B‐type natriuretic peptide‐guided treatment for heart failure

Abstract Background Heart failure is a condition in which the heart does not pump enough blood to meet all the needs of the body. Symptoms of heart failure include breathlessness, fatigue and fluid retention. Outcomes for patients with heart failure are highly variable; however on average, these patients have a poor prognosis. Prognosis can be improved with early diagnosis and appropriate use of medical treatment, use of devices and transplantation. Patients with heart failure are high users of healthcare resources, not only due to drug and device treatments, but due to high costs of hospitalisation care. B‐type natriuretic peptide levels are already used as biomarkers for diagnosis and prognosis of heart failure, but could offer to clinicians a possible tool to guide drug treatment. This could optimise drug management in heart failure patients whilst allaying concerns over potential side effects due to drug intolerance. Objectives To assess whether treatment guided by serial BNP or NT‐proBNP (collectively referred to as NP) monitoring improves outcomes compared with treatment guided by clinical assessment alone. Search methods Searches were conducted up to 15 March 2016 in the Cochrane Central Register of Controlled Trials (CENTRAL) in the Cochrane Library; MEDLINE (OVID), Embase (OVID), the Database of Abstracts of Reviews of Effects (DARE) and the NHS Economic Evaluation Database in the Cochrane Library. Searches were also conducted in the Science Citation Index Expanded, the Conference Proceedings Citation Index on Web of Science (Thomson Reuters), World Health Organization International Clinical Trials Registry and ClinicalTrials.gov. We applied no date or language restrictions. Selection criteria We included randomised controlled trials of NP‐guided treatment of heart failure versus treatment guided by clinical assessment alone with no restriction on follow‐up. Adults treated for heart failure, in both in‐hospital and out‐of‐hospital settings, and trials reporting a clinical outcome were included. Data collection and analysis Two review authors independently selected studies for inclusion, extracted data and evaluated risk of bias. Risk ratios (RR) were calculated for dichotomous data, and pooled mean differences (MD) (with 95% confidence intervals (CI)) were calculated for continuous data. We contacted trial authors to obtain missing data. Using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, we assessed the quality of the evidence and GRADE profiler (GRADEPRO) was used to import data from Review Manager to create a 'Summary of findings' table. Main results We included 18 randomised controlled trials with 3660 participants (range of mean age: 57 to 80 years) comparing NP‐guided treatment with clinical assessment alone. The evidence for all‐cause mortality using NP‐guided treatment showed uncertainty (RR 0.87, 95% CI 0.76 to 1.01; patients = 3169; studies = 15; low quality of the evidence), and for heart failure mortality (RR 0.84, 95% CI 0.54 to 1.30; patients = 853; studies = 6; low quality of evidence). The evidence suggested heart failure admission was reduced by NP‐guided treatment (38% versus 26%, RR 0.70, 95% CI 0.61 to 0.80; patients = 1928; studies = 10; low quality of evidence), but the evidence showed uncertainty for all‐cause admission (57% versus 53%, RR 0.93, 95% CI 0.84 to 1.03; patients = 1142; studies = 6; low quality of evidence). Six studies reported on adverse events, however the results could not be pooled (patients = 1144; low quality of evidence). Only four studies provided cost of treatment results, three of these studies reported a lower cost for NP‐guided treatment, whilst one reported a higher cost (results were not pooled; patients = 931, low quality of evidence). The evidence showed uncertainty for quality of life data (MD ‐0.03, 95% CI ‐1.18 to 1.13; patients = 1812; studies = 8; very low quality of evidence). We completed a 'Risk of bias' assessment for all studies. The impact of risk of bias from lack of blinding of outcome assessment and high attrition levels was examined by restricting analyses to only low 'Risk of bias' studies. Authors' conclusions In patients with heart failure low‐quality evidence showed a reduction in heart failure admission with NP‐guided treatment while low‐quality evidence showed uncertainty in the effect of NP‐guided treatment for all‐cause mortality, heart failure mortality, and all‐cause admission. Uncertainty in the effect was further shown by very low‐quality evidence for patient's quality of life. The evidence for adverse events and cost of treatment was low quality and we were unable to pool results.


Review question
We aimed to discover whether using B-type natriuretic-guided treatment or a health plan alone is more e ective for managing patients with heart failure.

Background
Heart failure is a complex condition that occurs when the heart does not pump blood e ectively enough to meet the needs of the body. It is caused by a range of diseases that impair the structure and function of the heart and may result in breathlessness, fatigue and fluid retention. People with heart failure are frequently users of general practice and hospitals, particularly as inpatients. Furthermore, they have reduced life expectancy, although medicines and other treatments can improve the chance of survival.
B-type natriuretic peptide (NP) is a substance produced in the heart. The measurement of NP can be used to indicate the condition of the heart. For some time, NP has been used for diagnosing heart failure and predicting what is likely to happen. We wanted to discover if NP may also o er a way to manage and make the best use of medicines.
The evidence was unclear as to whether number of deaths from any cause varied between patients with heart failure using NP-guided treatment compared with those using a health plan alone. Nor was it clear as to whether there were less deaths when the results were separated into patients older or younger than 75 years old (age results only included three studies). Furthermore, we found that the evidence was unclear whether the number of deaths from heart failure alone varied between the NP-guided treatment or health plan alone groups.
We found that hospital admission due to heart failure may be reduced in the patients using NP-guided treatment compared with a health plan alone. Based on these results we would expect that out of 1000 patients with heart failure who are guided by a health plan alone, 377 would experience an admission to hospital due to heart failure. Whereas, between 230 and 301 patients would experience an admission to hospital due to heart failure if they received NP-guided treatment. However, the evidence was unclear as to whether the numbers of hospital admission from any cause were a ected.
There was limited information about either harms to patients, or the cost of the treatment. It was not possible to combine the results from these studies for these outcomes. However, four of the six studies commented that they found no di erence in harms or less di erence in harms between the patients using NP-guided treatment compared with a health plan alone, the other two studies did not comment. Four studies reported results on costs, three of these reported there may be lower costs in the NP-guided treatment groups compared with health plan groups. Lower costs appeared to be due to less cost for hospital stays. However, one study reported that NP-guided treatment was unlikely to be cost-e ective.
The evidence was unclear as to if a benefit was shown in the replies to quality-of-life surveys when comparing between NP-guided treatment and health plan only groups.

Quality of evidence
Overall evidence for death from all causes, from heart failure alone and for hospital admission was of low quality. For harm to patients and cost outcomes the quality of evidence was low, whilst evidence for patients' quality of life surveys was very low. For all outcomes there was little evidence due to the way the studies were conducted. In addition, for harm to patients and cost of treatment there were di erences in the type of information available. 16 studies reported on all-cause mortality (n = 3292), but only 15 studies are included in the meta-analysis (n = 3169). For one study data could not be extracted or obtained in a format useable in the review.

Serial BNP or NT-proBNP-guided treatment
Funnel plot analysis suggests possible lack of small studies (beneficial control effect). Insufficient to justify downgrading the quality of evidence.
Heart failure mortality 3/4 studies suggested reduced cost in the intervention groups. One study suggested NP-guided treatment was unlikely to be cost-effective. Lower score indicates better quality of life *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio;

Quality of life
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality : We are very uncertain about the estimate.

