Monitoring of somatosensory evoked potentials during surgical procedures on the thoracoabdominal aorta

J

THoRAc CARDIOVASC SURG

1987;94:275-85

Monitoring of somatosensory evoked potentials during surgical procedures on the thoracoabdominal aorta IV. Clinical observations and results Thirty-three patients undergoing operations on the descending thoracic or thoracoabdominalaorta were monitored to evaluate causes and effects of spinal cord ischemia as manifested by changes in

somatosemoryevoked potentials. Maintenance of distal aortic perfusionpressure(>60 mm Hg) by either shlDlt or bypass techniques in 17 patients resulted in preservation of somatosensoryevoked potentials and a normal postoperative neurologic status, irrespective of the interval of thoracic aortic cross-clamping (range23 to 105 minutes~ In 16 other patients in whomcross-clamptime ranged from 16 to 124 minutes, evoked potentialloss was observed because of (1)failure to provide distal perfusion(n = 8), (2)inadequate maintenance of distal perfusionpressure «60 mm Hg) despiteshlDlt/bypass (n = 6), or (3)interruption of critical intercostal arteries (n = 2). Incidenceof paraplegia in the entire group was 15.1 % (5/33) and was limited to only those patients in whom evoked potential loss occurred (5/16, 31.2 %) (p = 0.02). Loss of somatosensory evoked potentials for more than 30 minutes resulted in a 71.2% (5/7) incidence of paraplegia, whereas no neurologic deficit was noted in patients (0/26) in whomevoked potential loss was either prevented or limited in duration to 30 minutes (P < 0.001 versus loss for more than 30 minutes). Intraoperative monitoring of somatosensory evoked potentials is a sensitive indicator of spinal cord ischemia. Simple aortic cross-clamping, failure to maintain distal perfusion pressure above 60 mm Hg, and inability to reimplant critical intercostals in a timely fashion result in a high rate of paraplegia if duration of spinal cord ischemia as measured by somatosensory evoked potentials exceeds 30 minutes. Routine evoked potential monitoringduring thoracoabdominal proceduresappears useful in assessing the adequacy of spinal cord perfusion. Furthermore, it can alert the surgeon to the necessity for critical intercostal artery reimplantation as weD as the need for adjustment or regulation of distal aortic perfusion.

Joseph N. Cunningham, Jr., M.D., John C. Laschinger, M.D., and Frank C. Spencer, M.D., New York and Brooklyn, N. Y.

Significant controversy remains concerning the cause of paraplegia and the usefulness of adjunctive techniques designed to prevent such complications during operations on the thoracoabdorninal aorta. The disconcerting prevalenceof this catastrophic complication has From the Divisionsof Cardiovascular and Thoracic Surgery, Departments of Surgery, New York University Medical Center, New York, N. Y., and Maimonides Medical Center, Brooklyn, N. Y. Received for publication July 22, 1986. Accepted for publication Aug. 23, 1986. Address for reprints: Joseph N. Cunningham, Jr., M.D., Director, Department of Surgery, Maimonides Medical Center, 4802 10th Ave. Brooklyn, N. Y. 11219.

remained high, despite numerous specific alterations in surgical and anesthetic techniques.':" Most information concerning causation and incidence of paraplegia has been obtained from retrospective studies.r" 20-22 We have previously clarified the importance of adequate distal perfusion and reimplantation of critical intercostal arteries after aortic cross-clamping as a means of preventing paraplegia, both clinically and experimentally. The present report summarizes our enlarged clinical experience with somatosensory evoked potential (SEP) methodology in a prospective group of patients undergoing surgical procedures on the descending thoracic and thoracoabdominal aorta. The goals of this report were to further define (1) what intraoperative 275

The Journal of Thoracic and Cardiovascular Surgery

2 7 6 Cunningham, Laschinger, Spencer

Table I. Summary of preoperative patient data Patient No.

Age (yr)

I

73 83 65

2 3

4

58

5

76

6

58

7 8 9 10 II 12 13 14

73

60 52

75

76 52

15 35

16

2 67

17

42

18

74 62 68 60 62

15

19 20 21

22 23 24 25

26

65

68 46 60

27

70

28

69

29 30 31 32 33

59 58

67 19 66

Etiology

M M F M F M F M F M M M M F F M F M M F M M M F M M M M M M M M F

Atherosclerotic Traumatic (old) Atherosclerotic Atherosclerotic Atherosclerotic Atherosclerotic Atherosclerotic Dissection (chronic) Atherosclerotic Atherosclerotic Atherosclerotic Atherosclerotic Congenital coarctation Congenital coarctation Congenital coarctation Atherosclerotic Dissection (chronic) Traumatic (old) Atherosclerotic Atherosclerotic Atherosclerotic Atherosclerotic Dissection (chronic) Atherosclerotic Traumatic (old) Mycotic Atherosclerotic Atherosclerotic Atherosclerotic Atherosclerotic Atherosclerotic Iatrogenic Atherosclerotic

events result in spinal cord ischemia, (2) which techniques or adjuncts may be useful in prevention of ischemia, and (3) which events or conditions result in the development of permanent paraplegia postoperatively.

