Clinical Nutrition (2008) 27, 614e622 available at www.sciencedirect.com http://intl.elsevierhealth.com/journals/clnu ORIGINAL ARTICLE Effects of short-term oral administration of dietary marine oils in patients with inflammatory bowel disease and joint pain: A pilot study comparing seal oil and cod liver oil Linn A. Brunborg a,*, Tor M. Madland b, Ragna A. Lind c, Gülen Arslan a,c, Arnold Berstad c, Livar Frøyland a a National Institute of Nutrition and Seafood Research (NIFES), P.O. Box 2029 Nordnes, N-5817 Bergen, Norway Section for Rheumatology, Institute of Medicine, Haukeland University Hospital, Bergen, Norway c Section for Gastroenterology, Haukeland University Hospital, Bergen, Norway b Received 13 January 2006; accepted 20 January 2008 KEYWORDS n-3 PUFA; IBD; Seal oil; Cod liver oil; Short-term oral supplementation; LTB4 Summary Background: Very long chain n-3 polyunsaturated fatty acids have modulating effects on inflammatory mechanisms. Seal and fish oils are rich in n-3 polyunsaturated fatty acids, and possibly therefore high doses of nasoduodenally administered seal oil rapidly relieved inflammatory bowel disease (IBD)-associated joint pain in two recent studies. In the present study, we compared the effects of short-term oral administration of seal oil and cod liver oil on IBD-related joint pain, leucotriene B4 level, serum fatty acid profile and IBD activity. Methods: Thirty-eight patients with IBD-related joint pain were included in the study; 21 had Crohn’s disease and 17 ulcerative colitis. Ten milliters of seal oil (n Z 18) or cod liver oil (n Z 20) was self-administered orally 3 times a day for 14 days before meals in a double-blind setting. Results: There were no significant differences between the two intervention groups or between Crohn’s disease and ulcerative colitis patients. There was a tendency toward improvement in several joint pain parameters after both seal oil and cod liver oil administration. Further, plasma leucotriene B4 concentration, serum S n-6 to S n-3, and arachidonic acid (20:4n-6) to eicosapentaenoic acid (20:5n-3) ratios were significantly reduced after administration of seal oil and cod liver oil. * Corresponding author. Tel.: þ47 41456840; fax: þ47 55905299. E-mail address: linn.anne.brunborg@nifes.no (L.A. Brunborg). 0261-5614/$ - see front matter ª 2008 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. doi:10.1016/j.clnu.2008.01.017 Seal oil and cod liver oil in IBD 615 Conclusion: No significant differences in the two treatment groups were seen; in both groups, the changes in several joint pain parameters, leucotriene B4 level of plasma, and serum fatty acid profile were putatively favourable. ª 2008 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. Introduction Inflammatory bowel disease (IBD) is a chronic inflammatory disease of the gastrointestinal tract, often associated with joint pain (arthralgia) with or without clinical indications of inflammation in the joints (arthritis).1,2 In addition to pain, morning stiffness of joints, disturbed sleeping pattern and low quality of life are typical. The two major forms of IBD, Crohn’s disease and ulcerative colitis, share several clinical features, but are two distinct diseases with different pathophysiologies and responses to treatment. However, for joint pain, the response to treatment is very similar. Although joint pain is relieved, treatment with non-steroidal anti-inflammatory drugs (NSAIDs) has been implicated in aggravation of the intestinal inflammation in both conditions.3e5 It is therefore a requisite to find alternative treatments in order to improve everyday life for patients with IBD-related arthralgia. Typical western diets contain excessive amounts of n-6 polyunsaturated fatty acids (PUFAs) compared to n-3 PUFAs, in particular linoleic acid (LA, 18:2n-6), which is the precursor of arachidonic acid (AA, 20:4n-6). High AA to eicosapentaenoic acid (EPA, 20:5n-3) ratios in the blood and tissues may promote the pathogenesis of inflammatory diseases by facilitating the production of pro-inflammatory components, like eicosanoids.6e8 EPA, docosapentaenoic acid (DPA, 22:5n-3), and docosahexaenoic acid (DHA, 22:6n-3) are very long chain (VLC, i.e. fatty acids with 20 or more carbons) n-3 PUFAs and are present in high amounts in both seal oil and fish oil. VLC n-3 PUFAs have several health promoting properties, including modulating effects on immune responses and inflammation.9e12 Seal oil contains approximately three times more DPA than cod liver oil.13 DPA is a more potent inhibitor of platelet aggregation, compared to e.g. EPA and DHA, possibly due to inhibition of cyclooxygenase (COX) and enhanced formation of 12-hydroxyeicosatetraenoic acid (12-HETE) activating lipoxygenase (LOX).14 VLC n-3 PUFAs in seal oil are located mainly at the end positions (sn-1 or