Methods

This study was designed as a retrospective cohort study in patients with STEMI who were ≥18 years. From October 21, 1999, until November 23, 2012, all patients with STEMI were admitted consecutively to the invasive catheterization laboratory at the Copenhagen University Hospital, Rigshospitalet. From 1999 until May 2011, Rigshospitalet covered the capital region of Denmark, with a catchment area of 1.7 million inhabitants, which corresponds to 30% of the entire Danish population. Beginning on June 1, 2011, Rigshospitalet covered all of eastern Denmark, with a catchment area of 2.5 million subjects, which corresponds to 45% of the entire Danish population. During this period (1999 to 2012), all patients with STEMI (n = 5,373) were included consecutively, and the clinical data were stored in a local hospital PCI registry. Of the 5,373 patients with STEMI, 4,875 had STEMI without VF, 410 had VF before, and 88 had VF during PPCI. In the VF before PPCI group, 14 patients (3.4%) had only sustained (≥30 seconds) pulseless ventricular tachycardia (VT) with clinical cardiac arrest, and in the VF during PPCI group, only 7 patients (7.9%) had sustained (≥30 seconds) pulseless VT with clinical cardiac arrest. Because of the low number of sustained pulseless patients with VT and cardiac arrest, and also because VT degenerates rapidly to VF, we considered these patients in the respective VF groups. Patients with nonsustained VT (<30 seconds) or with VT without cardiac arrest were not considered as cases. Patients with VF before PPCI who continued to have VF during PPCI were enrolled in the study and categorized as VF before PPCI. Patients who were non-Danish citizens (n = 96) and Danish citizens (n = 46) who moved out of Denmark were excluded from the survival analysis because of the lack of follow-up. One patient was excluded due to an unknown date of death. Therefore, of 5,373 patients with STEMI, 143 (15 had VF before and 3 during PPCI) were excluded, and follow-up was conducted with 5,230 patients with STEMI, of whom 4,750 patients had STEMI without VF, 395 had STEMI with VF before PPCI, and 85 had STEMI with VF during PPCI. For patients who had multiple admissions during the study period, only the first admission was considered. Patients with out-of-hospital cardiac arrest were included on admission to a PCI site after resuscitation by trained emergency medical service personnel.

To qualify for the study, patients had to have cardiac symptoms lasting ≤12 hours and acute ST-segment elevation on an electrocardiogram. All patients underwent angiography and subsequent PPCI. Acute STEMI on an electrocardiogram was defined as an ST-segment elevation of >0.1 mV in 2 adjacent electrocardiographic leads from V ₄ to V ₆ or in limb leads II, III, and aVF; as a segment elevation of >0.2 mV in leads V ₁ to V ₃ ; or as a left bundle branch block. At our PCI center, only patients with cardiac symptom and with “presumed new” LBBB who were clinically unstable were transferred to the PCI center for primary PCI; and if the cause of LBBB was not acute MI, then the patients were not included in the STEMI registry. VF before PPCI was defined as VF before the guided catheter insertion for PPCI, and VF during PPCI was defined as VF occurring after guided catheter insertion but before the end of the procedure.

All subjects with permanent residence in Denmark receive a unique and personal Civil Registration Number (CRN) that enables linkage between nationwide health care–related registers on an individual level. We used the CRN to retrieve information on each patients’ vital status (alive or date of death) from The Danish Civil Registration System, where all subjects residing in Denmark are registered from birth or the time of immigration. Furthermore, the CRN was used to obtain prehospital reports and discharge summaries. Baseline demographics, medical history, angiographic findings, and treatment characteristics were collected from the hospital PCI registry ( Tables 1 and 2 ). These data were entered into the PCI registry by the PPCI operator and assistants in the catheterization laboratory during the PPCI procedure.

The baseline and presenting characteristics were computed as the median or proportion for the cases and controls, and descriptive differences were assessed using the Wilcoxon rank-sum test for continuous variables and the chi-square test or Fisher’s exact test (where appropriate) for categorical variables. A 2-tailed p value ≤0.05 was considered statistically significant.

