Aortic stenosis (AS) is associated with an increased risk of bleeding, but little is known about the risk of bleeding during percutaneous coronary intervention (PCI) in patients with AS. In the era of transcutaneous aortic valve implantation, understanding the bleeding risks associated with AS is critical. This retrospective study included 7,926 patients who underwent PCI from 2004 to 2013. Patients were categorized according to the presence of significant AS: moderate or severe AS (n = 354) and mild or no AS (n = 7,572). The National Cardiovascular Data Registry (NCDR) definition of a bleeding event (transfusion, prolonged hospital stay, or decrease in hemoglobin >3.0 mg/dl) was used as the primary outcome, and the NCDR PCI Bleeding Risk Score was used to control for the underlying risk of bleeding because of baseline characteristics. Patients with AS had significantly higher NCDR PCI Bleeding Risk Scores and higher rates of bleeding events. Logistic regression showed that the NCDR PCI Bleeding Risk Score did predict bleeding outcomes. There was not, however, an independent association between AS and bleeding outcomes. The addition of AS to the risk score using net reclassification improvement did not enhance the model’s ability to predict bleeding (p = 0.71). These data suggest that the NCDR PCI Bleeding Risk Score appropriately adjusts for bleeding risks in patients with AS. In conclusion, although patients with AS are more likely to have bleeding complications after PCI, the increased risk of bleeding is driven by the patients’ baseline co-morbidities rather than their AS.
There is a relative paucity of data on outcomes of patients with aortic stenosis (AS) who undergo percutaneous coronary intervention (PCI). Goel et al found no difference in mortality in patients with severe AS versus those without severe AS who underwent PCI provided that the patients did not have a left ventricular ejection fraction <30%. A search of published reports, however, does not reveal any studies examining whether patients with AS are more likely to have a bleeding complication after PCI. Because patients with AS often undergo cardiac catheterization with PCI, establishing whether these patients are more likely to have bleeding complications is important and clinically useful.
Methods
This single-center, retrospective study comprised 7,926 patients who underwent PCI at MedStar Washington Hospital Center from March 18, 2004, to January 21, 2013. An ongoing registry of PCI procedures is maintained at this institution. No external funding was used to support this work. All patients gave written consent for cardiac catheterization with PCI. In all cases, the interventional strategy and the use of adjunctive devices and pharmacotherapy were at the discretion of the interventional cardiologist. All reported outcomes are in-hospital outcomes obtained by evaluation of the institutional registry and electronic medical record.
With respect to baseline patient characteristics, a chart review is performed on all patients who undergo PCI by subjects who have been trained on data definitions to determine their co-morbidities. Chronic renal insufficiency is defined in the database as a serum creatinine of >2.0 mg/dl on admission or a previous diagnosis of or treatment for chronic renal insufficiency. History of coronary artery disease is defined as a history of myocardial infarction, angina pectoris, or revascularization procedure. Evidence for a myocardial infarction is defined as elevated cardiac biomarkers ≥2 times the upper limit of normal or ischemic electrocardiographic changes. Heart failure was diagnosed if there was an evidence of fluid retention from a cardiac cause and could be diagnosed if any of the following were present: paroxysmal nocturnal dyspnea, cardiac rales, pulmonary congestion on x-ray, or an S3 gallop. The presence of edema or dyspnea alone was insufficient to make a diagnosis of heart failure.
The presence of AS was assessed using either echocardiogram or cardiac catheterization. If a patient had both an echocardiogram and a catheterization, the data from the echocardiogram were given preference. The aortic valve areas measured by echocardiography used to define mild, moderate, and severe AS were >1.5 cm 2 , 1.0 to 1.5 cm 2 , and <1.0 cm 2 , respectively. The mean gradients measured by cardiac catheterization used to define mild, moderate, and severe AS were <25 mm Hg, 25 to 40 mm Hg, and >40 mm Hg, respectively. For the purposes of this analysis, patients were categorized according to the presence of hemodynamically significant AS: moderate or severe AS and mild or no AS.
Clinical end points in this study included hematoma >4 cm, hematocrit decrease >15, need for blood transfusion, gastrointestinal bleeding, retroperitoneal bleeding, major hematoma (hematoma with hematocrit decrease), major bleeding, and a bleeding event as defined by the National Cardiovascular Data Registry (NCDR) (transfusion, prolonged hospital stay, or decrease in hemoglobin >3.0 mg/dl). Major bleeding is defined in the database for this study using the Randomized Evaluation in PCI Linking Angiomax to Reduced Clinical Events II (REPLACE-2) definition of major bleeding, which includes a hematocrit decrease ≥12%, hemoglobin decrease of ≥4 g/dl, transfusion of ≥2 units of packed red blood cells, and retroperitoneal, gastrointestinal, or intracranial bleeding. Gastrointestinal bleeding is defined as an evidence of an upper (coffee ground emesis, endoscopy demonstrating active bleeding) or lower (melena, hematochezia, or endoscopy demonstrating an active bleeding site) gastrointestinal bleed.
