National Academies Press: OpenBook

Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries (1957)

Chapter: Electrical Evidence of Regeneration

« Previous: Recovery of Motor Function
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 203
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 204
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 205
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 206
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 207
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 208
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 209
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 210
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 211
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 212
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 213
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 214
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 215
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 216
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 217
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 218
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 219
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 220
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 221
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 222
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 223
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 224
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 225
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 226
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 227
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 228
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 229
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 230
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 231
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 232
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 233
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 234
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 235
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 236
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 237
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 238
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 239
Suggested Citation:"Electrical Evidence of Regeneration." National Research Council. 1957. Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries. Washington, DC: The National Academies Press. doi: 10.17226/18485.
×
Page 240

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Chapter IV ELECTRICAL EVIDENCE OF REGENERATION Harry Grundfest, Y. T. Oester, and Gilbert W. Beebe A. INTRODUCTION Numerous electrodiagnostic tests have proved useful in the study of pe- ripheral nerves, both experimentally and clinically. Their clinical value is especially great preoperatively in evaluating the chance of spontaneous recovery, at operation when the surgeon is confronted with the necessity for deciding whether to lyse and leave alone or resect and suture, and later when there may be doubt concerning the progress of regeneration following suture. Inclusion of electrodiagnostic tests in the standard list of follow-up observations was prompted by no strong expectation that they would prove as useful in the evaluation of late results as they are in making early assess- ments, but rather by the belief that the opportunity should not be missed for comparing their results with those obtained in the study of voluntary movement. Also, there was some hope that electrical studies might throw additional light upon cases classified as failures in motor recovery on the basis of examination of voluntary movement. In retrospect it would also appear that the addition of the electrodiagnostic battery to the standard workup most probably served to improve the quality of the motor studies. Discrepancies between voluntary movement and electrical responses prompted more careful study of individual muscles, and at times alerted examiners to disparities between anatomical reinnervation and functional recovery which could be turned to patients' advantage. Since the follow-up studies were made about 5 years after injury, neither early nor interim observations could be procured under the design adopted for the study. However, in chapter XII, where electrodiagnostic tests are reviewed from the standpoint of their contribution to surgical diagnosis, early and longitudinal observations are given on the basis of a subsequent study of casualties from the Korean episode, made by Nulsen at Valley Forge General Hospital. In accordance with the original plan for the analysis of the follow-up data, a complete, separate study was made of the electrical observations along the lines already described in chapter III for voluntary motor recovery. 203

The standard muscles specified for routine electrodiagnostic tests are those listed in table 48, plus the following: Paroiual and sciatie-peroneal Extensor digitorum brevis Peroneus brevis Tibial and sciatic-tibial Flexor digitorum brevis The content of the present chapter is of more limited scope, however, and for these reasons: (1) The selection of muscles for certain of the electro- diagnostic tests depends so much upon the results of voluntary stimulation that no fair comparison of any two sets of cases could be made without prior assurance of the comparability of their selection, and in practice this means that the analysis of responses to these tests must rest upon the studies of voluntary movement; (2) to the extent that electrical and voluntary data correlate, a separate presentation of motor recovery seems superfluous; (3) the electrical data are so much less extensive than those on voluntary movement that explorations based on the latter are considerably more powerful; and (4) the electrical data are much more subject to center variation than the observations on voluntary movement. Furthermore, the chief clinical value of the electrodiagnostic tests at the point in time at which the follow-up study was done lies in the contribution they make to an understanding of the factors associated with absence of voluntary movement. Accordingly, the electrodiagnostic data are studied here, not as independent evidence on the presence or absence of nerve regeneration, but as auxiliary information of especial value in illuminating some of the factors which may be responsible for the absence of voluntary movement. B. METHODOLOGY Essentially four electrodiagnostic tests comprise the standard battery: 1. Stimulation of the nerve. 2. Chronaxiemetry. 3. Galvanic tetanus ratio. 4. Electromyography. In the coding of data for the statistical analysis provision had also been made for recording the EMG response to stimulation of the nerve, but this observation was made on only one percent of the individual muscles in- volved in the study and merits no further consideration. Finally, although not an electrodiagnostic test in the same sense as the four listed above, direct muscle stimulation was performed as a means of assessing muscular atrophy 1. Stimulation of Nerve Monopolar or bipolar electrodes are applied to the skin overlying a motor nerve or are inserted subcutaneously to lie close to or enter the sheaths of the nerve. Electrical stimuli of controllable, graded intensity are applied through the electrodes, either as brief, single shocks or as a train of repetitive 204

excitation. In the different follow-up centers they were variously derived from the capacity-type chronaxiemeter, single or multiple shock square pulse generators, or from an alternating current source (including faradic stimulation). The type of stimulus employed is relatively unimportant provided the strength can be controlled so that an adequate and safe stimulus is delivered to the nerve. The test ascertains whether the nerve is in functional continuity with a muscle capable of contraction. The sources of discontinuity may be: a. Absence of anatomical regeneration in the nerve; b. Absence of reinnervation of the muscle; or c. Absence of muscle fibers. The first two conditions could be detected by observing presence or absence of nerve impulses, but this was not done in the present study. Chronaxi- emetry, which was extensively carried out, provides the same information in a technically more convenient manner. In the course of the latter test also direct stimulation of the muscle provides information regarding the presence or absence of muscle as well as the nature of the response of the muscle fibers. Stimulation of the nerve rapidly delimits the functionally intact from the nonfunctional neuromuscular groups. Together with observations on voluntary movement it also provides a clear-cut differentiation between those neuromuscular groups which are not used by the patient because of anatomical defects and those which are not used because of psychic or learning factors. The latter categories of nonuse should not be minimized. In the course of this study it has been found that even experienced observers sometimes attribute to failure of anatomic regeneration a lack of movement apparently caused only by some psychic factor or by a failure in relearning the use of a muscle or even a digit. Of particular value in early diagnosis is the stimulation of the nerve at two levels, one above and one below the site of injury. In some instances following gunshot injury, when the nerve is in gross continuity, local propa- gation-block may cause a temporary paralysis which simulates anatomical loss of continuity. This condition may, however, be temporary, lasting up to 3 to 4 months. When temporary, it does not result in neural degenera- tion and the muscles involved remain excitable to neural stimulation applied below, but not above, the site of block. When block without degeneration is present some 2 weeks after injury it would appear to be preferable to delay surgical intervention until some opportunity for recovery has been provided, since recovery, if it occurs, then takes place without major disruption in the organization of the neuromuscular or the sensory complex, whereas after neurosurgery this occurs inevitably to some extent (5, 44, 61, 66). In all, observations were made on 6,264 individual muscles, about 40 percent of all those involved in the present study. There were, in addition, 185 individual muscles in which the observation could not be made because an adequate stimulus could not be tolerated. The extent to which direct 205

Table 114.—Percentage of Affected Muscles Examined by Nerve Stimulation, Complete Sutures on All Seven Major Nerves, for Muscles With Studies of Voluntary Movement, by Center Center None Voluntary movement Any Total Total Nerve stimu- lation done ' Total Nerve stimu- lation done ' Total Nerve stimu- lation done ' Num- ber Per- cent Num- ber Per- cent Num- ber Per- cent Boston 224 259 583 728 116 92 1 539 403 9 41.1 0.4 92.5 55.4 5.4 629 747 1,733 1,897 897 204 3 32.4 853 1,006 2,316 2*625 1,063 296 4 2,220 1,086 34.7 0.4 95.9 41.4 3.2 New York . ... 1,681 683 25 0.4 97.0 36.0 2.8 34 Total . . 1,960 1,044 53.3 5*903 2,596 44.0 7,863 3,640 46.3 1 Exclusive of those few in which nerve stimulation was attempted but the current could not be tolerated. nerve stimulation was performed in each center, and the presence of bias in the selection of cases for such stimulation, are shown in table 114, which is confined to 7,863 muscles supplied by the 7 major nerves, complete nerve sutures, standard muscles affected by injury, cases with movement not affected by tendon transplant, by loss of muscle substance by direct injury, or by sacrifice of a nerve branch, and muscles in which an examination of voluntary movement was made. Only in the New York center was direct nerve stimulation routine and the selection of muscles for testing quite unbiased, and in only three centers was nerve stimulation done on any appreciable number of cases. In the aggregate 46 percent of the muscles included in table 114 were examined by direct nerve stimulation, but somewhat more often (53 percent) in cases with no voluntary movement than in cases with voluntary movement (44 percent). The muscular activity induced by stimulation of the nerve was scaled as follows, for each muscle: No contraction. Contraction Barely visible. Similar to voluntary contraction. Stronger than voluntary contraction. Of the entire set of 6,264 muscles observed following stimulation of the nerve, contractions were observed in 71.4 percent. By type of contraction 206

