- Email: [email protected]
Abnormal Lipoprotein and Apolipoprotein Pattern in Lipoprotein Glomerulopathy
ORIGINAL INVESTIGATIONS
Abnormal Lipoprotein and Apolipoprotein Pattern in Lipoprotein Glomerulopathy Shinichi Oikawa, MD, Norihiro Suzuki, MD, Eriko Sakuma, MD, Takao Saito, MD, Kazuyuki Namai, MD, Hidetoshi Kotake, MD, Yutaka Fujii, MD, and Takayoshi Toyota, MD • Recently, two cases of renal disease were observed in which there was an abnormal accumulation of lipids, "lipoprotein thrombi," in the glomerular capillary lumen. This disease has been designated as lipoprotein glomerulopathy. Four other cases have been diagnosed independently by renal histology in other clinical laboratories. All six patients showed proteinuria (1.6 to 10 gjd), normal lecithin-cholesterol acyltransferase (LeAT) activity, type III hyperlipoproteinemia-like lipoprotein profiles, and significantly (P < 0.01) higher levels of plasma apolipoprotein (apo) E (>10 mgj dL) compared with the control patients with hyperlipidemic nephrotic syndrome without lipoprotein thrombi and type lib hyperlipoproteinemia without renal disease. Lipoprotein glomerulopathy is not familial type III hyperlipoproteinemia (dysbetalipoproteinemia), because apolipoprotein E3 is present. Apo E isoforms were all rare: five cases of E2j3 and one case of E4j4. These results suggest that excessive apo E is associated with apo E isoform and lipoprotein metabolic derangement in such a renal disease. Further studies are needed on the relationship between the apo E hyperlipoproteinemia and the formation of lipoprotein thrombi. © 1991 by the National Kidney Foundation, Inc. INDEX WORDS: Lipoprotein thrombi; lipoprotein glomerulopathy; apolipoprotein E.
A
BNORMAL accumulation of lipids in the kidney has been reported in hereditary disorders, eg, lipid storage diseases and lecithincholesterol acyltransferase (LCAT) deficiency.I,2 Low LCA T activity in hepatorenal syndrome may induce lipid deposition in kidney,3 but it is rare to find lipid storage in glomeruli on routine renal biopsies. Two cases of renal disease with glomerular lipids deposition, "lipoprotein thrombi," have been reported and designated as lipoprotein glomerulopathy.4,5 Another case was partially described in a review by Faraggiana and Churg, I and Koitabashi et al reported this case after a long-term follow-up.6 Three other cases were diagnosed independently in other laboratories and Shibata et ar reported one of these. In all six cases, the glomerular lipoprotein thrombi strongly resembled each other, and these patients had proteinuria and hyperlipoproteinemia (phenotype III). We studied lipoprotein components, plasma apolipoprotein profiles, and apolipoprotein (apo) E isoforms to clarify the relationship between glomerular changes and lipoprotein/ apolipoprotein abnormalities in the lipoprotein glomerulopathy. PATIENTS AND METHODS
Patients and Controls Six patients were diagnosed with lipoprotein glomerulopathy based on the renal histology, and the presence of subcutaneous
edema, proteinuria, and hypoproteinemia as shown in laboratory data (Table I). Cases 1,52,73,6 and 54 have been reported previously. Case 1 was the elder sister of case 2. Glomerular histological changes in all six cases were similar, and were diagnosed as lipoprotein glomerulopathy independently by different laboratories. Serum creatinine concentration and creatinine clearance were normal. Six nephrotic patients (four males and two females) from 15 to 47 years of age (31 ± 5 [mean ± SE]), 38 type lIb hyperiipoproteinemic patients (22 males and 16 females) from 21 to 67 years (52 ± 2), seven familial type III hyperlipoproteinemic patients (dysbetalipoproteinemia, one male and six females) from 28 to 65 years (48 ± 4) served as controls. Renal histology of control nephrotic patients showed membranoproliferative glomerulonephritis (n = 1), focal glomerulosclerosis (n = 2), and minimal change disease (n = 3). Glomerular lipoprotein thrombi were not observed in any of these six nephrotic patients. The plasma cholesterol (Ch) level 5.69 to 12.93 mmolfL (220 to 500 mg/dL) and triglyceride (TG) level 2.258 to 5.645 mmoljL (200 to 500 mgjdL) in type lIb hyperiipoproteinemic patients was matched with the levels of lipoprotein glomerulopathy patients. Familial type III hyperlipoproteinemia (dysbetalipoproteinemia) was diagnosed as apo E3 deficiency.
From the Second and Third Departments 0/ Internal Medicine, and the Department o/Public Health, Tohoku University School o/Medicine; and the Department o/Internal Medicine, Mizusawa General Hospital, Japan. Address reprint requests to Shinichi Oikawa, MD, The Third Department 0/ Internal Medicine, Tohoku University School o/Medicine, 1-1 Seiryo cho, Sendai 980, Japan. © 1991 by the National Kidney Foundation, Inc. 0272-6386/91/1805-0003$3.00;0
American Journal of Kidney Diseases, Vol XVIII, No 5 (November), 1991: pp 553-558
553
OIKAWA ET AL
554 Table 1. Laboratory Data of Lipoprotein Glomerulopathy Patients
Case No.
1
2 3 4
5 6
Age/Sex
Urine Protein (g/d)
Serum Total Protein (g/dL)
Albumin (%)
i'illobulin (%)
BUN (mmol/L)
Creatinine (I'mol/L)
Uric Acid (I'mol/L)
Creatinine Clearance (mL/min)
57/F 40/F 211M 271M 371M 201M
1.6 9 8 10 7 8
6.5 5.5 5.2 4.4 5.2 4.0
63.0 54.6 55.2 60.3 57.3 53.0
14.0 14.5 4.4 8.9 12.0 4.8
5.4 8.2 8.9 3.9 10.0 4.6
61 76 69 61 114 114
321 303 672 410 476 470
82 118 134 86 89
77
Abbreviation: BUN, blood urea nitrogen.
RESULTS
Laboratory Methods
All six patients with specific renal lesions initially had proteinuria. Case 15 showed mild proteinuria on admission, while the other five patients showed marked proteinuria and hypoproteinemia, although renal function remained within the normal range. No patients had diabetes mellitus, hypothyroidism, or LCAT deficiency (LCAT activities were normal, ranging from 70 to 117 nmol/mL/h). The lipoprotein components are shown in Table 2. Hypertriglyceridemia or hypercholesterolemia was observed in all cases, and the VLDL and IDL fractions were increased. This pattern suggests hyperlipoproteinemia phenotype III as defined by Hazzard et al l4 or Fredrickson et al. 15 Plasma apolipoproteins are shown in Table 3, and apo E was increased in all cases oflipoprotein glomerulopathy.
