Iron deficiency anaemia is strongly associated with poor outcomes after cardiac surgery. However, pre‐operative non‐anaemic iron deficiency (a probable anaemia precursor) has not been comprehensively examined in patients undergoing cardiac surgery, despite biological plausibility and evidence from other patient populations of negative effect on outcome. This exploratory retrospective cohort study aimed to compare an iron‐deficient group of patients undergoing cardiac surgery with an iron‐replete group. Consecutive non‐anaemic patients undergoing elective coronary artery bypass grafting or single valve replacement in our institution between January 2013 and December 2015 were considered for inclusion. Data from a total of 277 patients were analysed, and were categorised by iron status and blood haemoglobin concentration into iron‐deficient (n = 109) and iron‐replete (n = 168) groups. Compared with the iron‐replete group, patients in the iron‐deficient group were more likely to be female (43% vs. 12%, iron‐replete, respectively); older, mean (SD) age 64.4 (9.7) vs. 63.2 (10.3) years; and to have a higher pre‐operative EuroSCORE (median IQR [range]) 3 (2–5 [0–10]) vs. 3 (2–4 [0–9]), with a lower preoperative haemoglobin of 141.6 (11.6) vs. 148.3 (11.7) g.l⁻¹. Univariate analysis suggested that iron‐deficient patients had a longer hospital length of stay (7 (6–9 [2–40]) vs. 7 (5–8 [4–23]) days; p = 0.013) and fewer days alive and out of hospital at postoperative day 90 (83 (80–84 [0–87]) vs. 83 (81–85 [34–86]), p = 0.009). There was no evidence of an association between iron deficiency and either lower nadir haemoglobin or higher requirement for blood products during inpatient stay. After adjusting the model for pre‐operative age, sex, renal function, EuroSCORE and haemoglobin, the mean increase in hospital length of stay in the iron‐deficient group relative to the iron‐replete group was 0.86 days (bootstrapped 95%CI −0.37 to 2.22, p = 0.098). This exploratory study suggests there is weak evidence of an association between non‐anaemic iron deficiency and outcome after cardiac surgery after controlling for potentially confounding variables.