J Thorac Cardiovasc Surg 2006;131:853-861
© 2006 The American Association for Thoracic Surgery
Surgery for Acquired Cardiovascular Disease |
Magnesium as a neuroprotectant in cardiac surgery: A randomized clinical trial
Sunil K. Bhudia, MD
a
,
Delos M. Cosgrove, MD
a
,
Richard I. Naugle, PhD
b
,
Jeevanantham Rajeswaran, MSc
c
,
Buu-Khanh Lam, MD
a
,
Emily Walton, BSc
b
,
John Petrich, RPh
d
,
Roberta C. Palumbo, RN
a
,
A. Marc Gillinov, MD
a
,
Carolyn Apperson-Hansen, MStat
c
,
Eugene H. Blackstone, MD
a
,
c
,
*
a Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, Ohio
b Department of Psychiatry and Psychology, Cleveland Clinic, Cleveland, Ohio
c Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, Ohio
d Pharmacy Department, Cleveland Clinic, Cleveland, Ohio
Received for publication April 4, 2005; revisions received October 20, 2005; accepted for publication November 21, 2005.
* Address for reprints: Eugene H. Blackstone, MD, Section of Clinical Research, Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, 9500 Euclid Ave/JJ40, Cleveland, OH 44195 (Email: blackse{at}ccf.org).
 |
Abstract
|
|---|
OBJECTIVE: We sought to evaluate magnesium as a neuroprotectant in patients undergoing cardiac surgery with cardiopulmonary bypass.
METHODS: From February 2002 to September 2003, 350 patients undergoing elective coronary artery bypass grafting, valve surgery, or both were enrolled in a randomized, blinded, placebo-controlled trial to receive either magnesium sulfate to increase plasma levels 1
to 2 times normal during cardiopulmonary bypass (n = 174) or no intervention (n = 176). Neurologic function, neuropsychologic function, and depression were assessed preoperatively, at 24 and 96 hours after extubation (neurologic) and at 3 months (neuropsychologic, depression). Neurologic scores were analyzed using ordinal longitudinal methods, and neuropsychologic and depression inventory data were summarized by principal component analysis, followed by linear regression analysis using component scores as response variables.
RESULTS: Seven (2%) patients had a postoperative stroke, 2 (1%) in the magnesium and 5 (3%) in the placebo group (P = .4). Neurologic score was worse postoperatively in both groups (P < .0001); however, magnesium group patients performed better than placebo group patients (P = .0001), who had prolonged declines in short-term memory and reemergence of primitive reflexes. Three-month neuropsychologic performance and depression inventory score were generally better than preoperatively, with few differences between groups (P > .6); however, older age (P = .0006), previous stroke (P = .003), and lower education level (P = .0007) were associated with worse performance.
CONCLUSIONS: Magnesium administration is safe and improves short-term postoperative neurologic function after cardiac surgery, particularly in preserving short-term memory and cortical control over brainstem functions. However, by 3 months, other factors and not administration of magnesium influence neuropsychologic and depression inventory performance.
Abbreviations and Acronyms CABG = coronary artery bypass grafting; CPB = cardiopulmonary bypass; NMDA =
N-methyl-D-aspartate; WPNS = Western Perioperative Neurologic Scale
|
Introduction
|
|---|
Neurologic and neuropsychologic dysfunction are major causes of mortality and morbidity after cardiac surgery
1,2
and are associated with prolonged hospital stay and use of intermediate- or long-term care facilities.
3
Neuropsychologic dysfunction is evident in 50% to 80% of patients at hospital discharge, 20% to 50% at 6 weeks, and 10% to 30% at 6 months postoperatively
4-7
; thus, early neuropsychologic dysfunction is implicated in late cognitive deterioration.
8
Although measures to prevent brain injury have been studied,
9,10
strategies to minimize injury, as advocated for stroke victims,
11
have received less attention. After an ischemic neurologic insult, glutamate is released, activating N-methyl-D-aspartate (NMDA), leading to increased sodium and calcium ion conduction across cell membranes and exacerbation of neurologic injury. Magnesium sulfate, a noncompetitive NMDA antagonist with clinical benefit in acute stroke,
11
has not been investigated as a neuroprotectant in cardiac surgery. Therefore, we conducted a single-institution randomized clinical trial to determine whether increased magnesium during cardiac surgery and in the first 24 hours postoperatively is neuroprotective.
|
Patients and Methods
|
|---|
Study Design
From February 2002 to September 2003, 350 patients (mean age 64 ± 12 years) undergoing elective on-pump coronary artery bypass grafting (CABG), valve surgery, or combined CABG and valve surgery were enrolled in a randomized, blinded, placebo-controlled clinical trial at Cleveland Clinic to receive either magnesium to increase plasma levels to between 3.6 and 4.8 mg · dL–1 during the operation and for 24 hours thereafter (magnesium group, n = 174)
12-14
or no intervention (placebo group, n = 176; Appendix E1). Patients with preoperative atrial fibrillation or renal impairment were excluded (Figure E1). Patients were randomized 1:1 at operation, with a block size of 2 and 4, independently for each of 11 operating rooms. Preoperative and operative characteristics of the 2 groups were well balanced (Table 1). The Pharmacy Department (J.P.) prepared study medications, the anesthetic team administered the 24-hour infusion, and the perfusion teams dosed the cardiopulmonary bypass (CPB) circuit. All were blinded to the study except perfusion teams, who were required to monitor and maintain magnesium levels during CPB.
