Key result
This commentary discusses conflicting evidence regarding the association between intraoperative mean arterial pressure targets and postoperative stroke during cardiac surgery.
Answers are grounded in this paper and cite their sources.
This editorial emphasizes the need to consider unmeasured covariates, such as aortic atherosclerosis and CPB technique variations, when interpreting retrospective associations between intraoperative hypotension and postoperative stroke.
I have long been concerned that anesthesiologists, at least occasionally, are unwittingly accepting intraoperative mean arterial pressures (MAPs) that may bring the central nervous system very close to the thresholds for ischemic injury.1 The recent report by Sun et al. offered a caution about intraoperative hypotension and was therefore a result that I was inclined to welcome.2 They reported, on the basis of a retrospective examination of a large electronic database, an association between MAPs less than 65 mmHg before, during, or after cardiopulmonary bypass (CPB) and the occurrence of postoperative stroke. However, in spite of my biases, I found myself with some uncertainties about the strength of their observations. I am concerned about unmeasured (or merely unreported) covariates. In part, the prompt for that concern was the nearly simultaneous publication of an investigation by Vedel et al. (about which Sun et al. inevitably had no opportunity to comment).3 Vedel et al. assigned cardiac surgery patients to MAPs of either 40 to 50 or 70 to 80 mmHg on CPB. The investigation was randomized and prospective. The CPB parameters, which were reported in detail, were common to the two groups, with the exception of pressors and inotropes in the latter group. The primary endpoint was the volume of new diffusion weighted imaging lesions in the postoperative period. There were no differences between the groups in that endpoint or in several secondary endpoints. While relatively small numbers (98 and 99 patients per group) might be suspected of having resulted in type II statistical errors, every trend (diffusion weighted imaging lesion volume, stroke, cognitive dysfunction at the time of discharge, mortality) was in favor of the low pressure group.I write to seek the opinions of Sun et al. as to possible explanations for the apparently contradictory results of the two studies. Their retrospective trial reports a stroke association with MAP less than 65 mmHg while the prospective, randomized trial reports no disadvantage to a MAP of 40 to 50 mmHg. That prompted my concern about covariates. I have several questions. First, why was there so much MAP variation, especially during CPB, within one cardiac surgery group? Were the lower MAPs in some patients a function of the perceived fragility of the aorta? The references cited by Sun et al. confirm the importance of aortic atherosclerosis as a conspicuous stroke risk factor.4,5 Was an assessment of the severity of atherosclerotic disease of the aortic arch performed and recorded in their patients? Sun et al. provide little information about CPB technique and/or its variation among practitioners and over time. Were differences in the practices of individual surgeons, anesthesiologists, and perfusionists contributors to outcome differences among patients? More specifically, were the CPB techniques the same for all patients? As a reflection of the possible influence of variations in CPB technique, was the stroke rate constant over the 6-yr study period? I hope that this letter will present Sun et al. an opportunity to strengthen their work by providing information about potential covariates.As a minor additional concern, to which I seek no response, I think that there are some references to the literature that are misrepresentations of the cited papers, or at a minimum will be misunderstood. An assertion offered in the Introduction is one such: "Optimal blood pressure thresholds for stroke prevention have been described as MAP within 30% of baseline in a noncardiac, nonneurologic surgery cohort,6 and…MAP of greater than 66 mmHg…during CPB.7" Neither of the cited papers makes recommendations about target MAPs for the prevention of stroke. Bijker et al. reported an association between stroke and intraoperative MAP greater than 30% below baseline.6 However, they opined that "these associations should be interpreted cautiously and do not automatically imply causality." Joshi et al. identified the average lower limit of autoregulation on CPB to be 66 mmHg (notably, with a very wide CI and a poor correlation between baseline MAP and lower limit of autoregulation).7 Their conclusion was that MAP targets based on clinical guestimates of a probable lower limit of autoregulation are likely to be frequently very wide of the mark. However, they did not make MAP recommendations. While my comments may appear to some to be mere caviling over semantics, my experience has been that too many third party reviewers take hold of misstatements of this nature in the published literature and, in the face of adverse outcomes, make them the basis of standard of care criticisms of clinicians. The semantics matter.The author declares no competing interests.
No takes yet. Share an insight, caveat, or question.
John C. Drummond (2019) conducted a letter in Cardiac surgery. Intraoperative hypotension was evaluated. This commentary discusses conflicting evidence regarding the association between intraoperative mean arterial pressure targets and postoperative stroke during cardiac surgery.
Synapse has enriched 10 related papers. Consider them for comparative context: