Low fat dietary intervention with ω-3 fatty acid supplementation in multiple sclerosis patients

ARTICLE IN PRESS

Prostaglandins, Leukotrienes and Essential Fatty Acids 73 (2005) 397–404 www.elsevier.com/locate/plefa

Low fat dietary intervention with o-3 fatty acid supplementation in multiple sclerosis patients$ Bianca Weinstock-Guttmana,, Monika Baierb, Youngmin Parkc, Joan Feichtera, Peterkin Lee-Kwena, Eileen Gallaghera, Jaya Venkatramanc, Kulwara Meksawanc, Suzanne Deinehertd, David Pendergaste, Atif B. Awadc, Murali Ramanathanf, Frederick Munschauera, Richard Rudickg a

Baird Multiple Sclerosis Center for MS Treatment and Research, Jacobs Neurological Institute, State University of New York, 100 High Street, Buffalo General Hospital-E2, Buffalo, NY 14203, USA b University of Alabama at Birmingham, Birmingham, Alabama, USA c Department of Exercise and Nutrition Sciences, State University of New York, Buffalo, NY, USA d Department of Exercise Nutrition Sciences, Women and Children Hospital, Buffalo NY, USA e Department of Physiology and Biophysics, USA f Department of Pharmacology, State University of New York, Buffalo, NY, USA g Mellen Center for MS treatment and Research, Cleveland Clinic Foundation, Cleveland, OH, USA Received 26 January 2005; accepted 15 May 2005

Abstract Objectives: To determine whether a low fat diet supplemented with o-3 positively affects quality of life (QOL) in relapsingremitting MS (RRMS) patients. In this 1-year long double-blind, randomized trial, patients were randomized to two dietary interventions: the ‘‘Fish Oil’’ (FO) group received a low fat diet (15% fat) with o-3 FOs and the ‘‘Olive Oil’’ (OO) group received the AHA Step I diet (fat p30%) with OO supplements. The primary outcome measure was the Physical Components Summary Scale (PCS) of the Short Health Status Questionnaire (SF-36). Additional measures using MS specific QOL questionnaires, neurological status and relapse rate were obtained. Results: 31 RRMS patients were enrolled, with mean follow up over 117SD 2.9 months. Clinical benefits favoring the FO group were observed on PCS/SF-36 (P ¼ 0:050) and MHI (P ¼ 0:050) at 6 months. Reduced fatigue was seen on the OO diet at 6 months (P ¼ 0:035). The relapse rate decreased in both groups relative to the rates during the 1 year preceding the study: mean change in relapse rate in the FO group: 0.797SD 1.12 relapses/year (P ¼ 0:021) vs. 0.697SD 1.11 (P ¼ 0:044) in the OO group. This study suggests that a low fat diet supplemented with o-3 PUFA can have moderate benefits in RRMS patients on concurrent disease modifying therapies. r 2005 Elsevier Ltd. All rights reserved.

1. Introduction Multiple sclerosis (MS) is a chronic and disabling disease of the central nervous system with unknown $ Supported in part by the NMSS grant (PP0620T), Mellen Center Foundation and ‘‘The Jog for the Jake’’ grant. Corresponding author. Tel.: +1 716 859 7051. E-mail address: [email protected] (B. Weinstock-Guttman).

0952-3278/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.plefa.2005.05.024

etiology. Epidemiological studies suggest that unidentified environmental factors contribute to the etiology of MS [1–3] and diet is a commonly postulated factor because strong associations have been observed between increased MS prevalence and diets high in meat and dairy products and low in fish [4–6]. These epidemiological findings have provided a rationale for a number of subsequent clinical trials aimed to establish beneficial effect of dietary interventions in

