Brenner and Rector's The Kidney, 8th ed.

CHAPTER 61. Plasmapheresis

Ernesto Sabath   Bradley M. Denker



Historical Perspective 2071



Technical Aspects 2071



Plasma Separation Techniques 2071



Venous Access 2073



Anticoagulation 2073



Replacement Fluid 2073



Complications 2073



Plasmapheresis in Renal Diseases 2074



Anti-Glomerular Basement Membrane Disease 2075



Rapidly Progressive Glomerulonephritis 2076



Lupus Nephritis 2076



Mixed Cryoglobulinemia 2077



Thrombotic Thrombocytopenic Purpura and Hemolytic Uremic Syndrome 2077



Recurrent Focal Segmental Glomerulosclerosis 2078



Renal Transplantation 2078



Conclusion 2079

The term plasmapheresis is derived from the Greek word apheresis that means “taking away” or removal. The modern medical terms therapeutic plasma exchange (TPE) or plasmapheresis (PE) involves the therapeutic removal of macromolecules from the plasma of patients with various medical conditions. The plasmapheresis process involves re-placement of plasma with albumin or other solutions (crystalloid or colloid) to maintain a normovolemic state. This chapter will review the history of plasmapheresis, technical aspects of the plasmapheresis process, and the major renal conditions for which it has therapeutic benefit.


It is unclear when the notion of therapeutic removal of blood components first originated, but it was flourishing even before Hippocrates in the fifth century BC. Bloodletting to remove evil humors was a commonplace medical practice, in part due to lack of understanding of disease processes and the paucity of effective therapies. By the Middle Ages, both surgeons and barbers were specializing in this bloody and often painful practice, and even as late as the nineteenth century bloodletting was employed for nearly every infectious and malignant malady afflicting patients in the United States and Europe.[1] The first “true” plasmapheresis procedure involving the removal of the “bad” blood and replacement with a “clean” solution was performed in 1914 by Abel, Rowntree, and Turner at The John Hopkins Hospital. The procedure was called “vivi-diffusion” and demonstrated the principle that the blood of a living animal could be dialyzed outside the body and then returned to the circulation. [2] [3] In 1960 Schwab and Fahey performed the first therapeutic manual plasmapheresis to reduce elevated globulin levels in a patient with macroglobulinemia.[4] The number of clinical indications for plasmapheresis has been growing since then. In the early days, the utility of plasmapheresis was based on anecdotal or uncontrolled studies. More rigorous reexamination of this treatment modality and the better understanding of the pathophysiology of some diseases in the past several decades have reaffirmed the role of plasmapheresis in the management of some conditions. However, the number of clinical conditions that have been rigorously studied with prospective and randomized controlled trials remains small, and decisions for the implementation of plasmapheresis (an invasive and potentially dangerous procedure) rests on anecdotal and uncontrolled studies in many circumstances.


For most conditions, the aim of the procedure is the removal of pathologic autoantibodies or toxins and the initial treatment goal is to exchange 1 to 1.5 times the plasma volume per plasmapheresis procedure. This will lower plasma macromolecule levels by 60% to 75%, respectively.

A formula that can be used to estimate plasma volume in an adult is[5]:

Estimated plasma volume (in liters) = 0.07 weight (in kg) × (100 - hematocrit [Hct])

For removal of components predominantly restricted to the plasma space, the use of higher exchange volumes will require significantly longer procedure times without additional clinical benefit. The ultimate clinical success of the procedure depends on the abundance of the abnormal protein and its rate of production. Unless the removal of the protein by plasmapheresis is combined with additional therapies (usually immunosuppressive or cytotoxic) to eliminate or reduce the source of the abnormal protein(s), the procedure is unlikely to provide clinical benefit. The time required to suppress abnormal protein production can take several weeks, which is why plasmapheresis protocols often require daily pheresis (or near daily) for prolonged times.

Plasma Separation Techniques

There are two major modalities to separate the plasma from the blood during a plasmapheresis procedure: (1) centrifugation and (2) membrane filtration ( Fig. 61-1 ). The mechanism and equipment for these two techniques vary significantly and are usually utilized by blood bank or hemodialysis facilities, respectively. The plasmapheresis method and location for performing the procedure depends on the specific disease and clinical presentation as well as the capabilities of different health care facilities. If severe renal failure is present and dialysis is required, then the membrane filtration method can be done in combination with conventional hemodialysis. If renal failure is not part of the clinical picture, then either method of plasmapheresis can be utilized.

FIGURE 61-1  Centrifugal separator (A) and membrane filtration systems (B) for plasma exchange. A, Blood is pumped into the separator container. As the centrifuge revolves, different blood components are separated into discrete layers, which can be harvested separately. Plasma is pumped out of the centrifuge into a collection chamber. Red cells, leukocytes, and platelets are returned to the donor, along with replacement fluid. B, Blood is pumped into a biocompatible membrane that allows the filtration of plasma while retaining cellular elements.  (From Madore F, Lazarus JM, Brady HR: Therapeutic plasma exchange in renal diseases. J Am Soc Nephrol 7(3):367–386, 1996.)



The centrifugation method uses centrifugal force to separate whole blood into plasma and cellular fractions according to their density (see Fig. 61-1 ). The centrifugation process can be either intermittent or continuous. In intermittent centrifugation, sequential volumes of whole blood are removed and centrifuged; the cellular fraction returned to the patient and the process is repeated until the desired volume of plasma is removed. The blood is pumped from the patient at a flow rate of up to 100 mL/min into the processing unit that consists of a bell-shaped bowl that rotates at high speed. The denser cellular blood components are centrifuged against the lateral walls while the plasma is removed through a central outlet on the top of the bowl. Each cycle removes about 500 to 700 mL of plasma and usually it is necessary to perform the process five or six times to achieve 2.5 to 4.0 L (1–1.5 plasma volumes) during a session. At the conclusion of each segment, the packed cells are emptied from the bowl and returned to the patient. The advantages of intermittent centrifugation include the relative simplicity of operation, portability of machines, and the convenience of a single-needle peripheral venipuncture. The disadvantages include the time (the procedure typically takes more than 4 hours) and the relatively large extracorporeal volume removed each time. In the continuous flow centrifugation system, the blood is pumped continuously into a rapidly rotating bowl, where plasma and cells are separated. Plasma is removed at a specified rate and the cells plus replacement fluid are returned to the patient in a continuous manner. This method is faster and is more suitable for hemodynamically unstable patients; however, it is more costly and requires two venipunctures or insertion of a dual lumen central venous catheter.

The membrane filtration technique is based on a synthetic membrane filter composed of different pore sizes. Similar to a hemodialysis filter, the plasmapheresis filter is composed of many hollow fiber tubes made of a membrane material with relatively large pore sizes (0.2–0.6 mm in diameter) and arranged in parallel. Blood is pumped through the hollow fiber tubes and the large pores are sufficient to allow passage of plasma (proteins and plasma water) while retaining cells within the hollow fiber lumen. The plasma is drained off while the cells are returned to the patient through a typical hemodialysis circuit. This technique can be done using conventional or continuous hemodialysis equipment, with a blood flow rate of 100 (±20) mL/min and an optimal transmembrane pressure of less than 70 mm Hg. Plasma is removed at a rate of 30 to 50 mL/min and the infusion rate of replacement fluid is adjusted to maintain intravascular volume. Potential disadvantages of membrane filtration include activation of complement and leukocytes by the artificial membrane, and the need for a central large bore catheter to obtain adequate blood flow rates. Both centrifugation and membrane filtration are safe and efficient plasmapheresis techniques and the main differences lie in the cost and expertise needed to operate. [6] [7] [8]

In recent years these basic techniques have been modified and/or coupled to other separation modalities. Cytapheresis is the removal of leukocytes or platelets in hematologic conditions. The target white blood cell count in hyperleukocytosis is less than 100,000/mL and for thrombocytosis, the goal is fewer than one million platelets/mL. Cytapheresis can also be performed for sickle cell crisis. In this setting, the goal is the removal of more than 50% of hemoglobin S, and replacement with normal allogeneic red cells. Other variations include the use of secondary membrane plasma fractionation, as used in double filtration plasmapheresis (also called cascade filtration), thermofiltration, and cryofiltration apheresis. Alternatively, affinity columns can be used for processing the separated plasma with adsorption columns such as protein A columns to remove IgG antibodies and immune complexes, or chemical affinity columns such as dextran sulfate that have negative charges and are used to remove antibodies or other positively charged plasma substances such as low-density lipoproteins (LDL) and very low-density lipoproteins (VLDL).[9]

