COVID-19

Let's face it, there's no way you could producing an ICU exam paper in the current critical care environment without making some reference to the world-eating pandemic. When should we expect this? The CICM exam papers do tend to lag approximately eighteen months behind world events, which makes it surprising that vaccine induced immune thrombotic thrombocytopenia made it into the second paper of 2021 (about six to nine months too early, some would have said). In any case the college seem to have abandoned their practice of publishing past exam papers, leaving their trainees without even that surrogate curriculum. 

This, unfortunately, means that even when COVID 19 questions do appear in the exam, we may never find out exactly what they were, except from half-remember recollections of traumatised exam candidates. Those undergoing exam preparation are presently without a rudder.  It is for this group that this brief summary exists. Where possible, or wherever important, distinctions are drawn between the early access alpha variant and the subsequent DLC. Wherever possible, the most current IDSA guidelines were used to form this summary.

Natural history

• Day -5: Exposure (important to recall that the Delta variant is 40-60% more transmissible, owing to the 1000-fold increase in viral load)
• Day 0: Onset of symptoms (the vast majority of people will have no symptoms, but those who do, will experience them within 11.5 days, according to Lauer et al, 2020). 
• Day 2-5 after onset of symptoms: Radiological changes develop. For patients whose symptoms were severe enough to bring them to hospital, 63% have some CXR changes by days 0-2, and 72% have changes by day 3-5 (Vancheri et al, 2020). For symptomatic patients, day 3-5 also seems to be the timing of hospital admission.
• Days 5-10: rapid clinical deterioration may occur. The time course of this disease does not have a polite stepwise progression- the change from moderate disease to severe or critical can be rather rapid. This is a change oten associated with a "cytokine storm" and the development of an acute lung injury which was seen around day 8 in this very early Wuhan case series (Feb 2020). A rising CRP is usually seen in association with this, though the precise cut-off is not well-established. The onset of dyspnoea 6.5 day after the onset of symptoms can be followed by ARDS in as little as 2.5 days 
• Day 13-15 after the onset of symptoms: Radiological features are typically maximal by this stage. Yun et al (2021) found that CT evidence of pulmonary damage was maximal on day 13 for the moderate and day 15 for the severe cases. 
• ICU stay was quite prolonged in some of the earlier studies, and it remains to be seen what it will be in this era of effective therapies. 

Definition of disease severity

• Mild illness: minimally symptomatic. Not requiring oxygen. 80% of patients fall into this category.
• Moderate illness: defined by the requirement of oxygen, which is between 0 and 4 L/min. Symptoms are more severe and are described as "prostration", fever, radiological signs etc, but there are no "laboratory indicators" of severe disease, meaning a raised CRP. These are the people that get dexamethasone and remedesivir, but who are otherwise well enough to remain on a general medical ward.
• Severe illness: Defined by a resp rate over 30, PF ratio of less than 300, and deteriorating oxygenation (sats less than 92% on the aforementioned 4L/min). These are patients who would usually need to come to the ICU and get humidified high flow oxygen or NIV.
• Critical illness: This most severe spectrum includes patients with any additional organ failure, and with evolving respiratory failure (PF ratio less than 200, or those who deteriorate on non-invasive support and require intubation). In many of the earlier studies of the Alpha strain, the mortality was virtually confined to this severity group. For instance, Chinese data (Wu et al, 2020) had a total case mortality rate of 2.3%, but 49% among the critically ill group. 

• The effect of Delta variant infection on the severity of disease is still being worked out, but this Lancet observational study of mostly unvaccinated patients (2021) suggests that being infected with Delta strain makes it rather more likely that you will present to hospital, with 5.7% vs 4.2% attending "emergency care" within 14 days of positive results.​

Diagnosis

From the IDSA guidelines, each of which is linked:

