Abstract
Background
Coronary microvascular functional and structural obstruction (CMVO) remains a major complication in patients with ST-segment elevation myocardial infarction (STEMI). This study was designed to develop and validate a nomogram model to predict CMVO risk during primary percutaneous catheterization procedure.
Material/Methods
Starting January 2014 to December 2016, a cohort of eligible candidates were enrolled and divided into a training or a validation database. Each database was divided into MO or NMO subgroups based on TIMI myocardial perfusion grade results after recanalization. Independent factors were identified by multivariate logistic regression, from which the nomogram was plotted. The echocardiography measurement of the left ventricular ejection fraction (LVEF) was arranged within 7 days after the procedure.
Results
A nomogram was built for CMVO risk prediction for the first time. There were 446 participants in the training database with 319 cases in the NMO subgroup and 127 participants in the MO subgroup. The validation database included 99 participants with 25 cases in the NMO subgroup and 74 in the MO subgroup. The risk model was developed by 6 independently significant factors: age, symptom onset to balloon time, Killip classification, admission activated clotting time, neutrophil/lymphocyte ratio, and glucose value. Internal receiver operating characteristic displayed favorable performance with concordance index of 0.925, while external validation area under curve was 0.939. There were significant differences in LVEF values during hospitalization between the subgroups of each database (both P<0.001).
Conclusions
The nomogram model consisting of 6 factors could predict CMVO risk accurately for STEMI patients undergoing primary percutaneous catheterization.
MeSH Keywords: Microvessels; Models, Cardiovascular; Myocardial Infarction; Nomograms; Percutaneous Coronary Intervention
Background
ST-segment elevation myocardial infarction (STEMI), a typical symptom mostly generated from acute intracoronary thrombus formation and coronary blood flow cessation, remains a dominating threat to public health throughout the world. In patients presenting with acute STEMI, the main target of the reperfusion strategy is to promptly implement the intervention on the culprit artery. Regardless of epicardial repatency after revascularization, a considerable number of patients suffer from inadequate perfusion at the myocardial tissue level, resulting from coronary microvascular functional and structural obstruction (CMVO) during the perioperative period [1]. CMVO, previously identified and understood as “no-reflow phenomenon” (NRF), would add to the intervention strategy the incidence of adverse complications and events, thus affecting patients’ short and long-period prognosis [2,3]. Therefore, attempts to decrease microvascular obstruction risk via mechanical and pharmaceutical pretreatments in the early stage of catheterization have been proposed and achieved promising results [4,5]. Nevertheless, for patients developing subsequent microvascular impairment, no specific or definitive therapeutic approaches for attenuation have proven to be valid from large scale tests, highlighting the urgent need to address timely recognition of this condition.
In recent years, many trials have been carried out to analyze and investigate the possible related indicators of CMVO or NRF [6]. However, due to the differences in study protocols, sample size, and auxiliary measurements, there have been no consistent conclusions so far. In addition, limited by the testing costs and time, some clinical or laboratory indexes are still not suitable or available in present practice. Among the current existing findings, time from chest pain symptom onset to balloon (SO-B) has been confirmed as one of the most acknowledged predictors of microvascular perfusion, which intensifies the significance of early reperfusion in the setting of STEMI [7,8]. Nevertheless, on the basis of animal experiments and earlier clinical trials, an assumption has been put forward that multiple process, including ischemia-reperfusion injury, distal embolization, and individual susceptibility are involved in CMVO development, and in an integrated manner contribute to deteriorating microvascular perfusion [9,10]. Accordingly, one single indicator may not be valid enough to evaluate the perfusion state of microvasculature. In order to determine CMVO probability in primary percutaneous coronary catheterization and thus rendering early and rapid identification of high-risk patients, we systematically screened the possible clinical and angiographic information retrospectively, with an attempt to construct and validate a predictive nomogram model for the first time.
Material and Methods
Population
Starting January 2014 to December 2016, the participants hospitalized to the Cardiology Division of the Second Hospital of Hebei Medical University were recruited in this retrospective analysis. The qualified candidates were enrolled based on the following criteria: 1) diagnosis of STEMI was based on the current guideline [11]; 2) the duration lasted less than 24 hours from the symptom onset till balloon inflation; 3) patients and their direct relatives agreed to undergo the emergent intervention in the Chest Pain Unit of our Institute. Meanwhile, our exclusion criteria included: 1) Killip IV grade; 2) intravenous or intracoronary fibrinolytic pharmaceutic treatment; 3) history of myocardial infarction; 4) rejecting the angiographic examination or preparing an elective intervention; 5) multiple lesions not advisable for stenting; 6) immunological or rheumatic diseases, end-stage hepato-renal failure, and cancer; 7) serious mechanical or operative complications; and 8) contraindication for the anticoagulant or antiplatelet agents. According to the Helsinki Declaration, the design and protocol has been approved by the Institutional Ethics Committee (Approval Letter No. 2018-P044). Each recruited member has agreed to take part in this study and signed the informed consent.
Study design and procedures
Patients’ demographic data (gender, age, body mass index, Killip grade), previous history (hypertension, diabetes mellitus, dyslipidemia and smoking) and present illness description were recorded in medical files while hospitalized. Electrocardiogram and conventional laboratory examinations [complete blood cell, myocardial damage indicators, biochemical indices, glycosylated hemoglobin A1c (HbA1c), serum brain natriuretic peptide (BNP), and activated clotting time (ACT)] for STEMI patients were performed on admission. ACT assay test was detected by the double-channel mechanical plunger (ACT plus, Medtronic Inc., Minneapolis, MN, USA) on reaction temperature of 37°C.
Patients received 300 mg aspirin, 600 mg clopidogrel or 180 mg ticagrelor, and unfractionated heparin (UFH) 70–100 U/kg as pretreatment prior to the procedure. The coronary angiographic examination was conducted in keeping with the technical principle via radial, ulnar, or femoral access. Angiographic review and analysis were accomplished by no less than 2 skillful interventional cardiologists in different working teams. ACT levels were kept 250–300 seconds under the application of UFH conventionally, while Bivalirudin (initial bolus of 0.75 mg/kg, then intravenous infusion with 1.75 mg/kg/hour for 3–4 hours after procedure) served as an alternative for patients at high hemorrhage risk. Percutaneous coronary intervention (PCI) would be preferred when the stenosis degree of infarct related artery (IRA) was beyond 75%. When the lesions were not suitable or the patients refused stenting, transluminal coronary angioplasty (PTCA) would be adopted. Reperfusion therapy delay [for instance, SO-B and first medical contact to balloon (FMC-B) time] would be calculated when the balloon was inflated. The choice of interventional device, procedure (numbers and pressure of predilation and postdilation, thrombus aspiration, and temporary pacemaker implantation) and adjuvant medication (tirofiban, anisodamine, etc.) were left to the operators’ decisions. Some invasive indices were also assessed and listed as previously introduced, such as thrombolysis in myocardial infarction (TIMI) grade [12], Gibson’s intracoronary thrombus scores [13], TIMI myocardial perfusion grade (TMPG) [14], and corrected TIMI frame count (cTFC). Culprit artery cTFC was counted at the rate of 15 frames per second in accordance with Gibson’s method [15]. After the procedure, all the participants were administered with anticoagulant and antithrombotic therapy, b-receptor blocker, statins, nitrates, and angiotensin-converting enzyme inhibitors or angiotensin receptor blocker as stated in the guideline. The echocardiography measurement of left ventricular ejection fraction (LVEF) was arranged in 7 days after the emergent catheterization.
