AA (Inflammatory) Amyloidosis

Updated: Dec 06, 2015
  • Author: Richa Dhawan, MD, CCD; Chief Editor: Herbert S Diamond, MD  more...
  • Print


Amyloidosis comprises of a heterogeneous group of diseases in which normally soluble plasma proteins are deposited in the extracellular space in an abnormal, insoluble, fibrillar form.

Amyloid A (AA) amyloidosis is the most common form of systemic amyloidosis worldwide. [1] It is characterized by extracellular tissue deposition of fibrils that are composed of fragments of serum amyloid A (SAA) protein, a major acute-phase reactant protein, produced predominantly by hepatocytes. AA amyloidosis occurs in the course of a chronic inflammatory disease of either infectious or noninfectious etiology, hereditary periodic fevers, and with certain neoplasms such as Hodgkin lymphoma and renal cell carcinoma. [2]

In developing countries, the most common instigator of AA amyloidosis is chronic infection; in industrialized societies, rheumatic diseases, such as rheumatoid arthritis (RA), are the usual stimuli. The United States is a major exception to this in that immunoglobulin-related amyloid light chain type (AL) of amyloidosis is more frequent than AA as the cause of systemic amyloid deposition.

The major sites of involvement in AA amyloidosis are the kidney, liver, and spleen. Clinically overt disease typically develops when renal damage occurs, manifesting as proteinuria, nephrotic syndrome, or derangement in renal function.

The tissue fibril consists of a 7500-dalton cleavage product of the SAA protein, which is an acute phase reactant, and like C-reactive protein, is synthesized by hepatocytes under the transcriptional regulation of cytokines including interleukin (IL)-1, IL-6 and tumor necrosis factor (TNF). [3, 4] Under the influence of the inflammatory cytokine IL-6, hepatic transcription of the messenger ribonucleic acid (mRNA) for SAA may increase 1000-fold when exposed to an inflammatory stimulus.

Intact circulating SAA (molecular weight 12,500 dalton) is complexed with high-density lipoproteins (HDL). During the course of inflammation, the apolipoprotein SAA (apoSAA) apparently displaces apolipoprotein A1 (apoA1) from the HDL particles and facilitates HDL-cholesterol uptake by macrophages.

Several lines of evidence have indicated that the conversion of SAA into amyloid fibrils occurs through its specific interaction with heparan sulphate, a ubiquitously expressed glycosaminoglycan component of the extracellular matrix. SAA specifically binds to heparan sulfate (HS) glycosaminoglycan, a common constituent of all types of amyloid deposits that has been shown to facilitate conformational transition of a precursor to beta-pleated sheet structure. [5]

The protein has also been shown to be chemotactic for neutrophils, and it stimulates degranulation, phagocytosis, and cytokine release in these cells.

Until relatively recently, the erythrocyte sedimentation rate (ESR) and the serum C-reactive protein (CRP) level were used to monitor inflammation clinically. Current data suggest that, under some circumstances, changes in SAA may be a better measure. Increases in both CRP and SAA have been associated with active atherosclerotic coronary artery disease and cited as evidence for the inflammatory nature of that disease process. SAA also has been used to monitor the dissemination of malignancy.

For information on other types of amyloidosis, see the article Amyloidosis, Overview in eMedicine’s Rheumatology volume.



Chronic or acute, recurrent, substantial elevations of SAA are necessary but not sufficient for the development of amyloidosis. The median plasma concentration of SAA in healthy persons is 3 mg/L, but the concentration can increase to more than 2000 mg/L during the acute-phase response. Many individuals with long-standing inflammatory disease, although severely compromised by their primary condition, clearly do not develop tissue amyloid deposition. What determines any patient's risk for the development of this complication of inflammation is not known. Therapy, genetic factors, and environmental factors have all been proposed as possible contributors to the response of the primary disease.

Genes and proteins involved:

Three protein isoforms of SAA are noted (ie, SAA 1, 2, 4). Each isoform is encoded by its own gene in a cluster on band 11p15.1 that also includes a pseudogene (SAA3P). SAA1 has 3 alleles (SAA1.1, SAA1.3, SAA1.5), defined by amino acid substitutions at positions 52 and 57 of the molecule. [6]

The frequency of these alleles varies between populations and may be associated with the occurrence of AA amyloidosis in diseases such as rheumatoid arthritis. Also, it may have a role in determining the level of SAA in blood, clearance, susceptibility to proteolytic cleavage, severity of disease, and response to treatment. Seventy-six percent of Caucasians have SAA1.1, whereas only 5% have SAA1.3. In the Japanese population, the 3 alleles occur at approximately the same rate. Patients with a 1.1/1.1 genotype have a 3-fold to 7-fold increased risk of amyloidosis. But overall, the actual significance of the SAA genotype remains undefined. [5]

Cellular and extracellular tissue factors :

Mononuclear phagocytes might play a role in degradation of SAA and initiation of development of AA amyloidosis.

