Coronary perfusion pressure is associated with adverse outcomes in advanced heart failure

Abstract

Background:

Myocardial perfusion is an important determinant of cardiac function. We hypothesized that low coronary perfusion pressure (CPP) would be associated with adverse outcomes in heart failure. Myocardial perfusion impacts the contractile efficiency thus a low CPP would signal low myocardial perfusion in the face of increased cardiac demand as a result of volume overload.

Methods:

We analyzed patients with complete hemodynamic data in the Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness trial using Cox Proportional Hazards regression for the primary outcome of the composite risk of death, heart transplantation, or left ventricular assist device [(LVAD). DT × LVAD] and the secondary outcome of the composite risk of DT × LVAD and heart failure hospitalization (DT × LVADHF). CPP was calculated as the difference between diastolic blood pressure and pulmonary artery wedge pressure. Heart failure categories (ischemic vs non-ischemic) were also stratified based on CPP strata.

Results:

The 158 patients (56.7 ± 13.6 years, 28.5% female) studied had a median CPP of 40 mmHg (IQR 35–52 mmHg). During 6 months of follow-up, 35 (22.2%) had the composite primary outcome and 109 (69.0%) had the composite secondary outcome. When these outcomes were then stratified based on the median, CPP was associated with these outcomes. Increasing CPP was associated with lower risk of both the primary outcome of DT × LVAD (HR 0.96, 95% CI 0.94–0.99 p = .002) and as well as the secondary outcome of DT × LVADHF (p = .0008) There was significant interaction between CPP and ischemic etiology (p = .04).

Conclusion:

A low coronary artery perfusion pressure below (median) 40mmHg in patients with advanced heart failure undergoing invasive hemodynamic monitoring with a pulmonary artery catheter was associated with adverse outcomes. CPP could useful in guiding risk stratification of advanced heart failure patients and timely evaluation of advanced heart failure therapies.

Introduction

Heart failure (HF) is a widely prevalent medical condition with a high burden of morbidity and mortality.^1,2 Although there have been recent advances in the diagnosis and treatment of HF,^3,4 outcomes among symptomatic patients remain poor.⁵ In the US, HF accounts for over one million hospital admissions annually. Approximately 50% of US adults die within 5 years of initial HF diagnosis.^6,7 The mechanisms underlying HF progression are heterogeneous, complex, and poorly understood.⁸ However, the final common pathway in HF progression is heavily influenced by a contractile mechanical inefficiency state⁹ and the decoupling of myocardial oxygen consumption with the main determinants of systolic function (ventricular-wall tension, heart rate, and contractile state).^10,11 The consequence of these maladaptive changes in a failing heart is increased myocardial oxygen demand in the face of reduced oxygen tissue delivery.¹²

Coronary perfusion pressure (CPP) is a major determinant of myocardial tissue perfusion which has a direct impact on myocardial performance.¹³ CPP is conveniently calculated as the difference between systemic diastolic blood pressure (BP) and left ventricular end diastolic pressure (LVEDP).¹⁴ Pulmonary artery wedge pressure, PAWP, a surrogate marker for LVEDP (in the absence of mitral or pulmonary vein stenosis),¹⁵ is independently associated with mortality in heart failure with reduced ejection fraction.¹⁶ There is dearth of data evaluating the clinical utility of CPP in advanced HF patients. We therefore sought to evaluate CPP, as a hemodynamic risk stratification tool, in the advanced HF patients enrolled in the Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness (ESCAPE) trial.

Methods

The ESCAPE trial was a multicenter study that was conducted at 26 sites in the United States and Canada between 2000 and 2003. The trial was designed to test the efficacy of pulmonary artery catheter (PAC) in hemodynamically guided therapy of hospitalized patients with acute decompensated advanced HF. Patients were randomly matched to either treatment guided by clinical assessment alone or therapy or clinical assessment and a PAC. The results of the trial have been previously published.¹⁷ Briefly, advanced heart failure patients were randomly assigned on a 1:1 basis to either therapy guided by clinical assessment only or therapy guided by clinical assessment and PAC. Therapy goal in the clinical assessment group was resolution of clinical signs and symptoms of congestion. The PAC group had similar treatment goal, with the addition of a target pulmonary artery wedge pressure (PAWP) of 15 mmHg and right atrial pressure of 8 mmHg. In the PAC group, hemodynamics were measured twice at baseline and at least twice daily thereafter. To evaluate the relationship between CPP with the primary and secondary outcomes in the ESCAPE database, we used the final hemodynamic measurements and analyzed patients in the PAC cohort with complete invasive hemodynamics. The method of hemodynamic measurements has been previously reported in the original ESCAPE trial.¹⁷ We calculated CPP from the final hemodynamic values prior to the removal of the PAC as the difference between aortic diastolic blood pressure and pulmonary artery wedge pressure. The present study is a subgroup analysis of patients who were enrolled in the PAC group. We used a de-identified public version of the ESCAPE database from the National Heart, Lung, and Blood Institute to study the association of CPP and clinical outcomes.

