Galectin-3 as a Predictor of Long-term Survival after Isolated Coronary Artery Bypass Grafting Surgery

Abstract

Background:

Galectin-3 is a well-established biomarker of adverse clinical outcomes but its prognostic value for long-term survival after cardiac surgery is not well understood. Elevated levels of Galectin-3 have been found to be significantly associated with higher risk of death in both acute decompensated and chronic heart failure populations. Its prognostic value for long-term survival after cardiac surgery is not known.

Methods:

A sample of patients contributing to the Northern New England Cardiovascular Disease Study Group Cardiac Surgery Registry from 2004–2007 were enrolled in a prospective biomarker cohort (N = 1,690). Preoperative Galectin 3 levels were measured and categorized by quartile. We used Kaplan-Meier survival analysis and Cox regression models adjusting for variables in the STS ASCERT probability calculator to evaluate the association between elevated Galectin-3 levels and survival to 6 years.

Results:

Preoperative Galectin-3 levels ranged from 1.72 to 28.89 ng/mL (mean=8.96; median=8.06; interquartile range=5.42, 11.08). Crude survival decreased by increasing quartile. After adjustment, serum levels of Galectin-3 in the highest quartile of the cohort were associated with significantly decreased survival compared to the lowest quartile (HR: 2.22, 95% CI: 1.40 – 3.54, p = 0.001). There was no decrease in survival for the middle quartiles (HR: 1.36; 95%CI: 0.87 – 2.12; p = 0.177).

Conclusions:

There is a significant association between elevated preoperative Gal-3 levels and risk of mortality after isolated coronary artery bypass grafting surgery. An assessment of the relationship between preoperative serum biomarkers and long-term survival can be used for risk stratification or estimating postsurgical prognosis.

Heart disease is the leading cause of death in the United States^{1, 2}. Over 253,000 people undergo cardiac surgery to treat coronary artery and valve disease costing over $8 billion annually^{1, 3}. Patients undergoing cardiac surgery suffer from a particularly high-risk for mortality. Most prediction models for coronary artery bypass grafting (CABG) surgery are limited to in-hospital or 30-day endpoints. We applied a well-established long-term survival model using the Society of Thoracic Surgeons (STS) Collaboration on the Comparative Effectiveness of Revascularization Strategy (ASCERT) Long-Term Survival Probability risk calculator⁴.

Galectin-3 is a member of the lectin family involved in numerous physiological and pathological processes, including inflammation and fibrosis, which are pivotal pathophysiological mechanisms in the development and progression of HF.^5–8 An increase of Gal-3 stimulates the release of various mediators and promotes cardiac fibroblast proliferation, collagen deposition and ventricular dysfunction.^{9, 10} Furthermore, the involvement of Gal-3 in the development of fibrosis has also been demonstrated in the heart, liver and kidney.^{11, 12} Gal-3 was also found to be significantly up-regulated in hypertrophied hearts of patients with aortic stenosis and in the plasma of patients with acute and chronic HF.^13–16

These observations suggest that circulating Gal-3 is useful in identifying patients at risk for developing a poor postoperative prognosis, however little is known about the clinical utility of Gal-3 as it relates to CABG and long-term survival. Identifying CABG patients who may benefit from additional intervention or altered postoperative management is of interest to both cardiothoracic surgeons and hospitals. The use of novel biomarkers could aid in patient risk stratification and estimating postoperative prognosis. We hypothesize that elevated levels of Gal-3 are significantly associated with worse long-term outcomes and short long-term survival. Therefore, the aim of this study was to evaluate the relationship between pre-and postoperative Gal3 levels and long-term survival among patients undergoing isolated CABG.

Patient and Methods

NNE Biomarker Study

The Northern New England Cardiovascular Disease Study Group (NNECDSG)¹⁷, is a longitudinal cohort initiated in 1987 to prospectively study the management of patients with cardiovascular disease. The NNECDSG is a voluntary, regional consortium of clinicians, research scientists, and hospital administrators from eight medical centers in New Hampshire, Maine and Vermont. All hospitals in this consortium submit data on cases with validation of procedure numbers and mortality conducted every two years. Registry data includes patient characteristics, procedural indication, clinical variables and in-hospital outcomes. All institutional review boards for each contributing hospital reviewed and approved the data collection for the NNECDSG as well as the conduct of this study.

