Preprint: Prediction of Long COVID Based on Severity of Initial COVID-19 Infection: Differences in predictive feature sets... 2023 Socia et al

Full title: Prediction of Long COVID Based on Severity of Initial COVID-19 Infection: Differences in predictive feature sets between hospitalized versus non-hospitalized index infections

Abstract

Long COVID is recognized as a significant consequence of SARS-COV2 infection. While the pathogenesis of Long COVID is still a subject of extensive investigation, there is considerable potential benefit in being able to predict which patients will develop Long COVID. We hypothesize that there would be distinct differences in the prediction of Long COVID based on the severity of the index infection, and use whether the index infection required hospitalization or not as a proxy for developing predictive models.

We divide a large population of COVID patients drawn from the United States National Institutes of Health (NIH) National COVID Cohort Collaborative (N3C) Data Enclave Repository into two cohorts based on the severity of their initial COVID-19 illness and correspondingly trained two machine learning models: the Long COVID after Severe Disease Model (LCaSDM) and the Long COVID after Mild Disease Model (LCaMDM). The resulting models performed well on internal validation/testing, with a F1 score of 0.94 for the LCaSDM and 0.82 for the LCaMDM.

There were distinct differences in the top 10 features used by each model, possibly reflecting the differences in type and amount of pathophysiological data between the hospitalized and non-hospitalized patients and/or reflecting different pathophysiological trajectories in the development of Long COVID. Of particular interest was the importance of Plant Hardiness Zone in the feature set for the LCaMDM, which may point to a role of climate and/or sunlight in the progression to Long COVID.

Future work will involve a more detailed investigation of the potential role of climate and sunlight, as well as refinement of the predictive models as Long COVID becomes increasingly parsed into distinct clinical phenotypes.

https://www.medrxiv.org/content/10.1101/2023.01.16.23284634v2

 

As noted above, the pathophysiology of long COVID is still very unclear, and the empirical basis of the diagnosis of long COVID is in a state of evolution. Given that some authors [5] have identified instances of long COVID as a functional disorder without a clear biological signature, we sought to investigate whether there is a relationship between the presence of pre-existing functional disorder(s) and the risk of developing long COVID. We utilized the MIMIC-IV database [6] to identify the most common comorbidities for patients presenting with a set of six functional disorders (fibromyalgia, chronic pain, chronic fatigue, mast cell activation disorder, familial dysautonomia, and irritable bowel syndrome). We then classified the comorbidities into one of three categories: metabolic, psychiatric, or functional. For each patient, the number of prior diagnoses in each category was tabulated; thus, diagnostic history was compressed to three potential variables.​

No idea why they separate "functional" and psychiatric when they mean the same thing, in the sense that they don't really mean anything else than "haven't figured out".

Trying to do Big data research with data that are both invalid and incomplete yields nothing useful. The reference to some researchers having "identified" functional disorders is the awful debunked study by Ranque. I guess they're still in denial about dysautonomia, that was basically the easiest of them all and they still screw it up completely.

 

There reference to MIMIC-IV leads to https://physionet.org/content/mimiciv/1.0/ (recently superseded by https://physionet.org/content/mimiciv/2.2/). These refer to Medical Information Mart for Intensive Care (MIT's Electronic Health Record machine learning). Garbage-In-Garbage-Out.


For some more useful descriptors —

(For "fibromyalgia", "chronic pain", "chronic fatigue" etc see all of https://www.s4me.info)

For Mast cell activation disorder —

• AAAAI Mast Cell Disorders Committee Work Group Report: Mast cell activation syndrome (MCAS) diagnosis and management (2019).

Familial dysautonomia —

• https://omim.org/entry/223900
• https://en.wikipedia.org/wiki/Familial_dysautonomia
• https://rarediseases.info.nih.gov/diseases/7581/familial-dysautonomia

 

Familial dysautonomia is considered "functional"?!

 

The paper's results are tantalizing is scant on details. For example, there's a correlation with climate but which direction? I laughed at their use of Plant Hardiness Zone--do they think we're flowers?--but it's interesting science.

 

A UK developed coding using just four categories of hardiness was standard in horticulture for a long time, but US research has largely replaced this, although the allocation of individual species to a particular zone is often misleading, however the zones themselves as defined in the US are based on a very substantial body of work: Development of a New USDA Plant Hardiness Zone Map for the United States and yields a very pretty map: https://planthardiness.ars.usda.gov/

For COVID 19 which has a major respiratory component, climate variation might have especial significance, though population density will be confounder in any map based analysis.

 

More incidence in colder climate. Which mirrors many autoimmune diseases.

It would be great if someone could actually do a comprehensive study on this, but the main factor probably has to do with the same reason why respiratory illnesses are more common in the winter in cold climates: people spend more time indoors with other people. Which would imply a linear relation with number of infections, a very strong hypothesis given the, you know, germ theory of disease. Well, germs and nasty stuff, the immune system has to try and clean up either way.

I'm not sure where medicine still is on the issue of autoimmune diseases, but just like babies there's no immaculate conception here, if something starts a chain it still has to be there and start it, sticking to "auto" as equivalent to "happens spontaneously" was a very bad assumption. Everyone knows assumptions are bad, yet somehow they rule almost everything in medicine. Explains a lot, frankly.

 

Most well-known being the latitudinal relationship in MS.

Rather than eg vitamin D, this effect could relate to near infra-red promotion of subcellular melatonin to control ROS in the mitochondria. Brings us back to the discussion on convalescent hospitals and solariums.

See this MedCram video - link to 10 minute section on mitochondria and melatonin and cytochrome C oxidase.

Perhaps what this might indicate is that having damaged or suboptimal mitochondria while you're a teen and first getting infected with EBV/HHV6 etc, leads to reduced (mitochondrial) immune control of these latent viruses at that time. This could set up for MS or other diseases later in life.

---
Review article Melatonin: Both a Messenger of Darkness and a Participant in the Cellular Actions of Non-Visible Solar Radiation of Near Infrared Light (Biology, 2023)