Among patients hospitalized with acute myocardial infarction (AMI), cardiogenic shock (CS) is the leading cause of death, complicating up to 10% of admissions. Introduction of early revascularization strategies and mechanical ventricular support have seen short-term mortality associated with CS fall from 70% to 80% in the 1970s to approximately 50% to 60% in the 1990s. Previous studies reported a higher incidence of CS after AMI in women (11.6% vs 8.3%). The aims of this study were to determine hospital mortality outcomes and gender differences following primary percutaneous coronary intervention (PPCI) in the setting of CS. Data were collected prospectively among all patients undergoing PPCI for AMI at a large UK tertiary cardiac center between April 2008 and October 2011. A sample of 2,864 patients (women: 844 [29.5%]) underwent PPCI, of which 141 (4.9%) had a confirmed diagnosis of CS. Eighty-one of 2,019 [4.0%] male patients (mean age: 64.2 years) and 60 of 844 [7.1%]) female patients (mean age: 69.9 years) with CS underwent PPCI (p <0.001). The overall hospital mortality was 35.5% with no gender difference (male: 35.8% vs female: 35%, p >0.99). In conclusion, this analysis demonstrates that in the contemporary PPCI era, there is a reduction in the incidence of CS with reduced hospital mortality rates and no gender difference. The absence of a gender difference is remarkable because higher proportions of women presented with CS and were older than their male counterparts. Long-term follow-up data are required to determine if this difference is sustained.
Death from cardiogenic shock (CS) remains the leading cause of mortality in patients presenting to hospital with acute myocardial infarction (AMI). Advances in the management of patients with AMI have seen a fall in mortality with the introduction of primary percutaneous coronary intervention (PPCI). Despite this fall in overall mortality, death from CS, which occurs in up to 10% of AMI patients, remains high at approximately 40%. Women hospitalized with AMI are more likely to die than men, and much of this can be attributed to their different risk profile (older age, more hypertension, hypercholesterolemia, diabetes mellitus, and congestive heart failure). The higher prevalence of CS among women raises important issues with regard to targeting treatment. The aims of the present study were to record hospital mortality and explore gender differences among patients presenting with AMI undergoing PPCI using contemporary treatment strategies in the setting of CS.
Methods
Consecutive AMI patients undergoing PPCI in the setting of CS between April 2008 and October 2011 at the Freeman Hospital, United Kingdom, were included in the analysis. Cardiogenic shock was defined as the presence of systolic blood pressure <90 mm Hg sustained for at least 30 minutes in the absence of hypovolemia following a diagnosis of AMI. The Freeman Hospital, Newcastle-upon-Tyne Hospitals, National Health Service Foundation Trust is a high-volume tertiary percutaneous coronary intervention (PCI) center in the United Kingdom serving a population of approximately 2 million and 7 satellite hospitals and performing >3,000 procedures per year, of which >800 are PPCI. The majority of PPCI admissions are direct from the community, with 28% being transferred from other hospitals.
Patients presenting with chest pain suggestive of myocardial ischemia lasting >30 minutes accompanied by ST-segment elevation or new left bundle branch block on the electrocardiogram were considered for PPCI if they presented within 12 hours of symptom onset. Twelve-lead electrocardiograms performed at the time of first professional medical help were transmitted to the Freeman hospital coronary care unit where the decision to accept patients for PPCI was made by the on-call cardiology team. Patients were administered 300 mg of soluble aspirin on first medical contact and then transferred directly to the cardiac catheterization laboratory. On arrival to Freeman hospital, either 600 mg of clopidogrel or 60 mg of prasugrel (in patients with no contraindication to prasugrel) was administered, along with standard doses of heparin or bivalirudin according to operator choice. Coronary angiography was performed via the radial or femoral artery. Coronary revascularization and the use of intra-aortic balloon pump were at the discretion of the attending cardiologist. Patients requiring ventilation and ionotropic support were admitted to the cardiac intensive care unit following the PCI procedure, and clinically stable patients were admitted to the coronary care unit.
This was an observational cohort study. Data were collected prospectively for all patients undergoing PCI. The primary source of data was our local coronary artery disease database (Dendrite), which contains information on each PCI procedure performed at our institution. Baseline characteristics, clinical presentation, procedure details, and procedural complications were prospectively collected at the end of each procedure by the operator and entered in the dedicated PCI database. Postprocedural complications, clinical data, and medications were updated on discharge.
