This current study shows that intravenous infusion of omega-3 PUFAs in the form of Lipidem® (B. Braun, Melsungen, Germany) induces a fairly rapid and marked increase in EPA and DHA in plasma PC and a small increase of EPA in erythrocytes. Amongst these changes, the elevation of EPA in plasma PC occurred the earliest. Interestingly in the control group not receiving fish oil-type omega-3 PUFAs there was a small decline in plasma PC EPA. The effect of the infusion of omega-3 PUFAs on plasma PC and erythrocyte EPA was reversed when the infusion was stopped. In contrast, elevated DHA was retained in plasma PC beyond the end of the infusion period. The retention of DHA in plasma PC was more marked in some individuals that others. Overall these data indicate that infused omega-3 PUFAs, especially EPA, are rapidly incorporated into plasma PC and also into erythrocytes and that turnover of EPA and DHA in plasma PC is different so that there is retention of DHA after supply is terminated even when EPA has returned to its starting level. Thus, this study provides further evidence that EPA and DHA may be handled differently in the body. The preferential retention of DHA has been demonstrated in plasma phospholipids , platelets , white blood cells  and erythrocytes [12, 31] after oral dosing of omega-3 PUFAs. These observations might indicate a special functional significance or importance of DHA over EPA
The rapid appearance of EPA and DHA with intravenous infusion is an advantage over oral supply where appearance of these fatty acids is slower [12–15, 32]. This is most likely because oral intake of omega-3 PUFAs must be followed by the processes of digestion and absorption before the fatty acids can be incorporated into plasma lipids and of further processing before they can be incorporated into cellular lipids. These processes all take time and depend on other factors such as the fat content and macronutrient composition of the meal. In contrast, intravenous infusion provides the fatty acids directly into the bloodstream introducing them into blood lipids, like PC, and directly exposing cells very quickly. A second aspect of intravenous infusion that will favour incorporation of EPA and DHA is the dose that can be administered. Studies of oral dosing typically use 1 to 4 g EPA + DHA per day, although higher doses have been used in some studies. In contrast, intravenous administration can easily provide more than 10 g of EPA + DHA on a daily basis. The higher dose will promote quicker incorporation and also a higher level of incorporation than is possible with oral supply. It is important to note that this high level of intravenous omega-3 PUFAs was well tolerated in all patients and there were no adverse reactions shown by study participants. Again this is an advantage over oral supply where moderate to high doses of fish oil can be associated with adverse gastrointestinal reactions.
The increase in EPA and DHA, which was mirrored by a decrease in the omega-6 fatty acid linoleic acid, resulting in decrease in the omega-6 to omega-3 PUFA ratio could be functionally important especially with regard to inflammation [7–10] and perhaps also immune function  and blood coagulation . The infusion of EPA + DHA promotes an anti-inflammatory and anti-coagulatory environment that would be an advantage in many patient groups including acute severe pancreatitis, sepsis, head trauma and even in advance of major gastrointestinal or hepatic surgery. In the current study the functional implications of the fatty acid changes described were not investigated.
Erythrocytes from patients receiving intravenous fish oil did not show an elevation of DHA, despite the small elevation of EPA. This suggests a slower rate of incorporation of DHA than EPA into erythrocytes and that the period of infusion (72 hr) was insufficient for net DHA incorporation to occur. Browning et al.  demonstrated slower appearance of DHA than EPA into erythrocytes of healthy subjects consuming oral supplements containing EPA and DHA. Once again these observations indicate a slower turnover of DHA than EPA in cells.
The control group received a mixture of soybean oil and medium-chain triglycerides. This mix is fairly rich in linoleic acid, although less so than traditional soybean oil lipid emulsions . One interesting observation is that plasma PC EPA decreased slightly but significantly in the control group. Given the role of EPA in limiting inflammation and preventing coagulation, the decline seen in the control group suggests an undesirable effect of infusing lipid that does not contain preformed EPA.
In the clinical setting, the overall aim of the infusion protocol used here would be to enrich cells and tissues with biologically active omega-3 PUFAs in order to slow or reduce tumour growth and to promote a favourable response to a subsequent insult such surgery. Such a favourable response might involve prevention of excessive inflammation and reducing the likelihood of immune paralysis. Here, erythrocyte fatty acids were measured as a surrogate for those in tissue. At the end of the infusion period EPA had increased in erythrocytes of those patients receiving the intravenous fish oil. However, the infusion period was stopped some days before surgery and this resulted in a reversal of the infusion-induced fatty acid composition change in erythrocytes. This may also have occurred in tissues including the liver. It will be important in future studies to prolong the infusion period up to the time of surgery, in order to maximise the likelihood of establishing a beneficial impact of omega-3 fatty acids on the response to surgery, and to sample liver tissue in order to confirm that its fatty acid composition is modified. Furthermore, it will be important to link changes in plasma, blood cell and tissue fatty acid composition to biological effects such as the concentrations of lipid mediators and cytokines and to clinical outcomes.