sn-3 position) of the triacylglycerol (TAG) molecule, whereas they are located mainly in the middle position (sn-2) of fish oil TAG.15e17 During digestion and throughout the circulation, fatty acids are mainly hydrolysed from the sn-1 and sn-3 positions by position specific pancreatic and lipoprotein lipases.18 Thus, VLC n-3 PUFAs from seal oil may be more readily available for lipolysis compared to n-3 PUFAs from fish oil, and intake of seal oil may have a different impact on inflammatory and immunological reactions. Eicosanoids are hormone-like compounds, made from C20 fatty acids (mainly AA in current western diets). Generally, eicosanoids derived from EPA (VLC n-3 PUFA) are less potent triggers of inflammatory and immunological reactions, compared to the corresponding compounds derived from AA (VLC n-6 PUFA). Hence, increased dietary intake of n-3 fatty acids may shift the balance of the eicosanoid production to a putatively less inflammatory profile. Orally administered fish oil, rich in VLC n-3 PUFAs, ameliorates rheumatic arthritis (RA) after long-term treatment (2e12 months).8,19,20 In a recent pilot study, we found that 10 days’ duodenal administration of seal oil to patients with IBD and associated joint pain reduced joint pain and disease activity significantly.21,22 With this in mind, a study with emphasis on comparing the effects of short-term oral administration of seal oil or cod liver oil, both rich in VLC n-3 PUFAs, on IBD-related joint pain was undertaken. Materials and methods Patients Forty-five patients with IBD and associated joint pain were included in the study between April 2003 and November 2003, 38 of these patients (84.4%) completed the study. Seven subjects violated the study protocol: one because of influenza (seal oil), one because of lack of time (cod liver oil), while the remaining five gave no reason for dropping out (two from the seal oil group and three from the cod liver oil group). Clinical and demographic characteristics of Table 1 Characteristics of the patients Characteristic Disease Crohn’s disease Ulcerative colitis Disease duration IBD, years (mean  SD) Objective arthritis/spondylitis Disease duration arthritis, years (mean  SD) Medical treatment NSAIDs (non-steroidal anti-inflammatory drugs) Disease-modifying anti-rheumatic drugs Drugs that reduce bowel inflammation Age (years (range)) Sex (male:female) a Seal oil (n Z 18) Cod liver oil (n Z 20) 12 6 14  8 11 9 12  9 6 88 9 12  9 4 4 0 1a 13b 13c 48.1 (30e72) 5:13 47.7 (33e73) 10:10 Sodium aurothiomalate. Prednisolone (3), Mesalazine (5), Mesalazine þ Azathioprine (3), Azathioprine (1), Sulfasalazine (1). c Prednisolone (1), Mesalazine (6), Mesalazine þ Azathioprine (2), Azathioprine (1), Sulfasalazine (3). b 616 L.A. Brunborg et al. the patients are included in Table 1. Fifteen of the 38 patients had radiographic proven spondylitis or at some time objective arthritis, defined as swelling of peripheral joints assessed by the rheumatologist or by radiography. The patients were allowed to continue their ordinary medication, including NSAIDs and disease-modifying anti-rheumatic drugs (DMARDs) during the intervention. Thus, 25 of the 38 patients were receiving either NSAIDs and/or DMARDs. The remaining 13 patients received no medication (Table 1). Study design The patients were randomly allocated to intervention with either seal oil (n Z 18) or cod liver oil (n Z 20). During the 14 days’ treatment, all patients stayed at home and followed a normal Western diet. Ten milliters of seal oil (Rieber skinn A/S, Bergen, Norway) or cod liver oil (Peter Møller, Oslo, Norway) was self-administered orally before meals three times a day for 14 days. The seal oil group thus received a total amount of 2.4 g of EPA, 1.1 g of DPA and 2.6 g of DHA per day. The seal oil was refined oil from harp seal (Phagophilus groenlandicus). The cod liver oil group received 2.3 g of EPA, 0.3 g of DPA and 3.7 g of DHA. Fatty acid composition, levels of vitamins A and E, and thiobarbituric acid reactive substance (TBARS, a secondary oxidation product) in the seal and cod liver oils are presented in Table 2. The level of dioxins in the seal and cod liver oils did not exceed the current EU upper limit for dioxins in marine oils for human consumption (2 pg WHO-TEQ/g fat).23 The patients were examined the day before the intervention started and the day after the treatment period ended. Table 2 Major fatty acids, vitamin A and E concentration and TBARS in seal oil and cod liver oil Fatty acid Seal oil (g/100 g) S Saturated 16:1n-7 18:1n-9 20:1n-9 S Monoenes 18:2n-6 20:4n-6 S n-6 18:3n-3 20:5n-3 22:5n-3 22:6n-3 S n-3 n-6/n-3 Total vitamin A a-Tocopherol TBARS 14.2 14.0 14.9 7.7 48.9 1.5 0.6 2.2 0.6 7.9 3.7 8.6 23.9 0.1 0.3 mg/100 g 4.5 mg/100 g 3.6 nmol/g WW Cod liver oil (g/100 g) 15.4 6.1 13.7 10.2 46.0 1.5 0.4 2.2 