Open Access Research article

Relationship between intraoperative regional cerebral oxygen saturation trends and cognitive decline after total knee replacement: a post-hoc analysis

Fátima Salazar1*, Marta Doñate2, Teresa Boget2, Ana Bogdanovich1, Misericordia Basora1, Ferran Torres3, Isabel Gracia1 and Neus Fàbregas1

Author Affiliations

1 Department of Anesthesia, Hospital Clinic de Barcelona, (Universitat de Barcelona), Barcelona, Spain

2 Department of Neuropsychology, Hospital Clinic de Barcelona, (Universitat de Barcelona), Barcelona, Spain

3 Statistics & Methodology Support Unit (USEM), IDIBAPS, (Hospital Clinic de Barcelona), Laboratory of Biostatistics & Epidemiology (Universitat Autonoma de Barcelona), Barcelona, Spain

For all author emails, please log on.

BMC Anesthesiology 2014, 14:58  doi:10.1186/1471-2253-14-58

Published: 21 July 2014



Bilateral regional brain oxygen saturation (rSO2) trends, reflecting intraoperative brain oxygen imbalance, could warn of brain dysfunction. Various types of cognitive impairment, such as memory decline, alterations in executive function or subjective complaints, have been described three months after surgery. Our aim was to explore the potential utility of rSO2 values as a warning sign for the development of different types of decline in postoperative psychological function.


Observational post-hoc analysis of data for the patient sample (n = 125) of a previously conducted clinical trial in patients over the age of 65 years undergoing total knee replacement under spinal anesthesia. Demographic, hemodynamic and bilateral rSO2 intraoperative values were recorded. An absolute rSO2 value of <50% or a reduction of >20% or >25% below baseline were chosen as relevant cutoffs. Composite function test scores were created from baseline to three months for each patient and adjusted for the mean (SD) score changes for a control group (n = 55). Tests were used to assess visual-motor coordination and executive function (VM-EF) (Wechsler Digit Symbol-Coding and Visual Reproduction, Trail Making Test) and memory (Auditory Verbal Learning, Wechsler Memory Scale); scales were used to assess psychological symptoms.


We observed no differences in baseline rSO2 values; rSO2 decreased significantly in all patients during surgery (P < 0.0001). Seventy-five patients (60%) had no sign of cognitive decline or psychological symptoms. Twenty-one patients (16.8%) had memory decline, 3 (2.4%) had VM-EF decline, and 33 (26.4%) had psychological symptoms. Left and right rSO2 values were asymmetric in patients who had memory decline (mean [SD] left-right ratio of 95.03 [8.51] vs 101.29 [6.7] for patients with no changes, P = 0.0012). The mean right-left difference in rSO2 was also significant in these patients (-2.87% [4.73%], lower on the right, P = 0.0034).


Detection of a trend to asymmetry in rSO2 values can warn of possible postoperative onset of memory decline. Psychological symptoms and memory decline were common three months after knee replacement in our patients over the age of 65 years.

Cerebral oximetry; Near-infrared spectroscopy; Postoperative memory dysfunction; Postoperative visual-motor coordination and executive function; Postoperative psychological decline