Effects of losartan vs candesartan in reducing cardiovascular events in the primary treatment of hypertensionJournal of Human Hypertension (2010) 24, 263–273 & 2010 Macmillan Publishers Limited All rights reserved 0950-9240/10 $32.00 Effects of losartan vs candesartan inreducing cardiovascular events in theprimary treatment of hypertension SE Kjeldsen1, J Sta˚lhammar2, P Hasvold3, J Bodegard3, U Olsson4 and D Russell51Department of Cardiology, Oslo University Hospital, Ulleva˚l, Oslo, Norway; 2Department of Public Healthand Caring Sciences, Uppsala University, Uppsala, Sweden; 3Medical Department, AstraZeneca AS, Oslo,Norway; 4Statisticon AB, Uppsala, Sweden and 5Department of Neurology, Oslo University Hospital,Rikshospitalet, Oslo, Norway Although angiotensin receptor blockers have different cause register. There was no difference in blood pressure receptor binding properties no comparative studies with reduction when comparing the losartan and candesartan cardiovascular disease (CVD) end points have been groups during follow-up. Compared with the losartan performed within this class of drugs. The aim of this study group, the candesartan group had a lower adjusted hazard was to test the hypothesis that there are blood pressure ratio for total CVD (0.86, 95% confidence interval (CI) 0.77– independent CVD-risk differences between losartan and 0.96, P ¼ 0.0062), heart failure (0.64, 95% CI 0.50–0.82, candesartan treatment in patients with hypertension with- P ¼ 0.0004), cardiac arrhythmias (0.80, 95% CI 0.65–0.92, out known CVD. Seventy-two primary care centres in P ¼ 0.0330), and peripheral artery disease (0.61, 95% CI Sweden were screened for patients who had been 0.41–0.91, P ¼ 0.0140). No difference in blood pressure prescribed losartan or candesartan between the years reduction was observed suggesting that other mechan- 1999 and 2007. Among the 24 943 eligible patients, 14 100 isms related to different pharmacological properties of the patients were diagnosed with hypertension and prescribed drugs may explain the divergent clinical outcomes.
losartan (n ¼ 6771) or candesartan (n ¼ 7329). Patients Journal of Human Hypertension (2010) 24, 263–273; were linked to Swedish national hospitalizations and death ; published online 5 November 2009 Keywords: epidemiology; heart failure; cardiovascular diseases; blood pressure The RAS provides the most powerful regulation of blood pressure and angiotensin II is the primary Hypertension is an important risk factor for the mediator in this system. The binding of angiotensin development of cardiovascular disease (CVD), and blood pressure lowering treatments have preventive receptor produces a number of potentially harmful effects that include increased effects.Several studies comparing antihypertensive blood pressure, progression of atherosclerosis, myo- agents have shown differences in risk reduction in cardial-, and vascular hypertrophy. Losartan was the CVD with a similar blood pressure lowering effect, first ARB and has been shown to reduce the risk of suggesting that specific pharmacological mechan- stroke and new onset diabetes compared to ateno- isms may be invThe renin-angiotensin lol.A more recently developed ARB is candesartan, system (RAS) is targeted by some of the most widely which has also shown a proven benefit in the used antihypertensive medication classes: angioten- treatment of heart failure and the prevention of sin converting enzyme inhibitors and angiotensin stroke and new onset diabetAlthough they receptor blockers (ARBs). ARBs are increasingly belong to the same class of drugs, losartan and used in the treatment of hypertension because of candesartan have important pharmacological differ- fewer side effects than angiotensin converting ences.The binding to the AT enzyme inhibitors, combined with similar blood 1 receptor is tighter and lasts longer with candesartan compared to pressure lowering abilities.
Although losartan and candesartan have, in several large randomized clinical trials, documentedeffect on CVD-risk reduction, no head-to-head Correspondence: Professor SE Kjeldsen, Department of Cardiology, comparison studies have been performed. The Oslo University Hospital, Ulleva˚l, Kirkeveien 166, N-0407 Oslo, established approach to study differences between Norway.
Received 25 July 2009; revised 23 August 2009; accepted 3 the two drugs is a randomized controlled trial.
September 2009; published online 5 November 2009 Such a trial would, however, take 5–10 years and An observational study of 14 100 patients SE Kjeldsen et al involve large resourses that are usually unavailable.
nurses had an overall correct technique.When In addition, it is difficult to justify from a health analysing the extracted data, the baseline value for political perspective. The randomized controlled blood pressure was calculated as the mean of the trials also differ from a real life clinical setting as three last available blood pressures during the time there is a selection of both physicians and patients; period 15 month before index prescription until 14 the treatment alternatives depend on the study days after index prescription. Blood pressure at 6 protocol and the study itself may compromise months was calculated as the mean of recorded patient–doctor relationshipWe therefore studied blood pressures during the time period 2 weeks–6 a real life clinical situation by electronically extract- months after index prescription. From 6 months and ing patient data from electronic primary care onwards, consecutive blood pressure values were journals and mandatory Swedish national registers calculated as the mean of recorded blood pressures for death and hospitalization.
