Getting to Dry.

Abstract

HomeCirculationVol. 137, No. 16Getting to Dry Free AccessEditorialPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessEditorialPDF/EPUBGetting to Dry Sandip Zalawadiya, MD and Lynne Warner Stevenson, MD Sandip ZalawadiyaSandip Zalawadiya Vanderbilt Heart and Vascular Institute, Vanderbilt University Medical Center, Nashville, TN. Search for more papers by this author and Lynne Warner StevensonLynne Warner Stevenson Vanderbilt Heart and Vascular Institute, Vanderbilt University Medical Center, Nashville, TN. Search for more papers by this author Originally published17 Apr 2018https://doi.org/10.1161/CIRCULATIONAHA.118.032691Circulation. 2018;137:1684–1687Article, see p 1671Heart failure (HF) hospitalization presents a substantial burden to patients, hospitals, and payors. Hospitals face competing incentives from government payors because prolonged lengths of stay risk unreimbursed costs beyond the fixed reimbursement, whereas premature discharge increases the risk of 30-day readmission, for which hospitals can be penalized on a much broader scale. Clinicians have struggled to define the objective criteria regarding readiness for discharge from a HF hospitalization, which should address not only relief of congestion, but also evaluation of etiology, enhancement of guideline-directed therapies, review of disease trajectory, and consolidation of outpatient management. Once gaps in the transition process have been closed, the major cause of readmission is recurrent congestion. Regardless of how congestion is documented,1–3 patients who go home “wet” are likely to return “wetter”. Quality improvement and cost containment efforts are converging to design care pathways that begin early during hospitalization to ensure adequate decongestion by discharge.Titration Strategies Guided By Natriuretic Peptide LevelsEmployment of brain natriuretic peptide (BNP) levels as targets for therapy arose naturally from the robust predictive value of BNP levels and changes in levels, further supported by their close relationship with elevated cardiac filling pressures4,5 and recognized as a hemodynamic determinant of outcomes. Utility of a biomarker as a target for therapy requires not only clinical relevance, but also timely responsiveness to the therapy titration, and unique information beyond usual clinical assessments. Multiple trials have used natriuretic peptides as targets for adjustment of chronic outpatient therapy with neurohormonal antagonists6–10 (Figure). When uptitration was aggressive in response to persistently high levels of natriuretic peptides, outcomes were better than without such uptitration.9 However, when the design encouraged vigorous uptitration for both groups in the rigorous GUIDE-IT trial (Guiding Evidence Based Therapy Using Biomarker Intensified Treatment in Heart Failure) of ambulatory HF patients, neither outcomes nor BNP levels were better in the BNP-guided group.10Download figureDownload PowerPointFigure. Time intervals during which natriuretic peptide-guided strategies have been compared to clinical assessment alone to guide therapy for heart failure, underlined by the consistent power of natriuretic peptides to predict outcomes both using absolute and changing levels. Most trials have tested the strategy in the stable outpatient setting. The current study was randomized for the interval between clinical stability and discharge, but also tracked the period between admission and clinical stability. ADHF indicates acute decompensated heart failure; BNP, brain natriuretic peptide; GUIDE-IT trial, Guiding Evidence Based Therapy Using Biomarker Intensified Treatment in Heart Failure; HF, heart failure; NT-pro BNP, N-terminal pro B-type natriuretic peptide; PRIMA study, Can pro-Brain-Natriuretic Peptide Guided Therapy of Chronic Heart Failure Improve Heart Failure Morbidity and Mortality?; PRIMA II trial, NT-proBNP Target on the Reduction of Readmission and Mortality Rates; and STARBRITE trial, Strategies for Tailoring Advanced Heart Failure Regimens in the Outpatient Setting.While the chronic HF trials often targeted absolute levels of natriuretic peptides, the natriuretic peptide levels achieved by discharge have been used to assign more personalized targets after hospitalization. This was done for the first 3 months in the STARBRITE trial (Strategies for Tailoring Advanced Heart Failure Regimens in the Outpatient Setting).7 Diuretics were adjusted in the majority of patients in both groups, with similar frequency of increased and decreased doses and no difference in early outcomes. Using NT-proBNP (N-terminal pro-B-type natriuretic peptide), the PRIMA I trial (Can pro-Brain-Natriuretic Peptide Guided Therapy of Chronic Heart Failure Improve Heart Failure Morbidity and Mortality?) from the Netherlands compared clinically guided management to management guided by the target identified as the lowest value between discharge and 2-week postdischarge