Tag: long covid prognosis

Data-driven prognosis of long COVID in patients using machine learning

Abstract:

Long-COVID is a health condition in which individuals experience persisting, returning or new symptoms longer than 4 weeks after they have recovered from COVID-19 and this condition can even last for months. It can cause multi-organ failure and in some cases, it can even lead to death. The effects and symptoms of Long COVID can vary from person to person. Even though it’s rising globally, there is a limited understanding about its prediction, risk factors and whether its prognosis can be predicted in the initial first week of acute COVID-19. Artificial Intelligence (AI) and Machine Learning (ML) have aided the medical industry in a variety of ways including the diagnosis, prediction, and prognosis of many diseases.

This paper introduces a novel method to determine Long COVID in the early or first week of acute COVID-19 by considering the basic demographics, and symptoms during COVID-19, along with the clinical lab results of the patients hospitalized. In comparison with different ML models such as Logistic Regression, Support Vector Machine (SVM), XGBoost and Artificial Neural Network (ANN) to predict and classify the patients as Long COVID or Short COVID during the first week of COVID-19, ANN has outperformed the other models with an accuracy of 81% when considering the symptoms of COVID-19 and a 79% for the clinical test data. The predictive factors and the significant clinical tests for the Long COVID are also determined by using different methods like Chi-square Test and Pearson Correlation.

Source: S. S. Parvathy, Nagesh Subbanna, Sethuraman Rao, Rahul Krishnan Pathinarupothi, T. S. Dipu, Merlin Moni, Chithira V. Nair; Data-driven prognosis of long COVID in patients using machine learning. AIP Conf. Proc. 15 December 2023; 2901 (1): 060014. https://doi.org/10.1063/5.0178561 https://pubs.aip.org/aip/acp/article/2901/1/060014/2930006 (Full text available as PDF file)

Determinants of the onset and prognosis of the post-COVID-19 condition: a 2-year prospective observational cohort study

Abstract:

Background: At least 5-10% of subjects surviving COVID-19 develop the post-COVID-19 condition (PCC) or “Long COVID”. The clinical presentation of PCC is heterogeneous, its pathogenesis is being deciphered, and objective, validated biomarkers are lacking. It is unknown if PCC is a single entity or a heterogeneous syndrome with overlapping pathophysiological basis. The large US RECOVER study identified four clusters of subjects with PCC according to their presenting symptoms. However, the long-term clinical implications of PCC remain unknown.

Methods: We conducted a 2-year prospective cohort study of subjects surviving COVID-19, including individuals fulfilling the WHO PCC definition and subjects with full clinical recovery. We systematically collected post-COVID-19 symptoms using prespecified questionnaires and performed additional diagnostic imaging tests when needed. Factors associated with PCC were identified and modelled using logistic regression. Unsupervised clustering analysis was used to group subjects with PCC according to their presenting symptoms. Factors associated with PCC recovery were modelled using a direct acyclic graph approach.

Findings: The study included 548 individuals, 341 with PCC, followed for a median of 23 months (IQR 16.5-23.5), and 207 subjects fully recovered. In the model with the best fit, subjects who were male and had tertiary studies were less likely to develop PCC, whereas a history of headache, or presence of tachycardia, fatigue, neurocognitive and neurosensitive complaints and dyspnea at COVID-19 diagnosis predicted the development of PCC. The cluster analysis revealed the presence of three symptom clusters with an additive number of symptoms. Only 26 subjects (7.6%) recovered from PCC during follow-up; almost all of them (n = 24) belonged to the less symptomatic cluster A, dominated mainly by fatigue. Recovery from PCC was more likely in subjects who were male, required ICU admission, or had cardiovascular comorbidities, hyporexia and/or smell/taste alterations during acute COVID-19. Subjects presenting with muscle pain, impaired attention, dyspnea, or tachycardia, conversely, were less likely to recover from PCC.

Interpretation: Preexisting medical and socioeconomic factors, as well as acute COVID-19 symptoms, are associated with the development of and recovery from the PCC. Recovery is extremely rare during the first 2 years, posing a major challenge to healthcare systems.

