Recovery from acute COVID-19

By late 2022, over 700 million people had been infected with SARS-CoV-2 and more than six million had died. For many survivors, recovery from acute COVID-19 has been incomplete or temporary, characterized by persistent, recurrent, or new symptoms following the acute illness. Long COVID can follow asymptomatic, mild, moderate, severe, or critical disease, with risk increasing with greater acute severity. Nonetheless, the majority of Long COVID cases arise from the very large population who had asymptomatic or mild acute illness. As SARS-CoV-2 becomes endemic, the interplay between viral evolution, population immunity, and susceptibility will shape transmission, and even without highly pathogenic variants, the risk of Long COVID after mild illness remains significant. The paper aims to define Long COVID, review proposed mechanisms and clinical phenotypes, and provide a structured primary care approach to assessment, management, and rehabilitation.

The article synthesizes emerging evidence on Long COVID definitions, epidemiology, mechanisms, and phenotypes: - Definitions: The WHO consensus clinical case definition is supported, describing symptoms or symptom clusters persisting or developing at least 3 months from onset, lasting more than 2 months, impacting daily functioning, and not explained by alternative diagnoses. Overlap exists with ME/CFS, post-ICU syndrome, and POTS. - Epidemiology: Early pandemic estimates suggested up to 75% of critically ill, ~50% of hospitalized, and up to 30% of mild/asymptomatic cases might develop Long COVID. More recent estimates with vaccination and Omicron predominance suggest ~5%–10% prevalence across severities. In South Africa, 46.7% of hospitalized and 18.5% of non-hospitalized patients developed Long COVID across Beta, Delta, and Omicron waves, with Omicron-associated proportions markedly lower (~18.5% vs ~60% in prior waves). Less than 15% report complete recovery at 1 year. Risk factors include severe acute illness, reinfection, female sex, age, atopy, multimorbidity, obesity, diabetes, HIV, and other immunosuppression, though Long COVID occurs across all severities and demographics. - Mechanisms: Proposed mechanisms include viral persistence (RNA and antigens in multiple tissues), superantigen-driven immune activation, amyloidogenic platelet aggregates and microclots resistant to fibrinolysis, autoimmunity (anti-nuclear and extractable nuclear antigen autoantibodies), gut dysbiosis with prolonged inflammatory signatures, reactivation of latent herpesviruses (EBV, HHV-6), metabolic dysregulation (lipids, cytokines, metabolites), and neuroinflammation with elevated chemokines (e.g., CCL11) affecting neurogenesis, myelination, and neural circuitry. - Phenotypes: Common persistent symptoms include fatigue/post-exertional malaise (60%–70%), dyspnoea and cough (up to 60%), persistent anosmia/dysgeusia (20%–30% at ~8 months), cardiovascular symptoms (~10% of previously hospitalized) with increased MACE and arrhythmias up to 12 months, neuropsychiatric symptoms (40%–70%; >50% with cognitive impairment), musculoskeletal/rheumatic (~40% at 6 months), dermatologic manifestations persisting in some (e.g., COVID-toes, urticaria), and gastrointestinal complaints (10%–80%). - Increased risk of incident conditions: Post-COVID patients face elevated risks for discrete events and new diagnoses, including ischemic heart disease, dysrhythmias, myocarditis/pericarditis, heart failure, cerebrovascular events, thromboembolism, neurologic disorders (e.g., neuropathy, movement disorders), sensory deficits, cognitive disorders (e.g., Alzheimer’s), mental health conditions, rheumatological diseases, diabetes, chronic kidney and liver disease.

