Kidney International, Vol. 56, Suppl. 71 (1999), pp. S-206–S-209

Low-density lipoprotein apheresis retards the progression ofhyperlipidemic overt diabetic nephropathy

TOSHIYUKI NAKAO, MAKI YOSHINO, HIROSHI MATSUMOTO, TOMONARI OKADA, MYONGI HAN,HIROMI HIDAKA, TAMAMI SHINO, CHIKAYUKI YAMADA, YUME NAGAOKA,and TADASHI MIYAHARA

Department of Nephrology, Tokyo Medical University and Jikei Medical University School of Medicine, Tokyo, Japan

Low-density lipoprotein apheresis retards the progression of move a large amount of plasma lipid directly from pa-hyperlipidemic overt diabetic nephropathy. tients in a short time. If hyperlipidemia played an impor-

Background. Hyperlipidemia has recently received atten- tant role in the progression of renal diseases, it shouldtion as being involved in the progression of diabetic nephropa-be possible to improve renal injury by LDL-A. Somethy (DN). Low-density lipoprotein apheresis (LDL-A) can re-successful clinical results were already reported for pri-move a large amount of plasma lipid directly from the patients

in a short time. mary nephrotic syndrome caused by focal glomeruloscle-Methods. Fifteen type 2 diabetic patients with overt ne- rosis [4, 5]. LDL-A, however, has not yet been used in

phropathy received LDL-A in two different manners: short- patients with DN. In this study, we performed LDL-Aterm intensive therapy (SIT) for nine patients and long-termin hyperlipidemic overt DN patients to evaluate the ef-intermittent therapy (LIT) for six patients.fect of this therapy on the progression of DN.Results. The changes in the monthly decline rates of recipro-

cal serum creatinine (1/Cr) were 20.035 6 0.020 in the three-month period before SIT, 0.047 6 0.041 during and until two

METHODSweeks after SIT, and 20.035 6 0.015 after a period of twoweeks from the therapy. The mean duration of LIT in six Patientspatients was 8.2 6 7.4 months, and the mean monthly decline

Fifteen type 2 diabetic patients (nine males and sixrates of 1/Cr significantly decreased during the period of LITas compared with the six-month period before the treatment. females, mean age 60.3 6 6.4) with nephropathy were

Conclusion. LDL-A can retard the progression of overt DN, enrolled in this study. Their mean creatinine clearanceespecially when it is performed repeatedly for a long period was 35.6 6 19.2 ml/min, serum creatinine 2.1 6 1.4at two-week intervals.

mg/dl, and urinary protein excretion 6.8 6 2.9 g/day, andtheir renal function was progressively declining. Theirmean serum cholesterol was 298.4 6 49.1 mg/dl, triglycer-The prime pathogenetic factor of diabetic nephropa-ides 272.3 6 196.5 mg/dl, LDL 842.7 6 196.6 mg/dl, andthy (DN) is high blood glucose, and recent clinical andvery LDL (VLDL) 371.1 6 344.9 mg/dl, and their HbA1cepidemiological investigations have demonstrated thatwas 7.9 6 1.3%. The mean arterial pressure (MAP) wasthe incidence of outbreak of DN was higher in poor-99.4 6 6.5 mm Hg.control patients than in good-control patients with glyce-

All studies were conducted in accordance with themia [1]. In the progression of overt DN, in which protein-principles set out in the Declaration of Helsinki, anduria has progressed to the nephrotic range, however, theinformed consent was obtained from all patients.adequacy of blood glucose control has only a limited

impact [2]. The fall of glomerular filtration rate in ne- Low-density lipoprotein apheresisphrotic DN is usually rapid and appears to be linear with

The LDL-A system consists of polysulfone fibers (Sul-time. Therefore, the factors other than hyperglycemiaflux; Kaneka, Japan) as the plasma separator and a dex-have been suggested to contribute to such progression,tran sulfate cellulose column (Liposorba15LA; Kaneka)and hyperlipidemia has recently received attention asas the lipid absorber. One hundred to 150 ml per minuteone of the factors incriminated in this process [3].of blood flow was circulated extracorporeally from theLow-density lipoprotein apheresis (LDL-A) can re-femoral vein to the anticubital vein via the LDL-A sys-tem. In a single LDL-A procedure, a total of 4000 ml of

