Use of continuous venovenous hemofiltration for acute renal failure due to multiple myeloma cast nephropathy

Abstract

Renal insufficiency is associated with high morbidity and mortality in multiple myeloma. One of the common causes for acute renal failure in multiple myeloma is cast nephropathy. It is important to reduce the levels of light chains to improve renal failure and also the overall outcome. Plasmapheresis has failed to show any significant improvement in renal failure due to cast nephropathy as demonstrated in a recent randomized control trial. Here, we present a case series of three patients who were treated with continuous venovenous hemofiltration as a modality to remove these free light chains. There was improvement in renal failure in these patients with decrease in the levels of free light chains. These patients remained off hemodialysis on follow-up and two of them were able to undergo hematopoietic stem cell transplantation.

Keywords:

• Multiple myeloma
• Cast nephropathy
• Renal failure
• CVVH

Introduction

Renal insufficiency is present in almost half of multiple myeloma patients at presentation and is associated with a high morbidity and mortality.^1,2 The reversal of renal impairment is critical not only to prevent complications of uremia but also it was shown to be an important prognostic factor than response to chemotherapy and improves long-term survival.¹ Cast nephropathy accounted for 40–63% of the renal lesions in multiple myeloma patients in renal biopsy series.^3,4 These patients typically present with advanced renal failure. A recent study with plasma exchange showed that improvement in renal function with treatment only occurred if there was a significant clearance of light chains.⁴ The free light chains (FLC) found in myeloma are relatively small in size (kappa 25 kD and lambda 50 kD) and are distributed in similar concentrations within different body compartments.⁵ These characteristics have led to investigation of different methods to remove the FLC and hence improve renal survival.⁵ In small trials, plasmapheresis has been shown to remove the FLC,^6,7 but a recent randomized trial failed to show any significant improvement as compared to a control group.^2,8 The use of new high cutoff dialyzers (HCO) has been studied and noted to have benefit in removing FLC and improvement in renal function.⁵ We present three cases of multiple myeloma with acute renal failure caused by cast nephropathy. These patients were treated with continuous venovenous hemofiltration (CVVH) to remove FLC.

Case Presentations

Case 1

The first patient is a 38-year-old man with past medical history of colitis, who presented to his primary doctor with complaint of weakness and nausea. He was found to be in acute renal failure with a creatinine of 5.8 mg/dl and anemic with a hemoglobin of 10.3 mg/dl, and was referred to a nephrologist for further work-up. Both serum immunofixation electrophoresis and urine immunofixation electrophoresis showed monoclonal lambda light chains (serum IgA lambda 1150 mg/dl and urine IgA lambda 4361 mg/g). The remainder of the serological testing was negative. He had an unremarkable renal ultrasound, although bone survey indicated several thoracic and lumbar lytic lesions. He underwent a renal biopsy which indicated lambda light chain cast nephropathy with mild to moderate interstitial fibrosis and tubular atrophy. Bone marrow examination indicated multiple myeloma IgA lambda, 40% plasma cells that were CD20⁺, ISS stage III, Durie–Salmon stage II B. The patient was started on Bortezomib and dexamethasone for the myeloma and hemodialysis for the renal failure. The patient was subsequently changed to CVVH with a renaflo HF1200 hemofilter on a Baxter Aquarius system for trial to remove FLC. He underwent a total of 5 days of CVVH and three of those days were continuous 24-hour treatment and the rest 18 hours. He was discharged post the fifth treatment off renal replacement therapy (creatinine 3.2 mg/dl). Repeat immunofixation electrophoresis showed an IgA lambda of 520 mg/dl. He later underwent autologous bone marrow transplant which failed, and required an allogenic stem cell transplant. He did well after the second transplant and at follow-up 3 years after diagnosis, the patient was free from multiple myeloma and dialysis independent with a creatinine 2 mg/dl.

Case 2

The second patient is a 55-year-old woman with a past medical history of scoliosis and breast carcinoma, who presented to her primary medical doctor complaining of 2 days of nausea, vomiting, and poor oral intake. She was found to be in acute renal failure with a serum creatinine of 4 mg/dl and anemic with a hemoglobin of 9 mg/dl. Serum immunofixation electrophoresis indicated IgG kappa predominant (2900 mg/dl) and lamda (9.7 mg/dl) myeloma. The remainder of the serological work-up as well as renal imaging was normal. Her skeletal survey demonstrated numerous lytic bone lesions. A bone marrow biopsy indicated 40–50% involvement by plasma cell myeloma, IgG kappa with ISS stage III. She was started on lenalidomide, bortezomib, and dexamethasone. Due to ongoing emesis and progressive renal failure from presumed cast nephropathy (creatinine 6.24 mg/dl), she was admitted to the hospital and started on CVVH via renaflo HF1200 hemofilter on a Baxter Aquarius system and underwent CVVH for 18 hours per day for a total of 5 days. Her symptoms improved and her renal function after CVVH was with a creatinine of 2.43 mg/dl, and she was discharged home off renal replacement therapy. Six weeks later, her creatinine remained 2.3 mg/dl. Because of her renal function, her chemotherapy was switched to Vincristine, Adriamycin, and dexamethasone. She later underwent an autologous bone marrow transplant. At 1 year after CVVH, she is doing well with remission of her myeloma and her creatinine is 1.9 mg/dl and she no longer requires renal replacement therapy. Her free kappa was 73 mg/dl.

