Reliability of exophthalmos measurement and the exophthalmometry value distribution in a healthy Dutch population and in Graves' patients. An exploratory study.

AIM

The purpose of this study was to test the reliability of an exophthalmometer commonly used in the Netherlands; to determine the exophthalmometry value distribution with this instrument and to assess the upper exophthalmometry limits of normal in a healthy, adult, Caucasian, Dutch population. Furthermore, to assess the effects of gender and age on exophthalmometry readings in this group and in a group of Graves' patients by comparing healthy, adult, Caucasian, Dutch individuals with adult, Caucasian, Dutch Graves' patients.

METHODS

To test the reliability of our Hertel exophthalmometer, we determined the interobserver variation between two observers by measuring 160 eyes in healthy, adult, Caucasian, Dutch females and males (10 females and 10 males in each decade between 20 and 60 years of age). These data were also used for the assessment of the Hertel value distribution and for defining the upper limits of normal in these individuals by logistic regression analysis. The effects of disease, age and gender were established using these data plus data of a retrospective study of 393 adult, Caucasian, Dutch females (n=294) and males (n=99) with Graves' orbitopathy in whom Hertel values were measured with the same exophthalmometer.

RESULTS

Exophthalmometry using an Hertel exophthalmometer appeared reliable (Pearson correlation coefficient for interobserver variation 0.89; 96% of the Hertel values measured by two observers were within the limits (of 2 mm) of agreement). Hertel values usually show a normal distribution in healthy individuals and in Graves' patients and are sex- and age-dependent, but there was no dependence on age in this small series in adults. Logistic regression analysis revealed an upper limit of normal of 16 mm in females and 20 mm in males in our group, using the exophthalmometer described.

CONCLUSIONS

Exophthalmometry is reliable and absolute measurement of proptosis is feasible. International standardization of Hertel exophthalmometry is required in order to compare exophthalmometry data in the literature reliably.