What they are
Palm oil comes from the pulp of the oil palm fruit; about half its fat is saturated 12. Seed oils such as soybean, canola and corn oil supply linoleic acid, an omega-6 fat 3.
| Oil | Per 100 g |
|---|---|
| Palm | 43.5 g palmitic (16:0), 4.3 g stearic (18:0), 36.6 g oleic (cis-18:1 n-9), 9.1 g linoleic (18:2 n-6) and 15.9 mg α-tocopherol 2 |
| Soybean | 50.4 g linoleic, 6.79 g α-linolenic (18:3 n-3) 3 |
| Canola | 18.6 g linoleic, 9.14 g α-linolenic 3 |
| Corn | 53.2 g linoleic, 1.16 g α-linolenic 3 |
Carotenoids, chiefly α- and β-carotene, colour crude palm oil orange-red 4.
In the body
Digestion frees palm oil’s fatty acids, which gut cells repackage into fat particles 5. The body cannot make linoleic or α-linolenic acid and converts little of either into longer fatty acids 6.
In palm oil, most palmitic acid is esterified at the sn-1 and sn-3 positions 1. Colipase-anchored pancreatic lipase hydrolyses these esters, leaving sn-2-monoacylglycerol 7.
- Pancreatic lipase (PNLIP)Colipase (CLPS), 2 H2OCleaves sn-1 and sn-3 esters; releases two fatty acids
- Monoacylglycerol O-acyltransferase 2 (MOGAT2)Fatty acyl-CoAEnterocyte ER; over 75% of postprandial intestinal TAG synthesis
- Diacylglycerol O-acyltransferase 1 (DGAT1)Fatty acyl-CoAOnly DGAT detected in human intestine
- several stepsAssembly requires APOB and microsomal triglyceride transfer protein (MTTP)
Humans lack Δ12- and Δ15-desaturases 6. Under 1% of deuterated linoleic acid became arachidonic acid in one study 6. Conversion of α-linolenic acid is estimated at 8–12% to eicosapentaenoic and possibly under 1% to docosahexaenoic acid; more dietary linoleic acid lowers it 6.
In 36 studies of adults on Western-type diets, lowering dietary linoleic acid by up to 90% or raising it up to sixfold was not significantly correlated with changes in plasma phospholipid arachidonic acid 8.
- Long-chain-fatty-acid–CoA ligase 1 (ACSL1)ATP, CoA, Mg2+Releases AMP and diphosphate
- Acyl-CoA 6-desaturase (FADS2)O2, 2 reduced cytochrome b5First, rate-limiting desaturation; also acts on α-linolenoyl-CoA (n-3)
- several stepsMalonyl-CoA, 2 NADPHElongation cycle: ELOVL5, HSD17B12, HACD1–4, TECR; releases CO2
- Acyl-CoA (8-3)-desaturase, Δ5-desaturase (FADS1)O2, 2 reduced cytochrome b5
- Lysophospholipid acyltransferase 5 (LPCAT3)1-Acyl lysophospholipidReacylation in the Lands cycle; releases CoA
- Cytosolic phospholipase A2 (PLA2G4A)H2OCalcium-dependent; selective for the sn-2 arachidonoyl chain
- Prostaglandin G/H synthase 1 and 2 (PTGS1, PTGS2)2 O2, Haem b, Reducing substrate (AH2)Acetylsalicylic acid and other NSAIDsCyclooxygenase forms PGG2; peroxidase reduces PGG2 to PGH2
Prostaglandin G/H synthases, the targets of acetylsalicylic acid (aspirin), convert arachidonate to prostaglandin H2, precursor of 2-series prostaglandins and thromboxanes 9.
What studies show
In feeding trials, palm oil raised LDL cholesterol more than oils low in saturated fat 1. In 27 such trials (807 participants, 2 to 16 weeks), the difference was 0.24 mmol/L (95% CI 0.13 to 0.35), and 0.31 mmol/L (0.20 to 0.42) in randomised trials 1.
In Burkina Faso, 24 schools were randomised to usual lunches, one 60 mg vitamin A capsule, or 15 ml red palm oil three times a week 10. Over 12 months, serum retinol below 0.70 µmol/L fell from 40.4% to 14.9% of 114 pupils on the oil, 46.1% to 17.1% of 117 on the capsule and 23.6% to 14.2% of 106 controls 10. Controls had higher baseline retinol; endline retinol did not differ significantly 10.
Seed oils in place of saturated fat lower blood cholesterol, but trials have not shown fewer deaths 111213. In the Sydney Diet Heart Study, 458 men after a coronary event were randomised to safflower oil and safflower-oil margarine instead of saturated fat, or to no specific dietary instruction 14. Over a median 39 months, 17.6% of 221 against 11.8% of 237 died (hazard ratio 1.62, 95% CI 1.00 to 2.64; number needed to harm 17) 14.
The Minnesota Coronary Experiment randomised 9423 residents of mental hospitals and a nursing home to corn oil and corn-oil margarine instead of saturated fat, or to a control diet 13. Serum cholesterol fell 13.8% against 1.0% in the 2355 fed the diets for a year or more, but recovered life tables showed no mortality benefit 13.
In a Cochrane review of 19 trials of more omega-6 fat (6461 participants, 1 to 8 years), mortality was 167 per 1000 in both arms (risk ratio 1.00, 0.88 to 1.12; 10 trials) 11. In another, polyunsaturated instead of saturated fat for at least 2 years gave risk ratios of 0.79 (0.62 to 1.00; 8 trials) for cardiovascular events and 0.96 (0.82 to 1.13; 7 trials) for death 12.
In cohorts (median follow-up 4.9 to 30.2 years), the highest linoleic acid intake was associated with lower mortality than the lowest (relative risk 0.87, 0.81 to 0.94; 11 studies, 170,076 deaths) 15.
Processing
Refining makes palm oil nearly colourless and forms contaminants from its fat 419.
Conventional refining, with high-temperature deodorisation, removes some tocopherols and tocotrienols and destroys all carotenoids; a milder patented process gives red palm oil with 80% of them 4. Glycidyl esters form from diacylglycerols, up to 12% of palm oil, during deodorisation above 200 °C; 3-monochloropropanediol (3-MCPD) esters form on refining 19. Palm oil averaged 2,912 µg/kg 3-MCPD and 3,955 µg/kg glycidol from esters, rapeseed oil 232 and 166 µg/kg 19. In animals, ingested esters are extensively hydrolysed; glycidol, an epoxide, caused tumours at several sites in rats and mice, probably by genotoxicity 19.
Only high consumers in younger age groups slightly exceed the EFSA tolerable daily intake for 3-MCPD and its esters, 2 µg/kg body weight 20. For glycidol, margins of exposure below 25,000 indicate a health concern; for infants fed only formula, the margin was 2,100 at 95th-percentile occurrence 19.