- Email: [email protected]
Low-antigen-content diet in the treatment of patients with mixed cryoglobulinemia
Low-Antigen-Content Diet in the Treatment of Patients with Mixed Cryoglobulinemia CLODOVEO FERRI, M.D. Pisa, ItalyMAURIZIO PIETROGRANDE, M.D. Milano, ItalyRICCARDO CECCHECII, M.D., ANTONIO TAVONI, M.D. Pisa, Italy ALBERTO CEFALO, M.D., GIAMPIERO BUZZETTI, M.D. Milano, ItalyCLAUDIO VlTALI, M.D., STEFANO BOMBARDIERI, M.D. Pisa, Italy
PURPOSE: The effects of a low-antigen-content diet (LAC diet) versus a standard normocaloric diet on the signs and symptoms of mixed cryoglobulinemia (MC) were compared in a crossover randomized study. PATIENTSAND METHODS:The study consisted of 24 outpatients with MC, and was carried out in a 48-week period. After 18 weeks of either the LAC or the placebo diet, patients returned to a totally unrestricted diet for 12 w e e k s (washout period) and crossed over to the second half of the study for the other 18 weeks. RESULTS:After three weeks of the restricted LAC diet, the cryocrit decreased from 3.5 4- 3.4% (mean + SD) to 2.3 + 2.0% (p <0.01), and the circulating immune c o m p l e x levels decreased from 48 + 30% to 39 4- 34% (p <0.01). At the same time, the purpura score (p <0.05), glutamic pyruvic tran~aminase level (p <0.01), and g a m m a glutamyl transferase level (p <0.001) significantly improved. Splenic reticuloendothelial function, measured as the haftlife of heat-damaged autologous red cells, decreased from 41 + 21 minutes to 21 + 10 minutes (p <0.005). In contrast, no significant parallel clinical, biochemical, and immunologic changes occurred in
the same patients during the placebo (standard normocaloric) diet. CONCLUSION:These data show that an LAC diet decreases the amount o f circulating immune complexes in MC and can modify certain signs and symptoms of the disease. These effects of the LAC diet may be explained by postulating a functional restoration of the m o n o n u c l e a r phagocytic system.
ixed cryoglobulinemia (MC) is a systemic huM man disease characterized by the presence of large amounts of circulating cold precipitable complexes formed by monoclonal or polyclonal rheumatoid factor and polyclonal IgG [1,2]. Etiology of this condition is largely unknown, but current evidence indicates that most of the systemic organ involvement is mediated by the tissue deposition of circulating immune complexes [3,4]. Thus, the management of MC has mainly relied on reducing the load of immune complexes or their noxious effects, and generally includes the quantitative removal of circulating immune complexes in various ways (plasmapheresis and related technologies) [5,6], the use of cytotoxic agents, and corticosteroids [7]. In the present work, we have used a novel therapeutic approach, based on some theoretic assumptions [8] and on preliminary results of a study by one of us [9]. According to the proposed dynamic model, the circulating complexes in MC would compete for their catabolism with other inputs directed to the mononuclear phagocytic system. In normal conditions, such inputs include primarily blood-borne constituents and high-molecular-weight exogenous substances crossing the mucosal barrier of the gut [10,11] and reaching the systemic circulation via lymph and portal veins [12]. Thus, a dietary regimen capable of reducing this latter component could help to restore a saturated mononuclear phagocytic system, which in turn could be left free to remove the circulating immune complexes. We report the results of a randomized crossover study that compares the effects of a placebo "sham" diet and a dietary regimen of putative low-antigen content (LAC diet) on the signs and symptoms of MC. PATIENTS A N D METHODS Patients
From the Istituto di Patologia Medica I e Servizio di Reumatologia, Universitfi di Pisa, Pisa, Italy (CF, RC, AT, CV, SB); Centro Studi di Medicina Teoretica, Istituto di Medicina Interna, Universitfi di Milano, Milano, Italy (MP, AC); and Servizio di Dietologia, Ospedale di Niguarda, Milano, Italy (GB). Requestsfor reprints should be addressed to Clodoveo Ferri, M.D., Istituto di Patologia Medica I, Ospedale S. Chiara, Via Roma, 67, 56100 Pisa, Italy. Manuscript submitted January 20, 1989, and accepted in revised form July 20, 1989.
Of the 25 patients with MC who entered the study, 24 (19 women, five men) completed the trial. Patients' ages ranged from 42 to 68 years (mean, 56.3 years), and disease duration ranged from one to 15 years (mean, 6.4 years). Nine patients had type II cryoglobulin and 15 had type III cryoglobulin. Cryocrit varied from 0.3% to 21%. Seven of 24 patients had anti-hepatitis B virus (HBV) antibodies, and none had HBV-related antigens, either in the sera or in the cryoglobulins (Table I). Since no patient showed evidence of antigenemia but only signs of a previous HBV exposition (i.e., antibody to hepatitis B surface antigen), the putative role of HBV in these patients was not warranted and, therefore, they could be similarly admitted to the study. The diagnosis of MC, established in all cases at least 12 months before, was based on the presence of the typical clinical syndrome (purpura, arthralgias, weakness), mixed cryoglobulins with rheumatoid fac-
November 1989 The AmericanJournal of Medicine Volume87
519
DIETETIC THERAPY IN MIXED CRYOGLOBULINEMIA
TABLE I Epidemiologic Patient Number
and Immunologic
Data on 24 MC Patients before Diet Therapy
59F
fZ :i
42M 50M 44F 65F
CIC-K (%I
HBV Exposure
G-M-h’ G-M-A
iz
::: 5.6 1.4
G-M G-M-A G-M-A
ii i:
108 152 118 115
32 ii
: -
0.3
G-M-A
107
N”i 42
+
2.2
1.1 1.5 21 1.3 1.3 I’:!
11
44F 56F
RES (half-life minutes)
2:
:::
55F 43M 59F 66F
CHso
WJ/mU <50 105 154 <50 <50 127 <50 <50 124 126 129 126 <50 <50
:::
61M 54F 59F 51F
::
Cryopreclpitate Ig Comporltlon*
Cryocrlt (%I
Age/Sex
E 68F 51F 61F 67F 49F 61M 54F
/ FERRI ET AL
2:
;:[W:j-A
:::
G-M(k)-A G-M-A G-M
:7” ii ;;
115 141 174 <50 <50
:: :: ND
ii
Ii:, :
C-K = circulating immune complexes evaluated by means of bovine conglutinin as recognition unit (normal value < 40%); CHw = hemolytic complement iormal value 160-220 HU (hemolytic units]/mL); RES = reticuloendothelial spleen function expressed by (half-life minutes) (normal value < 30 minutes); tpatitis B virus exposure; ND = not done. IgA was constantly present in cryoprecipitate as trace amounts.
