To assess the safety and efficacy of deferred versus complete revascularization using a fractional flow reserve (FFR)–guided strategy in patients with diabetes mellitus (DM), we analyzed all DM patients who underwent FFR-guided revascularization from January 1, 2010, to December 12, 2013. Patients were divided into 2 groups: those with ≥1 remaining FFR-negative (>0.80) medically treated lesions [FFR(−)MT] and those with only FFR-positive lesions (≤0.80) who underwent complete revascularization [FFR(+)CR] and were followed until July 1, 2015. The primary end point was the incidence of major adverse cardiovascular events (MACE), a composite of death, myocardial infarction (MI), target lesion (FFR assessed) revascularization, and rehospitalization for acute coronary syndrome. A total of 294 patients, 205 (69.7%) versus 89 (30.3%) in FFR(−)MT and FFR(+)CR, respectively, were analyzed. At a mean follow-up of 32.6 ± 18.1 months, FFR(−)MT was associated with higher MACE rate 44.0% versus 26.6% (log-rank p = 0.02, Cox regression–adjusted hazard ratio [HR] 2.01, 95% confidence interval [CI] 1.21 to 3.33, p <0.01), and driven by both safety and efficacy end points: death/MI (HR 2.02, 95% CI 1.06 to 3.86, p = 0.03), rehospitalization for acute coronary syndrome (HR 2.06, 95% CI 1.03 to 4.10, p = 0.04), and target lesion revascularization (HR 3.38, 95% CI 1.19 to 9.64, p = 0.02). Previous MI was a strong effect modifier within the FFR(−)MT group (HR 1.98, 95% CI 1.26 to 3.13, p <0.01), whereas this was not the case in the FFR(+)CR group (HR 0.66, 95% CI 0.27 to 1.62, p = 0.37). Significant interaction for MACE was present between FFR groups and previous MI (p = 0.03). In conclusion, in patients with DM, particularly those with previous MI, deferred revascularization is associated with poor medium-term outcomes. Combining FFR with imaging techniques may be required to guide our treatment strategy in these patients with high-risk, fast-progressing atherosclerosis.
Fractional flow reserve (FFR) is presently the guideline-recommended invasive ischemic assessment of intermediate coronary lesions. However, despite proved superiority in primarily stable angina patients, an FFR-guided revascularization strategy has been extrapolated to high-risk patient subgroups, without robust clinical evidence. Incomplete revascularization is associated with worse outcomes compared with complete revascularization, particularly in patients with diabetes mellitus (DM) and multivessel disease. Based on the Fractional flow reserve versus Angiography for Multivessel Evaluation (FAME I, II) and DEFER studies, the outcome of residual coronary lesions that are hemodynamically nonsignificant is excellent and not improved by revascularization. However, the central premise of deferred revascularization centers on the assumption that these lesions, which are FFR >0.80, will remain quiescent over the short to medium term. Unfortunately, DM is associated with greater atherosclerotic burden and accelerated and therapy-refractive atherosclerosis, and thus, the longevity of an FFR >0.80 is unknown in a DM-only population. To date, a direct comparison of complete versus deferred revascularization in patients with DM using FFR has not been described. Therefore, to study the impact of deferred versus complete revascularization based on an FFR-guided strategy in all-comer patients with DM, and in particular, to study the longevity of a negative FFR in patients with DM, we retrospectively analyzed the outcomes of patients with DM in our center, where FFR-guided revascularization represents the standard of care.
Methods
From a total of 3,379 patients who underwent FFR-guided revascularization from January 1, 2010, to December 31, 2013, we identified all consecutive patients with DM and followed these patients until July 2015. All patients had a previous diagnosis of DM, defined by patient history and classified by treatment with diet, exercise, oral antidiabetic medication, or insulin.
Baseline demographics were obtained by means of the electronic patient record, in addition to data relating to the FFR measurement and baseline angiography. Follow-up events were obtained from the electronic patient record based on subsequent clinical review, by telephone contact with primary care physicians and referring hospitals or direct contact with patients where required. Follow-up was complete in all FFR-assessed patients.