Source: Mateu L, Tebe C, Loste C, Santos JR, Lladós G, López C, España-Cueto S, Toledo R, Font M, Chamorro A, Muñoz-López F, Nevot M, Vallejo N, Teis A, Puig J, Fumaz CR, Muñoz-Moreno JA, Prats A, Estany-Quera C, Coll-Fernández R, Herrero C, Casares P, Garcia A, Clotet B, Paredes R, Massanella M. Determinants of the onset and prognosis of the post-COVID-19 condition: a 2-year prospective observational cohort study. Lancet Reg Health Eur. 2023 Sep 5;33:100724. doi: 10.1016/j.lanepe.2023.100724. PMID: 37954002; PMCID: PMC10636281. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10636281/ (Full text)

Sequential multi-omics analysis identifies clinical phenotypes and predictive biomarkers for long COVID

Abstract:

The post-acute sequelae of COVID-19 (PASC), also known as long COVID, is often associated with debilitating symptoms and adverse multisystem consequences. We obtain plasma samples from 117 individuals during and 6 months following their acute phase of infection to comprehensively profile and assess changes in cytokines, proteome, and metabolome.

Network analysis reveals sustained inflammatory response, platelet degranulation, and cellular activation during convalescence accompanied by dysregulation in arginine biosynthesis, methionine metabolism, taurine metabolism, and tricarboxylic acid (TCA) cycle processes.

Furthermore, we develop a prognostic model composed of 20 molecules involved in regulating T cell exhaustion and energy metabolism that can reliably predict adverse clinical outcomes following discharge from acute infection with 83% accuracy and an area under the curve (AUC) of 0.96.

Our study reveals pertinent biological processes during convalescence that differ from acute infection, and it supports the development of specific therapies and biomarkers for patients suffering from long COVID.

Source: Wang K, Khoramjoo M, Srinivasan K, Gordon PMK, Mandal R, Jackson D, Sligl W, Grant MB, Penninger JM, Borchers CH, Wishart DS, Prasad V, Oudit GY. Sequential multi-omics analysis identifies clinical phenotypes and predictive biomarkers for long COVID. Cell Rep Med. 2023 Oct 18:101254. doi: 10.1016/j.xcrm.2023.101254. Epub ahead of print. PMID: 37890487. https://www.cell.com/cell-reports-medicine/fulltext/S2666-3791(23)00431-7 (Full text)

Long-term Prognosis at 1.5 years after Infection with Wild-type strain of SARS-CoV-2 and Alpha, Delta, as well as Omicron Variants

Highlights:

  • Trajectory of long COVID in SARS-CoV-2 wild-type, Alpha, Delta, and Omicron
  • Similar patterns of symptoms and severity of long COVID across all four variants
  • No clinically significant decline in median severity up to 1.5 years after infection
  • More than 50% of long COVID patients failed to improve using any outcome measure
  • Patients infected with Omicron may experience severe non-improving long COVID

Abstract:

Objectives: Knowledge is limited on how changing SARS-CoV-2 variants may translate into different characteristics and affect prognosis of patients with long COVID, especially following Omicron variants. We compared long-term prognosis of patients in a Danish Post COVID Clinic infected with wild-type strain, Alpha, Delta, or Omicron variants as well as the pre-Omicron compared to the Omicron period.

Methods: At enrollment a Post COVID symptom Questionnaire (PCQ), and standard health scores, were registered, and repeated four times until 1.5 years after infection. PCQ was the primary outcome to assess severity of long COVID, and delta PCQ to assess failure to improve.

Results: A total of 806 patients were enrolled. Patients infected with Omicron and Delta variants presented with more severe long COVID (median PCQ 43 in Delta vs 38 in wild-type, P=0.003) and health scores (EQ5D-index was 0.70 in Omicron vs 0.76 in wild-type, P=0.009 and 0.78 pre-Omicron, P=0.006). At 1.5 year after infection patients had no clinically meaningful decline in severity of long COVID, and 57% (245/429) of patients failed to improve 1.5 years after infection, with no differences between variants.

