The Importance of Delirium in Patients with Severe Aortic Stenosis Referred for SAVR or TAVR

Rationale for Study: Delirium has been associated with poor long-term outcomes including mortality and functional decline after major cardiac surgery. Transcatheter aortic valve replacement (TAVR) imposes less physiologic stress than a surgical approach and may be associated with lower rates of delirium and faster functional recovery. The authors investigated the association of delirium with functional outcomes after TAVR and surgical aortic valve replacement (SAVR).

Funding: National Center for Research Resources (KL2 TR001100-01), National Center for Advancing. Translational Sciences (UL1 TR001102), National Heart, Lung, and Blood Institute (T32-HL007374).


Study Design: Prospective cohort study.

Study population: A single-center (Beth Israel Deaconess Medical Center; Boston, MA) ancillary, study of functional and cognitive outcomes. Main study was the Frailty Assessment Before Cardiac Surgery and Transcatheter Interventions study (FRAILTY-AVR). Current study included patients 70 years and older undergoing TAVR or SAVR for severe aortic stenosis (AS). In this study, 446 patients were screened, 246 patients enrolled, and 187 included in the final analysis. Patients were followed up for 12 months after valve replacement with serial telephone interviews to assess function.


  1. Daily postoperative Delirium Assessment utilizing the Confusion Assessment Method (CAM).
  2. A validated CAM-based severity scale was applied (CAM-S, range 0-19), with higher scores representing worse delirium and further categorized in three groups: no; mild; and severe delirium.
  3. The motoric type of delirium was also defined: mixed or hyperactive; hypoactive; and normoactive delirium.


  1. Prolonged hospitalization (≥9 days).
  2. Institutional discharge (skilled nursing facility or rehabilitation).
  3. Poor functional recovery or death at 6 months.
  4. Change in functional status over 12 months.

Functional status score (range 0-22): Composite of the ability to perform 22 activities independently. Included 7 activities of daily living (ADL), 7 instrumental ADL, and 8 physical tasks.

Statistical Analysis: Standard descriptive statistics. Within the SAVR and TAVR cohorts, the outcomes risk was calculated by delirium status and severity using logistic regression. Longitudinal change in functional status over 12 months by delirium status was evaluated with linear mixed-effects model. Analyses were adjusted for age, sex, and Charlson Comorbidity Index (CCI). The investigators created a measure of Minimum Clinically Important Difference (MCID) in CAM-S; it was defined as the difference in the mean of the maximum CAM-S score between those who had poor functional recovery or death at 6 months versus those who did not.


  1. Patients who underwent SAVR had a lower burden of predisposing risk factors for delirium than those having TAVR. The incidence of delirium however was higher in SAVR patients than TAVR patients (50.7% vs 25.5%; P < .001), but SAVR patients had a shorter duration and lower severity of delirium as compared to TAVR patients.
  2. Higher delirium severity was associated with a greater risk of prolonged hospitalization and discharge to a post-acute care setting after SAVR and TAVR. There was also a greater incidence of poor functional recovery or death at 6 months in TAVR patients.
  3. The mean maximum CAM-S score for patients who experienced poor recovery or death at 6 months versus those who did not was 8.0 versus 5.7 for SAVR (MCID 2.3), and 6.9 vs 4.3 (MCID 2.6) for TAVR.
  4. All 3 SAVR patients and 7 out of 18 TAVR patients who died during the 12 months follow up experienced post-operative delirium. Patients who developed delirium after SAVR and TAVR had lower baseline mean functional status scores than those without delirium in their respective cohort.
    In the SAVR group, the difference in function in patients with delirium resolved by 6 months after surgery as compared to those without, but severe delirium patients had a slower recovery.
    In the TAVR cohort, patients with delirium remained functionally impaired throughout the 12 months follow up period as compared to those without, and those with severe delirium experienced significant and persistent functional decline post-operatively.

Study limitations:

  1. Modest sample size.
  2. Inability to compare SAVR and TAVR cohorts' outcomes due to the observational nature of this study.
  3. Unknown role of perioperative narcotics and anesthetic use in delirium incidence.

Take Home Message: Recognizing post-operative delirium as a marker for delayed or impaired functional recovery and adverse outcomes in older adults undergoing SAVR and TAVR. Consideration should be given to pre-operative delirium prevention strategies and post-operative interventions aimed to optimize recovery in patients who develop delirium after surgical or percutaneous aortic valve replacement.


  1. Shi SM, Sung M, Afilalo J, et. al. Delirium incidence and functional outcomes after transcatheter and surgical aortic valve replacement. J Am Geriatr Soc 2019;67:1393-1401.

Clinical Topics: Cardiac Surgery, Geriatric Cardiology, Invasive Cardiovascular Angiography and Intervention, Valvular Heart Disease, Aortic Surgery, Cardiac Surgery and VHD, Interventions and Structural Heart Disease

Keywords: Geriatrics, Transcatheter Aortic Valve Replacement, Aortic Valve, Activities of Daily Living, Skilled Nursing Facilities, Research Personnel, Prospective Studies, Aortic Valve Stenosis, Delirium, Heart Valve Prosthesis, Hospitalization, Comorbidity, Cognition

< Back to Listings