tor activity, and the absence of other known infectious, neoplastic, or systemic disorders [l]. Patients were consecutively admitted to the study when they had: (1) at least two of the typical symptoms (purpura, arthralgias, weakness),(2) a measurablecryocrit level, (3) involvement of at least one main organ (liver, renal, and/or neurologic involvement). In all patients, the absenceof other medical or nutritional problems wasascertained and informed consent was obtained. In the last 12 months before the study, 19 patients were treated with a low dosageof steroids (2 to 8 mg/day of B-methyl prednisolone), while five were untreated. In all cases,the prestudy corticosteroid dosage was maintained unchanged during the course of the trial. TABLE II LAC Diet Protocol*
Category
Basal Diet
Cereals Simple carbohydrates Fats Dairy products Meats, fish Em Vegetables Fruits Salt/spices Alcohol, beverage Additives/preservatives
I
Rice Sugar Olive oil None Turkey None Green, potatoes Lzrs, apples
Reintroductions II Ill IV
v
Ail All All
activity HBV =
Diets To prepare a diet with a reduced content of macromolecular antigens, we mostly relied on the current knowledge in food allergy [13]. The LAC diet wassimilar to the restricted hypoallergic diet, with somemodifications. Foods rich in putative macromolecular antigens, like meat, were further restricted; vegetables, fruits, and rice were allowed for the opposite reason. After three weeks of the restricted diet, the various classesof food were gradually reintroduced, according to a fixed protocol (Table II). The placebo diet was a standard dietetic regimen without restrictions on the variety of foods. However, in order for the patients to remain unaware of the identity of the diet they were receiving, close instructions on both regimens were given to them concerning the quantity of each food and how it had to be subdivided in the various meals. Both the placebo and the LAC diets were individually designedin order to be normocaloric (30 kcal/kg/day) , and equilibrated with respect to proteins (17% versus 16.5% of kcal), fats (32% versus 37%), carbohydrates (51% versus 46.5%), vitamins, mineral salts, and oligoelements.
t = clinico-immunologic evaluation reported in Figures 1. 2. and 3. l All diets are equilibrated in elementary components; only the basal diet is moderately hypocalcemic. All patients followed a normocaloric diet according to their ideal body weight.
Study Protocol The study was carried out in a 48-week period in outpatients with MC, randomly assignedto one of two groups. After the first 18 weeks of the LAC or placebo diet, patients returned to a totally unrestricted diet for 12 weeks (washout period) and crossed over to the second half of the study for the other 18 weeks. In order to ensure close adherence to the diet, all patients receiving either the placebo or the LAC diet were given an instruction booklet and a clinical diary for recording the subjective symptom variations and other parameters such as body weight, blood pressure, and drug ingestion. During the study, the patients on
520
87
Water None
Ail Yes All Ail All All Ail
Coffee, tea
Dme (weeks) P
I
November
1999
3
The American
6
Journal
g ‘f
of Medicine
15t
l8
Volume
DIETETIC
the LAC diet never met those on the placebo diet. A clinical and laboratory evaluation was carried out in all cases at the beginning and end of the study and at regular three-week intervals. At each visit, the following data were recorded by the same physician (C.F.) by means of the patient’s self-evaluation and the physical examination: presence and extent of the purpura (score 0 = absent; 1 = limited to lower legs; 2 = limited to entire legs; 3 = diffuse); presence and severity of arthralgias, weakness, peripheral neuropathy, Raynaud’s phenomenon (score 0 = absent; 1 = mild; 2 = moderate; 3 = severe); liver damage activity as measured by serum glutamic pyruvic transaminase (GPT) and gamma glutamyl transferase (GGT) levels; nephropathy by means of serum creatinine levels and proteinuria; arterial hypertension; body weight and drug therapy modifications. At the same three-week intervals, routine blood chemistry studies were carried out by standard methods. Immunologic Evaluation The following immunologic parameters were monitored: (1) cryocrit level was measured as percentage of packed cryoglobulins after cold centrifugation of the serum [14]. Briefly, serum samples,separated at 37”C, were left at 4°C for seven days and then centrifuged in a calibrated test tube at 1,200 X G for 15 minutes at 4°C. The composition of the cryoglobulins was determined by immunodiffusion on .Ouchterlony plates against specific antisera (Behring Institute, West Germany); (2) circulating immune complexes were measured by a competitive enzyme immunoassay using Escherichia coli beta-galactosidase in the probe with bovine conglutinin (CIC-K) as recognition unit [15]. Results were expressed as percent inhibition, and values above 40% were considered to be abnormal; (3) total hemolytic complement activity (CH& was detected by the method of Kent and Fife [16]. At the beginning of the study and after the first three-week period, reticuloendothelial splenic function was evaluated by the method of Bowring et al [17] in 13 patients on the LAC diet and 14 on the placebo diet. Briefly, autologous erythrocytes were heat-damaged at 49°C for 20 minutes, radiolabeled with technetium-99m, and reinjected in the same patient. The half-life was calculated from a disappearance curve, which was obtained by counting radioactivity in red blood cells at different times after the injection. Half-life values above 30 minutes were considered abnormal. Statistical Analysis Statistical analysis of the results was performed by computer using the following: Student’s t-test for paired and unpaired data, chi-square method with Yates’ correction, Wilcoxon non-parametric test. In addition, analysis of variance testing for assessmentof the variance within each group was performed.
RESULTS Table III summarizesthe main clinical features and organ involvement of the 24 patients who completed the study. On the whole, there was a good acceptance of the restricted dietary regimen, although it implied avoiding popular foods like bread and pasta. Results of routine blood chemistry studies remained practically unaffected during the study.
THERAPY
IN MIXED
CRYDGLDBULINEMIA
/ FERRI ET AL
TABLEIII Clinical Dataon 24 MCPatientsTreated with LACDiet Symptoms
Number
Purpura’ Weakness Arthralgias Raynaud’s phenomenon Neuropathy Live;Fvolvementf
13124 lb/24
t GGT Histology-CPH CAH Renal involvement tSerum creatinine and/or proteinuria histology (MPGN) Arterial hypertension*
i;:: 6/24 a/24 4124 L/24 2124 2/24 7124
%
l:;;: ::;z
I
GPT = glutamic pyruvic transaminase; GGT = glutamyl transferase; CPH = chronic persistent hepatitis; CAH = chronic active hepatitis: MPGN = membranoproliferative glomerulonephritis. l During the patient follow-up, purpura was present in all cases. + On the basis of histologic lesions or persistent hepatic enzyme (GPT and GGT) elevation. + Diastolic blood pressure greater than or equal to 90 mm Hg
purpura . :.:.:.:.:.:.:.:.:..:’ . . . . . . . . . . . ..: . .
