Abstract
Background/Objectives: While COVID-19 severity is strongly associated with older age, younger adults may also develop complications. This study investigated factors associated with transfer to ICU in hospitalized younger adults during May–December 2021. Materials and methods: Eighty-six consecutive patients aged ≥18 and <50 years were admitted to the University Multi-profile Hospital for Active Treatment of Infectious and Parasitic Diseases “Prof. Ivan Kirov”, Sofia, Bulgaria between 12 May 2021, and 7 December 2021. Results: Among 86 patients (median age 40 years; 55.8% female), common symptoms included fever (89.5%), cough (91.9%), headache (76.7%), and gastrointestinal symptoms (83.7%). Laboratory abnormalities were frequent: lymphopenia (median 0.94 × 109/L), elevated CRP (28.9 mg/L), LDH (333 U/L), and ferritin (198.5 µg/L). Severe andcritical disease occurred in 17.5% of cases, with 43% requiring oxygen therapy. In descriptive/univariate analyses, ICU transfer was more frequent among patients with hypoxemia (SaO2 < 90%) and higher CRP, D-dimer, vomiting and ferritin. In a Firth penalized logistic regression model (rare-event setting), endocrine/metabolic comorbidity showed a directionally consistent association with ICU transfer; gastrointestinal comorbidity signals were not interpretable because of extremely small cell counts. Conclusions: Younger adults usually present with mild to moderate COVID-19; however, a substantial minority develop severe illness. Hypoxemia and elevated inflammatory and coagulation markers were consistently associated with worse outcomes. Endocrine/metabolic comorbidity showed the most consistent association with transfer to ICU in this cohort, although all comorbidity estimates should be interpreted cautiously, given the small number of events.
1. Introduction
COVID-19, caused by SARS-CoV-2, has affected millions worldwide, with age and comorbidities identified as major risk factors for severe outcomes [1,2,3]. While older patients are at the highest risk, younger adults can also develop life-threatening complications [4,5]. This is clinically relevant as younger adults form a large proportion of the workforce and play a central role in viral transmission.
Clinical presentations in younger patients include fever, cough, headache, and gastrointestinal symptoms [6,7,8]. Laboratory markers such as lymphopenia, elevated CRP, LDH, ferritin, and D-dimer are consistently linked to adverse outcomes [9,10,11]. However, data specific to younger adults remain limited, particularly in Eastern Europe. Younger adults are generally underrepresented in clinical research, as most studies focus on high-risk older populations. Yet, younger individuals frequently lack chronic comorbidities, meaning that atypical symptoms or less recognized risk factors may play a disproportionately important role in clinical deterioration [12].
In addition to confirming the role of inflammatory and coagulation markers, our findings indicate that endocrine/metabolic comorbidity was associated with severe outcomes in this age group, while gastrointestinal comorbidity signals were exploratory and limited by very small cell counts. Given the low event rate and comorbidity heterogeneity, these results should be interpreted cautiously and considered hypothesis-generating.
2. Materials and Methods
2.1. Methods
We retrospectively analyzed 86 consecutive adult patients aged ≥18 and <50 years admitted with confirmed SARS-CoV-2 infection (observed age range in this cohort: 18–48 years). Demographic, clinical, laboratory, imaging, treatment, and outcome data were collected from hospital records. Endocrine/metabolic comorbidity was selected a priori for penalized regression based on clinical relevance and prior evidence. There were no missing values for variables included in the analyses; therefore, all analyses were conducted on the full cohort (n = 86) using a complete-case approach. Because ICU transfer occurred in only five patients, the primary inferential analysis was restricted to a minimal, clinically prespecified set of candidate factors, and Firth penalized logistic regression was applied.
2.2. Study Design and Population
This was a retrospective observational cohort study conducted at University Multi-profile Hospital for Active Treatment of Infectious and Parasitic Diseases “Prof. Ivan Kirov”, Sofia, Bulgaria. A total of 86 consecutive adult patients aged ≥18 and <50 years hospitalized with COVID-19 were included (observed age range: 18–48 years). All consecutive adult admissions meeting the eligibility criteria during the study period were assessed for inclusion. Patients were excluded if the medical record was incomplete, the diagnosis of SARS-CoV-2 infection was uncertain, or if the patient was pregnant or postpartum. No other exclusion criteria were applied. The final cohort comprised 86 patients, of whom 5 required transfer to ICU. Diagnosis of COVID-19 was confirmed by a positive reverse transcriptase polymerase chain reaction (RT-PCR) or a positive rapid antigen test for SARS-CoV-2 in accordance with national guidelines. The study period was 12 May 2021, to 7 December 2021. Viral sequencing was not performed as part of this study; therefore, variant attribution at the individual level was not available. The cohort falls within the pre-Omicron period in Bulgaria/Europe (mid–late 2021), when Delta became the dominant circulating variant regionally. Findings should therefore be interpreted in the context of that period.
