Neem

Azadirachta indica A.Juss. is the accepted name in Kew’s Plants of the World Online (POWO), first published by the French botanist Adrien de Jussieu in the Mémoires du Muséum d’Histoire Naturelle (1830, issued 1831). Linnaeus had described the tree in 1753 as Melia azadirachta, and POWO lists 14 synonyms. Its origin is debated. POWO gives the native range as Assam to Indo-China (Assam, Bangladesh, Myanmar, Thailand, Laos, Cambodia, and Vietnam) and treats the rest of India, Sri Lanka, Pakistan, and Nepal as introduced, along with much of West and East Africa, Iran, the Gulf, the Caribbean, Mexico, and Hawaii. EPA describes it as native to arid regions of India, LiverTox to India and Sri Lanka, and a 1992 US National Research Council (NRC) report to Assam and Burma, while noting the uncertainty. The NRC estimated 18 million neem trees on the Indian subcontinent, and over the past century or so the tree was carried to West Africa, the Caribbean, and Fiji; near Mecca, a forest of 50,000 neems was planted to shade pilgrims on the plains of Arafat. Older literature sometimes confused neem with its relative Melia azedarach, the chinaberry or Persian lilac.

The NRC report describes a broad-leaved evergreen up to 30 m tall, with a rounded crown as much as 20 m across, a short trunk with thick, strongly furrowed bark, and deep roots that throw up suckers. It drops its leaves only in extreme drought and tolerates heavy lopping. The compound, feathery leaves, which the NRC compares to those of walnut or ash, carry toothed, curved leaflets and taste intensely bitter. Small white flowers with a honey-like scent grow in loose clusters from the leaf axils and draw many bees. The fruit is a smooth, olive-like drupe up to almost 2 cm long that turns yellow or greenish yellow when ripe, with sweet pulp around a seed whose kernel holds most of the active compounds. Trees fruit after three to five years, can yield up to 50 kg of fruit a year, and may live for more than two centuries. Neem thrives in heat above 50 °C and on poor, dry soils with 400 to 1,200 mm of rain a year, but it cannot survive frost or waterlogging. The medicinal parts are the leaves, bark, twigs, flowers, seeds, and seed oil.

Neem’s chemistry is dominated by limonoids. The NRC names azadirachtin, salannin, meliantriol, and nimbin as the most important for pest control, with azadirachtin responsible for about 90% of the effect on most insects; kernels contain on average 2 to 4 mg of azadirachtin per gram, with a high of 9 mg reported from Senegal. Nimbidin, the main bitter principle of alcohol extracts, makes up about 2% of the kernel. Azadirachtin was isolated in 1968, according to Veitch and colleagues (Angewandte Chemie 2008), and its structure is so complex that chemists took more than 20 years to synthesize it (Fernandes and colleagues, Fitoterapia 2019). The seed oil, which can be up to 50% of the kernel’s weight, is mostly triglycerides of oleic, stearic, linoleic, and palmitic acids, with sulfur compounds that give the garlic smell. LiverTox notes that free fatty acids account for about 20% of the oil and bitter compounds such as nimbin and nimbidin for about 2%, and that these form during storage, so commercial samples vary. More than 140 compounds have been isolated from the tree, including the limonoids gedunin and nimbolide, and the bark contains about 14% tannins.

Neem belongs to the oldest layer of Indian medicine. The NRC report notes that the earliest Sanskrit medical writings refer to the benefits of its fruits, seeds, oil, leaves, roots, and bark, that each has been used for centuries in Ayurveda and Unani medicine, and that the classical Ayurvedic texts attributed to Charaka and Sushruta mention neem for malaria. Brahmachari (ChemBioChem 2004) describes its uses as known in India since Vedic times. Leaves were applied as poultices or decoctions for boils, ulcers, and eczema, and the oil for ringworm and slow-healing sores; LiverTox adds that small doses of the oil were taken for asthma, intestinal worms, arthritis, and leprosy. Farmers mixed neem leaves into grain stores to keep weevils away, so generations ate traces of neem every day, and the NRC records that Mahatma Gandhi ate a neem-leaf chutney despite its bitterness. Neem honey is prized in parts of Asia.

