NEW YORK (Reuters Health) – A large review of electronic health records by Boston-based researchers has revealed 33 phenotypes that are indicative of long COVID-19 among people who contracted the virus but did not need hospital care.
Many of the phenotypes – such as alopecia, anosmia, fatigue, shortness of breath and chest pain – have been well documented as signs and symptoms of post-acute sequelae of COVID-19 (PASC). Others, however, are fairly new and include new-onset diabetes and neurological diagnoses, the researchers report.
“What’s really important is that these are all non-hospitalized patients. They are all patients that were treated at home, and they still have some ongoing symptoms months later,” Dr. Hossein Estiri of Harvard Medical School and Massachusetts General Hospital told Reuters Health by phone.
“There does seem to be some association between COVID-19 and an increased diagnostic rate of some pretty severe diseases” like dementia, chronic kidney disease and diabetes, he said.
In addition, more than 63% of PASC phenotypes were in COVID-19 patients under 65, providing another reason for younger people to be vaccinated against COVID-19, the researchers say in a paper in BMC Medicine.
The study team applied a computational framework for knowledge discovery from EHR data to identify phenotypes associated with a past positive PCR test for COVID-19.
They evaluated newly recorded phenotypes in two temporal windows – three to six months and six to nine months – after the positive or negative COVID-19 test in more than 96,000 non-hospitalized patients.
“We were able to evaluate over 1,600 phenotypes and identify a small number of phenotypes (with confidence scores) that associate with a past COVID-19 infection,” they report.
Among the 33 phenotypes identified, new-onset anosmia and dysgeusia, alopecia, chest pain, chronic fatigue syndrome and shortness of breath are some of the most significant indicators of a past COVID-19 infection and are some of the earliest associations with PASC seen with “high confidence” in the three-to-six-month window after infection.
Alopecia and non-specific chest pain were not found with high confidence in the six-to-nine-month window after infection, but anosmia and chronic fatigue syndrome continued to be important phenotypes in both time periods, the researchers found.
Additionally, several phenotypes were identified with similarly high confidence including type-2 diabetes, pneumonia, proteinuria and syncope and collapse.
The researchers say several neurological phenotypes (vascular dementia, dementia and neurological disorders) were often diagnosed after COVID and appear to have an increased association with the infection.
The neurological-disorder phenotype includes several ICD codes, and in a random sampling of patients with this phenotype, most had the R41.89 ICD code for “other symptoms and signs involving cognitive function and awareness.”
Collectively, these phenotypes suggest ongoing cognitive dysfunction, the researchers note. In some cases, cognitive problems may be so severe they lead to an initial formal diagnosis of dementia at higher rates among those with a history of COVID.
“While many of these patients may have already shown some signs of memory loss, the formal diagnosis of dementia did not come until after infection, suggesting that the viral illness may have contributed to a worsening of their condition and the formal declaration of this diagnosis,” they note.
Type-2 diabetes was another “important” phenotype identified, supporting some prior research.
“Several studies have pointed out possible pathophysiological relationships between COVID-19 and diabetes. And the increased incidence of a number of metabolic diseases has been found with those after a COVID-19 diagnosis. Our study indicates that the metabolic disorder may be so significant as to lead to a formal diagnosis of diabetes mellitus,” they note.
The disease-of-the-nail phenotype, which includes leukonychia, onycholysis, onychomadesis, Mees’ lines, Muehrcke’s lines and Beau’s lines, have been associated with infections and renal or liver dysfunction previously. “Our results suggest this association is widespread and likely a result of systemic infection including renal injury,” the researchers say.
Their results also link proteinuria to COVID-19 infection among male patients younger than age 65. COVID-19 has previously been associated with acute kidney injury and proteinuria is a known surrogate for kidney disease. “The identification of proteinuria as an association with COVID-19 in the young patient cohort suggests the insult of COVID-19 to the kidneys persists months after the infection has resolved,” the team writes.
“Our understanding of COVID-19 and its chronic sequelae is evolving, and new risks are unknown. We do not know who might develop the post-COVID syndrome, how long the symptoms last, and whether COVID-19 prompts the presentation of chronic diseases,” they note.
“Accurate identification of phenotypes will be important to guide future research and the healthcare system to focus its efforts and resources on adequately controlled age- and gender-specific sequelae of a COVID-19 infection.”
This work was supported by grants from the National Human Genome Research Institute and the National Library of Medicine. The authors have declared no conflicts of interest.
SOURCE: https://bit.ly/3BiZELa BMC Medicine, online September 27, 2021.
Source: Read Full Article