Preoperative intra-aortic balloon pump improves the clinical outcomes of off-pump coronary artery bypass grafting in left ventricular dysfunction patients

Severe left ventricular (LV) dysfunction patients undergoing off-pump coronary artery bypass grafting (OPCAB) are often associated with a higher mortality. The efficacy and safety of the preoperative prophylactic intra-aortic balloon pump (IABP) insertion is not well established. 416 consecutive patients with severe LV dysfunction (ejection fraction ≤35%) undergoing isolated OPCAB were enrolled in a retrospective observational study. 191 patients was enrolled in the IABP group; the remaining 225 patients was in control group. A total of 129 pairs of patients were propensity-score matched. No significant differences in demographic and preoperative risk factors were found between the two groups. The postoperative 30-day mortality occurred more frequently in the control group compared with the IABP group (8.5% vs. 1.6%, p = 0.02). There was a significant reduction of low cardiac output syndrome in the IABP group compared with the control group (14% vs. 6.2%, p = 0.04). Prolonged mechanical ventilation (≥48 h) occurred more frequently in the control group (34.9% vs. 20.9%, p = 0.02). IABP also decreased the postoperative length of stay. Preoperative IABP was associated with a lower 30-day mortality, suggesting that it is effective in patients with severe LV dysfunction undergoing OPCAB.

Scientific RepoRts | 6:27645 | DOI: 10.1038/srep27645 dysfunction that underwent OPCAB and performed a propensity-matched analysis to determine whether these patients benefited from preoperative prophylactic IABP insertion.

Material and Methods
Patient population. The present study was a historical, single-center, observational cohort study conducted at Beijing Anzhen Hospital, Capital Medical University. A total of 18,719 consecutive adult patients underwent isolated OPCAB in this hospital from January 2009 to December 2014. The left ventricular ejection fraction (LVEF) of all patients was calculated from an echocardiography assessment performed before the OPCAB operation. A total of 446 (2.4%) patients were identified as having severe LV dysfunction (LVEF ≤ 35%). Among those patients, 30 cases that presented with hemodynamic instability or severe femoral artery stenosis were excluded (Fig. 1). Finally, 416 patients were categorized according to preoperative prophylactic IABP insertion into either the IABP group or the control group. Data were retrospectively extracted from an institutional registry of OPCAB patients and ICU clinical database. This study received approval from the Beijing Anzhen Hospital Research Ethics Committee and all operations were performed in accordance with relevant guidelines and regulations. Because this was a retrospective observational study, the institutional review board of the Beijing Anzhen Hospital Ethics Committee waived informed consent. Data collection was subjected to the supervision of the Beijing Municipal Administration of Hospitals.
Surgical technique. The detailed information of the OPCAB procedure has been previously described 7 .
Six experienced cardiac surgeons (≥ 200 cases/year) performed all of the procedures. A cardiopulmonary bypass (CPB) circuit was placed on stand-by during the procedures. Conversion to CPB was considered if there was any evidence of hemodynamic instability concerns, such as ventricular arrhythmia, hypotension (systolic pressure ≤ 80 mmHg), and cardiac arrest during OPCAB procedures. IABP support. Patients in the IABP group received IABP support prior to the induction of anesthesia, followed by continuous IABP during the entire procedure, and postoperatively if needed. Each patient provided written consent for the IABP insertion. The control patients did not receive IABP support preoperatively. IABP treatment was initiated in the control group when cardiac index (CI) could not be maintained at a level ≥ 2.0 L/ min/m 2 , regardless of a high inotrope status.
Two types of IABP system were used at our center, namely, the Arrow System and the Datascope CS300 System. IABP balloon was selected according to the height of the patients. The IABP balloon was connected to a Datascope or Arrow pump. IABP insertion through the best femoral artery was possible in all cases according to the results of an ultrasound examination of bilateral lower extremities, and the corrected placement was assessed by a chest X-ray. The IABP variables were settled at a 1:1 balloon inflation synchronized with the electrocardiogram or aortic blood flow. Heparin was systematically used for anticoagulation. IABP support was terminated with hemodynamic stability after the operation.
Primary and secondary outcomes and definitions. The primary endpoint was postoperative 30-day mortality (death occurring within 30 days after surgery). The secondary endpoints were major postoperative complications, such as low cardiac output syndrome (LCOS), myocardial infarction, bleeding requiring reoperation, tracheotomy, renal failure requiring dialysis, stroke, and ICU stay, and postoperative length of stay (LOS) 21 . The IABP-related complications were also recorded and analyzed for safety purposes.
Statistical analysis. Continuous variables are shown as the mean and standard deviations or the median and interquartile ranges, and categorical variables are shown as frequencies and percentages. Continuous variables were compared using Student's t-test or Mann-Whitney U-test. Categorical variables were compared using the Chi-square test or a Fisher's exact test. Limb ischemia requiring surgical intervention 1 (0.8%) 0 (0%) 1.00 Severe bleeding 0 (0%) 0 (0%) 1.00 Balloon failure or leak 0 (0%) 0 (0%) 1.00 Propensity score matching was undertaken by estimating the probability of receiving preoperative IABP support (that is, the propensity score) using a logistic regression model including the potentially confounding covariates shown in Table 1. Each IABP group patient was matched with a non-IABP-supported patient based on the propensity score by the method of nearest neighbor matching within a caliper (caliper = 0.25 × SD[logitP] 22 ) using the psmatch2 command in Stata 23 . After matching, continuous variables that followed a normal distribution were compared using the paired sample t-test; otherwise, the Wilcoxon's matched-pairs signed-rank test was used. The McNemar's test was used to evaluate the comparability of categorical factors after matching. The in-hospital survival rates after surgery of matched samples were shown as Kaplan-Meier survival cures and the effect of IABP vs. non-IABP was presented as a hazard ratio (HR) with associated 95% confidence intervals (CIs) from the Cox regression model.
Statistical analyses were conducted using Stata software version 11 (Stata Corp, College Station, Texas, USA). Two-sided testing was used with a p-value significance level of less than 0.05.

