Impact of timing of intraaortic balloon counterpulsation on mortality in cardiogenic shock – a subanalysis of the IABP-SHOCK II trial

Abstract

Background

Conflicting results exist on whether initiation of intraaortic balloon pumping (IABP) before percutaneous coronary intervention (PCI) has an impact on outcome in this setting. Our aim was to assess the outcome of patients undergoing IABP insertion before versus after primary PCI in acute myocardial infarction complicated by cardiogenic shock.

Methods

The IABP-SHOCK II-trial randomized 600 patients with acute myocardial infarction and cardiogenic shock to IABP-support versus control. We analysed the outcome of patients randomized to the intervention group regarding timing of IABP implantation before or after PCI.

Results

Of 600 patients included in the IABP-SHOCK II trial, 301 were randomized to IABP-support. We analysed the 275 (91%) patients of this group undergoing primary PCI as revascularization strategy surviving the initial procedure. IABP insertion was performed before PCI in 33 (12%) and after PCI in 242 (88%) patients. There were no differences in baseline arterial lactate (p = 0.70), Simplified Acute Physiology Score-II-score (p = 0.60) and other relevant baseline characteristics. No differences were observed for short- and long-term mortality (pre vs. post 30-day mortality: 36% vs. 37%, odds ratio 0.99, 95% confidence interval (CI) 0.47–2.12, p = 0.99; one-year mortality: 56% vs. 48%, hazard ratio 1.08, 95% CI 0.65–1.80, p = 0.76; six-year-mortality: 64% vs. 65%, hazard ratio 1.00, 95% CI 0.63–1.60, p = 0.99). In multivariable Cox regression analysis timing of IABP-implantation was no predictor for long-term outcome (hazard ratio 1.08, 95% CI 0.66–1.78, p = 0.75).

Conclusions

Timing of IABP-implantation pre or post primary PCI had no impact on outcome in patients with acute myocardial infarction complicated by cardiogenic shock.

Introduction

For many years there was uncertainty regarding the efficacy of intraaortic balloon pump (IABP) support in patients with acute myocardial infarction (AMI) complicated by cardiogenic shock. The randomized Intra-aortic Balloon Pump in Cardiogenic Shock II (IABP-SHOCK II) trial revealed no benefit from IABP with respect to the primary and all secondary endpoints in the setting of AMI complicated by cardiogenic shock.1–4 However, advocates of IABP still argue a possible improved outcome when IABP is implanted prior to percutaneous coronary intervention (PCI).⁵ It is speculated that IABP may improve left ventricular unloading as well as coronary artery perfusion, leading to reduction in infarct size and, therefore, stabilize haemodynamic parameters during PCI and reduce shock severity. However, data supporting this hypothesis are very limited and conflicting.^6,7 The aim of the present analysis was to assess the impact of IABP timing on outcome in patients with AMI complicated by cardiogenic shock.

Methods

Study design and population

We retrospectively analysed data from the open-label multicentre randomized IABP-SHOCK II trial (Clinicaltrials.gov Identifier: NCT00491036) trial, which randomly assigned 600 patients with cardiogenic shock complicating AMI undergoing early revascularization in 1:1 ratio to IABP or no IABP from 2009 until 2012. The exact design and results of the original trial have been published previously.1–4 Cardiogenic shock was defined by a systolic blood pressure <90 mmHg for >30 min or need for inotropes or vasopressors to maintain a systolic blood pressure >90 mmHg in the absence of hypovolaemia, signs of pulmonary congestion and signs of impaired organ perfusion with at least one of the following: (a) altered mental status, (b) cold, clammy skin, (c) urine output <30 ml/h or (d) serum lactate >2 mmol/l. Early revascularization with either PCI or coronary artery bypass graft (CABG) together with optimal medical treatment according to current guidelines was the intended therapeutic strategy in all patients.

For the present analysis, all patients who were randomized to the IABP group in the main trial and were treated with primary PCI and IABP were included. Patients crossing over from one to the other treatment group were excluded. Other exclusion criteria were primary conservative treatment, early death before end of primary PCI and CABG as primary treatment strategy. Clinical follow-up was performed at 30 days, one year and six years following the procedure. Patients or their legally authorized representatives gave written informed consent. The trial was approved by the local ethics committees and it was conducted according to the Declaration of Helsinki.

