- Email: [email protected]
Comparison of selectivity of LDL removal by double filtration and dextran-sulfate cellulose column plasmapheresis, and changes of subfractionated plasma lipoproteins after plasmapheresis in heterozygous familial hypercholesterolemia
Comparison of Selectivity’of LDL Removal by Double Filtration and Dextran-Sulfate Cellulose Column Plasmapberesis, and Changes of Subfractionated Plasma Lipoproteins After Plasmapheresis in Heterozygous Familial Hypercholesterolemia Yasuhiko
Homma,
Yoshikazu
Mikami,
Hiromitsu
Tamachi,
Noriaki Nakaya,
Haruo Nakamura,
and Yuichiro Goto
The possibility of selective removal of low density lipoprotein (LDL) by double filtration (DF) and dextran-sulfate cellulose (DSC) column plasmapheresis in hypercholesterolemia and the acute recovery process of the subfractionated plasma lipoproteins after plasmapheresis in heterorygous familial hypercholesterolemia were investigated. Sixty-six percent of the LDL cholesterol and 42% of the HDL cholesterol were removed by 2.5 L DF plasmapheresis with the second filters having average pore diameters of 30 nm and 40 nm. Fifty-nine percent of the LDL cholesterol was removed by 2.5 L DSC column plasmapheresis, while HDL cholesterol did not change. Therefore, DSC column phasmapheresis could remove LDL much more specifically than DF plasmapheresis. VLDL increased rapidly and reached the preplasmapheresis level within four days after plasmapheresis. IDL returned to the preplasmapheresis level in 2 weeks. The LDL, level was approximately 80% of the preplasmapheresis level on the 14th day. LDL, reached the peak at the seventh day. HDL2 and HDL, moved in the same manner and reached the peak on the seventh day after DF plasmapheresis. o 1987 by Grune & Stratton, Inc.
T
HE INCIDENCE of familial hypercholesterolemia (FH) has been reported to be one per 1,OOO,OOOpopulation for the homozygous and one per 500 population for the heterozygous types, respectively.lk2 The most serious complication of FH is coronary heart disease (CHD). FH is highly resistant to medical treatment. Plasmapheresis, portacaval shunt, and partial ileal bypass have been applied to the treatment of the homozygous and heterozygous forms of FH.3-‘5 Of these, plasmapheresis has been used most commonly in therapy for both the homozygous and heterozygous forms of FH. The disappearance of xanthoma, the decrease of angina1 attacks, and improvement of coronary stenosis by coronary angiography (CAG) have been reported as a result of relatively long-term phasmapheresis.5V7,” It is beneficial for FH patients if low density lipoprotein (LDL) is selectively removed. However, simple plasma exchange also removes high density lipoprotein (HDL), which is believed to be an antiatherogenic lipoprotein. Thus, plasmapheresis by the double filtration (DF) with a second filter of the proper pore diameters or by LDL binding column has been in the process of development for the selective removal of apoprotein B containing lipoproteins.639V’0 The diameter of HDL is around 10 nm and that of the apoprotein B containing lipoproteins is greater than 20 nm.‘63’7 Therefore, it is theoretically possible to remove apoprotein B containing lipoproteins by DF plasmapheresis with a second filter having an average pore diameter of 20 nm. Dextransulfate cellulose (DSC) can selectively bind apoprotein B containing lipoproteins.” The first aim of this study was to determine the efficacy of the second membrane filter of three different pore sizes for DF plasmapheresis and of the DSC column for the selective removal of apoprotein B containing lipoproteins. Plasmapheresis has the advantage of facilitating study on the metabolic interrelationship between lipoproteins without the use of radioisotopes by looking at the recovery process of plasma lipoproteins after plasmapheresis. There have been no reports on the metabolic interrelationship between subfractionated plasma lipoproteins in FH. The second aim of this study was to estimate the changes in subfractionated
Metabolism, Vol 36, No 5 (May). 1987: pp 419-425
plasma lipoproteins in heterozygous FH after plasmapheresis in order to speculate the metabolic interrelationship between them. MATERIALS
AND METHODS
Subjects Six heterozygous FH patients and one case of nonfamilial type IIb hyperlipidemia (Table 1) were subjected to plasmapheresis. Details concerning patient 1 have already been published.” Diagnosis of FH was made by plasma lipoprotein profile, Achilles tendon thickness, and monogenic inheritance from family analysis. The coronary stenosis of patients 3 and 5 was minimal, but any kind of drugs could not decrease their plasma total cholesterol levels less than 300 mg/dL. Therefore, plasmapheresis was prescribed for them to prevent the progress of coronary stenosis. Patients 5 and 6 were brothers. Patients 2,4,6, and 7 had severe coronary stenosis and had been receiving periodic plasmapheresis for the improvement of coronary stenosis. Patient 7 was nonfamilial hypercholesterolemic. She had frequent angina1 attacks even though her aortocoronary grafts were patent. Drug therapy was not satisfactory. Therefore, plasmapheresis was prescribed for her. For the investigation of the recovery process of plasma lipoproteins after plasmapheresis, no drugs had been administered for more than two months before plasmapheresis to patients 1,2, 3, 5, and 6. They admitted to Tokai University Hospital and were kept on a diet of 1,600 kcal/d (protein 20%, carbohydrate 55%. fat 25%, cholesterol 500 mg/d, P/S ratio: I .2) without medication. Procedure
of Plasmapheresis
In patient 1, the venous blood was introduced from the femoral vein to a cell separator, IBM-2997. The plasma was separated from the blood cells by continuous centrifugation at 1,600 rpm. The
From the Department of Internal Medicine, Tokai University School of Medicine, and the Department of Internal Medicine, National Defense Medical College, Japan. Address reprint requests to Yasuhiko Homma. MD, Department of Internal Medicine, Tokai University School of Medicine, Boseidai, Isehara-shi, 259-l I, Japan. o 1987 by Crune & Stratton, Inc. 0026-0495/87/3605-0003$03.00/0
419
420
HOMMA ET AL
Table 1. Clinical and Laboratory
Data of Plasmaoheresis-Treated
Patients Normal Value5 of Plasma
Subject Nos. 1
2
3
Lipids and
4’
5
6
7’
Apoproteinst
M
M
F
M
M
M
F
Age (vr)
30
32
55
61
41
38
51
BL (cm)
167
168
148
161
161
165
155
BW (kg)
65
50
43
59
52
60
54
488
337
351
290
345
391
270
176 + 26
146
117
111
59
245
187
197
87 f 29
9
9
13
2
16
18
20
31
13
13
4
15
9
11
LDL,-C (mg/dL)
324
160
171
128
128
189
109
LDL,-C (mg/dL)
40
47
35
47
72
77
52
HDL,-C (mg/dL)
10
16
17
8
19
22
11
HDL,-C (mg/dL)
8
17
18
11
21
18
8
Al
82
115
107
76
172
120
73
137 + 25
All
15
26
30
24
45
38
19
31 -c6
174
165
180
226
247
195
TC (mg/dL) TG (mg/dL) VLDL-C (mg/dL) IDL-C (mg/dL)
925 5+2 79 * 33 34 + 8 24 + 10 33 + 8
Apoprotein (mg/dL)
B
374
CII Clll E Diagnosis
7.0
3.6
12.7
10.1
8.0
6.8
FH
FH
AP
Ml
79 c 20
2.4
9.3
7.7
5.4
18.0
13.0
7.0
7.4 f 2.9
6.2
4.6
7.7
12.0
5.9
4.0 r 1.2
3.5
FH
8.4
FH
FH
FH
Ml
AP
MI
2.7
3.4 * 1.2
MI ACbypass
Abbreviations: FH, familial hypercholesterolemia; Ml, myocardial infarction: AP, angina pectoris; AC, aortocoronary bypass. *Analysis were made on the day of plasmapheresis, a day after probucol withdrawal. tValues are given as mean + SD. Data from Homma et al” and Goto et al?
separated plasma was discarded into a plastic bag. Blood cells were mixed with plasmanate and returned to the patient through the antecubital vein. ACD-A was used as an anticoagulant. Two and a half liters of plasma were removed at the rate of 1,000 mL/h. For DF plasmapheresis, the venous blood was introduced from the femoral vein at a speed of 100 mL whole blood/min to a membrane cell separator, KM 8500 (Kurare, Japan). The plasma that was separated from the blood cells with the first membrane filter (PVA filter SB-I, Kurare) was introduced to the second membrane filter
Table 2.
