European Journal of Clinical Nutrition (2009) 63, 821–827 & 2009 Macmillan Publishers Limited All rights reserved 0954-3007/09 $32.00 www.nature.com/ejcn

ORIGINAL ARTICLE Time- and dose-dependent effect of psyllium on serum lipids in mild-to-moderate hypercholesterolemia: a meta-analysis of controlled clinical trials

Z-h Wei1, H Wang1, X-y Chen, B-s Wang, Z-x Rong, B-s Wang, B-h Su and H-z Chen

Department of Pharmacology and Biostatistics, Institute of Medical Sciences, Shanghai Jiaotong University School of Medicine, Shanghai, China

Objectives: Evidences from randomized clinical trials and meta-analysis have claimed an association between the use of soluble dietary fiber from psyllium and a cholesterol-lowering effect. However, there is still uncertainty as to the dose–response relationship and its long-term lipid-lowering efficacy. This meta-analysis was primarily conducted to address the dose–response relationship between psyllium and serum cholesterol level and time-dependent effect of psyllium in mild-to-moderate hypercholesterolemic subjects. Methods: Twenty-one studies, which enrolled a total of 1030 and 687 subjects receiving psyllium or placebo, respectively, were included in the meta-analysis. The studies were randomized placebo-controlled trials, double blinded or open label, on subjects with mild-to-moderate hypercholesterolemia. The dose of psyllium was between 3.0 and 20.4 g per day and intervention period was more than 2 weeks. Any type of diet background was permitted. Diet lead-in period was between 0 and 8 weeks. Results: Compared with placebo, consumption of psyllium lowered serum total cholesterol by 0.375 mmol/l (95% CI: 0.257– 0.494 mmol/l), and LDL cholesterol by 0.278 mmol/l (95% CI: 0.213–0.312 mmol/l). With random-effect meta-regression, a significant dose–response relationship were found between doses (3–20.4 g/day) and total cholesterol or LDL cholesterol changes. Regression model of total cholesterol was À0.0222 þ 0.2061 Â log (dose þ 1), and that of LDL cholesterol was 0.0485 þ 0.1390 Â log (dose þ 1). There was a time effect of psyllium on total cholesterol (equation: 6.3640–0.0316 Â treatment period) and on LDL cholesterol (equation: 4.3134–0.0162 Â treatment period), suggesting that psyllium reduced serum total cholesterol more quickly than LDL cholesterol. Conclusions: Psyllium could produce dose- and time-dependent serum cholesterol-lowering effect in mild and moderate hypercholesterolemic patients and would be useful as an adjunct to dietary therapy for the treatment of hypercholesterolemia. European Journal of Clinical Nutrition (2009) 63, 821–827; doi:10.1038/ejcn.2008.49; published online 5 November 2008

Keywords: hypercholesterolemia; total cholesterol; LDL cholesterol; meta-analysis; randomized controlled trials

