Cervical Spinal Cord Injury and the Need for Cardiovascular Intervention

PAPER Cervical Spinal Cord Injury and the Need for Cardiovascular Intervention

John F. Bilello, MD; James W. Davis, MD; Mark A. Cunningham, MD; Tammi F. Groom, PA-C; Debbie Lemaster, BSN; Lawrence P. Sue, MD

Hypothesis: The level of cervical spinal cord injury Results: Eighty-three patients met the criteria for CSCI (CSCI) can be used to predict the need for a cardiovas- and quadriplegia, 62 in the high (C1-C5) and 21 in the cular intervention. low (C6-C7) level. There was no significant difference between the 2 groups in mean±SD age (38.2±17.8 vs Design: Retrospective review. Data included level of spi- 34.7±15.6 years; P=.43), mean±SD Injury Severity Score nal cord injury, Injury Severity Score, lowest heart rate, (35.7±17.5 vs 32.5±11.2; P=.44), mean±SD admission and systolic blood pressure in the first 24 hours and in- base deficit (−0.7±3.6 vs 0.7±2.7; P=.06), or mortality tensive care unit course. The level of CSCI was divided (12 [19%] of 62 patients vs 2 [10%] of 21 patients; P=.29). into high (cord level C1-C5) or low (cord level C6-C7). Neurogenic shock was present in 19 (31%) of the 62 pa- Neurogenic shock was defined as bradycardia with hy- tients with high CSCI and in 5 (24%) of the 21 patients potension. Statistical analysis was performed with the with low CSCI (P=.56). There was a marked difference t test and the ␹2 test. in the use of a cardiovascular intervention between those with a high and those with a low CSCI: 15 (24%) of 62 Setting: Level I trauma center. patients vs 1 (5%) of 21 patients (P=.02). Two patients with C1 through C5 spinal cord injuries required car- Patients: The patients studied were those with quad- diac pacemakers. riplegia who experienced a CSCI and were admitted to the hospital between December 1, 1993, and October 31, Conclusions: There was no significant difference in the 2001. frequency of neurogenic shock by injury level. Patients with a high CSCI (C1-C5) had a significantly greater re- Interventions: Pressors, chronotropic agents, and pace- quirement for a cardiovascular intervention compared makers. with patients with lower injuries (C6-C7).

Main Outcome Measure: Use of a cardiovascular intervention in the presence of neurogenic shock. Arch Surg. 2003;138:1127-1129

ERVICAL SPINAL cord in- heart itself.4 Pharmacologic and electri- jury (CSCI) is associated cal interventions (ie, pacemakers) may be with dysfunction of the necessary in these patients if fluid resus- sympathetic nervous sys- citation alone cannot maintain adequate tem and cardiovascular tissue perfusion. Failure to adequately treat deficits,C including severe bradycardia, asys- neurogenic shock may result in further is- tole, and loss of peripheral vascular tone. chemic injury of an already compro- These effects have been replicated in ani- mised nervous system (secondary injury).4 mal studies.1-3 Direct myocardial injury and Aggressive treatment with fluids and pres- a decrease in contractility (dp/dt [ie, change sors, and appropriate invasive monitor- in pressure per change in time]) may also ing (arterial and central venous access), is contribute to hypotension.1,3 paramount.5 Patients with a CSCI may develop To our knowledge, no study has spe- neurogenic shock, characterized by brady- cifically examined whether the level of cardia and hypotension from decreased pe- CSCI can be used to determine which pa- From the Department of ripheral vascular resistance and cardiac tients are most susceptible to neurogenic Surgery, University Medical output. These changes are related to in- shock and the requirement for cardiovas- Center, University of California creased vagal tone, decreased sympa- cular interventions (CVIs), namely, the ad- San Francisco–Fresno Campus. thetic input, and possibly changes in the ministration of pressors, chronotropic

(REPRINTED) ARCH SURG/ VOL 138, OCT 2003 WWW.ARCHSURG.COM 1127

©2003 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 Results are expressed as mean±SD. Analysis was per- Relationship of CSCI Level to Age, ISS, formed by ␹2 test and paired t test. PϽ.05 was considered sta- Admission Base Deficit, and Mortality* tistically significant.