Key words: diabetic nephropathy, hyperlipidemia, LDL apheresis. plasma was treated during two to three hours.Two different types of treatment, short-term intensive 1999 by the International Society of Nephrology

S-206

Nakao et al: LDL-apheresis for diabetic nephropathy S-207

therapy (SIT) and long-term intermittent therapy (LIT),were instituted. In SIT, a total of six LDL-A procedureswere performed twice a week during three weeks, andin LIT, LDL-A was performed once every two weeksfor as long a period as possible. SIT was conducted in thefirst nine consecutive patients, and LIT was conducted inthe next six consecutive patients. There were no statisti-cally significant differences between SIT and LIT pa-tients in creatinine clearance (35.5 6 19.2 vs. 23.7 6 8.6ml/min), urinary protein excretion (6.6 6 3.5 vs. 8.3 62.5 g/day), serum creatinine (2.4 6 1.4 vs. 2.4 6 0.5mg/dl), serum total cholesterol (264.0 6 40.8 vs. 303.5 6

Fig. 1. Effects of short-term intensive therapy (SIT) of low-density53.3 mg/dl), serum triglycerides (264.6 6 268.3 vs.lipoprotein apheresis on the decline rates of renal function in overt246.7 6 70.7 mg/dl), and serum VLDL (336.0 6 399.0 diabetic nephropathy. Improvement of monthly decline rates of recipro-

vs. 369.2 6 199.3 mg/dl). Serum LDL was significantly cal serum creatinine (1/Cr) were observed during SIT. These effects,however, were only transient, and from two weeks after the last therapy,higher in LIT than SIT patients (1000.3 6 151.9 vs.their renal function began to decline again at the same rates as before.767.1 6 172.0 mg/dl, P , 0.05).

Reduction rates of serum cholesterol, triglycerides,LDL, and VLDL by single LDL-A were assessed in nine

Table 1. Effects of long-term intermittent therapy of LDL-apheresispatients. Serum LDL and VLDL levels before, immedi-on the retardation of the renal function decline in overtately after, and two weeks after a single LDL-A were diabetic nephropathy

compared in seven patients. Serial changes in serum cre-Decline rate (1/Cr/month)atinine, creatinine clearance, and urinary protein excre- Duration

Patient month Before Duringtion were measured during the study period in all pa-1 22 20.032 20.010tients.2 11 20.060 20.006Statistical analyses were performed using the paired3 5 20.037 20.026

t-test and Student’s t-test. 4 5 20.108 20.0465 4 20.037 20.0016 2 20.034 20.015Mean 8.267.4 20.05160.030 20.02260.015aRESULTS

P 5 0.022Changes in serum lipids concentrations by low-densitylipoprotein apheresis

Serum lipids concentrations were significantly de-creased by single LDL-A, and the reduction rates of These effects, however, were only transient, and from

two weeks after the last therapy, their renal functionserum cholesterol, triglycerides, LDL, and VLDL were58.6 6 7.3, 57.9 6 11.1, 73.4 6 14.9, and 75.8 6 16.7%, began to decline again at the same rates as before. The

changes in the monthly decline rates of reciprocal serumrespectively. Serum LDL levels before, immediatelyafter, and at two weeks after a single LDL-A were 677.9 6 creatinine (1/Cr) were 20.035 6 0.020 in the three-

month period before SIT, 0.047 6 0.041 during and until75.6 mg/dl, 168.3 6 40.2 mg/dl, and 615.3 6 38.9 mg/dl,respectively, and serum VLDL levels at those times were two weeks after SIT, and 20.035 6 0.015 after the period

of two weeks from the therapy (Fig. 1).397.6 6 205.9 mg/dl, 165.3 6 129.9 mg/dl, and 489.1 6192.1 mg/dl, respectively. Serum LDL and VLDL con-

