Abstract
Introduction: The diagnosis of amyloidosis requires histological confirmation of Congo-red (CR) deposits. The tissue source is preferably fat aspiration and/or bone marrow (BM) biopsy, but at times organ biopsy is required.
Methods: We studied 612 patients with systemic immunoglobulin light chain amyloidosis to characterise the tissues used to establish the diagnosis.
Results: The median number of tissue samples was 3. About 95% of BM biopsies were stained for CR, while 79% of patients had fat aspiration CR-stained. CR stain sensitivity was 69% in BM, 75% in fat aspiration and 89% for both sources combined. In comparison, CR sensitivity was 97–100% for heart, renal and liver biopsies. About 42% of patients with renal involvement, 21% of patients with liver involvement and 13% of patients with heart involvement underwent organ biopsy, when a less invasive biopsy would have established the diagnosis. Predictors for the requirement for organ biopsy were male sex, limited organ involvement and lack of fat aspiration.
Discussion: Fat aspiration is underutilised for histologic confirmation of amyloidosis. A high rate of organ biopsies represents a failure to recognise the disease. Early awareness of amyloidosis in patients with organ dysfunction may lead to more judicious use of organ biopsies in this disease.
Fat pad aspiration is underutilised to establish the diagnosis of amyloidosis.
Bone marrow and fat pad aspiration obviates the need for invasive biopsies.
The excessive use of organ biopsy in AL amyloidosis reflects failure to recognise the disease early in its course.
Key messages
Introduction
The diagnosis of systemic immunoglobulin light chain (AL) amyloidosis is challenging, with failure to recognise the disease being a major diagnostic problem [Citation1]. Once the diagnosis is considered, histological confirmation is required, by demonstration of extracellular deposits of amorphous material that is Congo-red (CR) avid and has an apple-green birefringence under polarised light. The most accessible tissue source for diagnosis is bone marrow (BM) biopsy, which is performed to assess the underlying plasma cell disorder. When coupled with fat pad aspiration, a minimally-invasive procedure, the yield of CR-positivity approaches 90% [Citation2–4]. When amyloid deposits cannot be detected with these source tissues, more invasive biopsy from an organ suspected to be involved can be performed, although this carries a higher risk of complications [Citation5–7]. Alternatives to organ biopsies exist [Citation8,Citation9], including fat pad surgical biopsy or salivary gland biopsy, although their utilisation remains low.
This study explores biopsy practice in the diagnosis of patients with AL amyloidosis. We hypothesise that the failure to suspect amyloidosis results in patients undergoing organ biopsy when a diagnosis might have been established using less invasive methods. Our objective is to improve the diagnostic algorithm by making it more convenient for patients, reduce the risk from more invasive procedures and reduce costs.
Methods
Six-hundred and twelve consecutive patients with newly diagnosed symptomatic AL amyloidosis seen at Mayo Clinic between 1 January 2007 and 31 August 2015 were included. The study had three goals. First, to explore the spectrum of tissue biopsy evaluation of patients with AL amyloidosis. Second, to characterise the sensitivity of amyloid detection at each sampling site and correlate it with patient characteristics. Third, to define the role of organ biopsy in the evaluation of our population. The institutional review board approved the study in accordance with the declaration of Helsinki and Minnesota state law.
All biopsies performed in the evaluation of AL amyloidosis and stained for CR were included. Our centre is a referral centre for amyloidosis and samples were both internal and external. In the case of gastrointestinal (GI) tract involvement, we considered separately biopsies obtained by upper endoscopy from those obtained by colonoscopy. Samples from the same site collected on two separate occasions (repeat samples) were analysed as one sample, and if at least one of the repeat samples was positive for CR, it was regarded as positive. Repeat samples were further analysed in a sub-analysis.
Samples were stained with CR and were examined under polarised light for apple-green birefringence, which was considered positive. Negative samples were those without apple-green birefringence or had equivocal staining [Citation10]. Of the 1729 samples reported in this study, 1643 (95%) were confirmed by our pathologists. When samples were obtained from outside institutions and were reviewed, the Mayo pathology review was considered for this study. In those samples that were not available for review (n = 86), we used the outside report on CR avidity and apple-green birefringence.
Definition of organ involvement was established using consensus criteria [Citation11]. These criteria define organ involvement either by organ-proven biopsy or using specific criteria for the main four involved organs: heart, kidney, liver and peripheral nerve. We, therefore, evaluated the need for organ biopsy in these four organs.