We prospectively enrolled patients undergoing isolated CABG into the NNE Biomarker Study from 2004 to 2007 (N = 1,690). Patients were excluded from the analysis if they died during the index admission (n=40) or if their recorded date of discharge occurred after their recorded date last known alive (n=2). We removed patients with missing biomarker data (n=130), leaving 1,560 in our final analysis. The Committee for the Protection of Human Subjects at Dartmouth College (IRB) approved this study for both the prospective cohort with patient consent and the linkage of readmission and mortality events.

Primary Outcomes: Long-term Survival

The cohort was linked to state all-payer inpatient claims using name, gender, social security number, date of birth and residence at the time of surgery. Maine and Vermont completed links internally. We used probabilistic linking for New Hampshire all-payer inpatient claims. We achieved complete ascertainment for Vermont and Maine; five percent of New Hampshire patients were not matched in the inpatient claims. The primary outcome of this study was long-term survival after cardiac surgery from the index admission. All patients enrolled in the biomarker study were linked to the National Death Index to determine all-cause mortality.

Biomarker collection and measurement

Blood samples were preoperatively collected prior to incision at each participating site in a 10-mL serum tube from 2003–2007. Blood was allowed to clot at room temperature for 20 minutes to separate out the red blood cells, the tubes were centrifuged at 3500 rpm for 20 minutes, and the sera stored at the respective medical centers below −80 degree Celsius until transportation on dry ice to the Laboratory for Clinical and Biomedical Research in Colchester, Vermont where they were stored at −80 degree Celsius until measurement. Frozen serum was analyzed at a central laboratory, at the same time for biomarker measurement. Markers were measured by Roche Diagnostics Elecsys 2010. Adjusted Galectin-3 values were calculated to correct for a reagent shift in the assays.

Statistical Analysis

Patient, clinical and procedural characteristics were compared using chi-square tests and continuous data were compared between groups using the student’s t-test or Wilcoxon Rank-Sum tests, were appropriate. Kaplan-Meier and log-rank techniques were used to conduct a time-to-event analysis. We right-censored our analyses at 6-years due to the lack of events after 6 years. Patients were stratified by pre-and postoperative biomarker levels by quartile. We elected to treat the biomarker levels as categorical rather than continuous variables due to the graded relationship of quartile categories and mortality incidence. Following Kaplan-Meier analysis, quartiles 2 and 3 were found to have similar survival and were therefore combined into one category.

Multivariable Cox proportional hazard models were constructed to assess the relationship of pre-and-postoperative biomarker levels and long-term survival, using biomarker quartile cut points. Adjustment was carried out using the American College of Cardiology Foundation STS ASCERT Long-Term Survival Probability Calculator.⁴ This model corrects for age, sex, body surface area, preoperative intra-aortic balloon pump, preoperative myocardial infarction, prior CABG, prior percutaneous intervention, smoking status, atrial fibrillation, unstable angina, history of congestive heart failure, ejection fraction, left main stenosis > 50%, hypertension, diabetes, vascular disease, COPD, preoperative serum creatinine, and emergent and urgent priority status. All analyses were conducted using Stata 13.1 (College Station, TX).

Results

Patient characteristics

Overall, 241 (15.5%) patients died within 6-years of cardiac surgery. Patient preoperative and operative characteristics and comorbidities are described in Table 1. Mean age was 65 ± 10 years, 79% were male. Preoperative mean Gal-3 plasma levels were 8.9 ± 5.4 ng/mL (IQR: 5.42 – 11.1). Postoperative mean Gal-3 levels were 10.2 ±13.4 ng/mL (IQR: 5.7 – 12.0).

Table 1.

Characteristics of patients by survival status after CABG or valve replacement in NNE.

