Field JM, Hazinski MF, Sayre MR, et al
Circulation. 2010;122:S640-S656

The year 2010 marks the 50th anniversary of the introduction of cardiopulmonary resuscitation (CPR). During these past 50 years, tremendous research has been conducted to evaluate techniques, medications, and devices designed to advance the care of victims of cardiac arrest. The American Heart Association (AHA) developed the first CPR guidelines in 1966 and since that time has published frequent updates of the guidelines to help educate the public and medical establishment about optimal care for patients with cardiac arrest and other emergency cardiovascular conditions.

This past November, the newest set of guidelines pertaining to CPR and emergency cardiovascular care were published by the AHA in a supplement issue of Circulation. The guidelines consist of 16 parts. They address not only cardiac arrest, but also post-arrest care, dysrhythmias, acute coronary syndromes, stroke, cardiac arrest in special situations (eg, pregnancy, pulmonary embolism, etc), pediatric considerations, and ethics. Part I is a summary statement of the major changes in cardiac arrest and emergency cardiovascular care since the previous set of guidelines, which were published in 2005. The highlights of this "Executive Summary" are summarized below. For purposes of brevity, this discussion will focus on adult patients with acute cardiac conditions (cardiac arrest and dysrhythmias), excluding acute coronary syndromes, stroke, and pediatric considerations. The reader should note that the bulk of guideline recommendations, as in past years, are concentrated on victims of primary cardiac arrest and are not necessarily relevant to victims of pulmonary arrest (eg, drowning, drug overdose, etc).

Study Summary

Change from "A-B-C" to "C-A-B." A major change in basic life support is a step away from the traditional approach of airway-breathing-chest compressions (taught with the mnemonic "A-B-C") to first establishing good chest compressions ("C-A-B"). There are several reasons for this change.

• Most survivors of adult cardiac arrest have an initial rhythm of ventricular fibrillation (VF) or pulseless ventricular tachycardia (VT), and these patients are best treated initially with chest compressions and early defibrillation rather than airway management.
• Airway management, whether mouth-to-mouth breathing, bagging, or endotracheal intubation, often results in a delay of initiation of good chest compressions. Airway management is no longer recommended until after the first cycle of chest compressions -- 30 compressions in 18 seconds. The 30 compressions are now recommended to precede the 2 ventilations, which previous guidelines had recommended at the start of resuscitation.
• Only a minority of cardiac arrest victims receive bystander CPR. It is believed that a significant obstacle to bystanders performing CPR is their fear of doing mouth-to-mouth breathing. By changing the initial focus of resuscitation to chest compressions rather than airway maneuvers, it is thought that more patients will receive important bystander intervention, even if it is limited to chest compressions.

Basic life support. The traditional recommendation of "look, listen, and feel" has been removed from the basic life support algorithm because the steps tended to be time-consuming and were not consistently useful. Other recommendations:

• Hands-only CPR (compressions only -- no ventilations) is recommended for the untrained lay rescuers to obviate their fears of mouth-to-mouth ventilations and to prevent delays/interruptions in compressions.
• Pulse checks by lay rescuers should not be attempted because of the frequency of false-positive findings. Instead, it is recommended that lay rescuers should just assume that an adult who suddenly collapses, is unresponsive and not breathing normally (eg, gasping) has had a cardiac arrest, activate the emergency response system, and begin compressions.
• Pulse checks by healthcare providers have been de-emphasized in importance. These pulse checks are often inaccurate and produce prolonged interruptions in compressions. If pulse checks are performed, healthcare providers should take no longer than 10 seconds to determine if pulses are present. If no pulse is found within 10 seconds, compressions should resume immediately.
• The use of end-tidal CO₂ (ETCO₂) monitoring is a valuable adjunct for healthcare professionals. When patients have no spontaneous circulation, the ETCO₂ is generally ≤ 10 mm Hg. However, when spontaneous circulation returns, ETCO₂ levels are expected to abruptly increase to at least 35-40 mm Hg. By monitoring these levels, interruptions in compressions for pulse checks become unnecessary.

CPR devices. Several devices have been studied in recent years, including the impedance threshold device and load-distributing band CPR. No improvements in survival to hospital discharge or neurologic outcomes have been proven with any of these devices when compared with standard, conventional CPR.