Data were analysed with descriptive statistics. Continuous and ordinal variables were summarised with median and interquartile range (IQR), while categorical variables were reported as counts and percentages. The Kruskal–Wallis test and the Pearson χ2 test were used to ascertain differences between continuous and categorical variables, respectively. Factors associated with negative CR staining in BM/fat aspiration and factors associated with requirement for organ biopsy were analysed and reported as odds ratio (OR) with their 95% confidence interval (CI). p Values less than .05 were considered significant. Statistical analysis was performed on JMP software (SAS, Cary, NC).
Results
The baseline characteristics of the study population can be viewed in . The median age of the patients was 65. The median number of involved organs was 2, with the most commonly involved organs being the heart (85%), kidneys (60%), nerve (24%) and liver (21%).
Table 1. Baseline characteristics of the study population.
Biopsy work-up
The median number of biopsies per patient was 3 (IQR 2–3). The number of biopsies obtained was influenced by several factors. Patients ≥65 years were less likely to have more than three biopsies compared to their younger counterparts (17% vs 26%; OR 0.6, 95% CI 0.4–0.9; p = .01). Patients with more than two involved organs were more likely to have more than three biopsies compared to patients with 1–2 involved organs (31% vs 16%; OR 2.3, 95% CI 1.6–3.5; p < .001).
Sample source
Bone marrow/fat
Five-hundred and eighty patients (95% of the study population) had BM sample CR-stained, while 486 patients (79%) had fat pad aspiration obtained for CR staining. Altogether, 98% of the study population had at least a BM sample or fat aspiration stained for CR (of the remaining 2% of patients, the source for CR stain was a renal biopsy in seven patients, a GI biopsy in four patients and a tongue biopsy in one patient). Seventy-six per cent of patients had both BM and fat aspiration that were CR-stained, and 89% had at least one specimen positive for CR (thus theoretically making the second biopsy unnecessary).
Organ source and other sources
Organ biopsies are listed in . The most common organ biopsy was renal performed in 207 of the patients (56% of patients with renal involvement). This was followed by upper endoscopic biopsies in 101 patients (the defining criterion for GI involvement) and heart biopsy performed in 96 patients (18% of patients with heart involvement).
Table 2. Sensitivity of Congo red staining by specimen source.
Twenty-eight biopsies were obtained from soft tissue and other sources, which included eight soft tissue biopsies at various locations, seven salivary gland biopsies and five skin biopsies.
One-hundred and seventy two samples were duplicates obtained from the same source, representing 10% of all specimens. This mostly consisted of 42 repeated fat aspirations and 38 repeated BM samples.
Congo-red sensitivity by sample source
About 80% of the specimens were CR-positive. In patients with 1–2 biopsies, a positive stain was seen in 83% of the specimens, compared to a 77% positive rate in those with more than two specimens obtained, reaching statistical significance but not clinically relevant (mean difference 6%, 95% CI 2–10%; p = .003).
Bone marrow/fat
Sixty-nine per cent of patients with BM biopsies that were subjected to CR staining showed amyloid deposits in their samples, while 75% of patients with fat pad aspiration had a CR-positive aspiration. For those patients with both fat pad and BM samples that were CR-stained, 89% of them at least one sample that was CR-positive (59% in both samples, 13% in BM sample only and 17% in fat aspiration only). Of the 126 patients (21% of study population) without a fat pad aspiration performed, 64 patients had CR-positive deposits in their BM samples. Equivocal CR staining was seen in 17 fat aspirations and in 9 BM samples.
Of the 84 fat aspirations which represents 42 fat repeats, 55% were concordant (36% both positive, 19% both negative) while 45% were discordant on repeat. Of the 76 BM repeat samples, 63% were concordant (42% were both positive and 21% both negative), while 37% of the duplicates were discordant. After excluding repeat samples, sensitivity of CR staining in BM biopsy was 68% and in fat aspiration 75%. Of the discordant samples, 70% were read positive in our institution and negative in the outside institution.
Organ source and other sources
The rate of CR positivity by organ can be viewed in . The yield of CR positivity was high, ranging from 54% in rectal biopsy to 100% in heart biopsy, with most organs having a yield of >80%. Equivocal CR staining was seen in one nerve sample.
