Low Fasting Serum Triglyceride Level as a Precocious Marker of Autoimmune Disorders

Silvia Iannello, MD, Antonina Cavaleri, MD, Paolina Milazzo, MD, Santi Cantarella, MD, Francesco Belfiore, MD

Medscape General Medicine. 2003;5(3) 

Abstract and Introduction

The authors recently reported the occurrence of low fasting serum triglyceride (TG) and high free fatty acid (FFA) levels in idiopathic pulmonary fibrosis. TG estimation in diverse groups of patients with autoimmune disease or hyperactive immune response confirmed the occurrence of a similar decrease of TG. In some patients, serum FFA level was also evaluated. TG value in lean and obese patients was compared with that in lean (n = 108) and obese (n = 208) control subjects without autoimmune disease. In patients affected by autoimmune chronic thyroiditis with enhanced concentration of antithyroglobulin antibodies and without thyroidal failure (n = 24), lean and obese patients had reduced TG (-69/%, P < .01 and -52%, P < .0001, respectively). Both lean and obese patients affected by chronic active B or C hepatitis (n = 26), with autoantibodies and without signs of hepatic insufficiency or cirrhosis, presented reduced TG (-57%, P < .01 and -61%, P < .001, respectively). A marked TG decrease (-73%, P < .001) was observed in the lean patients affected by lupus-like syndrome (n = 7). The lean and obese patients with systemic lupus erythematosus or rheumatoid arthritis (n = 11) showed TG decrease (-66%, P < .01 and -55%, P < .05, respectively). In patients affected by anamnestic allergy or atopic dermatitis/asthma (n = 66), both lean and obese, TGs were reduced (-67%, P < .0001 and -62%, P < .001, respectively). In isolated cases of diverse autoimmune diseases (scleroderma, APECED [autoimmune polyendocrinopathy, candidiasis, and ectodermal dystrophy], urticaria or urticarial vasculitis, Reiter or Sjogren syndromes, ulcerative colitis or Crohn's disease, multiple sclerosis or Guillain-Barré syndrome) (n = 14), decreased TG was also observed both in the lean and obese subjects (-59%, P < .01 and -57%, P < .01, respectively). Concerning FFA (n = 69), value in lean patients (n = 22) vs that in lean controls (n = 18) was increased (520 ± 31 vs 299 ± 30 mcEq/L, +74%, P < .001), whereas value in obese patients (n = 18) vs that in obese control subjects (n = 11) was decreased (542 ± 34 vs 774 ± 62, -30%, P < .01). This opposite behavior of FFA in lean and obese patients needs to be confirmed. Data in this study seem to indicate that low TG value may be a precocious marker of autoimmunity or immune system hyperreactivity.

The occurrence of low fasting serum TG and high FFA levels was recently reported in chronic interstitial or fibrosing pulmonary disease.[1] Forty-four patients with fibrosing pulmonary disease, compared with 110 control subjects, showed a 61% reduction of TG (P < .001) and a 63% increase of serum FFA (P < .01).[1] In the presence of pulmonary fibrosis, TG levels were also low in patients with type 2 diabetes (-53%, P < .001) and in obese subjects (-69%, P < .01),[1] despite the known fact that TGs are often elevated in diabetes and obesity. The mechanisms of this change remain to be clarified.[1]

This datum prompted the authors to systematically estimate TG value in all patients with several autoimmune diseases or hyperactive immune response. In some patients, serum FFA level was also evaluated. The results of these observations are reported here.

Subjects and Methods

Since it is well known that TG levels in serum vary to a great extent for interindividual variability, the study included numerous control subjects (a total of 337 individuals). This group of control subjects was subdivided into 2 subgroups because there is the need to compare lean patients with lean control subjects and obese patients with obese control subjects, since it is well known that the values of TG and FFA are different in lean and obese subjects. The 2 groups considered were:

(a) A group of lean control subjects (body mass index [BMI] < 27 kg/m2), which includes all the 128 consecutive patients hospitalized during 2000 for minor illnesses without autoimmune diseases; and

(b) A group of obese control subjects (BMI > 27 kg/m2), which includes 209 patients with overweight or frank obesity, but without autoimmune disorder or other illnesses, received during 2000-2002 on ambulatory or day hospital for metabolic evaluation and dietary treatment.

Characteristics of these subjects are presented in . Moreover, 11 lean and 41 obese control subjects with diabetes were studied in order to have a suitable control groups for the few patients with autoimmune disorder who were also diabetic. The parameters of the lean and obese control subjects were used for comparisons with the lean and obese patients with autoimmune disorder, respectively.

  Table 1. TG, FFA, and Other Parameters in Control Subjects Studied (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean subjects 128 20/108 40 ± 4 25.13 ± 0.34 1.60 ±0.09 299 ± 30 (n = 18)
Obese subjects 208 86/122 43 ± 1 34.16 ± 0.36 1.81 ± 0.07 774 ± 62 (n = 11)
P value (student's t test)     NS < .001 < .05 < .001
P value (student's t test in LOG-transf. data)     NS < .001 < .05 < .001

The study includes diverse groups of patients affected by several autoimmune disorders or diseases and by hyperactive immune response, as described as follows:

(a) Twenty-four female patients affected by chronic autoimmune thyroiditis subdivided into 2 subgroups, one composed of 5 lean subjects (BMI < 27 kg/m2) and the other of 19 overweight or obese subjects (BMI > 27 kg/m2). These patients presented nodularity of thyroid with euthyroid state, slight increase in serum thyroid-stimulating hormone (TSH), and enhanced concentration of antithyroglobulin antibodies, mainly antithyroid peroxidase (anti-TPO or antimicrosomal) antibodies. In some cases, histologic confirmation of diagnosis was obtained by needle biopsy. The characteristics of these subjects are shown in .

  Table 2. TG, FFA, and Other Parameters in Patients With Chronic Autoimmune Thyroiditis (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 5 0/5 49 ± 6 23.89 ± 1.12 0.49 ± 0.10 518 ± 78
P value (student's t test)     NS NS < .01 < .01
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 5)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 19 2/17 43 ± 3 33.83 ± 1.58 0.87 ± 0.08 564 ± 77
P value (student's t test)     NS NS < .001 NS
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 5)

(b) Thirty patients affected by chronic thyreopathies without autoantibodies, prevailingly patients with diffuse or multinodular nontoxic goiter or with a slight increased concentration of TSH (without thyroid failure) and absence of circulating antimicrosomal or antithyroglobulin antibodies. In several cases, histologic confirmation of diagnosis was obtained by needle biopsy. These patients were subdivided into 2 subgroups, one composed of 9 lean subjects and the other of 21 overweight or obese subjects. The characteristics of these subjects are shown in .

  Table 3. TG, FFA, and Other Parameters in Patients With Chronic Thyreopathies Without Autoantibodies (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 9 1/8 50 ± 6 23.68 ± 0.78 1.15 ± 0.12 418 ± 33
P value (student's t test)     NS NS NS NS
P value (student's t test in LOG-transf. data)     NS NS NS NS (n = 3)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 21 0/21 48 ± 2 34.39 ± 0.90 1.34 ± 0.10 560 ± 47
P value (student's t test)     NS NS < .05 < .05
P value (student's t test in LOG-transf. data)     NS NS < .05 < .05 (n = 5)

(c) Twenty-six patients affected by chronic active B or C hepatitis who complained of fatigue, malaise, myalgias, hypergammaglobulinemia, and positive markers (and in some cases autoantibody positive titer), but who were free from signs of hepatic insufficiency and hypoalbuminemia, portal hypertension, jaundice with hyperbilirubinemia, and cirrhosis. These patients were subdivided into 2 subgroups, one composed of 7 lean subjects and the other of 19 overweight or obese subjects. The characteristics of these subjects are shown in .

