The Impact of Mean Arterial Pressure on Functional Outcome Post-Acute Spinal Cord Injury: A Scoping Systematic Review of Animal Models

Abstract

The aim of this work was to perform a scoping systematic review on the animal literature surrounding mean arterial blood pressure (MAP) and functional outcomes post-acute spinal cord injury (ASCI). We performed a systematic review of the literature by searching: MEDLINE, BIOSIS, EMBASE, Global Health, SCOPUS, and Cochrane Library from inception to January 2015. We also performed a hand search of various published meeting proceedings. Through a two-step review process, using two independent reviewers, we selected articles for the final review based on pre-defined inclusion/exclusion criteria. Ten studies were included within the final systematic review. A variety of animal models were used within these studies. All included studies had some objective means of documenting functional outcome post-manipulation of the MAP. Four studies could be considered to be “positive studies,” showing some neurological improvement or beneficial effect to having the blood pressure manipulated. Two studies displayed worse functional outcomes secondary to episodes of hypotension. Four studies failed to demonstrate a relationship between MAP and functional outcome within the animal models. This review concludes that, within the animal literature, there is insufficient evidence to draw a conclusion about the effect of MAP on neurological outcome in animal models of ASCI.

Introduction

The management of acute spinal cord injury (ASCI) has long been a controversial topic. Experimental evidence suggests that after the initial episode of cord injury and compression, there is an active secondary phase of injury.^1,2 Secondary injury mechanisms include vascular changes, such as reduction in blood flow, loss of autoregulation, neurogenic shock, hemorrhage, loss of microcirculation, vasospasm, and thrombosis. Other secondary injury mechanisms include biochemical changes leading to necrosis, electrolyte shifts, edema, and loss of energy metabolism. There is some experimental evidence to suggest that there might be a critical period immediately post-injury when some of the secondary mechanisms of injury may be countered.^3
–5

In order to counter the decreased blood flow observed in ASCI, current therapies aim to increase the blood pressure (BP) immediately post-injury. There is some evidence to suggest that an increase in BP leads to significant improvement in axonal function both in the motor and somatosensory tracts of the cord.⁶ The evidence, however, is insufficient to definitively recommend universally increasing the mean arterial pressure (MAP) in patients presenting with ASCI. Only Class III evidence exists that MAP-directed therapy, goal 85–90 mm Hg, in patients with ASCI improves neurological outcome.⁷ Strong scientific evidence is lacking with respect to the actual target MAP that should be achieved in ASCI patients as well as the length of time the MAP should be targeted post-injury. Current American Association of Neurological Surgeons (AANS) guidelines⁷ claim that there is insufficient evidence to support treatment guidelines. The “options” given include maintaining a MAP of 85–90 for first 7 days post-ASCI “to improve spinal cord perfusion.”

As is often the case, data from animal studies have been extrapolated to humans and have played a role in the development of these recommendations. There is a vast body of literature on animal models of ASCI and its management. Therefore, the first step in performing a comprehensive review of the literature and data on the management of ASCI should begin with a review of the animal study precursors that set the stage for later clinical studies.

The purpose of this scoping systematic review was to look at all the animal studies conducted that looked at hemodynamic parameters in animal models of ASCI and the effect of post-injury MAP on the neurological outcomes.

Methods

A scoping systematic review using the methodology outlined in the Cochrane Handbook for Systematic Reviewers was conducted.⁸ The data are reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).⁹ The review questions and search strategy were decided upon by the primary author (B.S.) and supervisor (N.B.).

Search question, population, inclusion, and exclusion criteria

The question posed for systematic review was: What is the effect of post-injury MAP on functional outcome in animals with acute traumatic spinal cord injury (SCI)? If a clear impact on outcome is identified, our secondary question was: What MAP is optimum post-injury in order to ensure a positive function outcome?

The functional outcome was defined as neurological exam (e.g., incline plane test), motor evoked potentials (MEPs), spinal evoked potentials (SEPs), and dorsal column evoked potentials (DCEPs).

Our inclusion criteria took into consideration the likely heterogeneity of the studies. We only included animal models of traumatic spinal cord injury (SCI). The studies had to address BP/MAP/hemodynamic parameters directly. They had to document neurological outcome/recovery in relation to BP, with neurological outcome measures being defined as physical exam or physiological studies (evoked potentials). All studies included had to have 5 or more experimental subjects.

We excluded all human studies. Studies looking exclusively at neuroprotective agents without documenting hemodynamic parameters were also excluded.

Studies that did not address neurological recovery/functional outcome were not included. Nontraumatic models (e.g., models for ischemia) were not a part of the inclusion criteria.

The primary outcome measure documented is effect of MAP on functional neurological outcomes, as defined by neurological exam (e.g., incline plane test), recovery/change in MEPs, or SEPs. There were no specific secondary outcome measures for our review. Any secondary measures were documented as they appeared in each individual study. Secondary outcomes documented in parent studies included spinal cord blood flow (SCBF) as well as morphometric and histopathological studies. Some studies included in this review looked at the above-mentioned secondary outcomes as their primary outcomes with neurological outcome being documented as a secondary outcome.

Search strategy

MEDLINE, BIOSIS, EMBASE, Global Health, SCOPUS, and Cochrane Library from inception to January 2015 were searched using our pre-conceived list of synonyms for “traumatic spinal cord injury,” “spinal cord perfusion pressure,” and “functional outcome.” The search strategy for MEDLINE can be seen in Supplementary Appendix A (see online supplementary material at http://www.liebertpub.com), with a similar search strategy utilized for the other databases.

