, but the differences were not statistically significant.
9.25 + 3.80* brief (30 s) crush did n o t differ f r o m t h o s e w i t h o u t
129 + 4
123 + 7
110
126 + 2
i n t e r v e n i n g n e r v e crush ( T a b l e II). A t 2 h f o l l o w i n g i n t e r v e n t i o n N B F t e n d e d t o w a r d h i g h e r v a l u e s and M R
n=5
4 3 h crush n=2
t o w a r d l o w e r v a l u e s in t h e a n i m a l s that h a d u n d e r g o n e crush injury, but t h e d i f f e r e n c e s w e r e n o t statistically
* Postcrush (prolonged) compared to the 2 h postcontrol group 2 (Table If) difference is not statistically significant (P > 0.05).
significant (Fig. 1). T w o o r 3 h of p r o l o n g e d crush (Table I I I ) did not result in significant c h a n g e s in N B F o r M R -
-
t h e r e was a non-significant t r e n d t o w a r d h i g h e r N B F
of serial H C c u r v e s within n e r v e and has b e e n p r e v i o u s l y
values and l o w e r M R v a l u e s t h a n t h e ' e x p e c t e d ' result
r e p o r t e d 3'3]. N B F values w e r e similar to t h o s e r e c o r d e d
(i.e. the 2 h N B F m e a s u r e m e n t in g r o u p 2). R e s u l t s of
using H C u'28 and o t h e r t e c h n i q u e s 2°. T h e initial N B F
arterial b l o o d gases a n d M A P are p r o v i d e d in T a b l e I.
and
MR
result
(before
intervention)
did not
differ
Oxygen tension and p H
significantly a m o n g t h e groups. P o s t c r u s h o l i g e m i a o r i s c h e m i a was n o t identified in
Individual oxygen tension measurements were variable
t h e s e studies. N B F and M R v a l u e s i m m e d i a t e l y f o l l o w i n g
b o t h within a n d a m o n g e x p e r i m e n t s . D e e p e r s a m p l i n g r e c o r d e d l o w e r values.
TABLE IV
Endoneurial oxygen tension (Torr)
10
Results are means + S.E.M.
o~ w~-w
[ ] PRE [] POST
z
Group
Before
Immediate post
2 h post
1 Brief (30 s) crush (n = 9) Nerve 54.3 + 9.7 a Blood 118.4 + 4.4 2 Control (n = 4) Nerve 67.3 -+ 16.1b Blood 121.5 -+ 5.4 5 Brief (30 s) crush (n = 3) Nerve 47.9 + 21.8 c 55.0 + 11.8d 35.2 + 11.6e Blood 175.3 + 10.8 180.4 + 6.8 185.7 + 10.3 6 Prolonged (3 h) crush (n = 4) Nerve 44.4 _+ 11.7~ 47.6 + 18.4g Blood 153.9 __+12.0 130.4 _+9.8 a vs b; c vs d, e; f vs g differences not statistically significant (P > 0.05).
rl-
0.05).
not crush - - suggesting differing mechanisms of injury. Of interest, N B F also does not decline in a severed nerve trunk, but the mechanism of injury is different 25. There may be transient ischemia during compression or crush which might induce direct injury to fibers p r o v i d e d it is maintained for an a d e q u a t e length of time. We have, however, e n c o u n t e r e d no evidence of 'no reflow' to account for continued fiber damage. Studies of nerve injury have frequently identified e d e m a at 24 h, a possible mechanism of microvascular compression and ischemia 15. Rydevik et al. 21 suggested that compressive injury p e r m a n e n t l y d a m a g e d vasa n e r v o r u m because they failed to observe perfusion by Evans blue in a segment of nerve that had been compressed. Quantitative flow measurements, however, were not conducted and precompressive nerve 'mobilization' may have c o m p r o m i s e d the microcirculation. A l t h o u g h we did not measure N B F at 24 h, a late e d e m a t o u s 'constrictive' effect b e y o n d 2 h may be less likely. Nerve e d e m a develops rapidly and the mechanism of e d e m a t o u s microvessel constriction has been disputed 12. Several mechanisms, distinct from 'no reflow' could account for the later development of perfusion defects - - these will require separate investigation. N B F m e a s u r e m e n t s postcrush were far above the residual low hydrogen clearance values o b t a i n e d during epochs of severe ischemia or circulatory arrest 4'22, and were much higher than those r e c o r d e d with 'no reflow' after severe ischemia 22. N B F m e a s u r e m e n t s were taken in the middle of a 3-mm length of evenly compressed nerve - - well b e y o n d the spatial sensitivity of the hydrogen microelectrode, r e p o r t e d to be 30 a m in diameter 14'26. If only partial 'no reflow' were to occur, the r a n d o m sampling of e n d o n e u r i u m by our electrode tip (as a result of nerve distortion from crush) would have identified a subset of animals with p o o r perfusion - - this was not observed. N B F and M R postcrush did not significantly differ from serial values in animals without an intervening injury - - in fact, 2 h after crush, N B F t e n d e d to rise. This may represent postinjury reactive hyperemia, as noted in o t h e r injured tissue beds 29. Shupeck et al. 24 observed high N B F values 48 and 72 h after nerve grafting, perhaps also an expression of postinjury hyperemia. E n d o n e u r i a l oxygen m e a s u r e m e n t s p r o v i d e d separate evidence that tissue oxygenation, which is critically d e p e n d e n t on perfusion 8 was intact. In groups 5 and 6, further experiments with exclusive attention toward multiple oxygen m e a s u r e m e n t s helped to confirm our findings. Nerve was s a m p l e d at multiple depths and mean values derived because of the topographical variation we observed, as also r e p o r t e d in e n d o n e u r i a l e d e m a due to galactose l°. M e a n oxygen tensions were higher than values r e p o r t e d w']-~'28 in animals but not human sural