2.3. Data Collection
Clinical, laboratory, radiological, and treatment-related data were extracted from hospital medical records using a standardized collection form. Variables recorded included:
Demographics: age, sex.
Clinical presentation: fever, cough, headache, myalgia, gastrointestinal symptoms, dyspnea, anosmia, ageusia.
Vomiting was recorded as present if documented in the admission history during the initial physical examination note (patient-reported). No standardized grading (frequency/severity) was available in the charts; vomiting was therefore analyzed as a binary variable (yes/no). For factors associated with transfer to ICU analyses, the primary vomiting variable was ‘vomiting at presentation,’ defined as vomiting documented on admission and prior to initiation of in-hospital therapies. Clinical notes were not standardized with respect to whether vomiting was cough-induced (post-tussive) versus independent; this distinction could not be extracted reliably.
Comorbidities: hypertension, diabetes, cardiovascular, pulmonary, renal, neurological, autoimmune, oncologic, and endocrine diseases.
Vital signs and oxygen saturation (SaO2) at admission.
Laboratory tests: complete blood count (CBC), C-reactive protein (CRP), lactate dehydrogenase (LDH), ferritin, urea, creatinine, liver function tests (AST, ALT), D-dimer, INR, arterial blood gases. Arterial blood gas parameters (pO2, pCO2) were obtained from capillary blood samples, which typically yield lower pO2 values than arterial measurements and were not used for WHO severity classification.
Imaging findings: chest X-ray patterns (interstitial, infiltrative, ground-glass opacities).
Microbiological tests: throat swabs for bacterial/fungal co-infections.
Treatment: antiviral drugs, antibiotics, corticosteroids, anticoagulants, oxygen therapy, and intravenous infusions.
Outcomes: length of hospital stay, disease severity, and discharge status.
Vaccination status was not consistently recorded in the charts and could not be analyzed.
Endocrine and metabolic comorbidities included obesity (when explicitly documented as a comorbidity in the epicrisis, e.g., obesity grade II or III), diabetes mellitus (type 1 or type 2), and thyroid disorders (hypothyroidism, Hashimoto thyroiditis, or Graves’ disease). These conditions were analyzed as a combined category due to the limited sample size and low number of severe outcome events. Obesity was identified exclusively based on explicit clinician documentation in the admission or discharge epicrisis. Body mass index (BMI) values were not systematically recorded and were therefore not recalculated retrospectively. WHO BMI criteria are referenced solely for conceptual definition and not for quantitative analysis in the present study.
Other comorbidity categories included cardiovascular (arterial hypertension, hypertensive heart disease, valvular disorders), gastrointestinal (gastro-oesophageal reflux disease, peptic ulcer disease), respiratory (bronchial asthma, chronic sinusitis or allergic rhinitis), renal and urological (nephrolithiasis, chronic pyelonephritis, history of nephrectomy), neurologic (epilepsy, post-traumatic paresis), autoimmune and dermatologic (psoriasis), oncologic conditions, and HIV infection. Inflammatory bowel disease and other chronic inflammatory gastrointestinal disorders requiring immunosuppressive or biologic therapy were not documented in this cohort.
For the purposes of regression analyses, binary variables were created to indicate the presence or absence of any cardiovascular, gastrointestinal, or endocrine/metabolic comorbidity. Patients could contribute to more than one comorbidity category.
2.4. Definitions
Severity of disease was classified according to WHO guidance:
Mild: SaO2 ≥ 94% on room air;
Moderate: SaO2 90–94%;
Severe: SaO2 < 90%;
Critical: requiring mechanical ventilation, shock, or associated with multi-organ failure.
Primary outcome (endpoint): ICU transfer, defined as transfer from the study hospital to an intensive care unit during the index hospitalization
Laboratory abnormalities were defined using the hospital laboratory reference ranges.
Comorbidities were considered present if they were documented in the patient’s history, discharge summaries, or prior specialist evaluations. Consequently, detailed BMI distributions and obesity severity categories could not be analyzed. Obesity was considered present when explicitly documented in the patient’s medical history or discharge summary; body mass index was not systematically recalculated for the purposes of this study. Cardiovascular and endocrine comorbidities were analyzed separately due to their hypothesized association with inflammatory dysregulation in COVID-19.