ARTICLE IN PRESS 398

B. Weinstock-Guttman et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 73 (2005) 397–404

MS. Swank [7,8] conducted a 35-year long, nonrandomized, retrospective study in MS patients and found that a diet with very low saturated fat content (average saturated fat 15–17 g/day) supplemented with cod liver or vegetable oils (10–40 g/day) provided long term benefits on mortality, relapse severity and disability, particularly if initiated during the earliest stages of MS. Two prospective double-blind controlled dietary studies that compared supplementation with linoleic acid (average 20 g/day) to oleic acid (average 11 g/day) for up to 2 years in relapsing-remitting MS (RRMS) patients reported clinical benefits with shorter and less severe relapses in the group receiving linoleic acid supplementation [9,10]. Although a third study with similar design failed to support a clinical benefit [11], a re-analysis of the raw data from all three double-blind randomized linoleic acid trials that included only the RRMS patients (the secondary progressive patients were excluded) confirmed the benefits of the linoleic acidsupplemented diet in decreasing the relapse severity and duration and also identified benefits in preventing neurological deterioration in patients with mild disability (EDSSp2.0) and short disease duration (o5 years) [9–12]. Another dietary intervention trial in MS was a double-blind study that compared the effect of o3 fatty acids at a dose of 2.8 g (EPA+DHA) per day to placebo (OO) and found only beneficial trends (P ¼ 0:07) on decreasing the duration, frequency and severity of MS relapses in favor of the group receiving o-3 fatty acids [13]. The results of all these dietary studies have had only a limited impact on MS therapeutics. During the last decade, two pharmacological immunomodulatory agents (interferon-b and glatiramer acetate) have been proven in large Phase III trials to be effective in reducing the attacks of MS and preventing long-term disability [14–16]. However, these therapies are only partially effective and there is considerable variability in patient responses. In the meantime, the spectrum of immunomodulatory effects mediated by polyunsaturated fatty acids (PUFA) has become better defined [17–22] and clinical trials of o-3 fatty acid supplemented dietary interventions have shown beneficial effects in other autoimmune diseases such as systemic lupus, rheumatoid arthritis, Crohn’s disease and psoriasis [23–26]. The traditional view of MS pathogenesis holds that MHC Class II-restricted CD4+ T-cells that recognize CNS components are the predominant pathogenic mediators and act by secreting inflammatory cytokines such as interleukin-1 (IL-1), IL-2, tumor necrosis factora (TNF-a), and interferon-g (IFN-g) [27,28]. Supplementation with long chain o-3 PUFAs (6 g/day; 86% EPA+DHA) for 6 months in MS patients and healthy controls decreased the secretion of the pro-inflammatory cytokines, IL-1b TNF-a, IL-2 and IFN-g by stimulated

peripheral blood mononuclear cells as well as reduced secretion of the inflammatory eicosanoids, prostaglandin E2 and leukotrienes LTB4, which are known to be increased in MS patients [29,30]. These studies demonstrated potentially beneficial immunologic effects in MS after o-3 PUFA ingestion and revived the interest in the use of o-3 PUFA supplementation as adjuncts to the approved immunomodulatory pharmacotherapies for MS, which as noted earlier, are only partially effective. The objective of the present study was to determine whether a low fat diet supplemented with o-3 long chain PUFA positively affects quality of life in RRMS patients already receiving a disease modifying therapy (DMT). Although the putative therapeutic agent in this study was considered to be the o-3 PUFA and its derivatives, the potential therapeutic effects related to a low fat diet itself (p15% total fat) cannot be clearly separated from the present study design.

2. Materials and methods This was a double-blind, randomized study to determine the effect of a low fat diet supplemented with fish oil (FO) vs. olive oil (OO) in RRMS patients. The MS patients studied were recruited from the Baird MS Center and from an additional suburban neurological facility. The study was approved by the local Institutional Review Board. 2.1. Inclusion and exclusion criteria With informed consent, subjects (age range of 18–60 years), with clinically diagnosed MS (Washington Panel criteria) were enrolled. The inclusion criteria included stable disease in the preceding 2 months and one exacerbation or more in the preceding 3 years. The concomitant use of disease modifying therapies (i.e., interferon beta products such as Avonexs, Betaserons) or glatiramer acetate (Copaxones) was permitted provided the patients were maintained on it for at least 2 months. To be eligible, the patients’ pre-study diet had to contain more than 30% of total calories from fat (the average consumption in US is about 35% calories form fat) as determined by a 7-day food record (7-DFR). 2.2. Dietary interventions After enrollment patients were randomly assigned to receive either the FO or OO supplementation. The FO group received 6 FO capsules per day containing 1 g FO (65% o-3; EPA 1.98 g and DHA 1.32 g/day) (EPAX 5500 EE, Tishcon Corp.) as well as the recommendation for a very low fat diet intake. Total fat intake including the o-3 PUFA supplements had not to exceed 15% of the total daily calories consumed. The OO group

ARTICLE IN PRESS B. Weinstock-Guttman et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 73 (2005) 397–404