Introduction (Petchetti et al., 2007). It has been speculated that a deficiency of soluble fibers in the Western diet might be In recent years, there has been growing interest in the use of contributing to the epidemics of coronary heart disease dietary fibers in health maintenance and disease prevention (CHD), diabetes mellitus and colonic cancer. Hypercholes- terolemia has been identified as a prominent independent risk factor in the development of CHD. Soluble fibers, Correspondence: Professor H-z Chen, Department of Pharmacology and including those from psyllium husk, have been shown to Biostatistics, Institute of Medical Sciences, Shanghai Jiaotong University School augment the cholesterol-lowering effects of a low-fat diet in of Medicine, 200025, Shanghai, China. persons with hypercholesterolemia. Increasing dietary fiber E-mail: [email protected] has been recommended as a safe and practical approach for 1These authors contributed equally to this work. Received 27 December 2007; revised 21 August 2008; accepted 18 cholesterol reduction (NCEP ATPIII, 2001 and Brunner et al., September 2008; published online 5 November 2008 2007). Time- and dose-dependent effect of psyllium on serum lipids Z-h Wei et al 822 Psyllium is one of the highest sources of soluble mucilagi- terms psyllium or dietary fiber and blood lipids to identify nous dietary fiber derived from seed husk of Plantago psyllium, the clinical trials involving psyllium from 1966 to August an annual plant grown in Mediterranean region, India, China 2005. In addition, FDA document, published reviews, and other regions. It is mainly marketed in two forms, an reference lists from clinical trials and conference abstracts over-the-counter bulk-forming laxative under the trade name were also examined. Metamucil and psyllium-enriched health foods (Singh, 2007). Psyllium is considered to be useful as an adjunct to dietary therapy (step 1 or step 2 American Heart Association (AHA) Trial selection diet) in the treatment of patients with mild-to-moderate For inclusion in the meta-analysis, studies had to meet hypercholesterolemia. The earliest clinical study (Garvin the following criteria: (1) Trials were conducted on et al., 1965) in a non-placebo-controlled pattern suggested human adults for the treatment of mild-to-moderate hyper- a significant cholesterol-lowering effect of psyllium cholesterolemia, not secondary to any recognized cause; (Metamucil). Then the first double-blind, placebo-controlled (2) they were controlled and had either a randomized trial of psyllium was performed and it showed its effectiveness crossover or a parallel design; (3) they provided lipid in mild-to-moderate hypercholesterolemia (Anderson et al., changes in the treatment and control groups to permit 1988). Subsequently, numerous clinical trials have investi- the calculation of the treatment effect; (4) they had a gated the treatment effect of psyllium as an adjunct to minimum intervention period of 2 weeks; (5) they had a low- dietary modification. There was a universal feeling in all these fat diet or controlled usual diet background. Disagreements investigators that psyllium was a hypocholesterolemic agent. regarding criteria for admission to meta-analysis were However, most of these studies were poorly designed and only resolved by discussion and consensus. In all 22 of 39 studies enrolled a small number of patients; thus the results should were selected for special review. Two studies (Stoy et al., be considered inconclusive. In 1997, Olson et al.(1997) 1993; Spence et al., 1995), which met above criteria, were conducted a meta-analysis involving 404 subjects with mild- excluded because they did not provide enough data to to-moderate hypercholesterolemia and found a beneficial compute the effect size and one (Wolever et al., 1994) was effect of psyllium-enriched cereal products on blood total excluded because subjects in the study were already cholesterol and low-density lipoprotein (LDL) cholesterol. represented in another published study by the same group. Psyllium in a form other than cereal (for example, crackers, As two research reports (Sprecher et al., 1993; Jenkins et al., bulk laxative) was not included in this study. More recently, 1997) involved two sub-groups, the data were treated as two Brown et al. (1999) performed another meta-analysis to separate studies. quantify the cholesterol-lowering effect of major dietary fibers including pectin, oat bran, guar gum and psyllium and found that various soluble fibers reduced total and LDL cholesterol Data extraction by similar amounts, but the effect was small within the Data were extracted from the published reports by two practical range of intake. The latest meta-analysis, conducted independent reviewers and disagreements resolved by dis- on individuals with mild-to-moderate hypercholesterolemia, cussion with the third reviewer. For each trial, the followings demonstrated that psyllium supplementation could signifi- were documented: country of origin, study population, cantly lower serum total and LDL cholesterol concentrations number and type of subjects, initial cholesterol concentra- in subjects consuming a low-fat diet (Anderson et al., 2000a). tion, type of psyllium preparation and used dose regimen, Despite the wealth of information that is available on the nature of control group, background diet, method of association between the consumption of soluble dietary fiber allocation, extent of blinding, study design (parallel or from psyllium and a cholesterol-lowering effect, the dose– crossover), treatment length, method of analysis, laboratory response relationship and its long-term lipid-lowering method used to measure lipid levels, withdrawal and efficacy have not been defined. Thus it was of great interest so on. to conduct a meta-analysis with more updated evidence to precisely determine the effect size of psyllium on decreasing serum lipids and explore the dose–response relationship and Meta-analysis long-term effects of psyllium on serum total cholesterol and The value of total cholesterol, LDL cholesterol, high density LDL cholesterol levels in individuals with mild-to-moderate lipoprotein (HDL) cholesterol and triglycerides were hypercholesterolemia. presented in units of mmol/l. Results reported in mg/dl were converted to mmol/l. For studies with parallel group designs, lipid-lowering effect size was calculated Materials and methods by subtracting the mean change in the control group from that in the treatment group. The mean change was mean Identification of previous studies baseline subtracted by mean end point value. For crossover A computerized literature search was conducted on the studies, the effect size was represented by the difference MEDLINE, EMBASE, BIOSIS and PASCAL databases, using the of post-treatment lipid concentrations of the treatment

European Journal of Clinical Nutrition Time- and dose-dependent effect of psyllium on serum lipids Z-h Wei et al 823 Table 1 Characteristics of the studies included in the meta-analysis

Sources Country Study Diet lead-in Initial TC Background Type of Psyllium dose Duration design period (d) concentration diet treatment (g/d) (d) (mmol/l)

