If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Cardiac screening using coronary computed tomography angiography (CCTA) in kidney transplant candidates before transplantation yields both diagnostic and prognostic information. Whether CT-derived fractional flow reserve (FFRCT) analysis provides prognostic information is unknown.
This study aimed to assess the prognostic value of FFRCT for predicting major adverse cardiac events (MACE) and all-cause mortality in kidney transplant candidates.
Methods
Among 553 consecutive kidney transplant candidates, 340 CCTA scans (61%) were evaluated with FFRCT analysis. Patients were categorized into groups based on lowest distal FFRCT; normal >0.80, intermediate 0.80–0.76, and low ≤0.75. In patients with ≥50% stenosis, a lesion-specific FFRCT was defined as; normal >0.80 and abnormal ≤0.80.
The primary endpoint was MACE (cardiac death, resuscitated cardiac arrest, myocardial infarction or revascularization). The secondary endpoint was all-cause mortality.
Results
Median follow-up was 3.3 years [2.0–5.1]. MACE occurred in 28 patients (8.2%), 29 patients (8.5%) died.
When adjusting for risk factors and transplantation during follow-up, MACE occurred more frequently in patients with distal FFRCT ≤0.75 compared to patients with distal FFRCT >0.80: Hazard Ratio (HR): 3.8 (95%CI: 1.5–9.7), p < 0.01.
In the lesion-specific analysis with <50% stenosis as reference, patients with lesion-specific FFRCT >0.80 had a HR for MACE of 1.5 (95%CI: 0.4–4.8), p = 0.55 while patients with lesion-specific FFRCT ≤0.80 had a HR of 6.0 (95%CI: 2.5–14.4), p < 0.01.
Abnormal FFRCT values were not associated with increased mortality.
Conclusion
In kidney transplant candidates, abnormal FFRCT values were associated with increased MACE but not mortality. Use of FFRCT may improve cardiac evaluation prior to transplantation.
Coronary artery disease (CAD) is frequent in patients with severe chronic kidney disease (CKD) and is a leading cause of mortality and morbidity among kidney transplant candidates, even after transplantation.
Cardiac disease evaluation and management among kidney and liver transplantation candidates: a scientific statement from the American heart association and the American College of Cardiology foundation: endorsed by the American society of transplant surgeons, American society of transplantation, and national kidney foundation.
Medical treatment, pre-transplantation coronary revascularization and individualized pre- and postoperative management may reduce the risk of cardiovascular events.
Consequently, although no screening strategy has been validated to improve cardiovascular outcomes, most institutions follow the recommendations from the American Heart Association and the American College of Cardiology Foundation (AHA/ACC) with non-invasive cardiac stress testing prior to kidney transplantation.
Cardiac disease evaluation and management among kidney and liver transplantation candidates: a scientific statement from the American heart association and the American College of Cardiology foundation: endorsed by the American society of transplant surgeons, American society of transplantation, and national kidney foundation.
However, recent studies have questioned the benefit of invasive coronary angiography (ICA) and revascularization guided only by non-invasive cardiac stress testing in patients with severe CKD.
In contrast to anatomy-based coronary computed tomography angiography (CCTA), non-invasive cardiac stress testing provides no information on non-calcified coronary plaque burden and plaque characteristics for individualized medical risk management. Previous studies have suggested that CCTA is feasible in kidney transplant candidates and can be performed in a combined one-bolus contrast investigation of both the coronary arteries and the pelvic vessels for pre-surgical planning.
Coronary CTA rules out obstructive CAD in a high proportion of CKD patients and has higher sensitivity for obstructive CAD than both clinical risk factors and non-invasive cardiac stress testing.
Additionally, CCTA improves prediction of major adverse cardiac events (MACE) and mortality compared to clinical risk factors, non-invasive cardiac stress tests and anatomy-based ICA.
Prognostic value of risk factors, calcium score, coronary CTA, myocardial perfusion imaging, and invasive coronary angiography in kidney transplantation candidates.
Coronary CTA is limited by moderate specificity for obstructive CAD, often caused by arrhythmia, severe coronary calcification and overestimation of luminal diameter reduction.
Coronary computed tomography angiography in diagnosing obstructive coronary artery disease in patients with advanced chronic kidney disease: a systematic review and meta-analysis.
Computed tomography derived fractional flow reserve (FFRCT) allows for evaluation of the hemodynamic severity of a stenosis from standard CCTA images and has been validated in the general population. Furthermore, FFRCT shows improved prediction of MACE compared to luminal diameter reduction by CCTA in the general population
Noninvasive fractional flow reserve derived from computed tomography angiography for coronary lesions of intermediate stenosis severity: results from the DeFACTO study.
; however, FFRCT has not been evaluated in larger cohorts of CKD patients and its potential as a risk stratification and prognostic tool has never been investigated in kidney transplant candidates.
The aim of this study was to evaluate the prognostic value of adding FFRCT to a diagnostic CCTA in patients with CKD stages 4 or 5 undergoing pre-transplant cardiac evaluation.