This study was approved by the institutional review board, and patients provided written informed consent.
Magnesium Dosing
Patients randomized to the magnesium group received 780 mg (32 mmol) of MgSO4 in 100 mL of normal saline intravenously over 15 minutes during anesthesia induction, followed by 3160 mg (130 mmol) in 100 mL of normal saline over 24 hours; the CPB circuit was primed with MgSO4 to a concentration of 3.6 mg · dL–1. The plasma magnesium level was measured every 15 minutes during CPB. If it decreased to less than 3.6 mg · dL–1, MgSO4 was administered to achieve levels between 3.6 and 4.8 mg · dL–1 (Figure 1).
View larger version (17K):
[in this window]
[in a new window]
|
Figure 1. Perioperative magnesium levels. Open circles (placebo group) and filled circles (magnesium group) are mean values within different time intervals. Solid lines are evolution of predicted mean values over time for each group.
|
|
Patients randomized to the placebo group received normal saline intravenously in the CPB circuit and over 24 hours in bags and syringes indistinguishable from those used for the magnesium group.
Conduct of Operations
Anesthesia was induced intravenously with a combination of midazolam, 3 to 5 mg; fentanyl, up to 0.25 mg; and sodium thiopental, 3 to 5 mg · kg–1 (in patients with mitral or aortic valve stenosis or left ventricular ejection fraction 
30%, etomidate, 0.3-0.5 mg · kg–1, was used instead of sodium thiopental). It was maintained with inhaled isoflurane, 1% to 2%, supplemented with fentanyl up to 1 mg.
CPB was instituted with a conventional roller-pump system and vacuum-assisted venous return
15
using a heparin-coated circuit that included a 25-µm arterial filter. Arterial flow rate was 2.0 to 2.4 L · min–1
· m–2, with mean arterial pressure maintained between 50 and 70 mm Hg. Body temperature was either kept at 37°C or allowed to drift to 32°C to 34°C. Blood-gas management was by alpha-stat strategy.
Assessments
Safety
Postoperative mortality and morbidity data were collected by research nurses and entered into the Cardiovascular Information Registry concurrently with patient care. This registry is approved for use in research by the institutional review board.
Neurologic assessment
The Western Perioperative Neurologic Scale (WPNS)
16
was used to assess neurologic performance preoperatively and at 24 and 96 hours after extubation. Fourteen tests were performed by an individual certified to administer them (S.K.B.) and were grouped into 4 domains:
- • Mentation (4 tests): level of consciousness; orientation to time, place, and person; short-term memory; and speech
- • Motor function (5 tests): motor strength in right and left upper and lower limbs, cranial nerves (facial motor function)
- • Sensory (4 tests): sensation in right and left upper and lower limbs
- • Cerebellar function (4 tests): movement, gait, reflexes, primitive reflexes (glabellar tap, snout response, suck response, palmomental reflex, grasp reflex). For clinical reasons, gait was not assessed 24 hours after extubation.
Neuropsychologic and depression assessment
A neuropsychologic assessment battery and depression inventory were administered by a neuropsychologist (R.I.N.) or neuropsychology technician (E.W.) preoperatively and 3 months postoperatively. Tests included the following:
- • Hopkins Verbal Learning Test (HVLT; total, discrimination index, and delayed recall) to measure memory
- • Controlled Oral Word Association Test (COWAT) and Boston Naming Test (BNT) to measure word fluency and confrontation naming
- • Digit Symbol and Symbol Search, subtests of the Wechsler Adult Intelligence Scale, third edition, to produce the Processing Speed Index, and Trail-Making Test (Forms A and B) to measure timed visual-motor speed
- • Grooved Pegboard (for dominant and nondominant hand) to measure timed manual dexterity
- • Beck Depression Inventory, second edition, to measure depressive symptomatology.
Level of education and intelligence were determined preoperatively by total completed years of education and reading ability, as assessed by the Wide Range Achievement Test (WRAT, third edition).
Two hundred seventy-three (78%) patients completed the 3-month neuropsychologic reassessment (the study was powered for 300 patients). Patient and procedure variables of those who were assessed were well balanced between the magnesium and placebo groups (Table E1), but those who refused reassessment were more likely to be female or to have had noncardiac surgery during follow-up (Table E2). Five (1%) patients died before reassessment.
View this table:
[in this window]
[in a new window]
|
TABLE E2. Characteristics of patients who completed 3-month neuropsychologic battery compared with patients who did not
|
|
Data Analysis
All analyses were performed with SAS statistical software (version 8.2; SAS Institute, Inc, Cary, NC).
Neurologic assessment
Longitudinal ordinal logistic regression was used to analyze 24- and 96-hour WPNS scores, with preoperative score, randomization group, and time of testing forced into the model as previously described.
17
Only a few patients had low scores, and therefore scores were collapsed into 3 categories to meet the proportional odds assumption of the model.