this total breaks down as follows: 12.9 percent barely visible, 53.4 similar to voluntary contraction, and 5.1 stronger than voluntary contraction. All statistical analysis has been based on the percentages with at least visible contraction and with no contraction. 2. Chronaxiemetry Principles The theoretical basis of chronaxiemetry can be simply stated, although many of the details remain unclear. With the exception of certain synapti- cally activated systems, every excitable tissue (e. g., nerve, muscle, sense organ) can respond to an electrical stimulus of adequate strength. Stimu- lus thresholds, however, are a function of duration (figure 17). A liminal stimulus of very brief duration will be of higher intensity than one of long duration. However, all excitable tissues possess the property that a stimu- lus of relatively long ("infinite") duration cannot be effective when reduced Figure 17. Change in chronaxie with the size of the stimulating electrode1 I 200 100 § 1000 1 - 200 100 CAPILLARY CATHODE O-021/.F-O-OS.'i AGCI-WIRE CATHODE NERVE JjJJSCLE NERVE MUSCLE I I I I I 0.001 0.002 0.005 0.01 0.02 0.05 0.1 0.2 0.5 O.I 'The single nerve-muscle fiber preparation of the frog retrolingual membrane; monopolar stimulation of either the nerve or the muscle fiber. The abscissa plots the logarithm of the various capacities used in the stimulating circuit. 1.0 jjF is equivalent to 4.0 msec. (or in the older form a). The ordinates plot (in relative units) the logarithm of the minimal voltage to which a given capacitor was charged to excite the tissue. When the punctate stimulating cathode was a fine saline-filled capillary* the upper Itrength duration curves were obtained yielding chronaxies of 0.04 msec. for stimulation of the nerve fiber* and 0.12 msec. for one of the muscle fibers innervated by that nerve fiber. The lower curves were obtained when the same pair of units were stimulated with a large chlorided silver wire placed close to the nerve or muscle fiber. The chron- axie of the latter, 2.1 msec. now had increased almost 20-fold, while the chronaxie of the nerve fiber (0.08 msec.), had only doubled. (From Grundfest (28), reprinted by permission from the Journal of Physiology.) 207

below an intensity which is specific for the type and condition of the tissue and for the nature of the stimulation. The minimal threshold for a stimulus of infinite duration defines the "rheobase" (38). It provides an easily de- termined quantity, the intensity of the stimulus when the duration of the latter need not be accurately determined, provided it is sufficiently long. Given the rheobase, which furnishes an intensity unit of the stimulus, a single point on the curve of the strength-duration relationship can then be determined which serves as a useful characterization of that curve. That point is the "chronaxie," and it may be defined as the minimal duration of a stimulus with an intensity twice the rheobase. The chronaxie is charac- teristic of the excitable tissue. Numerous studies have been published (6) on the chronaxie for different neuromuscular groups in normal and diseased conditions. In man the chronaxie of such normal units usually falls in the range below 1 msec. (millisecond) and most frequently below 0.5 msec. When neuromuscular transmission is eliminated as in nerve degeneration (also in neuromuscular block by curare, etc.) the chronaxies obtained may be in the range of scores of msec. The explanation of this difference has generated considerable dispute. Lapicque (38) believed that the anatomi- cal or pharmacological changes produced an alteration in the strength- duration relation which was reflected in the increase of the chronaxie. Others (28, 65) demonstrated, however, that the normal strength-duration curve of innervated muscle fibers is much more dependent on the conditions of stimulation than is that of the nerve fibers, a major factor being the area of the tissue which is being effectively stimulated. In motor nerve fibers this effective area of stimulation is limited by their myelinated, noded struc- ture, whereas the muscle fiber can be electrically stimulated everywhere on its surface. Therefore, under the conditions of percutaneous stimulation, where the effective electrode is a large surface, the chronaxie of direct elec- trical stimulation of the muscle is large, while that of the nerve is small. Normally, percutaneous stimulation applied to a motor point excites prima- rily nerve fibers, yielding a low chronaxie. When the muscle is denervated, however, only direct stimulation of its fibers is possible, and under the con- ditions of stimulation the chronaxie obtained is high. As reinnervation proceeds, the chronaxie decreases from the large values of the denervated muscle through various intermediate values, and tends to return toward the normal, low value when reinnervation of the whole muscle has taken place. No entirely satisfactory explanation of this phe- nomenon is available. At least one factor resides in the nature of the ob- servations. The measurements are carried out by suitable stimulation of a motor point, with the observer striving for a minimal contraction of the muscle under study. In visual observation the response is probably com- posed of neurally evoked activity of a number of motor units as well as of directly stimulated nonreinnervated muscle fibers. It is, therefore, likely that the magnitude of the chronaxie will depend on the relative proportions of th e differently excited elements and will decrease progressively as that proportion shifts more and more in the direction of completely neural acti- 201

vation. However, this may not be the complete explanation. Denervated fibers may have themselves undergone prolongation of chronaxie as their excitable properties were altered. If the return to normal excitability were gradual after reinnervation the observed chronaxie would likewise decrease gradually. Usefulness Whatever the theoretical explanations of the effects might be, it is evident from this study as well as from earlier work that chronaxie is closely corre- lated with regeneration of the neuromuscular complex. The progressive decline of the chronaxie in the course of successive examinations, therefore, gives an early indication of progress in reinnervation, and is accordingly a useful index. Indeed, it is also possible sometimes to detect the approximate degree of reinnervation of a given muscle. For example, the regrowth of only a small bundle in a motor nerve may be detected by the appearance of a localized region of the muscle possessing a lowered chronaxie while the rest of the muscle, which remains denervated, exhibits a high chronaxie. In the hands of an experienced observer the differences in these values, and also in the appearance of rapid twitches of the reinnervated muscle fibers and slower contraction of still denervated muscle fibers, furnish valuable information. At a single examination, such as characterized this study, chronaxiemetry has a more restricted usefulness, since it can then only reflect the current state of reinnervation or its absence. However, it provides in this connection some critically useful information: a. The character and degree of the response of directly excited muscle fibers provide a rough estimate of the state and amount of activable muscle tissues. b. The magnitude of the chronaxie gives an estimate of the degree of reinnervation. It is a more quantitative test of the intactness of the neuromuscular organization than is neural stimulation, but is more time-consuming and requires a more highly trained operator. c. The finding of a low chronaxie in the face of functional paralysis may indicate the nature of this loss of function. As with nerve stimulation, but in a more quanti- tative manner, it may indicate presence of a propagation block in the nerve, of psychogenic paralysis* of loss of the learned processes of muscle movement (by virtue of disruption of these patterns by long absence of use* or by virtue of anatomical alterations in the pathways), or of malingering. Methodology The chronaxiemeters used in this study fall into two classes, which do not differ basically in the data they can furnish. The electrical pulses generated may be rectangular as produced by a variety of electronic means. The amplitude may be controlled to provide a source of variable voltage or, as in the Golseth-Fizzell instrument (26) used by the Boston and Chicago centers, the stimulator provides a source of controllable current which is maintained at the desired level of output despite variation in the resistance offered by the electrode or by the tissues of the patient. In theory it may be preferable to use the latter "constant current" type of stimulation, but under actual conditions the advantage is doubtful. In 209

clinical measurements the stimulus is applied percutaneously, and the pathway for the current includes a variety of tissues, in addition to the nerve and muscle. Each of these tissues is polarizable and probably to a different extent. The instantaneous stimulating current through the excitable tissue will therefore be complexly determined by the polarization of the different components. At any rate, it would appear that measure- ments with "constant current" and "constant voltage" pulse generators are essentially equivalent. Short rectangular pulses (i. e., those below 0.5 msec. in duration) are easily distorted by the electrical properties of the stimulating electrodes and by the tissues of the patients. Furthermore, reproducibility of dura- tions to the second decimal place (i. e., 0.01 msec. in the lowest range of durations) is poor with most electronic circuits. Some centers, therefore, have used stimulators which deliver discharges of condensers charged to a controlled voltage. The theory of the capacitative chronaxiemeter is discussed fully by Lapicque (38). His measurements and many others have demonstrated that the magnitude of the capacity is directly converti- ble into durations for any given stimulating circuit. The capacitor chronaxiemeters used in this study were capable of delivering pulses varying accurately by increments of 0.01 msec. The techniques of chronaxiemetry are described in various textbooks (6, 38) and will not be dealt with here. It is, however, necessary to stress that the apparently simple procedure hides a number of pitfalls. Obser- vation by some of us of the measurements made during and after World War II by technicians in military hospitals indicates that much more than routine training is required. The measurements depend to some extent on the accurate placing of electrodes, on correct treatment of these, on the rate at which observations are repeated, and on the experience of the operator in observing and evaluating the response. Even in the hands of the usually experienced examiners of the different follow-up centers results varied too greatly to permit the data of all five centers to be pooled directly. Table 115 illustrates this variation on the basis of data on the abductor digiti quinti following complete suture of the ulnar nerve. There were in all 5,581 chronaxie determinations on individual muscles, or about 36 percent of all the muscles involved in the study. If attention is confined to the 7 major nerves, complete sutures, standard muscles affected by injury, muscles with movement not influenced by sacrifice of a nerve branch, by direct loss of substance through injury, or by tendon transplant, and muscles in which studies of voluntary movement were made, it appears that 44 percent were tested by chronaxiemetry. However, as may be seen in table 116, the centers varied widely in their resort to chronaxiemetry, and in every instance their selection of muscles was a biased one. In the aggregate 25 percent of the muscles incapable of voluntary movement were tested by this means in comparison with 51 percent of the muscles capable of voluntary contraction. 210

