Soursop Leaf for Blood Sugar and Blood Pressure

Evidence tier: preliminary (animal only) for both claims. Rodent studies do report lower fasting glucose and modest blood-pressure reduction with Annona muricata leaf extract, and the proposed mechanisms are reasonable. There are no adequate randomised human trials of soursop leaf for type 2 diabetes or hypertension.

This page is written differently from a normal benefits article, and deliberately so. Soursop leaf tea for diabetes and high blood pressure is an enormously common use in West Africa, Indonesia and the Caribbean, and the population using it is largely people already taking metformin, insulin, an ACE inhibitor or a calcium-channel blocker. For that reader, the useful framing is not "here is a benefit to pursue" — it is "here is an interaction to expect". An animal-tier hypoglycaemic effect is not strong enough to recommend, but it is plausible enough that combining it with glucose-lowering medication deserves thought.

There is also the matter that this use produces exactly the exposure pattern of greatest concern: a chronic condition invites a daily, indefinite tea habit, which is what the neurotoxicity article is about. Diabetes and hypertension are lifelong. So is the exposure, if you treat them this way.

The Claim and Where It Comes From
Blood Sugar: What the Studies Actually Found
Proposed Glucose Mechanisms
Why Streptozotocin Rats Are Not People
Blood Pressure: What the Studies Actually Found
Proposed Blood-Pressure Mechanisms
The Interaction Risk, in Practical Terms
The Metformin Problem Specifically
Recognising Hypoglycaemia and Low Blood Pressure
Why Substitution Is the Real Danger Here
The Fruit, the Potassium and the Sugar
Options With Better Evidence
If You Are Going to Use It Anyway
Key Research Papers
Connections

The Claim and Where It Comes From

Across the tropics, soursop leaf tea is one of the most commonly named plant remedies for both diabetes and high blood pressure. In Nigerian, Ghanaian and Cameroonian ethnobotanical surveys, Annona muricata appears repeatedly for both indications. In Indonesian Jamu practice, daun sirsak is taken for hypertension, gout and "high uric acid". In the Caribbean, guanábana or corossol leaf tea is a household remedy for blood pressure and "nerves".

Two features of that pattern are worth noting before looking at the evidence:

The traditional indication is broad and old, and its geographic consistency across continents that had no contact for most of history suggests people were observing something real — though "something real" could be sedation, diuresis, a placebo effect in a symptom-poor condition, or a genuine metabolic action.
Neither diabetes nor hypertension produces symptoms you can feel improving. This matters enormously. A remedy for pain gives immediate feedback; a remedy for a number gives none. A person can drink soursop tea for years while their HbA1c climbs and feel that it is working. This is why the substitution risk is higher for these two conditions than for almost anything else on this site.

Blood Sugar: What the Studies Actually Found

The glucose literature on Annona muricata is almost entirely rodent work, and almost all of it uses the same model.

The typical design. Rats are given streptozotocin, a compound that destroys pancreatic beta cells and produces a diabetes-like state, then treated with aqueous or ethanolic leaf extract for a period of days to weeks, with fasting glucose, body weight, insulin and various oxidative-stress and organ markers measured. Comparison arms sometimes include a standard drug such as glibenclamide or metformin.

What has been reported. Reduced fasting blood glucose relative to untreated diabetic controls; improvements in markers of oxidative stress; in some studies, histological reports of preserved or partially restored pancreatic islet architecture; and improvements in lipid parameters. Florence and colleagues reported antidiabetic and antioxidant effects of an aqueous Annona muricata extract in streptozotocin-induced diabetic rats in the Journal of Ethnopharmacology in 2014, which is among the more frequently cited examples. Adewole and colleagues reported morphological changes and hypoglycaemic effects on pancreatic beta cells in streptozotocin-treated diabetic rats in an African biomedical journal in the mid-2000s.

Also reported, in vitro: inhibition of α-glucosidase and α-amylase, the intestinal enzymes that break dietary carbohydrate into absorbable sugars. This is the same target class as the drug acarbose.

What has not been reported: an adequately designed randomised controlled trial in people with type 2 diabetes, with HbA1c or fasting glucose as a primary endpoint, adequate numbers, blinding and a control arm. Searching for one returns nothing that meets that description. Evidence tier: preliminary (animal and in vitro).

