OUP user menu

Fluorescence Microscopy Is Superior to Polarized Microscopy for Detecting Amyloid Deposits in Congo Red–Stained Trephine Bone Marrow Biopsy Specimens

Alan Marcus MD, Evita Sadimin MD, Maurice Richardson MD, Lauri Goodell MD, Billie Fyfe MD
DOI: http://dx.doi.org/10.1309/AJCP6HZI5DDQTCRM 590-593 First published online: 1 October 2012


The classic gold standard for detecting amyloid deposits is Congo red–stained bright field and polarized microscopy (CRPM). A prior study showed that Congo red fluorescence (CRF) microscopy had increased sensitivity compared with traditional CRPM when analyzing fat pad specimens. The purpose of the current study was to determine the sensitivity of CRF for evaluating Congo red–stained bone marrow biopsy specimens, and to compare these results with those of CRPM. We compared the CRPM and the CRF analyses of 33 trephine bone marrow biopsy specimens with clinical or morphologic suspicion of amyloid deposits. These results were verified against immunohistochemical staining with anti–amyloid P antibody. CRF achieved 100% sensitivity, and CRPM achieved 75% sensitivity. Both groups showed 100% specificity compared with amyloid P immunohistochemical staining. The results show that CRF is a sensitive method to analyze trephine bone marrow biopsy specimens for amyloid deposits.

Key Words
  • Plasma cell dyscrasia
  • Trephine bone marrow biopsy
  • Amyloid
  • Congo red
  • Fluorescence microscopy
  • Polarized microscopy
  • Immunohistochemistry

Amyloid is an extracellular proteinaceous deposit that results from many different precursor proteins. Regardless of the precursor protein, all amyloid is derived from 3 components: proteoglycans, amyloid P, and fibrillary proteins.1 Amyloid light chain amyloidosis is often associated with a plasma cell dyscrasia. Patients with amyloid light chain amyloidosis typically exhibit symptoms such as nephrotic range proteinuria, congestive heart failure secondary to restrictive cardiomyopathy, idiopathic hepatomegaly, and idiopathic peripheral neuropathy.

Amyloidosis is diagnosed by detecting amyloid deposits on histologic sections. Both bone marrow biopsy and subcutaneous fat pad biopsy can be used to evaluate for amyloidosis. However, of the 2, only bone marrow biopsy has the potential to reveal both the amyloid deposits and a potential underlying plasma cell dyscrasia.2 A review of the literature found that amyloid is present in 1.4% of patients diagnosed with either multiple myeloma or smoldering multiple myeloma who are without clinical symptoms of amyloid light chain amyloidosis, and in 40.5% of bone marrow biopsies in patients diagnosed with multiple myeloma.3,4

The gold standard for amyloid deposit diagnosis is detecting green birefringence via polarization.5 However, numerous studies have shown that alkaline Congo red stained–sections analyzed with polarized light microscopy (CRPM) can result in false-negative results, especially in the presence of smaller quantities of amyloid.6

The study by Giorgadze et al1 demonstrated that in abdominal fat pad needle aspiration biopsies, Congo red fluorescence (CRF) had superior sensitivity compared with CRPM. Seventy-eight biopsy specimens were analyzed, of which 12 (15.4%) were found to contain amyloid when CRPM was used. Seventy-four of these cases were reclassified using CRF, and 29 (39.2%) were found to contain amyloid. Colocalization of CRF and amyloid P immunohistochemical staining confirmed the presence of amyloid in the areas that were seen under fluorescence, but did not show green birefringence with polarization.1

Our study determines the sensitivity of CRF for detecting amyloid in bone marrow biopsy specimens, and compares the sensitivity of CRF with that of CRPM. Thirty-three trephine bone marrow biopsy specimens from patients with a diagnosis of plasma cell dyscrasia or with suspicion of plasma cell dyscrasia were retrospectively analyzed via CRF after prior analysis via CRPM. As in the study of Giorgadze et al,1 the results were verified against immunohistochemical staining with anti–amyloid P antibody.

Materials and Methods

With institutional review board approval, this retrospective study reviewed 33 trephine bone marrow biopsy specimens with suspicion for amyloidosis from the archives of Robert Wood Johnson University Hospital, New Brunswick, NJ. Each of these cases had either a clinically suspected or a confirmed diagnosis of plasma cell dyscrasia.

Bone marrow processing was performed using formalin fixation, rapid decalcification in 14% hydrochloric acid (1.5 hours), and routine histologic processing. Congo red staining (Ventana Medical System, Tucson, AZ) was performed using the Ventana Nexus Special Stainer. All sections were 8-μm thick, and examined under light, polarized, and fluorescence microscopy with an Olympus microscope (Olympus America, Center Valley, PA). The U-POT drop-in polarizer from Olympus was used for polarization, and the blue excitation filter for fluorescein isothiocyanate was used for fluorescence.