Meeting proceedings for the last 10 years were also searched, looking for ongoing and unpublished work based on MAP-directed therapy to maintain spinal cord perfusion pressure (PP) in animal models of traumatic SCI. The meeting proceedings of the following professional societies were searched: Canadian Neurological Sciences Federation (CNSF); AANS, Congress of Neurological Surgeons (CNS); European Neurosurgical Society (ENSS); World Federation of Neurological Surgeons (WFNS); American Neurology Association (ANA); American Academy of Neurology (AAN); European Federation of Neurological Science (EFNS); World Congress of Neurology (WCN); Society of Critical Care Medicine (SCCM); Neurocritical Care Society (NCS); World Federation of Societies of Intensive and Critical Care Medicine (WFSICCM); American Society for Anesthesiologists (ASA); World Federation of Societies of Anesthesiologist (WFSA); Australian Society of Anesthesiologists; International Anesthesia Research Society (IARS); Society of Neuroscience in Anesthesia and Critical Care (SNACC); Society for Neuroscience in Anesthesiology and Critical Care; the Japanese Society of Neuroanesthesia and Critical Care (JSNCC); the North American Spinal Society (NASS); the Canadian Spine Society (CSS); and the Eurospine Society.

Finally, reference lists of any review articles or systematic reviews on spinal cord PP goals in acute traumatic SCI were manually searched for any missed articles.

Study selection

Utilizing two reviewers (B.S. and F.Z.), a two-step review of all articles returned by our search strategies was performed. First, the reviewers independently screened all titles and abstracts of the returned articles to decide whether they meet the inclusion criteria. Second, full text of the chosen articles was then assessed to confirm that they met the inclusion criteria and that they document functional neurological outcome post-MAP-directed therapy. Any discrepancies between the two reviewers were resolved by a third party (N.B.).

Data collection

Data were extracted from the selected articles and stored in an electronic database. Data fields include: species, number of subjects, study design, primary endpoints, trauma model, treatment/manipulation of MAP, duration of BP manipulation, outcome assessment technique, primary outcome, and secondary outcome.

Bias assessment

Two reviewers (B.S. and F.Z.) independently assessed the bias of the individual manuscript included within this review. Bias was assessed by applying the RTI risk assessment questionnaire¹⁰ to each study. Any discrepancies were resolved through discussion, and a third party if needed (N.B.). The results of the bias assessment are summarized in Table 3 and can be seen in more detail in Supplementary Appendix B (see online supplementary material at http://www.liebertpub.com).

Formal grading of the evidence was not conducted for a variety of reasons. First, most evidence grading systems are designed for clinical studies and may or may not be applicable to animal studies. Second, there were various animal species studied in a variety of heterogeneous conditions/methods; thus, applying level of evidence would potentially prove meaningless given these facts. Finally, our goal was not to try and impose “level of evidence” statements on such a heterogeneous group of articles, but to provide a systematically conducted scoping review that highlights all of the scattered literature on the topic of interest.

Statistical analysis

A meta-analysis was not performed in this study because of the heterogeneity of the data within the animal studies.

Results

The above-mentioned strategy yielded a total of 2925 results from the databases and gray literature search. After manually removing the duplicates, we were left with 2789 results. A review of the titles and, when of interest, their abstracts, yielded 13 possible titles of interest. We then manually checked the references of these titles, which yielded another 42 results of interest, bringing the total number to 55. All of these titles were then fully reviewed and 10 studies were selected based on our pre-determined inclusion and exclusion criteria. Figure 1 demonstrates the results of the above-listed search strategy.

FIG. 1.

PRISMA flow diagram of search results. BP, blood pressure.

Summary of evidence

Species

There was a diverse selection of animals in the studies we obtained in our search. Lambs (1), dogs (2), cats (3), rats (2), pigs (1), and rabbits (1) were used.

Primary outcome measures documented

As mentioned above, we set out to find studies that documented some kind of neurological outcome either in the form of physical exam findings or physiological measures of evoked potentials. Predictably, the studies accumulated had diverse methods and techniques for documenting outcome measures.

Spinal evoked potential and dorsal column evoked potential measurement

Of the 10 studies,^{6,11

–19} five of them used somatosensory evoked potential (SSEP) or DCEP as a measure of neurological outcome. Dyste and colleagues¹⁷ determined SEPs and SCBF at 0.5, 1, 1.5, and 2.5 h post-injury. Griffiths and colleagues¹⁹ looked at SCBF and DCEP during subacute cord compression and studied the effect of progressive hypotension during cord compression. As a part of their outcome assessment, Hukuda and colleagues¹⁶ looked at SEPs before, immediately after, and 30 min post-SCI. They repeated measurement of SEPs at weekly intervals for 8 weeks post-injury. Haghighi and colleagues¹³ recorded SEPs at baseline and immediately post-trauma. Subsequent recordings were made every 5 min for up to 2 h. Fehlings and colleagues⁶ concomitantly measured SSEPs and MEPs while measuring SCBF. Recordings were performed immediately after therapeutic agent infusion and at 1 and 2 h after cessation of drug delivery.

Cortical evoked potential measurement

Of the studies obtained, two studies used cortical evoked potentials (CEPs) as a measure of neurological outcome. Hardy and colleagues¹⁵ studied how increasing weight on the spinal cord affected CEPs and how a subsequent increase in MAP affected this conduction block. Similarly, Brodkey and colleagues¹⁴ tried to demonstrate the interaction between the effects of pressure applied to the spinal cord and reduced arterial perfusion of the cord, as shown by effects on CEPs.

Neurological exam

Of the 10 studies in this review,^{6,11

–19} only three studies used a physical neurological exam to document neurological outcome. In addition to measuring SEPs, Hukuda and colleagues¹⁶ did weekly neurological exams for up to 8 weeks. The strength of the dogs was graded on a 1–4 scale. Gambardella and colleagues¹⁸ obtained neurological exams at 1, 24, and 48 h post-injury. They used the Tarlov scale to document their neurological exams. Dolan and colleagues¹⁹ assessed functional recovery on a weekly basis for 8 weeks post-injury using the incline plane technique.

Functional outcome in response to mean arterial blood pressure

Of the 10 studies that our search strategy yielded, four could be considered to be “positive studies”;^14,15,18,19 these were actually studies that showed some neurological improvement or beneficial effect to having the BP manipulated. The studies performed by Griffiths and collegaues,¹⁹ Brodkey and colleagues.¹⁴ and Gambardella and colleagues¹⁸ all showed improved neurological outcome when the BP was manipulated and maintained at a normotensive level. Hardy and colleagues¹⁵ was the only study that showed a benefit of maintaining the systemic BP at a hypertensive level.