Anderson et al. (1988) US Parallel 14 4.78–8.28 Usual diet Metamucil 10.2 56 Bell et al. (1989) US Parallel 42 5.35–7.01 NCEP step 1 Metamucil 10.2 56 Bell et al. (1990) US Parallel 42 5.35–7.01 NCEP step 1 Cereal 3.0 42 Neal and Balm (1990) US Parallel 49 46.21 NCEP step 1 Metamucil 20.4 91 Levin et al. (1990) US Parallel 56 45.17 NCEP step 1 Metamucil 10.2 112 Anderson et al. (1991) US Parallel 56 5.17–7.76 NCEP step 1 Metamucil 10.2 56 Anderson et al. (1992) US Parallel 7 5.17–7.76 NCEP step 1 Cereal 12.0 42 Everson et al. (1992) US Crossover — 46.2 Usual diet Metamucil 15.3 40 Sprecher et al. (1993) US Parallel 56 45.7 Low/high fat Metamucil 10.2 42 Summerbell et al. (1994) UK Parallel 21 5.2–7.8 Low fat Cereal 9.6 42 Wolever et al. (1994) Canada Crossover 28 46.21 NCEP step 2 Cereal 6.7 14 Roberts et al. (1994) Australia Crossover 28 6.00–7.75 Low fat Cereal 10.2 42 Maciejko et al. (1994) US Crossover 56 45.7 NCEP step 1 Metamucil 10.2 42 Jenkins et al. (1997) Canada Crossover — 46.72 NCEP step 2 Cereal 9.4 30 Romero et al. (1998) Mexico Parallel — 45.7 Usual diet Cookies 15.0 42 Davidson et al. (1998) US Parallel 56 LDL 3.36–5.68 NCEP step 1 Food 3.4/6.8/10.2 108 MacMahon and Carless (1998) UK Parallel 56 6.5–7.8 Low-fat Ispaghula husk 7.0/10.5 84 Anderson et al. (2000b) US Parallel 56 LDL 3.36–4.91 AHA step I Metamucil 10.2 182 Flannery and Raulerson (2000) US Parallel — 45.7 Usual diet Food 6.0 112

Abbreviations: TC: total cholesterol; LDL: LDL cholesterol.

and control periods. A technique described by Anne (2002) Results was used to pool the effect sizes from individual studies. A positive value indicated a greater lipid reduction relative Description of trials to placebo. Study characteristics of all published trials included in the Tests of heterogeneity were performed using Cochrane Q meta-analysis were listed in Table 1. The studies were test to determine whether they estimated for common undertaken in following five countries: United States, United treatment effect. The significant level was a ¼ 0.2. But to Kingdom, Canada, Australia and Mexico. They were either determine whether or not the Q statistic was significant, a parallel or crossover design. Subjects were mild-to-moderate random-effect model was assumed. hypercholesterolemia (total cholesterol 44.78 mmol/l or Pooled effect size and its 95% confidence interval were LDL cholesterol 43.36 mmol/l). Four studies had no dietary estimated by a fixed-effect or random-effect model. The lead-in period, whereas others had 2–8 weeks dietary lead-in subgroup analyses were performed by dose (10.2 g/day), period. Most subjects consumed either a low-fat diet or an psyllium-enriched foods, background diet (step I diet), NCEP step I or II diet during diet lead-in and therapy periods, treatment period (8 weeks), design type (parallel design) whereas those in four studies consumed controlled usual and clinical trial conducted in USA. Difference of lipid- diet. The therapy period ranged from 14 to 182 days. The lowering effect between psyllium-enriched foods and bulk form was psyllium bulk laxatives or psyllium-enriched foods laxative was tested by meta-regression (Berkey et al., 1995; with the dosage of 3–20.4 g/day. Data from 1717 persons Houwelingen et al., 2002). The possibility of publication bias with hypercholesterolemia were included in this meta- was detected with funnel plot (Egger et al., 1997). analysis. When more than one dose was studied, the mean lipid change across all doses was used to provide an average effect size. However, in the dose–response analysis, each dose was Effects of psyllium on lipid represented separately. Only clinical trials with parallel In the full dose range, psyllium significantly reduced group design were included in dose–response analysis, as both serum total cholesterol and LDL cholesterol concen- the net change of lipid could only be estimated in this kind trations: 0.375 mmol/l (95% CI: 0.257–0.494 mmol/l) and of trials. 0.278 mmol/l (95% CI: 0.213–0.312 mmol/l) respectively, A random-effect meta-regression was performed to address among adults with mild-to-moderate hypercholesterolemia dose–response relationships. The long-term effect of psy- (Table 2). The heterogeneity test of total cholesterol was llium was explored by mixed-effect model for repeated statistically significant, and that of LDL cholesterol was non- measure data. significant. Psyllium also significantly reduced HDL choles- All statistical analyses were performed using software terol, but by a much smaller amount: 0.0353 mmol/l (95% package SAS 9.13. CI: 0.0003–0.0514 mmol/l). Psyllium intake did not signifi-

