Brain Research, 579 (1992) 32-42 © 1992 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/92/$05.00
32
BRES 17663
The effect of direct current field polarity on recovery after acute experimental spinal cord injury Michael G. Fehlings and Charles H. Tator Canadian Paraplegic Association Spinal Cord Injury Research Laboratory, Playfair Neuroscience Unit and Division of Neurosurgery, Toronto Hospital, Western Division, and University of Toronto, Toronto (Canada) (Accepted 10 December 1991)
Key words: Rat; Electrical stimulation; Horseradish peroxidase; Motor-evoked potential; Somatosensory-evoked potential
Recent evidence indicates that direct current (DC) fields promote recovery of acutely injured central and peripheral nervous system axons. The polarity of the applied DC field may play an important role in modulating these effects. In the present study, the effect of DC field polarity on recovery of injured spinal cord axons was examined anatomically, electrophysiologically and behaviourly in a rat model. After a 53 g clip compression injury of the cord at T1, 30 adult rats were randomly and blindly allocated to one of three groups (n = 10 each): one group received implantation of a DC stimulator (14/~A) with the cathode caudal to the injury site; the second group received implantation of a similar stimulator with the cathode rostral to the injury site; and the third group received a sham (O/~A) stimulator. Clinical neurological function was assessed by the inclined plane technique and axonal function was assessed by motor- and somatosensory-evoked potentials (MEP and SSEP). A quantitative assessment of axonal integrity was performed by counting neurons in the brain retrogradely labelled by the axonal tracer horseradish peroxidase (HRP) and by counting axons at the injury site. The inclined plane scores (P < 0.0001), MEP amplitude (P < 0.02), counts of neurons retrogradely labelled by HRP (P < 0.0001), and axon counts at the injury site (P < 0.01) were significantly greater in the group treated with a DC field with the cathode caudal to the lesion than in the other two groups. Conversely, the cathode rostral DC field caused a decrease in the number of neurons retrogradely labelled by HRP (P < 0.05) compared to the sham and cathode caudal groups. These data confirm our previous finding that DC fields promote recovery of acutely injured spinal cord axons. Furthermore, the polarity of the applied field is of critical importance to this effect. INTRODUCTION
field p o l a r i t y on r e c o v e r y of m a m m a l i a n spinal c o r d axons s u b j e c t e d to a s e v e r e c o m p r e s s i v e injury. This ques-
There
is c o n s i d e r a b l e
e v i d e n c e that direct c u r r e n t
tion is o f f u n d a m e n t a l i m p o r t a n c e in u n d e r s t a n d i n g the
( D C ) fields p r o m o t e the r e c o v e r y and r e g e n e r a t i o n of injured p e r i p h e r a l 27'32'37'47 and central n e r v o u s system
m e c h a n i s m of action o f D C fields on injured axons.
axons 4'5'6'7'12'15'26'31'33'34'45. It has b e e n s h o w n in vitro
MATERIALS AND METHODS
that the polarity of the D C field is o f critical i m p o r t a n c e to the m e c h a n i s m of t h e s e effects 19'24'25'3°. F o r e x a m p l e , the t r o p h i c effects of D C fields o n n e u r o n s in vitro are polarity specific19'24'25'3°: n e u r i t e s p r e f e r e n t i a l l y g r o w tow a r d the c a t h o d e , w h e r e a s a n o d a l l y facing n e u r i t e s und e r g o a b s o r p t i o n 25. A n u m b e r of studies c o n d u c t e d in v i v o h a v e s h o w n similar effects. F o r e x a m p l e , a caudally n e g a t i v e D C field has b e e n r e p o r t e d to r e d u c e the extent of r e t r o g r a d e d e g e n e r a t i o n of t r a n s e c t e d reticulospinal axons in the l a m p r e y spinal c o r d 36. R e c e n t l y , it has b e e n o b s e r v e d that a rostrally n e g a t i v e , but not caudally n e g a t i v e D C field, p r o m o t e s r e c o v e r y of a s c e n d i n g axons in the g u i n e a pig f o l l o w i n g spinal c o r d h e m i s e c tion 7. T h u s , t h e r e is increasing e v i d e n c e that the polarity of an a p p l i e d D E field influences its effects o n axons, In the p r e s e n t e x p e r i m e n t , we e x a m i n e the effect of D C
Under halothane anaesthesia, 30 adult female Wistar rats (284 + 21 g) underwent a C 7 - T 1 laminectomy and received a 1-min clip compression injury of the cord at T 1 with a modified aneurysm clip exerting a force of 53 g, a lesion which results in severe paraparesis or complete paraplegia 28. After injury the rats were randomly and blindly assigned to receive one of the following DC stimulators: sham (0/~A; n = 10); cathode caudal to lesion (14/~A; n = 10); or cathode rostral to lesion (14/~A; n = 10). Each DC stimulator consisted of a 6.0 V battery, a 3.6 kfl resistor, and a junctional field-effect transistor designed to maintain a constant output of 14 ktA over a wide range of resistances 12. The stimulators were connected to a silastic pad by 3.5 cm of insulated platinum wire. One disc-shaped electrode (surface area 2 mm 2 protruded from each end of the silastic pad, and the two electrodes were separated by a distance of 10 mm. With microsurgical techniques, the electrode pad of the stimulator was sutured to the dura. The caudal electrode was placed distal to the injury site under the lamina of T 2 with the rostral electrode positioned under the lamina of C 6. The stimulator apparatus was secured to subcutaneous tissue, and the wound closed.
Correspondence: C.H. Tator, Lab. 12-423, McLaughlin Wing, Playfair Neuroscience Unit, Toronto Western Hospital, 399 Bathurst St., Toronto, Ontario, M5T 2S8, Canada.
33 The current output and voltage of the DC stimulators was tested prior to insertion and following the 8-week duration of the experiment by techniques described elsewhere 12. The current output of each stimulator was measured in vitro, using a digital multimeter, over a wide range of imposed resistances (1 kfl-20 MQ), with the unit immersed in saline to mimic in vivo conditions. During removal of some stimulators at postmortem, the disc-shaped stimulating electrodes became detached. In these cases, current output was measured from the distal end of the platinum wire directly, All outcome measures were assessed in a 'blinded' fashion. From the 1st to the 8th postoperative week, the clinical neurological function of each rat was assessed weekly using the inclined plane technique 35. In this study, in contrast to a previous investigation 12, rats were positioned laterally with respect to the long axis of the inclined plane to minimize the contribution of forelimbs to the inclined plane score. Thus the inclined plane scores are approximately 10-15 ° less in this study than we have reported elsewhere 12 for similar severities of injury. At 8 weeks after injury, electrophysiological assessment was performed by recording motor- and somatosensory-evoked potentials (MEP and SSEP) 12-14. Under halothane anaesthesia, and with the animal placed in a Kopf craniospinal stereotaxic frame, a right parietal craniectomy was performed to expose the hindlimb region of the sensorimotor cortex (SMC) 8"13'17. A 0.8 mm platinum ball electrode was placed extradurally over the SMC, with a reference Ag/AgCI disc electrode under the hard palate. To elicit MEPs, anodal stimuli (15 mA, 300 ms, 4,2. Hz) were applied to the SMC, and the responses differentially recorded from needle electrodes in the contralateral biceps femoris muscle. The SSEPs were elicited by applying cathodal stimuli at 4.2 Hz to the dorsum of the left hindpaw and recorded from the SMC. A total of 50 MEP and 1024 SSEP responses were recorded at a bandwidth of 30-3000 Hz, averaged and replicated, The peak-to-peak amplitude and onset latency of the evoked responses were calculated as described elsewhere 13. Immediately following electrophysiological recording and under the same anesthetic, each rat underwent a complete cord transection at T s and then a Gelfoam pledget impregnated with a 33% solution of horseradish peroxidase (HRP) in 2% dimethylsulfoxide was placed against the rostral stump of the cord. Forty-eight hours after HRP introduction, the animal was transcardially perfused with 10% dextran, 1.25% glutaraldehyde/1.25% paraformaldehyde fixative and sucrose-phosphate buffer~ , after which the entire brain and spinal cord were removed. The brain was subsequently sectioned in the coronal plane at 40/~m with a cryostat. Every fifth section was stained with thionin, and the section adjacent to every fifth section was processed for HRP reactivity using the chromogen tetramethylbenzidine. At 100 x magnification, the number of HRP-labelled cells was counted in the motor cortex and brainstem by two independent observers, blinded as to the treatment code and using previously established criteria 2s. Serial 40/~m longitudihal sections of the HRP introduction site at T s were also processed for HRP reactivity, An additional experiment was performed on five uninjured rats to control for non-specific labelling by HRP due to vascular uptake or diffusion through the subarachnoid space. These rats underwent complete transection of the cord at T 1, were allowed to survive 8 weeks and then underwent placement of the HRP-impregnated pledget against the rostral stump of the cord transected at T s as outlined above. There were no labelled cells in the cortex or brainstem of any of these five control rats, thus ruling out the possibility of non-specific labelling, The segment of spinal cord containing the injury site was sectioned transversely at 400/~m with a vibrotome, postfixed in 2% osmium tetroxide/3% sodium ferrocyanide solution, immersed in 2% aqueous uranyl acetate and embedded in araldite plastic. These blocks were sectioned serially in the transverse plane at 1/tm with an ultramicrotome and stained with Toluidine blue. Counts of surviving myelinated axons at the injury site were achieved by a modification of the line sampling technique of Blight 1'2 (Fig.l). After examination of the serial sections of the in-
jury site, the section containing the largest lesion area from each rat was selected for analysis. Counts were also performed from sections 10/~m rostral and caudal to the section designated to contain the maximum lesion area in eight randomly selected rats. The counts differed by less than 10% (P < 0.05) in each instance. At 25 x magnification, the pial surface of the section was traced with the aid of a camera lucida onto a digitizing tablet interfaced to a microcomputer and reproduced on paper by a plotter. Then 48 radial sample lines were drawn by the plotter from the centre of the cord at even intervals of 2~/48 radians. For axon counting, the slides were then placed on a rotating slide holder attached to the main stage of a Zeiss Universal photomicroscope such that the X and Y movements of the main stage were independent of the rotating movements of the slide holder. A line which could be viewed simultaneously with the microscopic image through the camera lucida was drawn on a sheet of paper and provided a reference line for axon counting. At 1000 x magnification, the number, diameter, and distance from the pia of axons intercepted by each of the 48 radial sampling lines were measured with the aid of a digitizing tablet interfaced to a microcomputer. The estimated total number of axons in the section was derived using the following algorithm based on line-sampling theory"2: ( L - D ) tan ¢p-d
S = ZZ
• ij
d + 2R where S = total axon count in cross section; i = sample line number;/' = any axon intercepted by sample line; n = number of axons intercepted on a sample line; L = sample line length; D = distance of axon from pia; d = diameter of axon; ~ = angle between sample lines = 2~/48 radians; R = zone of uncertainty around axon edge = -+0.1 mm. The inclined plane data were analyzed using multivariate analysis of variance (MANOVA) to account for repeated measures effects38. Post-hoc analyses were performed using one way analysis of variance (ANOVA) and Tukey's HSD test. A variance-stabilizing (square root) transformation of the cell count, axon count and evoked potential amplitude data was performed prior to parametric analyses to correct for unequal variances42. The current output of the DC stimulators before and after 8 weeks of in vivo stimulation were evaluated using paired t-tests. The results have been expressed as mean -+ S.E.M. and differences were considered significant at P < 0.05.
