Subsyndromal delirium after cardiac surgery: a clinically relevant but overlooked condition

This invited commentary refers to ‘Comparison of the confusion assessment method for intensive care units seven-item scale for intensive care units and the intensive care delirium screening checklist in predicting adverse outcomes for patients undergoing cardiovascular surgery’, by H.-S. Hsieh and P.-L. Chou, https://doi.org/10.1093/eurjcn/zvaf136. Delirium remains a frequent and serious complication after cardiac surgery with an incidence of ∼23% after open cardiac surgery1 and commonly leading to consequences including prolonged hospitalization, adverse events, and long-term cognitive decline.2,3 In clinical practice, delirium is often assessed using standardized bedside screening instruments such as the Confusion Assessment Method (CAM), the CAM-ICU, or the Intensive Care Delirium Screening Checklist (ICDSC). Despite the use of these validated tools, detection remains challenging. In particular, hypoactive and mixed delirium subtypes are frequently under-recognized, as their clinical manifestations are more subtle and easily overlooked by healthcare professionals. In a previous study published in the European Journal of Cardiovascular Nursing, the use of single-channel electroencephalogram (EEG) was shown to improve recognition of these often-missed cases, suggesting that objective measures may capture a broader range of delirium phenotypes than traditional screening instruments.4,5 In this current study by Hsieh and Chou,6 the approach towards assessing delirium is different: comparing two widely used screening instruments, the CAM-ICU-7 and ICDSC in cardiovascular surgical patients, with attention to both post-operative delirium (POD) and subsyndromal delirium (SSD).Their findings demonstrate that patients with both POD and SSD, as detected with the ICDSC, experienced notably worse outcomes than those with POD alone, including longer ICU and hospital stays and higher rates of complications. This could indicate that SSD is not a benign or transient phenomenon, but rather a clinically relevant state associated with adverse prognosis. Although the results of Hsieh and Chou are based on a relatively small cohort of 84 patients, they align closely with a larger study by Mailhot et al.,7 who studied 346 cardiac surgery patients, using the ICDSC as a screening tool. Mailhot et al. found that combined SSD and delirium was linked to the poorest outcomes. This aligns with the findings by Hsieh and Chou. However, Mailhot et al.7 also accounted for patients’ pre-operative state, which further strengthened the validity of this association. Patients with SSD had a higher risk profile pre-operatively and experienced serious complications such as sepsis, renal dysfunction, reoperation, and respiratory failure. The study raised the possibility that SSD may either represent a more severe form of delirium or act as a marker of underlying frailty and comorbidity.7 Importantly, the relationship between SSD and complications is likely bidirectional, making causal inference challenging. On the one hand, SSD may increase vulnerability to adverse events by impairing patient cooperation or increase vulnerability to invasive procedures. On the other, complications themselves may act as triggers that precipitate or worsen delirium symptoms. Thus, SSD should be considered a clinically meaningful condition along with manifest delirium. For clinicians, systematic screening with tools sensitive to SSD, such as the ICDSC, could be valuable to identify patients at risk of deterioration. Still, whether early recognition and targeted preventive interventions in SSD can change the trajectory and reduce the risk of manifest delirium is unknown. Larger prospective studies are warranted to clarify the prognostic role of SSD and to determine if it should be incorporated into future delirium management pathways. In the study by Hsieh and Chou, the inclusion criteria are narrow excluding patients with dementia, psychiatric illness, or emergent surgery. While such exclusions strengthen internal validity, they also reduce external validity, since these groups are those at highest risk of POD in real-world practice. Patients with pre-existing cognitive impairment or depression, for example, have been identified as particularly vulnerable to delirium in cardiac populations.8,9 Current NICE guidelines emphasize that these patients warrant systematic delirium risk assessment and tailored preventive strategies.10 The absence of these high-risk groups in research therefore limits the generalizability of the findings to everyday ICU settings, where heterogeneity is the norm and where delirium screening tools must perform under less controlled conditions. Additionally, delirium research has matured to a point where the association between delirium and adverse outcomes is firmly established. The next step may therefore be to design studies that mirror the complexity of our clinical populations more closely, even if this comes at the expense of some methodological control. By deliberately including patients who are often excluded from trials but dominate real-world ICUs, future research can provide evidence that is not only statistically robust but also clinically transferable, ensuring that screening tools and interventions are tested under the conditions in which they are actually used. Hsieh and Chou provide valuable evidence that delirium screening is not only about the choice of instrument but also about the type of delirium that is detected and how this relates to patient outcomes. The ICDSC, by capturing both POD and SSD and linking these states to adverse prognosis, appears well suited for cardiovascular surgical patients. At the same time, several knowledge gaps remain and we still lack consensus on definitions and management strategies for SSD. Finally, the future of delirium screening potentially lies in combining subjective clinical tools such as the ICDSC with objective measures like EEG previously described by Ten Broeke et al.,5 thereby uniting sensitivity with feasibility. Ultimately, delirium screening should be viewed not as the goal, but as a means to enable timely intervention, support recovery, and improve long-term outcomes patients in clinical practice. Nadja Buch Petersson (Conceptualization [equal], Writing—original draft [lead]) and Britt Borregaard (Conceptualization [equal], Writing—review & editing [supporting]) None declared.