Predictors for CR negativity
lists factors that are associated with CR-negativity in BM biopsy and fat aspiration. Factors which were found to be associated with a CR-negative stain in fat aspiration were: male sex (OR 1.5, 95% CI 1.0–2.4), cardiac involvement (OR 0.4, 95% CI 0.2–0.6), BMPCs <10% (OR 2.0, 95% CI 1.3–3.0), difference between involved to uninvolved light chain (dFLC) < 18 mg/dL (OR 1.6, 95% CI 1.1–2.5) and kappa light chain restriction (OR 1.9, 95% CI 1.2–3.1). In BM samples, two factors were associated with a negative CR stain: 1–2 involved organs (OR 1.7, 95% CI 1.5–2.5) and hepatic involvement (OR 0.6, 95% CI 0.4–1.0).
Table 3. Predictors for bone marrow and fat pad aspiration negative Congo-red staining.
Re-assessing the need for organ biopsies
In light of the high yield of positive-CR staining in BM and/or fat samples, we examined the need for organ biopsy as part of the diagnostic process in four clinically affected organs (in which consensus definition of organ involvement does not mandate an organ biopsy). The percentage of concomitant positive CR staining in BM/fat samples in patients with CR-positive in an organ biopsy was calculated. The results are in . For patients with organ biopsy from the heart, kidney or liver, a concomitant positive-CR stain in BM and/or fat was seen in 72–79% of patients. This represents in our opinion the proportion of organ biopsies that could have been avoided. For the four patients with positive-CR stain in a nerve biopsy, the CR positivity in marrow/fat was two. When assessing other sources with CR positivity other than BM/fat, the results did not change, except for the four undergoing nerve biopsy which concomitant CR positivity increased the CR yield outside the nerve tissue from two to three. Given the different rates of organ biopsies, the absolute numbers of avoidable organ biopsy, were: 154 patients with renal; 70 patients with heart; 27 patients with liver and three patients with nerve. Overall, among the 612 patients, we believe than 254 organ biopsies could have been rendered unnecessary for diagnosis.
Table 4. Overall yield of Congo-red (CR) staining in patients with CR-positive organ biopsy.
Predictors for organ-only CR positivity
We analysed the predictors for the need for an organ biopsy to confirm the diagnosis of amyloidosis (i.e. the only source for positive-CR stain). All patients with organ biopsy and at least one additional biopsy outside the organ were analysed (n = 318). The results of this analysis are presented in . The following were found to significantly predict the need for organ biopsy: male sex (OR 2.0, 95% CI 1.04–3.3); limited organ involvement (1–2 organs, OR 3.4, 95% CI 1.7–6.8) and failure to perform fat pad aspiration (OR 3.4, 95% CI 1.9–5.9). Failure to perform BM CR staining was marginally associated with a need for organ biopsy (OR 2.9, 95% CI 0.9–9.7; p = .09).
Table 5. Predictors for the need for organ biopsy.
Discussion
This study, assessing the practice of tissue biopsy for establishment of the diagnosis of AL amyloidosis, showed that while CR staining was universally performed on almost all BM biopsies, fat pad aspiration was only done in 79% of patients. Failure to perform fat aspiration was a predictor for the need for organ biopsy. This simple and innocuous procedure should be performed in all patients when the diagnosis is suspected, unless amyloid deposits have already been demonstrated elsewhere. When combining CR-staining in both fat pad aspiration and BM biopsy the positive stain rate is 89%, thus only 11% of patients should require an organ biopsy, when amyloidosis is suspected. Nevertheless, many patients underwent organ biopsy. This was particularly true with renal biopsies, which were performed in over half of patients with renal involvement.
Amyloidosis has many manifestations and patients are seen by a median of three physicians from different practice fields, including primary care providers, nephrologists, cardiologists, haematologists, gastroenterologists, rheumatologists and neurologists [Citation12]. This wide distribution of initial contacts makes disease recognition a challenge. If a suspicion for amyloidosis is not raised at initial evaluation, then an organ biopsy is more likely to be performed. However, if amyloidosis is suspected, screening with a free light chain assay is a powerful indicator of AL amyloidosis. An excess of light chain should be followed by BM biopsy and fat pad aspiration, revealing amyloid deposits in 89% of the patients [Citation2–4]. Although organ biopsy is occasionally required, this is an exception and is not required in most instances. Other than establishing a diagnosis, in most cases organ biopsy does not provide the practitioner with additional prognostic or therapeutic information which will alter disease management.