Study design and outcomes

This cohort included patients with complete hemodynamic variables for the calculation of CPP. We calculated CPP from the final hemodynamic values prior to the removal of the PAC. CPP was calculated using the equation: (systemic diastolic blood pressure - PAWP), a previously validated method.^18,19

Associations between CPP and clinical outcomes were analyzed both with CPP as a continuous variable and as a binary variable based on the median CPP. CPP groups were then evaluated for the (1) primary outcome of death, transplant and LVAD (DT × LVAD), or (2) the composite secondary outcome of DT × LVAD and heart failure rehospitalization (DT × LVADHF) at 6 months of follow up. Heart failure categories based on the underlying etiology (ischemic vs non-ischemic) were also stratified according to the CPP strata.

Statistical analysis

Descriptive analyses were used for baseline characteristics. Categorical variables were presented as frequencies with percentages, whereas continuous variables were described using medians and interquartile ranges (IQR). Analyses of categorical variables were conducted using the χ2 test or Fisher’s exact test. The Wilcoxon rank sum test was used to test for differences in continuous variables. Kaplan-Meier plots were used to model survival free of heart transplant/LVAD and or hospitalization between CPP stratas. The log rank test was used to compare differences between groups. Multivariable Cox proportional hazards regression and multivariable logistic regression were used to model associations of CPP with clinical endpoints during follow-up. Receiver operating characteristic (ROC) analysis for determination of the area under the curve (AUC) was also used. The results were expressed as hazard ratio (HR) [95% confidence interval (CI)]. A significance level of 0.05 was used. The proportional hazards assumption was confirmed by evaluating survival curves. Statistical analysis was performed using SAS 9.4 (SAS Institute, Cary, NC).

Results

Cohort definition and aggregate outcomes

The ESCAPE trial database contains records for 433 patients, who were randomized 1:1 to PAC-guided therapy versus management guided by clinical assessment without a PAC. The present analysis was based on 158 patients in the PAC arm who had hemodynamic data available at discharge for calculation of coronary perfusion pressure. Of these 158 patients, 35 (22.2%) died and 32 (20.2%) had LVAD placement or orthotopic heart transplantation during the follow-up period of 180 days. A total of 109 (60.0%) had the secoondary outcome of DT × LVADHF during follow-up.

Baseline characteristics of cohort by coronary perfusion pressure group

The distribution of CPP is shown in Figure 1. Using the median CPP of 40 mmHg as the threshold for stratification, above which CPP was considered high and ≤40 mmHg as low CPP. The clinical characteristics and co-morbid conditions of the two groups are shown in Table 1. There were 26 (32.9%) female patients in the high CPP group versus 19 (24.1%) in the low CPP group (p = .29). There was no significant difference between age, presence of comorbid conditions, or length of stay.

Figure 1.

Frequency distribution of coronary perfusion pressure. The frequency distribution of coronary perfusion pressure is shown.

Table 1.

Baseline patient demographics.

Clinical variable	High CPP (>40 mmHg) Median (IQR) n = 79	Low CPP (≤40 mmHg) Median (IQR) n = 79	p-value
Age (years)	58 (46–66)	57 (49–68)	.57
Sex (male)	53 (67.1)	60 (75.9)	.29
BMI (kg/m2)	27.1 (22.6–33.9)	27.7 (24.4–32.3)	.50
BUN	24 (17–32)	35 (26–51)	<.001
Creatinine	1.3 (0.9–1.6)	1.5 (1.1–2)	.01
Ischemic cardiomyopathy	42 (53.2)	41 (51.9)	1.00
Diabetes mellitus	23 (30.1)	27 (35.3)	.49
Hypertension	41 (51.9)	38 (48.1)	.75
COPD	17 (21.5)	9 (11.4)	.13
Stroke	9 (11.4)	8 (10.1)	1.00
CABG	24 (30.4)	24 (30.4)	1.00
Atrial fibrillation	21 (26.6)	24 (30.4)	.73
Length of stay (days)	4 (3–9.5)	7 (4–11)	.09

BMI: body mass index; CPP: Coronary perfusion pressure; CABG: Coronary artery bypass graft; COPD: Chronic obstructive pulmonary disease.