2. Materials and methods
2.1 Study population
The present study included kidney transplant candidates referred to a single transplantation center, Aarhus University Hospital, from the Central and North Denmark Regions. The study included two cohorts; the ACToR study cohort, which consisted of patients prospectively included between February 2011 and February 2014 10, and a retrospective cohort of patients undergoing first-time screening with CCTA between March 2014 and September 2019. Both cohorts included kidney transplant candidates with either diabetes, age >40 or dialysis >5 years that were referred for systematic cardiovascular screening, with first choice of screening modality being CCTA. Patients were evaluated if they had CCTA performed, few patients at the transplantation center were not referred for primary CCTA mainly due to; established cardiovascular disease or prior different CAD imaging modality. All the included patients underwent a pre-transplant CCTA, in accordance with the inclusion criteria in the ACToR study, which was continued as the local guideline after study completion.
Present study was approved by the relevant local and national Danish authorities.
In all CCTA scans of sufficient quality, FFRCT was calculated. The FFRCT results were not clinically available or used in patient management. Patients were excluded if 1) the scan quality was insufficient for FFRCT analysis, 2) FFRCT analysis of a major vessel was not possible either due to a processing error, or prior revascularization by either stent or coronary artery bypass graft (Fig. 1).
Fig. 1Study flow of patient inclusion. In total, 553 kidney transplantation candidates completed CCTA in the time period 2011–2019. The final study cohort (n = 340) represents patients who completed both CCTA and FFRCT analysis with complete follow-up. ACToR refers to the cohort from the Angiographic CT of Renal Transplantation Candidate–Study; CCTA coronary computed tomography angiography; CVD Cardiovascular disease; FFRCT computed tomography derived fractional flow reserve; ICA invasive coronary angiography; KTX-CTA refers to the cohort of patients with CCTA since ACToR completion. ∗Primary excluded CCTAs had the following artifacts: motion (n = 77), high image noise (n = 27), field of view too large (n = 24), clipping of the heart (n = 8), prior coronary stent or coronary bypass surgery (n = 12), poor image contrast (n = 1) or unspecified (n = 5). † Artifacts in returned CCTAs: motion (n = 21), high image noise (n = 7), misalignment (n = 8), incomplete data (n = 8) artifact in the right coronary artery (n = 2) or prior coronary stent (n = 1). ‡ The incomplete FFRCT was excluded as a major vessel could not be analyzed due to: process deviations (n = 8) or prior coronary stent (n = 4).
Information on CAD clinical risk factors as defined by the ACC/AHA was obtained from patient records: diabetes mellitus, prior cardiovascular disease, > 1 year on dialysis, left ventricular hypertrophy, age >60 years, smoking, hypertension, and dyslipidemia. Outcomes were obtained from the Western Denmark Heart Registry
Patient were followed until death or the end of follow-up which was December 31st, 2019.
2.3 Outcomes
The primary endpoint was time from the day of CCTA to the first MACE event defined as either cardiac death, resuscitated cardiac arrest, myocardial infarction (MI) or revascularization. Planned revascularization related to the baseline cardiac evaluation was not included as a MACE and such patients were not excluded from the study. The secondary endpoint was all-cause mortality. Outcomes were adjudicated by a cardiologist blinded to the study imaging results.
2.4 Computed tomography acquisition and interpretation
The CCTAs were performed with a dual-source scanner (SOMATOM Definition Flash or SOMATOM Force, Siemens Healthcare, Erlangen, Germany) or a 320-slice volume CT scanner (Aquillion One, Toshiba Medical Systems, Japan). An electrocardiogram gated non-enhanced scan was performed to access coronary artery calcium score (CACS). Contrast-enhanced CCTA was acquired with prospective gating. Tube settings were dependent on patient weight, and current modulation was applied. All patients received glyceryl nitrate (0.8 mg) sublingually before CCTA. If appropriate, Metoprolol and/or Ivabradine were administered to optimize target heart rate of <65 beats/min.
Analysis of coronary calcification and stenosis severity were previously performed in the ACToR study cohort.
Prognostic value of risk factors, calcium score, coronary CTA, myocardial perfusion imaging, and invasive coronary angiography in kidney transplantation candidates.
The remaining CCTA scans were analyzed in a core laboratory using commercially available software by an experienced cardiologist, who also analyzed CCTA scans in the ACToR study, readers were blinded to clinical imaging interpretation and outcomes. Coronary segments were visually analyzed according to standard clinical practice. In line with previous reported results,
Prognostic value of risk factors, calcium score, coronary CTA, myocardial perfusion imaging, and invasive coronary angiography in kidney transplantation candidates.
stenosis with ≥50% luminal diameter reduction on CCTA was considered suggestive of obstructive CAD. While CCTA were considered to be without obstructive CAD if coronary plaques were absent or lesions had a maximum stenosis of <50% luminal diameter reduction. The cardiologist also evaluated the diagnostic quality of CCTAs prior to submission for FFRCT analysis (Fig. 1).