received the American Heart Association Step I diet which is a controlled low cholesterol diet (total fat not exceeding 30% of total daily calories and saturated fats o10%), with ‘‘placebo’’ capsules containing equivalent of olive oil supplements (6 capsules of 1 g OO/day). Patients knew the percentage of dietary fat but did not know the assignment of capsules oil supplementation. All patients received 400 units of Vitamin E, one multivitamin tablet (not containing any PUFA) and at least 500 mg calcium per day. No additional supplements or changes in symptomatic therapies were allowed during the study. Participants were required to meet the dietitian weekly for the first 4 weeks, bi-weekly for the second month, and then were followed by monthly visits until the end of the study to monitor eating behavior. If the meeting did not occur, the patient would be reached by telephone for a dietary consultation. 2.3. Clinical assessments and sampling schedule The patients had a neurological examination including Extended Disability Status Scale (EDSS) [31] at baseline, 1 month, then every 3 months till the end of 1year study. Patients were also evaluated at the time of a new exacerbation, within 7 days of the new symptoms development. The examining practitioner performing the EDSS was blinded to the patient’s dietary intervention. The treating physician that was not blinded to the subjects’ dietary intervention was responsible for the relapse documentation. Thus, the relapse rate measurement but not the other clinical outcomes may be influenced by the blinding protocol. Patients completed their 7-DFR and a diary for subjective side effects. The food records and diary were reviewed with the dietitian and nurse at follow up visits or by telephone. Patients completed the QOL questionnaires at baseline, 1 month, 6 months and 1 year. Plasma samples for immunological tests and toxicity monitoring including CBC/diff, comprehensive metabolic panel, and lipid profile (cholesterol, triglycerides, LDL and HDL) were collected at baseline, 1, 3, 6, 9, and 12-month time points.

399

from its 8 subgroups using factor analytic methods. Scales are set up so that a higher score indicates better health. MFIS is a questionnaire designed to probe MSspecific quality of life, particularly perceiving the impact of fatigue on a variety of daily activities. The 3 subscales scores included in MFIS can be added to provide a total score ranging from 0 to 84. Higher scores indicate increased perception of fatigue. MHI is a widely accepted measure of overall emotional functioning that covers a wide range of negative and positive emotions. The MHI has 4 subscales and total scores range from 0 to 100, higher scores indicating better mental health. 2.5. Other outcome measures The relapse rate and change in EDSS were used as secondary outcome measures. 2.5.1. Immunological mediators Changes in multiple immunological parameters were assessed including: soluble adhesion molecules (sICAM1, sVCAM-1), prostaglandin PGE2, leukotrienes LTB4 and plasma cytokines (IL-1b, IL-4, IL-12, TNF-a, IFNg), leptin and chemokines (IL-8, RANTES, MCP-1) levels. All parameters were determined by the double sandwich ELISA (Enzyme Linked Immunosorbent Assay) kits, using manufacturer recommended protocols (R&D Systems, Minneapolis, MN). 2.5.2. Serum fatty acid profiles Serum fatty acid composition was examined using a gas liquid chromatograph (GLC) after methylation by the method of Lepage and Roy [34]. The GLC was equipped with 30 m EC-Wax capillary column (Alltech, Deerfield, IL). Nitrogen was used as carrier gas. Column temperatures was maintained at 190 1C and the injection port and the detector temperature were set at 250 and 260 1C, respectively. Peaks were identified by their retention times using fatty acid methyl ester standards and quantitated by an integrator. Values were expressed as percentages of total fatty acids.

2.4. Primary outcome measures The primary outcome measure was the Physical Component Scale (PCS) of the Short Form Health Survey Questionnaire (SF-36). Additional secondary outcome measures included: the Modified Fatigue Impact Scale (MFIS) and the Mental Health Inventory (MHI) [32,33]. Short health status questionnaire (SF-36) is a healthrelated quality-of-life questionnaire used as a standard health survey that consists of a 100-point scale. Two summary scales, Physical and Mental, have been derived

2.5.3. Statistical analysis Repeated measures analyses were used to analyze the change over time in the Physical Component of the SF36 as a measure of the change in quality of life. Treatment group, factors related to emotional status, fatigue and EDSS are known to effect quality of life and thus were examined initially for their effects on their primary response variable and statistically modeled accordingly. Next, the remaining factors of interest were considered for inclusion in the model based on the univariate results and on their association with each

ARTICLE IN PRESS 400

B. Weinstock-Guttman et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 73 (2005) 397–404

other. Similar analyses were performed for the secondary outcome measures.

3. Results Thirty-one patients were enrolled in this study. Fig. 1 summarizes the patient enrollment and attrition patterns in the study. Eight patients discontinued the study prematurely: 2 patients could not tolerate the diet (one from each group after one month in the study) and 2 patients were not compliant during the first 2 months; data from these 4 patients were censored; 5 discontinued because active disease (4 were from OO and 1 from FO group, data is available for all these patients at 3 and 6 months) and 1 patient became pregnant at 9 months (OO-group). Clinical and immunological parameters including adhesion molecules, leukotrienes, and prostaglandins were obtained from 27 patients; additional cytokine/chemokine levels were obtained from 19 patients. Baseline demographic and clinical characteristics were similar in both groups (Table 1). Out of 27 patients, 5 were on glatiramer acetate (Copaxones), one on IFNb1b (Betaserons), 20 on IFNb-1a (Avonexs) and one patient was on no DMT. Patients’ therapy type did not differ between the 2 groups. Most of the patients were recently diagnosed and started on disease modifying therapies with mean time on DMT of 1.1170.07 years (range of 2 months–5 years). Mean followup period on the diets was 11 months (SD72.9 months). Compliance to the diet based on individual food records was 69.2% from FO group and 66.7% in OO group, showing both groups complied with the diets. No significant clinical or immunological