  Table 4. TG, FFA, and Other Parameters in Patients With Chronic Active B or C Hepatitis (mean ± SEM)


Patient initials Gender M/F Age (years) Virus type BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean Patients (BMI < 27 kg/m2)
  P.I.* F 17 B 26.36 0.57  
  P.M.G.* F 24 B 22.73 0.36  
  M.G.† M 42 C 24.60 0.96  
  C.G.‡ M 59 C 27.00 1.03  
  G.G. F 60 C 24.42 0.57  
  T.F.† F 60 C 26.98 0.50  
  D.M.F.† M 66 C 25.00 0.75  
Mean ± SEM n = 7 3M/4F 47 ± 7   25.30 ± 0.59 0.68 ± 0.09 ---
P value (student's t test)     NS   NS < .01  
P value (student's t test in LOG-transf. data)     NS   NS < .001  
Obese Patients (BMI > 27 kg/m2)
  P.C.* F 21 B 29.77 0.26 385
  F.F. F 26 B 31.2 0.44  
  C.A. M 27 B 31.68 0.41 393
  L.A. F 33 C 33.44 0.55  
  P.D. F 45 C 29.34 0.69  
  R.L. F 48 B 38.12 0.59 360
  P.A.* M 49 B 30.86 0.74 382
  M.C.‡ F 50 C 36.96 0.55  
  M.C.† M 52 C 28.80 1.04  
  D.L.F. F 53 C 54.22 0.73  
  M.S.† M 54 C 27.50 0.95  
  C.B.† F 58 C 30.50 0.98  
  G.A.† F 62 C 28.60 0.95  
  C.R. F 63 C 32.44 0.96  
  L.M.† M 67 C 28.37 0.79  
  S.A.† F 70 B,C 31.62 0.53  
  D.P. F 71 C 38.48 0.57  
  L.M.V.† F 72 C 27.20 1.01  
  M.S.† F 75 C 29.00 1.02  
Mean ± SEM n = 19 5M/14F 52 ± 4   32.53 ± 1.43 0.71 ± 0.05 380 ± 7
P value (student's t test)     < .05   NS < .001 < .01
P value (student's t test in LOG-transf. data)     < .01   NS < .001 < .001

(d) Seven prevailingly lean patients with lupus-like syndrome (ie, affected by rheumatologic disorders; chronic fatigue; arthralgias with pain, swelling, and stiffness of the fingers; or mixed connective tissue disease with overlapping features, different from the classical rheumatoid arthritis and that did not fulfill the American Rheumatism Association criteria for systemic lupus erythematosus (SLE).[2] The characteristics of these subjects are shown in .

  Table 5. TG, FFA, and Other Parameters in Patients With Lupus-like Syndrome, SLE, and Rheumatoid Arthritis (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lupus-like Syndrome
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 7 1/6 32 ± 2 25.22 ± 2.08 0.43 ± 0.05 574 ± 81
P value (student's t test)     NS NS < .001 < .01
P value (student's t test in LOG-transf. data)     NS NS < .0001 < .01 (n = 4)
SLE and Rheumatoid Arthritis
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 6 2/4 40 ± 7 22.78 ± 2.18 0.55 ± 0.05 598 ± 58
P value (student's t test)     NS NS < .01 < .01
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 3)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 5 2/3 51 ± 3 35.50 ± 2.18 0.81 ± 0.06 ---
P value (student's t test)     NS NS < .05  
P value (student's t test in LOG-transf. data)     NS NS < .01  

(e) One patient affected by SLE that fulfilled the American Rheumatism Association criteria for SLE[2] and 10 patients affected by rheumatoid arthritis, with a chronic polyarthritis involving peripheral joint with symmetric distribution and deformities. These patients presented increased erythrocyte sedimentation rate, normocytic anemia, positivity for serum Waaler-Rose reaction, and increased rheumatoid factor. They were subdivided into 2 subgroups, one composed of 6 lean subjects and the other of 5 overweight or obese subjects. In the patient L.L., who was a dental technician exposed to ceramic silica dust, the disease was associated with lung interstitial disorder.[3] The HLA typing of this patient showed the HLA-A2-A31, HLA-B18-B51, and HLA-DR3-DR11. Some of these haplotypes are associated or correlated with susceptibility to diverse autoimmune diseases, including rheumatoid arthritis.[3] The characteristics of these subjects are shown in .

  Table 6. TG, FFA, and Other Parameters in Patients With Anamnestic Allergy or Atopic Dermatitis/Asthma (mean +/= SEM)


N

Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 23 1/22 30 ± 3 23.13 ± 0.61 0.53 ± 0.04 455 ± 41
P value (student's t test)     < .05 NS < .001 < .05
P value (student's t test in LOG-transf. data)     NS NS < .0001 < .05 (n = 4)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 43 7/36 35 ± 2 31.76 ± 0.62 0.68 ± 0.03 654 ± 45
P value (student's t test)     < .001 NS < .001 NS
P value (student's t test in LOG-transf. data)     < .001 NS < .0001 NS (n = 3)

(f) Sixty-six patients with hyperactive immune response disclosed by anamnestic allergy or atopic dermatitis/asthma. These patients were subdivided into 2 subgroups, one composed of 23 lean subjects and the other of 43 overweight or obese subjects. The characteristics of these patients are shown in .

  Table 7. TG, FFA, and Other Parameters in Patients With Some Diverse Autoimmune Diseases (mean +/= SEM)


Patient initials Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Scleroderma L.E. F 50 22.22 0.41 800
APECED syndrome P.S. F 42 13.67 0.77 441
Urticaria or urticarial vasculitis B.B. F 28 38.95 1.09  
  C.G. F 47 49.03 0.59  
  D.G.M. F 54 20.51 0.27  
  S.G. F 61 26.01 0.92 460
Reiter syndrome A.G. M 31 23.50 0.47  
Sjogren syndrome C.G.* F 72 17.09 0.68  
Ulcerative proctocolitis L.P.† F 21 33.80 0.64  
  L.M.† F 28 28.50 0.58  
  P.G. F 52 25.54 0.67  
Crohn's disease R.F.‡ F 36 25.57 0.72  
  R.G.‡ M 44 24.55 0.91  
Multiple sclerosis C.G. F 46 27.33 0.26  
Guillain-Barré syndrome S.C.* F 75 33.20 1.48  
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 9 2M/7F 49 ± 4 22.97 ± 1.42 0.65 ± 0.07 567 ± 117
P value (student's t test)     NS NS < .01 < .01
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 3)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 6 6F 41 ± 8 35.14 ± 3.26 0.77 ± 0.18 ---
P value (student's t test)     NS NS < .01  
P value (student's t test in LOG-transf. data)     NS NS < .001  

(g) A group of patients with various autoimmune disorders was also investigated, as detailed below. One female patient was affected by scleroderma with features of CREST syndrome (calcinosis, Raynaud's phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasia). This patient presented with diffuse cutaneous thickening of proximal and distal extremities without digital ulcers, anemia, hypergammaglobulinemia, and antinuclear antibodies. She was treated with iloprost (a prostacyclin analogue).