Proposed Glucose Mechanisms

Four mechanisms are proposed in the literature. They are worth understanding because they differ in how plausible they are and in what they would imply for a person.

Carbohydrase inhibition. Slowing α-glucosidase and α-amylase reduces the rate at which starch becomes glucose, blunting post-meal peaks. This is the most straightforwardly plausible mechanism: it acts in the gut lumen, needs no systemic absorption, and has a licensed drug analogue. It would, however, mainly affect post-meal glucose rather than fasting glucose — which does not match the rodent findings especially well.
Beta-cell protection. The claim that leaf extract protects or regenerates pancreatic beta cells rests on antioxidant activity offsetting streptozotocin's oxidative damage. This is plausible in the streptozotocin model specifically, because that model's injury is oxidative. It is a much weaker basis for type 2 diabetes in humans, where beta-cell failure arises from chronic metabolic stress, amyloid deposition and lipotoxicity rather than from an acute oxidative insult.
Improved peripheral glucose uptake. Sometimes proposed, rarely demonstrated with adequate methods in this literature.
Reduced hepatic glucose output. Occasionally proposed. Note the awkward implication: this is broadly metformin's mechanism, and metformin achieves part of it through mild complex I inhibition — the same target the acetogenins hit far more potently. Any glucose-lowering that arises from mitochondrial inhibition is not a benign mechanism to be pleased about; it is the toxic mechanism showing up as a metabolic effect. See the metformin section.

That last point is the one this page most wants a reader to notice. A glucose-lowering effect mediated by respiratory-chain inhibition is not a nutrient effect. It is closer to what happens in metabolic poisoning, and it is not something to seek out at an unknown dose.

Why Streptozotocin Rats Are Not People

Nearly the entire glucose literature rests on one model, so its limitations are the literature's limitations.

Streptozotocin diabetes is closer to type 1 than type 2. It works by chemically destroying beta cells, producing insulin deficiency. Human type 2 diabetes is primarily insulin resistance with progressive beta-cell failure. An intervention that protects cells against an acute chemical toxin may do nothing about insulin resistance.
An antioxidant advantage is built into the model. Streptozotocin injures through reactive oxygen species, so almost any antioxidant-rich plant extract looks good in this system. Dozens of plants show "antidiabetic" activity here; very few survive human testing.
The doses are large. Extract doses are commonly hundreds of milligrams per kilogram of body weight. Scaled naively to a human, that is grams of concentrated extract daily — far more than a cup of tea, and far into the range where the neurotoxicity concern dominates.
The duration is short. Weeks. Diabetes is managed over decades, and a drug's value lies in what it does to complications over that span, which no rodent study addresses.
Endpoints are surrogates. Fasting glucose in a rat is not HbA1c in a person, and neither is retinopathy, nephropathy or cardiovascular death — the outcomes that actually matter.

Blood Pressure: What the Studies Actually Found

The blood-pressure literature is smaller and also rodent-based.

The most commonly cited work is by Nwokocha and colleagues in Pharmaceutical Biology in 2012, examining possible mechanisms of the hypotensive effect of Annona muricata in normotensive Sprague-Dawley rats. Related work has examined leaf extract in anaesthetised rat preparations, measuring arterial pressure and heart rate directly.

What has been reported: modest, dose-related reductions in arterial blood pressure, generally attributed to peripheral vasodilation rather than to changes in heart rate or cardiac contractility. Some studies report the effect persisting after autonomic blockade, which points toward a direct action on vascular smooth muscle or endothelium.

What has not been reported: a randomised controlled trial in people with hypertension, with ambulatory or office blood pressure as a primary endpoint. Evidence tier: preliminary (animal).

One further honesty point specific to this claim: a reduction in blood pressure in a normotensive animal is not the same finding as treating hypertension. Human antihypertensives are judged by whether they reduce strokes, heart attacks and kidney failure over years — not by whether a number moves in a healthy rat over an afternoon.