The studies performed by Barrios and colleagues¹² and Haghighi and colleagues¹³ showed evidence that hypotension in the immediate post-injury period was detrimental to neurological outcome. These studies did not address hypertensive conditions in the post-injury period, given that the primary goal of these studies was not the impact of BP in the immediate post-injury period.

The remaining four studies^6,11,15,17 yielded by the search strategy can be classified as “negative studies.” These studies demonstrated no benefits of higher MAPs compared to control groups in the immediate post-injury period. The control groups in these studies consisted of hypotensive and normotensive groups as a comparison. Despite having higher MAPs, none of these studies demonstrated a statistically significant difference in neurological recovery.

A detailed account of the study design, endpoints, and functional outcome results can be seen in Tables 1, 2, and 3.

Table 1.

Study and Species Characteristics

Study	Species	No. of subjects	Study design	Primary endpoints
Dyste and colleagues¹⁷	Lambs	39	Anesthetized lambs subjected to epidural cord compression at T13 at 200 mm Hg for 80 min. Nine animals received hetastarch, 8 animals mannitol, and 10 animals received phenylephrine to raise MAP by 20–40%. Twelve control animals received saline. SEPs and SCBF determined at 0.5, 1.5, and 2.5 h after compression. Animals killed at 2.5 h.	Attempt to augment SCBF following cord injury, either with hetastarch, mannitol-induced volume expansion, or phenylephrine induced-hypertension
Griffiths and colleagues¹⁹	Dogs	12	SCBF and dorsal column conduction, assessed by DCEP, were measured during subacute cord compression. Three groups. Group 1: 6 dogs, no manipulation of map. Increase in pressure in steps of 20–30 mm Hg. Group 2: 4 dogs, induced hypotension. Pressure increased in increments of 5–10 mm Hg and recordings made until conduction failure. Group 3: 2 dogs, used to assess effect of progressive hypotension on DCEP. DCEP studied as MAP reduced.	Examine whether conduction defects caused by ischemia or other factors Secondary: To study the inter-relationships of SCBF, PP, and cord conduction during focal compression
Hukuda and colleagues¹⁶	Dogs	36	T5 and T6 laminectomy. Sudden inflation of 400 mm Hg, a tourniquet that encircled the spinal cord epidurally. This inflation pressure maintained for 5 min. SEPs recorded before, immediately after, and 30 min post-trauma. Intermittent hypertension and hypercarbia, which consisted of 15 min of hypertension and 95% O₂/5% CO₂ gas ventilation and 170–200 mm Hg induced by IV norepinephrine. This was alternated with 10 min of normotensive air ventilation for 3 h, beginning 3 h post-injury. Remainder of dogs were controls. Neurological and SEP examinations repeated at weekly intervals up to 8 weeks.	Assess the therapeutic effects of combined hypertension and hypercarbia against acute spinal cord trauma
Hardy and colleagues¹⁵	Cats	7	Abolishing CEP by applying increasing weight to the spinal cord. The effects of raising the BP on the CEPs were then assessed.	To determine the effect of hypertension on conduction block secondary to compression
Brodkey and colleagues¹⁴	Cats	5	Applied weight to spinal cord and also decreased the BP in abdominal aorta first separately, then both forms of trauma at same time. Each mode of trauma was first applied separately for 15–30 min. When it was shown that it alone would not affect CEP, the other mode was added and the time to the block of the CEP was noted. After the block of the CEP was demonstrated, the first mode of trauma was removed and the time required for the CEP to return was recorded. Then, experiment repeated, but in reverse order. In last 2 cats, only one block produced.	To demonstrate interaction between the effects of pressure applied to the cord and of reduced arterial perfusion of the cord, as shown by effects on CEP
Haghighi and colleagues¹³	Cats	10	Spinal evoked responses were abolished after weight drop injury of 100 gm-cm. Five cats were treated with nimodipine and 5 were treated with saline. Both groups received equal volumes of fluid. SEPs were recorded at baseline and immediately post-trauma. Subsequent recordings were made every 5 min up to 2 h.	To evaluate the effect of nimodipine in functional recovery of the spinal cord as measured by SEPs
Gambardella and colleagues¹⁸	Rabbit	80	Five groups, 16 in each group. T12–L1 laminectomy. SCI induced with modified Sugita vascular clip. Group 1: anesthetized and subjected to all procedures, but no SCI and no pharmacological treatment. Group 2: spinal cord compression 2 min and 5 min. Eight in each subgroup. Received normal saline. Group 3: same division as group 2. Received nimodipine (1.5 mcg/kg/min IV for 2 h). MAP maintained near control levels with adrenaline. Group 4: same division and groups 2 and 3. Nimodipine alone, same dose as above for 2 h. Group 5: same division as groups 2 and 3. Adrenaline alone for 2 h, same dose as group 3. To maintain MAP near control. Neuroexam performed at 1, 24, and 48 h after the end of spinal cord compression.	To assess the effect of combination of nimodipine and adrenaline treatment on neurological outcome and morphometric axonal changes following secondary spinal cord damage. Neurological outcome measured using Tarlov scale.
Dolan and colleagues¹¹	Rats	36	Acute spinal cord compression at T1 by a 180-gm clip for 1 min; this produced profound hypotension. Three groups. Untreated animals served as controls. Normotensive group with MAPs kept between 100 and 120 mm Hg, and hypertensive group with MAPs maintained between 125 and 150 mm Hg. MAP raised for 1 h post-injury. Functional recovery assessed weekly for 8 weeks post-operatively by the inclined plane method.	To assess long-term functional value of elevating systemic blood pressure post-SCI. Functional recovery of treated animals compared to that of untreated hypotensive animals.
Barrios and colleagues¹²	Pig	15	Surgical procedures under general anesthesia. Spinal cord and nerve roots exposed from T6 to T10. Three groups, 5 subjects per group. Separation group (cord translation), root stump pull, and torsion groups. An electromechanical external device was used to apply the displacing forces. The three displacement forces were repeated after sectioning the adjacent nerve roots. The experiments were then carried about under induced hypotension.	To determine the limits of cord displacement before disappearance of neurophysiologic signals Influence of type of force, section of the roots, and induced hypotension on cords tolerance also assessed
Fehlings and colleagues⁶	Rats	30	Thirty adult Wistar rats were anesthetized. Laminectomy was made from C6 to T2 and each rat received 1-min extradural clip compression injury of the cord at T1 with modified aneurysm clip, a force of 53 gm. SCBF was measured by hydrogen clearance technique. MEPs and SSEPs were concomitantly recorded with the SCBF. Rats were randomly and blindly assigned to the following groups (5 per group): placebo and saline, placebo and dextran 40, nimodipine 0.02 mg/kg and saline, nimodipine 0.02 mg/kg and dextran 40, nimodipine 0.05 mg/kg and saline, and nimodipine 0.05 mg/kg and dextran 40. Two Harvard infusion pumps were used to deliver a fixed volume of drug at a constant rate over 1 h; post-infusion SCBF and EP recordings were made immediately following the infusion and at 1 and 2 h after cessation of drug delivery.	To assess whether administration of nimodipine and dextran 40, either alone or in combination, could increase post-traumatic SCBF and improve axonal function in the cord after acute SCI

