AIS impairment scale, injury characteristics, intraoperative data after spinal cord injury in patients from a TRACK-SCI cohort
DOI:10.34945/F5KG6Z
DATASET CITATION
Chou A., Beattie M. S., Bresnahan J. C., Burke J. F., Almeida C. A., Dhall S. S., DiGiorgio A. M., Duong-Fernandez X., Haefeli J., Hemmerle D. D., Huie J., Kyritsis N., Manley G. T., Moncivais S., Omondi C., Pascual L. U., Singh V., Talbott J. F., Thomas L. H., Torres-Espin A., Weinstein P. R., Whetstone W. D., Pan J. Z., Ferguson A. R. (2022) AIS impairment scale, injury characteristics, intraoperative data after spinal cord injury in patients from a TRACK-SCI cohort. ODC-SCI:727 http://doi.org/10.34945/F5KG6Z
ABSTRACT
STUDY PURPOSE: The aim of the data collection was to identify predictors of whether a patient would improve on the ASIA Impairment Scale (AIS) grade between time of hospital admission and hospital discharge. Predictors investigated include injury characteristics, anesthetics and vasopressors used during surgery, MRI BASIC score, length of surgery and hospital stay, and intraoperative hemodynamics. The data were collected at a level I trauma center (Zuckerberg San Francisco General Hospital) and span both a retrospective and prospective study for the purpose of predictive modeling and external validation respectively.
DATA COLLECTED: The dataset include: (1) various injury characteristics such as the level of injury, presence of polytrauma, type of injury (e.g. penetrating, hemorrhagic, central cord, cervical, fracture, vertebral artery), and whether the patient had previous history of SCI or TBI; (2) vasopressor and anesthetic usage during surgery; (3) length of surgery and hospital stay; (4) intraoperative fraction of inspired oxygen; (5) the duration the patient was within and outside MAP goals in the ICU; (6) MRI BASIC Score; and (7) various intraoperative hemodynamics. In particular, the hemodynamics were derived from time series data of heart rate and systolic, diastolic, and mean arterial blood pressure during surgery. This includes summary statistics of mean, standard deviation, skew, and kurtosis. For mean arterial blood pressure specifically, the duration that a patient was outside a specific threshold (upper or lower) was also determined and recorded in the dataset.
CONCLUSIONS: Predictive models optimized from the data show that initial injury severity (either AIS at time of hospital admission or MRI BASIC score) were highly predictive variables for whether the patient improved in outcome by the time of hospital discharge. Additionally, intraoperative hyper- and hypotension (high and low MAP) were critical predictors; patients that spent more time in a hyper and/or hypotensive state were less likely to improve. Hypertension in particular was more predictive of worse outcome than hypotension. The lower MAP threshold predicting worse outcome was around 74-76 mmHg and the upper MAP threshold was around 103-104 mmHg.
KEYWORDS
Spinal Cord Injury; Surgery; Hemodynamic; Blood pressure; Vasopressor; AIS score; Human
PROVENANCE / ORIGINATING PUBLICATIONS
Chou, A., Torres-Espin, A., Kyritsis, N., Huie, J.R., Khatry, S., Funk, J., Hay, J., Lofgreen, A., Shah, R., McCann, C., Pascual, L.U., Amorim, E., Weinstein, P.R., Manley, G.T., Dhall, S.S., Pan, J.Z., Bresnahan, J.C., Beattie, M.S., Whetstone, W.D., Ferguson, A.R., and the TRACK-SCI Investigators. (2021). Expert-integrated automated machine learning uncovers hemodynamic predictors in spinal cord injury. bioRxiv , 2021.09.27.461544.. doi:10.1101/2021.09.27.461544.
Originating publication (biorxiv preprint).
Torres-Espín A., Haefeli J., Ehsanian R., Torres D., Almeida C. A., Huie J., Chou A., Dirlikov B., Suen C. G., Nielson J. L., Kyritsis N., Duong-Fernandez X., Thomas L. H., Hemmerle D. D., Morozov D., Sanderson N., Talbott J. F., Manley G. T., Dhall S. S., Whetstone W. D., Bresnahan J. C., Beattie M. S., McKenna S. L., Pan J. Z., Ferguson A. R. (2021) ASIA Impairment Scale, level of injury, intraoperative time series mean arterial pressure and heart rate after spinal cord injury in patients in a multi-site retrospective TRACK-SCI cohort: site 2 of 2. ODC-SCI:246 http://doi.org/10.34945/F5MG68. doi:10.34945/F5MG68.
Some overlap in subjects and variables (subject IDs are not matched). The provenanced dataset also includes intraoperative hemodynamics at the 5min-interval granularity from which the summary statistics were derived.