Proposed Blood-Pressure Mechanisms

Direct vasodilation — relaxation of vascular smooth muscle, possibly through calcium-channel effects, which would put it loosely in the same functional category as a calcium-channel blocker.
Endothelial nitric-oxide-mediated effects — proposed, not firmly demonstrated in this plant.
Diuresis — many traditional "blood pressure teas" are mild diuretics, and drinking a cup of hot water increases urine output on its own. This is an underrated explanation for the perceived effect of a great many herbal infusions.
Sedation. Soursop leaf's best-attested traditional use is as an evening sedative (see the sleep article). A calmer, sleepier person has lower blood pressure while calm. That is a real physiological effect and not a trivial one, but it is not treatment of hypertension — the pressure returns with the day.

The plausible mechanisms here are mostly mild and non-specific. That is consistent with the traditional reputation and consistent with the modest rodent findings. It is not consistent with the idea that leaf tea can replace an antihypertensive.

The Interaction Risk, in Practical Terms

This is the section that matters most for the actual audience of this page. If soursop leaf has any glucose- or pressure-lowering activity in humans — and animal data suggest it might — then the practical consequence is additive effects with medication that does the same thing. Nobody is monitored for that, because nobody tells their doctor about tea.

With glucose-lowering medication:

Insulin and sulfonylureas (glibenclamide/glyburide, gliclazide, glimepiride, glipizide) can cause hypoglycaemia on their own. Adding an unquantified glucose-lowering agent on top increases that risk. These are the two classes where an added hypoglycaemic effect can actually be dangerous rather than merely untidy.
Metformin — see the next section; there is a mechanistic reason beyond additive glucose lowering.
SGLT2 inhibitors, DPP-4 inhibitors and GLP-1 agonists carry lower intrinsic hypoglycaemia risk, but any additive effect still confounds dose adjustment.
The confounding problem cuts both ways. If you start drinking the tea and your glucose changes, neither you nor your prescriber can tell whether your medication needs adjusting or the tea is doing it. Then if you stop the tea — you go on holiday, the packet runs out — the change reverses without anyone knowing why.

With blood-pressure medication:

Additive hypotension with ACE inhibitors, ARBs, calcium-channel blockers, beta-blockers, alpha-blockers and diuretics. The practical consequence is orthostatic dizziness — light-headedness on standing — which in an older person is a fall risk, and falls are far more immediately dangerous than a slightly raised blood pressure.
Diuretics plus a mildly diuretic tea can nudge fluid balance and electrolytes, which is worth knowing if you already have low potassium or sodium.

General principle: the risk here is not that soursop leaf is a powerful drug. It is that it is an unquantified one. Every legitimate medication comes with a dose you can adjust. Leaf tea comes with a variable, unstated, unmeasurable amount of active material, which makes it impossible to manage sensibly in combination with something that is being titrated.

The Metformin Problem Specifically

This deserves separating out because it is a mechanistic issue and not just additive arithmetic.

Metformin's principal action is understood to involve mild inhibition of mitochondrial complex I in the liver, which reduces hepatic gluconeogenesis and shifts the cellular energy balance in a way that activates AMPK-dependent signalling. That mild, reversible, hepatically concentrated complex I inhibition is central to how the most widely prescribed diabetes drug in the world works — and metformin is a genuinely good drug with decades of outcome data.

Annonaceous acetogenins inhibit the same enzyme complex, far more potently, without the tissue selectivity. Metformin is a hydrophilic cation that requires specific transporters to enter cells, which is why its action concentrates in liver and gut and why it does not accumulate in brain. Acetogenins are lipophilic, enter cells passively, cross the blood–brain barrier and accumulate in fatty tissue.

Two conclusions follow:

Stacking them is unattractive on mechanism alone, quite apart from any additive glucose lowering. If you are already taking a drug that gently inhibits complex I where you want it inhibited, adding a potent one that inhibits it everywhere is the wrong direction.
It reframes the "benefit". If part of soursop's glucose-lowering in rodents is respiratory-chain inhibition, then the metabolic effect and the neurotoxic effect are the same effect measured in different tissues — which is the theme running through this entire hub. A glucose reduction achieved that way is not a benefit at an unknown dose.

Recognising Hypoglycaemia and Low Blood Pressure

Practical, because if you are taking medication and start a tea, these are the things that would tell you something is happening.

