A real remedy for indigestion and oral bacteria that's being sold as a cure for H. pylori, which it isn't.
Mechanisms, studies and evidence grades for each claim above.
Like the orange juice entry, this one is an adjudication. Four posts, three voices, and the claims escalate as they go:
Dalton was also the maximalist voice in the orange juice exchange. The pattern repeats here: real studies, each stretched one or two inferential steps past what it measured. Alfred's post is more consequential, because its advice replaces a treatment. Kruse's is a different category of claim, dealt with briefly at the end.
↑ back to summaryMastic is the resin that bleeds from cuts in the bark of Pistacia lentiscus var. chia, a shrub that produces commercial resin essentially only in southern Chios. It is sold as raw "tears" to chew, as capsules (typically 350–500 mg), as an essential oil, and as "mastiha water." Its bioactive fractions are triterpenoids (masticadienonic and isomasticadienonic acids, oleanonic acid) and a volatile fraction dominated by α-pinene and myrcene. The genus matters for safety: Pistacia is the pistachio genus. See Cautions.
Chewed resin and swallowed capsules are different interventions. Chewing delivers the volatiles and the mechanical stimulus to the mouth, which is where the dental data come from. Capsules deliver triterpenoids to the stomach and gut, which is where the dyspepsia and IBD data come from. Most posts treat the two as interchangeable.
↑ back to summaryEvery PMID below was checked against its own abstract, not against the post that cited it.
| Claim | What the literature shows | Does it hold up? |
|---|---|---|
| "Cleared H. pylori in 38% of patients… Published in the NEJM" | The NEJM item is Huwez et al. 1998, a one-page correspondence reporting in-vitro killing of H. pylori (PMID 9874617). The 38% is Dabos et al. 2010, Phytomedicine (PMID 19879118): 52 patients across four arms. 1.05 g three times daily for 14 days eradicated 5/13 (38%). 350 mg tid eradicated 4/13. Mastic plus pantoprazole eradicated 0/13. Standard triple therapy eradicated 10/13. The mastic-only breath-test changes were not statistically significant (p=0.08 and p=0.064). | Does not hold. The number is real but comes from a different paper in a different journal. It rests on 13 patients, needed three times the stated dose, missed significance, and its mastic-plus-PPI arm cleared nobody. The "NEJM" label borrows the prestige of an in-vitro letter. |
| Mastic kills H. pylori in humans | Bebb et al. 2003 (PMID 12888582), from the Nottingham group that wrote the 1998 NEJM letter: 1 g four times daily for 14 days. "Mastic had no effect on H. pylori status in any of the eight completed patients", and breath-test values were unchanged (19.1 → 18.7, p=0.8). | Contradicted. The authors of the original in-vitro finding tested it in people at four times Alfred's dose and found nothing. None of the posts mention this study. |
| Useful alongside standard H. pylori treatment | Tulsian et al. 2026, AIIMS Rishikesh, single-blind RCT, n=64 (PMID 41832367). Bismuth quadruple therapy + mastic eradicated 85%, vs 67% for quadruple therapy alone, p=0.19. Dyspepsia scores improved more with mastic (p=0.001). | Plausible adjunct, unproven. This is the honest version of the H. pylori story: mastic with antibiotics, not instead of them. It missed its primary endpoint, but the symptom benefit is consistent with the dyspepsia data below. |
| Ulcer healing | Al-Habbal et al. 1984, double-blind vs lactose placebo, 38 duodenal-ulcer patients, 1 g/day for 2 weeks (PMID 6395994). Endoscopic healing 70% vs 22%; symptom relief 80% vs 50%, p<0.01. | Holds, but it is old and small. This is a genuinely striking result, and it is the trial the "1 g/day" dose actually comes from. It predates routine H. pylori testing, so it says nothing about eradication. It has not been replicated in the forty years since. |
| Bloating, reflux, nausea: "gut healing" | Dabos et al. 2010, randomised double-blind placebo-controlled, n=148, Rome II functional dyspepsia, 350 mg tid for 3 weeks (PMID 19961914). Marked improvement in 77% vs 40% (p<0.02), including heartburn and epigastric pain. Kleftaki et al. 2025, three-way crossover, 1.4 g/day for 4 weeks (PMID 41197785): improved bloating, epigastric burning and heartburn compared with a no-intervention period. | Holds: the best evidence on this page. The 2010 trial is properly blinded and reasonably sized. The 2025 crossover has no placebo, so its comparison flatters any treatment in a symptom-driven condition with a large placebo response. |