RESULTS Twenty-six rats (nine cathode caudal, nine sham and
eight cathode rostral) survived the 8-week assessment period. Two (one sham, one cathode rostral) of the 26 surviving rats died before completion of electrophysiological a s s e s s m e n t a n d , t h e r e f o r e , e v o k e d p o t e n t i a l a n d H R P d a t a w e r e o b t a i n e d f r o m 24 r a t s ( n i n e c a t h o d e c a u dal, e i g h t s h a m , s e v e n c a t h o d e r o s t r a l ) . T h e c o r d f r o m the two rats which died prior to completion of electrop h y s i o l o g i c a l a s s e s s m e n t w e r e p o s t f i x e d f o r 7 d a y s in glu t a r a l d e h y d e a n d t h e i n j u r y sites w e r e p r o c e s s e d in t h e s a m e m a n n e r as f o r t h e o t h e r r a t s , as d e s c r i b e d a b o v e . In vitro testing of the DC stimulators revealed that the c u r r e n t o u t p u t o f t h e s t i m u l a t o r s d i d n o t d i f f e r signific a n t l y ( P > 0.05) b e f o r e ( 1 4 . 2 0 + 0.10 ~tA) o r a f t e r (13.74 + 0 . 2 2 / ~ A ) t h e 8 w e e k p e r i o d o f in v i v o s t i m u l a t i o n . I n all a n i m a l s , t h e e l e c t r o d e l e a d s w e r e i n t a c t t o
34
Piol Surfoce
I i
,48
x I
'x
32
line 2 /
/
C
~N=x-d
', line I
O
M,..
, o÷2M
Fig. 1. The line sampling technique for quantitative morphometric analysis of spinal cord sections: A: at 25×, the pial surface of each section is traced with the aid of a camera lucida on to a digitizing tablet and reproduced on paper by a plotter. Then, 48 radial sample lines are drawn by the plotter each extending from the center of the cord at equal intervals. B: for each axon intercepted by a sample line, the distance from the pia, the axon diameter, and myelin thickness are measured at 1000× with the aid of a digitizing tablet. C: derivation of the formula for computing the estimated number of axons within a sector, where R = 0.1 ~m and is a correction factor based on the limit of resolution of the light microscope. D: the method of calculating the myelination index.
gross inspection, although it was usually not possible to inspect the entire length of the electrode wires which at autopsy were encased in dense scar. There was no visible external evidence of electrode induced neural damage in any of the 26 rats which survived the 8-week assessment period. Multivariate analysis of the inclined plane data (Fig. 2) revealed that the scores varied significantly with time ( P < 0.0001) and were significantly different among the three experimental groups (time x group interaction: P < 0.02). At 8 weeks, the inclined plane scores for the sham, cathode rostral and cathode caudal groups were 33.0 + 1.9, 28.9 + 2.8 and 47.5 + 2.3, respectively. One way A N O V A and Tukey's HSD test showed that the inclined plane scores of the cathode caudal group significantly exceeded those of the sham and cathode rostral groups at 7 (P < 0.001) and 8 weeks (P < 0.0001) postinjury, while the scores of the latter two groups were similar at all time points (P > 0.05). The MEP data are summarized in Fig. 3. Two of the
eight rats in the sham group and two of the seven animals in the cathode rostral group had recordable MEPs, although they were of low amplitude less than (10/~V). In contrast, MEPs were present in seven of nine rats treated with a cathode caudal DC field (Z2 = 6.25; df = 2; P < 0.05) and these responses exceeded 20/~V in five rats (Fig. 3a). One-way A N O V A revealed that the MEP amplitude was significantly different (P < 0.02) among the three treatment groups. Post-hoc analysis with Tukey's test HSD revealed that the MEP amplitude of the cathode caudal group significantly exceeded (P < 0.05) that of the sham and cathode rostral groups. The MEPs of the latter two groups were similar (P > 0.05). Fig. 4 summarizes the SSEP data. Representative waveforms are depicted in Fig. 4a-d. The distribution of rats with no detectable SSEP (n = 18), SSEPs with a peakpeak amplitude of less than 2 I~V (n = 4) and SSEPs with an amplitude of more than 2/~V (n = 2) was similar among the three groups 2'2 = 1.4; P > 0.05; Fig. 4e). Parametric analysis of SSEP amplitude by univariate
35
techniques showed that the p e a k - p e a k amplitude was similar (P > 0.05) among the three experimental groups, The site of H R P introduction was located 34.7 + 0.6 m m distal to the injury site as measured at postmortem, Microscopic examination revealed that the H R P introduction site, including the zone of extracellular diffusion, extended 9.4 + 1.3 m m rostral from the introduction site and was, therefore, approximately 25 mm distal to the injury site in each rat. HRP-labelled neurons were confined to the following locations: layer V of the SMC, in a distribution similar to that described previously2S; the magnocellular red nucleus; the lateral, medial and inferior vestibular nuclei; the reticular formation (nucleus gigantocellularis, nucleus paragigantocellularis and the caudal aspect of the pontine reticular formation); and the raphe nuclei (primarily nucleus raphe magnus). Fig. 5 shows examples of labelled neurons from the red nucleus and nucleus raphe magnus from the three experimental groups. As shown in Table I, the cathode caudal group had a significant increase in the total number (from the section adjacent to every fifth section as described above) of retrogradely labelled neurons (P < 0.001) and greater numbers of labelled cells in the red nucleus (P < 0.001), raphe nuclei (P < 0.01) and reticular formation (P < 0.03) than the other two groups. In contrast, there was
no difference in the number of labelled neurons in the vestibular nuclei or motor cortex. Post-hoc analysis revealed that the total number of retrogradely labelled neurons was greater (P < 0.05) in the sham group (13.7 + 6.6) than the cathode rostral group (1.0 + 0.5). Examination of the Toluidine blue stained sections of the injury site (Fig. 6) revealed that both the sham and cathode rostral groups had a paucity of axons in the white matter and that the few residual axons were of small caliber (0.5-2.0/~m), thinly myelinated, and concentrated in the outer 250/~m of the cord. In a number of sections from rats treated with a cathode caudal DC field, there was a striking increase in the number of surviving myelinated axons, particularly in the dorsolateral and ventrolateral tracts of the cord. The distribution of residual axons in the cathode caudal group appeared otherwise qualitatively similar to that of the other two groups. The quantitative analysis of counts of myelinated axons at the injury site is summarized in Table II. Each section through the injury site was partitioned into six equal sectors (using the 48 radial sampling lines) to derive an estimate of the distribution of residual axons within the dorsal, dorsolateral (both sides pooled), ventral and ventrolateral (both sides pooled) tracts, respectively. The group treated with the cathode caudal DC
Inclined Plane Score vs Week Postinjury 60,
50' CathodeCaudal
®
40"
O
----O-- smm 30"
~--
•~
.IL
20
CathodeRostral
" p 0.05) among the three experimental groups (pooled mean for myelination index =0.68 + 0.01). DISCUSSION The present study supports previous findings that DC fields promote recovery after acute experimental spinal cord injury in rats 12'15'34'45. In addition, these data confirm the importance of DC field polarity in promoting functional recovery of injured central axons in vivo. In the present investigation, with a caudally negative field, an improvement was observed in four outcome parameters: inclined plane scores; motor-evoked potential amplitude; the number of retrogradely labelled rubrospinal, raphespinal and reticulospinal neurons; and the number of residual, intact axons at the injury site. In contrast, a DC field with the cathode oriented rostral to the lesion failed to improve any of these outcome measures. In-
deed, a cathode rostral DC field accentuated the extent of degeneration of descending axons in the spinal cord, as shown by the reduction in the number of retrogradely labelled neurons in this group as compared with the sham group (Table I). The findings in the present study are congruent with a number of other investigations which have examined the issue of field polarity. For example, numerous in vitro studies have found that neurites preferentially grow toward the cathode of an applied DC field whereas the growth of anodally facing neurites is retarded 19'24'25'3°. Roederer et al. 36 examined the influence of DC fields on retrograde degeneration of reticulospinal axons in the lamprey cord. A cathode caudal DC field reduced retrograde degeneration of severed descending axons whereas reversal of field polarity accentuated the extent of retrograde degeneration. Politis and Zanakis 34 studied the effect of delayed application of a DC field on recovery after contusion injury of the thoracic cord of rats.
