The Role of Renin—Angiotensin System Blockade in the Management of Hypertension Associated With the Cardiometabolic Syndrome

Authors

  • Arya M. Sharma MD, PhD,

    1. From McMaster University, Hamilton, Ontario, Canada;1 and Franz Volhard Clinical Research Center, University Clinic Charité Campus Buch, Humboldt University, Berlin, Germany2
    Search for more papers by this author
  • and 1 Stefan Engeli MD 2

    1. From McMaster University, Hamilton, Ontario, Canada;1 and Franz Volhard Clinical Research Center, University Clinic Charité Campus Buch, Humboldt University, Berlin, Germany2
    Search for more papers by this author

Arya M. Sharma, MD, PhD, Canada Research Chair in Cardiovascular Obesity Research and Management, Michael G. deGroote Medical School, McMaster University, 1280 Main Street West, Hamilton, Ontario, Canada L8S 4L8
E-mail: sharma@ccc.mcmaster.ca

Abstract

The mounting epidemic of overweight and obesity has made understanding the relationship between excess weight and associated comorbidities more urgent. Obesity is one of the strongest predictors of the development of hypertension and is an independent risk factor for cardiovascular disease, renal disease, and diabetes mellitus. The concomitant presence of obesity and hypertension, as commonly occurs in the cardiometabolic syndrome, magnifies the risk for cardiovascular and renal disease. The term “obesity—hypertension” thus serves to underscore the link between these two deleterious conditions and to emphasize the imperative for clinical intervention. Adipose tissue is now known to produce hormones and cytokines that promote inflammation, lipid accumulation, and insulin resistance. In addition, adipose tissue contains all the components of the renin—angiotensin system (RAS), which is upregulated in the presence of obesity. Evidence implicates activation of the systemic and adipose tissue RAS, as well as the sympathetic nervous system, as key obesity-related mechanisms of hypertension and other components of the cardiometabolic syndrome. RAS blockade therefore becomes a potential therapeutic strategy in patients with obesity-related hypertension and in persons with the cardiometabolic syndrome. Clinical trials of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers conducted in predominantly overweight/obese populations have demonstrated significant reductions in cardiovascular and renal disease risk among a range of at-risk patients. RAS blockade also is associated with a reduced risk of new-onset diabetes compared with other classes of antihypertensive therapy. Randomized, controlled trials conducted specifically in patients with obesity and hypertension are needed to determine the optimal therapeutic approach for these patients.

An increasing prevalence of hypertension has occurred in conjunction with a dramatic increase in the prevalence of overweight and obesity in the United States and worldwide.1–4 Data from the first National Health Examination Survey and the National Health and Nutrition Examination Surveys (NHANES) show that the frequency of obesity—measured by body mass index (BMI) or abdominal adiposity—more than doubled between 1960 and 2000.4 According to NHANES data for 1999–2000, almost two thirds of US adults are overweight (defined as BMI 25–29.9 kg/m2), and of these almost one third are obese (BMI ≥30 kg/m2).5,6 Excess weight is one of the strongest predictors of the development of hypertension,7 which affects approximately 50 million persons in the United States and approximately one billion worldwide.8 The prevalence of elevated blood pressure (BP) (≥140/90 mm Hg) increases progressively with increasing BMI among men and women, from 15% among persons with BMI <25 kg/m2 to approximately 40% among those with BMI ≥30 kg/m2,9 The likelihood of having hypertension is nearly nine times greater among men and 10 times greater among women with BMI ≥30 kg/m2 vs. <25 kg/m29 Despite well-recognized links to stroke, cardiovascular (CV) disease, heart failure, and end-stage renal disease (ESRD), hypertension is controlled in only about one quarter of patients.1,10

Obesity is independently associated with numerous adverse CV, renal, and metabolic outcomes, including coronary heart disease, heart failure, and type 2 diabetes mellitus (Table I).2,3,6,11,12 Further, abdominal adiposity has been shown to be an even stronger risk factor for myocardial infarction than BMI.13 As would be expected given such adverse outcomes, obesity independently increases the risk of CV and all-cause mortality.6,12