| "Pro-motility": gastroparesis | Singh et al. 2024, double-blind non-inferiority RCT vs levosulpiride, n=38 diabetic gastroparesis, 8 weeks (PMID 38533281). 4-hour gastric emptying on scintigraphy rose from 76.6% to 97.2%. The 24-week follow-up reports sustained symptom improvement (PMID 38454582). | Interesting, one small group. An objective scintigraphy endpoint is better than a questionnaire. But it is one centre and 19 patients per arm, with no placebo arm. It also reports an unexplained +1 g/dL haemoglobin rise, and implausibly broad "pleiotropic" wins (HbA1c, LDL). Worth replicating before relying on it. |
| Crohn's disease activity, gut inflammation | Kaliora et al. 2007: 10 patients, open-label, no placebo, 4 weeks. CDAI 223 → 136, with IL-6 and CRP down (PMID 17278198). The placebo-controlled follow-up, Papada et al. 2019 (n=60, 2.8 g/day, 3 months, PMID 30450689), had a between-group quality-of-life primary outcome. The abstract reports only a within-group improvement, plus lower faecal lysozyme and fibrinogen. | Weak. The dramatic CDAI drop comes from the uncontrolled pilot. The randomised trial reports secondary biomarkers and within-arm change, which usually means the primary between-group comparison did not come out. A 2023 systematic review calls for "more high-quality clinical trials" (PMID 37001176). |
| Reduces S. mutans and lactobacilli | Aksoy et al. 2006 (n=25, PMID 16343417) and 2007 (orthodontic patients, PMID 17029554). Chewing for 15 minutes cut salivary S. mutans and total bacteria vs placebo/paraffin gum, with the gap widening over 135 minutes. | Holds, for hours. This is a consistent, controlled effect. It is measured in saliva over two hours, not as caries incidence over years. |
| Reduces plaque and gingivitis | Takahashi et al. 2003, double-blind, placebo gum (PMID 12747455). Plaque index 2.69 vs 3.15 and gingival index 0.44 vs 0.66 over 7 days, in a trial where participants stopped brushing. The plaque arm had n=10 dental students. | Holds narrowly. The effect is real, and it was shown in an experimental-gingivitis model without toothbrushing. That is a sensitive model, not a demonstration that mastic adds anything to normal hygiene. |
| "Regenerates cavities" | Biria et al. 2009, Frontiers in Dentistry. An in-situ crossover: 15 adults wore appliances holding thin sections of lab-demineralised enamel and chewed 5 sticks a day for 3 weeks. Lesion area fell with both mastic and xylitol gum, "their effects were similar." | Overstated. "Regenerate cavities" implies reversing a cavitated lesion. The study shows remineralisation of early, non-cavitated demineralisation, which any saliva-stimulating gum does. Mastic matched xylitol and did not beat it. |
| Stimulates saliva | Assy et al. 2026, n=30 (PMID 42614181). Chewing mastic raised salivary flow more than chewing parafilm (p<0.0001), and even sniffing mastic or α-pinene volatiles raised flow. | Holds. This is the cleanest mechanistic finding on the dental side. The terpene volatiles add something beyond the mechanical effect of chewing. |
| "A potent antitumor agent… oral tumors effectively shrink" | The quoted phrase is the title of a 2011 study in Asian Pacific Journal of Cancer Prevention (PMID 22126583). It tested mastic extract on one oral squamous-cell carcinoma cell line (YD-10B) in a dish, and saw growth inhibition and caspase-3 activation. I found no oral-tumour animal study. The xenograft data that exist are colorectal and prostate tumours in immunodeficient mice. | Does not hold. "Tumors shrink" describes an experiment that, as far as I can find, was never done in the oral cancer context. Countless compounds kill cancer cells in culture at achievable-in-a-dish concentrations. It is the first rung of a ladder most candidates never climb. |
| Alertness and mood (the adolescent EEG pilot) | Magkou & Iordanis 2026, Journal of High School Science. The corresponding author is a secondary-school student, supervised by a biology teacher. n=30, 15 per arm, 3 minutes of chewing, unblinded, measured with a consumer EEG headband. Global Vigor rose within the mastic arm, and between-group differences were not significant. The mastic was the Chios Gum Mastic Growers' Association's own product. | Not evidence of a mastic effect. The authors say so themselves, which is to their credit. What it shows is that chewing makes people feel more alert, which the gum literature already knew. It is a well-reported school project, and it should be cited as one. |
Net: a gradient. The evidence is solid for oral bacteria, saliva and functional dyspepsia. It is suggestive for ulcer healing and gastroparesis. It is weak for IBD. It is absent or contradicted for H. pylori monotherapy and oral cancer. Across four posts, the claims with the weakest evidence got the most emphasis. (Moderate for dyspepsia and short-term oral antibacterial effects; Weak–Contested for everything else)
↑ back to summaryThis is the post that deserves the most care. It reached the most people, and it is the one where following the advice could cause harm.