Electrical therapies

• Patients with VF or pulseless VT should receive chest compressions until a defibrillator is ready. Defibrillation should then be performed immediately.
• Chest compressions for 1.5-3 minutes before defibrillation in patients with cardiac arrest longer than 4-5 minutes have been recommended in the past, but recent data have not demonstrated improvements in outcome.
• Transcutaneous pacing of patients who are in asystole has not been found to be effective and is no longer recommended.

Advanced cardiac life support. Good basic life support, including high-quality chest compressions and rapid defibrillation of shockable rhythms, is again emphasized as the foundation of successful advanced cardiac life support. The recommendations for airway management have undergone 2 major changes: (1) the use of quantitative waveform capnography for confirmation and monitoring of endotracheal tube placement is now a class I recommendation in adults; and (2) the routine use of cricoid pressure during airway management is no longer recommended.

As they did in 2005, the AHA acknowledges once again that as of 2010, data are "still insufficient ...to demonstrate that any drugs improve long-term outcome after cardiac arrest."

Several important changes in recommendations for dysrhythmia management have occurred:

• For symptomatic or unstable bradydysrhythmias, intravenous infusion of chronotropic agents (eg, dopamine, epinephrine) is now recommended as an equally effective alternative therapy to transcutaneous pacing when atropine fails;
• As noted above, transcutaneous pacing for asystole is no longer recommended; and
• Atropine is no longer recommended for routine use in patients with pulseless electrical activity or asystole.

Post-cardiac arrest care. Post-cardiac arrest care has received a great deal of focus in the current guidelines and is probably the most important new area of emphasis. There are several key highlights of post-arrest care:

• Induced hypothermia, although best studied in survivors of VF/pulseless VT arrest, is generally recommended for adult survivors of cardiac arrest who remain unconscious, regardless of presenting rhythm. Hypothermia should be initiated as soon as possible after return of spontaneous circulation with a target temperature of 32°C-34°C.
• Urgent cardiac catheterization and percutaneous coronary intervention are recommended for cardiac arrest survivors who demonstrate ECG evidence of ST-segment elevation acute myocardial infarction regardless of neurologic status. There is also increasing support for patients without ST-segment elevation on ECG who are suspected of having acute coronary syndrome to receive urgent cardiac catheterization.
• Hemodynamic optimization to maintain vital organ perfusion, avoidance of hyperventilation, and maintenance of euglycemia are also critical elements in post-arrest care.

Viewpoint

The AHA 2010 guidelines represent significant progress in the care of victims of cardiac arrest. Most important is the stronger emphasis on post-cardiac arrest care. Induced hypothermia is underscored, and perhaps the most important advance is the recommendation for urgent percutaneous coronary intervention in survivors of cardiac arrest. The wealth of data thus far indicate that post-arrest percutaneous coronary intervention may be the most significant advance toward improving survival and neurologic function since defibrillation was first introduced decades ago.

In reviewing these guidelines, I must admit, however, that I was disappointed that AHA hesitated to adopt the concepts of "cardiocerebral resuscitation" (CCR). CCR also promotes the "C-A-B" approach to resuscitation, but it fosters even further delays in airway intervention -- withholding any form of positive pressure ventilations, in favor of persistent chest compressions, for as long as 5-10 minutes after the cardiac arrest. The current guidelines recommend withholding positive pressure ventilation for a mere 18 seconds. First described in 2002,^[1] CCR has been studied more recently as well and demonstrated marked improvements in rates of resuscitation and neurologic survival.^[2-4] I think that CCR should be incorporated into basic life support protocols for victims of primary cardiac arrest as quickly as possible to further improve outcomes.

Optimal management of cardiac arrest in the current decade can be summarized simply by "the 4 Cs": Cardiovert/defibrillate, CCR, Cooling, and Catheterization

James B. Byrd, MD, MSc; Chan Zeng, PhD; Heather M. Tavel, BS; David J. Magid, MD, MPH; Patrick J. O'Connor, MD; Karen L. Margolis, MD; Joe V. Selby, MD, MPH; P. Michael Ho, MD, PhD

American Heart Journal. 2011;162(2):340-346. © 2011 Mosby, Inc.