Library
Trusted evidence. Informed decisions. Better health.
Cochrane Database of Systematic Reviews

Description of the condition
Heart failure is a condition in which the heart does not pump enough blood to meet all the needs of the body. It is caused by dysfunction of the heart due to muscle damage (systolic or diastolic dysfunction), valvular dysfunction, arrhythmias or other rare causes (NICE 2014). Clinically, it is a syndrome in which patients have typical symptoms (e.g. breathlessness, ankle swelling, and fatigue) and signs (e.g. elevated jugular venous pressure, pulmonary crackles, and displaced apex beat).The diagnosis can be di icult as many of the symptoms of heart failure are non-discriminating so the demonstration of an underlying cardiac cause is central to the diagnosis. Identification of the underlying cardiac problem is also crucial for therapeutic reasons, as the precise pathology determines the specific treatment used (e.g. valve surgery for valvular disease, specific pharmacological therapy for le ventricular systolic dysfunction, etc.) (McMurray 2012).
Heart failure due to le ventricular systolic dysfunction (LVSD) is caused by impaired le ventricular contraction, and is usually characterised by a reduced le ventricular ejection fraction (LVEF). Heart failure with preserved ejection fraction (HFPEF) is usually associated with impaired le ventricular relaxation, rather than le ventricular contraction, and is characterised by a normal or preserved le ventricular ejection fraction (NICE 2010).
Approximately 1% to 2% of the adult population in developed countries has heart failure, with the prevalence rising to ≥10% among persons 70 years of age or older (McMurray 2012). The prevalence is expected to rise in future as a result of an ageing population, improved survival of people with ischaemic heart disease and more e ective treatments for heart failure (Owan 2006).
Heart failure has a poor prognosis: 30% to 40% of patients diagnosed with heart failure die within a year -but therea er the mortality is less than 10% per year. There is evidence of a trend of improved prognosis in the past 10 years. The six-month mortality rate decreased from 26% in 1995 to 14% in 2005. Within the NHS, heart failure accounts for a total of 1 million inpatient bed-days -2% of all NHS inpatient bed-days -and 5% of all emergency medical admissions to hospital. Hospital admissions because of heart failure are projected to rise by 50% over the next 25 years, largely as a result of the ageing population. This is despite a progressive decline of the age-adjusted hospitalisation rate at 1% to 1.5% per annum since 1992/1993 (NICE 2010).

Description of the intervention
All patients with chronic heart failure require monitoring, which should include a detailed clinical assessment and a review of medication, including the need for titration and optimisation in line with guidelines and to pick up possible side e ects. The pharmacological treatment options for patients with LVSD (New York Heart Association (NYHA) functional class II-IV) include diuretics, angiotensin-converting enzyme (ACE) inhibitors (angiotensin receptor blockers if ACE inhibitors are not tolerated), beta-blockers and mineralocorticoid receptor antagonists (MRA).
The frequency of monitoring depends on the clinical status and stability of the patient. The monitoring interval should be short (days to two weeks) if the clinical condition or medication has changed, but is required at least six-monthly for stable patients with proven heart failure.
The intervention requires monitoring of B-type natriuretic peptide concentrations to guide treatment of heart failure with the aim of enhancing the management of individual patients. Btype natriuretic peptide, along with NT-proBNP, is a natriuretic peptide secreted when the heart stretches. B-type natriuretic peptide has a shorter half life of 20 minutes compared to the one to two hours for NT-proBNP, and both can be increased in patients with systolic or diastolic dysfunction (Atisha 2004). Both biomarkers have demonstrated diagnostic and prognostic utility in heart failure (Clerico 2007;Doust 2005;McMurray 2012NICE 2014. Monitoring NP concentration provides feedback to the physician about intravascular volume status, which can be used in combination with the patient's clinical condition to facilitate treatment decisions.

How the intervention might work
BNP and NT-proBNP (collectively referred to as NP) are biomarkers for heart failure which have been demonstrated to have diagnostic and prognostic utility (Clerico 2007;Doust 2005, McMurray 2012, NICE 2014. The precursor, preproBNP is cleaved to proBNP within the cardiomyocyte and stored in secretory granules; proBNP is cleaved to NT proBNP and BNP upon secretion into the bloodstream in response to an increase in intracardiac volume (Chen 2010;Ichiki 2013). Monitoring NP concentrations provides feedback to the physician about intravascular volume status, which can be used in combination with the patient's clinical condition to facilitate treatment decisions.

Why it is important to do this review
To date, five out of seven systematic reviews with meta-analyses have demonstrated that NP-guided treatment reduces all-cause mortality in patients with congestive heart failure compared with usual clinical care (Felker 2009 Monitoring with NP is recommended by NICE only for some patients by a specialist a er hospital admission or when up-titration of medication is problematic (NICE 2010). It is not recommended by the European Society of Cardiology (ESC) guideline (McMurray 2012) due to uncertainty about whether it is a more e ective approach than simply optimising treatment (combinations and doses of drugs, devices) according to guidelines.
In this review, we examined the seven outcomes described above and in addition included heart failure mortality, which has not been examined previously. In addition, we aimed to evaluate whether factors such as age, gender, severity of symptoms or stage of heart failure, and context of care (community or hospital) predicted whether a patient will benefit from NP monitoring, furthermore whether monitoring leads to a greater change in NP. However, only one of these pre-specified subgroup analyses was possible due to lack of data or inconsistency in reporting for these factors. Four further subgroup analyses were considered post-hoc: baseline LVEF, duration of follow-up, type of control, and type of biomarker.

O B J E C T I V E S
Our objectives are: 1. to assess whether treatment guided* by serial BNP or NT-proBNP (collectively referred to as NP) monitoring improves outcomes compared with treatment guided by clinical assessment alone; 2. to assess the extent to which improved outcomes are explained by up-titration of medication and/or reductions in BNP levels; and 3. to determine which groups of patients benefit most from monitoring in terms of their age, gender, severity of symptoms or stage of heart failure (with the use of the NYHA classification), and baseline NP.
*Treatment guided within this review refers to lifestyle and medication changes for the management of heart failure (i.e. no device therapy or transplantation).

Types of studies
All randomised controlled trials of BNP-or NT-proBNP-guided (collectively NP-guided) treatment of heart failure, in both inhospital and out-of-hospital settings, reporting a clinical outcome.
No restriction on length of follow-up.

Types of participants
All patients 18 years and older who are being treated for heart failure.

Types of interventions
Comparison of treatment guided by NP levels versus treatment guided by clinical assessment alone.

Primary outcomes
The primary outcome was all-cause mortality.

Search methods for identification of studies Electronic searches
We searched the following databases on 15 March 2016: 1. Cochrane Central Register of Controlled Trials (CENTRAL) in the Cochrane Library (2016, Issue 2), 2. MEDLINE (OVID, 1946to 15 March 2016, 3. Embase (OVID, 1974to 14 March 2016, 4. Database of Abstracts of Reviews of E ects (DARE) in the Cochrane Library (2015, Issue 2), 5. NHS Economic Evaluation Database (NHSEED) in the Cochrane Library (2015, Issue 2), and 6. Science Citation Index Expanded and the Conference Proceedings Citation Index on Web of Science (Thomson Reuters, 1945to 15 March 2016. Search filters limiting searches to randomised controlled trials were applied to MEDLINE and Embase (Lefebvre 2011). See Appendix 1 for the detailed search strategies. We applied no date or language restrictions.