Venous Access

Successful implementation of the plasmapheresis procedure requires reliable venous access. The clinical scenario, especially the possibility for long-term venous access, and the type of plasmapheresis being used are important factors to consider when deciding on peripheral or central venous access. A peripheral vein allows a maximum flow of up to about 50 mL/min to 90 mL/min, so a single venous access is adequate for intermittent centrifugation whereas continuous centrifugation techniques will require two venous access sites. For short-term procedures, this may be adequate but loss of venous access from recurrent intravenous catheters and phlebotomy in chronically ill patients is a major problem. If long-term (more than 1-2 weeks) plasmapheresis is planned, then a central venous catheter is required. When the membrane filtration technique is used, a central venous catheter is necessary in order to sustain blood flows over 70 mL/min. Central venous access can be achieved through the femoral, internal jugular, or subclavian veins; the femoral vein should be avoided if the treatment of the patient will be ambulatory.[10] In patients that require life-long therapy such as LDL-apheresis either an arterio-venous (AV) fistula or AV graft should be considered.[12] Central venous catheters have numerous long-term complications including catheter thrombosis, catheter infection, central vein thrombosis, and reactions/complications from the anti-coagulants (usually heparin). These issues will not be discussed here, but are discussed in Chapter 58 .


To prevent activation of the coagulation system within the extracorporeal circuit the plasmapheresis procedure requires anticoagulation. For centrifugation procedures, the acid-citrate-dextrose (ACD) solution (1/9 volume of solute per volume of solution) given as a continuous intravenous infusion is the most frequently used anticoagulant. The infusion rate is adjusted according to the blood flow rate (target ratio range from 1:10 to 1:25). When the venous flux and infusion rate of citrate are slow there is an increased risk for catheter clotting. In this circumstance, heparin (if not contraindicated) can be used alone or in combination with citrate. For membrane filtration plasmapheresis procedures, the use of standard unfractionated heparin is preferred and the required dose of heparin is about twice that needed for hemodialysis because a significant amount of infused heparin is removed along with the plasma. However, heparin may enhance systemic anticoagulation more than is expected because of the additional effect of dilution of clotting factors by the non-plasma replacement solutions. The initial loading dose of heparin (40 U/kg) is usually administered intravenously, followed by a continuous infusion (20 U/kg per hour) adjusted to maintain an adequate anticoagulation in the circuit. [6] [7] For patients who are receiving standard oral anticoagula-tion, additional low-dose anticoagulation with either regional citrate or heparin should to be added to facilitate the treatment and prevent clotting during the procedure. The heparin dose can usually be reduced by at least 50% in this situation.[11] In critically ill patients with coagulation abnormalities, the use of regional citrate is recommended.[12] Hirudin and lepirudin (thrombin inhibitors) are effective and safe alternatives for those patients with increased risk for thrombosis but who have contraindications for heparin administration. [13] [14]

Replacement Fluid

The choice of replacement fluids include 5% albumin, fresh frozen plasma (FFP) (or other plasma derivatives like cryosupernatant), and crystalloids (e.g., 0.9% saline, Ringer's lactate). Albumin is the most commonly used solution in plasmapheresis, and is generally combined with 0.9% saline on a 50:50 (vol%) basis. Albumin does not contain calcium or potassium and also lacks coagulation factors and immunoglobulins. It is safe and has never has been associated with transmission of hepatitis or HIV viruses. Fresh frozen plasma contains complement and coagulation factors and is the replacement fluid of choice in patients with thrombotic thrombocytopenic purpura (TTP) because the infusion of normal plasma may contribute to the replacement of the deficient plasma factor ADAMTS-13 (discussed later). Plasma may also be preferable in patients at risk of bleeding (e.g., those with liver disease or disseminated intravascular coagulation) or those requiring intensive therapy (e.g., daily exchanges for several weeks) because frequent replacements with albumin solution will eventually result in post-plasmapheresis coagulopathy and a net loss of immunoglobulins. The disadvantages of using FFP include the risk of viral disease transmission, and citrate overload.


Adverse events in plasmapheresis are common although death is rare. Table 61-1 summarizes the common complications related to the procedure. In the Swedish registry of therapeutic hemapheresis there were no deaths reported in over 14,000 procedures.[15] However, this registry recorded approximately a 4% adverse event rate each year that included medical symptoms, vascular access problems, and technical problems. Only about 1% of adverse events resulted in termination of the plasmapheresis procedure and interventions were needed in 65% of events. Paresthesias and hypotension were the most common events (22% and 20.5%, respectively). Vascular access problems occurred in only 0.44% but in nearly three quarters of these patients, therapy had to be interrupted.[15]

TABLE 61-1   -- Complications of Plasmapheresis

Vascular access





Catheter infections

Replacement fluids

Anaphylactoid reactions to fresh frozen plasma




Transmission of viral infections












Removal of erythropoietin and drugs bound to plasma proteins





Plasmapheresis can lead to a reduction in blood pressure that is usually due to a decrease in intravascular volume. The volume of extracorporeal whole blood is greater with intermittent centrifugation techniques and hypotension is more common than with continuous modalities. Hypotension can also occur in response to complement-mediated reactions to the membrane filter or sensitivity to ethylene oxide that is used as a membrane sterilant. Fresh frozen plasma used for replacement fluids is also associated with anaphylactoid reactions rarely resulting in death. FFP reactions are most often characterized by fever, rigors, urticaria, wheezing, and hypotension; cardiopulmonary collapse is rare.


The development of shortness of breath or dyspnea suggests the presence of pulmonary edema due to fluid overload. Noncardiogenic edema can rarely occur as a component of anaphylactic reactions; massive pulmonary emboli have been reported if the blood components being reinfused are not adequately anticoagulated.

Citrate-Induced Complications-Hypocalcemia and Metabolic Alkalosis

Citrate is infused either as the anticoagulant for the extracorporeal system or in the FFP administered as a replacement fluid. It binds to free calcium to form soluble calcium citrate, thereby lowering the free, but not the total, serum calcium concentration. Potential symptoms of hypocalcemia include perioral and distal extremity paresthesias. Symptoms can be anticipated and reduced by either intravenous or oral calcium if the plasmapheresis therapy will last longer than 1 hour. Another complication of citrate administration is the development of metabolic alkalosis in patients with concurrent renal failure. Metabolism of the excess citrate generates bicarbonate, the excretion of which is limited by the renal failure.


Replacement regimens utilizing saline and albumin solutions can result in a 25% reduction in the plasma potassium concentration in the post pheresis period. This can be minimized by adding 4 meq potassium per liter to the replacement solution.

Coagulation Abnormalities

Plasma exchange with albumin replacement produces a predictable decrease in clotting factors that may predispose to bleeding. A single plasma volume exchange increases the prothrombin time by 30% and the partial thromboplastin by 100%. These changes to return toward normal within several hours but with repeated plasmapheresis sessions these abnormalities can persist. Therefore, three to four units of FFP should be included with the replacement fluid each week or sooner in patients at risk for bleeding.


Removal of immunoglobulins and complement could result in an immunodeficient state. However, in a randomized, controlled trial of plasmapheresis in patients with lupus nephritis, patients receiving plasmapheresis were not more prone to infection than other patients receiving a similar immunosuppressive regimen but not plasma exchange.[16] Nevertheless, repeated apheresis treatments with albumin replacement will deplete the patient's reserve of immunoglobulins for several weeks. If an infection occurs, a single infusion of (100 to 400 mg/kg) intravenous immune globulin will restore the plasma immunoglobulin concentration toward normal. Although estimates for the risk of viral transmission by the use of FFP are low, the large volumes from multiple donors increase the risk in patients receiving long-term plasmapheresis therapy. Use of large-volume plasma units collected from a single donor and the use of hepatitis B vaccine may reduce the risk of virally transmitted infections.