Patient baseline characteristics.
A total of 416 patients were included in this study (Fig. 1), with 191 patients in the IABP group and 225 patients in the control group. There were significant differences documented in the demographics and comorbidities of the patients (Table 1). After the propensity-score matching, 129 matched pairs were obtained (Fig. 1). No differences in demographics or preoperative risk factors were found between the two groups (Table 1). Figure 2 shows a Love plot for the absolute differences in the baseline covariates before and after matching; a jitter plot for propensity-score distribution is also presented.
OPCAB procedural characteristics are detailed in Table 2. Importantly, the mean number of anastomoses were comparable between the two groups (p = 0.22). However, there were few patients converted to on-pump CABG in the IABP group (0 vs. 6 [4.6%], p = 0.04).
Postoperative 30-day mortality. Clinical outcomes are presented in Table 2. The 30-day mortality was 1.6% in IABP group compared to 8.5% in the control group (p = 0.02). The Kaplan-Meier survival curves of the two groups before and after matching are shown in Fig. 3. The preoperative prophylactic IABP insertion was an independent predictor of survival after adjusting for the propensity score using the Cox regression model (HR 0.17, CI 0.04-0.79, p = 0.02).
Postoperative complications. The postoperative complications are summarized in Table 2. There was a significant reduction in postoperative LCOS in the IABP group (14 vs. 6.2%, p = 0.039). There were no significant differences in the required transfusion of red blood cells (2 vs. 2 units, p = 0.07) and fresh frozen plasma (600 vs. 600 ml, p = 0.37). The postoperative myocardial infarction, reoperation for bleeding, tracheotomy, hemodialysis, and neurologic events were comparable between the two groups ( Table 2).  There were no statistically significant differences in the duration of mechanical ventilation (47.1 ± 49.1 vs. 37.9 ± 40.9 h, p = 0.16). However, the required prolonged mechanical ventilation occurred more frequently in the control group (34.9 vs. 20.9%, p = 0.02). ICU stay was comparable between the two groups. The mean postoperative LOS was shorter in the IABP group (interquartile range) of 9 (7-12) vs. 10 (7-16) days (p = 0.03) ( Table 2).
No significant interactions were observed between preoperative IABP insertion and any of the 12 sub-groups with respect to the in-hospital mortality, as shown in the hazard-ratio plots in Fig. 4.

IABP-related complications.
Post-cardiotomy IABP was used in 21 patients in the control group, including 3 who required weaning from CPB, and 18 who occurred LCOS. The mean duration of IABP support was shorter in the IABP group (135.9 ± 80.9 vs. 76.4 ± 33.4 h, p < 0.001). There were no cases of IABP-related mortality. No severe bleeding at the IABP insertion site, or balloon failure in any of the patients. Lower limb ischemia requiring surgical intervention was observed in 1 patient (0.8%) in the IABP group (Table 2).