Neuropsychologic assessment
Twelve test scores (Digit Symbol and Symbol Search were used instead of the Processing Speed Index) were available for analysis at both preoperative and postoperative assessment. WRAT-3 reading score, preoperative scores, and randomization group, were included in all analyses.
Because neuropsychologic tests are somewhat correlated, we used principal component analysis to condense the information into a few uncorrelated variables (components), one of which was the depression inventory by itself, with minimal loss of information.
18
We applied this technique (PROC PRINCOM, SAS) to preoperative test scores of all 350 patients, identifying 4 principal components that accounted for 72% of total variation (Table E3), and used the component coefficients to calculate preoperative component scores and those at 3-month follow-up. We then performed multivariable linear regression using component scores as response variables to assess the effect of magnesium. All preoperative and intraoperative variables (Table 1) were considered in these analyses, along with interaction between age and randomization group.
Other variables
Plasma magnesium levels were measured preoperatively, during CPB, at 2 to 6 AM on postoperative day 1, and at 24 ± 6 hours. During CPB, blood flow, mean arterial pressure, temperature, and hematocrit levels were recorded. Temporal trends of these variables were similar for each group (P > .08). In addition, Cardiovascular Information Registry personnel collected preoperative, operative, and postoperative data concurrently with patient care.
Presentation
Continuous variables are presented as mean ± standard deviation and as 15th, 50th (median), and 85th percentiles when values were skewed; comparisons were made using the Wilcoxon rank-sum test. Categorical data are described using frequencies and percentages; comparisons were made using the 
2 test or Fisher's exact test when the frequency was less than 5. Actual and estimated percentages of patients with the highest neurologic scores are presented with 68% confidence limits, equivalent to ±1 standard error.
|
Results
|
|---|
Safety
There were 2 (1%) hospital deaths, 1 in each group (P = 1). However, 5 (1%) patients died within their 3-month enrollment, none from stroke or neurologic injury. Mortality was similar in the magnesium (1 death) and placebo (4 deaths) groups (P[log rank] = .2), as were other in-hospital morbidities, including transfusion requirements (Table 2). Length of intensive care unit, postoperative, and total hospital stays were similar.
Neurologic Assessment
Seven (2%) patients had a postoperative cerebrovascular accident, 2 (1%) in the magnesium group and 5 (3%) in the placebo group (P[log rank] = .3). Preoperative WPNS scores were similar in both groups (Figure 2).
View larger version (23K):
[in this window]
[in a new window]
|
Figure 2. Overall and domain scores of the Western Perioperative Neurologic Scale preoperatively and at 24 and 96 hours after extubation in the magnesium (filled circles) and placebo (open circles) groups. Shown are percentages of patients reaching the maximum overall and domain scores (scores indicated on vertical axes). Unadjusted percentages with 68% confidence bars are given for patients with complete scores. A, All 16 neurologic tests. B, Mentation (5 tests). C, Motor (4 tests). D, Sensory function (4 tests). E, Cerebellar function (3 tests).
|
|
WPNS scores decreased by 24 hours postoperatively in both groups (P < .0001); by 96 hours, the magnesium group had recovered nearly to preoperative performance, but the placebo group had not (P = .003). Thus, neurologic decline or death from neurologic injury (which did not occur within 3 months in either group), the study's composite primary end point, was less at 96 hours in the magnesium group than the placebo group (P = .01).
Other factors associated with improved WPNS were higher preoperative score, younger age, and shorter ischemic time (Table 3).
View this table:
[in this window]
[in a new window]
|
TABLE 3. Factors associated with better neurologic performance (higher Western Perioperative Neurologic Scale Scores)
*
|
|
Of the WPNS domains, sensory and motor functions appeared to be relatively unaffected in the magnesium (P > .3) and placebo (P = .07) groups. The mentation score (consciousness, orientation, short-term memory, and speech) decreased in both groups by 24 hours (P < .0005); by 96 hours, it had returned to preoperative levels (P = .1) in the magnesium group but remained depressed in the placebo group. Cerebellar function (movement, gait, reflexes, and primitive reflexes) remained at preoperative levels in the magnesium group (P > .5), but decreased in the placebo group by 24 hours and was not restored by 96 hours (P < .002); the difference between groups was unlikely to be due to chance (P = .003). Decrease in cerebellar function score was mainly due to reappearance of primitive reflexes, such as snout and suck responses and grasp reflexes (see the Methods section). The older the patient in the magnesium group, the better the cerebellar function score; in contrast, the older the patient in the placebo group, the worse the score (Figure 3
and Table 4).
View larger version (16K):
[in this window]
[in a new window]
|
Figure 3. Estimated percentage of patients with total cerebellar function score of 9 (best possible score) as a function of age. The nomogram represents a solution of the multivariable equation in Table 4 (date of operation set to June 2003). Dashed lines are estimated 68% confidence intervals.
|
|
Neuropsychologic and Depression Assessment
Neuropsychologic scores in multiple domains and depression inventory scores were generally somewhat better 3 months postoperatively than preoperatively because of familiarity with test procedures during retesting (Figure E2
and Table E4), and changes were similar between groups (P > .6). However, older patients (P = .006) and those with previous stroke (P = .003), extensive coronary disease (P = .02), or lower education level (P = .0007) performed worse at 3-month follow-up than preoperatively (Tables E5-E8).