Conclusions: More than half of patients referred to a Post COVID Clinic failed to improve in long COVID severity 1.5 years after infection regardless of variants of SARS-CoV-2.

Source: Jane Agergaard , Jesper Damsgaard Gunst , Berit Schiøttz-Christensen , Lars Østergaard , Christian Wejse , Long-term Prognosis at 1.5 years after Infection with Wild-type strain of SARS-CoV-2 and Alpha, Delta, as well as Omicron Variants, International Journal of Infectious Diseases (2023), doi: https://doi.org/10.1016/j.ijid.2023.10.022 https://www.ijidonline.com/article/S1201-9712(23)00760-9/fulltext (Full text)

A National Evaluation of Outcomes in Long COVID Services using Digital PROM Data from the ELAROS Platform

Summary:

Key findings of this service evaluation study:
• Patient characteristics: A sample of 5,318 patients from 14 participating NHS LC sites were analysed. The sample had a female:male ratio of 2.1:1. The average age was 48.4 yrs, with 87% (of those whose ethnicity was recorded) of white ethnicity and 9% of Black or Asian ethnicity.
• Comorbidities: This sample of patients had a low prevalence of co-morbidities (7%) with a clear onset of new LC symptoms after their COVID-19 infection supporting the onset of a new condition in this cohort of previously healthy individuals.
• Duration of LC: The average duration of LC in this sample was 384 days (>12 months) at first assessment in an LC site, with symptoms still ongoing at presentation, with more than 90% of the sample being non-hospitalised patients.
• Digital platform: A total of 17,471 PROMs (C19-YRS and EQ-5D-5L) were completed by this sample of patients with at least 1,532 participants completing multiple assessments on the same PROM on the digital PROM platform. The completion of PROMs around the 3-month mark was low for both measures (11.7% for C19-YRS and 14.6% for EQ-5D-5L). The ones who completed PROMs both around the 3-month mark and the 6-month mark were 4.3% for C19-YRS and 5% for EQ-5D-5L. This limits the generalisability of the findings in this evaluation to all the LC population, but the findings remain valid for this cohort of individuals.
• New-onset disability: 3,395 patients who completed at least one C19-YRS questionnaire at first assessment showed significant new-onset symptom burden, functional disability, and deterioration of overall health since the COVID-19 infection.
• Comparison between LC and other chronic conditions: The cross-sectional EQ-5D-5L Index value of 3,438 patients suggests the burden and disability in LC are worse than that reported in the literature for Diabetes Mellitus, COPD, Heart Failure, and Multiple Sclerosis.
• 3-month follow-up: Among those who completed an initial C19-YRS assessment and another at 3 months, there was a statistically significant improvement in symptom burden, functional disability and overall health. Patients at 3 months however still had significant LC symptom
burden and disability compared to their pre-COVID-19 health status, i.e., their condition had improved, but they were far off from a complete recovery. Among those who completed EQ-5D- National Evaluation of Long COVID Service Outcomes using ELAROS Data (09 Oct 23) Page 3 of 31
5L, at first assessment and at 3 months, their EQ-5D-5L Index score did not show any statistically significant improvement, but the EQ-5D-5L VAS showed a statistically significant improvement.
• 6-month follow-up: Among those who completed measures at the first assessment, 3 months, and 6 months, C19-YRS and EQ-5D-5L VAS showed statistically significant improvement whereas EQ-5D-5L Index Value showed statistically significant deterioration. Patients at 6 months still had
significant LC symptom burden and disability compared to their pre-COVID-19 health status, i.e., their condition had improved but had not fully recovered. The follow-up changes in scores support the efficacy of interventions provided by LC services and suggest that continued specialist input is needed to manage these patients with persistent symptoms.
• C19-YRS (condition-specific measure) vs EQ-5D-5L (generic measure): The 3-month month follow-up changes in scores and responsiveness of PROMs highlight that C19-YRS is a more sensitive measure than EQ-5D-5L in this cohort of individuals with LC. This is in keeping with the literature recommending the use of condition-specific measures in addition to EQ-5D-5L.
• Vocational problems: 62% of this sample had their work role affected with them having to either be on sick leave, reduce hours, change roles, or quit roles. Only 21% were able to maintain their previous roles held prior to their COVID-19 infection. This is suggestive of considerable productivity loss and financial implications to the country.
• Fluctuating condition: In patients who completed multiple assessments, it was evident that LC is a fluctuant condition with no necessary linear trend of improvement or deterioration between the domains of symptom burden, functional disability, and overall health. This highlights the need to understand the triggers for the condition and invest in self-management and ongoing support from community healthcare services.
• Long-Term Condition: In most patients in this sample, LC has evidently become a Long-Term Condition (LTC) with fluctuations in their condition causing disability and significant deterioration of their overall health status seen even after 18 months of LC with no complete resolution or full recovery. There needs to be a national investment in managing this new LTC along with other LTCs.

