Conheça o Grupo Sabin
Clientes Médicos Empresas

busca

Opções de busca:
  • Notícias
  • Serviços
  • Unidades
  • Institucional
Encontre uma unidade
Serviços Sabin
Sabin » Home » DETERMINATION OF ACCURACY

DETERMINATION OF ACCURACY

DETERMINATION OF ACCURACY AND PRECISION OF TWO METHODS OF IGF-1 DETERMINATION AND ITS CORRELATION TO GH NADIR ON OGTT IN ACROMEGALIC PATIENTS
 
Luciana A Naves 1, Leonardo Abdalla Giacomini 1, Paula Pires Nascimento 1, Joao W C Rosa1, Luis Augusto Casulari 1, Sandra S S Costa2.
1 Unity of Endocrinology, University of Brasilia, Brasilia, DF, Brazil
2 Laboratório Sabin de Análises Clínicas, Brasília, DF, Brazil

ABSTRACT
The evaluation of insulin-like growth factor 1 in serum (IGF-1) is a very important tool on diagnosis and follow up of acromegalic patients. The standard method for the diagnosis of acromegaly is the measurement of GH nadir (GHn) during an oral glucose tolerance
test (OGTT). Determination of IGF-I levels is useful as it correlates with clinical features of acromegaly and with the 24-hour mean GH levels. According to the most recent consensus, a random GH <0.4 microg/l and IGF-I in the age- and gender-matched normal range exclude the diagnosis of acromegaly. If either of these criteria are not achieved, an OGTT should be performed, and then GHn <1 microg/l during OGTT excludes acromegaly. Discrepancies between IGF-1 and GH levels have been reported in 15% of acromegalic patients, and impose some pitfalls on clinical evaluation. The aim of this study was to evaluate reproducibility and interassay variation coecients of these two dierent methods used for IGF-1 determination and to compare the results to GH nadir on GTTo in acromegalic patients. Methods: We studied 35 acromegalic patients recruited to collect fast blood samples to IGF-1 analysis and to perform oGTT with GH determination. All patients signed the informed consent. IGF-1 analysis was performed in triplicate by automated IMMULITE assay system (DPC, Los Angeles) and radioimmunoassay (RIA) with extraction. The precision and the interassay coecient of variation were calculated for each method. These results were compared and correlated to GHn (IMMULITE 2000) on oGTT. Results: The patients were stratified by clinical status, as controlled (13 patients) and active (22 patients). The results have shown a very high correlation between the IGF-1 values obtained in the two assays (r=0.94, p<0.001). The mean coefficient of variation intraassay of IGF-1 for chemiluminescence was 4.94%, while for RIA, the CV was 9.67%.
The correlation between GHn on oGTT and chemiluminescence was r=0.71 and the RIA was r=0.66. Conclusion: IGF-1 determination by RIA had a significantly higher variation interassay than chemiluminescence, that seems to have a better correlation to GH nadir on oGTT in acromegalic patients.

INTRODUCTION
The diagnosis of Acromegaly is based on clinical and hormonal evaluation. The standard method is the measurement of GH nadir (GHn) during an oral glucose tolerance test (OGTT). Determination of IGF-I levels is useful as it correlates with clinical features of
acromegaly and with the 24-hour mean GH levels. According to the most recent consensus, a random GH <0.4 microg/l and IGF-I in the age- and gender-matched normal range exclude the diagnosis of acromegaly. If either of these criteria are not achieved, an
OGTT should be performed, and then GHn <1 microg/l during OGTT excludes acromegaly.
Discrepancies between IGF-1 and GH levels have been reported in 15% of acromegalic patients, and impose some pitfalls on clinical evaluation. Since IGF-I is inuenced by several other hormones and physiologic factors as well as GH, a knowledge of its regulation
is essential to understanding how to properly interpret the measurements (Clemmons et al, 2007). Only for the automated chemiluminescence immunoassay Nichols Advantage have age and sex specic reference values based on a large population
(Ivan et al,2005). This this study was designed to compare the results using two dierent methods, in a population of acromegalic patients with discrepant data concerning GH nadir on oGTT and IGF-1

OBJECTIVES
The aim of this study was to evaluate reproducibility and interassay variation coecients of these two dierent methods used for IGF-1 determination and to compare the results to GH nadir on GTTo in acromegalic patients.


SUBJECTS AND METHODS
We studied 35 acromegalic patients (20 women, age 40 + 2,5 years; and 15 men, age 43 + 1,8 years) recruited to collect fast blood samples to IGF-1 analysis and to perform oGTT with GH determination. Most of these patients were submitted to surgery, surgery
and radiotherapy or surgery and octreotide. All patients signed the informed consent.

Table 1 – Dierent treatments according to gender in 35 acromegalic patients

 

METHODS
IGF-1 analysis was performed in triplicate by automated IMMULITE assay system (DPC, Los Angeles) and radioimmunoassay (RIA) with extraction. The precision and the interassay coecient of variation were calculated for each method. Oral glucose tolerance test was performed after load of 100g of glucose and samples collected basal, 30,60,90,120 minutes after load. GHn was determined by chemiluninescence (IMMULITE 2000).


RESULTS
The patients were stratied by clinical status, as controlled (13 patients) and active (22 patients). The results have shown a very high correlation between the IGF-1 values obtained in the two assays (r=0.94, p<0.001). The mean coefficient of variation intraassay of IGF-1 for chemiluminescence was 4.94%, while for RIA, the CV was 9.67%. The correlation between GHn on oGTT and chemiluminescence was r=0.71 and the RIA was r=0.66.

Figure 1: Correlation between IGF-1 levels determined by RIA and chemiluminescence(r=0.94 p<0.001)

        


Table 2 - Precision of the two methods, calculated after triplicate determinations
for each method

 


DISCUSSION
Several technical criteria are required for successful laboratory estimation of IGF-I values. These include elimination of interference of IGF-I-binding proteins (IGFBP), utilization of adequate numbers of normal subjects to dene the normal ranges and importantly
the use of high anity, high specicity antisera that allow precise and reproducible measurements of the biologically active peptide. Cross comparisons of various commercial assays show that the results generally are similar when values are in the
normal range (Clemmons,2007).
However, the assays have dierent performance characteristics when concentrations are either above or below the normal range.
To obtain cross laboratory standardization for values outside the normal range requires utilization of similar, high-quality reagents and techniques that are reasonably comparable. Without this degree of standardization, cross comparisons among various
reference laboratories are likely to continue to show wide divergence for values that are above or below the 95% confidence interval. A future goal should be the development of standard procedures and reagents that eliminate this degree of variability (Clemmons,2007; Anckaert et al,2006).
In conclusion, the automated IMMULITE IGF-I assays are reliable tools in the diagnosis of pathologies of the GH/IGF axis and in the follow-up of their therapies. IGF-1 determination by RIA had a signicantly higher variation interassay than chemiluminescence, that seems to have a better correlation to GH nadir on oGTT in acromegalic patients


REFERENCES
1-Clemmons DR. IGF-I assays: current assay methodologies and their limitations. Pituitary. 2007 Apr 11
2-Anckaert E, Schiettecatte J, Vanbesien J, Smitz J, Velkeniers B, De Schepper J. Variability among ve dierent commercial IGF-1
immunoassays in conditions of childhood-onset GH decienc y and GH therapy. Acta Clin Belg. 2006 Nov-Dec;61(6):335-9
3-Emlingher MW, Kune W, Weber MM, Ranke MB. Reference ranges for two automated chemiluminescent assays for serum insulin-like growth factor I (IGF-I) and IGF-binding protein 3 (IGFBP-3).
Clin Chem Lab Med. 2004;42(6):654-64.
4-Brabant et al. Serum insulin-like growth factor I reference values for an automated chemiluminescence immunoassay system: results from a multicenter study.Horm Res. 2003;60(2):53-60


Laboratório Sabin de Análises Clínicas
Brasília Shopping, Torre Sul sala 104
Brasília - DF
tel. (61) 3329-8000
www.sabinonline.com.br
sabin@sabinonline.com.br


Empresas Parceiras