value. Following patients through the year, both groups had similar small declines in NT-proBNP below target by 12 months, with no difference in rehospitalization or death.8The most rapid changes in natriuretic peptide levels are seen during hospitalization,4,5 which presents an obvious setting in which to test them as targets to guide acute therapy. Multiple studies have shown that reductions in BNP between admission and discharge predict better outcomes, whether defined by an absolute or relative reduction, with common thresholds being absolute reduction of 250 pg/mL in BNP or 30% reduction in either BNP or NT-proBNP.11,12 The PRIMA II study (NT-proBNP Target on the Reduction of Readmission and Mortality Rates) in this issue of Circulation provides new information about the changes in natriuretic peptides during effective therapy of patients hospitalized with heart failure.13PRIMA-II Highlights the Target of Clinical StabilityStrategy trials are among the most difficult, because of the need to define parallel treatments as equivalently as possible, despite diverging decision nodes. The PRIMA II trial was cleverly designed to test whether a strategy guided to reduce NT-pro BNP by at least 30% would improve outcomes for hospitalized patients after clinical assessment suggested stability. It was not designed to answer whether this strategy would have been useful if initiated earlier on admission, but the thoughtful collection of data also allowed us to examine this question also.Although the criteria for clinical stability did not specify components of congestion, the dramatic early response before randomization reflects highly expert clinical care in decongestion at the selected investigating sites in the Netherlands, Portugal, France, and Spain. At a median of 5 days, 64% of patients had already achieved at least a 30% reduction in NT-proBNP even though the degree of reduction was not yet known to the clinicians. Furthermore, in the patients for whom no further intervention was needed after randomization, the total decrease in NT-proBNP during hospitalization was 61% (interquartile range, 47% to 76%). This is a more dramatic reduction than previously reported, except in the previous PRIMA-I trial in the Netherlands, in which the in-hospital reduction was 62%.8 We can speculate if the outcome of either of the PRIMA trials would have been different with a higher target of 60% biomarker reduction.If this protocol had provided BNP levels to guide therapy from the time of admission instead of at the time of clinical stability, the guided strategy could actually have decreased the duration and success of decongestion for responsive patients, stopping at 30% rather than the 61% improvement in NT-proBNP seen. Although it cannot be proven that the further reduction was beneficial, the outcome curves in their Figure 4 indicate that this responsive group had substantially better 6 month outcomes than those patients achieving less NT-proBNP reduction later (median, 42%). The worse outcomes occurred in this group despite their extra hospital days and further protocol-driven titration of neurohormonal antagonists.Matching Therapies to GoalsAcute Goal of DecongestionIn current practice, the short-term relief of congestion is achieved primarily with enhanced diuretic therapy. Intravenous nitroglycerin and nesiritide and other short-term intravenous vasodilator therapies can accelerate early symptom relief if rebound can be avoided, leaving open the possibility that judicious use of vasodilators may play a greater short-term role in relieving congestion, if they can be translated smoothly into the discharge regimen. Trials are ongoing with introduction of sacubitril/valsartan after hospitalization with advanced heart failure.Diuretic therapy is also the quickest and most reliable current therapy to decrease elevated natriuretic peptide levels, although there is a lag of a few days before the full reduction is seen.4,5 In the previous PRIMA-I trial, there was a further 15% reduction of NT-proBNP in the first 2 weeks after discharge.8 Although sometimes used in the current trial to further decrease NT-proBNP levels, diuretics were not recommended in the algorithm until after uptitration of neurohormonal antagonist therapy.Optimal Therapy for Chronic Disease ModificationHospitalization provides the opportunity to review and optimize recommended therapies to decrease disease progression and improve long-term outcomes. Therapies already in place should be continued whenever possible without hindering the achievement of decongestion. For patients not on optimal therapies, medications initiated in the hospital are more likely to be continued and uptitrated. However, renin-angiotensin system blockers and beta blockers do not relieve congestion and may decrease renal perfusion, glomerular filtration, and diuretic efficiency. Increasing these agents would not likely achieve the NT-proBNP targets in hospital