SAMI Score Study (Symptoms, Admission, MICU, and Intubation) Associates Mortality with Different Phenotypes of COVID-19

Abstract

Rationale: Diverse presentations of SARS-CoV-2 infection exist, with some studies differentiating as many as five phenotypes. Each of the phenotypes describes varying symptoms, inflammatory markers, and lung physiology - many of which require testing to diagnose. The two most common phenotypes, L-type and H-type, advance along a spectrum indicating an evolving illness. Recently, this physiology has been explained as a shift from an acute viral illness to progressive inflammatory response. The objective of this study is to identify differences in the SAMI Score between death and non-death cohorts, from data collected in a large, multi-center healthcare system for adult (age >18) patients. Methods: All patients enrolled in this retrospective study were identified by a positive nasal or oropharyngeal swab for SARS-CoV-2 PCR in the ED between March and September 2020. Symptomatic data was collected based on ED admission histories and prior medical records. The SAMI score was calculated based on duration of days from symptom onset to hospital admission (SA score), time to ICU admission (AM score), and time to intubation (MI score). Patient cohorts were separated by mortality. Results: Out of 510 patients, 227 patients met inclusion criteria for enrollment. In the death cohort (43 patients, mean age 75, 53% female), the mean SA score and AM score were 5.6d and 1.9d, respectively. The average MI score was 2.3d and the SAMI score was 11.6d among those intubated. In the non-death cohort (184 patients, mean age 60, 49% female), the mean SA score was 6.3d while the AM score was 1.3d. The average MI score and SAMI score were 0.4d and 6.0d, respectively. Overall, the mean MI score between the two groups (2.3d death group, 0.4d non-death group) was statistically significant, p = 0.045. Similarly, the mean SAMI score between the two groups (11.6d vs. 6.0d respectively) was also statistically significant, p = 0.026. There was no significant difference between the average SA score (p = 0.63) or AM score (p = 0.37). Conclusion: Patients who required rapid intubation after symptom onset (short SAMI score) are associated with less mortality than those requiring a prolonged time to intubation (long SAMI score). Physiologically, rapid intubation and a short MI score suggest a recoverable acute disease state. Conversely, prolonged time to intubation may be indicative of a progressive irreversible process. This research allows for further sub-group analysis to determine if inflammatory markers are higher in the group with longer SAMI scores.