Ample clinical evidence from well-nourished infants in developed countries is available to recommend an optional, adequate intake level of omega-3 DHA in infant artificial formulas to support development of neural tissue [5], confirming the idea that a properly functioning brain cannot be built without a dietary supply of omega-3 fatty acids and balanced omega-6 fatty acids, especially LA [6]. Unlike other omega-3 LC-PUFAs, circulating DHA levels in adults are unresponsive to supplementation with any precursor, including ALA, although some response has been observed in young infants [7].

In another study, Hsieh et al. [9] reduced omega-6 LA and increased omega-3 ALA, facilitated in part using high oleic peanuts to yield 13% of the total fatty acids from each of LA and ALA (HO-RUTF), with a similar total PUFA content to the control (C-RUTF) [9]. The two studies had a different ‘standard’ RUTF used as control, with a higher LA and a lower ALA content in the study from Hsieh et al. [9] compared to the study from Jones et al. [8]. The experimental RUTFs in both studies had similar LA contents (13.1% vs. 14.4%) but differed in the ALA content (13.1% vs. 6.2%; Table 1).

Both studies reported plasma phospholipid DHA, a form that is receptor-transported into the brain, at 28 days of treatment. The Jones et al. [8] study showed that both control and F-RUTF decreased DHA status, by −11% and −21%, respectively, although these differences from baseline were not significant. Consistent with this observation, Hsieh et al. [9], using a larger sample size, showed a significant decrease in their control group (−25% in DHA with C-RUTF). This decrease was avoided in their experimental group (HO-RUTF, +4% increase, not significant), indicating that the form of DHA transported most efficiently to the brain remained stable through the initial recovery period.

The results of the two studies are consistent with the hypothesis that standard RUTF results in a decline in DHA status. The two experimental groups were very similar in their LA content and differed only in ALA content. However, the difference in ALA between the two experimental diets (13.1% vs. 6.23%) is unlikely to explain the results as human and animal studies show that no amount of any omega-3 precursor – ALA, stearidonic acid, EPA, or omega-3 docosapentaenoic acid – improves DHA status [7]. Differences in other nutrients may play a role, including mineral status, which influences function of the iron-containing desaturases required for endogenous synthesis of DHA [10,11].

Reduction of omega-6 LA intake, as in the experimental group in the Hsieh et al. [9] study, has been observed to increase DHA status in at least three human studies [12] as expected from decades of animal studies. The intake range for effects is not well established in humans, especially malnourished children, and it is likely to differ based on age and physiological state, among other factors. Importantly, the amount of omega-6 LA required to prevent frank deficiency symptoms in otherwise well-nourished infants is less than 1% of energy but with seed oils it is often more than 10-fold this amount.

Both studies raise and attempt to address the serious issue of omega-3 adequacy in RUTF for severely malnourished children. Both studies demonstrated the safety and acceptability of the experimental RUTFs. Neither study was designed to identify a formulation producing optimal DHA status or measured neurodevelopment. It has long been known that omega-6 grows brawn, while omega-3 grows brains [6]. Although neither study was powered to detect effects on recovery from SAM, there is every reason to believe that oil formulations altering the relative proportions of the major fatty acids LA, ALA, and oleic, among others, will support energy needs.