~~~
:.:.:.:.:.:.:.:.:.:.:. .‘.‘.‘.‘.‘.‘.~.~.‘.‘.’ .‘.~.~.~.~.~.~.~.~.~.~
I L&Z-diet
**
I
I
0
i LI placebo diet
I
I
12
16wks
:iiiii restricted
diet period J
rlgure J. rurpura score (mean f SEM) variations during LAC and placebo diets. Statistical analysis was performed using the Wilcoxon non-parametric test. l p
Clinical Variations During the LAC diet, an improvement of the clinical conditions was observed both subjectively and objectively. The patients claimed an improvement in purpura, weakness,fatigabihty, and arthralgias; to a lesser extent, however, this improvement wasalso present during the placebo diet. Of the 24 patients studied, fresh purpura, weakness,and arthralgias were present in 16,17, and 16 subjects at the beginning of the LAC diet, and in 10, 19, and 16, respectively, before the placebo diet. During the treatment, purpura improved or disappeared in 10 of 16 (63%) subjects versus two of 10 (20%) (LAC diet versus placebo diet), weakness improved or disappeared in eight of 17 (47%) versus three of 19 patients (16%), and arthralgias improved or resolved in four of 16 (25%) versus three of 16 patients (19%). None of these differences were significant (Fisher’s exact test). In contrast, the score for purpura significantly decreased during the LAC diet (Figure l), and these changes were particularly evident after the first six to 12 weeks of treatment. In addition,
November
1989
The American
Journal
of Medicine
Volume
87
521
DIETETIC THERAPY
IN MIXED CRYOGLOBULINEMIA
1::::::;:: e::::: :;:::: ;..:_,.:. :.:.::.::. .:::::.,:: :.:::,:: :...::::. .:._::. :::i:::;: _::::::: _.:.::. .:. :.: .::_.:.: :: :;:.: .:::.:: ** *** ...*t+
/ FERRI ET AL
GGT
GPT
3
: base1 . LAC-diet
3
12
o placebo diet
basal
IEwks jji:j
restricteddiet
l
LAC-diet
12
0 placebo diet
IBwks :.:: : :::_:
restricted
-I
-r
basal
3
Immunologic Parameter Variations Among the immunologic parameters, cryocrit showeda clear-cut decreasein all patients on the LAC diet, and was more pronounced after the first three weeks (Figure 3); this reduction persisted during the subsequent weekswhen the reintroductions were carried out. In all patients, a more transient reduction was observed in CIC-K levels only at the end of the initial three weeksof the restricted regimen. No com-
November
1989
The American
Journal
12
diet period
purpura reappeared or worsened in seven of 24 patients during the placebo diet, but never during the experimental diet. During the LAC diet, biochemical evidence of liver damage was reduced. The improvement was particularly evident after the first three weeks of the restricted diet and persisted for GGT during the subsequent reintroductions (Figure 2). With regard to the other clinical symptoms, no variations were observed in Raynaud’s phenomenon, peripheral neuropathy, arterial hypertension, or nephropathy. Patients with or without associatedsteroid treatment showed similar changes in the symptoms and signs of the disease.
522
1Bwks
CIC-K
60
3
12
period
CRYOCRIT
beaal
3
Figure 2. Liver enzyme variations (GPT = glutamic pyruvic transaminase; GGT = gamma glutamyl transferase) during LAC and placebo diets. 1, 2 = multiple of normal values (mean f SEM). Statistical analysis was performed using the Wilcoxon non-parametric test. **p<0.01, ***p<0.001.
of Medicine
Volume
ISwks
Figure 3. Cryocrit and circulating immune complex (CIC-K) variations during LAC and placebo diets (mean f SEM). Statistical analysis was performed using the paired Student’s t-test. h.v. = normal values. l p
parable changeswere observed in the circulating immune complex moiety during the placebo diet (Figure 3). CHssremained unchanged during both dietary regimens. The clearance of heat-damaged erythrocytes was prolonged (half-life greater than 30 minutes) in nine of 12 patients before the LAC diet and in eight of 13 before the placebo diet. In the first group, the halflife significantly decreased in 10 of 12 patients (p <0.005), and was within the normal range in eight of 12patients at the end of the intensive dietary regimen. Such consistent consensualresults were not present in the placebo diet group (Figure 4). Analysis of variance testing confirmed the significant reduction of cryocrit as well as GGT, GPT, and purpura score in the patients treated with the LAC diet.
COMMENTS The current study demonstrates that the LAC diet exerts positive effects on somesignsand symptoms of MC. Among the clinical features, the most remarkable improvement was observed in purpura, and in the biochemical markers of liver damage. Among the immu87
DIETETIC
nologic parameters, cryoglobulins and, to a lesser extent, circulating immune complexes were significantly reduced, and reticuloendothelial splenic function was improved. It is possible to exclude that these positive results could be ascribed to factors other than diet alone. First of all, some patients were untreated during the study; in the others, the low glucocorticoid dosage was not modified. Second, since the study was carried out in outpatients, the unpredictable effect of bed rest can be excluded. This is still a controversial aspect of the studies involving the use of diet in the management of connective tissue diseases [18]. Hospitalization would in fact favor a closer adherence to the dietary regimen, but bed rest could at the same time profoundly affect certain symptoms such as purpura or arthralgias. Finally, to completely avoid a placebo effect, the patients were told that they were testing the effects of two different therapeutic diets. Above ah, however, clinical, biochemical, and immunologic improvement occurred only during the LAC diet and not during the placebo diet, and symptoms and signs of MC slowly recurred with the progressive return to a totally unrestricted diet. The attempt to manipulate the immune system with various dietary regimens is not new, and several hypotheses were proposed to explain the ensuing results [18,19]. Most of these mechanisms can be reasonably excluded on the basis of the available data. It is known, for instance, that starvation might modify the immune system in experimental conditions [20] or the time course of immune-mediated diseases such as rheumatoid arthritis in humans [21]. Although there was a slight decrease of body weight during the LAC diet in some of our patients, this is certainly not the case. In the last 10 years it has been reported that a diet deprived of certain polyunsaturated fatty acids, known precursors of arachidonic acid, profoundly affects organ involvement in experimental animals with systemic immune complex diseases [22-241. Again, this mechanism can hardly explain the results obtained in the present study, since both the LAC diet and the placebo diet had a normal content of polyunsaturated fatty acids. An immune reaction to food antigens has long been claimed to explain the beneficial effects observed in isolated patients with certain diseases such as rheumatoid arthritis [25,26]. In this study, however, in no instance did we observe rechallenge phenomena, as could be expected in food allergy. In addition, IgE levels were constantly within normal limits in 16 unselected patients with MC in a previous study [27]. Finally, it seems unlikely that our patients had abnormal intestinal permeability to macromolecules, mediated by a partial IgA deficiency [28]. IgA levels were within normal limits in these patients; moreover, in a previous study by one of us, patients with MC showed a normal salivary IgA content [27]. On the other hand, the clinical improvement and the effects observed on cryoglobulins and the phagocytic system during the LAC diet could at least in part be explained by the model on the basis of the preliminary study [9]. If we consider functional activities such as plasma opsonization and phagocytosis of bloodborne material as one open biologic system [8,29], we can build a model of plasma clearance of particles and marcromolecules, not necessarily formal [30]. The
THERAPY
IN MIXED CRYOGLOBULINEMIA
(13 pts)
(12 pts)
RES
50
T;
40
min
3o
1
placebo-diet
LAC-diet 901
/ FERRI ET AL
\
-n.v
20 I
I
.