2.5. Statistical Analysis
Categorical variables were expressed as frequencies (n) and proportions (%). Continuous variables were presented depending on their distribution: mean with standard deviation (SD) for normally distributed variables, and median with interquartile range (IQR, 25th–75th percentiles) for non-Gaussian distributions. Normality of distribution was assessed using the Kolmogorov–Smirnov test. Mann–Whitney U test or Fisher’s Exact test were used to compare two groups. Firth penalized logistic regression was also applied. Statistical significance was defined as p < 0.05. All analyses were performed using IBM SPSS software, version 29 as well as R version 4.5.5. Given the small number of ICU transfer events (N = 5), univariate comparisons are exploratory/descriptive and were not adjusted for multiple testing. We therefore interpret p-values as descriptive signals rather than confirmatory evidence.
2.6. Ethical Considerations
This study was a retrospective, non-interventional analysis of routinely collected clinical data conducted at the University Multi-profile Hospital for Active Treatment of Infectious and Parasitic Diseases “Prof. Ivan Kirov”, Sofia, Bulgaria.
According to Bulgarian national regulations and institutional policy, retrospective analyses of existing medical documentation performed without patient re-contact, intervention, or collection of additional data, and using fully anonymized datasets, do not require formal ethics committee review.
The exemption from formal ethics review and permission for use of anonymized patient data for scientific publication were formally certified by the Executive Director of the institution, who is the responsible institutional authority (Official institutional certificate No. 39/09.01.2026).
2.7. Data Availability
The dataset analyzed during the current study is available from the corresponding author upon reasonable request. At the time of submission, the data are not deposited in a public repository; accession numbers will be provided upon acceptance for publication.
3. Results
3.1. Clinical Characteristics
A total of 86 younger adult patients ranging from 18 to 48 years old with confirmed COVID-19 were analyzed. The median age was 40 years (IQR 34–44), and 55.8% were female. The demographic and clinical characteristics of younger adults hospitalized with COVID-19 are summarized in Table 1.
Table 1.
Demographic and Clinical Features of Younger Adults with COVID-19.
3.1.1. Symptoms at Presentation
The most common clinical features included:
Fever (89.5%, with 2.3% presenting high-grade fever ≥ 39.8 °C);
Cough (91.9%, lasting >10 days in 44.2%);
Headache (76.7%);
Musculoskeletal/joint pain (69.8%);
Gastrointestinal (GI) symptoms (83.7%), including loss of appetite (67.4%), nausea (50.0%), vomiting at presentation (25.6%), and diarrhea (37.2%);
Olfactory/gustatory disturbances were frequent (75.6%), predominantly combined anosmia and ageusia (53.5%). Dyspnea occurred in 34.9%, ranging from exertional (18.6%) to at rest (9.3%).
3.1.2. Comorbidities
Cardiovascular conditions were the most frequent comorbidities, with arterial hypertension documented in nine patients, along with isolated rhythm or structural cardiac abnormalities in a small number of cases.
Endocrine and metabolic comorbidities were heterogeneous and included autoimmune thyroiditis (Hashimoto’s disease) in six patients, diabetes mellitus in two patients, obesity in three patients, hypothyroidism in one patient, and Graves’ disease in one patient. Obesity severity could not be systematically stratified because BMI values were not uniformly available in the medical records.
Gastrointestinal comorbidities were infrequent and consisted of gastro-oesophageal reflux disease and peptic ulcer disease, each observed in one patient. Given the extremely sparse cell counts for gastrointestinal comorbidity (2 patients in total), GI-comorbidities estimates were considered exploratory and were not used for primary inference.
Respiratory conditions included chronic sinusitis (two patients), allergic rhinitis or hay fever (two patients in total), and bronchial asthma (one patient). Renal and urological disorders (chronic calculous pyelonephritis, nephrolithiasis, history of nephrectomy), neurological or psychiatric conditions, autoimmune and hematologic disorders, oncologic disease, and HIV infection were each present in a small number of patients.
Overall, eleven patients had more than one documented comorbidity.
3.2. Laboratory Findings
Baseline laboratory parameters of the younger adult COVID-19 cohort are summarized in Table 2. Key abnormalities included lymphopenia (median 0.94 × 109/L, IQR 0.65–1.22), elevated CRP (28.9 mg/L, IQR 5.2–83.8), LDH (333 U/L, IQR 225–494), and ferritin (198.5 µg/L, IQR 88.6–535.5). ALT and AST were moderately increased, while creatinine remained within reference range.