0.8 7.5 1.1 12.4 25.1 0.1 4.8 mg/100 g 309 mg/100 g 13.8 nmol/g WW Total vitamin A Z all-trans retinol þ 13-cis retinol þ 3-dehydroretinol. Abbreviations: TBARS Z thiobarbituric acid reactive substances. n.d. Z not detected. WW Z wet weight. Ethical considerations The study was approved by the Regional Committee for Medical Research Ethics and all the patients gave written informed consent before inclusion. Analytical and clinical methods Effect on joint pain was the primary clinical response variable, and the number of tender, swollen and painful joints was assessed by a rheumatologist before and after treatment, using a 44 joint count.24 The back mobility was tested by the simple ‘‘finger to floor distance’’, i.e. the distance between the tip of the third finger and the floor after the patient bends down with the knees straight.25 A 100 mm horizontal Visual Analogue Scale (VAS) ranging from 0 (very well) to 100 (very poor) was applied for the patient’s selfevaluation of rheumatic disease activity and intensity of pain last week.26 The patients also filled in the modified health assessment questionnaire (MHAQ).27 This method of quantifying functional impairment is based on self-evaluation of activities of daily life, giving a composite score from 1 (very well) to 4 (very poor). Venous blood samples were drawn from the overnight fasting patients in two vacutainers without anticoagulant (gel vial), and one with EDTA as an anticoagulant. LTB4 concentration in blood plasma was analysed according to an LTB4 enzyme-linked immunosorbent assay (ELISA) kit from R&D systems (Minneapolis, USA), which has not been tested for cross-reactivity with LTB5. The serum fatty acid composition of the total lipids was determined according to the method described by Lie and Lambertsen,28 with some modifications; the fatty acids were esterified in 20% boron trifluoride (BF3) in methanol, and biological sample parallels were analysed. The fatty acid composition was calculated using an integrator (Turbochrom Navigator, Version 6.1), connected to the Trace-GC 2000 gaseliquid chromatograph (‘‘cold on column’’ injection,    6025 C/min16025 C/min19025 C/min220  C), equipped with a 50 m CP-sil 88 (Chromopack) fused silica capillary column (id: 0.32 mm). Identification of the fatty acids was ascertained by standard mixtures of methyl-esters (Nu-Chek, Elyian, USA).21 Total cholesterol was determined according to Roche Modular Cholesterol CHOD-PAP. High density lipoprotein (HDL) cholesterol was determined according to Roche Modular HDL-Cholesterol plus. Low density lipoprotein (LDL) cholesterol was determined according to Roche Modular LDL-Cholesterol plus, with no pretreatment, and triacylglycerols (TAG) were determined according to Roche Modular Triacylglycerols GPO-PAP. Lipid peroxidation in the intervention oils and in serum from the patients was analysed by an in vitro method by determining TBARS.29,30 IBD activity was scored using the HarveyeBradshaw simple index for Crohn’s disease31 and the Walmsley simple clinical colitis activity index for ulcerative colitis.32 These IBD indexes consist of four clinical criteria: symptoms, physical signs, general well-being, and extra-intestinal complications. A score equal to or higher than six indicates active disease. Faecal calprotectin concentration, another measure of IBD activity,33 was determined according to Seal oil and cod liver oil in IBD 617 the Eurospital Calprestâ, code 9031 (Medinor, Oslo, Norway), an enzyme-linked immunosorbent assay (ELISA) kit. A simple self-constructed questionnaire about intake of seafood (n-3 PUFA supplement, fish and other seafood) was filled in by the patients at inclusion. There were six alternative answers for each meal (breakfast, lunch and dinner); every day Z score six, three to four days a week Z score five, one to two times a week Z score four, three to four times a month Z score three, one to two times a month Z score two, and never Z score one. Statistical analysis Data were analysed using the GraphPad Prism 4 (GraphPad Software Inc, San Diego, USA) statistical software package. Throughout the text, values are presented as mean  standard error of the mean (SEM) for normally distributed data, otherwise median and range are given. Student’s t-test (two sided) or ManneWhitney test was used for group comparisons. Within group differences were evaluated by the paired sample t-test (two sided) or Wilcoxon sign rank test. P-values 0.05 were regarded as indicating statistically significant differences. The influence of the medication on clinical and analytical parameters between the two intervention oils was tested using SIRIUS for Windows (version 6.5). Principal component analysis (PCA)34 was applied to identify and visualise the main trends by a low number of orthogonal factors, the so-called principal components (PC1, PC2, and PC3). A high positive or negative loading reveals a significant variable in the actual PCA model. Sample influences by the medication in a similar way, or with similar levels of fatty acids are located in the same area in the biplot. The data matrix was pretreated by auto scaling, i.e. dividing each data column by its standard deviation (SD), thereby preventing large variables with large absolute variance masking the small variables with small absolute variance. Results Patient characteristics at baseline At baseline, the cod liver oil group had a slightly higher intake of seafood compared to the seal oil group (10.1  0.9 vs. 9.2  0.6, P Z 0.3 data not shown). Consistently, the seal oil group had higher S n-6 to S n-3, and AA to EPA ratios (Table 4). LTB4 plasma concentrations at baseline were comparable in the seal oil group and cod liver oil group (Fig. 1). Eight of 18 patients in the seal oil group and four out of 20 receiving cod liver oil had an IBD index equal to or higher than 6, indicating active IBD (Fig. 2). The mean baseline IBD index in the cod liver oil group was borderline significantly lower compared to the seal oil group (3.9  0.5 vs. 5.4  0.7, P Z 0.06). Baseline faecal calprotectin values were within the normal range, thus indicating mild to moderately active intestinal inflammation (Table 5). There were no significant differences in joint pain assessment parameters (Table 3) or in concentration of blood lipids (Table 5) between the groups at baseline. Changes following intervention Joint pain There were no significant differences between the seal oil and cod liver oil group in joint pain parameters after treatment. However, as compared with baseline, the cod liver oil group showed a significant improvement in joint pain intensity (P Z 0.02) and the patient’s global assessment of rheumatic complaints (P Z 0.007) after 900 800 Seal oil Cod liver oil P = 0.02 P = 0.02 LTB4 concentration (pg/ml) 700 600 500 400 300 200 100 0 Before After Before After Figure 1 LTB4 concentrations in blood plasma before and after intervention. Plot of LTB4 concentrations before and after intervention with seal oil (n Z 18) or cod liver oil (n Z 20). Data are presented as individual results. There were no significant differences between the groups at baseline or after treatment. 618 L.A. Brunborg et al. Simple Index for Crohn's disease or Simple Clinical Colitis Activity Index for ulcerative colitis 13 Seal oil P = 0.08 12 Cod liver oil P = 0.08 11 10 9 8 7 6 5 4 3 2 1 0 Before After Before After Figure 2 IBD-score before and after intervention. Scatter plot of IBD-score before and after intervention with seal oil (n Z 18) or cod liver oil (n Z 20). Vertical lines denote within group mean. Simple Index for Crohn’s disease and Simple Clinical Colitis Activity Index for ulcerative colitis are pooled in the two groups. intervention (Table 3). The seal oil group showed a tendency toward improvement in these parameters; however it was not significant. In addition, there was a trend toward improvement in back pain intensity and modified health assessment questionnaire (MHAQ) in both groups, although these changes were not significant. The cod liver oil group showed a tendency toward improvement in the tender joint count, and the seal oil group in back mobility (‘‘finger to floor distance’’); however none of the changes were significant. (Table 4). Similarly, administration of seal oil or cod liver oil gave significant reductions in ratios of S n-6 to S n-3 and AA to EPA (both P < 0.01). After treatment there were significant group differences in DHA levels (cod liver oil group highest concentration) and in S n-6 to S n-3 and AA to EPA ratios (cod liver oil group lowest ratios) (all P < 0.01). However, if S n-6 to S n-3 and AA to EPA ratios were calculated as change from baseline, then the differences between the groups were not significant. Both the seal oil and the cod liver oil led to a small, though insignificant, rise in AA concentration. LTB4 Administration of seal oil and cod liver oil both lowered the mean LTB4 plasma levels (both P Z 0.02) (Fig. 1), and there was no significant difference between the groups after administration. IBD activity The IBD activity score tended to improve following treatment in both intervention groups (P Z 0.08 for both) (Fig. 2), the difference between the groups was not significant (P Z 0.4). Fatty acid composition Intake of seal oil or cod liver oil was accompanied by a significant rise in S n-3 fatty acids, EPA, DPA and DHA concentrations compared to baseline (all P < 0.01) Blood lipids Total cholesterol and LDL-cholesterol concentrations were significantly lowered following seal oil administration (P Z 0.02 and 0.03, respectively) (Table 5). Cod liver oil Table 3 Joint parameters e Median (range) Seal oil (n Z 18) Before Swollen