Dataset Information

Assessment of Machine Learning to Estimate the Individual Treatment Effect of Corticosteroids in Septic Shock.

ABSTRACT:

Importance

The survival benefit of corticosteroids in septic shock remains uncertain.

Objective

To estimate the individual treatment effect (ITE) of corticosteroids in adults with septic shock in intensive care units using machine learning and to evaluate the net benefit of corticosteroids when the decision to treat is based on the individual estimated absolute treatment effect.

Design, setting, and participants

This cohort study used individual patient data from 4 trials on steroid supplementation in adults with septic shock as a training cohort to model the ITE using an ensemble machine learning approach. Data from a double-blinded, placebo-controlled randomized clinical trial comparing hydrocortisone with placebo were used for external validation. Data analysis was conducted from September 2019 to February 2020.

Exposures

Intravenous hydrocortisone 50 mg dose every 6 hours for 5 to 7 days with or without enteral 50 μg of fludrocortisone daily for 7 days. The control was either the placebo or usual care.

Main outcomes and measures

All-cause 90-day mortality.

Results

A total of 2548 participants were included in the development cohort, with median (interquartile range [IQR]) age of 66 (55-76) years and 1656 (65.0%) men. The median (IQR) Simplified Acute Physiology Score (SAPS II) was 55 [42-69], and median (IQR) Sepsis-related Organ Failure Assessment score on day 1 was 11 (9-13). The crude pooled relative risk (RR) of death at 90 days was 0.89 (95% CI, 0.83 to 0.96) in favor of corticosteroids. According to the optimal individual model, the estimated median absolute risk reduction was of 2.90% (95% CI, 2.79% to 3.01%). In the external validation cohort of 75 patients, the area under the curve of the optimal individual model was 0.77 (95% CI, 0.59 to 0.92). For any number willing to treat (NWT; defined as the acceptable number of people to treat to avoid 1 additional outcome considering the risk of harm associated with the treatment) less than 25, the net benefit of treating all patients vs treating nobody was negative. When the NWT was 25, the net benefit was 0.01 for the treat all with hydrocortisone strategy, -0.01 for treat all with hydrocortisone and fludrocortisone strategy, 0.06 for the treat by SAPS II strategy, and 0.31 for the treat by optimal individual model strategy. The net benefit of the SAPS II and the optimal individual model treatment strategies converged to zero for a smaller number willing to treat, but the individual model was consistently superior than model based on the SAPS II score.

Conclusions and relevance

These findings suggest that an individualized treatment strategy to decide which patient with septic shock to treat with corticosteroids yielded positive net benefit regardless of potential corticosteroid-associated side effects.

SUBMITTER: Pirracchio R

PROVIDER: S-EPMC7729430 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ImportanceThe combination of ascorbic acid, corticosteroids, and thiamine has been identified as a potential therapy for septic shock.ObjectiveTo determine whether the combination of ascorbic acid, corticosteroids, and thiamine attenuates organ injury in patients with septic shock.Design, setting, and participantsRandomized, blinded, multicenter clinical trial of ascorbic acid, corticosteroids, and thiamine vs placebo for adult patients with septic shock. Two hundred five patients were enrolled between February 9, 2018, and October 27, 2019, at 14 centers in the United States. Follow-up continued until November 26, 2019.InterventionsPatients were randomly assigned to receive parenteral ascorbic acid (1500 mg), hydrocortisone (50 mg), and thiamine (100 mg) every 6 hours for 4 days (n = 103) or placebo in matching volumes at the same time points (n = 102).Main outcomes and measuresThe primary outcome was change in the Sequential Organ Failure Assessment (SOFA) score (range, 0-24; 0 = best) between enrollment and 72 hours. Key secondary outcomes included kidney failure and 30-day mortality. Patients who received at least 1 dose of study drug were included in analyses.ResultsAmong 205 randomized patients (mean age, 68 [SD, 15] years; 90 [44%] women), 200 (98%) received at least 1 dose of study drug, completed the trial, and were included in the analyses (101 with intervention and 99 with placebo group). Overall, there was no statistically significant interaction between time and treatment group with regard to SOFA score over the 72 hours after enrollment (mean SOFA score change from 9.1 to 4.4 [-4.7] points with intervention vs 9.2 to 5.1 [-4.1] points with placebo; adjusted mean difference, -0.8; 95% CI, -1.7 to 0.2; P = .12 for interaction). There was no statistically significant difference in the incidence of kidney failure (31.7% with intervention vs 27.3% with placebo; adjusted risk difference, 0.03; 95% CI, -0.1 to 0.2; P = .58) or in 30-day mortality (34.7% vs 29.3%, respectively; hazard ratio, 1.3; 95% CI, 0.8-2.2; P = .26). The most common serious adverse events were hyperglycemia (12 patients with intervention and 7 patients with placebo), hypernatremia (11 and 7 patients, respectively), and new hospital-acquired infection (13 and 12 patients, respectively).Conclusions and relevanceIn patients with septic shock, the combination of ascorbic acid, corticosteroids, and thiamine, compared with placebo, did not result in a statistically significant reduction in SOFA score during the first 72 hours after enrollment. These data do not support routine use of this combination therapy for patients with septic shock.Trial registrationClinicalTrials.gov Identifier: NCT03389555.

Project description:ObjectivesWe previously reported gene expression-based endotypes of pediatric septic shock, endotypes A and B, and that corticosteroid exposure was independently associated with increased mortality among pediatric endotype A patients. The Vasopressin vs Norepinephrine as Initial Therapy in Septic Shock trial tested the efficacy of vasopressin as initial vasopressor therapy for septic shock among adult patients, when compared with norepinephrine. Patients who reached a prespecified dose of either vasopressor were further randomized to receive hydrocortisone or placebo. A proportion of patients in the Vasopressin vs Norepinephrine as Initial Therapy in Septic Shock trial had transcriptomic data generated at baseline using whole blood-derived messenger RNA. We used the publicly available transcriptomic data from the Vasopressin vs Norepinephrine as Initial Therapy in Septic Shock trial to assign the study subjects to pediatric septic shock endotype A or B, and tested the hypothesis that hydrocortisone treatment is associated with increased mortality among patients in endotype A.DesignSecondary analysis of publicly available transcriptomic data.SettingMultiple adult ICUs.PatientsAdults with septic shock randomized to hydrocortisone (n = 47) or placebo (n = 50).InterventionsRandomization to the Vasopressin vs Norepinephrine as Initial Therapy in Septic Shock trial experimental arms.Measurements and main resultsEndotype A patients receiving hydrocortisone had a mortality rate of 46%, whereas endotype A patients receiving placebo had a mortality rate of 22% (p = 0.105). In contrast, the mortality rates for endotype B patients receiving hydrocortisone or placebo were 19% and 22%, respectively. The odds of death were more than three times greater in endotype A patients receiving hydrocortisone than endotype A patients receiving placebo (p = 0.05).ConclusionsThis exploratory analysis provides further evidence that corticosteroid exposure may be associated with increased mortality among septic shock endotype A patients.