Methods and patient population Patient population (Table I). Thirty-three consecutive patients undergoing surgical procedures for lesions of the descending thoracic aorta between October 1981 and October 1984 at New York University Medical Center, New York, New York, and Maimonides Medical Center, Brooklyn, New York, form the basis for this report. Etiologies of thoracic aortic lesions were variable and included atherosclerosis, trauma, mycosis, chronic dissection, congenital coarctation, and iatrogenesis (aneurysm at previous site of patent ductus). Origins of the left carotid and/or left subclavian arteries were involved in 20 patients, and extensive thoracoabdominal disease necessitating reimplantation of mesenteric and renal

Extent of lesion Left subclavian origin to T, Left carotid origin to T, Left subclavian origin to bifurcation Left subclavian origin to T IO Distal to left subclavian to T. Left carotid origin to T, Distal to left subclavian to bifurcation Left subclavian origin to T 10 Left subclavian origin to T 12 T, to T. Left subclavian origin to T l2 Left subclavian origin to T, Postductal to T, Postductal to T, Postductal to T, Left subclavian origin to Til Left subclavian origin to T 12 Left subclavian origin to T, Left subclavian origin to T IO Left subclavian origin to T" Distal to left subclavian to T l2 T, to T" Left carotid origin to T, Left subclavian origin to T IO T, to T. T. to T IO T, to T l2 Left subclavian origin to T, Left subclavian origin to T 10 Distal to left subclavian to T IO T, to T. Left subclavian origin to T, Left subclavian origin to T"

vessels was encountered in two patients. Ages ranged from 15 to 83 years, and patients were predominantly male (23 male and 10 female). . Operative data and techniques. Decisions regarding operative technique and use of shunt or bypass devices were made preoperatively according to the individual preference of the several surgeons* participating in this study and were not altered by intraoperative SEP findings. Dacron tube grafts were inserted in 25 patients by implantation of a cluster of intercostal arteries, as described by Crawford and associates.v' Dacron rooftop patches were employed after excision of saccular aneurysms in four patients, and primary end-to-end aortic anastomoses were accomplished in three patients with congenital coarctations. One patient underwent an *J. N. Cunningham, F. C. Spencer, O. W. Isom, K. Krieger, N. Trehan, A. Culliford, and S. Colvin.

Volume 94 Number 2

Monitoring of SEPs, IV 27 7

August 1987

Cortical Response •

200

" consecutivr.e=~~_~=-_-(~~;;~~~~

stimuli

Signal Input Signal

Signal

AVERAGER

GENERATOR

•

• •

300V

0.6 msec 2.3 cps

Fig. 1. Schematic representation of clinical technique of SEP monitoring.

BASELINE

Axe 4 -1 0

min ",",--...

Axe 7 -20 min

REPERFUSION 20 min

Fig. 2. Type I SEP response: Loss of SEP because of spinal cord ischemia after proximal aortic cross-clamping (AXC) without distal aortic perfusion. extra-anatomical right axillofemoral bypass before excision of a mycotic aneurysm at the level of T 9-10' The technique of spinal cord protection during aortic repair varied. A simple "clamp/repair" technique was used in eight patients without distal aortic perfusion. Femoral-femoral oxygenator bypass was used in the majority of the remaining patients during aortic repair or replacement. Distal perfusion was provided by a tridodecylmethylammonium chloride (TDMAC), heparin-coated shunt in three patients, and no attempt at distal perfusion was made in the three patients with coarctation because of documentation of adequate distal perfusion via large collaterals. Adequacy of distal perfusion techniques was determined by placement of radial and femoral arterial pressure lines in all patients. When heparinized shunts were employed, intraoperative assessment of shunt flow was achieved by insertion of an electromagnetic flowmeter probe* coated with TDMAC into the shunt. *Gould Inc., Cardiovascular Products, Oxnard, Calif.

Evaluation of neurologic status and spinal cord conduction. Thorough neurologic examinations were performed in all patients preoperatively, 24 hours postoperatively, and before hospital discharge. Examinations were performed by a single neurologist to avoid interobserver differences. A clinical SEP system* was used to monitor spinal cord conduction as reflected by SEP generation (Fig. 1).23 SEP responses were graded as types I to IV on the basis of previously established nomenclatures': Type I (Fig. 2). This response typically characterizes the rapid deterioration of spinal cord conduction with increased latency and diminution of SEP amplitude progressing to total absence of conduction 7 to 30 minutes after proximal aortic cross-clamping when the "clamp/repair" technique was used. Type II (Fig. 3). SEP remains normal after aortic cross-clamping because of adequate distal perfusion *TN-3000, Tracor Analytic Inc., Oak Grove Village, 111.; Nicolet Biomedical Instruments, Cranford, N. J.