joint count (0e44) Tender joint count (0e44) Finger to floor distance (cm) Joint pain intensity (VAS mm) Back pain intensity (VAS mm) Patient’s global assessment (VAS mm) MHAQ (1e4) ( 0 5 4.5 36.5 33 48 1.25 ( (0e2) (0e15) (0e20) (6e73) (1e72) (3e93) (1e2.4) Cod liver oil (n Z 20) After Before( After 0 5 1.5 30.5 23 31.5 1.2 0 5 5 44 29 42 1.2 0 3 5 27 14.5 20 1.1 (0e2) (0e25) (0e25) (3e79) (0e81) (1e90) (1e2.1) (0e10) (0e33) (0e45) (4e79)a (1e69) (2e90)a (1e2.3) Baseline values. Different superscripts indicate significant differences between before and after measures within treatment. (0e6) (0e24) (0e32) (0e84)b (0e90) (0e89)b (1e1.9) Seal oil and cod liver oil in IBD 619 Table 4 Fatty acid composition (mean  SEM, mg fatty acid/g sample) in serum before and after intervention with seal oil or cod liver oil Fatty acid Seal oil (n Z 18) Before S Saturated S Monoenes 18:2n-6 20:3n-6 20:4n-6 S n-6 18:3n-3 20:5n-3 22:5n-3 22:6n-3*y S n-3 n-6/n-3y 20:4n-6/20:5n-3y ( 1.2  0.1 0.95  0.09 1.3  0.1 0.06  0.01 0.23  0.03 1.6  0.1 0.03  0.01 0.05  0.01 0.02  0.01 0.11  0.01 0.21  0.02 8.4  0.5 5.5  0.6 Cod liver oil (n Z 20) After P-value Before( After P-value 1.31  0.06 1.01  0.07 1.5  0.1 0.04  0.01 0.28  0.02 1.9  0.1 0.04  0.01 0.38  0.03 0.05  0.01 0.27  0.01 0.74  0.05 2.7  0.1 0.82  0.05 0.22 0.41 0.26 0.04 0.09 0.25 0.16 0.01 0.01 0.01 0.01 0.01 0.01 1.2  0.1 1.0  0.1 1.3  0.1 0.05  0.01 0.22  0.02 1.6  0.1 0.03  0.01 0.07  0.01 0.03  0.01 0.14  0.01 0.26  0.03 7.2  0.6 5.0  0.7 1.41  0.08 1.01  0.06 1.59  0.09 0.04  0.01 0.27  0.02 2.0  0.1 0.04  0.01 0.44  0.03 0.04  0.01 0.34  0.02 0.87  0.06 2.3  0.1 0.66  0.05 0.56 0.81 0.36 0.01 0.31 0.36 0.68 0.01 0.01 0.01 0.01 0.01 0.01 ( Baseline values. *Denotes significant group difference at baseline. y Denotes significant group difference after treatment. four months in UC patients resulted in reduction of LTB4 concentration, improved histological indexes and increased weight in the patients. Significant improvement of clinical disease activity and lower drug requirement in UC patients was reported by Aslan and Triadafilopoulos.37 James and Cleland19 suggested that diets rich in n-3 PUFAs and low in n-6 PUFAs might increase the efficacy of anti-inflammatory medication and thus be drug-sparing in rheumatoid arthritis. The patients in the present study received approximately 6.5 g of VLC n-3 PUFAs in addition to an otherwise basically unchanged diet and medication. The newly discovered anti-inflammatory, vago-vagal reflex, in which endogenously released intestinal cholecystokinin (CCK) activates the vagal nerve and thereby suppresses the levels of tumor-necrosis factor (TNF) a and interleukin (IL)-6. This may be one mechanism by which duodenally administered VLC PUFAs relieves joint pain. The reflex can modulate the immune response and control inflammation in vagally innervated organs by activation of nicotinic receptors.38e40 To what extent the anti-inflammatory pathway is activated by oil administration and whether it is more strongly activated by duodenal compared to oral administration of the oils remain to be investigated. Dietary fatty acid composition affects the body fatty acid composition, and may influence the biosynthesis of reduced serum triacylglycerol concentration significantly after treatment (P Z 0.02). Again, there were no significant differences between the groups after intervention. Discussion In the present IBD patients, the joint pain and the disease activity tended to improve. The fatty acid profile of serum and the LTB4 level of plasma changed favourably, both in the seal oil group and in the cod liver oil group, when administered orally for 14 days. No significant differences between the two types of oils were revealed. Also, the effect of seal oil was apparently less pronounced in the present study than in our two prior studies.21,22 The main difference between the present and the prior studies was the mode of seal oil administration. While we used per oral administration in the present study, nasoduodenal tube administration was used in the former studies. Therefore, the importance of the mode of administration for local and systemic effects of seal oil warrants further studies. Lorenz et al.35 reported a moderate reduction in proinflammatory eicosanoids and limited morphological improvement in IBD patients following oral fish oil administration. Stenson et al.36 showed that fish oil supplementation for Table 5 Blood parameters and calprotectin (mean  SEM) Normal range Seal oil (n Z 18) Before Total cholesterol LDL-cholesterol HDL-cholesterol Triacylglycerol Calprotectin in faeces ( 3.3e7.7 mmol/L 1.8e5.7 mmol/L 1.0e2.3 mmol/L 0.45e2.45 0e50 mg/kg ( Cod liver oil (n Z 20) After a 5.4  0.2 3.4  0.2a 1.41  0.08 1.6  0.2 16  3 b 5.1  0.2 3.2  0.2b 1.50  0.08 1.3  0.2 17  3 Before( After 5.3  0.3 3.3  0.3 1.4  0.1 1.6  0.2a 19  8 5.3  0.3 3.4  0.3 1.6  1.1 1.1  0.2b 27  13 Baseline values. Different superscripts indicate significant differences between before and after measures within treatment. 