Chou A, Torres-Espin A, Kyritsis N, Huie JR, Khatry S, Funk J, Hay J, Lofgreen A, Shah R, McCann C, Pascual LU, Amorim E, Weinstein PR, Manley GT, Dhall SS, Pan JZ, Bresnahan JC, Beattie MS, Whetstone WD, Ferguson AR; TRACK-SCI Investigators. Expert-augmented automated machine learning optimizes hemodynamic predictors of spinal cord injury outcome. PLoS One. 2022 Apr 7;17(4):e0265254. doi: 10.1371/journal.pone.0265254. PMID: 35390006; PMCID: PMC8989303. doi:10.1371/journal.pone.0265254.
Originating publication
RELEVANT LINKS
NOTES
- Vasopressor columns are taken directly from medical records and may include different names or grammatical conventions when reporting vasopressor usage. We recommend data users to standardize the vasopressor names during data processing for future analyses.
-Vasopressor columns also include cells with just "." entries. These are equivalent to missing values in the columns as encoded in the original medical records. We recommend data users to replace these with NA during future analyses.
|
|
DATASET INFO
Contact: Pan Jonathan (jonathan.pan@ucsf.edu), Ferguson Adam (adam.ferguson@ucsf.edu)
Lab: TRACK-SCI De-ID Lab
ODC-SCI Accession:727
Records in Dataset: 159
Fields per Record: 126
Last updated: 2022-04-25 (See changelog)
Date published: 2022-03-24
Downloads: 42
Files: 2
LICENSE
Creative Commons Attribution License (CC-BY 4.0)
FUNDING AND ACKNOWLEDGEMENTS
NIH/NINDS grants R01NS088475 (ARF), R01NS122888 (ARF), UH3NS106899 (ARF), U24NS122732 (ARF), Department of Veterans Affairs grants 1I01RX002245 (ARF) and I01RX002787 (ARF), Wings for Life Foundation (ARF), Craig H. Neilsen Foundation (ARF), Department of Defense grants SC150198 (MSB) and SC190233 (MSB), C.H. Neilsen Foundation (MSB)
CONTRIBUTORS
- Chou, Austin [ORCID:0000-0003-4328-5811]
- University of California, San Francisco
- Beattie, Michael S. [ORCID:0000-0001-9463-3631]
- University of California, San Francisco
- Bresnahan, Jacqueline C. [ORCID:0000-0003-2243-7054]
- University of California, San Francisco
- Burke, John F. [ORCID:0000-0002-6190-5116]
- University of California, San Francisco
- Almeida, Carlos A. [ORCID:0000-0001-9997-8719]
- University of California, San Francisco
- Dhall, Sanjay S. [ORCID:0000-0002-6891-2722]
- University of California, San Francisco
- DiGiorgio, Anthony M. [ORCID:0000-0002-5710-691X]
- University of California, San Francisco
- Duong-Fernandez, Xuan [ORCID:0000-0001-8362-4166]
- University of California, San Francisco
- Haefeli, Jenny [ORCID:0000-0002-3546-728X]
- University of California, San Francisco
- Hemmerle, Debra D. [ORCID:0000-0003-2796-6107]
- University of California, San Francisco
- Huie, J. Russell [ORCID:0000-0001-5594-4277]
- University of California, San Francisco
- Kyritsis, Nikos [ORCID:0000-0001-7801-5796]
- University of California, San Francisco
- Manley, Geoffrey T. [ORCID:0000-0002-0926-3128]
- University of California, San Francisco
- Moncivais, Sara [ORCID:0000-0001-5506-9804]
- University of California, San Francisco
- Omondi, Cleopa [ORCID:0000-0001-8003-1146]
- University of California, San Francisco
- Pascual, Lisa U. [ORCID:0000-0001-6228-6217]
- University of California, San Francisco
- Singh, Vineeta [ORCID:0000-0003-3470-3532]
- University of California, San Francisco
- Talbott, Jason F. [ORCID:0000-0003-3943-5530]
- University of California, San Francisco
- Thomas, Leigh H. [ORCID:0000-0002-7593-0240]
- University of California, San Francisco
- Torres-Espin, Abel [ORCID:0000-0002-9787-8738]
- University of California, San Francisco
- Weinstein, Philip R. [ORCID:0000-0003-4038-745X]
- University of California, San Francisco
- Whetstone, William D. [ORCID:0000-0002-8568-6640]
- University of California, San Francisco
- Pan, Jonathan Z. [ORCID:0000-0001-5814-3707]
- University of California, San Francisco
- Ferguson, Adam R. [ORCID:0000-0001-7102-1608]
- University of California, San Francisco
|