Background Studies focused on C-reactive protein (CRP) as a marker of difficult laparoscopic cholecystectomy are limited to small case series. The aim of this study was to evaluate the association between preoperative CRP concentration and difficulty of laparoscopic cholecystectomy in patients admitted with a biliary emergency presentation. Methods Patients with an emergency admission for biliary disease treated between 2012 and 2017 with a documented preoperative CRP level were analysed. Elective patients and those with other concurrent causes of increased CRP concentration were excluded. The intraoperative difficulty grade was based on the Nassar scale. Statistical analysis was conducted to determine the association of preoperative CRP level with difficulty grading, adjusted for the interval to surgery. Results A total of 804 emergency patients were included. The mean preoperative peak CRP level was 64 center dot 7 mg/l for operative difficulty grade I, 69 center dot 6 mg/l for grade II, 98 center dot 2 mg/l for grade III, 217 center dot 5 mg/l for grade IV and 193 center dot 1 mg/l for grade V, indicating a significant association between CRP concentration and Nassar grade (P < 0 center dot 001). Receiver operating characteristic (ROC) curve analysis showed an area under the curve of 0 center dot 78 (95 per cent c.i. 0 center dot 75 to 0 center dot 82), differentiating patients with grade I-III from those with grade IV-V operative difficulty. ROC curve analysis found a cut-off CRP value of 90 mg/l, with 71 center dot 5 per cent sensitivity and 70 center dot 5 per cent specificity in predicting operative difficulty of grade IV or V. Logistic regression analysis found preoperative peak CRP level to be predictive of Nassar grade I-III versus grade IV-V operative difficulty, also when adjusted for timing of surgery (odds ratio 5 center dot 90, 95 per cent c.i. 2 center dot 80 to 12 center dot 50). Conclusion Raised preoperative CRP levels are associated with greater operative difficulty based on Nassar scale grading.