The Journal of Thoracic and Cardiovascular Surgery

2 7 8 Cunningham, Laschinger, Spencer

BASELINE

AXC 20 min

or - - -

\1"'--...

REPERFUSION 15 min

POST OP 24 hr

Fig. 3. Type II SEP response: Maintenance of normal SEP by adequate distal perfusion after proximal aortic cross-clamping (AXC).

BASELIN( AXC 10 min

REIMPLANT PROXIMAL INTERCOSTAL REIMPLANT DISTAL INTERCOSTAL POST OP 24 hr

----~ - " -...... _

Fig. 4. Type III SEP response (distal bypass instituted after aortic cross-clamping (AXCj): loss and return of SEPs after exclusion and reimplantation of critical intercostals.

(> 60 mm Hg) with mechanical devices (Gott shunt, left atrial-femoral artery bypass, femoral vein-femoral artery oxygenator bypass) or extensive collateralization (coarctation with significant internal mammary arteryintercostal artery flow). Type III (Fig. 4). In the presence of adequate distal perfusion after proximal aortic cross-clamping, sudden loss of SEPs after placement of a distal aortic crossclamp signifies spinal cord ischemia resulting from exclusion of critical intercostal arteries. Removal of the distal cross-clamp results in the rapid return of SEPs. This characteristic response alerts the surgeon to the need for a rapid reimplantation of critical intercostal vessels after exclusion of a diseased aortic segment. Type IV (Fig. 5). Loss of SEPs under circumstances of inadequate distal perfusion occurs slowly and occultly and represents a "fade out" of spinal cord conduction because of marginal or inadequate circulation to the

spinal cord after aortic cross-clamping. This response is typified by patients who are inadequately perfused (< 60 mm Hg) in a retrograde fashion because of technical or anatomical factors (extensive aneurysmal disease or aortoiliac syndrome), or both. All results are expressed as mean ± standard error of the mean, and statistical analysis was performed by Fisher's exact test unless otherwise indicated. Results

SEP measurements in 33 patients allowed clinical identification of four previously described patterns of response after aortic cross-clamping (Table 11),23·27 As illustrated in Fig. 6, 11 patients underwent operations without use of artificial shunt or bypass techniques. Three of these 11 patients had aortic coarctation and exhibited significant evidence of "self-shunting" via collaterals with mean distal aortic pressures remaining

Volume 94

Monitoring of SEPs, IV

Number 2 August 1987

279

BASELINE ~--

Distill P 70 mmHg

- -

-

DistillP ... 60mmHg ... __- -...._ -...._ - - - - -......_ _" , , -.......

Fig. 5. Type IV SEP response: Gradual loss of normal SEPs associated with inadequate distal aortic pressure «60 mm Hg).

DISTAL PERfUSION USED

OISIAL PERfUSION

20 •• Hg

INAOEQUATE

PRESSURE

5EP

RESPONSE

l't'PE

INrORHATION GAINED ABOUT CRITICAl INTERCOSTAL LOCAl ION BY SEP HONITORING

PARAPlEcr A INCIDENCE

I

II (no! 7)

I

I

NONE

3/8 (37.

j"'' "'.",

III (n.2)

I

'1' ' ' "..",

VESSELS ARISE OUTSIDE

VESSELS ARISE WITHIN

""~"

rxcruero

~S)

0/17 (05)

0/2 (05)

T' IV

Nt

I

2/6 (JJ. 35)

Fig. 6. Patterns of SEP responses (types I to IV) and the incidence of paraplegia in 33 patients undergoing procedures on the thoracoabdominal aorta.

greater than 60 mm Hg throughout the cross-clamp interval. A temporary shunt or distal bypass technique was employed in 22 patients. Incidence of paraplegia in the entire group of 33 patients was 15.1% (5/33). Paraplegia was not observed in any of the 19 patients in whom adequate distal aortic perfusion pressure (> 60 mm Hg) was maintained by distal bypass or collateral circulation after aortic cross-clamping (p = 0.01). In contrast, the incidence of paraplegia was excessive (35.7%, 5/14 patients) when adequate distal aortic perfusion was not achieved after proximal aortic crossclamping. Disappearance of SEPs with clamp/repair technique (type I response). Simple aortic cross-clamping without an attempt at distal aortic perfusion in eight patients resulted in rapid loss of SEPs within 7 to 30

minutes (mean 17.0 ± 8 minutes). In the absence of distal shunting or adequate collaterals, mean distal aortic perfusion pressure was less than 20 mm Hg in this group. The increase in latency and diminution of amplitude noted soon after proximal cross-clamping rapidly progressed, which indicates cessation of spinal cord impulse conduction via the dorsal columns. Conduction slowly returned over a 30 to 50 minute period after unclamping in five patients who subsequently had normal postoperative neurologic function. In the remaining three patients, SEPs failed to reappear immediately and permanent paraplegia was subsequently observed. Overall incidence of permanent paraplegia for this clamp/repair group was 37.5% (3/8 patients). Maintenance of normal SEPs associated with adequate distal aortic perfusion (type II response). Distal

The Journal of Thoracic and Cardiovascular Surgery

2 8 0 Cunningham, Laschinger, Spencer

Table D. Summary

0/ intraoperative and postoperative patient

data

Aortic cross-elamp placement Patient No.