Among all eligible participants, ID numbers were given according to the admission time. The candidates administered from January 2014 to June 2016 were scheduled to create a training database, while the patients enrolled from July 2016 to December 2016 were planned to validate the established model. Each database was divided into subgroups based on TMPG results after the recanalization, namely MO subgroup with 0–2 grade and NMO subgroup with 3 grade.
Statistical analysis
Kolmogorov-Smirnov’s test was utilized to evaluate those continuous data’s normality. If the data were distributed normally, they would be indicated as mean ± standard deviation (SD), or else they would be reported as median (25th quartile, 75th quartile). For categorical data, they would be presented as numbers and proportions. The statistical difference between continuous data were compared by Student’s t-test for normally distributed values and Mann-Whitney U test for those non-normally distributed. Proportions were assessed by chi-square test or Fisher’s exact test if the expected frequency was <5. Binary multivariate logistic regression was used to recognize the independent elements that influence myocardial perfusion by a backward step-down process in the training database. The multicollinearity among all potential factors was estimated. Then, those risk factors were put into use for development of nomogram. Statistical calculations and analysis were conducted by SPSS Software (Version 23.0, SPSS Inc., Chicago, IL, USA). A 2-tailed P value less than 0.05 was considered statistical significance.
Generated from multivariate logistic regression analysis of the training database, a nomogram was established with R Software (Version 3.4.4 for Windows, http://www.r-project.org/) and RStudio (https://www.rstudio.com/). The construction of nomogram was performed by running the programming codes and a series of R packages (Hmisc, grid, lattice, Formula, ggplot2, rms, haven, pROC, etc.), which were downloaded and installed from R Software online platform. The nomogram’s discriminant performance was assessed and analyzed by concordance index (C-index), and illustrated as receiver operating characteristic (ROC) and area under curve (AUC) [16]. Calibration curve was derived to test calibration ability. Five-fold cross-validation method was used for further internal validation afterwards, while external validation was carried out by substituting the validation data in the developed model.
Results
Group enrollment
Since January 2014 to June 2016, a total of 446 eligible STEMI participants formed the training database. Among those cases, 127 patients developed CMVO with an incidence of 28.48%. Beginning July 2016 to December 2016, a total of 99 eligible patients were enrolled to construct the validation database. Amongst these patients, 25 cases were diagnosed with CMVO with an incidence of 25.25%.
Baseline clinical characteristics between both subgroups in the training database
The demographic characteristic, baseline information, and clinical laboratory indices of the candidates in the NMO and the MO subgroups of the training database, and the relevant details of comparisons are described in Table 1. No evident difference was noticed between the 2 subgroups on gender, body mass index, previous history, life signs, red blood cell count, platelet count, electrolyte, lipid values, and HbA1c (all P>0.05). The average age of the MO subgroup participants was higher than that of the NMO subgroup (62.06±13.95 versus 57.68±12.29, P=0.019). Patients in the MO subgroup shared a higher proportion of Killip 3 grade and a lower proportion of Killip 1 grade upon admission (44.88% versus 10.34%, P<0.001; 10.34% versus 58.31%, P<0.001). Besides, GRACE, as well as CRUSADE scores of the MO subgroup were significantly higher [135.14±26.42 versus 151.28±32.41, P<0.001; 20 (13, 30) versus 27 (15, 44), P<0.001]. The following laboratory items were also distinctly different between the 2 subgroups, including white blood count, neutrophil count, lymphocyte count, neutrophil/lymphocyte ratio (NLR), high sensitive C-reactive protein (hs-CRP), cardiac troponin I (cTnI), creatine kinase-MB isozyme (CK-MB), ACT, serum creatinine, evaluated glomerular filtration rate (eGFR), glucose, D-dimer, and BNP (all P<0.05).
Table 1.
Baseline clinical characteristics between the 2 subgroups in the training database.
| Variables | NMO group (n=319) | MO group (n=127) | P value |
|---|---|---|---|
| Age (years) | 57.68±12.29 | 62.06±13.95 | 0.019 |
| Male, n (%) | 271 (84.95) | 100 (78.74) | 0.124 |
| BMI (kg/m2) | 25.59±3.15 | 26.06±3.68 | 0.274 |
| SBP (mmHg) | 133.15±26.78 | 128.93±23.95 | 0.123 |
| DBP (mmHg) | 77.29±16.85 | 78.47±15.30 | 0.494 |
| Heart rate (bpm) | 76.31±16.44 | 79.05±18.71 | 0.128 |
| Killip grade | |||
| Grade I, n (%) | 186 (58.31) | 21 (16.54) | <0.001 |
| Grade II, n (%) | 100 (31.35) | 49 (38.58) | 0.144 |
| Grade III, n (%) | 33 (10.34) | 57 (44.88) | <0.001 |
| History of CAD, n (%) | 177 (55.49) | 64 (50.39) | 0.347 |
| Hypertension, n (%) | 204 (63.95) | 74 (58.27) | 0.264 |
| Diabetes mellitus, n (%) | 85 (26.65) | 44 (35.43) | 0.065 |
| Hyperlipidemia, n (%) | 137 (42.95) | 50 (39.37) | 0.490 |
| Smoking, n (%) | 122 (38.24) | 60 (47.24) | 0.081 |
| Laboratory test on admission | |||
| WBC count (109/L) | 8.50 (10.20, 12.40) | 10.12 (12.00, 14.55) | <0.001 |
| Neutrophil count (109/L) | 8.00 (6.50, 9.80) | 10.30 (8.55, 12.50) | <0.001 |
| Lymphocyte count (109/L) | 1.5 (1.20, 2.00) | 1.06 (0.80, 1.35) | <0.001 |
| N/L ratio | 5.30 (3.94, 7.00) | 9.73 (7.81, 12.69) | <0.001 |
| Erythrocyte count (1012/L) | 4.45±0.87 | 4.33±0.92 | 0.197 |
| Platelet count (1012/L) | 205.37±63.55 | 209.79±55.03 | 0.549 |
| High sensitive CRP (mg/L) | 3.20 (1.60, 5.70) | 4.40 (2.05, 8.40) | 0.016 |
| ACT (s) | 160 (141, 182) | 124 (98, 138) | <0.001 |
| CK-MB (U/L) | 107 (33, 280) | 154 (61, 308) | 0.075 |
| Cardiac troponin I (ng/mL) | 4.5 (2.7, 10.4) | 11.4 (5.5, 29.5) | <0.001 |
| Serum creatinine (μmol/L) | 72 (61.2, 82.7) | 75 (68.9, 96.0) | 0.046 |
| eGFR (mL/min/1.73 m2) | 99.6 (84.40, 120.00) | 91.76 (63.24, 111.86) | 0.011 |
| Serum potassium (mmol/L) | 4.02 (3.76, 4.31) | 3.94 (3.69, 4.40) | 0.845 |
| Triglyceride (mmol/L) | 1.29 (0.87, 1.85) | 1.26 (0.81, 1.64) | 0.547 |
| Total cholesterol (mmol/L) | 4.22 (3.59, 4.91) | 4.05 (3.58, 5.46) | 0.908 |
| LDL cholesterol (mmol/L) | 2.81 (2.20, 3.39) | 2.70 (2.23, 3.40) | 0.905 |
| HDL cholesterol (mmol/L) | 1.02 (0.88, 1.26) | 1.10 (0.95, 1.30) | 0.061 |
| HbA1c (%) | 6.30±1.23 | 6.53±1.32 | 0.085 |
| Glucose (mmol/L) | 9.61 (7.30, 10.85) | 12.74 (10.23, 14.42) | <0.001 |
| D-dimer (μg/mL) | 0.11 (0.08, 0.20) | 0.19 (0.12, 0.37) | <0.001 |
| BNP (pg/mL) | 58.5 (30,108) | 95 (35,177.5) | 0.025 |
| Preprocedural medication | |||
| DAPT, n (%) | 304 (95.30) | 118 (92.91) | 0.314 |
| Statins, n (%) | 131 (41.07) | 43 (33.86) | 0.159 |
| Beta-blocker, n (%) | 24 (7.52) | 8 (6.30) | 0.665 |
| GRACE score | 135.14±26.42 | 151.28±32.41 | <0.001 |
| CRUSADE score | 20 (13, 30) | 27 (15, 44) | <0.001 |
Normally distributed continuous data are indicated as mean ± standard deviation; or else they would be reported as median (25th quartile, 75th quartile). Categorical data are shown as numbers and proportions (%). MO and NMO subgroups based on TIMI myocardial perfusion grade results after recanalization. MO subgroup with 0–2 grade and NMO subgroup with 3 grade. ACT – activated clotting time; BMI – body mass index; BNP – brain natriuretic peptide; CAD – coronary artery disease; CK-MB – creatine kinase-MB isozyme; CRP – C-reactive protein; DAPT – dual antiplatelet therapy; DBP – diastolic blood pressure; eGFR – evaluated glomerular filtration rate; HbA1c – Glycosylated hemoglobin A1c; HDL – high density lipoprotein; LDL – low density lipoprotein; SBP – systolic pressure; WBC – white blood cell.