Effects of long-term intermittent therapycentrations were significantly lower immediately afterthan before a single LDL-A (P , 0.001); however, there Mean duration of LIT in six patients was 8.2 6 7.4

months. During the period of LIT, the mean monthlywas no significant difference between before and two-week levels after a single LDL-A. decline rates of 1/Cr significantly decreased compared

with the six-month period before the therapy (Table 1).Effects of short-term intensive therapy There were no significant changes in either their aver-

aged monthly measured HbA1c (8.1 6 1.5 vs. 8.0 6A patient whose serum creatinine was 6.1 mg/dl at thestart of SIT required hemodialysis because of overhydra- 1.6%) or MAP (100.2 6 5.6 vs. 106.8 6 10.2 mm Hg)

between a six-month period before LDL-A and duringtion two weeks after the beginning of SIT and droppedout from the study. In the remaining eight patients, creat- the period of LDL-A. Figure 2 shows a typical case of

a 48-year-old woman, whose renal function decline wasinine clearance was increased in five. Urinary proteinexcretion was reduced also in five, and serum creatinine effectively suppressed over a 22-month period by receiv-

ing LDL-A once every two weeks.was lowered in seven patients by two weeks after SIT.

Nakao et al: LDL-apheresis for diabetic nephropathyS-208

Fig. 2. A typical case of a 48-year-old womanwho received long-term intermittent therapy(LIT). Her renal function decline was effec-tively suppressed over a period of 22 monthsby receiving low-density lipoprotein apheresisonce every two weeks.

DISCUSSION whether rapid lipid lowering by LDL-A is effective inthe retardation of progression of overt DN. In this study,Overt DN usually shows an irreversible linear declineour preliminary results suggest that LDL-A could retardof the glomerular filtration rate. Recently, however,the progression of overt DN, if it is performed repeatedlyblood pressure control, dietary protein restriction, andat an interval of two weeks.angiotensin-converting enzyme inhibition have been es-

A rapid decrease of circulating lipids immediatelytablished as effective interventions to slow the progres-after a single LDL-A procedure and lowering of time-sion of overt DN [6].averaged concentrations of serum lipids by repeatingExperimental evidence suggests that hyperlipidemiaLDL-A might reduce the load of lipids on renal tissueparticipates in the progression of glomerular injury, andand attenuate lipid-induced renal injury [14]. In additiona more rapid decline of renal function was observed into the previously mentioned mechanism, bradykinin,DN patients with hyperlipidemia than those without itwhich increases during LDL-A [15], might alter renal[7, 8]. Lipid-lowering therapy using antilipemic agentshemodynamics, contributing to improved renal function.such as HMG-CoA reductase inhibitors may have a ben-Although a larger randomized controlled study is neededeficial effect on the retardation of progression of DNto prove LDL-A as an effective therapy for DN, these[9–11]. However, this effect is controversial, and other

reports indicated no influence of antilipemic agents in data encourage further investigations.the progression of DN [12].

Low-density lipoprotein apheresis is an extracorporeal ACKNOWLEDGMENTSprocedure that specifically removes apo B-containing li- The authors are indebted to Professor J. Patrick Barron of thepoproteins directly from the blood. In this study, we International Medical Communications Center of Tokyo Medical Uni-

versity for his review of the manuscript.performed LDL-A for DN using a system with columnscontaining immobilized dextran sulfate. As plasma pas-

Reprint request to Toshiyuki Nakao, M.D., Department of Nephrol-ses through the column, apo B-containing lipoproteins ogy, Tokyo Medical University, 6-7-1, Nishishinjuku, Shinjukuku,