Changes in Plasma Lipids and Apoproteins
Immediately
After 2,500-ml
Plasmapheresis P Value
Double Filtratmn
Plasma Exchange
Number of treatments
(Eva1 HD 3A, 4A, Kurare or AC 1760, Asahi Medical, Japan). Plasma components that could not pass the second filter were discarded in a plastic bag. The plasma that passed through the second filter was returned to the patients through the antecubital vein. The same volume of 5% human albumin as the discarded plasma was added to the plasma returning to the patients. The approximate volume of the discarded plasma was 500 mL. The flow rate of the plasmapheresis was 1,000 mL/h at the second filter, and 2% L plasma was treated at each plasmapheresis. Average pore
DoubleFiltration 40 nm’
30 nm*
4
5
L
DSC Columnt
v DSC ColumnS
9
Total cholesterol
44.7
37.0
+ 9.9
36.5
” 7.5
44.9
* 5.9
VLDL-cholesterol
30.0
25.0
? 23.7
22.1
+ 2.7
16.6 * 13.4
IDL-cholesterol
24.7
28.6
+ 6.0
33.7
+ 6.0
27.8
+ 10.7
LDL-cholesterol
37.0
35.4
+ 9.7
30.8
t
1.6
40.8
+ 6.2
‘NS
HDL-cholesterol
30.9
55.2
+ 14.5
60.8
+ 5.7
97.6
i 3.0
coo
NS NS NS
Apoproteins§ A-l
37.9
56.9
+ 11.4
71.7
2 9.9
86.9
+ 6.5
<.oo 1
A-II
57.7
60.1
+ 5.9
67.2
* 12.6
88.1
* 11.0
1.00 1
B
26.3
34.8
f 8.7
38.2
+ 12.7
37.5
* 6.9
NS
C-II
56.0
43.4
* 14.1
44.1
+ 13.3
49.1
+ 13.1
NS
C-III
50.0
46.0
+ 7.7
45.3
+ 11.8
48.9
zk 12.4
NS
E
50.0
25.0
+ 23.7
35.2
? 10.0
39.3
k 16.0
NS
*Average pore diameter of the second filter membrane. tMean
t SD; values are expressed as the percentage of preplasmapheresis levels.
$Paired r-test. SApoproteins were measured at one plasma exchange in patient 1.
421
LDL REMOVAL BY DF AND DCS PLASMAPHERESIS
diameters of the second filter were 20 nm (Eva1 HF 3A), 30 nm (Eva1 HF 4A), and 40 nm (AC 1760). Heparin was used as an anticoagulant. The sterile DSC column (LA-01, 400 mL pack) for DSC column plasmapheresis was supplied from Kaneka, Japan. The second membrane of the DF plasmapheresis mentioned above was substituted for by a DSC column. The procedure of DSC column plasmapheresis was the same as the DF method. Albumin supply was unnecessary. Plasmapheresis was performed in the fasting state. DF plasmapheresis with the second filter having average pore diameters of 30 nm and 40 nm and DSC column plasmapheresis were performed alternatively to the same patients numbered 2, 3,4, 5. 6. and 7.
Laboratory Procedure Blood sampling was performed at every 500 mL plasmapheresis. Fasting blood was collected for up to 2 weeks. LDL was divided into two fractions at d = 1.045. Very low density lipoprotein (VLDL), intermediate density lipoprotein (IDL), LDL, (1.019 < d < 1.045), LDL, (1.045 < d < 1.063) HDL, (1.063 < d < 1.125) and HDL, I .21) were harvested ultracentrifugally by the method (I.125
of apoprotein B were reduced in the plasma by DSC column plasmapheresis. There were no significant differences in the removal rates of LDL-C, apoprotein B, CII, CIII, and E between DF plasmapheresis, with the second filter having an average pore diameter of 30 nm and 40 nm and DSC column plasmapheresis. About 90% of plasma apoprotein AI and AI1 was restored by DSC column plasmapheresis. DSC column bound lipoproteins were eluted by 1 .O mol/L NaCl solution. Apoproteins B, CII, CIII, and E were detected but apoprotein AI and AI1 were not detectable rn that eluate. Total cholesterol, LDL-C, apoprotein B, CD, CIII, and E decreased almost linearly throughout DSC plasmapheresis as well as DF plasmapheresis. Plasma apoprotein AI and AI1 decreased slightly at an early stage and stayed constant throughout DSC column plasmapheresis. Trends of the apoprotein AI and AI1 changes were similar to the changes in plasma albumin and y-globulin. Plasma y-globulin decreased markedly by DF plasmapheresis, but the intravenous y-globulin supply was unnecessary for 2.5 L plasmapheresis. The decrease of plasma albumin and y-globulin was minimal by DSC column plasmapheresis. There were no significant side effects during the plasmapheresis by either method. The recovery process of subfractionated lipoproteins after DF plasmapheresis was studied in patients 1, 2, 3, 5, and 6. All were inpatieints and kept on a 1.600-kcal diet without
Total
cholesterol
Plasmapheresis (PP) 500-
RESULTS
in plasma lipoprotein cholesterol and apoproteins after 2.5 L plasmapheresis are shown in Table 2. Plasma exchange was performed to patient 1. Sixty-three percent of LDL-C and 69% of HDL-C were removed by plasma exchange. The removal rate of LDL-C and HDL-C was almost the same between plasma exchange and DF plasmapheresis with the second filter having an average pore diameter of 20 nm. Therefore, we stopped using this second filter membrane and used the second filter having an average pore diameter of 30 nm or 40 nm. Sixty-three percent of total cholesterol (TC) and 66% of LDL-C were removed on average by DF plasmapheresis with the second membrane filter having an average pore diameter of 30 nm and 40 nm. On that occasion, the average removal rate of HDL-C was 42%, which was significantly lower than that of LDL-C (P c .Ol, paired t-test). The removal rate of apoprotein AI and AI1 was significantly lower than that of apoprotein B (P c .Ol, paired t-test) by DF plasmapheresis with the second filter having an average pore diameter of 30 nm and 40 nm. The removal rate of LDL-C with the second filter of 40 nm diameter was same as that with the second filter of 30 nm diameter. The removal rate of HDL-C, and apoproteins AI and AI1 with the second filter of 40 nm diameter was slightly lower than that with the second filter of 30 nm diameter. Fifty-five percent of TC and 59% of LDL-C were removed on average by DSC column plasmapheresis, while the plasma HDL-C level did not decrease at all. Thirteen percent of apoprotein AI and 12% of apoprotein AI1 and 63%
400.
Changes
-
300 200. 100. = B 1
O-
LDL-cholesterol .fi E f e s ; % a
3001 200IOOO-
40. 20.
HDL- cholesterol
9’
60.
kA+=====! J-1
01
Eeforsl
2
3 4
5 6
7 6
9 IO II 12 13 14
Days Fig 1. The plasma levels (mean t SD) of total cholesterol. LDL cholesterol, and HDL cholesterol after DF plasmapheresis (PP). Five FH patients (1,2. 3, 5, and 6) were examined.