Source: Dr Manoj Sivan, et al.  A National Evaluation of Outcomes in Long COVID Services using Digital PROM Data from the ELAROS Platform. National Evaluation of Long COVID Service Outcomes using ELAROS Data (09 Oct 23) https://locomotion.leeds.ac.uk/wp-content/uploads/sites/74/2023/10/National-Evaluation-of-LC-Service-Outcomes-using-ELAROS-Data-09-10-23.pdf (Full text)

How long is Long-COVID? Symptomatic improvement between 12 and 18 months in a prospective cohort study

Abstract:

Introduction COVID-19 infection can precede, in a proportion of patients, a prolonged syndrome including fatigue, exercise intolerance, mood and cognitive problems. This study aimed to describe the profile of fatigue-related, exercise-related, mood-related and cognitive-related outcomes in a COVID-19-exposed group compared with controls.

Methods 113 serving UK Armed Forces participants were followed up at 5, 12 (n=88) and 18 months (n=70) following COVID-19. At 18 months, 56 were in the COVID-19-exposed group with 14 matched controls. Exposed participants included hospitalised (n=25) and community (n=31) managed participants. 43 described at least one of the six most frequent symptoms at 5 months: fatigue, shortness of breath, chest pain, joint pain, exercise intolerance and anosmia. Participants completed a symptom checklist, patient-reported outcome measures (PROMs), the National Institute for Health cognitive battery and a 6-minute walk test (6MWT). PROMs included the Fatigue Assessment Scale (FAS), Generalised Anxiety Disorder-7 (GAD-7), Patient Health Questionnaire-9 (PHQ-9) and Patient Checklist-5 (PCL-5) for post-traumatic stress.

Results At 5 and 12 months, exposed participants presented with higher PHQ-9, PCL-5 and FAS scores than controls (ES (effect size) ≥0.25, p≤0.04). By 12 months, GAD-7 was not significantly different to controls (ES <0.13, p=0.292). Remaining PROMs lost significant difference by 18 months (ES ≤0.11, p≥0.28). No significant differences in the cognitive scales were observed at any time point (F=1.96, p=0.167). At 5 and 12 months, exposed participants recorded significantly lower distances on the 6MWT (ηp2≥0.126, p<0.01). 6MWT distance lost significant difference by 18 months (ηp2<0.039, p>0.15).

Conclusions This prospective cohort-controlled study observed adverse outcomes in depression, post-traumatic stress, fatigue and submaximal exercise performance up to 12 months but improved by 18-month follow-up, in participants exposed to COVID-19 compared with a matched control group.