Procedural and angiographic features between both subgroups in the training database
The angiographic features in the training database are listed in Table 2. SO-B and FMC-B time of the MO subgroup were apparently delayed compared with that in the NMO subgroup [7.0 (5.0, 9.1) versus 4.5 (3.5, 6.0), P<0.001; 2.5 (1.5, 4.8) versus 2.0 (1.2, 3.0), P=0.007]. Likewise, initial TIMI blood flow and thrombus score of IRA were different between the 2 subgroups. After the intervention, obvious differences were displayed regarding blood flow perfusion indicators, including TIMI 3 grade proportion (50.39% versus 100%, P<0.001) and cTFC [22 (18, 29) versus 36 (37,53), P<0.001]. Other angiographic and procedural details, like IRA distribution, intervention type, stent information, medication, and supplementary treatment were comparable for the subgroups (all P>0.05).
Table 2.
Procedural and angiographic features between the 2 subgroups in the training database.
| Variables | NMO group (n=319) | MO group (n=127) | P value |
|---|---|---|---|
| Onset to balloon (hours) | 4.5 (3.5, 6.0) | 7.0 (5.0, 9.1) | <0.001 |
| FMC to balloon (hours) | 2.0 (1.2, 3.0) | 2.5 (1.5, 4.8) | 0.007 |
| Myocardial wall, n (%) | |||
| Anterior wall | 150 (47.02) | 68 (53.54) | 0.214 |
| Others | 169 (52.98) | 59 (46.46) | 0.214 |
| Stenosed artery number, n (%) | |||
| 1 | 70 (21.94) | 21 (16.54) | 0.201 |
| 2 | 107 (33.54) | 46 (36.22) | 0.591 |
| 3 | 142 (44.52) | 60 (47.24) | 0.601 |
| Intervention pattern, n (%) | |||
| PCI | 279 (87.46) | 110 (86.61) | 0.809 |
| PTCA | 40 (12.54) | 17 (13.39) | 0.809 |
| Initial TIMI flow, n (%) | |||
| 0 | 172 (53.92) | 114 (89.76) | <0.001 |
| 1 | 45 (14.10) | 11 (8.66) | 0.117 |
| 2 | 51 (15.99) | 2 (1.58) | <0.001 |
| 3 | 51 (15.99) | 0 (0.00) | <0.001 |
| Initial thrombus score, n (%) | |||
| 0–1 | 10 (3.13) | 0 (0.00) | 0.044 |
| 2 | 34 (10.66) | 1 (0.79) | <0.001 |
| 3 | 70 (21.94) | 3 (2.36) | <0.001 |
| 4 | 33 (10.34) | 9 (7.09) | 0.288 |
| 5 | 172 (53.93) | 114 (89.76) | <0.001 |
| Final TIMI flow, n (%) | |||
| 0 | 0 (0.00) | 0 (0.00) | — |
| 1 | 0 (0.00) | 2 (1.58) | 0.025 |
| 2 | 0 (0.00) | 61 (48.03) | <0.001 |
| 3 | 319 (100) | 64 (50.39) | <0.001 |
| IRA-cTFC | 22 (18, 29) | 46 (34, 57) | <0.001 |
| Stent number per patient, n (%) | |||
| 1 | 242 (86.74) | 91 (82.73) | 0.310 |
| ≥2 | 37 (13.26) | 19 (17.27) | 0.310 |
| Stent length (mm) | 23 (18, 29) | 24 (18, 29) | 0.143 |
| Stent diameter (mm) | 3.0 (2.60, 3.50) | 2.75 (2.50, 3.00) | 0.204 |
| Predilation pressure (atm) | 14 (12, 16) | 14 (12, 16) | 0.193 |
| Predilation numbers | 2 (1, 3) | 2 (1, 3) | 0.348 |
| Stent expansion pressure (atm) | 14 (15, 16) | 14 (14, 16) | 0.509 |
| Postdilation pressure (atm) | 14 (12, 16) | 14 (11, 16) | 0.467 |
| Postdilation numbers | 1 (0, 2) | 1 (0, 2) | 0.960 |
| Thrombus aspiration, n (%) | 38 (11.91) | 9 (7.09) | 0.134 |
| Temporary pacemaker, n (%) | 7 (2.19) | 5 (3.94) | 0.305 |
| Collateral circulation, n (%) | 72 (22.57) | 35 (27.56) | 0.266 |
| Contrast media volume (mL) | 160 (130, 180) | 160 (145, 195) | 0.181 |
| Procedural medication, n (%) | |||
| Tirofiban | 275 (86.21) | 101 (79.53) | 0.085 |
| Bivalirudin | 60 (18.81) | 32 (25.20) | 0.154 |
| Anisodamine | 70 (21.94) | 28 (22.05) | 0.981 |
Normally distributed continuous data are indicated as mean ± standard deviation; or else they would be reported as median (25th quartile, 75th quartile). Categorical data are shown as numbers and proportions (%). cTFC – corrected TIMI frame count; FMC – first medical contact; IRA – infarct related artery; PCI – percutaneous coronary intervention; PTCA – percutaneous transluminal coronary angioplasty; TIMI – thrombolysis in myocardial infarction.
Baseline clinical and procedural characteristics between both subgroups in the validation database
The statistical analysis of the validation database shared similar conclusions with the training database. The correlative results of baseline information and interventional data are listed in Tables 3 and 4. Compared to their counterparts, members in the MO subgroup were of elderly age (68.36±11.41 years versus 60.20±1.57 years, P=0.003), poorer heart function with higher Killip 3 classification proportion (44.00% versus 20.00%, P<0.001), elevated risk score and corresponding test indexes (all P<0.05). In respect to the angiographic and procedural features, the perfusion intervals (SO-B and FMC-B time) were significantly shorter in the NMO subgroup [6.5 (4.0, 9.3) versus 3.5 (2.5, 5.0), P=0.001; 2.0 (1.0, 3.5) versus 1.5 (1.0, 2.5), P=0. 108]. Moreover, post-procedural TIMI 3 grade percentage and cTFC in the NMO subgroup were superior to those of the MO subgroup [40.00% versus 100%, P<0.001; 26 (22, 28) versus 40 (33, 60), P<0.001].