620 6.0 CB20 n-6/n-3 4.0 SB4 20:4n-6 SB8 Comp. 2 (15.7%) inflammatory components, and thus modulate the inflammatory response in tissues. Levels of EPA and DHA in serum were significantly increased after seal oil or cod liver oil intervention compared to baseline in the present study (Table 4). Belluzzi et al.41 reported that incorporation of EPA and DHA into red blood cell membrane phospholipids displaces AA from the membranes. This gives less AA available for cyclooxygenase and lipoxygenase enzyme systems producing pro-inflammatory eicosanoids. Shimizu et al.42,43 reported that highly purified EPA ethyl esters significantly lowered leukocyte and rectal LTB4 levels in children with UC. Increased concentrations of EPA and DHA, in addition to decreased AA in membranes have been shown to decrease the formation of LTB4.44 In the present study there was a slight rise in AA concentration after intervention with seal oil and cod liver oil, which may be due to a replacement of AA in membrane phospholipids (i.e. more free AA in the circulation) with a simultaneously decreased formation of AA derived. Kew et al.45 reported a decreased production of prostaglandin E2 (PGE2) in rats fed with EPA or DHA in the outer positions of TAG, but this was not seen in rats given VLC n-3 PUFA in the middle position of TAG. In addition, seal oil more effectively lowered thromboxane (TX) A2 production compared to fish oil.46 In the present study, both seal oil and cod liver oil intake were followed by significant reduction in LTB4 plasma levels. Whether administration of large amounts of VLC n-3 PUFAs could be harmful in IBD needs consideration, because double bonds in PUFAs are relatively unstable, and are easily oxidised by free radicals. It was established many years ago that inflamed tissue undergoes peroxidation more quickly than normal tissues.47 There are numerous reports suggesting that the chronically inflamed colon may be subjected to considerable oxidative stress, and Sedghi et al.48 have shown elevated breath ethane and pentane in IBD patients, suggesting increased lipid peroxidation. In vitro studies have shown that n-3 and n-6 PUFAs increase the susceptibility of lipid peroxidation in the small intestine. However, except for slight nausea in some patients, no side-effects were reported and none of the patients had exacerbation of their bowel disease. Before administration, both study oils were measured based on degree of peroxidation by determining the TBARS (thiobarbituric acid reactive substances) concentration (a measure of secondary lipid peroxidation products29,30). The TBARS concentrations were within the limits of non-oxidised oils. To determine whether administration of marine oils resulted in increased lipid peroxidation in the patients, the TBARS concentration of serum was measured. There was no indication that the degree of peroxidation was increased after either seal oil or cod liver oil administration. The cod liver oil contained almost 70 times higher a-tocopherol (vitamin E) level than seal oil. a-Tocopherol in cooperation with vitamin C acts as an antioxidant, scavenging lipid peroxidation of PUFAs in cell membranes. Whether the large difference in a-tocopherol level between the two oils contributed to the clinical response is not known. To study the influence of the different medications the various patients continued taking during the intervention period, we performed a principal component analysis (PCA) Fig. 3. This multivariate approach demonstrated that apparently the different medication the patients continued L.A. Brunborg et al. 2.0 0.0 -2.0 -4.0 -6.0 -4.0 T CB13 SB14 AQ CB4 RV MH To SB16 CB3 LTB4 CB6 LV SB1TARMMEDCB10 SB13 SB7 SB19 SB2 SB18 SB3SB17 CB2 FF SB 10 CB15 CB14 CB5 SB11 CB16 NS AI D DM ARD CB12 CB18 SB5 CB9 CB7 CB19 SB 9 SB15 SO CB1 PREDNI SB6 CB 17 CB11 CB8 22 :5n-3 5n3-3 2220: :6 n- -2.0 0.0 2.0 4.0 6.0 8.0 Comp. 1 (26.3%) Figure 3 Principal component analysis of selected parameters. Biplot with scorings and loadings (PC1 vs. PC2) revealed from principal component analysis of joint pain parameters, selected fatty acids and medication of individual patients of all patients in the seal oil group at baseline (SA), and after intervention (SB), and all of the patients in the cod liver oil group at baseline (CA), and after intervention (CB). Abbreviations: Dmard e disease modifiying anti-rheumatic drugs; FF e finger to floor distance; LV e joint VAS; MHAQ e modified health assessment questionnaire; NSAID e non-steroidal anti-inflammatory drugs; RV e back VAS; S e swollen joints; T e tender joints; Tarmmed e bowel inflammation medicine; To e total patient’s global