This is a narrative clinical guidance article synthesizing current evidence and proposing a structured primary care approach for assessment and management of Long COVID. Assessment timing and setting: - For mild/moderate cases managed in the community with progressive improvement, routine follow-up is not necessary; new, persistent, worsening, or recurrent symptoms should prompt consultation. - All patients (hospitalized or not) with ongoing concerns should have an initial screening and assessment visit approximately 12 weeks after acute infection, earlier if life-threatening symptoms arise. Prefer dedicated multidisciplinary clinics when available. Arrange follow-up within 6 weeks after discharge for hospitalized patients. - Early specialist referral is advised for severe multi-domain symptoms, atypical presentations, or severe functional impairment. Initial assessment components: - Confirm basis of initial COVID-19 diagnosis, acute illness severity, treatments, complications (e.g., VTE, AKI, myocardial injury, ARDS, delirium), and symptom trajectories. - Establish pre-COVID baseline health as comparator. Use validated questionnaires for symptom characterization, functional status, and social/occupational performance (e.g., WHO post-COVID-19 functional scale, MOCA). - Exclude serious conditions: myocarditis, myocardial ischemia/microvascular angina, arrhythmias, interstitial lung disease, VTE/PE, cerebral venous thrombosis, hepatitis. - Red flags warrant urgent referral: resting SpO2 < 94%, ≥4% exertional desaturation, unexplained tachycardia, hypotension, severe orthostatic hypotension, altered mental status. Investigations: - No labs for those fully recovered. For prior abnormal chest radiograph, repeat imaging at 6 weeks for current/former smokers. - For suspected Long COVID: CBC, renal function/electrolytes, LFTs (incl. albumin), CRP, ferritin, thyroid function, HbA1c; further tests guided by symptoms. No current role for PCR/antigen/antibody testing to diagnose Long COVID. Focused symptom-based management and evaluation: - Respiratory: At 12 weeks, obtain chest radiograph, spirometry if available, resting and post-exertion oximetry. Assess PE probability (e.g., Wells score); low probability with normal D-dimer excludes PE; otherwise refer for CTPA or V/Q imaging. Refer to pulmonology for persistent radiographic abnormalities, hypoxemia, exertional desaturation, or persistent symptoms beyond 6 months. Consider pulmonary rehab and breathing control. Use inhaled/oral corticosteroids only via shared decision-making with pulmonology. - Cardiovascular: At 12 weeks, obtain 12-lead ECG, 10-minute standing test, troponin; NT-proBNP if heart failure suspected. Abnormalities prompt cardiology referral. Persistent symptoms beyond 6 months warrant echocardiography, stress testing, Holter, tilt testing, CTCA, MPS, angiography, or CMR as appropriate. Treat chest discomfort with short NSAID course if pulmonary/cardiac evaluations are normal. For POTS/orthostatic intolerance: compression stockings, hydration, behavioral modification; pharmacotherapy (salt loading, rate-control, alpha agonists, mineralocorticoids) only with cardiology input. - Neurocognitive: Screen with MOCA. Severe/progressive deficits: neuropsychological assessment. Brain imaging only if focal deficits, mass concern, or per established algorithms. Assess sleep and mental health. - Fatigue/post-exertional malaise: Evaluate for systemic contributors (sleep disorders, endocrine/metabolic, nutritional, autoimmune, cardiac/respiratory diseases, anemia, medication effects). Screen with FAS or VAFS; use 6-minute walk or Timed Up-and-Go. Consider CK, herpesvirus panels, morning cortisol. Assess against ME/CFS criteria and manage accordingly. Self-management: rest, sleep hygiene, energy conservation techniques; structured, titrated return-to-activity programs, individualized for severe post-exertional malaise. - Olfactory/gustatory: Recommend olfactory training; address smoking cessation; assess for iron deficiency if anemic; exclude sinusitis/nasal polyps; consider nasal steroid spray; refer to ENT if refractory. Self-management and rehabilitation: - Encourage supported self-management, symptom diaries/apps, and participation in support groups. Multidisciplinary rehab teams should deliver individualized plans with shared decision-making, standardized monitoring, defined outcomes, referral pathways, and home-based options. Modify exercise programs for those with severe post-exertional malaise, exertional desaturation, orthostatic intolerance, or cardiac impairment. Pharmacologic prevention/treatment evidence: - Vaccination: Two-dose vaccination reduces Long COVID risk by 34%; vaccination after developing Long COVID may improve symptoms in ~20% of patients. - Reinfection increases Long COVID risk in a dose-dependent fashion; continue non-pharmacologic prevention. - Early antiviral therapy: Nirmatrelvir-ritonavir within 5 days of acute illness reduces Long COVID risk by 26%. - Metformin initiated within 4 days of symptom onset and titrated to 1500 mg/day over 2 weeks reduces Long COVID risk by 40%. - Post-discharge VTE prophylaxis: Rivaroxaban 10 mg daily for 35 days in high-risk recently hospitalized patients improves outcomes. Return to activity and work: - Use individualized, phased return-to-work plans based on occupational/functional assessments, with communication among clinicians, rehab teams, and employers. Cardiology review is required for those with myocarditis before return to intensive activity, especially athletes.

