Gut 55:1666-1667
  • Letter

Anti-TNF-α for treatment of amyloidosis associated with Crohn’s disease

  1. A Fernández-Nebro1,
  2. I Ureña2,
  3. M V Irigoyen2,
  4. R García-Vicuña3
  1. 1Servicio de Reumatología, Pabellón “Hospital Civil”, Hospital Regional Universitario Carlos Haya, Málaga, Spain
  2. 2Servicio de Reumatología, Hospital Regional Universitario Carlos Haya, Málaga, Spain
  3. 3Servicio de Reumatología, Hospital Universitario La Princesa, Madrid, Spain
  1. Correspondence to:
    Dr A Fernández-Nebro
    Servicio de Reumatología, Pabellón “Hospital Civil”, Hospital Regional Universitario Carlos Haya, Plaza del Hospital Civil s/n, 29009 Málaga, Spain; afernandezn{at}

    We wish to comment on the case reported by Iizuka and colleagues (Gut 2006;55:744–5) of a patient who experienced a reduction in creatine levels from 4 to 2 mg/dl after three weeks of treatment with infliximab for amyloidosis associated with Crohn’s disease. Unfortunately, no details were provided of other concomitant factors, such as volume depletion or drugs, which may have affected baseline creatinine levels and their later course, irrespective of amyloidosis, or of the results of proteinuria observed in 1992.

    In our experience,1 anti-TNF-α therapy has a rapid effect on proteinuria but not on renal function. In our patients it took several months (the most rapid decrease was from 1.6 to 0.94 mg/dl in three months) for serum creatinine levels to decrease by (mean (SD) percentage since baseline) 8.5 (16.6)% (range −21.4 to 39.4) after 70 (44) weeks (range 2–120), and only 36% of patients experienced a 20% reduction or greater.

    We should also like to comment on the choice of anti-TNF for the management of amyloidosis in patients with quiescent Crohn’s disease but active spondyloarthropathy.

    One of our patients was diagnosed by clinical, endoscopic, and histological criteria with spondyloarthropathy associated with Crohn’s disease in 1988 at the age of 20 years. His intestinal disease was moderately active at the beginning whereas his spondyloarthropathy remained highly active, causing severe restriction in the whole spine, shoulders, and hips. He underwent arthroplasty of both hips in 1995 and 1997. In 1998, when his intestinal disease had been inactive for more than five years, he developed nephrotic syndrome (proteinuria of 11.72 g/day and normal renal function). Rectal biopsy was negative but kidney biopsy showed extensive glomerular and arteriolar AA amyloid deposits. At the beginning of 2001 his renal function started to decline and treatment was initiated with etanercept 25 mg, twice weekly, with excellent clinical and analytical results (fig 1). Twelve months later repeated episodes of intestinal activity commenced, with later worsening of the proteinuria and renal function. After 29 months, etanercept was replaced by infliximab (5 mg/kg) but his status continued to worsen despite haemodialysis and ileal resection, with several interruptions of infliximab until his death 14 months later with severe pancytopenia due to progression of amyloidosis, as shown at autopsy.

    Figure 1

     Serum C reactive protein (CRP), serum creatinine (Cr(s)), and proteinuria (PU) in our patient after anti-tumour necrosis factor treatment.

    We chose etanercept initially because a case had been reported of a good response of AA amyloidosis to etanercept in 20012 and the results of the clinical trial showing lack of efficacy of etanercept in Crohn’s disease had not been published.3 This case illustrates the requirement in secondary amyloidosis for exhaustive control of any inflammatory activity, and highlights the fact that delay in providing effective treatment may prove fatal. Etanercept also lacks effectiveness for preventing reactivation of Crohn’s disease. Accordingly, we should always choose infliximab for this type of patient, even if Crohn’s disease has been inactive for a long period.


    Author’s reply

    1. S Konno4,
    2. M Iizuka4,
    3. Y Horie4,
    4. H Itou4,
    5. K Shindo4,
    6. S Watanabe4
    1. 4Department of Internal Medicine, Akita University School of Medicine, Akita, Japan

      In response to the comments of Fernandez-Nebro et al to our case (Gut 2006;55:744–5), we would like to make some additional observations.

      We demonstrated a patient with systemic amyloidosis associated with Crohn’s disease significantly responding to infliximab. Specifically, infliximab not only decreased serum amyloid protein level but also ameliorated renal function of the patient. With regard to the effect of infliximab on renal function, Anelli and colleagues1 also demonstrated a patient with rheumatoid arthritis and renal amyloidosis whose serum creatinine level significantly improved (from 2.8 to 1.3 mg/dl) after treatment with infliximab. However, it is notable that serum creatinine level of our patient rapidly improved after treatment with infliximab.

      Fernandez-Nebro et al suggested an association of other concomitant factors, such as volume depletion or drugs, which can affect baseline creatinine levels. However, the patient had not received any drugs that might affect baseline creatine levels for at least six months before treatment with infliximab. Also, we suggested that volume depletion might be involved in the progression of his renal dysfunction as the patient was admitted to our hospital because of acute progression of renal dysfunction during exacerbation of Crohn’s disease. Thus he underwent parental nutrition soon after admission. Central venous pressure of the patient on the 10th hospital day was within normal limits (5.5 cm H2O). An appropriate volume of infusion was provided for the patient but his renal function did not improve until he received an infusion of infliximab.

      Based on these findings, we considered that progression of his renal dysfunction was not caused by drugs or volume depletion alone and that infusion of infliximab was clearly involved in improvement of his renal function. Progression of renal dysfunction in our patient appeared to be correlated with exacerbation of Crohn’s disease and a high concentration of serum amyloid (762 μg/ml; normal range <8 μg/ml). It has been suggested that tumour necrosis factor blocking agents might not only reduce the synthesis of amyloid precursors but also slow amyloid deposition.2 In addition, Bosca and colleagues3 demonstrated that infliximab decreased the degree of amyloid deposits. Although the detailed mechanism is unknown, infliximab may work more effectively on patients with acute progression of renal dysfunction with high serum amyloid levels, such as in our patient.

      Our patient received an additional infusion of infliximab eight weeks later and serum creatine levels have been less than 3 mg/dl for approximately 41 weeks. Unfortunately, we were not able to assess the effect of infliximab on proteinuria because he could not collect total volume of urine before treatment with infliximab.

      As Fernandez-Nebro et al demonstrated, systemic amyloidosis is a serious and fatal complication of Crohn’s disease. It is critical to control the intestinal inflammation of Crohn’s disease to prevent progression of renal dysfunction due to amyloidosis. In this context, we also believe that infliximab should be the first choice for patients with systemic amyloidosis associated with Crohn’s disease.


      • Conflict of interest: None declared.