Tokyo 160-0023, Japan.such as LDL, VLDL, and Lp(a) are selectively adsorbedby the ligands on the surface of the adsorbent. Therefore,this system can remove a large amount of lipids from REFERENCESplasma. In this study, plasma LDL and VLDL levels 1. The Diabetes Control and Complication Trial Researchwere reduced to 80 to 70% of pretreatment levels within Group: The effect of intensive treatment of diabetes on the devel-

opment and progression of long-term complications in insulin-two to three hours. This system was applied first fordependent diabetes mellitus. N Engl J Med 329:977–986, 1993patients with heterozygous familial hypercholesterol- 2. Hasslacher C, Stech W, Wahl P, Ritz E: Blood pressure and

emia and is now also used for the treatment of coronary metabolic control as risk factors for nephropathy in type 1 insulin-dependent diabetes. Diabetologia 28:6–11, 1985arteriosclerosis and arteriosclerosis obliterans. These ex-

3. Guijarro C, Keane WF: Lipid-induced glomerular injury. Nephronperiences have shown that LDL-A performed by this67:1–6, 1994

system is a safe and effective method [13]. 4. Muso E, Yashiro M, Matsushima M, Yoshida H, Sawanishi K,Sasayama S: Dose LDL-apheresis in steroid-resistant nephroticFew studies have been undertaken to investigate

Nakao et al: LDL-apheresis for diabetic nephropathy S-209

syndrome affect prognosis? Nephrol Dial Transplant 9:257–264, cholesterol and blood pressure correlate with regression of protein-uria. Am J Kidney Dis 27:809–818, 19961994

11. Sasaki T, Kurata H, Nomura K, Utsnomiya K, Ikeda Y: Amelio-5. Yokoyama K, Sakai S, Yamaguchi Y, Suzuki Y, Hinoshita F,ration of proteinuria with pravastatin in hypercholesterolemic pa-Hara S, Yamada A, Ogura Y, Kawaguchi Y, Sakai O: Completetients with diabetes mellitus. Jpn J Med 29:156–163, 1990remission of the nephrotic syndrome due to focal glomerular scle-

12. Hommel E, Andersen P, Gall MA, Nielsen F, Jensen B, Rossingrosis achieved with low density lipoprotein adsorption alone. Neph- P, Dyerberg J, Parving HH: Plasma lipoproteins and renal func-ron 72:318–320, 1996 tion during simvastatin treatment in diabetic nephropathy. Diabe-

6. American Diabetes Association: Diabetic nephropathy. Diabetes tologia 35:447–451, 1992Care 21:S50–S53, 1998 13. Gordon BR, Kelsey SF, Bilheimer DW, Brown DC, Dau PC,

7. Mulec H, Johnsen SA, Wiklund O, Bjorck S: Cholesterol: A Gotto AM, Lllingworth DR, Jones PH, Leitman SF, Prihodarisk factor in diabetic nephropathy. Am J Kidney Dis 22:196–201, JS, Stein EA, Stern TN, Zavoral JH, Zwiener RJ: Treatment

of refractory familial hypercholesterolemia by using an automated1993dextran sulfate cellulose adsorption system. Am J Cardiol 70:1010–8. Krolewski AS, Warram JH, Christlieb R: Hypercholesterolemia:1016, 1992A determinant of renal function loss and death in IDDM patients

14. Kean WF, Mulcahy WS, Kasiske BL, Kim Y, O’Donnel MP:with nephropathy. Kidney Int 45:S125–S131, 1994Hyperlipidemia and progressive renal disease. Kidney Int 39:S41–9. Lam KSL, Cheng IKP, Janus ED, Pang RWC: Cholesterol-low- S48, 1991

ering therapy may retard the progression of diabetic nephropathy. 15. Kojima S, Ogi M, Yoshimata Y, Kuramochi M, Ikeda J, Naga-Diabetologia 38:604–609, 1995 nawa M, Hatakeyama H: Changes in bradykinin and prostaglan-

10. Ellis D, Lloyd C, Becker DJ, Forrest KYZ, Orchard TJ: The dins plasma levels during dextran-sulfate low-density-lipoproteinapheresis. Int J Artif Org 20:178–183, 1997changing course of diabetic nephropathy: Low-density lipoprotein