Paraguay

Tereré, Guaraní plant knowledge, stevia and the difference between a living tradition and a tested treatment.

Evidence at a glance

Cancer: Very hot drinks, including hot mate, have been linked to esophageal cancer risk; cold tereré is a different temperature exposure. IARC has not established that cold mate causes cancer, but that does not establish it prevents cancer. Neither tereré, lapacho nor stevia is a proven cancer treatment.

Diabetes and cardiovascular health: Human studies of stevia-derived compounds are mixed for blood pressure and largely negative for direct glucose lowering. A small mate-tea crossover study reported some short-term cardiometabolic changes, not fewer heart attacks or strokes.

Safety: Herbal mixtures are not one standardized product. A surveyed traditional medicinal plant belongs to the Aristolochia genus, whose aristolochic-acid-containing species raise serious kidney and cancer risks. Concentrated lapacho products may also pose bleeding concerns.

Evidence types and boundaries: UNESCO cultural documentation; a 100-person ethnobotanical study; randomized trials of isolated steviol glycosides and mate; official IARC/FDA/NCI safety assessments; laboratory studies. Many clinical trials were performed outside Paraguay. This is a selective, non-systematic evidence review, checked November 1, 2026.

Tereré and pohã ñana: culture before chemistry

A thermos of cold water, yerba mate, a shared vessel and a metal or cane straw can represent far more than a beverage. In Paraguay, tereré is closely connected with pohã ñana—knowledge and use of medicinal plants—and with Guaraní and Paraguayan community life. The herbs, preparation and meaning differ across families and practitioners. UNESCO inscribed the practices and traditional knowledge of tereré in the culture of pohã ñana on its intangible heritage list in 2020, emphasizing intergenerational knowledge, social connection and hospitality.

A 2018 ethnobotanical study by Kujawska and colleagues interviewed 100 Paraguayan migrants living in Misiones, Argentina, during 2014–2015 and documented 97 species added to hot or cold mate beverages. It described Allophylus edulis, Aristolochia triangularis, and species of Achyrocline among prominent examples. This is valuable evidence of reported use, not a controlled test of benefit or safety. It is also one migrant sample—not a description of every Indigenous nation, community or household in Paraguay.

Tradition is worth documenting on its own terms. The scientific question is separate: which precisely identified plant, in which preparation and dose, helped which condition in a well-controlled human study? Without that distinction, a long list of plant names can look like a list of proven treatments when it is not.

Research boundary: Cultural documentation tells us what was used and why it mattered. It cannot demonstrate an effect on cancer, hypertension, arthritis, bronchitis, diabetes or chronic pain. Any future field documentation should respect local knowledge holders, attribution, consent and benefit sharing.

Cancer: temperature risk is not the same as an herbal cure

1. Very hot mate versus cold tereré

The International Agency for Research on Cancer’s 2016 evaluation classified drinking very hot beverages (above approximately 65°C) as probably carcinogenic to humans (Group 2A), with the concern centered on esophageal cancer. This is a hazard classification supported by epidemiological and experimental evidence; it is not a claim that every hot-drink consumer will develop cancer or that botanical mate itself is the established cause.

Tereré is commonly prepared with cold water, so it does not deliver the same heat exposure. IARC reported insufficient evidence to classify mate consumed at temperatures that are not very hot as carcinogenic (Group 3, not classifiable). Importantly, Group 3 does not mean proven cancer-safe or cancer-protective. Tobacco and alcohol also matter substantially to esophageal cancer risk; observational studies require care about confounding.

2. Laboratory effects do not establish anticancer treatment

Lapacho/pau d’arco, derived from the inner bark of several related South American trees, is sometimes promoted for cancer. Memorial Sloan Kettering’s evidence review describes anticancer activity in laboratory experiments, but no demonstrated clinical cancer-control benefit. It also reports no improvement with isolated lapachol in an older chronic myelocytic leukemia investigation. Small oral-mucositis studies with a particular lapacho-based product address treatment side effects, not tumor shrinkage or survival; they cannot be generalized to commercial bark tea.

A 2014 original laboratory study of Lafoensia pacari, a bark used in Brazilian and Paraguayan traditional settings, observed changes in cultured tumor cells, including apoptosis-related signals. That paper does not report an effective cancer regimen in humans. Extracts, cell concentrations and routes matter; neither a cell-culture response nor a plant’s traditional use establishes efficacy or safety in a person receiving cancer treatment.

Practical interpretation: No verified randomized evidence from these materials establishes that tereré herbs, stevia, lapacho or Lafoensia pacari prevent recurrence, shrink tumors or extend survival. Do not replace surgery, radiation, systemic treatment or evidence-based supportive care with an untested botanical.