A blood sample was obtained from the antecubital vein of each subject after an overnight fast, and plasma was collected after centrifugation at 3,000 rpm for 10 minutes. Very-lowdensity lipoprotein (VLDL: d < 1.006), intermediate-density lipoprotein (IDL: d = 1.006 to 1.019), low-density lipoprotein (LDL: d = 1.019 to 1.063), high-density lipoprotein 2 (HDL2: d = 1.063 to 1.125), and high-density lipoprotein 3 (HDL3: d = 1.125 to 1.21) were separated by the methods of Hatch and Lees. s TG and Ch in the plasma samples and each lipoprotein were measured enzymatically by commercially available test kits (Triglyceride E-Test and Cholesterol E-Test, Wako Jyunyaku, Osaka, Japan). Plasma apo A-I, A-II, B, CII, C-III, and E were measured by single radial immunodifusion methods (Daiichi Kagaku Yakuhin, Tokyo, Japan).9 Apo E phenotype determination using VLDL was performed by our modification of the method lO of Kashyap et alII in five cases. In the other case, apo E phenotyping from plasma was determined by the immunoblotting methods l2 .13 using anti-apo E antibody (Daiichi Kagaku Yakuhin). LCAT activity in each plasma sample was measured by the Anasorb LCAT Kit (Wako Jyunyaku).
Table 2. Lipoprotein Profile of Lipoprotein Glomerulopathy Patients
Case No.
2
3 4
5 6
TG Ch TG Ch TG Ch TG Ch TG Ch TG Ch
Plasma
VLDL
IDL
LDL
HDL2
145 241 216 238 396 238 211 182 327 196 338 458
89 40 109 74 330 114 128 51 219 103 168 104
17 28 44 51 24 23 25 21 44 33 55 81
25 122 45 80 23 80 37 88 42 39 96 234
13 43 9 16 7 6 13 10 11 8 9 21
NOTE. Values are mg/dl. Index for diagnosis of type III hyperlipoproteinemia. >0.42 (by Hazzard et a1 14 ) for diagnosis of type III hyperlipoproteinemia. t >0.3 diagnostic, and 0.25 to 0.3 suggestive (by Fredrickson et aI 15 ). *
HDL3
1 9 9 17
11 15 9 13 11 13 9 17
Index VLDL-Ch/ VLDL-TG*
VLDL-Ch/ Plasma TGt
0.45
0.28
0.68
0.34
0.35
0.29
0.40
0.24
0.47
0.32
0.62
0.31
555
APOLIPOPROTEIN IN LIPOPROTEIN GLOMERULOPATHY Table 3. Plasma ApOlipoprotein Profile of Lipoprotein Glomerulopathy Patients Case No.
A-I
A-II
B
C-II
C-III
E
1 2 3 4 5 6
141 116 123 85 98 103
30 23 24 24 31 22
95 123 115 102 106 188
4.8 10.1 8.7 5.1 10.0 6.3
10.8 17.8 23.7 5.0 15.0 11.9
14.0 15.2 13.7 11 .7 10.3 19.3
NOTE. Values are mg/dl.
The comparison of lipoprotein lipids and plasma apolipoprotein profiles between lipoprotein g1omerulopathy patients and control patients (nephrotic syndrome, type lIb hyperlipoproteinemia and familial type III hyperlipoproteinemia) is shown in Table 4. The VLDL and IDL frac-
tions were increased, and LDL-Ch was decreased in all lipoprotein g1omerulopathy patients. Plasma apo A-I and B were significantly lower in the g1omerulopathy patients than in control subjects with nephrotic syndrome and type lIb hyperlipoproteinemia. There was no difference in apo C-II levels between the g1omerulopathy and control groups. C-III levels were slightly higher in nephrotic syndrome than in the other groups. The apo E level in lipoprotein glomerulopathy patients was markedly higher than in control nephrotic patients and hyperlipoproteinemic patients. Apolipoprotein VLDL analysis on five cases is shown in Fig 1. All patients had apo E3 and rare isoform patterns (E2/3 in four cases and E41 4 in one case). Phenotyping from plasma in case 6 showed E2/3 (data not shown).
Table 4. Lipoproteins and Apolipoproteins of All Subjects Control
Patient Group (n)
Urine protein (g/d) Serum total protein (g/dL) Lipids (mg/dL) Plasma TG Ch VLDL TG Ch IDL TG Ch LDL TG Ch TG HDL2 Ch TG HDL3 Ch Apolipoproteins (mg/dL) A-I A-II B C-II C-III E
(A) Nephrotic Syndrome (n = 6)
(B) Hyperlipoproteinemia (no renal disease) (n = 38)
(C) Oysbetalipoproteinemia (n = 7)
(0) Lipoprotein Glomerulopathy (n = 6)
5.7 ± 0.5
7.3 ± 1.2
4.2 ± 0.3
5.1 ± 0.4
196 346 115 49 18 27 44 207 10 17 8 14
± ± ± ± ± ± ± ± ± ± ± ±
23 26 21 11 3 8 7 16 1
175 33 162 7.6 19.0 6.9
± ± ± ± ± ±
16 4 10 1.3 2.7 0.8
6
2
± ± ± ± ± ± ± ± ± ± ± ±
14 7 14 5 2 1 3 8
116 ± 29 ± 155 ± 7.9 ± 17.5 ± 6.9 ±
3
287 265 207 63 21 19 41 149 10 18 8 16
426 330 251 156 40 51 27 70 (29 (30
± ± ± ± ± ± ± ± ± ±
102 47 45 35 8 11 6 12 4)'
272 259 174 81 35 40 45 107 10
4)'
17
8 14
4 0.5 1.4 0.3
(116 (26 93 8.5 14.4 19.2
± 14)t ± 8); ± 14 ± 0.8 ± 2.2 ± 3.3
± ± ± ± ± ± ± ± ± ± ±
39 41 36 13 6 9 11 28 1
P Value (0) v (A)
.05
(0)
v (B)
.05 .05
.02
6
1
c±:
111 ± 26 ± 122 ± 7.5 ± 14.0 ± 14.0 ±
8 2 14 1.0 2.6 1.3
.01 .05
.05
.001
.001
NOTE. Values are mean ± SEM. There were no significant differences between lipoprotein glomerulopathy cases (D) and dysbetalipoproteinemia (C). , Values are shown as HDL (HDL2 ± HDL3)' t Calculation from 5 cases. ; Calculation from 3 cases.