422
HOMMA ET Al
medication. Medication was stopped at least two months before plasmapheresis. The serial changes in the levels of plasma TC, LDL-C, and HDL-C after plasmapheresis are shown in Fig 1. The plasma TC and LDL-C levels were 80.3 t 9.7% and 79.3 t 12.4% of the preplasmapheresis levels on the 14th day after plasmapheresis, respectively. The plasma HDL-C level reached the preplasmapheresis level within a week. Plasma apoprotein levels up to 2 weeks after plasmapheresis are illustrated in Fig 2. Apoprotein B reached the preplasmapheresis level at the tenth day after plasmapheresis, while apoprotein AI, AII, CII, CID, and E did so at the seventh day. The plasma levels of cholesterol in VLDL, IDL, LDL,, LDL?, HDLI, and HDLJ up to two weeks after plasmapheresis are shown in Fig 3. VLDL-C returned very rapidly to the value before plasmapheresis, ie, within four days, while IDL-C needed 2 weeks to do so. The plasma LDL-C level increased gradually, up to 83.7 + 18.9% of the preplasmapheresis level on the 14th day after plasmapheresis. LDL*-C returned to the value before plasmapheresis at the seventh day. HDL,-C needed a week to return to the preplasmapheresis level, and HDLS-C reached the peak on the seventh day.
Plasmqlhaw*s (PP)
Y
@
PP
IDL-C
PP
LDLI-C
2
I4 g g s ‘c: e E
200.
f
T
i
100. 0
f
IOO-
iE
50.
PP
LDL2-C
PP
HDLz -C
PP
HDL3-C
L O3
+
2 T I
I
5
Plasmaphmsis[PP) T
IO
IO 2 0!
APO-AI I
VLDL-C 20.
I!I
IO 1 6efoml
30
2 3 4
5 6
7 6 9 KJ II 121314
Days
20 IO 40 1
The cholesterol levels (mean f SD) in VLDL. IDL. LDL,. Fig 3. LDL,, HDL,, and HDL, after DF plasmapheresis (PPl. Five FH patients (1.2.3. 6. and 6) were examined.
400 300. f
200.
1
‘0:
5 F f
IO
f
5
E I 0
di 20
APO-C II
APO-C III
PP
T I I
1 I
IO
l
0I IO
Ape-E
PP
f 50 I S&ml
2 3 4
5 6
7
6 9 IO II I2 I3 14
Days Fig 2. Plasma apoprotein levels (mean k SD) after DF plasmapheresis (PP). Five FH patients (1,2,3, 6, and 6) were examined.
Phospholipids of the subfractionated lipoproteins were also measured in patients 2, 3, 5, and 6 (Fig 4). The changing patterns of phospholipids were very similar to those of cholesterol in all lipoprotein fractions. LDL,-PL reached the peak much earlier than LDL,-PL as LDL*-C did so. The changes in apoprotein B of VLDL, IDL, LDL,, and LDLz and in apoprotein AI in HDL2 and HDL, are demonstrated in Fig 5. Apoprotein AI and B in the subfractionated lipoproteins were measured in patients 1, 2, 3, 5, and 6. The changes in apoprotein B were very similar to those in cholesterol and phospholipids in VLDL, IDL, LDL,, and LDL*. Average LDL,-apo B was 83.7% of the preplasmapheresis level on the 14th day after plasmapheresis. LDL2apo B reached the peak at the seventh day and was 84.7% of the preplasmapheresis level on that day. Plasma HDL,-apo AI was extremely low in those patients. It reached the peak at seventh day after DF plasmapheresis but the peak was slightly lower than the preplasmapheresis level. Plasma HDL,-apo AI was also low in those patients and returned to the preplasmapheresis level on the seventh day. The changing patterns of apoprotein AI in HDL, and HDLS were similar to those of cholesterol and phospholipids.
423
LDL REMOVAL BY DF AND DCS PLASMAPHERESIS
VLDLPiOSmODhCWOSiS (PP)
IDL-
PP
1-
VLDL-apoB
PL
Pl0smOdlNasislPP)
PL
IDL-apoB PP
PC -&
LDL2-PL
PP
HDL2-PL
PP
HDLe-opo
Al
I 1 HDLs-
PL
60
PP z
HDLs-apoAl T
40
-i
20
Beforel 2 3 4
5 6 7
8 9 1011 121314
Days Befod
2 3 4
5 6
7
6
9 1011 121314
Days Fig 4. The phospholipids levels (mean + SD) in VLDL, IDL, LDL,. LDL,, HDL,, and HDL, after DF plasmapheresis (PP). Four FH patients (2.. 3. 5, and 6) were examined.
DISCUSSION
The possibility of selective removal of apoprotein B containing lipoproteins by DF plasmapheresis and DSC plasmapheresis and the acute recovery process of plasma lipoproteins after plasmapheresis were investigated in hypercholesterolemia. Theoretically, VLDL, IDL, and LDL can be selectively removed by DF plasmapheresis if the second filter has a diameter of 20 nm. This is because the diameter of HDL is around 10 nm and the diameters of other apoprotein B containing lipoproteins are greater than 20 nm,15J6 However, it was impossible to remove apoprotein B containing lipoproteins selectively by DF plasmapheresis even though a second filter of large pore diameter was used. The removal rates of HDL-C, apoprotein AI, and AI1 were significantly lower than those of LDL-C and apoprotein B when a second filter having an average pore size of 30 nm or 40 nm was used, but the selectivity in the removal of apoprotein B containing lipoproteins was not satisfactory by DF plasmapheresis. This was probably due to the adhesion of plasma components to the second filter. The reduction of the flow rate, the use of a wide membrane, and the use of the second filter with a large pore size must be considered in an effort to increase the selective removal of apoprotein B
Fig 5. The levels (mean + SD) of apoprotein 6 and apoprotein Al in subfraotionated lipoproteins after DF plasmapheresis (PP). Five FH patients (1.. 2. 3. 5. and 6) were examined.
containing lipoproteins by DF plasmapheresis. The decrease of plasma HDL-C, apoprotein AI, and AII seemed to be smaller by DF plasmapheresis, with the second filter having an average pore diameter of 40 nm than with that having an average pore diameter of 30 nm. Therefore, the use of the second filter with a larger pore size must be tried first for the improvement of the selectivity of the removal of apoprotein B containing lipoproteins by DF plasmapheresis. The selectivity in the removal of apoprotein B containing lipoproteins by DSC column plasmapheresis was excellent. When the DSC column was eluted by 1.0 moi/L NaCl solution, bound lipoproteins could be detached. Apoprotein CD, CID, and E were also eluted as well as apoprotein B. Therefore, VLDL and IDL were also bound to the DSC column. Plasma LDL-C and apoprotein B decreased almost linearly up to 2.5 L plasma treatment by the plasmaphereseis with 400 mL DSC column. The plasma TC levels were satisfactorily low after 2.5 L DSC column plasmapheresis. The capacity of 400 mL DSC column was enough for the treatment of those patients. However, the replacement of DSC column might be necessary during plasmapheresis for the treatment of homozygous or large heterozygous FH patients. The combination therapy with the DSC column plasmapheresis and drugs was expected to normalize the plasma LDL levels of heterozygous FH.
