|
|
| |||||||||||||||||||||||||||||||
Background The n-3 fatty acids in fish oil affect eicosanoid and cytokine production and therefore have the potential to alter renal hemodynamics and inflammation. The effects of fish oil could prevent immunologic renal injury in patients with IgA nephropathy.
Methods In a multicenter, placebo-controlled, randomized trial we tested the efficacy of fish oil in patients with IgA nephropathy who had persistent proteinuria. The daily dose of fish oil was 12 g; the placebo was a similar dose of olive oil. Serum creatinine concentrations, elevated in 68 percent of the patients at base line, and creatinine clearance were measured for two years. The primary end point was an increase of 50 percent or more in the serum creatinine concentration at the end of the study.
Results Fifty-five patients were assigned to receive fish oil, and 51 to receive placebo. According to Kaplan-Meier estimation, 3 patients (6 percent) in the fish-oil group and 14 (33 percent) in the placebo group had increases of 50 percent or more in their serum creatinine concentrations during treatment (P = 0.002). The annual median changes in the serum creatinine concentrations were 0.03 mg per deciliter (2.7 µmol per liter) in the fish-oil group and 0.14 mg per deciliter (12.4 µmol per liter) in the placebo group. Proteinuria was slightly reduced and hypertension was controlled to a comparable degree in both groups. The cumulative percentage of patients who died or had end-stage renal disease was 40 percent in the placebo group after four years and 10 percent in the fish-oil group (P = 0.006). No patient discontinued fish-oil treatment because of adverse effects.
Conclusions In patients with IgA nephropathy, treatment with fish oil for two years retards the rate at which renal function is lost.
Methods
The study was a double-blind, placebo-controlled, randomized trial performed in 21 centers making up the Mayo Nephrology Collaborative Group. The study protocol was approved by the institutional review board at each center, and all patients gave written informed consent.
Selection of Patients
Patients were eligible for the study if they had biopsy-proved IgA nephropathy. The diagnosis was based on a semiquantitative evaluation of all lesions in the categories of focal (mesangial) and diffuse proliferative glomerulonephritis, and the diagnosis was confirmed by immunofluorescence studies showing mesangial deposition of IgA8. All renal-biopsy samples were examined independently by one pathologist, who was not provided with any clinical information about the patients. Patients given a morphologic diagnosis were enrolled in the study if their urinary protein excretion was 
1 g per day (or their serum creatinine concentration had increased by 25 percent or more during the preceding six months), their serum creatinine concentration was 
3.0 mg per deciliter (265 µmol per liter), and their expected survival was two years or more. Patients were excluded if they had systemic lupus erythematosus, chronic liver disease, or antiglomerular basement membrane glomerulonephritis or were pregnant or lactating. One patient each had dermatitis herpetiformis, idiopathic hypertrophic subaortic stenosis, and hereditary homocystinuria in addition to IgA nephropathy.
Study Design
The patients were randomized within strata determined by serum creatinine concentrations (either normal, defined as 1.2 mg per deciliter [106 µmol per liter] in men and 0.9 mg per deciliter [80 µmol per liter] in women, or elevated, defined as above the normal values but 3.0 mg per deciliter), the average of two 24-hour urinary protein values (either 1.0 to 3.4 g per day or 3.5 g per day), and the average of two blood-pressure readings by the same person on two different days during office visits (either normotensive, defined as a systolic pressure <140 mm Hg and a diastolic pressure <85 mm Hg in a patient not being treated for hypertension, or hypertensive, defined as a systolic pressure 140 mm Hg, a diastolic pressure 85 mm Hg, or current treatment for hypertension).