556
OIKAWA ET AL
E4
I,
E4
-
E3 ,
~~.
E3 . , E2 ~ E1
•••••-• ...."
Cs
2/2
2/3
Dys-f3-lipoproteinemia
2/3
2/3
2/3
Fig 1. Polyacrylamide gel electrophoresis of apolipoprotein VLDL. Dysbetalipoproteinemia (apo E3 deficiency) is shown in the left two lanes. Each lipoprotein glomerulopathy case (cases 1 to 5) had apo E3 •
4/4
4 5 3 Lipoprotein glomerulopathy 1
2
Plasma apo E level showed a significant correlation with VLDL-Ch concentration in both the control groups with nephrotic syndrome and type lIb hyperlipoproteinemia (P < 0.01, Fig 2). However, in the lipoprotein glomerulopathy patients, there was a discrepancy between apo E and VLDL-Ch levels, as seen in dysbetalipoproteinemia.
The apo E level showed a positive correlation with the VLDL-Ch concentration in control subjects, but not in the case of lipoprotein glomerulopathy that resembled familial type III hyperlipoproteinemia (dysbetalipoproteinemia) caused by apo E3 deficiency.21 In dysbetalipoprotein-
DISCUSSION
In the present study, patients who had the specific renal glomerular lesion, "lipoprotein thrombi,,,4-7 were found to have a characteristic lipoprotein composition and plasma apolipoprotein pattern. Hyperlipoproteinemia and plasma apolipoprotein profiles in chronic renal diseases have been reported. 16 Their characteristics are increased VLDL-TG, LDL-TG, and HDL-TG, and decreased HDL-Ch. When creatinine clearance is decreased to less than 1.17 mL/s (70 mLI min), apo C-I1 and C-III are increased. In the present study, TG levels oflipoprotein glomerulopathy patients whose creatinine clearance was normal were higher than levels of control nephrotic patients. The lipoprotein profile associated with lipoprotein glomerulopathy was phenotype III hyperlipoproteinemia. This phenotype is not unique for renal disease 17 and develops secondarily in some metabolic disorders (eg, diabetes mellitus,18,19 hypothyroidism 20). Some control subjects (nephrotic patients without glomerular lipoprotein thrombi) had type III hyperlipoproteinemia diagnosed by chemical indices; however, their plasma apo E concentration was lower than that associated with lipoprotein glomerulopathy.
0
~~
_ Isoi
:!!
'"
Ji,
130
]
110
~
'?
'-'3 >
0
•
150 140
e
120
• •
90 80 60 50
30
o·
.
/ 1. ~.
011\
03
.
j.;{'..~.
/
, 0,
4
•
•
•
o
•
70
•
• o·
•
100
40
0
0
5
•
'i
6
7
"
•
o
• "
,
•
o
ii
,
,
,
II,i
9 10 11 12 13 14 15 16 17 18 19 26 Apo E
1"II/dR)
Fig 2. Relationship between VLDL cholesterol and the apo E concentration in a nephrotic control group, familial type III hyperlipoproteinemia, and lipoprotein glomerulopathy patients. Single univariate correlation (r = 0.4540) in 44 cases including nephrotic syndrome (0, n = 6) and type lib hyperlipoproteinemia (e, n = 38) was significant (P < 0.01). Lipoprotein glomerulopathy patients (.... , n = 6) digressed from this regression line and closely resembled dysbetalipoproteinemia (0, n = 7). uThis case of dysbetalipoproteinemia showed a high level of VLDL-Ch (280 mg/dL) and apo E (35.7 mg/dL). *Two patients in this group were diagnosed as having type III hyperlipoproteinemia according to the chemical indices of lipoprotein lipids profile.
557
APOLIPOPROTEIN IN LIPOPROTEIN GLOMERULOPATHY
emia, a deficiency of apo E or isoprotein E3, as well as the presence of apolipoprotein E2 variants, has been reported. 21 -24 No evidence suggesting dysbetalipoproteinemia was found by analyzing the apo E isoform pattern among lipoprotein glomerulopathy cases. The isoform pattern, which occurred E2/3 in five cases and E4/4 in one case, was rare,25-28 and Davignon et al reported that the E2/3 occurred in only 7.5% to 15.7% and E4/4 in only 0.4% to 5.8% ofcases. 29 Different clinical profiles in lipoprotein glomerulopathy and familial type III hyperlipoproteinemia were found . Familial type III hyperlipoproteinemia is associated closely with atherosclerosis. 3D No patient in the present study had a history of coronary heart disease or peripheral atherosclerosis, and aU had normal electrocardiograms at the time of the study. Concerning the histological and functional change in lipoprotein glomerulopathy, hyperlipoproteinemia might modify the lesion and compromise glomerular function. 31 ,32 However, these glomerular changes have not been reported in familial hypercholesterolemia caused by LDL
receptor dysfunction,33 familial type III hyperlipoproteinemia,33 or combined hyperlipoproteinemia. 34 Apo E hyperlipoproteinemia is common in chylomicronemia, which is frequently associated with pancreatitis and diabetes mellitus,35 but not with renal lesions. Apo E in lipoprotein glomerulopathy was markedly elevated, although renal function and proteinuria were not different from control nephrotic patients. These findings suggested that apo E hyperlipoproteinemia in this rare renal disease was associated with apo E isoform and lipoprotein metabolic derangement of the disease. It is likely that hyperlipoproteinemia coupled with apo E hyperlipoproteinemia decreases the renal function and causes renal pathology. However, further studies are needed on the relationship between apo E hyperlipoproteinemia and the formation of lipoprotein thrombi. ACKNOWLEDGMENT We are grateful to Drs Y oshihito Hara, Toshikatsu Shibata, Yasushi Koitabashi, Takashi Ishida, Tetsuya Mitarai, Yuzo Watanabe, and Hitoshi Terasaki for supplying plasma from their patients with lipoprotein glomerulopathy, and to Professor Hiroshi Sakaguchi for his advice.