One female patient was affected by APECED or type I polyglandular autoimmune syndrome (an autosomal recessive disorder), which had appeared in childhood and was characterized by hypoevolute soma and genitalia, adrenal hypofunction, vitiligo, genital candidiasis and dental/nail dystrophy.

Four female patients were affected by urticaria or urticarial vasculitis (induced by a variety of factors), who showed systemic involvement, prevailingly characterized by recurrent urticarial painful lesions, dyspnea and cough, angioedema of the tongue and soft palate, malaise, arthralgias, and abdominal pain.

One male patient had Reiter syndrome, characterized by fatigue, malaise, conjunctivitis, mucocutaneous lesions (mainly oral ulcers), spondyloarthropathy, and urethritis complicated by chlamydial infection.

One aged female patient was affected by Sjogren syndrome, characterized by fatigue, malaise, xerostomia, dry eyes, arthralgias, renal failure, increased rheumatoid factor titer, and elevated erythrocyte sedimentation rate. The patient was also affected by severe insulin-dependent diabetes.

Three patients were affected by ulcerative proctocolitis and 2 by Crohn's disease, in whom the syndromes were accompanied by arthritic disorder.

One adult female patient was affected by chronic multiple sclerosis with vertigo, visual blurring, diplopia, spasticity with ataxia, and severe disability.

One aged female patient was affected by severe Guillain-Barré syndrome, characterized by rapidly evolving flaccid quadriplegia and lumbosacral pain, developed 15 days after an acute febrile respiratory disease. The recovery was poor despite high-dose intravenous immunoglobulin therapy for 5 days associated with methylprednisolone treatment. She was obese, hypertensive, diabetic (treated with insulin), and also affected by chronic autoimmune thyroiditis (confirmed by needle-biopsy).

Most of the patients affected by autoimmune diseases were female subjects, as was expected. The control subjects and the patients followed their habitual diet (about 2000-2200 Kcal) with a balanced content of carbohydrates (55% to 60%), fats (15% to 25%, prevailingly olive oil) and proteins (15% to 25%). Dietary regimens were comparable between control subjects and patients. None of the patients evaluated (apart from the diabetic patient with Sjogren syndrome) showed clinical or laboratory evidences of liver or kidney insufficiency, nor had history of glucocorticoid or levothyroxine treatment, nor alcohol abuse or oral contraceptive therapy. None was treated with cytotoxics, or with drugs influencing the nutritional state, or with lipid or blood pressure lowering agents, or with other drugs capable of affecting TG or FFA levels.

Serum TG level was measured in the morning, after 12 hours of fasting. In 18 lean and 11 obese control subjects as well as in 22 lean and 18 obese patients without autoimmune disease, fasting serum FFA level was also measured in the morning, after 12 hours of fasting (Tables 1-7). Since it is known that TG level serum vary to a great extent for intraindividual variability, in many patients TG assays were repeated after 2-3 months.

The assay method used for serum TG quantitative determination was the standard enzymatic-colorimetric method, in that TGs were hydrolyzed to glycerol and FFA through the action of lipase, and the reaction was coupled with 3 enzymatic phases catalyzed by glycerol kinase, glycerol oxidase, and peroxidase. The assay included a reagent blank, and the measurement of absorbance was made at Hg 520 nm, at 37° C. The sample material was processed without storage

The method "NEFA C" (by Wako Chemicals GmbH, Neuss, D) was used for serum FFA determination. This is a colorimetric-enzymatic analysis kit by the use of acyl-CoA synthetase, acyl-CoA oxidase, and peroxidase. The assay included a reagent blank, and the measurement of absorbance was made at Hg 550 nm, at 37° C. Serum was separated immediately, and the samples (keep frozen) were processed as soon as possible.

Data were statistically analyzed by means of the student's t test for unpaired data. Considering that data might not show a normal distribution, statistical tests were also calculated after LOG-transformation of data. A value of P < .05 was considered as statistically significant. Data were expressed as mean ± standard error of the mean.

Results

Data obtained in lean and obese control subjects are shown in . The changes between these lean and obese subjects are statistically significant for BMI, TG, and FFA. Concerning the diabetic control groups, mean value of TG was 2.13 ± 0.52 and 2.20 ± 0.17 mmol/L in lean and obese diabetic subjects, respectively. Results obtained in patients as compared with control subjects are shown in Tables 1-7.

  Table 1. TG, FFA, and Other Parameters in Control Subjects Studied (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean subjects 128 20/108 40 ± 4 25.13 ± 0.34 1.60 ±0.09 299 ± 30 (n = 18)
Obese subjects 208 86/122 43 ± 1 34.16 ± 0.36 1.81 ± 0.07 774 ± 62 (n = 11)
P value (student's t test)     NS < .001 < .05 < .001
P value (student's t test in LOG-transf. data)     NS < .001 < .05 < .001

(a) Concerning chronic autoimmune thyroiditis, the lean and obese patients had a reduced TG level (-69% and -52%, respectively) compared with TG value of the lean and obese control subjects without autoimmune disease ( and ). The FFA value showed an increase (+73%) in the lean patients and a slight statistically significant decrease (-27%) in the obese patients ( and ).

  Table 1. TG, FFA, and Other Parameters in Control Subjects Studied (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean subjects 128 20/108 40 ± 4 25.13 ± 0.34 1.60 ±0.09 299 ± 30 (n = 18)
Obese subjects 208 86/122 43 ± 1 34.16 ± 0.36 1.81 ± 0.07 774 ± 62 (n = 11)
P value (student's t test)     NS < .001 < .05 < .001
P value (student's t test in LOG-transf. data)     NS < .001 < .05 < .001

  Table 2. TG, FFA, and Other Parameters in Patients With Chronic Autoimmune Thyroiditis (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 5 0/5 49 ± 6 23.89 ± 1.12 0.49 ± 0.10 518 ± 78
P value (student's t test)     NS NS < .01 < .01
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 5)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 19 2/17 43 ± 3 33.83 ± 1.58 0.87 ± 0.08 564 ± 77
P value (student's t test)     NS NS < .001 NS
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 5)

  Table 1. TG, FFA, and Other Parameters in Control Subjects Studied (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean subjects 128 20/108 40 ± 4 25.13 ± 0.34 1.60 ±0.09 299 ± 30 (n = 18)
Obese subjects 208 86/122 43 ± 1 34.16 ± 0.36 1.81 ± 0.07 774 ± 62 (n = 11)
P value (student's t test)     NS < .001 < .05 < .001
P value (student's t test in LOG-transf. data)     NS < .001 < .05 < .001

  Table 2. TG, FFA, and Other Parameters in Patients With Chronic Autoimmune Thyroiditis (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 5 0/5 49 ± 6 23.89 ± 1.12 0.49 ± 0.10 518 ± 78
P value (student's t test)     NS NS < .01 < .01
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 5)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 19 2/17 43 ± 3 33.83 ± 1.58 0.87 ± 0.08 564 ± 77
P value (student's t test)     NS NS < .001 NS
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 5)

(b) In the patients affected by chronic thyreopathies without autoantibodies, TG and FFA were not changed in the lean patients, whereas in the obese patients, a slight (but statistically significant) decrease of TG (-26%) and FFA (-27%) was observed ( and ). In evaluating this datum, it should be considered that the absence of autoantibodies does not necessarily exclude an autoimmune disease. Indeed, it is known that a percentage of autoimmune patients do not present autoantibodies, if they are screened in the phase preceding the appearance of autoantibodies. Both patient groups presented nearly normal values of total cholesterol.