from 0.5 years before and 0.5 years after the specific The aim of this study was to test the hypothesis that losartan and candesartan have different effects The primary composite end point consisted of on CVD-risk reduction in a real life setting with CVD morbidity, CVD mortality, and elective coron- primary hypertensive treatment.
ary revascularization procedures. CVD was definedas heart failure (International Classification ofDiseases (ICD): I50, ICD-9: 428), cardiac arrhythmias (ICD-10: I46–I48, ICD-9: 427), peripheral arterydisease (ICD-10: I70, I71, I74, ICD-9: 440, 441, The study protocol was reviewed and approved by 444), chronic ischemic heart disease (ICD-10: I20.9, the regional research ethics committee in Uppsala, I25.1, ICD-9: 413–414), myocardial infarction (ICD- Sweden and registered with ClinicalTrials.gov, 10: I21–I23, ICD-9: 410, 411, 429), stroke (ICD-10: number NCT00620178.
I61, I63–I64, G45, ICD-9: 431–434, 435), unstable Patients eligible for the study were males or angina (ICD-10: I20.0, ICD-9: 411), and coronary females, more than 17 years of age who were revascularization. Cardiovascular mortality was de- prescribed candesartan (Anatomical Therapeutic fined as death due to CVD with the above diagnoses.
Chemical (ATC) classification system: C09CA06, Diabetes was registered separately (ICD-10: E10– C09DA06) or losartan (ATC: C09CA01, C09DA01) E14, ICD-9: 250).
for hypertension in the period 1 January 1999–31 In Sweden, a patient has only one general December 2007. The first prescription found in the practitioner who follows the patient. Both nurses primary care journals within the study period was and physicians in a primary care centre have access defined as the index prescription (treatment onset).
to the electronic patient journal and are responsible Exclusion criteria were a prior history of CVD and for documenting their examinations. The patient on-going malignancy (with the exception of basal journal is continuously updated with external data cell or squamous cell carcinoma of the skin) found (that are radiologic results, ultrasound examina- in primary care journals or the Swedish Hospital tions, hospital discharge data, and laboratory data).
Discharge Register. Patients who were prescribed The selected 72 primary care centres had to use the vitamin K antagonists, digitalis glycosides, or ni- same patient journal system to be eligible for trates within 15 months prior to the index prescrip- participation.Generally, the use of electronic tion were also considered to have potential CVD and patient journals, both in private and public care, is were excluded. The remaining exclusion criterion very common in Sweden. The selected journal was the prescription of an RAS inhibitor, other than system was the most widely used patient journal the study medications within 1 week after index system, having approximately 20 000 users, 57% of all primary care centres in Sweden. No formal In Sweden, blood pressure measurements are randomization of the primary care centres was performed according to standardized methods, conducted. However, an effort was made in the using the manual Korotkoff method or automatic recruitment of study sites to ensure a representative measurements. Sweden has extensive use of well- selection of primary care centres. Selection of study calibrated blood pressure equipment and measure- sites was based on the following criteria to ensure a ments are mostly performed by public health nurses.
representative sample of the Swedish population: a The patients are told to avoid coffee and tobacco mix of rural and urban areas; public and private care 30 min before the examination. After 5 min rest in providers; small, midsized, and large primary care either the supine or sitting position an appropriate centres. The use of ARBs was not considered when sized cuff is placed on the over arm. Heart rate is selecting study sites, and the centres were included measured for 1 min before blood pressure measure- by a written invitation. The final selection of centres ment. Communication with the patient is normally for the study included 39 public primary care not recommended. When several readings are centres located in 5 of the 21 Swedish county performed, the calculated mean is recorded. A councils and 33 private care centres located in 7 of detailed report of blood pressure readings in the the counties. With regards to size, 13 (18%) small Swedish primary care concluded that public health clinics, 12 (17%) midsize clinics, and 47 (65%) large Journal of Human Hypertension
An observational study of 14 100 patientsSE Kjeldsen et al clinics were recruited. With rural areas defined as tivity analysis, only discontinuation before a CVD communities inhabiting o15 000 people, 36 of the event could be used. The patient numbers therefore sites were considered rural and 36 sites were differs slightly from the discontinuation reported for defined as urban. This sample represents 4.7% of the total observation time. An HR that remains the total number of public care centres and 14.7% of similar, compared to the primary survival models, the total number of private care centres in Sweden.
will support the main results. By solving the Age or other characteristics of the physicians was (1880x/6771x ¼ 1819/7329), not considered in the selection.
found to be 265. All statistical analyses were Patient data were extracted from primary care performed using R version centre servers using a software system, which hasalso been used in earlier studiThe programscanned all patient journals at the participating primary care centres and identified all patients Among 24 943 eligible patients; 14 100 (56.5%) who were prescribed losartan or candesartan. The were included in the study The number patient's social security number was replaced with of discontinued patients was significantly higher in a study ID when processing the data. Data from 100 the losartan group compared to the candesartan patient journals at 25 primary care centres were manually compared with the study database to Po0.0001). Patients who were initially treated with ensure that the software system had extracted losartan were also more frequently converted to correct patient data. No discrepancies were found.
another RAS inhibitor compared with candesartan- The date of admission and discharge diagnoses treated patients 13.9% (n ¼ 939) vs 10.8% (n ¼ 788, and causes of death were collected from the Po0.0001) during the study. Other reasons for Swedish Hospital Discharge and Cause of Death discontinuations were death (2.3% (n ¼ 155) vs RegistThe merging of data from primary care 2.1% (n ¼ 156, P ¼ 0.1588), in the losartan group and hospital registers was performed by the Swed- and the candesartan group, respectively) or cessa- ish National Board of Health and Welfare.
tion of study drug prescription (15.2% (n ¼ 1029] vs14.6% (n ¼ 1071, P ¼ 0.3768), in the losartan groupand the candesartan group, respectively).