Diabetes: stevia is a sweetener, not a proven glucose-lowering medicine

Stevia rebaudiana—ka’a he’ẽ—has long-standing ties to Paraguay and Guaraní use. This history is distinct from modern products containing isolated, purified steviol glycosides such as stevioside and rebaudioside A. Whole leaves, homemade infusions, crude extracts and highly purified sweeteners are not interchangeable.

Paraguay-based randomized study, 2008: Barriocanal and colleagues, affiliated with the National University in Asunción, reported a double-blind, placebo-controlled study of stevioside at 250 mg three times per day for three months in groups with type 1 diabetes, type 2 diabetes or normal/low-normal blood pressure. The published abstract (PMID 18397817) found no significant glycemic, HbA1c or blood-pressure benefit attributable to treatment. The accessible abstract does not provide reliable per-group randomized and analyzed denominators, so these are not verified here. This regimen was research exposure, not a suggested dose.

Larger diabetes trial, 2008: A 16-week double-blind randomized trial (PMID 18555575) assigned 60 adults with type 2 diabetes to 1,000 mg/day of rebaudioside A and 62 to placebo. HbA1c changes were 0.11% versus 0.09% (p=0.355), with no significant between-group differences in fasting glucose, blood pressure or body weight. The report was assessed from its original abstract, not full text.

Independent formulation and comparator, 2020: An open-access randomized trial in Iran assigned 39 adults with type 2 diabetes to stevia-extract-sweetened tea or sucralose-sweetened tea; 34 completed eight weeks (15 stevia, 19 control). Between-group glucose, HbA1c, insulin and lipid differences were not significant. The comparison was against another non-sugar sweetener, not a test of replacing sugary drinks in everyday life. The original article reports support from the National Nutrition and Food Technology Research Institute and no declared author conflicts.

What this means: Using a high-purity stevia sweetener in place of added sugar can change the sugar content of a drink; that does not make stevia an antidiabetic medication. Do not infer glucose control, diabetic complication prevention or cancer prevention from its sweetness or from cell studies.

Hypertension: an encouraging result, then important qualifications

Two earlier placebo-controlled studies in Chinese adults with hypertension reported reductions in blood pressure using isolated stevioside capsules. A one-year trial (PMID 10971305) randomized 106 participants (60 stevioside, 46 placebo), testing 250 mg three times a day. A two-year trial (PMID 14693305) enrolled 174 adults with mild essential hypertension; 168 completed, with 500 mg three times daily used in the treatment group. In the two-year report, average treatment-group systolic pressure changed from 150 to 140 mmHg and diastolic from 95 to 89 mmHg, with favorable comparison against placebo. These values came from primary paper abstracts; complete risk-of-bias extraction and all adverse-event denominators were not verified.

That is a genuine clinical signal, but its interpretation is narrower than “stevia lowers blood pressure.” The plant-derived compound was given in concentrated capsules at research doses, not as ordinary sweetened tereré. Different populations, formulations and trials—including the negative Paraguay study—have not shown uniform effects. None of these studies establishes prevention of strokes, heart attacks or heart-failure admissions. Blood-pressure medication should not be stopped or substituted based on these findings.

Yerba mate itself: short-term cardiometabolic markers

A 2025 original randomized controlled crossover study in Spain included 52 nonhabitual mate consumers (25 with normal cholesterol and 27 with elevated cholesterol). Over an eight-week intervention, participants consumed three servings of mate made from approximately 9 grams of leaves per day or a decaffeinated isotonic control drink. Some blood-pressure and inflammatory-marker measures improved after mate consumption. LDL cholesterol decreased in those with normal cholesterol; among those with high cholesterol, control-drink changes were similar, weakening the claim of a mate-specific lipid effect. Total body weight did not significantly change. The study was supported by Spanish public research funding; the authors declared no conflicts and registered the trial retrospectively (NCT06729905).

This trial tested a particular yerba mate tea preparation outside Paraguay. It did not test cold tereré with added pohã ñana herbs, did not measure cardiovascular events and is the same trial already discussed in our Argentina chapter, not independent corroboration. Short-term improvements in laboratory markers must not be recast as proven protection from chronic disease.

Respiratory, autoimmune and other chronic illnesses: where the evidence stops

Traditional accounts associate some herbal preparations with digestive complaints, colds, bronchitis, inflammation, painful joints and other chronic symptoms. These uses deserve culturally attentive documentation, but the 2018 plant survey did not randomly assign people to treatments or establish effects on chronic obstructive pulmonary disease, asthma, arthritis, autoimmune disease or neurologic conditions. An antimicrobial result in a dish likewise cannot prove relief from a human respiratory infection.