Data are presented as percentages for categorical variables and as means ± SD for continuous variables and for counts. Comparisons were performed using the Fisher’s exact test for categorical variables, the unpaired t test with unequal variance assumption for continuous variables with bell-shaped empirical cumulative distribution function, or Wilcoxon 2-sample rank-sum test for continuous variables with highly skewed empirical cumulative distribution function, and Poisson regression for counts. We applied binomial test to test if the unadjusted hospital mortality rates following primary PCI is <50% (the lower bound of the range [50% to 60%] of the short-term mortality due to CS in 1990s). Simple (multiple) logistic regressions were used to test the individual (joint) effects of standard clinical, angiographic, and procedural factors on hospital mortality. All tests except binomial test are 2-sided tests. Binomial test is 1-sided test (the null hypothesis is that hospital mortality is equal to 50%; the alternative hypothesis is that hospital mortality is <50%). A p value <0.05 was considered statistically significant. All analysis was performed using SAS (9.3, SAS Inc., Cary, North Carolina).
Results
The baseline characteristics of patients presenting with and without CS are displayed in Table 1 . In total, 2,864 patients (women: 844 [29.5%]) underwent PPCI. Of 141 (4.9%) presenting with CS, 81 of 2,019 [4.0%] were men and 60 of 844 [7.1%] were women. The only significant difference in baseline characteristics of CS patients was the increased age of female patients (mean age: 69.9 years vs 64.2 years, p = 0.02). The angiographic and procedural characteristics are displayed in Table 2 . The only significant differences in angiographic and procedural characteristics for CS patients were seen for infarct location (more women had inferior infarcts; men: 24.7% vs women: 49.2%, p = 0.004), ventilated patients (men: 33.8% vs women 13.2%, p = 0.008), and cardiac arrest (more men experienced cardiac arrest: 64.2% vs 35.6%, p = 0.001).
| Variable | Total With CS n = 141/2,864 (5%) | Male With CS ∗ n = 81/2,019 (4%) | Female With CS n = 60/844 (7%) | p Value † | Total Without CS n = 2,722/2,864 (95%) | Male Without CS n = 1,938/2,019 (96%) | Female Without CS n = 784/844 (93%) | p Value † |
|---|---|---|---|---|---|---|---|---|
| Age (Mean ± SD yrs) | 66.6 ± 13.8 | 64.2 ± 12.3 | 69.9 ± 15.0 | 0.02 | 63.1 ± 13.4 | 60.9 ± 12.8 | 68.3 ± 13.3 | <0.001 |
| Hypertension | 61/124 (49.2%) | 33/71 (46.5%) | 28/53 (52.8%) | 0.59 | 1,134/2,647 (42.8%) | 748/1,885 (39.7%) | 386/762 (50.7%) | <0.001 |
| Diabetes mellitus | 19/128 (14.8%) | 10/72 (13.9%) | 9/56 (16.1%) | 0.80 | 292/2,669 (10.9%) | 200/1,900 (10.5%) | 92/769 (12.0%) | 0.30 |
| Total cholesterol >5.2 mmol/L | 42/124 (33.9%) | 22/71 (31.0%) | 20/53 (37.7%) | 0.45 | 898/2,647 (33.9%) | 630/1,885 (33.4%) | 268/762 (35.2%) | 0.39 |
| Previous angina pectoris | 33/124 (26.6%) | 17/69 (24.6%) | 16/55 (29.1%) | 0.68 | 539/2,680 (20.1%) | 377/1,912 (19.7%) | 162/768 (21.1%) | 0.42 |
| Previous myocardial infarction | 22/124 (17.7%) | 12/71 (16.9%) | 10/53 (18.9%) | 0.82 | 365/2,673 (13.7%) | 280/1,910 (14.7%) | 85/763 (11.1%) | 0.02 |
| Previous PCI | 7/135 (5.2%) | 3/77 (3.9%) | 4/58 (6.9%) | 0.46 | 180/2,715 (6.6%) | 150/1,932 (7.8%) | 30/783 (3.8%) | <0.001 |
| Previous coronary artery bypass graft | 2/136 (1.5%) | 1/78 (1.3%) | 1/58 (1.7%) | >0.99 | 57/2,713 (2.1%) | 46/1,931 (2.4%) | 11/782 (1.4%) | 0.14 |
| Smoker (ever smoker) | 79/101 (78.2%) | 47/58 (81.0%) | 32/43 (74.4%) | 0.47 | 1,987/2,613 (76.0%) | 1,463/1,867 (71.5%) | 524/746 (70.2%) | <0.001 |
| Pulmonary disease | 16/117 (13.7%) | 9/66 (13.6%) | 7/51 (13.7%) | >0.99 | 331/2,622 (12.6%) | 207/1,873 (11.1%) | 124/749 (16.6%) | <0.001 |
∗ The incidence rate of CS among male patients is significantly different from that among female patients (p value <0.001).