Statistical methods To assess the effects of the difference in disconti- The data were processed and analysed at an nuation, 265 discontinued subjects from the losartan independent statistical contract company, and the group were omitted to obtain the proportion of study database was managed by the Department of discontinued subjects equal to the candesartan Public Health and Caring Sciences, Uppsala Uni- group, and the survival models were re-run. The versity, Uppsala, Sweden.
265 discontinued subjects were selected in three The included patients were only eligible for the different ways: subjects with shortest exposure time, analysis as long as they continued treatment with subjects with longest exposure time, and randomly the losartan or candesartan. The observation period in 1000 repeated analyses. None of these analyses ended on the date when the patient died, started a changed the conclusions as compared with the new RAS inhibiting drug or until the last valid day primary survival model (data on file).
of the index prescription.
The proportion of included patients per year, from All descriptive data are given as mean or propor- 1999 to 2007, was similar when comparing the tion. The difference in continuous and categorical losartan and candesartan group (data on file). All 72 data was analysed with t-test and w2 test, respec- primary care centres prescribed both losartan tively. We used proportional hazards regression to and candesartan, although in various ratios. The compute hazard ratio (HR) and 95% confidenceinterval (CI). In the primary survival model, adjust-ments were made for age, gender, diabetes, andprescription index year. Additional adjustments in 24943 patients started prescription of losartan (13001) or candesartan (11942) from 1999 to 2007 the primary survival model were considered assensitivity analyses. Kaplan–Meier curves were usedto illustrate risk development during follow-up. If one 10843 patients were excluded:• 5792 (44.6%) losartan and 4144 (34.7%) candesartan patients with a patient had several end points, only the first was used history of cardiovascular disease and/or prescription of warfarin/digitalis/nitrates before index prescription in the survival model. Although the maximum • 386 (3.2%) losartan and 379 (2.9%) candesartan patients with observation period was 108 months, the x axis for malignancy• Prescribed another RAAS* inhibitor in the first week after index the plots was truncated at 96 months because of low prescription, losartan 59 (0.5%) and candesartan 83 (0.7%) number of observations after this time.
To assess the difference in discontinuation we removed x number of discontinued subjects fromthe losartan group, which gave a similar proportion 6771 (52.1%) losartan patients 7329 (61.4%) candesartan patients of discontinued subjects as that in the candesartangroup (that is 24.8%). When performing the sensi- Figure 1 Patient flow.
Journal of Human Hypertension
An observational study of 14 100 patients SE Kjeldsen et al Table 1 Baseline data from 14 100 hypertensive patients without previous cardiovascular disease Body mass index (kg/m2) Systolic blood pressure Diastolic blood pressure Total cholesterol (mmol l1) LDL cholesterol (mmol l1) HDL cholesterol (mmol l1) Triglycerides (mmol l1) Glucose (mmol l1) Serum creatinine (mmol l1) Potassium (mmol l1) Calcium channel blockersa, Can. 5860 4011 3908 2558 1703 1228 837 481 224 b-blockers, n (%) 80.0 62.6 70.2 71.9 71.1 71.5 73.6 70.2 72.3 Oral glucose lowering Los. 5370 3662 3740 2510 1663 1172 792 453 236 79.3 61.7 70.8 73.9 72.4 71.5 67.1 71.6 72.2 Antithrombotics, n (%) Blood pressure during follow-up. *No, number of Angiotensin receptor patients with blood pressure readings; w, per cent blood pressure readings among patients at risk; Los, losartan; Can, candesartan.
Angiotensin converting enzyme inhibitorsb, n (%) end point was lower in the candesartan group Numbers in brackets represent standard deviation, where no other compared to the losartan group and the description is given.
adjusted HR was 0.86 (95% CI 0.77–0.96, P ¼ 0.0062) bDiscontinued treatment before index prescription.
illustrates the risk development among six separate end points. The cumulative incidenceof heart failure, cardiac arrhythmias, and peripheral prescription ratio for losartan/candesartan had a artery disease was lower with candesartan than linear range from 7 to 85% among the centres.
losartan panels a, b, and c). Compared to The losartan group were older ( þ 0.7 years), losartan, the adjusted HR was lower for heart failure had lower systolic and diastolic blood pressu- 0.64 (95% CI 0.50–0.82, P ¼ 0.0004), cardiac ar- res (1/1 mm Hg), had higher blood glucose rhythmias 0.80 (95% CI 0.65–0.98, P ¼ 0.0330), and ( þ 0.1 mmol l1), had higher HbA1c ( þ 0.1%), and peripheral artery disease 0.61 (95% CI 0.41–0.91, had a higher prevalence of diabetes ( þ 2.8%), were P ¼ 0.0140) in the candesartan group less frequently treated with thiazides (2.3%) and Cardiac arrhythmias were mainly because of atrial b-blockers (2.0%) and more frequently treated with fibrillation (n ¼ 193, 91.9%), which had a separate glucose lowering drugs ( þ 1.7%), statins ( þ 1.3%), adjusted HR of 0.77 (95% CI 0.62–0.95, P ¼ 0.0170).