Endpoints

Primary endpoint for the present analysis was 30-day mortality. Secondary endpoints were one-year and six-year mortality, arterial lactate change within the first 48 h after randomization and in-hospital adverse events including stroke, recurrent AMI, implantation of active mechanical circulatory support, need for renal replacement therapy, vascular complications and bleeding complications according to the Global Use of Strategies to Open Occluded Arteries definition.⁸

Statistical analysis

Categorical variables were expressed as numbers and percentages. Continuous variables were expressed as median with interquartile range (IQR). For comparison of categorical variables, the Chi², and for continuous variables, the Mann–Whitney U tests were used, respectively. For assessment of the primary endpoint 30-day mortality Chi²-testing and for mid- and long-term follow-ups after one and six years Kaplan–Meier analysis with log-rank testing was applied. Furthermore, a stepwise Cox-regression model was built including the following variables: gender, age, prior stroke, peripheral artery disease, arterial lactate at admission, diabetes, hypertension, premedication, baseline serum creatinine, baseline haemoglobin, presence of sinus rhythm at baseline, ST-segment elevation, left bundle branch block, extent of coronary artery disease, mechanical ventilation prior to admission, Thrombolysis In Myocardial Infarction (TIMI) flow prior to and after PCI. Additionally, logistic regression analysis (with and without centre effects) was done to identify independent predictors for IABP insertion before PCI. All p-values were calculated by two-tailed tests. Statistical significance was defined at p < 0.05. Statistical analysis was performed by an independent statistician at the Institut für Herzinfarktforschung, Ludwigshafen, Germany using SAS version 9.3 (Cary, North Carolina, USA).

Results

Baseline characteristics

A total of 275 out of 600 patients from the IABP-SHOCK II trial were included for the present analysis. IABP was implanted prior to PCI in 33 (12%) patients and after PCI in the remaining 242 (88%) patients. The study flow is illustrated in Figure 1. Length of IABP therapy was similar in both groups (median 3 (IQR 2–3) days vs. 2 (IQR 2–4) days; p = 0.50). The baseline characteristics were well balanced with no differences between the two groups (Table 1).

Procedural details

Coronary angiography showed no differences regarding the number of diseased vessels and culprit lesion related arteries (Table 2). In patients with IABP implantation prior to PCI, drug-eluting stents were significantly more often chosen as revascularization technique compared with patients with IABP implantation after PCI. Furthermore, thrombus aspiration was more frequently performed in patients with IABP initiation before PCI. TIMI flow before as well as after PCI was not significantly different in both groups.

Impact of IABP timing on outcome

No difference in the primary endpoint 30-day mortality was observed between the two groups (36% vs. 37%, odds ratio (OR) 0.99, 95% confidence interval (CI) 0.47–2.12, p = 0.99). Timing of IABP support had also no influence on the time to one-year (hazard ratio 1.08, 95% CI 0.65–1.80; p = 0.76) or six-year mortality (hazard ratio 1.00, 95% CI 0.63–1.60; p = 0.99; Figure 2).

After adjustment for baseline and procedural risk factors no association of IABP implantation timing with 30-day (OR 0.98, 95% CI 0.39–2.48, p = 0.97) and six-year mortality was found (hazard ratio 1.08, 95% CI 0.66–1.78, p = 0.75). Arterial lactate levels during the first 48 h following hospital admission were similar among patients with IABP before compared with IABP after PCI (Figure 3). With respect to hospital complications no differences were identified among the two groups (Figure 4).

Predictors for timing of IABP insertion

A logistic regression model was built in order to identify possible predictors for IABP insertion timing. ST-elevation myocardial infarction (STEMI) or new left bundle branch block seemed to serve as trigger for IABP initiation prior to PCI (Figure 5(a)). Since timing of IABP insertion is also influenced by the specific strategy predefined by the respective centres we included a centre effect adjustment to the model. After adjustment no factors were found to be significantly associated with the timing of IABP support (Figure 5(b)).

Discussion

The present study assessed the timing of IABP support and its impact on outcome of patients with cardiogenic shock complicating AMI. The main finding is that timing of IABP did not significantly influence 30-day mortality. Furthermore, IABP insertion timing has no effect on other secondary parameters such as arterial lactate or mid-term to long-term follow-up all-cause mortality. In addition, no specific predictors are present for the timing of IABP insertion.