European Journal of Clinical Nutrition Time- and dose-dependent effect of psyllium on serum lipids Z-h Wei et al 824 Table 2 Pooled treatment effect of psyllium on serum lipid

Variables No. of studies No. of subjects Effect sizea (95% CI) (mmol/l) Heterogeneity (Q)

Total cholesterol 21 1717 0.375 (0.257–0.494) 0.0001 LDL cholesterol 21 1696 0.278 (0.213–0.312) 0.8825 HDL cholesterol 20 1430 0.0353 (0.0003–0.0514) 0.9980 Triglyceride 19 1397 À0.0077 (À0.0907–0.0346) 0.9251

aEffect size for each trial was computed as the difference of lipid level change between psyllium and placebo groups. The pooled effect size was calculated with standard, fixed- or random-effect model. Any 95% CI that included zero indicated no significant result.

Anderson(1988) 0.730(0.119,1.341) n=26 Anderson(1988) 0.710(0.018,1.402) n=26 Bell(1989) 0.310(0.026,0.594) n=75 Bell(1989) 0.260(-0.051,0.571) n=75 Bell(1990) 0.220(-0.204,0.644) n=38 Bell(1990) 0.710(0.018,1.402) n=38 Neal(1990) 0.340(-0.161,0.841) n=54 Neal(1990) 0.440(-0.083,0.963) n=54 Levin(1990) 0.270(0.011,0.529) n=58 Levin(1990) 0.350(0.070,0.630) n=58 Anderson(1991) 0.390(0.047,0.733) n=52 Anderson(1991) 0.280(-0.080,0.640) n=52 Anderson(1992) 0.430(0.022,0.838) n=44 Anderson(1992) 0.550(0.126,0.974) n=44 Everson(1992) 0.260(-0.105,0.625) n=40 Everson(1992) 0.230(-0.223,0.684) n=40 Specher(1993)I 0.240(-0.042,0.522) n=81 Sprecher(1993)I 0.240(-0.077,0.557) n=81 Specher(1993)II 0.350(-0.007,0.707) n=37 Sprecher(1993)II 0.470(0.057,0.883) n=37 Summerbell(1993) 0.190(-0.215,0.595) n=37 Summerbell(1994) 0.280(-0.179,0.739) n=37 Roberts(1994) 0.20(0.011,0.389) n=158 Roberts(1994) 0.210(0.011,0.409) n=162 Wolever(1994) 0.370(0.051,0.689) n=84 Wolever(1994) 0.430(0.097,0.763) n=84 Maciejko(1994) 0.240(-0.476,0.956) n=36 Maciejko(1994) 0.190(-0.589,0.969) n=36 Jenkins(1997)I 0.230(-0.089,0.549) n=64 Jenkins(1997)I 0.270(-0.090,0.630) n=64 Jenkins(1997)II 0.210(-0.275,0.695) n=54 Jenkins(1997)II 0.250(-0.235,0.735) n=54 Romero(1998) 1.060(0.495,1.625) n=20 Romero(1998) 0.980(0.438,1.522) n=20 MacMahon(1998) 0.240(0.05,0.430) n=266 MacMahon(1998) 0.215(-0.024,0.454) n=283 Davidson(1998) 0.220(-0.015,0.455) n=196 Davidson(1998) 0.143(-0.122,0.408) n=196 Anderson(2000) 0.270(0.105,0.43) n=248 Anderson(2000) 0.990(0.788,1.192) n=248 Flannery(2000) 0.130(-0.375,0.635) n=28 Flannery(2000) 0.160(-0.383,0.703) n=28 -1.0 -0.5 00.5 1.0 1.5 2.0 -1 -0.5 0 0.5 1.0 1.5 2.0 Figure 2 The effect size (and 95% CI) of psyllium on LDL Figure 1 The effect size (and 95% CI) of psyllium on total cholesterol of all studies included in the meta-analysis. The cholesterol of all studies included in the meta-analysis. The horizontal lines denoted the 95% CIs. horizontal lines denoted the 95% CIs.