and antithrombotics ( þ 0.8%) compared with the Chronic ischemic heart disease, myocardial in- candesartan group ( farction, and stroke showed similar cumulative Some blood pressure recordings were absent at all incidence in both groups panels d, e, time points. shows the similar blood and f, respectively). The losartan group showed a pressure levels that were recorded during the small non-significant, increased incidence of the follow-up, and also the number of missing values.
following events: chronic ischemic heart disease, The frequency of blood pressure recordings was myocardial infarction, stroke, hospitalization for equal in both groups. When calculating 95% CIs for unstable angina, elective coronary revascularization, the blood pressures values at each time point, no cardiovascular mortality, total mortality, and new significant differences between the losartan and onset diabetes compared with the candesartan group candesartan group were observed (data on file).
No differences in risk for these events During the study (median follow-up 2.0 years, were found in proportional hazards regression maximal follow-up 9.0 years, and 36 339 patient models. Chronic ischemic heart disease was the years), 676 CVD events occurred in the losartan exception and had an unadjusted lower risk in group, and 575 in the candesartan group the candesartan group (0.80 (95% CI 0.66–0.99, The cumulative incidence of the primary composite P ¼ 0.0350)) compared with the losartan group.
Journal of Human Hypertension
An observational study of 14 100 patientsSE Kjeldsen et al Table 2 Clinical outcomes obtained from primary care journals and Swedish national discharge and death registers n (%) (n ¼ 6771) n (%) (n ¼ 7329) Primary composite end pointb,c 0.79 (0.71–0.89) 0.86 (0.77–0.96) 0.58 (0.45–0.74) 0.64 (0.50–0.82) Cardiac arrhythmiasc 0.73 (0.60–0.90) 0.80 (0.65–0.98) Peripheral artery diseasec,d 0.61 (0.38–0.83) 0.61 (0.41–0.91) Chronic ischemic heart disease 0.80 (0.66–0.99) 0.86 (0.70–1.05) Myocardial infarctionc 0.85 (0.66–1.08) 0.93 (0.73–1.19) 0.88 (0.70–1.10) 0.95 (0.76–1.19) Hospitalization for unstable angina 0.77 (0.43–1.36) 0.80 (0.45–1.42) Elective coronary revascularization 0.74 (0.37–1.48) 0.78 (0.39–1.58) Cardiovascular mortality 0.83 (0.60–1.16) 0.93 (0.66–1.29) 0.96 (0.77–1.20) 1.06 (0.85–1.32) New onset diabetes 0.92 (0.79–1.08) 0.90 (0.77–1.05) aAdjusted for age, gender, diabetes, and index-year.
bPrimary composite end point consists of heart failure, cardiac arrhythmias, peripheral artery disease, chronic ischemic heart disease, myocardialinfarction, stroke, elective coronary revascularization, hospitalization for unstable angina, or cardiovascular mortality.
cIncludes fatal events.
dIncludes aortic aneurysms.
eIschemic, haemorrhagic stroke, and transient ischemic attacks.
Primary composite endpoint
Cumulative incidence (%) Adjusted risk reduction 14.4% p=0.0062 Unadjusted risk reduction 20.6% p<0.0001 Number at risk
6771 5812 4548 3913 3188 2591 2090 1738 1458 1169 7329 6291 4860 4091 3385 2742 2242 1875 1580 1302 1021 Figure 3 Kaplan–Meier curves of primary composite end point.
shows that the use of both losartan and were up-titrated to 16 and 16/12.5 mg. Within the candesartan was according to prescribing recom- first 6 months losartan was up-titrated by 6.1% and mendations for hypertension. The losartan group candesartan by 13.2%.
generally started with 50 mg and up-titrated to fixed The frequency of fixed combination tablets combination 50/12.5 mg. The candesartan group, on (ARB þ hydrochlorothiazide) in the losartan group the other side, mainly started with 4 and 8 mg in rose from 24.7 to 60.5% ( þ 35.9%) compared with patients with higher blood pressure at baseline, þ 1/ an increase from 13.4 to 33.4% ( þ 20.0%) in the þ 1 mm Hg. Within the first 6 months these patients candesartan group.
Journal of Human Hypertension
An observational study of 14 100 patients SE Kjeldsen et al Peripheral artery disease
Cumulative Incidence (%) Cumulative Incidence (%) Cumulative Incidence (%) Chronic ischemic heart disease
Cumulative Incidence (%) Cumulative Incidence (%) Cumulative Incidence (%) Figure 4 Kaplan–Meier curves for separate end points. Los, losartan; Can, candesartan.
shows the use of other antihypertensive treated with any glucose lowering drug (n ¼ 11 596) medications (thiazides, calcium channel blockers, and were analysed separately. When adjusting for age, b-blockers) increased in both groups during follow-up.
gender, and index year, the risk of CVD remained The use of thiazides (both separate and in fixed similar (HR 0.86, 95% CI 0.76–0.98, P ¼ 0.0210) in combination tablets) was more frequent in the losartan patients without diabetes.
group compared to the candesartan group. Slightlymore prescriptions of oral glucose lowering drugs atbaseline and during follow-up were observed in the losartan group compared to the candesartan group.