424
HOMMA
There are three types of apoprotein B containing lipoproteins adsorbing materials that have been used clinically: heparin-agarose, anti-LDL antibody-sepharose, and DSC.6*9.“’The removal of apoprotein B containing lipoproteins by heparin-agarose was made by batch chromatography.’ The removal capacity of apoprotein B containing lipoproteins was almost the same between heparin-agarose and DSC.” It must be possible to produce a same typed heparin-agarose column as a DSC column, but the cost of a heparin-agarose column is probably much higher than that of a DSC column. A 400 mL column of DSC bound about 7 g of cholesterol while a 500 mL column of anti-LDL column bound 2 g of cholesterol.9~‘0 Two columns were used in rotation, and adsorption and desorption were performed simultaneously during a treatment by anti-LDL column plasmapheresis because of the relatively low binding capacity of the antibody column.’ The desorption of apoprotein B containing lipoproteins was unnecessary during a treatment by DSC column plasmapheresis. Therefore, the method of DSC column plasmapheresis is simpler that that of antiLDL plasmapheresis. We have never used a DSC column repeatedly but it is possible to use a DSC column repeatedly because the desorbing of apoprotein B containing lipoproteins is easily made by NaCl solution (NaCl concentration ~0.4 mol/L). Anti-LDL antibody and DSC have a possibility to cause allergic reactions, but such side effects have not been reported. We must carefully observe possible side effects of the long-term treatment by plasmapheresis. DSC column is factory-made and the cost of DSC column will be reduced if it will be used widely. We believe that DSC column plasmapheresis is the best method of plasmapheresis to remove the plasma apoprotein B containing lipoproteins specifically at this moment. The changes in cholesterol, phospholipids, and apoproteins were similar in every lipoprotein fraction after plasmapheresis. Therefore, the alterations in chemical compositions of lipoprotein molecules seemed to be very little during the recovery process of plasma lipoproteins after plasmapheresis. All VLDL is converted to LDL via IDL by delipidation by
ET AL
lipoprotein lipase and hepatic triglyceride lipase in normal subjects.16 However, a part of plasma LDL is directly secreted from the liver in FH.24325 VLDL is converted rapidly to IDL in normal subjects.26 Plasma VLDL increase was very rapid after plasmapheresis. The baseline values of plasma IDL was high and IDL increase was relatively slow in heterozygous FH patients. This finding may suggest that VLDL is only a part of the precursor of IDL or the conversion of IDL to LDL was very slow due to the low hepatic triglyceride lipase acitivity in heterozygous FH.*’ Teng et al recently reported the good precursor-product relationships between VLDL and IDL, and between IDL and light LDL in the limited cases of heterozygous FH.25 Therefore, the second possibility might be the cause of slow IDL increase in heterozygous FH. Apoprotein B reached the plateau much earlier in LDL, than in LDL,, as well as cholesterol and phospholipids. The discrepancy of recovery speed between LDL, and LDL, probably indicates that LDL, and LDL2 have some different metabolism in heterozygous FH. Isotope kinetic study must be undertaken for further study. Teng et al demonstrated the direct secretion of heavy LDL to the plasma in heterozygous FH from the isotope kinetic study.25 Plasma HDL-C levels are normal or slightly low in heterozygous FH.’ These heterozygous FH patients had the plasma levels of low HDL2-C and very low HDL,-C. HDL, and HDL, increased in the same manner, and they reached the preplasmapheresis level in a week. HDL, is converted to HDL, in vitro.2s~29However, it is impossible to study the metabolic interrelationship between HDL, and HDL, in vivo only by the measurement of cholesterol, phospholipids, and apoprotein ALi which are transferred very rapidly between HDL2 and HDL,.‘55’6 Drug therapy is usually made for the treatment of FH between each plasmapheresis. It is important to look at the recovery process of plasma lipoproteins without medication at first as the basal data in order to select the most effective drug to normalize plasma LDL levels of FH by the combination therapy with plasmapheresis and drugs.
REFERENCES 1.
Goldstein JL, Brown MS: Familial
hypercholesterolemia,
in
Stanbury JB, Wyndgarden JB, Frederickson SD, et al (eds): The Metabolic Basis of Inherited Disease. New York, McGraw-Hill, 1982, pp 672-212 2. Mobuchi H, Tatami R, Haba T, et al: Homozygous familial hypercholesterolemia in Japan. Am J Med 65:290-297, 1978 3. Thompson GR, Lowenthal R, Myant NB: Plasma exchange in the management of homozygous familial hypercholesterolemia. Lancet 1:1208-1211, 1975 4. Simons LA, Gibson JC, lsbister JP, et al: The effects of plasma exchange on cholesterol metabolism. Atherosclerosis 31:195-204, 1978 5. Berger GMB, Miller JL, Bonnici F, et al: Continuous flow plasma exchange in the treatment of homozygous familial hypercholesterolemia. Am J Med 65:243-25 I, 1978 6. Lupien PH. Moorjani S, Lou M, et al: Removal of cholesterol from blood by affinity binding to heparin-agarose: Evaluation of treatment in homozygous familial hypercholesterolemia. Pediatr Res 14:113-l 17. 1980
7. Thompson GR, Myant NB, Kilpatrick D, et al: Assessment of long-term plasma exchange for familial hypercholesterolemia. Br Heart J 43:680-688, 1980 8. Stein EA, Blueck CJ, Wesselman A, et al: Repetitive intermittent flow plasma exchange in patients with severe hypercholesterolemia. Atherosclerosis 38:149-164, 1981 9. Stoffel W, Borberg H, Greve V: Application of specific extracorporeal removal of low density lipoprotein in familial hypercholesterolemia. Lancet 2:1005-1007, 1981 10. Yokoyama S, Hayashi R, Satani M, et al: Selective removal of low density lipoprotein by plasmapheresis in familial hypercholesterolemia. Arteriosclerosis 5:613-622, 1985 11. Saal SD, Parker TS, Gordon BR, et al: Removal of lowdensity lipoproteins in patients by extracorporeal immunoadsorption. Am J Med 80:583-589, 1986 12. Bilheimer DW, Goldstein JL, Grundy SM, et al: Reduction in cholesterol and low density lipoproteins synthesis after portacaval shunt in a patient with homozygous familial hypercholesterolemia. J Clin Invest 56:1420-1430, 1975
LDL REMOVAL
425
BY DF AND DCS PLASMAPHERESIS
13. Forman MB, Baker SD, Mieny homozygous familial hypercholesterolemia
CJ, et al: Treatment of with portacaval shunt.
Atheroscl41:349-361, 1982 14. Sterzl T, Chase HP, Ahrens EH, et al: Portacaval shunt in patients with familial hypercholesterolemia. Ann Surg 198:273-282, 1983 I?. Scott H W: lleal bypass in the control of hyperlipidemia and atherosclerosis. Arch Surg 113:62-70, 1978 Ih. Jackson RL, Morrisett JD, Gotto AM: Lipoproteins structure and metabolism. Physiol Rev 56:259-316. 1976 Ii. Eisenberg S: High density lipoprotein metabolism. J Lipid Res 25:1017-1058, 1984 18. Homma Y, Watanabe K, Tamachi H, et al: The change in plasma lipoproteins in a case of heterozygous familial hypercholesterolemia after plasmapheresis. Jpn J Med 23:205-210, 1984 19. Homma Y. Nakaya N, Nakamura H, et al: Increase in the density of lighter low density lipoprotein by hepatic triglyceride lipase. Artery 13:19-31. 1985 20. Goto Y. Akanuma Y. Harano Y, et al: Determination by the SRID method of normal values of serum apolipoproteins (A-l, A-II, B.C.II, C-III and E) in normolipidemic healthy Japanesesubjects. J Clin Biochem Nutr 1:73-88. 1986 21. Have1 RJ, Eder HA, Bragdon JH: The distribution and chemical composition of ultracentrifugally separated lipoproteins in human plasma. J Clin Invest 34:1346-1353, 1955 22. Autoanalyzer manual. New York, Technicon Autoanalyzer Inc, 1981
23. Takayama
M, ltoh S, Nagasaki
T, et al: A new enzymatic
method for determination of serum choline-containing lipids. Clin Chim Acta 79:93-98, 1977
phospho-
24. Janus ED, Nicoll A, Wootton R, et al: Quantitative studies of very low density lipoprotein: Conversion to low density lipoprotein in normal controls and primary hyperlipidemic states and the role of direct secretion of low density lipoprotein in heterozygous familial hypercholesterolemia. Eur J Clin Invest 10:149-l 59. 1980 25. Teng B, Sniderman hyperapobetalipoproteinemia. density lipoprotein with normal and 771663-672.