The patients were randomly assigned to receive dietary supplementation with either fish oil or placebo for two years. From June 1988 to March 1991, the fish-oil supplement (MaxEPA, Seven Seas Health Care, Hull, United Kingdom) was given in a dosage of 6 one-gram soft-gelatin capsules twice daily, providing 1.87 g of eicosapentaenoic acid and 1.36 g of docosahexaenoic acid -- the two major n-3 polyunsaturated fatty acids found in fish oil. The matched placebo was olive oil (12 one-gram gelatin capsules per day, containing <0.05 mg each of eicosapentaenoic acid and docosahexaenoic acid per gram). After March 1991, the fish-oil supplement was obtained from the Fish Oil Test Materials Program, which is sponsored by the National Institutes of Health and the National Oceanic and Atmospheric Administration and is based at the Southeast Fisheries Science Center, National Marine Fisheries Service, Charleston, South Carolina. This supplement was given in a daily dosage of 12 one-gram vacuum-deodorized gelatin capsules of menhaden oil that provided 1.68 g of eicosapentaenoic acid and 0.97 g of docosahexaenoic acid. Both the fish oils and the placebo contained 0.2 mg of tertiary butylhydroquinone per gram of oil as an antioxidant, 2 mg of tocopherol per gram of oil, and 0.1 percent peppermint oil.
Patients with hypertension were treated with the angiotensin-converting-enzyme inhibitor enalapril (Vasotec, Merck, Sharp & Dohme Laboratories, West Point, Pa.). If the target blood pressure of 140/85 mm Hg was not achieved with enalapril, other antihypertensive agents were added at the physician's discretion. Each patient's blood pressure was measured and recorded by the same person at each visit and at about the same time of day, with the patient seated; no tobacco or caffeine was consumed for 30 minutes before the measurement. To avoid potentially detrimental hemodynamic effects, mild dietary sodium restriction (a limit of 90 mmol per day) was advised. Diuretic therapy was avoided unless the patient had severe edema.
At study entry, complete medical histories were taken and physical examinations were performed in all patients. Initial clinical and laboratory results were forwarded to the coordinating center. Follow-up patient examinations and measurements of renal function, protein excretion, and plasma lipids and phospholipids were scheduled after 1.5, 6, 12, 18, and 24 months of treatment. At the end of this period, the treatment code was broken, and the patients and physicians were informed of the treatment and given the option of continuing, starting, or stopping treatment with fish oil. Compliance with the coded medication regimen was determined by capsule counts and measurements of the fatty-acid profiles of plasma phospholipids. Except for one center, which recruited 44 patients, the centers recruited fewer than 10 patients each, so there were insufficient numbers for an analysis of an effect of the study center. To reduce variability, at each scheduled visit we measured serum creatinine; plasma total, high-density lipoprotein (HDL), and low-density lipoprotein (LDL) cholesterol; plasma triglycerides; and 24-hour urinary excretion of creatinine, protein, and nitrogen at a central laboratory using standard methods. The fatty-acid composition (including n-6 and n-3 fatty acids) of plasma phospholipids was measured by capillary gas-liquid chromatography14,15. The patients were asked to avoid aspirin, medicines containing aspirin, and other nonsteroidal antiinflammatory drugs for one week before these laboratory tests.
Study End Points
The primary end point of the study was the loss of renal function, defined as an increase of 50 percent or more in the serum creatinine concentration during the 2-year treatment period (which in some patients was actually as short as 1.7 years or as long as 2.3 years). Secondary end points included the annual rate of change in the serum creatinine concentration, 24-hour creatinine clearance, and urinary protein excretion. The treatment code was broken for patients who had an increase of 50 percent or more in the serum creatinine concentration; treatment could then be continued or stopped at the discretion of the physician. These patients were considered to have left the study at that time. All the patients, including those who left the study, were followed after the treatment period ended (until April 1994) to identify cases of end-stage renal disease, defined as a requirement for dialysis or renal transplantation, or death.