cantly affect serum triacylglycerol concentrations. Figures 1 or LDL cholesterol levels. The regression between log term of and 2 showed the net effect of consumption of psyllium on dose and measurement was estimated with random-effect total cholesterol and LDL cholesterol levels for each study regression model. The regression model of total cholesterol included in the meta-analysis. To explore heterogeneity, was À0.0222 þ 0.2061  log (dose þ 1), and that of LDL funnel plots for total cholesterol and LDL cholesterol were cholesterol was 0.0485 þ 0.1390  log (dose þ 1). Although drawn and shown in Figure 3. Egger test displayed that the two models were both statistically significant (both the asymmetry ¼À0.3999 (P ¼ 0.7058) for total cholesterol, P-values were smaller than 0.0001), the regression coefficients suggesting that publication bias was not confirmed, but the were relatively small and the independent variable was phenomenon of ‘large effect and small sample’ could be log-transformed. Thus greatly increasing dosages of psyllium concluded. The asymmetry of LDL cholesterol was 0.9424 might not correspondingly produce a clinically significant (P ¼ 0.0399), indicating that both publication bias and the difference in the lipid-lowering effect across a dose range of phenomenon of ‘large effect and small sample’ existed 3–20.4 g/day. On the other hand, we fitted a weighted linear simultaneously. model to explore the relationship between the percent reduction of LDL cholesterol (calculated as ðbaselineÀ end pointÞ=baselineÂ100%) and dose of psyllium (Figure 4). Dose–response relation of psyllium Results showed that there was a moderate correlation, There was a significant dose–response relationship between and the regression equation was: 2.200 þ 0.685  dose doses (3–20.4 g/day) and changes of serum total cholesterol (P ¼ 0.0454). According to this equation, consumption of

European Journal of Clinical Nutrition Time- and dose-dependent effect of psyllium on serum lipids Z-h Wei et al 825 12 7.5

10 7.0 Anderson(2000) 8

6 6.5 Placebo 4 Romero(1998) 6.0 Precision (1/SE) Precision 2 Anderson(1988) 5.5 Psyllium 0 0.2 0.4 0.6 0.8 1.0 1.2 Effect size (mmol/L) (mmol/L) level cholesterol Total 5.0 0 5 10 15 20 25 3 14 Treatment (week)

12 5.2 10 4.8 8

6 4.4 Romero(1998) Placebo 4 Precision (1/SE) Precision 4.0 2 Anderson(1988) Psyllium 0 0.2 0.4 0.6 0.8 1.0 1.2 3.6 Effect size (mmol/L) LDL cholesterol level (mmol/L) level LDL cholesterol 3.2 Figure 3 Funnel plot of all trials in the meta-analysis on total 0 5 10 15 20 25 30 cholesterol-lowering (upper) and LDL cholesterol-lowering (lower) Treatment (week) effects of psyllium. Figure 5 The time-effect relation of psyllium on serum total cholesterol (upper) and LDL cholesterol (lower) levels. m: placebo, K: psyllium. The line was estimated by random-effect model 30 for repeated measurements. The time effect was fixed in the model. 25

20 a mixed-effect model for a repeated measurement. Through- 15 out the treatment period, serum total cholesterol and LDL 10 cholesterol levels of psyllium group reduced faster than that of the placebo group (Po0.0001). The equation of total 5 cholesterol was 6.3640–0.0316 Â treatment period, and that 0 of LDL cholesterol was 4.3134–0.0162 Â treatment period, Percent Reduction in LDL-C (%) Percent 0 3 6 9 12 15 18 21 suggesting that, along with time, total cholesterol reduced Dose (g/day) more quickly than LDL cholesterol. Figure 4 Relation between the percent reduction in LDL cholesterol (LDL-C) and dose of psyllium. The percent change of LDL cholesterol was calculated as ðbaseline À end pointÞ=baselineÂ100%: The line Type of intervention was estimated by weighted linear model. Plot showed a positive and Psyllium, as either bulk laxatives or enriched foods, could moderate association between dose of psyllium and percent decrease in LDL cholesterol. significantly reduce serum total cholesterol and LDL choles- terol levels. The effect sizes of psyllium as bulk laxatives on total cholesterol and LDL cholesterol were 0.442 mmol/l (95% CI: 0.210–0.674 mmol/l) and 0.300 mmol/l (95% psyllium 5, 10 and 15 g/day could result in 5.6, 9.0 and CI: 0.167–0.430 mmol/l) respectively, whereas that of 12.5% decrease of LDL cholesterol level respectively. psyllium as enriched foods were 0.320 mmol/l (95% CI: 0.162–0.477 mmol/l) and 0.260 mmol/l (95% CI: 0.180–0.340 mmol/l), respectively. The form of psyllium Long-term effects of psyllium on lipid seemed to have no significant influence on lipid-lowering There was a long-term effect of psyllium on total cholesterol effects (P ¼ 0.2367 for total cholesterol and P ¼ 0.5688 for and LDL cholesterol (Figure 5). The equation was fitted using LDL cholesterol).