  Table 1. TG, FFA, and Other Parameters in Control Subjects Studied (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean subjects 128 20/108 40 ± 4 25.13 ± 0.34 1.60 ±0.09 299 ± 30 (n = 18)
Obese subjects 208 86/122 43 ± 1 34.16 ± 0.36 1.81 ± 0.07 774 ± 62 (n = 11)
P value (student's t test)     NS < .001 < .05 < .001
P value (student's t test in LOG-transf. data)     NS < .001 < .05 < .001

  Table 3. TG, FFA, and Other Parameters in Patients With Chronic Thyreopathies Without Autoantibodies (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 9 1/8 50 ± 6 23.68 ± 0.78 1.15 ± 0.12 418 ± 33
P value (student's t test)     NS NS NS NS
P value (student's t test in LOG-transf. data)     NS NS NS NS (n = 3)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 21 0/21 48 ± 2 34.39 ± 0.90 1.34 ± 0.10 560 ± 47
P value (student's t test)     NS NS < .05 < .05
P value (student's t test in LOG-transf. data)     NS NS < .05 < .05 (n = 5)

(c) The results in patients affected by chronic active B or C hepatitis (a group of patients older than their control subjects), without signs of hepatic insufficiency or cirrhosis, are shown in and . Both the lean and obese patients showed reduced values of TG (-57% and -61%, respectively). In the obese patients, a reduced FFA level (-51%) was recorded. These patients presented nearly normal values of total cholesterol.

  Table 1. TG, FFA, and Other Parameters in Control Subjects Studied (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean subjects 128 20/108 40 ± 4 25.13 ± 0.34 1.60 ±0.09 299 ± 30 (n = 18)
Obese subjects 208 86/122 43 ± 1 34.16 ± 0.36 1.81 ± 0.07 774 ± 62 (n = 11)
P value (student's t test)     NS < .001 < .05 < .001
P value (student's t test in LOG-transf. data)     NS < .001 < .05 < .001

  Table 4. TG, FFA, and Other Parameters in Patients With Chronic Active B or C Hepatitis (mean ± SEM)


Patient initials Gender M/F Age (years) Virus type BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean Patients (BMI < 27 kg/m2)
  P.I.* F 17 B 26.36 0.57  
  P.M.G.* F 24 B 22.73 0.36  
  M.G.† M 42 C 24.60 0.96  
  C.G.‡ M 59 C 27.00 1.03  
  G.G. F 60 C 24.42 0.57  
  T.F.† F 60 C 26.98 0.50  
  D.M.F.† M 66 C 25.00 0.75  
Mean ± SEM n = 7 3M/4F 47 ± 7   25.30 ± 0.59 0.68 ± 0.09 ---
P value (student's t test)     NS   NS < .01  
P value (student's t test in LOG-transf. data)     NS   NS < .001  
Obese Patients (BMI > 27 kg/m2)
  P.C.* F 21 B 29.77 0.26 385
  F.F. F 26 B 31.2 0.44  
  C.A. M 27 B 31.68 0.41 393
  L.A. F 33 C 33.44 0.55  
  P.D. F 45 C 29.34 0.69  
  R.L. F 48 B 38.12 0.59 360
  P.A.* M 49 B 30.86 0.74 382
  M.C.‡ F 50 C 36.96 0.55  
  M.C.† M 52 C 28.80 1.04  
  D.L.F. F 53 C 54.22 0.73  
  M.S.† M 54 C 27.50 0.95  
  C.B.† F 58 C 30.50 0.98  
  G.A.† F 62 C 28.60 0.95  
  C.R. F 63 C 32.44 0.96  
  L.M.† M 67 C 28.37 0.79  
  S.A.† F 70 B,C 31.62 0.53  
  D.P. F 71 C 38.48 0.57  
  L.M.V.† F 72 C 27.20 1.01  
  M.S.† F 75 C 29.00 1.02  
Mean ± SEM n = 19 5M/14F 52 ± 4   32.53 ± 1.43 0.71 ± 0.05 380 ± 7
P value (student's t test)     < .05   NS < .001 < .01
P value (student's t test in LOG-transf. data)     < .01   NS < .001 < .001

(d) The results in the patients affected by lupus-like syndrome (who were prevailingly lean), compared with their control subjects, are shown in and . A marked decrease of TG (-73%) and increase (+74%) of FFA value were observed.

  Table 1. TG, FFA, and Other Parameters in Control Subjects Studied (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean subjects 128 20/108 40 ± 4 25.13 ± 0.34 1.60 ±0.09 299 ± 30 (n = 18)
Obese subjects 208 86/122 43 ± 1 34.16 ± 0.36 1.81 ± 0.07 774 ± 62 (n = 11)
P value (student's t test)     NS < .001 < .05 < .001
P value (student's t test in LOG-transf. data)     NS < .001 < .05 < .001

  Table 5. TG, FFA, and Other Parameters in Patients With Lupus-like Syndrome, SLE, and Rheumatoid Arthritis (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lupus-like Syndrome
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 7 1/6 32 ± 2 25.22 ± 2.08 0.43 ± 0.05 574 ± 81
P value (student's t test)     NS NS < .001 < .01
P value (student's t test in LOG-transf. data)     NS NS < .0001 < .01 (n = 4)
SLE and Rheumatoid Arthritis
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 6 2/4 40 ± 7 22.78 ± 2.18 0.55 ± 0.05 598 ± 58
P value (student's t test)     NS NS < .01 < .01
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 3)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 5 2/3 51 ± 3 35.50 ± 2.18 0.81 ± 0.06 ---
P value (student's t test)     NS NS < .05  
P value (student's t test in LOG-transf. data)     NS NS < .01  

(e) The results in the patients affected by SLE or rheumatoid arthritis, compared with their control subjects, are shown in and . In the lean patients, decrease of TG (-66%) and increase of FFA (+100%) values were observed. In the obese patients (in whom assay of FFA could not been performed), TG value was decreased (-55%).

  Table 1. TG, FFA, and Other Parameters in Control Subjects Studied (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean subjects 128 20/108 40 ± 4 25.13 ± 0.34 1.60 ±0.09 299 ± 30 (n = 18)
Obese subjects 208 86/122 43 ± 1 34.16 ± 0.36 1.81 ± 0.07 774 ± 62 (n = 11)
P value (student's t test)     NS < .001 < .05 < .001
P value (student's t test in LOG-transf. data)     NS < .001 < .05 < .001

  Table 6. TG, FFA, and Other Parameters in Patients With Anamnestic Allergy or Atopic Dermatitis/Asthma (mean +/= SEM)


N

Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 23 1/22 30 ± 3 23.13 ± 0.61 0.53 ± 0.04 455 ± 41
P value (student's t test)     < .05 NS < .001 < .05
P value (student's t test in LOG-transf. data)     NS NS < .0001 < .05 (n = 4)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 43 7/36 35 ± 2 31.76 ± 0.62 0.68 ± 0.03 654 ± 45
P value (student's t test)     < .001 NS < .001 NS
P value (student's t test in LOG-transf. data)     < .001 NS < .0001 NS (n = 3)

(f) The results in the patients affected by anamnestic allergy or atopic dermatitis/asthma (who were younger subjects than their controls) are shown in and . In the lean patients, reduced value of TG (-67%) and increased value of FFA (+52%) were observed. In the obese patients, a reduced value of TG (-62%) was also observed, whereas the value of FFA remained unchanged.