Statistical Analysis
Univariate base-line comparisons between the fish-oil and placebo groups were made with the chi-square test (for nominal data) and the rank-sum test (for ordinal data). The cumulative percentage of patients whose serum creatinine concentrations increased by 50 percent or more in two years was calculated with the Kaplan-Meier method16. Data on patients who did not have an increase of this magnitude and who did not complete the study were censored at the time of their last visit. Univariate comparisons of the cumulative percentages whose serum creatinine concentrations increased by 50 percent or more were made with the log-rank test17. The formal test for the effect of treatment on the primary end point was performed with proportional-hazards analysis,18 with the stratification factors (hypertension, an elevated serum creatinine concentration, and urinary protein excretion 3.5 g per day) included as adjusting predictor variables. Linear regression analysis was used to estimate the annual rates of change (slopes) in serum creatinine concentration, creatinine clearance, urinary protein excretion, and plasma lipid concentrations for each patient. The rank-sum test was used to compare the distributions of the slopes in the two groups.
The cumulative percentage of patients with the end point of death or end-stage renal disease (after a maximal follow-up period of 5.8 years) was also estimated by the Kaplan-Meier method16. This analysis included events occurring after the two-year treatment period. We also compared survival curves using the log-rank test17. All statistical tests were two-sided.
Results
Between June 1988 and December 1991, 142 patients with idiopathic IgA nephropathy were evaluated for this study. One hundred six patients met the clinical criteria for eligibility and were willing to enter the study (44 from the Mayo Clinic in Rochester, Minnesota, and 62 from the 20 other centers in the collaborative group). Fifty-five patients were assigned to receive fish oil, and 51 to receive placebo. The clinical and laboratory characteristics of the patients in the two groups were similar (Table 1), as were the illnesses that preceded the clinical onset of IgA nephropathy, such as upper respiratory infection and gastroenteritis, and the clinical events that occurred at onset, including proteinuria, hematuria, hypertension, and edema. For example, 14 patients in the fish-oil group and 13 in the placebo group had episodic macroscopic hematuria before entry.
|
Seventy-five patients completed the two years of treatment, and 31 did not. Six patients in each group did not comply with treatment, and one patient in each group was withdrawn from the study because of intercurrent illness. Seventeen patients (3 in the fish-oil group and 14 in the placebo group) reached the primary end point of the study -- an increase of 50 percent or more in the serum creatinine concentration at two years. The respective Kaplan-Meier estimates were 6 percent and 33 percent (P = 0.002) (Figure 1). The difference between treatments remained significant when adjusted for the three stratification factors (relative risk in the fish-oil group, 0.18; 95 percent confidence interval, 0.05 to 0.63; P = 0.008). The primary end point occurred significantly more frequently in the placebo group at all levels of the three stratification factors (hypertension, higher serum creatinine concentration, and proteinuria in the nephrotic range) (Table 2).
|
|
During the study, 61 percent of the patients (34 patients in the fish-oil group and 31 in the placebo group) received antihypertensive treatment. Thirty-two patients were treated with enalapril, and 29 were treated with enalapril and other drugs, including calcium antagonists, adrenergic antagonists, centrally acting alpha2-adrenergic agonists, and diuretic agents. Four patients received only drugs other than enalapril. The changes in blood pressure did not differ significantly in either group between the patients who were hypertensive at base line and those who were normotensive (Table 3). Five patients (three in the fish-oil group and two in the placebo group) had adverse reactions to enalapril -- persistent hyperkalemia (serum potassium concentrations, >6 mmol per liter) in two patients, cough in two patients, and angioedema in one patient. Despite the adverse effects, all the patients continued to take fish oil or placebo.
|
During an average follow-up of three years, 19 of the 106 patients (5 in the fish-oil group and 14 in the placebo group, P = 0.006) died or required repeated dialysis or renal transplantation (Figure 2). Two of these patients (one in each group) died of metastatic lung cancer; the patient in the placebo group had end-stage renal disease. Among the remaining 18 patients in whom end-stage renal disease ultimately developed, 2 of the 4 in the fish-oil group and 9 of the 14 in the placebo group had reached the primary end point during the two-year treatment period. Five of the nine patients in the placebo group who reached the primary end point during the two-year period later received fish oil, which did not seem to slow the progression of their disease.