In this study, 14 100 patients in a primary preven-tion setting were treated with either losartan(n ¼ 6771) or candesartan (n ¼ 7329). They fulfilled Sensitivity analyses the inclusion criteria of being hypertensive and When additionally adjusting for lipid lowering without previously known CVD. Patients were drugs, thiazides, b-blockers, and antithrombotics followed for a median of 2 years with a total of the risk of CVD in the candesartan group remained 36 339 patient years, using the combination of significantly lower (HR 0.84, 95% CI 0.75–0.94, patient journal data in 72 primary care centres and P ¼ 0.0030). When adding systolic blood pressure to the mandatory Swedish national hospital discharge the adjustments (age, gender, diabetes, index year, and cause of death registers. There was a substantial lipid lowering drugs, thiazides, b-blockers, and blood pressure reduction after treatment with both antithrombotics), the number of patients available drugs, with no difference in blood pressure between for a survival analysis was reduced by 20% the two treatment groups during the follow-up (n ¼ 11230) because of absent blood pressure values.
period. Our results showed that candesartan was However, despite this loss of patients in the survival more effective than losartan in reducing the risk of model, the risk remained similar (HR 0.87, 95% CI primary composite end point, HR 0.86 (95% CI 0.77–0.98, P ¼ 0.0250) compared with the primary 0.77–96, P ¼ 0.0062).
We used a method that uses the strengths of To further evaluate potential effects of differences electronic patient registers, which cover large at baseline, patients without a history of diabetes or patient groups, making them increasingly accessible Journal of Human Hypertension
An observational study of 14 100 patientsSE Kjeldsen et al % Losartan dose titration
% Candesartan dose titration
Figure 5 Changes of losartan and candesartan doses during follow-up. Losartan 12.5 mg and losartan/hydrochlorothiazide 100/12.5 mgwere used in very small numbers (o2% respectively) and were therefore excluded in the figure.
% Calcium channel blockers % Oral glucose lowering drugs Concomitant medications during 8 years of follow up. *Thiazides (ATC C03A A, C03A B, C09D A01, C09D A06), wdihydropyridine derivates (ATC C08 CA).
Journal of Human Hypertension
An observational study of 14 100 patients SE Kjeldsen et al to real life observational studies. However, though Patients were also scanned for hospitalizations this method provides large number of patients in caused by CVD in the Swedish Hospital Discharge short time and at a rather low cost, some limitations Register, which has had mandatory registration are inevitable.
since 1984.The combination of these two search One of the limitations with this method is the techniques should therefore have lowered the risk of absence of recorded baseline data in the electronic skewed CVD prevalence at baseline as all patients patient primary care journal. Blood pressure record- with documented CVD in the primary care journals, ings were only registered in 80% of all the patients hospital records, or indicated by drug prescriptions at baseline, leaving 20% unaccounted for, which were excluded. Thus, we feel that we have reason- was similar in both groups. The candesartan group able well excluded the risk of confounding by had a slightly higher baseline blood pressure indication. The increased prescriptions of antith- compared to the losartan group. When adjusting rombotics and statins at baseline in the losartan for blood pressure we lost 20% of the patient group could have also signalled a systematic failure material in the survival model, but the results in CVD detection before inclusion. However, having remained similar. This type of large register study more diabetic patients in the losartan group may will always be associated with incomplete data.
have translated into more prescriptions of drugs However, the large number of patients may com- used in the primary prevention of CVD. Further- pensate for this weakness.
more, the HR for CVD was unchanged compared to When including patients over a longer time span, the primary survival model when additionally the possibility of variations in the patient handling adjustments were made for baseline differences in may be a confounder. For example, an important lipid lowering drugs, thiazides, b-blockers, and potential confounding factor could have been varia- tion in hypertensive treatment, favouring inclusion The study had a follow-up time of median 2.0 either in the losartan or candesartan group. The years, and 47% of the patients were followed for annual frequency of inclusion to the losartan or 2 years. Compared to the LIFE study (93% over candesartan group from 1999 to 2007 was, however, 2 years), this follow-up intensity may seem low and similar. This suggests that there was a similar stresses the need of large patient groups when assessment of hypertensive patients with no differ- performing a register study with this method.The ences between groups regarding the time point of explanation for this drop in patient numbers in our study could partly be due to the lack of adherence to Some confounders are difficult to measure and the primary treatment of hypertension (approxi- therefore not usable in a survival model. These mately 30%), which was also described in other real confounders may have changed the baseline risk life studiRandomized controlled trials have among the patients (that are the recording of also had a similar follow-up time as those in our diagnoses, new indications, reimbursement, guide- study. In the JUPITER trial the median follow-up lines, marketing, and shift in the ICD system or time was 1.9 years and only 43% of the patients continuous updating of the software for electronic were followed for 2 years.
