PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of clinbiorevLink to Publisher's site
 
Clin Biochem Rev. 2009 August; 30(3): 141–151.
PMCID: PMC2755003

Reporting of Quantitative Protein Electrophoresis in Australia and New Zealand: a Call for Standardisation

Abstract

Background

The lack of guidelines on reporting standards for protein electrophoresis may have led to significant differences in reports from different laboratories.

Objective

To determine the extent of variation in reporting of protein electrophoresis results in Australia and New Zealand.

Method

Questionnaires were distributed to laboratories throughout Australia and New Zealand asking about protein electrophoresis practices and reporting.

Results

Extensive variation was found in the following reporting practices: (a) units for urine Bence Jones protein (BJP); (b) reporting absence of a paraprotein rather than a normal pattern; (c) numerical reporting of all protein fractions or only the paraprotein; (d) warning of possible inaccuracy in the serum immunoglobulin result of the paraprotein type; (e) co-migration of a paraprotein with a normal serum protein; (f) use of a confirmatory test when a known paraprotein is no longer detectable.

Conclusions

A working party should be established to make recommendations on the reporting of protein electrophoresis. Implementation of such recommendations should reduce both report variation between laboratories and the risk of misinterpretation of reports.

Introduction

A number of guidelines have been published relating to diagnosis, treatment and monitoring of plasma cell dyscrasias.17 Although these often include recommendations about related clinical laboratory aspects, they fail to offer systematic reporting standards for serum and urine protein electrophoresis. In Western Australia (WA), noticeable differences in the reporting of protein electrophoresis between laboratories became apparent when public sector laboratories were developing a single laboratory information management system. In addition, clinicians working in both public and private sector hospitals had noted differences in paraprotein reporting between the pathology practices they used. There was, therefore, a significant impetus to standardise local electrophoresis reporting, but little evidence to recommend one approach over another.

To assess whether variation in reporting was widespread within Australia and New Zealand, two questionnaires were distributed during 2008 under the auspices of the Australasian Association of Clinical Biochemists (AACB). This paper summarises the results and indicates where variation could be reduced.

Method

The first questionnaire (Q1) on electrophoresis reporting practices was distributed in March/April 2008 by the AACB WA Branch Quality Control Sub-committee (QCSC) to all laboratories in Australia and New Zealand with representatives on branch QCSCs. This questionnaire generally asked for “free text” responses. A summary was returned to respondents and a brief presentation made at a workshop in September 2008. Many delegates at the workshop were unaware of this questionnaire so a second questionnaire (Q2), based upon Q1 but using a multiple choice approach, was distributed in December 2008 to all delegates. The two questionnaires are reproduced in the Appendix.

Results are presented as the number of responses in each category for each questionnaire. Because questions differed, not every category has responses shown for both questionnaires.

Results

General

Both questionnaires received 19 replies, with eight respondents replying to both, resulting in replies from 30 individuals or laboratories. Respondent demographics are shown in Table 1. The Royal College of Pathologists of Australasia Quality Assurance Programme for Paraproteins has approximately 56 participants. Thus questionnaire responses may reflect the practice of just over half the laboratories in the region.

Table 1
Questionnaire respondents by region.

Serum Electrophoresis

The majority of respondents report on all of the common serum protein electrophoresis fractions (Q1: 74%; Q2: 79%) (Table 2). From Q2, of those that report most fractions, 67% do so quantitatively.

Table 2
Serum protein fractions reported by respondents.

When albumin is reported numerically along with other fractions, Q1 showed that 11 respondents report the value from an automated chemistry method while six report that from densitometry.

“Normal” Serum Electrophoresis Pattern

Commenting upon the absence of a paraprotein or a normal protein pattern is usual (Table 3), the majority specifically referring to the absence of a paraprotein.

Table 3
Comments used when no paraprotein detected in serum.

Paraprotein Band Visible

Reporting of quantifiable paraproteins is relatively uniform. Reports of a newly-found paraprotein generally include band size, type and position (Table 4). Subsequently most respondents report only band size and type, with the position mentioned less frequently. A reference in the comment to a previous report is made by about half the respondents but most use cumulative reports (Table 4).

