Receiver Operating Characteristic Curve of Total Serum IgE in Allergic Patients

Section: Original Articles
Issue
Vol. 5 No. 2 (2017): Volume 5 Issue 2
Published
Sep 10, 2017
Pages
77-82

Abstract

Background and objective: Validity of continuous variable test like total serum IgE is performed by constructing Receiver Operating Characteristic (ROC) curve. ROC curve is a technique for visualizing, organizing and selecting classifiers based on their performance. The study aims to determine the performance and validity of total serum IgE in various allergic diseases, age groups and gender. To select the optimal cut off value that has the best discriminative capability. To assess the sensitivity, specificity, predictive value and likelihood ratio of the selected cut-off points. Materials and method: A diagnostic Receiver Operator Characteristic ROC curvestudy was conducted at Allergy Clinic, Al-Jamhouri Teaching Hospital in Mosul, Iraq on 751 individuals (561 patients with various allergic disorders and 190 healthy non-allergic subjects) ranged in age from 10 to 59 years. Classification of study samples into allergic or healthy groups was done by a committee panel of two qualified and expert allergiologist. A blood sample was taken for estimating total serum IgE by ELISA. Results: The constructed ROC curve of overall sample showed that total serum IgE has moderate accuracy in allergic diseases and the AUC SE was 0.7300.022 with 95% confidence interval (0.686-0.773). Urticaria ROC curve had the maximum AUC (0.7420.030) and asthma had the minimum AUC (0.720 0.027). The maximum AUC was detected in the age group 10-19 years and the least one was in age 50-59 years. No difference was found in the estimated AUC in both gender. The optimal cut-off point of total serum IgE was 96 IU/ml in the overall allergic patients and varies between 95 IU/ml (in urticaria) and 115 IU/ml (in rhinitis). Assessment of selected overall cutoff point of total serum IgE reveals the following: sensitivity (79.3%), specificity (61.0%), validity (74.7%), PPV (50.0%), NPV (85.7%), LH+ (2.03), LH-(0.34). Conclusion: Total serum IgE has moderate accuracy in diagnosis of various allergic diseases. Its accuracy is much better in younger patients in comparison to elderly. A cut offpoint of total serum IgE(1005 IU/ml) was the best classifier. Total IgE is a useful screening rather than a confirmatory test.

References

  1. Baldacci, S.; Omenaas, E.; Oryszczyn, MP. (2001). Allergy markers in respiratory epidemiology. EuResp J. 17. P.p.773-90
  2. Clark, RD.; Webester-Clark, DJ. (2008). Managing bias in ROC curves. J ComputMol Des. 22. P.p.141-46
  3. Fluss, R.; Farggi, D.; Reisser, B. (2005). Estimation of Yoden index and its associated cut of point. Biom J. 47.P.p.458-72 Greiner, M.; Pfeiffer, D.; Smith, RD. (2000). Principles and practical application of the receiver-operating characteristic analysis for diagnostic tests. Prev Vet Med. 45. P.p.23-41
  4. Hanley, JA.; and McNeil, BJ. (1992). The Meaning and Use of the Area Under a Receiver Operating Characteristic (ROC) Curve. Radiology. 143. P.p. 29-36.
  5. Isolauri, E.; Huurre, A.; Salminen, S.; Impivaara, O. (2004). The allergy epidemic extends beyond the past few decades. ClinExper Allergy. 34. P.p.100710
  6. Jarvis, D.; Burney, P. (1998). The epidemiology of allergic diseases. BMJ. 316. P.p.607
  7. Kovac, K.; Doding, S.; Tiesic-Drunkovic, D.; Roos, M. (2007). Correlation y and serum IgE in asthmatic children sensitized to DermatophagoidsPeteronyssinus. Arch Med Res. 38. P.p.99-105
  8. Kumar R, Indrayan. (2011). A Receiver Operating Characteristic (ROC) Curve for Medical Researchers. Indian Ped. 48. P.p.277-67
  9. Obuchowski, NA.; Lieber, ML.; Wians, FH. (2004). ROC Curves in Clinical Chemistry: Uses, Misuses, and Possible Solutions. ClinChem. 50. P.p.111825
  10. Owen, CE. (2007). Immunoglobulin E: role in asthma and allergic disease: lessons from the clinic. PharmacolTher. 113. P.p.121-33.
  11. Sing, T.; Sander, O.; Beerenwinkl, N. (2005). ROC: visualizing, classifier, performance. Bioinformatics. 21. P.p.3940-41
  12. Whiting P, Ruljes AW, Reitsma JB, Glas AS, Bossuyt PM, Kleijnen J (2004). Sources of variation and bias in studies of diagnostic accuracy a systematic review. Ann Intern Med.140. P.p.189-202.
  13. Zhou, XH.; Obuchowski, NA.; McClish, DK. (2002). Statistical methods in diagnostic medicine. (1st ed.). New York: John Wiley & Sons. P.p. 15164
  14. Zweig, MH.; Campbell, G. (1993). Receiver Operating Characteristic (ROC) Plots: A Fundamental Evaluation Tool in Clinical Medicine. ClinChem. 39. P.p.561-77

Author

Haitham Bader Fathi

Dept. of Medicine, Ninevah College of Medicine/ University of Mosul

Identifiers

Download this PDF file
PDF

Statistics