patient journal systems). Adjusting for all these Our main finding was a risk reduction in CVD potential confounders is difficult and we, therefore, mainly determined by heart failure, cardiac ar- made adjustments for the year of index prescription rhythmias, and peripheral artery disease. The risk to minimize the possible effects of temporal changes of heart failure was reduced by 36% in the with regard to the above parameters.
candesartan group. Heart failure may be a result of Registry data are to a certain degree associated left ventricular hypertrophy and/or myocardial with absent laboratory data (blood samples and infarction, and is therefore responsive to the blood pressure measurements) at specific time myocardial remodelling properties of a drug. Both intervals. However, in our study the frequency of losartan and candesartan have beneficial remodel- laboratory data did not differ markedly between the ling effects on hypertrophic myocardium.The two groups, suggesting similar need of medical risk of cardiac arrhythmias in the candesartan group attention at baseline.
was 20% lower and may be explained by the lower We have adjusted for differences in observed CVD incidence of heart failure. Alternatively, the lower risk between the two groups at inclusion. However, incidence of atrial fibrillation may explain the differences in undetected risk at inclusion may also lower risk of heart failureWe believe that the be of concern. When excluding patients because of lower risk of new heart failure and cardiac arrhyth- earlier CVD, more patients (9.9%) were excluded mias in the candesartan group may be due to a from the losartan group. Consequently, the risk of more potent inhibition of AT1 receptors as this undetected CVD may have been higher in this group, occurred at the same level of blood pressure control requiring a reliable method of finding CVD before and in patient groups with similar baseline char- inclusion. To exclude patients with earlier CVD, patient journals in primary care were searched for Furthermore, candesartan has reported positive CVD diagnoses on average 5.8 years before inclusion effects on risk markers such as high oxidative stress and drugs associated with CVD before inclusion.
and increased coagulability compared to losartan, Journal of Human Hypertension
An observational study of 14 100 patientsSE Kjeldsen et al which may explain the lower risk for CVD compli- have potentially reduced CVD risk in this group.
cations.Interestingly, the risk of peripheral artery Other concomitant antihypertensive treatments disease was lower (39%) in the candesartan group (calcium channel blocker and b-blocker) did not compared to the losartan group. A possibility is that differ during the observation period. We have the peripheral circulation may be reduced in therefore no reason to believe that the use of patients with heart failure leading to earlier is- additional antihypertensive drugs could have chemic symptoms in the lower limbs.
influenced our results.
There was no difference between the two treatment groups in blood pressure throughoutthe study, with an average blood pressure ofapproximately 145/85 mm Hg in both groups.
The blood pressures in this study are office We believe that our study method provides a new readings and a potential difference in 24 h blood tool that can be used to study existing treatments, pressure has not been investigated and can there- providing rapid results at a low cost. However, it fore not be excluded. Slightly elevated baseline requires the wide use of similar electronic patient blood pressure in the candesartan group suggests journal systems in primary care and a long tradition that this group may have had a higher baseline with nationwide hospitalization and cause of death CVD risk. However, after adjustment for baseline registers. Sweden offers this combination and pro- systolic blood pressure, only small effects in the vides the opportunity to study differences between survival analyses were seen. The similar blood treatments, not feasible to assess in randomized pressure in both groups after inclusion may clinical trials.
explain why we did not observe any differences The results of this study suggest that there is a in the risk of stroke. The LIFE study, which blood pressure independent risk reduction in CVD compared losartan and atenolol, did show a blood with candesartan compared to losartan in the pressure independent reduction in stroke risk in primary treatment of hypertension. This suggests favour of losartan, but atenolol may be an inferior that pharmacological differences within the ARB drug for stroke prevention.Our results suggest class may translate into important clinical effects.
that candesartan and losartan have similar effects Furthermore, the study of a real life situation may on stroke prevention.
provide an additional method that can be used to Up-titration of the ARB dose from the index assess existing and future treatments.
prescription was somewhat higher in the candesar-tan group compared to the losartan group, whichmay be seen in light of the different start dosagealternatives and the next dose step. The losartangroup was generally started with ARB only and up-titrated by adding thiazides, without changing the What is known about the topic ARB dose. The candesartan group, on the other side, K Angiotensin receptor blockers have different pharmacological properties. Losartan and candesartan have was mainly started with lower ARB dosages in significant differences regarding receptor binding when patients with significantly higher blood pressure at baseline. Within the first 6 months those patients K Differences in blood pressure reduction have been reported were up-titrated by both increasing ARB dosage and between However, no study has been performed toassess possible differences in risk of cardiovascular disease adding thiazides. The initial ARB dosage was within this class of drugs.
therefore relatively lower during the first 6 months K The use of electronic patient journals increases continuou- in the candesartan group compared with the sly with time, both in primary-, and hospital care. Those losartan group. The ARB use was therefore consid- large databases are accessible and are increasingly used for ered reasonably comparable in the two groups after epidemiological studies, also in other What this study adds In this study, the initial and continued use of K The choice between two different ARBs (losartan and thiazides was more frequent in the losartan group candesartan) did not seem to affect the blood pressure levels compared to the candesartan group. The increased during treatment in a real life clinical setting. However,increased use of thiazides in the losartan group was observed use of thiazides in the losartan group support to achieve equal blood pressure reduction.