1 2 3 4 5

6 7

8 9 10 11

12 13

14 15 16 17

18 19

20 21 22 23 24 25 26 27 28 29 30 31

32 33

Shunt used None None None None None None None None Fern-fern bypass Fern-fern bypass TDMAC-heparin TDMAC-heparin "Self shunt" "Self shunt" "Self shunt" Fern-fern bypass Fern-fern bypass Fern-fern bypass Fern-fern bypass Fern-fern bypass Fern-fern bypass Fern-fern bypass Fern-fern bypass Fern-fern bypass Fern-fern bypass Extra-anatomic (ax-fern) Fern-fern bypass Fern-fern bypass TDMAC-heparin Fern-fern bypass Fern-fern bypass Fern-fern bypass Fern-fern bypass

Proximal Pre-left subclavian Pre-left carotid Pre-left subclavian Post-left subclavian Post-left subclavian Pre-left carotid Post-left subclavian Pre-left subclavian Pre-left subclavian

T. Pre-left subclavian Pre-left subclavian Post-left subclavian Post-left subclavian Post-left subclavian Pre-left subclavian Pre-left subclavian Pre-left subclavian Pre-left subclavian Pre-left subclavian Post-left subclavian

T, Pre-left carotid Pre-left subclavian

I

Distal

T,

29

T9

37

Bifur.

58

TIO TIO T.

Bifur. Til

T I2

65 65 27 23 26 62 68

T9 T,

T,

T, T I2

T12 T, T12 Tl2 T12 Tl2 T,

35

35 55 53

47 26

Distal aortic pressure (mm Hg)

20 20 20 20 20 20 20 20 65 60 65 65 70 60 60 65 60 60

60 70 75 60 70 60 60 65 70

Til

105 61

L,

40

T.

30 85

37

68 62 67 69

30 45

T,

Tl2

Pre-left subclavian Pre-left subclavian Post-left subclavian Pre-left subclavian Pre-left subclavian

16

75

T9

T.

23 23 84 124

T12 T9

T.

T,

Duration (min)

Til

TIO T9 T, Tl2

37

20 20 30

Legend: Fern-fern. Femoral-femoral. Ax-fern. Axillofemoral. SEP. Somatosensory evoked potentials. AXe, Aortic cross-clamp,

aortic perfusion pressure was maintained at 60 to 80 mm Hg throughout the entire cross-clamp interval in a group of 17 patients. Femoral vein-femoral artery oxygenator bypass was employed in 11 patients, heparinized shunts in three patients, and three patients with congenital coarctations exhibited "self-shunting" through large collaterals. Because normal SEPs were maintained throughout the entire cross-clamp interval (aortic cross-clamping range 23 to 105 minutes), intercostal reimplantation was not performed in any patient in this group. No paraplegia occurred despite the fact that over half (9/17) of these cases involved extensive aneurysms terminating below the diaphragm. Use of SEPs to identify critical intercostal vessels (type III response). Two patients in whom SEPs were

first maintained after initial cross-clamping and adequate distal aortic perfusion were noted to exhibit lossof SEP after placement of the distal aortic cross-damp. Because this phenomenon implied inclusion of critical intercostal vessels within the confines of proximal and distal cross-damps, appropriate measures were taken to reimplant and quickly reperfuse this important source of spinal cord blood supply. SEPs disappeared initiallyin both patients (total duration of SEP loss 30 and 13 minutes) but quickly returned after intercostal reperfusion. Both patients remained neurologically intact postoperatively. They are the only two patients in this entire series in whom positive identification of isolated groups of "critical intercostals" were made. "Fade-out" of SEPs associated with inadequate distal perfusion (type IV response). In six patients,

Volume 94 Number 2

Monitoring of SEPs, IV

AU9ust 1987

28 1

Table m. Relationship of variables to incidence of postoperative paraplegia SEP response type

Time of SEP loss post-AXe (min)

Duration of SEP loss

I I I I I I I

8 9 22 20 19 22 29 7

21 28 36 3 4 62 95 110

I II II II II II II II II II II II II II II II II II III III

IV IV IV IV IV IV

Incidence of postop. paraplegia Variable

%

I No/total

Extent of aneurysm 10.0 a. Above T, 17.4 b. T. to bifurcation Use of shunt/bypass 8.0 a. Yes 37.5 b. No Distal perfusion pressure after AXC 0 a. >60 mm Hg 35.7 b. <60 mm Hg Spinal cord ischemia (SEP loss) after 0 a. No 31.2 b. Yes Duration of spinal cord ischemia 0 a. 30 min or less 71.2 b. >30 min

Significance of difference*

1/10 4/23

NS

2/25 3/8

NS

0/19 5/14 AXC 0/17 5/16

P = 0.01

0/26 5/7

P = 0.001

P = 0.02

Legend: Significance of differences calculated by Fisher's exact test. AXe, Aortic cross-clamp. SEP, Somatosensory evoked potentials. .p value for incidence of paraplegia, a versus b, for each variable.

