✓ The presence of lactic acidosis in the cerebrospinal fluid of patients suffering brain injury as the result of trauma, subarachnoid hemorrhage, neoplasia, or ischemia has been well documented. The authors theorized that this acidosis becomes harmful in itself, and that treatment with an alkalinizing agent (tris(hydroxymethyl)aminomethane: tromethamine) capable of penetrating the blood-brain barrier would be efficacious. Fifteen pairs of mongrel cats were subjected to a 2.85-atmosphere fluid-percussion injury (LD80), and were supported by respirators for up to 72 hours prior to being placed in cages for an additional 4 days of observation. Experimental cats underwent continuous infusion of tromethamine (begun 10 minutes after injury); control animals were infused with an equal volume of lactated Ringer's solution. Twenty percent of the control group survived until sacrificed on Day 7 post-injury. Survival in the tromethamine group was 60% (p < 0.05), and morbidity also appeared to be reduced in the treated cats. Intracranial pressure (ICP) in treated cats was 60% (p < 0.05) of that in the control cats after respirator support for 3 days. Tromethamine infusion was associated with improved survival, decreased morbidity, and decreased ICP when compared with results in control animals. The literature with regard to central nervous system acidosis has been reviewed in an attempt to clarify and define this problem.
With an overview of CNS acidosis
Michael J. Rosner and Donald P. Becker
Experimental observations and a theoretical model
Michael J. Rosner and Donald P. Becker
✓ Laboratory observations made in cats with fluid-percussion head injuries have suggested that plateau waves or Lundberg “A-waves” are not independent of systemic circulatory events. Four distinct phases in the evolution of the plateau wave have been identified, and each related to a circulatory change in a causal manner. The first phase is the premonitory drift phase where intracranial pressure (ICP) gradually increases prior to the plateau proper. This phase is caused by a slow gradual decline in systemic arterial blood pressure (SABP) which increases ICP by autoregulatory vasodilation and reduces cerebral perfusion pressure (CPP) to a range of 70 to 80 mm Hg. The second phase is the plateau phase initiated at a CPP of about 70 to 80 mm Hg, and is characterized by a rapid increase in ICP as CPP falls further to 40 to 50 mm Hg. The plateau lasts as long as the CPP remains stable and above ischemic levels. The third phase is the ischemic response, characterized by CPP being returned toward normal by increases in SABP in response to very low CPP's. The fourth phase is the resolution, characterized by a rapid decline in the ICP to baseline levels with stabilization of the SABP and CPP, and is best explained by autoregulatory vasoconstriction.
Plateau waves appear to occur as the result of intact or mostly intact autoregulation responding to changes in CPP. The series of events that follow are best explained by what is known of normal autoregulation; the various properties of plateau waves are viewed and explained as the expected and logical consequences of an unstable CPP acting upon a generally intact cerebrovascular bed in the face of elevated ICP and decreased compliance.
Mario J. Cardoso and Michael K. Rosner
Minimally invasive lumbar spine surgery has dramatically evolved over the last decade. Minimally invasive techniques and transforaminal lumbar interbody fusion (TLIF) often require a steep learning curve. Surgical techniques require pre-positioning the patient in maximal kyphosis to optimize visualization of the disc space and prevent unnecessary retraction of neural structures. The authors describe their experience in validating the surgical technique recommendation of Wilson frame–induced kyphosis.
Over the past 6 months, data obtained in 20 consecutive patients (40 total levels) undergoing minimally invasive TLIF were reviewed. In each patient, preincision intraoperative radiographs were reviewed at L4–5 and L5–S1 with the patient on a Wilson frame in maximal lordosis and then in maximal kyphosis. The change in disc space angle at L4–5 and L5–S1 after changing from maximal lordosis to maximal kyphosis was reviewed. Descriptive statistics were calculated for sagittal plane angular measures at L4–5 and L5–S1 in lordosis and kyphosis, including absolute differences and percentage of change between positions. Inferential statistics were calculated using paired t-tests with α= 0.05.
Twenty patients underwent single- or multilevel minimally invasive TLIF. Inducing kyphosis with the Wilson frame aided in optimizing exposure and decreasing the need for neural structure retraction. Both L4–5 and L5–S1 showed statistically significant (p < 0.001) and clinically meaningful changes with increased segmental flexion in the kyphotic position. At L4–5 the mean increase in flexion was 4.5° (95% CI 2.9–6.0°), representing an average 47% change. The mean increase in flexion at L5–S1 was 3.2° (95% CI 2.3–4.2°), representing an average 20.8% change. In lordosis the mean angle at L4–5 was 10.6 ± 4.4° and at L5–S1 was 17 ± 7.0°. In kyphosis the mean angle at L4–5 was 6.1 ± 4.5° and at L5–S1 was 13.8 ± 6.5°. Additionally, there was a statistically significant difference (p < 0.05) in percentage of change between the 2 levels, with L4–5 showing a greater change (27% more flexion) between positions, but the absolute mean difference between the levels was small (1.3°).