AD. Soutar AK, et al: Metabolic basis of Turnover of apolipoprotein B in low
and its precursors and subfractions compared familial hypercholesterolemia. J Clin Invest
1986
26. Reardon MF, Fidge NH, Nestel PJ: The catabolism of very low density lipoprotein B apoprotein in man. .I Clin Invest 61:850860. 1978 27. Nikkill EA, Huttunen JK, Ehnholm C: Low post-hepdrin plasma hepatic lipase activity in familial type IIa hyperlipoproteinemia. Ann Clin Res 8:63-67, 1976 28. Patsch JR. Gotto M, Olivecrona T, et al: Formation of high density lipoprotein-like particles during lipolysis of very low density lipoproteins in vitro. Proc Natl Acad Sci USA 75:45 19-4523. 1978 29. Grow TE. Fried M: Interchange of apoprotein components between the human plasma high density lipoprotein subclass HDL, and HDL, in vitro. J Biol Chem 253:8034-804 I, 1978
Homma,
Yoshikazu
Mikami,
Hiromitsu
Tamachi,
Noriaki Nakaya,
Haruo Nakamura,
and Yuichiro Goto
The possibility of selective removal of low density lipoprotein (LDL) by double filtration (DF) and dextran-sulfate cellulose (DSC) column plasmapheresis in hypercholesterolemia and the acute recovery process of the subfractionated plasma lipoproteins after plasmapheresis in heterorygous familial hypercholesterolemia were investigated. Sixty-six percent of the LDL cholesterol and 42% of the HDL cholesterol were removed by 2.5 L DF plasmapheresis with the second filters having average pore diameters of 30 nm and 40 nm. Fifty-nine percent of the LDL cholesterol was removed by 2.5 L DSC column plasmapheresis, while HDL cholesterol did not change. Therefore, DSC column phasmapheresis could remove LDL much more specifically than DF plasmapheresis. VLDL increased rapidly and reached the preplasmapheresis level within four days after plasmapheresis. IDL returned to the preplasmapheresis level in 2 weeks. The LDL, level was approximately 80% of the preplasmapheresis level on the 14th day. LDL, reached the peak at the seventh day. HDL2 and HDL, moved in the same manner and reached the peak on the seventh day after DF plasmapheresis. o 1987 by Grune & Stratton, Inc.
T
HE INCIDENCE of familial hypercholesterolemia (FH) has been reported to be one per 1,OOO,OOOpopulation for the homozygous and one per 500 population for the heterozygous types, respectively.lk2 The most serious complication of FH is coronary heart disease (CHD). FH is highly resistant to medical treatment. Plasmapheresis, portacaval shunt, and partial ileal bypass have been applied to the treatment of the homozygous and heterozygous forms of FH.3-‘5 Of these, plasmapheresis has been used most commonly in therapy for both the homozygous and heterozygous forms of FH. The disappearance of xanthoma, the decrease of angina1 attacks, and improvement of coronary stenosis by coronary angiography (CAG) have been reported as a result of relatively long-term phasmapheresis.5V7,” It is beneficial for FH patients if low density lipoprotein (LDL) is selectively removed. However, simple plasma exchange also removes high density lipoprotein (HDL), which is believed to be an antiatherogenic lipoprotein. Thus, plasmapheresis by the double filtration (DF) with a second filter of the proper pore diameters or by LDL binding column has been in the process of development for the selective removal of apoprotein B containing lipoproteins.639V’0 The diameter of HDL is around 10 nm and that of the apoprotein B containing lipoproteins is greater than 20 nm.‘63’7 Therefore, it is theoretically possible to remove apoprotein B containing lipoproteins by DF plasmapheresis with a second filter having an average pore diameter of 20 nm. Dextransulfate cellulose (DSC) can selectively bind apoprotein B containing lipoproteins.” The first aim of this study was to determine the efficacy of the second membrane filter of three different pore sizes for DF plasmapheresis and of the DSC column for the selective removal of apoprotein B containing lipoproteins. Plasmapheresis has the advantage of facilitating study on the metabolic interrelationship between lipoproteins without the use of radioisotopes by looking at the recovery process of plasma lipoproteins after plasmapheresis. There have been no reports on the metabolic interrelationship between subfractionated plasma lipoproteins in FH. The second aim of this study was to estimate the changes in subfractionated
Metabolism, Vol 36, No 5 (May). 1987: pp 419-425
plasma lipoproteins in heterozygous FH after plasmapheresis in order to speculate the metabolic interrelationship between them. MATERIALS
AND METHODS
Subjects Six heterozygous FH patients and one case of nonfamilial type IIb hyperlipidemia (Table 1) were subjected to plasmapheresis. Details concerning patient 1 have already been published.” Diagnosis of FH was made by plasma lipoprotein profile, Achilles tendon thickness, and monogenic inheritance from family analysis. The coronary stenosis of patients 3 and 5 was minimal, but any kind of drugs could not decrease their plasma total cholesterol levels less than 300 mg/dL. Therefore, plasmapheresis was prescribed for them to prevent the progress of coronary stenosis. Patients 5 and 6 were brothers. Patients 2,4,6, and 7 had severe coronary stenosis and had been receiving periodic plasmapheresis for the improvement of coronary stenosis. Patient 7 was nonfamilial hypercholesterolemic. She had frequent angina1 attacks even though her aortocoronary grafts were patent. Drug therapy was not satisfactory. Therefore, plasmapheresis was prescribed for her. For the investigation of the recovery process of plasma lipoproteins after plasmapheresis, no drugs had been administered for more than two months before plasmapheresis to patients 1,2, 3, 5, and 6. They admitted to Tokai University Hospital and were kept on a diet of 1,600 kcal/d (protein 20%, carbohydrate 55%. fat 25%, cholesterol 500 mg/d, P/S ratio: I .2) without medication. Procedure
of Plasmapheresis
In patient 1, the venous blood was introduced from the femoral vein to a cell separator, IBM-2997. The plasma was separated from the blood cells by continuous centrifugation at 1,600 rpm. The
From the Department of Internal Medicine, Tokai University School of Medicine, and the Department of Internal Medicine, National Defense Medical College, Japan. Address reprint requests to Yasuhiko Homma. MD, Department of Internal Medicine, Tokai University School of Medicine, Boseidai, Isehara-shi, 259-l I, Japan. o 1987 by Crune & Stratton, Inc. 0026-0495/87/3605-0003$03.00/0
419
420
HOMMA ET AL
Table 1. Clinical and Laboratory
Data of Plasmaoheresis-Treated
Patients Normal Value5 of Plasma
Subject Nos. 1
2
3
Lipids and
4’
5
6
7’
Apoproteinst
M
M
F
M
M
M
F
Age (vr)
30
32
55
61
41
38
51
BL (cm)
167
168
148
161
161
165
155
BW (kg)
65
50
43
59
52
60
54
488
337
351
290
345
391
270
176 + 26
146
117
111
59
245
187
197
87 f 29
9
9
13
2
16
18
20
31
13
13
4
15
9
11
LDL,-C (mg/dL)
324
160
171
128
128
189
109
LDL,-C (mg/dL)
40
47
35
47
72
77
52
HDL,-C (mg/dL)
10
16
17
8
19
22
11
HDL,-C (mg/dL)
8
17
18
11
21
18
8
Al
82
115
107
76
172
120
73
137 + 25
All
15
26
30
24
45
38
19
31 -c6
174
165
180
226
247
195
TC (mg/dL) TG (mg/dL) VLDL-C (mg/dL) IDL-C (mg/dL)
925 5+2 79 * 33 34 + 8 24 + 10 33 + 8
Apoprotein (mg/dL)
B
374
CII Clll E Diagnosis
7.0
3.6
12.7
10.1
8.0
6.8
FH
FH
AP
Ml
79 c 20
2.4
9.3
7.7
5.4
18.0
13.0
7.0
7.4 f 2.9
6.2
4.6
7.7
12.0
5.9
4.0 r 1.2
3.5
FH
8.4
FH
FH
FH
Ml
AP
MI
2.7
3.4 * 1.2
MI ACbypass
Abbreviations: FH, familial hypercholesterolemia; Ml, myocardial infarction: AP, angina pectoris; AC, aortocoronary bypass. *Analysis were made on the day of plasmapheresis, a day after probucol withdrawal. tValues are given as mean + SD. Data from Homma et al” and Goto et al?
separated plasma was discarded into a plastic bag. Blood cells were mixed with plasmanate and returned to the patient through the antecubital vein. ACD-A was used as an anticoagulant. Two and a half liters of plasma were removed at the rate of 1,000 mL/h. For DF plasmapheresis, the venous blood was introduced from the femoral vein at a speed of 100 mL whole blood/min to a membrane cell separator, KM 8500 (Kurare, Japan). The plasma that was separated from the blood cells with the first membrane filter (PVA filter SB-I, Kurare) was introduced to the second membrane filter
Table 2.