Polymorphisms of the mannose-binding lectin 2(MBL-2) gene leading to decreased levels of functional MBL have been related to defective macrophage function. This suggests that genetic background may affect the ability of mononuclear phagocytes to effectively process and degrade SAA proteins.

Additional tissue factors, such as enzymes found in the extracellular matrix, are likely to be involved in the proteolytic processing of SAA. Matrix metalloproteinases (MMPs) are involved in generation of SAA N-terminal fragments. In vitro studies confirmed that human SAAs and AA amyloid fibrils are susceptible to proteolytic cleavage by MMPs, generating fragments of different sizes. Studies have demonstrated that susceptibility to MMP-1 degradation is highly dependent on SAA1 genotype. [7]

The factors responsible for determining the site of deposition in any form of amyloidosis have not been identified. AA fibrils have been generated in tissue cultures by incubating SAA with macrophages. Deposits are frequently found in tissues with large numbers of phagocytic cells, notably the liver and spleen, but other affected organs, such as the kidneys, do not have the same cellular composition. Some data, derived from analysis of renal biopsy specimens, have suggested that glycoxidative modification of proteins, probably the AA protein itself, may also play a role in AA deposition in kidneys.




United States

The absolute prevalence of AA amyloidosis is difficult to ascertain because it depends on both the occurrence of predisposing inflammatory disorders and the proportion of individuals with those conditions who develop tissue amyloid deposition. The diseases in which AA amyloidosis has been reported are noted below, as are the frequencies (when such data are available).

AA amyloidosis is far less common in the United States than in other countries, even in the setting of the same inflammatory disease. The variation in the occurrence of amyloid in a particular disease in different geographic locales may reflect genetic background, differences in treatment of the primary disease, or factors that are not currently understood.


As in the United States, the frequency of AA amyloidosis is determined by the prevalence of the associated diseases, as well as the incidence of amyloid deposition in those conditions. For instance, in some Middle Eastern countries, the prevalence of familial Mediterranean fever (FMF) is higher than anywhere else in the world. The frequency of renal amyloidosis in some populations with untreated FMF is almost 100%. In those countries, amyloidosis represents a significant proportion of all renal disease.

Most available data to approximate the epidemiology of AA amyloidosis are derived from autopsies. The overall autopsy incidence of AA amyloidosis in western nations ranges from 0.50-0.86%. [8]

Currently, rheumatic diseases such as rheumatoid arthritis (RA), ankylosing spondylitis (AS), psoriatic arthritis, and juvenile idiopathic arthritis are the most frequent causes (70%) of AA amyloidosis. The reported prevalence of amyloidosis in RA varies with the diagnostic procedure used (that is, autopsy, kidney biopsy or subcutaneous fat aspiration), the clinical status (preclinical or symptomatic disease), and the type of study (case series or population-based study).

A study from Finland of the autopsy records of 1,666 patients with RA revealed a prevalence of amyloidosis of 5.8%, while a 10-year study of 1,000 patients with RA showed that 3.1% died of amyloidosis.

The most common cause of renal involvement in ankylosing spondylitis is AA amyloidosis (62%), followed by IgA nephropathy (30%). [9] Although its prevalence might be in decline, renal AA amyloidosis is a serious complication of AS, with a median survival time after onset of dialysis of 2.37 years, and with a 5-year survival rate of only 30%.

The prevalence of the asymptomatic phase of AA amyloidosis in RA can range between 0.5% and 14%.

Autopsy studies from the Netherlands have suggested a minimal prevalence of amyloidosis of approximately 1 per 75,000 population. Because 30-40% of amyloidosis cases in Western Europe is of the AL type, the estimated prevalence of AA amyloidosis is 1 per 100,000 population. Both the duration and severity of the inflammatory disease correlate with the frequency of amyloidosis as a complication.

The occurrence of multiple alleles encoding the predominant fibril precursor raised the issue of whether each allele had the same propensity to form amyloid. If an amyloidogenic allele were more common in a particular population, then the frequency of amyloidosis in inflammatory disease would be expected to be higher.

Three studies have indicated that a particular inherited form of SAA1 is associated with an increased frequency of amyloidosis in the course of a single inflammatory disease. In Japanese people, in whom the SAA 1.5 allele is far more common than in whites (37.4% vs 5.3%), the 1.5 allele is enriched among patients with RA and amyloidosis. Individuals with RA and a single 1.5 gene have twice the risk for developing amyloid as those with no 1.5 alleles. People who are homozygous for the 1.5 allele have a relative risk of 4.48 compared with those with RA who lack any 1.5 alleles. The mechanism of the association may reside in the fact that the SAA 1.5 allele is associated with higher SAA levels in Japanese patients. The duration of the inflammatory disease prior to the development of amyloidosis appeared to be inversely related to the dose of the allele.

