Anti-inflammatory effects of perioperative statin therapy

Perioperative physicians seek to reduce the risk of adverse perioperative events, focusing in particular on adverse cardiovascular events such as myocardial infarction (MI) and cardiac death. To this aim, extensive and often invasive intra- and postoperative hemodynamic monitoring have become the standard of care. Obviously, monitoring in itself does not prevent adverse events as over 5% of high-risk surgical patients still suffer from perioperative myocardial ischemia and infarction.1,2 Therefore, efforts have been undertaken to reduce the risk of perioperative MI by risk stratification during preoperative assessment and subsequent initiation of preventive medical treatment early before surgery in patients identified as high-risk.3 Preventive medical treatments that have been investigated include preoperative initiation of beta-blockers, alpha2-agonists, acetylsalicylic acid, and statins.2-6 The effect of initiating prophylactic treatment with beta-blockers to reduce perioperative MI seems counterbalanced by the occurrence of other major adverse events, such as stroke and death.2 The results of studies evaluating perioperative prophylactic treatment of high-risk patients with acetylsalicylic acid and alpha2-agonists seem encouraging, but current evidence is insufficient to advocate their widespread use.4 Statins were considered another promising class of drugs in the prevention of perioperative MI among high-risk surgical patients with cardiovascular risk factors.5,6