Table 115.—Center Variation in Chronaxie Determinations on Abductor Digiti Quinti Following Ulnar Nerve Suture New York and San Francisco Boston and Chicago Philadel- phia Chronaxie, in msec. Percent 60.5 Percent 1.0 Percent 17.9 00 01-02 27.2 54.7 39.3 03-07 11.1 22.1 31.0 08-12 1.2 13.7 9.7 0 8.4 2. 1 Total . . . 100 0 99 9 100 0 Number of muscles 162 95 145 Table 116.—Percentage of Affected Muscles Studied by Chronaxiemetry, Complete Sutures on All Seven Major Nerves, Muscles Studied for Voluntary Movement, by Center Voluntary movement Center None Any Total Total Chronaxie- metry done Total Chronaxie- metry done Total Chronaxie- metry done Num- ber Per- cent Num- ber Per- cent Num- ber Per- cent Boston 224 259 583 728 166 25 47 228 166 26 11.2 629 747 1,733 218 170 1,349 733 520 34.7 22.8 77.8 38.6 58.0 853 243 217 1*577 899 546 28.5 21.6 68.1 34.2 51.4 Chicago 18.1 39.1 22.8 15.7 1,006 2*316 2,625 1,063 New York Philadelphia 1*897 897 Total 1,960 492 25.1 5,903 2*990 50.7 7,863 3,482 44.3 3. Galvanic Tetanus Ratio The tetanus ratio (TR) is based upon two quantities which may be ob- tained in the course of stimulating muscles directly by means of galvanic current: (a) the current required to produce a sustained contraction, or tetanus; and (b) the rheobase or current required to produce a minimal contraction. In human skeletal muscles with normal innervation the tetanus ratio is usually about 3.5 to 4.0, and the normal range is generally

taken as 3.0 to 5.0. During a period of complete denervation the ratio approaches and may actually reach unity. During a period of reinnerva- tion the ratio rises sharply, generally markedly above normal values, and as reinnervation progresses it slowly falls to within the normal range. Determinations are made by applying conventional electrodes to the skin overlying the muscle to be tested. A constant-current generator, with an output in the range of milliamperes, and capable of maintaining a fixed output in the face of changes in the resistance of local tissues, is used to stim- ulate a minimal twitch. The current is then increased until the muscle responds with a sustained contraction (tetanus) during the entire time the current is applied, usually 4 seconds. The current required to produce tetanus is then divided by the current required to produce the least muscle twitch, and the resulting ratio is the tetanus ratio. Although the tetanus ratio was included in the standard battery of tests agreed to by all investigators, determinations were not often made and for a variety of reasons. In all, determinations were made on 1,138 muscles, regardless of type of injury, extent of surgery, or sampling area. On muscles affected by complete suture, for lesions in the representative sample, 11 muscles were tested often enough to permit some exploration of bias asso- ciated with the selection of muscles for testing. Contraction on voluntary stimulation was employed in this investigation, which showed that the TR was more often sought in the presence of voluntary contraction than in the presence of none, so that the set of muscles with TR values is not itself an unbiased set. For example, 18.3 percent of upper extremity muscles ob- served to contract voluntarily have TR readings in contrast to 12.6 percent of those with no voluntary contraction; this discrepancy has a probability of .03 in the statistical test employed here. For the lower extremity the dis- crepancy is smaller and by itself well within the range of chance, but both discrepancies, considered jointly, have a probability of about .025. Where- as in the particular representative sample of upper extremity muscles used here 11 percent failed to contract voluntarily, in the set with TR readings the figure is 7.7 percent. The bias is not, therefore, large enough to be very troublesome. Of greater interest is the relation between voluntary contraction and TR. Although the present series is not large enough to exhibit this relationship hi any detail by nerve, it does include an adequate number of readings on the abductor digiti V and, moreover, as may be seen from figure 18, the same average relationship appears to characterize this muscle and all others taken together. An adjustment has been made for the selection of muscles for TR determinations, so that the curves of figure 18 are free from this defect. In the region of TR readings below 3.0 the percentage with volun- tary contraction rises rapidly from about 50 to 90 percent, and continues to rise thereafter but more slowly. In the region of TR about 6.0 or more voluntary contraction is very nearly complete. Although the TR readings were studied in relation to a number of char- acteristics of injury and treatment, for the reasons already mentioned the 212

213

results will not be given here. However, it may be useful to provide some of the TR data in descriptive form. Table 117 provides average values for muscles with reasonable numbers of readings, and table 118 provides a fre- quency distribution for the abductor digiti V which was the most extensively studied muscle. Table 117.—Mean Tetanus Ratio Readings Following Complete Suture, for Selected Muscles, Lesions in Representative Sample Number tested Mean tetanus ratio Nerve Muscle Median Opponcns 45 3.8 Ulnar Fl. car. uln 43 4.0 Fl dig prof 4 & 5 30 4 0 Abd dig V . 137 3.7 Add. poll 34 3.5 1st dors. inteross 69 3.5 Radial Ext. car. rad Ext. dig 22 29 3.4 3.1 Tib ant . . . 23 3 3 Tibial Inteross 24 3.2 Sciatic-peroneal Tib. ant 26 2.8 Table 118.—Distribution of TR Readings on Abductor Digiti V Following Complete Suture TR Number Percent TR Number Percent 8-1 2 .... 4 2.9 5 8-6 2 2 1 5 1.3-1.7 5 3.6 6 3-6.7 1 0.7 1.8-2.2 8 5.8 6.8-7.2 2.3-2.7 11 8.0 7.3-7.7 2 1.5 2 8-3 2 . . 34 24 8 7.8-8 2 2 1 5 3 3-3 7 .... 16 11.7 8.3-8.7 . 1 0 7 3 8-4 2 .... 27 19.7 8.8-9.2 1 0.7 4 1-4 7 10 7 3 4.8-5.2 9 6.6 Total 137 99.9 5.3-5.7 4 2.9 4. Electromyography Principles As part of their activity the excitable tissues generate electrical responses. These are characteristic of the different type of tissue, but in nerve and muscle the general property is a transient, brief change of potential which 214

at the source (in the excitable membrane) has an amplitude of approxi- mately 100 mv. (millivolts), lasting from about 0.5 msec. (in some nerve fibers) to several msec. (in some muscle fibers) and up to fractions of a second (as in heart muscle). The full value of this potential is measurable only under special conditions (e. g., with a microelectrode inserted into the cell and a suitable recording system). However, a fraction of the potential can be obtained by recording from electrodes closely applied to the responding units. This fraction may vary from some microvolts (single nerve or muscle fiber active in situ) up to nearly the full value of the impulse (isolated giant axons in oil). The ready availability of electronic amplifiers has made it possible to study these impulses, and electromyography is now widely practiced. In many cases the amplified potentials are recorded by direct writers which are commonly used for electroencephalography. These recording devices, however, respond very poorly to the brief impulses of muscle fibers. As a gross index of presence or absence of massed, prolonged activity the records obtained with the direct writers are very useful, but they cannot provide information regarding fine detail or occasional isolated responses. All the electromyographic studies of the various follow-up centers were, therefore, recorded on more suitable equipment employing the cathode ray oscillo- graph and photography of its traces. The equipment of the different centers varied, some having units commercially available, others being individually designed and constructed according to various specifications. The multiplicity of these designs makes it impossible to describe them here. (For such descriptions see Grundfest (29), Dickinson (19), Whitfield (86).) The electromyographs used by the various centers meet research require- ments as to sensitivity and faithfulness of recording. Methodology During voluntary activity or on electrical stimulation of a normal inner- vated muscle, the neural impulses arriving at the muscle set up end-plate activity and a propagated electrical response in the activated muscle fibers. A single motor nerve fiber, by virtue of its terminal branching, innervates a number of muscle fibers which generally respond together as the motor unit. However, both the length of the neural path in the motor unit and the calibers of the different terminal branches may differ to some extent. Therefore, an impulse coursing in a nerve fiber exerts its terminal effect, the activation of the different muscle fiber components of the motor unit, with some asynchrony. A recording electrode in the vicinity of the motor unit, therefore, is subjected to a composite of potential changes which collectively lasts longer than would the electrical response of a single muscle fiber. In the case of a single electrical stimulation of the nerve the motor unit usually responds only once. However, in voluntary activity the spinal integrative activity usually causes repetitive discharge of the motoneuron, and similar repetitive responses in the motor unit. The rate of the dis- 215

charge will depend upon the amplitude of the voluntary effort and its duration upon the persistence of the effort. When a single motor unit is in the recording field, as may be achieved by proper choice of recording conditions, by anatomical isolation (as after destruction of most motor units), or by functional isolation (as with extremely weak voluntary effort or electrical stimulation), the response recorded from the motor unit has a rather constant amplitude and form, the deviation being ascribable to baseline instability in the recording equipment, or to shifts of the electrode (both classified as artifacts), or to variation in the number of muscle fibers of the motor unit responding to the neural impulse, and in their relative asynchrony. The potential reco-ded by the electrodes is a reflection of the current flowing in the medium surrounding the generators of the excitable tissues. The intrinsic response of the muscle fiber is a relatively simple pulse (external negativity) moving away from the site of initiation (at the end- plate following neural excitation) at a velocity characteristic of the muscle fiber. The generators of the impulse (the muscle fibers) in this case con- stitute a shifting site in a mass of tissue (the rest of the muscle, fluid, etc.). An impulse generated and propagated in this type of volume conductor is therefore recorded by the electrodes not as a unidirectional pulse but as a complex depending upon the relative change in current flow at the elec- trodes. The recording situation in a volume conduction is described briefly by Brazier (8) in Fulton's Textbook of Physiology (25), and in much greater detail by Lorente de N6 (45). In general when only one electrode is close to the site of activity a single traveling unidirectional pulse is then altered into a triphasic response indicative of the approach, arrival, and departure of the impulse. However, the electrical responses of muscle and of nerve are not so simple (30) but are composed of the spike and of after- potentials which may be both negative and positive in sign. These com- plications will affect the form of the response recorded from the volume (8, 25, 45). Furthermore, when several motor units are active, but not synchronously, the algebraic summation of the different phases of their responses will complicate the record still more. In addition, when the muscle fibers do not all lie in parallel within the volume (as is true of many muscles) the impulses coursing in them will travel in divergent paths. The relative contributions of one motor unit and another will therefore vary. For these and other reasons the amplitude of the recorded response may have no strict relation to the number of active motor units, and the form of the response may become complicated because of factors which have no relation to the function of the motor units. Another class of complications is introduced by the electrical require- ment that a potential can be recorded only between two electrodes and not by a single electrode. Thus, when both electrodes are close to the source of the activity, the recorded potential must be the momentary difference of potential between them. In a volume conductor having a moving generator the difference potential can become very complicated 216