reports of an improved blood pressure lowering K When comparing the hypertensive primary treatment with effect with candesartan compared with losartan two of the most used ARBs, losartan vs candesartan, when given as monotherapy.Recently, the differences in the risk of cardiovascular disease was LIFE study group reported that concomitant thia- observed. The candesartan group had lower risk ofcardiovascular disease compared with the losartan group.
zide therapy was associated with reduced cardio- K In countries with nationwide use of electronic patient vascular morbidity and mortality, independent of journals, it is possible to combine separate patient databases blood pressure and electrocardiographic evidence from primary-, hospital care, and cause of death register.
of left ventricular hypertrophy.Consequently, the Assessment of established treatment in a real life primarycare setting is possible to be performed rapid and rather use of more concomitant thiazide treatment in inexpensive, using hard disease events as end points.
patients treated with losartan should therefore Journal of Human Hypertension
An observational study of 14 100 patients SE Kjeldsen et al Conflict of interest Prognosis in the Elderly (SCOPE): principal results ofa randomized double-blind intervention trial. J Hyper- The study was designed by a steering committee tens 2003; 21(5): 875–886.
(authors of this paper) and AstraZeneca. The steering 9 Lindholm LH, Persson M, Alaupovic P, Carlberg B, committee had full access to all data and had full Svensson A, Samuelsson O. Metabolic outcome during independence to interpret them. Professor David 1 year in newly detected hypertensives: results of the Russell, Professor Sverre E Kjeldsen, and Jan Sta˚l- Antihypertensive Treatment and Lipid Profile in a hammar did not receive any financial compensation North of Sweden Efficacy Evaluation (ALPINE study).
for their work in this study and do not declare any J Hypertens 2003; 21(8): 1563–1574.
10 Van Liefde I, Vauquelin G. Sartan-AT1 receptor conflict of interest. Johan Bodegard and Pa˚l Hasvold interactions: in vitro evidence for insurmountable are employed by AstraZeneca and participated as antagonism and inverse agonism. Mol Cell Endocrinol non-voting members in the steering committee. Urban 2009; 302(2): 237–243.
Olsson was responsible for database managing and is 11 Fabiani ME, Dinh DT, Nassis L, Casley DJ, Johnston CI.
employed by an independent statistical company In vivo inhibition of angiotensin receptors in the who were invoiced by AstraZeneca.
rat kidney by candesartan cilexetil: a comparisonwith 12 Ojima M, Inada Y, Shibouta Y, Wada T, Sanada T, Kubo K et al. Candesartan (CV-11974) dissociates slowlyfrom the angiotensin AT1 receptor. Eur J Pharmacol We are very thankful for the ideas, encouragement, 1997; 319(1): 137–146.
and scientific support provided by Anders Ljunggren, 13 Stanley K. Design of randomized controlled trials.
AstraZeneca Mo¨lndal. We thank Bo Lidman, Pygargus Circulation 2007; 115(9): 1164–1169.
AB, for his engagement in the data collection. Sven 14 Nallamothu BK, Hayward RA, Bates ER. Beyond the Olof Jansson was involved in designing the study, randomized clinical trial: the role of effectiveness protocol writing, and preparation of the paper. The studies in evaluating cardiovascular therapies. Circu- study was funded by AstraZeneca AS Norway.
lation 2008; 118(12): 1294–1303.
15 Bjørkenstam C. The Causes of Death Register. 2008 1 Lawes CM, Vander Hoorn S, Rodgers A. Global burden 16 Edberg A, Forsberg L, Jacobsson A, Nyqvist K. The of blood-pressure-related disease, 2001. Lancet 2008; National Patient Register. 2008 (cited; available from: 2 Wing LM, Reid CM, Ryan P, Beilin LJ, Brown MA, Jennings GL et al. A comparison of outcomes with angiotensin-converting—enzyme inhibitors and diure- 17 Drevenhorn E, Hakansson A, Petersson K. Blood tics for hypertension in the elderly. N Engl J Med 2003; pressure measurement—an observational study of 348(7): 583–592.
21 public health nurses. J Clin Nurs 2001; 10(2): 3 The ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. Major out- 18 Profdoc Software. 2008 (cited; Medical Office) Available comes in high-risk hypertensive patients randomized 19 Lidman B, Linder R. Working Model. 2007 (cited; calcium channel blocker vs diuretic: The Antihyper- tensive and Lipid-Lowering Treatment to Prevent 20 Lindgren P, Borgstrom F, Stalhammar J, Alemao E, Yin Heart Attack Trial (ALLHAT). JAMA 2002; 288(23): DD, Jonsson L. Association between achieving treat- ment goals for lipid-lowering and cardiovascular 4 Dahlof B, Devereux RB, Kjeldsen SE, Julius S, Beevers G, events in real clinical practice. Eur J Cardiovasc Prev de Faire U et al. Cardiovascular morbidity and mortality Rehabil 2005; 12(6): 530–534.
in the Losartan Intervention For Endpoint reduction in 21 Ringborg A, Martinell M, Stalhammar J, Yin DD, hypertension study (LIFE): a randomised trial against Lindgren P. Resource use and costs of type 2 diabetes atenolol. Lancet 2002; 359(9311): 995–1003.
in Sweden—estimates from population-based register 5 Gavras H. Update on the clinical pharmacology of cande- data. Int J Clin Pract 2008; 62(5): 708–716.
sartan cilexetil. Am J Hypertens 2000; 13(1 Pt 2): 25S–30S.