Table 4
Reporting practice when a paraprotein is detected in serum.

Previous Paraprotein No Longer Detectable

For patients who had a detectable paraprotein that is currently not visible, the majority of respondents refer to “previously” in a comment (Table 5). Fourteen respondents (in Q2) indicated that the absence of a previously detected paraprotein is confirmed by another method. Although not a specific question in Q1, five respondents reported using a confirmatory test.

Table 5
Reporting practice when a previous paraprotein is no longer detectable in serum.

Use of Serum Immunoglobulin Results in Electrophoresis Reporting

Immunochemical tests for serum immunoglobulins are generally only performed upon specific request and not routinely included as part of serum electrophoresis (Table 6a). Occasionally they are performed as a check for paraprotein identification or quantification. If immunoglobulins are reported with a paraprotein band quantification, more than 50% of respondents report both. A comment concerning the possible inaccuracy of immunochemical testing is used by about 50% of respondents (Table 6b).

Table 6a
Role of serum immunochemical immunoglobulin assays.
Table 6b
Role of serum immunochemical immunoglobulin assays.

Co-migrating Bands

The issue of quantification of co-migrating bands was included in both questionnaires, although the responses suggested significant ambiguity in the interpretation of the questions. The following comments should be read bearing this in mind.

When a paraprotein migrates with a normal protein fraction e.g. in the beta region, its concentration is reported as the total band concentration, i.e. “band plus normal fraction”, by the majority of respondents (Table 7). Other practices that are used by a few respondents include reporting the paraprotein as the total band value only under certain circumstances or subtraction of the median underlying protein value from the total band. A number of respondents provide an immunochemical immunoglobulin result as well as, or as an alternative to, the densitometric paraprotein concentration.

Table 7
Reporting practice with paraproteins co-migrating with normal protein fractions.

Urine Electrophoresis

Sample Collection

Q2 asked respondents about urine collections they recommend and those they usually receive for Bence Jones protein (BJP). Eight respondents recommend first morning specimens, seven recommend random samples, one recommends both first morning and random samples and five recommend 24-hour collections. By comparison, respondents indicated that they mostly receive random samples (16/17 responses). No answer to this question was received from two respondents.

Units

Urine total protein units are predominantly, but not universally, g/L and/or g/day (Table 8).

Table 8
Units for urine total protein.

The majority of respondents report urine BJP quantitatively with the units being predominantly g/L and g/day (Table 9). However, a number of respondents use other units and, in total, nine different units are in use. One respondent apparently uses mass units for creatinine.

Table 9
Units for Bence Jones protein.

Discussion

These two questionnaires have shown that there is significant variation in the reporting of electrophoresis results. The following aspects stand out as potential targets for standardisation: (a) units for urine BJP quantification; (b) reporting “normal pattern” rather than “no paraprotein detected”; (c) whether all protein zones should be reported numerically, or just the paraprotein, perhaps with gamma globulins or albumin; (d) the use of a comment warning of the possible inaccuracy of the serum immunoglobulin of the same type as the paraprotein; (e) the approach to quantifying paraproteins that co-migrate with normal serum protein fractions; (f) the use of a confirmatory test (immunofixation) to confirm that a previously detectable paraprotein is no longer present. Guideline recommendations provide limited help for these questions.

Diagnostic guidelines for the use of urine BJP in myeloma staging and response to treatment use either g/24h6,8 or mg/24h.2,7 The majority of respondents employ units based on g rather than mg although nine different units are in use. Controversy also exists about the preferred urine specimen for quantifying urine BJP excretion. There is no consistent agreement that, for example, 24-hour collections9,10 are preferred to morning or random specimens with results expressed per unit creatinine.11 These issues are reflected in the variation between the recommended and received urine collections reported and in the range of units used.

