HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Charlson, M. E. Articles by Gold, J. P. Search for Related Content PubMed Articles by Charlson, M. E. Articles by Gold, J. P.

J Thorac Cardiovasc Surg 1997;113:1118-1120
© 1997 Mosby, Inc.

LETTERS TO THE EDITOR

Improvement of outcomes after coronary artery bypass

Albert Einstein College of Medicine
Montefiore Medical Center
East 210th St.
Bronx, NY 10467

Reply to the Editor:

The members of the Cornell Coronary Artery Bypass Outcomes Trial (CCABOT) Group appreciate the opportunity to expand on the points raised by Reves and colleagues. The authors have raised three major concerns. The first issue is the "high event" rate. The overall rate of stroke in our cohort was 4.8%, with 7.2% occurring in the control group. Few studies have formally trained neurologic observers blinded to intraoperative randomization performing serial evaluations, as were done in our cohort. Our study neurologists performed neurologic evaluations before the operations, 24 to 28 hours after the operations, 6 days after the operations, and 6 months after the operations. Such uniform surveillance has been shown unequivocally to increase detection rates. ¹

In many ways, our patients were similar demographically and clinically to those in the Bypass Angioplasty Revascularization Investigation (BARI) trial. ² However, we now have evidence that the stroke rate is primarily a function of advanced aortic atheromatous disease as imaged by transesophageal echocardiography. In our study, among patients with atheroma grades I or II, no strokes had occurred at 1 week after the operations, whereas the 1-week stroke rates were 5.5%, 10.5%, and 45.5% for grades 1II, IV, and V atheroma, respectively. ³ Thus, if the BARI trial had higher proportions of patients with low-grade atheroma than our trial, this difference in case mix would explain the differences in rates of stroke. It should be emphasized that high mean arterial pressure and atheroma grade are independently related to outcome.

The issue of multiple comparisons is widely known; our colleagues, however, have provided contradictory arguments about this. They first argue that there is no license to test each outcome at the 0.05 level. The original sample size of 248 was calculated to take into account aggregate outcomes, a total complication rate (including mortality, cardiac, neurologic, and cognitive complications, and deterioration in functional status) in the range of 35%, and an estimated delta at 20%. That is, a 20% drop in the incidence of any of the adverse outcomes would constitute a clinically important difference between the two management strategies. At the conclusion of our trial, confidence intervals were provided for each of the specific outcomes, as well as the aggregate outcomes, and these provide the best way to assess the clinical and statistical significance of the findings. ⁴ Because stroke was the most common adverse event, it received the greatest emphasis in the analysis.

Second, they argue that the outcomes are clearly strongly related and not mutually exclusive. We agree. In fact, that is precisely why we provided the analysis of the combined cardiac, neurologic, and mortality rates, and planned to do so from the outset. After arguing that the outcomes are strongly related and not mutually exclusive, the writers argue against the combination of these outcomes and suggest that this combination was post hoc. As noted previously, this was planned from the outset, for that very reason. In fact, we have presented the data with a formal statistical analysis both ways.

Reves and coworkers comment on the mechanistic issue relating mean arterial pressure and cerebral perfusion. The data on the relationship between cerebral perfusion pressure (CPP) and cerebral blood flow (CBF) (i.e., autoregulation) are predominantly accumulated from the studies involving global blood flow using the Fick principle. Reves and coworkers have extensively contributed to the literature on the variability of individual blood flow-pressure relationships and the importance of regional factors in the development of neurologic injury after cardiopulmonary bypass (CPB). These comments are reflected in the following passages from their research group*. "The lower limit of the autoregulatory threshold in individual hypertensive patients will vary: therefore, a higher perfusion pressure on CPB may be required to assure adequate cerebral perfusion." ⁵ They further comment: "Although global CBF remained intact in these studies, the presence of extracranial and intracranial obstructive vascular disease in cardiac surgery patients, and its impact on care, await further investigation and the development of better techniques to assess regional CBF and metabolism during CPB." ⁶ "The mechanism(s) of increased neurologic risk, which again is unknown, may involve increased microemboli, cerebrovascular disease, and/or aberrations in cerebral autoregulation with incomplete global ischemia." ⁷ They add, "in those patients with altered autoregulation (chronic hypertension, insulin-dependent diabetes) or severe cerebrovascular disease, the lower level of CPP at which a change in CPP effects a change in CBF is unknown. The absolute CPP and pump flows necessary to avoid the need for compensatory mechanisms is likely to vary between patients and be altered by other factors such as temperature, hematocrit and anesthetic state," ⁸ and that "the data apply only to patients managed with alpha-stat blood gas strategy and to patients without symptoms or history of hypertension and neurovascular disease." ⁹

Our findings and commentary agree with many of these statements and were directed at the lack of data on regional arterial blood flow during CPB as reflected in the work of Reves and coworkers, as well as others. We sincerely appreciate the opportunity to further clarify and expand on these areas, as well as to address the specific methodologic concerns that were raised. It is only through continued critique and collaboration that these important areas can be more fully explored.

12/8/79444

References

1. Charlson ME, MacKenzie CR, Ales K, Gold JP, Fairclough G, Shires GT. Surveillance for postoperative myocardial infarction after noncardiac operations. Surg Gynecol Obstet 1988;167:407-14.[Medline]
   
2. Bypass Angioplasty Revascularization Investigation (BARI) Investigators. Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. N Engl J Med 1996:335:217-25.
   
3. Hartman GS, Yao FSF, Bruefact M, Barbut D, Peterson JC, Purcell MH, et al. Severity of aortic atheromatous disease diagnosed by transesophageal echocardiography predicts stroke and other outcomes associated with coronary artery surgery: a prospective study. 1996;83:701-8.
   
4. Braitman LE. Confidence intervals assess both clinical significance and statistical significance. Ann Intern Med 1991;114-515.
   
5. Schell RM, Kern FH, Greeley WJ, Schulman SR, Frasco PE, Croughwell ND, et al. Cerebral blood flow and metabolism during cardiopulmonary bypass. Anesth Analg 1993;76:854.
   
6. Schell RM, Kern FH, Greeley WJ, Schulman SR, Frasco PE, Croughwell ND, et al. Cerebral blood flow and metabolism during cardiopulmonary bypass. Anesth Analg 1993;76:855.
   
7. Schell RM, Kern FH, Greeley WJ, Schulman SR, Frasco PE, Croughwell ND, et al. Cerebral blood flow and metabolism during cardiopulmonary bypass. Anesth Analg 1993;76:854-5.
   
8. Schell RM, Kern FH, Greeley WJ, Schulman SR, Frasco PE, Croughwell ND, et al. Cerebral blood flow and metabolism during cardiopulmonary bypass. Anesth Analg 1993;76:858.
   
9. Aladi LJ, Croughwell N, Smith LR, Reves JG. Cerebral blood flow autoregulation is preserved during cardiopulmonary bypass in isoflurane-anesthetized patients. Anesth Analg 1991;72:52.
   

This Article Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Charlson, M. E. Articles by Gold, J. P. Search for Related Content PubMed Articles by Charlson, M. E. Articles by Gold, J. P.