Summary
Background
In primary percutaneous coronary intervention (pPCI), conflicting data exist on the relative importance of patient presentation time (time from symptom onset (SO) to first medical contact [FMC]) and transfer time (time from FMC to sheath insertion).
Objectives
To evaluate the impact of transfer time on mortality in an unselected ST-elevation myocardial infarction (STEMI) population treated with pPCI.
Methods
In a well-organized urban network, using mobile intensive care units (MICU) whenever possible, the impact of transfer time on inhospital mortality was evaluated in 703 unselected consecutive STEMI patients transferred for pPCI.
Results
Our STEMI population included patients with cardiogenic shock (5.3%) and out-of-hospital cardiac arrest (3.7%). Longer transfer times were found to be associated with a stepwise increase in mortality ranging from 2.99% in the first quartile (Q1) up to 8.65% in the fourth quartile (Q4) ( P = 0.005). This result was noted in patients presenting early (≤ 2 h of SO, 0.96% for Q1 vs. 9.8% for Q4, P = 0.006) but not in late presenters (> 2 h of SO, 7.00% for Q1 vs. 7.8% for Q4, P = 0.85). After adjustment for confounding variables such as the severity of patients, the relationship between mortality and transfer time was no longer apparent.
Conclusions
In a well-organized urban network dedicated to pPCI, including unselected STEMI patients, transfer time does not appear to be a major contributor to mortality. The relationship of transfer time to mortality seems to be dependent on presentation time and patients’ clinical severity.
Résumé
Contexte
Il existe peu de données sur l’importance pronostique relative du délai de présentation du patient (début des symptômes [DS] – premier contact médical) et du temps de transfert (premier contact médical-insertion du désilet) dans le SCA ST+ traité par angioplastie primaire (AP).
Objectif
Évaluer l’impact indépendant du temps de transfert sur la mortalité intrahospitalière dans le SCA ST+ traité par AP.
Matériel et méthodes
Dans un réseau urbain organisé utilisant le SAMU, l’impact du temps de transfert sur la mortalité intrahospitalière a été évalué chez 703 patients SCA ST+ non sélectionnés transférés pour AP.
Résultats
La population de l’étude comprenait des chocs cardiogéniques (5,3 %) et des arrêts cardiaques extrahospitaliers (3,7 %). L’allongement du temps de transfert était associé à une augmentation progressive de la mortalité (2,99 % pour le premier quartile [Q1] jusqu’à 8,65 % pour le quatrième quartile [Q4]; p = 0,005). Cette relation était encore plus marquée chez les patients se présentant précocement (≤ 2 h du DS, 0,96 % pour Q1 vs 9,8 % pour Q4; p = 0,006), mais non significative pour les patients se présentant tardivement (> 2 h du DS, 7,00 % pour Q1 vs 7,8 % pour Q4; p = 0,85). En analyse multivariée, le temps de transfert n’était plus associé à la mortalité.
Conclusion
Dans un réseau urbain organisé dédié à l’AP, le temps de transfert ne semble pas être un déterminant majeur de la mortalité. La relation entre le temps de transfert et la mortalité précoce apparaît fortement dépendante du délai de présentation et de la sévérité clinique.
Background
European and American guidelines support primary percutaneous coronary intervention (pPCI) as the treatment of choice for patients with acute ST-segment elevation myocardial infarction (STEMI), especially when delivered within 12 hours of symptom onset (SO) . Despite guidelines and quality improvement programmes, reducing time to reperfusion in STEMI patients remains important but challenging . Numerous studies, including randomized trials and meta-analyses, have investigated the benefits of reducing ischaemic time in STEMI patients treated with pPCI. Most have suggested a benefit in reducing time to reperfusion to salvage the myocardium involved, subsequently lowering the risk of death . Factors delaying time to reperfusion in pPCI have been identified, but vary according to countries, populations and facilities of the STEMI networks involved . Moreover, little is known about the importance of the relationship between patient presentation time (the time from SO to first medical contact [FMC]) and transfer time (the time from FMC to sheath insertion). Field triage appears to be a useful approach for reducing transfer time and allowing fast delivery of prehospital pharmacological therapies, thus potentially improving outcomes after pPCI . The French Health Care System has been using field triage with mobile intensive care units (MICUs) for a long time. An on-board physician, trained in the diagnosis of acute myocardial infarction, triages the patients and facilitates rapid transfer to a pPCI centre (direct transfer) while at the same time, administers prehospital antiplatelet therapy loading and anticoagulation. This system is predominant in Paris, with full coverage of the city area where pPCI is the exclusive mode of reperfusion for STEMI.
While it would seem logical that a reduction in the time from FMC to catheterization (transfer time) would result in improved mortality through reduced ischaemic time, the translation of this general finding in our specific urban STEMI network using MICUs routinely remains uncertain. Additionally, determinants of mortality such as presentation time are often not considered in studies of time delays for pPCI and the most severe patients are often excluded from such analyses. We therefore tested the hypothesis that transfer time for pPCI is still a major contributor to inhospital death in our specific STEMI network. For that purpose, we evaluated the independent impact of transfer time on inhospital mortality in unselected consecutive STEMI patients routinely treated with pPCI in a high-volume PCI centre.