Abstract

Background The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure recommends that clinicians consider the use of multidrug therapy to increase likelihood of achieving blood pressure goal. Little is known about recent patterns of combination antihypertensive therapy use in patients being initiated on hypertension treatment.
Methods We investigated combination antihypertensive therapy use in newly diagnosed hypertensive patients from the Cardiovascular Research Network Hypertension Registry. Multivariable logistic regression was used to assess the relationship between combination antihypertensive therapy and 12-month blood pressure control.
Results Between 2002 and 2007, a total of 161,585 patients met criteria for incident hypertension and were initiated on treatment. During the study period, an increasing proportion of patients were treated initially with combination rather than with single-agent therapy (20.7% in 2002 compared with 35.8% in 2007, P < .001). This increase in combination therapy use was more pronounced in patients with stage 2 hypertension, whose combination therapy use increased from 21.6% in 2002 to 44.5% in 2007. Nearly 90% of initial combination therapy was accounted for by 2 combinations, a thiazide and a potassium-sparing diuretic (47.6%) and a thiazide and an angiotensin-converting enzyme inhibitor (41.4%). After controlling for relevant clinical factors, including subsequent intensification of treatment and medication adherence, combination therapy was associated with increased odds of blood pressure control at 12 months (odds ratio compared with single-drug initial therapy 1.20; 95% CI 1.15–1.24, P < .001).
Conclusions Initial treatment of hypertension with combination therapy is increasingly common and is associated with better long-term blood pressure control.

Introduction

The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC-7) recommends consideration of initial hypertension therapy with >1 drug, particularly when blood pressure (BP) is >20/10 mm Hg above goal, to increase the likelihood of achieving BP goal in a timely fashion.^[1] Moreover, clinical trials demonstrate that most hypertensive patients will not achieve goal BP with a single drug alone, further supporting the use of multidrug therapy.^[2–4] Combining drugs with complementary mechanisms of action permits use of lower doses of each, reducing the risk of dose-dependent adverse events. However, to date, little is known about the use of combination therapy for the initial treatment of hypertension in routine practice.

Prior studies have assessed prevalent combination therapy use for hypertension and found increasing use over time.^[5,6] However, many of these studies are dated, report trends before the release of the JNC-7 report, or provide little detail regarding the most commonly occurring drug combinations. Weycker et al^[7] found that fewer patients with stage 1 than with stage 2 hypertension at time of diagnosis achieved adequate BP control (BP <140/90) at 360 days. However, use of combination therapy was not reported, nor its association with BP control.^[7] There is a lack of contemporary data regarding the most commonly used combination therapy for newly diagnosed patients with hypertension and the association between initial combination therapy use and BP control.

Accordingly, we assessed trends in the use of combination versus single-agent therapy as initial treatment for patients with incident hypertension using the Cardiovascular Research Network (CVRN) Hypertension Registry from 2002 to 2007. Next, we identified the most common combination agents used over time and stratified use of combination therapy by stage of hypertension at time of treatment initiation. Finally, we assessed the association between combination agent use and BP control 12 months after initiation of hypertension therapy among patients with incident hypertension

Definition of the Hypertensive Cohort

This study was conducted within the CVRN, a consortium of research organizations affiliated with the Health Maintenance Organization Research Network and sponsored by the National Heart, Lung, and Blood Institute. The CVRN Hypertension Registry includes all adult patients with hypertension at 3 large integrated health care delivery systems, HealthPartners of Minnesota, Kaiser Permanente Colorado, and Kaiser Permanente Northern California. The algorithm for entry into the cohort has been described previously.^[8] Briefly, patients entered into the registry upon meeting ≥1 of the following criteria: (1) at least 2 consecutive elevated BP measurements (ie, ≥140 mm Hg systolic and/or 90 mm Hg diastolic or ≥130/80 mm Hg in the presence of diabetes mellitus or chronic kidney disease); (2) 2 diagnostic codes for hypertension (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes: 401.x to 405.x) recorded on separate dates; (3) 1 diagnostic code for hypertension plus prescription for an antihypertensive medication; or (4) 1 elevated BP measurement plus 1 diagnostic code for hypertension.