In the United Kingdom, heterozygosity or homozygosity for the SAA 1.1 allele is associated with a greater risk for amyloidosis in whites with juvenile chronic arthritis; however, in patients with adult RA, the increase was not statistically significant.


In some cases, usually of infectious origin, the clinical consequences of amyloid deposition may dissipate with reduction or disappearance of the tissue deposits if the inflammatory disease can be suppressed totally or eliminated. If treatment of the primary disease is unsuccessful, death of organ failure secondary to the amyloid deposition is the rule. In patients treated at centers in the United States, the United Kingdom, and Europe from 1956-1992, renal failure or sepsis was the mode of exitus in one half to three quarters of AA amyloidosis cases, with a median survival of 24-36 months. Series that are more current show a longer survival, which is based largely on the increased availability of renal replacement therapy.

The progression of amyloidosis is related to the production and concentration of the circulating amyloidogenic precursor protein. The concentration of the acute phase protein SAA during follow-up correlates with deterioration of renal function, amyloid burden, and mortality in AA amyloidosis.

In a study of 374 patients with AA amyloidosis who were followed for 15 years, the median survival after diagnosis of amyloidosis in those with a sustained acute phase response was 133 months. As per this study, the risk of death was 17.7 times as high among patients with SAA concentrations in the highest eighth, or octile, (≥155 mg/L) as among those with concentrations in the lowest octile (< 4 mg/L).

In general, amyloidosis shortened the median life span 7.7 years, and survival strongly depended on controlling the underlying inflammatory process. Amyloid deposits regressed in 60% of patients who had a median SAA concentration of less than 10 mg/L, and survival among these patients was superior to survival among those in whom amyloid deposits did not regress. Sustained increased concentration of SAA is the most significant risk factor in AA amyloidosis, whereas reduction of SAA concentration improves survival and is associated with arrest or even regression of amyloid deposits. [10, 11, 12]

As per the Finnish Registry for Kidney Diseases, 502 patients with amyloidosis were identified entering RRT from 1987-2002. Eighty percent of these patients had amyloidosis associated with an underlying rheumatic disease. The 5-year survival rates among patients with the RA, AS, and juvenile idiopathic arthritis were 18%, 30%, and 27%, respectively. [13]

Cardiac amyloidosis appears to be a predictor of worse outcome with a 5-year survival of 31% versus 63% for patients without cardiac involvement in a retrospective series of 42 patients from Japan. [14]

The degree of renal involvement is important, with patients who have elevated creatinine levels doing worse compared with patients with a normal creatinine. The pattern of renal involvement is also important. Specifically, glomerular involvement with amyloid and fibrosis appear to have clinical course characterized by deteriorating renal function compared to patients with other types of renal involvement. Generally, however the median survival is over 5 years. [15]

In a multicentric retrospective survey to assess the graft and patient survival in 59 renal recipients with AA amyloidosis, the recurrence rate of AA amyloidosis nephropathy was estimated at 14%. There was significant decrease in the 5-year and 10-year survival of patients in the AA amyloidosis group compared with the control group. Also, AA amyloidosis transplanted patients exhibited a high proportion of infectious complications after transplantation. [16]


Very few appropriately controlled data address the question of racial prevalence of AA amyloidosis, other than observations suggesting that an increased frequency of AA amyloidosis occurs in the course of RA, which is related to variation in the distribution of particularly amyloidogenic SAA1 alleles among different ethnic groups. Within a single medical center in California, autopsies of patients of similar economic status with different ethnic origins displayed differences in the frequency of AA amyloidosis. In that series, AA amyloidosis was more common in Hispanic patients of Mexican origin than in either whites or African Americans.


In the United States, AA amyloidosis is more common in females, reflecting the fact that the major predisposing disease, RA, is predominantly a disorder of younger women and middle-aged men; hence, women are apt to have the disease for a longer period than men.

  • Despite the statistical female predominance in terms of overall numbers of AA amyloidosis cases, males seem to have an earlier average age of onset.

  • FMF is more common in males than in females (male-to-female ratio, 60:40), but the frequency of renal amyloidosis in people who are affected appears to be similar.


The age of onset of amyloidosis is related to the age of onset of the inflammatory disease, its severity, and the duration of the disease within the constraints imposed by the alleles of SAA carried by the patient. Thus, in the course of juvenile rheumatoid arthritis (JRA), amyloidosis occurs in teenagers. When it is a consequence of adult RA, it develops in late middle age. In the course of inadequately treated FMF, the renal amyloidosis is also of relatively early onset.