  Table 1. TG, FFA, and Other Parameters in Control Subjects Studied (mean ± SEM)


N Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Lean subjects 128 20/108 40 ± 4 25.13 ± 0.34 1.60 ±0.09 299 ± 30 (n = 18)
Obese subjects 208 86/122 43 ± 1 34.16 ± 0.36 1.81 ± 0.07 774 ± 62 (n = 11)
P value (student's t test)     NS < .001 < .05 < .001
P value (student's t test in LOG-transf. data)     NS < .001 < .05 < .001

  Table 7. TG, FFA, and Other Parameters in Patients With Some Diverse Autoimmune Diseases (mean +/= SEM)


Patient initials Gender M/F Age (years) BMI (kg/m2) TG (mmol/L) FFA (mcEq/L)
Scleroderma L.E. F 50 22.22 0.41 800
APECED syndrome P.S. F 42 13.67 0.77 441
Urticaria or urticarial vasculitis B.B. F 28 38.95 1.09  
  C.G. F 47 49.03 0.59  
  D.G.M. F 54 20.51 0.27  
  S.G. F 61 26.01 0.92 460
Reiter syndrome A.G. M 31 23.50 0.47  
Sjogren syndrome C.G.* F 72 17.09 0.68  
Ulcerative proctocolitis L.P.† F 21 33.80 0.64  
  L.M.† F 28 28.50 0.58  
  P.G. F 52 25.54 0.67  
Crohn's disease R.F.‡ F 36 25.57 0.72  
  R.G.‡ M 44 24.55 0.91  
Multiple sclerosis C.G. F 46 27.33 0.26  
Guillain-Barré syndrome S.C.* F 75 33.20 1.48  
Lean Patients (BMI < 27 kg/m2)
Mean ± SEM n = 9 2M/7F 49 ± 4 22.97 ± 1.42 0.65 ± 0.07 567 ± 117
P value (student's t test)     NS NS < .01 < .01
P value (student's t test in LOG-transf. data)     NS NS < .001 < .05 (n = 3)
Obese Patients (BMI > 27 kg/m2)
Mean ± SEM n = 6 6F 41 ± 8 35.14 ± 3.26 0.77 ± 0.18 ---
P value (student's t test)     NS NS < .01  
P value (student's t test in LOG-transf. data)     NS NS < .001  

(g) The results in patients affected by other autoimmune diseases (scleroderma, APECED, urticaria or urticarial vasculitis, Reiter or Sjogren syndromes, ulcerative proctocolitis or Crohn's disease, and chronic multiple sclerosis or Guillain-Barré syndrome) compared with lean and obese control subjects, are discussed above. In the lean and obese subjects, decreased value of TG was observed (-59% and -57%, respectively). In the lean patients, FFA value was increased (+90%). (In the obese patients, FFA assay could not be performed).

It is noteworthy that the lowest TG values were observed in the patients affected by lupus-like syndrome (0.43 ± 0.05 mmol/L), in 1 female patient (21 years old) with a chronic active type B hepatitis, myalgia, and strong familiarity for type B hepatitis (TG = 0.26), in 1 female patient (50 years old) affected by scleroderma (TG = 0.41), and in 2 female patients with autoimmune thyroiditis: one (36 years old) with rheumatologic complaints who developed a severe autoimmune thyroiditis following a twin pregnancy (TG = 0.30) and another (34 years old) with elevated titer of thyroidal autoantibodies (TG = 0.20). A very low value of TG was also evidenced in an adult female patient with chronic multiple sclerosis (TG = 0.26).

The low TG values were confirmed for some patients in various determinations, repeated after 2-3 months. Concerning the FFA level, data obtained when all the lean patients were compared with the lean control subjects, and all the obese patients were compared with the obese controls are as follows: (a) value in the lean patients (n = 22) vs that in the lean controls (n = 18) was increased (520 ± 31 vs 299 ± 30 mcEq/L, +74%, P < .001 in both statistical calculations); (b) value in the obese patients (n = 18) vs that in the obese control subjects (n = 11) was decreased (542 ± 34 vs 774 ± 62 mcEq/L, -30%, P < .01 in both statistical calculations). Obviously, this opposite behavior of FFA in lean and obese patients needs to be confirmed in a larger series of patients with autoimmune disorder or hyperreactivity of the immune system. FFA values obtained in the lean and obese control subjects in this study were very close to that reported in a previous metabolic study.[4] No correlation existed between TG and FFA values.

Discussion

Previous data suggest an unexpected association between fasting low TG (and high FFA levels) and chronic interstitial lung disease or pulmonary fibrosis.[1] The data in this paper confirm a link between decreased TG levels and some other autoimmune diseases or allergic hyperreactivity syndromes. TG were found to be also low in patients with autoimmunity, who were also obese, when compared with the obese subjects without autoimmune disorder.

These data seem to be in contrast with epidemiologic studies showing that the Greenland population has low TG (perhaps as a result of high dietary intake of polyunsaturated fatty acids or PUFA n-3) and a low incidence of some chronic diseases, including some immune disorders such as thyrotoxicosis, bronchial asthma, multiple sclerosis, and psoriasis, in comparison with western European populations.[5]

The available data from literature concerning the relationships between TG or FFA and some different autoimmune or dysreactive disorders are summarized and discussed below.

(a) Concerning chronic autoimmune thyroiditis and (b) chronicthyreopathies without autoantibodies, plasma TG metabolism is very well known with reference to the advanced states of thyroid disease, when functional failure is well developed.[6] The reported changes include primarily a significant increase in cholesterol levels in the most severely hypothyroid patients as well as a graded increase in mean serum TG as thyroid failure increases.[7] The data in this paper seem to be the first observation of a TG decrease in the early stage of chronic autoimmune thyroiditis, before the appearance of thyroid failure.

(c) Concerning chronic active B or C hepatitis (in the initial stage, before the appearance of liver failure or cholestasis), available data are scanty. In a Japanese study,[8] among 4256 visitors to a Tokyo Health Control Center for their health examination, 463 asymptomatic subjects (11%) showed abnormal liver function including elevation of ALT value. Ultrasonography and HCV antibody measurement had been applied to 362 cases in order to screen the etiology of liver dysfunction. HCV antibody-positive hepatitis showed high levels of total protein, aminotransferases, and low levels of TG as well as of albumin, A/G, total cholesterol, gamma-GT, and cholinesterase.