View larger version (51K):
[in this window]
[in a new window]
|
Figure E2. Unadjusted preoperative and 3-month follow-up neuropsychologic test scores presented for magnesium and placebo groups. Top whisker, Maximum score; bottom whisker, minimum score; top line in box, 85th percentile; middle line, median; bottom line, 15th percentile; +, mean. A, Hopkins Verbal Learning Test (HVLT). B, HVLT delayed recall. C, HVLT discrimination index. D, Controlled Oral Word Association Test (COWAT). E, Boston Naming Test. F, Digit Symbol. G, Symbol Search. H, Grooved Pegboard, dominant hand (D). I, Grooved Pegboard, nondominant hand (ND). J, Trail-Making Test Form A. K, Trail-Making Test Form B. L, Beck Depression Inventory second edition.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE E4. Neuropsychologic and depression inventory performance 3 months postoperatively compared with preoperatively in entire study population
|
|
|
Discussion
|
|---|
There is great interest in developing strategies to prevent neurologic dysfunction in patients undergoing cardiac surgery.
1,2,19-21
However, rather than focusing on prevention of neurologic insult, this study addresses the issue of minimizing injury from such insults. To our knowledge, magnesium has not been used previously as a neuroprotectant in cardiac surgery. The rationale for its use is that neuronal damage after injury is attributable to release of glutamate from neurons and glia,
22
and magnesium is a noncompetitive NMDA receptor antagonist.
23
Glutamate receptor modulation has shown promise in animal and clinical studies. Remacemide, an NMDA antagonist, reduced neuropsychologic dysfunction after cardiac surgery but at the cost of important side effects (ataxia, dizziness, drowsiness, fatigue, headache, nausea, and abdominal pain).
24
Magnesium shows no such toxicity. In addition, magnesium modulates the ischemic brain injury cascade at other levels. These include abilities to inhibit neurotransmitter release
23
and cerebral vasospasm,
25
antagonize voltage-gated calcium ion channels,
26
and attenuate production of reactive oxygen species.
27
During CPB, serum magnesium levels might be depleted, and hypomagnesemia can persist for 24 hours postoperatively.
28
In animal models, brain magnesium deficiency before injury exacerbates functional deficits,
29
and limiting postinjury magnesium decline attenuates functional deficit after brain injury.
30
The level of serum magnesium required for neuroprotection is not well established, but animal studies indicate that a concentration of 3.62 mg · dL–1 reduces histologic infarct volume in ischemic brain injury.
12
In acute stroke, increasing serum magnesium levels to 1 to 2 times normal is beneficial.
11
Key Findings
Safety
No adverse effects of elevated magnesium levels were observed in this study.
Neurologic assessment
The primary end point of the study was achieved, with the better overall neurologic function in the magnesium group unlikely to be due to chance. Interestingly, the WPNS demonstrated that magnesium reduced cerebellar complications, especially return of primitive reflexes. After perioperative stroke or subtle brain injury, these primitive reflexes reappear in up to 39% of patients.
4
Their reappearance is considered a sensitive marker of brain injury in "silent" brain ischemia,
31
advanced human immunodeficiency disease,
32
neuropsychiatric disorders,
33
Alzheimer disease (where it is a poor prognostic sign),
34,35
and after cardiac surgery,
4,36
particularly in the elderly.
37
The phenomenon is complex and perhaps best interpreted as nonspecific loss of cortical control over brainstem functions.
38
Mentation, which includes short-term memory, was better preserved in patients receiving magnesium, with older patients benefiting more than younger ones. This might reflect greater brain reserve in younger patients.
We hypothesize, without supporting data, that rehabilitation of cardiac surgical patients might be enhanced if these subtle but sensitive and important deficits are reduced.
Neuropsychologic and depression assessment
Magnesium did not influence neuropsychologic function or depressive symptomatology 3 months postoperatively, but other systemic factors, including older age, did. Neuropyschologic dysfunction and depression after cardiac surgery are highly variable. Timing of assessment is crucial, with more variability in dysfunction observed early after surgery. In this study, neuropsychologic performance was generally better at 3 months than preoperatively, probably a manifestation of the test–retest phenomenon.
39
|
Limitations
|
|---|
Typical of randomized clinical studies, only a small fraction of this center's cardiac surgical population was included (Figure E1), which might introduce bias. A major deterrent to participation was need to return for 3-month follow-up neuropsychologic and depressive symptom testing. Also, because pretesting was required, patients undergoing emergency and urgent operations were not included; such patients are particularly vulnerable to neurologic injury. Recognizing this limitation, we designed the study to include not only patients undergoing CABG, as has been the case in most studies of neurologic injury after cardiac surgery,
3-8
but also patients undergoing heart valve surgery, in whom opportunity for neurologic injury from particulate matter might be even greater. CABG and valve surgery account for the majority of cardiac operations performed.