Low blood sugar — typically below about 4 mmol/L (roughly 70 mg/dL), and important if you are on insulin or a sulfonylurea:

Shakiness, sweating, a pounding or racing heart.
Sudden hunger, nausea, pallor.
Anxiety or irritability that arrives out of nowhere.
Difficulty concentrating, slurred speech, blurred vision.
At night: sweating through the sheets, vivid dreams, waking with a headache. Relevant here because soursop tea is usually drunk in the evening.
Action: take fast-acting carbohydrate immediately, recheck in fifteen minutes, and tell your prescriber it happened. Severe hypoglycaemia with confusion or loss of consciousness is an emergency.

Blood pressure too low:

Light-headedness or a grey-out on standing up.
Unusual fatigue, blurred vision, difficulty concentrating.
Falls, or near-falls, particularly getting out of bed at night.
Action: sit or lie down, and mention it to your prescriber alongside what you have started taking. Do not simply stop prescribed antihypertensives on your own.

Why Substitution Is the Real Danger Here

The most likely serious harm from soursop leaf in a diabetic or hypertensive reader is not an interaction. It is replacement — and it is uniquely easy to do with these two conditions, for a reason worth stating explicitly.

Diabetes and hypertension are largely asymptomatic until they cause damage. Nothing hurts. There is no cough to notice easing, no pain to notice returning. You cannot feel an HbA1c of 9%, and you cannot feel a blood pressure of 165/100. So a person who stops metformin and drinks tea instead feels exactly the same — often better, because they have stopped a drug with gastrointestinal side effects and started an evening ritual they enjoy. The absence of any felt consequence is read as evidence that the change was fine.

Meanwhile, the damage that these conditions do is silent and cumulative in the same way:

Uncontrolled glucose damages the retina, the kidneys and peripheral nerves over years. Diabetic neuropathy and retinopathy are not reversible once established.
Uncontrolled blood pressure raises the risk of stroke, heart failure and kidney disease over the same timescale, with no warning symptom before the event.
By the time either condition announces itself, the harm is done and the treatment decision cannot be revisited.

Both conditions are also monitored by numbers, which is the good news: HbA1c and a home blood-pressure cuff will tell you what is actually happening regardless of how you feel. If someone insists on trying a plant remedy, the minimum reasonable condition is that they keep measuring — and there is a note on that in the last section.

The Fruit, the Potassium and the Sugar

The fruit is a genuinely different question from the leaf, and it comes up often enough with these two conditions to be worth a short section.

Soursop pulp is a moderately sugary fruit — on the order of 66 kcal per 100 g, mostly carbohydrate. As whole fruit in a varied diet, unremarkable. As a large glass of champola or es sirsak made with condensed milk and syrup, it is a substantial sugar load, and someone managing diabetes should count it as one. Whole fruit beats juice for the same reason it always does: the fibre and the chewing slow absorption.
It contains useful potassium and magnesium, both of which are genuinely relevant to blood pressure and both of which are better obtained from a varied diet of fruit and vegetables than from any single item.
Fruit is the lowest-acetogenin part of the plant, which is why the safety concerns on this hub apply to it least.
Eating the fruit is not a treatment for either condition. It is food. That is not a criticism — it is the whole point.

Options With Better Evidence

Since the honest verdict on soursop leaf here is "animal-tier evidence and a neurological hazard", it is only fair to note where a reader could put the same effort with better return. None of these is a substitute for prescribed treatment either, and none should be started without telling a prescriber.

The interventions with the strongest human evidence are not herbs. For glucose: dietary carbohydrate quality and quantity, weight reduction where relevant, and physical activity — walking after meals has measurable effects on post-meal glucose. For blood pressure: sodium reduction, potassium-rich whole foods, weight, alcohol reduction, exercise and sleep. These have outcome data that no botanical comes close to.
Among herbs with human trial data for glucose, cinnamon, fenugreek and bitter melon have at least been tested in people, with mixed and modest results — which is still a tier above soursop leaf.
For blood pressure, hibiscus and garlic have small human trials behind them, again with modest effects.
Crucially, none of these carries a documented neurodegeneration signal. That is the comparison that matters. If a reader wants a tea to drink nightly for years, this is a strong reason to pick one of the others.