Table I.  Increased Risks Associated With Obesity
Cardiovascular
 Hypertension
 Coronary heart disease
 Congestive heart failure
 Stroke
 Cardiac arrhythmias
 Cardiac autonomic neuropathy
 Cardiovascular mortality
 Endothelial dysfunction
 Left ventricular hypertrophy
Metabolic
 Insulin resistance
 Hyperinsulinemia
 Impaired glucose tolerance
 Type 2 diabetes mellitus
 Metabolic syndrome
 Dyslipidemia
Renal
 Hyperfiltration
 Microalbuminuria
 Structural changes in kidney
Other
 Altered pulmonary function
 Sleep apnea
 Obesity hypoventilation
 Gallbladder disease
 Osteoarthritis
 Cancers of endometrium, breast, prostate, colon
 All-cause mortality

“Obesity—hypertension,” a term that underscores the link between these two deleterious conditions, may be seen in the context of the cardiometabolic syndrome, the prevalence of which also is on the rise, particularly among women, due to increasing rates of both obesity and hypertension.14 Whether this cluster of metabolic risk factors, most definitions of which also include dyslipidemia and insulin resistance/ impaired glucose tolerance, comprises a clinically meaningful syndrome remains controversial; however, the benefit of interventions targeted toward reducing one or more of the component risk factors is well recognized.15–17 The American Heart Association/National Heart, Lung, and Blood Institute (AHA/ NHLBI) has published an updated version of the National Cholesterol Education Program's (NCEP's) definition of the metabolic syndrome,16,18 and a similar definition was recently proposed by the International Diabetes Federation (IDF).16,17 Obesity, particularly abdominal obesity, conveys even greater risk in conjunction with other components of the cardiometabolic syndrome.18 Thus, waist circumference, which is a surrogate for visceral adipose tissue, is integral to both the IDF and AHA/NHLBI definitions.16,17

The presence of one or two components of the metabolic syndrome doubles CV risk, whereas the co-occurrence of all components more than triples the risk.19,20 The term “cardiometabolic syndrome” serves to emphasize the importance of this relationship. The metabolic syndrome increases the risk of cardiac and extracardiac target-organ damage21 and CV and all-cause mortality.19 A strong and consistent relationship has been observed between the presence of the metabolic syndrome and increased risk of myocardial infarction, stroke, and left ventricular hypertrophy and remodeling.21,22 In addition, the metabolic syndrome is associated with renal impairment,23,24 with a significant, graded relationship between the number of metabolic risk factors and the presence of kidney disease.23

Mechanisms Contributing to Obesity—Hypertension

Although the mechanisms whereby obesity contributes to the development of hypertension are not fully elucidated, mounting evidence implicates multiple neurohormonal, hemodynamic, renal, and metabolic pathways (Figure).7,11,25 Importantly, adipose tissue functions as an endocrine organ to produce hormones and cytokines, such as leptin, tumor necrosis factor-α, interleukin-6, C-reactive protein (CRP), plasminogen activator inhibitor-1, and nonesterified fatty acids, which promote inflammation, lipid accumulation, and insulin resistance.2,26 Visceral adipose tissue secretes products directly into the portal vein circulation, which may be an important factor in the development of obesity-associated metabolic disturbances such as insulin resistance and type 2 diabetes, high triglycerides, and decreased levels of high-density lipoprotein. Obesity is also associated with down-regulation of adiponectin, a protein that enhances insulin sensitivity and inhibits atherosclerosis, and reduced adiponectin is associated with increased levels of inflammatory markers such as CRP and tumor necrosis factor-α.11 In addition, endothelial abnormalities such as decreased NO responsiveness are common in obesity, possibly due to increased production of endothelin-1.

Figure Figure.

Figure Figure.

    Mechanisms of obesity—hypertension. RAS=renin—angiotensin system; SNS=sympathetic nervous system;=increase. Adapted from Hypertension. 2004;44:12–19.11

    Activation of the sympathetic nervous system (SNS) is a typical feature of obesity and appears to be a key mechanism of obesity-related hypertension, resulting in peripheral vasoconstriction, altered baroreflex activity, and increased arterial pressure.7,11,25 Importantly, excess weight is associated with increased renal SNS activity, which contributes to two principal mechanisms of hypertension: sodium retention and volume expansion.11 Increasing evidence implicates leptin, a hormone secreted by adipocytes in proportion to the degree of adiposity, as a mediator between obesity and increased SNS activity. Plasma leptin levels, which are commonly elevated in persons with hypertension, are associated with increased heart rate, plasma renin activity, aldosterone levels, and angiotensinogen levels.11,27