b
a
1
a
b p: ns.
pc ,005 I
J
Figure
4. Reticuloendothelial
after (a) the first three-week analysis was performed using mal values.
splenic function (RES) before (b) and period of dietetic treatment. Statistical the paired Student’s t-test. n.v. = nor-
model proposed describes an open system, the input to which is represented in part by cellular structures and organic proteins and in part by exogenous macromolecules. Under physiologic steady-state conditions, the system is sufficiently efficient to clear all input material from the plasma. During MC, the system shifts to a new steady state in which it is no longer able to clear the input (red blood cells, platelets, low-density lipoprotein, macromolecules of intestinal origin, and also cryoglobulins). In this case, any reduction in the input load (e.g., LAC diet) favors a functional rescue of the system and enhances its compensatory reserve. It is obvious that the proposed model will be verified or falsified by future studies [31]. In conclusion, the current study suggests a safe new therapeutic approach in the management of MC. Although the significant improvement of certain signs and symptoms is certainly encouraging, this shortterm trial cannot clarify how much such a therapeutic approach can modify the natural history of this chronic disease. We are therefore currently evaluating the effect of a long-term dietary regimen that alternates an LAC diet with a totally unrestricted diet.
REFERENCES 1. Meltzer M, Franklin EC, Elias K. McCluskey RT. Cooper N: Cryoglobulinemia. A clinical and laboratory study. II. Cryoglobulins with rheumatoid factor activity. Am J Med 1966; 40: 837-856. 2. Brouet JC. Clauvel JP. Danon F. Klein M, Seligman M: Biological and clinical significance of cryoglobulins. Am J Med 1974; 57: 775-786. 3. McIntosh RM. Griswold WR. Chernack WE. et at Cryoglobulins. Ill. Further studies on the nature, incidence. clinical, diagnostic, prognostic and immunopathotogic significance of cryoglobulins in renal disease. Q J Med 1975; 44: 285-307. 4. Maggiore Q. L’Abbate A. Bartolomeo F. et&The role of circulating cryoglobulins and HBsAg in the pathogenesis of glomerulonephritis of essential mixed cryoglobulinemia. In: Chiandussi L. Bartok E, Sherlock S. eds. Systemic effect of HBsAg immunocomplexes. Padova: Piccin. 1980; 242-251. 5. Ferri C. Moriconi L. Gremignai G, et al: Treatment of the renal involvement in mixed cryoglobulinemia with prolonged plasma exchange. Nephron 1986; 43: 246-253. 6. Ferri C. Gremignai G, Bombardieri S. et al: Plasma exchange in mixed cryoglobulinemia: the effect on renal. liver and neurologic involvement. Ric Clin Lab 1986; 16: 403-411. 7. Gorevic PD. Kassab HJ. Levo Y. et al: Mixed cryoglobulinemia: clinical aspects and long-term follow-up of 40 patients. Am J Med 1980; 69: 287-308. 8. Dioguardi N. ed: II fegato un sistema aperto. Milano: Masson ltalia Editori. 1984. 9. Pietrogrande M. Cefalo A. Nicora F, Marchesini D: Dietetic treatment of essential mixed cryoglobulinemia. Ric Clin Lab 1986: 16: 413-416. 10.Walker WA, Is&baker KJ: Uptake and transport of macromolecules by the
November
1989
The American
Journal
of Medicine
Volume
87
523
DIETETIC THERAPY
IN MIXED CRYOGLOBULINEMIA
/ FERRI ET AL
intestine: possible role in clinical disorders. Gastroenterology 1974: 67: 531-550. 11. Walker WA: Mechanisms of antigen handling by the gut. Clinics in Immunology and Allergy 1982: 2: 15-40. 12. Bock SA. Lee WJ. Remigio LK. May CD: Studies of hypersensitivity reactions to loods in infants and children. J Allergy Clin lmmunol 1978: 62: 327-331. 13. Zanussi C: Food allergy treatment. Clinics in Immunology and Allergy 1982: 2: 221-240. 14. Bombardieri S, Paoletti P. Ferri C. Di Munno 0. Fornai E. Giuntini C: Lung involvement in essential mixed cryoglobulinemia. Am J Med 1979; 66: 748-756. 15. Manta F. Migfiorini P. Bombardieri S. Celada F: An enzymatically active antigen-antibody probe to measure circulating immune complexes by competition, I. Use of Escherichia co/i&galactosidase in the probe and of bovine conglutinin as the complex-binding reagent. Clin lmmunol lmmunopathol 1980; 16: 131-141. 16. Kent GF. Fife EH: Precise standardization of reagents for complement fixation. Am J Trop Med Hyg 1963: 12: 103-109. 17. Bowing CS. Glass HI. Lewis SM: Rate of clearance by the spleen of heatdamaged erythrocytes. J Clin Pathol 1976; 29: 852-857. 18. Ziff M: Diet in the treatment of rheumatoid arthritis. Arthritis Rheum 1983; 26: 457-461. 19. Good FfA. Wet A, Fernandes G: Nutritional modulation of immune responses. Fed Proc 1980; 39: 3098-3104. 20. Park BH. Good RA: The influence of nutrition efficiency and aging. In: Park BH, Good RA. eds. Principles of modern immunobiology. Philadelphia: Lea and Febiger, 1974; 189-199.
524
November
1989
The American
Journal
of Medicine
Volume
21. Skoldstam L. Larsson L. Lindstrom FD: Effects of fasting and lactovegetarian diet on rheumatoid arthritis. Stand J Rheumatol 1979; 8: 249-255. 22. Diaz A, Pope R. Fischnach M. Talal N. FernandezG: Enhancement of circulating autoantibodies and immune complex levels in autoimmune prone mice by high dietary fat intake. Arthritis Rheum 1983; 26: 520-525. 23. Kelly VE. Ferretti A, lzui S. Strom JB: A fish oil diet rich in eicosapentanoic acid reduces cyclooxygenase metabolites. and suppresses lupus in MRL-Lpr mice. J lmmunol 1985; 134: 1914-1919. 24. Kremer JM. Jubiz W. Michalek A. et al: Fish-oil fatty acid supplementation in active rheumatoid arthritis. Ann Intern Med 1987: 106: 497-503. 25. Parke AL, Hughes GRW: Rheumatoid arthritis and food: a case study. Br Med J 1981: 282: 2027-2029. 26. Walport MJ, Parke AL. Hughes GRW: Food and connective tissue disease. Clinics in Immunology and Allergy 1982; 2: 113-120. 27. Pietrogrande M. personal communication. 28. Cuningham-Ruudless C. Brandeis WE, Good f?A, Day MK: Milk precipitating, circulating immune complexes and IgA deficiency. Proc Natl Acad Sci USA 1978: 75: 3387-3389. 29. Von Bertalanffy L. ed: The organism considered as physical system. In: General system theory. Harmondsworth, England: Penguin University Books, 1973: 127. 30. National Institutes of Health. Evaluation of models for biomedical research. In: US National Institutes of Health, ed. Models for biomedical research. Bethesda, Maryland, 1984. 31. Popper KR. ed: The logic of scientific discovery. New York: Basic Books, 1959.