European Journal of Clinical Nutrition Time- and dose-dependent effect of psyllium on serum lipids Z-h Wei et al 826 Table 3 Pooled treatment effect on total and LDL cholesterol in subgroups of trials defined by study design features

Subgroup No. of trials Effect sizea (95% CI) mmol/l

TC LDL

Dose ¼ 10.2 g/day 10 0.388 (0.193,0.583) 0.267 (0.185, 0.309) Psyllium-enriched foods 11 0.320 (0.162, 0.477) 0.260 (0.180, 0.340) Step I diet 9 0.420 (0.211, 0.629) 0.271 (0.175, 0.320) Duration ¼ 8 weeks 8 0.357 (0.203, 0.512) 0.283 (0.164, 0.343) Parallel design 15 0.415 (0.264, 0.567) 0.292 (0.215, 0.331) Study in USA 14 0.396 (0.238, 0.554) 0.287 (0.203, 0.329)

aEffect size for each subgroup of total cholesterol was computed by random-effect model and that of LDL cholesterol by fixed-effect model. TC: total cholesterol; LDL: LDL cholesterol.

Subgroup analysis one of the most effective lipid-lowering agents. In 1998, FDA Effect size of each subgroup was similar to the pooled effect ruled that labels on certain foods containing soluble fiber from size of all studies in this meta-analysis (Table 3). psyllium seed husk, such as certain breakfast cereals, might claim that these foods, as part of a diet low in saturated fat and cholesterol, may reduce the risk of coronary heart disease. Discussion Newer dietary approaches combining cholesterol-lowering drugs or foods may offer another option. The synthesized In this meta-analysis of 21 studies, subjects who admini- effects of a combination therapy including psyllium and low- strated psyllium bulk laxatives or enriched foods had dose statins (Moreyra et al., 2005) or plant sterols (Shrestha significantly decreased total cholesterol, LDL cholesterol et al., 2006) have been demonstrated in clinical trials. and HDL cholesterol concentrations. Compared with place- We believed that the results of this meta-analysis repre- bo control, the mean decrease of total cholesterol, LDL sented a conservative estimate of psyllium effect on lipid- cholesterol and HDL cholesterol were 0.375, 0.278 and lowering. We could conclude that there was a publication 0.0353 mmol/l, respectively. But for triglycerides, there was bias in this meta-analysis, based on the results of funnel little effect. plots. The publication bias was mainly caused by ‘large effect We established a dose–response relationship between but small trial’ phenomenon. The effect size would be larger psyllium and serum total and LDL cholesterol levels. if all those small trials were published and selected into the However, it was a logarithmic relationship, which indicated meta-analysis. that greatly increasing dosage of psyllium did not produce a Consuming psyllium as a bulk-forming fiber laxative is correspondingly large effect size. Although psyllium was well functionally equivalent to consuming psyllium-enriched foods. tolerated, high dosage (20.4 g per day) might cause diarrhea Clinical studies evaluating psyllium administered as bulk (Neal and Balm, 1990). laxatives were conducted in a fashion similar to the studies Long-term intake of psyllium could cause sustained conducted with psyllium-enriched foods, including patient decrease in the total and LDL cholesterol levels. It had been selection criteria and regimen, thus efficacy should be compar- estimated that a fall of 0.6 mmol/l in total cholesterol would able for bulk laxative and enriched foods studies. The results of reduce subsequent cardiovascular risk by 50% among those this meta-analysis confirmed that the effect of these two forms aged 40 and by 20% among those aged 70 (Betteridge, 1994). on serum lipid levels has a non-significant difference. Having Every 1% reduction in LDL cholesterol could reduce the risk both available options would provide alternatives to improve of CHD by 1.2–2.0% (Katan et al., 2003). Based on the result of the compliance of a fiber intake recommendation. Moreover, this meta-analysis, after consuming psyllium for 20 weeks, psyllium is one of the most commonly used over-the-counter serum total cholesterol level could be reduced from baseline drug for the treatment of chronic constipation, especially in the level (about 6.36 mmol/l) to 5.73 mmol/l and LDL cholesterol elders who have a high incidence of hypercholesterolemia, concentration could be dropped from 4.31 to 4.0 mmol/l with thus psyllium would provide another potential benefit for the decrease rate as 9%. And after psyllium consumption for them due to its efficacy in lowering serum cholesterol levels at about 1 year and a half, the LDL cholesterol could get to thedoseincommonuse. 3.1 mmol/l (the upper limit of the reference range). CHD is a major cause of death in United States and most Western countries. Blood cholesterol is a major risk factor of Acknowledgements coronary heart disease. Dietary and pharmacological reduc- tions in total and LDL cholesterol decrease the risk of This work was partially supported by Cooperative Program coronary events, so dietary has been recommended as a safe of Rhone-Alps Region and Shanghai. We thank Professor and practical approach for cholesterol reduction. Psyllium is Jean-Pierre Boissel and Dr Michael Cucherat (Institute of