    Activation of the systemic and adipose tissue renin—angiotensin system (RAS) also has been implicated as a mechanism of obesity-related hypertension.11,25 Adipose tissue contains all the components of the RAS, and upregulation of the genes for renin, angiotensin-converting enzyme (ACE), and the angiotensin II type 1 receptor has been observed in association with obesity—hypertension.28 Expression of the angiotensin II type 1 receptor gene is two-fold greater in fat cells of obese hypertensive vs. obese normotensive persons.28 Animal studies indicate that adipocyte-derived angiotensin II functions locally to affect adipocyte growth and differentiation, as well as lipid synthesis and storage in adipocytes.29 Angiotensinogen produced in adipocytes may be released into the bloodstream, contributing to the circulating pool of angiotensinogen and angiotensin II.29 Tis concept is supported by a recent human study demonstrating that weight loss of 5% reduces activation of both the circulating and the adipose-tissue RAS to nearly normal levels.30 In this study, the reduction in circulating angiotensinogen levels was closely related to the reduction in waist circumference and adipose angiotensinogen gene expression. Moreover, decreased RAS activity was associated with a BP reduction of 7/2 mm Hg.30

    Obesity-related activation of the renal SNS and RAS has deleterious effects on the kidney, even in the absence of hypertension, including increased renal tubular reabsorption, glomerular hyperfiltration, and albumin excretion.7,27,31 In addition, obesity exerts direct effects through physical compression of the kidneys. Renal abnormalities associated with obesity—hypertension are similar to those commonly seen in patients with early type 2 diabetes.7,27 As in patients with diabetes, renal impairment may lead to more severe hypertension and, ultimately, glomerulosclerosis and kidney failure.7,31 The two predominant precursors of ESRD are hypertension and diabetes, both of which are closely linked to obesity.7,27

    Effects of RAS Blockade in Clinical Trials

    Although many of the major trials of RAS blockade have not used overweight or obesity as enrollment criteria or analyzed results in the subgroup of obese subjects, it is notable that mean BMI was ≥28 kg/m2 in a number of clinical trials conducted in high-risk populations.32–35 Because patients with overweight/obesity-related hypertension have a high incidence of left ventricular hypertrophy, heart failure, hyperfiltration, and micro-albuminuria, the effect of RAS blockade on CV and renal outcomes is an important treatment consideration.36

    Cardioprotective Effects. The Heart Outcomes Prevention Evaluation (HOPE)32,37 and MICRO-HOPE37 trials demonstrated cardioprotective, renoprotective, and vasculoprotective effects of the ACE inhibitor ramipril in high-risk patients with and without preexisting diabetes. In the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) study,38 the angiotensin II type 1 receptor blocker (ARB) losartan was shown to be superior to atenolol for reducing the risk of CV morbidity and mortality in hypertensive patients with left ventricular hypertrophy. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT)34 and the Valsartan Antihypertensive Long-term Use Evaluation (VALUE),35 however, which were also conducted in patients at high risk for CV events, found similar effects of various classes of antihypertensive agents, including both RAS-blocking and non—RAS-blocking agents, on the primary composite CV end point.

    Renoprotective Effects. Growing evidence suggests that ARBs delay the progression of early stage and more advanced renal disease in type 2 diabetic patients, with and without coexisting hypertension. In patients with type 2 diabetes and microalbuminuria (mean BMI, 30 kg/m2), irbesartan at doses of 300 mg and 150 mg daily reduced the relative risk of developing diabetic nephropathy by 70% (95% confidence interval [CI], 39%–86%; p <0.001) and 39% (95% CI, −8% to 66%; p=0.08), respectively, compared with placebo.39 More recently, a smaller study in subjects with type 2 diabetes and microalbuminuria investigated doses of irbesartan up to 900 mg daily.40 The 900-mg dose reduced urinary albumin excretion 15% more than the 300-mg dose (p=0.02), independent of changes in BP or glomerular filtration rate.40 In the Microalbuminuria Reduction with Valsartan (MARVAL) trial,41 valsartan significantly improved renal function, as assessed by decline in urinary albumin excretion rate, compared with amlodipine in patients with type 2 diabetes and microalbuminuria with or without hypertension (mean BMI, ?31 kg/m2). At 24 weeks, despite similar BP effects, urinary albumin excretion rate was 56% of baseline in the valsartan group vs. 92% of baseline in the amlodipine group (p <0.001). A significantly higher percentage of patients reverted to normoalbuminuria with valsartan than with amlodipine (29.9% vs. 14.5%; p=0.001).41