Minimally invasive TLIF is challenging and requires a significant learning curve. The recommended surgical technique of inducing kyphosis with the Wilson frame prior to incision significantly optimizes exposure. The authors' experience demonstrates that this technique is essential when performing minimally invasive lumbar spinal fusions.
Part 2: The effects of low cerebral perfusion pressure and autoregulation
W. John Gray and Michael J. Rosner
✓ The pressure-volume index (PVI) was measured in six adult cats while cerebral perfusion pressure (CPP) was reduced from normal levels to below the autoregulatory range by a continuous infusion of adenosine triphosphate. Anesthesia was induced with methohexital and maintained with an N2O:O2 (70%:30%) mixture. Body temperature, hematocrit, and PaCO2 were held constant throughout each experiment. Cerebral blood flow (CBF) was measured by the hydrogen clearance method. At CPP levels over 50 mm Hg, CBF remained relatively constant despite changes in CPP. Within this range, the PVI varied directly with CPP (PVI = 0.24 ml + 0.0013 mm Hg CPP). Below the autoregulatory range, CBF fell progressively with further decreases in CPP; in this range, PVI was found to increase as CPP fell (PVI = 0.84 ml − 0.0071 mm Hg CPP). These results indicate that the PVI is a complex function of CPP, varying directly with CPP within the autoregulatory range and indirectly with CPP below the autoregulatory range.
Michael J. Rosner and Irene B. Coley
✓ Previous investigations have suggested that intracranial pressure waves may be induced by reduction of cerebral perfusion pressure (CPP). Since pressure waves were noted to be more common in patients with their head elevated at a standard 20° to 30°, CPP was studied as a function of head position and its effect upon intracranial pressure (ICP).
In 18 patients with varying degrees of intracranial hypertension, systemic arterial blood pressure (SABP) was monitored at the level of both the head and the heart. Intracranial pressure and central venous pressure were assessed at every 10° of head elevation from 0° to 50°. For every 10° of head elevation, the average ICP decreased by 1 mm Hg associated with a reduction of 2 to 3 mm Hg CPP. The CPP was not beneficially affected by any degree of head elevation. Maximal CPP (73 ± 3.4 mm Hg (mean ± standard error of the mean)) always occurred with the head in a horizontal position. Cerebrospinal fluid pressure waves occurred in four of the 18 patients studied as a function of reduced CPP caused by head elevation alone. Thus, elevation of the head of the bed was associated with the development of CPP decrements in all cases, and it precipitated pressure waves in some. In 15 of the 18 patients, CPP was maintained by spontaneous 10- to 20-mm Hg increases in SABP, and pressure waves did not occur if CPP was maintained at 70 to 75 mm Hg or above.
It is concluded that 0° head elevation maximizes CPP and reduces the severity and frequency of pressure-wave occurrence. If the head of the bed is to be elevated, then adequate hydration and avoidance of pharmacological agents that reduce SABP or prevent its rise are required to maximize CPP.
Michael J. Rosner, Heber H. Newsome and Donald P. Becker
✓ Catecholamine release was studied in cats as a function of injury severity in the fluid-percussion model of brain injury. Hyperglycemia was also studied as a function of catecholamine response and injury severity. Epinephrine (E) and norepinephrine (NE) increased as a function of injury. This increase was maximal at about a 3.0-atmosphere (atm) injury level and amounted to a 500-fold increase for E and 100-fold increase for NE. Both catecholamines increased maximally by 10 seconds postinjury. Glucose increased to about 350% of baseline at 500 seconds postinjury and also increased as a function of injury severity. Results suggested that the sympathoadrenal discharge was capable of a graded response which was maximal at about 3.0 atm. Associated with this increase were hypertension, bradyarrhythmias, tachyarrhythmias, and hyperglycemia.
Mario J. Cardoso and Michael K. Rosner
In this study, the authors review the technique for inserting the Prestige ST in a contiguous multilevel cervical disc arthroplasty in patients with radiculopathy and myelopathy. They describe the preoperative planning, surgical technique, and their experience with 10 patients receiving a contiguous Prestige ST implant. They present contiguous multilevel cervical arthroplasty as an alternative to multilevel arthrodesis.
