This page follows a source whose reuse terms permit translation, preserving its structure and line of argument in English.
About this resource
This page closely follows the retrospective observational study published by Thompson, Malcolm, and Tegzes in 2022. Existing medical literature had been weighted toward severe intoxication cases; this study instead examined sessions documented by a single kambô practitioner in routine work and described commonly observed physical responses in relation to the number and location of application points.
This was not a prospective clinical trial confirming safety. The absence of a “serious adverse event” in the records does not mean that kambô is generally safe. The study concerned ritual application of unfractionated skin secretion to burn points and must be distinguished from cell and animal research on individual peptides.
Study design
The authors retrospectively analyzed records of kambô rituals conducted in the southwestern United States from August 1, 2018 through the end of June 2019. The dataset contained 147 unique participants and 241 sessions. Mean age was 38.9 years, and 100 participants (68%) were women. 93 people had records for 1 session, 23 for 2, 26 for 3, 5 for 4, and 1 for 5 sessions.
Demographics, motivations, and medical history came from pre-session forms; the practitioner recorded the number and location of points and physical responses during each session. Facial swelling and sweating used graded scales, while syncope, shaking, immediate defecation, and similar observations were recorded as present or absent. Because no dedicated validated scales existed, the study used its own categories and practitioner ratings.
The practitioner screened in advance and normally excluded people who reported severe cardiovascular disease, pregnancy, epilepsy, impaired liver or kidney function, severe mental illness, and other specified conditions. A test point was used for first-time participants, who were instructed to drink 1–2 L of water immediately beforehand. Rapé or sananga was sometimes offered before the session, but the records did not identify who used them.
Content
Participants could provide multiple motivations: 90 (61.2%) named spiritual purposes, 64 (43.5%) physical purposes, and 27 (18.4%) psychological purposes. These were reasons for participation, not medical indications or assessments of effect. Some participants self-reported chronic disease or infection, but the study did not independently verify diagnosis or measure disease-specific outcomes.
The mean number of points per session was 4.2 ± 2.2, the median was approximately 4, and the range was 1–11. The paper analyzed point count as a proxy for “dose,” but did not measure point size, the amount of secretion placed on each point, drying state, source animal, or peptide concentration. Equal point counts do not necessarily mean equal actual exposure.
Most sessions used a single application site, while 36 sessions involved multiple sites. The authors descriptively compared point count and site with records of facial swelling, sweating, syncope, shaking, defecation, and other responses. There were no biologic measurements such as blood pressure, heart rate, electrolytes, or kidney and liver function.
Results
The principal observed responses were facial swelling in 70.5%, sweating in 53.3%, immediate defecation in 45.2%, and syncope in 10.4%. Sessions involving multiple application sites had higher ratings of facial swelling and sweating than those involving a single site. Participants receiving more points had more facial swelling (p < 0.001) and sweating (p = 0.008), whereas differences in syncope or shaking by point count were small.
No serious sequela requiring hospitalization or death occurred in the recorded sample. The authors nevertheless argued that the observed responses themselves carried risk: facial swelling could affect breathing, syncope could cause injury from falls, intense vomiting could cause aspiration or esophageal injury, and water intake could contribute to hyponatremia.
The study suggested that point count may be associated with some physical responses, but because actual chemical amount was not measured, this is not a rigorous dose–response relationship. The absence of serious events might also reflect conservative practitioner selection, the test point, periods when co-used substances were avoided, and the scope of records rather than an intrinsic safety of kambô.
The observations focused primarily on responses readily visible during the session. No outcomes compared participants’ subjective benefits, changes in disease, or longer-term physical health, so ritual efficacy cannot be evaluated from these results.
Limitations
These were retrospective records from a single practitioner, without a control group, randomization, or blinding. The practitioner both rated participants and participated in the research, creating potential observation and recording bias. The high proportions of women and people reporting autoimmune disease may reflect the practitioner’s background and participant network, and the sample cannot be generalized to other regions or Indigenous practices.
Point count was a crude proxy for exposure, and secretion composition was not measured. Response scales were neither standardized nor validated, and individual co-use of rapé or sananga was not recorded. Without blood tests, blood pressure, or ECG, asymptomatic electrolyte abnormalities could not be detected. Guidance emails were sent after 2 days, but delayed events that participants did not report were absent from the data.
It is possible to state “0 serious events” as an observation within 241 sessions, but not to prove that rare events do not exist, establish a safety rate, or demonstrate benefit in treating disease.
Safety
Facial swelling, vomiting, sweating, and defecation—frequent in this study—are related to shifts in fluid and electrolytes, and syncope can cause immediate trauma. When heavy water intake, vomiting, and antidiuretic effects coincide, the danger of hyponatremia documented in prior case reports cannot be ignored. Because presession selection occurred, the study provides no safety information about excluded high-risk people.
The paper did not establish a safe number of points, home administration, methods of combining substances, or emergency response. Frequencies in practitioner records should not be used as a guide for self-administration.
Source and rights
Original source: Thompson C, Malcolm B, Tegzes J. Use of Phyllomedusa bicolour secretion during kambô ritual: observational responses, dosage, and risk of adverse events. Toxicology Communications. 2022;6(1):6–12. DOI 10.1080/24734306.2021.2006524.
The original article is published under Creative Commons Attribution 4.0 (CC BY 4.0). This page follows the study methods, principal numbers, and limitations identified by the authors without reproducing photographs, tables, participant forms, or original prose verbatim.