The twig toothbrush is one of neem’s most enduring uses. The NRC describes millions of people in India and Africa breaking off a twig every morning, chewing the end into a brush, and scrubbing their teeth and gums, with neem the most popular choice; neem later became the active ingredient in popular toothpastes in Germany and India. Folk practice spread with the tree. In Haiti crushed leaves were rubbed into maggot-infested wounds, in India and Bangladesh villagers put neem oil on the hair to kill lice, and in Nigeria and Haiti neem-leaf teas were drunk for malaria. The NRC also records a darker side: in parts of Asia neem oil was given to small children as a cure-all, much as castor oil once was in the West, and the deaths of a few young children in Malaysia in the 1980s were linked to it. Medical reports from Malaysia, Singapore, India, and Sri Lanka have described the same poisoning since 1981.

Modern science arrived through pest control. Indian scientists reported in 1928 and 1929 that ground neem kernels repelled desert locusts, and in 1959 the German entomologist Heinrich Schmutterer saw neem trees standing green after a locust swarm in Sudan stripped everything else. The US Department of Agriculture began testing neem in the 1970s, and in 1985 EPA approved a commercial neem insecticide, Margosan-O, for certain nonfood uses. India also became the scene of a landmark fight over traditional knowledge. The European Patent Office had granted the US Department of Agriculture and the company W.R. Grace a patent on a neem-oil fungicide; Vandana Shiva’s Research Foundation for Science, Technology and Ecology, the European Parliament’s Greens, and the International Federation of Organic Agriculture Movements challenged it, and the BBC reported that it was revoked in 2000 and the appeal lost in March 2005, on the grounds that the use had long been known in India.

Neem’s best-understood effects are on insects. The NRC report explains that azadirachtin resembles ecdysones, the hormones that control molting, and blocks their release, so treated larvae cannot molt and die without breeding; it also makes plants so distasteful that many leaf-eating insects will starve rather than feed. EPA’s fact sheet summarizes azadirachtin as deterring feeding and interfering with feeding, molting, mating, and egg laying, and notes that clarified neem oil, with the azadirachtin removed, also controls fungal diseases such as mildews and rusts. Head lice are a different story. Al-Quraishy and colleagues (Parasitology Research 2015) found that the neem seed extract shampoo Licener, made after the oil had been pressed out, killed lice by flooding their breathing tubes and suffocating them within 3 to 10 minutes, a physical effect similar to that of silicone lice products rather than a hormonal one.

Most other mechanisms come from the laboratory. The NRC cites neem preparations that were toxic to 14 kinds of disease-causing fungi in culture, including ringworm fungi and Candida, and neem oil that suppressed Staphylococcus aureus but not many other bacteria. Nagini and colleagues (Medical Research Reviews 2024) review the limonoids azadirachtin, gedunin, epoxyazadiradione, and nimbolide, which slow cancer cell growth and block inflammatory signaling in cells and animals; MSKCC stresses that there are no human studies showing neem treats cancer. In animals, a bark extract reduced stomach acid, an effect attributed to blocking the stomach’s proton pump, and MSKCC notes that a neem extract inhibited the liver enzymes CYP3A4/5, CYP2C8, and CYP2C9 in the test tube, with unknown clinical relevance. Sharma and colleagues (Journal of Ethnopharmacology 2026) conclude that poor absorption, inconsistent extracts, and a lack of rigorous human trials still block neem’s translation into medicine.

The toxicity and fertility effects are better characterized than the benefits. LiverTox attributes neem oil poisoning to mitochondrial damage: in isolated rat mitochondria the oil uncoupled the energy-producing chain, and in mice it caused fatty liver and kidney changes resembling the Reye-like illness seen in children. Whether a component of the oil or a contaminant is responsible is unresolved; mice given neem oil at up to 1.6 g per kg a day for four weeks showed little toxicity, and the NRC reported that oil from clean, mold-free kernels was not toxic to rats at 5,000 mg per kg. On reproduction, Patil and colleagues (Phytomedicine 2021) reviewed 27 studies showing that neem leaf, seed, bark, and flower extracts reduced sperm production, blocked ovulation, altered hormones, or ended pregnancies in animals, effects that appeared temporary and reversible. Mukherjee and colleagues (Contraception 1996) found that a purified neem seed extract, Praneem, given by mouth for six days ended established pregnancies in baboons and bonnet monkeys. The NRC reports that neem oil made human sperm motionless within 30 seconds in the laboratory.