For chronic illnesses, the absence of a well-conducted trial is neither proof that a tradition has no value nor proof that it works. It is a specific evidence gap. Future studies should identify the plant to species level, document how it is prepared, compare with appropriate care, use meaningful symptom and functional outcomes and actively collect adverse events. For people receiving immunosuppressive or cancer treatment, unidentified multiple-herb mixtures create additional medication-interaction uncertainty.

Safety: the most consequential details

Aristolochia exposure deserves a firm warning. The 2018 Paraguay-related ethnobotanical survey lists Aristolochia triangularis. The US National Cancer Institute identifies aristolochic acids in many Aristolochia plants as a cause of serious kidney injury and urinary-tract cancers. The concentration in the species or preparations described in that survey was not measured in the source reviewed. The prudent recommendation is to avoid Aristolochia-containing remedies; cultural significance does not neutralize this established genus-level hazard.
Stevia formulation and regulation: The FDA has not questioned GRAS notices for certain high-purity steviol glycosides under stated food-use conditions, but does not consider whole stevia leaf or crude extracts GRAS for use as sweeteners in conventional foods. This is a US food-regulatory distinction, not a judgment about a culture or a claim that a sweetener treats disease.
Lapacho/pau d’arco: MSK reports nausea, vomiting and possible increased bleeding risk when combined with anticoagulant or antiplatelet medicines; the clinical significance of that interaction is uncertain. Bark preparations vary by species and extraction method.
Yerba mate and mixtures: Mate contains caffeine; stimulant effects may be relevant to palpitations, anxiety, insomnia, pregnancy and some medications. Mixed-herb tereré adds risks that cannot be inferred from the caffeine or mate-leaf trial alone. Very hot servings introduce a distinct thermal cancer concern; cold tereré does not reproduce that specific temperature exposure.
Care coordination: Discuss supplement labels, herb species, preparation and timing with the oncology or chronic-disease team, especially before surgery, during chemotherapy, on anticoagulants, with kidney impairment, or while taking blood-pressure or glucose-lowering medicines. Unverified interactions should be described as uncertain, not automatically absent.

The studied quantities above describe published research conditions, not dosing advice. Seek medical assessment rather than attempting to self-treat concerning symptoms or a suspected cancer.

Questions worth bringing to the care team

Is this an ordinary food or beverage, a purified steviol glycoside, a concentrated bark extract or a mixture of plants with unknown species?
Is the hoped-for outcome a symptom, blood-pressure reading, glucose level, inflammatory marker or an actual cancer outcome? Has that specific endpoint been tested in people?
Does a trial compare this exact formulation against placebo or usual care, and do the results represent meaningful differences rather than only changes from baseline?
Could the preparation interact with anticoagulants, antiplatelet drugs, cancer therapy, blood-pressure medicines or diabetes treatment?
Who holds the cultural knowledge being described, and has that context been represented accurately and respectfully?

Explore related records in the Evidence Explorer, the Cancer-Control Research collection, and our methods. This page is educational, not individual medical advice. Neither laboratory results nor traditional use substitute for verified human treatment evidence.

Primary sources and official assessments

UNESCO (2020) — inscription of tereré and pohã ñana traditional knowledge; heritage documentation, not a clinical efficacy study.
Kujawska et al. (2018), DOI 10.1155/2018/6849317; PMID 29725356 — original survey of 100 Paraguayan migrants and 97 plants; published original abstract consulted; full interview-level data not independently extracted.
IARC Monographs Volume 116 (2016) — authoritative cancer-hazard assessment of very hot beverages and mate.
Barriocanal et al. (2008), PMID 18397817, DOI 10.1016/j.yrtph.2008.02.006 — randomized stevioside study from Paraguay; original abstract reviewed.
Maki et al. (2008), PMID 18555575, DOI 10.1016/j.fct.2008.05.007 — 122-person randomized rebaudioside A diabetes trial; original abstract reviewed.
Ajami et al. (2020), PMCID PMC7103435 — 39 randomized / 34 completed stevia versus sucralose diabetes study; full text examined.
Chan et al. (2000), PMID 10971305 and Hsieh et al. (2003), PMID 14693305 — original randomized stevioside hypertension reports; original abstracts reviewed.
Bravo et al. (2025), PMID 40263915, DOI 10.1002/mnfr.70065 — full original crossover mate trial; shared with Argentina chapter, not an independent second trial.
Lafoensia pacari cell-line study (2014), PMID 25311274 — original preclinical experiment; no human cancer efficacy data.
NCI, aristolochic acids (reviewed 2022); MSK, pau d’arco; FDA, stevia and sweeteners — risk interpretation and product distinctions.

Source search and page appraisal: November 1, 2026. Original-paper abstract access is clearly differentiated from full-text review; no clinical trial results were inferred from registrations or plant-use surveys. Findings may change as new human studies emerge.