Methods

All consecutive patients presenting to the cardiac catheterization laboratory for coronary stent implantation at the Institute of Cardiology of Rio Grande do Sul from 8:00 a.m. to 6:00 p.m. from November 2007 to March 2008 were considered for inclusion in this study. Those patients presenting during night or weekend hours, when physicians work on call, were not eligible. This institution is a dedicated center for cardiology, performing approximately 2,500 percutaneous coronary interventions and >7,000 diagnostic catheterization procedures per year. Exclusion criteria included participation in studies involving antiplatelet treatment, unsuccessful procedures, major cardiovascular events during hospitalization, and refusal to sign informed consent. The study was approved by the institutional review board. No extramural funding was used to support this work. The authors are solely responsible for the design and conduct of this study, all study analyses, drafting and editing of the report, and its final contents. Clinical, socioeconomic, and angiographic characteristics of the patients were prospectively collected and included in a dedicated database. Hypertension was considered present in patients with a medical diagnosis or those using antihypertensive drugs. Dyslipidemia was considered present in patients with a medical diagnosis or those using antilipemic drugs. Depression was considered present in patients using antidepressives or in those reporting this diagnosis as assessed by a physician.

All patients received aspirin 100 mg/day and clopidogrel 75 mg/day. A bolus dose of clopidogrel 300 mg was administered ≥6 hours before the procedure. In urgent procedures, a clopidogrel 600-mg loading dose was administered immediately before stenting. The Institute of Cardiology has protocols for patient care after stenting procedures. At hospital discharge, patients treated with bare metal stents were instructed to maintain the use of thienopyridines for ≥30 days and aspirin indefinitely. Patients treated with drug-eluting stents were instructed to maintain the use of thienopyridines for ≥1 year and aspirin indefinitely. At hospital discharge, all patients received a prescription and a detailed explanation about the importance of 30-day adherence to dual antiplatelet therapy. Stent implantation procedures followed the protocol of the Institute of Cardiology, similar to those described in the literature. Clinical aspects such as type and number of stents implanted, use of other devices, and additional pharmacology were left to the discretion of the operators.

All patients were interviewed 30 days after stent implantation to assess adherence to medical treatment. The main outcome of the study was discontinuation of thienopyridines before the 30-day follow-up assessment as reported by the patient at the 30-day visit. Those who discontinued their medication were questioned about the reasons. The Morisky questionnaire was given to all patients to assess the degree of adherence to drug therapy. This test evaluates drug adherence in an objective and standardized way, taking into account the behavior of the patient regarding use of the prescribed medicine and not just drug interruption. It is based on 4 questions that involve forgetfulness, negligence, interrupting the drug after clinical improvement, and reuse of the medicine for worsening symptoms. Adherence was classified from 0 to 4. Patients with grade 0 totally adhered to their medication; those with higher grades adhered less. In this study, the adaptation of Goldberg et al was used, which allows the patient the chance to talk about the causes of discontinuation. Delay in dose taking was considered when a medication was taken >2 hours from the time it should have been taken.

Characteristics of patients who discontinued thienopyridine treatment were compared with those who continued use until day 30. Data were analyzed with SPSS 11.0 for Windows (SPSS, Inc., Chicago, Illinois). Continuous variables were expressed as mean ± SD and categorical variables as percentage. Comparisons between groups were done using t test for continuous variables and chi-square test or Fischer’s exact test for categorical values. Sample size was calculated assuming a discontinuation rate of 20% with 3.5% variation and a 95% confidence interval and was determined to be 400 patients. A p value <0.05 was considered statistically significant.

Variables associated with discontinuation in univariate analysis or classified as predictors of discontinuation in previous studies were included in multiple logistic regression models. The model with better calibration was chosen according to the Hosmer–Lemeshow goodness-of-fit test. Based on this model, a risk score was developed for discontinuation of thienopyridine and points were assigned proportionately to the odds ratios. Continuous variables were categorized according to cutpoints that allowed the best prediction of the primary outcome. Diagnostic accuracy of the score to predict drug discontinuation was assessed through c-statistics. Diagnostic accuracy of the risk score to predict total medication adherence (defined by a Morisky score 0) was also evaluated.

Developing and testing a model on the same set of patients is likely to result in biased estimates. To estimate the impact of this, a bias-corrected c-index was generated using 1,000 bootstrapped samples by implementing the R functions Validate and Calibrate. A graphic representation of the calibration curve using the original dataset and bias-corrected samples was also generated.