Searching other resources
We contacted authors of relevant studies, performed citation searches and reviewed references of all full text papers retrieved. We also contacted experts in the field when relevant. We identified any ongoing trials that were registered with the World Health Organization International Clinical Trials Registry Platform (http://apps.who.int/trialsearch/) and ClinicalTrials.gov (http:// clinicaltrials.gov) on 15 March 2016.

Selection of studies
We screened the title and abstract of articles obtained from the search results (LW/JM/NP/CB) for studies that met the inclusion criteria as well as any articles in which there was uncertainty. For each article, two review authors (LW/JM/NP/CB) independently reviewed the studies for final inclusion/exclusion. In cases where it was still unclear, we contacted the study authors for clarification. We resolved disagreements by consensus or thirdparty adjudication (CH/RP).

Data extraction and management
We used data abstraction forms specifically designed for this review to abstract data on participants, interventions, and outcomes. For each study two review authors (LW/JM/NP) extracted trial results independently. We resolved di erences between authors' results by discussion and, when necessary, in consultation with a third review author (CH/RP). Where data were insu iciently reported in the published paper, we wrote to the original authors for clarification and further information.
components assessed included allocation concealment, random sequence generation, blinding of participants, personnel, and outcome assessment, incomplete outcome data, selective reporting and source of funding. We reported our judgement for each component using Cochrane's tool for 'Risk of bias' assessment (Higgins 2011).

Unit of analysis issues
No included studies had nonstandard designs such as cross-over or cluster-randomised. If a study compared more than one type of control group then the intervention group data were split equally between the control groups for both outcome events and sample size.
For continuous outcomes, if the study provided data as medians and interquartile ranges then medians were assumed to equate to the mean and the interquartile ranges were converted to standard deviations by dividing the di erence between the two values divided by 1.35 (approximate relationship between the two assuming a normal distribution). The mean di erence and standard deviation were calculated assuming a correlation of 0.5 (Higgins 2011).

Dealing with missing data
Where data were insu iciently reported in the published paper, we wrote to the original authors for clarification and further information. We analysed only the available data and discussed the impact of the missing data on our findings.

Assessment of heterogeneity
Where we pooled data, we used the I 2 statistic to quantify the level of statistical heterogeneity (Higgins 2011) .

Assessment of reporting biases
We assessed publication bias by the use of funnel plots where there were su icient studies, and reasons for asymmetry were considered if it was noted. We addressed other potential reporting biases in the Discussion.

Data synthesis
Where appropriate, we pooled data from all the studies using the analysis so ware in Review Manager (RevMan) version 5.3. For dichotomous outcomes, we combined data using a fixed-e ect model with the Mantzel-Haenzel method to determine a summary estimate of the risk ratio (RR) with 95% confidence intervals (CI). For continuous outcomes, we used a fixed-e ect model with the inverse variance method to produce a mean di erence (MD) with 95% CI for the summary estimate. Where substantial heterogeneity (I 2 ≥ 50%) was present, we considered potential explanations and where applicable used a random-e ects model to test the robustness of the findings and also considered not combining the results and presenting a descriptive analysis.

Subgroup analysis and investigation of heterogeneity
We considered subgroup analyses for the following: 1. age; 2. severity of heart failure (New York Heart Association (NYHA) classification); 3. baseline NP; 4. target NP; 5. achieved NP decrease (as a percentage of baseline); 6. patients treated in the community compared with those treated in secondary care; 7. gender.

Sensitivity analysis
We incorporated the results of the 'Risk of bias' assessment into our interpretation of the results by performing sensitivity analyses in which we excluded studies with the highest level of or unclear bias and included low risk of bias studies only.

Results of the search
The search identified 3394 references. Once duplicates were removed, the titles and abstracts of the remaining 3379 references were screened using our inclusion /exclusion criteria and 3044 removed as not relevant to the review. Full texts were examined for the remaining 335 references and from these 18 studies were included in this review (see Figure 1). Full details of all the studies are given in the Characteristics of included studies, Table 1, Table 2, Characteristics of excluded studies, and Characteristics of ongoing studies. Each study is identified by the name of the first author and year of publication of the main results paper (Study ID). Additional references are listed together with this main publication under the study ID.
Two of the 18 studies (Berger 2010; Lainchbury 2010) had three comparison arms comparing NP-guided treatment both to clinical assessment and to usual care. For usual care there were no scheduled visits and the participants were managed in primary care. Studies recruited 3660 participants ranging from 41 to 499 participants per study. The average age of participants in all the studies ranged from 62 to 80 years old. Studies followed up participants from baseline to between one and 54 months.  Table 2).
Two studies (Beck-da-Silva 2005; Li 2015), compared the e ect of NP-guided treatment with clinical assessment exclusively for the up-titration of beta-blockers. Beck-da-Silva 2005 changed the dose of bisoprolol, but all other drugs remained unchanged, during a three-month follow-up period. Li 2015 started and increased the dose of metoprolol succinate over one month; for these patients intravenous cardiotonic, vasodilator or diuretic was applied if signs or symptoms of heart failure were observed.
Beck-da-Silva 2005 was the only study to report an algorithm where medication (beta blocker) was decreased for patients whom the BNP measurement was increasing, but the clinical assessment was worse.
All, bar three studies (Eurlings 2010, Lainchbury 2010; Schou 2013), reported inclusion criteria for classifying participants according to the New York Heart Association (NYHA) functional classification. This classifies patients with heart disease into four stages based on limitations on physical activity, symptoms with ordinary physical activity and status at rest. Stage four indicating the highest severity of symptoms. At baseline, most studies grouped participants by

Excluded studies
Thirty-five references are included in the Characteristics of excluded studies tables where the title or abstract or both appeared to suggest a relevant study to this review. Of these 68% were excluded as the study was not a randomised control trial. Other reasons included not NP-guided treatment (20%), trial terminated, not treatment for heart failure, or not a baseline heart failure population.

Figure 3. 'Risk of bias' graph: review authors' judgements about methodological quality presented as percentages across all included studies. Allocation
All studies clearly stated the study was randomised, but not all studies reported on how randomisation was completed or if allocation concealment was achieved. Five studies confirmed sequence generation and allocation concealment and methods were judged to be at low risk of bias ( The remaining studies were classified as unclear.

Blinding
Blinding of participants and study personnel was only judged to be low risk if both were blinded to the treatment allocation; only one study met this standard (Lainchbury 2010). Five studies did not report or it was unclear whether participants or personnel were blinded to treatment allocation (Anguita 2010;Li 2015;Persson 2010;Shochat 2012;Skvortsov 2015). In all the remaining studies one or more of these groups were not blinded. Blinding of outcome assessments was not achieved or not reported in the majority of studies; only five studies blinded outcome assessment (Berger 2010;Eurlings 2010;Karlstrom 2011;Lainchbury 2010;Schou 2013).