Fig. 6. A: this section taken from the ventrolateral funiculus (near the pial edge) of a sham-treated rat illustrates a paucity of intact myelinated axons (arrows). In addition, this photomicrograph illustrates astrogliosis, n u m e r o u s degenerating myelin profiles (d) and microcystic degeneration (mc) with macrophage infiltration. This section is representative of both s h a m and cathode rostral groups (calibration bar = 2 gin). B: this section taken from the ventrolateral funiculus (100 g m from the pial edge) illustrates n u m e r o u s intact myelinated axons (arrow) ranging in diameter from approximately 0 . 5 - 4 . 0 / ~ m . A few degenerating myelin profiles are also evident. This section is one of the best examples of axon survival and is from an animal treated with a cathode caudal DC field (calibration bar = 2 pm).
40 While rats treated with both a caudally negative and rostrally negative DC field showed greater recovery of inclined plane performance, only the cathode rostral group had a greater number of axons in the dorsal columns as assessed by immunocytochemical analysis of neurofilament protein. Recently, Borgens et al. 7 examined the effect of DC fields on functional recovery of guinea pigs with spinal cord hemisection. Only animals treated with a cathode rostral DC feld showed recovery of the cutaneous trunci muscle reflex, a measure which reflects the integrity of ascending axons in the spinal cord. In contrast, animals treated with sham or cathode caudal stimulators did not recover this reflex. Thus there is convincing evidence that DC field polarity greatly influences the recovery and possibly regeneration of injured axons in the spinal cord. Although the data from the present study support the conclusion that DC field polarity strongly influences the survival of axons after compressive injury it is unclear why stimulation with the cathode rostral to the lesion did not promote recovery of SSEPs or enhance survival of axons in the dorsal columns. It is possible that the sublaminar placement of the disc-shaped stimulating electrodes may have produced compression of the dorsal columns and interfered with axonal conduction, Alternatively, the dorsally applied DC fields may have induced electrolytic injury to the dorsal columns without obvious external evidence of damage. Our laboratory has recently reported that DC stimulation with currents as low as 3/~A can have neurotoxic effectsTM which are most striking in the dorsal columns. A histopathological examination of the spinal cord subjacent to the stimulating electrodes was not conducted in the present study and thus it is not possible to state whether there was demyelination or axonal loss at these sites. We are presently examining different electrode configurations to produce more homogeneous dispersion of current and thus avoid the potentially damaging effects of focal stimulation over the dorsum of the cord by sublaminar electrodes. In addition, a detailed neuropathological examination of the spinal cord adjacent to the DC-stimulating electrodes is underway in a separate investigation. The location and distribution of retrogradely labelled neurons (Table I) in the brainstem and sensorimotor cortex seen after introduction of the tracer into the cord at T 8 is consistent with previous descriptions of the main descending pathways in the rat spinal cord 28'29'43'44. Furthermore, the dependence of return of hindlimb locomotor function on non-pyramidal tracts is again confirmed in the present study 9'28. A review of Table I and Figs. 2 and 3 will show that despite the virtually complete ablation of the corticospinal tract, rats treated with a cathode caudal DC field achieved considerable return
of function as judged by inclined plane performance and MEP amplitude. It is noteworthy that counts of retrogradely labelled neurons in the red nucleus, raphe nuclei and reticular formation were increased in this group, suggesting that these tracts may in part subserve hindlimb locomotor function. The beneficial effects of DC fields on recovery after acute spinal cord injury could be due to: (a) enhanced survival of sublethally injured axons; (b) enhanced collateral sprouting by intact fibres; or (c) regeneration of transected axons. Although it is theoretically possible that DC fields could have promoted regeneration of transected axons, this appears improbable because the severed fibers would have had to grow at least 3.5 cm (from the injury site at T 1 to the site of HRP introduction at T s, and to have formed functional synapses with the motor neurons innervating trunk or lower limb musculature below this level. To date functionally significant axonal regeneration has not been convincingly demonstrated in the mammalian spinal cord 1°'22. To examine whether DC fields could promote central axonal regeneration would require an alternative injury model and other outcome measures such as transection of selected tracts of the cord or complete cord transection followed by the examination of axonal integrity by fiber tracing methods. A compression model was selected for the present study as it accurately simulates one of the most common mechanisms of acute human spinal cord injury 11. Indeed, the spinal cord is rarely transected after injury in humans. It is possible that DC fields enhanced axonal sprouting at the site of cord injury. It has been shown that DC fields promote neuritic outgrowth ~9'24'25'3° and enhance cell survival2° in vitro, and there is some evidence that this may also occur in vivo5-7. Although sprouting may partly explain the increased counts of myelinated axons at the site of cord injury, sprouting into deafferented zones of the cord is usually by small unmyelinated fibres 46. However, axonal sprouting would not result in an increase in the number of retrogradely labelled neurons (Table I), since the site of HRP introduction was over 30 mm from the injury site. Nevertheless, it is possible that axonal sprouting could augment local spinal cord reflex mechanisms 16 and thus improve behavioural assessments of outcome and possibly evoked potential measurements. Examination for sprouting in future studies could be assessed by electron microscopic analysis of anterogradely labelled terminals at the site of cord injury and at sites rostral and caudal to the lesion and by immunocytochemical techniques to localize different neurotransmitter systems46. The likeliest explanation for our findings is that the DC fields attenuated the process of secondary injury and
41 p r o m o t e d recovery of sublethally injured fibers in the spinal cord. Borgens et al. a m e a s u r e d "injury currents" of 0.5 mAJcm 2 consisting mainly of N a + and Ca 2+, which • e n t e r e d the acutely transected ends of l a m p r e y spinal cord axons. M o r e recently, Kerns et al. 21 used the vibrating p r o b e technique to confirm the presence of such "injury currents" at the tips of transected m a m m a l i a n p e r i p h e r a l nerve axons. It is n o t e w o r t h y that the intracellular influx of cations, in particular Ca 2÷, appears to • play a key role in axonal d e g e n e r a t i o n 39'4°. In the lamprey, R o e d e r e r et al. a6 d e m o n s t r a t e d that a distally negative D C field r e d u c e d the extent of r e t r o g r a d e degeneration of transected reticulospinal axons. The authors explained their result by hypothesizing that D C fields r e d u c e d the entry of N a ÷ and Ca 2+ into the cut ends of the spinal cord axons. R e c e n t experiments by Strautman et al. 41 which d e m o n s t r a t e that a caudally negative D C field attenuates intracellular entry of Ca 2÷ into transected axons support this hypothesis. It is possible that similar "injury currents" could result from compressive injuries which cause focal disruption of the myelin sheath and a x o l e m m a of an axon or cause transection of m a j o r branches of the main axon. A n intriguing possibility is
that the D C field could o v e r c o m e conduction block at points of axonal bifurcation by alteration in m e m b r a n e potential 23. Currently, our l a b o r a t o r y is measuring the magnitude of the current density at various depths within the spinal cord to d e t e r m i n e w h e t h e r the applied field is of sufficient magnitude to attenuate posttraumatic "injury currents" in the rat spinal cord. In conclusion, D C stimulation appears to be a p r o m ising technique for p r o m o t i n g recovery after acute exp e r i m e n t a l spinal cord injury. The present study establishes the i m p o r t a n c e of field polarity in vivo. A d d i t i o n a l work is required to establish safe levels of D C stimulation, in view of recent reports from our l a b o r a t o r y of potential neurotoxicity TM,and to further define the mechanism of action.
Acknowledgements. Dr. Fehlings was a Fellow of the Medical Research Council of Canada (MRC) during the tenure of this investigation. This research was supported by grants from the MRC and from the University Research Incentive Fund of the Ontario Ministry of Colleges and Universities. The direct current stimulators were provided by Medtronic, Inc. for which grateful acknowledgement is made. The authors thank D. Nedelcu, R. Van Bendegem and M. Moncada for technical assistance.