MAP, mean arterial pressure; SCBF, spinal cord blood flow; EP, evoked potential; DCEP, dorsal column evoked potential; SEP, spinal evoked potential; MEP, motor evoked potential; SCI, spinal cord injury; CEP, cortical evoked potential; mg, milligram, kg: kilogram; gm, gram; mcg: microgram; mm Hg, millimeter of mercury; IV, intravenous; BP, blood pressure; PP, perfusion pressure; SSEPs, somatosensory-evoked potentials.

Table 2.

Model Characteristics and Experimental Design

Study	Trauma model	Treatment; manipulation of MAP	Duration of BP manipulation/intervention	Outcome assessment technique
Dyste and colleagues¹⁷	Epidural balloon inflated and maintained at 200 mm Hg for 80 min with a syringe pump	Nine animals received hetastarch 20 mL/kg IV bolus of 6% solution followed by maintenance infusion at 80 mL/h. Eight animals mannitol at 1 g/kg IV bolus of 20% solution followed by an infusion of 1g/kg/h. Ten animals received phenylephrine; they received an IV infusion of a 16% solution of phenylephrine at a rate sufficient to raise and maintain MAP by 20–40% over the compression MAP. Twelve control animals received saline at 80 mL/h.	Entire experiment (2.5 h)	Bipolar electrodes at C7 and L7 to stimulate and to record SEPs. SCBF was measured using the radiolabeled microsphere technique. SEPs and SCBF determined at 0.5, 1.5, and 2.5 h after compression
Griffiths and colleagues¹⁹	Ventral, midline balloon compression at L2	MAP reduced and maintained at 70–80 mm Hg in groups 2 and 3 using trimetaphan camsylate and sodium nitroprusside	Until conduction failure; time not specified	DCEP assessed. Hydrogen gas technique used to measure SCBF.
Hukuda and colleagues¹⁶	Inflation of tourniquet that encircled spinal cord epidurally to 400 mm Hg; pressure maintained for 5 min. Injury produced at T5–T6 level.	Three hours post-injury, 15 min of hypertension, and 95% O₂/5% CO₂ gas ventilation were alternated with 10 min of normotensive air ventilation. MAPs maintained at 170–200 mm Hg and were induced with IV infusion of 0.002% norepinephrine.	3 h	Neurological exam (grade 1–4) as well as SEP repeated for up to 8 weeks. The thoracic cords were examined pathologically at the end of the experiment, comparing the size of hemorrhagic lesions in control and treatment groups.
Hardy and colleagues¹⁵	Increasing weight to the spinal cord (38–53 gm)	MAP raised to 150–200 mm Hg with 1–2 mg of levarterenol bitartrate	30–60 min	CEP measurement
Brodkey and colleagues¹⁴	Clamp on thoracic aorta to reduce perfusion pressure. Low thoracic laminectomy, weight holding apparatus positioned to deliver weight (18–38 gm) over entire dorsal surface of cord and intact dura.	MAP reduced to 60 mm Hg in thoracic aorta while weight was applied to spinal cord	MAP maintained low until disappearance of CEPs	Time to return of CEP
Haghighi and colleagues¹³	Weight drop model; 100 gm-cm impact injury at T13 using modified Allen's trauma device	Five cats were treated with nimodipine; 10 mcg/kg bolus followed by 1 mcg/kg/min infusion Five cats were treated with saline. Both groups received equal volumes of fluid. Nimodipine-treated group had MAPs significantly less than saline group (p < 0.05; MAPs dropped by 32% compared to baseline).	2 h	Recovery of SEPs
Gambardella and colleagues¹⁸	Modified Sujita vascular clip applied at T12–L1 level; applied perpendicular so that a 60.5–64.2% reduction in diameter in order to cause incomplete spinal cord injury. The clip was left on for 2 and 5 min in the trauma groups.	Group 1: No injury, no pharmacological treatment Group 2: Received NS post-injury Group 3: Received nimodipine (1.5 mcg/kg/min IV). MAP maintained near control levels with adrenaline (1:100,000 solution saline; 40 drops/min). Group 4: Nimodipine only, same dose as above Group 5: Adrenaline only, same dose as above to maintain MAPs at control levels Cord compression decreased MAP to 81 ± 3.5 in all animals; NS did not restore BP to control levels. Treatment with Nimodipine alone decreased BP further by 20–40 mm Hg. MAP maintained within physiological limits (120 mm Hg) in adrenaline alone and adrenaline+nimodipine groups.	2 h	Neurological outcome was evaluated by comparing scores obtained with Tarlov scale at 1, 24, and 48 h after the end of spinal cord compression. Morphometric studies were carried out 48 h post-injury; the number of myelinated axons per unit area was compared in the groups.
Dolan and colleagues¹¹	Modified aneurysm clip exerting compression force of 180 gm for 1 min at T1	Normotensive group with MAPs kept between 100 and 120 mm Hg, and hypertensive group with MAPs maintained between 125 and 150 mm Hg using sufficient noradrenaline; stopping infusion in treated animals after 60 min caused an immediate fall in the mean BP to levels below those of control animals (mean of 76 ± 1 mm Hg in normotensive and 77 ± 1 in hypertensive group compared to 84 ± 2 in control).	Noradrenaline was infused for 60 min, then MAP was monitored for 15 min.	Functional recovery assessed weekly for 8 weeks post-injury using inclined plane technique
Barrios and colleagues¹²	Cord translation group, root stump pull group, and torsion group. Electromechanical device used to apply displacing forces. Procedures performed at the T6–T10 level.	Induced hypotension to MAP of 45 mm Hg (50% decrease from the baseline MAP in pigs) using 0.1–0.5 mg of esmolol, followed after 2 min by nitroprusside sodium at 0.001–0.01 mg/min	Until loss neurophysiological signals (exact duration of hypotension unspecified)	Direct cord-to-cord evoked potentials recorded. Amount of displacement of cord until disappearance of CEPs was measured for each group with unharmed roots, severed roots, and under induced hypotension.
Fehlings and colleagues⁶	Modified aneurysm clip; 1 min of 53 gm of extradural compression at T1	MAPs not directly manipulated, but they were measured in the different groups. MAP was significantly increased (p < 0.02) following infusion of placebo and saline (74 ± 8.3), placebo, and dextran 40 (85 ± 7.7 mm Hg) and nimodipine 0.02 mg/kg and dextran 40 (88.4 ± 8.4 mm Hg). However, at 1 h after drug infusion, the MAP was again similar among the six treatment groups (p > 0.05).	1 h	MEPs and SSEPs were recorded concomitantly with the SCBF 1 h before and at 1, 2, 3, and 4 h post-SCI.