Drug Removal

Substantial drug removal by plasma exchange occurs for those drugs that are highly protein bound and therefore primarily limited to the vascular space. Among drugs used to manage renal diseases, prednisone is not substantially removed whereas cyclophosphamide and azathioprine are removed to some extent. This potential problem can be circumvented by administering the drug after a plasma exchange treatment.

Angiotensin Converting Enzyme Inhibitors

Flushing, hypotension, abdominal cramping, and other gastrointestinal symptoms have been reported during plasmapheresis in patients receiving angiotensin converting enzyme inhibitors. In one report of 299 consecutive patients undergoing TPE, these atypical symptoms occurred in all 14 patients receiving an ACE inhibitor versus only 7% of those not treated with this medication.[17]


The mechanism(s) for clinical improvement of renal diseases by plasmapheresis depends on the pathophysiology of the underlying disease. The removal of small molecules and toxins with large volumes of distribution is more efficient with hemodialysis or hemofiltration. Plasmapheresis should be considered when the pathogenic factor(s) are large molecular weight substances or the patient is deficient in a plasma component. Table 61-2 lists the pathologic factors that may be removed with plasmapheresis. The molecular weight complexes amenable to plasmapheresis are most often abnormal proteins (typically auto-antibodies seen in numerous diseases) or (monoclonal immmunoglobulins seen in plasma cell dyscrasias). Less commonly, plasmapheresis can be used to remove lipoproteins or cells (platelets or white blood cells) in a modification of the procedure called cytapheresis. Plasmapheresis may also provide some benefit when there is high-grade immune complex formation leading to acute glomerulonephritis. In addition to removal of toxic proteins or replacement of deficient ones with plasma exchange, there appear to be other beneficial mechanisms. Splenic reticuloendothelial blockade in patients with vasculitis was reversed more frequently in patients who received plasmapheresis in addition to immunosuppressive therapy.[18] The removal of fibrinogen by plasmapheresis and the replacement of humoral factors may also play a beneficial role. The clinical indications for plasmapheresis according to category are summarized in Table 61-3 .

TABLE 61-2   -- Pathologic Factors Removed by Plasmapheresis


Immune complexes

Myeloma protein


Complement products

ADAMTS-13 (metalloproteinase)


Protein-bound toxins




TABLE 61-3   -- Summary of Renal Diseases Treated with Plasmapheresis



Antiglomerular basement membrane disease


Rapidly progressive glomerulonephritis


Hemolytic uremic syndrome




Renal transplant rejection


Desensitization for renal transplantation


Recurrent FSGS




Systemic lupus erythematosus



FSGS, focal and segmentary glomerulosclerosis; TTP, thrombotic thrombocytopenia purpura. Category I: Standard Primary Therapy; Category II: Supportive Therapy; Category III: When the evidence of benefit is unclear; Category IV: When there is no current evidence of benefit or for research protocols.




Anti-Glomerular Basement Membrane Disease

Anti-glomerular basement membrane (anti-GBM) antibody disease is a disorder in which circulating antibodies are directed against an antigen in the glomerular basement membrane, resulting in rapidly progressive glomerulonephritis. When accompanied by pulmonary hemorrhage, the condition is known as the Goodpasture syndrome. The target for the anti-GBM antibodies is the non-collagen (NC)-1 domain of the α3 chain of type IV collagen. Circulating anti-GBM antibodies are detected in more than 90% of patients, and the titer of circulating antibodies correlates with disease activity. [19] [20] Historically, untreated patients do not recover renal function and have substantial mortality, particularly from pulmonary hemorrhage. The rationale for the use of plasma exchange is the rapid removal of the pathogenic autoantibodies. Cyclophosphamide and corticosteroids are essential to reduce additional antibody synthesis, and to reduce inflammation in the short term. A rapid reduction in anti-GBM antibody levels is necessary in view of the speed of glomerular damage, and this cannot be achieved by drug therapy alone. The use of plasmapheresis was introduced for the management of anti-GBM disease in 1975,[21] but plasmapheresis regimens for anti-GBM disease have never been assessed in prospective randomized controlled trials. However, numerous uncontrolled studies and series of patients published in the past 25 years suggest the beneficial effect of plasmapheresis in survival and renal preservation rates. Some of these studies are summarized in Table 61-4 .

TABLE 61-4   -- Renal Recovery According to Initial Creatinine Concentration in Treated Patients with Anti-Glomerular Basement Membrane Antibody Disease

Study, Year, (Reference)

Patients (No)

Patients with Independent Renal Function at 1 Year




Initial Cr Concentration <600 μmol/L (<6.8 mg/dl)

Initial Cr Concentration ≥600 μmol/L (≥6.8 mg/dl) or dialysis dependent


Bouget et al, 1990[22]




Most patients received PE

Herody et al, 1993[23]




Most patients received PE

Merkel et al, 1994[24]




25 patients received PE

Andrews et al, 1995[25]




All patients had Cr concentration ≥600 mmol/L, only 8 patients received treatment

Daly et al, 1996[26]




23 patients received PE

Levy et al, 2001[27]




All patients received PE, C, and CFM

Saurina et al, 2003[28]




24 patients received treatment with C, CFM, and PE


Cr, creatinine; PE, plasmapheresis; C, corticosteroids; CFM, cyclophosphamide.




In an early, uncontrolled study of the clinical course in 20 patients with Goodpasture syndrome,[29] eight patients were treated with plasmapheresis and immunosuppression, four patients with immunosuppression alone, and eight patients with non-specific therapy. The plasmapheresis group had less severe renal failure, shorter duration of alveolar hemorrhage, and lower mortality. The only randomized study comparing immunosuppressive therapy (prednisone and cyclophosphamide) alone or in combination with plasmapheresis was small (17 patients).[30] Only two of eight patients who received plasmapheresis became dialysis dependent compared with six of nine in the immunosuppression alone group. A large uncontrolled study reported long-term outcomes in 71 patients.[27] All the patients received a standard immunosuppressive regimen of plasmapheresis, oral prednisolone, and oral cyclophosphamide. Plasma exchange (50 mL/kg to a maximum of 4 L) was performed by using a centrifugal cell separator daily for at least 14 days or until anti-GBM antibody was undetectable. Human albumin (5%) with added calcium and potassium was used as replacement fluid and FFP (150 mL to 300 mL at the end of the exchange) was used in patients with recent surgery or renal biopsy and those with pulmonary hemorrhage. Overall patient survival was 81% at 1 year of follow-up (95% in those with creatinine less than 5.7 mg/dL and 65% in those who presented with dialysis-dependent renal failure). In patients who presented with a creatinine concentration >5.7 mg/dL (but did not require immediate dialysis), renal survival was 82% at 1 year and 69% at last follow-up. There was only 8% renal survival in patients presenting with dialysis dependent renal failure, and all patients who required immediate dialysis and had 100% crescents on renal biopsy remained dialysis dependent. In conclusion, all patients with anti-GBM antibody disease and severe renal failure who do not require immediate dialysis should be immunosuppressed and immediately receive intensive plasmapheresis. Because pulmonary hemorrhage is associated with high mortality, plasmapheresis should be initiated in all Goodpasture patients with pulmonary hemorrhage regardless of the severity of the renal failure.