Incomplete outcome data
For the primary outcome, all-cause mortality, eight studies (Anguita 2010;Berger 2010;Jourdain 2007;Li 2015;Schou 2013;Shah 2011;Skvortsov 2015;Troughton 2000) were judged to be low risk with regard to incomplete outcome data, in fact they all had no attrition except for Skvortsov 2015 where the numbers and reasons were fully reported. The remaining studies either did not report attrition, or the studies did confirm attrition with break down by intervention arm, but did not explain how missing data were handled. For those studies reporting dropouts, the overall attrition rates were no more than 23%.
All of the studies, bar four, completed intention-to-treat (ITT) analyses; Beck-da-Silva 2005 did not complete an ITT analysis, whilst Anguita 2010; Jourdain 2007 and Li 2015 did not report whether this method was used.

Selective reporting
Nine out of 18 studies reported on all stated outcomes and were considered low risk for reporting bias. Six studies have not yet reported on some secondary outcomes (Berger 2010 on heart failure mortality and all-cause admission, Eurlings 2010 on allcause admission, Persson 2010 and Maeder 2013 on quality of life, Schou 2013 and Shah 2011 on treatment costs). Lainchbury 2010 partially reported quality of life data. Skvortsov 2015 is currently awaiting further publications. It was not possible to assess reporting bias for Shochat 2012 as data were provided from conference abstracts and direct contact with the author and any pre-specified outcomes were not stated.
One study (Lainchbury 2010) was solely funded from a national research body and therefore considered at low risk of bias from the funding source.

E ects of interventions
See: Summary of findings for the main comparison Does treatment guided by serial BNP or NT-proBNP monitoring improve outcomes compared to treatment guided by clinical assessment alone? Sixteen studies (Anguita 2010;Beck-da-Silva 2005;Berger 2010;Eurlings 2010;Jourdain 2007;Karlstrom 2011;Krupicka 2010;Lainchbury 2010;Maeder 2013;Persson 2010;Pfisterer 2009;Schou 2013;Shah 2011;Shochat 2012;Skvortsov 2015;Troughton 2000) with 3292 participants recruited, reported results for all-cause mortality. Follow-up ranged from one month to four and a half years. However, data for Maeder 2013 was presented as survival curves and it was not possible to extract or obtain data for this study. Therefore meta-analysis was only possible for the remaining 15 studies: During the follow-up period, 265 (18%) participants died in the NP-guided treatment groups compared to 368 (22%) in the control groups. When the data were pooled for all studies using a fixed-e ect model, the evidence favoured the guided treatment groups, but overall the evidence showed uncertainty (risk ratio (RR) 0.87, 95% confidence interval (CI) 0.76 to 1.01; patients = 3169; studies = 15; low quality of evidence). Heterogeneity was low (I 2 = 16%).
The two studies that did not report results for all-cause mortality were Januzzi 2011 and Li 2015.

Heart failure mortality
(See Analysis 1.2) Only six studies (Jourdain 2007;Karlstrom 2011;Krupicka 2010;Li 2015;Skvortsov 2015;Troughton 2000) with 853 participants recruited reported results for heart failure mortality. In the NPguided treatment groups, 34 participants died and in the control groups 38 participants died due to heart failure, representing 8% and 9% respectively. Similar to all-cause mortality, the pooled result, using a fixed-e ect model, favoured the intervention, but overall, the evidence showed uncertainty (RR 0.84, 95% CI 0.54 to 1.30; participants = 853; studies = 6; low quality of evidence). The heterogeneity was low (I 2 = 21%).

All-cause admission
(See Analysis 1.4) Six studies (Beck-da-Silva 2005;Jourdain 2007;Karlstrom 2011;Schou 2013;Shah 2011;Troughton 2000) with 1142 participants recruited reported data for all-cause admission. During the follow-up, 304 (53%) participants experienced an event requiring admission in the NP-guided treatment groups. This compared to 327 (57%) participants in the control groups. The pooled results for all studies, with a fixed-e ect model, favoured the intervention, but overall, the evidence showed uncertainty (RR 0.93, 95% CI 0.84 to 1.03; participants = 1142; studies = 6; low quality of evidence). No heterogeneity was identified (I 2 = 0%). Lainchbury 2010 commented that no di erence was seen between intervention and control groups for all-cause admission, but the data were not provided.

Adverse events
(See Table 3) Six studies (Januzzi 2011;Krupicka 2010;Maeder 2013;Persson 2010;Pfisterer 2009;Troughton 2000) with 1144 participants reported number of adverse events during follow-up. Maeder 2013 did not report the number of adverse events broken down by intervention group, only as a total for the study. For the remaining five studies, the NP-guided treatment groups (511 participants) experienced 215 compared to 184 adverse events in the control groups (510 participants). Meta-analysis was not viable for this outcome since it was possible to have multiple events per individual. Therefore, the results have been tabulated. Quality of evidence was low.
Nevertheless, three studies (Januzzi 2011;Pfisterer 2009;Troughton 2000) commented there was no di erence between the NP-guided treatment and control groups: Januzzi 2011 reported that there was no significant di erences between the groups, whilst Pfisterer 2009 and Troughton 2000 reported P values greater than 0.05. Maeder 2013 reported the number of patients experiencing a serious adverse event did not di er between the groups. Two studies (Januzzi 2011; Krupicka 2010) reported a complete breakdown of the nature of the adverse events, whilst Pfisterer 2009 and Maeder 2013 only highlighted two areas (renal impairment and hypotension). For Maeder 2013, adverse events for renal failure were more frequent in the NP-guided group, where as events were less frequent for hypotension compared to the control group. However, both Januzzi 2011 and Pfisterer 2009 confirmed no di erence between the groups based on specific adverse events. Incomplete data meant it was not possible to comment on the most frequent types of adverse events.

Cost
Four studies (Berger 2010;Januzzi 2011;Maeder 2013;Pfisterer 2009) presented data on costs, two only as conference abstracts. It was not possible to pool results for these four studies because the outcome measure di ered for each study. Pfisterer 2009 reported on total overall costs per intervention arm: $20,949 for the NT-proBNP-guided treatment group versus $23,928 in the symptomguided group (control). Generally, costs were comparable, the main di erence occurred in the residency costs (staying in a nursing home or home for the elderly): $4157 in the NT-proBNP-guided treatment group versus $7564 in the symptom-guided group.
Januzzi 2011 examined the mean costs in the duration of the study. Overall costs for the NT-proBNP group totaled $35,262 ($451 per day) versus overall costs for the standard of care management (control) group of $42, 629 ($580 per day). Similar to Pfisterer 2009, the lower costs in the NT-proBNP group was predominantly due to Library Trusted evidence. Informed decisions. Better health.
Cochrane Database of Systematic Reviews inpatient costs. Januzzi et al concluded that costs were reduced by approximately 20% in the NT-proBNP-guided treatment group over the 10-month follow-up period.
In Berger 2010 an economic analysis was completed for a subgroup of participants (n = 190) who had complete follow-up data. This analysis suggested NP-guided treatment was cost-e ective and cheaper than in the usual care control group (for the multidisciplinary care control group this was cost neutral).
In contrast to the above three studies Maeder 2013 reported NPguided therapy as unlikely to be cost-e ective. Overall costs being $38,876 per patient for the NP-guided group compared to $21,419 per patient in the control group over 18 months.
Quality of evidence was low. participants recruited using the Minnesota Living with Heart Failure questionnaire. Lainchbury 2010 is only represented by one data set as data were only reported for the usual care control group. The pooled evidence for all studies, using a fixed-e ect model, marginally favoured NP-guided groups, but overall, the evidence showed uncertainty (mean di erence (MD) -0.03, 95% CI -1.18 to 1.13; very low quality of evidence). Heterogeneity was judged to be substantial (I 2 = 75%).