2.5 CT–derived fractional flow reserve
All CCTA scans of diagnostic quality were transferred to a blinded, off-site external core laboratory for computation of FFRCT (HeartFlow, Version 2, Redwood City, CA). Computation of FFRCT is done as a post hoc software analysis of initial CCTA images, without any need for a new scan or contrast exposure. All FFRCT values were recorded in the major coronary arteries of ≥1.8 mm in diameter including side branches and occluded vessels.
The pre-specified analysis included a primary evaluation of patients on a per-patient level according to the lowest distal FFRCT value in any vessel analyzed by FFRCT. Based on this, patients were categorized into three groups according to clinical interpretation of FFRCT: 1) normal FFRCT > 0.80; 2) intermediate FFRCT 0.80 to 0.76; and 3) low FFRCT ≤ 0.75 or occluded vessels.
Coronary CT angiography-derived fractional flow reserve testing in patients with stable coronary artery disease: recommendations on interpretation and reporting.
Coronary CT angiography-derived fractional flow reserve testing in patients with stable coronary artery disease: recommendations on interpretation and reporting.
Where only FFRCT values in segments with ≥50% stenosis on CCTA were obtained. Based on the FFRCT value obtained approximately 2 cm distal to a ≥50% stenosis, patients were categorized into the following groups: 1) < 50% stenosis; 2) ≥ 50% stenosis with corresponding normal FFRCT > 0.80; and 3) ≥ 50% stenosis with corresponding abnormal FFRCT ≤ 0.80 (Fig. 2).
Fig. 2Risk of MACE according to distal and lesion-specific FFRCTvalues. Depicted is the evaluation process and hazard ratios (HR) of the central findings. (Panel A) Patients had CCTA with visual assessment, (arrow; stenosis ≥50%) (Panel B) then extraction of distal FFRCT values in all vessels ≥1.8 for primary analysis. In patients with stenosis of ≥50%, FFRCT was obtained approximately 2 cm distal to the lesion and used for lesion-specific analysis. (Panel C) Shown are HRs with (95% CI) and p values, calculated using multivariate analysis adjusting for ≥3 risk factors and kidney transplantation during follow-up. For distal values the HRs are for lower FFRCT groups as compared with the normal FFRCT group. For lesion-specific values, HRs for patients with <50% stenosis are compared to patients with ≥50% stenosis, and either normal or abnormal lesion-specific FFRCT (Table 3), subgroup analysis of patients with stenosis are presented in Table S4. CI denotes Confidence Interval; CCTA coronary computed tomography angiography; CX circumflex artery; FFRCT computed tomography derived fractional flow reserve; LAD Left anterior descending artery; RCA right coronary artery.
Continuous variables were expressed as median with total or interquartile range (IQR) or mean with standard deviations (SD). Dichotomous or categorical variables were reported as n (%). Baseline characteristics were compared using Pearson chi-square test or Fisher exact test; means and medians using the Kruskal-Wallis test. Time-to-event analysis was performed with patients categorized according to distal and lesion-specific FFRCT values. The cumulative incidence of the primary outcome, MACE, was estimated using the Aalen-Johansen method with death as competing risk, while the Kaplan Meier method was used for mortality. Event rates were compared using a log-rank test. Univariate and multivariate Cox regression of hazard ratios (HR) were performed for each of the FFRCT approaches, with number of AHA risk factors ≥3 as a covariate and kidney transplantation during follow-up as a time-dependent covariate. The models were examined for the proportional hazards assumption, which was satisfied according to Schoenfeld residuals tests. Harell's C-Statistic was used to evaluate the addition of FFRCT to base prognostic models of ≥3 risk factors, CACS and presence of ≥50% stenosis on CCTA.
For all statistical analyses, 95% confidence intervals (CI) were reported when appropriate and a two-tailed p value < 0.05 was considered statistically significant. The statistical analyses were performed using Stata 16 (StataCorp, College Station, Texas).
3. Results
3.1 Study population
As CCTA was introduced as primary CAD screening modality, 76% of all kidney transplantation candidates referred for CAD screening completed CCTA, main reasons for non-completion were established CAD or very high CACS. Among the 553 consecutive patients referred for kidney transplant evaluation with CCTA performed, 340 patients (61%) had coronary CCTA analyzable for FFRCT (Fig. 1). Of the non-analyzed CCTAs, 154 (28%) were excluded locally for technical reasons. An additional 59 (11%) patients did not have successful FFRCT analysis as CCTAs were inadequate according to the external lab. Patients who didn't have FFRCT performed due to the presence of artifacts, generally had a higher heart rate, more risk factors, higher CACS scores and more frequently established cardiovascular disease, see Table S1.
Among the 340 patients with FFRCT analysis, 213 (63%) were men, mean age was 53 (±12) years and 107 (32%) were on dialysis. Patient and imaging baseline characteristics are presented in Table 1, a comparison of characteristics between groups is presented in Table S2.
Table 1Baseline patient and imaging characteristics (n = 340).