Results

During the study period, 405 patients were included and clinical follow-up was assessed and completed in 400 patients (99%) an average of 30 ± 4 days after the procedure. Only 5 patients were lost from the follow-up period. Discontinuation of thienopyridine was reported by 66 patients (16.5%); 102 additional patients (25%) presented adherence problems, according to the Morisky questionnaire. Median time (25th, 75th percentiles) of thienopyridine discontinuation was 10 days (9, 16).

Clinical characteristics of study patients are listed in Table 1 . The 2 groups were similar with the exception of a trend toward a smaller percentage of diabetics and a significantly larger percentage of patients with congestive heart failure in those who discontinued thienopyridines. These patients more frequently demonstrated the following characteristics: unmarried, lack of private health insurance, and an income <2 times the minimum wage ($256.00 per month). Patients who discontinued thienopyridines were also more frequently referred from the emergency department with a clinical presentation of an acute coronary syndrome and had a lower educational level. Depressive symptoms were assessed and were similar between groups. Regarding stent implantation, 3% received drug-eluting stents and 97% received bare metal stents. This difference is due mainly to health insurance policies in Brazil, which are restrictive for drug-eluting stents. Patients who received drug-eluting stents did not report discontinuation of thienopyridines compared with a rate of 17% in those receiving bare metal stents (0 of 12 vs 66 of 388 patients, p = 0.23).

The following reasons were given for discontinuing thienopyridine therapy: 41 patients said cost of medication (when data were collected, the monthly cost of clopidogrel was equivalent to $50.00), 11 patients said they had not received sufficient information about the importance of the dual antiplatelet regimen, 10 patients received advice from another physician to stop the drug, 1 patient had an allergy, 1 reported urinary bleeding, 1 had a hemorrhagic stroke, and 1 refused to accept physician advice to take the drug. Patients who discontinued thienopyridines frequently stated that they were not aware of the importance of use compared with those who maintained treatment (50% vs 26%, p <0.001). There was no significant difference between patients who maintained thienopyridine use and those who discontinued use with regard to access to free supplies of the drug. Rate of 30-day stent thrombosis was 0.5%; there were 2 cases of probable stent thrombosis by the Academic Research Consortium definition (sudden death). One of these patients who died within 30 days from the procedure had discontinued thienopyridine before death, whereas the other patient was still using clopidogrel.

Using the Morisky questionnaire, we found that 102 patients had problems with adherence but did not discontinue the medication. Of these, 15% had better adherence (classifications 1 and 2), although they did stop taking the medication ≥1 time during the month. However, 11% demonstrated serious problems regarding adherence (classifications 3 and 4) with time delays or not taking the thienopyridine several times per week.

Table 2 presents the final multiple regression model for discontinuation of thienopyridine with the best calibration for this population according to Hosmer–Lemeshow goodness-of-fit test (chi-square 1.92, p = 0.96). Other variables included in multiple regression models but not retained in the final model were age, sex, chronic heart failure, and emergency procedures.

Table 2

Multiple logistic regression analysis of candidate variables associated with thienopyridine discontinuation (n = 400)

Characteristic	Beta	OR	95% CI	p Value
Unmarried	0.908	2.48	1.01–6.07	0.046
Lack of health insurance	1.543	4.68	1.05–21	0.04
Education (years) ⁎	−0.112	0.89	0.42–1.86	0.76
Acute coronary syndrome	0.838	2.31	1.28–4.14	0.004
Diabetes mellitus	−0.788	2.20	1.02–4.69	0.04
Salary
<2 times minimum wage	2.108	8.23	2.70–25	<0.001
2–3 times minimum wage	1.795	4.46	1.25–16	0.02
>3 times minimum wage	1.795	4.46	1.25–16	0.02
Constant	−5.709

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. Global Variability in Angina Pectoris and Its Association With Body Mass Index and Poverty
2. Sleep-Disordered Breathing in Patients With the Brugada Syndrome
3. Usefulness of Postexercise Ankle-Brachial Index to Predict All-Cause Mortality
4. Carcinoid Heart Disease Without the Carcinoid Syndrome but With Quadrivalvular Regurgitation and Unsuccessful Operative Intervention

Stay updated, free articles. Join our Telegram channel

Join