Methods
Study design and patient population
Consecutive patients admitted for pPCI at the catheterization laboratory of the Pitié-Salpêtrière University Hospital, Paris (France) between June 2004 and February 2007 were included in the e-PARIS registry, a web-based registry used to gather data from patients referred to our institution. Patients discharged without a final diagnosis of STEMI were excluded. We identified 703 patients with a confirmed STEMI, including patients with cardiogenic shock and out-of-hospital cardiac arrest. STEMI was defined as the presence of chest discomfort or symptoms of myocardial ischaemia, associated with new or presumed new electrocardiographic abnormalities in the ST-segment (elevation at the J point of at least 0.2 mV in leads V1, V2 and V3 and at least 0.1 mV in at least two contiguous leads), or new left bundle branch block, associated with elevation of cardiac enzymes at least three times above the upper limit of normal. The Pitié-Salpêtrière University Hospital is part of the Paris STEMI network, which comprises seven PCI centres that are open 24 hours/day, 7 days/week. Our centre and the corresponding MICU team cover the south and east part of the city, which represents approximately one-quarter of the Parisian population. We have on-site cardiac surgery available and we are the invasive hub for five non-PCI hospitals, including three emergency departments (EDs) that provide 90% of the interhospital transfer patients.
First medical contact in the urban ST-elevation myocardial infarction network
The FMC can be made through two pathways. The first pathway is MICU contact (field triage); the dispatch centre can be reached by calling the dedicated number for medical emergencies (‘15’). All the calls are operated by a medical regulator who makes the decision to send a fully equipped MICU team on-site, with an on-board emergency doctor who is trained in 18-lead ECG interpretation, decides the reperfusion strategy and administers prehospital treatment. The second pathway is ED contact, where patients are triaged for pPCI after hospital admission (‘walk-ins’), before ambulance transfer to the PCI centre.
Treatment delay definition
A total of seven key-time points were identified and systematically collected: SO or the time of occurrence of the permanent ischaemic symptoms; FMC or time of first physical medical contact with the patient or performance of the first electrocardiogram; activation call or first contact with the on-site pPCI cardiologist; catheterization laboratory door time or time of patient arrival to the catheterization laboratory; intervention time or arterial sheath insertion; time of first balloon inflation; time of thrombolysis in myocardial infarction (TIMI) 3 flow grade.
These time points were used to characterize two important time delays: presentation time (the time delay from SO to FMC); and transfer time (the time delay from FMC to sheath insertion).
Two additional time delays were collected but not used in the analysis: ischaemic time (the time delay from SO to attainment of TIMI 3 flow or the end of PCI [only in patients where TIMI 3 flow was obtained]); and abciximab time (the time delay from FMC to abciximab bolus administration [only in patients treated with abciximab]).
Primary percutaneous coronary intervention procedure
Our catheterization laboratory is open 24 hours/day, 7 days/week and can be directly activated by the ED or the MICU. Primary PCI is performed by the on-call senior interventional cardiologist according to contemporary interventional guidelines in STEMI presenters. All patients received at least aspirin and anticoagulation during transfer. The use of glycoprotein IIb/IIIa inhibitors was strongly encouraged for all patients triaged to pPCI and a loading dose of clopidogrel (900 mg) was given as soon as possible, according to our local protocol and based on previous studies . Thromboaspiration was used as often as possible if angiographically indicated, followed by systematic stent implantation (unless considered inappropriate by the physician). Subsequent medical treatment included anti-ischaemic, lipid-lowering and antithrombotic drugs, according to current treatment guidelines.
Baseline and procedural data
Baseline data were prospectively collected for all patients and entered in the web-based registry e-Paris. Data regarding medication at admission and during follow-up were recorded, as well as inhospital events until discharge. Our angiographic core laboratory reviewed all angiographic films and blindly evaluated TIMI flow grade, TIMI frame count and TIMI myocardial blush grade. Special attention was given to comorbidities and a risk profile was defined for each patient according to the TIMI risk score for STEMI .
Study objectives
Our main objective was to evaluate the impact of transfer time on inhospital death, defined as death from any cause during the initial hospitalization period. Secondary objectives were to evaluate the impact of presentation time on mortality, defining early presenters as patients with an SO – FMC delay ≤ 120 minutes and late presenters as those with an SO – FMC delay > 120 minutes. We also evaluated the effect of field triage on the different key-time intervals.