Ten patients didn't have a non-contrast scan to reduce radiation exposure. For a comparison of the primary analysis groups see Table S2. CAD denotes coronary artery disease; CCTA coronary computed tomography angiography.
Values presented as n (%), mean (Standard Deviation) or median [interquartile range].
a Estimated glomerular filtration rate was calculated using the CKD-EPI equation.
b Smoking, never: Twenty patients had no recorded history of smoking, and was categorized as never.
c Ten patients didn't have a non-contrast scan to reduce radiation exposure. For a comparison of the primary analysis groups see Table S2. CAD denotes coronary artery disease; CCTA coronary computed tomography angiography.
Ten patients didn't have a non-contrast scan to reduce radiation exposure. Number of MACE and all-cause mortality during a median follow-up time of 3.3 years according to CCTA, distal and lesion-specific FFRCT values. During follow-up, 41 (12%) patients experienced either a MACE or mortality endpoint. CACS denotes coronary artery calcium score, CCTA coronary computed tomography angiography, CAD coronary artery disease; FFRCT computed tomography derived fractional flow reserve; MACE major adverse cardiac event NSTEMI, Non-ST-elevation myocardial infarction; STEMI, ST-elevation myocardial infarction.
119 (36.1%)
152 (46.1%)
59 (17.9%)
MACE, n = 27 (8.2%)
5 (4.2%)
9 (5.9%)
13 (22.0%)
Cardiac death
0
4
5
Resuscitated cardiac arrest
4
1
1
STEMI
0
4
1
NSTEMI
1
0
3
Revascularization
0
0
3
Mortality, n = 28 (8.5%)
8 (6.7%)
8 (5.3%)
12 (20.3%)
CCTA – Stenosis evaluation
No CAD
Stenosis < 50%
Stenosis ≥ 50%
Number of patients
95 (27.9%)
130 (38.2%)
115 (33.8%)
MACE, n = 28 (8.2%)
5 (5.3%)
4 (3.1%)
19 (16.5%)
Cardiac death
0
2
8
Resuscitated cardiac arrest
4
0
1
STEMI
0
2
3
NSTEMI
1
0
4
Revascularization
0
0
3
Mortality, n = 29 (8.5%)
5 (5.3%)
9 (6.9%)
15 (13.0%)
Computed Tomography Derived Fractional Flow Reserve and Clinical Endpoints including MACE components, Values presented as n (%) or total cumulative incidence (%).
a Ten patients didn't have a non-contrast scan to reduce radiation exposure. Number of MACE and all-cause mortality during a median follow-up time of 3.3 years according to CCTA, distal and lesion-specific FFRCT values. During follow-up, 41 (12%) patients experienced either a MACE or mortality endpoint. CACS denotes coronary artery calcium score, CCTA coronary computed tomography angiography, CAD coronary artery disease; FFRCT computed tomography derived fractional flow reserve; MACE major adverse cardiac event NSTEMI, Non-ST-elevation myocardial infarction; STEMI, ST-elevation myocardial infarction.
Computed Tomography Derived Fractional Flow Reserve and Clinical Endpoints including MACE components, Values presented as n (%) or total cumulative incidence (%).
Patients were followed for a median of 3.3 years (IQR: 2.0–5.1) and no patients were lost to follow-up. During follow-up 211 (62%) patients were transplanted. Thirteen patients (3.8%) were revascularized using percutaneous coronary intervention (PCI) following baseline evaluation. All baseline revascularized patients had distal FFRCT ≤ 0.75, and 12 (92%) had lesion-specific FFRCT ≤ 0.80. The primary endpoint, MACE, occurred in 28 (8.2%) patients. Ten patients had a cardiac death event as index MACE. Twelve patients suffered an MI at the index event, with 10 having distal FFRCT ≤ 0.75, two STEMI's were fatal. The three late revascularizations occurred in patients with both low distal and lesion-specific FFRCT. Among patients with normal or intermediate distal FFRCT 10 MACE occurred with cardiac death (n = 5) or resuscitated cardiac arrest (n = 4) as the main events. The secondary endpoint, all-cause mortality, occurred in 29 (8.5%) patients (Table 2a, Table 2ba, b, Table S3).
3.2 Distal FFRCT analysis
3.2.1 MACE
In the primary FFRCT analysis, the risk of MACE increased with decreasing FFRCT values (Figure S1). Using FFRCT > 0.80 as reference, the risk of MACE was significantly increased for patients with a FFRCT ≤ 0.75 (HR: 3.80; 95% CI: 1.51–9.58; p < 0.01), and non-significantly increased for patients with a FFRCT of 0.80 to 0.76 (HR: 1.70; 95% CI: 0.48–6.03; p = 0.41) (Fig. 3A, Table 3). The HRs were similar after adjusting for having ≥3 risk factors and transplantation during follow-up (Table 3, Fig. 2). In a subgroup analysis distal FFRCT did not predict MACE in patients with <50% stenosis on CCTA (Table S3). Addition of distal FFRCT to a base prognostic model of ≥3 risk factors did increase C-statistic from 0.54 to 0.68; p < 0.01, but did not cause a significant increase in C-statistic for the CCTA models (Table 4).