Table 3.
Baseline clinical characteristics between the 2 subgroups in the validation database.
| Variables | NMO group (n=74) | MO group (n=25) | P value |
|---|---|---|---|
| Age (years) | 60.20±1.57 | 68.36±11.41 | 0.003 |
| Male, n (%) | 62 (83.78) | 18 (74.00) | 0.242 |
| BMI (kg/m2) | 24.84±2.41 | 25.74±3.06 | 0.136 |
| SBP (mmHg) | 138.58±24.61 | 133.0±29.96 | 0.374 |
| DBP (mmHg) | 81.27±15.91 | 78.20±17.93 | 0.421 |
| Heart rate (bpm) | 74.32±16.71 | 79.52±20.57 | 0.209 |
| Killip grade | |||
| Grade I, n (%) | 46 (62.16) | 5 (20.00) | <0.001 |
| Grade II, n (%) | 21 (28.38) | 9 (36.00) | 0.473 |
| Grade III, n (%)6 | 7 (9.46) | 11 (44.00) | <0.001 |
| History of CAD, n (%) | 38 (51.35) | 10 (40.00) | 0.326 |
| Hypertension, n (%) | 40 (54.05) | 12 (48.00) | 0.600 |
| Diabetes mellitus, n (%) | 17 (22.97) | 10 (40.00) | 0.098 |
| Hyperlipidemia, n (%) | 30 (40.54) | 11 (44.00) | 0.655 |
| Smoking, n (%) | 31 (41.89) | 14 (56.00) | 0.221 |
| Laboratory test on admission | |||
| WBC count (109/L) | 10.19 (8.71, 12.46) | 11.90 (9.80, 14.50) | 0.034 |
| Neutrophil count (109/L) | 8.17 (6.34, 9.75) | 10.31 (8.10, 12.60) | 0.003 |
| Lymphocyte count (109/L) | 1.52 (1.29, 1.85) | 1.00 (0.90, 1.30) | <0.001 |
| N/L ratio | 5.25 (4.11, 6.87) | 9.69 (7.97, 11.70) | <0.001 |
| Erythrocyte count (1012/L) | 4.33±0.84 | 4.36±0.51 | 0.915 |
| Platelet count (1012/L) | 211 (179, 254) | 213 (184, 244) | 0.894 |
| High sensitive CRP (mg/L) | 3.50 (1.50,10.85) | 5.5 (3.50, 11.00) | 0.087 |
| ACT (s) | 166 (140, 182) | 108 (95, 122) | <0.001 |
| CK-MB (U/L) | 82 (37, 174) | 177 (43, 383) | 0.004 |
| Cardiac troponin I (ng/mL) | 5.35 (2.1, 16.0) | 11.4 (6.5, 30.5) | <0.001 |
| Serum creatinine (μmol/L) | 85.84±19.92 | 77.33±15.35 | 0.054 |
| eGFR (mL/min/1.73 m2) | 95.56 (86.42, 117.95) | 80.00 (69.04, 97.85) | 0.011 |
| Serum potassium (mmol/L) | 3.90 (3.60, 4.10) | 3.85 (3.50, 4.25) | 0.762 |
| Triglyceride (mmol/L) | 1.34 (0.89, 2.14) | 1.24 (0.85, 1.85) | 0.784 |
| Total cholesterol (mmol/L) | 4.46±1.04 | 4.30±0.73 | 0.498 |
| LDL cholesterol (mmol/L) | 2.91 (2.59, 3.35) | 2.78 (2.38, 3.21) | 0.192 |
| HDL cholesterol (mmol/L) | 1.03±0.19 | 1.07±0.25 | 0.437 |
| HbA1c (%) | 6.26±1.30 | 6.49±1.07 | 0.458 |
| Glucose (mmol/L) | 9.16 (7.52, 11.14) | 12.39 (10.56, 16.33) | 0.005 |
| D-dimer (μg/mL) | 0.10 (0.07, 0.20) | 0.24 (0.11, 0.57) | <0.001 |
| BNP (pg/mL) | 70 (30,158) | 100 (60,277.5) | 0.004 |
| Preprocedural medication | |||
| DAPT, n (%) | 69 (93.24) | 23 (92.00) | 1.000 |
| Statins, n (%) | 33 (44.59) | 9 (36.00) | 0.545 |
| Beta-blocker, n (%) | 4 (5.41) | 2 (8.00) | 0.641 |
| GRACE score | 136 (120, 159) | 156 (144, 184) | 0.001 |
| CRUSADE score | 20 (11, 27) | 32 (20, 43) | <0.001 |
Normally distributed continuous data are indicated as mean ± standard deviation; or else they would be reported as median (25th quartile, 75th quartile). Categorical data are shown as numbers and proportions (%). MO and NMO subgroups based on TIMI myocardial perfusion grade results after recanalization. MO subgroup with 0–2 grade and NMO subgroup with 3 grade. ACT – activated clotting time; BMI – body mass index; BNP – brain natriuretic peptide; CAD – coronary artery disease; CK-MB – creatine kinase-MB isozyme; CRP – C-reactive protein; DAPT – dual antiplatelet therapy; DBP – diastolic blood pressure; eGFR – evaluated glomerular filtration rate; HbA1c – Glycosylated hemoglobin A1c; HDL – high density lipoprotein; LDL – low density lipoprotein; SBP – systolic pressure; WBC – white blood cell.
Table 4.
Procedural and angiographic features between the 2 subgroups in the validation database.