22 Ringborg A, Lindgren P, Martinell M, Yin DD, Schon S, 6 Yusuf S, Teo KK, Pogue J, Dyal L, Copland I, Stalhammar J. Prevalence and incidence of Type 2 Schumacher H et al. Telmisartan, ramipril, or both in diabetes and its complications 1996–2003—estimates patients at high risk for vascular events. N Engl J Med from a Swedish population-based study. Diabet Med 2008; 358(15): 1547–1559.
2008; 25(10): 1178–1186.
7 Young JB, Dunlap ME, Pfeffer MA, Probstfield JL, Cohen-Solal A, Dietz R et al. Mortality and morbidity for Statistical Computing. Fountation for Statistical reduction with Candesartan in patients with chronic Computing: Vienna Austria, 2008.
heart failure and left ventricular systolic dysfunction: 24 Corrao G, Zambon A, Parodi A, Poluzzi E, Baldi I, results of the CHARM low-left ventricular ejection Merlino L et al. Discontinuation of and changes in fraction trials. Circulation 2004; 110(17): 2618–2626.
drug therapy for hypertension among newly-treated 8 Lithell H, Hansson L, Skoog I, Elmfeldt D, Hofman A, patients: a population-based study in Italy. J Hypertens Olofsson B et al. The Study on Cognition and 2008; 26(4): 819–824.
Journal of Human Hypertension
An observational study of 14 100 patientsSE Kjeldsen et al 25 Ridker PM, Danielson E, Fonseca FA, Genest J, Gotto 30 Andersson OK, Neldam S. The antihypertensive effect AM Jr, Kastelein JJ et al. Rosuvastatin to prevent and tolerability of candesartan cilexetil, a new genera- vascular events in men and women with elevated tion angiotensin II antagonist, in comparison with C-reactive protein. N Engl J Med 2008; 359(21): 2195–2207.
losartan. Blood Press 1998; 7(1): 53–59.
26 McMurray JJ, Ostergren J, Swedberg K, Granger CB, 31 Lacourciere Y, Asmar R. A comparison of the efficacy Held P, Michelson EL et al. Effects of candesartan and duration of action of candesartan cilexetil and in patients with chronic heart failure and reduced losartan as assessed by clinic and ambulatory blood left-ventricular systolic function taking angioten- pressure after a missed dose, in truly hypertensive sin-converting-enzyme inhibitors: the CHARM-added patients: a placebo-controlled, forced titration study.
trial. Lancet 2003; 362(9386): 767–771.
Candesartan/losartan study investigators. Am J Hyper- 27 Aksnes TA, Schmieder RE, Kjeldsen SE, Ghani S, tens 1999; 12(12 Pt 1–2): 1181–1187.
Hua TA, Julius S. Impact of new-onset diabetes 32 Okin PM, Devereux RB, Hille DA, Kjeldsen S, mellitus on development of atrial fibrillation and heart Lindholm LH, Edelman JM et al. Concomitant hydro- failure in high-risk hypertension (from the VALUE chlorothiazide therapy in hypertensive patients is trial). Am J Cardiol 2008; 101(5): 634–638.
associated with reduced cardiovascular morbidity 28 Koh KK, Han SH, Chung WJ, Ahn JY, Jin DK, Kim HS and mortality independent of blood pressure and et al. Comparison of effects of losartan, irbesartan, and electrocardiographic left ventricular hypertrophy: the candesartan on flow-mediated brachial artery dilation LIFE study. Circulation 2008; 118(18): 886.
and on inflammatory and thrombolytic markers inpatients with systemic hypertension. Am J Cardiol2004; 93(11): 1432–1435, A1410.
This work is licensed under the Creative 29 Lindholm LH, Carlberg B, Samuelsson O. Should beta blockers remain first choice in the treatment No Derivative Works 3.0 Licence. To view a copy of primary hypertension? A meta-analysis. Lancet of this licence, visit http://creativecommons.org/ 2005; 366(9496): 1545–1553.
Journal of Human Hypertension
SCHERING CORP. v. GENEVA PHARMACEUTICALS Cite as 339 F.3d 1373 (Fed. Cir. 2003) the parties intended that all files that were 1. Federal Courts O766 officially kept would be purged as agreed. Grant of summary judgment is re- viewed without deference. 2. Patents O72(1) Patent is invalid for anticipation if sin-
Physical Therapy Rehabilitation Following TLIF A Case Series Approach By Michael R. Noonan P.T., D.P.T November 18, 2005 The spinal fusion procedure was introduced as a treatment option for chronic LBP nearly 70 years ago; however the literature reveals divergent opinions about when fusion is indicated and how it should be performed. Furthermore, the significance of the role of postoperative rehabilitation following spinal fusion may be underestimated and there exists no consensus on the design of a program specific for rehabilitation (1-Christensen FB 2004).