Source: Barker-Davies RM, O’Sullivan O, Holdsworth DA, et alHow long is Long-COVID? Symptomatic improvement between 12 and 18 months in a prospective cohort studyBMJ Mil Health Published Online First: 03 October 2023. doi: 10.1136/military-2023-002500 https://militaryhealth.bmj.com/content/early/2023/10/03/military-2023-002500.abstract (Full text available as PDF file)

Correlation between Hepatocyte Growth Factor (HGF) with D-Dimer and Interleukin-6 as Prognostic Markers of Coagulation and Inflammation in Long COVID-19 Survivors

Abstract:

In general, an individual who experiences the symptoms of Severe Acute Respiratory Syndrome Coronavirus 2 or SARS-CoV-2 infection is declared as recovered after 2 weeks. However, approximately 10–20% of these survivors have been reported to encounter long-term health problems, defined as ‘long COVID-19’, e.g., blood coagulation which leads to stroke with an estimated incidence of 3%, and pulmonary embolism with 5% incidence.
At the time of infection, the immune response produces pro-inflammatory cytokines that stimulate stromal cells to produce pro-hepatocyte growth factor (pro-HGF) and eventually is activated into hepatocyte growth factor (HGF), which helps the coagulation process in endothelial and epithelial cells. HGF is a marker that appears as an inflammatory response that leads to coagulation.
Currently, there is no information on the effect of SARS-CoV-2 infection on serum HGF concentrations as a marker of the prognosis of coagulation in long COVID-19 survivors. This review discusses the pathophysiology between COVID-19 and HGF, IL-6, and D-dimer.
Source: Zaira B, Yulianti T, Levita J. Correlation between Hepatocyte Growth Factor (HGF) with D-Dimer and Interleukin-6 as Prognostic Markers of Coagulation and Inflammation in Long COVID-19 Survivors. Current Issues in Molecular Biology. 2023; 45(7):5725-5740. https://doi.org/10.3390/cimb45070361 https://www.mdpi.com/1467-3045/45/7/361 (Full text)

Determinants of the Onset and Prognosis of the Post-COVID-19 Condition: A 2-Year Prospective Cohort Study

Abstract:

Background: At least 5-10% of subjects surviving COVID-19 develop the post-COVID-19 condition (PCC) or “Long COVID”. The clinical presentation of PCC is heterogeneous, its pathogenesis is being deciphered, and objective, validated biomarkers are lacking. It is unknown if PCC is a single entity or a heterogeneous syndrome with overlapping pathophysiological basis. In a large crossectional evaluation, the RECOVER study in the US identified four clusters of subjects with PCC according to their presenting symptoms. The long-term clinical implications of PCC remain unknown.

Methods: We conducted a 2-year prospective cohort study of subjects surviving COVID-19, including individuals fulfilling the WHO PCC definition and subjects with full clinical recovery. We systematically collected post-COVID-19 symptoms using prespecified questionnaires and performed additional diagnostic imaging tests when needed. Factors associated with PCC were identified and modeled using logistic regression. Unsupervised clustering analysis was used to group subjects with PCC according to their presenting symptoms. Factors associated with PCC recovery were modelled using a direct acyclic graph approach.

Findings: The study included 548 individuals, 341 with PCC, followed for a median of 23 months (IQR 16·5 – 23·5), and 207 subjects fully recovered. In the model with the best fit, subjects who were male and had tertiary studies were less likely to develop PCC, whereas a history of headache, or presence of tachycardia, fatigue, neurocognitive and neurosensitive complaints and dyspnea at COVID-19 diagnosis predicted the development of PCC. The cluster analysis revealed the presence of three symptom clusters with an additive number of symptoms. Only 26 subjects (7·6%) recovered from PCC during follow-up; almost all of them (n=24) belonged to the less symptomatic cluster A, dominated mainly by fatigue. Recovery from PCC was more likely in subjects who were male, required ICU admission, or had cardiovascular comorbidities, hyporexia and/or smell/taste alterations during acute COVID-19. Subjects presenting with muscle pain, impaired attention, dyspnea, or tachycardia, conversely, were less likely to recover from PCC.

Interpretation: Preexisting medical and socioeconomic factors, as well as acute COVID-19 symptoms, predict the development of and recovery from the PCC. Recovery is extremely rare during the first 2 years, posing a major challenge to healthcare systems.