FFR assessment was performed systematically in all patients with intermediate coronary lesions ranging from 40% to 80% diameter stenosis, where no previous noninvasive test of ischemia was performed or when these were inconclusive. FFR was not performed for culprit lesions in MI, lesions with Thrombolysis In Myocardial Infarction flow <3, or when the operator deemed a lesion to be clearly of hemodynamic significance.
FFR was performed using a standard coronary pressure wire (PressureWire Certus; St. Jude Medical, St Paul, Minnesota; or Combowire; Volcano Corporation, Rancho Cordova, California). Both baseline FFR and maximum hyperemic FFR values were noted for each lesion. After steady-state hyperemia was achieved, the FFR was calculated as the ratio of mean distal intracoronary pressure measured by the pressure wire and the mean arterial pressure measured through the coronary guiding catheter. A cut-off value of ≤0.80 was taken to imply a functionally significant coronary stenosis and the patient underwent revascularization as appropriate. A lesion with an FFR value >0.80 was adjudicated as a functionally nonsignificant leading to deferred revascularization and further medical treatment.
Visual assessment of reference vessel diameter, diameter stenosis, American Heart Association lesion type, and the presence of calcification and diffuse disease were noted for all lesions by 2 independent interventional cardiologists. Both reviewers were blinded to the clinical outcomes. A third interventional cardiologist was used in cases where discordance arose. In addition, the Syntax Score (SS) was calculated retrospectively, based on the index (time of FFR-measurement) coronary angiogram, by scoring all lesions >1.5 mm with at least 50% diameter stenosis using the previously described algorithm. For those patients with previous coronary artery bypass graft (CABG), no SS was calculated.
To assess the safety and efficacy of deferred versus complete revascularization using an FFR-guided strategy in patients with DM and to assess the longevity of a negative FFR in patients with DM, 2 groups were formed, according to the presence or absence of any remaining FFR-negative lesions (>0.80). The first group comprised patients in whom ≥1 FFR-negative lesion (>0.80) remained and were further treated medically [FFR(−)MT], whereas the second group included patients with only FFR-positive lesions (≤0.80), which underwent complete angiographic and functional revascularization [FFR(+)CR]. The local institutional review board approved this study and waived the requirement for written consent to an institutional registry.
The primary end point was the incidence of major adverse cardiovascular events (MACE) defined as a composite of death, myocardial infarction (MI), target lesion revascularization (TLR), or rehospitalization for acute coronary syndrome (ACS). A composite of death or MI, in addition to rehospitalization for ACS, represented the safety end points, whereas the efficacy end point was represented by TLR. Data relating to mortality were obtained from the Dutch national civilians register. Target lesion was defined as the lesion(s) in which the FFR was performed, with TLR referring to revascularization in that lesion(s) whether treated by index revascularization or by medical therapy. MI and periprocedural MI were defined according to the established guidelines. Recent MI was defined as occurring <6 months before the FFR assessment. Rehospitalization for ACS refers to urgent presentation to the Emergency Department for MI or unstable angina pectoris requiring an unscheduled hospitalization. Complete revascularization was defined as the absence of any remaining lesions with an FFR >0.80 and any remaining coronary lesions >50% diameter stenosis in a viable myocardial territory, as determined by the operator.
Continuous variables are summarized as mean ± SD. Discrete variables are summarized as frequency (group percentage). Group comparisons were tested using Student’s t test or Mann–Whitney U test for continuous variables and Pearson’s chi-square test for discrete data. Kaplan-Meier (KM) estimates were used to estimate survival curves and event rates, and the log-rank test was used to establish differences between groups. Cox proportional hazards multiple regression models were used to estimate differences in time to event between the 2 groups expressed as hazard ratios (HRs) with 95% confidence intervals, adjusted for several patient characteristics. In the exploratory model, gender, age, renal insufficiency, previous MI, previous PCI, type of DM, levels of HbA1c, smoking, reference vessel diameter, diameter stenosis, the presence of calcific and diffuse disease, FFR value, multivessel disease, and left ventricular ejection fraction were analyzed. Formal interaction testing was performed to determine whether the presence of identified effect modifiers influenced the relative risk for the occurrence of MACE in both groups. A p value <0.05 was considered significant. All analyses were conducted using SPSS 23 (SPSS Inc., Chicago, IL).