87
PURPOSE: The effects of a low-antigen-content diet (LAC diet) versus a standard normocaloric diet on the signs and symptoms of mixed cryoglobulinemia (MC) were compared in a crossover randomized study. PATIENTSAND METHODS:The study consisted of 24 outpatients with MC, and was carried out in a 48-week period. After 18 weeks of either the LAC or the placebo diet, patients returned to a totally unrestricted diet for 12 w e e k s (washout period) and crossed over to the second half of the study for the other 18 weeks. RESULTS:After three weeks of the restricted LAC diet, the cryocrit decreased from 3.5 4- 3.4% (mean + SD) to 2.3 + 2.0% (p <0.01), and the circulating immune c o m p l e x levels decreased from 48 + 30% to 39 4- 34% (p <0.01). At the same time, the purpura score (p <0.05), glutamic pyruvic tran~aminase level (p <0.01), and g a m m a glutamyl transferase level (p <0.001) significantly improved. Splenic reticuloendothelial function, measured as the haftlife of heat-damaged autologous red cells, decreased from 41 + 21 minutes to 21 + 10 minutes (p <0.005). In contrast, no significant parallel clinical, biochemical, and immunologic changes occurred in
the same patients during the placebo (standard normocaloric) diet. CONCLUSION:These data show that an LAC diet decreases the amount o f circulating immune complexes in MC and can modify certain signs and symptoms of the disease. These effects of the LAC diet may be explained by postulating a functional restoration of the m o n o n u c l e a r phagocytic system.
ixed cryoglobulinemia (MC) is a systemic huM man disease characterized by the presence of large amounts of circulating cold precipitable complexes formed by monoclonal or polyclonal rheumatoid factor and polyclonal IgG [1,2]. Etiology of this condition is largely unknown, but current evidence indicates that most of the systemic organ involvement is mediated by the tissue deposition of circulating immune complexes [3,4]. Thus, the management of MC has mainly relied on reducing the load of immune complexes or their noxious effects, and generally includes the quantitative removal of circulating immune complexes in various ways (plasmapheresis and related technologies) [5,6], the use of cytotoxic agents, and corticosteroids [7]. In the present work, we have used a novel therapeutic approach, based on some theoretic assumptions [8] and on preliminary results of a study by one of us [9]. According to the proposed dynamic model, the circulating complexes in MC would compete for their catabolism with other inputs directed to the mononuclear phagocytic system. In normal conditions, such inputs include primarily blood-borne constituents and high-molecular-weight exogenous substances crossing the mucosal barrier of the gut [10,11] and reaching the systemic circulation via lymph and portal veins [12]. Thus, a dietary regimen capable of reducing this latter component could help to restore a saturated mononuclear phagocytic system, which in turn could be left free to remove the circulating immune complexes. We report the results of a randomized crossover study that compares the effects of a placebo "sham" diet and a dietary regimen of putative low-antigen content (LAC diet) on the signs and symptoms of MC. PATIENTS A N D METHODS Patients
From the Istituto di Patologia Medica I e Servizio di Reumatologia, Universitfi di Pisa, Pisa, Italy (CF, RC, AT, CV, SB); Centro Studi di Medicina Teoretica, Istituto di Medicina Interna, Universitfi di Milano, Milano, Italy (MP, AC); and Servizio di Dietologia, Ospedale di Niguarda, Milano, Italy (GB). Requestsfor reprints should be addressed to Clodoveo Ferri, M.D., Istituto di Patologia Medica I, Ospedale S. Chiara, Via Roma, 67, 56100 Pisa, Italy. Manuscript submitted January 20, 1989, and accepted in revised form July 20, 1989.
Of the 25 patients with MC who entered the study, 24 (19 women, five men) completed the trial. Patients' ages ranged from 42 to 68 years (mean, 56.3 years), and disease duration ranged from one to 15 years (mean, 6.4 years). Nine patients had type II cryoglobulin and 15 had type III cryoglobulin. Cryocrit varied from 0.3% to 21%. Seven of 24 patients had anti-hepatitis B virus (HBV) antibodies, and none had HBV-related antigens, either in the sera or in the cryoglobulins (Table I). Since no patient showed evidence of antigenemia but only signs of a previous HBV exposition (i.e., antibody to hepatitis B surface antigen), the putative role of HBV in these patients was not warranted and, therefore, they could be similarly admitted to the study. The diagnosis of MC, established in all cases at least 12 months before, was based on the presence of the typical clinical syndrome (purpura, arthralgias, weakness), mixed cryoglobulins with rheumatoid fac-
November 1989 The AmericanJournal of Medicine Volume87
519
DIETETIC THERAPY IN MIXED CRYOGLOBULINEMIA
TABLE I Epidemiologic Patient Number
and Immunologic
Data on 24 MC Patients before Diet Therapy
59F
fZ :i
42M 50M 44F 65F
CIC-K (%I
HBV Exposure
G-M-h’ G-M-A
iz
::: 5.6 1.4
G-M G-M-A G-M-A
ii i:
108 152 118 115
32 ii
: -
0.3
G-M-A
107
N”i 42
+
2.2
1.1 1.5 21 1.3 1.3 I’:!
11
44F 56F
RES (half-life minutes)
2:
:::
55F 43M 59F 66F
CHso
WJ/mU <50 105 154 <50 <50 127 <50 <50 124 126 129 126 <50 <50
:::
61M 54F 59F 51F
::
Cryopreclpitate Ig Comporltlon*
Cryocrlt (%I
Age/Sex
E 68F 51F 61F 67F 49F 61M 54F
/ FERRI ET AL
2:
;:[W:j-A
:::
G-M(k)-A G-M-A G-M
:7” ii ;;
115 141 174 <50 <50
:: :: ND
ii
Ii:, :
C-K = circulating immune complexes evaluated by means of bovine conglutinin as recognition unit (normal value < 40%); CHw = hemolytic complement iormal value 160-220 HU (hemolytic units]/mL); RES = reticuloendothelial spleen function expressed by (half-life minutes) (normal value < 30 minutes); tpatitis B virus exposure; ND = not done. IgA was constantly present in cryoprecipitate as trace amounts.
