Skip to search boxSkip to navigationSkip to main content

Measurement of monoclonal immunoglobulin protein concentration in serum protein electrophoresis: Comparison of Automated vs Manual/Human Readings

  • Alex Clavijo
    ,
  • Nathan Ryan
    ,
  • ,
  • Gurmukh Singh(corresponding author)
*Corresponding author for this work
  • Department of Pathology
    ,
  • Department of Epidemiology and Bio-Statistics
    ,
Scholary Output:
Contribution to journal
Article
Research
Peer-review

Abstract

Background: Protein concentration of monoclonal immunoglobulin in plasma-cell myeloma/multiple myeloma provides an estimate of the tumor mass and allows for monitoring of the response to treatment. Accurate and reproducible estimates of the monoclonal immunoglobulin concentration are important for patient care. Objective: To address the optimum method for estimation of the concentration of monoclonal immunoglobulins. Methods: Serum protein electrophoresis and immunofixation electrophoresis were conducted by using the Helena SPIFE Touch instrument. Estimation of the protein concentration of monoclonal immunoglobulin in the gamma region by computer-assisted reading was compared with the reading by technologists and pathology residents, in 300 gels. The data were compared using t-testing and analysis of variance. Results: Computer-generated readings had a consistent positive bias. The correlation coefficient of the average reading by technologists and residents with the computer generated value was 0.997. The average positive bias by the computer reading was 0.29 g per dL. The intercept on the regression analysis was 0.22 g per dL. The reading by the computer was significantly higher than each of the human-interpreted readings. The readings by the 3 human groups were not significantly different amongst them. The main reason for the higher reading by the computer was inclusion of a greater area on the anodal size of the peak on the densitometric scan. Conclusions: Human- A nd computer-interpreted readings of the protein concentration of monoclonal immunoglobulin have a high degree of correlation. The consistent positive bias by the computer reading occurred due to inclusion of a greater area of the densitometric scan on the anodal side of the peak. We suggest that vendors should adjust such computer programs to provide readings comparable to those generated by expert humans. We recommend manual delineation of the monoclonal peaks for measuring the concentration of monoclonal immunoglobulins.

Publication Information

Output type

Scholary Output:
Contribution to journal
Article
Research
Peer-review

Original language

English (US)

Pages from-to (Number of pages)

Pages 252-258 (7 pages)

Journal (Volume, Issue Number)

Lab Medicine (Volume 51, Issue 3)

Publication milestones

  • Published - 2020

Publication status

Published - 2020

ISSN

0007-5027

Publication IDs

  • Scopus: 85084328895
  • PubMed: 32374393

Publication metrics

Metrics

Fractional count
2
Fractional count
0.50
Fractional count
2
Fractional count
0.50
Fractional count
2
Fractional count
1
SciVal
FWCI
0.41
SciVal
Author count
4
SciVal
citations
1
SciVal
Paper percentile
67
Scopus
citations

PlumX, opens in new tab

Captures
9
Citation count
13
Social media
7