47 nerve 16. Unlike our studies, oxygen tension measurements in previous studies were m a d e within nerve s u r r o u n d e d by oil b u b b l e d with 100% N 2. McManis and Low 15 identified subperineurial diffusion of oxygen into (and out of) nerve - - exclusion of bathing oxygen by bubbling N 2 m a y have lowered oxygen tensions whereas atmospheric diffusion may have increased the readings in our studies. O u r arterial oxygen tensions were also higher than those of previous studies - - this has been predicted to increase endoneurial tensions, particularly if sampling is near an arteriole 8. Values were significantly above those associated with conversion to anaerobic metabolism 9. Preserved p H measurements suggested that anaerobic driven lactate generation did not occur. Antim o n y p H microelectrodes have drawbacks in size (with a possibility of inducing further nerve injury) and accuracy 27 despite their rugged suitability in the present experiments - - r e p e a t e d calibrations were required. F u r t h e r studies will be required to compare their accuracy with that using o t h e r methods. W h y is severe peripheral nerve ischemia from vascular ligation associated with 'no reflow' but not 3 h of
continuous compressive crush injury? O n e mechanism of 'no reflow' injury has been thought to be due to generation of injurious free radical oxygen species during reperfusion 5. E n d o t h e l i u m m a y be a target of free radical injury with m e m b r a n e p e r o x i d a t i o n and m i c r o t h r o m b osis. In a substantial length of nerve trunk or ischemic limb, one might expect large quantitative differences in free radical generation and effect in comparison to a localized segment of reperfused nerve. In the crush model, n e a r b y normal perfused tissue, in contiguity with the injured segment m a y 'wash o u t ' injurious products of reperfusion limiting their concentrations and duration of action. Nearby generated prostacyclin and/or other vasodilatory substances (substance P, for example), may then prevent microthromboses and 'no reflow'. These suggestions, however, require experimental substantiation.
REFERENCES
injury and increased endoneurial fluid pressure: a 'miniature compartment syndrome', J. Neurol. Neurosurg. Psychiat., 46 (1983) 1119-1124. 14 McManis, P.G., Low, P.A. and Yao, J.K., Relationship between nerve blood flow and intercapillary distance in peripheral nerve edema, Am. J. Physiol., 251 (1986) E92-E97. 15 McManis, P.G. and Low, P.A., Factors affecting the relative viability of centrifascicular and subperineurial axons in acute peripheral nerve ischemia, Exp. Neurol., 99 (1988) 84-95. 16 Newrick, P.G., Wilson, A.J., Jakubowski, J., Boulton, A.J.M. and Ward, J.D., Sural nerve oxygen tension in diabetes, Br. Med. J., 293 (1986) 1053-1054. 17 Nukada, H., Post-traumatic endoneurial neovascularization and nerve regeneration: a morphometric study, Brain Research, 449 (1988) 89-96. 18 Ochoa, J., Fowler, T.J. and Gilliat, R.W., Anatomical changes in peripheral nerves compressed by a pneumatic tourniquet, J. Anat., 113 (1972) 433-455. 19 Powell, H.C., Myers, R.R., Costello, M.L. and Lampert, P.W., Endoneurial fluid pressure in wallerian degeneration, Ann. Neurol., 5 (1979) 550-557. 20 Rundquist, I., Smith, Q.R., Michel, M.E., Ask, P., Oberg, P.A. and Rapoport, S.I., Sciatic nerve blood flow measured by laser doppler flowmetry and (C14) iodoantipyrine, Am. J. Physiol., 248 (1985) H311-H317. 21 Rydevik, B., Lundborg, G. and Bagge, U., Effects of graded compression on intraneural blood flow. An in vivo study on rabbit tibial nerve, J. Hand. Surg., 6 (1981) 3-12. 22 Schmelzer, J.D., Zochodne, D.W. and Low, P.A., Ischemic and reperfusion injury of rat peripheral nerve, Proc. Natl. Acad. Sci., 86 (1989) 1639-1642. 23 Schmid-Schonbein, G.W., Capillary plugging by granulocytes and the no-reflow phenomenon in the microcirculation, Fed. Proc., 46 (1987) 2397-2401. 24 Shupeck, M., Ward, K.K., Schmelzer, J.D. and Low, P.A., Comparison of nerve regeneration in vascularized and conventional grafts: nerve electrophysiology, norepinephrine, prostacyclin, malondialdehyde and the blood-nerve barrier, Brain Research, 493 (1989) 225-230.