    In the Irbesartan Diabetic Nephropathy Trial (IDNT),42 irbesartan protected hypertensive patients (mean BMI, ?31 kg/m2) with type 2 diabetes-related nephropathy against renal disease progression. Irbesartan reduced the relative risk of ESRD by 23% (95% CI, −3% to 43%) vs. both amlodipine and placebo (p=0.07). In the irbesartan studies, the renoprotective effect was independent of its BP-lowering effect.39,42 Losartan also demonstrated significant renoprotection in patients with type 2 diabetes and nephropathy without hypertension (mean BMI, ?30 kg/m2) in the Reduction in Endpoints in NIDDM With the Angiotensin II Antagonist Losartan (RENAAL) study.43 The risk of the primary composite outcome (doubling of serum creatinine, ESRD, or death) was significantly reduced by 16% in the losartan group compared with placebo (95% CI, 2%–28%; p=0.02). There was no between-group difference in the death rate; however, losartan significantly reduced the incidence of doubling of creatinine by 25% (95% CI, 8%–39%; p=0.006) and ESRD by 28% (95% CI, 11%–42%; p=0.002).43

    Metabolic Effects. An accumulating body of evidence indicates that therapy with ACE inhibitors or ARBs significantly reduces the risk of new-onset diabetes compared with other antihypertensive classes (Table II).32–35,37,44–46 In ALLHAT,34 in which patients had a mean BMI of approximately 30 kg/ m2, treatment with the ACE inhibitor lisinopril was associated with a significantly lower rate of new-onset diabetes compared with the diuretic chlorthalidone (p <0.001). A substudy33 of LIFE that demonstrated a reduction in new-onset diabetes with losartan vs. atenolol suggests that preservation of insulin sensitivity may explain the reduced diabetes risk with the ARB.47 Results of VALUE demonstrated a significantly (p <0.0001) lower risk of new-onset diabetes with valsartan compared with amlodipine, an agent that is considered to have a neutral effect on glucose metabolism.35

    Table II.  Effect of Renin-Angiotensin System Blockade on the Relative Risk of New-Onset Diabetes
    StudyTreatmentSubject CharacteristicsMean BMI (kg/m2)RRR (% [95% CI])p Value
    HOPE32,37Ramipril vs. placeboHigh risk28.434 (15–49)<0.001
    ALLHAT34Lisinopril vs. chlorthalidoneHigh risk with hypertension29.830 (NR)<0.001
    PEACE44Trandolapril vs. placeboStable CAD, normal or slightly reduced LV functionNR17 (4–28)0.01
    LIFE33Losartan vs. atenololHypertension, LV hypertrophy28.025 (12–37)0.001
    VALUE35Valsartan vs. amlodipineHigh risk, hypertension28.6523 (14–31)<0.0001
    CHARM46Candesartan vs. placeboHeart failure2822 (4–36)0.020
    BMI=body mass index; RRR=relative risk reduction; CI=confidence interval; NR=not reported; PEACE=Prevention of Events with Angiotensin Converting Enzyme Inhibition; CAD=coronary artery disease; LV=left ventricular; CHARM=Candesartan in Heart Failure-Assessment of Reduction in Mortality and Morbidity; other trial acronyms expanded in text

    Numerous mechanisms have been proposed for the antidiabetogenic effects of RAS blockade, including increased insulin sensitivity and preserved β-cell function. RAS-blocking agents may improve perfusion to skeletal musculature and pancreatic islets as well as increase plasma levels of adiponectin.45,48 Some ARBs modestly activate peroxisome proliferator-activated receptor-γ (an insulin sensitizer), which stimulates adiponectin production.49