Table 2.
Laboratory Characteristics of Younger Adults with COVID-19.
Arterial blood gases revealed reduced pO2 (mean 64.2 mmHg, SD 10.5) and mild hypocapnia (pCO2 mean 34.2 mmHg, SD 3.9).
In descriptive comparisons, patients with endocrine/metabolic comorbidity showed higher inflammatory and tissue-injury markers. Gastrointestinal comorbidity signals were based on very small numbers, and these comparisons were not powered for robust subgroup inference.
3.3. Imaging and Microbiology
Chest radiography demonstrated:
Infiltrative changes in 66.3%;
Interstitial changes in 11.6%;
Ground-glass opacities in 4.7%;
No abnormalities in 17.4%.
Microbiological analysis of throat swabs revealed normal flora in the majority (83.7%). Pathogens were isolated in 16.3%, most frequently Candida albicans (10.5%) and Haemophilus influenzae (3.5%). Bacterial co-infection with Staphylococcus aureus and Pseudomonas aeruginosa was rare (1.2% each).
3.4. Treatment and Outcomes
Patients with cardiovascular and endocrine/metabolic comorbidities were more likely to receive oxygen therapy, anticoagulation and systemic corticosteroids, reflecting their more severe clinical course. Most patients received antibiotics (96.5%), typically Ceftriaxone 2 × 2 g IV or Levofloxacin 2 × 500 mg IV throughout hospitalization. Antibiotics were indicated for suspected bacterial coinfection (clinical, radiographic, or laboratory evidence) or severe disease requiring oxygen.
Antiviral therapy with Remdesivir (Veklury®) was administered in 7.0%, using a 5-day regimen (200 mg IV day 1, followed by 100 mg IV daily). Inclusion criteria were confirmed infection with hypoxemia (SaO2 ≤ 94%) or radiographic pneumonia.
Corticosteroids were given to 25.6%, anticoagulants to 93.0%, oxygen supplementation to 43.0%, and IV infusions to 62.8%.
Median hospital stay was 8.5 days (IQR 6–11). Disease severity distribution was:
Mild: 46.5%.
Moderate: 36.0%.
Severe: 12.8%.
Critical: 4.7%.
Outcomes were favorable, with improvement in 94.2%. Five patients (5.8%) required transfer to ICU, but no deaths occurred.
3.5. Factors Associated with ICU Transfer
Transfer to ICU occurred in 5/86 (5.8%) patients. This endpoint was rare; therefore, all comparisons should be interpreted as hypothesis-generating. In univariate analyses, vomiting at presentation and several laboratory markers (CRP, ALT, AST, LDH, ferritin, D-dimer, and pO2) were more pronounced among transferred to ICU patients, and antiviral therapy use was more frequent among transferred patients, consistent with confounding by indication.
Given the very small number of transfer events and sparse cell counts, conventional multivariable logistic regression yielded unstable estimates with very wide confidence intervals, consistent with quasi-complete separation; therefore, these models were considered exploratory and not used for primary inference.
Direct comparisons between patients with improved outcomes and those transferred are presented in Table 3. Vomiting, dyspnea, SaO2 < 90%, and endocrine/metabolic comorbidity were more frequent among transferred patients, alongside higher inflammatory/tissue-injury markers and lower pO2.
Table 3.
Clinical, laboratory, and comorbidity characteristics by clinical outcome (improved vs. transferred to ICU).
To address sparse-data bias, Firth penalized logistic regression was performed with endocrine/metabolic comorbidity as a prespecified predictor. Endocrine/metabolic comorbidity remained associated with transfer to ICU (penalized log-odds coefficient 3.45; p = 0.0006), corresponding to an odds ratio of approximately 31. Given sparse data, the OR should be viewed as an indicator of outcome clustering rather than a stable quantitative estimate. The magnitude should be interpreted cautiously, given N = 5 events. Simple 2 × 2 contingency tables are provided in Table 3 to allow direct inspection of sparse counts. Larger datasets are required to evaluate gastrointestinal comorbidity and multi-predictor models reliably in this age group.
Endocrine/metabolic comorbidity was associated with markedly increased odds of transfer to ICU in a penalized model accounting for sparse data; however, the magnitude of this association should be interpreted cautiously given the very small number of events. Results of the Firth penalized logistic regression analysis examining the association between endocrine/metabolic comorbidity and transfer to the ICU are presented in Table 4.
Table 4.
Firth Penalized Logistic Regression Model Examining Endocrine/Metabolic Comorbidity and Transfer to ICU.