Characteristics	Died (n=241)	Survived (n=1,319)	Std. Diff. (d)	p-value‡
Age, mean years (SD)	69.9 (10.2)	64.5 (9.9)	0.54	<0.001
Women	30%	22%	0.20	0.002
Body Mass Index				<0.001
< 18.5 kg/m2	2.2%	0.4%	0.16
18.5–24.9 kg/m2	18%	17%	0.04
25.0–29.9 kg/m2	41%	40%	0.02
30.0–34.9 kg/m2	24%	28%	0.09
≥ 35.0 kg/m2	15%	15%	0.01
Preoperative IABP	6.7%	4.2%	0.11	0.071
Preoperative MI				<0.001
No	48%	57%	0.20
< 24 hours	1.1%	1.9%	0.06
> 24 hours, < 7 days	22%	18%	0.09
> 7 days, < 365 days	13%	9.1%	0.14
> 365 days	16%	14%	0.07
Prior CABG	4.2%	2.1%	0.12	0.037
Prior PCI	3.4%	3.7%	0.02	0.820
Smoker	25%	23%	0.07	0.318
Atrial Fibrillation	16%	5.6%	0.34	<0.001
Unstable angina	57%	54%	0.06	0.335
Congestive heart failure	29%	7.8%	0.57	<0.001
Ejection Fraction (SD)	49.7 (13.3)	54.6 (11.5)	0.39	<0.001
Left main, ≥ 50% stenosis	40%	33%	0.15	0.020
Hypertension	84%	80%	0.09	0.184
Diabetes	50%	36%	0.29	<0.001
Vascular Disease	42%	25%	0.39	<0.001
COPD	22%	12%	0.28	<0.001
Preoperative serum creatinine, mg/dL (SD)	1.32 (0.82)	1.13 (1.05)	0.18	0.007
Priority				<0.001
Non-urgent	24%	32 %	0.17
Urgent	75%	66%	0.18
Emergent	1.5%	2.0%	0.04
Received PRBC	65%	34%	0.65	<0.001
PRBC, units (SD)	3.12 (4.49)	0.95 (1.73)	0.64	<0.001
AKIN Network Stage				0.021
No AKI	43%	68%	0.52
AKIN 1	39%	29%	0.22
AKIN 2	7.6%	2.0%	0.27
AKIN 3	10%	9.4%	0.42
Low cardiac output	15.8%	4.6%	0.24	<0.001

AKI=acute kidney injury, AKIN=Acute Kidney Injury Network, BMI=body mass index (kg/m²), BSA=body surface area (m²), CABG=coronary artery bypass graft, CHF=congestive heart failure, COPD=chronic obstructive pulmonary disease, IABP=intra-operative balloon pump, MI=myocardial infarction, PCI=percutaneous coronary intervention, pRBC=packed red blood count.

Within our cohort, we observed a significant association with patients who displayed classic cardiovascular risk factors and risk of long-term mortality. These patients were older on average, had prior CABG, had atrial fibrillation, congestive heart failure (CHF), decreased ejection fraction, more likely to be diabetic, have vascular disease, COPD, and received more PRBC units.

Univariate Analyses

In the univariate analyses, pre-and postoperative Gal-3 were significantly associated with long-term mortality after CABG surgery for those in the highest quartile (HR: 3.73, 95% CI: 2.43 – 5.73, p=<0.001; HR: 2.70, 95% CI: 1.95 – 3.72, p=<0.001, respectively) (Table 2). The postoperative lower quartile did not demonstrate a significant relationship with elevated Gal-3 and increased hazard of long-term mortality.

Table 2.

Simple and adjusted effects of preoperative and postoperative serum biomarker levels on long-term survival.

Biomarker	% Dead	Unadjusted				Adjusted‡
		HR	95% CI	P-value		HR	95% CI	p-value
Preoperative
Quartile 1	8.5%	1.00	-	-		1.00 (ref)	-	-
Quartile 2–3	12%	(ref)	1.14 – 2.70	0.010		1.36	0.87 – 2.12	0.177
Quartile 4	27%	1.76	2.43 – 5.73	<0.001		2.22	1.40 – 3.54	0.001
		3.73
Postoperative
Quartile 1	7.8%	1.00	-	-		1.00 (ref)	-	-
Quartile 2–3	13%	(ref)	0.70 – 1.39	0.936		0.84	0.59 – 1.19	0.328
Quartile 4	29%	0.98	1.95 – 3.72	<0.001		1.66	1.16 – 2.37	0.005
		2.70