Fig. 3Hazard curves for clinical endpoints according to distal FFRCTvalues. Shown are the results of the unadjusted time-to-event analysis for computed tomography derived fractional flow reserve distal (FFRCT) values by the primary endpoint, MACE (major cardiac adverse event) (Panel A) and secondary endpoint (all-cause mortality) (Panel B). See Table 3 for hazard ratios.
Table 3Cox regression analysis of clinical endpoints.
MACE
All-cause mortality
3a. FFRCT– distal values, all patients (n = 340)
Univariate analysis
FFRCT > 0.80
Ref.
Ref.
FFRCT 0.80 to 0.76
1.70 (0.48–6.03); p = 0.41
1.57 (0.60–4.07); p = 0.36
FFRCT ≤ 0.75
3.80 (1.51–9.58); p < 0.01
1.16 (0.50–2.69); p = 0.73
Multivariate analysis
FFRCT > 0.80
Ref.
Ref.
FFRCT 0.80 to 0.76
1.55 (0.43–5.50); p = 0.50
1.39 (0.53–3.62); p = 0.50
FFRCT ≤ 0.75
3.79 (1.49–9.65); p < 0.01
0.97 (0.41–2.30); p = 0.95
Number of risk factors ≥3
0.89 (0.41–1.93); p = 0.77
2.10 (0.93–4.77); p = 0.74
Kidney transplantation (post)
0.19 (0.09–0.43); p < 0.01
0.07 (0.03–0.18); p < 0.01
3b. FFRCT– lesion-specific values, all patients (n = 340)
Univariate analysis
Stenosis <50%
Ref.
Ref.
Stenosis ≥50%
-
Lesion-specific FFRCT > 0.80
2.14 (0.66–6.95); p = 0.21
2.05 (0.79–5.33); p = 0.14
-
Lesion-specific FFRCT ≤ 0.80
6.16 (2.68–14.09); p < 0.01
2.08 (0.90–4.81); p = 0.09
Multivariate analysis
Stenosis <50%
Ref.
Ref.
Stenosis ≥50%
-
Lesion-specific FFRCT > 0.80
1.45 (0.43–4.83); p = 0.55
1.07 (0.53–3.62); p = 0.90
-
Lesion-specific FFRCT ≤ 0.80
5.99 (2.50–14.37); p < 0.01
1.35 (0.56–3.26); p = 0.50
Number of risk factors ≥3
0.68 (0.30–1.53); p = 0.35
1.91 (0.82–4.38); p = 0.13
Kidney transplantation (post)
0.20 (0.08–0.45); p < 0.01
0.07 (0.02–0.18); p < 0.01
3c. FFRCT– lesion-specific values, only patients with ≥ 50 stenosis (n = 115)
Univariate analysis
Lesion-specific FFRCT > 0.80
Ref
Ref
Lesion-specific FFRCT ≤ 0.80
2.83 (0.94–8.54); p = 0.07
1.01 (0.36–2.85); p = 0.98
Multivariate analysis
Lesion-specific FFRCT > 0.80
Ref
Ref
Lesion-specific FFRCT ≤ 0.80
3.22 (1.02–10.28); p < 0.05
1.51 (0.48–4.80); p = 0.48
Number of risk factors ≥3
0.98 (0.36–2.66); p = 0.97
1.41 (0.42–4.74); p = 0.58
Kidney transplantation (post)
0.43 (0.16–1.17); p = 0.10
0.08 (0.02–0.34); p < 0.01
Unadjusted and Adjusted Cox Regression Analysis of Clinical Endpoints, Shown are hazard ratios (HR) with (95% confidence intervals) and p values, calculated using univariate unadjusted analysis and multivariate analysis adjusting for ≥ 3 risk factors and kidney transplantation during follow-up. Table 3a contains distal values where HRs are for lower FFRCT groups as compared with the normal FFRCT group. Table 3b contains lesion-specific values, HRs for patients with no stenosis are compared with patients with ≥ 50% stenosis, and either normal or abnormal lesion-specific FFRCT. Table 3c contains subgroup analysis of patients with normal lesion-specific FFRCT compared to abnormal lesion-specific FFRCT. Abbreviations as in Table 2.
Table 4Prognostic models with computed tomography derived fraction flow reserve.