Quality of Life
Pfisterer 2009 also reported results for quality of life using the Short Form 12 and Duke Activity Status Index questionnaires; though not included due to incompatibility, both of these showed an improvement in both guided treatment and control groups with no di erences in the degree of improvement.
In Karlstrom 2011, changes in quality of life for participants was measured using the Swedish and Norwegian Short Form Health Survey 36; 68% from the NP-guided group and 74% from the control group completed the survey at both the start and end of the study. For these participants NP-guided treatment did not improve quality of life compared to clinical assessment alone.
Participants in Persson 2010 completed the Kanas City Cardiomyopathy Questionnaire at baseline and follow-up. This symptom score tool contains a quality of life element. In Persson 2010, the scores improved in both groups (+3.6 (SEM 1.65) in the NT-proBNP group and +6.2 (SEM 1.66) in the control group). There was no di erences between the groups (P = 0.28).

Subgroup analysis
Except for age, it was not possible to explore subgroups within the study populations. Data were reported for severity of heart failure, baseline NT-proBNP, target NT-proBNP, achieved NT-proBNP/BNP drop and gender, but generally only as totals, in varying categories, or as averages, for intervention and control groups (Table 1,  Table 2). Post hoc, consideration was given to subgrouping by le ventricular ejection fraction, (LVEF), but this too was not reported in an appropriate form (Table 1). All studies were completed under supervision of the hospital, except for Berger 2010 and Lainchbury 2010 where supervision was jointly in hospital and the community, and therefore subgroup analysis for this factor was not completed.
Subgroup analysis was only possible by age for three studies (Eurlings 2010;Lainchbury 2010;Shochat 2012) and only for the primary outcome of all-cause mortality (see Analysis 3.1). From the three studies, including Lainchbury 2010 with two control groups, there were 830 participants. For this analysis, the age threshold was set as equal or greater than 75 years old versus under 75 years old, though the data from Eurlings 2010 are reported marginally di erent as greater than 74 versus equal to or less than 74 years old. When the data from these three studies were pooled, the evidence showed uncertainty for either age subgroup. However, whilst showing uncertainty for either age subgroup the results suggest that for participants equal to or greater than 75 years old, the e ect favoured the control groups (RR 1.23, 95% CI 0.96 to 1.57; participants = 410; studies = 3) whilst for participants less than 75, the e ect favoured the guided-treatment groups ((RR 0.73, 95% CI 0.49 to 1.10; participants = 420; studies = 3) (Analysis 3.1).
Despite data not being available to pool, three further studies did comment on the age of participants in their results. Januzzi 2011 concluded for their study that 'no interaction between NT-proBNPguided care and age was found (P = 0.11)'. Persson 2010 commented 'levels of NT-proBNP tended to decrease more in patients younger than 75 years than in patients older than 75 years (change -2.4% ≥75 versus -20.3% <75 years, P = 0.06). Finally, Pfisterer 2009 reported that in the first six months the BNP levels decreased similarly for both guided treatment and control groups and were similar for participants under 75 and equal to or over 75 years of age. Though Pfisterer 2009 did state that "there was a significant interaction between treatment and age groups, i.e. patients aged ≥ 75 years in the NT-proBNP group had a smaller relative benefit on NT-proBNP levels (p = 0.04) and symptoms (p = 0.05) than younger patients". At eighteen months, the interaction between treatment and age was significant for mortality (P = 0.01, Cox regression adjusting for baseline characteristics) indicating that 'NT-proBNPguided treatment di ered significantly between younger and older patients'.
Post hoc subgroup analysis was carried out to explore whether data from two studies (Berger 2010; Lainchbury 2010) using usual care di ered to all other studies using clinical assessment as the comparator to NP-guided treatment (Analysis 2.1). This was only possible for two outcomes. For the primary outcome of all-cause mortality, the evidence showed very little di erence for either subgroup (
Post hoc we also explored the assumption that the two biomarkers were su iciently biologically and clinical similar to evaluate together. We investigated this by separating the pooled data by each biomarker. For all-cause mortality (Analysis 7.1), heart failure mortality (Analysis 7.2), all-cause admission (Analysis 7.4) and quality of life (Analysis 7.5), the pooled data for each biomarker showed uncertainty and were similar to the overall pooled result for each outcome. For heart failure admission, using a fixede ect model, the result grouping the trials by BNP ( 3. In view of the substantial heterogeneity we tested the robustness of this finding using a random-e ects model and found that the pooled result for studies using the BNP marker continued to favour NP-guided treatment but now showed uncertainty (BNP: RR 0.68, 95% CI 0.43 to 1.05; participants = 600; studies = 4; NT-proBNP: RR 0.65, 95% CI 0.48 to 0.89; participants = 1328; studies 6).

Sensitivity analysis
Risk of bias within the studies varied across the aspects of bias assessed. Blinding of participants and study personnel appeared to be poor (see Figure 2 and Figure 3), nevertheless, it was not always practical to blind participants and personnel in some studies. High risk in this category could still mean one party was blinded. Blinding of outcome assessment and attrition was judged to potentially impact on the pooled results. Sensitivity analyses were completed restricting studies to those with low risk of bias for blinding of outcome assessment (Berger 2010;Eurlings 2010;Karlstrom 2011;Lainchbury 2010;Schou 2013) and for attrition (Anguita 2010;Berger 2010;Jourdain 2007;Li 2015;Schou 2013;Shah 2011;Skvortsov 2015;Troughton 2000). For all outcomes, the analyses produced a similar e ect to the main findings (see Table 4). Though there was only one study (Karlstrom 2011) assessed as low risk for detection bias for heart failure mortality and therefore no comparison with the main findings could be made in this instance.

Summary of main results
We found the evidence for NP-guided treatment in patients with heart failure showed uncertainty for all-cause mortality or heart failure mortality. Furthermore, it showed uncertainty for all-cause mortality when examining subgroups under or over 75 years of age. Heart failure admission was reduced, but evidence for allcause admission showed uncertainty. In addition, the evidence showed uncertainty for NP-guided treatment improving quality of life. We were not able to pool results for adverse events and cost. All results were pooled from low-quality evidence except the outcome quality of life where the quality level of evidence was very low (see Summary of findings for the main comparison). The up-or down-titration of medication varied across studies in terms of the guidelines or algorithms used and changes in medication; neither was the reporting of NT levels consistent across studies. This meant we were unable to evaluate the impact of either of these for heart failure admission.

Overall completeness and applicability of evidence
Our review included 18 studies, which recruited 3660 participants. The age of the participants in the studies may have favoured younger patients as the average age of participants ranged from 62 to 80 years old; however, New York Heart Association (NYHA) functional classification varied su iciently across trials to ensure a broad range of severity. We were unable to assess a number of important subgroups; particularly, severity of heart failure at baseline, which may underpin an important e ect of NP-guided treatment on mortality outcomes. A systematic review in heart failure patients including 19 studies reported for each 100 pg/ mL increase in BNP there was an associated 35% increase in the relative risk of death (Doust 2005). Further to this, subgroup analysis of baseline NP, and NP decrease, which could underpin the mechanism of e ect, was not possible. In addition, a number of analyses were limited by lack of reporting: only six studies reported on all-cause admission, there were limited data on costs and only six studies reported on adverse events.