Respondents were equally divided between those that comment directly that a paraprotein is “not detected”, or indirectly by reporting a “normal” electrophoresis pattern. Many samples, especially those from hospital patients, will not have strictly “normal” electrophoresis patterns even in the absence of a paraprotein. As most requests are to determine the presence of paraprotein, it may be that a direct comment is preferable to the indirect comment. There is no guidance in the literature that we are aware of on this question. Related to this, some laboratories do not report the quantification of the alpha and beta globulins, instead reporting just the presence or absence of paraprotein. Again, given the major purpose relates to paraprotein detection, whether relevant clinical information is provided by quantifying other protein fractions is uncertain. Omission of this data would simplify the report. Nevertheless the majority of respondents report most of the fractions numerically.

The All Wales Clinical Biochemistry Audit Group recommended that only the non-paraprotein serum immunoglobulin results should be reported, as a means of demonstrating immune paresis,12 and this is supported by others.13 Consistent with this, about 50% of respondents withheld the paraprotein-type immunoglobulin result when the band was quantified. On the other hand, Durie et al. recommended the use of the serum immunoglobulin result of the paraprotein type in defined circumstances, particularly for IgA and low concentration paraproteins, for the purpose of aiding paraprotein quantification.6 This recommendation has been strengthened by the recent demonstration of good agreement between immunochemical and densitometric assessments of IgA paraproteins, but questionable associations for IgG and IgM paraproteins.14 It is also recommended that the same measurement of paraprotein concentration, either densitometric or immunochemical, should be used consistently for an individual patient, with the alternative result, if generated, suppressed.15 Whether this is widely practised is not known. Nevertheless, given the well-known risk of misleading results,13 it would seem prudent to append a disclaimer to the immunoglobulin results. However, this appeared not to be done universally, perhaps to avoid unnecessary crowding on the report.

The data from the questions relating to paraproteins that co-migrate with normal protein fractions were difficult to interpret, but indicate significant variation in practice. It would be helpful to define the most clinically valid estimate of paraprotein concentration, for example, either band value, with or without subtraction of underlying proteins, or the paraprotein-type immunoglobulin result. The recent demonstration of good agreement between serum IgA and densitometric IgA paraprotein measurements appears to validate the latter approach14 but this needs confirmation with different immunochemical systems and electrophoresis dyes. For other paraprotein types, however, other techniques may be needed.

Reporting on the disappearance of a previously detectable paraprotein could be informed by current guidelines that require absence of the original paraprotein in serum and urine by both electrophoresis and immunofixation.2,10 The questionnaire data showed that the practice of confirmatory testing in this situation is not always performed. It was also apparent that a number of laboratories do not specifically refer to previous reports, although some of these may issue cumulative reports, possibly rendering this less important.

This report has limitations. The questions in Q1 allowed free text responses, whereas Q2 was in a multiple-choice format. This sometimes led to difficulty in classifying the responses, but a significant number of laboratories would have been excluded if only one questionnaire was reported on here. In addition, it is possible that respondents interpreted some questions differently. Thus responses and classi cations may not always have reflected the respondents’ true practice. The authors did not confirm the intended interpretation of the questions with the respondents. This report has not addressed all the questions that were, or could have been, asked, mainly for reasons of space or relevance of the outcomes. For example, neither questionnaire specifically addressed the reporting of new monoclonal bands appearing after stem cell transplant treatment of myeloma,16,17 although this is recognised to be a potentially contentious area. A strength of this report is that responses came from most of the major centres in Australia and New Zealand, probably reflecting the practice in about half of the laboratories performing protein electrophoresis in these countries.

Issues with the reporting of protein electrophoresis are not confined to Australia and New Zealand, as demonstrated by the inclusion of a workshop on electrophoresis interpretation as part of the American Association for Clinical Chemistry Annual Meeting in 2009.18 The findings of this workshop may well be useful in the development of guidelines in Australasia.

This report has highlighted several aspects of the reporting of protein electrophoresis that vary substantially between laboratories. These would be amenable to standardisation if best practices could be recommended. We suggest, therefore, the formation of a working party to draw up guidelines addressing issues such as those raised in this report. Reducing the variability in reporting practices between laboratories will ease the task of clinicians and reduce the chance of misinterpretation of reports, thus providing an improved service to patients.

