Multiple studies have shown that the two different citrate concentrations in common use as the anticoagulant in blood collection for haemostasis assays can affect the results obtained with the prothrombin time assay. It is clear from the literature that there is considerable variability in the results obtained using different instrument-reagents combinations, but the clinical relevance of these differences is unclear. Most of the studies have used an optical system for end-point detection. This study reports on the citrate sensitivity using mechanical end-point detection. Using two different reagents, one previously shown to be citrate sensitive on optical systems (Neoplastin CI plus) and a citrate-insensitive reagent (Neoplastin CI), we demonstrate that the effect of using different citrate concentrations (0.105 or 0.129 mol/l) has statistically significant but clinically irrelevant effects on the International Normalized Ratio using a mechanical instrument (STA)-reagent combination (mean percentage difference in results, 1.9 and 3.8% respectively). This demonstrates that the citrate effect is both instrument type and reagent dependent. Every reagent and instrument combination needs to be tested to see whether any citrate effect exists. In a secondary study, it was shown that the international reference rabbit thromboplastin (CRM 149(s)) was not citrate-concentration sensitive.