10 20 25 36 25 51 45 18

30 13

5 49 43 II 15 51

distal shunt/bypass techniques were employed after proximal and distal aortic cross-clamping, but adequate distal aortic perfusion pressure (>60 mm Hg) was not maintained because of technical problems. Distal extent of aneurysm varied from T 1 to T 12 and duration of aortic cross-clamptime ranged from 30 to 85 minutes. In these patients shunt or bypass techniques were limited by the presence of either extensive distal disease, which prevented the maintenance of adequate retrograde perfusion distal to the lower aortic cross-clamp, or overzealous use of nitroprusside (Nipride), which resulted in distal hypotension despite measured distal flows of 3 to 5 L/min. A key observation in this group of patients was that SEP tracings gradually decayed or "faded-out" over periods as long as 51 minutes (mean 33.3 ± 13 minutes). This time course of SEP disappearance was

significantly more prolonged than the more rapid disappearance of SEPs seen after sudden interruption of identifiable critical intercostals or interruption of flow from a simple clamp/repair approach to resection (p = < 0.05, type IV versus types I and III) (Table 11). Restoration of distal aortic perfusion with a pressure greater than 70 mm Hg after unclamping resulted in prompt return of SEPs and a normal postoperative neurologic status in three patients. The duration of total SEP loss after the "fade-out phenomenon" in these patients was relatively brief (5, 11, and 15 minutes). Duration of SEP loss was more prolonged (43, 49, and 51 minutes) in three other patients, of whom two had permanent postoperative paraplegia.

Relationship of variables to incidence of postoperative paraplegia (Table III). Five variables were analyzed to determine their association with postoperative paraplegia. The relationship of distal extent of aneurysm was closely scrutinized as a contributing factor to postoperative neurologic injury. Of the five cases of paraplegia, two involved patients with extensive aneurysms extending to the aortoiliac bifurcation and the remaining three occurred at or above the level of the diaphragm. No statistically significant difference in incidence of postoperative paraplegia was found when upper thoracic aneurysms (above T s) were compared to more extensive thoracoabdominal aneurysms (T 9 to bifurcation).

2 8 2 Cunningham, Laschinger, Spencer

The failure to use a Gott shunt or peripheral perfusion device per se was not found to be a statistical predictor of subsequent neurologic injury. Although the 37.5% occurrence of paraplegia seen without employment of such devices did not reach statistical significance, there was a strong suggestion that adequacy of such techniques was quite important. Maintenance of adequate distal perfusion pressure (>60 nun Hg) after crossclamping by either "self-shunting" or shunt/bypass techniques was associated with a 0% (0/19) incidence of paraplegia, a statistically significant observation when compared to the higher incidence of paraplegia in inadequately perfused patients (35.7%, 5/14; p = 0.01). The loss of SEPs during the cross-clamp interval was a significant predictor of subsequent paraplegia. Analysis of these data revealed that duration of such ischemia during the cross-clamp interval was the most sensitive predictor of postoperative neurologic status. Paraplegia did not occur in any patient in whom SEP loss was limited in duration to 30 minutes or less. In contrast, a high incidence of permanent neurologic injury (71%, 5/7 patients) occurred after operation when SEP disappearance exceeded 30 minutes after aortic cross-clamping (p < 0.001). Discussion SEP monitoring: A measure of spinal cord viability. Adams and Van Geertruyden," in 1956, first analyzed paraplegia after surgical attempts to repair lesions of the thoracoabdominal aorta. This important work identified both duration of aortic cross-clamping and variability of spinal cord blood supply as the primary determinants of injury. Numerous authors have suggested additional causes of neurologic injury, but unfortunately these reports are limited by their retrospective nature and the previous absence of technology to accurately determine the intraoperative status of spinal cord blood flow and conduction.t- 28-32 The development and clinical use of SEP monitoring during surgical procedures on the descending thoracic and thoracoabdominal aorta has provided a sensitive means for intraoperative detection of spinal cord ischemia. 23-27, 29 Key observations concerning the temporal occurrence of neurologic injury associated with aortic operations have been made possible by delineation of SEP changes during the operative procedure. For example, the type I SEP response (no distal perfusion after cross-clamping) indicated that SEPs disappeared in 17 ± 8 minutes and disappearance was associated with a significant incidence of paraplegia (71%) if such