difference was seen based on the compliance status. When assessing the weight loss, the FO group lost on average, 5.97SD 13.7 lbs while OO group gained an average of 0.57SD 11.2 lbs. As shown in Fig. 2, a clinical benefit (P ¼ 0:050) in PCS/SF-36 between the FO and the OO groups was seen at 6 months. Although a higher PCS was maintained at 12 months in the FO group compared to OO group, the difference did not reach statistical significance. The global SF-36 remained unchanged in the FO group while in the OO it had a tendency to worsen. There was a weak trend towards an increase (worsening) in EDSS (mean increase +0.35 EDSS points) in the OO group vs. a decrease of 0.07 EDSS points in the FO group. As shown in Fig. 3, a clinical benefit (P ¼ 0:050) was seen in the MHI scale at the 6-month time point favoring the FO group; however, this difference was not maintained at 12 months (Fig. 3). A significant difference was seen in the fatigue scale MFIS (P ¼ 0:035) at 6 months and a trend was maintained for 12 months (P ¼ 0:059); however, the benefit favored the OO group (Fig. 4).

Table 1 Baseline characteristics

Gender F Age-mean Disease duration/year EDSS SF-36/PCS: MHI MFIS

Group 1 N ¼ 14

Group 2 N ¼ 13

85.7% 45.1(7.7) 6.9(5.9) 1.9(0.6) 43.1(8.4) 86.3(16.2) 50.6(21.3)

84.6% 39.9(10.0) 4.6(3.5) 2.0(1.3) 40.8(7.8) 79.0(15.6) 38.3(15.1)

*No significant difference on baseline clinical parameters.

Flow Chart Assessed for eligibility N=31 Randomized N-31 Allocated to Fish Oil (15% fat) N=15

Allocated to Olive Oil (30% fat) N=16

Discontinued intervention (n=3) (1=not tolerated, 1 noncompliant, 1 active disease)

Analyzed (n=13) Excluded from analysis (n=2; 1=not tolerated, 1 noncompliant; D/C at 1 month)

Fig. 1. Flow chart.

Discontinued intervention (n=7) (2=noncompliant;4 active disease at 3 and 6 m;1 became pregnant at 9 months)

Analyzed (n=14) Exclude from analysis (n=2 noncompliant; D/C at 1 month)

ARTICLE IN PRESS B. Weinstock-Guttman et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 73 (2005) 397–404 Table 2 Change in serum lipids and fat intake

Trend of the PCS 46

45.4

44

43.7

43.1

43.4

42 40.9

39.3 38.4

38 36 34 Baseline

1 Month Overall

6 Month

15% Diet

12 Month

30% Diet

p=0.05 at 6 months

Fig. 2. Trend of the PCS.

Diet group ¼ FO

Diet group ¼ OO

Mean

SD

Mean

SD

21.6 25.9 23.1 10.8 10.3 7.6 12.9 8.1 13.9 437.7 720.7 691.6

120.3 148.5 25.0 58.0 47.1 -11.2 21.8 25.7 3.9 1765 1602 162.2

38.7 93.6 93.3 13.6 17.0 19.7 8.5 7.6 8.5 855.8 563.5 553.1

P-value

LDL at 1.visit 120.9 LDL at last visit 125.1 LDL change 4.2 HDL at 1. visit 49.9 HDL at last visit 50.6 HDL change 1.8 Fat at 1. visit 19.6 Fat at last visit 19.0 Fat change 0.6 Calories at 1.visit 1470 Calories at last visit 1467 Calories change 2.7

NS NS NS NS NS 0.0318 NS 0.0384 NS NS NS NS

Trend of Mental Health Inventory

95

93.2

90 86.3

91.6

86.3 84.1

80

Variable

40.4

40

85

401

79.8

79

81.8

75 70 65 60 Baseline

1 Month Overall

15% Diet

6 Month

12 Month

30% Diet

P=0.05 at 6 months Fig. 3. Trend of mental health inventory.

Trend of MFIS

70 60

58.8

50

50.6

48.9

40

38.3

37.5

30

51.8

33.8

37.3

20 10 0 Baseline

1 Month Overall

15% Diet

6 Month

12 Month

30% Diet

P=0.0348 at 6 months; p=0.0591 at 12 months Fig. 4. Trend of MFIS.