Most patients with anti-GBM disease have no problems with subsequent renal transplantation although up to 50% may show linear IgG staining of the glomerular basement membrane.[31] The delay of renal transplantation for 12 months after the disappearance of anti-GBM antibodies and the immunosuppression required to maintain a functioning renal allograft are thought to be the main reasons why recurrences are very rare.[32]

Rapidly Progressive Glomerulonephritis

Rapidly progressive glomerulonephritis (RPGN) is characterized by rapid deterioration in renal function occurring over a period ranging from a few days to a few weeks. Untreated RPGN leads to end-stage renal insufficiency. It is characterized histologically by severe inflammation and necrosis of the glomeruli and, frequently, by glomerular crescent formation (crescentic glomerulonephritis, GN). There are three major subgroups: (1) anti-GBM disease and Goodpasture syndrome (discussed earlier); (2) immune complex mediated processes where there is immune-deposition usually resulting from auto-immune processes such as systemic lupus erythematosus, post-infectious processes, mixed cyroglobulinemia, and IgA nephropathy, and (3) pauci-immune diseases that are most often (about 80% of patients) associated anti-neutrophil cytoplasmic antibodies (ANCA; known as Wegener's granulomatosis [WG] or microscopic polyarteritis [MP]). A therapeutic role for plasmapheresis in anti-GBM disease is discussed earlier, and there is limited data for its role in immune complex-mediated processes (see later). The rationale for using plasmapheresis in pauci-immune ANCA-associated diseases (Wegener and microscopic polyarteritis) was initially based on the similarity of the renal pathology to Goodpasture disease. The first use of plasmapheresis for the management of Wegener RPGN was reported in 1977 when the combination of plasmapheresis, oral prednisolone, and cyclophosphamide was associated with rapid renal recovery in five of nine patients.[33]

However, several studies through the 1990s have not demonstrated an additional benefit for the use of plasmapheresis in the treatment of ANCA-associated diseases. For example, The Hammersmith Hospital reported a controlled trial of plasmapheresis in focal necrotizing glomerulonephritis on 48 patients randomized to conventional treatment with oral steroids, cyclophosphamide, and azathioprine with or without intensive plasmapheresis (at least five exchanges in the first 7 days). There was no benefit in patients with moderate or severe renal disease who were not dialysis dependent at presentation.[34] The Canadian Apheresis Study Group randomized 32 patients with non anti-GBM RPGN to receive intravenous methylprednisolone, followed by oral prednisolone, and azathioprine with or without plasmapheresis (10 exchanges in the first 16 days). Again, there was no demonstrable benefit of plasmapheresis in the non-dialysis dependent patients. However, a non-significant trend in benefit was seen in the dialysis dependent patients with three of four patients receiving plasmapheresis coming off dialysis compared with only two of seven in the control group.[35] In 62 patients with Churg-Strauss syndrome or polyarteritis nodosa, there was no additional benefit in patients who received plasmapheresis in addition to cyclophosphamide and steroids.[36] More recently, a multicenter study of 39 patients with RPGN randomized in two groups to receive immunosuppression alone or immunosuppression and plasmapheresis showed no additional benefit for plasmapheresis on renal or patient survival.[37] Recently, the results from the multicenter European Vasculitis Study Group were reported.[38] This randomized controlled clinical trial compared plasma exchange with intravenous methylprednisolone in ANCA-associated vasculitis in patients with severe renal involvement (creatinine > 500 mmol/L or 5.7 mg/dL). All patients received oral cyclophosphamide for 3 months followed by azathioprine. Treatment with plasmapheresis was associated with lower incidence of dialysis dependence at 12 months. Taken together these studies provide strong support for additional therapy with plasmapheresis in patients with severe ANCA-associated glomerulonephritis and advanced renal failure.[39] In patients presenting without acute renal failure, there is less evidence for a beneficial role of plasmapheresis. However, plasmapheresis should be initiated in any patient with ANCA-associated disease that co-presents with anti-GBM disease, and in any patient with diffuse pulmonary alveolar hemorrhage. [40] [41] [42]

Lupus Nephritis

Renal disease is a common and potentially serious complication of systemic lupus erythematosus. Class IV lupus nephritis is associated with poor renal outcomes unless treated with immunosuppressive protocols that utilize cytotoxic therapy.

In 1992, the Lupus Nephritis Collaborative Study Group published a large, randomized, controlled multicenter trial comparing a standard therapy regimen of prednisone and cyclophosphamide with a regimen of standard therapy plus plasmapheresis in patients with severe lupus nephritis. Forty-six patients were randomized to the standard therapy group and 40 were randomized to the plasmapheresis group. Histological categories included lupus nephritis types III, IV, and V. Plasmapheresis was carried out three times per week for 4 weeks, and drug therapy was standardized. The mean follow-up was 136 weeks. Although patients treated with plasmapheresis had a significantly more rapid reduction of serum concentrations of antibodies to double-stranded DNA and cryoglobulins, the addition of plasmapheresis did not improve clinical outcome. Renal failure developed in 8 of 46 patients (17%) in the standard therapy group compared with 10 of 40 patients (25%) in the plasmapheresis group, and 6 of 46 (13%) died in the standard therapy group compared with 8 of 40 (20%) in the plasmapheresis group. Results were unchanged after an extended follow-up of 277 weeks. These results show that the addition of plasmapheresis to a regimen of oral cyclophosphamide and oral corticosteroids confers no additional clinical benefit in patients with severe lupus nephritis despite the rapid fall of serum anti-DNA antibodies.[43]

Another small trial had 18 patients randomly assigned to receive either six courses of intravenous cyclophosphamide over 8 months along with prednisone or three daily plasmaphereses prior to each of these courses of cyclophosphamide. In each group, 2 of 9 patients developed end-stage renal disease, and 3 of 9 patients went into renal remission at 24 months.[44] However, selected patients with aggressive lupus nephritis might benefit from plasmapheresis. We have utilized it for patients with lupus nephritis relapsing on conventional therapy, and an uncontrolled study of 14 patients with severe lupus utilized plasmapheresis and pulse intravenous cyclophosphamide followed by oral cyclophosphamide and prednisone.[45] All 14 patients responded and eight remained off therapy for 5 to 6 years. One patient had a major relapse and two others had a minor relapse at 2 and 3 years. The main side effects were herpes zoster and four women developed irreversible amenorrhea.

Mixed Cryoglobulinemia

Cryoglobulinemia refers to the presence of serum proteins that precipitate at temperatures below 37°C and re-dissolve on rewarming. More than 80% of patients with mixed cryoglobulinemia are infected by hepatitis C virus (HCV) and cryoglobulinemia is found in all patients with HCV-related membranoproliferative glomerulonephritis. HCV-associated cryoglobulinemia and glomerulonephritis is related to the glomerular deposition of immune complexes. The renal manifestations may range from isolated proteinuria to overt nephritic or nephrotic syndrome with variable progression toward chronic renal insufficiency.[46] There are no randomized, controlled studies of plasmapheresis for cryoglobulinemia. However, the goal of removing pathogenic cryoglobulins is rational, and there are numerous successful case reports and uncontrolled studies showing that plasmapheresis may benefit the treatment of severe, active disease as manifested by progressive renal failure, coalescing purpura, or advanced neuropathy. [47] [48] One approach to the management of severe acute flares of cryoglobulinemia with glomerulonephritis or vasculitis is combination of antiviral therapy using peginterferon and ribavirin for 48 weeks, adding corticosteroids and cyclophosphamide as needed to control severe symptoms. In the most severe cases, plasmapheresis (exchanges of 3 L of plasma 3 to 4 times per week for 2 to 3 weeks) can be helpful. Uncontrolled studies with more than five patients showed that plasmapheresis induced rapid reduction in the cryocrit, improved renal function in 55 to 87% of patients, and improved survival (≈25% mortality rate) compared with historical data (≈55% mortality rate).[6] The unique characteristics of cryogobulins has led to modifications of the plasmapheresis technique to enhance their removal. Cryofiltration cools the plasma in an extracorporeal circuit allowing for more efficient removal of the pathogenic proteins. However, this technique is most efficiently performed by a continuous process requiring a specialized machine designed for this purpose. Alternatively, one can perform a two-step procedure in which the patient's own plasma can be reinfused after incubation in the cold to precipitate out the abnormal proteins.[49]