REFERENCES 1 Blight, A.R., Cellular morphology of chronic spinal cord injury in the cat: analysis of myelinated axons by line-sampling, Neuroscience, 10 (1983) 521-543. 2 Blight, A.R. and Decrescito, V., Morphometric analysis of experimental spinal cord injury in the cat: the relation of injury intensity to survival of myelinated axons, Neuroscience, 19 (1986) 321-341. 3 Borgens, R.B., Jaffe, L.E. and Cohen, M.J., Large and persistent electric currents enter the transected lamprey spinal cord, Proc. Natl. Acad. Sci., USA, 77 (1980) 1209-1213. 4 Borgens, R.B., Roederer, E. and Cohen, M.J., Enhanced spinal cord regeneration in lamprey by applied electrical fields, Science, 213 (1981) 611-617. 5 Borgens, R.B., Blight, A.R. and Murphy, D.J., Transected dorsal column axons within the guinea pig spinal cord regenerate in the presence of an applied electrical field, J. Comp. Neurol., 250 (1986) 168-180. 6 Borgens, R.B., Blight, A.R. and McGinnis, M.E., Behavioural recovery induced by applied electric fields after spinal cord hemisection in guinea pig, Science, 238 (1987) 366-369. 7 Borgens, R.B., Blight, A.R. and McGinnis, M.E., Functional recovery after spinal cord hemisection in guinea pigs: the effects of applied electric fields, J. Comp. Neurol., 296 (1990) 634-653. 8 Donaghue, J.P. and Wise, S.P., The motor cortex of the rat: cytoarchitecture and microstimulation mapping, J. Comp. Neurol., 212 (1982) 76-88. 9 Eidelberg, E., Story, J.L., Walden, J.G., Meyer, B.L., Anatomical correlates of return of locomotor function after partial spinal cord lesions in cats, Exp. Brain Res., 42 (1981) 81-88. 10 Fehlings, M.G. and Tator, C.H., A. review of spinal cord regeneration, Adv. Trauma, 3 (1988) 223-240. 11 Fehlings, M.G. and Tator, C.H., A review of experimental models of acute spinal cord injury. In L. Illis (Ed.), Spinal Cord Dysfunction, Oxford University Press, Oxford, 1988, pp. 3-33. 12 Fehlings, M.G., Tator, C.H. and Linden, R.D., The effect of
13 14
15 16 17 18 19 20
21
22 23
an applied direct current field on recovery from acute experimental spinal cord injury, J. Neurosurg., 68 (1988) 781-792. Fehlings, M.G., Tator, C.H., Linden, R.D. and Piper, I.R., Motor and somatosensory evoked potentials recorded from the rat, EEG Clin. Neurophysiol., 69 (1988) 65-78. Fehlings, M.G., Tator, C.H. and Linden, R.D., The relationships among the severity of spinal cord injury, motor and somatosensory evoked potentials and spinal cord blood flow, EEG Clin. Neurophysiol., 74 (1989) 241-259. Fehlings, M.G., Wong, T.H., Tator, C.H. and Tymianski, M., The effect of a direct current field on axons after experimental spinal cord injury, Can. J. Surg., 32 (1989) 188-191. Freed, W.J., de Medinaceli, L. and Wyatt, R.J., Promoting functional plasticity in the damaged nervous system, Science, 222 (1985) 1522-1544. Hall, R. and Lindholm, E., Organization of the motor and somatosensory cortex of the albino rat, Brain Res., 66 (1974) 2338. Hurlbert, R.J., Theriault, E. and Tator, C.H., Neurotoxicity from direct current stimulation, Soc. Neurosci. Abstr., 16(2) (1990) 1115. Jaffe, L.E and Poo, M.M., Neurites grow faster towards the cathode than the anode in a steady field, J. Exp. Zool., 209 (1979) 115-128. Kaplan, ES., Maguire, T.G., Lee, U.M., Seizer, M.E., Spindler, K., Black, J. and Brighton, C.T., Direct extracellular electrical stimulation influences density dependent aggregation of fetal rat cerebrocortical neurons in vitro, Neurosci. Lett., 94 (1988) 33-38. Kerns, J.M., Pakovic, I.M., Fakhouri, A.J., Wickersham, K.L. and Freeman, J.A., An experimental implant for applying an electrical field to peripheral nerve, J. Neurosci. Methods, 19 (1987) 217-223. Kiernan, J.A., Hypotheses concerned with axonal regeneration in the mammalian nervous system, Biol. Rev., 54 (1979) 15197. Krathamer, V., Modulation of conduction at points of axonal
42
24
25
26
27
28
29 30 31
32
33
34
35
bifurcation by applied electric fields, 1EEE Trans Biomed. Engin., 37 (1990) 515-519. Marsh, G. and Beams, H.W., In vitro control of growing chick nerve fibers by applied electric currents, J. Cell Comp. Physiol., 27 (1946) 139-157. McCaig, C.D., Spinal neurite reabsorption and regrowth in vitro depend on the polarity of an applied electric field, Development, 100 (1987) 31-41. McCaig, C.D., Nerve growth in a small applied electric field and the effects of pharmacological agents on rate and orientation, J. Cell Sci., 95 (1990) 617-622. McDevitt, L., Fortner, P. and Pomeranz, B., Application of a weak electric field to the hindpaw enhances sciatic motor nerve regeneration in the adult rat, Brain Res., 416 (1987) 308-314. Midha, R., Fehlings, M.G., Tator, C.H., Saint-Cyr, J.A. and Guha, A., Assessment of spinal cord injury by counting corticospinal and rubrospinal neurons, Brain Res., 410 (1987) 299308. Murray, H.M. and Gurule, M.E., Origin of the rubrospinal tract of the rat, Neurosci. Lett., 14 (1979) 19-23. Patel, N.B. and Poo, M.M., Orientation of neurite outgrowth by extracellular electric fields, J. Neurosci., 2 (1982) 19-23. Politis, M.J. and Zanakis, M.E, Treatment of the damaged rat hippocampus with a locally applied electric field, Exp. Brain Res., 71 (1988) 223-226. Politis, M.J. Zanakis, M.E and Albala, B.J., Facilitated regeneration in the rat peripheral nervous system using applied electric fields, J. Trauma, 28 (1988) 1375-1381. Politis, M.J., Zanakis, M.E and Albela, B.J., Mammalian optic nerve regeneration following the application of electric fields, J. Trauma, 28 (1988) 1548-1552. Politis, M.J. and Zanakis, M.F., The short term effects of delayed application of electric fields in the damaged rodent spinal cord, Neurosurgery, 25 (1989) 71-75. Rivlin, A.S. and Tator, C.H., Objective clinical assessment of motor function after experimental spinal cord injury in the rat, J. Neurosurg., 47 (1977) 577-581.
36 Roederer, E., Goldberg, N.H. and Cohen, M.J., Modification of retrograde degeneration in transected spinal axons of lamprey by applied DC current, J. Neurosci., 3 (1983) 153-160. 37 Roman, G.C., Strahlendorf, H.K., Coates, P.W. and Rowley, B.A., Stimulation of sciatic nerve regeneration in the adult rat by low intensity electric current, Exp. Neurol., 98 (1987) 222232. 38 SAS Institute, Inc., SAS User's Guide Statistics, Version 5 edn., SAS Institute, Inc, Cary, NC, 1985. 39 Schlaepfer, W.W., Calcium induced degeneration of axoplasm in isolated segments of rat peripheral nerve, Brain Res., 69 (1974) 203-215. 40 Schlaepfer, W.W. and Haster, M.B., Characterization of the calcium-induced disruption of neurofilaments in rat peripheral nerve, Brain Res., 168 (1979) 299-309. 41 Strautman, A.E, Cork, R.J. and Robinson, K.R., The distribution of free calcium in transected spinal axons and its modulation by applied electrical fields, J. Neurosci., 10 (1990) 35643575. 42 Snedecore, G.W. and Cochrane, W.G., Statistical Methods, 7th edn. Iowa State University Press, Ames, IA, 1980, pp. 287-288. 43 Tracey, D.J., Ascending and descending pathways in the rat spinal cord. In G. Paxinos (Ed.), The Rat Spinal Cord, Vol. 2, Hindbrain and Spinal Cord, Academic Press, Sydney, 1985, pp. 311-324. 44 Waldron, H.A. and Gwyn, D.G., Descending nerve tracts in the spinal cord of the rat. 1. Fibers from the midbrain, J. Comp. Neurol., 137 (1969) 143-154. 45 Wallace, M.C., Tator, C.H. and Piper, I.R., Recovery of spinal cord function induced by direct current stimulation of the injured rat spinal cord, Neurosurgery, 20 (1987) 878-884. 46 Wang, S.D., Goldberger, M.E. and Murray, M., Plasticity of spinal systems after unilateral dorsal rhizotomy in the adult rat, J. Comp. Neurol., 304 (1991) 555-568. 47 Zanakis, M.E, Differential effects of various electrical parameters on peripheral and central nerve regeneration, Acupunct. Electrother. Res., 15 (1990) 185-191.