The Journal of Thoracic and Cardiovascular Surgery

changes were not reversed within 30 minutes. This correlates well with previous empiric clinical observations that the incidence of paraplegia is rare if aortic cross-clamp time is less than 30 minutes but increases in frequency with longer ischemic times. Appreciation of these data emphasizes the need to adjust the surgical approach to aortic operations with the intent of limiting total spinal cord ischemic time as indicated by SEP response. What has been wrong with shunts in the past? Historical evidence from previous clinical studies suggests that the mere use of shunt or bypass techniques after proximal aortic cross-clamping does not necessarily prevent paraplegia. However, the absence of paraplegia in this study when patients underwent adequate shunting or distal aortic perfusion (distal aortic pressure> 60 nun Hg, normal SEPs) suggests that regulation of these devices on the basis of SEP monitoring techniques is effective in preventing neurologic injury. The most likely explanation for failure of shunts or distal bypass to eliminate postoperative paraplegia after aortic operations is related to two factors: basic inadequacy of the shunt or perfusion device and inappropriate identification of critical intercostal vessels. Small shunts provide inadequate distal aortic perfusion, which may result in undetected ischemia and subsequent neurologic injury. We have not employed heparinized aortofemoral shunts for years because of the unpredictable capabilities of these devices caused by factors such as tubing size, proximal aortic driving pressure, and kinking. Similarly, the use of left ventricular-aortic shunts is suboptimal and hampered by the abnormal physiologic effect produced by regurgitant flow. Optimal distal aortic perfusion pressure and flow are best achieved with a direct perfusion system. This may be accomplished through the appropriate employment of heparinless left atrial-femoral artery bypass or standard oxygenator femoral-femoral bypass. Regardless of the circumstances of distal aortic perfusion, appropriate flow and pressure must be achieved to prevent ischemia. Our previous experience clearly indicates that requirements for "optimal" distal perfusion vary greatly from patient to patient and the level of appropriate retrograde perfusion can be determined only by on-line monitoring of SEPs and distal aortic pressure. Interruption or exclusion of critical intercostal segments is then an identifiable event signaled by sudden loss of SEPs after placement of a distal aortic cross-clamp. Importance of controlled distal vasodilatation during aortic cross-clamping. Theoretically, after aortic

Volume 94 Number 2 August 1987

cross-clamping much of the blood supply to the spinal cord must come through the collateral circulation. Because blood flowing through the collateral circulation will inevitably have a higher vascular resistance, the use of higher perfusion pressures as can be provided by appropriate bypass techniques will improve delivery of spinal cord blood supply. As mentioned, this may necessitate a large perfusion cannula (32 Fr.) and high flow rates. Care should be taken to avoid dilatation of splanchnic and collateral vessels below the aortic crossclamp. Because most cardiac anesthesiologists are trained to use nitroprusside to prevent proximal hypertension, this drug is commonly used after aortic crossclamping for aortic operations. However, in all probability, once nitroprusside is started it may be impossible to maintain an adequate distal pressure to provide appropriate collateral flow to the spinal cord, even when distal perfusion devices are employed. Theoretically, nitroprusside should not be needed at all if the shunt is adequate, and one should simply perfuse enough blood through the shunt or the bypass to keep the proximal pressure normal and provide adequate distal flow. The idea of a "nitroprusside steal" is more than theoretical and may be responsible for the inordinately high flow rates required to maintain normal pressures and SEPs in some patients in our series and the failure of high flow rates to maintain adequate pressure in others. Certainly such alterations in regional blood flow have been reported by others and are reflected by diminution in critical blood supply to the spinal cord, renal cortex, and other areas." Although there probably are situations in which nitroprusside must be used after aortic cross-clamping, the most ideal situation is simply to provide adequate distal flow and use no dilators whatsoever. Single artery of Adamkiewicz: Myth or fact? The long-standing theory that a single critical artery or pair of intercostals exists and must be preserved or reimplanted to prevent paraplegia will likely be more thoroughly evaluated with further experience and use of constant SEP monitoring. It is unclear from historical and present data exactly where critical pairs of intercostal vessels are anatomically located, and it is unlikely that a single artery is critical to overall spinal cord blood supply. It is generally recognized that the incidence of paraplegia is highest after thoracoabdominal aneurysm repair and interruption of intercostals between the diaphragm and renal arteries. Nonetheless, sporadic reports of neurologic injury record a 3% to 8% incidence of paraplegia after repair of upper thoracic aneurysms alone. Therefore, the variable incidence of paraplegia

