Disorientation as a delirium feature in non-intubated patients: development and evaluation of diagnostic accuracy of the 'Confusion Assessment Method for Intermediate Care Unit' (CAM-IMC) - a prospective cohort study.

Background: Disorientation is an early indicator of developing postoperative delirium (POD), which is associated with increased mortality and cognitive decline. The well-established "Confusion-Assessment-Method-for-Intensive-Care-Unit" (CAM-ICU) for diagnosing POD in intubated patients cannot make use of the feature 'disorientation', as this requires verbal communication. Other tools such as the 4AT test for disorientation but are not established in ICU settings. We therefore combined test-variables of the CAM-ICU (level of consciousness, fluctuating mental status and inattention) with verbal testing for disorientation to develop and enhance diagnostic accuracy of the "Confusion Assessment Method for Intermediate Care Unit" (CAM-IMC). In the present study we describe the development and the evaluation of the diagnostic accuracy of the CAM-IMC.

Methods: We conducted a prospective cohort-study to develop and evaluate the diagnostic accuracy of the CAM-IMC and disorientation for diagnosing POD in non-intubated patients undergoing elective cardiac surgery. All patients were eligible during data collection period. Exclusion criteria were preexisting brain-organic disease, age < 50 years, preoperative intubation, and insufficient language skills. Patients were assessed for POD using the CAM-IMC as the index-test by two independent examiners over three postoperative days. Reference-testing was conducted by experienced reference-raters. The primary outcome was the diagnostic test-performance.

Results: Among 178 eligible patients, 624 paired observations were completed with 155 patients. Of these, 9% experienced POD. Sensitivity and specificity were 0.96 (CI-95%: 0.87-1.00) and 0.94 (CI-95%: 0.92-0.96), respectively. Area-Under-the-Receiver-Operating-Characteristic-Curve (AUROC; equivalent to c-statistic) for CAM-IMC with a cut-off at three points was 0.95 (CI-95%: 0.93-0.98). The interrater reliability was 0.80 (CI-95%: 0.69-0.91).

Conclusion: The CAM-IMC demonstrates excellent test performance for diagnosing POD in non-intubated patients by combining features of the CAM-ICU with 'disorientation'. Given an aging community with an increasing delirium risk, the CAM-IMC provides a highly structured assessment tool for POD. It enables early and accurate detection of delirium, which is critical for timely intervention and improved patient outcomes. The CAM-IMC appears to be a useful tool to be implemented in units for not-intubated patients and seems to be the perfect match where the CAM-ICU is already in use for monitoring POD.

Trial Registration: DRKS00026980 (German registry of clinical studies).