Another limitation is that neuropsychologic assessment was made only at 3 months and not earlier (when it is less stable) or longitudinally over years.
8
Despite prior agreement, a substantial number of patients refused postoperative neuropsychologic testing, which could introduce bias (Table E2). Neurologic assessment was made by using a formal, sensitive, and validated instrument.
16
However, neurologic assessment was not performed 3 months postoperatively. Thus, we do not know whether patients who manifested neurologic abnormalities before hospital discharge improved.
Finally, the study was designed to discover a possible short-term neuroprotective effect of magnesium elevation. It was not designed to assess duration of the effect or effects on convalescence or long-term outcomes.
|
Conclusions
|
|---|
Increasing serum magnesium levels during CPB and for 24 hours thereafter is safe. It offers short-term neurologic benefits, particularly in preserving short-term memory and preventing reemergence of primitive reflexes. However, patient factors primarily influenced 3-month neuropsychologic function and depressive symptomatology, not use of magnesium.
The trial suggests that future research be directed toward ascertaining duration of detectable neurologic benefit and determining whether the observed short-term benefits of elevated magnesium levels translate into helping patients recover more quickly and completely after cardiac surgery or confer long-term neurologic benefit.
|
Discussion
|
|---|
Dr Ralph J. Damiano (St Louis, Mo). Congratulations on a beautifully designed, prospective randomized study. Studies like this are sorely needed in our specialty to allow us, as surgeons, to optimize patient management. And I congratulate you on undertaking a prospective study in what has become an even more difficult and perhaps confusing area.
I have a number of questions for you, and I will ask them one at a time to give you a chance to respond.
First of all, why do you think the principal effect of magnesium was on cerebellar function? It had very little effect certainly on motor and sensory function, and even its effect on mentation was not dramatic. Is this due to some anatomic variability in the distribution of NMDA receptors? How do you explain that from a mechanistic standpoint?
Dr Bhudia. Thank you, Dr Damiano. I think the main effect of magnesium was to minimize any insult. If one was to experience any motor dysfunction, then the likelihood of a huge brain insult, as in stroke, is high. It is similar for the sensory function.
What we are picking up in the cerebellar dysfunction are subtle changes, for example the primitive reflexes. This cerebellar dysfunction was more obvious in those receiving the placebo. These subtle changes are not obvious if we do not look for them. We believe that the magnesium helps reduce the effect of these small insults to the brain rather than the larger insults, which manifest as stroke.
One of the possible mechanisms for magnesium as a neuroprotectant is through its antagonist property on the NMDA receptor. The distribution of NMDA receptors is different across the brain, but I do not think the reason for reduction in cerebellar dysfunction is related to the NMDA receptor distribution.
Dr Damiano. Is there any clinical relevance to your findings? Because there were no changes between the groups at 3 months, does the small difference in cerebellar function have any real effect in terms of patient recovery?
As a second part of this question, in your manuscript you suggested that the early benefits of magnesium on cerebellar function might accelerate patient recovery, yet length of stay in this randomized study was exactly the same. Were there any objective measures of accelerated recovery in the magnesium group?
Dr Bhudia. As for the short-term benefit, although the difference was mainly in the primitive reflexes, short-term memory was also better in those receiving magnesium. We therefore argue that although the length of stay of patients was similar in both groups, these subtle changes can influence other factors. If short-term memory is better, then getting back to routine tasks and going back to work might occur sooner, even though at 3 months we did not see any difference between the 2 groups.
Dr Damiano. But is that something you are supposing and have no objective data to support?
Dr Bhudia. No, we did not have any objective data to support it.
Dr Damiano. Also it might be of interest to the audience to point out that you are looking at fairly subtle differences in neurocognitive testing.
Dr Bhudia. That's right.
Dr Damiano. And even though there is a difference in cerebellar function, did you notice a difference in gait in these patients or anything objective that would make that subtle difference actually have a clinical effect?
Dr Bhudia. The testing of cerebellar function, and all the other modalities, was very objective. Although the changes might be subtle, they could make a difference to the overall well-being of the patient.
Dr Damiano. There certainly is a lot of literature that says that magnesium prevents atrial fibrillation, yet in your study you mentioned that there was no difference in atrial fibrillation between groups. How do you explain how this finding relates to the rest of the literature, which is fairly supportive that magnesium administration prevents atrial fibrillation?
Dr Bhudia. We were puzzled about this as well. Patients in the magnesium group had a similar incidence of atrial fibrillation as those in the placebo group, although this was not what we set out to look for as the primary outcome. Also, other postoperative outcomes, such as incidence of bleeding, infection, renal impairment, and respiratory insufficiency, were similar in the 2 groups, implying that it was safe to give magnesium at these higher levels.
One speculation as to why magnesium did not reduce the incidence of atrial fibrillation is that the highest level of serum magnesium occurred during aortic clamping. The magnesium would therefore distribute to the body, including the brain, and not the myocardium.
Dr Damiano. Finally, just quickly, at the Cleveland Clinic, is everybody receiving magnesium right now?
Dr Bhudia. That is the next step.
Dr Guo-Wei He (Hong Kong, China). Congratulations for your elegant study. I have 2 questions.