• Rapid antigen testing (Hanson et al, 2024)
  • Lateral flow assay (Koczula & Gallotta 2016) - a low cost portable device
  • Looks for characteristic COVID nucleocapsid protein
  • Specificity is 99%, sensitivity 81%-89% in symptomatic individuals within the first 5 days of illness, and ~ 54% after 5 days
  • First line where there is no time to wait for the result and /or PCR is not available, i.e.  better than not testing
• PCR (Hanson et al, 2024)
  • Nucleic acid amplification testing (NAAT), or rapid reverse transcriptase–polymerase chain reaction (RT-PCR)
  • Looks for COVID RNA; specific genetic material (eg. the Cobas automated assay platform uses O‐gene, a very COVID-specific nonstructural region, and "E‐gene", which is an envelope region common to all Sarbecovirus subgenus)
  • Specificity 98%, sensitivity ~90%  (thus, repeat testing has low yield)
  • First line test, preferred over RAT
  • Can remain positive well into the recovery phase, as viral RNA may still be present in the absence of viable virus
• Cycle threshold (Ct value)
  • The RT-PCT test reverse-transcribes the viral RNA into complimentary DNA, which is then unravelled into single strands using heat, cooled in the presence of polymerase to replicate multiple copies, and then heated again to denature it into single strands, and so on. 
  • Each such cooling/heating cycle roughly doubles the number of specific DNA fragments being amplified
  • At a point, after a certain number of cycles, the number of the replicated molecules reaches the threshold value, beyond which the PCR test is considered positive. Or, it continues cycling, and the lab tech declares the result negative when a certain number of cycles has occurred without the threshold value being reached (eg. 40).
  • Thus, a low cycle time suggests a larger initial dose of viral RNA, which suggests active infection, as Ct values were empirically found to be inversely proportional to the viral load (Aranha et al, 2021)
  • It is still considered "a nonstandardized instrument signal value" and is not recommended by the ISDA (Harris 2024)
  • In immunocompromised patients, it is probably better to use this value instead of some arbitrary timeframe to deisolate patients, as immunosuppressed patients (eg. the haematology population from the study by Harris et al, 2024) tend to clear the viral more slowly and remain infectious for longer.
• Viral culture is the laborious and expensive gold standard which everybody is trying to avoid by doing all the methods listed above. It works on the principle that there is no better proof of infectivity than infecting something with your sample, the "something" in this case usually being Vero E6 cells -  a clone of African green monkey renal epithelial cells which almost everyone ended up using for COVID (Pires De Souza et al, 2022). If the sample contains live virus, the cells will become infected and begin to demonstrate detectable changes. This is not quick - perhaps 4-5 days - but the certainty it affords for the calibration of other tests has allowed people to make bedside decisions on the basis of empirical timeframes and Ct values:
  • "Although SARS-CoV-2 RNA identification in respiratory (up to 83 days) and stool (35 days) can be prolonged, duration of viable virus is relatively short-lived (up to a maximum of 8 days from symptom onset)" - Jefferson et al (2021)
• Serology: This measures the humoral immune response to current or past infection. Antibodies can be seen around 2 weeks after the onset of symptoms, which means they are not particularly good for the immediate diagnosis of acute illness.

Biomarkers

It is the custom of CICM Second Part Examiners to ask their candidates to "critically evaluate" something that's getting dragged around in the recent literature, and the use of biomarkers to predict the course of COVID19 pneumonia is definitely a ripe low-hanging fruit for that sort of SAQ. The best overview of this topic is probably Weidmann et al (2021)

Rationale

• CVOVID-19 is characterised by an exaggerated inflammatory response, which is held responsible for the most serious manifestations of that disease
• Secondary and surrogate inflammatory markers (CRP, procalcitonin, ferritin) and directly  measured cytokine levels (IL-6) are increased in the course of the inflammatory response, and should correlate with its severity
• Newer disease-modifying agents target various steps in this inflammatory response
• Ergo, there may be merit in tracking inflammatory marker changes to assess disease progression , predict deterioration and define patients who would derive maximum benefit from disease-modifying agents
• Also, biomarkers of (for example) cardiac damage can predict patients who go on to develop myocarditis