Renal Failure Associated with Multiple Myeloma

Renal disease is a common problem in multiple myeloma with 20%-50% of patients having a plasma creatinine concentration above 1.5 mg/dL (133 mmol/L). Renal impairment can be caused by a variety of factors, including precipitation of myeloma light chains within renal tubules and direct toxicity to tubule epithelium. Other factors frequently implicated include hypercalcemia, hyperuricemia, amyloidosis, hyperviscosity, infections, and chemotherapeutic agents. The removal of nephrotoxic Bence Jones proteins with plasmapheresis could prevent the development of renal failure, but a recent report failed to document significant reductions in patients with elevated serum free light chains.[50] An early study of 29 patients with acute renal failure and multiple myeloma included 24 patients on dialysis and five with a creatinine concentration > 5 mg/dL. The patients were randomly divided into two groups: 15 patients received plasmapheresis plus standard therapy and 14 patients standard therapy alone. In the plasmapheresis group, 13 of 15 patients recovered renal function (Cr < 2.5 mg/dl) versus 2 of 14 in the control group.[51] However, Johnson and colleagues[52] reported no difference in patient survival or in recovery of renal function in a study of 21 participants who were randomly assigned either to plasmapheresis plus chemotherapy or to chemotherapy alone. The mortality at 6 months was 20% in each group, which increased to 60% to 80% at 12 months.[52] In the largest study to date, 104 patients with multiple myeloma and acute renal failure were randomly assigned to conventional therapy plus five to seven plasma exchanges of 50 mL per kg of body weight of 5% human serum albumin for 10 days or conventional therapy alone. The primary end point (death, dialysis, or GFR less than 30 ml/min) occurred in 33 of 57 (57.9%) patients in the plasmapheresis group and in 27 of 39 (69.2%) patients in the control group.[53] Although no benefit was demonstrated, questions remain about whether subgroups of patients may benefit from plasmapheresis. Nevertheless, these studies suggest caution in considering plasmapheresis for the acute renal failure associated with multiple myeloma.[54]

Thrombotic Thrombocytopenic Purpura and Hemolytic Uremic Syndrome

These clinical disorders share a spectrum of abnormalities in numerous organ systems and are characterized by the presence of thrombocytopenia and microangiopathic hemolytic anemia. When the prominent features are hemolytic anemia, thrombocytopenia, and advanced renal failure, the syndrome is typically labeled hemolytic uremic syndrome (HUS). If there is a finding of neurologic symptoms with fever and perhaps less severe renal failure, the syndrome is classically labeled TTP. However, these labels are artificial and both syndromes are characterized with pathologic changes of endothelial injury and platelet microthrombi. With two exceptions, the causes for this disorder remain unknown and can be see as complications of drug therapy (mitomycin, cyclosporine, ticlopidine), autoimmune disorders (systemic lupus erythematosus [SLE], anti-phospholipid antibody syndrome), and in pregnancy. One exception is the HUS syndrome associated with hemorrhagic diarrhea caused by E Coli O157:H7. In this disease, the toxin induces colonic vascular injury leading to systemic absorption and activation of numerous pathways leading to endothelial cell damage. Platelet microthrombi are particularly prominent in the glomerular capillaries often leading to severe renal failure. Although the disease is often self-limited in children and a role for plasmapheresis is not clear, plasmapheresis is often utilized for severe or persistent disease (especially in adults). In the largest uncontrolled trial of an outbreak in Scotland, 22 patients were identified with E Coli O157:H7.[55] Plasmapheresis could only be performed in 16 patients, and five deaths occurred (31%), whereas five of six patients (83%) who did not receive plasmapheresis died of the disease. There is evidence that plasmapheresis improves outcomes with HUS-TTP resulting from ticlopidine (50% versus 24% mortality[56]), but no literature supporting a beneficial role for plasmapheresis in patients with HUS-TTP secondary to cancer chemotherapy or bone marrow transplantation.[57]

The mechanism of some forms of TTP is now partially understood and reveals how plasmapheresis with plasma exchange is beneficial. Genetic studies of congenital TTP led to the identification of defects in a metalloproteinase called von Willebrand factor-cleaving protease (now called ADAMTS13: A Disintegrin-like And Metalloprotease with ThromboSpondin type 1 repeats).[58] TTP has been shown to occur in the setting of accumulation of unusually large von Willebrand factor (ULVWf). Multimers of VWf normally accumulate on the endothelial cell membrane and are rapidly cleaved into normal-sized multimers by the ADAMTS13 protease. In some patients, ADAMTS13 deficiency leads to accumulation of unusually large VWf multimers resulting in platelet microthrombus formation and subsequent microangiopathic hemolytic anemia. An inhibitory autoantibody to the ADAMTS13 metalloproteinase has been found at varying titers among a high percentage of patients with the idiopathic form of this disease. [59] [60] By removing autoantibodies to ADAMTS13 and replacing with normal plasma (containing ADAMTS13 activity), plasmapheresis can reverse the TTP syndrome caused by ADAMTS13 deficiency. However, ADAMTS13 deficiency may be necessary but is not sufficient to account for many cases of TTP. Furthermore, enzyme activity is significantly reduced in numerous other conditions including infection, cancer, cirrhosis, uremia, SLE, disseminated intravascular coagulopathy (DIC), and many others.

Prior to the introduction of plasma infusion and plasmapheresis, the disease typically progressed rapidly and was almost uniformly fatal (90% fatality rate).[61] In 1977 it was discovered that infusion of FFP or plasmapheresis with FFP replacement was able to reverse the course of disease. [62] [63] The efficacy of plasma exchange in the management of TTP-HUS in adults has been demonstrated in two trials that included 210 patients. [64] [65] Plasma exchange with fresh frozen plasma was more effective than plasma infusion alone. At 6 months, the remission rate was 78% versus 31% and survival rate with these two procedures 78% versus 50%. Patients treated with plasma exchange received approximately three times as much plasma as those treated with plasma infusion alone (where the amount of plasma administration was limited by the risk of volume overload). Therefore, it is possible that the benefit seen with plasma exchange may have been due to infusion of more plasma rather than to the removal of a toxic substance.

The duration of plasmapheresis in HUS-TTP is not known, but is performed daily until the platelet count has risen to normal and evidence for hemolysis (schistocytes, LDH elevation) has resolved.[57] There is a wide range of required exchanges (3–145) reported with an average of 7 to 16 exchanges necessary to induce remission. [57] [61] [64] [65] The American Association of Blood Banks recommends daily plasmapheresis until the platelet count is above 150,000/l for 2 to 3 days, and The American Society for Apheresis recommends daily PE until the platelet count is above 100,000/l and LDH level near to normal.[66] When present, neurologic symptoms improve quickly and the serum LDH tends to improve over the first 1 to 3 days. Improvement in the platelet count may not be seen for several days, and improvements in renal function often take longer. Patients requiring dialysis at presentation may be able to recover enough function to discontinue dialysis, but many patients have residual chronic kidney disease. When a normal platelet count has been achieved, plasma exchange is gradually tapered by increasing the interval between treatments. This allows more efficient treatment until a durable remission is achieved. Many patients (one third to one half) will abruptly develop thrombocytopenia and increased evidence for hemolysis when daily plasma exchanges are tapered or stopped. These patients may benefit from the addition of immunosuppressive therapy (cyclosporine, prednisone) although there is little data validating any benefits.

Recurrent Focal Segmental Glomerulosclerosis

Focal segmental glomerulosclerosis (FSGS) is a common glomerular disease resulting in end-stage renal disease and is the most common disease to reoccur in a renal allograft (20%). The mechanisms of early proteinuria related to FSGS recurrence after transplantation are unclear. The immediate reappearance of proteinuria after transplantation suggests that a non-dialyzable circulating factor may be present altering glomerular permeability. [67] [68] Removal of such a factor by immunoadsorption with protein A columns or plasma ex-change may account for the remission of the disease in some patients.[69] The circulating factor may be a non immunoglobulin with a molecular weight of approximately 30 to 50 kilodaltons, [67] [68] [69] but has not been definitively identified and is found in normal plasma.[69] In some cases, a factor in normal serum may be lost causing increased glomerular permeability. [70] [71] [72]Variable levels of this constituent of normal plasma may explain the discrepant results observed among studies evaluating the circulating factor. Early studies showed benefits of plasmapheresis in FSGS but early relapse. Better results were obtained when the number of plasmapheresis sessions was increased to nine.[73] Six of nine patients treated within 1 week of the onset of proteinuria had a mean reduction in protein excretion from 11.5 down to 0.8 g/day and these remissions were sustained for up to 27 months.