Regarding the plasma concentration of magnesium, how did you decide your magnesium level of plasma concentration should be increased to 1.5 to 2 times? How did you decide on this index?
Second, in your placebo group, do you also use the magnesium cardioplegia? I believe if you use blood cardioplegia, it contains 8 to 10 mEq of magnesium that would be absorbed into the blood also. Therefore, the question is this: Did you measure the plasma concentration of magnesium in those placebo group patients as well to make a fair comparison?
Dr Bhudia. Thank you for your comments. The question about how much magnesium to give and what level to bring it up to was one we had to answer early on. Some animal studies have shown that a serum magnesium level of 3.6 mg · dL–1, approximately 1.5 times normal range in human subjects, is the neuroprotective dose. That is how we decided on 3.6 mg · dL–1 to 4.8 mg · dL–1. The upper limit, 4.8 mg · dL–1, was high, but not high enough to cause side effects.
Regarding whether we measured magnesium levels in the placebo group, the answer is yes. One of our graphs showed that the 2 curves of magnesium levels were significantly different.
|
Appendix E1
|
|---|
Details of Trial Design Hypothesis
Keeping serum magnesium levels greater than 1 times adult normal range during CPB and 24 hours thereafter in patients undergoing cardiac surgery has no neuroprotective effect.
Trial Design
Randomized, blinded, placebo-controlled, single-site clinical study in which patients undergoing elective cardiac surgery were randomized to have serum magnesium levels increased to 1 to 2 times normal level versus placebo. The trial included the following interactions with patients, interventions, and assessments:
- 1 Three to 4 days before surgery: consent, followed by neurologic, neuropsychologic, and depressive assessment
- 2 Day of operation: randomization to treatment or placebo
- 3 Induction of anesthesia: intravenous bolus of magnesium sulfate or placebo over 15 minutes, followed by infusion over 24 hours
- 4 Off CPB: possible magnesium sulfate bolus of 1 g (surgeon practice); this is given occasionally to prevent cardiac arrhythmias, with the surgeon making the decision
- 5 24 hours after surgery: magnesium sulfate or placebo discontinued
- 6 24 hours after extubation: neurologic assessment
- 7 Extubation: neurologic assessment
- 8 96 hours after surgery: neurologic assessment
- 9 90 days after surgery: neuropsychologic and depressive inventory assessment.
Patient Population
Patients were recruited from the 7-county Cleveland region if they were admitted to Cleveland Clinic for elective on-pump cardiac surgery. Other patients from outside this region who were willing to come back for 3-month assessment were also recruited.
Inclusion Criteria
Patients 18 years or older scheduled to undergo elective CABG, valve surgery, or both, on CPB who consented for the trial.
Exclusion Criteria
- 1 Preoperative atrial fibrillation.
- 2 Preoperative renal impairment, defined as plasma creatinine level greater than 2.0 or on chronic renal dialysis.
- 3 Participation in another trial.
- 4 Inability to read, write, or comprehend English.
- 5 Refusal to consent.
In toto, other than refusal to consent, 29 patients were excluded: 1 with diabetic retinopathy was unable to read, 1 was unable to understand instructions, 2 discontinued the assessment (withdrew consent), 2 had a baseline assessment but then withdrew consent, 5 had baseline assessment but either had no operation or a percutaneous procedure, and 18 underwent an operation more extensive than the one contemplated at the time of baseline assessment that did not meet inclusion criteria.
Sample Size
Neurologic decline or death from neurologic injury was anticipated to be 35% in the control group. Assuming this would be 20% in the treatment group, a 2-group test of independent proportions, a type I error of 0.05, and a 2-tailed alternative, a total sample size of 300 would achieve at least 80% power to detect this difference. This was augmented by 50 patients to guard against overly optimistic anticipated differences and the possibility that not all consenting patients would complete all tests.
Magnesium Dose
Magnesium dosing was derived from the knowledge that (1) serum magnesium concentrations 1 times normal (3.62 mg · dL–1) were associated with reduced histologic infarct volume in ischemic injury animal models,
E1
(2) concentrations 2 to 3 times normal (4.2 mg · dL–1 to 6.8 mg · dL–1) with reduced occurrence of preeclamptic and eclamptic seizure,
E2
and (3) concentrations 1 to 2 times normal with important clinical benefit in acute stroke.
E3
|
Appendix E2
|
|---|
Analyses of Neuropsychologic and Depressive Inventory Scores
Principal components analysis of preoperative test scores identified 4 components accounting for 72% of total variation: principal component 1 for 42%, 2 for 13%, 3 for 9%, and 4 for 8%. These 4 components can be interpreted as follows (Table E3):
- 1 Weighted average of all tests except Beck Depression Inventory, second edition. A lower score signifies worse performance.
- 2 Mainly represents average of 3 trials of the Hopkins Verbal Learning Test scores. A lower score signifies worse performance.
- 3 Represents Beck Depression Inventory, second edition. A higher score signifies higher levels of depressive symptomatology.
