Objectives <p>The purpose of this study was to propose a new difficulty score for the treatment of multiple gingival recessions (GRs) with Modified Coronally Advanced Tunnel (MCAT), and to test the score’s reliability.</p> Material and methods <p>A difficulty score was developed for the assessment and grading of 13 relevant anatomical parameters at baseline. Six experienced dental practitioners evaluated existing GRs in three patients. The scale was assessed at two levels: for each single tooth, and seven days later for whole quadrants. Accuracy was estimated using Fleiss’ kappa statistics and intraclass correlation coefficients (ICC) to determine the inter- and intra-examiner agreement.</p> Results <p>During the evaluation of individual teeth, all parameters exhibited very good to excellent agreement, with Fleiss’ kappa statistics ranging between 0.74 and 1. Only recession type, CEJ detectability, and cervical step displayed minor variations. The total MCAT score also produced a minor discrepancy (kappa = 0.78). During the evaluation of whole quadrants, excellent reproducibility was recorded for individual parameters (kappa 0.9—1). Minor variation was observed for recession type and CEJ detectability. Excellent reliability in the total MCAT scores was noted between examiners (kappa = 0.9). Overall intra-examiner agreement values were very high (0.75—1).</p> Conclusions <p>The reliability of the proposed difficulty score was assessed between very good and excellent. Some difficulties arose in the evaluation of recession type, CEJ detectability and cervical step.</p> Clinical relevance <p>The presented score may be considered for assessing the difficulty of multiple GRs treatment using MCAT in order to improve the process of clinical decision-making.</p>