The mean serum creatinine in the high CPP group was 1.30 (0.9–1.6) versus 1.50 (1.1–2.0) in the low CPP group (p = .01). The final hemodynamic measurements between the two groups are depicted in Table 2. The median systemic systolic blood pressure among patients in the high CPP group was significantly higher than that of the low CPP group [106 mmHg (IQR 95–115) versus 97 mmHg (IQR 90–110), p = .003]. There was no difference in the systemic diastolic blood pressure between groups. Patients in the high CPP group had significantly lower pulmonary systolic pressure, pulmonary diastolic pressure, pulmonary artery wedge pressure and right atrial pressures, 40 mmHg (IQR 34–50) versus 48 mmHg (IQR 42–58), 17 mmHg (IQR 14–24) versus 23 mmHg (IQR 20–26), 14 mmHg (IQR 11–18) versus 19 mmHg (IQR 17–24), 7 mmHg (IQR 4–10) versus 10 mmHg (6–15), (p < .001 for all) respectively. In the multivariable model for the primary outcome, the interaction between CPP and ischemic etiology was statistically significant (p = .04).

Table 2.

Final hemodynamic variables.

Hemodynamic variables (at discharge)	High CPP (>40 mmHg) Median (IQR) n = 79	Low CPP (≤40 mmHg) Median (IQR) n = 79	p-value
Systolic BP (mmHg)	106 (95–115.50)	97.50 (90–110)	.003
Diastolic BP (mmHg)	61 (54–69.50)	58 (50–66)	.100
PASP (mmHg)	40 (34–50)	48 (42–58)	<.001
PADP (mmHg)	17 (14–24)	23 (20–26)	<.001
Mean PAP (mmHg)	25 (20–32)	32 (28–38)	<.001
PAWP (mmHg)	14 (11–18)	19 (17–24)	<.001
Mixed venous blood gas (%)	63 (54–65.5)	60 (54–69)	.970
Cardiac index (L/min/m2)	2.4 (1.9–2.71)	2.3 (2–2.7)	.690
Pulmonary vascular resistance (W.U.)	2.36 (1.58–3.47)	2.73 (1.72–3.89)	.320
Right ventricular stroke work index (g*m/m2)	7.43 (4.28–9.6)	8.01 (6.29–11.6)	.050
Right atrial pressure (mmHg)	7 (4–10)	10 (6–15)	.001

CPP: Coronary perfusion pressure; BP: blood pressure; PADP: Pulmonary artery diastolic pressure; PASP: Pulmonary artery systolic pressure; PAP: Pulmonary artery pressure; PAWP: Pulmonary artery wedge pressure; RAP: Right Atrial Pressure; W.U.: Woods Units.

Clinical outcomes and coronary perfusion pressure stratas

The associations of CPP with 6-month clinical outcomes are shown in Table 3. There were markedly fewer deaths (11 [13.9%]) in the CPP >40 mmHg group compared with the CPP ≤40 mmHg group (24 [30.8%]) (p = .02). There were also significantly fewer patients reaching the secondary endpoint of DTxLVADHF in the high CPP group (44 [55.7%]) compared with the low CPP group (65 [82.3%]) (p = .0003). There were 12 (15.2%) events of DT × LVAD in the high CPP group versus 32 (40.5%) in the low CPP group (p = .0006).

Table 3.

Coronary perfusion pressure and 6-month outcomes.

Outcome variable	High CPP (>40 mmHg): No. of patients (%) n = 79	Low CPP (≤40 mmHg): No. of patients (%) n = 79	p-value
Death	11 (13.92)	24 (30.38)	.02
Death, transplant, or LVAD	12 (15.19)	32 (40.51)	.0006
Death, transplant, LVAD, or HF hospitalization	44 (55.70)	65 (82.28)	.0005

DT × LVAD: Death, Transplant, or Left Ventricular Assist Device; DT × LVADHF: Death, Transplant, Left Ventricular Assist Device, or Rehospitalization; LVAD: Left Ventricular Assist Device.

Association of coronary perfusion pressure on clinical outcomes

To examine the impact of CPP on overall survival during the 6 months of follow-up in the ESCAPE trial, bivariable logistic regression analysis was used to determine the association of CPP with survival at 6 months (Table 4). The models showed that an increase of 10 mmHg in CPP was associated with a lower risk of death at 6 months follow-up (HR: 0.75 [0.58– 0.98]; p = .036). Similarly, an increase of 10 mmHg in CPP was associated with a lower risk of DT × LVAD (HR: 0.68 [0.53–0.87]; p = .0018). The overall survival time free of DT × LVAD are shown for high and low CPP groups in Figure 2. The receiver operating characteristic for the primary outcome of DT × LVAD with the covariate of CPP had an AUC of 0.67 for this outcome (Figure 3).

