Implementation of thyroid function tests algorithms by clinical laboratories: A four-year experience of good clinical and diagnostic practice in a tertiary hospital in Greece.

INTRODUCTION: Thyroid Function Tests (TFTs) are among the most commontly ordered tests. Significant overuse of TFTs can occur when instead of using a single TSH test to screen for thyroid disease a full panel (TSH plus FT4 and FT3) is ordered. The aim of our study was to evaluate the effectiveness of the application of a scientifically-established laboratory-controlled algorithm for TFTs to physician's orders for inpatients and to address potential pitfalls of such an approach. MATERIALS AND METHODS: We collected and analyzed Laboratory Information System data of the TFTs performed between April 2009 and March 2016 in a 739-bed tertiary teaching hospital. Between April 2013 and March 2016, we applied a laboratory controlled algorithm for inpatient TFT assays after TSH and did not perform further tests, unless a justified bypass was requested by the treating physician. RESULTS: Algorithm application led to significant reductions of TFTs executed per TSH ordered. Compared to the four years preceding the intervention, executed FT4/TSH tests decreased from 93 to 18%, FT3/TSH from 92 to 18%, anti-TG/TSH from 18 to 4% and anti-TPO/TSH from 11 to 3%. Simultaneously, FT4, FT3, anti-TG, and anti-TPO tests ordered in outpatients also displayed a significant gradual decrease. CONCLUSIONS: Hospital-based laboratories can safely apply a generally accepted TFTs algorithm on physician's orders without any compromise in diagnostic/therapeutic accuracy, thus achieving significant direct cost-reduction and increased physician awareness on current TFT ordering practices. Such an approach, combined with collaboration with ordering physicians, can safeguard patients from the consequences of low-value care practices.

Gingival condition of women with gestational diabetes on a Greek island.

The purpose of this study was to investigate the condition of gingival tissues in women who had gestational diabetes mellitus. The study group consisted of 64 women with gestational diabetes and the control group of 88 healthy pregnant women. The plaque index and gingival index and their relationship to a series of variables (socio-demographic, diet, oral hygiene, oral condition) were evaluated. Clinical evaluations were carried out between 34 and 36 weeks of pregnancy. The women with gestational diabetes had a higher mean plaque index and higher mean gingival index than the healthy pregnant women (p = 0.048 and p = 0.029, respectively). Linear regression analysis showed an association between gingival index and plaque index (p < 0.001). Gestational diabetes was not significantly related to gingivitis when other factors that differed between the two groups were taken into account. The results of the study suggest that gingival inflammation seems to be more prevalent in women with gestational diabetes compared to healthy pregnant women, while the plaque accumulation remains the main cause of gingival inflammation.