The current analysis included 161,585 patients with incident hypertension between 2002 and 2007 who were started on therapy with antihypertensive medications. Incident hypertension was defined as having 1 year or longer health plan membership with drug coverage before meeting hypertension registry entry criteria, without a prior hypertension diagnosis, and without prior pharmacy dispensing or claims filed for antihypertensive medications. Individuals who qualified for entry into the cohort but whose initial BP upon cohort entry was normal (n = 16,038) or missing (n = 41,473) were excluded. We excluded patients whose initial BP upon entry into the cohort was normal to help ensure that the patients included in the analysis did indeed have as-yet-untreated incident hypertension. Only individuals with at least 180 days of health plan membership with drug coverage after entry into the cohort were included in the current analysis. Baseline BP was defined as the BP closest to the date of entry into the cohort and could include BP measurements up to 90 days before cohort entry date. In addition, the 1.4% of patients prescribed 3 or medication classes as initial treatment for a new diagnosis of hypertension were not included in the analysis because these patients may have prevalent rather than incident hypertension.

Combination Therapy

Combination antihypertensive therapy, the primary exposure of interest, was defined as treatment initiation with 2 antihypertensive medications for patients with incident hypertension. Treatment initiation was defined as the first prescription or set of prescriptions filled after entry in the hypertension registry in patients without prior diagnosis of hypertension. Initiation of medication was determined using pharmacy dispensing data, which include all medications dispensed from health plan outpatient pharmacies, or claims filed for medications.

Blood Pressure Control at 12 Months

Blood pressure control at 12 months was the primary outcome of interest and was assessed using the BP measurement closest to 12 months after initiation of hypertension treatment (and within a 6- to 18-month window). Blood pressure was considered uncontrolled if BP >140/90 mm Hg (or >130/80 mm Hg in patients with diabetes mellitus or chronic kidney disease). Systolic BP (SBP) and diastolic BP (DBP) data at 12 months were available for 77.4% of the cohort.

Statistical Analysis

The BP levels, stage of hypertension, and treatment with either combination or single-agent therapy as well as the specific combination were described for patients with incident hypertension being started on treatment during the study period. Student's t and χ² tests were used to compare baseline variables in patients treated with combination versus single-agent therapy. Multivariable logistic regression assessed demographic and clinical factors associated with initial combination antihypertensive drug therapy. The model included site of care, gender, race, smoking status, year of entry into the cohort, and comorbid compelling indications for particular antihypertensive drug classes (eg, diabetes, myocardial infarction, heart failure, chronic kidney disease, or stroke). International Classification of Diseases, Ninth Revision, Clinical Modification codes were used to define myocardial infarction (410.*), congestive heart failure (428.*), and ischemic stroke (433.*, 434.*, and 436). Diabetes mellitus was defined by the following: (1) 2 outpatient diagnoses or 1 primary inpatient discharge diagnosis of diabetes mellitus (ICD-9-CM code 250.x), (2) prescription for any antidiabetic medication other than metformin or thiazolidinediones, (3) prescription for metformin or a thiazolidinedione plus a diagnosis of diabetes mellitus, or (4) hemoglobin A1c value >7% or 2 fasting plasma glucose values >7 mmol/L (126 mg/dL) on separate dates. Chronic kidney disease was defined as 2 consecutive serum creatinine values that yield estimated glomerular filtration rates <60 mL/min when the Modification of Diet in Renal Disease equation is applied or by an ICD-9-CM diagnostic code for chronic kidney disease (codes 585.1–585.9). To assess the association between combination therapy and BP control at 12 months, multivariable logistic regression models were constructed adjusting for these variables, as well as BP stage and combination antihypertensive agent use. Data from patients who died or who were missing a BP measurement within a 6-month window around the 12-month time point were censored.

To further assess the robustness of our findings, we performed a series of sensitivity analyses. First, we assessed the association between fixed-dose (ie, 2 antihypertensive medications in a single pill) or free (ie, 2 antihypertensive medications as 2 separate pills) combinations versus single-agent therapy and BP control. Second, we evaluated the association between angiotensin-converting enzyme (ACE)/thiazide combination therapy versus single-agent and BP control. Third, we adjusted for adherence to antihypertensive medications in the evaluation of the association between combination therapy and BP control. Medication adherence was calculated as total days' supply dispensed divided by the number of days between first and last fills for that drug.^[9] Patient-level adherence with antihypertensive drugs was calculated as the time-weighted average of adherence for each antihypertensive drug.