tor activity, and the absence of other known infectious, neoplastic, or systemic disorders [l]. Patients were consecutively admitted to the study when they had: (1) at least two of the typical symptoms (purpura, arthralgias, weakness),(2) a measurablecryocrit level, (3) involvement of at least one main organ (liver, renal, and/or neurologic involvement). In all patients, the absenceof other medical or nutritional problems wasascertained and informed consent was obtained. In the last 12 months before the study, 19 patients were treated with a low dosageof steroids (2 to 8 mg/day of B-methyl prednisolone), while five were untreated. In all cases,the prestudy corticosteroid dosage was maintained unchanged during the course of the trial. TABLE II LAC Diet Protocol*
Category
Basal Diet
Cereals Simple carbohydrates Fats Dairy products Meats, fish Em Vegetables Fruits Salt/spices Alcohol, beverage Additives/preservatives
I
Rice Sugar Olive oil None Turkey None Green, potatoes Lzrs, apples
Reintroductions II Ill IV
v
Ail All All
activity HBV =
Diets To prepare a diet with a reduced content of macromolecular antigens, we mostly relied on the current knowledge in food allergy [13]. The LAC diet wassimilar to the restricted hypoallergic diet, with somemodifications. Foods rich in putative macromolecular antigens, like meat, were further restricted; vegetables, fruits, and rice were allowed for the opposite reason. After three weeks of the restricted diet, the various classesof food were gradually reintroduced, according to a fixed protocol (Table II). The placebo diet was a standard dietetic regimen without restrictions on the variety of foods. However, in order for the patients to remain unaware of the identity of the diet they were receiving, close instructions on both regimens were given to them concerning the quantity of each food and how it had to be subdivided in the various meals. Both the placebo and the LAC diets were individually designedin order to be normocaloric (30 kcal/kg/day) , and equilibrated with respect to proteins (17% versus 16.5% of kcal), fats (32% versus 37%), carbohydrates (51% versus 46.5%), vitamins, mineral salts, and oligoelements.
t = clinico-immunologic evaluation reported in Figures 1. 2. and 3. l All diets are equilibrated in elementary components; only the basal diet is moderately hypocalcemic. All patients followed a normocaloric diet according to their ideal body weight.
Study Protocol The study was carried out in a 48-week period in outpatients with MC, randomly assignedto one of two groups. After the first 18 weeks of the LAC or placebo diet, patients returned to a totally unrestricted diet for 12 weeks (washout period) and crossed over to the second half of the study for the other 18 weeks. In order to ensure close adherence to the diet, all patients receiving either the placebo or the LAC diet were given an instruction booklet and a clinical diary for recording the subjective symptom variations and other parameters such as body weight, blood pressure, and drug ingestion. During the study, the patients on
520
87
Water None
Ail Yes All Ail All All Ail
Coffee, tea
Dme (weeks) P
I
November
1999
3
The American
6
Journal
g ‘f
of Medicine
15t
l8
Volume
DIETETIC
the LAC diet never met those on the placebo diet. A clinical and laboratory evaluation was carried out in all cases at the beginning and end of the study and at regular three-week intervals. At each visit, the following data were recorded by the same physician (C.F.) by means of the patient’s self-evaluation and the physical examination: presence and extent of the purpura (score 0 = absent; 1 = limited to lower legs; 2 = limited to entire legs; 3 = diffuse); presence and severity of arthralgias, weakness, peripheral neuropathy, Raynaud’s phenomenon (score 0 = absent; 1 = mild; 2 = moderate; 3 = severe); liver damage activity as measured by serum glutamic pyruvic transaminase (GPT) and gamma glutamyl transferase (GGT) levels; nephropathy by means of serum creatinine levels and proteinuria; arterial hypertension; body weight and drug therapy modifications. At the same three-week intervals, routine blood chemistry studies were carried out by standard methods. Immunologic Evaluation The following immunologic parameters were monitored: (1) cryocrit level was measured as percentage of packed cryoglobulins after cold centrifugation of the serum [14]. Briefly, serum samples,separated at 37”C, were left at 4°C for seven days and then centrifuged in a calibrated test tube at 1,200 X G for 15 minutes at 4°C. The composition of the cryoglobulins was determined by immunodiffusion on .Ouchterlony plates against specific antisera (Behring Institute, West Germany); (2) circulating immune complexes were measured by a competitive enzyme immunoassay using Escherichia coli beta-galactosidase in the probe with bovine conglutinin (CIC-K) as recognition unit [15]. Results were expressed as percent inhibition, and values above 40% were considered to be abnormal; (3) total hemolytic complement activity (CH& was detected by the method of Kent and Fife [16]. At the beginning of the study and after the first three-week period, reticuloendothelial splenic function was evaluated by the method of Bowring et al [17] in 13 patients on the LAC diet and 14 on the placebo diet. Briefly, autologous erythrocytes were heat-damaged at 49°C for 20 minutes, radiolabeled with technetium-99m, and reinjected in the same patient. The half-life was calculated from a disappearance curve, which was obtained by counting radioactivity in red blood cells at different times after the injection. Half-life values above 30 minutes were considered abnormal. Statistical Analysis Statistical analysis of the results was performed by computer using the following: Student’s t-test for paired and unpaired data, chi-square method with Yates’ correction, Wilcoxon non-parametric test. In addition, analysis of variance testing for assessmentof the variance within each group was performed.
RESULTS Table III summarizesthe main clinical features and organ involvement of the 24 patients who completed the study. On the whole, there was a good acceptance of the restricted dietary regimen, although it implied avoiding popular foods like bread and pasta. Results of routine blood chemistry studies remained practically unaffected during the study.
THERAPY
IN MIXED
CRYDGLDBULINEMIA
/ FERRI ET AL
TABLEIII Clinical Dataon 24 MCPatientsTreated with LACDiet Symptoms
Number
Purpura’ Weakness Arthralgias Raynaud’s phenomenon Neuropathy Live;Fvolvementf
13124 lb/24
t GGT Histology-CPH CAH Renal involvement tSerum creatinine and/or proteinuria histology (MPGN) Arterial hypertension*
i;:: 6/24 a/24 4124 L/24 2124 2/24 7124
%
l:;;: ::;z
I
GPT = glutamic pyruvic transaminase; GGT = glutamyl transferase; CPH = chronic persistent hepatitis; CAH = chronic active hepatitis: MPGN = membranoproliferative glomerulonephritis. l During the patient follow-up, purpura was present in all cases. + On the basis of histologic lesions or persistent hepatic enzyme (GPT and GGT) elevation. + Diastolic blood pressure greater than or equal to 90 mm Hg
purpura . :.:.:.:.:.:.:.:.:..:’ . . . . . . . . . . . ..: . .