Changes in Plasma Lipids and Apoproteins
Immediately
After 2,500-ml
Plasmapheresis P Value
Double Filtratmn
Plasma Exchange
Number of treatments
(Eva1 HD 3A, 4A, Kurare or AC 1760, Asahi Medical, Japan). Plasma components that could not pass the second filter were discarded in a plastic bag. The plasma that passed through the second filter was returned to the patients through the antecubital vein. The same volume of 5% human albumin as the discarded plasma was added to the plasma returning to the patients. The approximate volume of the discarded plasma was 500 mL. The flow rate of the plasmapheresis was 1,000 mL/h at the second filter, and 2% L plasma was treated at each plasmapheresis. Average pore
DoubleFiltration 40 nm’
30 nm*
4
5
L
DSC Columnt
v DSC ColumnS
9
Total cholesterol
44.7
37.0
+ 9.9
36.5
” 7.5
44.9
* 5.9
VLDL-cholesterol
30.0
25.0
? 23.7
22.1
+ 2.7
16.6 * 13.4
IDL-cholesterol
24.7
28.6
+ 6.0
33.7
+ 6.0
27.8
+ 10.7
LDL-cholesterol
37.0
35.4
+ 9.7
30.8
t
1.6
40.8
+ 6.2
‘NS
HDL-cholesterol
30.9
55.2
+ 14.5
60.8
+ 5.7
97.6
i 3.0
coo
NS NS NS
Apoproteins§ A-l
37.9
56.9
+ 11.4
71.7
2 9.9
86.9
+ 6.5
<.oo 1
A-II
57.7
60.1
+ 5.9
67.2
* 12.6
88.1
* 11.0
1.00 1
B
26.3
34.8
f 8.7
38.2
+ 12.7
37.5
* 6.9
NS
C-II
56.0
43.4
* 14.1
44.1
+ 13.3
49.1
+ 13.1
NS
C-III
50.0
46.0
+ 7.7
45.3
+ 11.8
48.9
zk 12.4
NS
E
50.0
25.0
+ 23.7
35.2
? 10.0
39.3
k 16.0
NS
*Average pore diameter of the second filter membrane. tMean
t SD; values are expressed as the percentage of preplasmapheresis levels.
$Paired r-test. SApoproteins were measured at one plasma exchange in patient 1.
421
LDL REMOVAL BY DF AND DCS PLASMAPHERESIS
diameters of the second filter were 20 nm (Eva1 HF 3A), 30 nm (Eva1 HF 4A), and 40 nm (AC 1760). Heparin was used as an anticoagulant. The sterile DSC column (LA-01, 400 mL pack) for DSC column plasmapheresis was supplied from Kaneka, Japan. The second membrane of the DF plasmapheresis mentioned above was substituted for by a DSC column. The procedure of DSC column plasmapheresis was the same as the DF method. Albumin supply was unnecessary. Plasmapheresis was performed in the fasting state. DF plasmapheresis with the second filter having average pore diameters of 30 nm and 40 nm and DSC column plasmapheresis were performed alternatively to the same patients numbered 2, 3,4, 5. 6. and 7.
Laboratory Procedure Blood sampling was performed at every 500 mL plasmapheresis. Fasting blood was collected for up to 2 weeks. LDL was divided into two fractions at d = 1.045. Very low density lipoprotein (VLDL), intermediate density lipoprotein (IDL), LDL, (1.019 < d < 1.045), LDL, (1.045 < d < 1.063) HDL, (1.063 < d < 1.125) and HDL, I .21) were harvested ultracentrifugally by the method (I.125
of apoprotein B were reduced in the plasma by DSC column plasmapheresis. There were no significant differences in the removal rates of LDL-C, apoprotein B, CII, CIII, and E between DF plasmapheresis, with the second filter having an average pore diameter of 30 nm and 40 nm and DSC column plasmapheresis. About 90% of plasma apoprotein AI and AI1 was restored by DSC column plasmapheresis. DSC column bound lipoproteins were eluted by 1 .O mol/L NaCl solution. Apoproteins B, CII, CIII, and E were detected but apoprotein AI and AI1 were not detectable rn that eluate. Total cholesterol, LDL-C, apoprotein B, CD, CIII, and E decreased almost linearly throughout DSC plasmapheresis as well as DF plasmapheresis. Plasma apoprotein AI and AI1 decreased slightly at an early stage and stayed constant throughout DSC column plasmapheresis. Trends of the apoprotein AI and AI1 changes were similar to the changes in plasma albumin and y-globulin. Plasma y-globulin decreased markedly by DF plasmapheresis, but the intravenous y-globulin supply was unnecessary for 2.5 L plasmapheresis. The decrease of plasma albumin and y-globulin was minimal by DSC column plasmapheresis. There were no significant side effects during the plasmapheresis by either method. The recovery process of subfractionated lipoproteins after DF plasmapheresis was studied in patients 1, 2, 3, 5, and 6. All were inpatieints and kept on a 1.600-kcal diet without
Total
cholesterol
Plasmapheresis (PP) 500-
RESULTS
in plasma lipoprotein cholesterol and apoproteins after 2.5 L plasmapheresis are shown in Table 2. Plasma exchange was performed to patient 1. Sixty-three percent of LDL-C and 69% of HDL-C were removed by plasma exchange. The removal rate of LDL-C and HDL-C was almost the same between plasma exchange and DF plasmapheresis with the second filter having an average pore diameter of 20 nm. Therefore, we stopped using this second filter membrane and used the second filter having an average pore diameter of 30 nm or 40 nm. Sixty-three percent of total cholesterol (TC) and 66% of LDL-C were removed on average by DF plasmapheresis with the second membrane filter having an average pore diameter of 30 nm and 40 nm. On that occasion, the average removal rate of HDL-C was 42%, which was significantly lower than that of LDL-C (P c .Ol, paired t-test). The removal rate of apoprotein AI and AI1 was significantly lower than that of apoprotein B (P c .Ol, paired t-test) by DF plasmapheresis with the second filter having an average pore diameter of 30 nm and 40 nm. The removal rate of LDL-C with the second filter of 40 nm diameter was same as that with the second filter of 30 nm diameter. The removal rate of HDL-C, and apoproteins AI and AI1 with the second filter of 40 nm diameter was slightly lower than that with the second filter of 30 nm diameter. Fifty-five percent of TC and 59% of LDL-C were removed on average by DSC column plasmapheresis, while the plasma HDL-C level did not decrease at all. Thirteen percent of apoprotein AI and 12% of apoprotein AI1 and 63%
400.
Changes
-
300 200. 100. = B 1
O-
LDL-cholesterol .fi E f e s ; % a
3001 200IOOO-
40. 20.
HDL- cholesterol
9’
60.
kA+=====! J-1
01
Eeforsl
2
3 4
5 6
7 6
9 IO II 12 13 14
Days Fig 1. The plasma levels (mean t SD) of total cholesterol. LDL cholesterol, and HDL cholesterol after DF plasmapheresis (PP). Five FH patients (1,2. 3, 5, and 6) were examined.
422
HOMMA ET Al
medication. Medication was stopped at least two months before plasmapheresis. The serial changes in the levels of plasma TC, LDL-C, and HDL-C after plasmapheresis are shown in Fig 1. The plasma TC and LDL-C levels were 80.3 t 9.7% and 79.3 t 12.4% of the preplasmapheresis levels on the 14th day after plasmapheresis, respectively. The plasma HDL-C level reached the preplasmapheresis level within a week. Plasma apoprotein levels up to 2 weeks after plasmapheresis are illustrated in Fig 2. Apoprotein B reached the preplasmapheresis level at the tenth day after plasmapheresis, while apoprotein AI, AII, CII, CID, and E did so at the seventh day. The plasma levels of cholesterol in VLDL, IDL, LDL,, LDL?, HDLI, and HDLJ up to two weeks after plasmapheresis are shown in Fig 3. VLDL-C returned very rapidly to the value before plasmapheresis, ie, within four days, while IDL-C needed 2 weeks to do so. The plasma LDL-C level increased gradually, up to 83.7 + 18.9% of the preplasmapheresis level on the 14th day after plasmapheresis. LDL*-C returned to the value before plasmapheresis at the seventh day. HDL,-C needed a week to return to the preplasmapheresis level, and HDLS-C reached the peak on the seventh day.