In recurrent FSGS after renal transplantation, beneficial results have been reported in children treated with plasmapheresis and cyclophosphamide. In one study of 11 children with recurrent FSGS post-transplant, nine were administered plasmapheresis (6 to 10 times over 15 to 24 days). Seven had a persistent remission (follow-up of 32 months).[74] A second study of six children with recurrent FSGS found that plasmapheresis combined with cyclophosphamide resulted in complete or partial remissions all six patients.[75] In adults, there also are no controlled trials, but the outcome of recurrent FSGS in 23 patients with renal allografts (13 patients treated with plasmapheresis and 10 historical controls) showed that after a median follow-up of 3.5 years 2 patients in the plasmapheresis-treated group and 10 in the non-treated had lost their allografts.[76] In these patients with recurrent proteinuria, FSGS recurred within 4 weeks of transplantation (77%) and plasmapheresis was initiated within 14 days of recurrence (85%). Recently, a series of nine patients treated with plasmapheresis found complete or partial remission in eight cases; however, plasma exchange did not preclude graft loss secondary to recurrent FSGS in six patients (66%) and there was no correlation with timing of initiation of plasmapheresis.[77]

Renal Transplantation

Plasmapheresis has been utilized in several different clinical scenarios involving kidney transplantation. These include ABO blood group incompatible transplants, positive T-cell cross-match, acute humoral rejection, and recurrent focal glomerulosclerosis (FSGS) in the transplant (discussed earlier).

ABO Incompatible Kidney Transplantation

ABO-incompatible kidney transplantation is an increasingly attractive option for renal allograft candidates whose only living donor is blood group incompatible. The ABO blood group consists of four common categories (A, B, AB, and O), with types A and O most frequently found in the US population. The proteins comprising these antigens are on red blood cells, lymphocytes, and platelets, in addition to epithelial and endothelial cells. Historically, blood group incompatibility was an exclusion for renal transplantation because blood group antibodies arise against those antigens not native to the host. For example, antibodies to both A and B antigens are found in an individual with blood type O (no A, B antigens), whereas an individual with blood type AB has no preformed antibodies to A or B antigens. Mismatched renal transplantation in this situation can lead to acute humoral rejection. However, reported outcomes with ABO-incompatible renal transplants from Japan have been excellent,[78] and these results have kindled renewed interest in ABO-incompatible renal transplantation with a variety of desensitization protocols. The outcomes of 441 ABO-incompatible kidney transplants performed at 55 centers across Japan showed no significant difference in graft survival when compared with historic recipients of ABO-compatible living donor organs. This is the largest study to date, and provides follow-up as long as 9 years.

Current desensitization protocols utilize intravenous immune globulin (IVIG) with and without plasmapheresis to remove IgG and IgM antibodies against the ABO group of the potential recipient. These treatments are done along with a variety of immunosuppression protocols using tacrolimus, mycophenolate and steroids, and occasionally splenectomy (see Chapter 65 for more details). In one study, 10 of 18 patients receiving an ABO-incompatible kidney received pre-transplant plasmapheresis (in addition to Thymoglobulin antibody induction, tacrolimus, mofetil, and prednisone). There were no episodes of hyperacute rejection, and antibody-mediated rejection occurred in 28% of ABO-incompatible recipients, but was reversible with plasmapheresis, intravenous immunoglobulin, and increasing immunosuppression in all patients except one. [79] [80] Recently, a comparison of plasmapheresis versus high-dose IVIG desensitization was reported.[81] This study compared high-dose IVIG with two plasmapheresis protocols and found that patients who received plasmapheresis were more likely to achieve a negative cross-match. However, the number of patients was small (13–32 in each group), the study was not randomized, and there were other differences in treatment that limit comparisons among the groups.

Positive T-cell Cross-Match

The most recent data available in the United States indicates that about 20% of patients awaiting kidney transplanta-tion have reactive antibodies. Primary sensitization results from exposure to foreign HLA antigens via transplantation, transfusion, and/or pregnancy. Patients who have antibodies reactive with donor lymphocytes are at increased risk for hyperacute or acute antibody-mediated rejection and graft loss. There are currently two HLA antibody reduction protocols for which efficacy has been demonstrated: high-dose IVIG and plasmapheresis combined with IVIG. Plasmapheresis is used to remove anti-HLA antibodies and is followed by infusion of low doses of IgG during hemodialysis. The hypothesis is that low-dose IgG will have additional beneficial immunomodulating effects. At the same time that plasmapheresis is started, patients begin treatment with tacrolimus, ± steroids (and antimicrobial prophylaxis). Plasmapheresis is continued thrice weekly until the T-cell CDC crossmatch is negative; transplant then takes place within 24 hours. Plasmapheresis and low-dose IgG are usually repeated several times during the first two post-transplant weeks to remove any rebounding antibody. Plasmapheresis-based protocols are usually not suitable for highly sensitized patients awaiting deceased donor transplantation because the availability of suitable organs is unpredictable and plasmapheresis is both difficult and very expensive to continue indefinitely. As soon as it is stopped, anti-HLA antibody titers will rebound. [82] [83]

Acute Humoral Rejection

Acute humoral rejection is characterized by a severe allograft dysfunction associated with the presence of circulating donor-specific antibodies. Very poor outcomes are seen with acute humoral rejection and treatment with pulse steroids and antilymphocyte therapy is often ineffective.[84] Removal of the donor specific antibodies with plasmapheresis has been successful when combined with tacrolimus and mofetil.[85] It is now proposed that the combination of plasmapheresis and IVIG may lead to short-term recovery from acute antibody-mediated rejection in more than 80% of cases. [86] [87]


The use of plasmapheresis to manage a variety of kidney diseases has grown significantly in recent years. In some cases, the rationale and benefit are supported by clinical studies, but in many cases the benefits are not well established. Nevertheless, the concept of removing plasma containing pathogenic antibodies is now well established for some renal (as in Goodpasture disease) and non-renal conditions (primary autoimmune autonomic failure[88]). Additional studies are needed to determine the potential benefits for plasmapheresis in these other conditions.


1. Kambic HE, Nose Y: Historical perspective on plasmapheresis.  Ther Apher  1997; 1(1):83-108.

2. Abel JJ, Rowntree LG, Turner BB: Plasma removal with return of corpuscles (plasmaphaeresis). The Journal of Pharmacology and Experimental Therapeutics Vol. V. No. 6, July, 1914.  Transfus Sci  1990; 11(2):166-177.

3. Abel JJ, Rowntree LG, Turner BB: On the removal of diffusable substances from the circulating blood by means of dialysis. Transactions of the Association of American Physicians, 1913.  Transfus Sci  1990; 11(2):164-165.

4. Schwab PJ, Fahey JL: Treatment of Waldenstrom's macroglobulinemia by plasmapheresis.  N Engl J Med  1960; 263:574-579.

5. Zarkovic M, Kwaan HC: Correction of hyperviscosity by apheresis.  Semin Thromb Hemost  2003; 29(5):535-542.

6. Madore F: Plasmapheresis. Technical aspects and indications.  Crit Care Clin  2002; 18(2):375-392.

7. Madore F, Lazarus JM, Brady HR: Therapeutic plasma exchange in renal diseases.  J Am Soc Nephrol  1996; 7(3):367-386.

8. Reimann PM, Mason PD: Plasmapheresis: Technique and complications.  Intensive Care Med  1990; 16(1):3-10.

9. Stegmayr BG: A survey of blood purification techniques.  Transfus Apher Sci  2005; 32(2):209-220.

10. Stegmayr B, Wikdahl AM: Access in therapeutic apheresis.  Ther Apher Dial  2003; 7(2):209-214.

11. Ziai F, Benesch T, Kodras K, et al: The effect of oral anticoagulation on clotting during hemodialysis.  Kidney Int  2005; 68(2):862-866.

12. Linenberger ML, Price TH: Use of cellular and plasma apheresis in the critically ill patient: Part 1: technical and physiological considerations.  J Intensive Care Med  2005; 20(1):18-27.

13. Hassell K: The management of patients with heparin-induced thrombocytopenia who require anticoagulant therapy.  Chest  2005; 127(2 Suppl):1S-8S.