| Variables | NMO group (n=74) | MO group (n=25) | P value |
|---|---|---|---|
| Onset to balloon (hours) | 3.5 (2.5, 5.0) | 6.5 (4.0, 9.3) | 0.001 |
| FMC to balloon (hours) | 1.5 (1.0, 2.5) | 2.0 (1.0, 3.5) | 0.108 |
| Myocardial wall, n (%) | |||
| Anterior wall | 28 (37.83) | 12 (48.00) | 0.371 |
| Others | 46 (62.17) | 13 (52.00) | 0.371 |
| Stenosed artery number, n (%) | |||
| 1 | 23 (31.08) | 6 (24.00) | 0.501 |
| 2 | 25 (33.78) | 8 (32.00) | 0.870 |
| 3 | 26 (35.14) | 11 (44.00) | 0.428 |
| Intervention pattern, n (%) | |||
| PCI | 66 (89.18) | 22 (88.00) | 1.000 |
| PTCA | 8 (10.81) | 3 (12.00) | 1.000 |
| Initial TIMI flow, n (%) | |||
| 0 | 42 (56.76) | 22 (88.00) | 0.005 |
| 1 | 21 (28.38) | 3 (12.0) | 0.099 |
| 2 | 6 (8.11) | 0 (0.00) | 0.333 |
| 3 | 5 (6.75) | 0 (0.00) | 0.326 |
| Initial thrombus score, n (%) | |||
| 0–1 | 4 (5.41) | 0 (0.00) | 0.569 |
| 2 | 9 (12.16) | 0 (0.00) | 0.107 |
| 3 | 12 (16.22) | 2 (8.00) | 0.508 |
| 4 | 14 (18.92) | 3 (12.00) | 0.549 |
| 5 | 35 (47.29) | 20 (80.00) | 0.004 |
| Final TIMI flow, n (%) | |||
| 0 | 0 (0.00) | 3 (12.00) | 0.015 |
| 1 | 0 (0.00) | 2 (8.00) | 0.062 |
| 2 | 0 (0.00) | 11 (44.00) | <0.001 |
| 3 | 74 (100.00) | 9 (36.00) | <0.001 |
| IRA-cTFC | 26 (22, 28) | 40 (33, 60) | <0.001 |
| Stent number per patient, n (%) | |||
| 1 | 59 (88.06) | 17 (77.27) | 0.165 |
| ≥2 | 8 (11.94) | 5 (22.73) | 0.165 |
| Stent length (mm) | 24 (18, 33) | 24 (16, 31) | 0.937 |
| Stent diameter (mm) | 3.0 (2.75, 3.00) | 2.75 (2.50, 3.00) | 0.911 |
| Predilation pressure (atm) | 14 (12, 16) | 14 (12, 16) | 0.378 |
| Predilation numbers | 2 (1, 3) | 2 (1, 3) | 0.272 |
| Stent expansion pressure (atm) | 14 (14, 16) | 14 (12, 16) | 0.379 |
| Postdilation pressure (atm) | 16 (14, 18) | 14 (12, 16) | 0.502 |
| Postdilation numbers | 0 (0, 2) | 1 (0, 2) | 0.134 |
| Thrombus aspiration, n (%) | 25 (33.78) | 7 (28.00) | 0.593 |
| Temporary pacemaker, n (%) | 0 (0.00) | 1 (4.00) | 0.253 |
| Collateral circulation, n (%) | 13 (18.92) | 5 (20.00) | 0.771 |
| Contrast media volume (mL) | 150 (130, 180) | 160 (150, 190) | 0.233 |
| Procedural medication, n (%) | |||
| Tirofiban | 63 (85.14) | 20 (80.00) | 0.542 |
| Bivalirudin | 15 (20.27) | 9 (36.00) | 0.113 |
| Anisodamine | 14 (18.92) | 6 (24.00) | 0.584 |
Normally distributed continuous data are indicated as mean ± standard deviation; or else they would be reported as median (25th quartile, 75th quartile). Categorical data are shown as numbers and proportions (%). MO and NMO subgroups based on TIMI myocardial perfusion grade results after recanalization. MO subgroup with 0–2 grade and NMO subgroup with 3 grade. cTFC – corrected TIMI frame count; FMC – first medical contact; IRA – infarct related artery; PCI – percutaneous coronary intervention; PTCA – percutaneous transluminal coronary angioplasty; TIMI – thrombolysis in myocardial infarction.
Cardiac function
All patients accepted cardiac ultrasound evaluation after the procedure in hospital. In the training database, LVEF of the NMO subgroup was higher than that of the MO subgroup (57.94±4.46 versus 48.33±6.67, P<0.001). Similarly, the LVEF value of the NMO subgroup in the validation database was also elevated (57.20±4.70 versus 48.04±9.37, P<0.001).
Logistic regression analysis of CMVO
Binary logistic regression analysis was conducted to recognize the CMVO independent risk factors in STEMI patients undergoing a primary catheterization. In accordance with the regression model, the following indicators were highly associated with the occurrence of CMVO: age >65 years (X1), admission Killip classification (X2), SO-B time (X3), baseline ACT level (X4), baseline NLR >7.0 (X5) and glucose value >12.0 mmol/L (X6). Regression formula was established as follows: Logit P=−7.580+1.055×X1+1.461×X2+0.278×X3−0.734×X4+1.423×X5+0.893×X6. The predictive rate of the model was calculated to be 86.3%. The result of logistic regression analysis is described in Table 5. Nomogram was plotted on the regression results with R software and relevant packages (Figure 1).
Table 5.
Predictors of CMVO by binary logistic regression analysis in the training database.
| Variables | CMVO | |||||
|---|---|---|---|---|---|---|
| Coefficient | SE | Wald | OR | 95% CI | P value | |
| Age (years) >65 | 1.055 | 0.301 | 12.286 | 2.871 | 1.592–5.179 | 0.000 |
| Symptom onset to balloon (hours) | 0.278 | 0.048 | 33.609 | 1.320 | 1.202–1.450 | 0.000 |
| Killip grade | 1.461 | 0.213 | 47.228 | 4.312 | 2.842–6.541 | 0.000 |
| Admission ACT (seconds) | −0.037 | 0.007 | 29.712 | 0.964 | 0.951–0.977 | 0.000 |
| NLR >7.0 | 1.423 | 0.315 | 20.441 | 4.148 | 2.239–7.687 | 0.000 |
| Admission glucose (mmol/L) >12.0 | 0.893 | 0.304 | 8.643 | 2.441 | 1.346–4.426 | 0.003 |
| Constant Term | −6.113 | 1.215 | 25.307 | 0.002 | 0.000 | |
ACT – activated clotting time; CI – confidence interval; CMVO – coronary microvascular functional and structural obstruction; NLR – neutrophil/lymphocyte ratio; OR – odds ratio; SE – standard error.
Figure 1.
Nomogram of microvascular obstruction risk in patients with ST-segment elevation myocardial infarction (STEMI) undergoing primary percutaneous catheterization. The present nomogram indicates the possible risk prediction of coronary microvascular functional and structural obstruction (CMVO) in patients with STEMI undergoing primary percutaneous catheterization. For clinical use, draw a straight line from the corresponding value of each variable to the top point axis repeatedly. Add the points together, obtaining the total points and the risk probability according to the bottom axis.
Validation of nomogram model
The C-index of nomogram model built by the training database was 0.925. AUC value, identical to C-index level, was 0.925 [95% confidence interval (CI): 0.900–0.949] as well. After 5-fold cross internal validation, the corresponding AUC values were 0.945, 0.943, 0.898, 0.914, and 0.901 respectively, with the mean AUC value of 0.902. The calibration curve is shown in Figure 2. In the training database, the mean absolute error (MAE) between the actual and evaluated possibilities was 0.026. External validation was carried out with validation database information. The correlated AUC was 0.939 (95% CI: 0.894–0.984), sharing the sensitivity of 0.811 and specificity of 0.960.
Figure 2.
Calibration curve for the coronary microvascular functional and structural obstruction (CMVO) risk model in the training database.
Discussion
STEMI has become a major public health issue and is among the primary causes of cerebro-cardiovascular deaths worldwide. As for STEMI patients, rapid reperfusion is of great significance in restoring epicardial coronary flow, salvaging jeopardized myocardium, limiting infarction extension and ventricular remodeling. Although considerable advancements have been achieved in therapeutic strategies, there remains room for further exploration and enhancement. Despite IRA repatency, a portion of patients fail to benefit due to insufficient perfusion of the microvasculature, which is termed as CMVO or NRF. As from accumulated literature, impaired myocardial perfusion is closely associated with poorer cardiac function, increasing mortality and worsening prognosis [17,18]. Similarly, in our study, distinct differences in LVEFs during hospitalization were observed between the study subgroups, emphasizing the importance of timely evaluation and recognition of patients at high risk in the early setting.
