2144. Vital signs are vital in identifying high risk postoperative patients
Session: Poster Abstract Session: Healthcare Epidemiology: Surgical Site Infections
Saturday, October 6, 2018
Room: S Poster Hall
Posters
  • IDSA 2018 Vital Signs are Vital Poster 3.pdf (481.3 kB)
  • Background: Changes in vital signs are frequently the first sign to point to pathology in the postoperative setting. There is no prediction model that exists that evaluates risk of postoperative complication in real-time. We are interested in understanding if we are able to risk stratify patients after surgery using novel predictors, trajectories of the various vital signs and evaluating their ability to risk stratify patients.

    Methods: We reviewed patients who underwent pancreatectomy at an academic health system from 1/15-2/18. Postoperative complications were abstracted using definitions set by the National Surgical Quality Improvement Program (NSQIP) and vital signs, including pain score, were extracted from the Data Warehouse. Group based trajectory modeling, a technique used to identify distinct clusters of patients with similar trajectories, was used to group patients with similar temperature, heart rate, blood pressure and pain scores. Postoperative complications were tabulated for each risk group and chi-square test was used to compare categorical variables.

    Results: 195 patients with pancreatectomy were evaluated and the rate of NSQIP complications was 35.4%. Pancreatectomy patients clustered into two distinct clusters for temperature, heart rate, systolic blood pressure and pain score. All four of these vital signs were able to stratify infectious and inflammatory complications between low and high risk groups but only systolic blood pressure was significant in stratifying readmission risk and heart rate and pain score for stratifying sepsis risk (Table 1).

    Conclusion: Trends of vital signs may be important predictors of complications. Some vital signs may be better at predicting distinct complications. More work is required to understand if different covariates within trajectory analysis can be combined to further enhance risk stratification for any and specific postoperative complications.

     

     

    Sepsis %

    Any complication %

    Readmission %

    Temperature High

    6.1

    27.0*

    14.8

    Temperature Low

    6.8

    41.9*

    16.2

    HR High

    2.8*

    25.7*

    13.8

    HR Low

    11.3*

    42.5*

    17.5

    SBP High

    6.9

    23.0*

    9.2*

    SBP Low

    5.9

    41.2*

    20.6*

    Pain score High

    3.5*

    27.0*

    15.7

    Pain score Low

    10.8*

    41.2*

    14.9

    Table 1: Rates of complications by trajectory analysis. * = significant at p<0.05

     

    Eric Bhaimia, DO1, Urmila Ravichandran, MS2, Elias Baied, DO1, Frances Lahrman, DO3, Huma Saeed, MD1, Katherine Kaplar, DO4, Ronak Parikh, DO4, Jennifer Paruch, MD5, Rema Padman, PhD6, Jennifer Grant, MD7 and Nirav Shah, MD, MPH8, (1)Internal Medicine, University of Chicago (NorthShore), Evanston, IL, (2)NorthShore University HealthSystem, Evanston, IL, (3)Northshore University HealthSystem, Evanston, IL, (4)University of Chicago, Chicago, IL, (5)Surgery, NorthShore University HealthSystem, Evanston, IL, (6)Healthcare Informatics, Carnegie Mellon University, Pittsburgh, PA, (7)Infectious Disease, NorthShore University HealthSystem, Evanston, IL, (8)Infectious Diseases / Informatics, NorthShore University Health Systems, Evanston, IL

    Disclosures:

    E. Bhaimia, None

    U. Ravichandran, None

    E. Baied, None

    F. Lahrman, None

    H. Saeed, None

    K. Kaplar, None

    R. Parikh, None

    J. Paruch, None

    R. Padman, None

    J. Grant, None

    N. Shah, None

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 3rd with the exception of research findings presented at the IDWeek press conferences.