European Journal of Clinical Nutrition Time- and dose-dependent effect of psyllium on serum lipids Z-h Wei et al 827 Clinical Pharmacology, Lyon University I, France) for Houwelingen HC, Arends LR, Stijnen T (2002). Advanced methods in expert advice. meta-analysis: multivariate approach and meta-regression. Stat Med 21, 589–624. Jenkins DJ, Wolever TM, Vidgen E, Kendall CW, Ransom TP, Mehling CC et al. (1997). Effect of psyllium in hyper- References cholesterolemia at two monounsaturated fatty acid intakes. Am J Clin Nutr 65, 1524–1533. Adult Treatment Panel III (2001). Executive Summary of the Third Katan MB, Grundy SM, Jones P, Law M, Miettinen T, Paolrtti R (2003). Report of the National Cholesterol Education Program (NCEP) Efficacy and safety of plant stanols and sterols in the management Expert Panel on Detection, Evaluation, and Treatment of High of blood cholesterol levels. Mayo Clin Proc 78, 965–978. Blood Cholesterol in Adults. JAMA 285, 2486–2497. Levin EG, Miller VT, Muesing RA, Stoy DB, Balm TK, LaRosa JC Anderson JW, Allgood LD, Lawrence A, Altringer LA, Jerdack GR, (1990). Comparison of psyllium hydrophilic mucilloid and Hengehold DA et al. (2000a). Cholesterol-lowering effects of cellulose as adjuncts to a prudent diet in the treatment of psyllium intake adjunctive to diet therapy in men and women mild to moderate hypercholesterolemia. Arch Intern Med 150, with hypercholesterolemia: meta-analysis of 8 controlled trials. 1822–1827. Am J Clin Nutr 71, 472–479. Maciejko JJ, Brazg R, Shah A, Patil S, Rubenfire M (1994). Psyllium for Anderson JW, Davidson MH, Blonde L, Brown WV, Howard WJ, the reduction of cholestyramine-associated gastrointestinal symp- Ginsberg H et al. (2000b). Long-term cholesterol-lowering effects toms in the treatment of primary hypercholesterolemia. Arch Fam of psyllium as an adjunct to diet therapy in the treatment of Med 3, 955–960. hypercholesterolemia. Am J Clin Nutr 71, 1433–1438. MacMahon M, Carless J (1998). Ispaghula husk in the treatment of Anderson JW, Floore TL, Geil PB, O’Neal DS, Balm TK (1991). hypercholesterolemia: a double-blind controlled study. J Cardio- Hypocholesterolemic effects of different bulk-forming hydrophilic vasc Risk 5, 167–172. fibers as adjuncts to dietary therapy in mild to moderate Moreyra AE, Wilson AC, Koraym A (2005). Effect of combining hypercholesterolemia. Arch Intern Med 151, 1597–1602. psyllium fiber with simvastatin in lowering cholesterol. Arch Intern Anderson JW, Riddell-Mason S, Gustafson NJ, Smith SF, Mackey M Med 165, 1161–1166. (1992). Cholesterol-lowering effects of psyllium-enriched cereal as Neal GW, Balm TK (1990). Synergistic effects of psyllium in an adjunct to a prudent diet in the treatment of mild to moderate the dietary treatment of hypercholesterolemia. South Med J 83, hypercholesterolemia. Am J Clin Nutr 56, 93–98. 1131–1137. Anderson JW, Zettwoch N, Feldman T, Tietyen-Clark J, Oeltgen P, Olson BH, Anderson SM, Becker MP, Anderson JW, Hunninghake DB, Bishop CW (1988). Cholesterol-lowering effects of psyllium Jenkins DJ et al. (1997). Psyllium-enrich cereals lower blood total hydrophilic mucilloid for hypercholesterolemia. Arch Intern Med cholesterol and LDL cholesterol, but not HDL cholesterol, in 148, 292–296. hypercholesterolemic adults: results of a meta-analysis. J Nutr 127, Anne W (2002). Meta-Analysis of Controlled Clinical Trial. The Atrium, 1973–1980. Southern Gate, John Wiley & Sons Ltd: Chichester, West Sussex, Petchetti L, Frishman WH, Petrillo R, Raju K (2007). Nutriceuticals in England. cardiovascular disease: psyllium. Cardiol Rev 15, 116–122. Bell LP, Hectorne KJ, Reynolds H, Balm TK, Hunninghake DB (1989). Roberts DC, Truswell AS, Bencke A, Dewar HM, Farmakalidis E Cholesterol-lowering effects of psyllium hydrophilic mucilloid: (1994). The cholesterol-lowering effect of a breakfast cereal adjunct therapy to a prudent diet for patients with mild to containing psyllium fibre. Med J Aust 161, 660–664. moderate hypercholesterolemia. JAMA 261, 3419–3423. Romero AL, Romero JE, Galaviz S, Fernandez ML (1998). Cookies Bell LP, Hectorn KJ, Reynolds H, Hunninghake DB (1990). Cholester- enriched with psyllium or oat bran lower plasma LDL cholesterol ol-lowering effects of soluble-fiber cereals as part of a prudent in normal and hypercholesterolemic men from Northern Mexico. diet for patients with mild to moderate hypercholesterolemia. J Am Coll Nutr 7, 601–608. Am J Clin Nutr 52, 1020–1026. Shrestha S, Volek JS, Udani J, Wood RJ, Greene CM, Aggarwal D et al. Berkey CS, Hoaglin DC, Mosteller F, Colditz GA (1995). A random- (2006). A combination therapy including psyllium and plant effects regression models for meta-analysis. Stat Med 14, 395–411. sterols lowers LDL cholesterol by modifying lipoprotein metabo- Betteridge DJ (1994). Cholesterol lowering and CHD prevention: lism in hypercholesterolemic individuals. J Nutr 136, 2492–2497. time to get on with it. Br J Clin Pract 48, 115–117. Singh B (2007). Psyllium as therapeutic and drug delivery agent. Brown L, Rosner B, Willett WW, Sacks FM (1999). Cholesterol- Int J Pharm 334, 1–14. lowering effects of dietary fiber: a meta-analysis. Am J Clin Nutr 69, Spence JD, Huff MW, Heidenheim P, Viswanatha A, Munoz C, 30–42. Lindsay R et al. (1995). Combination therapy with colestipol and Brunner EJ, Rees K, Ward K, Burke M, Thorogood M (2007). Dietary psyllium mucilloid in patients with hyperlipidemia. Ann Intern advice for reducing cardiovascular risk. Cochrane Database Syst Rev Med 123, 493–499. 17, CD002128. Sprecher DL, Harris BV, Goldberg AC, Anderson EC, Bayuk LM, Davidson MH, Maki KC, Kong JC, Dugan LD, Torri SA, Hall HA et al. Russell BS et al. (1993). Efficacy of psyllium in reducing serum (1998). Long-term effects of consuming foods containing psyllium cholesterol levels in hypercholesterolemic patients on high- or seed husk on serum lipids in subjects with hypercholesterolemia. low-fat diets. Ann Intern Med 119, 545–554. Am J Clin Nutr 67, 367–376. Stoy DB, LaRosa JC, Brewer BK, Mackey M, Meusing RA (1993). Egger M, Davey Smith G, Schneider M, Minder CE (1997). Bias in Cholesterol-lowering effects of ready-to-eat cereal containing meta-analysis detected by a simple, graphical test. Br Med J 315, psyllium. J Am Diet Assoc 93, 910–912. 629–634. Summerbell CD, Manley P, Barnes D, Leeds A (1994). The effects of Everson GT, Daggy BP, McKinley C, Story JA (1992). Effects of psyllium on blood lipids in hypercholesterolaemic subjects. JHum psyllium hydrophilic mucilloid on LDL cholesterol and bile acid Nutr Diet 7, 147–151. synthesis in hypercholesterolemic men. J Lipid Res 33, 1183–1192. Wolever TM, Jenkins DJ, Mueller S, Boctor DL, Ransom TP, Patten R Flannery J, Raulerson A (2000). Hypercholesterolemia: a look at low- et al. (1994). Method of administration influences the serum cost treatment and treatment adherence. J Am Acad Nurse Pract 12, cholesterol-lowering effect of psyllium. Am J Clin Nutr 59, 462–466. 1055–1059. Garvin JE, Forman DT, Eiseman WR, Phillips CR (1965). Lowering of Wolever TM, Jenkins DJ, Mueller S, Patten R, Relle LK, Boctor D et al. serum cholesterol by oral hydrophilic colloid. Proc Soc Exp Biol Med (1994). Psyllium reduces blood lipids in men and women with 120, 744–746. hyperlipidemia. Am J Med Sci 307, 269–273.

European Journal of Clinical Nutrition