Note that multiple mechanisms contribute to microvascular obstruction development together, a single element may not be convincing enough in risk prediction and stratification. Therefore, risk scores or models, consisting of various indicators are favored for accurate detection and diagnosis. In this study, potential factors chosen from earlier hypothesis and practical experience have been tested for their possible correlations with CMVO from the institute database system. The logistic regression results showed that age, Killip classification, symptom onset to balloon time, initial ACT level, NLR, and admission glucose value were independently responsible for microcirculatory injury in STEMI patients undergoing emergent catheterization.
Among all these factors, CMVO development is most likely driven by prolonged reperfusion time, known as symptom onset to balloon time. Consequently, prompt diagnosis and revascularization are of remarkable importance in preventing vasculature damage, improving prognosis and decreasing the incidence of cardiac events.
With the development of angiographic techniques and device, advanced age is no longer an absolute contraindication for coronary intervention. It is not rare for elderly STEMI patients receiving primary catheterization, particularly when they share the contradiction of thrombolysis, renal failure, or high bleeding risk. However, earlier evidence has proven the hypothesis with the similar conclusion of this study that, CMVO incidence would increase if the patient’s age was over 65 years [19,20], possibly owing to preexisting pathophysiological basis, such as severe and diffuse coronary lesions, serious calcification, and endothelial dysfunction [21].
As for myocardial infarction patients, Killip classification acts as a simple and fundamental assessment for myocardial ischemic injury and hemodynamics. Higher Killip grade is recognized as poorer functional capacity and extensively damaged myocardium as well as microvasculature. As De Luca et al. reported, Killip 3 grade is an independent indicator of myocardial perfusion impairment after emergent PCI [22]. Jeong et al. has also suggested that baseline levels of serum BNP, which is commonly used parameter for left ventricular function, is linked to NFR [23]. Nevertheless, we failed to verify this assumption, considering that serum BNP level could not elevate in the early phase. Moreover, patients with Killip grade >4 were excluded, taking into account the fact that those cardiac shock candidates required additional rescue support and adjuvant treatment, leading to unbalanced comparisons between the subgroups.
Intracoronary thrombus burden is directly related to microvascular perfusion in myocardial infarction patients. Apart from the thrombus burden score, the interactions of thrombotic or coagulant laboratory parameters with CMVO were worthy of consideration. Compared with activated partial thromboplastin time and D-dimer, ACT is the most frequently utilized point of care test in monitoring whole blood coagulation and anticoagulant effects of heparin during a coronary intervention procedure, sharing advantages of shorter testing time, smaller blood sample size, and less sample degradation [24]. Levels of initial ACT prior to primary catheterization were assessed to predict IRA patency and coronary coagulating effects in our center conventionally, but without definite reference cutoff values and supporting evidence. Additionally, the relationship of this bedside index with patients’ prognosis is controversial. Tolleson et al. reported that for patients receiving contemporary intervention, an ACT of 200 seconds to 250 seconds was efficient to reduce ischemic events, while bleeding events increased with ACT levels [25]. Participants in the EASY trial were administered with abciximab and UFH to determine the relationship between the final ACT value in emergent PCI and outcomes, suggesting that with abciximab treatment, a higher final ACT value was beneficial in reducing post procedure myocardial injury without increasing bleeding [26]. The FUTURA/OASIS-8 trial conducted by Ducrocq et al. pointed out that peak ACT ≤300 seconds values of NSTE-ACS patients in PCI procedure are at high thrombotic risk with no clear hemorrhagic threat [27]. Rajpurohit et al. studied and reviewed 12 055 candidates who underwent conventional PCI, proposing that ACT values before intervention was not independently related to short- and long-period survival [28]. In consequence, for STEMI members, whether baseline ACT values are independently associated with ischemic, hemorrhagic risks and cardiac events, is worth further investigation. In this study, based on multivariate logistic regression result, we found that the initial ACT prior to the procedure had a close relationship with microvascular obstruction occurrence, with the possible cutoff values listed in the nomogram. Despite its advantages over other laboratory indexes, ACT value is influenced by various conditions (hypothermia, antiplatelet treatment, and hemodilution). Whereas this study was performed in a single center, the range of ACT values might be different in other centers due to different personnel and testing device.
Inflammatory activation plays an essential role during coronary atherosclerosis, along with microvascular dysfunction. With the growing understanding of the pathophysiological mechanisms, neutrophil-platelets are thought to be activated to exaggerate inflammatory cascade reaction by releasing mediators, radical oxygen, and to aggregate to block the coronary lumen, deteriorating microvascular perfusion after restoring the blood flow of the culprit vessel. Among all the routine blood indexes, NLR has been confirmed as the most significantly independent parameter of CMVO [29,30], which is consistent with the present study findings.
In a myocardial infarction setting, the glucose level rises as a result of sympathetic nerve activation and increase in plasma catecholamine. It has been implied that, regardless of diabetes history, admission with hyperglycemia has a close connection with mortality [31]. Yildiz et al. reported that glucose on admission was independent of high TFC, regardless of diabetes history [32]. Recently, a study revealed that hyperglycemia affects coronary thrombus burden and inflammatory status, resulting in worse prognosis in patients undergoing interventional treatment [33]. As seen in the present study, we obtained comparable results, supporting that hyperglycemia on admission directly correlated to CMVO among the STEMI patients undergoing primary recanalization.
Limited by sample size and evaluation conditions, there were some statistically different indicators between the 2 subgroups, including hs-CRP, eGFR, D-dimer, diabetes history, and collateral circulation, which were shown not to be responsible for CMVO independently. Additionally, some predictors with obvious significance were not entered into the regression analysis owing for several reasons. First, the scores that incorporated a series of clinical elements, like GRACE, CRUSADE, SYNTAX, and CHA2DS2-VASc scores, are presumably not suitable and practical in constructing new models, considering that they might influence the weight of risk variables, regardless of their correlation with NRF incidence [34,35]. Second, given the fact that cardiac biomarker values were strongly correlated with ischemic time and affected by a range of factors (infarct size, collateral circulation, and several other conditions) [10], initial CK-MB and cTnI were excluded. Furthermore, in spite of the confirmed independence of microvascular obstruction, initial coronary thrombus burden was eliminated due to its multicollinearity with ACT values in regression analysis.
In addition to the potential factors, previous models or estimating scores of NFR have been recently created. Dogan et al. reported that hyperglycemia, prolonged ischemic time, and low neutrophil count were attributed to the development of the risk model [36]. Bayramoglu et al. built a predictive model covering age, LVEF value, SYNTAX score, stent length, thrombus burden score, Killip classification and reperfusion time [37]. The retrospective study conducted by Wang et al. showed that age, pre-PCI thrombus score, pain to PCI time, Killip class, neutrophil level, admission glucose, and collateral circulation could be adopted to establish a no reflow model [38]. Regardless of those results, the existing models were usually displayed as counting scores.
To our knowledge, this is the first study to present a nomogram model as a novel and valid representation form to identify the possibility of CMVO. Unlike the former systems, nomogram is a specific and graphical tool that integrates quantified independent predictors with corresponding risk or prognosis possibility according to the individual condition [39,40]. Markedly, C-index or AUC value over 0.75 is recognized as a reliable validation. The present nomogram performed well in calculating the risk probability, and its validity was tested by internal and external validation and the calibration curve, highlighting the predictive accuracy of the established model and nomogram. The comprehensive nomogram might be reasonable and sensible in facilitating the physicians assessing the potential CMVO risk in the early stage of STEMI, offering therapeutic and mechanical guidance in clinical practice.