Results
From a total of 3,379 patients who underwent FFR-guided revascularization, we identified 294 consecutive patients with DM who had FFR measurement in 385 intermediate coronary lesions ( Figure 1 ). A total of 205 patients with at least 1 remaining FFR >0.80 lesion formed the FFR(−)MT group, and 89 patients were included in the FFR(+)CR group, having had complete revascularization of all lesions. The mean length of follow-up was 32.6 ± 18.1 months (±SD). Baseline clinical and angiographic characteristics are listed in Tables 1 and 2 , respectively. The average age of patients was 69.2 ± 9.8 years, which was similar in both groups. Overall, baseline characteristics were well matched in both groups; however, there were more male patients (75.3% vs 61%, p = 0.02) and more current smokers (33.7% vs 20%, p = 0.01) in the FFR(+)CR group. Patients in the FFR(−)MT group had a lower mean SS compared with the FFR(+)CR group (10.84 ± 6.96 vs 17.34 ± 12.44, p = 0.001).
| FFR(-)MT (n=205) | FFR(+)CR (n=89) | p-value | |
|---|---|---|---|
| Age (years) | 69.7 ± 9.6 | 68.0 ± 10.2 | 0.23 |
| Men | 125 (61%) | 67 (75%) | 0.02 |
| Diabetes mellitus | 205 (100%) | 89 (100%) | NA |
| Insulin-treated | 87 (42%) | 36 (40%) | 0.75 |
| Left ventricle ejection fraction | 51.8 ± 10.2 | 49.6 ± 11.2 | 0.16 |
| Multi-vessel coronary artery disease | 115 (56%) | 38 (43%) | 0.04 |
| Family history of coronary artery disease | 61 (30%) | 40.4 (36%) | 0.07 |
| Hypertension | 197 (96%) | 85 (96%) | 0.76 |
| Hypercholesterolemia | 198 (97%) | 85 (97%) | >0.99 |
| Current smoker | 41 (20%) | 30 (34%) | 0.01 |
| Renal insufficiency | 30 (15%) | 15 (17%) | 0.63 |
| HbA1c | 53.7 ± 10.5 | 52.8 ± 10.3 | 0.59 |
| Prior myocardial infarction | 93 (45%) | 45 (51%) | 0.41 |
| Remote myocardial infarction | 47 (23%) | 20 (23%) | 0.93 |
| Recent myocardial infarction | 46 (22%) | 25 (28%) | 0.30 |
| Prior percutaneous coronary intervention | 82 (40%) | 46 (52%) | 0.06 |
| Prior coronary artery bypass graft | 29 (14%) | 15 (17%) | 0.55 |
| FFR(-)MT (n=205) | FFR(+)CR (n=89) | p-value | |
|---|---|---|---|
| Clinical syndrome at time of FFR performance: | |||
| Acute coronary syndrome | 73 (36%) | 33 (37%) | 0.81 |
| Non-acute coronary syndrome ∗ | 132 (64%) | 56 (63%) | 0.81 |
| Mean Syntax Score | 10.84 ± 6.96 | 17.34 ± 12.44 | <0.01 |
| Low scores (0-22) | 165 (81%) | 54 (61%) | <0.01 |
| Intermediate scores (23-32) | 7 (3%) | 7 (8%) | 0.13 |
| High scores (≥33) | 4 (2%) | 13 (15%) | <0.01 |
| Unclassified, prior coronary artery bypass graft | 29 (14%) | 15 (17%) | 0.55 |
| FFR performed in one lesion | 143 (70%) | 76 (85%) | 0.01 |
| FFR performed in two lesions | 49 (24%) | 10 (11%) | 0.01 |
| FFR performed in three lesions | 13 (6%) | 3 (3%) | 0.41 |
| Mean FFR result | 0.86 ± 0.06 | 0.73 ± 0.06 | <0.01 |
