Myocardial Injury and Intraoperative Tissue Oximetry in Patients Undergoing Spine Surgery

Significance:

Each year 10 million people worldwide suffer from myocardial injury after non-cardiac surgery
(MINS). The majority of these events are undetected because they are asymptomatic and there
is sparse knowledge about triggering causes. Major surgery is associated with a high risk of
postoperative complications. These include covert stroke, myocardial injury and -infarction,
and are all major contributors to mortality. The 30-day post-operative mortality in patients
suffering from covert stroke is increased up to eight-fold compared with matched controls and
is 10% in patients with MINS, which makes it a substantial public health problem.
Approximately 33% of patients undergoing major spine surgery suffer from major postoperative
complications, including stroke and myocardial infarction. However, knowledge is lacking
regarding which perioperative factors predict myocardial injury and -infarction, covert and
overt stroke, and mortality in high-risk patients such as those undergoing major spine
surgery.

Measure of tissue oxygenation with near-infrared spectroscopy (NIRS) is non-invasive and
uncomplicated. Tissue desaturation is common in spine surgery patients. Previous studies
found that a decrease in tissue oxygenation (StO2) was associated with postoperative
complications such as creatinine elevation, hypotension and prolonged hospitalization.
However, current knowledge of how tissue oxygenation affects other clinical outcomes is
lacking.

Thus, to address these knowledge gaps the investigators propose a cohort study to examine the
association between tissue oxygenation and MINS in patients undergoing major spine surgery.

Methods This is a prospective observational cohort study. Participants are patients
undergoing major spine surgery. Measurements include blood samples, non-invasive tissue
oximetry and collection of general perioperative factors from the anesthesia machine. First
measurements are made on the day of surgery and further measurements are conducted on the
first two postoperative days. Secondary outcomes from medical records (e.g. 30-day mortality)
are collected retrospectively.

Patients will be screened from Operation Room Surgery Schedule at University of California,
San Francisco (UCSF) prior to surgery. The informed consent will be obtained at the
preoperative area about 30-60 min prior to surgery. Eligible patients will sign a written
consent form and HIPAA Authorization for study participation. If the subject is determined to
have diminished or lack a capacity to consent, a surrogate/legally authorized representative
will need to be present to authorize their participation in the study and complete the
applicable forms as required by UCSF and Californian law.

Data from medical records are entered by certified study personnel into REDCap (a HIPAA
compliant online data collection tool) to assure all data is complete and stored correctly.
Included in REDCap is a data dictionary with descriptions of every individual variable.
Standard Operating Procedures (SOP's) for patient recruitment and enrollment, data collection
and data extraction have been created and are available. Tissue Oximetry data is collected
separately with the laboratory's oximeters. After each case/study the data is uploaded to an
online secure box. As all data are collected case-by-case in the course of the study, all
missing data are registered immediately as missing in REDCap and will be treated as missing
data in the final data analysis.

Sample size:

The sample size assessment is based on previous studies investigation tissue oximetry and
clinical outcomes.

Effect size: 0.36 (SD=0.50) based on Meng et al. (Br J Anaesth. 2017 Apr 1;118(4):551-562.
doi: 10.1093/bja/aex008.):

Muscle tissue oxygenation (SmO2) time weighted area under the curve (AUC) (%*min*hr-1 (SD)):

<3 complications: 1.4 (0.8) >3 complications: 1.9 (1.2) p-value: 0.07 Proportion of patients
with >3 composite complications = 33% Alpha: 0.05 Beta: 0.2 (power=0.8) Sample size: 68. We
expect to include 70 patients in this study.

By convention the 5% level of alpha and 20% level of beta have been used. As this study is
explorative and hypothesis-testing, the investigators consider the above calculation to
contain sufficient information.

Statistical analysis:

Primary analysis will be multiple regression with adjustment for age, sex and pre-existing
conditions including cardiovascular disease, coronary heart disease, heart failure, diabetes
mellitus and obstructive respiratory diseases. The following pre-specified potential
confounders will be tested separately: age, gender, BMI, smoking status, history of coronary
artery disease, congestive heart failure, stroke, transient ischemic attacks, chronic
obstructive pulmonary disease, insulin therapy, hypertension, preoperative creatinine
elevation above normal, and length of surgery. Since this relationship may or may not be
linear, the data will first visually be assessed by plotting the estimated probability (on
the logit scale) of having the outcome as a function of StO2, using a univariable logistic
regression incorporating a smooth (thin-plate regression spline) term for StO2 (smoothing
parameter obtained via cross-validation). The fit of each model will be assessed by the
Hosmer-Lemeshow goodness-of-fit test. STATA software will be used for the statistical
procedures.

Inclusion Criteria:

- Male or female ≥18 years

- Patient is undergoing elective surgery of the spine

- Surgery is scheduled to last ≥ 2 hours and involve instrumentation

Exclusion Criteria:

- Patient is < 18 years

- Patient is undergoing emergent or urgent surgery

- American Society of Anesthesiologist (ASA) status > IV

- Patient is undergoing non-instrumental surgery, such as laminectomy alone

- Patient is undergoing spine surgery for tumor or infection