(d) Concerning lupus-like syndrome and (e) SLE or rheumatoid arthritis, numerous literature data are available, some confirming results in this paper, others opposite. Data in this study seem to be confirmed in 2 clinical studies by Svenson and coworkers[9,10] and in a research study by Lakatos and Harsagyi.[11] In the former study, Svenson[9] prospectively analyzed serum lipids and lipoprotein patterns of 33 patients with active chronic inflammatory arthritides. Before any treatment, low TG concentration (both in very-low-density lipoprotein (VLDL) and high-density lipoprotein (HDL) cholesterols) was found. In the other study by Svenson,[10] in 48 patients with untreated active rheumatoid arthritis, TG levels in VLDL and HDL were reduced by 10% to 30% and significantly correlated with inflammatory activity (C-protein reactivity [CPR]). The authors suggest that it is the degree of inflammatory activity that governs the altered lipoprotein metabolism in untreated active chronic inflammatory arthritides.[10] The relationships between CPR and VLDL indicate that the VLDL particles may be altered by the inflammatory process, and that the increased elimination may take place through the "scavenger pathway".[10] In patients with rheumatoid arthritis (26 men and 103 women), compared with TG values observed in controls (625 men and 749 women), Lakatos and Harsagyi[11] reported low TG level (with high LDL and low HDL cholesterol). This lipid pattern did not change when the patients were treated with steroidal or nonsteroidal anti-inflammatory drugs.

Other data, however, are at variance with results in this paper. In fact, increased TG levels were reported in patients with SLE (who had also depressed HDL and elevated VLDL cholesterol).[12,13] Ettinger and coworkers[14] also found high TG levels in 46 female patients with SLE. Elevated TG levels were suggested to result from the presence of autoantibodies to LPL (which occurred in 47% of SLE patients and in a similar percentage of patients with polymyositis or systemic sclerosis).[15] More recently, elevated TG levels (and other lipid abnormalities) were observed in 53 premenopausal (mean 34.5 years old) SLE patients and 45 controls; TG, as well as total cholesterol and LDL-cholesterol, were significantly correlated with proteinuria.[16] The authors concluded that proteinuria is a good predictor of dyslipoproteinemia in these patients.

The above observations on SLE and rheumatoid arthritis indicate that both high and low TG levels have been observed. These apparent contrasting findings could perhaps be explained by hypothesizing a different behavior of TG during the course of these disease, with low TG levels in the early stages and high TG levels in the advanced stages. This is supported by the very low TG level reported in the early stage (ninth day) of adjuvant-induced arthritis in rats.[17] It could be hypothesized that the development of SLE and rheumatoid arthritis (and perhaps other autoimmune diseases) is associated with a reduction in TG level, and that in the advanced stages of these diseases, appearance of antibodies toward LPL (which results in reduced TG clearance) and of proteinuria (which induces increased hepatic lipoprotein synthesis) is associated with an increase in TG levels.

(f) Concerning anamnestic allergy or atopic dermatitis/asthma, no data on the behavior of TG and FFA levels are available. The only finding marginally related to this study is the observation that in prepubertal and pubertal atopic patients, the proportions of linoleic acid in total plasma lipids and phospholipids were significantly increased and those of oleic acid reduced.[18] The n-3/n-6 fatty acid ratio of the TG fraction was also lower in atopic patients.[18] In these subjects, however, no significant decreases in the proportions of dihomo-gamma-linolenic acid and arachidonic acid were observed in plasma lipids, suggesting that delta-6-desaturase activity was not impaired in the atopic subjects.[18] This observation in atopia provides a good explanation for the beneficial effects of raising the n-3/n-6 ratio of dietary oils for suppressing allergic hyperreactivity or inflammatory process and repairing chronic fibrosing mechanisms in humans.

(g) With regard to other autoimmune diseases (scleroderma, APECED, urticaria or urticarial vasculitis, Reiter or Sjogren syndromes, ulcerative proctocolitis or Crohn's disease, and chronic multiple sclerosis or Guillain-Barré syndrome), only data concerning Sjogren disease are available. In 41 patients with primary Sjogren syndrome, docosahexaenoic acid was the EFA whose levels correlated (inversely) most closely with the clinical disease status.[19] Moreover, sigma n-3 EFA/sigma n-6 EFA ratios correlated significantly to the quantitative estimates of immunopathologic and clinical disease status.[19] These data are in agreement with current understanding of pro- and anti-immunoinflammatory roles of EFA.[19]

Concerning low TG level and autoimmunity, interesting data are reported in autoimmune diabetes. It is noteworthy that autoantibodies to glutamic acid decarboxylase (GAD) are considered the most useful marker for autoimmune diabetes.[20,21] GAD levels and their relationships with clinical features and pancreatic beta-cell function were studied in 140 young Chinese type 1 diabetic patients.[22] Patients who had antibodies to GAD had lower plasma TG and higher concentrations of plasma HDL-cholesterol as well as lower BMI and blood pressure.[22] In a large study on latent autoimmune diabetes in adults performed in western Finland, the prevalence of GAD was 9.3% among 1122 type 2 diabetic patients, 3.6% among 558 impaired glucose tolerance subjects, and 4.4% among 383 nondiabetic control subjects.[23] The GAD-positive patients (compared with GAD-negative subjects) had lower TG concentrations (1.40 ± 1.18 vs 1.75 /- 1.25 mmol/L, P = .003) as well as lower fasting C-peptide concentrations, insulin response to oral glucose, and blood pressure values.[23]

Possible role of polyunsaturated fatty acids (PUFA). Based on the observation of the low prevalence of lung disease among Greenland Eskimos as result of their diet high in PUFA, a protective role of the PUFA n-3 in lung disease is reported in an extensive review by Schwartz.[24] Incorporation of PUFA n-3 into lipid membranes lowers the production of inflammatory eicosanoids (a) through competition with arachidonic acid as a constituent of lipid membranes, (b) through competition with arachidonic acid as a substrate for prostaglandin-endoperoxide synthase (cyclo-oxygenase activity), and (c) through inhibition of the conversion of linolenic acid to arachidonic acid.[25] Gamma-linolenic acid has similar effects.[24] When PUFA n-3 are the substrate of the cyclo-oxygenase pathway (instead of arachidonic acid), they may modulate inflammatory activity, reducing leukotriene B4, which is a proinflammatory mediator responsible for production of eicosanoids and neutrophil recruitment.[24] Some data reported from the lung would suggest that inflammatory cells (neutrophils and macrophages) might affect TG metabolism. These cells contain much more arachidonic acid, which is accumulated in a large quantity into a TG pool.[26] The accumulation of arachidonic acid into TG could be a marker of cell activation, suggesting a central role of the TG pool in arachidonic acid metabolism of activated inflammatory cells.[26] Probably, in human inflammatory cells, TG is not only a storage site for arachidonic acid, but may also be an important regulator of arachidonic acid metabolism and eicosanoid biosynthesis.[27]

Fibrosing mechanism in autoimmune disease. Data in this study are in agreement with the previous published data about low TG in pulmonary fibrosis,[1] since most autoimmune disorders are fibrosing diseases that can evolve toward chronic interstitiopathy or pulmonary fibrosis.[28,29] Fibrosis is a pathologic process characterized by the replacement of normal tissue by mesenchymal cells and the extracellular matrix produced by these cells.[30] The sequence of events leading to fibrosis of an organ involves the subsequent processes of injury with inflammation and disruption of the normal tissue architecture, followed by tissue repair with accumulation of mesenchymal cells in the area of derangement.[30] Inflammatory cells (mainly mononuclear phagocytes), platelets, and endothelial cells (and, in pulmonary fibrosis, type II pneumocytes) play direct and indirect roles in tissue injury and repair.[30] In pulmonary fibrosis, several cytokines participate in the local injury and inflammatory reaction, such as interleukin (IL)-1, IL-8, monocyte chemotactic protein-1, and tumor necrosis factor-alpha.[30] Other cytokines are involved in tissue repair and fibrosis, such as platelet-derived growth factor (PDGF), insulin-like growth factor-1 (IGF-1), transforming growth factor-beta (TGF-beta), and basic-fibroblast growth factor.[30] Recent data directly demonstrated how acute tissue injury in the lung, initiated by a highly proinflammatory cytokine, IL-1-beta, converts to progressive fibrotic changes.[31] TGF-beta1 has been reported to modify fibroblast growth factor-2 production in type II cells in the lung.[32] Interferon-gamma, a Th1 cytokine, downregulates the expression of TGF-beta1 and is a therapeutic agent in some diverse autoimmune diseases.[33] In several other autoimmune diseases (besides primary pulmonary fibrosis), these cytokines also play a key role in acute local injury/inflammatory reaction and chronic repair.[34,35]