^‡Adjusted for STS ASCERT

^*HR= hazard ratio from Cox proportional hazards model; CI= confidence interval

We observed a significant relationship between elevated levels of pre-and postoperative Gal-3 and long-term mortality for patients with low cardiac output after surgery. For those with pre-and postoperative Gal-3 levels in the highest quartile, low cardiac output patients incurred a 2-fold increase of long-term mortality (OR: 1.93; 95% CI: 1.08 – 3.43; OR: 2.36; 95% Ci: 1.19 – 4.65, respectively). Patient characteristics by Gal-3 quartile are presented in Supplementary Table 1.

Adjusted analyses

Multivariable Cox proportional hazards regression analysis revealed serum levels of Galectin-3 in the highest quartile of the population measured preoperatively and postoperatively conferred increased risk of mortality (HR: 2.22, 95% CI: 1.40 – 3.54, p = 0.001; HR: 1.66, 95% CI: 1.16 – 2.37, p = 0.005, respectively) (Table 2). Pre- or postoperative serum levels of Galectin-3 in quartile 2/3 were not significantly associated with long-term survival. Full adjusted Cox proportional hazard model analyses by preoperative and postoperative Gal-3 tercile are presented in Supplementary Tables 2 and 3.

Survival analyses

The median follow-up time was 8 years (IQR: 7.5 – 9.7) and the 6-year survival rate was 86%. Adjusted Kaplan-Meier survival analysis demonstrated a significant (p<0.001) difference across quartiles of pre- and postoperative biomarker levels (Figures 1 and 2).

Figure 1. Adjusted Kaplan Meier curves for survival by quartile categories of preoperative Galectin-3 biomarker levels.

Kaplan-Meir survival analysis demonstrated a significant (p<0.001) difference across quartiles of preoperative biomarker levels (quartile 1 = blue line; quartiles 2/3 = red line; quartile 4 = green line). Adjustment was carried out using the STS ASCERT long-term mortality model.

Figure 2. Adjusted Kaplan-Meier curves for survival by quartile categories of postoperative Galectin-3 biomarker levels.

Kaplan-Meir survival analysis demonstrated a significant (p<0.001) difference across quartiles of preoperative biomarker levels (quartile 1 = blue line; quartiles 2/3 = red line; quartile 4 = green line). Adjustment was carried out using the STS ASCERT long-term mortality model.

The incidence of 6-year mortality was 1.7 per 100 person-years. The rate increased across elevated preoperative Gal-3 levels (Q1: 0.91; Q2/3: 1.32; Q4: 3.61) and postoperative Gal-3 levels (Q1: 0.86; Q2/3: 1.48; Q4: 3.24).

Subgroup analyses: Ejection fraction < 40

For those with low ejection fraction (N=163), pre and postoperative Gal-3 quartiles confirmed a trend of increased risk of long-term mortality following CABG surgery. Patients with elevated levels of preoperative Gal-3 conferred increased risk of long-term mortality (log rank p= 0.015), but after adjustment, the relationship was not significant (HR: 1.89; 95 % CI: 0.94 – 3.82). Similarly, patients with elevated postoperative Gal-3 had significantly increased risk of mortality (log rank p=0.001) but after adjustment the association was not statistically significant (HR: 2.10; 95 % CI: 0.89 – 4.89).

Comment

There are limited clinical data on the relationship between plasma Gal-3 and CABG outcomes. Our findings demonstrate that higher levels of Gal-3, a member of the lectin family of proteins, are associated with increased risk of long-term mortality for patients undergoing cardiac surgery. Previous studies have examined the prognostic value of Gal-3 in individuals with existing heart failure.¹⁸ Elevated levels of Gal-3 have also been shown to be strongly associated with kidney injury, a finding that has been corroborated by previous studies.^{2, 9, 19–21} To our knowledge, this is the first study to report the association of Gal-3 with long-term mortality for patients undergoing isolated CABG.