Table 4.

Cox proportional hazards regression for the outcome of death.

Parameter	Hazard ratio (95% confidence interval)	Chi-square	p-value
Coronary perfusion pressure (per 10 mmHg)	0.75 (0.58–0.98)	4.38	.036
Age (per year)	1.03 (1–1.05)	4.80	.028
Gender (female)	1.32 (0.68–2.53)	0.68	.41
BMI	0.98 (0.93–1.03)	0.65	.42
Baseline creatinine (per mg/dL)	2.08 (1.38–3.15)	12.13	.001
Length of stay (days)	1.06 (0.97–1.15)	1.63	.20
Ischemic heart disease	2.22 (1.12–4.38)	5.27	.022
Hypertension	0.73 (0.39–1.38)	0.95	.33
Diabetes	1.59 (0.83–3.06)	1.97	.16
COPD	1.51 (0.69–3.29)	1.08	.30
Stroke	0.88 (0.31–2.46)	0.06	.80
CABG	1.72 (0.91–3.26)	2.80	.094
Atrial fibrillation	1.58 (0.83–3.01)	1.91	.17
Systolic blood pressure	0.98 (0.96–1.01)	1.71	.19
Diastolic blood pressure	0.98 (0.95–1.01)	1.45	.23
PASP (per mmHg)	1.04 (1.01–1.06)	9.39	.002
PADP (per mmHg)	1.06 (1.03–1.10)	12.42	.0004
PAWP (mmHg)	1.12 (1.07–1.17)	22	<.0001
Mixed venous blood gas (per %)	0.98 (0.96–1)	3.93	.047
Pulmonary vascular resistance (W.U.)	1.22 (1.04–1.42)	6.28	.012
Cardiac index (per L/min/m2)	0.77 (0.44–1.33)	0.90	.34
Right atrial pressure (mmHg)	1.12 (1.08–1.17)	30.43	<.0001

DT × LVADHF: Death, Transplant, LVAD, or Heart Failure Rehospitalization; COPD: Chronic obstructive pulmonary disease; CABG: Coronary artery bypass graft; PADP: Pulmonary artery diastolic pressure; PASP: Pulmonary artery systolic pressure; PAP: Pulmonary artery pressure; PAWP: Pulmonary artery wedge pressure; RAP: Right Atrial Pressure; W.U.: Woods Units.

Figure 2.

Overall survival and survival free of DT × LVAD based on coronary perfusion pressure groups. The Kaplan-Meier survival curves for (a) survival free of DT × LVAD (b) survival free of DT × LVADHF are shown.

Figure 3.

ROC analysis. ROC curve is shown for the logistic regression model for 6 months DT × LVAD and coronary perfusion pressure. ROC: Receiver operating characteristic.

Discussion

The present study evaluated the relationship between CPP measured at final hemodynamic reading with 6-month clinical outcomes among patients with advanced HF enrolled in the ESCAPE trial. We found that patients with CPP >40 mmHg at final hemodynamic measurement had significantly improved adverse outcomes including death, DT × LVAD, and DT × LVADHF. This study demonstrates an independent association of CPP with adverse clinical outcomes in patients hospitalized with advanced HF. These findings have important implications in the risk stratification of advanced HF patients undergoing invasive hemodynamic monitoring. In the present study, low CPP status was significantly associated with increased risk of death, DTxLVADHF, and DT × LVAD at 6 months of follow-up. A low CPP was associated with a phenotypic cluster of unfavorable hemodynamics and clinical parameters that signal advanced disease status in HF. Specifically, low CPP status was significantly associated with hemodynamic markers of right ventricular dysfunction (elevated PASP, PADP, RAP, a reduced mixed venous saturation, and a low pulmonary artery capacitance). These clinical characteristics, either in isolation or in combination have been shown to parallel adverse outcomes in HF patients, independent of the severity of left ventricular failure.^16,20–23 Further, patients in the low CPP strata had reactive pulmonary vascular remodeling, as demonstrated by elevated pulmonary vascular resistance. Elevated pulmonary vascular resistance in HF is often a consequence of combined pre- and post-capillary pulmonary vascular remodeling, and is associated with increased adverse events.^24,25 Low CPP status was also associated with a higher baseline serum creatinine, another important determinant of prognosis in HF.^26,27 Myocardial perfusion is an important determinant of contractile function.^13,28 The myocardium has an intrinsic ability to autoregulate blood supply over a wide range of pressures. However, in animal models, endocardium autoregulation is impaired when CPP decreases to a threshold less than 40 mmHg.²⁹ In the present study, we found that a CPP below 40 mmHg was associated with adverse clinical events. Expectedly, patients with an ischemic etiology of HF had greater association with low CPP, possibly reflecting a lower coronary flow reserve in this cohort.³⁰