1 Ames, A., Wright, R.L., Kowada, M., Thurston, J.M. and Majno, G., Cerebral ischemia. I1. The no-reflow phenomenon, Am. J. Pathol., 52 (1968) 437-453. 2 Benstead, T.J., Dyck, P.J., Sangalang, V. and Steele, A., Ischemia-induced endothelial swelling of endoneurial microvessels, Neurology, Suppl. 1, 38 (1988) 402. 3 Day, T.J., Schmelzer, J.D. and Low, P.A., Aortic occlusion and reperfusion and conduction, blood flow and the blood-nerve barrier of rat sciatic nerve, Exp. Neurol., 103 (1989) 173-178. 4 Day, T.J., Lagerlund, T.D. and Low, P.A., Analysis of H 2 clearance curves used to measure blood flow in the rat sciatic nerve, J. Physiol., 414 (1989) 35-54. 5 Del Maestro, R.E, An approach to free radicals in medicine and biology, Acta Physiol. Scand., Suppl. 492 (1980) 153-168. 6 Denny-Brown, D. and Brenner, C., Paralysis of nerve induced by direct pressure and by tourniquet, Arch. Neurol. Psychiat., 51 (1944) 1-26. 7 Denny-Brown, D. and Brenner, C., Lesion in peripheral nerve resulting from compression by spring clip, Arch. Neurol. Psychiat., 52 (1944) 1-19. 8 Lagerlund, T.D. and Low, P.A., A mathematical simulation of oxygen delivery in rat peripheral nerve, Microvasc. Res., 34 (1987) 211-222. 9 Low, P.A., Schmelzer, J.D., Ward, K.K. and Yao, J.K., Experimental chronic hypoxic neuropathy: relevance to diabetic neuropathy, Am. J. Physiol., 250 (1986) E94-E99. 10 Low, P.A., Nukada, H., Schmelzer, J.D., Tuck, R.R. and Dyck, P.J., Endoneuriai oxygen tension and radial topography in nerve edema, Brain Research, 341 (1985) 147-154. I1 Low, P.A. and Tuck, R.R., Effects of changes of blood pressure, respiratory acidosis and hypoxia on blood flow in the sciatic nerve of the rat, J. Physiol., 347 (1984) 513-524. 12 Low, P.A., Endoneurial fluid pressure and microenvironment of nerve. In P.J. Dyck, P.K. Thomas, E.H. Lambert and R. Bunge (Eds.), Peripheral Neuropathy, Vol. 1, Saunders, Amsterdam, 1984, pp. 599-618. 13 Lundborg, G., Myers, R. and Powell, H., Nerve compression
Acknowledgements. Lisa Fregeau provided excellent secretarial assistance. Funding support was provided by the Medical Research Council of Canada, Canadian Diabetes Association, the Muscular Dystrophy Association of Canada, and Botterell, Principal's Development Fund Category B, and Research Initiation Grants from Queen's University. Ethicon Canada provided suture material.
48 25 Smith, D.R., Kobrine, A.I. and Rizzoli, H.V., Blood flow in peripheral nerves. Normal and post severance flow rates, J. Neurol. Sci., 33 (1977) 341-346. 26 Stosseck, K. and Lubbers, D.W., Local generation and measurement of hydrogen in tissue by means of bare platinum electrodes, In M. Kessler, D.E Bruley, L.C. Clark, D.W. Lubbers, I.A. Silver and J. Strauss, Oxygen Supply, University Park, Baltimore, 1971, pp. 113-114. 27 Thomas, R.C., Ion-sensitive Intracellular Microelectrodes: How To Make And Use Them, Academic, London, 1978. 28 Tuck, R.R., Schmelzer, J.D. and Low, P.A., Endoneurial blood flow and oxygen tension in the sciatic nerves of rats with
experimental diabetic neuropathy, Brain, 107 (1984) 935-950. 29 Westerman, R.A., Magerl, EW., Szolcsanyi, L, Handwerker, H.O., Low, A., Pratt, A., Widdop, R. and Kozak, W., Vasodilator axon reflexes. In P.M. Vanhoutte (Ed.), Vasodilatation: Vascular Smooth Muscle, Peptides, Autonomic Nerves and Endothelium, Raven, New York, 1988, pp. 107-112. 30 Woodhall, B. and Davis, C., Changes in the arteriae nervorum in peripheral nerve injuries in man, J. Neuropath. Exp. Neurol.. 9 (1950) 335-343. 31 Zochodne, D.W. and Ho, L., Activity-induced endoneurial hyperemia, Neurology, Suppl. 1, 40 (1990) 160.