CSCI Level RESULTS Cord Level Cord Level C1-C5 C6-C7 Variable (n = 62) (n = 21) P Value Eighty-three patients met the inclusion criteria for CSCI and quadriplegia during the study period, with 62 in the Age, y 38.2 ± 17.8 34.7 ± 15.6 .43 ISS 35.7 ± 17.5 32.5 ± 11.2 .44 high injury (C1-C5) category and 21 in the low injury Admission base deficit −0.7 ± 3.6 0.7 ± 2.7 .06 (C6-C7) category. There was no significant difference in Mortality, No. (%) 12 (19) 2 (10) .29 age, Injury Severity Score, admission base deficit, or mortality between the groups (Table). Abbreviations: CSCI, cervical spinal cord injury; ISS, Injury Severity Score. There was no significant difference in the presence *Data are given as mean ± SD unless otherwise indicated. of neurogenic shock between the C1 through C5 group and the C6 through C7 group (19 [31%] of 62 patients agents, and/or cardiac pacing. This study was per- vs 5 [24%] of 21 patients; P=.56). However, there was a formed to test the hypothesis that the level of CSCI can significant difference between the 2 groups in the need be used to predict the likelihood of requiring a CVI. for a CVI: 15 (24%) of the C1 through C5 group required an intervention for neurogenic shock, whereas only METHODS 1 (5%) of the C6 through C7 group needed an intervention (P=.02). In addition, the only 2 patients requiring The records of quadriplegic patients who experienced a CSCI at permanent pacemakers were in the higher cervical level University Medical Center, a level I trauma center, between De- injury group. None of the patients with C6 through C7 cember 1, 1993, and October 31, 2001, were reviewed. Level of spinal cord injury, age, sex, Injury Severity Score,6 admission base injury required a pacemaker. deficit,7-9 mechanism (blunt vs penetrating), associated inju- Dobutamine was used in 2 patients. One patient was ries, hemorrhage, previous medical conditions, and in-hospital a 69-year-old woman with a history of congestive heart mortality were noted. The lowest heart rate and systolic blood failure, and the other was a 90-year-old man with a low pressure in the first 24 hours after injury and during the remain- cardiac index. Both patients met invasive criteria for ino- ing intensive care unit course were recorded. All resuscitations tropic support, and both also required a vasoconstric- followed Advanced Trauma Life Support guidelines.10 Initial hy- tive agent (dopamine or phenylephrine hydrochloride potension was treated with crystalloid, followed by packed red [Neo-Synephrine]). blood cells as appropriate and control of hemorrhage. Resusci- tation was guided by response to therapy and the end points of systolic blood pressure of 90 mm Hg or higher, no further need COMMENT for transfusion, and correction of base deficit. Pressors and chronotropic agents were used only when hypotension was refrac- Cervical spinal cord injury can cause significant changes tory to intravenous fluid administration, and were titrated to keep in pulse, blood pressure, cardiac output, and rhythm. the systolic blood pressure at 90 mm Hg or higher. All patients who began treatment with pressors and chronotropic agents un- Within seconds to minutes after injury, there is a sys- derwent placement of central venous and/or pulmonary artery temic pressor response with a widened pulse pressure seen catheters and invasive arterial monitoring. in animals and humans, resulting from short-term out- The level of CSCI was divided into 2 categories: high (cord flow of sympathetic activity and adrenal hormones. This level C1-C5) and low (cord level C6-C7). C5/C6 was chosen pressor response quickly abates, and is followed by neu- as the level of division between the 2 categories because of in- rogenic shock with bradycardia and hypotension.11,12 nervation of the phrenic nerve, which terminates at the C5 level. Preganglionic sympathetic nerve fibers exit the spinal cord Previous researchers11,12 have used this level as the line of de- at the first through fourth thoracic level. Parasympa- marcation because of the respiratory pathophysiological fea- thetic outflow, however, is carried by the vagus nerve, tures that may contribute to neurogenic shock. Bradycardia was which originates in the medulla. Therefore, a CSCI may defined as a heart rate of less than 50 beats/min. Hypotension was defined as a systolic blood pressure lower than 90 mm Hg. completely transect spinal cardiac and vasomotor sym- Neurogenic shock was defined as the presence of simulta- pathetic fibers from above while the parasympathetic fi- 11,12,16-18 neous bradycardia and hypotension. The use of pressors (dopam- bers remain intact. We chose C5/C6 as the bor- ine hydrochloride, dobutamine hydrochloride, epinephrine hy- der between high and low cervical injuries because phrenic drochloride, norepinephrine bitartrate, and phenylephrine nerve innervation terminates at C5. Piepmeier11 and hydrochloride), chronotropic agents (atropine sulfate), and car- Lehmann12 and colleagues invoke hypoxia as a culprit diac pacing was recorded. Use of 1 or more of these treatments in the exacerbation of bradycardia and cardiac arrest in constituted a CVI. Patients with a CSCI from blunt trauma were patients with cervical cord injuries. Piepmeier et al fur- 13 given intravenous methylprednisolone sodium succinate. Pa- ther explain that hypoxic patients with normal cervical tients with a CSCI who experienced penetrating trauma did not cord function will increase their heart rate with the sym- receive corticosteroids.14,15 The Injury Severity Score is an expression of anatomic in- pathetic outflow that normally accompanies hypoxic- jury developed for comparison of trauma care that correlates induced tachypnea (pulmonary inflation reflex). This res- with outcome.6 The admission base deficit has been shown to piratory-related sympathetic outflow, which normally be an excellent index of shock, perfusion, and outcome for pa- overrides vagal parasympathetic effects on the heart, seems tients who experience trauma.7-9 to be blocked by cervical cord injuries that compromise