If You Are Going to Use It Anyway

People make their own decisions, and a page that only says "don't" is not useful to someone who has already decided. Harm reduction, then, in order of importance:

Do not stop or reduce prescribed medication. This is the one that actually determines whether you come to harm. Add, never replace, and change doses only with your prescriber.
Tell your prescriber or pharmacist. Bring the packet. They need it in your record, particularly if you are on insulin or a sulfonylurea.
Keep measuring. Home glucose monitoring, an HbA1c at the usual interval, a home blood-pressure cuff. Numbers are the only honest feedback these conditions provide.
Do not use it daily and indefinitely. This is where this page and the neurotoxicity page meet. Chronic conditions invite chronic use, and chronic use is the implicated exposure. If you use it, use it occasionally, not as a permanent nightly fixture.
Choose the lowest-exposure form. A brief infusion of leaf rather than powdered-leaf capsules or a tincture. Never seeds. Never a "standardised acetogenin" extract.
Do not use it at all if you have any movement disorder, a family history of parkinsonism, a mitochondrial condition, or are pregnant or breastfeeding.
Watch for orthostatic dizziness and hypoglycaemia in the first couple of weeks, as described above.

Key Research Papers

Per this site's citation policy, each reference links to a PubMed topic search rather than a numeric identifier, so a transcription error cannot silently point at the wrong paper. Titles, journals and years are stated in the text where they are known with confidence; where they are not, the finding is described and a topic search is given instead.

Florence and colleagues on antidiabetic and antioxidant effects of an aqueous Annona muricata extract in streptozotocin-induced diabetic rats, Journal of Ethnopharmacology, 2014. Animal. PubMed topic search
Adewole and colleagues on morphological changes and hypoglycaemic effects of Annona muricata leaf aqueous extract on pancreatic beta cells of streptozotocin-treated diabetic rats, mid-2000s. Animal. PubMed topic search
Nwokocha and colleagues on possible mechanisms of the hypotensive effect of Annona muricata in normotensive Sprague-Dawley rats, Pharmaceutical Biology, 2012. Animal. PubMed topic search
In-vitro work on Annona muricata inhibition of α-glucosidase and α-amylase, the carbohydrate-digesting enzymes targeted by acarbose. In vitro. PubMed topic search
Studies of Annona muricata leaf extract on lipid profile and oxidative-stress markers in diabetic rodent models. Animal. PubMed topic search
The clinical-trial gap for this indication: searching for randomised human trials of Annona muricata in type 2 diabetes or hypertension returns no adequately designed study. PubMed topic search
West African ethnobotanical surveys recording Annona muricata for diabetes and hypertension — the source of the traditional claim. Traditional use documentation. PubMed topic search
The literature on metformin's mechanism as a mild mitochondrial complex I inhibitor — essential context for why stacking a potent one is a poor idea. PubMed topic search
Lannuzel, Michel, Höglinger and colleagues, The mitochondrial complex I inhibitor annonacin is toxic to mesencephalic dopaminergic neurons by impairment of energy metabolism, Neuroscience, 2003 — the toxicity that has to be weighed against any metabolic benefit. Cell culture. PubMed topic search
Champy and colleagues, Quantification of acetogenins in Annona muricata linked to atypical parkinsonism in Guadeloupe, Movement Disorders, 2005 — how much acetogenin a daily infusion habit represents. PubMed topic search
The general literature on herb–drug interactions with antidiabetic and antihypertensive medication, including additive hypoglycaemia and hypotension. PubMed topic search
Reviews of the streptozotocin rat as a model of human diabetes, and of its limitations as a screening system for antidiabetic plant extracts. PubMed topic search

Live PubMed Searches

Connections

Safety and disclaimer. This page is health information, not medical advice. The glucose- and blood-pressure-lowering effects of soursop leaf are supported by animal studies only; there are no adequate human trials, and this is not a treatment for diabetes or hypertension. Never stop or reduce prescribed medication in favour of a herbal tea — both conditions damage eyes, kidneys, nerves, heart and brain silently, with no symptom to warn you. If you take insulin, a sulfonylurea or any antihypertensive, expect the possibility of additive effects and tell your prescriber or pharmacist before adding anything. Soursop leaf also carries a documented neurological hazard: long-term daily use is the exposure pattern of concern, so a chronic condition is the worst possible reason to start a permanent daily habit. Avoid entirely in pregnancy, breastfeeding, any movement disorder, or a known mitochondrial condition. See the safety article.

Back to Table of Contents