Source: Mateu, Lourdes and Tebe, Cristian and Loste, Cora and Santos, José Ramón and Lladós, Gemma and López, Cristina and España-Cueto, Sergio and Toledo, Ruth and Font, Marta and Chamorro, Anna and Muñoz-López, Francisco and Nevot, Maria and Vallejo, Nuria and Teis, Albert and Puig, Jordi and Fumaz, Carmina Rodríguez and Muñoz-Moreno, José Antonio and Prats, Anna and Estany-Quera, Carla and Coll-Fernández, Roser and Herrero, Cristina and Casares, Patricia and Garcia, Anna and Paredes, Roger and Clotet, Bonaventura and Massanella, Marta, Determinants of the Onset and Prognosis of the Post-COVID-19 Condition: A 2-Year Prospective Cohort Study. Available at SSRN: https://ssrn.com/abstract=4505315 or http://dx.doi.org/10.2139/ssrn.4505315 (Full text available as PDF file)

How Does Long-COVID Impact Prognosis and the Long-Term Sequelae?

Abstract:

Context: We reviewed what has been studied and published during the last 3 years about the consequences, mainly respiratory, cardiac, digestive, and neurological/psychiatric (organic and functional), in patients with COVID-19 of prolonged course.
Objective: To conduct a narrative review synthesizing current clinical evidence of abnormalities of signs, symptoms, and complementary studies in COVID-19 patients who presented a prolonged and complicated course.
Methods: A review of the literature focused on the involvement of the main organic functions mentioned, based almost exclusively on the systematic search of publications written in English available on PubMed/MEDLINE.
Results: Long-term respiratory, cardiac, digestive, and neurological/psychiatric dysfunction are present in a significant number of patients. Lung involvement is the most common; cardiovascular involvement may happen with or without symptoms or clinical abnormalities; gastrointestinal compromise includes the loss of appetite, nausea, gastroesophageal reflux, diarrhea, etc.; and neurological/psychiatric compromise can produce a wide variety of signs and symptoms, either organic or functional. Vaccination is not associated with the emergence of long-COVID, but it may happen in vaccinated people.
Conclusions: The severity of illness increases the risk of long-COVID. Pulmonary sequelae, cardiomyopathy, the detection of ribonucleic acid in the gastrointestinal tract, and headaches and cognitive impairment may become refractory in severely ill COVID-19 patients.
Source: Baroni C, Potito J, Perticone ME, Orausclio P, Luna CM. How Does Long-COVID Impact Prognosis and the Long-Term Sequelae? Viruses. 2023; 15(5):1173. https://doi.org/10.3390/v15051173 https://www.mdpi.com/1999-4915/15/5/1173 (Full text)

Epidemiological and clinical perspectives of long COVID syndrome

Abstract:

Long COVID, or post-acute COVID-19 syndrome, is characterized by multi-organ symptoms lasting 2+ months after initial COVID-19 virus infection. This review presents the current state of evidence for long COVID syndrome, including the global public health context, incidence, prevalence, cardiopulmonary sequelae, physical and mental symptoms, recovery time, prognosis, risk factors, rehospitalization rates, and the impact of vaccination on long COVID outcomes. Results are presented by clinically relevant subgroups.

Overall, 10-35% of COVID survivors develop long COVID, with common symptoms including fatigue, dyspnea, chest pain, cough, depression, anxiety, post-traumatic stress disorder, memory loss, and difficulty concentrating. Delineating these issues will be crucial to inform appropriate post-pandemic health policy and protect the health of COVID-19 survivors, including potentially vulnerable or underrepresented groups. Directed to policymakers, health practitioners, and the general public, we provide recommendations and suggest avenues for future research with the larger goal of reducing harms associated with long COVID syndrome.

Source: Huerne K, Filion KB, Grad R, Ernst P, Gershon AS, Eisenberg MJ. Epidemiological and clinical perspectives of long COVID syndrome. Am J Med Open. 2023 Jun;9:100033. doi: 10.1016/j.ajmo.2023.100033. Epub 2023 Jan 18. PMID: 36685609; PMCID: PMC9846887. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9846887/ (Full text)