indeed, and the potential (being the difference) may be very small, re- quiring high amplification and consequent instability of base line due to amplifier "noise." Some of the difficulties of recording with bipolar electrodes can be avoided by using monopolar recording. In this procedure only one electrode (the "active" one) is inserted into the region of activity. The second ("reference") is fixed at some relatively inactive region (skin, tendon, etc.). The potential so recorded is usually larger, and may sometimes appear to have a simple form. However, with monopolar recording it is essential that the reference electrode be in a truly "silent" region. This is not always possible to obtain. Furthermore, activity of elements anywhere in the volume between the active and reference elec- trode will contribute to the total potential. Therefore, it is frequently more difficult with monopolar leads to record activity confined to single units and sometimes monopolar recording may include activity of elements entirely unrelated to those under study. Another type of recording employs coaxial electrodes, taking advantage of the availability of small and sharp hypodermic needles. An insulated wire is inserted into the lumen of this and is fixed there. The outside of the needle shaft is carefully insulated, leaving only the tip as a conductor. The internal wire and the concentric needle form the two electrodes leading from the interior of the muscle to the recording amplifier. Since both electrodes are very close together the difference in potential produced by activity in their vicinity becomes rather small. The amplifiers em- ployed with such electrodes must be operated at relatively high gain and designed for low intrinsic noise. The advantage of coaxial electrodes, however, is that the difference in potential between the pair of electrodes is large only when they are very close to an active site. Therefore, the restriction of recording to the activity of one or a few motor units becomes easier, and to that extent also the form of the recorded potential becomes more significant. Localization of the recording of activity, and avoidance of pickup of extraneous potentials, whether these be generated in the body of the patient or externally (as a. c. pickup, etc.) can be furthered by the use of differential amplifiers (29). In these neither of the two electrodes (active and reference) is grounded, and an additional ground electrode is provided. The advantages of differential recording are discussed in various textbooks of electronics and by Grundfest (29) and Dickinson (19). In the New York center, the amplifier of the electromyograph was highly differential (1:30,000 or better). The description of the recorded electromyograms given above refers primarily to the response of one or more motor units. In the denervated muscle, the responses obtained can only be due to spontaneous or induced activity of individual muscle fibers. As described above, the response of an individual fiber in a volume conductor will be very small, and may also be briefer than that of the motor unit. For reasons not known, during 403930—67 16 217

certain stages of their life-cycle denervated muscle fibers may exhibit spontaneous activity, or fibrillation, which is therefore seen in the records as spontaneous, random potentials of low amplitude and rather short duratio i. Fibrillation may also be initiated by the act of insertion of the electrodes; or by various mechanical or chemical stimuli. Fibrillation ceases after reinnervation and its occurrence is therefore indicative of denervation, but its absence is no index of reinnervation. The indication of reinnervation must rather be sought in the presence of the larger, longer responses of the motor unit, elicited either by effort of the patient or on stimulation of the nerve. During the process of reinnervation the response of a motor unit may undergo complex changes but these are beyond the scope of the present analysis. An important finding of this study has been that reinnervated muscles usually show extremely large electrical responses from a few motor units (90). Similar though not as large responses are obtainable from muscles affected by various diseases (poliomyelitis, etc.), but are not found in normal muscles. In the recording volume of the muscle are also present the impulses of afferent and efferent nerve fibers and the potentials produced at the end- plates. Usually these are small compared with the responses of the muscle fibers, but might play a role in complicating the form of the recorded potentials. Clinical usefulness of the electromyogram The present state of knowledge of the electrophysiology of muscle limits to some extent the clinical usefulness of the electromyogram. Thus, as a clinical index the significance of fibrillation is confined to positive identi- fication of the existence of denervated muscle fibers. However, as stated earlier, absence of fibrillation does not indicate the absence of denervated fibers. A second limitation derives from the nature of the response. In a volume conductor this is a complex summation of the activity of many elements. When the experimental conditions are such as to present one or a few easily identifiable units, the result is informative only with respect to these few units—a sampling. Reconstruction of the events in the total population would require a large number of samples, preferably made simultaneously with many separate recording channels. In practice, however, recording is feasible with only a few channels. On the other hand when the recording conditions are such as to obtain the summated electrical activity of the entire muscle, the process of algebraic summation may complicate the data or even lead to erroneous conclusions, and the complexity of the^records^may defyjanalysis. Nevertheless, electromyog- raphy as a clinical test does have a sphere of usefulness. In the first place, absence of electromyographic response after neural stimulation, or on effort by the patient, particularly when sampling is adequate, is clear indication of persistent denervation. Secondly, presence of a few active motor units is indicated electromyographically with greater 218

accuracy than with other methods. Although such reinnervation may be of little utility to the patient it nevertheless can serve to indicate that conditions for neural regeneration have been favorable, and perhaps offer a clue to reasons for the absence of further recovery. Thus, some patients possess only slight motor unit activity when asked to perform with a re- innervated member. In some of these chronaxiemetry and nerve stimu- lation indicate good anatomic regeneration. The lack of voluntary activity frequently may be ascribed to loss of the organized pattern of activity in the central nervous system. Occasionally brief periods of voluntary activity during which the patient is permitted to observe for himself the appearance of the electrical responses on the oscillograph screen are sufficient to encourage in him the relearning process which is needed to develop functional activity. Examiners at the New York center are of the opinion that nonactivity because of learning loss may be frequent and that the use of the electromyogram as a reeducational tool might be useful. Unfortunately, systematic efforts in this direction were not possible. Electromyographic tests were done on 2,712 individual muscles, or about 17 percent of those studied here. The classification of responses is exhibited in table 119. Two centers (Boston and San Francisco) devoted consider- ably more effort to electromyographic tests than did the others. Even more important is the very great variation in the classification of responses by the individual centers. To some extent the variation may reflect differences in recording technique, for the electrodes used were bipolar in Boston, mono- polar in Chicago, surface and coaxial in Philadelphia, and coaxial in New York and San Francisco. Factors other than the mere choice of electrodes are undoubtedly involved in these disparities, however, some in the area of instrumentation and others in the interpretation of recordings by individual investigators. Efforts in the direction of standardization of equipment and interpretation were entirely too feeble to compete with the difficult problems in this field (11, 18). Because the number of observations was relatively small, and the center variation extreme, systematic statistical analysis on the electromyographic material has not been considered useful. 5. General Remarks on the Evaluation of the Results and Useful- ness of Electrodiagnostic Tests Electrodiagnostic tests provide information concerning principally the anatomical or lowest level of the complex neuromuscular system,1* whereas sensory and voluntary motor testing provides information on the recovery of the complex processes. The studies of World War II veterans reported here were made usually 4 to 6 years after definitive suture, so that ample time had elapsed for anatomical regeneration in most cases. Under such circumstances the usefulness of electrodiagnostic tests is rather limited 14 Aj noted immediately above, however, electromyography is able to provide some information regarding the more complex events. 219

Table 119.—Classification of Electromyographic Recordings at Each Follow-up Center Classification of muscles examined electromyographically Cei iter Bos- ton Chi- cago New York Phila- del- phia San Fran- cisco Total" Electrical silence Percent 6.1 Percent 11.7 Percent 7.1 Percent 1.8 Percent 1.7 Percent 4.8 Fibrillation potentials, spontaneous . . . 4. 5 7.6 2.7 .4 3.7 4.0 Fibrillation potentials, with insertion of 2 1 0 2.1 . 1 .6 g 3 .2 .3 Few motor unit potentials with voluntary effort 2.7 1.4 .2 .4 .8 Many motor unit potentials with volun- tary effort 18.5 20.0 6.5 31.1 1.1 10.7 55.3 37.9 5.6 54.2 9.6 26.2 Complex or other potentials 1.3 2.1 67.3 11.6 1.6 14.2 Spontaneous fibrillation potentials with motor unit potentials 4.8 12.4 7.5 .4 45.6 21.8 Spontaneous fibrillation potentials, motor unit potentials, and complex or other 1 3 1 7 6 13.0 5.8 Fibrillation potentials with insertion of needle or induced plus motor unit po- tentials . . 4.5 3.8 .2 23.6 10.8 Total 100.0 99.9 100.0 99.9 100.0 100.0 622 290 480 225 1,066 2*712 1 Includes readings on 29 "multiple lesions" not tabulated by center. whereas in the early stages of injury and therapy these tests have much greater value. Some demonstration of this may be found in the discussion of the Valley Forge material consisting of peripheral nerve injuries sus- tained in the Koreanfighting (pp. 576-588). The value of electrodiagnostic data in this series is also limited by the considerable center variation usually characteristic of these observations. C. ANALYSIS OF ELECTRODIAGNOSTIC DATA 1. Orientation and methodology Interest in electrodiagnostic tests done by the follow-up centers many years after injury lies in the expectation that they may provide information on some of the factors responsible for absence of voluntary movement. As 220