14. Lowentritt JE, Kahn MJ, Batuman V: Use of recombinant hirudin as an anticoagulant in plasmapheresis in a patient with heparin-induced thrombocytopenia.  Nephrol Dial Transplant  2002; 17(8):1533-1534.

15. Norda R, Stegmayr BG: Apheresis registry in Sweden: scope, techniques and indications for treatment. A report from the Swedish apheresis study group.  Transfus Apher Sci  2001; 24(1):49-55.

16. Pohl MA, Lan SP, Berl T: Plasmapheresis does not increase the risk for infection in immunosuppressed patients with severe lupus nephritis. The Lupus Nephritis Collaborative Study Group.  Ann Intern Med  1991; 114(11):924-929.

17. Owen HG, Brecher ME: Atypical reactions associated with use of angiotensin-converting enzyme inhibitors and apheresis.  Transfusion  1994; 34(10):891-894.

18. Lockwood CM, Worlledge S, Nicholas A, et al: Reversal of impaired splenic function in patients with nephritis or vasculitis (or both) by plasma exchange.  N Engl J Med  1979; 300(10):524-530.

19. Kalluri R, Wilson CB, Weber M, et al: Identification of the alpha 3 chain of type IV collagen as the common autoantigen in antibasement membrane disease and Goodpasture syndrome.  J Am Soc Nephrol  1995; 6(4):1178-1185.

20. Pusey CD: Anti-glomerular basement membrane disease.  Kidney Int  2003; 64(4):1535-1550.

21. Lockwood CM, Boulton-Jones JM, Lowenthal RM, et al: Recovery from Goodpasture's syndrome after immunosuppressive treatment and plasmapheresis.  Br Med J  1975; 2(5965):252-254.

22. Bouget J, Le Pogamp P, Perrier G, et al: [Anti-basement-membrane antibody mediated, rapidly progressive, glomerulonephritis.  Diagnostic and therapeutic strategy based on a retrospective study of 14 cases.] Ann Med Interne (Paris)  1990; 141(5):409-415.

23. Herody M, Bobrie G, Gouarin C, et al: Anti-GBM disease: Predictive value of clinical, histological and serological data.  Clin Nephrol  1993; 40(5):249-255.

24. Merkel F, Pullig O, Marx M, et al: Course and prognosis of anti-basement membrane antibody (anti-BM-Ab)-mediated disease: Report of 35 cases.  Nephrol Dial Transplant  1994; 9(4):372-376.

25. Andrews PA, Sheerin NS, Hicks JA, et al: Unusual presentations of anti-glomerular basement membrane antibody mediated disease are associated with delayed diagnosis and poor outcome.  Clin Nephrol  1995; 44(4):262-265.

26. Daly C, Conlon PJ, Medwar W, et al: Characteristics and outcome of anti-glomerular basement membrane disease: A single-center experience.  Ren Fail  1996; 18(1):105-112.

27. Levy JB, Turner AN, Rees AJ, et al: Long-term outcome of anti-glomerular basement membrane antibody disease treated with plasma exchange and immunosuppression.  Ann Intern Med  2001; 134(11):1033-1042.

28. Saurina A, Arrizabalaga P, Rodriguez R, et al: [Anti-glomerular basement membrane antibody mediated disease: Revision of 32 cases and follow-up at one year of the diagnosis.].  Nefrologia  2003; 23(5):415-422.

29. Simpson IJ, Doak PB, Williams LC, et al: Plasma exchange in Goodpasture's syndrome.  Am J Nephrol  1982; 2(6):301-311.

30. Johnson JP, Moore Jr J, Austin 3rd HA, et al: Therapy of anti-glomerular basement membrane antibody disease: Analysis of prognostic significance of clinical, pathologic and treatment factors.  Medicine (Baltimore)  1985; 64(4):219-227.

31. Kotanko P, Pusey CD, Levy JB: Recurrent glomerulonephritis following renal transplantation.  Transplantation  1997; 63(8):1045-1052.

32. Khandelwal M, McCormick BB, Lajoie G, et al: Recurrence of anti-GBM disease 8 years after renal transplantation.  Nephrol Dial Transplant  2004; 19(2):491-494.

33. Lockwood CM, Pinching AJ, Sweny P, et al: Plasma-exchange and immunosuppression in the treatment of fulminating immune-complex crescentic nephritis.  Lancet  1977; 1(8002):63-67.

34. Pusey CD, Rees AJ, Evans DJ, et al: Plasma exchange in focal necrotizing glomerulonephritis without anti-GBM antibodies.  Kidney Int  1991; 40(4):757-763.

35. Cole E, Cattran D, Magil A, et al: A prospective randomized trial of plasma exchange as additive therapy in idiopathic crescentic glomerulonephritis. The Canadian Apheresis Study Group.  Am J Kidney Dis  1992; 20(3):261-269.

36. Guillevin L, Lhote F, Cohen P, et al: Corticosteroids plus pulse cyclophosphamide and plasma exchanges versus corticosteroids plus pulse cyclophosphamide alone in the treatment of polyarteritis nodosa and Churg-Strauss syndrome patients with factors predicting poor prognosis. A prospective, randomized trial in sixty-two patients.  Arthritis Rheum  1995; 38(11):1638-1645.

37. Zauner I, Bach D, Braun N, et al: Predictive value of initial histology and effect of plasmapheresis on long-term prognosis of rapidly progressive glomerulonephritis.  Am J Kidney Dis  2002; 39(1):28-35.

38. de Lind van Wijngaarden RA, Hauer HA, Wolterbeek R, et al: Clinical and histologic determinants of renal outcome in ANCA-associated vasculitis: A prospective analysis of 100 patients with severe renal involvement.  J Am Soc Nephrol  2006; 17(8):2264-2274.

39. Buhaescu I, Covic A, Levy J: Systemic vasculitis: Still a challenging disease.  Am J Kidney Dis  2005; 46(2):173-185.

40. Gallagher H, Kwan JT, Jayne DR: Pulmonary renal syndrome: A 4-year, single-center experience.  Am J Kidney Dis  2002; 39(1):42-47.

41. Klemmer PJ, Chalermskulrat W, Reif MS, et al: Plasmapheresis therapy for diffuse alveolar hemorrhage in patients with small-vessel vasculitis.  Am J Kidney Dis  2003; 42(6):1149-1153.

42. Iwatani H, Uzu T, Kakihara M, et al: A case of Wegener's granulomatosis with pulmonary bleeding successfully treated with double filtration plasmapheresis (DFPP).  Clin Exp Nephrol  2004; 8(4):369-374.

43. Lewis EJ, Hunsicker LG, Lan SP, et al: A controlled trial of plasmapheresis therapy in severe lupus nephritis. The Lupus Nephritis Collaborative Study Group.  N Engl J Med  1992; 326(21):1373-1379.

44. Wallace DJ, Goldfinger D, Pepkowitz SH, et al: Randomized controlled trial of pulse/synchronization cyclophosphamide/apheresis for proliferative lupus nephritis.  J Clin Apher  1998; 13(4):163-166.

45. Euler HH, Guillevin L: Plasmapheresis and subsequent pulse cyclophosphamide in severe systemic lupus erythematosus. An interim report of the Lupus Plasmapheresis Study Group.  Ann Med Interne (Paris)  1994; 145(5):296-302.

46. Ferri C, Sebastiani M, Giuggioli D, et al: Mixed cryoglobulinemia: Demographic, clinical, and serologic features and survival in 231 patients.  Semin Arthritis Rheum  2004; 33(6):355-374.

47. D'Amico G: Renal involvement in hepatitis C infection: Cryoglobulinemic glomerulonephritis.  Kidney Int  1998; 54(2):650-671.

48. Dominguez JH, Sha E: Apheresis in cryoglobulinemia complicating hepatitis C and in other renal diseases.  Ther Apher  2002; 6(1):69-76.

49. Siami GA, Siami FS: Current topics on cryofiltration technologies.  Ther Apher  2001; 5(4):283-286.

50. Cserti C, Haspel R, Howell C, Dzik W: Light chain removal by plasmapheresis in myeloma associated renal failure.  Transfusion  2007; 47:511-514.

