Evolution of a Parsimonious Prognostic Index in COVID-19

Abstract

In the December 2021 issue of the Journal of Cardiac Critical Care, we proposed a promising prognostic index, shock index (SI) for risk-stratifying the patients ailing from coronavirus disease 2019 (COVID-19).1 While it is indeed gratifying to witness an increasing research interest in the subject,2–5 the elucidation of nuances of the accruing literature would be of potential interest to the readers of the Journal. Reiterating the basics of the classically described prognostic index, SI is computed as the ratio of heart rate (HR) to systolic blood pressure (SBP).6 Alongside being conducive to the prediction of hemodynamic instability in patients with myocardial infarction and sepsis, the former has also been propounded as a prognostic marker for the in-hospital mortality prediction.7 Notably, a normal SI (0.5–0.7) along with the absence of an elevated lactate level has been associated with a very low risk for severe sepsis at presentation.8 Talking specifically of SI-based prognostication in COVID-19, there have been four studies that analyzed the merit of SI among patients with COVID-19 over the past 2 years of the pandemic. The first study by Doğanay et al provided results in favor of the mortality predictive value of SI.2 In a retrospective evaluation of 489 COVID-19 patients using CHAID (Chi-Square Automatic Interaction Detection Technique" is a statistical tool used for comparing relationship between variables) analysis, they highlighted SI as the factor with the highest predictive value within any given age group with a cutoff of 0.93. They also noted an increase in the association of SI and mortality with advanced age (mortality in patients with SI>0.93 and age<56 years old1⁄426.9%, 56–77 years1⁄480.5%,>77 years1⁄491.4%). These results were corroborated by Kurt et al in their efforts to investigate the mortality predictive potential of SI and modified SI (HR/mean arterial pressure).3 Using the data of 464 patients evaluated retrospectively, they noted a higher predictive power of modified SI compared with SI (DeLong’s test, AUC difference1⁄4–0.020, p1⁄40.003). Similarly, a research endeavor by Ak et al compared various threshold values of SI (0.7, 0.8, 0.9, and 1.0) in 364 COVID-19 cases to assess the predictive performance in terms of intensive care unit (ICU) admission and mortality rate.4 A pre-hospital SI>0.9 was found to be predictive of a higher ICU admission and mortality rate. While these studies show a beneficial aspect of employing SI in risk stratification of COVID-19 patients, van Rensen et al could not elicit a similar result in their study. Their retrospective analysis of 411 patients revealed no discernible discriminative potential of SI for ICU admission and mortality in COVID-19 patients.5 Being a purely hemodynamic parameter, SI emerges as a useful metric for physiological status monitoring.9 At the same time, the inclusion of other vital-system monitoring parameters in SI can buttress the subsequent prognostic potential. The former becomes all the more pertinent in the context of a multifaceted syndrome, as seen in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Among the various modifications made to the classical SI, some appear to be more suitable than others in prognosticating poor outcomes in these cases. Inclusion of respiratory rate in respiratory adjusted shock index (RASI), age in ageadjusted shock index or Horowitz index (partial pressure of arterial oxygen/fractional inspired concentration of oxygen,