It is noteworthy that the progress toward autoimmune chronic pulmonary fibrosis is especially associated with increased expression of pulmonary IGF-1, mainly in interstitial and alveolar macrophages.[36,37] In lungs of patients with idiopathic fibrosis, this growth factor is known to promote fibroblast proliferation and differentiation as well as to stimulate collagen synthesis by these cells.[38] The role for interstitial macrophages as a source of IGF-1 is crucial, and the level of expression of IGF-I by interstitial macrophages was related to the degree of disease severity.[38] In rheumatoid arthritis, synovium is characterized by the presence of a number of secreted cytokines (including IGF-1, PDGF, and TGF-beta) of activated lymphocytes, macrophages, and other cell types (such as fibroblasts and endothelial cells) that have a key role in the rheumatoid inflammatory process.[34] The activity of these cytokines appears to account for many of the features of rheumatoid synovitis as well as the systemic manifestations of disease.[34]

IGF-1, the mediator strongly implicated in fibroblast proliferation and collagen deposition, is also involved in TG clearance. Total IGF-1 levels are lower in patients with atherogenic lipid profile, so contributing to the development of atherosclerosis.[39,40] Age- and sex-adjusted fasting free IGF-1 levels were inversely correlated with serum TG (but showed no relation with most cardiovascular risk factors).[41] IGF-1 concentrations would be subnormal in obesity, a disease in which TG levels are increased.[42] In both healthy young men and type 2 diabetic patients, IGF-1 lowers fasting and postprandial TG (without affecting the metabolism of intestine-derived TG-rich lipoproteins after a high-fat meal in normal subjects) and increases FFA level, most probably by decreasing insulin secretion and production of VLDL particles and possibly by increasing their turnover.[43,44] Recombinant IGF-1 administered subcutaneously to healthy subjects or patients with type 2 diabetes causes a drop in plasma levels of TG and VLDL.[45] IGF-1 may be envisaged for cases of insulin resistance[45] or altered lipid profile in type 2 (noninsulin-dependent) diabetic patients.[43] On these grounds, an increased release of IGF-1 by different tissues might be the cause of the metabolic pattern (low TG and high FFA levels) occurring in some patients with autoimmune disease.

Conclusion

Despite the data discussed above, the link between low TG and autoimmune disorder is not clear. Several tissues (adipose tissue, muscle, and lung) are rich in LPL and are the source of postheparin lipase activity, thus contributing to clear serum TG level and to take up FFA from circulating VLDL-TG.[46] It might be postulated that, in autoimmune diseases, an increased content and/or release of LPL by these tissues might lower serum TG concentration and increase FFA level. However, this is a mere hypothesis that needs to be confirmed with further experimental work. The observation that low TG levels are associated with several diverse autoimmune diseases and hyperactive immune response speaks in favor of a pathogenic role of TG lowering.

It could be postulated that low TG level may induce alterations that favor autoimmune processes. On the other hand, it could also be possible that both low TG (and high FFA) and autoimmune disorder are the result of a common unknown etiologic factor, either congenital or acquired.

The present study should be considered as a preliminary clinical observation that requires confirmation in a larger population of patients with autoimmune diseases and further investigation concerning the mechanisms involved (such as evaluation of postheparin plasma LPL activity, plasma lipoprotein status and metabolism, IGF-1 level, and possibly n-3/n-6 ratio in total plasma lipids and phospholipids).

Hypotriglyceridemia (sometimes severe), however, seems to be a precocious and constant marker of autoimmune disease or immune system hyperreactivity (or even of familiarity for autoimmune disorder). In addition, this marker is easy to assay and is very inexpensive. With the assessment of TG value in all patients, it would be possible to detect the subjects with low TG level who are at risk for autoimmunity, dysreactive disorder, initial interstitiopathy (without clear signs of pulmonary disease), or undiscovered/initial other autoimmune disease.

References

  1. Iannello S, Cavaleri A, Camuto M, et al. Low fasting serum triglyceride and high free fatty acid levels in pulmonary fibrosis: a previously unreported finding. MedGenMed. 2002;4(2). Available at: http://www.medscape.com/viewarticle/434218. Accessed August 5, 2003.

  2. Tan EM, Cohen AS, Fries JF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum. 1982;25:1271-1277. Abstract

  3. Iannello S, Camuto M, Cantarella S, et al. Rheumatoid syndrome associated with lung interstitial disorder in a dental technician exposed to ceramic silica dust. A case report and critical literature review. Clin Rheumatol. 2002;21:76-81. Abstract

  4. Belfiore F, Iannello S, Camuto M, Fagone S, Cavaleri A. Insulin sensitivity of blood glucose versus insulin sensitivity of blood FFA in normal, obese and obese-diabetic subjects. Metabolism. 2001;50:573-582. Abstract

  5. Kromann N, Green A. Epidemiologic studies in the Upernavik district, Greenland: incidence of some chronic disease 1950-1974. Acta Med Scand. 1980;208:401-406. Abstract

  6. Nikkila EA, Kekki M. Plasma triglyceride metabolism in thyroid disease. J Clin Invest. 1972;51:2103-2114. Abstract

  7. Kutty KM, Bryant DG, Farid NR. Serum lipids in hypothyroidism - a re-evaluation. J Clin Endocrinol Metab. 1978;46:55-56. Abstract

  8. Takemura Y, Kobayashi H, Kamachi M, et al. Pathogenesis-screening tests for liver dysfunction in the asymptomatic patients with elevated ALT values and their diagnostic efficacies in primary care medicine [Japanese]. Rinsho Byori. 1996;44:261-266. Abstract

  9. Svenson KL, Lithell H, Hallgren R, Vessby B. Serum lipoprotein in active rheumatoid arthritis and other chronic inflammatory arthritides. II. Effects of anti-inflammatory and disease-modifying drug treatment. Arch Intern Med. 1987;147:1917-1920. Abstract

  10. Svenson KL, Lithell H, Hallgren R, Selinus I, Vessby B. Serum lipoprotein in active rheumatoid arthritis and other chronic inflammatory arthritides. I. Relativity to inflammatory activity. Arch Intern Med. 1987;147:1912-1916. Abstract

  11. Lakatos J, Harsagyi A. Serum total, HDL, LDL cholesterol, and triglyceride levels in patients with rheumatoid arthritis. Clin Biochem. 1988;21:93-96. Abstract

  12. Ilowite NT, Samuel P, Ginzler E, Jacobson MS. Dyslipoproteinemia in pediatric systemic lupus erythematosus. Arthritis Rheum. 1988;31:859-863. Abstract