Prior studies have demonstrated the central role of CPP in influencing outcomes in specific conditions such as cardiac arrest patients. Maximal CPP was more predictive than other hemodynamic characteristics in predicting return of spontaneous circulation among individuals with cardiac arrest undergoing cardiopulmonary resuscitation.³¹ Moreover, a study involving patients with critical coronary artery stenoses undergoing percutaneous coronary interventions demonstrated that the use of mechanical circulatory support was associated with improvement in CPP, highlighting the target effects of mechanically unloading the left ventricle.³² In animal studies, increasing CPP was associated with increased left ventricular performance and contractility.^13,33 A more recent study investigating the prognostic role of CPP in patients with left ventricular systolic dysfunction undergoing percutaneous coronary intervention who had either complete revascularization or reasonable incomplete revascularization (RIR), found that both CPP and RIR had prognostic utility. When patients with left ventricular systolic dysfunction had CPP >42 mmHg, RIR was equivalent to complete revascularization in survival. However, when patients with left ventricular systolic dysfunction had CPP ≤42 mmHg, RIR had a significantly higher mortality risk than those with complete revascularization.³⁴ CPP is defined by two clinical variables, PAWP (a surrogate marker for LVEDP) and diastolic blood pressure. PAWP is a marker of left ventricular preload and directly affects myocardial wall stress, ventricular efficiency, and stroke volume.³⁵ Diastolic blood pressure on the other hand, is an important determinant of myocardial perfusion as 85% of left ventricular perfusion occurs during diastole.³⁶ The interactions of diastolic blood pressure with cardiovascular outcomes is complex, with a J shaped relationship in various clinical conditions.^37,38 However, in patients with reduced ejection fraction and particularly with ischemic etiology, a low diastolic blood pressure is associated with adverse outcomes.³⁹ In our study, there was no difference in systemic diastolic blood pressure between the two strata of CPP, suggesting that the effect of CPP on outcomes was mostly driven by PAWP.

From a clinical utility standpoint, CPP is easy to calculate, relies on PAWP and DBP, and is significantly and independently associated with clinical outcomes. CPP has the potential to be a useful hemodynamic target during invasive hemodynamic guidance for HF therapy. A lower CPP (cutoff ~40 mmHg) could forewarn physicians that a patient is likely to have an adverse event and stimulate changes to medications or accelerate consideration for LVAD placement or heart transplant. Although the ESCAPE trial failed to demonstrate that the routine use of pulmonary artery catheterization in HF was efficacious,¹⁷ the critical role that invasive hemodynamics play in advanced HF is increasingly recognized.⁴⁰ Further, the management of advanced HF has become increasingly reliant upon invasive hemodynamic monitoring with PAC.⁴¹ Therefore, application of novel hemodynamic indices such as CPP, may further enhance risk stratification of HF to facilitate timely utilization of advanced HF therapies.^42,43

Limitations

This was a subgroup analysis of the ESCAPE trial, limited to patients in the intervention group. Consequently, we do not know if the findings would have been the same if there were hemodynamic data on all patients. Further, the mean age of the patients enrolled in the ESCAPE trial was relatively younger than more recent contemporary HF trials. Even so, the current focus on hemodynamics as the basis of stratifying patients offset this limitation. Furthermore, the prevalence of HF is increasing with the aging population, [1] who are more likely to have concomitant coronary artery disease (low myocardial perfusion reserve) is also a compelling argument in favor of the importance of CPP in the risk stratification of HF patients.

Conclusions

In patients with advanced HF, low CPP was strongly associated with adverse events. CPP promises to be an important clinical tool for risk stratification in HF patients. Further clinical studies investigating the role of CPP in subgroups of heart failure patients are warranted. For example, future studies could explore the temporal changes of CPP during heart failure therapy and heart failure prognosis during invasive hemodynamic monitoring. Other lines of research could also explore the efficacy of pressors/inotropic medications based on CPP cut-offs.