Quality of the evidence
All included studies were reported as randomised, but not all reported on the methods of randomisation. Eight confirmed allocation concealment and were judged to be at low risk of bias, and the other 10 were classified as unclear. Blinding was o en poorly done with only one study reporting blinding of both participants and study personnel to treatment allocation, and only five studies reported blinding outcome assessors. Fourteen studies reported outcomes on an intention-to-treat basis and attrition bias, eight studies were judged to be low risk as seven studies had no losses to follow-up, and the one fully documented the reported losses.
Using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, we assessed the quality of the evidence and GRADE profiler (GRADEPRO) was used to import data from Review Manager to create a 'Summary of findings' (SoF) table. For overall quality of evidence, the primary outcome plus heart failure mortality, heart failure admission and all-cause admission were judged to have low quality and quality of life was judged to be very low quality indicating low/very low confidence in the pooled result, but that the result could vary and is likely to be a ected Library Trusted evidence. Informed decisions. Better health.
Cochrane Database of Systematic Reviews by future research. The quality of evidence for adverse events and cost, which were not pooled, were also judged to be low. Quality of evidence was downgraded predominantly for limitations in the study design and/or inconsistency in the data.

Potential biases in the review process
Whilst we did perform a thorough search with no date or language restrictions, it is possible some studies may have been overlooked in searching and study selection. We were unable to include data from one study for the primary outcome. Whilst only 15 studies contributed data for the funnel plot for all-cause mortality, the graph does display a slight asymmetry with a lack of smaller studies showing a beneficial control e ect. This suggests the potential for publication bias (see Figure 4).

Agreements and disagreements with other studies or reviews
At least 12 reviews have been undertaken on the e ects of NP- Five of the seven previous reviews reported NP reduced all-cause mortality in heart failure patients and the other two, similar to this review, reported no e ect for all-cause mortality. No previous review has examined heart failure mortality as an outcome. Allcause admission was analysed in three of the previous reviews and no e ect was reported in agreement with our findings. Similar to this review, five previous reviews have reported an e ect favouring NP-guided treatment when examining heart failure admission and all reported a moderate level of heterogeneity. Two reviews examined adverse events and reported no reduction in events for NP-guided patients compared to clinical assessment. To date, no review has examined costs, and only one previous review (Xin 2015) has reported on quality of life (see Table 5).
The meta-analysis published in 2014, Troughton 2014, included individual patient data (IPD) from nine trials and aggregate data sets from two trials and reported no e ect in all-cause mortality. Though, with the advantage of IPD Troughton and colleagues were able to adjust for patient characteristics and used Kaplan Meier curves to compare time to all-cause mortality between NPguided and clinically-guided treatment groups and they reported a reduction in all-cause mortality (hazard ratio (HR) = 0.62; 95% CI, 0.45 to 0.86; P = 0.004, nine IPD studies). Similar to Porapakkham 2010, but again using time to event data, mortality was reduced in those under 75 years of age (HR 0.62; 95% CI, 0.45 to 0.85; P = 0.004), but not in those 75 years and older (HR 0.98; 95% CI, 0.75 to 1.3; P = 0.96), and the test of interaction between age and treatment e ect was significant (P = 0.028). Hospitalisation due to heart failure was reduced in patients with NP-guided therapy, both using time to event data (HR 0.80, 95% CI 0.67 to 0.94, P = 0.009), however, there was no e ect for all-cause hospitalisation using time to event data (HR 0.94, 95% CIs 0.84 to 1.07, P = 0.38).
While not directly comparable to this review, De Vecchis 2014 included six randomised controlled trials (RCTs) (n = 1775 patients) in a systemic review of BNP peptide-guided versus symptomguided therapy in outpatients with chronic heart failure. This review reported guided therapy decreased a composite outcome of mortality and heart failure hospitalisations during the follow-up period (odds ratio (OR) 0.64; 95%CI: 0.43 to 0.95; P = 0.028, I 2 = not reported).
Some subgroup analyses have been completed by previous reviews which can be compared to this review's subgroup analyses (see Table 6). Only Porapakkham 2010 is directly comparable to this review and similarly reported for all-cause mortality in patients over 75 years old an uncertain result. However, in patients under 75 years, unlike this review, Porapakkham 2010 reported a significant e ect for NP monitoring compared to clinical assessment.
Li 2013 reported heart failure admissions were reduced in patients with higher baseline BNP ≥2114 pg/mL (RR, 0.53; 95% CI, 0.39to 0.72; P < 0.0001, I 2 = 21.8%). Furthermore, Li 2014 completed sensitivity analyses to show a reduction in all-cause mortality and heart failure admission was especially seen in patients with reduced ejection function.
This review is consistent with previous reviews in all outcomes except all-cause mortality. For this outcome, the first (chronological) five reviews (Felker 2009; Porapakkham 2010; Li 2013; Savarese 2013; Li 2014) found a reduction, while Troughton 2014 found a reduction a er adjustment for patient characteristics. The latest systematic review by Xin 2015 reported no e ect on this outcome, similar to this review. One of the latest published trial (Schou 2013) reports higher all-cause mortality in the NP-guided group. The pooled estimate of e ect based on exclusion of this study shows a reduction in all-cause mortality similar to previous systematic reviews. Therefore, the inconsistency in this estimate leads us to suggest that further evaluation is required.

Implications for practice
This review confirms the evidence base to date, with at least four systematic reviews and one individual patient meta-analysis published, of the e icacy of NP-guided treatment e ects on heart failure admission. Our post hoc analysis for this outcome demonstrates that e ects are observed in shorter studies, less than two years in duration. This e ect observed in the shorter studies could reflect the severity of the disease process whereby many patients would be hospitalised or experience adverse events with NP-guided treatment having an impact delaying short-term outcomes.
Although previous reviews consistently report a reduction for allcause mortality, our review, the largest to date reports low-quality evidence that long-term, all-cause mortality and heart failure mortality show uncertainty. Furthermore, low-quality evidence showed uncertainty for all-cause admissions and very low quality of evidence showed uncertainty for quality of life outcomes.

Implications for research
There are a number of significant ongoing trials, therefore we do not perceive the need for any more until these have reported their results; but the significance around our results may change Cochrane Database of Systematic Reviews in the light of new data. We will update our review once these new trials are published, and we recommend updating the IPD analysis and using these data to perform cost-e ective analyses. Cost-e ectiveness data would aid decision making, particularly as length of hospital stay and preventing readmissions are important for the health service. In addition, it is important to clearly describe the components of the intervention and of the control group, as subtle changes in the control group in combination with a lack of blinding could have significant e ects on treatment escalation and the overall e icacy of the intervention. In case a future update identifies an e ect in mortality, the potential mechanisms for this e ect, such as increased patient and physician adherence to treatment regimens, would need to be explored.