32
BRES 17663
The effect of direct current field polarity on recovery after acute experimental spinal cord injury Michael G. Fehlings and Charles H. Tator Canadian Paraplegic Association Spinal Cord Injury Research Laboratory, Playfair Neuroscience Unit and Division of Neurosurgery, Toronto Hospital, Western Division, and University of Toronto, Toronto (Canada) (Accepted 10 December 1991)
Key words: Rat; Electrical stimulation; Horseradish peroxidase; Motor-evoked potential; Somatosensory-evoked potential
Recent evidence indicates that direct current (DC) fields promote recovery of acutely injured central and peripheral nervous system axons. The polarity of the applied DC field may play an important role in modulating these effects. In the present study, the effect of DC field polarity on recovery of injured spinal cord axons was examined anatomically, electrophysiologically and behaviourly in a rat model. After a 53 g clip compression injury of the cord at T1, 30 adult rats were randomly and blindly allocated to one of three groups (n = 10 each): one group received implantation of a DC stimulator (14/~A) with the cathode caudal to the injury site; the second group received implantation of a similar stimulator with the cathode rostral to the injury site; and the third group received a sham (O/~A) stimulator. Clinical neurological function was assessed by the inclined plane technique and axonal function was assessed by motor- and somatosensory-evoked potentials (MEP and SSEP). A quantitative assessment of axonal integrity was performed by counting neurons in the brain retrogradely labelled by the axonal tracer horseradish peroxidase (HRP) and by counting axons at the injury site. The inclined plane scores (P < 0.0001), MEP amplitude (P < 0.02), counts of neurons retrogradely labelled by HRP (P < 0.0001), and axon counts at the injury site (P < 0.01) were significantly greater in the group treated with a DC field with the cathode caudal to the lesion than in the other two groups. Conversely, the cathode rostral DC field caused a decrease in the number of neurons retrogradely labelled by HRP (P < 0.05) compared to the sham and cathode caudal groups. These data confirm our previous finding that DC fields promote recovery of acutely injured spinal cord axons. Furthermore, the polarity of the applied field is of critical importance to this effect. INTRODUCTION
field p o l a r i t y on r e c o v e r y of m a m m a l i a n spinal c o r d axons s u b j e c t e d to a s e v e r e c o m p r e s s i v e injury. This ques-
There
is c o n s i d e r a b l e
e v i d e n c e that direct c u r r e n t
tion is o f f u n d a m e n t a l i m p o r t a n c e in u n d e r s t a n d i n g the
( D C ) fields p r o m o t e the r e c o v e r y and r e g e n e r a t i o n of injured p e r i p h e r a l 27'32'37'47 and central n e r v o u s system
m e c h a n i s m of action o f D C fields on injured axons.
axons 4'5'6'7'12'15'26'31'33'34'45. It has b e e n s h o w n in vitro
MATERIALS AND METHODS
that the polarity of the D C field is o f critical i m p o r t a n c e to the m e c h a n i s m of t h e s e effects 19'24'25'3°. F o r e x a m p l e , the t r o p h i c effects of D C fields o n n e u r o n s in vitro are polarity specific19'24'25'3°: n e u r i t e s p r e f e r e n t i a l l y g r o w tow a r d the c a t h o d e , w h e r e a s a n o d a l l y facing n e u r i t e s und e r g o a b s o r p t i o n 25. A n u m b e r of studies c o n d u c t e d in v i v o h a v e s h o w n similar effects. F o r e x a m p l e , a caudally n e g a t i v e D C field has b e e n r e p o r t e d to r e d u c e the extent of r e t r o g r a d e d e g e n e r a t i o n of t r a n s e c t e d reticulospinal axons in the l a m p r e y spinal c o r d 36. R e c e n t l y , it has b e e n o b s e r v e d that a rostrally n e g a t i v e , but not caudally n e g a t i v e D C field, p r o m o t e s r e c o v e r y of a s c e n d i n g axons in the g u i n e a pig f o l l o w i n g spinal c o r d h e m i s e c tion 7. T h u s , t h e r e is increasing e v i d e n c e that the polarity of an a p p l i e d D E field influences its effects o n axons, In the p r e s e n t e x p e r i m e n t , we e x a m i n e the effect of D C
Under halothane anaesthesia, 30 adult female Wistar rats (284 + 21 g) underwent a C 7 - T 1 laminectomy and received a 1-min clip compression injury of the cord at T 1 with a modified aneurysm clip exerting a force of 53 g, a lesion which results in severe paraparesis or complete paraplegia 28. After injury the rats were randomly and blindly assigned to receive one of the following DC stimulators: sham (0/~A; n = 10); cathode caudal to lesion (14/~A; n = 10); or cathode rostral to lesion (14/~A; n = 10). Each DC stimulator consisted of a 6.0 V battery, a 3.6 kfl resistor, and a junctional field-effect transistor designed to maintain a constant output of 14 ktA over a wide range of resistances 12. The stimulators were connected to a silastic pad by 3.5 cm of insulated platinum wire. One disc-shaped electrode (surface area 2 mm 2 protruded from each end of the silastic pad, and the two electrodes were separated by a distance of 10 mm. With microsurgical techniques, the electrode pad of the stimulator was sutured to the dura. The caudal electrode was placed distal to the injury site under the lamina of T 2 with the rostral electrode positioned under the lamina of C 6. The stimulator apparatus was secured to subcutaneous tissue, and the wound closed.
Correspondence: C.H. Tator, Lab. 12-423, McLaughlin Wing, Playfair Neuroscience Unit, Toronto Western Hospital, 399 Bathurst St., Toronto, Ontario, M5T 2S8, Canada.
33 The current output and voltage of the DC stimulators was tested prior to insertion and following the 8-week duration of the experiment by techniques described elsewhere 12. The current output of each stimulator was measured in vitro, using a digital multimeter, over a wide range of imposed resistances (1 kfl-20 MQ), with the unit immersed in saline to mimic in vivo conditions. During removal of some stimulators at postmortem, the disc-shaped stimulating electrodes became detached. In these cases, current output was measured from the distal end of the platinum wire directly, All outcome measures were assessed in a 'blinded' fashion. From the 1st to the 8th postoperative week, the clinical neurological function of each rat was assessed weekly using the inclined plane technique 35. In this study, in contrast to a previous investigation 12, rats were positioned laterally with respect to the long axis of the inclined plane to minimize the contribution of forelimbs to the inclined plane score. Thus the inclined plane scores are approximately 10-15 ° less in this study than we have reported elsewhere 12 for similar severities of injury. At 8 weeks after injury, electrophysiological assessment was performed by recording motor- and somatosensory-evoked potentials (MEP and SSEP) 12-14. Under halothane anaesthesia, and with the animal placed in a Kopf craniospinal stereotaxic frame, a right parietal craniectomy was performed to expose the hindlimb region of the sensorimotor cortex (SMC) 8"13'17. A 0.8 mm platinum ball electrode was placed extradurally over the SMC, with a reference Ag/AgCI disc electrode under the hard palate. To elicit MEPs, anodal stimuli (15 mA, 300 ms, 4,2. Hz) were applied to the SMC, and the responses differentially recorded from needle electrodes in the contralateral biceps femoris muscle. The SSEPs were elicited by applying cathodal stimuli at 4.2 Hz to the dorsum of the left hindpaw and recorded from the SMC. A total of 50 MEP and 1024 SSEP responses were recorded at a bandwidth of 30-3000 Hz, averaged and replicated, The peak-to-peak amplitude and onset latency of the evoked responses were calculated as described elsewhere 13. Immediately following electrophysiological recording and under the same anesthetic, each rat underwent a complete cord transection at T s and then a Gelfoam pledget impregnated with a 33% solution of horseradish peroxidase (HRP) in 2% dimethylsulfoxide was placed against the rostral stump of the cord. Forty-eight hours after HRP introduction, the animal was transcardially perfused with 10% dextran, 1.25% glutaraldehyde/1.25% paraformaldehyde fixative and sucrose-phosphate buffer~ , after which the entire brain and spinal cord were removed. The brain was subsequently sectioned in the coronal plane at 40/~m with a cryostat. Every fifth section was stained with thionin, and the section adjacent to every fifth section was processed for HRP reactivity using the chromogen tetramethylbenzidine. At 100 x magnification, the number of HRP-labelled cells was counted in the motor cortex and brainstem by two independent observers, blinded as to the treatment code and using previously established criteria 2s. Serial 40/~m longitudihal sections of the HRP introduction site at T s were also processed for HRP reactivity, An additional experiment was performed on five uninjured rats to control for non-specific labelling by HRP due to vascular uptake or diffusion through the subarachnoid space. These rats underwent complete transection of the cord at T 1, were allowed to survive 8 weeks and then underwent placement of the HRP-impregnated pledget against the rostral stump of the cord transected at T s as outlined above. There were no labelled cells in the cortex or brainstem of any of these five control rats, thus ruling out the possibility of non-specific labelling, The segment of spinal cord containing the injury site was sectioned transversely at 400/~m with a vibrotome, postfixed in 2% osmium tetroxide/3% sodium ferrocyanide solution, immersed in 2% aqueous uranyl acetate and embedded in araldite plastic. These blocks were sectioned serially in the transverse plane at 1/tm with an ultramicrotome and stained with Toluidine blue. Counts of surviving myelinated axons at the injury site were achieved by a modification of the line sampling technique of Blight 1'2 (Fig.l). After examination of the serial sections of the in-
jury site, the section containing the largest lesion area from each rat was selected for analysis. Counts were also performed from sections 10/~m rostral and caudal to the section designated to contain the maximum lesion area in eight randomly selected rats. The counts differed by less than 10% (P < 0.05) in each instance. At 25 x magnification, the pial surface of the section was traced with the aid of a camera lucida onto a digitizing tablet interfaced to a microcomputer and reproduced on paper by a plotter. Then 48 radial sample lines were drawn by the plotter from the centre of the cord at even intervals of 2~/48 radians. For axon counting, the slides were then placed on a rotating slide holder attached to the main stage of a Zeiss Universal photomicroscope such that the X and Y movements of the main stage were independent of the rotating movements of the slide holder. A line which could be viewed simultaneously with the microscopic image through the camera lucida was drawn on a sheet of paper and provided a reference line for axon counting. At 1000 x magnification, the number, diameter, and distance from the pia of axons intercepted by each of the 48 radial sampling lines were measured with the aid of a digitizing tablet interfaced to a microcomputer. The estimated total number of axons in the section was derived using the following algorithm based on line-sampling theory"2: ( L - D ) tan ¢p-d
S = ZZ
• ij
d + 2R where S = total axon count in cross section; i = sample line number;/' = any axon intercepted by sample line; n = number of axons intercepted on a sample line; L = sample line length; D = distance of axon from pia; d = diameter of axon; ~ = angle between sample lines = 2~/48 radians; R = zone of uncertainty around axon edge = -+0.1 mm. The inclined plane data were analyzed using multivariate analysis of variance (MANOVA) to account for repeated measures effects38. Post-hoc analyses were performed using one way analysis of variance (ANOVA) and Tukey's HSD test. A variance-stabilizing (square root) transformation of the cell count, axon count and evoked potential amplitude data was performed prior to parametric analyses to correct for unequal variances42. The current output of the DC stimulators before and after 8 weeks of in vivo stimulation were evaluated using paired t-tests. The results have been expressed as mean -+ S.E.M. and differences were considered significant at P < 0.05.
