BASELINE METABOLIC PROFILING AND RISK OF DEATH FROM COVID-19

  1. Glen H. Murata
  2. Allison E. Murata
  3. Heather M. Campbell
  4. Brent Wagner
  5. Benjamin H. Mcmahon
  6. Jenny T. Mao

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Objective

To derive a predicted probability of death (PDeathLabs) based upon complete value sets for 11 clinical measurements (CM) obtained on patients prior to their diagnosis of coronavirus disease (COVID-19). PDeathLabs is intended for use as a summary metric for baseline metabolic status in multivariate models for COVID-19 death.

Methods

Cases were identified through the COVID-19 Shared Data Resource (CSDR) of the Department of Veterans Affairs. The diagnosis required at least one positive nucleic acid amplification test (NAAT). The primary outcome was death within 60 days of the first positive test. We retrieved all values for systolic blood pressure (SBP), diastolic blood pressure (DBP), oxygen saturation (O2SAT), body mass index (BMI), estimated glomerular filtration rate (EGFR), alanine aminotransferase (ALT), serum albumin (ALB), hematocrit (HCT), LDL cholesterol (LDL) hemoglobin A1c (A1C), and HDL cholesterol (HDL) if they were done at least 14 days prior to the NAAT. Clinicians evaluate several attributes of CM that are of critical importance: metabolic control, disease burden, chronicity, refractoriness, tendency to relapse, temporal trends, and lability. We derived 1-3 parameters for each of these attributes: the most recent value (metabolic control); time-weighted average and abnormal area under a severity versus time curve (disease burden); time and number of readings above or below goal (chronicity); longest abnormal cluster and time/number of consecutive readings above goal if the last value was abnormal (refractoriness); number of abnormal clusters (tendency to relapse); long- and short-term changes (temporal trends); and coefficient of variation and mean deviation between consecutive readings (lability). We created computer programs to derive cumulative values for these 13 parameters for all 11 CM as each new value is added. A fitted logistic model was developed for each CM to determine which of the 13 parameters contributed to the risk of death. A main logistic model was developed to determine which of the 13 × 11 = 143 metabolic parameters were independently predictive of death. The resulting model was used to derive PDeathLabs for each patient and the area under its receiver operating characteristic (ROC) curve calculated. Single variable logistic models were also derived for age at diagnosis, the Charlson 2-year (Charl2Yr) and lifetime (CharlEver) scores, and the Elixhauser 2-year (Elix2Yrs) and lifetime (ElixEver) scores. Stata was used to compare the ROCs for PDeathDx and each of the other metrics.

Results

On September 30, 2021, there were 347,220 COVID-19 patients in the CSDR. 329,491 (94.9%) patients had CM performed at least 14 days prior to the COVID-19 diagnosis and form the basis for this report. 17,934 (5.44%) died within 60 days of the diagnosis. On the subset regressions, the number of significant parameters ranged from all 13 for SBP to 7 for HDL. 239,393 patients had complete sets of data for developing the main model. Of 143 candidate predictors, 49 parameters were identified as statistically significant, independent predictors of death. The most influential domains were the most recent value, disease burden, temporal trends, and tendency to relapse. The ROC area for PDeathLabs was 0.785 +/- 0.002. No difference was found in the ROC areas of PDeathLabs and age at diagnosis (0.783 +/- 0.002; P = NS). However, the ROC area for PDeathLabs was significantly greater than that of Charl2Yrs (0.704 +/- 0.002; P < 0.001), CharlEver (0.729 +/- 0.002; P < 0.001), Elix2Yrs (0.675 ± 0.002; P < 0.001), and ElixEver (0.707 +/- 0.002; P < 0.001). A poor prognosis was found for chronic systolic hypertension. On the other hand, a higher BMI was protective once SBP, DBP, HDL, LDL and A1C were considered.

Conclusions

Our study confirms that parameters derived for 11 CM are significant determinants of COVID-19 death. The most recent value should not be selected over other parameters for multivariate modeling unless there is a physiologic basis for doing so. PDeathLabs has the same discriminating power as age at diagnosis and outperforms comorbidity indices as a summary metric for pre-existing conditions. If validated by others, this approach provides a robust approach to handling CM in multivariate models.

Article activity feed

  1. Version published to 10.1101/2022.01.22.22269691v2 on medRxiv
  2. ScreenIT

    SciScore for 10.1101/2022.01.22.22269691: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    Ethicsnot detected.
    Sex as a biological variablenot detected.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    The following parameters were derived for each of the 11 CMs (above): Statistical methods: In summary, we analyzed the following 13 parameters for each CM: Value1, FUDaysAboveGoal, NumAboveGoal, AbnAUC, TimeWtAvg, NumClust, TimeAboveGoal, CtAboveGoal, MaxClustDays, CoeffVar, MeanValDiff, NetChange, and Lag3Dev.
    NumClust
    suggested: None

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: An explicit section about the limitations of the techniques employed in this study was not found. We encourage authors to address study limitations.

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    Results from scite Reference Check: We found no unreliable references.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  3. Version published to 10.1101/2022.01.22.22269691v1 on medRxiv