Carperitide
alpha-human atrial natriuretic peptide, alpha-hANP, recombinant human ANP, ANP(1-28), atrial natriuretic factor 1-28
Carperitide is recombinant human atrial natriuretic peptide, the 28-amino-acid hormone released by atrial myocytes when they are stretched. It has been licensed in Japan since 1995 for acute decompensated heart failure and is used there very widely, but it has never been approved anywhere else. The evidence behind it is thinner than its usage suggests, and several large observational analyses have raised the possibility of harm.
Mechanism
Carperitide is recombinant alpha-human atrial natriuretic peptide, ANP(1-28), identical to the mature circulating hormone. Like BNP it contains the seventeen-residue disulfide-closed ring that defines the natriuretic peptide family, and like BNP it signals through natriuretic peptide receptor A (NPR-A/guanylyl cyclase A), raising intracellular cyclic GMP and activating protein kinase G.
The downstream effects are the physiological counter-regulation to volume overload: arterial and venous dilation lowering afterload and preload; increased glomerular filtration through preferential afferent arteriolar dilation and efferent constriction; inhibition of sodium reabsorption in the inner medullary collecting duct producing natriuresis and diuresis; suppression of renin and aldosterone secretion; reduced sympathetic outflow and reduced endothelin release. Preclinical work has additionally attributed antifibrotic and anti-hypertrophic effects to NPR-A signalling in the myocardium, and animal studies have reported reduced infarct size after coronary occlusion and reperfusion.
The distinction from nesiritide is essentially which family member is supplied. ANP is atrial in origin and more responsive to acute volume changes; BNP is predominantly ventricular and rises with sustained wall stress. Both are cleared by NPR-C-mediated endocytosis and by neprilysin, and both have very short circulating half-lives requiring continuous infusion.
What the research shows
The efficacy evidence is small in scale and inconsistent in direction. The PROTECT multicentre randomised study (Hata et al., 2008) allocated 49 patients with acute decompensated heart failure to low-dose carperitide infusion (n=26) or standard treatment (n=23) and reported significantly fewer deaths and rehospitalisations over 18 months (11.5% versus 34.8%, p=0.0359). That is a positive signal, but from 49 patients, far too few to be treated as an outcome result.
The observational evidence runs in the opposite direction. Matsue and colleagues analysed 1,037 Japanese patients with acute heart failure, of whom 402 (38.7%) received carperitide, and after propensity score matching (367 pairs) found in-hospital mortality significantly higher with carperitide, odds ratio 2.13, rising to 2.93 in elderly patients. A 2025 systematic review and meta-analysis by Shiga and colleagues pooled nine studies (four randomised trials and five propensity score-matched cohorts) and found higher in-hospital mortality in the carperitide group overall (OR 1.38, 95% CI 1.07-1.78), while the randomised trials alone showed no significant mortality difference, which is itself the clearest possible statement of how little randomised data exist. No significant differences were found in long-term mortality, heart failure rehospitalisation, hypotension incidence or length of stay, and the authors concluded that carperitide may not be superior to non-carperitide therapy.
The honest summary is that a drug used in a large fraction of Japanese acute heart failure admissions for three decades has never been tested in a trial capable of showing whether it helps or harms.
Evidence assessment
Mixed evidence
Confirmed as 'mixed' on audit. Carperitide does hold a regulatory approval, but in a single jurisdiction and granted in 1995 largely on haemodynamic and neurohormonal surrogate data from small studies. The randomised controlled trial base is genuinely thin. The frequently cited PROTECT study randomised only 49 patients, a figure verified directly against the published record. Meanwhile the largest observational analyses point the other way: a propensity-matched cohort of 1,037 patients found in-hospital mortality more than doubled with carperitide, and a 2025 meta-analysis of nine studies found higher pooled in-hospital mortality (OR 1.38) when propensity-matched studies were included, with the randomised trials alone showing no significant difference. All three citations were verified against PubMed. Small, conflicting human trials with an unresolved safety signal is the definition of mixed.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
Effects of carperitide on the long-term prognosis of patients with acute decompensated chronic heart failure: the PROTECT multicenter randomized controlled study Preclinical only
Death or rehospitalisation occurred in 11.5% of carperitide patients versus 34.8% of controls (p=0.0359). Low-dose carperitide in the acute phase, higher systolic blood pressure and beta-blocker use independently predicted better long-term cardiac outcomes. The sample size is far too small to support an outcome conclusion.
Carperitide is associated with increased in-hospital mortality in acute heart failure: a propensity score-matched analysis Preclinical only
Carperitide was associated with significantly higher in-hospital mortality (odds ratio 2.13), with a stronger association in elderly patients (odds ratio 2.93). The authors called for rigorous randomised trials of the drug's safety and efficacy.
Safety
Hypotension is the expected and most frequently reported adverse effect, an unavoidable consequence of potent vasodilation, and it is the mechanism most often invoked to explain the mortality signal seen in observational analyses, although the 2025 meta-analysis did not find a significant difference in recorded hypotension events between groups, which complicates that explanation. Bradycardia, headache and nausea are reported.
The unresolved question is whether carperitide causes net harm in unselected acute heart failure. Propensity-matched data suggest it might, particularly in elderly patients; randomised data are too sparse to adjudicate. That uncertainty has persisted for decades precisely because the drug was approved on surrogate endpoints and no one subsequently ran the trial that would have settled it. Carperitide is not indicated where hypotension or cardiogenic shock is present, and it interferes with ANP assays because it is the hormone being measured.
Carperitide is a hospital intravenous infusion requiring continuous haemodynamic monitoring in a critical care or coronary care setting. It has no legitimate use outside that context, and material offered under this name outside Japanese regulated supply is of unverifiable identity.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | Not licensed and not available. Carperitide has never held a UK or EU marketing authorisation and is not used in British practice. UK management of acute decompensated heart failure relies on loop diuretics with nitrates where vasodilation is required. |
| United States | Not FDA-approved. Carperitide has never been licensed in the United States. Nesiritide was the natriuretic peptide pursued in the US market. |
| WADA (sport) | Not listed on the WADA Prohibited List. Natriuretic peptides are not named in any prohibited class. |
Questions
It was developed and licensed there in 1995, at a time when regulatory expectations for acute heart failure drugs centred on haemodynamic improvement rather than clinical outcomes, and it became embedded in Japanese practice. No sponsor pursued approval elsewhere. In the US and Europe the natriuretic peptide that was developed instead was nesiritide, which was approved in the US in 2001 and then failed decisively in ASCEND-HF.
Carperitide is recombinant human atrial natriuretic peptide (ANP, 28 residues, atrial in origin); nesiritide is recombinant human B-type natriuretic peptide (BNP, 32 residues, predominantly ventricular). They share the same disulfide-closed ring architecture and the same receptor, NPR-A, and produce essentially the same haemodynamic and renal effects. The main practical difference is regulatory geography, not pharmacology.
That question has not been answered. The only randomised outcome data come from small studies, the largest positive one enrolling 49 patients. Large propensity-matched observational analyses suggest in-hospital mortality may be higher with carperitide, and a 2025 meta-analysis found pooled in-hospital mortality elevated (OR 1.38) when those cohorts were included. Observational data cannot prove harm (sicker patients may be selected for treatment), but after three decades of widespread use the absence of a properly powered trial is itself the striking finding.
No. It has no marketing authorisation in the UK, EU or US and is not obtainable through any regulated pharmaceutical channel outside Japan. It is an intensive-care intravenous infusion requiring continuous blood pressure monitoring, not a medicine that could be used in any other setting.