if not already reached. In patients who had not reached target NT-proBNP reduction before randomization, 1 in 5 had contraindications to medication titration and some attempted increases were stopped for side effects. Although renin-angiotensin system inhibition by protocol was increased more often in the guided arm, beta blockers were decreased more often, perhaps due to limited blood pressure remaining.Good Outcomes Without Benefit from Titration To NT-proBNP LevelsThe outcomes in this elegantly designed trial provide further evidence of the remarkably effective clinical care, with a combined rate of death and HF readmission of only 36% by 6 months. The 30-day event rates are remarkably low, with combined rate of death and readmission less than 20%, compared to over 20% for readmission alone in the US experience.Outcomes were not different, however, between the patients whose therapy was guided by NT-proBNP and those with therapy guided by clinical assessment alone. Regardless of the utility of NT-proBNP, this lack of difference is not surprising as a majority of the patients already had at least a 30% decrease in NT-proBNP before randomization. The most relevant subgroup is comprised of the 140 patients whose decrease was < 30% by the time of randomization. Within this group, the event rate at 6 months was 49% with NT pro-BNP guidance versus 55% without; it is not known whether this difference might have been significant with larger sample sizes. It is notable that this group did have a substantially higher risk than those patients achieving earlier stabilization, confirming predictive value of natriuretic peptides at this point.Are We There Yet?It is hard to imagine another single biomarker with greater prognostic breadth and power than that offered by natriuretic peptides. Unfortunately, even this prevailing biomarker has not been useful to guide therapy to better outcomes than observed with meticulous guideline-directed titration based on clinical assessment and tolerability. The present study demonstrates lack of efficacy for the BNP-guided strategy in hospital as other recent trials such as PRIMA-I and GUIDE-IT have done for the strategy in outpatient practice.How can we tell when we have reached effective decongestion in hospital? Target weights have often drifted off because of nutritional changes over time. Absolute changes in net fluid output or in weight have to be interpreted in relation to the amount of congestion to be relieved because “better” is not usually “best”. The ESCAPE trial (The Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness) used clinical goals of normal jugular venous pressure, and resolution of edema and orthopnea, with and without additional targets of pulmonary capillary wedge pressure <16 mm Hg and right atrial pressure <8 mm Hg2. The EVEREST trial (Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study With Tolvaptan) arrived at the same trio of clinical findings after distillation from a grading scale that initially also included rales and fatigue.1 Although the jugular venous pressure is a central focus of the daily clinical assessment for most heart failure clinicians, critical analysis of the US Heart Failure Network trials of acute heart failure showed that when assessed by a variety of clinicians, it added little prognostic value to the orthodema score that graded orthopnea and edema at admission and discharge.3 The seminal work on hemoconcentration has emphasized the importance of adequate diuresis during hospitalization, but has not established utility for a clinical strategy titrated to achieve hemoconcentration.14 Regardless of the methods used to define congestion, this trial confirms many previous results that residual congestion identifies a clinical group at high risk for hospitalization and death. This nonresponder group deserves particular focus for new strategies beyond the algorithm of the current trial.Clinical assessment of congestion continues to prevail during daily rounds in the hospital, defying compression into any 1 dimension. There is currently no single click for “getting to dry”. It requires keeping watch on the laundry as it goes around. We can choose to set the time on the dryer, which will be too long for some and too short for others. We could start instead with the automatic dryness sensor, which does not distinguish reliably between damp and sodden, so it will also render some items too dry and leave others heavy. Because access to dryers is limited, we will have to stay close by and reach our hands in to frequently assess progress. Then, we take out and fold the laundry that is dry, return the rest, and reset it to go around again.DisclosuresNone.FootnotesThe opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.http://circ.ahajournals.orgLynne Warner Stevenson, MD, Cardiovascular Division, Vanderbilt University Medical Center, 1215 21st Avenue South, Nashville, TN 37232. E-mail [email protected]