The Association of Angiogenesis Markers With Acute Kidney Injury and Mortality After Cardiac Surgery

Abstract

Rationale & objective: The process of angiogenesis after kidney injury may determine recovery and long-term outcomes. We evaluated the association of angiogenesis markers with acute kidney injury (AKI) and mortality after cardiac surgery.

Study design: Prospective cohort.

Setting & participants: 1,444 adults undergoing cardiac surgery in the TRIBE-AKI (Translational Research Investigating Biomarker Endpoints for Acute Kidney Injury) cohort.

Exposures: Plasma concentrations of 2 proangiogenic markers (vascular endothelial growth factor A [VEGF] and placental growth factor [PGF]) and 1 antiangiogenic marker (soluble VEGF receptor 1 [VEGFR1]), measured pre- and postoperatively within 6 hours after surgery.

Outcomes: AKI, long AKI duration (≥7 days), and 1-year all-cause mortality.

Analytical approach: Multivariable logistic regression.

Results: Following cardiac surgery, plasma VEGF concentrations decreased 2-fold, and PGF and VEGFR1 concentrations increased 1.5- and 8-fold, respectively. There were no meaningful associations of preoperative concentrations of angiogenic markers with outcomes of AKI and mortality. Higher postoperative VEGF and PGF concentrations were independently associated with lower odds of AKI (adjusted ORs of 0.89 [95% CI, 0.82-0.98] and 0.69 [95% CI, 0.55-0.87], respectively), long AKI duration (0.65 [95% CI, 0.49-0.87] and 0.48 [95% CI, 0.28-0.82], respectively), and mortality (0.74 [95% CI, 0.62-0.89] and 0.46 [95% CI, 0.31-0.68], respectively). In contrast, higher postoperative VEGFR1 concentrations were independently associated with higher odds of AKI (1.56; 95% CI, 1.31-1.87), long AKI duration (1.75; 95% CI, 1.09-2.82), and mortality (2.28; 95% CI, 1.61-3.22).

Limitations: Angiogenesis markers were not measured after hospital discharge, so we were unable to determine long-term trajectories of angiogenesis marker levels during recovery and follow-up.

Conclusions: Higher levels of postoperative proangiogenic markers, VEGF and PGF, were associated with lower AKI and mortality risk, whereas higher postoperative antiangiogenic VEGFR1 levels were associated with higher risk for AKI and mortality.

Keywords: AKI duration; Acute kidney injury (AKI); VEGF-A; angiogenesis; angiogenic growth factor; biomarker; cardiac surgery; cytokine; mortality; placental growth factor (PGF); soluble VEGF receptor 1 (VEGFR1); vascular endothelial growth factor A (VEGF).

Figure 1.

A. Boxplots of vascular endothelial growth factor (VEGF), plasma placental growth factor (PGF), and VEGF receptor 1(VEGFR1) concentrations are presented at preoperative time point and on postoperative days 1 and 2. Each angiogenic marker had a distinct trajectory after cardiac surgery. VEGF initially decreased postoperatively on day 1 and approached baseline on day 2. Both PGF and VEGFR1 continued to increase postoperatively on day 1 and day 2. Each box represents the interquartile range (IQR), the horizontal lines represent the median, and the lower and upper whiskers represent the 5th and 95th percentile, respectively. For the VEGF boxplot, 123 values (400–2545) were excluded from visual presentation. For the PGF boxplot, 25 values (100–158) were excluded from visual presentation. For the VEGFR1 boxplot, 98 values (2000–16340) were excluded from visual presentation. All comparisons had a p-value <0.001 B. Boxplots of median fold change in angiogenic marker levels from preoperative baseline levels. VEGF concentrations decreased 2-fold on postoperative day 1, but approached preoperative levels on day 2. PGF concentrations increased 1.5-fold on day 1 and remained elevated to a similar extent on day 2. VEGFR1 concentrations increased 8-fold on day 1 and decreased to a 2-fold increase relative to preoperative levels on day 2. All comparisons had a p-value <0.001.

Figure 2.

Forest plots summarizing the associations of plasma vascular endothelial growth factor (VEGF), placental growth factor (PGF), and VEGF receptor 1 (VEGFR1) with the outcomes of AKI, long AKI duration, and one-year all-cause mortality after cardiac surgery. For the outcomes of AKI and AKI duration ≥ 7 day, odds ratios (ORs) are adjusted for age (years), gender, white race, CPB time >120 minutes, non-elective surgery, surgery type, preoperative eGFR, diabetes, hypertension, congestive heart failure, myocardial infarction, preoperative urine albumin-creatinine ratio, site, and corresponding preoperative marker. For the outcome of mortality, ORs are adjusted for the above covariates, as well as change in serum creatinine from the preoperative level to postoperative day 1 level. Postoperative VEGF concentrations were independently associated with lower risk of adverse outcomes with 11% lower odds of AKI, 35% lower odds of long duration of AKI, and 26% lower odds of mortality. Postoperative PGF concentrations were independently associated with lower risk of adverse outcomes with 31% lower odds of AKI, 52% lower odds of long duration of AKI, and 54% lower odds of mortality. In contrast, higher postoperative VEGFR1 concentrations were independently associated with higher risk for each outcome with 56% higher odds of AKI, 75% higher odds of long duration of AKI, and about 2 times higher odds of mortality.

Figure 3.

Receiver operating characteristic (ROC) curves showing the performance of the combined angiogenesis panel plus the clinical model for the outcomes of acute kidney injury (AKI), long duration of AKI and one-year mortality. When added to the clinical model, the combined postoperative angiogenesis marker panel significantly improved the area under the curve (AUCs) for AKI to 0.72 (95% CI, 0.69–0.75), for long duration of AKI to 0.88 (95% CI, 0.84–0.92) and for all-cause mortality to 0.74 (95% CI, 0.69–0.80). The changes in AUCs between the clinical model and the clinical plus combined angiogenesis panel were statistically significant for all outcomes. The clinical model included age, gender, race, cardiopulmonary bypass time >120 minutes, non-elective surgery, type of surgery (coronary artery bypass graft or valve surgery), preoperative eGFR, diabetes, hypertension, heart failure, myocardial infarction, preoperative urine albumin to creatinine ratio, and change in serum creatinine.