already indicated, any one of the following factors is sufficient to prevent a voluntary movement: a. Psychological factors, as seen in malingering and in hysterical paralysis. b. Failure to rclearn use of a muscle following injury. c. Failure of regeneration in the nerve itself. d. Lack of inner var ion of muscle. e. Absence of muscle fibers. If, in the absence of voluntary movement, direct nerve stimulation produces a contraction, it may be assumed that one of the first two factors has blocked voluntary movement. In some instances, of course, this assump- tion would be wrong, either because the voluntary movement is possible but was improperly elicited or observed, or because of corresponding error in reporting movement following nerve stimulation. The design of this study provides no estimates of the frequency with which such errors are made, but the existence of examining error is demonstrated by the fact that 6.3 percent of the muscles with observations on both voluntary move- ment and direct nerve stimulation were reported as moving voluntarily but not on direct nerve stimulation. However, apart from errors of observa- tion, it may be assumed that movement following electrical stimulation of the nerve, but not voluntarily, points to the operation of one of the first two factors. If voluntary movement is not possible, and direct nerve stimulation also fails to elicit movement, then some one or more of the last three factors will be considered at fault. Again the possibility of errors must be borne in mind, but precise estimates of their frequency are not obtainable from this study. In such instances the chronaxie determination will be helpful, for a low chronaxie, reflecting activity in the nerve, is indicative of regenera- tion in the nerve, and a high chronaxes is suggestive of failure. For the reasons discussed in the preceding section on methodology, the tetanus ratio and the electromyographic response are considered only briefly, and the main analysis is confined to direct nerve stimulation and chronaxie in relation to voluntary movement. Table 120 provides a summary of the New York data on all muscles examined for both voluntary movement and response to direct nerve stimulation. Although the data pertain to no single muscle, they may usefully serve as an introduction to the observa- tions and to the discussion of various ways of handling them. It may be noted, first of all, that the electrical test and voluntary movement correlate rather closely, as they should, but that more muscles (1,789) responded to electrical stimulation than to the voluntary effort at contraction (1,681). The 147 with movement on electrical but not on voluntary stimulation might be considered a measure of the extent to which voluntary contraction is impeded by poor motivation and relearning, but some or all may be a reflection of errors in the neurological examination. Moreover, the 39 responding voluntarily but not electrically are obviously defects in one test or both as noted earlier, but one hardly knows which. The measure of uncertainty is not large, for only 186 muscles or 8.4 percent of the total lie 221

in these two cells of the table, but since the chief interest lies not in the relation of the 186 to the total but rather to the 392 with no contraction in either test, some further refinement would be helpful. In this situation the chronaxie values are especially useful, and table 121 provides these data as an extension of table 120. Unfortunately, the selection of muscles for chronaxie determinations is quite a biased one. Eighty percent of the muscles observed to move voluntarily were studied as to chronaxie in contrast to 41 percent of the muscles not moving voluntarily. Therefore, it has been necessary to adjust the observations for this fact and to present the adjusted data in table 122 as the basis for all calculations. The adjust- ment consists in inflating the samples with chronaxie determinations in columns 1, 2, 4, and 5 of table 121 to correspond to the totals shown on line 1 of that table. In other words, the assumption is made that muscles on which chronaxie determinations were not made would be distributed exactly like those on which the determinations were made, provided that the response to both voluntary and direct nerve stimulation is fixed as in the column designations of the table. Table 120.—Voluntary Movement and Response to Direct Nerve Stimulation for All Affected Muscles Following Complete Sutures on All Seven Major Nerves, New York Center l Response to electrical stimulation of nerve Voluntary movement None Any Total None ... 392 39 147 1,642 539 1,681 Any. . Total 431 1,789 2,220 1 Only muscles with observations on both voluntary movement and response to direct nerve stimulation are included. The distribution of chronaxies in the 1,642 muscles whose reinnervation is demonstrated by both nerve stimulation and by voluntary response (col. 5 of table 122) is concentrated at the low end of the chronaxie scale. The distribution of the 392 muscles whose probable lack of reinnervation is demonstrated by these same tests (col. 1 of table 122) is less heavily con- centrated, but the low end of the scale is a region of low density. On the other hand, the distribution of chronaxies for the 147 muscles (col. 2 in table 1 22) with contraction following nerve stimulation but not voluntarily differs markedly from the other two. Very high chronaxies are absent, but otherwise the distribution is fairly uniform with only slight peakedness in the range of 3 to 4 msec. This distribution suggests that many of the 147

Table 121.—Voluntary Movement, Response to Direct Nerve Stimulation, and Chronaxie Determinations, for All Affected Muscles Following Complete Sutures on All Seven Major Nerves, New York CenterJ None Voluntary movement Any Total Chronaxie, msec. Contraction follow- ing nerve stimula- tion Contraction follow- ing nerve stimula- tion Contraction follow- ing nerve stimula- tion None 0) Any Total (3) None (4) Any (5) Total (6) None (7) Any (8) Total (9) (2) Total 392 268 124 0 1 0 1 7 6 16 12 16 11 54 147 48 99 10 12 10 15 20 11 9 7 4 1 0 539 316 223 10 13 10 16 27 17 25 19 20 12 54 39 18 21 2 1 0 0 1 4 6 4 2 1 0 1,642 1,681 335 1,346 601 352 133 90 78 49 22 8 8 2 3 431 286 145 2 2 0 1 8 10 22 16 18 12 54 1,789 365 1*424 609 363 143 105 97 56 25 11 10 2 3 2,220 651 1,569 611 Known 317 1*325 599 351 133 90 77 45 16 4 6 1 3 0 1 365 143 106 105 66 47 27 28 14 57 2 3 4 5 6 7 8 9 >10.. 1 Only muscles with observations on both voluntary movement and response to direct nerve stimulation are included. muscles had been reinnervated, and that failure of contraction on volun- tary stimulation occurred because either the contraction was too small, or the patients had not developed central command of the muscles. The very small group (39 out of 2,220 in col. 4, table 122) with voluntary con- traction but no response to electrical stimulation of the nerve might repre- sent errors in either test. None of the 39 muscles has a chronaxie value above 10 msec., which might be taken to suggest that the neurological examina- tion had not erred as often as the electrical. On the other hand an esti- mated 6 of these muscles (3 observed) had chronaxie values of 0 to 1 msec., and these may be examples of congenitally anomalous innervation. Loss of normal innervating supply would not have affected the anomalous, thereby perhaps leaving intact the ability to contract voluntarily. The remainder might also be explained similarly on the basis of reinnervation by ingrowth of nerves other than those of normal supply. In figure 19 are plotted, in cumulative percentage form, the four estimated distributions of chronaxie values shown in table 122. 223

Table 122.—Voluntary Movement, Response to Direct Nerve Stimulation, and Estimated Distribution of Ckronaxie Values, JOT All Affected Muscles Following Complete Sutures on All Seven Major Nerves, New York Centerl None Voluntary movement Any Total Chronaxie* msec. Contraction follow- ing nerve stimula- tion Contraction follow- ing nerve stimula- tion Contraction follow- ing nerve stimula- tion None (1) Any (2) Total (3) None (4) Any (5) Total (6) None (7) Any (8) Total (9) Total 392 0 3 0 3 22 19 51 38 50 35 171 147 15 18 15 22 30 16 13 11 6 1 0 539 15 21 15 25 52 35 64 49 56 36 171 39 4 2 0 0 2 7 11 7 4 2 0 1,642 742 435 165 112 95 56 20 5 7 1 4 1,681 746 437 165 112 97 63 31 12 11 3 4 431 4 1,789 757 453 180 134 125 72 33 16 13 2 4 2,220 761 458 180 137 149 98 95 61 67 39 175 0 1 5 0 3 24 26 62 45 54 37 171 2 3 4 5 6 7 8 9 £10 1 Only muscles with observations on both voluntary movement and response to direct nerve stimulation are included. Distributions of chronaxie values shown in table 121 have been inflated to those shown here by multiplying individual frequencies by the ratio of the total on the first line of that table to the number tested, shown on the third line. See text. Addition of chronaxiemetry to the other tests therefore immediately re- veals three important facts: (a) at follow-up the distributions of chronaxie values fail to reveal sharp discontinuities associated with reinnervation and continued denervation; (b) however, the distributions of muscles grouped as to contraction on voluntary stimulation and response to nerve stimulation are concentrated at quite different regions of the scale; and (c) the regions of concentration vary with the pattern of the results of the other two tests: with maximal evidence of regeneration the region of low chronaxie values is densely occupied; with no evidence of regeneration it is the region of high chronaxie values in which concentration occurs; and with mixed evidence of regeneration the distribution is more uniform. The distributions of table 122 and of figure 19 strongly suggest that chronaxie is not a specific measure or indicator of nerve regeneration but a probability measure of some kind, such that for one chronaxie the prob- ability is high that regeneration has occurred and for another the proba- 924

bility is low. Unfortunately it is not certain, for any particular muscle, whether regeneration has occurred or not; there is merely more or less evidence of regeneration or lack thereof. The particular interest in the chronaxie, therefore, lies in the possibility that it may aid in discriminating between cases with and without regeneration. The suggestion of table 122 is that the chronaxie provides little if any information not already inherent in the results of the other two tests, but one further step was taken on the basis of a cut in the chronaxie scale between high values and low values such as to maximize the agreement between the resulting chronaxie classi- fication and each of the other two tests. When various cuts were tried on the New York data it was found that the regions 0 to 5 and 6 or more produced the best agreement with both voluntary contraction and response Figure 19. Cumulative Percentage Distributions of Chronaxie Values by Response to Direct Nerve Stimulation and to Voluntary Stimulation, New York Data on All Muscles. 20 — 10 — 20 — 10 225