    RAS Blockade in Obesity—Hypertension

    The mechanisms of overweight/obesity-related hypertension, and accumulated clinical trial evidence in high-risk patients, point to RAS blockade as a useful therapeutic strategy to prevent CV and renal complications in this patient population. However, only three double-blind, randomized trials conducted to date directly assessed the effect of RAS blockade in obese hypertensive patients: Treatment in Obese Patients with Hypertension (TROPHY),50 Candesartan Role on Obesity and on Sympathetic System (CROSS),51 and a comparison of valsartan with atenolol.52

    The TROPHY trial randomized 232 obese patients (mean BMI, 32 kg/m2) with mild-to-moderate hypertension (baseline diastolic BP 90–109 mm Hg) to lisinopril, hydrochlorothiazide (HCTZ), or placebo for 12 weeks to assess the effect on B P, lipids, and glucose metabolism.50 Both active treatments significantly (p <0.05) lowered BP from baseline vs. placebo. There was no significant difference in mean BP reduction between the lisinopril and HCTZ groups at Week 12; however, the percentage of patients who achieved diastolic BP <90 mm Hg was significantly higher in the lisinopril group than in the HCTZ group (60% vs. 43%, respectively; p <0.05). Mean plasma glucose levels decreased in the lisinopril group (−3.8 mg/dL), but increased in the HCTZ group (+5.6 mg/dL; p <0.001). No effect on insulin or lipid profiles was seen with either drug.50

    The CROSS trial51 examined the anti-hypertensive, neuroadrenergic, and metabolic effects of candesartan and HCTZ in 127 obese (mean BMI, 34.4 kg/m2) hypertensive subjects (baseline diastolic B P, 95–115 mm Hg). Sympathetic activity was measured by plasma catecholamines and muscle sympathetic nerve activity; insulin sensitivity was calculated as the ratio of the area under the curve for glucose to that of insulin during a glucose tolerance test. Despite similar BP reductions with both drugs at 12 weeks, candesartan significantly increased insulin sensitivity (p <0.02) and reduced muscle sympathetic nerve activity (p <0.01) from baseline, whereas HCTZ decreased insulin sensitivity and had no significant effect on muscle sympathetic nerve activity.51

    The metabolic effects of treatment with valsartan or atenelol were investigated in a randomized, double-blind, parallel-group trial of 132 obese subjects (BMI, 30–40 kg/m2) with mild-to-moderate hypertension (baseline B P, 150– 179/95–109 mm Hg).52 HCTZ was added if BP remained >140/90 mm Hg with either monotherapy after 4 weeks. At the end of the 13-week study, B P, lipids, and high-sensitivity CRP levels were similar between groups. However, there were differences in glucose metabolism; valsartan reduced the homeostasis model assessment for insulin resistance index (−1.1±7.3) compared with an increase (+0.6±2.3) with atenolol (p=0.06). The calculated area under the concentration—time curves for this index were significantly smaller with valsartan vs. atenolol (p=0.02), indicating a greater cumulative effect of valsartan treatment on glucose metabolism.52

    Optimal Management of Obesity—Hypertension. Although obesity is recognized as a major CV disease risk factor,8 no specific guidelines for treatment of obesity—hypertension have been proposed.36 Given the burden of obesity on the heart and kidney and the high prevalence of other risk factors in obese patients, it seems reasonable to adopt a BP goal similar to that recommended for patients with existing diabetes or chronic kidney disease (<130/80 mm Hg).8

    The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) recommends weight loss as a primary intervention in patients with hypertension, supported by evidence showing that every 10-kg decrease in weight is associated with a 5- to 20-mm Hg decrease in systolic B P.8 Weight loss has multiple other beneficial effects, including the following: lowering serum cholesterol and triglycerides, raising high-density lipoprotein cholesterol, lowering glucose levels and reducing insulin resistance, and reducing proinflammatory (CRP) and prothrombotic (plasminogen activator inhibitor-1) factors.2 Because sustained weight loss can be an elusive goal in the clinical setting, however, a comprehensive approach involving lifestyle modifications and pharmacologic therapy will likely be needed.

    Most patients with hypertension require two or more medications to achieve their BP goal.8 When selecting therapy, physicians should consider not only the proven efficacy of antihypertensive agents, but also their potential effects on the risk profile of the patient with obesity—hypertension.