There was a decrease in relapse rate when compared with the 1 year prior to the study in both groups: FO group: 0.797SD 1.12 relapses/year (P ¼ 0:021) vs. OO group: 0.697SD 1.11 relapses/year (P ¼ 0:044). The lipid profile did not show significant changes during the 1-year study period, although a trend toward increased HDL was seen only in FO group (FO group:

HDL increase of +1.8 vs. a 11.2 decrease in OO group). The 1-year HDL difference between the groups did reach significance (P ¼ 0:032) (Table 2). No significant effect was observed in the levels of the soluble adhesion molecules sICAM and sVCAM during the study and between the two groups. However, a trend for a continuous decrease in PGE2 and LTB4 levels was seen primarily in the FO group (data not shown). Additional cytokines and chemokines data were obtained in 19 of the patients. No clear differences were seen in the pro-inflammatory cytokines (IL-1b, IFN-g and TNF-a) and leptin levels during the study and between the groups. The baseline and study values of IL-4 were not different between the groups, but a trend of an increase in IL-4 levels during the initial 6 months was seen only in the FO group. RANTES did not significantly change over the 12 months in neither group, although the FO diet group had an overall decrease of 21% while the OO diet had only a 1% decrease. Similarly, MCP-1 did not significantly change over the 12 months, however comparing the FO to OO diet there was 67% difference between the two groups favoring the FO group. 3.1. Serum fatty acid composition Fatty acid analysis chromatography was performed on the serum samples obtained during the study and reported as percentage of total fatty acids. The mean percentage changes between baseline and the last visit are seen in Table 3. Interestingly, the oleic acid (C18:1(o-9)) percentage decreased in both groups. Only the change in EPA (C20:5(3)) reached statistical significance between the two groups (an increase of 0.3670.58 in FO group vs. a decrease of 0:20  0:35 in OO group; P ¼ 0:027). The DHA (C22:6 (3)) increased in both groups. A decrease in saturated fatty acids (SFA) was seen in both groups, more in the OO group

ARTICLE IN PRESS 402

B. Weinstock-Guttman et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 73 (2005) 397–404

Table 3 Percent change in fatty acids (chromatography data) 15% diet mean (SD) 18:1(9) OO 20:5(3) EPA 22:5(3) 22:6(3) DHA Omega 3’s combined Sat. fatty acids

0.65 0.36 0.39 0.76 1.51 1.74

(5.47) (0.58) (0.58) (3.01) (2.88) (2.28)

30% diet mean (SD) 3.38 0.20 0.44 1.45 1.94 4.66

(8.94) (0.35) (0.92) (2.09) (1.74) (14.5)

P-value

0.6664 0.0270 0.8852 0.3648 0.4916 0.6256

but the difference did not reach significance. Spearman correlations were performed between the FA percentage levels at the different time points and the clinical parameters evaluated. The EPA levels at 6 and 9 months correlated well with EDSS at the last visit, (r ¼ 0:56; P ¼ 0:039 and r ¼ 0:78; P ¼ 0:022, respectively) and EPA at 9 months correlated with PCS at the last visit (r ¼ 0:76; P ¼ 0:027). The DHA levels at month 6 as well as the difference in DHA from baseline to the last visit correlated with MFIS at the last visit (067; P ¼ 0:008 and 0:48; P ¼ 0:050, respectively). SFA at baseline correlated with the MFIS at baseline (r ¼ 0:70; P ¼ 0:0002) and SFA at 6 months correlated with the MFIS at last visit (r ¼ 0:72; P ¼ 0:032), suggesting that higher SFA levels are associated with a higher perceived fatigue. Additional correlation analyses were performed between the EPA/DHA, SFA and the different adhesion molecules and cytokines levels. The most significant correlations were seen between decrease in LTB4, PGE2, ICAM and VCAM and the increase in o-3 primarily EPA (data not shown).

4. Discussion This study compared the effects of dietary interventions on quality-of-life measures in MS patients receiving a steady disease-modifying therapy. MS patients were randomly assigned to receive either a low fat diet (p15% fat) supplemented with o-3 PUFA or an American Heart Association Step I diet supplemented with OO. We used patient-centered outcome measures such as physical functioning quality of life, fatigue level, and emotional well-being. At the 6-month time point, the physical and mental status parameters (PCS and MHI) were significantly better in FO group compared to the OO group, which suggests that the FO group felt physically and emotionally healthier. Using chromatographic lipid profiling, we confirmed the expected increase in EPA o-3 levels in the FO group. The EPA levels at the 6- and 9-month time points were correlated with the EDSS and PCS benefits at the last visit. A trend favoring the FO group was maintained on PCS and MHI until the end of the study for all measurements.

