Background: Medical students are taught that wind, water and wound complications occur at specific post-operative times. This may influence the timing of work-up for specific complications. The goal of this study was to investigate the relationship between post-operative temperature curve and time to diagnosis of inflammatory complications.
Methods: We reviewed patients who underwent pancreatectomy at an academic health system from 1/15-2/18. Clinical data including complications were extracted using definitions set by the National Surgical Quality Improvement Program and temperature was extracted from the Data Warehouse. Time of diagnosis, as determined by labs, microbiologic cultures, radiology and procedures, was extracted for each complication. Group based trajectory modeling, a technique used to identify distinct clusters of temperature trajectories of patients in the postoperative setting was used to group patients into low and high temperature trajectories.
Results: Among 195 patients who underwent pancreatectomy, 35.5% (69/195) experienced at least one complication within 30 days of surgery. Of the patients who developed complications, 49% (n=34) and 51% (n=35) were classified into the low and high temperature trajectory groups based on their temperature trajectory. For most individual inflammatory complications, time to diagnosis was later in the low rather than high temperature groups (Figure 1) and this was significant when averaging all inflammatory complications (12.7d low and 8.6d high; p = 0.002). Time to diagnosis tended to be later in the high rather than low temperature trajectory but this was not statistically significant when averaging all non-inflammatory complications (11.7d low and 11.9d high; p = 0.95).
Conclusion: We identified earlier diagnosis of inflammatory complications in patients with elevated temperature trajectories. There was no difference in timing of diagnosis for non-inflammatory complications. Temperature trajectory modeling may allow for earlier diagnosis of patients at high risk for inflammatory complications.
Figure 1: Time to diagnosis by low (L) and high (H) temperature trajectories.
U. Ravichandran, None
F. Lahrman, None
H. Saeed, None
K. Kaplar, None
R. Parikh, None
E. Bhaimia, None
R. Padman, None
J. Grant, None
M. Talamonti, None
N. Shah, None