51. Zucchelli P, Pasquali S, Cagnoli L, Ferrari G: Controlled plasma exchange trial in acute renal failure due to multiple myeloma.  Kidney Int  1988; 33(6):1175-1180.

52. Johnson WJ, Kyle RA, Pineda AA, et al: Treatment of renal failure associated with multiple myeloma. Plasmapheresis, hemodialysis, and chemotherapy.  Arch Intern Med  1990; 150(4):863-869.

53. Clark WF, Stewart AK, Rock GA, et al: Plasma exchange when myeloma presents as acute renal failure: A randomized, controlled trial.  Ann Intern Med  2005; 143(11):777-784.

54. Gertz MA: Managing myeloma kidney.  Ann Intern Med  2005; 143(11):835-837.

55. Dundas S, Murphy J, Soutar RL, et al: Effectiveness of therapeutic plasma exchange in the 1996 Lanarkshire Escherichia coli O157:H7 outbreak.  Lancet  1999; 354(9187):1327-1330.

56. Bennett CL, Weinberg PD, Rozenberg-Ben-Dror K, et al: Thrombotic thrombocytopenic purpura associated with ticlopidine. A review of 60 cases.  Ann Intern Med  1998; 128(7):541-544.

57. George JN, Li X, McMinn JR, et al: Thrombotic thrombocytopenic purpura-hemolytic uremic syndrome following allogeneic HPC transplantation: A diagnostic dilemma.  Transfusion  2004; 44(2):294-304.

58. Levy GG, Nichols WC, Lian EC, et al: Mutations in a member of the ADAMTS gene family cause thrombotic thrombocytopenic purpura.  Nature  2001; 413(6855):488-494.

59. Rieger M, Mannucci PM, Kremer Hovinga JA, et al: ADAMTS13 autoantibodies in patients with thrombotic microangiopathies and other immunomediated diseases.  Blood  2005; 106(4):1262-1267.

60. Zheng XL, Kaufman RM, Goodnough LT, Sadler JE: Effect of plasma exchange on plasma ADAMTS13 metalloprotease activity, inhibitor level, and clinical outcome in patients with idiopathic and nonidiopathic thrombotic thrombocytopenic purpura.  Blood  2004; 103(11):4043-4049.

61. Ruggenenti P, Noris M, Remuzzi G: Thrombotic microangiopathy, hemolytic uremic syndrome, and thrombotic thrombocytopenic purpura.  Kidney Int  2001; 60(3):831-846.

62. Bukowski RM, King JW, Hewlett JS: Plasmapheresis in the treatment of thrombotic thrombocytopenic purpura.  Blood  1977; 50(3):413-417.

63. Byrnes JJ, Khurana M: Treatment of thrombotic thrombocytopenic purpura with plasma.  N Engl J Med  1977; 297(25):1386-1389.

64. Bell WR, Braine WG, Ness PM, Kickler TS: Improved survival in thrombotic thrombocytopenic purpura-hemolytic uremic syndrome. Clinical experience in 108 patients.  N Engl J Med  1991; 325(6):398-403.

65. Rock GA, Shumak KH, Buskard NA, et al: Comparison of plasma exchange with plasma infusion in the treatment of thrombotic thrombocytopenic purpura. Canadian Apheresis Study Group.  N Engl J Med  1991; 325(6):393-397.

66. Rock GA: Management of thrombotic thrombocytopenic purpura.  Br J Haematol  2000; 109(3):496-507.

67. Savin VJ, Sharma R, Sharma M, et al: Circulating factor associated with increased glomerular permeability to albumin in recurrent focal segmental glomerulosclerosis.  N Engl J Med  1996; 334(14):878-883.

68. Sharma M, Sharma R, McCarthy ET, Savin VJ: “The FSGS factor:” Enrichment and in vivo effect of activity from focal segmental glomerulosclerosis plasma.  J Am Soc Nephrol  1999; 10(3):552-561.

69. Vincenti F, Ghiggeri GM: New insights into the pathogenesis and the therapy of recurrent focal glomerulosclerosis.  Am J Transplant  2005; 5(6):1179-1185.

70. Coward RJ, Foster RR, Patton D, et al: Nephrotic plasma alters slit diaphragm-dependent signaling and translocates nephrin, Podocin, and CD2 associated protein in cultured human podocytes.  J Am Soc Nephrol  2005; 16(3):629-637.

71. Ghiggeri GM, Bruschi M, Candiano G, et al: Depletion of clusterin in renal diseases causing nephrotic syndrome.  Kidney Int  2002; 62(6):2184-2194.

72. Sharma R, Sharma M, McCarthy ET, et al: Components of normal serum block the focal segmental glomerulosclerosis factor activity in vitro.  Kidney Int  2000; 58(5):1973-1979.

73. Artero ML, Sharma R, Savin VJ, Vincenti F, et al: Plasmapheresis reduces proteinuria and serum capacity to injure glomeruli in patients with recurrent focal glomerulosclerosis.  Am J Kidney Dis  1994; 23(4):574-581.

74. Dall'Amico R, Ghiggeri G, Carraro M, et al: Prediction and treatment of recurrent focal segmental glomerulosclerosis after renal transplantation in children.  Am J Kidney Dis  1999; 34(6):1048-1055.

75. Cheong HI, Han HW, Park HW, et al: Early recurrent nephrotic syndrome after renal transplantation in children with focal segmental glomerulosclerosis.  Nephrol Dial Transplant  2000; 15(1):78-81.

76. Deegens JK, Andresdottir MB, Croockewit S, Wetzels JF: Plasma exchange improves graft survival in patients with recurrent focal glomerulosclerosis after renal transplant.  Transpl Int  2004; 17(3):151-157.

77. Pardon A, Audard V, Caillard S, et al: Risk factors and outcome of focal and segmental glomerulosclerosis recurrence in adult renal transplant recipients.  Nephrol Dial Transplant  2006; 21(4):1053-1059.

78. Takahashi K, Saito K, Takahara S, et al: Excellent long-term outcome of ABO-incompatible living donor kidney transplantation in Japan.  Am J Transplant  2004; 4(7):1089-1096.

79. Gloor JM, Lager DJ, Moore SB, et al: ABO-incompatible kidney transplantation using both A2 and non-A2 living donors.  Transplantation  2003; 75(7):971-977.

80. Stegall MD, Dean PG, Gloor JM: ABO-incompatible kidney transplantation.  Transplantation  2004; 78(5):635-640.

81. Stegall MD, Gloor J, Winters JL, et al: A comparison of plasmapheresis versus high-dose IVIG desensitization in renal allograft recipients with high levels of donor specific alloantibody.  Am J Transplant  2006; 6(2):346-351.

82. Magee CC: Transplantation across previously incompatible immunological barriers.  Transpl Int  2006; 19(2):87-97.

83. Montgomery RA, Zachary AA: Transplanting patients with a positive donor-specific crossmatch: A single center's perspective.  Pediatr Transplant  2004; 8(6):535-542.

84. Crespo M, Pascual M, Tolkoff-Rubin N, et al: Acute humoral rejection in renal allograft recipients: I. Incidence, serology and clinical characteristics.  Transplantation  2001; 71(5):652-658.

85. Pascual M, Saidman S, Tolkoff-Rubin N, et al: Plasma exchange and tacrolimus-mycophenolate rescue for acute humoral rejection in kidney transplantation.  Transplantation  1998; 66(11):1460-1464.

86. Abraham KA, Brown C, Conlon PJ, et al: Plasmapheresis as rescue therapy in accelerated acute humoral rejection.  J Clin Apher  2003; 18(3):103-110.

87. Ibernon M, Gil-Vernet S, Carrera M, et al: Therapy with plasmapheresis and intravenous immunoglobulin for acute humoral rejection in kidney transplantation.  Transplant Proc  2005; 37(9):3743-3745.

88. Schroeder C, Vernino S, Birkenfield AL, et al: Plasma exchange for primary autoimmune autonomic failure.  N Engl J Med  2005; 353(15):1585-1590.