The dietary interventions were a significant change relative to the patients’ pre-study diet and were selected to minimize potential placebo effects that can interfere with QOL assessments. Building on previous studies in MS patients with the knowledge that a diet high in total fats decreases the beneficial effects of PUFA supplementation, our primary treatment group (FO) received a low fat diet combined with o-3 PUFA so that the potential effect size would be increased due to contributions from both saturated fat reduction and the inclusion of o-3 PUFA. Post-study power calculations indicated that the sample size of 31 patients provided 61% power at a ¼ 0:05 to detect the effect size observed on the SF-36 primary outcome measure at 6 months. Our choice of patient-centered outcome measures was motivated by considerations of the sensitivity to the clinical symptoms that would be meaningful to patients at the early stages of MS. The Kurtzke EDSS Scale, which is commonly used in pharmacotherapy trials as an outcome measure was used only as a tertiary outcome measure because it is heavily influenced by ambulation and typically requires very large sample size. A four-arm study design capable of separating the effects of fat levels, o-3 PUFA, OO, and the American Heart Association Step I diet individually was not practical for this pilot study because of the very large sample sizes required for obtaining meaningful results, particularly when added on to ongoing therapy. Interestingly, a benefit on fatigue, as measured by the MFIS, was observed only in the OO group. The exact mechanisms for this are not clear but we systematically explore some potential explanations below. The simple caloric intake explanation for this finding can be excluded because the total mean daily calories (1467 in the FO group vs. 1602 in the OO group) were not significantly different between the groups. However, our lipid profiles indicated an unexpected increase in DHA levels in both groups and concomitantly, we also found significant correlations between DHA levels and the decrease in fatigue. Although this increase in DHA may mediate the benefits on fatigue, it invites an explanation for why the DHA levels increased in the OO group. The simplest explanation for the DHA source in the OO group is dietary intake from fish and other foods. However, the DHA increase may contain a contribution from the IFN-b treatment, which has been shown to induce specific release of unsaturated fatty acids such oleic acid, linoleic acid, arachidonic acid and possibly, EPA and DHA, from the cell membranes [35]. However, we should also mention a potential detrimental effect of high plasma levels of DHA: because the DHA molecule has six double bonds, it is susceptible to peroxidation and increases the risk for exhaustion of Vitamin E levels [36]. Although, Vitamin E supplementation could potentially counteract the DHA tendency to oxidation, this protective effect is still a matter of debate [37,38].

ARTICLE IN PRESS B. Weinstock-Guttman et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 73 (2005) 397–404

OO is rich in monounsaturated fatty acids and although in our study it was chosen as ‘‘placebo’’ supplement recent data from animal models and other human autoimmune diseases such as rheumatoid arthritis [39,40] indicate that OO can modulate the immune system. Additionally, OO also contains phenolic compounds capable of scavenging free radicals and protecting against lipid peroxidation [41]. These antiinflammatory and anti-oxidant effects of the OO diet may potentially explain the benefits seen on fatigue. During our study we also evaluated multiple immunological parameters but no significant differences were seen in the cytokines, chemokines and adhesion molecules levels between the two groups, although certain trends for a stronger anti-inflammatory effect was seen in FO group. The small number of patients and the concomitant immunomodulatory therapy (IFN or GA) may explain in part our limited immunological changes. It is reasonable to assume that the benefit seen during this 1-year study is not likely to be related to a delayed effect of DMT, but rather to a combined additive effect. We would like to emphasize that the benefits seen at 6 months were likely due to the nutritional intervention, since the study was randomized, and patients had been on DMT for an average of 1 year prior to entering the study. Likewise the trend toward a statistical benefit on physical and emotional status was maintained till the end of the study. Patients with chronic diseases such as MS often find themselves drawn to therapies that promise some control over their illness [42] and dietary modifications appear very attractive options. However, patients frequently cannot adhere to strict dietary regimens for extended periods. Although, our study required adherence to strict dietary requirements for 1 year, patient adherence was good, around 70% for both groups (based on their food record). However, we found higher than expected patient attrition from the study. This itself is an important finding that relates to future nutrition research in MS because it provides information necessary for study planning. The high attrition rate further documents the difficulties inherent to dietary studies in general and there is little in the literature about this problem in MS.

5. Conclusions In summary, our data demonstrated that both low fat diets might have potential to improve the subjective perception of physical and emotional disease burden in MS patients, although the very low fat diet supplemented with FO effects emerged as a more efficient and more prompt intervention. Both low fat dietary interventions were well tolerated and associated with a

403

decrease in number of relapses. However, despite the study limitations, including the small number of patients and two concurrent interventions, our results imply that low fat diet supplemented with o-3 PUFA may complement the beneficial effects of concurrent disease modifying therapies.