|
Renal function was better maintained in the patients in the fish-oil group, as indicated by the median annual changes in serum creatinine concentrations (measured in 103 patients) and creatinine clearance (102 patients) (Figure 3). The annual median changes in the serum creatinine concentrations were 0.03 mg per deciliter (2.7 µmol per liter) in the fish-oil group and 0.14 mg per deciliter (12.4 µmol per liter) in the placebo group (P = 0.001). The median annual change in 24-hour urinary protein excretion was -0.23 g (-15 percent) in the fish-oil group and -0.10 g (-7 percent) in the placebo group (P = 0.54). The median annual changes in plasma lipid concentrations were small, ranging from -8.3 to +0.8 mg per deciliter, and were similar in the two groups.
|
As a measure of compliance, plasma phospholipid fatty-acid profiles were measured in 98 patients (52 in the fish-oil group and 46 in the placebo group). The ratio of eicosapentaenoic acid (C20:5,n-3), the principal n-3 fatty acid, to arachidonic acid (C20:4,n-6), the principal n-6 fatty acid, increased significantly (P<0.001) in the fish-oil group but not in the placebo group (Figure 4). No adverse reactions (e.g., bleeding) required the discontinuation of treatment in any patient. The mean (±SD) bleeding time (measured in 18 patients in the fish-oil group and 16 in the placebo group) was 3.9 ±1.3 minutes at base line and 4.1 ±2.3 minutes after 12 months in the fish-oil group and 3.8 ±1.7 minutes at base line and 4.0 ±1.6 minutes after 12 months in the placebo group. The only complaint, by some patients in the fish-oil group, was of a fishy aftertaste despite the addition of peppermint oil to the fish oil.
|
Dietary fish-oil supplementation significantly slowed the rate of loss of renal function in patients with IgA nephropathy. During the two-year treatment period, only 6 percent of the fish-oil group had an increase in serum creatinine concentration of 50 percent or more, as compared with 33 percent of the placebo group, and the annual decline in renal function was slower in the fish-oil group as a whole. The beneficial effect of fish oil was independent of the presence of hypertension and the degree of elevation of serum creatinine and urinary protein excretion at base line.
The rationale for using fish oil in IgA nephropathy13 is that n-3 fatty acids alter several biologic systems, including eicosanoid13 and cytokine19 production, membrane fluidity and blood rheology,20,21 thromboxane A2-dependent platelet aggregation,22 and the activity of platelet-activating factor23. The net effects could reduce glomerular injury. In the long-term prognosis of IgA nephropathy, immunologic factors have little influence, and hypertension, arteriosclerosis, and aging are more important24. Fish consumption may reduce the incidence of stroke,25 and it reduces mortality from myocardial infarction26.
The n-3 fatty acids improve renal hemodynamics in renal-transplant recipients treated with cyclosporine27; reduce proteinuria in patients with chronic glomerular disease28; lower plasma triglyceride concentrations, reduce whole-blood viscosity, and improve red-cell flexibility in patients with lupus nephritis21; and reduce platelet hyperaggregation in patients with the idiopathic nephrotic syndrome29. The finding of increased plasma LDL cholesterol concentrations and concomitantly decreased HDL:LDL cholesterol ratios in the study by Hall et al.29 may preclude the use of fish oil as a hypolipidemic agent in patients with nephrotic syndrome, but in our study plasma cholesterol and triglyceride concentrations did not change in the patients treated with fish oil.