- Global and national burden: Approximately 100 million people globally and ~0.5 million in South Africa may already be living with Long COVID. - Prevalence estimates: Early pandemic estimates as high as 30% (mild/asymptomatic), ~50% (hospitalized), up to 75% (critical). With vaccination and Omicron predominance, overall prevalence is ~5%–10% post-infection across severities. - South Africa cohort: 46.7% of hospitalized and 18.5% of non-hospitalized developed Long COVID across Beta/Delta/Omicron; Omicron wave had markedly lower incidence (~18.5%) compared to Delta/Beta (~60%). - Risk factors: Severe acute COVID-19, reinfection, female sex, older age, atopy, multimorbidity, obesity, diabetes, HIV/immunosuppression. However, absolute burden is highest in middle-aged individuals without comorbidities who had mild/moderate acute disease. - Symptom course and recovery: Many symptoms recur or intensify around 1 month after infection; neurological/cognitive symptoms may increase over time; respiratory symptoms tend to improve slowly; GI symptoms often resolve by 12 months. Fewer than 15% report complete recovery at 1 year. - Phenotype prevalence: Fatigue/post-exertional malaise (60%–70%); dyspnoea/cough (up to 60%); persistent anosmia/dysgeusia (20%–30% at ~8 months); neuropsychiatric symptoms (40%–70%, >50% cognitive impairment); musculoskeletal/rheumatic (~40% at 6 months); cardiovascular symptoms (~10% of previously hospitalized) with increased MACE/arrhythmias up to 12 months; GI symptoms (10%–80%). - Mechanisms: Evidence supports roles for viral persistence, superantigen-mediated immune activation, microclots resistant to fibrinolysis, autoimmunity (anti-nuclear and ENA antibodies), dysbiosis with prolonged inflammation, herpesvirus reactivation (EBV/HHV-6), metabolic dysregulation (lipids, cytokines, metabolites), and neuroinflammation (elevated CCL11) affecting neurogenesis/myelination. - Elevated incidence of new diagnoses/events: Higher risks of ischemic heart disease, dysrhythmias, myocarditis/pericarditis, heart failure, cerebrovascular events, thromboembolism, neurological disorders (e.g., neuropathy, movement disorders), sensory abnormalities, cognitive disorders, mental health conditions, rheumatologic diseases, diabetes, CKD, and chronic liver disease. - Prevention/treatment signals: Two-dose vaccination reduces Long COVID risk by 34%; nirmatrelvir-ritonavir started within 5 days reduces risk by 26%; early metformin reduces risk by 40%; vaccination after onset may improve symptoms in ~20% of patients; post-discharge thromboprophylaxis with rivaroxaban 10 mg/day for 35 days improves outcomes in high-VTE-risk patients.

The paper integrates evolving evidence to frame Long COVID as a significant, heterogeneous post-viral syndrome with multi-system involvement and substantial functional impact. By delineating prevalence, risk factors, and symptom trajectories, it underscores the need for systematic follow-up and structured assessment in primary care, even among those with initially mild disease. The synthesis of proposed biological mechanisms (viral persistence, microclots, autoimmunity, dysbiosis, herpesvirus reactivation, metabolic and neuroinflammatory pathways) helps explain the diversity and temporal variability of clinical phenotypes, guiding a multi-pronged approach to evaluation and management. The proposed clinical pathway addresses critical gaps: early identification of red flags, judicious use of investigations, and symptom-specific management strategies spanning respiratory, cardiovascular, neurocognitive, fatigue, and sensory domains. Emphasis on self-management and multidisciplinary rehabilitation aligns with patient-centered care and the fluctuating nature of symptoms. Preventive and therapeutic signals—vaccination, early antiviral therapy, and metformin—offer actionable strategies to reduce Long COVID risk at the population level while disease-specific treatments remain under investigation. Overall, the framework supports primary care providers in triage, evaluation, and longitudinal management, aiming to reduce morbidity and facilitate safe return to activity and work.

Long COVID is a complex condition with incompletely elucidated pathogenesis, diverse and overlapping clinical phenotypes, and considerable intra- and inter-patient variability over time. A rational, holistic, patient-centered approach incorporating shared decision-making and multidisciplinary care is essential. Primary care clinicians should adopt structured assessment pathways to identify complications, manage symptoms, and coordinate rehabilitation. Preventive strategies, including vaccination, early antiviral therapy, and metformin during acute infection, show promise in reducing Long COVID risk. Continued research into mechanisms and targeted therapies is needed, alongside development of evidence-based guidelines and standardized care pathways.

- Definitions and diagnostics: No universally accepted clinical definition; current diagnosis relies on exclusion and clinical judgment. There is no validated diagnostic test for Long COVID, and current evidence does not support PCR/antigen/antibody testing for diagnosis in the post-acute phase. - Evidence base: Pathogenesis remains incompletely understood; many mechanistic insights are preliminary or indirect. There are limited randomized controlled trials for treatment; no approved pharmacotherapies specifically for established Long COVID to date. - Heterogeneity: Clinical phenotypes are diverse with temporal variability, complicating standardization of assessment and management. Estimates of prevalence and risk vary by variant, vaccination status, and study design. - Guidelines: While recommendations are provided, comprehensive evidence-based local guidelines are lacking; proposed algorithms are based on expert consensus and evolving literature rather than definitive trials. - Generalizability: Data drawn from varied settings, waves, and populations (including South African cohorts) may limit generalizability across regions and healthcare systems.