ObjectivesTo compare the determinants of trimethoprim-sulfamethoxazole resistance with established susceptibility values for fastidious Haemophilus spp., to provide recommendations for optimal trimethoprim-sulfamethoxazole measurement. MethodsWe collected 50 strains each of Haemophilus influenzae and Haemophilus parainfluenzae at Bellvitge University Hospital. Trimethoprim-sulfamethoxazole susceptibility was tested by microdilution, E-test and disc diffusion using both Mueller–Hinton fastidious (MH-F) medium and Haemophilus test medium (HTM) following EUCAST and CLSI criteria, respectively. Mutations in folA, folP and additional determinants of resistance were identified in whole-genome-sequenced isolates. ResultsStrains presented generally higher rates of trimethoprim-sulfamethoxazole resistance when grown on HTM than on MH-F, independent of the methodology used (average MIC 2.6-fold higher in H. influenzae and 1.2-fold higher in H. parainfluenzae). The main resistance-related determinants were as follows: I95L and F154S/V in folA; 3- and 15-bp insertions and substitutions in folP; acquisition of sul genes; and FolA overproduction potentially linked to mutations in -35 and -10 promoter motifs. Of note, 2 of 19 H. influenzae strains (10.5%) and 9 of 33 H. parainfluenzae strains (27.3%) with mutations and assigned as resistant by microdilution were inaccurately considered susceptible by disc diffusion. This misinterpretation was resolved by raising the clinical resistance breakpoint of the EUCAST guidelines to ≤30 mm. ConclusionsGiven the routine use of disc diffusion, a significant number of strains could potentially be miscategorized as susceptible to trimethoprim-sulfamethoxazole despite having resistance-related mutations. A simple modification to the current clinical resistance breakpoint given by the EUCAST guideline for MH-F ensures correct interpretation and correlation with the reference standard method of microdilution.