to nerve stimulation. That is, if the cut is made in this way there are 163 cases with chronaxie values of 0 to 5 in which voluntary contraction was not observed and 61 cases with chronaxie values of 6 or more in which voluntary contraction was observed, or a total disagreement of 224 cases, 10.1 percent of the entire sample of 2,220 muscles. When response to nerve stimulation is taken as the criterion there are 62 cases with chronaxie values of 0 to 5 in which contraction did not occur and 68 with values of 6 or more in which it did, or a total of 130 disagreements, 5.9 percent of the sample of 2,220 muscles. In each instance this number of disagree- ments is the smallest of the set obtained by varying the cut. For example, if the cut be made after 3 msec. rather than after 5 the percentage of disagreements rises from 10.1 to 13.4 when voluntary contraction is the criterion and from 5.9 to 12.5 when response to nerve stimulation serves as the criterion. More complex criteria might be evolved than these, as by paying attention not merely to the total number of disagreements but also to their nature, but as may be seen from table 123 it would appear Table 123.—Estimated Distribution of Affected Muscles by Voluntary Contraction, Response to Nerve Stimulation and Chronaxie Group, New York Data l Chronaxie group, msec. Response to direct nerve stimulation None Any Total A. No voluntary contraction 0-5 47 6 or more 345 Total 392 B. Voluntary contraction 0-5 15 6 or more 24 Total 39 C. All muscles 0-5 62 6 or more 369 Total 431 116 31 147 163 376 539 1*605 37 1,642 1*620 61 1,681 1,721 68 1,789 1,783 437 2,220 i Based on table 122. 226

that no further refinement is likely to be fruitful. If the discrimination between regeneration and lack thereof is predicated solely upon the response to voluntary stimulation, then 539 or 24.3 percent of the muscles will be called denervated. If response to nerve stimulation is the only basis then 431 or 19.4 percent of the muscles will be termed denervated. If both these tests are used in combined fashion, then 392 or 17.7 percent will appear to be definitely denervated and there will be some uncertainty about an additional 186 or 8.4 percent. If, now, the chronaxie information be added on the basis of the cut developed above it would be applied only to the latter group of 186 cases, so that the margin of its contribution is a small one to begin with. The addition of the chronaxie information adds 55 to the count of cases without regeneration and 131 to the count of cases with regeneration, so that the final split is 447 or 20.0 percent without regeneration and 1,773 or 80.0 percent with regeneration. This result differs too little from that obtained with nerve stimulation alone to be of any practical value in estimation, but one must not overlook the consider- able contribution made by the chronaxie information to the correct classi- fication of the 147 cases contracting on nerve stimulation but not volun- tarily and of the 39 cases contracting voluntarily but not on nerve stimu- lation. Although it must remain literally true that no individual muscle can be certified as surely not reinnervated, the amount of information on the probability of such reinnervation is considerable and it is of particular interest to plot, for each point on the chronaxie scale, each of several estimates of the proportion of muscles believed to have been reinnervated. This is done in figure 20 according to the following criteria of reinnervation: (a) Nerve stimulation only. (b) Voluntary stimulation only. (c) Both voluntary and nerve stimulation—both tests agree. (d) Voluntary and nerve stimulation plus chronaxie to dispose of disagreements between them. If table 122 is approached more generally, and it is not required that the chronaxie scale be arbitrarily divided into two regions, the following facts stand out: a. Muscles unable to contract either voluntarily or on nerve stimulation rarely have low chronaxies, and often have high values; almost all values of 10 msec. or more were read on muscles in this group. b. Muscles unable to contract voluntarily, but responding to direct nerve stimula- tion, have no chronaxies in the region of 10 or more msec., but are much less con- centrated in the 0 to 3 msec. region than muscles contracting voluntarily. c. Muscles contracting voluntarily but not on nerve stimulation have scattered chronaxie values, mostly high but none as high as 10 msec. d. Muscles contracting both voluntarily and on nerve stimulation only rarely (2.3 percent) have chronaxies of 6 msec. or more. On the basis of these observations one should have little hesitation in concluding that failure to move both voluntarily and on nerve stimulation 227

Figure 20. Percentage of Muscles With Specific Evidence of Regeneration, by Ckronaxie, New York Data on All Muscles PERCENT PERCENT 100 CONTRACTION ON BOTH VOL. S ELEC. STIMULATION WITH CHRONAXIE RESOLVING MIXED CASES CONTRACTION ON BOTH VOL. t ELEC. STIH ONTRACTION ON NERVE STIMULATION 1 CONTRACTION ON VOL. STIMULATION 20 — 10 4 5 CHRONAXIE* MSEC. 10 OR MIRE is almost wholly attributable to failure in regeneration; it cannot often be true that regeneration has occurred but muscle atrophy has been too extensive to support movement. It also seems clear that the great majority of muscles unable to move voluntarily, but responding to nerve stimulation, had been reinnervated; failure to move voluntarily must be attributed to psychological factors in the areas of motivation and learning. The few muscles with voluntary movement and no response to nerve stimulation present no clear-cut picture; certainly they do not look like other muscles able to move voluntarily in that their chronaxies are higher. One would be inclined to prefer the results of the nerve stimulation in most instances here. Finally, it is rare indeed that one would be suspicious of any muscle reported to have contracted both voluntarily and on direct nerve stimu- lation. If, now, table 123 is approached from the more general standpoint of the cogency of evidence for regeneration, and with an awareness of the conflicting testimony of the tests, its entries may be organized around two opposite poles representing the counts of muscles for which all three tests 228

agree, namely the 345 with minimal evidence of regeneration and 1,605 with maximal evidence. According to these estimates at least 15.5 percent of the muscles reflect failure of regeneration, and at least 72 percent reflect regeneration. The middle 12 percent can be allocated to one of the other only with some sacrifice of certainty, but on the basis of the observations already cited it would appear that its several parts may be allocated as follows: Probably not indicative of nerve regeneration 47 muscles for which there is at most weak chronaxie evidence of regeneration 24 muscles for which there is at most evidence of voluntary contraction. Probably indicative of nerve regeneration 116 muscles with no voluntary contraction, but response to nerve stimulation and low chronaxies 31 muscles with voluntary contraction, and moderately high chronaxies, which responded to nerve stimulation 15 muscles with voluntary contraction and low chronaxie, but failing to respond to nerve stimulation 37 muscles with both voluntary contraction and response to nerve stimulation, but high chronaxie. In summary, then, these considerations lead to an estimate of 1,804 muscles with, and 416 without, evidence of nerve regeneration, or 81 and 19 per- cent. These estimates are so close to those of 392 and 1,828 obtainable directly from table 120 that it would hardly seem necessary to present the chronaxie material in detail by muscle. Since direct muscle stimulation was usually done at the New York center it may be of interest to note that only 1 muscle among 1,552 studied by means of voluntary and direct nerve stimulation failed to contract on direct muscle stimulation; this was 1 of 119 muscles which failed to contract either voluntarily or on direct nerve stimulation. Among 405 muscles examined at Philadelphia there were 5 which failed to contract on direct stimulation, 3 among 51 which did not contract voluntarily or on direct nerve stimulation and 2 among 51 which contracted on direct nerve stimulation but not voluntarily. It seems plain that complete muscle atrophy is rare in this material, and that the failure of muscles to contract voluntarily cannot, therefore, be attributed to this factor. The Philadelphia observations are presented in table 124 to parallel those of the New York Center in table 121. Without further analysis it is plain that the Philadelphia chronaxie determinations are much less certainly discriminating than those of the New York center. Table 125 presents adjusted data for the Philadelphia center, based on table 124, the adjust- ment being the same as that performed on the New York data with one exception: account has been taken of the fact that examiners in the Philadelphia center selected muscles for direct nerve stimulation with some regard for their response to voluntary stimulation. The amount of infor- mation in table 124 is not large, and inflation to the totals used in table 125 provides only very approximate distributions with sharp discontinuities.

Table 124.—Voluntary Movement, Response to Direct Nerve Stimulation, and Chronaxie Determinations, for All Affected Muscles Following Complete Sutures on all Seven Major Nerves, Philadelphia Center1 None Voluntary movement Any Total Chronaxie, msec. Contraction follow- ing nerve stimula- tion Contraction follow- ing nerve stimula- tion Contraction follow- ing nerve stimula- tion None (1) Any (2) Total (3) None (4) Any (5) Total (6) None (7) Any (8) Total (9) Total 307 248 59 3 2 3 1 7 96 41 55 3 5 5 5 6 1 8 403 289 114 70 54 16 2 1 4 613 286 327 55 58 48 15 34 11 18 4 43 1 40 683 377 302 75 5 3 7 1 9 2 5 709 327 382 58 63 53 20 40 12 26 4 52 1 53 1,086 629 Unknown . . . 340 343 57 59 52 15 36 13 18 4 43 1 45 457 63 66 60 21 49 14 31 4 65 1 83 0 6 7 8 6 13 1 13 1 2 3 . . . . 4 2 2 5 6 5 7 8 13 9 22 13 9 >10 25 13 38 5 30 1 Only muscles with observations on both voluntary movement and response to direct nerve stimulation are included. When a division of the chronaxie scale was sought on the basis of mini- mizing disagreement with the results of direct nerve stimulation, the best division was found to be 0 to 10 v. 11 or more msec., but even on the basis of this division 27 percent of the muscles presented discrepancies between the two tests in comparison with only 6 percent in the New York center. In figure 21 the two centers are compared as to the relation between the several tests. In table 126 are shown the results of dividing the chronaxie scale in the optimum fashion just described. It will be noted that all 3 tests agree in 178 instances with minimal evidence of regeneration, and in 1,552 instances with maximal evidence. The corresponding percentages are 7 and 60, which may be compared with 16 and 72 in the New York data. If the cases in the remaining 6 cells of the table, about which the tests disagree to some extent, are allocated in the fashion described for the New York data the final estimate of the percentage with regeneration is found to be 77, slightly below the value of 81 calculated for the New York 230

