Dataset Information

Clinical factors associated with rapid treatment of sepsis.

ABSTRACT:

Purpose

To understand what clinical presenting features of sepsis patients are historically associated with rapid treatment involving antibiotics and fluids, as appropriate.

Design

This was a retrospective, observational cohort study using a machine-learning model with an embedded feature selection mechanism (gradient boosting machine).

Methods

For adult patients (age ≥ 18 years) who were admitted through Emergency Department (ED) meeting clinical criteria of severe sepsis from 11/2007 to 05/2018 at an urban tertiary academic medical center, we developed gradient boosting models (GBMs) using a total of 760 original and derived variables, including demographic variables, laboratory values, vital signs, infection diagnosis present on admission, and historical comorbidities. We identified the most impactful factors having strong association with rapid treatment, and further applied the Shapley Additive exPlanation (SHAP) values to examine the marginal effects for each factor.

Results

For the subgroups with or without fluid bolus treatment component, the models achieved high accuracy of area-under-receiver-operating-curve of 0.91 [95% CI, 0.86-0.95] and 0.84 [95% CI, 0.81-0.86], and sensitivity of 0.81[95% CI, 0.72-0.87] and 0.91 [95% CI, 0.81-0.97], respectively. We identified the 20 most impactful factors associated with rapid treatment for each subgroup. In the non-hypotensive subgroup, initial physiological values were the most impactful to the model, while in the fluid bolus subgroup, value minima and maxima tended to be the most impactful.

Conclusion

These machine learning methods identified factors associated with rapid treatment of severe sepsis patients from a large volume of high-dimensional clinical data. The results provide insight into differences in the rapid provision of treatment among patients with sepsis.

SUBMITTER: Song X

PROVIDER: S-EPMC8101717 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Cavitary pulmonary tuberculosis (TB) is associated with poor outcomes, treatment recurrence, higher transmission rates, and the development of drug resistance. However, reports on its clinical characteristics, associated factors, and treatment outcomes are lacking. Hence, this study sought to evaluate the clinical factors associated with cavitary pulmonary TB and its treatment outcomes. We retrospectively evaluated 410 patients with drug-susceptible pulmonary TB in a university hospital in Korea between 2014 and 2019. To evaluate the factors associated with cavitary TB, multivariable logistic regression was performed with adjustments for potential confounders. We also compared the treatment outcomes between patients with cavitary TB and those without cavitary TB. Of the 410 patients, 244 (59.5%) had non-cavitary TB and 166 (40.5%) had cavitary TB. Multivariable logistic analysis with forward selection method showed that body mass index (BMI) (adjusted OR = 0.88, 95% CI: 0.81-0.97), previous history of TB (adjusted OR = 3.45, 95% CI: 1.24-9.59), ex- or current smoker (adjusted OR = 1.77, 95% CI: 1.01-3.13), diabetes mellitus (adjusted OR = 2.72, 95% CI: 1.36-5.44), and positive results on the initial sputum acid-fast bacilli (AFB) smear (adjusted OR = 2.24, 95% CI: 1.26-3.98) were significantly associated with cavitary TB. Although treatment duration was significantly longer in patients with cavitary TB than in those with non-cavitary TB (248 (102-370 days) vs. 202 (98-336 days), p < 0.001), the recurrence rate after successful treatment was significantly higher in the patients with cavitary TB than in those with non-cavitary TB (0.4% vs. 3.0% p = 0.042). In conclusion, ex- or current smoker, lower BMI, previous history of TB, diabetes mellitus, and positivity of the initial AFB smear were associated with cavitary TB. The patients with cavitary TB had more AFB culture-positive results at 2 months, longer treatment duration, and higher recurrence rates than those with non-cavitary TB.

Project description:ImportanceSepsis induces profound metabolic derangements, while exogenous levocarnitine mitigates metabolic dysfunction by enhancing glucose and lactate oxidation and increasing fatty acid shuttling. Previous trials in sepsis suggest beneficial effects of levocarnitine on patient-centered outcomes.ObjectivesTo test the hypothesis that levocarnitine reduces cumulative organ failure in patients with septic shock at 48 hours and, if present, to estimate the probability that the most efficacious dose will decrease 28-day mortality in a pivotal phase 3 clinical trial.Design, setting, and participantsMulticenter adaptive, randomized, blinded, dose-finding, phase 2 clinical trial (Rapid Administration of Carnitine in Sepsis [RACE]). The setting was 16 urban US medical centers. Participants were patients aged 18 years or older admitted from March 5, 2013, to February 5, 2018, with septic shock and moderate organ dysfunction.InterventionsWithin 24 hours of identification, patients were assigned to 1 of the following 4 treatments: low (6 g), medium (12 g), or high (18 g) doses of levocarnitine or an equivalent volume of saline placebo administered as a 12-hour infusion.Main outcomes and measuresThe primary outcome required, first, a greater than 90% posterior probability that the most promising levocarnitine dose decreases the Sequential Organ Failure Assessment (SOFA) score at 48 hours and, second (given having met the first condition), at least a 30% predictive probability of success in reducing 28-day mortality in a subsequent traditional superiority trial to test efficacy.ResultsOf the 250 enrolled participants (mean [SD] age, 61.7 [14.8] years; 56.8% male), 35, 34, and 106 patients were adaptively randomized to the low, medium, and high levocarnitine doses, respectively, while 75 patients were randomized to placebo. In the intent-to-treat analysis, the fitted mean (SD) changes in the SOFA score for the low, medium, and high levocarnitine groups were -1.27 (0.49), -1.66 (0.38), and -1.97 (0.32), respectively, vs -1.63 (0.35) in the placebo group. The posterior probability that the 18-g dose is superior to placebo was 0.78, which did not meet the a priori threshold of 0.90. Mortality at 28 days was 45.9% (34 of 74) in the placebo group compared with 43.3% (45 of 104) for the most promising levocarnitine dose (18 g). Similar findings were noted in the per-protocol analysis.Conclusions and relevanceIn this dose-finding, phase 2 adaptive randomized trial, the most efficacious dose of levocarnitine (18 g) did not meaningfully reduce cumulative organ failure at 48 hours.Trial registrationClinicalTrials.gov Identifier: NCT01665092.