Plasmqlhaw*s (PP)
Y
@
PP
IDL-C
PP
LDLI-C
2
I4 g g s ‘c: e E
200.
f
T
i
100. 0
f
IOO-
iE
50.
PP
LDL2-C
PP
HDLz -C
PP
HDL3-C
L O3
+
2 T I
I
5
Plasmaphmsis[PP) T
IO
IO 2 0!
APO-AI I
VLDL-C 20.
I!I
IO 1 6efoml
30
2 3 4
5 6
7 6 9 KJ II 121314
Days
20 IO 40 1
The cholesterol levels (mean f SD) in VLDL. IDL. LDL,. Fig 3. LDL,, HDL,, and HDL, after DF plasmapheresis (PPl. Five FH patients (1.2.3. 6. and 6) were examined.
400 300. f
200.
1
‘0:
5 F f
IO
f
5
E I 0
di 20
APO-C II
APO-C III
PP
T I I
1 I
IO
l
0I IO
Ape-E
PP
f 50 I S&ml
2 3 4
5 6
7
6 9 IO II I2 I3 14
Days Fig 2. Plasma apoprotein levels (mean k SD) after DF plasmapheresis (PP). Five FH patients (1,2,3, 6, and 6) were examined.
Phospholipids of the subfractionated lipoproteins were also measured in patients 2, 3, 5, and 6 (Fig 4). The changing patterns of phospholipids were very similar to those of cholesterol in all lipoprotein fractions. LDL,-PL reached the peak much earlier than LDL,-PL as LDL*-C did so. The changes in apoprotein B of VLDL, IDL, LDL,, and LDLz and in apoprotein AI in HDL2 and HDL, are demonstrated in Fig 5. Apoprotein AI and B in the subfractionated lipoproteins were measured in patients 1, 2, 3, 5, and 6. The changes in apoprotein B were very similar to those in cholesterol and phospholipids in VLDL, IDL, LDL,, and LDL*. Average LDL,-apo B was 83.7% of the preplasmapheresis level on the 14th day after plasmapheresis. LDL2apo B reached the peak at the seventh day and was 84.7% of the preplasmapheresis level on that day. Plasma HDL,-apo AI was extremely low in those patients. It reached the peak at seventh day after DF plasmapheresis but the peak was slightly lower than the preplasmapheresis level. Plasma HDL,-apo AI was also low in those patients and returned to the preplasmapheresis level on the seventh day. The changing patterns of apoprotein AI in HDL, and HDLS were similar to those of cholesterol and phospholipids.
423
LDL REMOVAL BY DF AND DCS PLASMAPHERESIS
VLDLPiOSmODhCWOSiS (PP)
IDL-
PP
1-
VLDL-apoB
PL
Pl0smOdlNasislPP)
PL
IDL-apoB PP
PC -&
LDL2-PL
PP
HDL2-PL
PP
HDLe-opo
Al
I 1 HDLs-
PL
60
PP z
HDLs-apoAl T
40
-i
20
Beforel 2 3 4
5 6 7
8 9 1011 121314
Days Befod
2 3 4
5 6
7
6
9 1011 121314
Days Fig 4. The phospholipids levels (mean + SD) in VLDL, IDL, LDL,. LDL,, HDL,, and HDL, after DF plasmapheresis (PP). Four FH patients (2.. 3. 5, and 6) were examined.
DISCUSSION
The possibility of selective removal of apoprotein B containing lipoproteins by DF plasmapheresis and DSC plasmapheresis and the acute recovery process of plasma lipoproteins after plasmapheresis were investigated in hypercholesterolemia. Theoretically, VLDL, IDL, and LDL can be selectively removed by DF plasmapheresis if the second filter has a diameter of 20 nm. This is because the diameter of HDL is around 10 nm and the diameters of other apoprotein B containing lipoproteins are greater than 20 nm,15J6 However, it was impossible to remove apoprotein B containing lipoproteins selectively by DF plasmapheresis even though a second filter of large pore diameter was used. The removal rates of HDL-C, apoprotein AI, and AI1 were significantly lower than those of LDL-C and apoprotein B when a second filter having an average pore size of 30 nm or 40 nm was used, but the selectivity in the removal of apoprotein B containing lipoproteins was not satisfactory by DF plasmapheresis. This was probably due to the adhesion of plasma components to the second filter. The reduction of the flow rate, the use of a wide membrane, and the use of the second filter with a large pore size must be considered in an effort to increase the selective removal of apoprotein B
Fig 5. The levels (mean + SD) of apoprotein 6 and apoprotein Al in subfraotionated lipoproteins after DF plasmapheresis (PP). Five FH patients (1.. 2. 3. 5. and 6) were examined.
containing lipoproteins by DF plasmapheresis. The decrease of plasma HDL-C, apoprotein AI, and AII seemed to be smaller by DF plasmapheresis, with the second filter having an average pore diameter of 40 nm than with that having an average pore diameter of 30 nm. Therefore, the use of the second filter with a larger pore size must be tried first for the improvement of the selectivity of the removal of apoprotein B containing lipoproteins by DF plasmapheresis. The selectivity in the removal of apoprotein B containing lipoproteins by DSC column plasmapheresis was excellent. When the DSC column was eluted by 1.0 moi/L NaCl solution, bound lipoproteins could be detached. Apoprotein CD, CID, and E were also eluted as well as apoprotein B. Therefore, VLDL and IDL were also bound to the DSC column. Plasma LDL-C and apoprotein B decreased almost linearly up to 2.5 L plasma treatment by the plasmaphereseis with 400 mL DSC column. The plasma TC levels were satisfactorily low after 2.5 L DSC column plasmapheresis. The capacity of 400 mL DSC column was enough for the treatment of those patients. However, the replacement of DSC column might be necessary during plasmapheresis for the treatment of homozygous or large heterozygous FH patients. The combination therapy with the DSC column plasmapheresis and drugs was expected to normalize the plasma LDL levels of heterozygous FH.
