Frankincense (Boswellia): why the same mechanism can affect joints and gut at the same time

The resin extract of Boswellia serrata contains boswellic acids, above all 3-O-acetyl-11-keto-beta-boswellic acid, AKBA for short. In laboratory experiments, AKBA inhibited the enzyme 5-lipoxygenase, which forms leukotrienes from arachidonic acid.

Leukotrienes are messengers that attract white blood cells to a site and can amplify inflammation there. Leukotriene B4 has been studied particularly closely, a strong attractant for neutrophil granulocytes. In inflamed gut lining, a small tissue investigation from 1987 suggested that leukotriene B4 could be an essential attractant there. The authors themselves phrased it cautiously, and there were twelve tissue samples.

Important for context: this mechanism is well documented in the test tube. In human whole blood it could not be confirmed in the same way in a critical review, because the levels reached in blood lie clearly below the concentrations effective in the laboratory. Additional points of action are therefore being discussed, among them microsomal prostaglandin E synthase-1 and cathepsin G.

AKBA stands for 3-O-acetyl-11-keto-beta-boswellic acid. It is the best studied single compound in frankincense resin.

Structural investigations from Tübingen showed that the so-called 11-keto group in particular is decisive for the inhibition of 5-lipoxygenase. Boswellic acids without this group inhibited the enzyme only partially in the same experiments, other structurally related molecules not at all.

This is why serious manufacturers state the AKBA content, and the European Pharmacopoeia uses AKBA and KBA as markers for the quality of the resin. A product without any statement on content leaves you in the dark about what you are actually taking.

There are now several randomized, placebo-controlled studies on the knee. A 2020 meta-analysis pooled seven studies with 545 patients and found lower pain and stiffness scores compared with the control group, among them a weighted mean difference of minus 8.33 on the visual analogue scale. An analysis from 2024 with nine studies and 712 participants arrived at a comparable picture. It comes, however, from staff of a food supplement manufacturer and overlaps with the first one in the studies included, so it is not an independent second confirmation.

An investigation over 180 days additionally observed on MRI an increase in cartilage volume, cartilage thickness and joint space width compared with placebo. That is the most interesting single observation, but it comes from a manufacturer-related study and needs independent confirmation.

To be set against this is a large systematic review from 2018 that rated the quality of the evidence for dietary supplements in osteoarthritis overall as very low and found no clinically meaningful effects beyond the short term. Both belong in the same sentence.

The evidence is mixed. A small controlled investigation from 1997 with few participants reported remission rates in the same range as under the comparator sulfasalazine. I deliberately do not name the figures here as a comparison, because the study has clear weaknesses by today's methodological standards and was never replicated. It grounds a hypothesis, not a comparison with an approved medicine.

In active Crohn's disease the extract H15 met the study goal in a non-inferiority trial against mesalazine. H15 is, however, a medicinal product without a German marketing authorization that was obtainable only on medical prescription via individual import, so not the same as a frankincense capsule from online retail. Mesalazine and sulfasalazine are prescription medicines.

A large German study on maintaining remission in Crohn's disease over 52 weeks, by contrast, was stopped early because extract and placebo did not differ sufficiently. In collagenous colitis a small study was positive in the per-protocol result, but not in the stricter intention-to-treat analysis.

The honest summary therefore reads: frankincense might play a role in active inflammation, but it did not prove itself as a sole maintainer of remission in Crohn's disease. It is not a replacement for an established therapy and belongs under medical supervision in these conditions.

Because boswellic acids are fat-loving molecules and are poorly absorbed from the gut without fat.

In a crossover study in healthy men, the same capsule dose taken together with a standardized high-fat meal led to blood levels several times higher and to higher peak concentrations than in the fasting state. Two of the boswellic acids were detectable in blood at all only when taken with food. A later study confirmed the increase for 11-keto-beta-boswellic acid after a standardized meal.

Taking it on an empty stomach can therefore mean that even a high quality product stays without effect, simply because it never arrives. With this substance that is not a side issue.

First of all: in Germany, frankincense products are as a rule foods and not medicines. I am not giving an intake recommendation here, I am describing features that are documented in the studies.

Four things are worth a look. First the species: most studies were carried out with Boswellia serrata, not with other frankincense species, which have a different composition.