~~~
:.:.:.:.:.:.:.:.:.:.:. .‘.‘.‘.‘.‘.‘.~.~.‘.‘.’ .‘.~.~.~.~.~.~.~.~.~.~
I L&Z-diet
**
I
I
0
i LI placebo diet
I
I
12
16wks
:iiiii restricted
diet period J
rlgure J. rurpura score (mean f SEM) variations during LAC and placebo diets. Statistical analysis was performed using the Wilcoxon non-parametric test. l p
Clinical Variations During the LAC diet, an improvement of the clinical conditions was observed both subjectively and objectively. The patients claimed an improvement in purpura, weakness,fatigabihty, and arthralgias; to a lesser extent, however, this improvement wasalso present during the placebo diet. Of the 24 patients studied, fresh purpura, weakness,and arthralgias were present in 16,17, and 16 subjects at the beginning of the LAC diet, and in 10, 19, and 16, respectively, before the placebo diet. During the treatment, purpura improved or disappeared in 10 of 16 (63%) subjects versus two of 10 (20%) (LAC diet versus placebo diet), weakness improved or disappeared in eight of 17 (47%) versus three of 19 patients (16%), and arthralgias improved or resolved in four of 16 (25%) versus three of 16 patients (19%). None of these differences were significant (Fisher’s exact test). In contrast, the score for purpura significantly decreased during the LAC diet (Figure l), and these changes were particularly evident after the first six to 12 weeks of treatment. In addition,
November
1989
The American
Journal
of Medicine
Volume
87
521
DIETETIC THERAPY
IN MIXED CRYOGLOBULINEMIA
1::::::;:: e::::: :;:::: ;..:_,.:. :.:.::.::. .:::::.,:: :.:::,:: :...::::. .:._::. :::i:::;: _::::::: _.:.::. .:. :.: .::_.:.: :: :;:.: .:::.:: ** *** ...*t+
/ FERRI ET AL
GGT
GPT
3
: base1 . LAC-diet
3
12
o placebo diet
basal
IEwks jji:j
restricteddiet
l
LAC-diet
12
0 placebo diet
IBwks :.:: : :::_:
restricted
-I
-r
basal
3
Immunologic Parameter Variations Among the immunologic parameters, cryocrit showeda clear-cut decreasein all patients on the LAC diet, and was more pronounced after the first three weeks (Figure 3); this reduction persisted during the subsequent weekswhen the reintroductions were carried out. In all patients, a more transient reduction was observed in CIC-K levels only at the end of the initial three weeksof the restricted regimen. No com-
November
1989
The American
Journal
12
diet period
purpura reappeared or worsened in seven of 24 patients during the placebo diet, but never during the experimental diet. During the LAC diet, biochemical evidence of liver damage was reduced. The improvement was particularly evident after the first three weeks of the restricted diet and persisted for GGT during the subsequent reintroductions (Figure 2). With regard to the other clinical symptoms, no variations were observed in Raynaud’s phenomenon, peripheral neuropathy, arterial hypertension, or nephropathy. Patients with or without associatedsteroid treatment showed similar changes in the symptoms and signs of the disease.
522
1Bwks
CIC-K
60
3
12
period
CRYOCRIT
beaal
3
Figure 2. Liver enzyme variations (GPT = glutamic pyruvic transaminase; GGT = gamma glutamyl transferase) during LAC and placebo diets. 1, 2 = multiple of normal values (mean f SEM). Statistical analysis was performed using the Wilcoxon non-parametric test. **p<0.01, ***p<0.001.
of Medicine
Volume
ISwks
Figure 3. Cryocrit and circulating immune complex (CIC-K) variations during LAC and placebo diets (mean f SEM). Statistical analysis was performed using the paired Student’s t-test. h.v. = normal values. l p
parable changeswere observed in the circulating immune complex moiety during the placebo diet (Figure 3). CHssremained unchanged during both dietary regimens. The clearance of heat-damaged erythrocytes was prolonged (half-life greater than 30 minutes) in nine of 12 patients before the LAC diet and in eight of 13 before the placebo diet. In the first group, the halflife significantly decreased in 10 of 12 patients (p <0.005), and was within the normal range in eight of 12patients at the end of the intensive dietary regimen. Such consistent consensualresults were not present in the placebo diet group (Figure 4). Analysis of variance testing confirmed the significant reduction of cryocrit as well as GGT, GPT, and purpura score in the patients treated with the LAC diet.
COMMENTS The current study demonstrates that the LAC diet exerts positive effects on somesignsand symptoms of MC. Among the clinical features, the most remarkable improvement was observed in purpura, and in the biochemical markers of liver damage. Among the immu87
DIETETIC
nologic parameters, cryoglobulins and, to a lesser extent, circulating immune complexes were significantly reduced, and reticuloendothelial splenic function was improved. It is possible to exclude that these positive results could be ascribed to factors other than diet alone. First of all, some patients were untreated during the study; in the others, the low glucocorticoid dosage was not modified. Second, since the study was carried out in outpatients, the unpredictable effect of bed rest can be excluded. This is still a controversial aspect of the studies involving the use of diet in the management of connective tissue diseases [18]. Hospitalization would in fact favor a closer adherence to the dietary regimen, but bed rest could at the same time profoundly affect certain symptoms such as purpura or arthralgias. Finally, to completely avoid a placebo effect, the patients were told that they were testing the effects of two different therapeutic diets. Above ah, however, clinical, biochemical, and immunologic improvement occurred only during the LAC diet and not during the placebo diet, and symptoms and signs of MC slowly recurred with the progressive return to a totally unrestricted diet. The attempt to manipulate the immune system with various dietary regimens is not new, and several hypotheses were proposed to explain the ensuing results [18,19]. Most of these mechanisms can be reasonably excluded on the basis of the available data. It is known, for instance, that starvation might modify the immune system in experimental conditions [20] or the time course of immune-mediated diseases such as rheumatoid arthritis in humans [21]. Although there was a slight decrease of body weight during the LAC diet in some of our patients, this is certainly not the case. In the last 10 years it has been reported that a diet deprived of certain polyunsaturated fatty acids, known precursors of arachidonic acid, profoundly affects organ involvement in experimental animals with systemic immune complex diseases [22-241. Again, this mechanism can hardly explain the results obtained in the present study, since both the LAC diet and the placebo diet had a normal content of polyunsaturated fatty acids. An immune reaction to food antigens has long been claimed to explain the beneficial effects observed in isolated patients with certain diseases such as rheumatoid arthritis [25,26]. In this study, however, in no instance did we observe rechallenge phenomena, as could be expected in food allergy. In addition, IgE levels were constantly within normal limits in 16 unselected patients with MC in a previous study [27]. Finally, it seems unlikely that our patients had abnormal intestinal permeability to macromolecules, mediated by a partial IgA deficiency [28]. IgA levels were within normal limits in these patients; moreover, in a previous study by one of us, patients with MC showed a normal salivary IgA content [27]. On the other hand, the clinical improvement and the effects observed on cryoglobulins and the phagocytic system during the LAC diet could at least in part be explained by the model on the basis of the preliminary study [9]. If we consider functional activities such as plasma opsonization and phagocytosis of bloodborne material as one open biologic system [8,29], we can build a model of plasma clearance of particles and marcromolecules, not necessarily formal [30]. The
THERAPY
IN MIXED CRYOGLOBULINEMIA
(13 pts)
(12 pts)
RES
50
T;
40
min
3o
1
placebo-diet
LAC-diet 901
/ FERRI ET AL
\
-n.v
20 I
I
.
b
a
1
a
b p: ns.
pc ,005 I
J
Figure
4. Reticuloendothelial
after (a) the first three-week analysis was performed using mal values.
splenic function (RES) before (b) and period of dietetic treatment. Statistical the paired Student’s t-test. n.v. = nor-
model proposed describes an open system, the input to which is represented in part by cellular structures and organic proteins and in part by exogenous macromolecules. Under physiologic steady-state conditions, the system is sufficiently efficient to clear all input material from the plasma. During MC, the system shifts to a new steady state in which it is no longer able to clear the input (red blood cells, platelets, low-density lipoprotein, macromolecules of intestinal origin, and also cryoglobulins). In this case, any reduction in the input load (e.g., LAC diet) favors a functional rescue of the system and enhances its compensatory reserve. It is obvious that the proposed model will be verified or falsified by future studies [31]. In conclusion, the current study suggests a safe new therapeutic approach in the management of MC. Although the significant improvement of certain signs and symptoms is certainly encouraging, this shortterm trial cannot clarify how much such a therapeutic approach can modify the natural history of this chronic disease. We are therefore currently evaluating the effect of a long-term dietary regimen that alternates an LAC diet with a totally unrestricted diet.