data, but well above the values of 72 percent obtained from the data on voluntary contraction alone and 71 percent obtained from the observations on direct nerve stimulation. The numerical discrepancy between the estimates of the two centers is not too important because there are also differences between them in the emphasis they placed upon the examina- tion of specific muscles, and these differences are not taken into account here. The chief lesson of the analysis is the same for both centers, namely that the best estimate which can be made from the data on all three tests is about what may be obtained from the joint observations on voluntary Table 125.—Voluntary Movement, Response to Direct Nerve Stimulation, and Estimated Distribution of Chronaxie Values, for All Affected Muscles Following Complete Sutures on All Seven Major Nerves, Philadelphia Datal None Voluntary movement Any Total Chronaxie, msec. Contraction follow- ing nerve stimula- tion Contraction follow- ing nerve stimula- tion Contraction follow- ing nerve srimula- tion None (1) Any (2) Total (3) None (4) Any (5) Total (6) None (7) Any (8) Total (9) Total 555 28 19 28 9 66 173 9 16 16 16 19 3 25 728 193 24 1,686 284 299 247 77 175 57 93 21 222 5 72 1,879 308 311 296 77 199 81 93 21 222 5 84 748 52 31 77 9 90 24 47 1,859 293 315 263 93 194 60 118 21 250 5 75 2,607 345 346 340 102 284 84 165 21 373 5 144 9 173 0 37 35 44 25 85 3 72 1 12 49 2 3 4 24 24 5 6 47 7 8 123 28 151 123 9 10 57 9 3 60 9 12 69 9 11 12 75 9 84 12 77 89 87 86 13 14 12 10 22 12 10 22 15 16 47 19 66 12 21 33 59 40 99 17 >18 47 10 57 12 26 38 59 36 95 1 In view of selection of muscles for direct nerve stimulation as well as chronaxie, the estimates were developed to reflect the entire sample of muscles with tests for volun- tary movement, not merely those tested both for voluntary movement and induction of movement by direct nerve stimulation. 231

8 8 9 8 8 R 8 ft f ft I I 232

contraction and response to direct nerve stimulation on the following assumptions: a. Muscles which move voluntarily are innervated (i. <-.* errors of positive findings in neurological examination are absent); b. Muscles which do not move voluntarily but do contract on direct nerve stimu- lation are innervated (i. e., positive results of the electrical test are valid); and c. Only those muscles which contract in neither are denervated. In the New York center about 21 percent of the muscles examined by means of voluntary and direct nerve stimulation were also studied by means of electromyography, and in the Philadelphia center about 19 percent. In both centers there was a marked tendency to avoid electromyographic study of muscles unable to contract voluntarily or on direct nerve stimu- lation, as may be seen in table 127, which pertains to muscles affected by complete suture and on which both voluntary and direct nerve stimulation was performed. Although the eleciromyographic observations are quite sparse, it was thought that they might be useful in distinguishing between muscles which Table 126.—Estimated Distribution of Affected Muscles by Voluntary Contraction, Response to Nerve Stimulation and Chronaxie Group, Philadelphia Data Chronaxie group, msec. Response to direct nerve stimulation None Any Total A. No voluntary contraction (MO 377 11 or more 178 Total 555 B. Voluntary contraction 0-10 145 11 or more 48 Total 193 C. All muscles 0-10 522 11 or more 226 Total 748 135 38 173 512 216 728 1*552 134 1,686 1,697 182 1*879 1*687 172 1,859 2*209 398 2*607 403930—BT- -17 233

could not be moved voluntarily because of failure in regeneration and those in which absence of voluntary movement might be attributed to other factors. Electromyographic studies on the 33 muscles shown in table 127 as incapable of voluntary movement at the New York center usually showed no more than fibrillation potentials if direct nerve stimu- lation produced no contraction, and motor unit potentials if direct nerve stimulation resulted in contraction. Table 128 provides the details of this relationship. The very high correlation exhibited there lends considerable support to the validity of the interpretation that failure of voluntary move- ment is not a direct measure of failure of peripheral nerve regeneration. Table 127.—Choice of Muscles for Electromyographic Study in Relation to Voluntary Movement and Results of Direct Nerve Stimulation, by Center Results of stimulation New York Philadelphia Muscles with electromyo- graphic studies Muscles with electromyo- graphic studies Voluntary Direct nerve Total muscles Total muscles Number Percent Number Percent None None 120 98 19 1,319 17 16 2 291 14.2 16.3 10.5 22.1 51 51 15 288 4 3 2 68 7.8 5.9 13.3 23.6 None Contraction.. . . None Contraction Contraction Contraction.. . . Total 1,556 326 21.0 405 77 19. 0 Table 128.—Electromyographic Interpretations and Response to Direct Nerve Stimulation for Muscles With No Voluntary Movement, New York Data Electromyographic interpretation Response to direct nerve stimulation None Contraction Total At most fibrillation 13 1 14 Motor unit potentials 4 15 19 Total 17 16 33 The analysis of electromyographic data from the New York center was carried one step farther by introducing chronaxie as a fourth variable. It is unfortunate that the data are so few (table 129). They do suffice to 234

show a very intimate association between chronaxie and electromyographic interpretation: % cases with at most fibrillation had chronaxies of 6 or more, and tf%it cases with motor unit potentials had chronaxies below 6. In the main, also, table 129 reveals all 3 electrical tests to be in moderately good agreement and to point to the necessity for regarding muscles in- capable of voluntary contraction as a mixed group, some denervated and some not. The electromyographic data are too few to provide any in- dependent basis for making that distinction, but do at least roughly con- firm the distinction based on the results of nerve stimulation and chron- axiemetry. The only discrepancy of note consists of the 4 muscles with motor unit potentials but no movement either voluntarily or on nerve stimulation and chronaxies of 6 or more. One of these 4 was interpreted as "few motor unit potentials with voluntary effort" and 3 as "spontaneous fibrillation potentials plus some motor unit potentials, and no complex or other potentials." Since potentials of the latter variety were noted in only about 8 percent of the 284 with motor potentials and movement on both voluntary and direct nerve stimulation, these 3 cases are not typical of those with motor unit potentials and they will hardly serve to challenge the testimony of the other 2 electrical tests. Table 129.—Electromyographic Interpretations in Relation to Voluntary Stimu- lation, Direct Nerve Stimulation, and Chronaxie, New York Data Contraction Chronaxie 0 to 5 msec. Chronaxie 6 or more msec. Voluntary Direct nerve stimu- lation At most fibrilla- tion Motor unit po- tentials At most fibrilla- tion Motor unit po- tentials None None 1 0 0 2 0 8 2 284 12 1 0 0 4 7 0 5 None Some Some None Total Some 3 294 13 16 2. Estimated Influence of Factors Preventing Voluntary Contrac- tion The method outlined in the preceding section provides a tool for esti- mating likelihood of reinnervation and for estimating the frequency with which psychological and neurological processes outside the neuromuscular complex may have impeded the voluntary contraction of reinnervated muscles. Table 130 contains these estimates for muscles of chief interest following complete suture, and is confined to cases in the representative 235

jy« -3 Ji g-g 1 oo in NO 00 CM r- 3 RI 3 > a-lT! s I •* co ON o m ON Isg n *• U 1 §?! JS ill 151 1 u-X 1 § 1 T3 | O C ^ CM CM CM -as 9 °^3 * -H c3 5i: |f 1 1 CM r- m * r^ NO Sjf- H -* m CM CM GO O ^* CM CM a isi PI g oo eo m CM •* r- ^- eo CM -*• • * -o •* a li t M* o ;j ON o o * r*- CM 1 •2 O GO ON 00 ON 00 00 J4 M U C > g* o II d C? O If O S— ' O oo r* - co co c s -j ^ ' § 2 >-3 6 •^ o •7 •-H ^1 N_^ i i * s E h £ o* co m oo •-i NO rv) co NO oo *-i CM to CM g, ** S t; - 0 NO co IF i NO NO ON T3 «T3 i i co CM ^I " "* n § U J9 2 X a * •JP i T Q ^* •- — co -52 I i fc -^ 1 I 3 -3 |||| ^ S S88 in *-i ON 1 O C* GO h 1 1 E liil NO 00 CM y 1 S PJ CO *Q ON ON •» •!•? C C || E 1 i^gi ^ cN r*J oo t- oo •^j- in co s U IT) 00 OO 00 § §| i l§i 1 y NO c'l CT1 in ^ co 73 g 1 2 es CO iri oo oo co — — 1 1 JB 'o I i H •J .^— . r oo — ON 00 t*- i 1 £ " 00 OO 00 1 I IS r~ r- oo c s •o m in r* NO >§ 1 3 ON 00 CN] | •^ w •-» 1 1*1 3 ESS •" ^ O « — Comparative Esi I 1 I.S * : 2 *. : fc I?i E < 43 d Median .... T- ^ i i J J3 23«