Model
MACE
All-cause mortality
Harrell's C-Statistic (95% CI)
p value
Harrell's C-Statistic (95% CI)
p value
Risk Factors + FFRCT
Risk Factors ≥3
0.54 (0.45-0.64)
0.65 (0.56-0.74)
Risk Factors ≥3 + Distal FFRCT
0.68 (0.58–0.78)
<0.01
0.64 (0.54–0.75)
0.78
Risk Factors ≥3 + Lesion-specific FFRCT
0.71 (0.61–0.81)
<0.01
0.68 (0.58–0.77)
0.20
CACS + FFRCT
CACS
0.68 (0.58–0.79)
0.62 (0.50–0.74)
CACS + Distal FFRCT
0.73 (0.63–0.84)
0.07
0.64 (0.52–0.76)
0.73
CACS + Lesion-specific FFRCT
0.71 (0.60–0.83)
0.19
0.65 (0.54–0.77)
0.62
Stenosis ≥ 50% + FFRCT
Stenosis ≥50%
0.69 (0.60–0.78)
0.60 (0.50–0.69)
Stenosis ≥50% + Distal FFRCT
0.72 (0.61–0.82)
0.22
0.60 (0.49–0.72)
0.87
Stenosis ≥50% + Lesion-Specific FFRCT
0.71 (0.61–0.81)
<0.05
0.60 (0.50–0.70)
1.00
Harell's C-Statistic for coronary computed tomography angiography prognostic models. Shown are Harell's C-statistics for the time to event models with 95% confidence intervals, and p values for the difference between the combined model and the base model. CACS was split into three groups as in Table 2. Abbreviations as in Table 2.
There was no difference in mortality between distal FFRCT groups using the FFRCT > 0.80 group as reference vs. the FFRCT 0.80 to 0.76 group (HR: 1.57; 95% CI: 0.60–4.07; p = 0.36), and vs. the FFRCT ≤ 0.75 group (HR: 1.16; 95% CI: 0.50–2.69; p = 0.73) (Fig. 3B, Table 3,Table S3).
3.3 Lesion-specific FFRCT analysis
3.3.1 MACE
Using patients with <50% stenosis as reference, in patients with ≥50% stenosis and lesion-specific FFRCT of >0.80, there was no significant association with MACE: HR 2.14: 95% CI: 0.66–6.95); p = 0.21, whereas FFRCT of ≤0.80 revealed increased risk of MACE: HR: 6.16; 95% CI: 2.68–14.09; p < 0.01.
When the analysis was confined to the subgroup of patients with ≥50% stenosis (n = 115), the risk of MACE in patients with a lesion-specific FFRCT ≤ 0.80 compared to FFRCT > 0.80 was HR: 2.83; 95% CI: 0.94–8.54; p = 0.07. (Fig. 4A, Table 3,Table S4). Adjusting for ≥3 risk factors and transplantation during follow-up, the higher MACE rate in patients with lesion-specific FFRCT ≤ 0.80 vs. lesion-specific FFRCT > 0.80 was statistically significant (HR: 3.22, 95% CI: 1.02–10.28; p < 0.05), aside of this the results were similar (Table 3,Table S4, Fig. 2). Similar to distal FFRCT, addition of lesion-specific FFRCT increased the C-statistic for ≥3 risk factors model from 0.54 to 0.71; p < 0.01, and caused a non-significant change in the CACS model, while for 50% stenosis it increased the C-statistic from 0.69 to 0.71; p < 0.05 (Table 4).
Fig. 4Hazard curves for clinical endpoints according to lesion-specific FFRCTvalues. Shown are the results of the unadjusted time-to-event analysis for computed tomography derived fractional flow reserve lesion-specific (FFRCT) values by the primary endpoint, MACE (major cardiac adverse events) (Panel A) and secondary endpoint (all-cause mortality) (Panel B). See Table 3 for hazard ratios.
There was a trend for increased mortality in patients with lesion-specific FFRCT ≤ 0.80 compared to patients with< 50% stenosis (HR: 2.08; 95% CI: 0.90–4.81; p = 0.09). (Fig. 4B, Table 3,Table S4). Adjusted analysis did not show evidence of an association between lesion-specific FFRCT and mortality (Table 3,Table S5).
Regarding all-cause mortality, adding the FFRCT approaches to any of the base prognostic models did not alter C-statistic significantly (Table 4).
4. Discussion
Among 340 CKD patients without previous CAD referred for cardiac evaluation prior to kidney transplantation, a distal FFRCT ≤ 0.75 was associated with a higher incidence of MACE after adjusting for kidney transplantation and ≥3 risk factors compared to patients with a normal FFRCT. Furthermore, in secondary analysis, the combination of a ≥50% stenosis and a lesion-specific FFRCT ≤ 0.80 was associated with a higher incidence of MACE compared to both patients with< 50% stenosis and patients with ≥50% stenosis and a lesion-specific FFRCT > 0.80.
4.1 FFRCT and CAD screening in severe CKD patients
The present study differed in important ways from previous FFRCT studies.
Foremost, it only included patients with severe CKD. Secondly, the CCTA was used as part of a routine pre-operative evaluation in combination with an angiography of the pelvic vessels in a high-risk population rather than on the basis of clinically suspected CAD.
The definitive strategy regarding medical treatment or revascularization in cases of obstructive CAD in CKD patients remains to be established. The ISCHEMIA-CKD trial randomized 777 patients with advanced kidney disease (eGFR <30) and moderate to severe ischemia on stress testing to either an initial invasive strategy or medical treatment.