424
HOMMA
There are three types of apoprotein B containing lipoproteins adsorbing materials that have been used clinically: heparin-agarose, anti-LDL antibody-sepharose, and DSC.6*9.“’The removal of apoprotein B containing lipoproteins by heparin-agarose was made by batch chromatography.’ The removal capacity of apoprotein B containing lipoproteins was almost the same between heparin-agarose and DSC.” It must be possible to produce a same typed heparin-agarose column as a DSC column, but the cost of a heparin-agarose column is probably much higher than that of a DSC column. A 400 mL column of DSC bound about 7 g of cholesterol while a 500 mL column of anti-LDL column bound 2 g of cholesterol.9~‘0 Two columns were used in rotation, and adsorption and desorption were performed simultaneously during a treatment by anti-LDL column plasmapheresis because of the relatively low binding capacity of the antibody column.’ The desorption of apoprotein B containing lipoproteins was unnecessary during a treatment by DSC column plasmapheresis. Therefore, the method of DSC column plasmapheresis is simpler that that of antiLDL plasmapheresis. We have never used a DSC column repeatedly but it is possible to use a DSC column repeatedly because the desorbing of apoprotein B containing lipoproteins is easily made by NaCl solution (NaCl concentration ~0.4 mol/L). Anti-LDL antibody and DSC have a possibility to cause allergic reactions, but such side effects have not been reported. We must carefully observe possible side effects of the long-term treatment by plasmapheresis. DSC column is factory-made and the cost of DSC column will be reduced if it will be used widely. We believe that DSC column plasmapheresis is the best method of plasmapheresis to remove the plasma apoprotein B containing lipoproteins specifically at this moment. The changes in cholesterol, phospholipids, and apoproteins were similar in every lipoprotein fraction after plasmapheresis. Therefore, the alterations in chemical compositions of lipoprotein molecules seemed to be very little during the recovery process of plasma lipoproteins after plasmapheresis. All VLDL is converted to LDL via IDL by delipidation by
ET AL
lipoprotein lipase and hepatic triglyceride lipase in normal subjects.16 However, a part of plasma LDL is directly secreted from the liver in FH.24325 VLDL is converted rapidly to IDL in normal subjects.26 Plasma VLDL increase was very rapid after plasmapheresis. The baseline values of plasma IDL was high and IDL increase was relatively slow in heterozygous FH patients. This finding may suggest that VLDL is only a part of the precursor of IDL or the conversion of IDL to LDL was very slow due to the low hepatic triglyceride lipase acitivity in heterozygous FH.*’ Teng et al recently reported the good precursor-product relationships between VLDL and IDL, and between IDL and light LDL in the limited cases of heterozygous FH.25 Therefore, the second possibility might be the cause of slow IDL increase in heterozygous FH. Apoprotein B reached the plateau much earlier in LDL, than in LDL,, as well as cholesterol and phospholipids. The discrepancy of recovery speed between LDL, and LDL, probably indicates that LDL, and LDL2 have some different metabolism in heterozygous FH. Isotope kinetic study must be undertaken for further study. Teng et al demonstrated the direct secretion of heavy LDL to the plasma in heterozygous FH from the isotope kinetic study.25 Plasma HDL-C levels are normal or slightly low in heterozygous FH.’ These heterozygous FH patients had the plasma levels of low HDL2-C and very low HDL,-C. HDL, and HDL, increased in the same manner, and they reached the preplasmapheresis level in a week. HDL, is converted to HDL, in vitro.2s~29However, it is impossible to study the metabolic interrelationship between HDL, and HDL, in vivo only by the measurement of cholesterol, phospholipids, and apoprotein ALi which are transferred very rapidly between HDL2 and HDL,.‘55’6 Drug therapy is usually made for the treatment of FH between each plasmapheresis. It is important to look at the recovery process of plasma lipoproteins without medication at first as the basal data in order to select the most effective drug to normalize plasma LDL levels of FH by the combination therapy with plasmapheresis and drugs.
REFERENCES 1.
Goldstein JL, Brown MS: Familial
hypercholesterolemia,
in
Stanbury JB, Wyndgarden JB, Frederickson SD, et al (eds): The Metabolic Basis of Inherited Disease. New York, McGraw-Hill, 1982, pp 672-212 2. Mobuchi H, Tatami R, Haba T, et al: Homozygous familial hypercholesterolemia in Japan. Am J Med 65:290-297, 1978 3. Thompson GR, Lowenthal R, Myant NB: Plasma exchange in the management of homozygous familial hypercholesterolemia. Lancet 1:1208-1211, 1975 4. Simons LA, Gibson JC, lsbister JP, et al: The effects of plasma exchange on cholesterol metabolism. Atherosclerosis 31:195-204, 1978 5. Berger GMB, Miller JL, Bonnici F, et al: Continuous flow plasma exchange in the treatment of homozygous familial hypercholesterolemia. Am J Med 65:243-25 I, 1978 6. Lupien PH. Moorjani S, Lou M, et al: Removal of cholesterol from blood by affinity binding to heparin-agarose: Evaluation of treatment in homozygous familial hypercholesterolemia. Pediatr Res 14:113-l 17. 1980
7. Thompson GR, Myant NB, Kilpatrick D, et al: Assessment of long-term plasma exchange for familial hypercholesterolemia. Br Heart J 43:680-688, 1980 8. Stein EA, Blueck CJ, Wesselman A, et al: Repetitive intermittent flow plasma exchange in patients with severe hypercholesterolemia. Atherosclerosis 38:149-164, 1981 9. Stoffel W, Borberg H, Greve V: Application of specific extracorporeal removal of low density lipoprotein in familial hypercholesterolemia. Lancet 2:1005-1007, 1981 10. Yokoyama S, Hayashi R, Satani M, et al: Selective removal of low density lipoprotein by plasmapheresis in familial hypercholesterolemia. Arteriosclerosis 5:613-622, 1985 11. Saal SD, Parker TS, Gordon BR, et al: Removal of lowdensity lipoproteins in patients by extracorporeal immunoadsorption. Am J Med 80:583-589, 1986 12. Bilheimer DW, Goldstein JL, Grundy SM, et al: Reduction in cholesterol and low density lipoproteins synthesis after portacaval shunt in a patient with homozygous familial hypercholesterolemia. J Clin Invest 56:1420-1430, 1975
LDL REMOVAL
425
BY DF AND DCS PLASMAPHERESIS
13. Forman MB, Baker SD, Mieny homozygous familial hypercholesterolemia
CJ, et al: Treatment of with portacaval shunt.
Atheroscl41:349-361, 1982 14. Sterzl T, Chase HP, Ahrens EH, et al: Portacaval shunt in patients with familial hypercholesterolemia. Ann Surg 198:273-282, 1983 I?. Scott H W: lleal bypass in the control of hyperlipidemia and atherosclerosis. Arch Surg 113:62-70, 1978 Ih. Jackson RL, Morrisett JD, Gotto AM: Lipoproteins structure and metabolism. Physiol Rev 56:259-316. 1976 Ii. Eisenberg S: High density lipoprotein metabolism. J Lipid Res 25:1017-1058, 1984 18. Homma Y, Watanabe K, Tamachi H, et al: The change in plasma lipoproteins in a case of heterozygous familial hypercholesterolemia after plasmapheresis. Jpn J Med 23:205-210, 1984 19. Homma Y. Nakaya N, Nakamura H, et al: Increase in the density of lighter low density lipoprotein by hepatic triglyceride lipase. Artery 13:19-31. 1985 20. Goto Y. Akanuma Y. Harano Y, et al: Determination by the SRID method of normal values of serum apolipoproteins (A-l, A-II, B.C.II, C-III and E) in normolipidemic healthy Japanesesubjects. J Clin Biochem Nutr 1:73-88. 1986 21. Have1 RJ, Eder HA, Bragdon JH: The distribution and chemical composition of ultracentrifugally separated lipoproteins in human plasma. J Clin Invest 34:1346-1353, 1955 22. Autoanalyzer manual. New York, Technicon Autoanalyzer Inc, 1981
23. Takayama
M, ltoh S, Nagasaki
T, et al: A new enzymatic
method for determination of serum choline-containing lipids. Clin Chim Acta 79:93-98, 1977
phospho-
24. Janus ED, Nicoll A, Wootton R, et al: Quantitative studies of very low density lipoprotein: Conversion to low density lipoprotein in normal controls and primary hyperlipidemic states and the role of direct secretion of low density lipoprotein in heterozygous familial hypercholesterolemia. Eur J Clin Invest 10:149-l 59. 1980 25. Teng B, Sniderman hyperapobetalipoproteinemia. density lipoprotein with normal and 771663-672.
AD. Soutar AK, et al: Metabolic basis of Turnover of apolipoprotein B in low
and its precursors and subfractions compared familial hypercholesterolemia. J Clin Invest
1986
26. Reardon MF, Fidge NH, Nestel PJ: The catabolism of very low density lipoprotein B apoprotein in man. .I Clin Invest 61:850860. 1978 27. Nikkill EA, Huttunen JK, Ehnholm C: Low post-hepdrin plasma hepatic lipase activity in familial type IIa hyperlipoproteinemia. Ann Clin Res 8:63-67, 1976 28. Patsch JR. Gotto M, Olivecrona T, et al: Formation of high density lipoprotein-like particles during lipolysis of very low density lipoproteins in vitro. Proc Natl Acad Sci USA 75:45 19-4523. 1978 29. Grow TE. Fried M: Interchange of apoprotein components between the human plasma high density lipoprotein subclass HDL, and HDL, in vitro. J Biol Chem 253:8034-804 I, 1978