to Cv! CM co oo in eg ve co ^ eN vo oo r- ~ •* •* O •* oJ 00 CO O 00 » * # ^ r- m co .- 0) ^i C M "SS in NO c^ ^— • -—• N.X *H •* N m CM oo m 5^ '^ c** O T• r^ N C M 00 O ON Tj - CO cs cq rg *-• m ^- t- T-i T-I ON NO W Scs oN T« in co vo ^- C M cs C M r- •* cs o r- •* o OO NO 00 ?•* CM ON TT -• NO 00 •« ov ON r- eg CM in r- O •1 N 00 vO ^- l*- r* C M r- o oN §e^ •* 1*- t*- ^H r^ oo CM o ~ oo oo r- oo *-• *-- *-• ^ CO cO CO S25 *-• T-• T^ •-i -" T-- CM- * Ov CO CN 00 GO m vo •*• *-• r^ ^o vn i*- •* vo r- ON co in m m T|- co m Tf OO OO Ov CN O) 00 NO CO ON NO •O oo m r^ oo r~ ON r- *-i •* oo oN oN r^ oN CO 00 ON CO CO CO ON CO Ov •* cv r- oo cs co co »H ON O r- in f ov r-- m ON ON ON ON •-i•-iTH r^ ON r- •»? CNT ON t^ ON 00 t- f- - ON 00 NO 00 in oo oo co in in oo o r- o vo t*- vo r- i*- •* •<- 00 CM O C M n n co t^ C^ CO co r- *-i CM" --T co" * * 0O •* ov •-i oo £^ I^C* •f o « c- »5-v^ ON NO r- — co *-J •*' V vo vc JLIxt/ oN oo r^ o CO •,«, •* •* oo T-. C M '-. CM *-t *- CM •- N oj m co • T-. CM co * m cs CM •* in t-- o r- *-• ^^ ^- CO 1^ 0 U-. — T •* vO 0) O £££ cs vo in * oo m o CM *-i ^- C - 33 r- ^ in r- t^ m co vo ON CO * O *-^ NO ON NO r^ O C?. --0 m *H •-i -Tr •* •» c o T OO 00 O C- - O Ov 00 CM I*- ON oo ON r- Tr r^ CM r-- •* CM r- in in vo in C M in in >-i *-i •-i ON oo eo oo oo r- O »H l*- C M £££ o in o oN t^ m co in CM co in in •* co t*- NO CO NO ON CO CO CO' NO O oo in r^ NO r- ON NO m •* CO 00 d ON •*• CO O CO 00 00 O m ov m co NO * CM m ON NO P- CM CM • * ON •* CO •* *-i ^i NO •* T-« CO OO CM O co co m 00 O ON .-i •* r- ON NO CM ON C^ OO 00 r- vo •* vO » » » in •* CN in ON CM CM oo in NO v_x v- ' ^_x S- r~ in ro ov vo *-> oo r • oo S22: >- ON O T- 0 CM NO in m •* CO •M •* in co CM •* r- * TI •» m r* - 00 S 3 S « oo fs co ^- T* CO ON -^ O ON 00 00 ON $33 ON •* CO CM 00 *-« 00 r~ NO ill • bi g> : -Si* *!•§ "> c : §1 g> i ** * ji ill fe.*| iJ-^g1 s.a>^'° JJJ j S .tf^-9. 8 .Sf^ u -a ex « T3 J3 c C T3 J3 355 P^tSi O E E Pill 3 •• - < < O W « W O b t* Radial Tibial |-a 2 •a •43 II i |3 vS 1,1 •h-2 „ a }< u i i I C u S +«* i

sample. It also contains, for comparative purposes, the following additional estimates: a. The percentage reinnervated, using as the basis of estimate the percentage contracting voluntarily for all muscles in the representative sample, not merely those in which direct nerve stimulation was done; b. The percentage reinnervated, using as the basis of estimate the percentage con- tracting on direct nerve stimulation; and c. The percentage of cases in which voluntary contraction was blocked by psycho- logical and learning factors* estimated from data on all complete sutures, not merely those in the representative sample. The estimates based on either voluntary contraction alone or response to nerve stimulation alone do not differ appreciably in the aggregate, but occasionally large discrepancies are noted for individual muscles. When the estimates of columns 3 and 5 of table 130 were correlated, the correlation coefficient was found to be + .96,w and the equation for the best-fitting straight line is: Y=-7.90+1.1 IX where Y denotes the percentage contracting voluntarily and X the per- centage contracting on direct nerve stimulation. Figure 22 contains a plot of these percentages. The estimates in column 7, reflecting the as- sumptions stated above and the methodology developed in the preceding section, are considered to be the best estimates for the muscles tabulated, except possibly for the fact that there is some slight bias in the selection of muscles for testing by direct nerve stimulation. However, the bias is considered too small to warrant any further adjustment of the estimates of column 7. On the average, 23 percent of the muscles unable to contract volun- tarily were observed to move on direct nerve stimulation; for the repre- sentative sample of sutures the figure is 22 percent. Individual muscles are not represented by enough cases to permit an extensive comparison of muscles, and to compare nerves on the basis of all muscles combined raises problems of the independence of one muscle from another on the same limb. Accordingly, comparisons have been made to only a limited extent and on the basis of individual muscles in a fashion suited to the assumption of independence of all observations combined together. The specific comparisons made, and their results, are as follows: a. The various nerves, with each represented by the muscle with the largest number of cases in which voluntary movement was not possible* provided there be at least 25 such cases, i. e., Median—abductor pollicis brevis Ulnar—1 st dorsal interosseus Peroneal—extensor hallucis longus Sciatic-peroneal—extensor hallucis longus Sciatic-tibial—flexor hallucis longus These muscles are clearly not homogeneous as to the proportion, among all incapable of voluntary movement, which contract on direct nerve stimulation: the median and ulnar muscles more often contract on electrical srimulation. 15 It should be noted that this coefficient has been obtained on a set of percentages* i. e., averages, which mask a certain amount of individual variation. The corresponding coefficient applicable to individual muscles, as estimated from the four-fold table of voluntary contraction by response to nerve stimulation, is +.89. 238

Figure 22. Percentage of Affected Muscles Contracting Voluntarily After Com- plete Suture, and Percentage Contracting on Direct Nerve Stimulation PERCENT PERCENTA8E CONTRACTING VOLUNTARILY PERCENT WO 90 80 70 60 50 30 20 10 T I (EACH POINT REPRESENTS A MUSCLE) I I -LINE OF IDENTITY I I I I I I 10 20 30 40 50 60 70 80 PERCENTA6E CONTRACTING ON NERVE STIMULATION 90 100 90 80 70 60 30 20 10 100 b. The abductor pollicis brevis and the 1st dorsal interosseus are obviously quite similar. c. Peroneal and sciatic-peroneal lesions were compared on the basis of each muscle. For the extensor hallucis longus the peroneal lesions have an advantage, but for the other three peroneal muscles the percentage is higher for sciatic-peroneal. However, for only one muscle (extensor digitorum longus) is the discrepancy a statistically significant one (P about .03), and for the material as a whole it would seem best to take the position that the sciatic-peroneal and the peroneal do not differ. d. The two sciatic components were compared on the basis of the extensor hallucis longus and the flexor hallucis longus and appeared not to differ by more than chance expectation. The muscles chosen to represent the upper extremity (abductor pollicis brevis and 1st dorsal interosseus) are, of course, distal muscles while those chosen to represent the lower extremity in the first comparison (a) are proximal muscles, and there may be some confounding on that account. Distal muscles in the lower extremity were not tabulated in this fashion, 239

and proximal muscles in the upper extremity are represented by too few cases to settle the matter; we know only that the five sets of cases studied in the first comparison (a) differ quite significantly. The foregoing comparison provides a test of the hypothesis that the proportion of muscles able to contract on direct nerve stimulation, among muscles unable to contract voluntarily, is constant from nerve to nerve, muscle to muscle. Perhaps an even more meaningful hypothesis to test is that the proportion of muscles unable to contract voluntarily, among all those innervated, is constant from muscle to muscle and nerve to nerve. The latter proportion is given in columns 9 and 16 of table 130, from which comparisons may be made rather directly. Column 9 is limited to the representative sample of complete sutures, while column 16 applies to all complete sutures for which both voluntary and electrical stimulation was attempted. The evidence of the two columns is the same, and since column 16 rests on many more cases it may be used as the basis for any comparisons. It seems clear, without formal test, that the proportion of reinnervated muscles which do not contract on voluntary stimulation is much higher in the lower extremity than in the upper, and especially for muscles affected by sciatic lesions. However, among the muscles of the lower extremity the gastrocnemius and soleus (following injury to the sciatic-tibia!) has an exceptionally low percentage in relation to other muscles of the lower extremity. In summary, then, table 130 provides the best estimates available in this material of the likelihood of reinnervation and, given reinnervation, of the chance that contraction will not occur on voluntary stimulation. The latter varies by nerve and to some extent by muscle within the set innervated by a given nerve, in that it is much higher in the lower ex- tremity than in the upper, and that among muscles of the lower ex- tremity, the gastrocnemius and soleus are atypical. 240

Next: Recovery of Sensory Function »
Peripheral Nerve Regeneration: A Follow-Up Study of 3,656 World War II Injuries Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

In peacetime, the busiest civilian clinics do not see enough peripheral nerve injuries to permit authoritative conclusions to be drawn about their management. In World War I, large numbers of these injuries were skillfully cared for by a small group of pioneer neurosurgeons, but there was no comprehensive follow-up and the opportunity to use the experience to the fullest possible extent was lost.

The publication of Peripheral Nerve Regeneration: a Follow-Up Study marks the end of a huge clinical research program that began in 1943, in the course of World War II. The program was participated in by more than a hundred of the neurosurgeons who served in the Medical Corps, as well as by many neurologists, neuroanatomists, neurophysiologists, neuropathologists, physical therapists, statisticians, and representatives of the administrative personnel of every echelon of command in the Army Medical Corps. Later the program was also participated in by representatives of the Veterans Administration and the National Research Council.

The primary purpose of this study was to evaluate the suites of peripheral nerve injuries sustained in World War II, with the hope of standardizing such treatment for future wars and, where possible, for similar injuries of civilian life. The secondary purpose of this study was to discover nerve injuries among veterans of all services that still required remedial measures. Peripheral Nerve Regeneration: a Follow-Up Study describes the final level of regeneration in representative cases of complete suture, neurolysis, and nerve graft, examines the apparent influence of gross characteristics or the legion, and or associated injuries, upon final result, and evaluates predictions of final recovery based on gross and histologic study of tissue removed at operation. The report of this study of postwar nerve regeneration provides for the surgeons of the future a body of information upon which they may guide repair of injured peripheral nerves and initiate needed orthopedic rehabilitation.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!