The authors found no evidence that an initial invasive strategy reduced the risk of death or nonfatal MI as compared with medical treatment.
Although current study cohort differs from the cohort in ISCHEMIA-CKD trial, it could be argued based on their findings, that kidney transplant candidates should undergo CCTA to identify obstructive and non-obstructive lesions in order to institute relevant medical treatment and only undergo ICA if any symptoms of CAD persist. Such a strategy differs from current standard of care, which includes cardiac evaluation of CAD with potential revascularization prior to acceptance for the kidney transplantation waiting list. However, the benefit of prophylactic coronary revascularization compared to optimal medical treatment before kidney transplantation is not unequivocal, and current revascularization practice is based on current standard of care and an older study (4). Nevertheless, baseline cardiac evaluation of CAD prior to kidney transplantation wait-listing continues to be practice in many centers. An ongoing randomized study aims to determine whether regular screening tests for CAD after baseline CAD screening provides any benefits for the prevention of MACE when compared to screening prior to wait-listing only.
The present study demonstrates the prognostic value of such baseline evaluation using CCTA and FFRCT, which could enable cardiac risk stratification and individualized pre-operative management. Hence, an FFRCT strategy might enable identification of patients with hemodynamically obstructive stenosis in whom invasive testing and revascularization might be considered, or where further invasive testing might be avoided.
4.2 Mortality in severe CKD
Neither FFRCT model demonstrated a significant association to all-cause mortality. This is concordant with previous findings in the general population.
The main reason might be that half of the deaths in our study were not cardiovascular, and some of the cardiac deaths may be attributable to other causes than CAD (Table S6), despite our definition of cardiovascular mortality being in line with the general interpretation in severe CKD patients.
Other studies have found diverging prognostic value of cardiac imaging in CKD patients, in a previous study ≥50% stenosis on CCTA was associated with increased mortality, whereas myocardial perfusion imaging was not,
Prognostic value of risk factors, calcium score, coronary CTA, myocardial perfusion imaging, and invasive coronary angiography in kidney transplantation candidates.
Coronary CT angiography-derived fractional flow reserve testing in patients with stable coronary artery disease: recommendations on interpretation and reporting.
No evidence of increased risk of MACE or mortality was associated with this grey zone FFRCT, although the few events and fewer subjects in this group may limit any definite conclusions. A recent trial showed that revascularization of invasively measured FFR with values of 0.80 to >0.76 does not improve prognosis.
In addition, the large FFRCT ADVANCE Registry demonstrated similar MACE rates for patients with FFRCT values of 0.80 to >0.75 compared to 0.85 to >0.80.
The number of patients with abnormal lesion-specific FFRCT were too few to allow similar stratification.
4.4 Stenotic lesions and FFRCT
The interpretation of a low distal FFRCT value without stenotic lesions is challenging, and in our study, 90 (26.5%) patients had a distal FFRCT ≤ 0.80 with <50% stenosis on CCTA. This group did not reveal a higher incidence of MACE or mortality compared to patients with >0.80 FFRCT with <50% stenosis. The possible reasons for low distal FFRCT values in the absence of ≥50% stenosis were not investigated in this cohort, but potential explanations include the presence of diffuse non-obstructive CAD, microvascular disease, high CACS, or low coronary lumen volume/myocardial mass ratios.
Safety of coronary revascularization deferral based on fractional flow reserve and instantaneous wave-free ratio in patients with chronic kidney disease.
The aim was to examine if FFRCT values related to a CCTA-verified stenosis ≥50% would increase the prognostic value. Among patients with ≥50% stenosis and normal distal or lesion-specific FFRCT, we found a low incidence of MACE. On the contrary, the risk of MACE was increased in patients with ≥50% stenosis and abnormal FFRCT, although this was only significant in the adjusted analysis of lesion-specific FFRCT. This finding in conjunction with a significant increase in C-statistic suggests an incremental value of FFRCT in patients with ≥50% stenosis on CCTA (Table 3, Table 4, Figure S2, Tables S3–S5).
4.5 Safety concerns
In CKD patients not on dialysis, the use of iodinated contrast for CCTA may raise concerns for the risk of contrast-induced acute kidney injury or accelerated loss of kidney function. Nonetheless, iodinated contrast examination may be needed to evaluate the pelvic vessels for pre-surgical planning and CCTA can be performed during the same contrast injection.
Repeated contrast administration is associated with low risk of postcontrast acute kidney injury and long-term complications in patients with severe chronic kidney disease.
The study included kidney transplant candidates at a single transplantation center referred for CCTA as per local practice. A minor fraction of patients underwent alternative CAD screening methods due to mainly established CAD. Optimization and adherence to medical treatment after the baseline cardiac evaluation was not recorded and could not be adjusted for.