assessment VAS; VAS e visual analogue scale. taking during the intervention did not influence the trial in a particular direction. In conclusion, the present study could not reveal significant differences between seal oil and cod liver oil after short-term oral administration. The patients in both groups had a tendency toward improvement of IBD disease activity and IBD-related joint pain. The fatty acid profile of serum was shifted to a presumably more favourable composition, with a significant decrease in S n-6 to S n-3 and in AA to EPA ratios and level of LTB4. Conflict of interest statement The authors declare no conflict of interest. Acknowledgements Dr. Linn A. Brunborg and Dr. Arnold Berstad conceived the study, and they carried out the study, data analysis and drafted the manuscript with help from Dr. Tor M. Madland, Ragna A. Lind RN/MSc and Dr. Gülen Arslan. Dr. Livar Frøyland helped in the coordination and the drafting of the manuscript. All the authors read and approved the final manuscript. The study was financed by The Norwegian Fishermen Association, the Ministry of Fisheries, NIFES, Seal oil and cod liver oil in IBD aukeland University Hospital and the G.C. Rieber Foundations. The seal oil and cod liver oil were kind gifts from Rieber Skinn A/S and Peter Möller A/S, respectively. The patients are greatly thanked for their participation. The Gastroenterological Unit at aukeland University Hospital is also thanked, especially Aud Sissel Hjartholm and Gro Maria Olderøy for help with clinical analysis. The authors would also like to thank the Fat and vitamin laboratory at NIFES, especially Annbjørg Bøkevoll, Thu Thao Nguyen, Anne Karin Syversen, Felicia Dawn Couilliard and Vidar Fauskanger for excellent technical assistance. Johan Gorgas Brun at the Department of Rheumatology, Haukeland University Hospital is thanked for valuable and constructive discussions about statistics, rheumatic and haematological issues. The Seafood and Human Health Research Program at NIFES is also thanked, especially Tormod Bjørkkjær and Kine Gregersen for help with data collection, Marian Kjellevold Malde for constructive comments and Pedro Araujo for valuable comments and discussions on statistics. Valuable discussions and constructive comments from Rune Brunborg are also highly appreciated. References 1. Heuman R, Bolin T, Sjödahl R, et al. The incidence and course of perianal complications and arthralgia after intestinal resection with restoration of continuity for Crohn’s diesease. Br J Surg 1981;68:528e30. 2. Orchard T, Wordsworth B, Jewell D. Peripheral arthropathies in inflammatory bowel disease: their articular distribution and natural history. Gut 1998;42(3):387e91. 3. Kaufmann H, Taubin H. Nonsteroidal anti-inflammatory drugs activate quiescent inflammatory bowel disease. Ann Intern Med 1987;107:513e6. 4. Wallace J. How do NSAIDs cause ulcer disease? Bailieres Clin Gastroenterol 2000;14(1):147e59. 5. Wallace J. Pathogenesis of NSAID-induced gastroduodenal mucosal injury. Best Pract Res Clin Gastroenterol 2001;15(5): 691e703. 6. Simopoulos A. The importance of the ratio omega-6/omega-3 essential fatty acids. Biomed Pharmacother 2002;56(8): 365e79. 7. Simopoulos A. Omega-3 fatty acids in inflammation and autoimmune diseases. J Am Coll Nutr 2002;21(6):495e505. 8. Gil A. Polyunsaturated fatty acids and inflammatory disease. Biomed Pharmacother 2002;56:388e96. 9. Wallace J, Keenan C, Finn N. Antiinflammatory effects of a fish oil diet in a rat model of chronic colitis. Gastroenterologist 1980;96:83. 10. Empey L, Garg M, Fedorack R. Eicosapentaenoic acid (fish oil)enriched diet is protective in rat experimental colitis. Gastroenterologist 1989;96:83. 11. Calder P. Immunoregulatory and anti-inflammatory effects of n-3 polyunsaturated fatty acids. Braz J Med Biol Res 1998; 31(4):467e90. 12. Kremer J. N-3 fatty acid supplementation in rheumatoid arthritis. Am J Clin Nutr 2000;71:349e51. 13. Brox J, Olaussen K, Østerud B, et al. A long-term seal- and codliver oil supplementation in hypercholesterolemic subjects. Lipids 2001;36(1):7e13. 14. Akiba S, Murata T, Kitatani K, et al. Involvement of lipoxygenase pathway in docosapentaenoic acid induced inhibition of platelet aggregation. Biol Pharm Bull 2000;23(11):1293e7. 621 15. Brockerhoff H, Hoyle R, Hwang P. Incorporation of fatty acids of marine origin into triglycerides and phospholipids of mammals. BBA 1967;144:541e8. 16. Yoshida H, Kumamaru J, Mawatari M, et al. Lymphatic absorption of seal and fish oils and their effect on lipid metabolism and eicosanoid production in rats. Biosci Biotechnol Biochem 1996;60(8):1293e8. 17. Wanasundara U, Shahidi F. Positional distribution of fatty acids in triacylglycerols of seal blubber oil. J Food Lipids 1997;4(1): 51e64. 18. Small D. The effects of glyceride structure on absorption and metabolism. Annu Rev Nutr 1991;11:413e34. 