  13. Borba EF, Bonfa E, Vinagre CG, Ramires JA, Maranhao RC. Chylomicron metabolism is markedly altered in systemic lupus erythematosus. Arthritis Rheum. 2000;43:1033-1040. Abstract

  14. Ettinger WH, Goldberg AP, Applebaum-Bowden D, Hazzard WR. Dyslipoproteinemia in systemic lupus erythematosus. Effect of corticosteroids. Am J Med. 1987;83:503-508. Abstract

  15. Reichlin M, Fesmire J, Quintero Del Rio AI, Wolfson-Reichlin M. Autoantibodies to lipoprotein lipase and dyslipidemia in systemic lupus erythematosus. Arthritis Rheum. 2002;46:2957-2963. Abstract

  16. Formiga F, Meco JF, Pinto X, et al. Lipid and lipoprotein levels in premenopausal systemic lupus erythematosus patients. Lupus. 2001;10:359-363. Abstract

  17. Yamaguchi Y, Tanaka Y, Yamada K, Bando Y, Kunitomo M. Abnormal lipid metabolism in adjuvant arthritic rats. Jpn J Pharmacol. 1989;50:377-386. Abstract

  18. Sakai K, Okuyama H, Shimazaki H et al. Fatty acid compositions of plasma lipids in atopic dermatitis/asthma patients. Arerugi. 1994;43:37-43. Abstract

  19. Oxholm P, Asmussen K, Wiik A, Horrobin DF. Essential fatty acid status in cell membranes and plasma of patients with primary Sjogren's syndrome. Correlations to clinical and immunologic variables using a new model for classification and assessment of disease manifestations. Prostaglandins Leukot Essent Fatty Acids. 1998;59:239-245. Abstract

  20. Zimmet PZ, Tuomi T, Mackay IR, et al. Latent autoimmune diabetes mellitus in adults (LADA): the role of antibodies to glutamic acid decarboxylase in diagnosis and prediction of insulin dependency. Diabetic Med. 1994;11:299-303. Abstract

  21. von Boehmer H, Sarukhan A. GAD, a single autoantigen for diabetes. Science. 1999;284:1135-1137. Abstract

  22. Ko GT, Chan JC, Yeung VT, et al. Antibodies to glutamic acid decarboxylase in young Chinese diabetic patients. Ann Clin Biochem. 1998;35:761-767. Abstract

  23. Tuomi T, Carlsson A, Li H, et al. Clinical and genetic characteristics of type 2 diabetes with and without GAD antibodies. Diabetes. 1999;48:150-157. Abstract

  24. Schwartz J. Role of polyunsaturated fatty acids in lung disease. Am J Clin Nutr. 2000;7(suppl):393S-396S.

  25. Leaf A, Weber PC. Cardiovascular effects of n-3 fatty acids. N Engl J Med. 1988;318:549-557. Abstract

  26. Triggiani M, De Marino V, De Crescenzo G, Marone G. Arachidonic acid remodeling in human inflammatory cells migrating to the lung in vivo. Int Arch Allergy Immunol. 1997;113:190-192. Abstract

  27. Triggiani M, Oriente A, de Crescenzo G, Marone G. Biochemical functions of a pool of arachidonic acid associated with triglycerides in human inflammatory cells. Int Arch Allergy Immunol. 1995;107:261-263. Abstract

  28. Briggs DC, Vaughan RW, Welsh KI, et al. Immunogenetic prediction of pulmonary fibrosis in systemic sclerosis. Lancet. 1991;338:661-662. Abstract

  29. Ferri C, La Civita L, Fazzi P, et al. Interstitial lung fibrosis and rheumatic disorders in patients with hepatitis C virus infection. Br J Rheumatol. 1997;36:360-365. Abstract

  30. Martinet Y, Menard O, Vaillant P, Vignaud JM, Martinet N. Cytokines in human lung fibrosis. Arch Toxicol. 1996;18(suppl):127-139.

  31. Kolb M, Margetts PJ, Anthony DC, Pitossi F, Gauldie J. Transient expression of IL-1beta induces acute lung injury and chronic repair leading to pulmonary fibrosis. J Clin Invest. 2001;107:1529-1536. Abstract

  32. Li CM, Khosla J, Hoyle P, Sannes PL. Transforming growth factor-beta(1) modifies fibroblast growth factor-2 production in type II cells. Chest. 2001;120(suppl):60S-61S.

  33. Lawrence DA. Transforming growth factor-beta: a general review. Eur Cytokine Netw. 1996;7:363-374. Abstract

  34. Harrison's Principles of Internal Medicine. 13th edition. New York: McGraw-Hill Inc; 1994:1650.

  35. Harrison NK, Cambrey AD, Myers AR, et al. Insulin-like growth factor-I is partially responsible for fibroblast proliferation induced by bronchoalveolar lavage fluid from patients with systemic sclerosis. Clin Sci. 1994;86:141-148. Abstract

  36. Piguet PF, Van GY, Guo J. Heparin attenuates bleomycin but not silica-induced pulmonary fibrosis in mice: Possible relationship with involvement of myofibroblasts in bleomycin, and fibroblasts in silica-induced fibrosis. Int J Exp Pathol. 1996;77:155-161. Abstract

  37. Maeda A, Hiyama K, Yamakido H, Ishioka S, Yamakido M. Increased expression of platelet-derived growth factor A and insulin-like growth factor-I in BAL cells during the development of bleomycin-induced pulmonary fibrosis in mice. Chest. 1996;109:780-786. Abstract

  38. Uh ST, Inoue Y, King TE Jr, et al. Morphometric analysis of insulin-like growth factor-I localization in lung tissues of patients with idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 1998;58:1626-1635.

  39. Ferns GGA, Morani AS, Anggard EE. The insulin-like growth factors: their putative role in atherogenesis. Artery. 1991;18:197-225. Abstract

  40. Delafontaine P, Bernstein KE, Alexander RW. Insulin-like growth factor-1 gene expression in vascular cells. Hypertension. 1991;17:693-699. Abstract

  41. Janssen JAM, Stolk RP, Pols HAP, Grobbee DE, Lamberts SWJ. Serum total IGF-I, free IGF-I, and IGFBP-1 levels in an elderly population. Relation to cardiovascular risk factors and disease. Arterioscler Thromb Vasc Biol. 1998;18:277-282. Abstract

  42. Nam SY, Lee EJ, Kim KR, et al. Effect of obesity on total and free insulin-like growth factor (IGF)-1, and their relationship to IGF-binding protein (BP)-1, IGFBP-2, IGFBP-3, insulin, and growth hormone. Int J Obesity. 1997;21:355-359.

  43. Zenobi PD, Holzmann P, Glatz Y, Riesen WF, Froesch ER. Improvement of lipid profile in type 2 (non-insulin-dependent) diabetes mellitus by insulin-like growth factor I. Diabetologia. 1993;36:465-469. Abstract

  44. Bianda T, Zapf J, Froesch ER, Schmid C. Insulin-like growth factor-I lowers fasting and postprandial triglyceride levels without affecting chylomicron clearance in healthy men. J Intern Med. 1999;245:455-462. Abstract

  45. Binoux M. The IGF system in metabolism regulation. Diabetes Metab. 1995;21:330-337.

  46. Belfiore F. Enzyme Regulation and Metabolic Diseases. Basel, Switzerland: Karger; 1980.