In this observational study of regional data, we used a prospective cohort to assess the relationship of Gal-3 and long-term mortality for patients undergoing isolated CABG. We found that patients in the highest quartile of preoperative Gal-3 had over three-times the long-term mortality risk after CABG than patients in lower quartiles of the serum biomarker. Similarly, we found that patients in the highest quartile of postoperative Gal-3 had 66% greater risk of long-term mortality than those patients with lower levels of Gal-3 after surgery.

Our findings indicate that pre-and-postoperative Gal-3 values are associated with long-term mortality in cardiac surgery patients and could potentially be used as a tool to identify patients at higher risk of death before surgery and at discharge. This study highlights the importance of assessing the true value of emerging cardiac fibrosis biomarkers beyond clinical risk factors and natriuretic peptides, particularly in light of the newly obtained American College of Cardiology/American Heart Association class II recommendation for determination of prognosis in chronic HF.²²

Gal-3 is a well-established biomarker for cardiac fibrosis, ventricular dysfunction and poor prognosis in heart failure.^{23, 24} Gal-3 is a soluble beta-galactoside-binding protein secreted by activated macrophages. Its main action is to bind to and activate the fibroblasts that form collagen and scar tissue, leading to progressive cardiac fibrosis.²⁵ Gal-3 represents a link between inflammation and fibrosis, as circulating levels have been found elevated in several human fibrotic conditions, including pancreatitis and renal failure. In the heart, Gal-3 is thought to augment fibrosis and modulate immune response, both pivotal processes in maladaptive cardiac remodeling.²⁶

Several studies have shown that elevated concentrations of Gal-3 provide important prognostic information. Increasing evidence has implicated Gal-3 in the pathogenesis and progression of heart failure.^{27, 28} Studies have shown that circulating Gal-3 levels predict future events in patients with acute and chronic HF. Van Kimmenade et al found that an elevated Gal-3 level was the best independent predictor of 60-day mortality in patients with acute decompensated HF.¹⁶ The study by de Boer et al confirmed the value of Gal-3 as an independent prognostic marker.¹¹ Patients with elevated levels of Gal-3 in the highest quartile had three times higher risk of HF hospitalization and/or death. In a pooled analysis, Meijers et al found a significant relationship between near term hospitalization and unplanned rehospitalization with elevated levels of Gal-3 in HF patients.²⁹

Fibrosis is a fundamental component of the adverse structural modeling of myocardium present in the failing heart. There is previously reported limited benefit of Gal-3 observed in previous studies where remodeling and fibrosis was observed at an advanced stage, it is conceivable that Gal-3 could have a more prominent role in earlier stages of fibrosis pathobiology and ventricular remodeling.^{30, 31} Given that progressive cardiac fibrosis is a central aspect in the progression of cardiac dysfunction, as well as the primary substrate for lethal arrhythmias and sudden death, it is intuitive that a blood marker of cardiac fibrosis would be independently associated with cardiovascular mortality. This study shows that increased serum levels of Gal-3 were predictive of cardiovascular mortality.

Limitations

The NNE Biomarker Study was a large, multicenter, regional investigation that prospectively enrolled patients across multiple hospitals, ensuring the generalizability of its subsequent findings. Our data are from 2004 – 2007, when the prevalence of heart disease related deaths was slightly higher compared to current data (221.6 vs 168.5 age-adjusted deaths per 100,000 people).³² While age-adjusted mortality rates have declined in recent years, heart disease remains the leading cause of death in the United States.³²

An important aspect of this study is that the NNE mortality model included only patient characteristics and history, comorbid conditions, and patient clinical status variables immediately available to clinicians before surgery. We have limited our predicting variables to those known prior to operation. Therefore the variables alone cannot provide a perfect prediction for all-cause mortality.

Conclusion

We report that unadjusted elevated pre-and postoperative Gal-3 values are significantly associated with increased risk of mortality after cardiac surgery. Our findings suggest that Gal-3 can be used as a biomarker in risk stratification and estimating postsurgical prognosis. Clinical application of pre- and postoperative Gal-3 is beneficial when evaluating operative care strategies, discharge protocols and follow-up care after cardiac surgery.