(REPRINTED) ARCH SURG/ VOL 138, OCT 2003 WWW.ARCHSURG.COM 1128

©2003 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 ventilation (ie, at C5 and above). Lehmann et al explain age or concomitant injuries. This difference may affect this phenomenon as well in ventilator-dependent pa- allocation of intensive care unit resources between the 2 tients undergoing endotracheal suctioning. The mechani- groups. cal respiratory embarrassment that accompanies high (C1- C5) injuries may also contribute to the specific Accepted for publication May 3, 2003. cardiovascular complications associated with those in- This paper was presented as a poster at the 74th An- juries. nual Meeting of the Pacific Coast Surgical Association; Feb- Lehmann et al12 reviewed the presence of bradycar- ruary 16, 2003; Monterey, Calif; and is published after peer dia, hypotension, and cardiac arrest and the need for at- review and revision. ropine and pressor therapy by severity of CSCI, but not Corresponding author: John F. Bilello, MD, Depart- by level within the cervical spine. These researchers found ment of Surgery, University Medical Center, University of that 71% of patients with a severe CSCI who had Frankel California San Francisco, 445 S Cedar Ave, Fresno, CA scores of A (complete loss of neurologic function below 93702 (e-mail: [email protected]). the level of injury) and B (no motor function but sen- sory function below the injury level) experienced brady- REFERENCES cardia, with a heart rate lower than 45 beats/min. In addition, 68% of the patients with a severe CSCI were hypotensive and 35% required pressor therapy. This is 1. Tibbs PA, Young B, McAllister RG, Brooks WH, Tackett L. Studies of experimental cervical spinal cord transection, part I: hemodynamic changes after acute cer- comparable to 29% (24/83) of all our patients with a CSCI vical spinal cord transection. J Neurosurg. 1978;49:558-562. who required some sort of a CVI, including chronotro- 2. Maiorov DN, Fehlings MG, Krassioukov AV. Relationship between severity of spi- pic agents. nal cord injury and abnormalities in neurogenic cardiovascular control in con- Piepmeier et al11 also examined CSCI by Frankel scious rats. J Neurotrauma. 1998;15:365-374. 3. Guha AB, Tator CH. Acute cardiovascular effects of experimental spinal cord in- score severity, and placed a strong emphasis on brady- jury. J Trauma. 1988;28:481-490. cardic complications. Patients in that study were strati- 4. Dumont RJ, Okonkwo DO, Verma S, et al. Acute spinal cord injury, part I: patho- fied by level of cervical cord injury, with C1 through C4 physiologic mechanisms. Clin Neuropharmacol. 2001;24:254-264. regarded as high and C5 through C7 as low. They found 5. Dumont RJ, Verma S, Okonkwo DO, et al. Acute spinal cord injury, part II: con- no difference between groups with regard to bradycar- temporary pharmacotherapy. Clin Neuropharmacol. 2001;24:265-279. 6. Baker SP, O’Neil B, Hadden W Jr, et al. The Injury Severity Score: a method for dia, cardiac arrest, and the need for pressors. In our study, describing patients with multiple injuries and evaluating emergency care. J Trauma. a significant difference was found between the high and 1974;14:187-196. low injury groups and the need for pressors. The differ- 7. Rutherford EJ, Morris JA Jr, Reed GW, Hall KS. Base deficit stratifies mortality ence in findings may be explained by the difference in and determines therapy. J Trauma. 1992;33:417-423. 