REFERENCES I. Faraggiana T, Churg J: Renal lipidosis: A review. Hum Pathol 18:661-679, 1987 2. Gjone E: Familial lecithin:cholesterol acyltransferase deficiency-A new metabolic disease with renal involvement. Adv Nephrol \0:167-185, 1981 3. Sakaguchi H, Dachs S, Grishman E, et al: Hepatic glomerulosclerosis: An electron microscopic study of renal biopsies in liver disease. Lab Invest 14:533-545, 1965 4. Watanabe Y, Ozaki I, Yoshida F, et al: A case of nephrotic syndrome with glomerular lipoprotein deposition with capillary ballooning and mesangiolysis. Nephron 51 :265-270, 1989 5. Saito T, Sato H, Kudo K, et al: Lipoprotein g1omerulopathy: Glomerular lipoprotein thrombi in a patient with hyperlipoproteinemia. Am J Kidney Dis 13:148-153, 1989 6. Koitabashi Y, Ikoma M, Miyahira T, et al: Long-term follow-up ofa paediatric case oflipoprotein g1omerulopathy. Paediatr NephroI4:122-128, 1990 7. Shibata T, Kaneko N, Hara Y, et al: A case oflipoprotein g1omerulopathy. Light and electron microscopic observations of the glomerulus. Acta Pathol Jpn 40:448-457, 1990 8. Hatch Fr, Lees RS: Practical methods for lipoprotein analysis. Adv Lipid Res 6: 1-68, 1986 9. Goto Y, Akanuma Y, Harano Y, et al: Determination by the SRID method of normal values of serum apolipoproteins (A-I, A-II, B,C-I1, C-III and E) in normolipidemic healthy Japanese subjects. J C1in Biochem Nutr 1:73-88, 1986 10. Sano R, Abe R, Oikawa S, et al: ApoE-2/E-3 ratio of very low density lipoprotein in diabetes mellitus. Tohoku J Exp Med 146:131-136, 1985
II . Kashyap ML, Hynd BA, Robinson K, et al: Abnormal preponderance of sialilated apolipoprotein C-III in triglyceride rich lipoproteins in type V hyperlipoproteinemia. Metabolism 30:111-118,1981 12. Menzel HJ, Utermann G: Apolipoprotein E phenotyping from serum by Western blotting. Electrophoresis 7: 492-495, 1986 13. Havekes LM, de Kniff P, Beisiegel U, et al: A rapid micromethod for apolipoprotein E phenotyping directly in serum. J Lipid Res 28:455-463, 1987 14. Hazzard WR, Porte D Jr, Bierman EL: Abnormallipid composition of very low density lipoproteins in diagnosis of braod-beta disease (type III hyperIipoproteinemia). Metabolism 21 :1009-1019, 1972 15. Fredrickson OS, Morganroth J, Levy RI: Type III hyperIipoproteinemia: An analysis of two contemporary definitions. Ann Intern Med 82:150-157, 1975 16. Grutzmacher P, Marz W, Peschke B, et al: Lipoproteins and apolipoproteins during the progression of chronic renal disease. Nephron 50:\03-111, 1988 17. Gherardi E, Rota E, Calandra S, et al: Relationship among the concentration of serum lipoproteins and changes in their chemical composition in patients with untreated nephrotic syndrome. Eur J Clin Invest 7:563-570, 1977 18. Chance GW, Albutt EC, Edkins SM: Serum lipid and lipoproteins in untreated diabetic children. Lancet I: 11261128, 1969 19. Reaven GM, Greenfield MS: Diabetic hypertriglyceridemia. Evidence for three clinical syndromes. Diabetes 30: 66-75, 1981 (suppI2)
558 20. Hazzard WR, Biennan EL: Aggravation of broad f3 disease (type 3 hyperiipoproteinemia) by hypothyroidism. Arch Intern Med 130:822-828, 1972 21. Utennann G, Albrecht G, Steinmetz A: Polymorphism of apolipoprotein E and occurence of dysbetalipoproteinemia in man. Nature 269:604-607, 1977 22. Ghiselli G, Schaefer EJ, Gascon P, et al: Type III hyperiipoproteinemia associated with apolipoprotein E deficiency. Science 214:1239-1241, 1981 23. Utennann G, Jaeschke M, Menzel HJ: Familial hyperiipoproteinemia type III: Deficiency of a specific apolipoprotein (apo E-III) in the very-low-density lipoproteins. FEBS Lett 56:352-355, 1975 24. Rall SC, Jr, Weisgraber KH, Innerarity TL, et al: Identification of a new structural variant of human apolipoprotein E, E2 (Lysl46 -+ Gin), in a type III hyperiipoproteinemic subject with the E3/2 phenotype. J Clin Invest 72:1288-1297, 1983 25. Sano R, Abe R, Oikawa S, et al: Apo1ipoprotein E phenotypes ofnonno and hyperiipoproteinemia in Japanese. Tohoku J Exp Med 154:297-303, 1983 26. Sing CF, Davignon J: Role of the apolipoprotein E polymorphism in detennining nonnal plasma lipid and lipoprotein variation. Am J Hum Genet 37:268-285, 1985 27. Eto M, Watanabe K, Ishii K: A rapid flat gel isoelectric focussing method for the detennination of apolipoprotein E
OIKAWA ET AL
phenotypes and its application. Clin Chim Acta 149:21-28, 1985 28. Yamamura T, Yamamoto A, Hiramori K, et al: A new isofonn of apolipoprotein E-apo E-5-associated with hyperlipidemia and atherosclerosis. Atherosclerosis 50: 159-172, 1984 29. Davignon J, Gregg RE, Sing CF: Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis 8: 1-21, 1988 30. Morganroth J, Levy RI, Fredrickson DS: The biochemical, clinical and genetic features of type III hyperlipoproteinemia. Ann Intern Med 82:158-174,1975 31. Moorhead JF, Chan MK, Nahas M, et al: Lipid nephrotoxicity in chronic progressive glomerular and tubulo-interstitial disease. Lancet 2: 1309-1311, 1982 32. Diamond JR, Karnovsky MJ: Exacerbation of chronic aminonucleoside nephrosis by dietary cholesterol supplementation. Kidney Int 32:671-678, 1987 33. Brown MS, Goldstein JL: Familial hypercholesterolemia: Genetic, biochemical and pathophysiologic considerations. Adv Intern Med 20:273-296, 1975 34. Grundy SM, Chait A, Brunzell JD: Familial combined hyperiipidemia workshop. Arteriosclerosis 7:203-207, 1987 35. Havel RJ: Pathogenesis, differentiation, and management ofhypertriglyceridemia. Adv Intern Med 15:117-154, 1969
Abnormal Lipoprotein and Apolipoprotein Pattern in Lipoprotein Glomerulopathy Shinichi Oikawa, MD, Norihiro Suzuki, MD, Eriko Sakuma, MD, Takao Saito, MD, Kazuyuki Namai, MD, Hidetoshi Kotake, MD, Yutaka Fujii, MD, and Takayoshi Toyota, MD • Recently, two cases of renal disease were observed in which there was an abnormal accumulation of lipids, "lipoprotein thrombi," in the glomerular capillary lumen. This disease has been designated as lipoprotein glomerulopathy. Four other cases have been diagnosed independently by renal histology in other clinical laboratories. All six patients showed proteinuria (1.6 to 10 gjd), normal lecithin-cholesterol acyltransferase (LeAT) activity, type III hyperlipoproteinemia-like lipoprotein profiles, and significantly (P < 0.01) higher levels of plasma apolipoprotein (apo) E (>10 mgj dL) compared with the control patients with hyperlipidemic nephrotic syndrome without lipoprotein thrombi and type lib hyperlipoproteinemia without renal disease. Lipoprotein glomerulopathy is not familial type III hyperlipoproteinemia (dysbetalipoproteinemia), because apolipoprotein E3 is present. Apo E isoforms were all rare: five cases of E2j3 and one case of E4j4. These results suggest that excessive apo E is associated with apo E isoform and lipoprotein metabolic derangement in such a renal disease. Further studies are needed on the relationship between the apo E hyperlipoproteinemia and the formation of lipoprotein thrombi. © 1991 by the National Kidney Foundation, Inc. INDEX WORDS: Lipoprotein thrombi; lipoprotein glomerulopathy; apolipoprotein E.