ASCI, acute spinal cord injury; MAP, mean arterial pressure; SCBF, spinal cord blood flow; DCEP, dorsal column evoked potential; SEP, sensory evoked potential; SSEP, somatosensory evoked potential; SCI, spinal cord injury; CEP, cortical evoked potential; gm, gram; mg, milligram; mcg, microgram; kg, kilogram; cm, centimeter; mm, millimeter; mm Hg, millimeter of mercury; min, minute; NS, normal saline; IV, intravenous; mL, milliliter; h, hour; BP, blood pressure.

Table 3.

Functional Outcomes and Biases

Study	Primary outcome	Secondary outcome	Biases/shortcomings	Overall impression
Dyste and colleagues¹⁷	Hetastarch group: spinal cord blood flow not affected at any cord segment Mannitol: did not affect blood flow to any measured tissue PE: SCBF significantly greater than the baseline level at 30 min after compression (p < 0.05), but not at subsequent measures of flow. Regression analysis showed only that the PE group was a statistically significant predictor of SCBF.	MAP was highest in the PE group after initiation of treatment compared to all other groups and was statistically significantly different from control group (p < 0.01). Within 10 min of balloon inflation, the SEPs were obliterated. There was no recovery of SEP during the remainder of the experiment in any treatment group, including the PE group.	SEP data recorded during experiment do not give an accurate prediction of long-term outcome.	In the PE group, despite having MAPs higher than control group (MAP kept at baseline) and other groups, there was no difference in neurological recovery. Negative study
Griffiths and colleagues¹⁸	In group 1 (normotensive), the SCBF remained unchanged until a PP of 65–70 mm Hg. Below that pressure, flow declined and was 50% of control values at PP of 35 mm Hg. For cord pressures, flow remained constant until cord pressures increased to 55–60 mm Hg, above which it declined. In group 2 (hypotensive), the flow started to decrease at PP of 70 mm Hg, which corresponded to cord pressures of 15–20 mm Hg. Amplitude of the DCEP decreased significantly from control at each reduction of PP in both groups. In group 1, conduction failure occurred at cord pressures of 90–100 mm Hg and PP of 20–30 mm Hg. In group 2, conduction failure occurred when PP measured 30–40 mm Hg and cord pressures were 20–40 mm Hg. There was no significant difference in SCBF or PP between the two groups, but the level of PP at conduction failure was higher in group 2. Group 3 (hypotension only) reduction in amplitude of DCEP by 12% when PP reduced to 43 mm Hg. At these lower levels of PP, the amplitude of DCEP is greater than in groups 1 and 2.	MAP was significantly different between groups 1 and 2 (125 ± 8 in group 1, 71 ± 14 in group 2; p < 0.01). Cord pressure was significantly different as well (group 1, 96 ± 5; group 2, 28 ± 11); however, no difference in SCBF or PP.	Only 12 subjects in entire study. No actual neurological exam performed on subjects to assess neurological outcome. No long-term follow-up, which may be necessary in spinal cord injury.	DCEP conduction failure occurred at a higher PP in hypotensive group; may suggest some benefit to keeping subjects normotensive, but not sufficient evidence to support normotensive or hypertensive support.
Hukuda and colleagues¹⁶	During first week, treated dogs attained better neurological grades, but was not statistically significant. After first week, both groups demonstrated steady neurological improvement for next 2 weeks, but remained paraparetic thereafter. There was no significant difference when comparing sizes of spinal cord lesions between treatment and control groups.	SEP recovered in some dogs in both control and treatment groups. There was no difference observed in recovery rate between the treatment group and the controls. However, those dogs (from either group) in which SEPs recovered also attained better neurological grades (p < 0.001) and had smaller pathological lesions (p < 0.001) compared to those who had no recovery of SEPs.	Lack of follow-up SEP study in half of the dogs made the analysis somewhat inaccurate; treatment initiated 3 h post-injury, which may be too late for recovery of damaged neurons.	Despite hypertensive therapy, treated dogs did not have a better neurological outcome compared to controls. Negative study
Hardy and colleagues¹⁵	When BP was raised to 150–200 mm Hg, the CEP returned in 2–5 min. The CEP was lost when BP returned to normal in 5–18 min.	None	Only 7 subjects; neurological function not examined, only CEPs looked at; duration of study only 60 min, no long-term outcome assessment; the actual data of each individual subject not displayed.	Recovery of CEP with raised BP; insufficient evidence to support hypertensive therapy
Brodkey and colleagues¹⁴	Effects of pressure and ischemia are additive when CEP is concerned.	In 2 of the cats, after CEP had disappeared attributed to combined effects of weight and BP reduction, weight was removed and potential returned, but disappeared again after 5 and 15 min. When BP raised to normal levels, CEP promptly returned and did not disappear again when BP was subsequently reduced after 30 min of normotension.	Only 5 subjects; in 2 of the 5 subjects, only single block applied. Only CEP used, no neurological exam performed.	Some suggestion that maintaining normotension in post-injury period may be beneficial for CEP recovery
Haghighi and colleagues¹³	In control (saline) group, average time to return of SEPs post-injury was 50 ± 15 min. In nimodipine-treated cats, 3 did not show return of SEPs up to 4 h post-trauma. Of the remaining 2, SEPs returned, on average, 80 ± 10 minutes.	Spinal cord specimens used to document location of trauma to cord; histological changes were mild in all cases and did not correlate with outcome.	Not specifically looking at effects of MAP on neurological recovery; trying to determine neuroprotective effects of nimodipine	Clear suggestion that hypotension in post-injury period detrimental for neurological outcome