Statistical analysis
Continuous variables are presented as means ± standard deviations and were compared with Student’s t test. Categorical variables are expressed as rates or proportions and were compared by the Chi 2 test or Fisher’s exact test. According to the American College of Cardiology/American Heart Association Task Force on Performance Measures , time intervals are expressed as means with interquartile ranges (IQRs; 25th–75th percentiles) and the non-parametric Wilcoxon rank-sum test was adopted for group comparisons (field triage group [MICU] versus non-field triage group [ED]) of time-delay. To address the impact of transfer time on inhospital death we divided the transfer time into quartiles and a multivariable logistic regression model was fit to evaluate the independent variables associated with inhospital mortality with their adjusted effect estimates on inhospital mortality. The following variables (potential confounders) that were associated with inhospital mortality in the univariate analysis were included in the multivariable model: age, creatinine, smoking status, high blood pressure, renal insufficiency, TIMI risk score > 4, out-of-hospital cardiac arrest, heart rate, cardiogenic shock, troponin peak, initial TIMI 3 flow, multivessel disease, presentation time group and field triage. All tests were two-sided with a significance level fixed at 5%. We undertook all analyses with SAS software, version 9.0 (SAS Institute Inc., Cary, NC, USA). The study was approved by the local scientific ethical committee (CPP) at the Pitié-Salpêtrière University Hospital.
Results
Patient population and baseline characteristics
Baseline characteristics, treatments and procedural data for the 703 patients included in this study are shown in Tables 1 and 2 . Patients were compared according to prehospital triage, i.e. patients who were referred by the MICU (field triage, n = 476; 67.7%) with direct transfer from the field to the catheterization laboratory versus those referred by the ED (non-field triage, n = 227; 32.3%) with inter- or intrahospital transfer for pPCI. All-comer STEMI patients were recruited with high-risk profiles and frequent comorbidities. Clopidogrel loading dose was administered either in the ED or during transfer by MICU and was ≥ 600 mg in one-third of patients. Clopidogrel loading was completed in all patients to reach a final loading dose of 900 mg when discharged from the catheterization laboratory . Abciximab was used in 73.7% of patients and enoxaparin was the most common anticoagulant used. Nine out of 10 patients (88%) underwent pPCI with radial access. MICU patients differed from ED patients with respect to clinical presentation and risk profile; for example, MICU patients displayed a much higher rate of out-of-hospital cardiac arrests (4.9% vs. 1.3%; P = 0.017). Angiographic success was similar in both groups but complete ST resolution was higher in the MICU group than in the ED group (71% vs. 64.7%; P = 0.027).
| Total ( n = 703) | ED ( n = 227; 32.3%) | MICU ( n = 476; 67.7%) | P | |
|---|---|---|---|---|
| Age (years) | 63 ± 14 | 61.7 ± 15 | 63.3 ± 14 | 0.16 |
| Women | 154 (21.9) | 49 (21.6) | 115 (24.2) | 0.45 |
| BMI (kg/m 2 ) | 25.6 ± 4 | 25.3 ± 4 | 25.7 ± 4 | 0.22 |
| Obesity (BMI > 30 kg/m 2 ) | 98 (13.9) | 28 (12.3) | 70 (14.7) | 0.34 |
| Diabetes | 144 (20.5) | 53 (23.3) | 91 (19.1) | 0.19 |
| Dyslipidaemia | 303 (43.1) | 81 (35.7) | 232 (48.8) | 0.0009 |
| Smoker | 302 (43) | 94 (41.3) | 208 (43.7) | 0.53 |
| Hypertension | 302 (43) | 100 (44) | 202 (42.5) | 0.41 |
| Family history of CAD | 153 (21.7) | 48 (21.3) | 115 (24.1) | 0.32 |
| Previous MI | 101 (14.4) | 32 (14.2) | 69 (14.5) | 0.82 |
| Previous PCI | 90 (12.8) | 30 (13.3) | 60 (12.5) | 0.75 |
| Previous CABG | 20 (3) | 7 (3.0) | 13 (2.7) | 0.84 |
| CrCl (mL/minute) | 81.6 ± 40 | 82.5 ± 40 | 81.3 ± 40 | 0.80 |
| CrCl < 60 mL/minute | 206 (29.3) | 71 (31) | 145 (30.5) | 0.80 |
| Heart rate (bpm) | 78 ± 18 | 78 ± 17 | 77 ± 18 | 0.92 |
| Systolic BP (mmHg) | 128 ± 25 | 128 ± 26 | 128 ± 24 | 0.93 |
| Anterior MI | 336 (47.8) | 98 (43.2) | 228 (47.9) | 0.21 |
| Killip | 1.3 ± 0.7 | 1.3 ± 0.7 | 1.3 ± 0.7 | 0.91 |
| Killip class ≥ 2 | 128 (18.2) | 34 (15.2) | 94 (19.7) | 0.55 |
| Cardiogenic shock | 37 (5.3) | 10 (4.5) | 27 (5.6) | 0.57 |
| Out-of-hospital cardiac arrest | 26 (3.7) | 3 (1.3) | 23 (4.9) | 0.017 |
| Troponin I (peak), UI/L | 77.2 ± 90 | 66.7 ± 82.9 | 81 ± 116 | 0.08 |
| TIMI risk score | 3.8 ± 2.6 | 3.9 ± 2.6 | 3.8 ± 2.6 | 0.57 |
| TIMI risk score > 4 | 249 (35.4) | 79 (35.0) | 170 (35.8) | 0.84 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