Second the standardization: a serious product states the content of boswellic acids and, separately, the share of AKBA in percent. Third the form: an extract is not ground raw resin, whose active substance share can be low and strongly varying.

Fourth the purity check, because natural resins vary in their composition and impurities can occur. A manufacturer who additionally advises taking it with a meal containing fat has engaged with the pharmacokinetics. Price differences often, but not always, reflect these points.

From the included studies, the 2020 meta-analysis derives a duration of use of at least four weeks. That is a figure from that analysis, not an intake recommendation from me.

In a study over 30 days with a strongly AKBA-enriched extract, differences in the pain scores appeared after five days. Studies on changes in cartilage structure on MRI, by contrast, ran for 180 days.

Roughly speaking: symptoms and inflammatory markers move within weeks, structural questions need months. If nothing at all happens after several weeks of consistent intake with a meal containing fat, that is a good occasion to rethink the matter with a physician instead of increasing the dose on your own.

In the controlled studies, frankincense extract was mostly well tolerated. Mild complaints in the gastrointestinal area were reported most often, for example in the 2003 crossover study on knee osteoarthritis.

In the 52-week study in Crohn's disease the safety assessment showed no disadvantages compared with placebo, even though efficacy could not be demonstrated there. A systematic review on frankincense describes only mild adverse effects across toxicology and clinical studies.

Good tolerability in studies with selected participants does not, however, mean harmlessness for every person and every situation. Allergic and skin reactions to natural resins are possible. Anyone with pre-existing conditions or taking medication should clarify this with a physician beforehand.

Systematic interaction studies in humans are rare, and that is precisely the point.

Boswellic acids act on arachidonic acid metabolism, which also plays a role in the function of blood platelets. In one pharmacological investigation, platelet aggregation was therefore explicitly carried along as a measure. Anyone taking anticoagulants, platelet aggregation inhibitors, immunosuppressants, biologics or an ongoing therapy for inflammatory bowel disease should discuss intake with a physician beforehand.

Second, there are laboratory indications that boswellic acids can influence enzymes of the cytochrome P450 system, through which a large share of all medicines is broken down. In an investigation in rats they inhibited CYP3A4 in liver microsomes and raised the blood levels of a co-administered antidiabetic drug. A cell study in a human liver cell line found, with the resin extract of a different Boswellia species, the opposite, namely an increased formation of such enzymes. Whether and in which direction this becomes relevant in humans is not settled. For people on immunosuppressants such as ciclosporin or tacrolimus, this point matters in particular.

The same applies before planned surgery and with multiple medication. Missing data are not proof of harmlessness, they are a reason for caution.

For pregnancy and breastfeeding there are no robust safety data for frankincense extracts. For this reason, intake during this period is usually advised against.

The same applies to children and adolescents, because the available studies were carried out almost exclusively in adults and dosages cannot simply be scaled down.

If you are pregnant, breastfeeding or planning a pregnancy, that is a clear situation for a medical conversation before you start any herbal product. Herbal does not automatically mean low risk, in many cases it simply means little studied.

No. For inflammatory bowel diseases and for inflammatory joint diseases there are established therapies with an evidence base that is considerably broader than the one for frankincense.

Stopping them on your own can trigger a flare and have consequences that are hard to catch up with. The 52-week study in Crohn's disease also shows concretely that frankincense did not convince as a sole maintainer of remission in this situation.

The sensible order is the reverse: first the base therapy in calm waters, then the question of whether a plant building block might play a supporting role, and that in conversation with the people treating you, who know your situation and your laboratory values.

More than was long assumed. In people with spondyloarthritis, a group of inflammatory spinal and joint diseases, an inflammatory bowel disease is often found, and even more often a so-called subclinical gut inflammation without typical bowel symptoms.

Both disease groups share numerous genetic risk factors and show changes in the gut flora. The direction of causality is open, however, and is discussed controversially in rheumatology, because some therapies perform well at the joint and not at the gut, and because a treatment that prevents immune cells from entering the gut can trigger joint complaints in some people.

For practice this means: with joint complaints a look at the gut is worthwhile and the other way around, without that turning into a simple cause and effect story. The connection is well documented. The explanation is not yet.