After institutional board review approval was obtained, the authors performed a retrospective review of all contiguous multilevel cervical disc arthroplasties with the Prestige ST artificial disc between August 2007 and November 2009 at a single institution by a single surgeon. Clinical criteria included patients who had undergone a multilevel cervical disc arthroplasty performed for radiculopathy and myelopathy without the presence of a previous cervical fusion. Between August 2007 and November 2009, 119 patients underwent cervical arthroplasty. Of the 119 patients, 31 received a Hybrid construct (total disc resection [TDR]–anterior cervical decompression and fusion [ACDF] or TDR-ACDF-TDR) and 24 received a multilevel cervical arthroplasty. The multilevel cervical arthroplasty group consisted of 14 noncontiguous and 10 contiguous implants. This paper examines patients who received contiguous Prestige ST implants.
Ten men with an average age of 45 years (range 25–61 years) were treated. Five patients presented with myelopathy, 3 presented with radiculopathy, and 2 presented with myeloradiculopathy. Twenty-two 6 × 16–mm Prestige ST TDRs were implanted. Six patients received 2-level Prestige ST implants. Five patients received TDRs at C5–6 and C6–7, and 1 patient received TDRs at C3–4 and C4–5. One patient received a TDR at C3–4, C5–6, and C6–7 where C4–5 was a congenital block vertebra. Three patients (2 with 3-level disease and 1 with 4-level disease) received contiguous Prestige ST implants as well as a Prevail ACDF as part of their constructs. The mean clinical and radiographic follow-up was 12 months. There has been no case of screw backout, implant dislodgment, progressive kyphosis, formation of heterotopic bone, evidence of pseudarthrosis at the Prevail levels, or development of symptomatic adjacent level disease.
Multilevel cervical arthroplasty with the Prestige ST is a safe and effective alternative to fusion for the management of cervical radiculopathy and myelopathy.
Chris J. Neal and Michael K. Rosner
Minimal-access transforaminal lumbar interbody fusion (TLIF) has gained popularity as a method of achieving interbody fusion via a posterior-only approach with the aim of minimizing injury to adjacent tissue. While many studies have reported successful outcomes, questions remain regarding the potential learning curve for successfully completing this procedure. The goal of this study, based on a single resident's experience at the only Accreditation Council for Graduate Medical Education–approved neurosurgical training center in the US military, was to determine if there is in fact a significant learning curve in performing a minimal-access TLIF.
The authors retrospectively reviewed all minimal-access TLIFs performed by a single neurosurgical resident between July 2006 and January 2008. Minimal-access TLIFs were performed using a tubular retractor inserted via a muscle-dilating exposure to limit approach-related morbidity. The accuracy of screw placement and operative times were assessed.
A single resident/attending team performed 28 minimal-access TLIF procedures. In total, 65 screws were placed at L-2 (1 screw), L-3 (2 screws), L-4 (18 screws), L-5 (27 screws), and S-1 (17 screws) from the resident's perspective. Postoperative CTs were reviewed to determine the accuracy of screw placement. An accuracy of 95.4% (62 of 65) properly placed screws was noted on postoperative imaging. Two screws (at L-5 in the patient in Case 17 and at S-1 in the patient in Case 9) were lateral, and no revision was needed. One screw (at L-4 in Case 24) was 1 mm medial without symptoms or the need for revision. In evaluating the operative times, 2 deformity cases (Grade III spondylolisthesis) were excluded. The average operating time per level in the remaining 26 cases was 113.25 minutes. The average time per level for the first 13 cases was 121.2 minutes; the amount of time decreased to 105.3 minutes for the second group of 13 cases (p = 0.25).
In summary, minimal-access TLIF can be safely performed in a training environment without a significant complication rate due to the expected learning curve.
Part 1: The effects of cerebral perfusion pressure changes and anesthesia
W. John Gray and Michael J. Rosner
✓ The pressure-volume index (PVI) was measured as a function of cerebral perfusion pressure (CPP) in 12 adult cats. Anesthesia was induced with methohexital in six animals and with pentobarbital in six animals; all were maintained on an N2O:O2 (70%:30%) mixture. The CPP was either increased in 10-torr increments using norepinephrine or decreased by a combination of adenosine triphosphate and hemorrhage in subgroups. Three estimations of PVI were made at each level of CPP. The PaCO2, body temperature, and hematocrit were controlled at normal levels throughout. In both groups there was a linear relationship between PVI and CPP with increasing CPP being reflected by a rise in PVI. This relationship was more marked in the methohexital group: PVI = 0.37 ml + 0.0005 mm Hg CPP in the pentobarbital group, and PVI = 0.14 ml + 0.0019 mm Hg CPP in the methohexital group. These results indicate that the PVI is not independent of CPP but is a function of CPP and is profoundly influenced by anesthesia.