A
BNORMAL accumulation of lipids in the kidney has been reported in hereditary disorders, eg, lipid storage diseases and lecithincholesterol acyltransferase (LCAT) deficiency.I,2 Low LCA T activity in hepatorenal syndrome may induce lipid deposition in kidney,3 but it is rare to find lipid storage in glomeruli on routine renal biopsies. Two cases of renal disease with glomerular lipids deposition, "lipoprotein thrombi," have been reported and designated as lipoprotein glomerulopathy.4,5 Another case was partially described in a review by Faraggiana and Churg, I and Koitabashi et al reported this case after a long-term follow-up.6 Three other cases were diagnosed independently in other laboratories and Shibata et ar reported one of these. In all six cases, the glomerular lipoprotein thrombi strongly resembled each other, and these patients had proteinuria and hyperlipoproteinemia (phenotype III). We studied lipoprotein components, plasma apolipoprotein profiles, and apolipoprotein (apo) E isoforms to clarify the relationship between glomerular changes and lipoprotein/ apolipoprotein abnormalities in the lipoprotein glomerulopathy. PATIENTS AND METHODS
Patients and Controls Six patients were diagnosed with lipoprotein glomerulopathy based on the renal histology, and the presence of subcutaneous
edema, proteinuria, and hypoproteinemia as shown in laboratory data (Table I). Cases 1,52,73,6 and 54 have been reported previously. Case 1 was the elder sister of case 2. Glomerular histological changes in all six cases were similar, and were diagnosed as lipoprotein glomerulopathy independently by different laboratories. Serum creatinine concentration and creatinine clearance were normal. Six nephrotic patients (four males and two females) from 15 to 47 years of age (31 ± 5 [mean ± SE]), 38 type lIb hyperiipoproteinemic patients (22 males and 16 females) from 21 to 67 years (52 ± 2), seven familial type III hyperlipoproteinemic patients (dysbetalipoproteinemia, one male and six females) from 28 to 65 years (48 ± 4) served as controls. Renal histology of control nephrotic patients showed membranoproliferative glomerulonephritis (n = 1), focal glomerulosclerosis (n = 2), and minimal change disease (n = 3). Glomerular lipoprotein thrombi were not observed in any of these six nephrotic patients. The plasma cholesterol (Ch) level 5.69 to 12.93 mmolfL (220 to 500 mg/dL) and triglyceride (TG) level 2.258 to 5.645 mmoljL (200 to 500 mgjdL) in type lIb hyperiipoproteinemic patients was matched with the levels of lipoprotein glomerulopathy patients. Familial type III hyperlipoproteinemia (dysbetalipoproteinemia) was diagnosed as apo E3 deficiency.
From the Second and Third Departments 0/ Internal Medicine, and the Department o/Public Health, Tohoku University School o/Medicine; and the Department o/Internal Medicine, Mizusawa General Hospital, Japan. Address reprint requests to Shinichi Oikawa, MD, The Third Department 0/ Internal Medicine, Tohoku University School o/Medicine, 1-1 Seiryo cho, Sendai 980, Japan. © 1991 by the National Kidney Foundation, Inc. 0272-6386/91/1805-0003$3.00;0
American Journal of Kidney Diseases, Vol XVIII, No 5 (November), 1991: pp 553-558
553
OIKAWA ET AL
554 Table 1. Laboratory Data of Lipoprotein Glomerulopathy Patients
Case No.
1
2 3 4
5 6
Age/Sex
Urine Protein (g/d)
Serum Total Protein (g/dL)
Albumin (%)
i'illobulin (%)
BUN (mmol/L)
Creatinine (I'mol/L)
Uric Acid (I'mol/L)
Creatinine Clearance (mL/min)
57/F 40/F 211M 271M 371M 201M
1.6 9 8 10 7 8
6.5 5.5 5.2 4.4 5.2 4.0
63.0 54.6 55.2 60.3 57.3 53.0
14.0 14.5 4.4 8.9 12.0 4.8
5.4 8.2 8.9 3.9 10.0 4.6
61 76 69 61 114 114
321 303 672 410 476 470
82 118 134 86 89
77
Abbreviation: BUN, blood urea nitrogen.
RESULTS
Laboratory Methods
All six patients with specific renal lesions initially had proteinuria. Case 15 showed mild proteinuria on admission, while the other five patients showed marked proteinuria and hypoproteinemia, although renal function remained within the normal range. No patients had diabetes mellitus, hypothyroidism, or LCAT deficiency (LCAT activities were normal, ranging from 70 to 117 nmol/mL/h). The lipoprotein components are shown in Table 2. Hypertriglyceridemia or hypercholesterolemia was observed in all cases, and the VLDL and IDL fractions were increased. This pattern suggests hyperlipoproteinemia phenotype III as defined by Hazzard et al l4 or Fredrickson et al. 15 Plasma apolipoproteins are shown in Table 3, and apo E was increased in all cases oflipoprotein glomerulopathy.