Appendix

Questionnaire 1 (Q1)

Results for Q1 were collated from responses to the following questions.

What medium and stain do you use for SPEPs (e.g. cellulose acetate, agarose, Ponceau S etc.) ?
What system do you use for SPEPs (e.g. SPIFE, Sebia, capillary electrophoresis)?
Do you use split beta gels?
What SPEP fractions do you report (e.g. TP, alb, alpha-1, alpha-2, beta, globs, paraprotein)?
Do you use both TP and albumin from your automated instrument or just TP when quantifying from the gel?
If albumin is part of the SPEP report, do you report the SPEP albumin result or the automated albumin result?
If a SPEP appears “normal” do you have a standard comment with this (e.g. “No paraprotein evident.”)?
When there is a paraprotein band present how do you report it?
When there is a paraprotein band present do you report the position of the band?
If you have a known patient with a previous paraprotein band who is in remission and no paraprotein band is seen in the SPEP (i.e. it looks “normal”), do you comment? If so what is the comment please?
How do you report repeat SPEPs on the same patient with the same band present (i.e. do you just quantify it, add a comment e.g. “Previous identified band of… still seen”, etc)?
Are immunoglobulins included as part of a SPEP request or only if requested specically?
Do you quantify a paraprotein that is in the beta region from the gel and report a paraprotein band value?
If yes to the previous question, do you report the paraprotein band result taking into account, for example, the underlying transferrin if it is in the beta region? How do you calculate/quantitate it?
For this type of paraprotein band (i.e. in the beta region) do you report immunoglobulins routinely as well or only if requested? Do you have any comment with this?
If there is a paraprotein band quantitated from the SPEP (not just in the beta region) and reported with an automated result of immunoglobulins, do you have a comment suggesting, for example, that ‘immunoglobulins may not accurately reflect the paraprotein quantitation’?
Do you report paraprotein bands, BJP and free light chain bands if seen in a UPEP?
What units do you use for reporting urine BJP if it is present and do you quantify it? How many decimal places?
Are you a public or private laboratory?
Does a Chemical Pathologist, Haematologist or scientist review reports? Please state which. Do you receive specialist Haematologist requests frequently?
Is cumulative reporting used?
Are urine and serum results reported on the same report?
If there are incidental findings or if there is difficulty in deciding the presence/significance of a band, is this reported on the written report or only discussed verbally? If reported, what type of comments are issued (e.g. repeat in three months, etc.)?
Any references/books you could recommend that may be useful to other laboratories especially relating to your reporting format?
If permitted please attach a copy of an example/s of your laboratory report for comparison with the group (please delete the patient’s name beforehand and your identifying laboratory name).
Any other comments relating to reporting issues of concern to you?

SPEP = serum protein electrophoresis; UPEP = urine protein electrophoresis; TP = total protein; BJP = Bence Jones protein.

Questionnaire 2 (Q2)

Results for Q2 were collated from responses to the following questions.