Perioperative physicians seek to reduce the risk of adverse perioperative events, focusing in particular on adverse cardiovascular events such as myocardial infarction (MI) and cardiac death. To this aim, extensive and often invasive intra-and postoperative hemodynamic monitoring have become the standard of care. Obviously, monitoring in itself does not prevent adverse events as over 5% of highrisk surgical patients still suffer from perioperative myocardial ischemia and infarction. 1,2 Therefore, efforts have been undertaken to reduce the risk of perioperative MI by risk stratification during preoperative assessment and subsequent initiation of preventive medical treatment early before surgery in patients identified as high-risk. 3 Preventive medical treatments that have been investigated include preoperative initiation of beta-blockers, alpha 2agonists, acetylsalicylic acid, and statins. [2][3][4][5][6] The effect of initiating prophylactic treatment with beta-blockers to reduce perioperative MI seems counterbalanced by the occurrence of other major adverse events, such as stroke and death. 2 The results of studies evaluating perioperative prophylactic treatment of high-risk patients with acetylsalicylic acid and alpha 2 -agonists seem encouraging, but current evidence is insufficient to advocate their widespread use. 4 Statins were considered another promising class of drugs in the prevention of perioperative MI among high-risk surgical patients with cardiovascular risk factors. 5,6 Since the mid-1990s, when the results of the Scandinavian Simvastatin Survival Study (4S) were published, statins have become a cornerstone in the secondary prevention of cardiovascular disease. 7,8 Statins are particularly recommended for those patients with cardiovascular disease who do not meet the lipid-lowering goals through lifestyle approaches, as statins effectively lower cholesterol levels and decrease mortality by decreasing the incidence of MI and stroke. 8 Beyond lipid-lowering activity in the prevention of atherosclerosis, statins exhibit action by improving vascular endothelial function, modulating inflammatory responses, and maintaining plaque stability, thereby preventing thrombus formation. These so-called ''pleiotropic'' effects of statins are believed to occur within 24 hr after statin initiation and prior to the reduction in serum cholesterol levels (weeks). 9 The rapid onset of the pleiotropic effects was considered potentially useful to prevent perioperative MI, as plaque instability / disruption, most likely associated with perioperative inflammation, has been recognized as a relevant cause of MI that is potentially responsible for up to 50% of perioperative MIs. 10 If statin therapy can effectively diminish the inflammatory response to surgical trauma, a perioperative MI might be prevented. This effect would be comparable with maintaining plaque stability after acute coronary syndrome in the nonsurgical setting.
Several retrospective and nonrandomized studies showed that statin use is indeed associated with reduced mortality. 11 A small number of randomized clinical trials examining the effect of statins on cardiovascular outcome after noncardiac surgery have been published. In 2004, Durazzo et al. reported a threefold reduction (from 26% to 8%) in the occurrence of an adverse cardiovascular event or death within six months after surgery in a high-risk population of 100 patients undergoing vascular surgery. 6 In this trial, patients were given atorvastatin or placebo for an average of 30 days before surgery. The DECREASE III trial, which included 497 vascular surgery patients who were on placebo or fluvastatin for a median of 37 days, found comparable results (reduction in adverse events from 10% to 5%). 5 The DECREASE IV study was an open-label two-by-two factorial trial designed to include 6,000 intermediate cardiac risk patients who were assigned to bisoprolol, fluvastatin, combination treatment, or control therapy before surgery for a median of 34 days. 12 The trial was terminated early after inclusion of only 1,066 patients because of slow patient recruitment. Fluvastatin therapy was deemed not superior compared with placebo in preventing cardiac death or MI. Mainly based on these three studies, the 2009 European Society of Cardiology Guidelines (2009 Guidelines) recommended initiation of perioperative statin therapy in high-risk surgical patients from 30 to 7 days before surgery (Class I recommendation, level of evidence B). 3 However, it is often difficult to initiate drug therapy a week or more before surgery. Moreover, as the antiinflammatory effects of statins occur within 24 hr after initiation, statin treatment started shortly before surgery may be equally effective. Accordingly, Neilipovitz et al. studied whether a clinically relevant effect on inflammation, as assessed by pre-and postoperative C-reactive protein (CRP), could be detected if statins were initiated shortly before surgery. 13 In this issue of the Journal, they report the results of an interesting study designed to assess the anti-inflammatory effect of short-term atorvastatin use in patients at high cardiac risk undergoing noncardiac surgery, but without previous statin administration. The authors initiated the statins a week prior to surgery or on the day of surgery. Herein, the authors hypothesized that anti-inflammatory benefits of perioperative statins, as assessed by CRP, would be apparent within 48 hr after surgery and would be comparable with the benefit obtained in the nonsurgical setting (i.e., a decrease in postoperative CRP levels of 33%). Two active treatment groups, one in which treatment was initiated seven days before surgery (n = 26) and one in which treatment was initiated the day of surgery (n = 16), were compared with a placebo group (n = 17). Treatment was continued for seven days or until the time of discharge. The patients were a reasonably highrisk group and there was a predominance of patients undergoing vascular surgery. Unfortunately, 79% of the 1,037 eligible patients were on statins already, and patients had to be allocated to the treatment arms in unequal ratios due to serious logistical problems in timely enrollment. Furthermore, there was an uneven distribution of patients over the three small-sized trial arms, which may have influenced the results. Nevertheless, there were no significant differences among the groups for CRP values at any time. Moreover, the direction of the differences in mean CRP levels was not as expected as these levels were lowest in the placebo group even though these patients were certainly comparable with respect to preexisting comorbidity and underwent identical procedures.
The strength of the trial by Neilipovitz et al. lies in the fact that the authors evaluated both the clinical applicability of the recommendations of the 2009 Guidelines on initiating perioperative statin therapy and the effects of statins on the surgical stress response, i.e., whether statins effectively attenuate the normally occurring increase in postoperative CRP levels.
There are two important results. First, the recommendations of the 2009 Guidelines on the timing of statin initiation did not appear to be feasible, as many patients were scheduled for surgery within seven days after presenting at the outpatient clinics. One could argue that surgery can be postponed to make the guidelines applicable, but this is only reasonable if postponing surgery and initiating statin therapy makes an important difference in outcome. Thus far, this evidence does not seem strong enough. Moreover, it seems that many statin naïve patients should have been on statins because of their underlying disease, irrespective of the scheduled surgery. In fact, a study such as this trial ''catches'' only those patients in whom secondary prevention of cardiovascular diseases failed, at least partly, for whatever reason. Maybe the question should not be whether these patients should be on statins, perioperatively or permanently, but perhaps the question should be whether these patients should have been on a statin already. Neilipovitz et al. considered the trial as a pilot before committing to a large multicentre trial. However, apart from potentially serious problems with recruiting a sufficient number of patients for such a trial within a reasonable time period, the ethics of withholding statins from patients who otherwise should likely receive these drugs should be carefully considered.
Second, the authors tried to unravel further the underlying mechanism of action of perioperative statin therapy, i.e., the potential attenuation of the inflammatory response as a precursor for coronary plaque instability and rupture. Although limited by the small sample size and uneven distribution of patients over the trial arms, the results of this study suggest that a possible positive effect of statins on postoperative outcome is not achieved by reducing perioperative inflammation as measured by postoperative CRP levels. This outcome measure is of considerable importance as previous studies (e.g., DECREASE III) used CRP values taken immediately preoperatively, assuming that a low CRP level in treated patients before surgery implicated adequate treatment effects. 5 Although no dif-ferences in CRP levels were found in the current trial at any time, this does not necessarily imply that reduced inflammation in atherosclerotic plaques does not play a role. As far as we know, it is unknown whether systemic CRP levels do adequately reflect local inflammatory activity in plaques during recovery from such a major event as the surgical trauma. In other words, is measuring postoperative CRP sensitive and specific enough to reflect local effects of statins? Furthermore, the other atherosclerotic plaque stabilizing effects, such as increased expression of endothelial nitric oxide synthase, reduced production of reactive oxygen species, and improved thrombogenic profile, may be responsible for the reported beneficial effects of perioperative statin therapy. However, the evidence for these potentially beneficial effects is fairly poor.
As described above, with respect to statin therapy, the 2009 Guidelines are based mainly on three small-sized trials, one of which was terminated early. If the early terminated DECREASE IV study-which by the way did not show a benefit of statins-is not taken into account, only the trial by Durazzo et al. and the DECREASE III study remain. Within the past few months, the principal investigator of the latter study is no longer active because of scientific misconduct. Although one might argue that previous results reported by this researcher therefore become implausible, until now, an investigational committee instituted by the host institution has not recommended retracting any research paper. Thus, two clinical trials on perioperative statin use, including 597 patients in total, remain herein as a basis for the 2009 Guidelines Class I level B recommendation. Apparently, the 2009 Guidelines were based partly on intuitively expected effects of perioperative statins on postoperative outcome. The pilot study by Neilipovitz et al. suggests that this expected effect may not act via reducing the perioperative inflammatory response.
The available evidence warrants further research to elucidate both the expected effect (is there any) as well as the mechanism of action of perioperative statin therapy. Such studies need to be powered adequately, not only on clinical outcome but also on adverse effects. The 2009 Guidelines seem to consider the risk of serious adverse effects of statins (myopathy and rhabdomyolysis) acceptable; nevertheless, there is a need for studies that include an estimation of the incidence of these rare but disastrous effects during perioperative statin therapy. For now, as recommended by the 2009 Guidelines, those patients who are on statins already should continue taking them perioperatively. For those patients who are not on statins, perioperative physicians should ensure that they are treated adequately for cardiovascular disease or risk factors, irrespective of surgery.
Funding Departmental sources only.

Competing interests None declared.
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.