Amyloid P immunohistochemical staining (mouse monoclonal, NCL-AMP, Novocastra Laboratories, Buffalo Grove, IL) was performed using the Ventana Benchmark XT.


Thirty-three cases were examined. Four cases that were initially classified as negative (cases 1, 9, 11, and 26) via CRPM were found to contain amyloid deposits in the paracortical soft tissue and/or the marrow space on examination with CRF Image 1. In addition, 3 cases (cases 14, 15, and 22) that were described as focally positive in the paracortical soft tissue with CRPM were discovered to be positive in the paracortical soft tissue and the marrow on examination with CRF. The other 26 diagnoses were not altered on CRF examination. The results are delineated in Table 1.

Image 1

A, Congo red–stained specimen viewed via light microscopy (×1,000). B, Congo red–stained bright field and polarized microscopy findings (CRPM). CRPM displays faint focal apple green birefringence in a case that was originally defined as negative (×1,000). C, Congo red fluorescence (CRF). The amyloid stains red-orange and is unambiguously positive, in comparison with the image in B (×1,000).

View this table:
Table 1

The diagnoses rendered were verified by means of amyloid P immunohistochemical staining. CRF achieved 100% sensitivity, and CRPM achieved 75% sensitivity.

Follow-up of the 4 reclassified cases reveals that 2 of them required additional invasive procedures to obtain a diagnosis of amyloidosis. Case 1 required an abdominal fat pad aspiration, and case 11 needed an additional bone marrow biopsy.


The use of Congo red in pathology laboratories has been in a state of refinement since its conception.5,7-9 Different methods have been used to improve the sensitivity of Congo red to detect amyloid. Linke6 compared the sensitivity of 3 methods used to analyze alkaline Congo red–stained sections: CRPM, CRPM and immunohistochemistry using a battery of antibodies against amyloid (CRPM-IHC), and CRF. That study used samples from the kidney, gastrointestinal tract, cardiovascular system, central nervous system, skin, synovium, skeletal muscle, pancreas, and pituitary gland. All 3 methods were negative for amyloid when no amyloid was present (57 cases), and positive for amyloid when large amounts of amyloid were present (126 cases). However, when only small amounts of amyloid were present (211 cases) the sensitivity of the 3 methods varied. CRF was positive in 172 cases, CRPM-IHC was positive in 158 cases, and CRPM was positive in 84 cases. Thus, CRF achieved a better sensitivity.6

Our study examined the sensitivities of CRF and CRPM for detecting amyloid deposits in trephine bone marrow biopsy specimens. The literature reports that CRPM has approximately a 50% sensitivity.2,10 No reported sensitivity could be found for CRF. Our study found the sensitivity of CRPM to be 75%, and the sensitivity of CRF to be 100%. The improved sensitivity of CRF found the 4 false-negative results previously seen with CRPM.

In our study, the specificity of both CRPM and CRF were 100%. We were unable to find a report in the literature on the specificity of CRPM or CRF to detect amyloid deposits in bone marrow biopsy specimens. However, CRPM has been reported to have a specificity of 100% for detecting amyloid in subcutaneous fat aspirations.1,11 The clinical importance of this finding is that the high sensitivity and specificity of CRF would prevent patients from undergoing additional procedures to diagnose amyloidosis, as seen in several patients in our study.

Immunohistochemistry has several disadvantages as an amyloid screening test. The study by Linke6 demonstrated that CRF had a greater sensitivity than CRPM-IHC. It has been hypothesized that tissue surrounding the amyloid may prevent the passage of antibodies but allow the passage of the Congo red stain.6 In addition, the specificity of immunohistochemical staining for amyloid is questionable because the precursor proteins that form amyloid are normal constituents of the body. Thus, to guarantee specificity, immunohistochemistry requires colocalization with Congo red. In our study, both immunohistochemistry and CRF had the same sensitivity and specificity.5

CRF has several benefits. It is easy to perform because bright field, polarized, and fluorescent microscopies can be performed using the same microscope.1 As demonstrated in Image 1, it is our opinion that CRF is easier to interpret than CRPM when diagnosing amyloidosis. In addition, CRF has a high sensitivity, an important property for a screening test. This leads to early diagnosis, while amyloid deposits are still small. This is important for 2 reasons. First, it prevents additional biopsies to make the diagnosis of amyloidosis. Second, the body has a limited ability to clear existing amyloid deposits.2 Thus, once the diagnosis of amyloidosis is made, the underlying disease is sought and treatment is started. The first step in determining the underlying disease is classifying the precursor protein creating the amyloid deposit.5 More than 20 possible precursor proteins are known to form amyloid deposits.1 The specific precursor protein is determined with immunohistochemical stains and mass spectrometry.12 The treatment for amyloidosis varies, depending on whether it is localized or systemic and on the specific precursor protein/underlying disease.1


View Abstract