RESULTS Twenty-six rats (nine cathode caudal, nine sham and
eight cathode rostral) survived the 8-week assessment period. Two (one sham, one cathode rostral) of the 26 surviving rats died before completion of electrophysiological a s s e s s m e n t a n d , t h e r e f o r e , e v o k e d p o t e n t i a l a n d H R P d a t a w e r e o b t a i n e d f r o m 24 r a t s ( n i n e c a t h o d e c a u dal, e i g h t s h a m , s e v e n c a t h o d e r o s t r a l ) . T h e c o r d f r o m the two rats which died prior to completion of electrop h y s i o l o g i c a l a s s e s s m e n t w e r e p o s t f i x e d f o r 7 d a y s in glu t a r a l d e h y d e a n d t h e i n j u r y sites w e r e p r o c e s s e d in t h e s a m e m a n n e r as f o r t h e o t h e r r a t s , as d e s c r i b e d a b o v e . In vitro testing of the DC stimulators revealed that the c u r r e n t o u t p u t o f t h e s t i m u l a t o r s d i d n o t d i f f e r signific a n t l y ( P > 0.05) b e f o r e ( 1 4 . 2 0 + 0.10 ~tA) o r a f t e r (13.74 + 0 . 2 2 / ~ A ) t h e 8 w e e k p e r i o d o f in v i v o s t i m u l a t i o n . I n all a n i m a l s , t h e e l e c t r o d e l e a d s w e r e i n t a c t t o
34
Piol Surfoce
I i
,48
x I
'x
32
line 2 /
/
C
~N=x-d
', line I
O
M,..
, o÷2M
Fig. 1. The line sampling technique for quantitative morphometric analysis of spinal cord sections: A: at 25×, the pial surface of each section is traced with the aid of a camera lucida on to a digitizing tablet and reproduced on paper by a plotter. Then, 48 radial sample lines are drawn by the plotter each extending from the center of the cord at equal intervals. B: for each axon intercepted by a sample line, the distance from the pia, the axon diameter, and myelin thickness are measured at 1000× with the aid of a digitizing tablet. C: derivation of the formula for computing the estimated number of axons within a sector, where R = 0.1 ~m and is a correction factor based on the limit of resolution of the light microscope. D: the method of calculating the myelination index.
gross inspection, although it was usually not possible to inspect the entire length of the electrode wires which at autopsy were encased in dense scar. There was no visible external evidence of electrode induced neural damage in any of the 26 rats which survived the 8-week assessment period. Multivariate analysis of the inclined plane data (Fig. 2) revealed that the scores varied significantly with time ( P < 0.0001) and were significantly different among the three experimental groups (time x group interaction: P < 0.02). At 8 weeks, the inclined plane scores for the sham, cathode rostral and cathode caudal groups were 33.0 + 1.9, 28.9 + 2.8 and 47.5 + 2.3, respectively. One way A N O V A and Tukey's HSD test showed that the inclined plane scores of the cathode caudal group significantly exceeded those of the sham and cathode rostral groups at 7 (P < 0.001) and 8 weeks (P < 0.0001) postinjury, while the scores of the latter two groups were similar at all time points (P > 0.05). The MEP data are summarized in Fig. 3. Two of the
eight rats in the sham group and two of the seven animals in the cathode rostral group had recordable MEPs, although they were of low amplitude less than (10/~V). In contrast, MEPs were present in seven of nine rats treated with a cathode caudal DC field (Z2 = 6.25; df = 2; P < 0.05) and these responses exceeded 20/~V in five rats (Fig. 3a). One-way A N O V A revealed that the MEP amplitude was significantly different (P < 0.02) among the three treatment groups. Post-hoc analysis with Tukey's test HSD revealed that the MEP amplitude of the cathode caudal group significantly exceeded (P < 0.05) that of the sham and cathode rostral groups. The MEPs of the latter two groups were similar (P > 0.05). Fig. 4 summarizes the SSEP data. Representative waveforms are depicted in Fig. 4a-d. The distribution of rats with no detectable SSEP (n = 18), SSEPs with a peakpeak amplitude of less than 2 I~V (n = 4) and SSEPs with an amplitude of more than 2/~V (n = 2) was similar among the three groups 2'2 = 1.4; P > 0.05; Fig. 4e). Parametric analysis of SSEP amplitude by univariate
35
techniques showed that the p e a k - p e a k amplitude was similar (P > 0.05) among the three experimental groups, The site of H R P introduction was located 34.7 + 0.6 m m distal to the injury site as measured at postmortem, Microscopic examination revealed that the H R P introduction site, including the zone of extracellular diffusion, extended 9.4 + 1.3 m m rostral from the introduction site and was, therefore, approximately 25 mm distal to the injury site in each rat. HRP-labelled neurons were confined to the following locations: layer V of the SMC, in a distribution similar to that described previously2S; the magnocellular red nucleus; the lateral, medial and inferior vestibular nuclei; the reticular formation (nucleus gigantocellularis, nucleus paragigantocellularis and the caudal aspect of the pontine reticular formation); and the raphe nuclei (primarily nucleus raphe magnus). Fig. 5 shows examples of labelled neurons from the red nucleus and nucleus raphe magnus from the three experimental groups. As shown in Table I, the cathode caudal group had a significant increase in the total number (from the section adjacent to every fifth section as described above) of retrogradely labelled neurons (P < 0.001) and greater numbers of labelled cells in the red nucleus (P < 0.001), raphe nuclei (P < 0.01) and reticular formation (P < 0.03) than the other two groups. In contrast, there was
no difference in the number of labelled neurons in the vestibular nuclei or motor cortex. Post-hoc analysis revealed that the total number of retrogradely labelled neurons was greater (P < 0.05) in the sham group (13.7 + 6.6) than the cathode rostral group (1.0 + 0.5). Examination of the Toluidine blue stained sections of the injury site (Fig. 6) revealed that both the sham and cathode rostral groups had a paucity of axons in the white matter and that the few residual axons were of small caliber (0.5-2.0/~m), thinly myelinated, and concentrated in the outer 250/~m of the cord. In a number of sections from rats treated with a cathode caudal DC field, there was a striking increase in the number of surviving myelinated axons, particularly in the dorsolateral and ventrolateral tracts of the cord. The distribution of residual axons in the cathode caudal group appeared otherwise qualitatively similar to that of the other two groups. The quantitative analysis of counts of myelinated axons at the injury site is summarized in Table II. Each section through the injury site was partitioned into six equal sectors (using the 48 radial sampling lines) to derive an estimate of the distribution of residual axons within the dorsal, dorsolateral (both sides pooled), ventral and ventrolateral (both sides pooled) tracts, respectively. The group treated with the cathode caudal DC
Inclined Plane Score vs Week Postinjury 60,
50' CathodeCaudal
®
40"
O
----O-- smm 30"
~--
•~
.IL
20
CathodeRostral
" p 0.05) among the three experimental groups (pooled mean for myelination index =0.68 + 0.01). DISCUSSION The present study supports previous findings that DC fields promote recovery after acute experimental spinal cord injury in rats 12'15'34'45. In addition, these data confirm the importance of DC field polarity in promoting functional recovery of injured central axons in vivo. In the present investigation, with a caudally negative field, an improvement was observed in four outcome parameters: inclined plane scores; motor-evoked potential amplitude; the number of retrogradely labelled rubrospinal, raphespinal and reticulospinal neurons; and the number of residual, intact axons at the injury site. In contrast, a DC field with the cathode oriented rostral to the lesion failed to improve any of these outcome measures. In-
deed, a cathode rostral DC field accentuated the extent of degeneration of descending axons in the spinal cord, as shown by the reduction in the number of retrogradely labelled neurons in this group as compared with the sham group (Table I). The findings in the present study are congruent with a number of other investigations which have examined the issue of field polarity. For example, numerous in vitro studies have found that neurites preferentially grow toward the cathode of an applied DC field whereas the growth of anodally facing neurites is retarded 19'24'25'3°. Roederer et al. 36 examined the influence of DC fields on retrograde degeneration of reticulospinal axons in the lamprey cord. A cathode caudal DC field reduced retrograde degeneration of severed descending axons whereas reversal of field polarity accentuated the extent of retrograde degeneration. Politis and Zanakis 34 studied the effect of delayed application of a DC field on recovery after contusion injury of the thoracic cord of rats.