There are no prospective data on the scenario of IABP timing, yet a few retrospective studies have been performed. A retrospective single centre analysis found that IABP insertion prior to PCI significantly reduced 30-day mortality. However, the number of patients was limited (<30 patients per group) and subjects assigned to IABP after PCI had significantly larger AMI, probably influencing the results.⁶ Another analysis showed increased adjusted one-year mortality with IABP insertion after PCI. It must be noted that unadjusted one-year mortality by Kaplan–Meier was similar between the two groups and no information regarding the variables included into the multivariable model were given. Furthermore, the patients included seem to differ from a typical cardiogenic shock population since 30-day mortality in this analysis was <10%.⁷ In line with our results, Cheng et al. and Sjauw et al. did not find any difference between patients treated with IABP before versus after primary PCI with respect to hospital and long-term mortality.^9,10

Those who continue to adhere to IABP argue for improved haemodynamic condition, reduced myocardial oxygen demand and augmentation in coronary flow during PCI when IABP support is initiated before revascularization. However, clinical outcome data do not support any benefit with IABP use prior to PCI. During high-risk PCI no reduction in the primary endpoint major adverse cardiac and cardiovascular events (death, acute myocardial infarction, cerebrovascular event, or further revascularization at hospital discharge) was achieved with IABP use prior to PCI in the Balloon Pump-Assisted Coronary Intervention Study.¹¹ In STEMI patients without cardiogenic shock analysed in the Counterpulsation to Reduce Infarct Size Pre-PCI Acute Myocardial Infarction trial IABP prior to PCI did not reduce infarct size compared with unprotected revascularization.¹² Of note, there was even a strong trend towards larger infarct size in the IABP group. This may be related to longer door-to-balloon times due to the time required for IABP implantation before revascularization. Delay in revascularization was already shown to independently predict mortality in AMI complicated by cardiogenic shock.⁷

Based on the secondary outcome analysis we were also not able to find any benefit with IABP use before start of revascularization beyond mortality. Arterial lactate as measure for cardiogenic shock severity and tissue hypoxia over the first 48 h was not significantly influenced by IABP timing. Furthermore, no differences between the groups were observed with regard to hospital complications. Particularly, the rate of active mechanical circulatory support device implantation was similar, indicating no beneficial effect on severe left ventricular failure of IABP prior to PCI. The increase in stroke volume provided by intra-aortic counterpulsation is limited¹³ and, hence, may be not sufficient to improve clinical outcome beyond experimental measures in patients with AMI and poor haemodynamic condition. This is underlined by the results of a recently published large registry using propensity score matching in comparison of IABP with a microaxial active mechanical circulatory support device, showing no difference in outcome regarding device implantation before or after PCI.¹⁴ However, other observational reports suggest a benefit if such devices are inserted before revascularization.^15,16 Such an approach has been recently tested in the Door-To-Unload trial in STEMI patients without cardiogenic shock, which showed no difference in infarct size if reperfusion was delayed by roughly 30 min and if unloading was applied.¹⁷ However, this trial is hampered by the lack of a standard of care control group. A larger trial with clinical endpoints is currently planned.

We also performed a multivariable analysis in order to search for factors that potentially triggered IABP use prior to revascularization. After adjustment for a centre effect no parameters were found to be independently associated with timing of IABP initiation. The fact that no differences in baseline characteristics were observed between the groups contradicts the speculation that sicker patients with poorer haemodynamic condition received IABP prior to PCI as bailout contributing to a misleading outcome. Based on these data, the results of the main IABP-SHOCK II trial are most probably not hampered by the different time points of IABP initiation used.

Several limitations to this work need to be reported. The number of patients, particularly in the group with IABP prior to PCI, was limited. Therefore, the present trial may not provide enough power to adequately assess mortality. Since no differences were observed in secondary outcome measures, a beneficial effect of IABP insertion before revascularization seems to be unlikely despite the small population size. Due to the post hoc design of the study no randomization of IABP pre versus post PCI could be provided. As a result, selection bias regarding time of IABP initiation cannot be ruled out. However, baseline characteristics were similar between the groups and no factors were identified that had an impact on IABP timing. Other secondary analyses such as infarct size were not obtained.

Conclusion

Timing of IABP insertion was not found to have an impact on short- and long-term mortality in patients presenting with cardiogenic shock complicating AMI. Also, secondary measures such as arterial lactate and hospital complications did not differ between patients treated with IABP before versus after PCI.

Acknowledgements

GF and JL contributed equally to the manuscript. Clinical Trial Registration: www.clinicaltrials.gov Identifier NCT00491036.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the German Research Foundation (TH-1406-6-1), the German Heart Research Foundation, the German Cardiac Society, Arbeitsgemeinschaft Leitende Kardiologische Krankenhausärzte, the University of Leipzig Heart Centre, and by unrestricted grants from Maquet Cardiopulmonary as well as Teleflex Medical.

Conflict of interest: The authors have no conflicts of interest to declare.

References