4. Discussion
In our cohort of younger adults with COVID-19, several findings stood out as clear warning signs for clinical deterioration and subsequent ICU transfer. This pattern suggests that, even in younger age groups, baseline health status substantially modifies the clinical trajectory of COVID-19. This observation is consistent with larger regional syntheses demonstrating that cardiometabolic and endocrine-related comorbidities confer increased risk of severe COVID-19 outcomes independent of age, and that multimorbidity rather than chronological age alone is a key driver of disease severity [13].
Vomiting documented at presentation was associated with transfer to ICU in this cohort. However, vomiting in younger adults with COVID-19 may have several non-mutually exclusive explanations, including nonspecific manifestation of systemic illness, delayed presentation, cough-associated (post-tussive) vomiting in the context of high cough prevalence. Because the charts did not allow reliable attribution of vomiting to cough triggers or to ambulatory medication timing, mechanistic inference is limited. We therefore interpret vomiting primarily as a clinical warning sign that may co-occur with a higher inflammatory burden and/or more severe baseline respiratory compromise, rather than as evidence of a specific gastrointestinal pathophysiology.
Dyspnea and low oxygen saturation were also strong indicators of severity. In our cohort, most younger adults were previously healthy or had only one chronic condition, yet comorbidities clearly clustered among those with more severe disease. Cardiovascular disorders and endocrine/metabolic comorbidity categories were particularly overrepresented in patients who required ICU care, whereas patients with mild courses were more often free of underlying disease. This pattern suggests that, even in younger age groups, baseline health status substantially modifies the clinical trajectory of COVID-19. This is consistent with previous work, including the study by Xie et al., which showed that even modest reductions in SpO2 are closely connected to mortality. Our data suggest that this relationship holds true even in younger individuals, who are often assumed to be at lower risk [5,6].
Inflammatory and biochemical abnormalities—including elevated CRP, ALT, AST, LDH, ferritin, and reduced pO2—were consistently higher among transferred patients, aligning with patterns described in large multiwave cohorts. These patterns mirror what has been repeatedly described in severe COVID-19 across large multiwave cohorts, where respiratory compromise measures and biomarkers of systemic injury (including LDH and D-dimer) remain consistently informative for ICU admission and mortality prediction across variant periods [7,9,14,15,16,17,18].
Most importantly, comorbidity analysis showed an association between endocrine/metabolic comorbidity and severe disease, while gastrointestinal comorbidity signals were exploratory and should be interpreted with caution given the small number of events and heterogeneity of diagnoses. Endocrine/metabolic comorbidities were present in a substantial subset of patients and were disproportionately represented among those transferred to the ICU. Given the low overall event rate, effect estimates are inherently imprecise and reflect outcome clustering within a small number of affected patients rather than stable quantitative measures of risk. After addressing sparse-data bias using Firth penalized logistic regression, endocrine/metabolic comorbidity remained strongly associated with transfer, supporting a clinically meaningful signal despite the low event count.
Importantly, these comorbidities were individually infrequent and clinically heterogeneous. Autoimmune thyroiditis was the most commonly documented endocrine condition in this cohort, whereas diabetes mellitus and obesity were present in a smaller number of patients, and other endocrine disorders occurred sporadically. Obesity was identified from clinician-documented epicrisis data rather than standardized BMI measurement, precluding analysis of obesity severity or dose–response relationships. Gastrointestinal comorbidities were rare and consisted mainly of isolated cases of gastro-oesophageal reflux disease or peptic ulcer disease. Taken together, this distribution indicates that the observed association reflects aggregation of diverse cardiometabolic and thyroid-related vulnerabilities rather than the effect of a single dominant diagnosis.
These findings partially overlap with larger young-adult cohorts, in which obesity, diabetes, and cardiovascular disease have repeatedly emerged as major risk factors for hospitalization, severe pneumonia, and readmission. In our cohort, endocrine/metabolic comorbidity showed the most consistent association with deterioration, suggesting that cardiometabolic and thyroid-related vulnerability may be clinically relevant even in younger patients. Given the low prevalence of classic cardiometabolic conditions in this age group, endocrine/metabolic disorders may serve as early warning markers for deterioration in hospitalized younger adults. We also observed an association between the use of antiviral therapy—primarily remdesivir—and ICU transfer. This likely reflects delayed presentation and treatment initiation in our cohort rather than an adverse effect of the drug itself.
Candida albicans colonization showed a descriptive signal toward higher frequency among ICU transfers, but this finding is based on very small numbers and should be interpreted cautiously.