REFERENCES 1. Meltzer M, Franklin EC, Elias K. McCluskey RT. Cooper N: Cryoglobulinemia. A clinical and laboratory study. II. Cryoglobulins with rheumatoid factor activity. Am J Med 1966; 40: 837-856. 2. Brouet JC. Clauvel JP. Danon F. Klein M, Seligman M: Biological and clinical significance of cryoglobulins. Am J Med 1974; 57: 775-786. 3. McIntosh RM. Griswold WR. Chernack WE. et at Cryoglobulins. Ill. Further studies on the nature, incidence. clinical, diagnostic, prognostic and immunopathotogic significance of cryoglobulins in renal disease. Q J Med 1975; 44: 285-307. 4. Maggiore Q. L’Abbate A. Bartolomeo F. et&The role of circulating cryoglobulins and HBsAg in the pathogenesis of glomerulonephritis of essential mixed cryoglobulinemia. In: Chiandussi L. Bartok E, Sherlock S. eds. Systemic effect of HBsAg immunocomplexes. Padova: Piccin. 1980; 242-251. 5. Ferri C. Moriconi L. Gremignai G, et al: Treatment of the renal involvement in mixed cryoglobulinemia with prolonged plasma exchange. Nephron 1986; 43: 246-253. 6. Ferri C. Gremignai G, Bombardieri S. et al: Plasma exchange in mixed cryoglobulinemia: the effect on renal. liver and neurologic involvement. Ric Clin Lab 1986; 16: 403-411. 7. Gorevic PD. Kassab HJ. Levo Y. et al: Mixed cryoglobulinemia: clinical aspects and long-term follow-up of 40 patients. Am J Med 1980; 69: 287-308. 8. Dioguardi N. ed: II fegato un sistema aperto. Milano: Masson ltalia Editori. 1984. 9. Pietrogrande M. Cefalo A. Nicora F, Marchesini D: Dietetic treatment of essential mixed cryoglobulinemia. Ric Clin Lab 1986: 16: 413-416. 10.Walker WA, Is&baker KJ: Uptake and transport of macromolecules by the
November
1989
The American
Journal
of Medicine
Volume
87
523
DIETETIC THERAPY
IN MIXED CRYOGLOBULINEMIA
/ FERRI ET AL
intestine: possible role in clinical disorders. Gastroenterology 1974: 67: 531-550. 11. Walker WA: Mechanisms of antigen handling by the gut. Clinics in Immunology and Allergy 1982: 2: 15-40. 12. Bock SA. Lee WJ. Remigio LK. May CD: Studies of hypersensitivity reactions to loods in infants and children. J Allergy Clin lmmunol 1978: 62: 327-331. 13. Zanussi C: Food allergy treatment. Clinics in Immunology and Allergy 1982: 2: 221-240. 14. Bombardieri S, Paoletti P. Ferri C. Di Munno 0. Fornai E. Giuntini C: Lung involvement in essential mixed cryoglobulinemia. Am J Med 1979; 66: 748-756. 15. Manta F. Migfiorini P. Bombardieri S. Celada F: An enzymatically active antigen-antibody probe to measure circulating immune complexes by competition, I. Use of Escherichia co/i&galactosidase in the probe and of bovine conglutinin as the complex-binding reagent. Clin lmmunol lmmunopathol 1980; 16: 131-141. 16. Kent GF. Fife EH: Precise standardization of reagents for complement fixation. Am J Trop Med Hyg 1963: 12: 103-109. 17. Bowing CS. Glass HI. Lewis SM: Rate of clearance by the spleen of heatdamaged erythrocytes. J Clin Pathol 1976; 29: 852-857. 18. Ziff M: Diet in the treatment of rheumatoid arthritis. Arthritis Rheum 1983; 26: 457-461. 19. Good FfA. Wet A, Fernandes G: Nutritional modulation of immune responses. Fed Proc 1980; 39: 3098-3104. 20. Park BH. Good RA: The influence of nutrition efficiency and aging. In: Park BH, Good RA. eds. Principles of modern immunobiology. Philadelphia: Lea and Febiger, 1974; 189-199.
524
November
1989
The American
Journal
of Medicine
Volume
21. Skoldstam L. Larsson L. Lindstrom FD: Effects of fasting and lactovegetarian diet on rheumatoid arthritis. Stand J Rheumatol 1979; 8: 249-255. 22. Diaz A, Pope R. Fischnach M. Talal N. FernandezG: Enhancement of circulating autoantibodies and immune complex levels in autoimmune prone mice by high dietary fat intake. Arthritis Rheum 1983; 26: 520-525. 23. Kelly VE. Ferretti A, lzui S. Strom JB: A fish oil diet rich in eicosapentanoic acid reduces cyclooxygenase metabolites. and suppresses lupus in MRL-Lpr mice. J lmmunol 1985; 134: 1914-1919. 24. Kremer JM. Jubiz W. Michalek A. et al: Fish-oil fatty acid supplementation in active rheumatoid arthritis. Ann Intern Med 1987: 106: 497-503. 25. Parke AL, Hughes GRW: Rheumatoid arthritis and food: a case study. Br Med J 1981: 282: 2027-2029. 26. Walport MJ, Parke AL. Hughes GRW: Food and connective tissue disease. Clinics in Immunology and Allergy 1982; 2: 113-120. 27. Pietrogrande M. personal communication. 28. Cuningham-Ruudless C. Brandeis WE, Good f?A, Day MK: Milk precipitating, circulating immune complexes and IgA deficiency. Proc Natl Acad Sci USA 1978: 75: 3387-3389. 29. Von Bertalanffy L. ed: The organism considered as physical system. In: General system theory. Harmondsworth, England: Penguin University Books, 1973: 127. 30. National Institutes of Health. Evaluation of models for biomedical research. In: US National Institutes of Health, ed. Models for biomedical research. Bethesda, Maryland, 1984. 31. Popper KR. ed: The logic of scientific discovery. New York: Basic Books, 1959.
87