Fig. 6. A: this section taken from the ventrolateral funiculus (near the pial edge) of a sham-treated rat illustrates a paucity of intact myelinated axons (arrows). In addition, this photomicrograph illustrates astrogliosis, n u m e r o u s degenerating myelin profiles (d) and microcystic degeneration (mc) with macrophage infiltration. This section is representative of both s h a m and cathode rostral groups (calibration bar = 2 gin). B: this section taken from the ventrolateral funiculus (100 g m from the pial edge) illustrates n u m e r o u s intact myelinated axons (arrow) ranging in diameter from approximately 0 . 5 - 4 . 0 / ~ m . A few degenerating myelin profiles are also evident. This section is one of the best examples of axon survival and is from an animal treated with a cathode caudal DC field (calibration bar = 2 pm).
40 While rats treated with both a caudally negative and rostrally negative DC field showed greater recovery of inclined plane performance, only the cathode rostral group had a greater number of axons in the dorsal columns as assessed by immunocytochemical analysis of neurofilament protein. Recently, Borgens et al. 7 examined the effect of DC fields on functional recovery of guinea pigs with spinal cord hemisection. Only animals treated with a cathode rostral DC feld showed recovery of the cutaneous trunci muscle reflex, a measure which reflects the integrity of ascending axons in the spinal cord. In contrast, animals treated with sham or cathode caudal stimulators did not recover this reflex. Thus there is convincing evidence that DC field polarity greatly influences the recovery and possibly regeneration of injured axons in the spinal cord. Although the data from the present study support the conclusion that DC field polarity strongly influences the survival of axons after compressive injury it is unclear why stimulation with the cathode rostral to the lesion did not promote recovery of SSEPs or enhance survival of axons in the dorsal columns. It is possible that the sublaminar placement of the disc-shaped stimulating electrodes may have produced compression of the dorsal columns and interfered with axonal conduction, Alternatively, the dorsally applied DC fields may have induced electrolytic injury to the dorsal columns without obvious external evidence of damage. Our laboratory has recently reported that DC stimulation with currents as low as 3/~A can have neurotoxic effectsTM which are most striking in the dorsal columns. A histopathological examination of the spinal cord subjacent to the stimulating electrodes was not conducted in the present study and thus it is not possible to state whether there was demyelination or axonal loss at these sites. We are presently examining different electrode configurations to produce more homogeneous dispersion of current and thus avoid the potentially damaging effects of focal stimulation over the dorsum of the cord by sublaminar electrodes. In addition, a detailed neuropathological examination of the spinal cord adjacent to the DC-stimulating electrodes is underway in a separate investigation. The location and distribution of retrogradely labelled neurons (Table I) in the brainstem and sensorimotor cortex seen after introduction of the tracer into the cord at T 8 is consistent with previous descriptions of the main descending pathways in the rat spinal cord 28'29'43'44. Furthermore, the dependence of return of hindlimb locomotor function on non-pyramidal tracts is again confirmed in the present study 9'28. A review of Table I and Figs. 2 and 3 will show that despite the virtually complete ablation of the corticospinal tract, rats treated with a cathode caudal DC field achieved considerable return
of function as judged by inclined plane performance and MEP amplitude. It is noteworthy that counts of retrogradely labelled neurons in the red nucleus, raphe nuclei and reticular formation were increased in this group, suggesting that these tracts may in part subserve hindlimb locomotor function. The beneficial effects of DC fields on recovery after acute spinal cord injury could be due to: (a) enhanced survival of sublethally injured axons; (b) enhanced collateral sprouting by intact fibres; or (c) regeneration of transected axons. Although it is theoretically possible that DC fields could have promoted regeneration of transected axons, this appears improbable because the severed fibers would have had to grow at least 3.5 cm (from the injury site at T 1 to the site of HRP introduction at T s, and to have formed functional synapses with the motor neurons innervating trunk or lower limb musculature below this level. To date functionally significant axonal regeneration has not been convincingly demonstrated in the mammalian spinal cord 1°'22. To examine whether DC fields could promote central axonal regeneration would require an alternative injury model and other outcome measures such as transection of selected tracts of the cord or complete cord transection followed by the examination of axonal integrity by fiber tracing methods. A compression model was selected for the present study as it accurately simulates one of the most common mechanisms of acute human spinal cord injury 11. Indeed, the spinal cord is rarely transected after injury in humans. It is possible that DC fields enhanced axonal sprouting at the site of cord injury. It has been shown that DC fields promote neuritic outgrowth ~9'24'25'3° and enhance cell survival2° in vitro, and there is some evidence that this may also occur in vivo5-7. Although sprouting may partly explain the increased counts of myelinated axons at the site of cord injury, sprouting into deafferented zones of the cord is usually by small unmyelinated fibres 46. However, axonal sprouting would not result in an increase in the number of retrogradely labelled neurons (Table I), since the site of HRP introduction was over 30 mm from the injury site. Nevertheless, it is possible that axonal sprouting could augment local spinal cord reflex mechanisms 16 and thus improve behavioural assessments of outcome and possibly evoked potential measurements. Examination for sprouting in future studies could be assessed by electron microscopic analysis of anterogradely labelled terminals at the site of cord injury and at sites rostral and caudal to the lesion and by immunocytochemical techniques to localize different neurotransmitter systems46. The likeliest explanation for our findings is that the DC fields attenuated the process of secondary injury and
41 p r o m o t e d recovery of sublethally injured fibers in the spinal cord. Borgens et al. a m e a s u r e d "injury currents" of 0.5 mAJcm 2 consisting mainly of N a + and Ca 2+, which • e n t e r e d the acutely transected ends of l a m p r e y spinal cord axons. M o r e recently, Kerns et al. 21 used the vibrating p r o b e technique to confirm the presence of such "injury currents" at the tips of transected m a m m a l i a n p e r i p h e r a l nerve axons. It is n o t e w o r t h y that the intracellular influx of cations, in particular Ca 2÷, appears to • play a key role in axonal d e g e n e r a t i o n 39'4°. In the lamprey, R o e d e r e r et al. a6 d e m o n s t r a t e d that a distally negative D C field r e d u c e d the extent of r e t r o g r a d e degeneration of transected reticulospinal axons. The authors explained their result by hypothesizing that D C fields r e d u c e d the entry of N a ÷ and Ca 2+ into the cut ends of the spinal cord axons. R e c e n t experiments by Strautman et al. 41 which d e m o n s t r a t e that a caudally negative D C field attenuates intracellular entry of Ca 2÷ into transected axons support this hypothesis. It is possible that similar "injury currents" could result from compressive injuries which cause focal disruption of the myelin sheath and a x o l e m m a of an axon or cause transection of m a j o r branches of the main axon. A n intriguing possibility is
that the D C field could o v e r c o m e conduction block at points of axonal bifurcation by alteration in m e m b r a n e potential 23. Currently, our l a b o r a t o r y is measuring the magnitude of the current density at various depths within the spinal cord to d e t e r m i n e w h e t h e r the applied field is of sufficient magnitude to attenuate posttraumatic "injury currents" in the rat spinal cord. In conclusion, D C stimulation appears to be a p r o m ising technique for p r o m o t i n g recovery after acute exp e r i m e n t a l spinal cord injury. The present study establishes the i m p o r t a n c e of field polarity in vivo. A d d i t i o n a l work is required to establish safe levels of D C stimulation, in view of recent reports from our l a b o r a t o r y of potential neurotoxicity TM,and to further define the mechanism of action.
Acknowledgements. Dr. Fehlings was a Fellow of the Medical Research Council of Canada (MRC) during the tenure of this investigation. This research was supported by grants from the MRC and from the University Research Incentive Fund of the Ontario Ministry of Colleges and Universities. The direct current stimulators were provided by Medtronic, Inc. for which grateful acknowledgement is made. The authors thank D. Nedelcu, R. Van Bendegem and M. Moncada for technical assistance.