Discussion
This large regional study of propensity-matched patients indicated that preoperative prophylactic IABP insertion was associated with reduced postoperative 30-day mortality in severe LV dysfunction patients who underwent OPCAB. Furthermore, prophylactic IABP insertion resulted in significant reduction of postoperative LCOS incidence and shorter postoperative hospital stay.
From the perspective of pathophysiology, the positive effect of IABP insertion is believed to increase coronary blood flow while simultaneously decreasing myocardial oxygen demand. Consequently, preoperative prophylactic IABP assistance provides better hemodynamic stability in crucial times of higher oxygen demand when the heart is displaced in OPCAB procedures 24,25 .
Although there are certain advantages in theory, the results have been controversially debated in clinical practice. Some studies have shown a positive effect of preoperative prophylactic IABP insertion in improving the outcomes of high-risk patients 13,14 . The strongest evidence supporting preoperative IABP insertion for high-risk patients undergoing CABG comes from published meta-analysis studies 12,14,17 . However, many contemporary studies have challenged the effectiveness of preoperative IABP in high-risk patients undergoing CABG 15,16 . Worse outcomes were shown in a recent propensity-score matching study 16 , in which the preoperative IABP insertion in patients undergoing CABG after acute myocardial infarction was associated with increased in-hospital morbidity, greater transfusion requirements, and longer postoperative ICU stay.
The results of previous studies have been controversial for several possible reasons. First, there is no standard definition of a high-risk patient. Various conditions, including severe LV dysfunction, left main disease, diffuse coronary disease, and reoperation, have been suggested for preoperative prophylactic IABP insertion 14 . Second, the criteria for prophylactic IABP insertion have not been well defined 26 . A distinction is lacking between therapeutic use for patients with preoperative cardiogenic shock or hemodynamic instability and prophylactic use for patients with preoperative hemodynamic stability. In many previous studies including patients with hemodynamic instability, the indication has been more likely for therapeutic, rather than prophylactic insertion. Third, the results were also possibly affected by the severity of the patients who were selected to receive preoperative IABP support. Finally, most of the procedures were conducted in CABG patients. IABP insertion before surgery was suspended during CPB. The benefit from IABP support was relatively low.
Our study investigated for the clinical effects of preoperative prophylactic IABP insertion in patients with severe LV dysfunction that underwent selective OPCAB. The IABP group of patients received preoperative prophylactic IABP support to increase the safety of OPCAB procedures. The patients who received preoperative IABP support for hemodynamic instability, cardiogenic shock, and emergency operations were excluded. Comparatively, IABP was still working during OPCAB procedures in our study. These patients were more likely to benefit from preoperative IABP insertion. Therefore, preoperative prophylactic IABP insertion was associated with a lower rate of conversion to on-pump CABG, which has been associated with increased in-hospital mortality 9 .
The preoperative prophylactic IABP insertion in our study was not associated with an increased rate of IABP-relationship complications (that is, limb ischemia requiring surgical intervention, severe bleeding at the IABP insertion site, and embolism). The incidence rate of IABP-related complications was low, similar with a previous study 27 . Therefore, IABP insertion is safe in those high-risk patients.

Limitations.
Our study had several limitations. First, this study was subject to the limitations inherent in any retrospective, observational study from a single center. The nonrandomized design might have affected our results, owing to unmeasured confounds, procedural bias, or detection bias. Despite the benefits of propensity matching, it is possible that there are additional confounds that were not accounted for in our adjustment algorithm. The whole study depends on the accuracy of propensity score matching and many pitfalls may be hidden. Second, it is generally believed that the experience of the surgeon can influence the results of OPCAB. Six experienced cardiac surgeons performed the OPCAB procedures in this study. However, our hospital is an international center for cardiovascular clinical and research. All of the surgeons followed the same standard OPCAB procedure of our hospital. Furthermore, this study was also limited to patients undergoing isolated OPCAB. Patients requiring concomitant cardiac surgical procedures, and/or those with mechanical complications of acute myocardial infarction, such as acute mitral regurgitation or myocardial rupture, were excluded from this study. Therefore, the results of this study cannot be extended to these extreme high-risk patient populations. Finally, the present study was conducted in the setting of a high-volume tertiary cardiovascular center in a developing country; therefore, the results might not be generalizable to other centers in different situations.