A logistic regression model was constructed to identify only the risk factors that preceded STEMI and were associated with subsequent VF. The adjusted odds ratios (ORs) and accompanying 95% confidence intervals (CIs) were computed to determine the association of each variable with the risk of VF. An initial multivariable model was built by including any covariate that preceded STEMI (besides the Killip class at admission) with p <0.10 on the univariate test. Age, gender, smoking, preinfarction angina, and Killip class at admission were forced into the model based on their previous associations with VF. Other covariates were sequentially added to the model and retained if they were significant or if they changed the coefficient of a risk factor by >10%; covariates were excluded if they changed the standard error by >10%. The Hosmer-Lemeshow test was used to evaluate the final model, and the area under the receiver-operating characteristic curve was used to measure the prediction accuracy.

A Cox proportional hazard regression model was constructed to identify variables that were independently associated with all-cause mortality. To this model, VF before PPCI and during PPCI were then added as time dependent to determine the adjusted hazard ratio (HR) and 95% confidence interval (CI) of VF for mortality within 30 days of PPCI and 30 days after PPCI. We checked all the explanatory variables for proportionality and linearity and found that the model was appropriate for describing the covariate effects. For this purpose, we used all the baseline variables ( Table 1 ) and additional procedural variables, including procedure time, infarct location, Killip class at admission, and Thrombolysis In Myocardial Infarction (TIMI) flow grades. Because all the earlier mentioned variables were previously associated with VF, these variables were all forced into the model. We also tested for interaction between VF before and during PPCI to identify mortality difference over time within the 2 VF groups. Missing data for each variable are listed in Tables 1 and 2 . Therefore, we conducted sensitivity analyses, in which we compared the beta coefficients of the Cox regression analysis in complete cases with imputed data using multiple imputations by chained equations, where the basic idea is to use complete observations to represent incomplete observations. Survival curves over the follow-up period were drawn using the Kaplan-Meier estimator, and the groups were compared using the log-rank test. All patients were followed from the date of the STEMI until death or until May 03, 2013, whichever came first. All analyses were performed using either the Stata software package, V.12.0 (StataCorp, College Station, Texas) or the R (version 3.0.2) statistical programming language. The Danish Data Protection Agency approved the present study (j.nr.:2007-58-0015). Register-based studies do not require ethical approval in Denmark.

Results

In this study, we consecutively enrolled 5,373 patients with STEMI, of whom 410 had VF before PPCI and 88 had VF during PPCI. In total, VF before or during PPCI occurred in 498 patients (9.3%). The baseline characteristics and angiographic characteristics at the time of STEMI are outlined in Tables 1 and 2 , respectively. The results of the multivariate logistic regression analysis for VF before PPCI are summarized in Tables 3 and 4 . Younger age, anterior infarct, preprocedural TIMI flow grade 0 to I, and Killip class >I at admission were all positively associated with VF before PPCI ( Table 3 ), whereas inferior infarct, preprocedural TIMI flow grade 0 to I, and Killip class >I at admission were all positively associated with VF during PPCI ( Table 4 ).

Table 3

Adjusted odds ratios of variables associated with the risk of ventricular fibrillation before primary percutaneous coronary intervention

Variable	Contrast	Odds Ratio	95% Confidence Interval	P value
Age	Per 10 years of increase	0.84	0.73-0.97	0.020
Sex	Male vs. female	1.23	0.81-1.86	0.300
Infarct location	Anterior vs. non-anterior	1.46	1.04-2.02	0.020
TIMI flow before PPCI	TIMI 0-I vs. II-III	1.65	1.12-2.44	0.010
Smoking
Past	Never	0.98	0.60-1.65	0.900
Current	Never	1.10	0.72-1.72	0.600
Canadian Cardiovascular Society grading of angina pectoris
Class I-II	No previous angina	0.60	0.33-1.10	0.100
Class III-IV	No previous angina	0.66	0.39-1.12	0.100
Claudication	Yes vs. No	1.15	0.50-2.80	0.700
Killip class > I at admission	Killip class I	2.80	1.73-4.40	<0.001

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Predictors and Outcomes of Routine Versus Optimal Medical Therapy in Stable Coronary Heart Disease
2. The Impact of Metabolic Syndrome on Clinical Outcomes After Everolimus-Eluting Stent Implantation
3. Temporal Changes of Platelet Reactivity After Coronary Stenting—A Thing to Think About
4. Vortex Keratopathy: Fabry Related or Amiodarone Induced?

Stay updated, free articles. Join our Telegram channel

Join