Fourth, we compared therapy intensification between patients initiated on combination therapy versus single-agent therapy. Intensification of antihypertensive therapy was based on the number of times a drug dose was increased and/or the number of times an antihypertensive drug class was added (including the addition of a new class in substitution for another class). Thus, this intensification metric accounts for changes in the antihypertensive drug regimen after our initial assessment of single versus combination therapy. We hypothesized that patients started on combination therapy would have less need to have therapy intensified during the follow-up period because they were being initiated on more antihypertensive therapy. Finally, we adjusted for therapy intensification in the assessment of the association between combination therapy initiation and BP control. All analyses were 2 tailed, and P < .05 was considered significant. All statistical analyses were completed using SAS v9.1 (SAS Inc, Cary, NC).

The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the manuscript, and its final contents. Funding for the study was provided by the National Institutes of Health.

Baseline Characteristics of Patients With Incident Hypertension

Among 161,585 patients with incident hypertension being initiated on therapy during the study period, there was a decline in the proportion of individuals entering the cohort with stage 2 hypertension (defined as initial SBP ≥160 mm Hg or DBP ≥100 mm Hg), from 58.7% in 2002 to 40.0% in 2007 (P < .001) (Figure 1). There was a corresponding decrease in SBP during the same period from a mean SBP of 158.1 mm Hg for those entering the cohort in 2002 to a mean of 152.3 mm Hg for those entering in 2007 (Figure 2). The mean DBP for those entering the cohort in 2002 was 91.3 mm Hg. By 2004, the mean DBP decreased to 89.1 mm Hg, subsequently increasing slightly to 89.6 mm Hg in 2007.

[ CLOSE WINDOW ]

Figure 1.

Study participants' baseline BP upon entering the hypertension cohort. Between 2002 and 2007, there was an increase in the proportion of patients whose baseline BP was in the stage 1 range (blue line). There was a corresponding decrease in the proportion with a baseline BP in the stage 2 range (red line).

[ CLOSE WINDOW ]

Figure 2.

Study participants' mean BP upon entering the hypertension cohort. The error bars represent SD. Between 2002 and 2007, the mean SBP of patients with incident hypertension decreased. The mean DBP differed little in 2002 compared with 2007.

Initial Combination Therapy for Incident Hypertension

In both stage 1 and stage 2 hypertensive patients, the proportion initially treated with 2 drugs increased from 2002 to 2007. This increase was larger in patients with stage 2 hypertension. Among patients newly diagnosed with stage 2 hypertension, the proportion initially treated with 2 drugs rose from 21.6% in 2002 to 44.5% in 2007 (Figure 3). Nearly 90% of initial combination therapy was accounted for by 2 combinations of classes. A combination of a thiazide and a potassium-sparing diuretic accounted for 47.6% of initial combination therapy, and a combination of a thiazide and an ACE inhibitor accounted for another 41.4%. Of patients initiated on antihypertensive treatment with 2 drugs, 86% of first antihypertensive prescriptions were for fixed-dose combination products.

[ CLOSE WINDOW ]

Figure 3.

The use of 2-drug combination antihypertensive therapy as initial treatment of incident hypertension increased significantly between 2002 and 2007. The increase was larger among patients with stage 2 hypertension at time of diagnosis.

Less common combinations of antihypertensive medication classes included a thiazide and a β-blocker (5.3% of all combination therapy), and an ACE inhibitor and a β-blocker (2.6% of initial combination therapy). The remaining 3.1% of initial combination therapy was composed of a wide variety of infrequent combinations, such as ACE inhibitor/loop diuretic (0.54%), thiazide/calcium-channel blocker (CCB) (0.56%), ACE inhibitor/CCB (0.17%), and thiazide/angiotensin receptor blocker (ARB) (0.14%).

Patients treated with initial combination therapy (fixed-dose or free combinations) were slightly younger and had higher SBP and DBP than those treated initially with a single agent (Table I). Compared with patients with stage 1 hypertension, a higher proportion of patients with stage 2 hypertension were treated with combination therapy. Multivariable logistic regression identified several factors associated with combination therapy, including more recent entry into the cohort and prior myocardial infarction (Table II). Conversely, diabetes mellitus, history of ischemic stroke, and history of chronic kidney disease were associated with initiation of a single antihypertensive agent.