Gum health and plaque are where neem has the most human data. Dhingra and Vandana (International Journal of Dental Hygiene 2017) systematically reviewed neem mouthwashes and found only three randomized trials; all reported that neem rinses used with toothbrushing were about as effective as chlorhexidine, the standard antiseptic rinse, at reducing plaque and gum inflammation, but the studies were too different to pool and their methods and reporting were flawed. A broader review by Furquim dos Santos Cardoso and colleagues (Journal of Ethnopharmacology 2021) found four neem studies with results similar to chlorhexidine. In a 90-day double-blind trial in 60 teachers in India, Nimbulkar and colleagues (2020) found that a 2.5% neem gel and a 0.2% chlorhexidine gel reduced plaque, gingivitis, and mouth bacteria equally. For chewing sticks, Ramli and colleagues (Journal of Ethnopharmacology 2022) meta-analyzed randomized trials of miswak sticks from Salvadora persica or neem and found them comparable to a toothbrush, and better than brushing alone when added to it, though the trials reported technique and frequency poorly. Verdict: a reasonable add-on to brushing and fluoride toothpaste, not a replacement for them or for dental care.

Head lice have a small, product-specific evidence base. Abdel-Ghaffar and Semmler (Parasitology Research 2007) treated 60 heavily infested Egyptian children aged 4 to 15 with a neem seed extract shampoo and found it effective after 10 minutes, with no scalp irritation, but there was no comparison group. Abdel-Ghaffar and colleagues (Parasitology Research 2012) left the shampoo Licener on for 10 minutes in 12 children and 20 minutes in 8 more; no lice survived, no live lice were found a week later, and soaked eggs failed to hatch. In the only randomized trial, Semmler and colleagues (Parasitology Research 2017) compared Licener with a dimethicone product in 119 children older than 2 in two Egyptian villages; after one treatment, 100% of the neem group (60 of 60) and 94.7% of the dimethicone group (54 of 57) were cured, a difference that was not statistically significant. One author of that trial worked for Alpha-Biocare, the shampoo’s maker, and the same research group ran the earlier studies. Verdict: promising for that formulation, which appears to work by suffocation; it says nothing about rubbing plain neem oil into the hair.

Skin conditions rest on weak or indirect evidence. In an uncontrolled pilot study, Charles and Charles (Tropical and Geographical Medicine 1992) reported that a paste of neem and turmeric cured scabies in 97% of 814 people within 3 to 15 days. Elson and colleagues (PLoS Neglected Tropical Diseases 2019) randomized 96 Kenyan children with tungiasis, an infestation by burrowing sand fleas, to 20% neem seed oil in coconut oil or a potassium permanganate foot bath; neem oil killed no more fleas, about 40% after six days, but reduced inflammation and pain. In an uncontrolled series of 50 head and neck cancer patients, Franco and colleagues (Medical Oncology 2017) used a St John’s wort and neem oil product for radiation skin damage. We found no good trials of neem for acne, eczema, or psoriasis, and claims for dandruff and hair growth rest on laboratory and rat studies, such as one of a rosemary and neem extract by Hashem and colleagues (Scientific Reports 2024). Meanwhile, MSKCC lists allergic contact dermatitis from neem oil. Verdict: not established.

Malaria is a traditional use that research has not borne out. The NRC report describes neem-leaf teas drunk for malaria in India, Nigeria, and Haiti, but early tests in infected mice, ducks, and chickens were inconsistent and usually negative, although the limonoid gedunin was active against the parasite in cell culture. Tajbakhsh and colleagues (Malaria Journal 2021) reviewed 722 experiments on African antimalarial plants and found neem among the most often tested, the most often toxic, and the most often inactive, with only one clinical trial of any plant, not neem. We found no human trial showing that neem treats or prevents malaria. As a mosquito repellent, 2% neem oil in coconut oil gave 96 to 100% protection from Anopheles mosquitoes in an Indian field study (Sharma and colleagues 1995), and in Ethiopia, Abiy and colleagues (Malaria Journal 2015) found that 20% neem oil gave more than 70% protection and complete protection for 3 hours, against 8 hours for DEET. Verdict: neem is not a malaria treatment, and as a repellent it is weaker and shorter-lasting than standard products.