A C K N O W L E D G E M E N T S
Many thanks to the following for their help in clarifying study information or providing further data: Dr Tariq       ; v) addition of an ARB, at recommended doses; vi) addition of chlorthalidone 50 mg/d; vii) addition of digoxin 0.25 mg/d or adjusted to renal function; viii) other drugs: nitrates, amlodipine. If the target BNP is achieved the patient will follow the same treatment regimen as prior to the visit until the next scheduled visit. 2. Control: Visits same as intervention without BNP data and additional visit at two weeks; treatment guided by less or greater Framingham score of two, recent events, questions to patient and medical history. If target score achieved the patient follow the same treatment regimen as prior to the visit until the next scheduled visit.  (13) Interventions 1. NT-proBNP-guided intensive management (BM): > 2200 pg/mL at hospital discharge; minimum six visits in first quarter, eight in first year and 8 to 26 visits overall; structured clinical assessment including NT-proBNP data at outpatient clinic; as long as NT-proBNP remained above 2200 pg/mL drug treatments were dictated by a flow chart until maximum or tolerated doses of HF drugs were established. If NT-proBNP fell below 2200 pg/mL 3 or 6 months after discharge then patients reverted to following the treatment schedule for the control group ( Inclusion criteria: European Society of Cardiology (ESC) diagnostic guideline criteria for acute HF, NT-proBNP levels at admission were required to be at least 1,700 pg/mL, NT-proBNP levels during hospitalisation were required to decrease more than 10%, with a drop in NT-proBNP levels of at least 850 pg/ mL, from admission to discharge Exclusion criteria: Life-threatening cardiac arrhythmias during the index hospitalisation, urgent invasive or surgical intervention performed or planned during the index hospital admission, severe COPD with a forced expiratory volume in 1 s (FEV1) of 1 l/s, pulmonary embolism less than 3 months prior to admission, pulmonary hypertension not caused by le ventricular systolic dysfunction (LVSD), a non-HF-related expected survival of less than 1 year, and patients undergoing haemodialysis or CAPD Interventions 1. NT-proBNP-guided treatment: minimum three visits in first quarter, six in first year and estimated 10 visits overall; structured clinical assessment including NT-proBNP data; individual patient NT-proBNP target value was set as the lowest level at discharge or at 2 weeks follow-up. If NT-proBNP levels were more than 10% with a minimum of 850 pg/mL above this individual target level, NT-proBNP level was considered "o -target," and therapy was intensified according to the ESC HF treatment guidelines. They report changes in 10 different medications. Except for calcium channel blockers, all changes in drug therapies concern the start or increase of medication or change in the type of medication. It was not specifically stated if no/any action was taken if the patient was below or at target. 2. Clincially-guided (control): Visits same as intervention without NT-proBNP data, treatment dictated by clinical assessment alone. Interventions 1. NT-proBNP-guided treatment: minimum two visits in first quarter, quarterly visits up to a maximum of 12 months (median number of visits for both arms was five); however scheduled visits were every two weeks until optimal/maximal medical therapy was achieved; structured clinical assessment including NT-proBNP data at outpatient clinic; if NT-proBNP levels were higher than 1000 pg/mL the drug therapy was intensified irrespective of clinical status; choice of medication therapy for either intervention arm was made by the physician according to consensus guidelines (American College of Cardiology foundation/American Association task force on practical guidelines); no algorithm for drug titration as used; once the patient achieved ≤ 1000 pg/mL (NT-proBNP-targeted optimal medical regimen) or if the target was not achieved but reached clear therapeutic limit then the patient will cease two weekly visits and revert to quarterly schedule. 2. Standard of care treatment (control): Visits same as intervention without NT-proBNP data, treatment dictated by clinical assessment and managed according to consensus guidelines. Once the patient achieves optimal medical regimen they will cease two-weekly visits and revert to quarterly schedule. Blinding of participants and personnel (performance bias) All outcomes

Characteristics of included studies [ordered by study ID]
High risk Unblinded "patients were made aware of their BNP value in order increase motivation to adhere to treatment" Blinding of outcome assessment (detection bias) All outcomes Low risk "All endpoints were adjudicated using a predefined endpoint protocol by a committee with two experienced cardiologists who did not participate in the study and were blinded to the results" Incomplete outcome data (attrition bias) All outcomes

Unclear risk Numbers provided, but not reasons
Selective reporting (reporting bias) Low risk All outcomes reported as specified in the publication Other bias Unclear risk Source of funding: Swedish Heart-Lung foundation, Regional research foundation in south eastern Sweden, regional foundation in northern Sweden, and by unrestricted grant from Biosite International and Infiniti Medical AB who supplied BNP analysing equipment Interventions 1. NT-proBNP-guided treatment: minimum two visits in first quarter, five in first year and nine overall ; structured clinical assessment including NT-proBNP data at outpatient clinic; general education regarding HF; treatment triggered by NT-proBNP level greater than 150 pmol/L and/or a HF score greater than 2, for values below this threshold, treatment was not altered a. Algortihm for heart score >2: i) increase frusemide to 120 mg/day or optimisation of ACE inhibitor dose if sub optimal; ii) addition of digoxin 0.25 mg/day adjusted for creatinine clearance; iii) add spironolactone (up to 50 mg/day) in patients with persisting class III or IV symptoms; iv) increase frusemide with twice-daily doses up to a maximum of 500 mg twice daily with doubling increments; v) addition of bendrofluazide or metolazone b. Algortihm for NT-proBNP >150 p/mol, heart score stable: i) optimisation of ACE inhibitor to trial-based doses; ii) addition or titration of beta blockade to trial-based doses; iii) addition of further therapy as for the clinically-guided group 2. Clinically-guided (CG, control): Visits same as intervention without NT-proBNP data; treatment determined by HF score above or below 2 a. Algorithm for heart score < 2: i) optimisation of ACE inhibitor dose; ii) addition and titration or optimisation of beta-blocker dose b. Algorithm for heart score > 2: same as NT-proBNP-guided treatment 3. Usual care (UC, control): No visit schedule or structured follow-up; management in primary care with or without requested HF clinic referrals

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias)

Low risk Stratified by age (≤75 or > 75) in permuted blocks of 30
Allocation concealment (selection bias)

Unclear risk Not stated
Blinding of participants and personnel (performance bias) All outcomes Low risk "double blind", "Patients will be blinded as to their group allocation, and clinical assessments will be made by a physician also blinded. Intensification of drug treatment will be made by an unblinded physician in the research team" Blinding of outcome assessment (detection bias) All outcomes Low risk "double blind" Incomplete outcome data (attrition bias) All outcomes