Case 3

The third patient is a 54-year-old man who had a history of multiple myeloma which was in remission for 10 years and was found to be anemic. He was found to have serum immunofixation positive for IgG kappa (17 800 mg/dl) and lambda (2 mg/dl). He was started on chemotherapy with bortezomib and dexamethasone. After several doses of chemotherapy, he was found to be in acute renal failure with a serum creatinine of 6.05 mg/dl. He was admitted to the hospital and further serological and imaging tests failed to indicate other causes for his renal failure. On admission, his serum creatinine was 5.9 mg/dl, potassium was 5.2 mg/dl, and calcium was 9.2 mg/dl. He was started on CVVH via renaflo HF1200 hemofilter on a Baxter Aquarius system. He underwent daily CVVH for 18 hours for a total of 5 days. At the end of his treatment, his repeat creatinine was 2.58 mg/dl and IgG kappa 8220 mg/dl. The patient was able to continue his chemotherapy and did not require further renal replacement therapy. He was subsequently hospitalized in a different institution with septic shock and acute renal failure requiring several hemodialysis sessions. His renal function recovered but unfortunately he passed away from other complications.

Discussion

Renal failure associated with multiple myeloma has important ramifications above and beyond simple organ failure. It worsens morbidity and mortality, and interferes with treatment of the underlying disease.^1,2 Aside from limiting chemotherapy, renal failure has traditionally been an obstacle to hematopoietic cell transplantation. In individuals with normal renal function, 10–30 g/day of serum light chains can be filtered and removed through the kidneys, whereas in patients with multiple myeloma, the amount of FLC produced exceeds the normal renal metabolism.⁹ As renal failure worsens, FLC clearance decreases and the extension of the half life of the FLC to 2–3 days causes the cast nephropathy.⁹ Direct nephrotoxocity of paraproteins plays an important role in the etiology of renal failure in myeloma. It follows therefore, that removing FLC through a path other than the kidneys may prevent the development of renal failure, and its’ attendant complications. To this end, attempts have been made to remove FLC from the blood compartment directly via plasma exchange, with mixed results. An alternate approach is the removal of FLC via artificial filtration membrane. While membranes with small pore sizes can clear FLC by membrane binding,⁵ the Gambro HCO 1100, a large pore membrane, has been shown to provide the best removal of FLC in vitro. While small pore membranes have not been clinically investigated for removal of FLC, treatment with high cutoff (large pore) membranes has been well documented.⁵ Although limited to compassionate use care in the USA, the Gambro HCO 1100, successfully removed FLC and provided a survival benefit compared to treatment with chemotherapy alone.⁵

In our case series, we used the commercially available renaflo HF1200 membrane (Minntech, Minneapolis, MN, USA). The HF1200 has a molecular clearance cutoff of 65 kD as opposed to the Gambro HCO 1100 which has a molecular clearance cutoff of 45 kD. Theoretically, this should allow for a greater clearance of FLC. In fact, patients who underwent hemodialysis with a Gambro HCO 1100 had mean clearance rates of 10.8–19.3 ml/minute (depending on the blood flowrate) for lambda light chains.⁵ We were able to calculate the clearance for our latter two patients in whom we measured filtrate and serum FLC concentrations using a simple clearance formula.⁵ The clearance rates were 30 and 32 ml/minute reduction in kappa light chains respectively. We had significantly higher clearance rates even at a lower blood flow (200 ml/minute). We also utilized continuous hemofiltration, as opposed to extended hemodialysis sessions as was used by Hutchinson et al.⁵ While we were limited to a blood flow of 200 ml/minute, we were able to use longer filtration times (18 hours/day) and consecutive day treatment. We expect that this provides a major advantage, in that the production of FLC is constant, thus in the times between dialysis, the FLC can accumulate rapidly and further damage the kidneys. There are several disadvantages of using CVVH in removing FLC. Firstly, slower flowrates limit the amount of clearance of FLC. Secondly, there is a greater incidence of clotting on CVVH. Thirdly, it can be difficult for conscious and relatively asymptomatic patients to undergo a full day of dialysis, which was our rationale for 18 hours a day therapy, as opposed to a complete 24-hour cycle. While the large pore size of the renaflo membrane allows greater clearance of FLC, the consequence is that there is also potentially a greater clearance of albumin, especially since albumin is the same size as the pore (65 kD). Though all three patients’ albumin levels decreased during treatment, we did not measure the dialysate albumin concentration, and cannot determine how much was due to filtered albumin, as opposed to a decrease due to the patients’ illnesses.

There are limitations to our study. Only one patient (Case 1) had a renal biopsy demonstrating cast nephropathy. We were unable to calculate FLC clearance for this patient. Nonetheless, our patients all had improvement in their renal function and remained off dialysis. Two patients (Cases 1 and 2) were able to undergo hematopoietic stem cell transplantation, and are doing well at 4- and 1-year follow-up, respectively. The last patient died of chemotherapy related sepsis, but remained off dialysis. We believe that extracorporeal filtration is an exciting approach, which is likely to provide significant benefit to myeloma patients. Our specific approach, using continuous modalities with a large pore membrane, may be a feasible or even superior alternative to hemodialysis with the Gambro HCO membrane, which is not readily available for use in the USA.

Conclusion

Acute renal failure is associated with high morbidity and mortality in cases of multiple myeloma. CVVH may be used in the treatment of cast nephropathy as modality to remove the FLC and possibly improve long-term survival for these patients, who otherwise have a worse prognosis. However, larger studies are required to assess the efficacy of CVVH in achieving this goal.