Contraception and other internal uses remain experimental. Indian researchers developed Praneem formulations from neem seed or leaf extracts, usually combined with soapnut (Sapindus) extract: a vaginal cream greatly reduced pregnancies in rabbits and monkeys (Garg and colleagues, Contraception 1993), a pessary showed spermicidal action in post-coital tests in women (Raghuvanshi and colleagues 2001), and safety trials found single intrauterine doses tolerable in 18 sterilized women (Talwar and colleagues 1995) and a vaginal tablet tolerable over six months in a placebo-controlled trial of 100 women (Joshi and colleagues, Sexually Transmitted Infections 2008). None established how well neem prevents pregnancy in people, and Patil and colleagues conclude that clinical studies are lacking. Other human studies are small: in a small clinical study, Bandyopadhyay and colleagues (Life Sciences 2004) reported that a bark extract cut stomach acid by 77% and healed duodenal ulcers; Pingali and colleagues (Journal of Medicinal Food 2021) randomized 80 adults with metabolic syndrome to a leaf and twig extract or placebo for 12 weeks and reported dose-related falls in blood sugar and HbA1c; and an uncontrolled study of 60 people with HIV reported rising CD4 counts (Mbah and colleagues 2007). Verdict: none of these is established.

Neem comes in many forms, and they are not interchangeable. Leaves are used fresh, dried, powdered, or as teas and pastes; bark as powders and extracts; twigs as chewing sticks, chewed at one end into a brush; and seed kernels are pressed for oil, which the NRC describes as dark, bitter, and smelly when cold-pressed and mainly used for lamps, soaps, and other nonedible products. EPA distinguishes three registered pesticide ingredients: azadirachtin itself; clarified hydrophobic neem oil, the oil left after alcohol removes the azadirachtin; and cold-pressed neem oil, which can contain about 0.3% azadirachtin. Garden neem oil sprays are pesticides, not medicines. Products studied in people include leaf extract dental gels at 25 mg per gram (Pai and colleagues, Journal of Ethnopharmacology 2004) and 2.5% (Nimbulkar and colleagues), neem mouthwashes, the Licener shampoo, left on for 10 minutes in one study and applied on days 1 and 9 in the randomized trial, and 20% neem seed oil in coconut oil for sand flea infestation.

Oral doses come only from small trials and have no official standing. Bandyopadhyay and colleagues used 30 to 60 mg of a bark extract twice daily for up to 10 weeks, Pingali and colleagues 125, 250, or 500 mg of a leaf and twig extract twice daily for 12 weeks, and Mbah and colleagues 1 g a day of a leaf extract for 12 weeks. We found no medicinal dose set by any regulator and no fact sheet from the US National Center for Complementary and Integrative Health (NCCIH). Traditional malaria remedies used large amounts of leaves, about 500 g decocted in water or alcohol according to the NRC, which also warned that leaf teas may be harmful when drunk in quantity over long periods. Neem oil should not be taken by mouth in any amount, and the NRC report concluded in 1992 that neem oil and neem extracts should not be used internally until better tested.

Quality varies widely. The NRC notes that neem trees are genetically diverse, so neighboring trees can yield different mixtures, and that no standard of potency existed; Fernandes and colleagues found that azadirachtin content still varies greatly between seed batches. Fruits harvested in the wet season and left damp can grow the mold Aspergillus flavus, which produces aflatoxins, potent carcinogens that are extracted and concentrated along with the active ingredients. LiverTox raises the possibility that some poisoning outbreaks involved oil contaminated with seeds of look-alike plants, and the NRC warns that older sources confused neem with the chinaberry. Practical points: buy products labeled Azadirachta indica that state the plant part, keep neem oil and garden sprays in clearly labeled containers locked away from children, never decant them into food or drink bottles, and use only diluted, purpose-made products on skin and hair.