Advantages

• CRP, procalcitonin and ferritin are ubiquitous biomarkers which can be measured in most laboratories
• CRP is an inexpensive test
• IL-6 is dramatically elevated in COVID-19, and correlates with severity (Liu et al, 2019), which means it could be used to guide decisions about targeted anti-IL-6 therapy. CRP synthesis is stimulated by IL-6.
• Myocardial damage which might not be clinically apparent from patient haemodynamic performance or ECG changes might be revealed by serial troponins

Disadvantages

• CRP is a nonspecific biomarker of tissue injury and inflammation, which means in ICU the lower of the suggested cut-off values may lose their specificity (i.e. anything could be causing that CRP rise)
• Data regarding the use of CRP and other biomarkers is often derived from patient populations which were not treated with modern disease-modifying agents
• Disease modifying agents, especially those targeting the IL-6 pathway, may change the IL-6 and CRP response to COVID-19 rendering these biomarkers less predictive (i.e if you have received tociliozumab, your CRP will drop, but this does not mean your disease is improving)
• IL-6 testing is not available everywhere
• Myocardial damage which does not produce haemodynamic instability or arrhythmias, if it were detected by the troponins alone,  would also not attract any additional management, as there is no evidence-based therapy for COVID myocarditis. 
• Management protocols are not guided by inflammatory marker parameters but by patient clinical status

Evidence 

• CRP  can be a predictor of clinical deterioration with a cut-off of 18 mg/L (Yitao et al, 2021) and progression to severe disease with a cut-off of 26 mg/L (Wang et al, 2020)
• CRP can also identify patients at a higher risk of death, with a cut-off of 40 mg/L (Stringer et al, 2021)
• CRP levels over 150mg/L were found to be associated with an improved response to tocilizumab (i.e. patients with very high CRP who received tocilizumab were observed to have improved mortality upon post-hoc analysis of CORIMUNO-TOCI data by Mariette et al, 2021)
• At this point, there are no protocolised strategies to escalate or deescalate therapy on the basis of biomarker changes in COVID-19

Pharmacotherapy

What could CICM ask from their candidates here? The territory is murky. The least unfair question could be something like "Outline the evidence for the use of disease-modifying agents in severe COVID-19 pneumonia", because this would call for an overview of the agents and some brief statement regarding their relative efficacy and the evidence published in support. Thus:

Remedesivir:

• Indicated for patients who are hypoxic, but not requiring oxygen
• Dose is 200mg loading and then 100mg/daily IV, for a total of 5 days
• The evidence does not support the use of this drug in patients with worsening respiratory failure (eg. ventilated for more than 48 hours or requiring ECMO); maximum benefit seems to be in the moderate illness group
• Contraindicated in patients with renal failure or developing MOSF

Dexamethasone:

• Indicated  for patients who are hypoxic and requiring oxygen 
• Dose is 6mg daily, for up to 10 days; and it continues even if (especially if!) the patient becomes critically ill
• This has been the mainstay of treatment ever since Villar et al (2020) demonstrated a 15.3% reduction in absolute mortality in critically ill COVID-19 patients (from ~36% to ~21%)

Baricitinib, a Janus kinase (JAK) 1 and 2 inhibitor, inhibits the effect of IL-6 on the STAT3 pathway, thereby interrupting the downstream effects of the IL-6 mediated inflammatory cascade, including cytokine secretion, immunoglobulin production, and macrophage activation (especially metalloproteinase secretion which is thought to be responsible for a lot of the late lung damage). Oh, and it might inhibit the endocytosis of virus. 

• Indicated for patients who are requiring oxygen and who have deteriorated to the point of requiring ICU-level respiratory support
• Dose is 4 mg po daily or up to 14 days or until hospital discharge
• Benefit is substantial - Marconi et al (2021) found an absolute mortality improvement of 11.5% (17.5% vs 29.4%) when they added baricitinib to standard care, with the "severe" cohort most likely to benefit (HFNP, NIV, that sort of thing)
• Major complications are various cytopenias, thrombocytosis,  increased risk of VTE and increased risk of infection
• Major contraindication is end-stage renal failure (but eGFR 15-30ml is still ok with dose reduction)

Tocilizumab and Sarilumab,  IL-6 receptor blockers:

• Indicated for patients who have deteriorated to the point of requiring intubation
• Dose is 8mg/kg for adults, as a single dose (but with the option of a second dose if there is no clinical improvement). As it comes in very expensive 200mg vials, practically this means 600mg for most normal-sized patients (66-90kg). 
• Some evidence of improvement in ICU patients requiring organ support (Gordon et al, 2021, specifically identified the sickest group of patients as the ones most likely to benefit). The greatest benefit is thought to be within the first 24 hours of starting ICU-level organ support.
• Most will agree with the IDSA Guidelines (2021), that those receiving baricitinib should not go on to receive tocilizumab, which means inpatients who deteriorate rapidly and bypass the baracitinib stage (or those admitted in extremis via the emergency department) are the main target population.

Sotrovimab, a monoclonal antibody with the viral spike protein as the molecular target

• Early limited data from Gupta et al (2021) suggests that it reduces the risk of hospitalization by 85% (vs. placebo)
• Its role is though to be in the prevention of disease progression in at-risk individuals (older, with more comorbidities)

In summary:

• Oxygen dependent patients receive dexamethasone 6mg and remedesivir
• When they graduate to HFNP or NIV, they should get baricitinib
• If they get intubated, and they were already on barcitinib, they should finish that course.
• If they get intubated and they had not had any baricitinib, they should get tocilizumab or sarilumab instead

Outcomes

• The effect of age: people usually quote the exponential increase in mortality from Levin et al (2020), who found the following rates of death from "resource-rich" countries:
  • Age 10: 0.002% mortality
  • Age 25: 0.01% mortality
  • Age 55: 0.4% mortality
  • Age 65: 1.4% mortality
  • Age 75: 4.6% mortality
  • Age 85: 15% mortality, 
  • Age 90 and above: over 25% mortality

Vaccine induced immune thrombotic thrombocytopenia (VITT)

Most of this has come from the THANZ advisory statement:

Definition:

• An autoimmune-mediated thrombocytopenia caused by autoantibodies against platelet factor 4 (PF4), which is the same antigenic target as HIT except without the attached heparin.

Diagnostic criteria (all five need to be met):

• AstraZeneca vaccine 4 to 42 days prior to symptom onset
• Any venous or arterial thrombosis
• Thrombocytopenia (platelet count < 150 x 109/L)
• Positive PF4 “HIT” (heparin-induced thrombocytopenia) ELISA
• Markedly elevated D-dimer (more than 4 times upper limit of normal)

Screening:

• Suspicion is raised by:
  • AstraZeneca vaccine 4 to 42 days prior to symptom onset
  • Thrombocytopenia (platelet count < 150 x 109/L)
  • Low fibrinogen OR extremely high D-dimers (five times the upper limit of normal)
• If the imaging confirms venous or arterial thrombosis, VITT is probable and specific andti-VITT ttherapy should commence
• If imaging reveals no thrombosis whatsoever, VITT is still possible, and non-heparin anticoagulation therapy should commence

Confirmation:

• The standard HIT screen PF4 ELISA seems to detect the VITT antibody
• Functional antibody testing should also be carried out, but this usually requires a haemostasis expert and a well-outfitted lab, which means that treatment should be started early.

Important points for early management:

• Do not administer platelet transfusion
• Do not commence heparin-based anticoagulation

Specific management:

• IVIG g/kg daily for 2 days
• Corticosteroids (THANZ recommend methylpred)
• Plasma exchange

Non-heparin anticoagulation, just like HITS, options for which include: 

• Parenteral agents: fondaparinux, bivalirudin, argatroban, danaparoid
  • Bivalirudin or argatroban are the preferred agents, owing to their short duration of action, and theTHANZ guideline recommends these for PE and CVST.
• Oral agents: rivaroxaban, apixaban, dabigartan
• "duration should probably be time limited (3-6 months)" but nobody is entirely sure what duration of anticoagulation to recommend, and resolution of the antibodies and clot burden on imaging will probably be the most effective guide for this. In HIT, antibodies can still be detected after 3 months.