Blood Pressure Control at 12 Months After Initiation of Antihypertensive Therapy

After exclusion of patients whose 12-month BP control status was missing, the population included 124,984 patients with 12-month BP control data. The 12-month BP was controlled in 78,095 patients (62.5% of participants with incident hypertension). Initial treatment with 2 antihypertensive drugs was associated with higher odds of BP control at 12 months (odds ratio [OR] compared with single-drug initial therapy 1.16; 95% CI 1.12–1.20, P < .001) adjusting for demographics and clinical characteristics). The findings were consistent whether patients were prescribed fixed-dose combination agents (OR compared with single-drug initial therapy 1.16; 95% CI 1.12–1.20, P < .001) or 2 free-drug antihypertensive agents (OR compared with single-drug initial therapy 1.14; 95% CI 1.06–1.23, P < .001). In addition, prescription of the combination thiazide/ACE inhibitor was also associated with increased odds of BP control compared with initial single-agent therapy (OR 1.25; 95% CI 1.19–1.31, P < .001 for thiazide/ACE inhibitor compared with single-drug initial therapy).

Next, we compared adherence to antihypertensive medication between patients started on combination versus single-agent therapy. Although adherence to antihypertensive medications was statistically different between the 2 groups, the difference in adherence levels was not clinically significant (mean adherence 0.76 ± 0.28 vs 0.78 ± 0.26 for combination versus single-agent therapy; P < .001). Furthermore, the association between combination therapy and improved BP control remained consistent after adjusting for adherence to antihypertensive medications (OR 1.19; 95% CI 1.15–1.24, P < .001). Patients initiated on combination therapy were less likely to have an increase in the number of classes of antihypertensive medications during the year of follow-up compared with patients prescribed with single-agent therapy (32.1% vs 37.4%, for the percentage of patients with any increase in classes of antihypertensive medications; P < .001); however, there was no difference in the number of dosing increases of antihypertensive medications (24.6% vs 24.9%; P = .33). After additional adjustment for therapy intensification, combination therapy remained associated with increased odds of BP control (OR 1.20; 95% CI 1.15–1.24, P < .001). Baseline SBP and initial single versus combination therapy did not have an interactive effect on 12-month BP control (P = .75).

The objectives of this study were to assess trends in the use of combination versus single-agent therapy as initial treatment for patients with incident hypertension and to evaluate the use of combination therapy and BP control. We found that initial therapy with 2 drugs for new-onset hypertension was associated with better BP control at 12 months compared with single-agent therapy even after adjusting for, among other factors, subsequent therapy intensification and medication adherence. The association between 2-drug initial therapy for hypertension and better 12-month BP control was found irrespective of use of fixed-dose or free combination products. In addition, we found increasing use of combination therapy as first-line therapy for hypertension from 2002 to 2007, with a particularly large increase in 2006. This increase was larger in patients who were initially treated for stage 2 hypertension. Finally, between 2002 and 2007, fewer patients were treated initially for stage 2 hypertension, a finding perhaps related to earlier recognition of hypertension by clinicians. Our findings highlight not only the increasing role of combination antihypertensive agents in routine practice but also the potential long-term benefit of initiating 2 antihypertensive medications when hypertension is diagnosed.

Several studies, differing in important ways from the current study, have reported aspects of the pharmacoepidemiology of combination antihypertensive therapy. A study of Canadian office visits reported little change in the use of fixed-dose antihypertensive products as a proportion of all antihypertensive agents used between 1996 and 2006.^[10] Our finding of increasing use of combination therapy is consistent with other studies that have considered the more common^[5] free-drug combinations as well as the less common fixed-dose combinations.^[5,6] In what was perhaps the most detailed longitudinal epidemiologic study of combination therapy before the current study, Ma et al^[5] reported increasing use of several combinations between 1993 and 2004. Increasingly common combinations included ACE inhibitor or ARB in combination with diuretic (10% vs 27% of all antihypertensive drug visits), CCB (8% vs 17%), or β-blocker (9% vs 16%), as well as diuretic/β-blocker combination therapy (9% vs 16%).^[5] Overall, the most common 2-drug combinations in the study by Ma et al were diuretic/β-blocker, diuretic/ACE inhibitor or ARB, and ACE inhibitor or ARB/β-blocker. Although we also found a high proportion of patients treated with a combination of a diuretic and an ACE inhibitor, the other common combinations in the study by Ma et al were not commonly observed in our cohort. The practice patterns observed may be related to formulary considerations specific to managed care organizations. In contrast to the study of Ma et al, the current study focused on initiation of therapy. The association between certain comorbid compelling indications and initiation of a single drug is somewhat surprising. This finding remains unexplained but may be attributable to a variety of factors, including a desire to avoid a too rapid decrease in BP in patients who have a history of cerebrovascular accident. To our knowledge, the current study is the first large epidemiologic study reporting the association between initial treatment with combination antihypertensive therapy and long-term BP control.