The very large effect estimates observed for comorbidity categories must be interpreted with caution, as they are driven by a small number of outcome events and sparse data, which inherently limit precision. Moreover, escalation endpoints such as ICU transfer are inherently context-dependent and may be influenced by local triage practices, resource availability, and pandemic phase, as demonstrated in large emergency department linkage cohorts using composite severe outcomes (ICU admission or death) [19,20]. Although ICU transfer is a clinically meaningful endpoint, its rarity in this cohort (N = 5) and the known variability of escalation practices across settings limit generalizability; nevertheless, the alignment of this endpoint with established inflammatory and respiratory severity markers observed in large contemporary cohorts supports its clinical relevance. In this context, conventional odds ratios should not be viewed as stable quantitative measures of risk but rather as indicators of outcome clustering under rare-event conditions. After accounting for sparse-data bias using Firth penalized logistic regression, endocrine/metabolic comorbidity remained the only comorbidity category consistently associated with transfer to ICU, whereas gastrointestinal comorbidity signals were exploratory and not supported in penalized models. Accordingly, the primary conclusions of this study are based on penalized regression and coherent clinical patterns rather than on isolated univariate p-values. Larger, multicenter studies in younger populations are required to clarify the prognostic role of specific endocrine/metabolic diagnoses and to determine whether gastrointestinal conditions carry independent risk.
Limitations and Future Directions
This study has several limitations. First, it was conducted in a single center, which may limit the generalizability of the findings. Multiple univariate comparisons were performed to explore potential associations; these analyses are exploratory and were not adjusted for multiple testing. Accordingly, findings from univariate analyses should be interpreted cautiously and not as definitive causal relationships. Only five patients experienced ICU transfer, which substantially limits statistical power and precision. Consequently, multivariable modeling beyond a minimal prespecified set of predictors is not reliable, and effect estimates should be treated as hypothesis-generating. Additionally, ICU transfer is influenced by local escalation practices and capacity; therefore, external generalizability is limited. Because individual-level variant data were unavailable and the cohort predates Omicron dominance, external validity to later periods with different variants and clinical practices may be limited. Vaccination status was unavailable/incomplete and could not be adjusted for; this is a major limitation because vaccination modifies both disease severity and clinical management. Second, the sample size, although adequate for initial comparisons, may limit statistical power in detecting subtle associations. Additionally, obesity was derived from clinician-documented epicrisis data rather than standardized BMI measurements, and detailed BMI distributions or severity categories were unavailable. Vomiting was abstracted from retrospective clinical documentation as a binary symptom present at admission.
Future studies should aim to validate these findings in larger, multicenter cohorts and explore the role of genetic and immunological factors in shaping COVID-19 severity in young adults. Longitudinal follow-up will also be essential to assess the impact of post-COVID complications and to identify predictors of long-term outcomes in this age group.
5. Conclusions
Younger adults with COVID-19 usually experience mild or moderate disease; however, vomiting at presentation, hypoxemia (SaO2 < 90%), elevated inflammatory, liver markers and D-dimer were associated with more severe courses. Endocrine/metabolic comorbidity was associated with transfer to ICU; gastrointestinal comorbidity signals were limited by sparse data. Because vaccination status could not be assessed and individual-level variant data were unavailable, these findings should be interpreted as hypothesis-generating and primarily applicable to hospitalized younger adults in Bulgaria during the May–December 2021 (pre-Omicron) period.
These findings underscore the importance of early identification and close monitoring of younger patients with these risk factors, enabling timely interventions to prevent progression to critical illness.
Author Contributions
Conceptualization, P.T.; methodology, P.T. and E.N.; formal analysis, E.N.; investigation, P.T.; resources, P.T.; data curation, P.T.; writing—original draft preparation, P.T.; writing—review and editing, E.N. and N.Y.; supervision, E.N. and N.Y. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
This study was a retrospective, non-interventional analysis of routinely collected clinical data conducted at the University Multi-profile Hospital for Active Treatment of Infectious and Parasitic Diseases “Prof. Ivan Kirov”, Sofia, Bulgaria. According to Bulgarian national regulations and institutional policy, retrospective analyses of existing medical documentation performed without patient re-contact, intervention, or collection of additional data, and using fully anonymized datasets, do not require formal ethics committee approval. The exemption from formal ethics review and authorization for the use of anonymized patient data for scientific publication were formally certified by the Executive Director of the institution, who is the responsible institutional authority (Official institutional certificate No. 39/09.01.2026).