REFERENCES 1 Blight, A.R., Cellular morphology of chronic spinal cord injury in the cat: analysis of myelinated axons by line-sampling, Neuroscience, 10 (1983) 521-543. 2 Blight, A.R. and Decrescito, V., Morphometric analysis of experimental spinal cord injury in the cat: the relation of injury intensity to survival of myelinated axons, Neuroscience, 19 (1986) 321-341. 3 Borgens, R.B., Jaffe, L.E. and Cohen, M.J., Large and persistent electric currents enter the transected lamprey spinal cord, Proc. Natl. Acad. Sci., USA, 77 (1980) 1209-1213. 4 Borgens, R.B., Roederer, E. and Cohen, M.J., Enhanced spinal cord regeneration in lamprey by applied electrical fields, Science, 213 (1981) 611-617. 5 Borgens, R.B., Blight, A.R. and Murphy, D.J., Transected dorsal column axons within the guinea pig spinal cord regenerate in the presence of an applied electrical field, J. Comp. Neurol., 250 (1986) 168-180. 6 Borgens, R.B., Blight, A.R. and McGinnis, M.E., Behavioural recovery induced by applied electric fields after spinal cord hemisection in guinea pig, Science, 238 (1987) 366-369. 7 Borgens, R.B., Blight, A.R. and McGinnis, M.E., Functional recovery after spinal cord hemisection in guinea pigs: the effects of applied electric fields, J. Comp. Neurol., 296 (1990) 634-653. 8 Donaghue, J.P. and Wise, S.P., The motor cortex of the rat: cytoarchitecture and microstimulation mapping, J. Comp. Neurol., 212 (1982) 76-88. 9 Eidelberg, E., Story, J.L., Walden, J.G., Meyer, B.L., Anatomical correlates of return of locomotor function after partial spinal cord lesions in cats, Exp. Brain Res., 42 (1981) 81-88. 10 Fehlings, M.G. and Tator, C.H., A. review of spinal cord regeneration, Adv. Trauma, 3 (1988) 223-240. 11 Fehlings, M.G. and Tator, C.H., A review of experimental models of acute spinal cord injury. In L. Illis (Ed.), Spinal Cord Dysfunction, Oxford University Press, Oxford, 1988, pp. 3-33. 12 Fehlings, M.G., Tator, C.H. and Linden, R.D., The effect of
13 14
15 16 17 18 19 20
21
22 23
an applied direct current field on recovery from acute experimental spinal cord injury, J. Neurosurg., 68 (1988) 781-792. Fehlings, M.G., Tator, C.H., Linden, R.D. and Piper, I.R., Motor and somatosensory evoked potentials recorded from the rat, EEG Clin. Neurophysiol., 69 (1988) 65-78. Fehlings, M.G., Tator, C.H. and Linden, R.D., The relationships among the severity of spinal cord injury, motor and somatosensory evoked potentials and spinal cord blood flow, EEG Clin. Neurophysiol., 74 (1989) 241-259. Fehlings, M.G., Wong, T.H., Tator, C.H. and Tymianski, M., The effect of a direct current field on axons after experimental spinal cord injury, Can. J. Surg., 32 (1989) 188-191. Freed, W.J., de Medinaceli, L. and Wyatt, R.J., Promoting functional plasticity in the damaged nervous system, Science, 222 (1985) 1522-1544. Hall, R. and Lindholm, E., Organization of the motor and somatosensory cortex of the albino rat, Brain Res., 66 (1974) 2338. Hurlbert, R.J., Theriault, E. and Tator, C.H., Neurotoxicity from direct current stimulation, Soc. Neurosci. Abstr., 16(2) (1990) 1115. Jaffe, L.E and Poo, M.M., Neurites grow faster towards the cathode than the anode in a steady field, J. Exp. Zool., 209 (1979) 115-128. Kaplan, ES., Maguire, T.G., Lee, U.M., Seizer, M.E., Spindler, K., Black, J. and Brighton, C.T., Direct extracellular electrical stimulation influences density dependent aggregation of fetal rat cerebrocortical neurons in vitro, Neurosci. Lett., 94 (1988) 33-38. Kerns, J.M., Pakovic, I.M., Fakhouri, A.J., Wickersham, K.L. and Freeman, J.A., An experimental implant for applying an electrical field to peripheral nerve, J. Neurosci. Methods, 19 (1987) 217-223. Kiernan, J.A., Hypotheses concerned with axonal regeneration in the mammalian nervous system, Biol. Rev., 54 (1979) 15197. Krathamer, V., Modulation of conduction at points of axonal
42
24
25
26
27
28
29 30 31
32
33
34
35
bifurcation by applied electric fields, 1EEE Trans Biomed. Engin., 37 (1990) 515-519. Marsh, G. and Beams, H.W., In vitro control of growing chick nerve fibers by applied electric currents, J. Cell Comp. Physiol., 27 (1946) 139-157. McCaig, C.D., Spinal neurite reabsorption and regrowth in vitro depend on the polarity of an applied electric field, Development, 100 (1987) 31-41. McCaig, C.D., Nerve growth in a small applied electric field and the effects of pharmacological agents on rate and orientation, J. Cell Sci., 95 (1990) 617-622. McDevitt, L., Fortner, P. and Pomeranz, B., Application of a weak electric field to the hindpaw enhances sciatic motor nerve regeneration in the adult rat, Brain Res., 416 (1987) 308-314. Midha, R., Fehlings, M.G., Tator, C.H., Saint-Cyr, J.A. and Guha, A., Assessment of spinal cord injury by counting corticospinal and rubrospinal neurons, Brain Res., 410 (1987) 299308. Murray, H.M. and Gurule, M.E., Origin of the rubrospinal tract of the rat, Neurosci. Lett., 14 (1979) 19-23. Patel, N.B. and Poo, M.M., Orientation of neurite outgrowth by extracellular electric fields, J. Neurosci., 2 (1982) 19-23. Politis, M.J. and Zanakis, M.E, Treatment of the damaged rat hippocampus with a locally applied electric field, Exp. Brain Res., 71 (1988) 223-226. Politis, M.J. Zanakis, M.E and Albala, B.J., Facilitated regeneration in the rat peripheral nervous system using applied electric fields, J. Trauma, 28 (1988) 1375-1381. Politis, M.J., Zanakis, M.E and Albela, B.J., Mammalian optic nerve regeneration following the application of electric fields, J. Trauma, 28 (1988) 1548-1552. Politis, M.J. and Zanakis, M.F., The short term effects of delayed application of electric fields in the damaged rodent spinal cord, Neurosurgery, 25 (1989) 71-75. Rivlin, A.S. and Tator, C.H., Objective clinical assessment of motor function after experimental spinal cord injury in the rat, J. Neurosurg., 47 (1977) 577-581.
36 Roederer, E., Goldberg, N.H. and Cohen, M.J., Modification of retrograde degeneration in transected spinal axons of lamprey by applied DC current, J. Neurosci., 3 (1983) 153-160. 37 Roman, G.C., Strahlendorf, H.K., Coates, P.W. and Rowley, B.A., Stimulation of sciatic nerve regeneration in the adult rat by low intensity electric current, Exp. Neurol., 98 (1987) 222232. 38 SAS Institute, Inc., SAS User's Guide Statistics, Version 5 edn., SAS Institute, Inc, Cary, NC, 1985. 39 Schlaepfer, W.W., Calcium induced degeneration of axoplasm in isolated segments of rat peripheral nerve, Brain Res., 69 (1974) 203-215. 40 Schlaepfer, W.W. and Haster, M.B., Characterization of the calcium-induced disruption of neurofilaments in rat peripheral nerve, Brain Res., 168 (1979) 299-309. 41 Strautman, A.E, Cork, R.J. and Robinson, K.R., The distribution of free calcium in transected spinal axons and its modulation by applied electrical fields, J. Neurosci., 10 (1990) 35643575. 42 Snedecore, G.W. and Cochrane, W.G., Statistical Methods, 7th edn. Iowa State University Press, Ames, IA, 1980, pp. 287-288. 43 Tracey, D.J., Ascending and descending pathways in the rat spinal cord. In G. Paxinos (Ed.), The Rat Spinal Cord, Vol. 2, Hindbrain and Spinal Cord, Academic Press, Sydney, 1985, pp. 311-324. 44 Waldron, H.A. and Gwyn, D.G., Descending nerve tracts in the spinal cord of the rat. 1. Fibers from the midbrain, J. Comp. Neurol., 137 (1969) 143-154. 45 Wallace, M.C., Tator, C.H. and Piper, I.R., Recovery of spinal cord function induced by direct current stimulation of the injured rat spinal cord, Neurosurgery, 20 (1987) 878-884. 46 Wang, S.D., Goldberger, M.E. and Murray, M., Plasticity of spinal systems after unilateral dorsal rhizotomy in the adult rat, J. Comp. Neurol., 304 (1991) 555-568. 47 Zanakis, M.E, Differential effects of various electrical parameters on peripheral and central nerve regeneration, Acupunct. Electrother. Res., 15 (1990) 185-191.