Unclear risk Numbers provided, but not reasons
Selective reporting (reporting bias) High risk Planned outcomes specified in protocol. No follow-up quality of life data for usual care (UC) control group. Analyses for two secondary outcomes were completed and commented on, but data were not provided. Interventions 1. NT-proBNP-guided treatment: minimum four visits in first quarter, six in first year and six overall ; structured clinical assessment including NT-proBNP data at outpatient clinic, treatment intensified until at least a 50% reduction from baseline NT-proBNP, stepwise treatment to Swedish guidelines: a. Patients with NYHA II: base therapy included an ACE-inhibitor and a betablocker, Loop diuretics could be added and used based on signs of fluid retention. In patients who did not tolerate ACEinhibitor treatment, an ARB was to be used instead. b. Patients with NYHA III-IV: base therapy as for NYHA II, in patients with persistent CHF symptoms despite target or maximum tolerated doses of ACE-inhibitor and beta-blocker, additional therapy with an ARB or spironolactone (or eplerenone in the case of hormonal side effects) could be initiated. In addition, digoxin could be added as an option for extra symptom relief, although the main indication for this treatment was atrial fibrillation. 2. Not NT-proBNP group (control): Visits same as intervention without NT-proBNP data; same stepwise treatment used based on clinical assessment only It was not specifically stated if no or any action was taken if the patient was below or at target. Blinding of outcome assessment (detection bias) All outcomes Low risk "vital status and admissions evaluated by an independent endpoint committee whose members were unaware of the study group assignments" Incomplete outcome data (attrition bias) All outcomes Cochrane Database of Systematic Reviews Trial stopped early due to poor enrolment

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias) Low risk "stratified by site with randomisation blocks of 6 through a central telephone centre" Allocation concealment (selection bias) Low risk Email by author 7 October 2014 "opaque envelopes were used" Blinding of participants and personnel (performance bias) All outcomes High risk "Clinicians were aware of the treatment allocation but were blinded to BNP levels in patients in the congestion score strategy arm. Patients were blinded to the randomisation arm." Blinding of outcome assessment (detection bias) All outcomes

Cochrane Database of Systematic Reviews
Intervention provider: Specialist (HF clinic)

Outcomes
Review relevant: i) All-cause mortality; ii) HF mortality (data not confirmed); iii) HF admission (data not confirmed); iv) All-cause admission (data not confirmed) Additional outcomes: i) Cardiovascular mortality

Bias Authors' judgement Support for judgement
Random sequence generation (selection bias) Low risk "randomised' by computer" Allocation concealment (selection bias) Unclear risk Email from author 12 November 14 "computer generated".
Blinding of participants and personnel (performance bias) All outcomes Unclear risk Email from author 12 November 14 "Patients and physicians blinded to group allocation. Study co-ordinator not blinded but did not participate in study process". Correspondence with author makes evaluation of bias unclear as it is not known if participants and clinicians were blinded to the monitoring process (intervention).
Blinding of outcome assessment (detection bias) All outcomes Inclusion criteria: Hospital admission due to acute decompensation HF, NYHA class III -IV at admission, LVEF < 40%, high risk at hospital discharge (> 1400 pg/mL NT-proBNP) Exclusion criteria: Participant unable or unwilling to provide written informed consent, inoperable aortic or mitral valve disease, coronary revascularisation (PCI or CABG) within the previous 3 months, acute myocardial infarction in previous 6 month, inflammatory myocardium disease, serum creatinine > 220 mkmol/mL, severe obstructive or restrictive pulmonary disease, high degree atrioventricular block, alcohol abuse, oncology Library Trusted evidence. Informed decisions. Better health.

Characteristics of ongoing studies [ordered by study ID]
Trial name or title NCT01685840 'GUIDE-IT'

Setting: USA & Canada
Duration of study: 12-24 months Inclusion criteria: ≥18 years old, LVEF ≤ 40% within 12 months of randomisation, High risk HF (HF hospitalisation, treatment in emergency department, outpatient treatment with intravenous diuretics in the prior 12 months) AND NT-proBNP greater than 2000 pg/mL or BNP greater than 400 pg/mL at any time during the 30 days prior to randomisation, willing to provide informed consent Exclusion criteria: Acute coronary syndrome or cardiac revascularisation procedure within 30 days, cardiac resynchronisation therapy (CRT) within prior 3 months or current plan to implant CRT device, active myocarditis, hypertrophic obstructive cardiomyopathy, pericarditis, or restrictive cardiomyopathy, severe stenotic valvular disease, anticipated heart transplantation or ventricular assist device within 12 months, chronic inotropic therapy, complex congenital heart disease, end stage renal disease with renal replacement therapy, non cardiac terminal illness with expected survival less than 12 months, women who are pregnant or planning to become pregnant, inability to comply with planned study procedures, enrolment or planned enrolment in another clinical trial

Participants
Number of participants at baseline: 1100 (all groups) Interventions 1. NT-proBNP-guided treatment: Visits every two weeks until optimal doses of therapies achieved, then every three months. Titration of HF treatment using guideline recommended therapies with a target of achieving and maintaining NT-proBNP level <1000 pg/mL 2. Usual care: Visit schedule same as for first arm. Ttitration of HF treatment based on target doses of evidence-based guidelines (American Heart Association and American College of Cardiology) Cochrane Database of Systematic Reviews Inclusion criteria: > 18 years old, HF diagnosed on a first hospitalisation for acute exacerbation during the last 12 months, without high age limit, minimal knowledge of the French language (patient or his relatives), informed written consent, resides or is treated in Ile de France, insured under the social security system Exclusion criteria: Myocardial infarction or revascularisation or heart valve surgery < 3 months, inability to execute the feasibility test, major cognitive disorders do not allow access to the platform, patient does not have the necessary autonomy to use the equipment, patient enrolled in another clinical trial, renal failure with creatininemia clearance (cockcroft) <15 mL/min 24h/day oxygen Inclusion criteria: Acute decompensated HF (either de novo or acute-on-chronic HF) and NT-proB-NP levels of N1,700 ng/L (ie, 200 pmol/L) measured within 24 hours of hospital admission Exclusion criteria: COPD with FEV1 of <1 L, pulmonary embolism within 1 month before admission and pulmonary hypertension not caused by le ventricle dysfunction, undergoing CAPD/ haemodialysis patients, planned coronary artery bypass gra (CABG), percutaneous coronary intervention (PCI), cardiac resynchronisation therapy (CRT), and/or valvular surgery before randomisation, cardiogenic shock at admission requiring invasive treatment, history of STEMI, CABG, PCI, CRTand/or valvular surgery within 1 month before admission, signed informed consent for any current interventional study, presence of severe noncardiac-related life-threatening disease before inclusion with an expected survival of < 6 months after inclusion, unwillingness to give or mental or physical status not allowing written informed consent, circumstances that prevent follow-up (no permanent home address, transient, etc)

Participants
Number of participants at baseline: Intervention 170; Control 170 Interventions 1. NT-proBNP-guided treatment: minimum three plus visits in first quarter, four plus in first year, four plus visits overall, structured clinical assessment including NT-proBNP data in hospital, when patients achieve over 30% reduction in NT-proBNP values hospital discharge and follow-up occurs. Under 30% NT-proBNP measurements triggers a drug algorithm: For patients with reduced ejection fractions: i) up-titration or addition of ACE inhibitor, β-blocker, and/or aldosterone antagonist; ii) CRT for patients who meet current guideline criteria; iii) electrical cardioversion for newonset atrial fibrillation; iv) coronary artery angiography (CAG) or intervention when ischemia is suspected. For patients with preserved ejection fractions: i) adequately treat hypertension and myocardial ischaemia; ii) ventricular rate control in atrial fibrillation; iii) electrical cardioversion for new-onset atrial fibrillation; iv) CAG or intervention when ischaemia is suspected 2. Conventional therapy (control): Discharge and follow-up of the patients can be planned at the discretion of the treating physician, physicians are discouraged from taking NT-proBNP measurements