Informed Consent Statement
Individual informed consent for research participation was waived because the analysis was based exclusively on fully anonymized, routinely collected clinical data and involved no patient re-contact or intervention. All patients had provided written informed consent at hospital admission for diagnostic and therapeutic procedures and for processing of personal data as part of standard clinical care, in accordance with Bulgarian law. No identifiable individual-level data or images are reported; therefore, separate consent for publication was not required.
Data Availability Statement
The data presented in this study are available on reasonable request from the corresponding author. The data are not publicly available due to privacy and ethical restrictions.
Conflicts of Interest
The authors declare no conflicts of interest.
Abbreviations
| ALT | Alanine Aminotransferase |
| APTT | Activated Partial Thromboplastin Time |
| AST | Aspartate Aminotransferase |
| BE | Base Excess |
| CI | Confidence Interval |
| GI | Gastrointestinal |
| CRP | C-Reactive Protein |
| HCO3− | Bicarbonate |
| HCT | Hematocrit |
| HGB | Hemoglobin |
| ICU | Intensive Care Unit |
| INR | International Normalized Ratio |
| IQR | Interquartile Range |
| LDH | Lactate Dehydrogenase |
| Lym | Lymphocytes |
| MLR | Monocyte-to-Lymphocyte Ratio |
| Me | Median |
| OR | Odds Ratio |
| PCR | Polymerase Chain Reaction |
| PLR | Platelet-to-Lymphocyte Ratio |
| PLT | Platelets |
| pO2 | Partial Pressure of Oxygen |
| pCO2 | Partial Pressure of Carbon Dioxide |
| SaO2 | Arterial Oxygen Saturation |
| SD | Standard Deviation |
| SARS-CoV-2 | Severe Acute Respiratory Syndrome Coronavirus 2 |
| WBC | White Blood Cells |
References
- Guan, W.J.; Ni, Z.Y.; Hu, Y.; Liang, W.H.; Ou, C.Q.; He, J.X.; Liu, L.; Shan, H.; Lei, C.L.; Hui, D.S.C.; et al. Clinical characteristics of coronavirus disease 2019 in China. N. Engl. J. Med. 2020, 382, 1708–1720. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Huang, C.; Wang, Y.; Li, X.; Ren, L.; Zhao, J.; Hu, Y.; Zhang, L.; Fan, G.; Xu, J.; Gu, X.; et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 2020, 395, 497–506. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Pan, L.; Mu, M.; Yang, P.; Sun, Y.; Wang, R.; Yan, J.; Li, P.; Hu, B.; Wang, J.; Hu, C.; et al. Clinical characteristics of COVID-19 patients with digestive symptoms in Hubei, China. Am. J. Gastroenterol. 2020, 115, 766–773. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Sultan, S.; Altayar, O.; Siddique, S.M.; Davitkov, P.; Feuerstein, J.D.; Lim, J.K.; Falck-Ytter, Y.; El-Serag, H.B. AGA Institute rapid review of the gastrointestinal and liver manifestations of COVID-19, meta-analysis of international data, and recommendations for the consultative management of patients. Gastroenterology 2020, 159, 320–334.e27. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Xie, J.; Covassin, N.; Fan, Z.; Singh, P.; Gao, W.; Li, G.; Kara, T.; Somers, V.K. Association between hypoxemia and mortality in patients with COVID-19. Mayo Clin. Proc. 2020, 95, 1138–1147. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Lechien, J.R.; Chiesa-Estomba, C.M.; Place, S.; Van Laethem, Y.; Cabaraux, P.; Mat, Q.; Hans, S. Clinical and epidemiological characteristics of 1420 European patients with mild-to-moderate coronavirus disease 2019. J. Intern. Med. 2020, 288, 335–344. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Grant, M.C.; Geoghegan, L.; Arbyn, M.; Mohammed, Z.; McGuinness, L.; Clarke, E.L.; Wade, R.G. The prevalence of symptoms in 24,410 adults infected by the novel coronavirus (SARS-CoV-2; COVID-19): A systematic review and meta-analysis of 148 studies from 9 countries. PLoS ONE 2020, 15, e0234765. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Xiong, Q.; Xu, M.; Li, J.; Liu, Y.; Zhang, J.; Xu, Y.; Dong, W. Clinical sequelae of COVID-19 survivors in Wuhan, China: A single-centre longitudinal study. Clin. Microbiol. Infect. 2020, 27, 89–95. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Petrilli, C.M.; Jones, S.A.; Yang, J.; Rajagopalan, H.; O’Donnell, L.; Chernyak, Y.; Tobin, K.A.; Cerfolio, R.J.; Francois, F.; Horwitz, L.I. Factors associated with hospitalization and critical illness among 4,103 patients with COVID-19 in New York City. BMJ 2020, 369, m1966. [Google Scholar] [CrossRef] [Scilit]
