A38 Abstracts PCV5 NON-PERSISTENT USE OF ANTIHYPERTENSIVE DRUGS INCREASES RISK OF HOSPITALIZATIONS FOR STROKE BY NEARLY 20% Br eekv eldt-P ostma NS 1 , Siiskonen SJ 1 , Penning-van Beest FJA 1 , Erkens JA 1 ,Vincze G 2 , Falvey H 2 , Herings RMC 1 1 PHARMO Institute, Utrecht, The Netherlands, 2 Novartis Pharma AG, Basel, Switzerland OBJECTIVES: Low persistence with antihypertensive drug treat- ment (AHT) is expected to limit patient’s benefits in terms of a reduction of cardiovascular and cerebrovascular disease. This study investigated the relationship between persistence with anti- hypertensive drugs and the risk of stroke in clinical practice. METHODS: From the PHARMO Record Linkage System com- prising, among others, linked drug-dispensing and hospital records of >2 million inhabitants in The Netherlands, new users of AHT were identified in the period 1993–2002. Persistence with AHT was determined by summing the number of days of continuous treatment (gaps between dispensings of <60 days). Patients were classified as persistent if they remained on AHT for at least two years. The outcome of interest was the first hos- pital admission for stroke occurring two or more years after initiation of AHT therapy. Patients were classified as high, inter- mediate or low cardiovascular risk based on other cardiovascu- lar drug use and hospitalizations during the first two years of follow-up. RESULTS: The study included 98,485 patients of whom 16% were at high cardiovascular risk. About 50% (n = 48,548) of all patients were persistent with AHT for two years, and 2.1% (n = 2093) were hospitalized for stroke in the period of two or more years after initiation of AHT therapy. Multi- variate analyses showed that non-persistent users of AHT had a 16%–19% increased risk for stroke compared to persistent users (Low/intermediate risk group RRadj = 1.19; 95% CI: 1.07–1.32; high risk group RRadj = 1.16; 95% CI: 0.97–1.39). CONCLU- SION: In clinical practice antihypertensive drug treatment is used over too short a time interval to have maximum benefit from preventing stroke. PCV6 NON-PERSISTENT USE OF ANTIHYPERTENSIVE DRUGS INCREASES RISK OF HOSPITALIZATIONS FOR ACUTE MYOCARDIAL INFARCTION BY 10% IN PATIENTS WITH LOW OR INTERMEDIATE CARDIOVASCULAR RISK Br eekv eldt-P ostma NS 1 , Siiskonen SJ 1 , Penning-van Beest FJA 1 , Erkens JA 1 ,Vincze G 2 , Falvey H 2 , Herings RMC 1 1 PHARMO Institute, Utrecht, The Netherlands, 2 Novartis Pharma AG, Basel, Switzerland OBJECTIVES: Low persistence with antihypertensive drug treat- ment (AHT) may limit patient’s benefits in terms of a reduction of cardiovascular and cerebrovascular disease. This study in- vestigated the relationship between persistence with antihy- pertensive drugs and the risk of acute myocardial infarction (AMI) in clinical practice. METHODS: From the PHARMO record linkage system comprising, among others, linked drug- dispensing and hospital records of >2 million inhabitants in The Netherlands, new users of AHT were identified in the period 1993–2002. Persistence with AHT was determined by summing the number of days of continuous treatment (gaps between dis- pensings <60 days). Persistent patients remained on AHT for two years. The outcome of interest was the first hospital admission for AMI occurring two or more years after initiation of AHT therapy. Patients were classified as high, intermediate or low car- diovascular risk based on other cardiovascular drug use and hos- pitalizations during the first two years of follow-up. RESULTS: The study included 98,485 patients of whom 16% were at high cardiovascular risk. About 50% of all patients were persistent with AHT for two years and 1.5% was hospitalized for AMI in the period of two or more years after initiation of AHT. Multi- variate analyses showed that non-persistent use of AHT increased the risk for AMI in the low/intermediate risk group (RRadj = 1.12; 95% CI: 0.99–1.28), but not in the high risk group (RRadj = 0.90; 95% CI: 0.73–1.10). CONCLUSION: In clinical practice antihypertensive drug treatment is used over too short a time interval to have maximum benefit for preventing AMI in patients with low or intermediate cardiovascular risk. PCV7 RELATIONSHIPS BETWEEN VENOUS THROMBOEMBOLIC (VTE) PROPHYLACTIC TREATMENTS AND VTE COMPLICATIONS OR THROMBOCYTOPENIA AND AMONG VETERANS RECEIVING TOTAL HIP REPLACEMENTS Raisch D W , Campbell HM,Taylor Z Department of Veterans Affairs Cooperative Studies Program, Albuquerque, NM, USA OBJECTIVES: To compare rates of venous thromboembolic (VTE) complications and thrombocytopenia by VTE prophylac- tic treatments, among Department of Veterans Affairs (VA) patients receiving total hip replacement(THR). METHODS: From the VA national databases, we identified all THR patients between March 2003 and March 2004. Using inpatient and out- patient data; we collected demographics, diagnoses, health care utilization, and VTE prophylactic strategies for each patient. We followed patients for 1 year post-surgery, applying diagnostic codes to identify VTE complications (deep vein thrombosis (DVT)), pulmonary embolism (PE), post-thrombotic syndrome (PTS)), and thrombocytopenia. Using logistic regression; con- trolling for age, gender, obesity, congestive heart failure, and cancer; we compared VTE complications and thrombocytopenia by VTE prophylaxis (reference = enoxaparin alone). RESULTS: We found 1722 THRs with VTE prophylaxis: enoxaparin = 1005 (58.4%), warfarin = 345 (20.0%), dalteparin = 205 (11.9%), and the combination of enoxaparin with warfarin (enox/warf) = 167 (9.7%). Respectively, patients experiencing VTE complications (chi square p < 0.001) were: 26 (2.6%), 22 (6.4%), 5 (2.4%), and 34 (20.4%) or suffering thrombocytopenia (chi square p = 0.177) were: 6 (0.5%), 3 (0.9%), 1 (0.5%), and 4 (2.4%). Logis- tic regression revealed significantly greater (p < 0.001) VTE com- plications (odds ratio, 95% confidence intervals) with enox/warf (9.9, 5.6–17.2) or warfarin alone (2.6, 1.5–4.7) versus enoxa- parin alone. Significant covariates were age (p = 0.016) and cancer diagnosis (p = 0.018). Treatment with enox/warf was associated with significantly (p < 0.001) more PEs, 4.8% versus 0.5% (10.6, 3.7–33.6). Treatment with enox/warf or warfarin alone was asso- ciated with more (p < 0.001) DVTs, 17.4% versus 2.2% (9.4, 5.2–17.1) and 5.2% versus 2.2% (2.5, 1.3–4.8). There were no cases of PTS. Logistic regression results for thrombocytopenia were not significant (p = 0.174). CONCLUSION: Warfarin and enox/warf were significantly less effective VTE prophylactic strategies following THR than dalteparin or enoxaparin. Poten- tial limitations include the non-controlled, observational design, inclusion of primarily male VA patients, and constraints inherent in national VA data. PCV8 RISK OF MYOPATHY ASSOCIATED WITH THE USE OF STATINS AND POTENTIALLY INTERACTING MEDICATIONS— A RETROSPECTIVE ANALYSIS Shah S 1 , Shepherd MD 2 , Barner JC 2 , Wilson JP 2 , Lawson KA 2 1 Bayer Pharmaceuticals Corporation,West Haven, CT, USA, 2 University of Texas at Austin, Austin,TX, USA