† Two-sample t test (unequal variance) for continuous variable (e.g., age) and Fisher’s exact test for categorical variable (e.g., diabetes mellitus).
| Variable | Total n = 141/2,864 (5%) | Male n = 81/2,019 (4%) | Female n = 60/844 (7%) | p Value | Total Without CS n = 2,722/2,864 (95%) | Male Without CS n = 1,938/2,019 (96%) | Female Without CS n = 784/844 (93%) | p Value |
|---|---|---|---|---|---|---|---|---|
| No. of lesion attempted (mean ± SD) | 1.4 ± 0.7 | 1.5 ± 0.7 | 1.4 ± 0.8 | 0.75 ∗ | 1.2 ± 0.5 | 1.2 ± 0.5 | 1.2 ± 0.5 | 0.70 |
| No. of stents per lesion (mean ± SD) | 1.3 ± 0.8 | 1.4 ± 0.9 | 1.1 ± 0.7 | 0.23 ∗ | 1.3 ± 0.8 | 1.3 ± 0.8 | 1.2 ± 0.8 | 0.10 |
| No. of drug-eluting stents | 70/120 (58.3%) | 44/69 (63.8%) | 26/51 (51.0%) | 0.19 | 1,540/2,469 (62.4%) | 1,114/1,768 (63.0%) | 426/701 (60.8%) | 0.31 |
| Radial procedure | 45/141 (31.9%) | 28/81 (34.6%) | 17/60 (28.3%) | 0.47 | 1,942/2,717 (71.5%) | 1,418/1,935 (73.3%) | 524/782 (67.0%) | 0.001 |
| Femoral procedure | 96/141 (68.1%) | 53/81 (65.4%) | 43/60 (71.7%) | 0.47 | 758/2,717 (27.9%) | 509/1,935 (26.3%) | 249/782 (31.8%) | 0.004 |
| Location of infarct: anterior | 65/140 (46.4%) | 41/81 (50.6%) | 24/59 (40.7%) | 0.30 | 1,043/2,696 (38.7%) | 786/1,916 (41.0%) | 257/780 (33.0%) | <0.001 |
| Location of infarct: inferior | 49/140 (35.0%) | 20/81 (24.7%) | 29/59 (49.2%) | 0.004 | 1,392/2,696 (51.6%) | 953/1,916 (49.7%) | 439/780 (56.3%) | 0.002 |
| Three-vessel coronary disease (>75% stenosis) | 53/141 (37.6%) | 33/81 (40.7%) | 20/60 (33.3%) | 0.39 | 325/2,655 (12.2%) | 236/1,886 (12.5%) | 89/769 (11.6%) | 0.56 |
| Left main stem stenosis (>50%) | 35/141 (24.8%) | 25/81 (30.9%) | 10/60 (16.7%) | 0.08 | 146/2,704 (5.4%) | 102/1,926 (5.3%) | 44/778 (5.7%) | 0.71 |
| Door to balloon time (median [minimum, maximum]) (min) | 24 [5, 225] | 23 [5, 225] | 24 [9, 126] | 0.77 † | 25 [3, 239] | 25 [3, 239] | 25 [6, 174] | 0.13 |
| Use of IIb/IIIa | 109/140 (77.9%) | 64/80 (80.0%) | 45/60 (75.0%) | 0.54 | 2,196/2,709 (81.1%) | 1,615/1,932 (83.6%) | 581/777 (74.8%) | <0.001 |
| Ventilated patients | 33/130 (25.4%) | 26/77 (33.8%) | 7/53 (13.2%) | 0.008 | 20/2,500 (0.8%) | 10/1,774 (0.6%) | 10/726 (1.4%) | 0.05 |
| Intra-aortic balloon pump use | 68/139 (48.9%) | 44/79 (55.7%) | 24/60 (40.0%) | 0.09 | 36/2,690 (1.3%) | 27/1,917 (1.4%) | 9/773 (1.2%) | 0.71 |
| Cardiac arrest | 73/140 (52.1%) | 52/81 (64.2%) | 21/59 (35.6%) | 0.001 | 230/2,705 (8.5%) | 166/1,926 (8.6%) | 64/779 (8.2%) | 0.76 |
Among the 141 patients, 54 of 117 (46.2%) experienced complications (defined as ≥1 of death, reintervention, blood transfusion, and cerebrovascular accident bleed) during hospital stay with no gender difference (men: 44.9% vs women: 47.9%, p = 0.85; Table 3 ).