A high frequency of kidney biopsy was seen with renal involvement and suggests a low index of suspicion in patients with proteinuria, although renal AL amyloidosis is present in 10% of adult nephrotic syndrome [Citation13]. Fifty-six per cent of patients with renal amyloidosis underwent renal biopsy, but only in 14% of the renal patients was biopsy unavoidable (and could be reduced to 10% if fat aspiration was universally performed). Over-utilisation of renal biopsy for the diagnosis of renal amyloidosis was also demonstrated by an Italian study, where 85% of renal AL amyloidosis patients underwent renal biopsy, while fat aspiration was done in <50% of these patients [Citation14]. In comparison to renal biopsy, endomyocardial biopsy was less frequent than renal biopsy (performed in 18% of patients with cardiac involvement), but was required in only 5% of the cardiac patients. The reason why we saw fewer endomyocardial biopsies has two explanations. First, endomyocardial biopsy is a procedure with a small but significant complications [Citation5,Citation6], and therefore its use is restricted. Second, unlike renal involvement which manifests as proteinuria that cannot be distinguished from other causes, cardiac amyloidosis had distinct echocardiographic features which provide the clinician with a better clue to the underlying disease and can drive early screening with a free light chain assay.
Our ability to demonstrate overuse of organ biopsy stems from Mayo Clinic being a major referral centre for amyloidosis. Patients often have an organ biopsy performed locally and complete their evaluation in our centre, which entails determination of the extent of disease and typing the amyloid deposits. This evaluation includes a BM biopsy, fat pad aspiration and biopsies of organs, when there is no other way to confirm involvement (i.e. GI tract, muscle, lung) or occasionally in organs where the clinician felt it appropriate to biopsy to accurately assess the disease extent (e.g. a patient with cardiac amyloidosis who undergoes GI biopsy before any decision is made on pursuing heart transplantation). Therefore, there are occasional scenarios where organ biopsy is required even when histologic confirmation already exists.
Male sex was correlated in this study with a lower rate of positive CR stain in fat, as previously shown [Citation15,Citation16]. Male sex was also marginally associated with BM-negative CR stain. Therefore, males were more likely to require organ biopsy. A search for an alternative accessible tissue source in males is warranted. Disease burden parameters (BMPCs, dFLC) were positively correlated with fat CR-stain in this study. At least two prior studies linked early amyloidosis with higher likelihood of CR-negative stain in fat smears [Citation16,Citation17], although no data was provided on BMPCs or dFLC in such patients. The association between CR sensitivity in fat aspiration and disease burden is probably explained by the limited amount of tissue obtained in fat aspiration compared to a tissue biopsy. This also explains the higher number of equivocal stains in fat compared to BM seen in this study (17 fat samples compared to 9 BM biopsies). Inadequate fat sampling for CR staining has previously been reported to be 11% [Citation18]. Therefore, negative results are more likely with fat, are probably linked to the sampling technique, and are more likely to be negative when disease burden is low. Kappa light chain amyloidosis was more likely to result in a negative fat, while cardiac amyloidosis increased the likelihood of a positive CR in fat. These findings were reported previously [Citation15] and may represent differential tissue tropism. It has been proposed that in subgroups in which fat aspiration results in a high negative CR stain, a surgical fat pad biopsy to increase diagnostic sensitivity should be used [Citation19,Citation20]. This approach is rational, overcomes the technical barriers of fat aspiration and should be validated to assess its true utility. An alternative to this approach is to perform a thorough evaluation of three fat smears by two pathologists [Citation9], or use of salivary gland biopsy [Citation8,Citation21].
This study has several limitations. First, it retrospective, which impacts the completeness of the data (e.g. CR staining was unconfirmed in 27 patients). Second, we did not analyse the sequence in which biopsies were performed. Nonetheless, we feel that the conclusion of over-utilisation of organ biopsies is valid independent of the biopsy sequence, as the presence of CR deposits outside an organ is the key message.
In conclusion, this study reveals one consequence of inadequate disease suspicion in amyloidosis, with resultant overuse of organ biopsies, primarily renal but also liver and heart. Increased awareness of the disease among different medical disciplines coupled with commonly used screening blood tests (i.e. serum free light chain) can increase the pre-test probability of AL amyloidosis and lead to a more judicious use of organ biopsy. Certain sub-populations (male sex, limited disease extent) may require organ biopsy for histological confirmation or alternatively fat biopsy instead of fat aspiration.
Acknowledgement
The study was supported by the Jabbs foundation (Birmingham, UK), the Henry J. Predolin Foundation (USA) and the Specialized Programs of Research Excellence (SPORE), grant CA186781 from the National Cancer Institute, National Institutes of Health, USA.
Disclosure statement
The authors declare no conflict of interests.
References
- Muchtar E, Buadi FK, Dispenzieri A, et al. Immunoglobulin light-chain amyloidosis: from basics to new developments in diagnosis, prognosis and therapy. Acta Haematol. 2016;135:172–190.