Part 1
Is your laboratory
 PublicPrivate
Electrophoresis is performed in
 Immunology departmentBiochemistry departmentOther (specify)
Number of serum electrophoresis (SPEP) performed daily
 <1010–3031–60>60
Requests are predominantly from
 General PractitionersSpecialists
Approximate percentage of requests that are for monitoring of known monoclonal gammopathies
 <1010–2526–5051–7576–100
Diagnostic supplier (SPEP)
 BeckmanHelenaSebia (Siemens)Other (specify)
SPEP system
 SPIFEHydrasysCapillarysParagonOther (specify)
SPEP medium
 cellulose acetateagarosecapillaryother (specify)
SPEP stain
 Amido blackAcid violet/acid blueCoomassiePonceau SOther (specify)
Do your routine SPEP gels split the beta region ?
 YesNo
Does your lab currently offer serum free light chains?
 YesNo
If no, are you considering introducing this assay?
 NoYes, within 6 monthsYes, within 12 monthsYes, sometime
Part 2 -Serum electrophoresis
Routine reporting is done by
 Chemical PathologistClinical HaematologistSenior ScientistOther (specify)
Difficult cases are reviewed by
 Chemical PathologistClinical HaematologistSenior Scientist
Regions reported on
 Pre- albuminAlbuminAlpha 1Alpha 2Beta 1Beta 2GammaParaprotein
Is each region reported numerically?
 yesNo
Is the albumin quantitatedby
 SPEPSpectrophotometric
Quantitation on gels is achieved using
 Total proteinAlbuminBothTotal globulinsOther
For normal patterns is a comment specifying the absence of paraprotein included?
 YesYes, if requesting presence of paraproteinNo
When a paraprotein is first detected by your lab, your report to a GP includes
 Band sizeBand PositionBand TypeImmune paresis (present or not)
 Recommended further investigationsRecommended review period
For subsequent reports on a known paraprotein, your report to a GP includes
 Band sizeBand PositionBand TypeImmune paresis (present or not)
 Previous band sizeDate previously seenTrend commentSuggested review period
When a paraprotein is first detected by your lab, your report to a Specialist includes
 Band sizeBand PositionBand TypeImmune paresis (present or not)
 Recommended further investigationsRecommended review period
For subsequent reports on a known paraprotein, your report to a Specialist includes
 Band sizeBand PositionBand TypeImmune paresis (present or not)
 Previous band sizeDate previously seenTrend commentSuggested review period
If immune paresis is present, is it (semi) quantitated e.g mild/moderate/marked?
 YesNo
Are immunoglobulins reported
 routinelyonly if requested
Where immunoglobulins are reported and a paraprotein is present, is the quantitative immunoglobulin of the paraprotein type reported?
 YesNo, only as part of the paraprotein
If you answered yes to the previous question, is there a comment issued concerning the accuracy of the quantitation?
 YesNo
Are paraprotein bands which run concurrently with beta region bands reported?
 YesNo
If yes, quantitation is reported as
 Band+ Beta = X g/LNephelometric/immunoturbidimetric immunoglobulin resulttransferrin/C3 measured
What is your approach when a known paraprotein patient is in remission and no paraprotein band is apparent on the current gel
 Report as“normal” with no special commentConfirm absence of band by alternate method
 Comment on previous band included in current reportRecommend repeat
Are all de-novo incidental (small) bands seen on routine gels reported
 Yes, as quantitated. No extra commentYes, but with a commentNo, communicated verbally
If yes but with comment, does the comment include
 Band sizeProbable significanceRecommended review period
In deciding your action/report what factors do you consider
 None, reporting is standardBand sizePresence of immune paresisPresence of BJP
 Patient agePresence of anaemiaPresence of renal impairment
 Presence of hypercalcaemiaOther comments

Part 3 - Urine electrophoresis
Reporting is done by
 Chemical PathologistClinical HaematologistSenior ScientistOther (specify)
Are serum and urine electrophoresis reported on the same report?
 YesNo, completely separateNo, but reports are cross-referenced
Your recommended collection for BJP is
 first voidrandom void24 hr collection(other) timed collection
Urine collections for BJP are predominantly
 first voidrandom void24 hr collection(other) timed collection
Urine electrophoresis include quantitative levels of (if present)
 total proteinalbuminalpha globulinstransferrinbeta 2microglobulin
Urine electrophoresis include qualitative levels of (if present)
 total proteinalbuminalpha globulinstransferrinbeta 2microglobulinlight chains
Patterns are reported as (as appropriate)
 Selective glomerularNon-selective glomerularTubularMixedlight chains
Bence Jones Protein (BJP), is reported as
 BJP not detectedBJP detectedBJP detected and quantitated
If urine total protein is quantitated what units are used?
 g/Lmg/Lmg/mLg/daymg/dayg/mmol creatOther (specify)
If BJP is reported quantitatively what units are used?
 % of total proteing/Lmg/Lmg/mLg/daymg/dayOther (specify)

Part 4
What Guidelines do you use for reference when reporting or giving verbal advice
 IMF International Myeloma FoundationNCCN National Comprehensive Cancer Network
 BCSH British Committee for Standards in HaematologyOther (specify)

Footnotes

Competing Interests: None declared.