19. James M, Cleland L. Dietary n-3 fatty acids and therapy for rheumatoid arthritis. Semin Arthritis Rheum 1997;27(2): 85e97. 20. Cleland L, James M, Proudman S. The role of fish oils in the treatment of rheumatoid arthritis. Drugs 2003;63(9): 845e53. 21. Arslan G, Brunborg L, Frøyland L, et al. Effects of duodenal seal oil administration in patients with inflammatory bowel disease. Lipids 2002;37(10):935e40. 22. Bjørkkjær T, Brunborg L, Arslan G, et al. Reduced joint pain after short-term duodenal administration of seal oil in patients with inflammatory bowel disease: comparison with soy oil. Scand J Gastroenterol 2004;11:1088e94. 23. EU Council regulation (EC) no 2375/2001 amending commission regulation (EC) no 466/2001 setting maximum levels for certain contaminants in foodstuffs; 2001. 24. Van Riel, P. EULAR handbook of clinical assessment in rheumatoid arthritis. 2000, Nijmegen. 25. Pile K, Laurent M, Salmond C, et al. Clinical assessment of ankylosing spondylitis: a study of observer variation in spinal measurements. Br J Rheumatol 1991;30:29e34. 26. Scott J, Huskisson E. Graphic representation of pain. Pain 1976;2:175e84. 27. Pincus T, Summey J, Soraci Jr S, et al. Assessment of patient satisfaction in activities of daily living using a modified Stanford health assessment questionnaire. Arthritis Rheum 1983; 26(11):1346e53. 28. Lie Ø, Lambertsen G. Fatty acid composition of glycerophospholipids in seven tissues of cod (Gadus morhua), determined by combined high-performance liquid chromatography and gas chromatography. J Chromatogr 1991;565(1e2): 119e29. 29. Hamre K, Næss T, Espe M, et al. A formulated diet for Atlantic halibut (Hippoglossus hippoglossus) larvae. Aqua Nutr 2001; 7(2):123e32. 30. Schmedes A, Hølmer G. A new thiobarbituric acid (TBA) method for determining free malondialdehyde (MDA) and hydroperoxides selectively as a measure of lipid peroxidation. J AOAC 1989;66(6):813e7. 31. Harvey R, Brandshaw J. A simple index of Crohn’s-disease activity. Lancet 1980;1:514. 32. Walmsley R, Ayres R, Allan R. A simple clinical colitis activity index. Gut 1998;43(1):29e32. 33. Tibble J, Bjarnason I. Non-invasive investigation of inflammatory bowel disease. World J Gastroenterol 2001;7(4): 460e5. 34. Martens H, Næss T. Multivariate calibration. Chichester, UK: John Wiley & Sons Ltd; 1989. 35. Lorenz R, Weber PC, Szimnau P, et al. Supplementation with n-3 fatty acids from fish oil in chronic inflammatory bowel disease e a randomized, placebo-controlled, double-blind crossover trial. J Intern Med 1989;225:225e32. 36. Stenson W, Cort D, Rodgers J, et al. Dietary supplementation with fish oil in ulcerative colitis. Ann Intern Med 1992; 116(8):609e14. 622 37. Aslan A, Triadafilopoulos G. Fish oil fatty acid supplementation in active ulcerative colitis: a double-blind, placebo-controlled, crossover study. Am J Gastroenterol 1992;87(4):432e7. 38. Tracey K. The inflammatory reflex. Nature 2002;420:853e9. 39. Luyer M, Greve W, Hadfoune M, et al. Nutritional stimulation of cholecystokinin receptors inhibits inflammation via the vagus nerve. J Exp Med 2005;202(8):1023e9. 40. Tracey K. Fat meets the cholinergic antiinflammatory pathway. J Exp Med 2005;202(8):1017e21. 41. Belluzzi A, Brignola C, Campieri M, et al. Effects of new fish oil derivative and fatty-acid phospholipid-membrane pattern in a group of Crohn’s disease patients. Dig Dis Sci 1994;39(12): 2589e94. 42. Shimizu T, Igarashi J, Ohtuka Y, et al. Effects of n-3 polyunsaturated fatty acids and vitamin E on colonic mucosal leukotriene generation in rats with experimental colitis. Digestion 2001;63:49e54. 43. Shimizu T, Fujii T, Suzuki R, et al. Effects of highly purified eicosapentaenoic acid on erythrocyte fatty acid composition and L.A. Brunborg et al. 44. 45. 46. 47. 48. leukocyte and colonic mucosa leukotriene B4 production in children with ulcerative colitis. J Pediatr Gastroenterol Nutr 2003; 37(5):581e5. Teitelbaum J, Walker W. Review: the role of omega-3 fatty acids in intestinal inflammation. J Nutr Biochem 2001;12: 21e32. Kew S, Wells S, Thies F, et al. The effect of eicosapentaenoic acid on rat lymphocyte proliferation depends upon its position in dietary triacylglycerols. J Nutr 2003;133(12):4230e8. Ikeda I, Yoshida H, Tomooka M, et al. Effects of long-term feeding of marine oils with different positional distribution of eicosapentaenoic and docosahexaenoic acids on lipid metabolism, eicosanoid production, and platelet aggregation in hypercholesterolemic rats. Lipids 1998;33(9):897e904. Halliwell B. Lipid peroxidation, oxygen radicals, cell damage, and antioxidant therapy. Lancet 1984;23:1396e7. Sedghi S, Keshavarizan A, Klamut M, et al. Elevated breath ethane levels in active ulcerative colitis: evidence for excessive lipid peroxidation. Am J Gastroenterol 1994;89(12):2217e21.