    ACE inhibitors and ARBs target mechanisms of overweight/obesity-related hypertension and metabolic disorders, and thus may be particularly beneficial in patients with obesity—hypertension. In obese hypertensive patients, enalapril has been shown to significantly (p <0.05) decrease levels of norepinephrine, insulin, and leptin, whereas amlodipine lowers leptin and insulin levels to a lesser extent and has no effect on norepinephrine.53 ACE inhibitors improve endothelial function by a number of effects, including inhibition of bradykinin degradation and increasing NO production. ARBs also appear to improve endothelial dysfunction by increasing endothelial-dependent relaxation and inducing anti-inflammatory, anticoagulant, and antioxidant effects.54 In patients with heart failure, valsartan reduces brain natriuretic peptide and attenuates the increase in norepinephrine seen over time; these neurohormones are strong predictors of morbidity and mortality.55,56 RAS blockade with an ACE inhibitor, and to a greater extent with an ARB, increases adiponectin levels in association with increased insulin sensitivity.48,57 In an animal model of diabetes, valsartan increased insulin sensitivity and glucose uptake in skeletal muscle.58

    As discussed, ACE inhibitors and ARBs have been shown in clinical trials to reduce the risk of new-onset diabetes compared with other antihypertensive classes.34,35 ACE inhibitors and ARBs also delay the progression of renal disease in diabetic and nondiabetic patients.39,41–43

    Ongoing/Future Studies

    The Diabetes Reduction Assessment With Ramipril and Rosiglitazone (DREAM) study59 and the Nateglinide and Valsartan in Impaired Glucose Tolerance Outcomes Research (NAVIGATOR) trial60 are investigating the effects of an ACE inhibitor or an ARB, respectively, in high-risk patients. The primary end point of DREAM is new-onset diabetes. The two primary end points of NAVIGATOR are progression to diabetes and occurrence of a major CV event. NAVIGATOR has enrolled more than 9000 patients with impaired glucose tolerance and other CV disease risk factors; although overweight/obesity is not an entry criterion, the mean BMI of enrolled subjects is 30 kg/m2. Tis study should help address the issue of whether RAS blockade prevents the development of type 2 diabetes in patients with impaired glucose tolerance.

    The Ongoing Telmisartan Alone and in Combination with Ramipril Global Endpoint Trial (ONTARGET)61 is evaluating the effect of combination therapy with telmisartan and ramipril vs. telmisartan or ramipril monotherapy in high-risk patients with coronary heart disease, peripheral vascular disease, cerebrovascular disease, or diabetes with end-organ damage. A substudy, the Telmisartan Randomized Study in ACE Intolerant Subjects with Cardiovascular Disease (TRANSCEND)62 is assessing the efficacy of telmisartan in ACE-intolerant patients. Secondary outcomes in both studies include the incidence of new-onset diabetes.

    Conclusions

    The dramatic increase in rates of overweight and obesity, along with increases in the prevalence of hypertension, the metabolic syndrome, and diabetes mellitus, call for rigorous control of risk factors in high-risk populations. Although there are no specific treatment recommendations for persons with obesity—hypertension at present, weight loss, BP control, and modification of other risk factors are clearly indicated. A large body of clinical trial evidence demonstrates the benefit of RAS blockade for reducing the risk of adverse CV and renal outcomes, as well as new-onset diabetes, in patients with hypertension. Notably, most clinical trials of the antihypertensive, cardioprotective, and renoprotective effects of ACE inhibitors and ARBs were conducted in patient populations that would be classified as overweight or obese according to current criteria. Ongoing trials will provide further data on the potential protective effects of RAS blockade in persons with overweight/obesity-related hypertension, as well as greater insight into the mechanisms linking these highly prevalent conditions.

    Disclosure: Dr. Sharma holds a Canada Research Chair and is supported by grants from Canadian Institutes of Health Research, Institute of Nutrition, Metabolism and Diabetes, the Heart and Stroke Foundation, and the National Centres of Excellence, New Initiative Program. Dr. Sharma has received research funding and speaker honoraria and has served as a consultant to manufacturers of angiotensin-converting enzyme inhibitors and angiotensin receptor blockers, including Novartis, Boehringer-Ingelheim, Bristol-Myers Squibb, Sanofi-Aventis, Merck, and AstraZeneca.

    Ancillary