References

1. International Myeloma Working Group. Criteria for the classification of monoclonal gammopathies, multiple myeloma and related disorders: a report of the International Myeloma Working Group. Br J Haematol. 2003;121:749–57. [PubMed]
2. Anderson KC, Alsina M, Bensinger W, Biermann JS, Chanan-Khan A, Comenzo RL, et al. Multiple myeloma. Clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2007;5:118–47. [PubMed]
3. UK myeloma forum. British Committee for Standards in Haematology. Diagnosis and management of multiple myeloma. Br J Haematol. 2001;115:522–40. [PubMed]
4. Smith A, Wisloff F, Samson D. Guidelines on the diagnosis and management of multiple myeloma 2005. Br J Haematol. 2006;132:410–51. [PubMed]
5. Keren DF, Alexanian R, Goeken JA, Gorevic PD, Kyle RA, Tomar RH. Guidelines for clinical and laboratory evaluation of patients with monoclonal gammopathies. Arch Pathol Lab Med. 1999;123:106–7. [PubMed]
6. Durie BG, Kyle RA, Belch A, Bensinger W, Blade J, Boccadoro M, et al. Myeloma management guidelines: a consensus report from the Scientific Advisors of the International Myeloma Foundation. Hematol J. 2003;4:379–98. [PubMed]
7. Kyle RA, Rajkumar SV. Criteria for diagnosis, staging, risk stratification and response assessment of multiple myeloma. Leukemia. 2009;23:3–9. [PMC free article] [PubMed]
8. Shaheen SP, Talwalkar SS, Medeiros LJ. Multiple myeloma and immunosecretory disorders: an update. Adv Anat Pathol. 2008;15:196–210. [PubMed]
9. Levinson SS. Urine protein electrophoresis and immunofixation electrophoresis supplement one another in characterizing proteinuria. Ann Clin Lab Sci. 2000;30:79–84. [PubMed]
10. Durie BG, Harousseau JL, Miguel JS, Bladé J, Barlogie B, Anderson K, et al. International uniform response criteria for multiple myeloma. Leukemia. 2006;20:1467–73. [PubMed]
11. Graziani M, Merlini G, Petrini C. Guidelines for the analysis of Bence Jones protein. Clin Chem Lab Med. 2003;41:338–46. [PubMed]
12. All Wales Clinical Biochemistry Audit Group: Standards for the Investigation of Serum and Urine Paraproteins. http://acbwales.org.uk/audit/guidelines/paraprot.pdf (Accessed 18 May 2009).
13. Attaelmannan M, Levinson SS. Understanding and identifying monoclonal gammopathies. Clin Chem. 2000;46:1230–8. [PubMed]
14. Murray DL, Ryu E, Snyder MR, Katzmann JA. Quantitation of serum monoclonal proteins: relationship between agarose gel electrophoresis and immunonephelometry. Clin Chem. 2009 Jun 11; [Epub ahead of print] [PubMed]
15. Riches PG, Hobbs JR. Laboratory investigation of paraproteinaemia. J Clin Pathol. 1988;41:776–85. [PMC free article] [PubMed]
16. Nagashima T, Muroi K, Kawano-Yamamoto C, Komatsu N, Ozawa K. Paraproteinemia after hematopoietic stem cell transplantation. Leuk Lymphoma. 2004;45:135–7. [PubMed]
17. Tate JR, Mollee P, Gill D. The reporting of serum protein electrophoresis to clinicians. Clin Chim Acta. 2005;358:204–5. [PubMed]
18. American Association for Clinical Chemistry, Annual Meeting 2009. Serum Protein Electrophoresis: Standardized Interpretation, Algorithms, and High Resolution Capillary Electrophoresis. [(Accessed 28 May 2009)]. http://direct.aacc.org/am/AMProduct.aspx?ID=5315.

Articles from The Clinical Biochemist Reviews are provided here courtesy of The Australian Association of Clinical Biochemists