Monitoring of SEPs, IV

28 3

relative to location and extensiveness of aneurysm resection may be a function of a combination of factors including anatomical location of single or multiple critical intercostals as well as individual surgeon technique. It is likely that paraplegia after aortic operations occurs more frequently as increasing numbers of collaterals are excluded. This situation might result in a cumulative increase in resistance to spinal cord blood flow and diminished collateral supply. In this situation direct reimplantation of any large intercostal or cluster of intercostals, as suggested by Crawford and associates,' might be sufficient to improve regional blood supply. Paradoxically, however, prolongation of ischemic time associated with endeavors to reimplant large numbers of intercostals might increase the frequency of paraplegia. It seems more reasonable during repair of extensive aneurysm that the surgeon simply plan to reestablish flow to one or more major groups of intercostals within 30 to 40 minutes after the aorta is cross-clamped rather than postponing reestablishment of flow until all anastomoses are done. This concept will be tested with further experience but can only be evaluated appropriately if SEP monitoring is used. Conclusions The final common pathway to postoperative paraplegia is spinal cord ischemia, an event recognized by SEP monitoring. Once ischemia occurs as the result of any cause, the patient is at risk for postoperative paraplegia. The duration of this ischemic state is the single most important factor affecting the subsequent incidence of paraplegia. Ischemic intervals associated with less than 30 minutes of SEP loss were not observed to result in injury in this study, whereas longer periods were associated with a 71% incidence of paraplegia. Use of devices to maintain adequate distal flow and perfusion pressure prevents spinal cord ischemia and paraplegia unless critical intercostal arteries originate from within the excluded aortic segment. Exclusion of critical groups of intercostals can be appropriately identified by sudden loss of SEP responses after distal aortic cross-clamping when adequate distal perfusion is maintained. In such situations, spinal cord ischemic time can be minimized by early reimplantation of clusters of intercostals located in a "patch" of aorta. Perhaps other modalities of protection, such as direct perfusion of critical intercostals during cross-clamping, may prove to be important in preventing neurologic injury. It should be expected that unfavorable anatomy (e.g., aortoiliac disease, extensive thoracoabdominal aneurysm, or distal

The Journal of Thoracic and Cardiovascular Surgery

2 8 4 Cunningham, Laschinger, Spencer

aortic dissection) often results in an inability to provide adequate distal perfusion after proximal aortic crossclamping. In this situation SEP monitoring may not be helpful, and the incidence of paraplegia will vary with the length of total spinal cord ischemia occurring before reperfusion of blindly reimplanted intercostal arteries is accomplished. Certainly, large prospective studies using clinical monitoring of evoked spinal cord potentials during surgical procedures on the thoracoabdominal aorta are needed to elucidate completely the many unknown factors in production and prevention of paraplegia.

13.

14.

15.

16.

REFERENCES 1. Bahnson HT. Definitive treatment of saccular aneurysms of the aorta with excision of the sac and aortic suture. Surg Gynecol Obstet 1953;96:383-402. 2. DeBakey M, Cooley DA. Successful resection of aneurysms of the thoracic aorta and replacement by graft. JAMA 1953;152:673-6. 3. Crawford ES, Fenstermacher JM, Richardson W, Sandiford F. Reappraisal of adjuncts to avoid ischemia in the treatment of thoracic aortic aneurysms. Surgery 1970; 67:182-96. 4. Crawford ES, Waler HSJ III, Saleh SA, Normann NA. Graft replacement of aneurysm in descending thoracic aorta: results without bypass or shunting. Surgery 1981; 89:73-85. 5. Najafi H, Javid H, Hunter J, Serry C, Monson D. Descending aortic aneurysmectomy without adjuncts to avoid ischemia. Ann Thorac Surg 1980;30:326-35. 6. DeBakey ME, McCollum CH, Graham JM. Surgical treatment of aneurysms of the descending thoracic aorta: long-term results in 500 patients. J Cardiovasc Surg 1978; 19:571-6. 7. Wolfe WG, Kleinman LH, Wechsler AS, Sabiston DC Jr. Heparin-coated shunts for lesions of the descending thoracic aorta. Arch Surg 1977;112:1481-7. 8. Lawrence GH, Hessel EA, Sauvage LR, Krause AH. Results of use of the TDMAC-heparin shunt in surgery of aneurysms of the descending thoracic aorta. J THORAC CARDIOVASC SURG 1977;73:393-8. 9. Donahoo JS, Brawley RK, Gott VL. The heparin-coated vascular shunt for thoracic aortic and great vessel procedures: a ten-year experience. Ann Thorac Surg 1977; 23:507-13. 10. Connors JP, Ferguson TB, Roper CL, Weldon CS. The use of the TDMAC-heparin shunt in replacement of the descending thoracic aorta. Ann Surg 1975;181:735-41. 11. Frantz PT, Murray GF, Shallal JA, Lucas CL. Clinical and experimental evaluation of left ventriculoiliac shunt bypass during repair of lesions of the descending thoracic aorta. Ann Thorac Surg 1981;31:551-7. 12. DeMeester TR, Cameron JL, Gott VL. Repair of a

17.

18.

19.

20.

21. 22.

23.

24.

25.

26.