Acknowledgments To Mladen Golobic, MD, Ph.D., CCF for active involvement in the initial study design, and to Kristine Napier, MPH, RD for providing the dietary recommendations for this study.

References [1] C.M. Poser, The epidemiology of multiple sclerosis: a general overview, Ann. Neurol. 36 (Suppl. 2) (1994) S180–S193. [2] J.F. Kurtzke, Epidemiology of multiple sclerosis, in: P.J. Vinken, G.W. Bruyn, H.L. Klawans, J.C. Koetsier (Eds.), Handbook of Clinical Neurology, Demyelinating diseases, vol. 3, Elsevier, Amsterdam, 1985, pp. 259–287. [3] K. Lauer, Diet and multiple sclerosis, Neurology 49 (Suppl. 2) (1997) S55–S61. [4] R.L. Swank, O. Lerstad, A. Strjaum, J. Backer, Multiple sclerosis in rural Norway: its geographical and occupational incidence in relation to nutrition, N. Engl. J. Med. 246 (1952) 721–728. [5] M. Alter, M. Yamoor, M. Harshe, Multiple sclerosis and nutrition, Arch. Neurol. 31 (1974) 267–272. [6] K. Lauer, The risk of multiple sclerosis in the USA in relation to sociogeographic features: a factor-analytical study, J. Clin. Epidemiol. 47 (1994) 43–48. [7] R.L. Swank, B. Brewer Dugan, Effect of low saturated fat diet in early and late cases of multiple sclerosis, Lancet 336 (1990) 37–39. [8] R.L. Swank, J. Goodwin, Review of MS patient survival on a Swank low saturated fat diet, Nutrition 19 (2003) 161–162. [9] J.G.D. Millar, K.J. Zilkha, Double blind trial of linoleate supplementation of the diet in multiple sclerosis, Br. Med. J. (1973) 765–768. [10] D. Bates, P.R.W. Fawcett, Polyunsaturated fatty acids in treatment of acute remitting multiple sclerosis, Br. Med. J. (1977) 1390–1391. [11] D.W. Patty, Linoleic acid in multiple sclerosis: failure to show any therapeutic benefit, Acta Neurol. Scand. 58 (1978) 53–58. [12] R.H. Dworkin, D.W. Paty, Linoleic acid and multiple sclerosis; a reanalysis of three double blind trials, Neurology 34 (1984) 1441–1445. [13] D. Bates, A double-blind controlled trial of long chain n-3 polyunsaturated fatty acids in the treatment of multiple sclerosis, J. Neurol. Neurosurg. Psych. 58 (1989) 18–22. [14] IFNB Multiple Sclerosis Study Group, Interferon beta-1b is effective in relapsing-remitting multiple sclerosis. I Clinical results of a multicenter, randomized, double-blind, placebo-controlled trial, Neurology 43 (1993) 655–661. [15] L.D. Jacobs, D.L. Cookfair, R.A. Rudick, et al., Intramuscular interferon beta-1a for disease progression in relapsing multiple sclerosis, Ann. Neurol. 39 (1996) 285–294. [16] K.P. Johnson, B.R. Brooks, J.A. Cohen, et al., Copolymer 1 reduces relapse rate and improves disability in relapsing-remitting multiple sclerosis: results of a phase III multicenter, double-blind, placebo-controlled trial, Neurology 45 (1995) 1268–1276.