The results of treating patients with IgA nephropathy have been conflicting. Among 20 Japanese patients, renal function did not deteriorate in 9 who were treated with fish oil for one year but did deteriorate in 11 who were untreated9. In a study of 11 patients with advanced renal failure who were treated with fish oil and nonsteroidal antiinflammatory drugs, antiplatelet agents, and immunosuppressive drugs in Hong Kong,10 renal function continued to decline in all but 1, but it is unlikely that any sort of treatment would have had much effect in these patients because of the severity of their renal disease. In a randomized two-year trial in patients with IgA nephropathy in Australia, renal function declined at the same rate in 17 patients given fish oil and 20 untreated patients11. In a six-month randomized trial in Sweden, renal function deteriorated in 15 patients treated with fish oil and did not change in 17 given placebo (corn oil)12. The differences in the effects of fish oil in patients with IgA nephropathy may be due to differences in the disease itself, including differences in the rate of progression1,2,3,4,5; the number, place of residence, or ethnicity of the patients; and the composition of the fish oil used and the duration of treatment. Our results, in a larger number of subjects than heretofore studied, and obtained with a specific material meeting stated compositional specifications, provide new evidence of the value of fish oil in patients with IgA nephropathy.
Supported by grants from the Mayo Foundation, the Medical School Grant Program of Merck, Sharp & Dohme Research Laboratories, Ross Laboratories, and Seven Seas Health Care.
We are indebted to Dr. B.J. Holub, Department of Nutritional Sciences, University of Guelph, Guelph, Ontario, Canada, for the measurement of plasma phospholipids.
Source Information
From the Division of Nephrology, Department of Medicine (J.V.D., D.C.S.), the Section of Biostatistics (E.J.B., K.P.O.), and the Department of Laboratory Medicine and Pathology (K.E.H.), Mayo Clinic and Mayo Foundation, Rochester, Minn. The investigators of the Mayo Nephrology Collaborative Group are listed in the Appendix.
Address reprint requests to Dr. Donadio at the Division of Nephrology, Mayo Clinic, 200 First St., SW, Rochester, MN 55905.
References
3). Biochem Biophys Res Commun 1983;116:1091-1099. [CrossRef][Medline]
The Mayo Nephrology Collaborative Group comprises the following investigators: Clinical Coordinating Center -- J. Donadio, D. Spencer, K. Holley, and D. Wilson; Section of Biostatistics -- E. Bergstralh, K. Offord, C. Schleck; Mayo Clinic, Rochester, Minn. -- T. Larson, B. Morgenstern, D. Milliner, and J. Velosa; Intermed Consultants, Edina, Minn. -- W. Mairhofer, T. Slack, and J. Somerville; Duluth Clinic, Duluth, Minn. -- R. Hellman, T. Johnson, M. Maddy, G. Mix, and R. Russ; Marshfield Clinic, Marshfield, Wis. -- R. Dart, D. Duffy, and J. Parker; McFarland Clinic, Ames, Iowa -- J. Graves; Patch-Levy Partnership, Bangor, Me. -- D. Levy; Haywood Professional Center, Frederick, Md. -- M. Rubin; Renal and Endocrine Associates, Mesa, Ariz. -- D. Reichert; Lincoln, Nebr. -- S. Youngberg and S. Liggett; Gunderson Clinic, LaCrosse, Wis. -- P. Dahlberg and J. Singer; Hennepin County Medical Center, Minneapolis -- J. Opsahl; LaSalle Clinic, Neenah, Wis. -- D. Hathaway and M. Madden; Kansas City, Mo. -- J. Mertz and B. Wood; Columbia Nephrology Associates, Columbia, S.C. -- W. Edwards; Carbondale, Ill. -- M. Webel; Burns Clinic, Petoskey, Mich. -- G. Shaw; Renal Associates, Grand Rapids, Mich. -- M. Boelkins; West Michigan Nephrology, Muskegon, Mich. -- G. Downer and T. Bao-Min; Cumberland, Md. -- R. Welik; Mayo Clinic, Scottsdale, Ariz. -- R. Heilman and D. Wochos; and Mayo Clinic, Jacksonville, Fla. -- P. Fitzpatrick and C. Van Den Berg.
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