| Lesion characteristics: lesion level | |||
| Number of lesions assessed | 280 | 105 | |
| AHA/ACC lesion type classification: | |||
| Type A | 33 (12%) | 6 (6%) | 0.04 |
| Type B1 | 150 (54%) | 33 (31%) | <0.01 |
| Type B2 | 79 (28%) | 57 (54%) | <0.01 |
| Type C | 18 (6%) | 9 (9%) | 0.68 |
| Calcified lesion | 57 (20%) | 51 (49%) | <0.01 |
| Diffuse disease | 77 (28%) | 51 (49%) | <0.01 |
| Reference vessel diameter (mm) † | 2.94 ± 0.43 | 2.96 ± 0.42 | 0.58 |
| Diameter stenosis (%) † | 60.46 ±8.48 | 65.96 ± 9.09 | <0.01 |
∗ Includes stable angina and patients undergoing staged FFR of non-culprit lesions following ACS >1 month previously.
In the FFR(−)MT group, 122 patients had all lesions (157 lesions) assessed as FFR negative. A total of 83 patients had index revascularization (27 patients with a lesion assessed by FFR as ≤0.80 and 54 patients with non-FFR-guided revascularization of another lesions) in addition to deferred revascularization based on an FFR >0.80 of at least one other lesion. This treatment included 77 patients who underwent PCI and 6 patients in whom CABG was performed. In the FFR(−)MT group, all lesions <0.80 were revascularized. In the FFR(+)CR group, 88 patients (98.8%) (37 patients by CABG and 51 patients by PCI) underwent complete revascularization at index of 104 of 105 FFR ≤0.80 lesions, with 1 patient (1 lesion) not receiving index revascularization due to technically unsuccessful PCI ( Figure 1 ).
The results of the clinical outcomes are listed in Table 3 and Figure 2 . The primary end point was observed more frequently in the FFR(−)MT group (76 [KM event rate = 44.0%] vs 20 [KM event rate = 26.6%]), unadjusted p = 0.03, and after adjustment by multivariate Cox regression (HR 2.01, 95% confidence interval [CI] 1.21 to 3.33, p <0.01; Table 3 , Figure 2 ). Both safety end points death and/or MI and rehospitalization for ACS were significantly higher in the FFR(−)MT group ( Table 3 , Figure 2 ). Similarly, the efficacy end point, TLR, was also higher in the FFR(−)MT group ( Table 3 , Figure 2 ).
| FFR(-)MT (n=205) | FFR(+)CR (n=89) | FFR(-)MT KM Estimate (n=205) | FFR(+)CR KM Estimate (n=89) | Adjusted HR (95% CI) | Adjusted p-value | |
|---|---|---|---|---|---|---|
| Major adverse cardiac event | 76 (37%) | 20 (23%) | 44.0% | 26.6% | 2.01 (1.21-3.33) | <0.01 |
| Mortality | 36 (18%) | 11 (12%) | 23.8% | 14.8% | 1.78 (0.88-3.60) | 0.11 |
| Myocardial infarction | 13 (6%) | 3 (3%) | 7.2% | 3.5% | 1.81 (0.51-6.38) | 0.36 |
| Death or myocardial infarction | 46 (22%) | 13 (15%) | 28.6% | 16.9% | 2.02 (1.06-3.86) | 0.03 |
| Rehospitalization for acute coronary syndrome | 44 (22%) | 10 (11%) | 24.8% | 12.2% | 2.06 (1.03-4.10) | 0.04 |
| Target lesion revascularization | 29 (14%) | 4 (5%) | 17.6% | 8.2% | 3.38 (1.19-9.64) | 0.02 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