8. Davis JW, Parks SN, Kaups KL, et al. Admission base deficit predicts transfu- definition of high vs low. Piepmeier et al did not in- sion requirements and risk of complications. J Trauma. 1996;41:769-774. clude the C5 level in the high group, which could ac- 9. Davis JW, Kaups KL, Parks SN. Base deficit is superior to pH in evaluating clear- count for this difference. In both previously mentioned ance of acidosis after traumatic shock. J Trauma. 1998;44:114-118. studies,11,12 bradycardia and hypotension were exam- 10. Advanced Trauma Life Support. 6th ed. Chicago, Ill: Committee on Trauma, Ameri- ined separately, not together as one entity of neuro- can College of Surgeons; 1997. 11. Piepmeier JM, Lehman KB, Lane JG. Cardiovascular instability following acute genic shock. Both studies showed that severity of CSCI cervical spinal cord trauma. Cent Nerv Syst Trauma. 1985;2:153-160. correlates with cardiovascular complications, especially 12. Lehmann KG, Lane JG, Piepmeier JM, Batsford WP. Cardiovascular abnormali- bradycardia. Our study showed that the level of CSCI with ties accompanying acute spinal cord injury in humans: incidence, time course quadriplegia could define a subgroup of patients who are and severity. J Am Coll Cardiol. 1987;10:46-52. more likely to require CVIs. There was no significant dif- 13. Bracken MB, Shepard MJ, Collins WF, et al. A randomized trial of methylprednisolone or naloxone in the treatment of acute spinal cord injury: results of the ference in age, Injury Severity Score, admission base defi- Second National Acute Spinal Cord Injury Study (NASCIS II). N Engl J Med. 1990; cit, and mortality between the high and low CSCI groups 322:1405-1411. in our study. The 2 groups differed only in the need for 14. Levy ML, Gans W, Wijesinghe HS, et al. Use of methylprednisolone as an ad- a CVI. junct in the management of patients with penetrating spinal cord injury: out- There was a trend toward a higher mortality in the come analysis. Neurosurgery. 1996;39:1141-1149. 15. Prendergast MR, Saxe JM, Ledgerwood AM, et al. Massive steroids do not re- higher CSCI group (12 [19%] of 62 patients vs 2 [10%] duce the zone of injury after penetrating spinal cord injury. J Trauma. 1994;37: of 21 patients). This is consistent with previous stud- 576-579. ies19,20 in which persons with injuries above C5 have had 16. Volker KH, Sonntag VK, Douglas RA. Management of spinal cord trauma. Neu- a higher mortality. rosurg Clin N Am. 1990;1:729-750. 17. Kawamoto M, Sakimura S, Takasaki M. Transient increase of parasympathetic In conclusion, although there was no significant dif- tone in patients with cervical spinal cord trauma. Anaesth Intensive Care. 1993; ference in the frequency of neurogenic shock (bradycar- 21:218-221. dia with hypotension), patients with a high CSCI (C1- 18. Teasell RW, Arnold MO, Krassioukov A, Delaney GA. Cardiovascular conse- C5) had a significantly greater requirement for CVIs, quences of loss of supraspinal control of the sympathetic nervous system after including pacemakers, compared with patients with in- spinal cord injury. Arch Phys Med Rehabil. 2000;81:506-516. 19. Messard L, Carmody A, Mannarino E, Ruge D. Survival after spinal cord trauma: juries at a lower level (C6-C7). Patients with C1 through a life table analysis. Arch Neurol. 1978;35:78-83. C5 CSCIs require more intensive monitoring after in- 20. Sneddon DG, Bedbrook G. Survival following traumatic tetraplegia. Paraplegia. jury and are more likely to require a CVI regardless of 1982;20:201-207.

(REPRINTED) ARCH SURG/ VOL 138, OCT 2003 WWW.ARCHSURG.COM 1129