This study exhibits the following limitations: first, this retrospective trial was conducted from one single institutional dataset with a relatively small sample scale, hence C-index and AUC for internal as well as external validation might be higher. Therefore, multicenter prospective validations of this risk model are required to confirm this. Second, hyperglycemia impairs the microvascular integrity in complex mechanisms, and we did not investigate the possible correlations of glucose intolerance/insulin resistance indices and CMVO due to the lack of original data. In future studies, glucose tolerance test, glycosylated hemoglobin, free fatty acids, post prandial insulin concentrations, and other relative quantifications need to be addressed. Furthermore, the level of ACT is influenced by a series of factors, and the reference range in the present nomogram might be different depending on the testing equipment and surroundings.
Conclusions
In conclusion, the nomogram model based on 6 variables, including age, symptom onset to balloon time, Killip classification, admission ACT, neutrophil/lymphocyte ratio, and glucose value, could accurately predict coronary microvascular obstruction risk for STEMI patients undergoing primary percutaneous catheterization.
Acknowledgments
We thank the staff in the Cardiology Division and Cardiac Catheterization Laboratory of the Second Hospital of Hebei Medical University deeply for their contributions and devotions.
Abbreviations
- ACT
activated clotting time
- AUC
area under curve
- BMI
body mass index
- BNP
brain natriuretic peptide
- CI
confidence interval
- C-index
concordance index
- CK-MB
creatine kinase MB isozyme
- CMVO
coronary microvascular functional and structural obstruction
- cTFC
corrected TIMI frame count
- cTn
cardiac troponin
- FMC
B first medical contact to balloon
- hs-CRP
high sensitive C-reactive protein
- LVEF
left ventricular ejection fraction
- NLR
neutrophil/lymphocyte ratio
- NRF
no-reflow phenomenon
- PCI
percutaneous coronary intervention
- PTCA
percutaneous transluminal coronary angioplasty
- ROC
receiver operating characteristic
- SO-B
symptom onset to balloon
- STEMI
ST-segment elevation myocardial infarction
- TIMI
thrombolysis in myocardial infarction
- TMPG
TIMI myocardial perfusion grade
- UFH
unfractionated heparin
Footnotes
Source of support: This work was funded by the grant of National Key R&D Program of China, No. 2016YFC1301100
Conflict of interest
None.
References
- 1.Niccoli G, Burzotta F, Galiuto L, Crea F. Myocardial no-reflow in humans. J Am Coll Cardiol. 2009;54:281–92. doi: 10.1016/j.jacc.2009.03.054. [DOI] [PubMed] [Google Scholar]
- 2.Lerman A, Holmes DR, Herrmann J, Gersh BJ. Microcirculatory dysfunction in ST-elevation myocardial infarction: Cause, consequence, or both. Eur Heart J. 2007;28:788–97. doi: 10.1093/eurheartj/ehl501. [DOI] [PubMed] [Google Scholar]
- 3.Bolognese L, Carrabba N, Parodi G, et al. Impact of microvascular dysfunction on left ventricular remodeling and long-term clinical outcome after primary coronary angioplasty for acute myocardial infarction. Circulation. 2004;109:1121–26. doi: 10.1161/01.CIR.0000118496.44135.A7. [DOI] [PubMed] [Google Scholar]
- 4.Rezkalla SH, Stankowski RV, Hanna J, Kloner RA. Management of no-reflow phenomenon in the catheterization laboratory. JACC Cardiovasc Interv. 2017;10:215–23. doi: 10.1016/j.jcin.2016.11.059. [DOI] [PubMed] [Google Scholar]
- 5.Guo AQ, Sheng L, Lei X, Shu W. Pharmacological and physical prevention and treatment of no-reflow after primary percutaneous coronary intervention in ST-segment elevation myocardial infarction. J Int Med Res. 2013;41:537–47. doi: 10.1177/0300060513479859. [DOI] [PubMed] [Google Scholar]
- 6.Durante A, Camici PG. Novel insights into an “old” phenomenon: The no reflow. Int J Cardiol. 2015;187:273–80. doi: 10.1016/j.ijcard.2015.03.359. [DOI] [PubMed] [Google Scholar]
- 7.Sabin P, Koshy AG, Gupta PN, et al. Predictors of no- reflow during primary angioplasty for acute myocardial infarction, from Medical College Hospital, Trivandrum. Indian Heart J. 2017;69(Suppl 1):S34–45. doi: 10.1016/j.ihj.2016.12.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yip HK, Chen MC, Chang HW, et al. Angiographic morphologic features of infarct-related arteries and timely reperfusion in acute myocardial infarction: Predictors of slow-flow and no-reflow phenomenon. Chest. 2002;122:1322–32. doi: 10.1378/chest.122.4.1322. [DOI] [PubMed] [Google Scholar]
- 9.Niccoli G, Scalone G, Lerman A, Crea F. Coronary microvascular obstruction in acute myocardial infarction. Eur Heart J. 2016;37:1024–33. doi: 10.1093/eurheartj/ehv484. [DOI] [PubMed] [Google Scholar]
- 10.Reinstadler SJ, Stiermaier T, Fuernau G, et al. The challenges and impact of microvascular injury in ST-elevation myocardial infarction. Expert Rev Cardiovasc Ther. 2016;14:431–43. doi: 10.1586/14779072.2016.1135055. [DOI] [PubMed] [Google Scholar]
- 11.Ibanez B, James S, Agewall S, et al. 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: The Task Force for the management of acute myocardial infarction in patients presenting with ST-segment elevation of the European Society of Cardiology (ESC) Eur Heart J. 2018;39:119–77. doi: 10.1093/eurheartj/ehx393. [DOI] [PubMed] [Google Scholar]
- 12.Ganz W. The thrombolysis in myocardial infarction (TIMI) trial. N Engl J Med. 1985;313:1018. doi: 10.1056/NEJM198510173131611. [DOI] [PubMed] [Google Scholar]
- 13.Gibson CM, de Lemos JA, Murphy SA, et al. Combination therapy with abciximab reduces angiographically evident thrombus in acute myocardial infarction: a TIMI 14 substudy. Circulation. 2001;103:2550–54. doi: 10.1161/01.cir.103.21.2550. [DOI] [PubMed] [Google Scholar]
- 14.Gibson CM, Cannon CP, Murphy SA, et al. Relationship of TIMI myocardial perfusion grade to mortality after administration of thrombolytic drugs. Circulation. 2000;101:125–30. doi: 10.1161/01.cir.101.2.125. [DOI] [PubMed] [Google Scholar]
- 15.Gibson CM, Cannon CP, Daley WL, et al. TIMI frame count: A quantitative method of assessing coronary artery flow. Circulation. 1996;93:879–88. doi: 10.1161/01.cir.93.5.879. [DOI] [PubMed] [Google Scholar]
- 16.Balachandran VP, Gonen M, Smith JJ, DeMatteo RP. Nomograms in oncology: More than meets the eye. Lancet Oncol. 2015;16:e173–80. doi: 10.1016/S1470-2045(14)71116-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.de Waha S, Patel MR, Granger CB, et al. Relationship between microvascular obstruction and adverse events following primary percutaneous coronary intervention for ST-segment elevation myocardial infarction: An individual patient data pooled analysis from seven randomized trials. Eur Heart J. 2017;38:3502–10. doi: 10.1093/eurheartj/ehx414. [DOI] [PubMed] [Google Scholar]