| Variables | Total n = 141/2,864 (5%) | Men n = 81/2,019 (4%) | Women n = 60/844 (7%) | p Value ∗ |
|---|---|---|---|---|
| In-hospital complication † | 54/117 (46.2%) | 31/69 (44.9%) | 23/48 (47.9%) | 0.85 |
| Hospital mortality | 50/141 (35.5%) | 29/81 (35.8%) | 21/60 (35.0%) | >0.99 |
† Death, reintervention, blood transfusion, and cerebrovascular accident bleed.
The overall unadjusted hospital mortality rate was 35.5% with no gender difference (men: 35.8% vs women: 35.0%, p >0.99; Table 3 ). Binomial test showed mortality to be significantly (p <0.001) less than 50% (the lower bound of the range [50% to 60%] of the short-term mortality due to CS in 1990s). There was no difference between the risk that a male died of CS than the risk that a female died of CS, even after adjustment for cardiac arrest (odds ratio [OR] 1.033, 95% confidence interval [CI] 0.249 to 4.287, p = 0.96) or after stratifying cardiac arrest (with cardiac arrest, OR 1.44; 95% CI 0.519 to 3.998, p = 0.48 and without cardiac arrest, OR 1.94; 95% CI 0.532 to 7.072, p = 0.32). Tables 4 and 5 demonstrates the individual and joint effects of clinical, angiographic, and procedural factors on hospital mortality using simple and multiple logistic regression analysis respectively. On multivariate regression, cardiac arrest was a predictor of hospital mortality (OR 0.073; 95% CI 0.012 to 0.434, p = 0.004).
| Variables | OR (95% CI) | p Value |
|---|---|---|
| Age (yrs) | 1.021 (0.995–1.048) | 0.1193 |
| Gender (female vs male) | 0.966 (0.480–1.941) | 0.9216 |
| Hypertension (no vs yes) | 0.955 (0.445–2.048) | 0.9049 |
| Diabetes mellitus (no vs yes) | 0.462 (0.172–1.243) | 0.1262 |
| Hypercholesterolemia (no vs yes) | 2.006 (0.845–4.765) | 0.1146 |
| Previous angina (no vs yes) | 0.490 (0.201–1.144) | 0.0993 |
| Previous myocardial infraction (no vs yes) | 0.243 (0.093–0.633) | 0.0038 |
| Previous PCI (no vs yes) | 0.651 (0.139–3.043) | 0.5855 |
| Previous coronary artery bypass surgery (no vs yes) | 0.489 (0.030–8.001) | 0.6158 |
| Family history (no vs yes) | 1.555 (0.597–4.049) | 0.3655 |
| Medical history ∗ (no history vs having history) | 0.937 (0.425–2.063) | 0.8714 |
| Smoking status (never smokers vs ever smokers) | 0.436 (0.117–1.626) | 0.2165 |
| H/O pulmonary disease (no vs yes) | 0.704 (0.235–2.112) | 0.5315 |
| Type of stent used (bare metal vs drug eluting) | 1.000 (0.453–2.208) | >0.999 |
| Radial procedure (no vs yes) | 2.115 (0.957–4.672) | 0.0640 |
| Femoral procedure (no vs yes) | 0.473 (0.214–1.045) | 0.0640 |
| Location of infarct: anterior (no vs yes) | 0.906 (0.453–1.812) | 0.7811 |
| Location of infarct: infarct inferior (no vs yes) | 1.417 (0.676–2.969) | 0.3562 |
| 3 vessel disease (≤75% vs >75% stenosis) | 0.577 (0.285–1.171) | 0.1280 |
| Left main stem stenosis (≤50% vs >50%) | 0.657 (0.301–1.437) | 0.2931 |
| Door to balloon time (min) | 1.004 (0.991–1.016) | 0.5923 |
| Use of IIb/IIIa (no vs yes) | 2.068 (0.918–4.661) | 0.0797 |
| Ventilated patients (no vs yes) | 0.431 (0.191–0.971) | 0.0422 |
| Intra-aortic balloon pump use (no vs yes) | 0.878 (0.438–1.762) | 0.7149 |
| Cardiac arrest (no vs yes) | 0.234 (0.110–0.501) | 0.0002 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