Gambardella and colleagues¹⁸	In all animals, cord compression induced fifth-degree deficit. Saline improved the motor deficit slightly. Adrenaline-alone groups showed a better recovery in comparison with animals treated with saline alone. Nimodipine+adrenaline group showed the most improvement in neurological outcome. The groups that showed most improvement were those with higher MAPs (kept at baseline of 120 mm Hg).	Morphometric studies showed axonal number was significantly (p < 0.05–0.001) reduced in all groups subjected to spinal cord injury when compared to control group. In the group of animals treated with a combination of nimodipine and adrenaline, the concentration of axons in the regions of trauma or a region caudal to it was greater than in the other group of animals.	Possible protective effects of nimodipine may confound results when comparing nimodipine+adrenaline group to control group; duration of experiment only 48 h, which may not be enough time to for full neurological recovery.	Better neurological outcomes in treatment groups where MAPs were maintained at baseline (MAPs 120 mm Hg, normotensive). Hypertensive therapy not looked at.
Dolan and colleagues¹¹	Weekly mean inclined plan values for 8 weeks showed no statistically significant difference between the three groups at any time post-injury	None	Hypotension, which occurred after 60-min infusion was discontinued, may have had negative effects, offsetting any improvement caused by infusion. Sixty minutes of treatment may have been insufficient duration for treatment. Subjects may need longer MAP support for potential recovery.	No difference in neurological outcome in hypertensive vs. normotensive vs. control groups Negative study
Barrios and colleagues¹²	Group 1: separation group. Evoked potential disappeared with a displacement of 10.1 ± 1.6 mm with unharmed roots and 15.3 ± 4.7 mm after sectioning of four adjacent roots (p < 0.01). After induced hypotension, potentials lost at 4.0 ± 1.2 mm (p < 0.01). Group 2: root stump pull: Absence of potentials occurred at 20 ± 4.3 mm and increased to 23.5 ± 2.1 mm (p < 0.05) after cutting the two contralateral roots. With induced hypotension, signals lost at 5.3 ± 1.2 mm (p < 0.01). Group 3: Torsion group. Cord allowed torsion of 95.3 ± 0.2 degrees that increased to 112.4 ± 7.1 (p < 0.01) degrees if contralateral nerve roots cut. Under hypotension, loss of potentials at 20 ± 6.2 degrees (p < 0.01).	Pathological study performed showed that, independently from the spinal cord displacement method, none of the spinal cord samples showed macroscopic, histological, or immunohistochemical damage.	Cord manipulation took place immediately after decrease in MAP, which may not have allowed for spinal cord autoregulation. The sacrifice of the roots under hypotensive conditions may have affected the cord vascularization. Return of signal was not correlated with neurological function.	Clear suggestion that cord does not tolerate manipulation under hypotensive conditions Hypertensive conditions not looked at
Fehlings and colleagues⁶	Only the combination of nimodipine 0.02 mg/kg and dextran 40 increased SCBF at T1 (43.69 ± 6.09 mL/100 gm/min; p < 0.003) and improved MEPs (p < 0.0001) and SSEPs (p < 0.0001) post-ASCI.	Histopathology revealed that neither the total volume of hemorrhage at the injury site (p > 0.05) nor the percent hemorrhage by volume (p > 0.05) varied significantly among treatment groups.	Interventions were not directly targeting BP; the duration of the interventions were only 1 h; 1 h post-injury, the MAP in all groups were similar; only 4-h duration of experiment.	Despite having MAP that was higher in placebo and saline groups, placebo and dextran 40 groups compared to nimodipine 0.02 mg/kg and saline, nimodipine 0.05 mg/kg and saline and nimodipine 0.05 mg/kg and dextran 40 groups, there was no improvement in neurological outcome. In this scenario, MAP by itself did not contribute to neurological improvement.

MAP, mean arterial pressure; BP, blood pressure; PE, phenylephrine; SCBF, spinal cord blood flow; DCEP, dorsal column evoked potential; SEP, spinal evoked potential; SSEP, somatosensory evoked potentials; ASCI, acute spinal cord injury; CEP, cortical evoked potential; DCEP, direct cortical evoked potential; mL, milliliter; gm, gram; mg, milligram; kg, kilogram; min, minute; mm, millimeter; mm Hg, millimeter of mercury; PP, perfusion pressure.

Discussion

It is a currently utilized therapeutic practice, in many institutions, to target high MAPs immediately post-SCI in order to reduce the amount of secondary injury as well as improve neurological outcome. The targeting of MAPs and SBP post-ASCI is still a controversial subject, given that there exists little evidence to support the current treatment recommendations. Current recommendations of keeping MAP above 85–90 is based on level III evidence only. The reasoning behind this suggestion is that maintaining spinal cord perfusion immediately post-injury is important to reduce the effects of secondary injury that follows the initial insult. To our knowledge, there is no systematic review of the animal studies on which these recommendations are based. The purpose of this scoping systematic review was to accumulate all of the animal data available that address post-injury MAPs and systolic blood pressure (SBP) and try to assess the validity of these recommendations.

We limited our inclusion criteria to studies that actually measured some sort of neurological outcome, either by physical exam or with electrophysiological evidence. We purposefully excluded studies that only addressed SCBF only, in order to better correlate the effect of MAP with actual clinical outcomes.