- 4 Mainly represents the Boston Naming Test and Controlled Oral Word Association Test (word fluency). A lower score signifies worse performance.
|
Footnotes
|
|---|
Read at the Eighty-fifth Annual Meeting of The American Association for Thoracic Surgery, San Francisco, Calif, April 10-13, 2005.
|
References
|
|---|
- Cosgrove DM, Loop FD, Lytle BW, Baillot R, Gill CC, Golding LA, et al. Primary myocardial revascularization. Trends in surgical mortality. J Thorac Cardiovasc Surg. 1984;88:673-684.[Abstract]
- Estafanous FG, Loop FD, Higgins TL, Tekyi-Mensah S, Lytle BW, Cosgrove 3rd DM, et al. Increased risk and decreased morbidity of coronary artery bypass grafting between 1986 and 1994. Ann Thorac Surg. 1998;65:383-389.[Abstract/Free Full Text]
- Roach GW, Kanchuger M, Mangano CM, Newman M, Nussmeier N, Wolman R, et al. Multicenter Study of Perioperative Ischemia Research Group and the Ischemia Research and Education Foundation Investigators Adverse cerebral outcomes after coronary bypass surgery. N Engl J Med. 1996;335:1857-1863.[Abstract/Free Full Text]
- Shaw PJ, Bates D, Cartlidge NE, Heaviside D, Julian DG, Shaw DA. Early neurological complications of coronary artery bypass surgery. BMJ. 1985;291:1384-1387.[Abstract/Free Full Text]
- Hammon Jr JW, Stump DA, Kon ND, Cordell AR, Hudspeth AS, Oaks TE, et al. Risk factors and solutions for the development of neurobehavioral changes after coronary artery bypass grafting. Ann Thorac Surg. 1997;63:1613-1618.[Abstract/Free Full Text]
- Selnes OA, Goldsborough MA, Borowicz LM, McKhann GM. Neurobehavioural sequelae of cardiopulmonary bypass. Lancet. 1999;353:1601-1606.[Medline]
- Shaw PJ, Bates D, Cartlidge NE, French JM, Heaviside D, Julian DG, et al. Long-term intellectual dysfunction following coronary artery bypass graft surgery. a six month follow-up study. Q J Med. 1987;62:259-268.[Medline]
- Newman MF, Kirchner JL, Phillips-Bute B, Gaver V, Grocott H, Jones RH, et al. Longitudinal assessment of neurocognitive function after coronary-artery bypass surgery. N Engl J Med. 2001;344:395-402.[Abstract/Free Full Text]
- Sabik JF, Lytle BW, McCarthy PM, Cosgrove DM. Axillary artery. an alternative site of arterial cannulation for patients with extensive aortic and peripheral vascular disease. J Thorac Cardiovasc Surg. 1995;109:885-891.[Abstract]
- Banbury MK, Kouchoukos NT, Allen KB, Slaughter MS, Weissman NJ, Berry GJ, et al. Emboli capture using the Embol-X intraaortic filter in cardiac surgery. a multicentered randomized trial of 1,289 patients. Ann Thorac Surg. 2003;76:508-515.[Abstract/Free Full Text]
- Muir KW, Lees KR. A randomized, double-blind, placebo-controlled pilot trial of intravenous magnesium sulfate in acute stroke. Stroke. 1995;26:1183-1188.[Abstract/Free Full Text]
- Marinov MB, Harbaugh KS, Hoopes PJ, Pikus HJ, Harbaugh RE. Neuroprotective effects of preischemia intraarterial magnesium sulfate in reversible focal cerebral ischemia. J Neurosurg. 1996;85:117-124.[Medline]
- Thurnau GR, Kemp DB, Jarvis A. Cerebrospinal fluid levels of magnesium in patients with preeclampsia after treatment with intravenous magnesium sulfate. a preliminary report. Am J Obstet Gynecol. 1987;157:1435-1438.[Medline]
- Muir KW, Lees KR. Dose optimization of intravenous magnesium sulfate after acute stroke. Stroke. 1998;29:918-923.[Abstract/Free Full Text]
- Banbury MK, White JA, Blackstone EH, Cosgrove 3rd DM. Vacuum-assisted venous return reduces blood usage. J Thorac Cardiovasc Surg. 2003;126:680-687.[Abstract/Free Full Text]
- Murkin JM, Martzke JS, Buchan AM, Bentley C, Wong CJ. A randomized study of the influence of perfusion technique and pH management strategy in 316 patients undergoing coronary artery bypass surgery. II. Neurologic and cognitive outcomes. J Thorac Cardiovasc Surg. 1995;110:349-362.[Abstract/Free Full Text]
- McCarthy PM, Bhudia SK, Rajeswaran J, Hoercher KJ, Lytle BW, Cosgrove DM, et al. Tricuspid valve repair. durability and risk factors for failure. J Thorac Cardiovasc Surg. 2004;127:674-685.[Abstract/Free Full Text]
- Johnson RA, Wichern DW. Applied multivariate statistical analysis. 5th ed.. Upper Saddle River (NJ): Prentice Hall; 2002.
