Objectives The aims of the present study were to assess the influence of mild to moderate hyperglycaemia and body weight on brain 2-[¹⁸F]fluoro-2-deoxy-d-glucose ([¹⁸F]FDG) PET, and to what extent a simple algorithm for maintaining count density may compensate for these effects. Methods We prospectively included 63 patients undergoing routine brain [¹⁸F]FDG PET. Scan time and injected activity were adjusted in patients with hyperglycaemia or increased body weight. Measures of perceived image quality, image noise and image contrast were obtained in both standard scans and intervention scans. Results Elevated blood glucose and increased body weight were associated with reduced count density and increased image noise that in turn were associated with lower scores of perceived image quality. The proposed simple algorithm effectively maintained the image noise level and improved perceived image quality across the full range of elevated blood glucose values and body weights, although the effect of intervention on perceived image quality was attenuated by lower image contrast in patients with moderate hyperglycaemia. In patients with increased body weight or blood glucose, the fraction of scans with poor image quality decreased from 9/29 to 2/29 (P = 0.04) and the fraction with good image quality increased from 7/29 to 20/29 (P = 0.001) when applying the proposed algorithm. Conclusions Increasing blood glucose and body weight are associated with increased image noise in standard imaging conditions. Improving count density by prolonging scan time and increasing injected activity significantly improves image quality in hyperglycaemic patients, although the image contrast remains reduced in patients with most pronounced hyperglycaemia.