Our search strategy yielded 10 studies that met our inclusion criteria. It is not surprising that we found studies with significant heterogeneity. There were significant differences in study population, method of inducing injury, method of manipulating MAP/SBP, and measuring neurological outcome. Given these differences, a statistical analysis and regression model were deemed to be unhelpful.

Because of the heterogeneity of these studies, no definitive conclusion can be reached regarding the impact of MAP on neurological outcome in ASCI models. However, we did make certain observations; first of all, in the positive studies mentioned, there is only one study that shows a positive outcome for hypertension.¹⁵ The other three studies show a benefit of maintaining normotension as opposed to pushing the MAPs to a supraphysiological level. In fact, there is evidence that suggests that hypertension causes larger hemorrhage and edema in the spinal cord at the level of injury as well as the adjacent levels attributed to loss of autoregulation.¹⁹ Second, two studies on our list^12,13 invariably showed that hypotension in the setting of ASCI is most likely detrimental to the neurological outcome. This is not surprising, given that hypotension has been shown to be harmful in the setting of acute traumatic brain injury as well²¹ and should be avoided in these patients.

The negative studies yielded in our search failed to show any benefit for higher MAPs on neurological outcome. The main critique of these studies is usually 2-fold; first, the BP was manipulated for a period of hours only. Proponents of hypertension in the post-injury period argue that, in order to prevent secondary injury, the BP should be maintained elevated over a period of days,^23,24 which was based on an experimental animal SCI study that showed that, between days 3 and 5 post-injury, the spinal cord experienced the greatest degree of cord edema and vascular congestion.²² Second, the neurological outcome is also assessed after a period of hours at the end of the experiment. Most will argue that this is insufficient time for neurological recovery to occur after a severe ASCI and that a more-accurate assessment of neurological recovery would be after a period of weeks to months, when the damaged neuronal structures have had time to recover. Dolan and colleagues assessed functional recovery at the 8-week mark and did not find any difference between the normotensive, hypertensive, and control groups.¹¹ However, they only manipulated the MAPs over a 60-minperiod.

Limitations

This scoping systematic review was not without its limitations. The heterogeneity of the studies compiled is perhaps one of the main causes of the limitations. There were different animals used in most of the studies found; it is difficult to say what the impact of this species variation is on the response to manipulation of the MAP. Different mechanisms were used to induce traumatic SCI, and these different mechanisms can potentially lead to different secondary injury cascades, making it difficult to generalize results obtained. Heterogeneous outcome assessments made it difficult to come to an overall conclusion about the impact of MAP on functional outcome.

Also notable is the small number of studies we were able to obtain within the confines of our search; these studies did not contain a large number of experimental studies either, further limiting our ability to come to any certain conclusions. Given that these were animal studies, even if the above-stated limitations did not exist, we would not be able to extrapolate these results to humans.

Potential future directions

Despite the limitations described above, there is potential for future investigation in the area of MAP and ASCI. However, there should be some considerations in planning future animal studies. First, regardless of the animal species selected, the number of subjects needs to be larger than currently described. Further, there needs to be extensive stratification of the target MAP, with large volumes of animals in multiple MAP categories. Separation of MAP categories by 5 mm Hg, from MAPs of 50 up to 100 mm Hg (or more), may provide information on the “optimal” MAP. This concept of “bigger” and more in-depth studies carries with it cost and issues of space to care for the animals. Thus, it is imperative that multi-center collaboration occurs with future projects, in order to avoid reproducing more small studies that fail to add clarity to the question. Second, there needs to be control arms in future studies evaluating “standard” intensive care unit–based practices of maintaining homeostasis during the acute injury period. This would involve targeting normal MAP levels and maintenance of normal biochemical/physiological profiles. Hence, future should consider including biochemical and continuous physiological profiles on the enrolled animals in order to ensure there are no confounders leading to secondary injury to the spinal cord. Third, the mechanism of inducing trauma should mimic that observed in the clinical situation. The majority of ASCI observed in trauma patients is that of a crush-type injury, which, preceding any surgical intervention, is usually maintained upon presentation to the hospital. Future animal models must consider this fact and try best to replicate this scenario. Timing for surgical decompression in ASCI is still debated and depends on the presenting neurological grade and duration of deficits. Thus, in planning future animal models of ASCI, consideration should be given to the duration of compression and treatment both in models with early and late “decompression.” Fourth, a combination of functional outcome measures is likely best. Both electrophysiological and formal physical examination, conducted at frequent regular intervals, would yield useful information. Electrophysiological evaluation of dorsal column function in isolation is an insufficient measure of outcome. We know, from clinical experience with intraoperative monitoring of MEP and SSEP, that dramatic changes in monitoring do not routinely reflect the patients' post-operative neurological function and cannot be relied on solely for outcome prediction. Thus, applying these methods in isolation to animal models of ASCI, though informative, does not really reflect our current approach clinically and should be accompanied by objective neurological examination bases outcomes. Fifth, the duration of follow-up is a major issue, given that recovery from ASCI can take months to years. Plans for long-term animal models is a must. Short-term analysis of functional outcome, though interesting, does not really assist the treating clinician in determining whether the risk of prolonged vasopressor use is warranted, or whether any improvement observed is sustained. Finally, histological examination of the spinal cord in long-term models would provide valuable information as to the impact of MAP therapy on tissue fate over time. This may answer whether or not MAP therapy actually leads to healthier spinal cord in the long run.

Conclusion

This review concludes that, within the animal literature, there is insufficient evidence to draw a conclusion about the effect of MAP on neurological outcome in animal models of ASCI. The role of induced hypertension in the immediate post-injury period requires further study, given that this treatment is not without risk.

Footnotes

Acknowledgments

This work was made possible through salary support through the Cambridge Commonwealth Trust, University of Manitoba Clinician Investigator Program, R. Samuel McLaughlin Research and Education Award, the Manitoba Medical Service Foundation, and the University of Manitoba Faculty of Medicine Dean's Fellowship Fund.