- Gilman S. Cerebral disorders after open-heart operations. N Engl J Med. 1965;272:489-498.[Medline]
- Blackstone EH. Neurologic injury from cardiac surgery—an important but enormously complex phenomenon. J Thorac Cardiovasc Surg. 2000;120:629-631.[Free Full Text]
- Fox HM, Rizzo ND, Gifford S. Psychological observations of patients undergoing mitral surgery; a study of stress. Psychosom Med. 1954;16:186-208.[Abstract/Free Full Text]
- Obrenovitch TP, Urenjak J. Altered glutamatergic transmission in neurological disorders. from high extracellular glutamate to excessive synaptic efficacy. Prog Neurobiol. 1997;51:39-87.[Medline]
- Mayer ML, Westbrook GL, Guthrie PB. Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature. 1984;309:261-263.[Medline]
- Arrowsmith JE, Harrison MJ, Newman SP, Stygall J, Timberlake N, Pugsley WB. Neuroprotection of the brain during cardiopulmonary bypass. a randomized trial of remacemide during coronary artery bypass in 171 patients. Stroke. 1998;29:2357-2362.[Abstract/Free Full Text]
- Ema M, Gebrewold A, Altura BT, Zhang A, Altura BM. Alcohol-induced vascular damage of brain is ameliorated by administration of magnesium. Alcohol. 1998;15:95-103.[Medline]
- Iseri LT, French JH. Magnesium. nature's physiologic calcium blocker. Am Heart J. 1984;108:188-193.[Medline]
- Kowaltowski AJ, Naia-da-Silva ES, Castilho RF, Vercesi AE. Ca2+-stimulated mitochondrial reactive oxygen species generation and permeability transition are inhibited by dibucaine or Mg2+
. Arch Biochem Biophys. 1998;359:77-81.[Medline]
- Parra L, Fita G, Gomar C, Rovira I, Marin JL. Plasma magnesium in patients submitted to cardiac surgery and its influence on perioperative morbidity. J Cardiovasc Surg (Torino). 2001;42:37-42.[Medline]
- Saatman KE, Bareyre FM, Grady MS, McIntosh TK. Acute cytoskeletal alterations and cell death induced by experimental brain injury are attenuated by magnesium treatment and exacerbated by magnesium deficiency. J Neuropathol Exp Neurol. 2001;60:183-194.[Medline]
- McIntosh TK, Faden AI, Yamakami I, Vink R. Magnesium deficiency exacerbates and pretreatment improves outcome following traumatic brain injury in rats. 31P magnetic resonance spectroscopy and behavioral studies. J Neurotrauma. 1988;5:17-31.[Medline]
- Rao R, Jackson S, Howard R. Primitive reflexes in cerebrovascular disease. a community study of older people with stroke and carotid stenosis. Int J Geriatr Psychiatry. 1999;14:964-972.[Medline]
- Tremont-Lukats IW, Serbanescu R, Teixeira GM, Iriza E, Hernandez DE, Schneider C. Multivariate analysis of primitive reflexes in patients with human immunodeficiency virus type-1 infection and neurocognitive dysfunction. Ital J Neurol Sci. 1999;20:17-22.[Medline]
- Walterfang M, Velakoulis D. Cortical release signs in psychiatry. Aust N Z J Psychiatry. 2005;39:317-327.[Medline]
- Molsa PK, Marttila RJ, Rinne UK. Long-term survival and predictors of mortality in Alzheimer's disease and multi-infarct dementia. Acta Neurol Scand. 1995;91:159-164.[Medline]
- Franssen EH, Reisberg B. Neurologic markers of the progression of Alzheimer's disease. Int Psychogeriatr. 1997;9(suppl 1):297-306discussion 17-21.[Medline]
- Heyer EJ, Delphin E, Adams DC, Rose EA, Smith CR, Todd CJ, et al. Cerebral dysfunction after cardiac operations in elderly patients. Ann Thorac Surg. 1995;60:1716-1722.[Abstract/Free Full Text]
- Kobayashi S, Yamaguchi S, Okada K, Yamashita K. Primitive reflexes and MRI findings, cerebral blood flow in normal elderly. Gerontology. 1990;36:199-205.[Medline]
- Schott JM, Rossor MN. The grasp and other primitive reflexes. J Neurol Neurosurg Psychiatry. 2003;74:558-560.[Abstract/Free Full Text]
- Kneebone AC, Andrew MJ, Baker RA, Knight JL. Neuropsychologic changes after coronary artery bypass grafting. use of reliable change indices. Ann Thorac Surg. 1998;65:1320-1325.[Abstract/Free Full Text]
|
References
|
|---|
- Marinov MB, Harbaugh KS, Hoopes PJ, Pikus HJ, Harbaugh RE. Neuroprotective effects of preischemia intraarterial magnesium sulfate in reversible focal cerebral ischemia. J Neurosurg. 1996;85:117-124.[Medline]
- Thurnau GR, Kemp DB. Cerebrospinal fluid levels of magnesium in patients with preeclampsia after treatment with intravenous magnesium sulfate. a preliminary report. Am J Obstet Gynecol. 1997;158:1435-1438.
- Muir KW, Lees KR. Dose optimization of intravenous magnesium sulfate after acute stroke. Stroke. 1998;29:918-923.[Abstract/Free Full Text]
Top
Abstract
Introduction
