The present study was conducted to assess whether high-risk patients (hypertensive with LVH) treated by cardiac specialists in the community setting (outpatients) differ from patients treated by specialists in cardiac rehabilitation centers (inpatients) in terms of patient characteristics and comorbidities, process of care, and clinical outcomes. The principle findings of the study are as follows: First, inpatients had a higher rate of comorbidities and more advanced atherosclerotic disease. Second, control of hypertension of inpatients was already better on admission than in outpatients, and treatment intensity in this group was also higher during the observation period. Third, while blood pressure lowering was substantial in both groups, there were still a high proportion of patients that did not achieve treatment goals at discharge.
Current guidelines for the management of hypertension such as the one of the European Hypertension Society/European Society of Cardiology stress the importance of searching for comorbidities ("associated clinical conditions" and "target organ damage"), as they substantially influence prognosis of the patient . These associated clinical conditions (cerebrovascular disease, heart disease, renal disease, peripheral vascular disease and advanced retinopathy) or typical forms of target organ damage (LVH, arterial wall thickening e.g. in the carotids, or atherosclerotic plaques, nephropathy or microalbuminuria) have been clearly linked with elevated risk in epidemiological studies. Either of these findings puts the patient at a risk which is at least as high as in diabetes mellitus, or matches the combined presence of at least 3 conventional risk factors (such as higher age, smoking, dyslipidemia, abdominal obesity, family history of premature cardiovascular disease). Depending on the level of blood pressure, such a patient is at least at "high added risk" .
The present study focused primarily on LVH as this condition in cardiac specialist care, when diagnosed, should trigger intensive blood lowering treatment. This approach is clearly evidence-based, as a number of studies have documented substantial, however variably strong, regression of LVH with various antihypertensive drugs . The large-scale, long-term LIFE study with losartan is of particular interest as it showed, in line with Framingham  and HOPE  outcomes, that the greater regression of LVH was paralleled by a reduced incidence of CV events . Mainly based on this study, the ESC/ISH guidelines explicitly recommend AT1 blockers in patients with LVH .
Our study found that hypertensive patients with LVH when referred as inpatients were generally sicker than outpatients when taking into account comorbidities (especially in view of atherosclerotic complications). Obviously, referring primary care physicians trust that these patients will benefit from the characteristics of an inpatient setting (off-work atmosphere, additional educational elements, and generally more comprehensive treatment options).
Regarding general hypertension management and medication choice, cardiac specialists treating inpatients or outpatients seemed to follow guideline recommendations to a substantial extent. Antihypertensive treatment during the observation was intensified, as evidenced by the increased proportion of medically treated patients, by the increased number of drugs (2.5 at the discharge), and the preference of inhibitors of the renin angiotensin system. The LOWESS regression suggests that inpatients were somewhat more aggressively treated than outpatients; however, this was seen only in patients with SBP values above 160 mmHg. Notably, AT1 receptor blockers were much more frequently used in outpatients than in inpatients (54% versus 37%), whereas the opposite held true for ACE inhibitors (66% versus 37%). Potential reasons for this difference might include cost considerations. Interestingly outpatients seemed more difficult to manage, as they had much higher medication switch rates with lack of tolerability being three-fold increased compared to inpatients, and lack of compliance being substantially increased. It is known from the controlled study setting  as well as in primary care that antihypertensive medication changes due to a variety of reasons are the rule rather than the exception [27, 28]. All classes seem to have similar rates of non-response among patients, however, the newer drugs such as AT1 blockers seem to be associated under study as well as clinical practice conditions with better tolerability and consequently, higher persistence rates among treated patients [27, 29]. This might be an explanation why AT1 receptor blockers were preferred in outpatients. Nonetheless, previous reports from the primary care setting suggest that there is wide-spread reluctance of physicians to treat hypertension aggressively enough. Underlying reasons might be, at least in the elderly, the fear of doing harm by applying too-intensive treatment , and as noted in our study, compliance problems of patients if side effects are experienced . "Clinical inertia", a term that summarizes three related problems associated with inadequate management of chronic diseases (overestimation of care provided; use of 'soft' reasons to avoid intensification of therapy; and lack of education, training and practice organisation aimed at achieving treatment goals), may also play a major role, as has recently been suggested as a reason for the suboptimal hypertension treatment in the primary care sector in Germany . Reimbursement issues in Germany at least in the outpatient setting (fixed budget system ) may also contribute to underprescribing and undertreatment.
In terms of treatment outcomes, the mean absolute BP lowering effect achieved was substantial in both groups (SBP -22/-21 mmHg, DBP -9/-10 mmHg). In a recent metaanalysis of 354 randomised controlled trials including all current first-line antihypertensives, the mean BP lowering effect across all drug classes in the standard doses was SBP/DBP -9.1/-5.5 mmHg . Thus, even when accounting for the placebo effect which adds to the drug effect, the BP reduction achieved by cardiac specialists in our study was not inferior to that achieved under highly controlled study conditions. Further, they managed pre-treated patients, with the need to switch or add antihypertensive drugs, and had only a limited follow-up period to identify an optimized treatment for their patients. The mean average number of 2.5 drugs in both groups at discharge was still below the average of other observations and clinical studies to reach BP goals, where up to 5 different agents were needed . This is especially the case in patients with diabetes or nephropathy [36, 37], which made up a substantial fraction of individuals in both inpatients and outpatients in our study.
The general BP target of <140/90 mmHg were achieved by inpatients more frequently than by outpatients. This might be due to the fact that outpatient practitioners had to manage higher blood pressures at entry. However, even with comparable baseline values, in the inpatient setting more pronounced blood pressure reductions were achieved in the outpatient setting. While control rates as such in both groups were suboptimal, it has to be stressed that physicians had to treat "difficult" patients with multiple comorbidities within the constrictions of a challenging time frame.
The present study was not designed to answer the question whether cardiac specialists in the hospital setting compared to those in the community setting provide better care for patients. A number of studies compared certified cardiologist care with internists or primary care physicians, and found improved care for cardiology conditions, mainly in the treatment of patients with acute myocardial infarction or heart failure [38–43]. However, differences are multifactorial, and often a function of study design or patient selection . Treatment initiation in a hospital setting has been reported to be especially effective for cardioprotective therapies . As patients referred to rehabilitation centers usually have been pre-treated in acute care hospitals, they might benefit from better cardiac management. In terms of hypertension treatment, our study supports this view, because inpatients seemed to receive more intensive care. Further, expert physicians credentialed as "hypertension specialists" (by the German Hypertension League, similarly to the American Society of Hypertension Specialists Program ) were not identified nor were we able to analyse their treatment approaches.