A blood sample was obtained from the antecubital vein of each subject after an overnight fast, and plasma was collected after centrifugation at 3,000 rpm for 10 minutes. Very-lowdensity lipoprotein (VLDL: d < 1.006), intermediate-density lipoprotein (IDL: d = 1.006 to 1.019), low-density lipoprotein (LDL: d = 1.019 to 1.063), high-density lipoprotein 2 (HDL2: d = 1.063 to 1.125), and high-density lipoprotein 3 (HDL3: d = 1.125 to 1.21) were separated by the methods of Hatch and Lees. s TG and Ch in the plasma samples and each lipoprotein were measured enzymatically by commercially available test kits (Triglyceride E-Test and Cholesterol E-Test, Wako Jyunyaku, Osaka, Japan). Plasma apo A-I, A-II, B, CII, C-III, and E were measured by single radial immunodifusion methods (Daiichi Kagaku Yakuhin, Tokyo, Japan).9 Apo E phenotype determination using VLDL was performed by our modification of the method lO of Kashyap et alII in five cases. In the other case, apo E phenotyping from plasma was determined by the immunoblotting methods l2 .13 using anti-apo E antibody (Daiichi Kagaku Yakuhin). LCAT activity in each plasma sample was measured by the Anasorb LCAT Kit (Wako Jyunyaku).
Table 2. Lipoprotein Profile of Lipoprotein Glomerulopathy Patients
Case No.
2
3 4
5 6
TG Ch TG Ch TG Ch TG Ch TG Ch TG Ch
Plasma
VLDL
IDL
LDL
HDL2
145 241 216 238 396 238 211 182 327 196 338 458
89 40 109 74 330 114 128 51 219 103 168 104
17 28 44 51 24 23 25 21 44 33 55 81
25 122 45 80 23 80 37 88 42 39 96 234
13 43 9 16 7 6 13 10 11 8 9 21
NOTE. Values are mg/dl. Index for diagnosis of type III hyperlipoproteinemia. >0.42 (by Hazzard et a1 14 ) for diagnosis of type III hyperlipoproteinemia. t >0.3 diagnostic, and 0.25 to 0.3 suggestive (by Fredrickson et aI 15 ). *
HDL3
1 9 9 17
11 15 9 13 11 13 9 17
Index VLDL-Ch/ VLDL-TG*
VLDL-Ch/ Plasma TGt
0.45
0.28
0.68
0.34
0.35
0.29
0.40
0.24
0.47
0.32
0.62
0.31
555
APOLIPOPROTEIN IN LIPOPROTEIN GLOMERULOPATHY Table 3. Plasma ApOlipoprotein Profile of Lipoprotein Glomerulopathy Patients Case No.
A-I
A-II
B
C-II
C-III
E
1 2 3 4 5 6
141 116 123 85 98 103
30 23 24 24 31 22
95 123 115 102 106 188
4.8 10.1 8.7 5.1 10.0 6.3
10.8 17.8 23.7 5.0 15.0 11.9
14.0 15.2 13.7 11 .7 10.3 19.3
NOTE. Values are mg/dl.
The comparison of lipoprotein lipids and plasma apolipoprotein profiles between lipoprotein g1omerulopathy patients and control patients (nephrotic syndrome, type lIb hyperlipoproteinemia and familial type III hyperlipoproteinemia) is shown in Table 4. The VLDL and IDL frac-
tions were increased, and LDL-Ch was decreased in all lipoprotein g1omerulopathy patients. Plasma apo A-I and B were significantly lower in the g1omerulopathy patients than in control subjects with nephrotic syndrome and type lIb hyperlipoproteinemia. There was no difference in apo C-II levels between the g1omerulopathy and control groups. C-III levels were slightly higher in nephrotic syndrome than in the other groups. The apo E level in lipoprotein glomerulopathy patients was markedly higher than in control nephrotic patients and hyperlipoproteinemic patients. Apolipoprotein VLDL analysis on five cases is shown in Fig 1. All patients had apo E3 and rare isoform patterns (E2/3 in four cases and E41 4 in one case). Phenotyping from plasma in case 6 showed E2/3 (data not shown).
Table 4. Lipoproteins and Apolipoproteins of All Subjects Control
Patient Group (n)
Urine protein (g/d) Serum total protein (g/dL) Lipids (mg/dL) Plasma TG Ch VLDL TG Ch IDL TG Ch LDL TG Ch TG HDL2 Ch TG HDL3 Ch Apolipoproteins (mg/dL) A-I A-II B C-II C-III E
(A) Nephrotic Syndrome (n = 6)
(B) Hyperlipoproteinemia (no renal disease) (n = 38)
(C) Oysbetalipoproteinemia (n = 7)
(0) Lipoprotein Glomerulopathy (n = 6)
5.7 ± 0.5
7.3 ± 1.2
4.2 ± 0.3
5.1 ± 0.4
196 346 115 49 18 27 44 207 10 17 8 14
± ± ± ± ± ± ± ± ± ± ± ±
23 26 21 11 3 8 7 16 1
175 33 162 7.6 19.0 6.9
± ± ± ± ± ±
16 4 10 1.3 2.7 0.8
6
2
± ± ± ± ± ± ± ± ± ± ± ±
14 7 14 5 2 1 3 8
116 ± 29 ± 155 ± 7.9 ± 17.5 ± 6.9 ±
3
287 265 207 63 21 19 41 149 10 18 8 16
426 330 251 156 40 51 27 70 (29 (30
± ± ± ± ± ± ± ± ± ±
102 47 45 35 8 11 6 12 4)'
272 259 174 81 35 40 45 107 10
4)'
17
8 14
4 0.5 1.4 0.3
(116 (26 93 8.5 14.4 19.2
± 14)t ± 8); ± 14 ± 0.8 ± 2.2 ± 3.3
± ± ± ± ± ± ± ± ± ± ±
39 41 36 13 6 9 11 28 1
P Value (0) v (A)
.05
(0)
v (B)
.05 .05
.02
6
1
c±:
111 ± 26 ± 122 ± 7.5 ± 14.0 ± 14.0 ±
8 2 14 1.0 2.6 1.3
.01 .05
.05
.001
.001
NOTE. Values are mean ± SEM. There were no significant differences between lipoprotein glomerulopathy cases (D) and dysbetalipoproteinemia (C). , Values are shown as HDL (HDL2 ± HDL3)' t Calculation from 5 cases. ; Calculation from 3 cases.