Author Disclosure Statement

F.Z. has received salary support for dedicated research time, during which this project was partially completed. Such salary support came from: the Cambridge Commonwealth Trust; University of Manitoba Clinician Investigator Program; R. Samuel McLaughlin Research and Education Award; the Manitoba Medical Service Foundation; and the University of Manitoba Faculty of Medicine Dean's Fellowship Fund.

References

Tator

C.H.

(1997). Experimental and clinical studies of the pathophysiology and management of acute spinal cord injury. J. Spinal Cord Med., 19, 206–214.

Young

(1993). Secondary injury mechanisms in acute spinal cord injury. J. Emerg. Med., 11, 13–22.

Aebi

, Mohler

, Zach

G.A.

, and Morscher

(1986). Indication, surgical technique, and results of 100 surgically treated fractures and fracture-dislocations of the cervical spine. Clin. Orthop., 203, 244–247.

Bracken

, and Holford

(1993). Effects of timing of methylprednisolone or naloxone administration on recovery of segmental and long-tract neurologic function in NASCIS 2. J. Neurosurg., 79, 500–507.

Bracken

, Shepard

, Holford

, Leo-Summers

, Aldrich

E.F.

, Fazl

, Fehlings

, Herr

D.L.

, Hitchon

P.W.

, Marshall

L.F.

, Nockels

R.P.

, Pascale

, Perot

P.L.

Jr. , Piepmeier

, Sonntag

V.K.

, Wagner

, Wilberger

J.E.

, Winn

H.R.

, and Young

(1997). Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal cord injury. JAMA, 277, 1597–1604.

Fehlings

M.G.

, Tator

C.H.

, and Linden

R.D.

(1989). The effect of nimodipine and dextran on axonal function and blood flow following experimental spinal cord injury. J. Neurosurg., 71, 403–416.

(2013). Guidelines for the management of acute cervical spine and spinal cord injuries. Neurosurgery, 72, 2 Suppl., 1–259.

Higgins

J.P.T.

, and Green

(eds). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. Available at: http:handbook.cochrane.org. Accessed October 25, 2013 .

Moher

, Liberati

, Tetzlaff

, and Altman

D.G.

; PRISMA Group. (2009). Preferred reporting items for systematic reviews and meta-analysis: the PRISMA statement. Ann. Intern. Med., 151, 264–269.

10.

Viswanathan

, Berkman

. Development of the RTI Item Bank on Risk of Bias and Precision of Observational Studies. Methods Research Report. (Prepared by the RTI International–University of North Carolina Evidence-based Practice Center under Contract No. 290-2007-0056-I.) AHRQ Publication No. 11-EHC028-EF. Agency for Healthcare Research and Quality: Rockville, MD, September 2011. www.effectivehealthcare.ahrq.gov/reports/final.cfm (last accessed January 2016 ).

11.

Dolan

E.J.

, Tator

C.H.

(1980). The treatment of hypotension due to acute experimental spinal cord compression injury. Surg Neurol, 13(5):80–4.

12.

Barrios

, Pizá-Vallespir

, Burgos

, De Blas

, Montes

, Hevia

, Collazos-Castro

J.E.

, and Correa

(2014). Influence of hypotension and nerve root section on the ability to mobilize the spinal cord during spine surgery. An experimental study in a pig model. Spine J., 14, 1300–1307.

13.

Haghighi

S.S.

, Chehrazi

B.B.

, and Wagner

F.C.

(1988). Effect of nimodipine-associated hypotension on recovery from acute spinal cord injury in cats. Surg. Neurol., 29, 293–297.

14.

Brodkey

J.S.

, Richards

D.E.

, Blasingame

J.P.

, and Nulsen

F.E.

(1972). Reversible spinal cord trauma in cats. Additive effects of direct pressure and ischemia. J. Neurosurg., 37, 591–593.

15.

Hardy

R.W.

, Brodkey

J.S.

, Richards

D.E.

, and Nulsen

F.E.

(1972). Effect of systemic hypertension on compression block of spinal cord. Surg. Forum, 23, 434–435.

16.

Hukuda

, Mochizuki

, and Ogata

(1980). Therapeutic trial of combined hypertension and hypercarbia on experimental acute spinal cord injury. Neurosurgery, 6, 644–648.

17.

Dyste

G.N.

, Hitchon

P.W.

, Girton

R.A.

, and Chapman

(1989). Effect of hetastarch, mannitol, and phenylephrine on spinal cord blood flow following experimental spinal injury. Neurosurgery, 24, 228–235.

18.

Gambardella

, Collufio

, Caruso

, Abbate

, Germana

, and Tomasella

(1995). Experimental incomplete spinal cord injury: treatment with a combination of nimodipine and adrenaline. J. Neurosurg. Sci., 39, 67–74.

19.

Griffiths

I.R.

, Trench

J.G.

, and Crawford

R.A.

(1979). Spinal cord blood flow and conduction during experimental cord compression in normotensive and hypotensive dogs. J. Neurosurg., 50, 353–360.

20.

Rawe

S.E.

, Lee

W.A.

, and Perot

P.L.

(1978). The histopathology of experimental spinal cord trauma. J. Neurosurg., 48, 1002–1007.

21.

Chesnut

R.M.

, Marshall

L.F.

, Klauber

M.R.

, Blunt

B.A.

, Baldwin

, Eisenberg

H.M.

, Jane

J.A.

, Marmarou

, and Foulkes

M.A.

(1993). The role of secondary brain injury in determining outcome from severe head injury. J. Trauma, 34, 216–222.

22.

Yashon

, Bingham

W.G.

Jr. , Faddoul

E.M.

, and Hunt

W.E.

(1973). Edema of the spinal cord following experimental impact trauma. J. Neurosurg., 38, 693–697.

23.

Vale

F.L.

, Burns

, Jackson

A.B.

, and Hadley

M.N.

(1997). Combined medical and surgical treatment after acute spinal cord injury: results of a prospective pilot study to assess the merits of aggressive medical resuscitation and blood pressure management. J. Neurosurg., 87, 239–246.

24.

Levi

, Wolf

, and Belzberg

(1993). Hemodynamic parameters in patients with acute cervical cord trauma: description, intervention, and prediction of outcome. Neurosurgery, 33, 1007–1017.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB