'In vitro' binding of propofol to serum lipoproteins in thyroid dysfunction

Objective: To determine a regression relationship between the unbound fraction (fu percent) of propofol, a highly lipophilic intravenous anaesthetic agent, and demographic and biochemical variables in a thyroid dysfunction population. Methods: Serum samples from patients with hypo (n=33) and hyperthyroidism (n=33) and also from healthy volunteers (control group; n=9) were spiked with propofol to a total (bound + unbound propofol) concentration of 10 mug/ml. The unbound concentration was determined using ultrafiltration followed by high-performance liquid chromatography with fluorimetric detection and the unbound fraction percent (fu) and the binding ratio [bound/free concentration (B/F)] were calculated. Albumin, alpha-acid glycoprotein, cholesterol, triglycerides, HDL, LDL, VLDL lipoproteins, and T3 and T4 hormone concentrations were measured. B/F has a linear relationship with lipoprotein concentration - unlike that for fu which is curvilinear - so, we developed a linear regression model of B/F for propofol with the above biochemical variables and with gender. Results: The fu in hypothyroidism was significantly lower than in the healthy control (mean standard deviation 0.74 +/- 0.13% vs 0.87 +/- 10.12%, P<0.01) but the fu was no different in hyperthyroid subjects than controls (0.94 +/- 0.16% vs 0.87 +/- 0.12%, P > 0.05). HDL, LDL and VLDL lipoproteins were significant predictors of B/F (P < 0.05) and were capable of explaining 54% of the variability in the 'in vitro' binding of propofol in thyroid disease. Conclusion: These results could help explain the interindividual variability in the protein binding of propofol in thyroid dysfunction patients and suggest the inclusion of lipoproteins in covariate model development for the pharmacokinetic/pharmacodynamic parameters of propofol.