556
OIKAWA ET AL
E4
I,
E4
-
E3 ,
~~.
E3 . , E2 ~ E1
•••••-• ...."
Cs
2/2
2/3
Dys-f3-lipoproteinemia
2/3
2/3
2/3
Fig 1. Polyacrylamide gel electrophoresis of apolipoprotein VLDL. Dysbetalipoproteinemia (apo E3 deficiency) is shown in the left two lanes. Each lipoprotein glomerulopathy case (cases 1 to 5) had apo E3 •
4/4
4 5 3 Lipoprotein glomerulopathy 1
2
Plasma apo E level showed a significant correlation with VLDL-Ch concentration in both the control groups with nephrotic syndrome and type lIb hyperlipoproteinemia (P < 0.01, Fig 2). However, in the lipoprotein glomerulopathy patients, there was a discrepancy between apo E and VLDL-Ch levels, as seen in dysbetalipoproteinemia.
The apo E level showed a positive correlation with the VLDL-Ch concentration in control subjects, but not in the case of lipoprotein glomerulopathy that resembled familial type III hyperlipoproteinemia (dysbetalipoproteinemia) caused by apo E3 deficiency.21 In dysbetalipoprotein-
DISCUSSION
In the present study, patients who had the specific renal glomerular lesion, "lipoprotein thrombi,,,4-7 were found to have a characteristic lipoprotein composition and plasma apolipoprotein pattern. Hyperlipoproteinemia and plasma apolipoprotein profiles in chronic renal diseases have been reported. 16 Their characteristics are increased VLDL-TG, LDL-TG, and HDL-TG, and decreased HDL-Ch. When creatinine clearance is decreased to less than 1.17 mL/s (70 mLI min), apo C-I1 and C-III are increased. In the present study, TG levels oflipoprotein glomerulopathy patients whose creatinine clearance was normal were higher than levels of control nephrotic patients. The lipoprotein profile associated with lipoprotein glomerulopathy was phenotype III hyperlipoproteinemia. This phenotype is not unique for renal disease 17 and develops secondarily in some metabolic disorders (eg, diabetes mellitus,18,19 hypothyroidism 20). Some control subjects (nephrotic patients without glomerular lipoprotein thrombi) had type III hyperlipoproteinemia diagnosed by chemical indices; however, their plasma apo E concentration was lower than that associated with lipoprotein glomerulopathy.
0
~~
_ Isoi
:!!
'"
Ji,
130
]
110
~
'?
'-'3 >
0
•
150 140
e
120
• •
90 80 60 50
30
o·
.
/ 1. ~.
011\
03
.
j.;{'..~.
/
, 0,
4
•
•
•
o
•
70
•
• o·
•
100
40
0
0
5
•
'i
6
7
"
•
o
• "
,
•
o
ii
,
,
,
II,i
9 10 11 12 13 14 15 16 17 18 19 26 Apo E
1"II/dR)
Fig 2. Relationship between VLDL cholesterol and the apo E concentration in a nephrotic control group, familial type III hyperlipoproteinemia, and lipoprotein glomerulopathy patients. Single univariate correlation (r = 0.4540) in 44 cases including nephrotic syndrome (0, n = 6) and type lib hyperlipoproteinemia (e, n = 38) was significant (P < 0.01). Lipoprotein glomerulopathy patients (.... , n = 6) digressed from this regression line and closely resembled dysbetalipoproteinemia (0, n = 7). uThis case of dysbetalipoproteinemia showed a high level of VLDL-Ch (280 mg/dL) and apo E (35.7 mg/dL). *Two patients in this group were diagnosed as having type III hyperlipoproteinemia according to the chemical indices of lipoprotein lipids profile.
557
APOLIPOPROTEIN IN LIPOPROTEIN GLOMERULOPATHY
emia, a deficiency of apo E or isoprotein E3, as well as the presence of apolipoprotein E2 variants, has been reported. 21 -24 No evidence suggesting dysbetalipoproteinemia was found by analyzing the apo E isoform pattern among lipoprotein glomerulopathy cases. The isoform pattern, which occurred E2/3 in five cases and E4/4 in one case, was rare,25-28 and Davignon et al reported that the E2/3 occurred in only 7.5% to 15.7% and E4/4 in only 0.4% to 5.8% ofcases. 29 Different clinical profiles in lipoprotein glomerulopathy and familial type III hyperlipoproteinemia were found . Familial type III hyperlipoproteinemia is associated closely with atherosclerosis. 3D No patient in the present study had a history of coronary heart disease or peripheral atherosclerosis, and aU had normal electrocardiograms at the time of the study. Concerning the histological and functional change in lipoprotein glomerulopathy, hyperlipoproteinemia might modify the lesion and compromise glomerular function. 31 ,32 However, these glomerular changes have not been reported in familial hypercholesterolemia caused by LDL
receptor dysfunction,33 familial type III hyperlipoproteinemia,33 or combined hyperlipoproteinemia. 34 Apo E hyperlipoproteinemia is common in chylomicronemia, which is frequently associated with pancreatitis and diabetes mellitus,35 but not with renal lesions. Apo E in lipoprotein glomerulopathy was markedly elevated, although renal function and proteinuria were not different from control nephrotic patients. These findings suggested that apo E hyperlipoproteinemia in this rare renal disease was associated with apo E isoform and lipoprotein metabolic derangement of the disease. It is likely that hyperlipoproteinemia coupled with apo E hyperlipoproteinemia decreases the renal function and causes renal pathology. However, further studies are needed on the relationship between apo E hyperlipoproteinemia and the formation of lipoprotein thrombi. ACKNOWLEDGMENT We are grateful to Drs Y oshihito Hara, Toshikatsu Shibata, Yasushi Koitabashi, Takashi Ishida, Tetsuya Mitarai, Yuzo Watanabe, and Hitoshi Terasaki for supplying plasma from their patients with lipoprotein glomerulopathy, and to Professor Hiroshi Sakaguchi for his advice.