- 18.Hamirani YS, Wong A, Kramer CM, Salerno M. Effect of microvascular obstruction and intramyocardial hemorrhage by CMR on LV remodeling and outcomes after myocardial infarction: A systematic review and meta-analysis. JACC Cardiovasc Imaging. 2014;7:940–52. doi: 10.1016/j.jcmg.2014.06.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kirma C, Izgi A, Dundar C, et al. Clinical and procedural predictors of no-reflow phenomenon after primary percutaneous coronary interventions: Experience at a single center. Circ J. 2008;72:716–21. doi: 10.1253/circj.72.716. [DOI] [PubMed] [Google Scholar]
- 20.Harrison RW, Aggarwal A, Ou FS, et al. Incidence and outcomes of no-reflow phenomenon during percutaneous coronary intervention among patients with acute myocardial infarction. Am J Cardiol. 2013;111:178–84. doi: 10.1016/j.amjcard.2012.09.015. [DOI] [PubMed] [Google Scholar]
- 21.Herrmann J, Lerman A. The endothelium: Dysfunction and beyond. J Nucl Cardiol. 2001;8:197–206. doi: 10.1067/mnc.2001.114148. [DOI] [PubMed] [Google Scholar]
- 22.De Luca G, Gibson CM, Huber K, et al. Association between advanced Killip class at presentation and impaired myocardial perfusion among patients with ST-segment elevation myocardial infarction treated with primary angioplasty and adjunctive glycoprotein IIb-IIIa inhibitors. Am Heart J. 2009;158:416–21. doi: 10.1016/j.ahj.2009.06.029. [DOI] [PubMed] [Google Scholar]
- 23.Jeong YH, Kim WJ, Park DW, et al. Serum B-type natriuretic peptide on admission can predict the ‘no-reflow’ phenomenon after primary drug-eluting stent implantation for ST-segment elevation myocardial infarction. Int J Cardiol. 2010;141:175–81. doi: 10.1016/j.ijcard.2008.11.189. [DOI] [PubMed] [Google Scholar]
- 24.Spinler SA, Wittkowsky AK, Nutescu EA, Smythe MA. Anticoagulation monitoring part 2: Unfractionated heparin and low-molecular-weight heparin. Ann Pharmacother. 2005;39:1275–85. doi: 10.1345/aph.1E524. [DOI] [PubMed] [Google Scholar]
- 25.Tolleson TR, O’Shea JC, Bittl JA, et al. Relationship between heparin anticoagulation and clinical outcomes in coronary stent intervention: Observations from the ESPRIT trial. J Am Coll Cardiol. 2003;41:386–93. doi: 10.1016/s0735-1097(02)02767-5. [DOI] [PubMed] [Google Scholar]
- 26.Bertrand OF, Rodés-Cabau J, Rinfret S, et al. Impact of final activated clotting time after transradial coronary stenting with maximal antiplatelet therapy. Am J Cardiol. 2009;104:1235–40. doi: 10.1016/j.amjcard.2009.06.036. [DOI] [PubMed] [Google Scholar]
- 27.Ducrocq G, Jolly S, Mehta SR, et al. Activated clotting time and outcomes during percutaneous coronary intervention for non-ST-segment-elevation myocardial infarction: Insights from the FUTURA/OASIS-8 Trial. Circ Cardiovasc Interv. 2015;8(4) doi: 10.1161/CIRCINTERVENTIONS.114.002044. pii: e002044. [DOI] [PubMed] [Google Scholar]
- 28.Rajpurohit N, Gulati R, Lennon RJ, et al. Relation of activated clotting times during percutaneous coronary intervention to outcomes. Am J Cardiol. 2016;117:703–8. doi: 10.1016/j.amjcard.2015.12.003. [DOI] [PubMed] [Google Scholar]
- 29.Park JJ, Jang HJ, Oh IY, et al. Prognostic value of neutrophil to lymphocyte ratio in patients presenting with ST-elevation myocardial infarction undergoing primary percutaneous coronary intervention. Am J Cardiol. 2013;111:636–42. doi: 10.1016/j.amjcard.2012.11.012. [DOI] [PubMed] [Google Scholar]
- 30.Kurtul A, Murat SN, Yarlioglues M, et al. Increased neutrophil-to-lymphocyte ratio predicts persistent coronary no-flow after wire insertion in patients with ST-elevation myocardial infarction undergoing primary percutaneous coronary intervention. Clinics (Sao Paulo) 2015;70:34–40. doi: 10.6061/clinics/2015(01)07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Mellbin LG, Malmberg K, Norhammar A, et al. The impact of glucose lowering treatment on long-term prognosis in patients with type 2 diabetes and myocardial infarction: A report from the DIGAMI 2 trial. Eur Heart J. 2008;29:166–76. doi: 10.1093/eurheartj/ehm518. [DOI] [PubMed] [Google Scholar]
- 32.Yildiz A, Arat-Ozkan A, Kocas C, et al. Admission hyperglycemia and TIMI frame count in primary percutaneous coronary intervention. Angiology. 2012;63:325–29. doi: 10.1177/0003319711418957. [DOI] [PubMed] [Google Scholar]
- 33.Sardu C, D’Onofrio N, Mauro C, et al. Thrombus aspiration in hyperglycemic patients with high inflammation levels in coronary thrombus. J Am Coll Cardiol. 2019;73:530–31. doi: 10.1016/j.jacc.2018.10.074. [DOI] [PubMed] [Google Scholar]
- 34.Ipek G, Onuk T, Karatas MB, et al. CHA2DS2-VASc Score is a predictor of no-reflow in patients with st-segment elevation myocardial infarction who underwent primary percutaneous intervention. Angiology. 2016;67:840–45. doi: 10.1177/0003319715622844. [DOI] [PubMed] [Google Scholar]
- 35.Magro M, Nauta ST, Simsek C, et al. Usefulness of the SYNTAX score to predict “no reflow” in patients treated with primary percutaneous coronary intervention for ST-segment elevation myocardial infarction. Am J Cardiol. 2012;109:601–6. doi: 10.1016/j.amjcard.2011.10.013. [DOI] [PubMed] [Google Scholar]
- 36.Dogan NB, Ozpelit E, Akdeniz S, et al. Simple clinical risk score for no-reflow prediction in patients undergoing primary percutaneous coronary intervention with acute STEMI. Pak J Med Sci. 2015;31:576–81. doi: 10.12669/pjms.313.7484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Bayramoğlu A, Taşolar H, Kaya A, et al. Prediction of no-reflow and major adverse cardiovascular events with a new scoring system in STEMI patients. J Interv Cardiol. 2018;31(2):144–49. doi: 10.1111/joic.12463. [DOI] [PubMed] [Google Scholar]
- 38.Wang JW, Zhou ZQ, Chen YD, et al. A risk score for no reflow in patients with ST-segment elevation myocardial infarction after primary percutaneous coronary intervention. Clin Cardiol. 2015;38:208–15. doi: 10.1002/clc.22376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Wang Y, Li J, Xia Y, et al. Prognostic nomogram for intrahepatic cholangiocarcinoma after partial hepatectomy. J Clin Oncol. 2013;31:1188–95. doi: 10.1200/JCO.2012.41.5984. [DOI] [PubMed] [Google Scholar]
- 40.Touijer K, Scardino PT. Nomograms for staging, prognosis, and predicting treatment outcomes. Cancer. 2009;115:3107–11. doi: 10.1002/cncr.24352. [DOI] [PubMed] [Google Scholar]