27.

through-and-through gunshot wound of the aortic arch using a heparinized shunt. Ann Thorac Surg 1973;16:1938. Murray GF, Young WG. Thoracic aneurysmectomy utilizing direct left ventriculofemoral shunt (TDMACheparin) bypass. Ann Thorac Surg 1976;21:26-9. Cukingnan RA, Fee HJ, Carey JS. Repair of lesions of the descending thoracic aorta with the TDMAC-heparin shunt. J THORAC CARDIOVASC SURG 1978;75:227-31. May lA, Ecker RR, Iverson LIG. Heparinless femoral venoarteriai bypass without an oxygenator for surgery on the descending thoracic aorta. J THORAC CARDIOVASC SURG 1977;73:387-92. Hilgenberg AD, Rainer WG, Sadler TR. Aneurysm of the descending thoracic aorta: replacement with the use of shunt or bypass. J THORAC CARDIOYASC SURG 1981; 81:818-24. Connolly JE, Wakabayashi A, German JC, Stemmer EA, Serres EJ. Clinical experience with pulsatile left heart bypass without anticoagulation for thoracic aneurysms. J THORAC CARDIOVASC SURG 1971;62:568-76. Wakabayashi A, Connolly JE. Prevention of paraplegia associated with resection of extensive thoracic aneurysms. Arch Surg 1976;111:1186-9. Wakabayashi A, Connolly JE. Heparinless left heart bypass for resection of thoracic aortic aneurysms. Am J Surg 1975;130:212-8. Reul GJ, Cooley DA, Hallman GL, Reddy SB, Kuger ER III, Wukasch De. Dissecting aneurysms of the descending aorta: improved surgical results in 91 patients. Arch Surg 1975;110:632-40. Adams HD, Van Geertruyden HH. Neurologic complications of aortic surgery. Ann Surg 1956;144:574-610. Culliford AT, Ayvaliotis B, Shemin R, Colvin SB, lsom OW, Spencer FC. Aneurysms of the descending aorta. J THORAC CARDIOVASC SURG 1983;85:98-104. Cunningham IN Jr, Laschinger JC, Merkin HA, et al. Measurement of spinal cord ischemia during operations upon the thoracic aorta. Ann Surg 1982;196:285-96. Laschinger JC, Cunningham IN Jr, Catinella FC, Nathan 1M, Knopp EA, Spencer Fe. Detection and prevention of intraoperative spinal cord ischemia after cross-clamping of the thoracic aorta: use of somatosensory evoked potentials. Surgery 1982;92:1109-17. Laschnger JC, Cunningham IN Jr, Nathan 1M, Knopp EA, Cooper MM, Spencer FC. Experimental and clinical assessment of the adequacy of partial bypass in the maintenance of spinal cord blood flow during operations on the thoracic aorta. Ann Thorac Surg 1983;36:41726. Laschinger JC, Cunningham IN Jr, Nathan 1M, Krieger K, lsom OW, Spencer Fe. Intraoperative identification of vessels critical to spinal cord blood supply: use of somatosensory evoked potentials. Curr Surg 1984;41:107-9. Laschinger JC, Cunningham IN Jr, Isom OW, Nathan 1M, Spencer Fe. Definition of the safe lower limits of

Volume 94 Number 2 August 1987

28.

29.

30.

31.

aortic resection during surgical procedures on the thoracoabdominal aorta. J Am Coli Cardiol 1983;2:959-65. Brewer LA, Fosburg RG, Mulder GA, Verska JJ. Spinal cord complications following surgery for coarctation of the aorta. J THoRAc CARDIOVASC SURG 1966;64:368-81. Coles JG, Wilson GJ, Sima AF, et al. Intraoperative management of thoracic aortic aneurysm. J THORAC CARDIOVASC SURG 1983;85:292-9. Spencer FC, Zimmerman JM. The influence of ligation of intercostal arteries on paraplegia in dogs. Surg Forum 1958;9:340-2. Nylander WA Jr, Plunkett RJ, Hammon JW Jr, Oldfield

Monitoring of SEPs, IV

28 5

EH, Menacham WF. Thiopental modification of ischemic spinal cord injury in the dog. Ann Thorac Surg 1982;33:64-8. 32. Blaisdell FW, Cooley DA. The mechanism of paraplegia after temporary thoracic aortic occlusion and its relationship to spinal fluid pressure. Surgery 1962;51:351-5. 33. Gelman S, Reves JG, Fowler K, Samuelson PN, Lell WA, Smith LR. Regional blood flow during crossclamping of the thoracic aorta and infusion of sodium nitroprusside. J THoRAc CARDIOVASC SURG 1983;85:28791.

Bound volumes available to subscribers Bound volumes of THE JOURNAL OFTHORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1987 issues from the Publisher, at a cost of $49.00 ($66.00 international) for Vol. 93 (January-June) and Vol. 94 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact The C. V. Mosby Company, Circulation Department, 11830 Westline Industrial Drive, St. Louis, Missouri 63146, USA; phone (800) 325-4177, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.