Our main finding, the association between initial combination therapy and improved 12-month BP control, is consistent with clinical trials data. Feldman et al^[11] found improved 6-month BP control in patients randomized to a treatment algorithm involving initiation of 2 antihypertensive drugs compared with guideline-based management. In a large meta-analysis of antihypertensive therapy, the use of a second antihypertensive agent was associated with approximately 5 times the extra BP-lowering effect compared with doubling the dose of the initial medication.^[12] This increased efficacy likely arises from the fully additive effect provided by some drug combinations with complementary mechanisms of action.^[13] Moreover, this finding adds to the evidence supporting current guidelines, which recommend consideration of combination antihypertensive therapy in many situations.

The American Society of Hypertension (ASH) recently published a position paper in which various 2-drug combinations of antihypertensive agents were classified as preferred, acceptable, or less effective.^[13] The combination of a thiazide and a potassium-sparing diuretic was recognized as a form of combination antihypertensive therapy in both JNC-7 and the ASH guidelines. Preferred combinations include renin-angiotensin-aldosterone system (RAAS) inhibitor/diuretic and RAAS inhibitor/CCB. Although the data analyzed in this study predated the publication of this position paper, 94.2% of prescriptions in our study were accounted for by 3 combinations judged to be either preferred or acceptable. A small minority of combinations were of the less effective type, including ACE inhibitor/β-blocker (2.6% of all combinations). Although our findings suggest that clinical practice is largely consistent with expert consensus and clinical trials data, use of a RAAS antagonist in combination with a CCB, a preferred combination, was rare, accounting for <1% of combination therapy. In recommending RAAS/CCB combination therapy, the ASH position paper cites the ACCOMPLISH trial,^[14] which was published after the end of data collection for the present study. It is possible that the use of RAAS/CCB combination therapy has increased since the publication of ACCOMPLISH. Moreover, it is possible that specific formulary considerations at the sites involved in the study may have had an impact on the use of CCBs in combination with RAAS antagonists, although generic drugs are available in both of these classes.

This study has several limitations. The study cohort was composed of patients treated within integrated health care delivery systems, which could limit the applicability of the findings to other populations. The inclusion of multiple geographically separated sites with different health plans increases the likelihood that the results reflect practice in many areas. During the study period, hypertension practice guidelines and algorithms were made available via paper format and/or the Web within each health care system and disseminated through educational conferences. These guidelines were consistent with national guidelines (eg, JNC-7) on hypertension management. In addition, generic medications were encouraged over brand name medications, but medications representing all therapeutic classes for hypertension treatment were available. Clinicians were encouraged to follow evidence-based practice; however, there was no mandate to use a specific agent in the initial treatment of hypertension. An emphasis on the use of generic drugs may have resulted in greater use of the combination of hydrochlorothiazide and triamterene than would be observed in other practice settings. The 3 large health systems in which the study was conducted have preferred formularies, and there are financial incentives for both physicians and patients to avoid expensive and/or nongeneric drugs, including many branded combination therapy treatments. Our analysis, however, was performed at the level of antihypertensive classes, rather than individual products, and generic drugs were available in most of the classes during the period we studied, particularly in the later years. In addition, our analysis of the association between combination therapy and 12-month BP excluded data from patients for whom 12-month BP data were not available. It is possible that inclusion of those data, had they been available, might have altered the findings. Finally, the analysis only takes into account prescriptions that were filled rather than all prescribed medications. However, failure to fill prescriptions would be expected to bias the study toward a lack of effect of combination antihypertensive therapy on 12-month BP.

Initial treatment of newly diagnosed hypertensive patients with a combination of 2 antihypertensive drugs is becoming increasingly common, particularly among patients with stage 2 hypertension. After controlling for subsequent changes in therapy, initial treatment with 2 antihypertensive agents was associated with a higher odds of BP control at 1 year. Clinicians should consider initiation of 2 antihypertensive agents upon diagnosis of hypertension.