- Gertz MA, Lacy MQ, Dispenzieri A, et al. Autologous stem cell transplant for immunoglobulin light chain amyloidosis: a status report. Leuk Lymphoma. 2010;51:2181–2187.
- Kyle RA, Gertz MA. Primary systemic amyloidosis: clinical and laboratory features in 474 cases. Semin Hematol. 1995;32:45–59.
- Miyazaki K, Kawai S, Suzuki K. Abdominal subcutaneous fat pad aspiration and bone marrow examination for the diagnosis of AL amyloidosis: the reliability of immunohistochemistry. Int J Hematol. 2015;102:289–295.
- Deckers JW, Hare JM, Baughman KL. Complications of transvenous right ventricular endomyocardial biopsy in adult patients with cardiomyopathy: a seven-year survey of 546 consecutive diagnostic procedures in a tertiary referral center. J Am Coll Cardiol. 1992;19:43–47.
- Holzmann M, Nicko A, Kühl U, et al. Complication rate of right ventricular endomyocardial biopsy via the femoral approach: a retrospective and prospective study analyzing 3048 diagnostic procedures over an 11-year period. Circulation. 2008;118:1722–1728.
- Tondel C, Vikse BE, Bostad L, et al. Safety and complications of percutaneous kidney biopsies in 715 children and 8573 adults in Norway 1988–2010. Clin J Am Soc Nephrol. 2012;7:1591–1597.
- Foli A, Palladini G, Caporali R, et al. The role of minor salivary gland biopsy in the diagnosis of systemic amyloidosis: results of a prospective study in 62 patients. Amyloid. 2011;18:80–82.
- van Gameren II, Hazenberg BP, Bijzet J, et al. Diagnostic accuracy of subcutaneous abdominal fat tissue aspiration for detecting systemic amyloidosis and its utility in clinical practice. Arthritis Rheum. 2006;54:2015–2021.
- Gertz MA, Li CY, Shirahama T, et al. Utility of subcutaneous fat aspiration for the diagnosis of systemic amyloidosis (immunoglobulin light chain). Arch Intern Med. 1988;148:929–933.
- Gertz MA, Comenzo R, Falk RH, et al. Definition of organ involvement and treatment response in immunoglobulin light chain amyloidosis (AL): a consensus opinion from the 10th International Symposium on Amyloid and Amyloidosis, Tours, France, 18–22 April 2004. Am J Hematol. 2005;79:319–328.
- Lousada I, Comenzo RL, Landau H, et al. Light chain amyloidosis: patient experience survey from the amyloidosis research consortium. Adv Ther. 2015;32:920–928.
- Haas M, Meehan SM, Karrison TG, et al. Changing etiologies of unexplained adult nephrotic syndrome: a comparison of renal biopsy findings from 1976–1979 and 1995–1997. Am J Kidney Dis. 1997;30:621–631.
- Bergesio F, Ciciani AM, Santostefano M, et al. Renal involvement in systemic amyloidosis–an Italian retrospective study on epidemiological and clinical data at diagnosis. Nephrol Dial Transplant. 2007;22:1608–1618.
- Fernandez de Larrea C, Verga L, Morbini P, et al. A practical approach to the diagnosis of systemic amyloidoses. Blood. 2015;125:2239–2244.
- van Gameren II, Hazenberg BP, Bijzet J, et al. Amyloid load in fat tissue reflects disease severity and predicts survival in amyloidosis. Arthritis Care Res (Hoboken). 2010;62:296–301.
- Devata S, Hari P, Markelova N, et al. Detection of amyloid in abdominal fat pad aspirates in early amyloidosis: Role of electron microscopy and Congo red stained cell block sections. Cytojournal. 2011;8:11.
- Guy CD, Jones CK. Abdominal fat pad aspiration biopsy for tissue confirmation of systemic amyloidosis: specificity, positive predictive value, and diagnostic pitfalls. Diagn Cytopathol. 2001;24:181–185.
- Bogov B, Lubomirova M, Kiperova B. Biopsy of subcutaneus fatty tissue for diagnosis of systemic amyloidosis. Hippokratia. 2008;12:236–239.
- Guidelli GM, Bardelli M, Selvi E, et al. Punch biopsy for fat tissue collection in amyloidosis: is it time to stop needle aspiration? Rheumatology (Oxford). 2015;54:2109–2111.
- Suzuki T, Kusumoto S, Yamashita T, et al. Labial salivary gland biopsy for diagnosing immunoglobulin light chain amyloidosis: a retrospective analysis. Ann Hematol. 2016;95:279–285.