Secondly, a considerable portion of CCTA's did not have complete FFRCT analysis (39%). The exclusion rate was larger than for prospective dedicated FFRCT trials,
Several well-known factors might have contributed to the low FFRCT feasibility, like higher heart rate and established CAD in the excluded kidney transplant candidates. Additionally, this particular cohort had CCTA performed as a one-bolus contrast investigation of both coronary arteries and pelvic vessels, thus in some patients CCTA was carried out despite extreme levels of coronary calcification and high heart rates. On that note level of CACS was elevated in patients where FFRCT was not feasible. Selection bias might have occurred due to above mentioned factors and the overall results should be interpreted with caution in patients with severe coronary calcification, as most of the included patients did not have severe calcification. The ability of newer CT scanners to obtain good image quality enabling FFRCT analysis is likely to increase, in particular among this challenging patient group with sinus tachycardia and coronary calcification.
4.7 Clinical implications
While FFRCT has mainly been used as a tool to guide decision making regarding invasive referral and revascularization, FFRCT can also be employed as an additional risk stratification tool among kidney transplant candidates without established CAD, who are referred for CCTA as part of the work-up prior to kidney transplantation. Based on the present findings patients with <50% stenosis or ≥50% stenosis with FFRCT > 0.80 have a low risk of MACE, in which further diagnostic testing may have limited clinical value.
Whereas abnormal FFRCT values carry higher risk of adverse cardiac events primarily in patients with concomitant coronary stenosis. In kidney transplant candidates with CCTA, FFRCT provides prognostic information, in line with conventional models using CACS and visual stenosis assessment as previously proved,
Prognostic value of risk factors, calcium score, coronary CTA, myocardial perfusion imaging, and invasive coronary angiography in kidney transplantation candidates.
and addition of lesion-specific FFRCT may even improve MACE prediction. Further studies using plaque location, quantitative and qualitative plaque assessment and comparing the different methods are warranted.
5. Conclusions
In conclusion 61% of consecutive kidney transplant candidates, evaluated by CCTA, had successful FFRCT analysis. In these patients distal FFRCT ≤ 0.75 was a predictor of MACE. Lesion-specific analysis suggested that FFRCT ≤ 0.80 carried higher risk in patients with CCTA-verified ≥ 50% stenosis compared to patients with< 50% stenosis or patients with lesion-specific FFRCT > 0.80. FFRCT did not predict all-cause mortality in kidney transplant candidates. FFRCT adds prognostic information in the cardiac evaluation of kidney transplant candidates.
Data availability statement
The data underlying this article cannot be shared publicly due to the privacy of the participants.
Declaration of competing interest
A) Jonathan Nørtoft Dahl: The Health Research Fund of the Central Denmark Region sponsored part of the PhD Fellowship. No other disclosures.
B) Marie Bodilsen Nielsen: The Health Research Fund of the Central Denmark Region, The Augustinus Foundation (MBN). The Aarhus University PhD scholarship fund sponsored part of the PhD Fellowship. No other disclosures.
C) Simon Winther: SW collaborated sponsorship of the FFRCT analysis by the HeartFlow Inc, US. With no exchange of financial means.
D) Henrik Birn: None.
E) Laust Dupont Rasmussen: None.
F) Morten Bøttcher: None.
G) My Hanna Sofia Svensson: None.
H) Per Ivarsen: None.
I) Sripal Bangalore: None.
Funding
The authors, JND, MBN received grants for salaries as PhD fellows related to the project. Grants from The Health Research Fund of the Central Denmark Region (JND, MBN). The Augustinus Foundation (MBN). The Aarhus University PhD scholarship fund (MBN). However, the sponsors had no role in any part of design, conduct, collection, analysis, interpretation, preparation, review, or approval of the manuscript.
The remaining authors have nothing to disclose.
Acknowledgements
No acknowledgments to declare.
Appendix A. Supplementary data
The following is the supplementary data to this article:
Cardiac disease evaluation and management among kidney and liver transplantation candidates: a scientific statement from the American heart association and the American College of Cardiology foundation: endorsed by the American society of transplant surgeons, American society of transplantation, and national kidney foundation.
Prognostic value of risk factors, calcium score, coronary CTA, myocardial perfusion imaging, and invasive coronary angiography in kidney transplantation candidates.
Coronary computed tomography angiography in diagnosing obstructive coronary artery disease in patients with advanced chronic kidney disease: a systematic review and meta-analysis.
Noninvasive fractional flow reserve derived from computed tomography angiography for coronary lesions of intermediate stenosis severity: results from the DeFACTO study.
Coronary CT angiography-derived fractional flow reserve testing in patients with stable coronary artery disease: recommendations on interpretation and reporting.
Safety of coronary revascularization deferral based on fractional flow reserve and instantaneous wave-free ratio in patients with chronic kidney disease.
Repeated contrast administration is associated with low risk of postcontrast acute kidney injury and long-term complications in patients with severe chronic kidney disease.
00040-5&id=ga1.jpg){kind=link}
00040-5&id=gr1.jpg){kind=link}
00040-5&id=gr2.jpg){kind=link}
00040-5&id=gr3.jpg){kind=link}
00040-5&id=gr4.jpg){kind=link}