The NCDR PCI Bleeding Risk Score was used to control for the underlying risk of bleeding because of baseline patient characteristics. Multivariable logistic analysis of major bleeding was performed using a logistic regression model. The baseline covariates used for the logistic regression were those within the NCDR PCI Bleeding Risk model, including age, gender, previous congestive heart failure, glomerular filtration rate, peripheral vascular disease, no previous PCI, New York Heart Association class IV congestive heart failure, acute ST-elevation myocardial infarction, non–ST-elevation myocardial infarction, and cardiogenic shock. The covariates were analyzed as a single, combined variable: NCDR PCI Bleeding Risk Score.
Statistical analysis was performed using SAS, version 9.1.3 (SAS Institute Inc., Cary, North Carolina). Continuous variables and categorical variables are expressed as mean ± SD and percentages, respectively. The Student’s t test was used to compare continuous variables, and the chi-square or Fisher’s exact test was used to compare categorical variables. A p value <0.05 was considered statistically significant. The continuous net reclassification improvement (NRI) was calculated to study the effect of adding AS as an ordinal variable to the NCDR clinical risk algorithm when predicting bleeding events. The continuous NRI calculates the net proportion for events and nonevents, where predicted risk moves in the correct direction (increases for events and decreases for nonevents).
Results
Three hundred fifty-four patients with moderate or severe AS who underwent PCI were identified. There were 122 and 232 patients with moderate and severe AS, respectively. These patients were compared with 7,572 patients who had either mild or no AS and underwent PCI.
The baseline characteristics of the 2 groups of patients can be found in Table 1 , and the differences in procedure characteristics can be found in Table 2 . The patients with moderate or severe AS were older, more likely to be female, and more likely to be white. They were also more likely to have a history of coronary bypass grafting, peripheral vascular disease, heart failure, and chronic renal insufficiency. Patients with AS were less likely, however, to be current smokers, to have a family history of coronary artery disease, and to have a myocardial infarction during the current admission. There was no difference between patients with and without AS with respect to measured LV ejection fraction, a history of coronary artery disease, hypertension, diabetes, hyperlipidemia, previous PCI, or cardiogenic shock.
| Variable | Aortic stenosis | P value | |
|---|---|---|---|
| Moderate or Severe (n=354) | Mild or None (n=7,572) | ||
| Age (years) | 79.9 ± 11.0 | 65.06 ± 12.2 | < 0.001 |
| Women | 45.5% | 35.9% | < 0.001 |
| White | 79.6% | 56.6% | < 0.001 |
| Body mass index (kg/m 2 ) | 27.8 ± 6.3 | 30.0 ± 6.8 | < 0.001 |
| Hypertension | 90.7% | 88.2% | 0.135 |
| Diabetes Mellitus | 39.8% | 39.4% | 0.892 |
| Hyperlipidemia | 87.0% | 84.4% | 0.164 |
| Chronic renal insufficiency | 37.7% | 18.0% | < 0.001 |
| On hemodialysis | 5.8% | 4.5% | 0.236 |
| Current smoker | 11.6% | 26.7% | < 0.001 |
| Prior coronary bypass grafting | 28.3% | 19.4% | < 0.001 |
| Family history of coronary artery disease | 32.5% | 40.6% | 0.002 |
| Peripheral vascular disease | 27.9% | 16.1% | < 0.001 |
| History of coronary artery disease | 48.4% | 44.6% | 0.150 |
| Percutaneous coronary intervention | 27.5% | 30.6% | 0.212 |
| Heart failure | 37.9% | 16.7% | < 0.001 |
| Heart failure class III or IV | 20.7% | 8.4% | < 0.001 |
| Left ventricular ejection fraction (%) | 0.46 ± 0.17 | 0.45 ± 0.16 | 0.846 |
| Prior myocardial infarction | 25.3% | 22.5% | 0.214 |
| Myocardial infarction this admission | 27.8% | 38.1% | < 0.001 |
| Cardiogenic shock | 2.1% | 3.5% | 0.158 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
