Intra-aortic balloon counterpulsation in cardiogenic shock: Is it really the end of an indication?

The lack of hard evidence to support the routine use of intra-aortic balloon counterpulsation (IABP) in cardiogenic shock has resulted in a class-IIIa recommendation in the newest European guidelines (1). Thus, its use is not only unsupported, but actively discouraged. A level-III recommendation indicates “evidence or general agreement that the given treatment or procedure is not useful/effective, and in some cases may be harmful.” Also, the evidence is judged to be based on “multiple randomized clinical trials or meta-analyses.” In our opinion, such a classification of IABP in cardiogenic shock gives the IABP-SHOCK II trial too much impact, particularly considering the shortcomings of the trial and the positive observational support given for IABP in various other studies (2–5). A detailed discussion of whether this downgrading is supported by available data can be found in the online version and discussion of a much referred meta-analysis published in the European Heart Journal (4). Of particular interest, a similarly constructed meta-analysis (5) indicates that there are more positive merits to the IABP in cardiogenic shock than that suggested in the analysis from Sjauw et al. (4).

In our highly specialized health care system with all therapeutic options available, a patient in cardiogenic shock still faces a 30-day mortality risk close to 50%. What is well known to primary care physicians, anesthesiologists, cardiologists, and cardiac surgeons alike is that time is everything when these patients are planned for transportation and treatment outside and inside the hospital. In the setting of shock, the timing of any modality to stabilize hemodynamics, including IABP, is highly relevant. The whole issue of timing has, however, been somewhat overshadowed by the neutral results from the IABP-SHOCK II trial (2,6). In this trial, only a minority (13%) of the patients received IABP before revascularization, and IABP was in reality used as an additional therapy to catecholamines in the vast majority of patients (90%). Some would argue that this may preclude the potential benefits of mechanical circulatory support, such as reducing the need for inotropes/vasopressors and stabilizing the patient through complex revascularization procedures. Could the results from IABP-SHOCK II have been different if, per protocol, IABP was inserted before percutaneous coronary intervention (PCI)?

The aspect of timing has been observed in coronary surgery. In a meta-analysis, seven papers including approximately 300 patients were included (3). These small-number studies demonstrated a beneficial effect of preoperative IABP on mortality, low-output failure, and length of stay in ICU. Thus, in this setting of an alternate revascularization procedure, a preprocedural IABP seems to be of preference.

Interestingly, the randomized balloon pump-assisted coronary intervention study (BCIS-1) on prophylactic use of IABP during high-risk PCI showed improved outcomes in the long-term analysis with a 34% relative reduction in all-cause mortality (7). The authors discuss whether this effect could be due to less periprocedural ischemia and/or a more aggressive (and extensive) revascularization in patients on hemodynamic support, but the mechanisms behind the improved survival are as of yet unknown. However, this aspect of mechanical circulatory support may also be relevant in patients with cardiogenic shock. The benefit of primary complete early revascularization versus treatment of the culprit lesion only is still undetermined in cardiogenic shock. Guidelines are vague regarding this issue, but treatment of all critically narrowed vessels is partly recommended in the recently published 2014 European guidelines on myocardial revascularization (1). A scientific basis for these recommendations can be found in the recently published randomized preventive angioplasty in myocardial infarction trial which demonstrated that complete primary revascularization in ST elevation myocardial infarction reduced the risk of adverse cardiovascular events, including mortality, at nearly 2 years (8). These findings should warrant renewed interest in the concept of mechanical circulatory support to facilitate complicated revascularization procedures in cardiogenic shock. Of note, in IABP-SHOCK II, more than 70% of the patients had multi-vessel disease.

The IABP-SHOCK II trial leaves, in addition to the issue of timing, a number of questions unanswered. First, the study population in IABP-SHOCK-II with a median systolic blood pressure of 89–90 mmHg (only 1/3 oliguric) most likely in part represents patients with a mild or moderate degree of shock. The majority of patients seemed to be hemodynamically stabilized on medical treatment. Cardiogenic shock is a highly heterogenic condition, and the patients included in most studies range from those with profound shock unresponsive to medical treatment having a detrimental prognosis to patients with systolic blood pressure of > 100 mmHg on inotropic/vasopressor support with cold and clammy extremities, but no oliguria or elevated lactate levels. For the clinician this is a large gap. Could we expect that one size fits all of these patients? The observed lack of effect could theoretically reflect that IABP treatment was futile in patients with mild shock, while patients with the most profound shock needed a more powerful assist (i.e., ECMO). However, this is presently unknown. These uncertainties do point to the need for accurate patient assessment and prognostication. Strict inclusion of patients with defined hemodynamic data in such a trial is not an easy setup, and a simple randomization is particularly difficult in this patient group. Second, there was a trend toward benefit in the youngest cohort of patients, and these patients are the best candidates for an aggressive approach including advanced assist systems and transplantations. The study also had a 17% crossover to IABP, an increased use of left ventricular assist devices in the control group, and a 4.3% mortality before insertion of IABP. These question marks should be addressed before IABP is dismissed completely in the setting of severe coronary pathology and hemodynamic compromise.

By now, we know that the broad and routinely use of IABP is not of benefit in all present-day revascularized patients. A profound shock unresponsive to medical treatment has 100% mortality, and these patients constitute the important subset of patients that needs salvage in order to reduce the overall cardiogenic shock mortality. The treatment of these patients will always rely on early identification, timely upscaling of medical and mechanical treatment, and continuous assessment of the hemodynamic effects by each intervention.

Should the observations in IABP-SHOCK-II affect the use of IABP in surgical patients? IABP is recommended by guideline for use in patients with mechanical complications following myocardial infarction such as ventricular septal rupture or severe mitral regurgitation (papillary muscle rupture), in order to stabilize the circulation before surgery (1). In the first SHOCK trial registry, such mechanical complications accounted for 12% of the patients with cardiogenic shock (9). In addition, postcardiotomy cardiogenic shock is another entity where IABP has been in frequent use. These patient groups are not likely to be included in any future large-scale IABP trials, but the recent evidence does warrant a new and critical focus on the use of IABP in cardiac surgery. On the bright side, every vascular surgeon should be relieved by the fact that IABP-treated patients did not have more bleeding or vascular complications than controls (2).

Has the downgrading of guideline recommendations altered the use of IABP over the last year? There has never been a widespread and routine use of this treatment. Prior registry reports have always indicated an underuse with respect to the prior guideline recommendations (10,11). There are, to our knowledge, no published registry data that cover the recent development. Interestingly, in Norway, the monthly sale of intra-aortic balloon pump catheters from the market-dominating company has not declined in 2013 or 2014 (∼40/month) and has been quite stable since 2007 (personal communication). Whether this mirrors skepticism to the prior recommendations or to the recently published studies is of course unknown.

The evidence-based therapeutic armamentarium for clinicians to treat cardiogenic shock irrespective of cause is minimal. In fact, besides treatment of arrhythmias, respiratory support, and early revascularization, all recommendations are Class II with low levels of evidence. In 1954, Griffith et al. published their results on different treatments for cardiogenic shock following myocardial infarction at the Los Angeles County Hospital (12). Sixty years of medical research and development has passed since then, and it is noteworthy that their conclusion still seems valid.

“The promptness with which measures for combating shock are instituted is a key factor in recovery, overshadowing in importance the particular method or combination of methods used in bringing shock under control.”

Declaration of interest: The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.