Briefly, because there is little here to adjudicate. The claims are that chewing "liberates D+" (deuterium), that dentin acts as a piezoelectric "phased array" that clears deuterium from teeth and brain, and that this "close[s] the LES and kill[s] H. Pylori." No study I could find measures any of these links. There is no evidence that chewing changes tissue deuterium, that dentin piezoelectricity has any systemic effect, or that either kills bacteria in the stomach.
The one testable premise is backwards. "Fixes the GERD that allows H. pylori to grow" gets the epidemiology inverted. H. pylori infection is generally inversely associated with GERD. Some strains reduce gastric acid output, and reflux symptoms can appear or worsen after eradication. GERD does not create the conditions for H. pylori. (Speculative for the deuterium mechanism: no supporting data; Contested-to-contradicted for the GERD → H. pylori premise)
Deuterium depletion is a live topic on this site. The cancer paper engages with the mitochondrial and metabolic framing that some of Kruse's ideas grow out of. The problem is not that the framing is unorthodox. It is that this post attaches it to an intervention without a single measured step in between.
↑ back to summaryEvery juice entry on this index has an industry funder dominating its literature: POM Wonderful for pomegranate, the Florida Department of Citrus for orange juice. Mastic has the same structure. The Chios Gum Mastic Growers' Association, the cooperative that holds the Protected Designation of Origin and sells most of the world's mastic, appears as an author affiliation on the placebo-controlled IBD trial (PMID 30450689), on the MAST4HEALTH NAFLD trial (PMID 33629536), and on the 2025 dyspepsia crossover. That last paper's competing-interest statement declares none. The association's product was also the mastic used in the EEG pilot.
Two further points apply symmetrically. First, most of the positive human trials come from a small number of Greek groups, chiefly Harokopio University and Chios General Hospital. That is expected for a regional product, and it also means the literature has had little independent replication. The one independent group that tested the headline H. pylori claim in people is the one that found nothing. Second, the loudest promoter has a commercial stake too. Dalton's oral-health post ends in an affiliate call to action for a mastic brand ("comment GRECO"), and that brand's own website publishes a "what the research shows" page drawing on these same studies. As with orange juice, none of this makes a claim false. It is a reason to read the designs rather than the posts.
↑ back to summaryA useful oral and upper-GI adjunct that is being sold as a cure for the one thing it can't cure.
Verification status: every PMID on this page was checked against its published abstract rather than the posts citing it. That is how the following were found, and none of them appear in the four posts: the relocation of the "38%" from NEJM to Phytomedicine; the three-times-daily dose; the 0/13 mastic-plus-PPI arm; the null Nottingham trial; the within-group-only IBD result; and the one-cell-line basis of the "antitumor" quote. The Biria 2009 remineralisation study and the Magkou 2026 EEG pilot are not PubMed-indexed. The former was read from its published summary, the latter from the full PDF. Growers' Association affiliations were read from PubMed author records. Figures not independently verified: current H. pylori prevalence estimates and first-line eradication rates (standard guideline figures, not checked against a specific source here), and the absence of Chios-specific gastric-cancer data, which reflects a search that found nothing, not a confirmed null. No numbered reference list yet, consistent with the other entries in this index.