ARTICLE IN PRESS 404

B. Weinstock-Guttman et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 73 (2005) 397–404

[17] M.D. Peck, Interaction of lipids with immune function I: Biochemical effects of dietary lipids on plasma membranes, J. Nutr. Biochem. 5 (1994) 466–478. [18] M.D. Peck, Interaction of lipids with immune function II: Experimental and clinical studies of lipids and immunity, J. Nutr. Biochem. 5 (1994) 514–521. [19] S. Endres, Messengers and mediators: interactions among lipids, eicosanoids and cytokines, Am. J. Clin. Nutr. 57 (Suppl.) (1993) 798S–800S. [20] E. Soyland, M.S. Nenseter, L. Barrthen, C.A. Drevon, Very long chain n-3 and n-6 polyunsaturated fatty acids inhibit proliferation of human T lymphocytes in vitro, Eur. J. Clin. Invest. 23 (1993) 112–121. [21] T.H. Lee, R.L. Hoover, J.D. Williams, et al., Effects of dietary enrichment with eicosapentaenoic and docosahexaenoic acids on in vitro neutrophil and monocyte leukotrienes generation and neutrophil function, N. Engl. J. Med. 312 (1985) 1217–1224. [22] O. Holian, R. Nelson, Action of long-chain fatty acids on protein kinase C activity: comparison of omega-6 and omega-3 fatty acids, Anticancer Res. 12 (1992) 975–980. [23] J.M. Kremer, D.A. Lawrence, G.F. Petrillo, et al., Effects of highdose fish oil on rheumatoid arthritis after stopping nonsteroidal antiinflammatory drugs, Arthritis Rheumatism 38 (1995) 1107–1114. [24] R.D. Robinson, X. Li-Lian, C.T. Knoell, Alleviation of autoimmune disease by o-3 fatty acids, in effects of fatty acids and lipids in health and disease, World Rev. Nutr. Diet. 76 (1994) 95–102. [25] M.J. James, R.A. Gibson, L.G. Cleland, Dietary polyunsaturated fatty acids and inflammatory mediator production, Am. J. Clin. Nutr. 71 (2000) 343S–348S. [26] P.C. Calder, Fat chance of immunomodulation, Trends Immun. Today 6 (1998) 244–247. [27] E. Prat, R. Martin, The immunopathogenesis of multiple sclerosis, J. Rehabil. Res. Dev. 39 (2) (2002) 187–199. [28] K.C. O’Connor, A. Bar-Or, D.A. Hafler, The neuroimmunology of multiple sclerosis: possible roles of T and B lymphocytes in immunopathogenesis, J. Clin. Immunol. 21 (2) (2001) 81–92. [29] V. Gallai, P. Sarchielli, A. Trequattrini, M. Franceschini, A. Floridi, C. Firenze, A. Alberti, D. Di Benedetto, E. Stragliotto, Cytokine secretion and eicosanoid production in the peripheral blood mononuclear cells of MS patients undergoing dietary

[30]

[31]

[32]

[33]

[34] [35] [36]

[37]

[38]

[39]

[40] [41]

[42]

supplementation with n-3 polyunsaturated fatty acids, J. Neuroimmunol. 56 (2) (1995) 143–153. I. Neu, J. Mallinger, A. Wildfeuer, L. Mehlber, Leukotrienes in the cerebrospinal fluid of multiple sclerosis patients, Acta Neurol. Scand. 86 (6) (1992) 586–587. J.F. Kurtzke, Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS), Neurology 33 (11) (1983) 1444–1452. J.S. Fischer, N.G. La Rocca, D.M. Miller, P.G. Ritvo, H. Andrews, D. Paty, Recent developments in the assessment of quality of life in multiple sclerosis (MS), Mult. Scler. 5 (4) (1999) 251–259. B.G. Vickrey, R.D. Hays, B.J. Genovese, L.W. Myers, G.W. Ellison, Comparison of a generic to disease-targeted healthrelated quality-of-life measures for multiple sclerosis, J. Clin. Epidemiol. 50 (5) (1997) 557–569. G. Lepage, C.C. Roy, Direct transesterefication of all classes of lipids in a one-step reaction, J. Lipid Res. 27 (1986) 114–120. U.N. Das, Is there a role for saturated and long-chain fatty acids in multiple sclerosis, Nutrition 19 (2) (2003) 163–166. J.H. Song, K. Fujimoto, T. Miyazawa, Polyunsaturated (n-3) fatty acids susceptible to peroxidation are increased in plasma and tissue lipids of rats fed docosahexaenoic acid-containing oils, J. Nutr. 130 (2000) 3028–3033. E.E. Reich, K.S. Montine, M.D. Gross, L.J. Roberts II, L.L. Swift, J.D. Morrow, T.J. Montine, Interactions between apolipoprotein E gene and dietary alpha-tocopherol influence cerebral oxidative damage in aged mice, J. Neurosci. 21 (16) (2001) 5993–5999. J.P. Allard, R. Kurian, E. Agdassi, R. Muggli, D. Royall, Lipid peroxidation during n-3 fatty acid and vitamin E supplementation in humans, Lipid 32 (5) (1997) 535–541. A. Linos, V.G. Kaklamani, E. Kaklamani, Y. Koumantaki, E. Giziaki, S. Papazoglou, C. Montzoros, Dietary factors in relation to rheumatoid arthritis: a role for olive oil and cooked vegetables?, Am. J. Clin. Nutr. 70 (1999) 1077–1082. P. Yaqoob, Monounsaturated fatty acids and immune function, Eur. J. Clin. Nutr. 56 (2002) S9–S13. S.A. Wiseman, L.B. Tijburg, F.H. van de Put, Olive oil phenolics protect LDL and spare vitamin E in the hamster, Lipids 37 (11) (2002) 1053–1057. P. Ebers, in: D.W. Paty, G.C. Ebers (Eds.), Multiple Sclerosis, F.A. Davis Company, Pennsylvania, 1998, pp. 404–420.