REFERENCES I. Faraggiana T, Churg J: Renal lipidosis: A review. Hum Pathol 18:661-679, 1987 2. Gjone E: Familial lecithin:cholesterol acyltransferase deficiency-A new metabolic disease with renal involvement. Adv Nephrol \0:167-185, 1981 3. Sakaguchi H, Dachs S, Grishman E, et al: Hepatic glomerulosclerosis: An electron microscopic study of renal biopsies in liver disease. Lab Invest 14:533-545, 1965 4. Watanabe Y, Ozaki I, Yoshida F, et al: A case of nephrotic syndrome with glomerular lipoprotein deposition with capillary ballooning and mesangiolysis. Nephron 51 :265-270, 1989 5. Saito T, Sato H, Kudo K, et al: Lipoprotein g1omerulopathy: Glomerular lipoprotein thrombi in a patient with hyperlipoproteinemia. Am J Kidney Dis 13:148-153, 1989 6. Koitabashi Y, Ikoma M, Miyahira T, et al: Long-term follow-up ofa paediatric case oflipoprotein g1omerulopathy. Paediatr NephroI4:122-128, 1990 7. Shibata T, Kaneko N, Hara Y, et al: A case oflipoprotein g1omerulopathy. Light and electron microscopic observations of the glomerulus. Acta Pathol Jpn 40:448-457, 1990 8. Hatch Fr, Lees RS: Practical methods for lipoprotein analysis. Adv Lipid Res 6: 1-68, 1986 9. Goto Y, Akanuma Y, Harano Y, et al: Determination by the SRID method of normal values of serum apolipoproteins (A-I, A-II, B,C-I1, C-III and E) in normolipidemic healthy Japanese subjects. J C1in Biochem Nutr 1:73-88, 1986 10. Sano R, Abe R, Oikawa S, et al: ApoE-2/E-3 ratio of very low density lipoprotein in diabetes mellitus. Tohoku J Exp Med 146:131-136, 1985
II . Kashyap ML, Hynd BA, Robinson K, et al: Abnormal preponderance of sialilated apolipoprotein C-III in triglyceride rich lipoproteins in type V hyperlipoproteinemia. Metabolism 30:111-118,1981 12. Menzel HJ, Utermann G: Apolipoprotein E phenotyping from serum by Western blotting. Electrophoresis 7: 492-495, 1986 13. Havekes LM, de Kniff P, Beisiegel U, et al: A rapid micromethod for apolipoprotein E phenotyping directly in serum. J Lipid Res 28:455-463, 1987 14. Hazzard WR, Porte D Jr, Bierman EL: Abnormallipid composition of very low density lipoproteins in diagnosis of braod-beta disease (type III hyperIipoproteinemia). Metabolism 21 :1009-1019, 1972 15. Fredrickson OS, Morganroth J, Levy RI: Type III hyperIipoproteinemia: An analysis of two contemporary definitions. Ann Intern Med 82:150-157, 1975 16. Grutzmacher P, Marz W, Peschke B, et al: Lipoproteins and apolipoproteins during the progression of chronic renal disease. Nephron 50:\03-111, 1988 17. Gherardi E, Rota E, Calandra S, et al: Relationship among the concentration of serum lipoproteins and changes in their chemical composition in patients with untreated nephrotic syndrome. Eur J Clin Invest 7:563-570, 1977 18. Chance GW, Albutt EC, Edkins SM: Serum lipid and lipoproteins in untreated diabetic children. Lancet I: 11261128, 1969 19. Reaven GM, Greenfield MS: Diabetic hypertriglyceridemia. Evidence for three clinical syndromes. Diabetes 30: 66-75, 1981 (suppI2)
558 20. Hazzard WR, Biennan EL: Aggravation of broad f3 disease (type 3 hyperiipoproteinemia) by hypothyroidism. Arch Intern Med 130:822-828, 1972 21. Utennann G, Albrecht G, Steinmetz A: Polymorphism of apolipoprotein E and occurence of dysbetalipoproteinemia in man. Nature 269:604-607, 1977 22. Ghiselli G, Schaefer EJ, Gascon P, et al: Type III hyperiipoproteinemia associated with apolipoprotein E deficiency. Science 214:1239-1241, 1981 23. Utennann G, Jaeschke M, Menzel HJ: Familial hyperiipoproteinemia type III: Deficiency of a specific apolipoprotein (apo E-III) in the very-low-density lipoproteins. FEBS Lett 56:352-355, 1975 24. Rall SC, Jr, Weisgraber KH, Innerarity TL, et al: Identification of a new structural variant of human apolipoprotein E, E2 (Lysl46 -+ Gin), in a type III hyperiipoproteinemic subject with the E3/2 phenotype. J Clin Invest 72:1288-1297, 1983 25. Sano R, Abe R, Oikawa S, et al: Apo1ipoprotein E phenotypes ofnonno and hyperiipoproteinemia in Japanese. Tohoku J Exp Med 154:297-303, 1983 26. Sing CF, Davignon J: Role of the apolipoprotein E polymorphism in detennining nonnal plasma lipid and lipoprotein variation. Am J Hum Genet 37:268-285, 1985 27. Eto M, Watanabe K, Ishii K: A rapid flat gel isoelectric focussing method for the detennination of apolipoprotein E
OIKAWA ET AL
phenotypes and its application. Clin Chim Acta 149:21-28, 1985 28. Yamamura T, Yamamoto A, Hiramori K, et al: A new isofonn of apolipoprotein E-apo E-5-associated with hyperlipidemia and atherosclerosis. Atherosclerosis 50: 159-172, 1984 29. Davignon J, Gregg RE, Sing CF: Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis 8: 1-21, 1988 30. Morganroth J, Levy RI, Fredrickson DS: The biochemical, clinical and genetic features of type III hyperlipoproteinemia. Ann Intern Med 82:158-174,1975 31. Moorhead JF, Chan MK, Nahas M, et al: Lipid nephrotoxicity in chronic progressive glomerular and tubulo-interstitial disease. Lancet 2: 1309-1311, 1982 32. Diamond JR, Karnovsky MJ: Exacerbation of chronic aminonucleoside nephrosis by dietary cholesterol supplementation. Kidney Int 32:671-678, 1987 33. Brown MS, Goldstein JL: Familial hypercholesterolemia: Genetic, biochemical and pathophysiologic considerations. Adv Intern Med 20:273-296, 1975 34. Grundy SM, Chait A, Brunzell JD: Familial combined hyperiipidemia workshop. Arteriosclerosis 7:203-207, 1987 35. Havel RJ: Pathogenesis, differentiation, and management ofhypertriglyceridemia. Adv Intern Med 15:117-154, 1969