- Zhou, F.; Yu, T.; Du, R.; Fan, G.; Liu, Y.; Liu, Z.; Xiang, J.; Wang, Y.; Song, B.; Gu, X.; et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: A retrospective cohort study. Lancet 2020, 395, 1054–1062. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Zheng, Z.; Peng, F.; Xu, B.; Zhao, J.; Liu, H.; Peng, J.; Li, Q.; Jiang, C.; Zhou, Y.; Liu, S.; et al. Risk factors of critical & mortal COVID-19 cases: A systematic literature review and meta-analysis. J. Infect. 2020, 81, e16–25. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Sandoval, M.; Nguyen, D.T.; Vahidy, F.S.; Graviss, E.A. Risk factors for severity of COVID-19 in hospital patients age 18–29 years. PLoS ONE 2021, 16, e0255544. [Google Scholar] [CrossRef] [Scilit] [PubMed] [PubMed Central]
- Thompson, M.; Buttery, A.K.; Oh, S.X.; Chan, M.; Lee, B.H.; Iino, T.; Wang, Y.A.; Clarke, C. Risk factors for severe COVID-19 outcomes in the Asia-Pacific region: A literature review. Front. Public Health 2025, 13, 1562179. [Google Scholar] [CrossRef] [Scilit] [PubMed] [PubMed Central]
- Zhang, J.J.; Dong, X.; Liu, G.H.; Gao, Y.D. Risk and Protective Factors for COVID-19 Morbidity, Severity, and Mortality. Clin. Rev. Allergy Immunol. 2023, 64, 90–107. [Google Scholar] [CrossRef] [Scilit] [PubMed] [PubMed Central]
- Dessie, Z.G.; Zewotir, T. Mortality-related risk factors of COVID-19: A systematic review and meta-analysis of 42 studies and 423,117 patients. BMC Infect. Dis. 2021, 21, 855. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Zhang, L.; Li, X.; Wang, Z.; Zhao, L.; Gao, H.; Liu, C.; Bai, J.; Liu, T.; Chen, W.; Li, W.; et al. Risk factors for severe COVID-19 and development of a predictive model. BMC Pulm. Med. 2025, 25, 422. [Google Scholar] [CrossRef] [Scilit] [PubMed] [PubMed Central]
- Ruiz Giardin, J.M.; Garnica, Ó.; Mesa Plaza, N.; SanMartín López, J.V.; Farfán Sedano, A.; Madroñal Cerezo, E.; Duarte Millán, M.Á.; Izquierdo Martínez, A.; Rivas, L.; Rivilla, M.; et al. AI Predictive Model of Mortality and Intensive Care Unit Admission in the COVID-19 Pandemic: Retrospective Population Cohort Study of 12,000 Patients. J. Med. Internet Res. 2025, 27, e70674. [Google Scholar] [CrossRef] [Scilit] [PubMed] [PubMed Central]
- Narin Çopur, E.; Ergün, D.; Ergün, R.; Atik, S.; Türk Dağı, H.; Körez, M.K. Risk Factors Affecting the Severity, Mortality, and Intensive Care Unit Admission of COVID-19 Patients: A Series of 1075 Cases. Viruses 2025, 17, 429. [Google Scholar] [CrossRef] [Scilit] [PubMed] [PubMed Central]
- Boldt, K.; Rose, N.; Ware, S.; Dinh, M.M.; Paul, K.K.; Ho, Y.; Muscatello, D.J. Characteristics and predictors of severe outcomes of COVID-19 cases presenting to the emergency department of a major Australian referral hospital: A record linkage study. Emerg. Med. Australas. 2025, 37, e70040. [Google Scholar] [CrossRef] [Scilit] [PubMed] [PubMed Central]
- Towair, E.; Al Achkar, Z.; Haykal, R.; Kassar, A.; Bou Karam, E.; Salameh, P.; Chaaban, T. Accuracy of Different COVID-19 Severity Scores in ICU Transfer Prediction and Mortality Prediction Among Inpatients at Lebanese American University Medical Center-Rizk Hospital (LAUMC-RH). Cureus 2025, 17, e95269. [Google Scholar] [CrossRef] [Scilit] [PubMed] [PubMed Central]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2026 by the authors. Published by MDPI on behalf of the Germs. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.