Management of patients with SARS-CoV-2 infections with focus on patients with chronic lung diseases (as of 10 January 2022)

As depicted in Table 1, in hospitalized patients with non-critical COVID-19, therapeutic anticoagulation with LMWH should be considered, except if they have an increased bleeding risk or other contraindications. There is moderate evidence for this after two recent studies showing that therapeutic anticoagulation with LMWH was superior to prophylactic anticoagulation [21, 22]. In both studies the bleeding risk with therapeutic anticoagulation was increased; however, this was not significant and/or outweighed by the beneficial effects of therapeutic LMWH. In the HEP-COVID study, only patients with d‑dimer above 4‑fold upper limit of normal were included [22]. The ATTACC study did not use such a strict inclusion criterion and found less convincing beneficial effects; however, the authors calculated that in 1000 hospitalized patients, therapeutic vs. prophylactic anticoagulation may save 40 lives on the expense of 7 major bleeding events [21]. In contrast, in critically ill COVID-19 patients, therapeutic anticoagulation was inferior to prophylactic anticoagulation due to a higher complication risk. This refers to two studies that found no superiority for therapeutic anticoagulation in terms of the primary endpoint but an increased bleeding rate [22, 23]. In critically ill patients it may be indicated to use half-therapeutic anticoagulation, particularly if patients present with a high venous thromboembolism (VTE) risk profile.

If hospitalized patients are discharged early, prophylactic anticoagulation should be continued in the outpatient setting to complete a total time of anticoagulation of minimum 7 days. Anticoagulation for outpatients with COVID-19 is not recommended according to the recent ERS guidelines [28].

As depicted in Table 2, antibiotic prophylaxis is not recommended because the rate of opportunistic infections in COVID-19 ARDS is low [29]; however, primary coinfections must be excluded at hospital admission in agreement with the S3 guidelines for community acquired pneumonia [30]. If antibiotic treatment is indicated, the same dose and duration as in patients without COVID-19 should be used. These recommendations have been adapted from the German S3 guidelines [25]. It is important to note that the rate of nosocomial infections in severe COVID-19 ARDS is high, particularly in immunocompromised patients [31] but even in these patients the S3 COVID-19 guidelines did not recommend antibacterial or antifungal prophylaxis but a high level of awareness and routine search for such superinfections.

The main evidence is depicted in Table 3. In an early retrospective study from Wuhan, the application of corticosteroids was associated with a better survival [34]; however, this was also associated with a younger age and higher levels of inflammatory markers. This led to a controversial discussion about the pros and cons of corticosteroids, based on mostly negative studies with corticosteroids in influenza pneumonia. In COVID-19 ARDS, several prospective studies were performed and finally a large randomized controlled open-label study came to the conclusion that 6 mg dexamethasone per day for up to 10 days, improves survival of patients with progressive disease, who need supplemental oxygen or non-invasive or invasive ventilation [35]. The German S3 guidelines [25] and the recent ERS guidelines [28] recommended this treatment in this indication and we do not disagree; however, there were rather negative effects in less severe COVID-19 ARDS and therefore cortisone treatment should not be given to outpatients and those COVID-19 patients who do not deteriorate or need supplemental oxygen.

Tocilizumab, an interleukin‑6 (IL-6) antagonist approved for rheumatoid arthritis, is of particular importance because it received conditional recommendations by the German S3 [25] and the ERS guidelines [28]. The ASP has evaluated the evidence and commented these guidelines (Flick et al. Positionspapier-der-OeGP-zu-aIL-6R-bei-COVID_final_Mai2021_IC.pdf [ogp.​at]). There were two high-quality randomized controlled blinded studies which did not suggest improved mortality with tocilizumab as compared to placebo [43, 44] and there was a subgroup of the RECOVERY study where 4116 patients were 2:1 randomized to tocilizumab and usual care. In this latter study, the mortality was in favor of tocilizumab (OR = 0.85), relatively homogeneously distributed over all subgroups. Also, several secondary endpoints were in favor of tocilizumab [36]. Of note, this was an open-label platform trial (not the highest quality standard) and patients were only eligible if CRP was ≥75 mg/L and if there was no uncontrolled bacterial or fungal infection. Based on these data, we agree with the S3 and the ERS guidelines to give a conditional recommendation for the use of tocilizumab (Table 3).

Anakinra, an interleukin‑1 (IL-1) antagonist approved for rheumatoid arthritis, is also approved for COVID-19 by the EMA. According to the Committee for Medicinal Products for Human Use (CHMP) review, a study involving 606 hospitalized adults with moderate or severe COVID-19 pneumonia who had soluble urokinase plasminogen activator receptor (suPAR) levels of at least 6 ng/ml showed that anakinra was effective at treating COVID-19 [45‐47]. These patients received anakinra or placebo in addition to standard of care. Standard of care for most patients included low or high-flow oxygen and dexamethasone, and some also received remdesivir. The study showed greater clinical symptom improvements in patients treated with anakinra plus standard of care compared with those who received placebo plus standard of care. Anakinra reduced the risk of a patient’s condition worsening to more severe disease or death during the 28-day study period compared with placebo. The treatment benefit of anakinra compared to placebo was supported by an increase in the number of patients who fully recovered and a reduction in the number of patients whose condition worsened to severe respiratory failure or death; however, as the suPAR biomarker is not widely available, this treatment concept is difficult to apply in clinical practice.

Baricitinib, a Janus kinase inhibitor approved for rheumatoid arthritis and atopic dermatitis, has been tested in two phase III randomized placebo-controlled trials. In the first trial with 1033 patients, there was a significant beneficial effect on recovery time [39], in the other trial with 1525 patients, there was a significant reduction of mortality with baricitinib as compared to placebo (HR 0.62, p = 0.005) [40]. The quality of both trials was high and beneficial effects appeared substantial as compared to the adverse effects. Application for approval was submitted at EMA. The CHMP is in the review process for the indication COVID-19 ARDS with need for low-flow or high-flow oxygen (Table 3).

There have been a large number of studies using other repurposed drugs to treat COVID-19. The WHO Solidarity Trial, a parallel group randomized controlled open-label trial, investigated remdesivir, lopinavir, interferon, and hydroxychloroquine and found no significant beneficial effects of any drug on mortality, initiation of mechanical ventilation or hospital duration [48]. There were multiple other studies on other drugs and vitamin D3, which are summarized in Table 3. All the respective recommendations of the German S3 guidelines have been negative [25].

Two major approaches have been applied to cause direct antiviral effects: small interfering molecules with virustatic properties like remdesivir, molnupiravir and PF-07321332/ritonavir on the one hand and serum from convalescent COVID-19 patients or monoclonal antibodies against the spike protein. Except for convalescent plasma, the other approaches have been EMA approved or they are in review. All successful studies have been performed in unvaccinated patients with initiation of therapy in the first days after start of symptoms (Table 4).

Remdesivir is applied as intravenous infusion for 3 days and received EMA approval after a study in hospitalized COVID-19 patients [49] and a study in non-hospitalized patients at risk of a severe disease [50]; however, as already mentioned, in the Solidarity Trial, remdesivir did not reduce the mortality rate [48]. Molnupiravir (Lagevrio™) is applied as 4 capsules (800 mg) BID for 5 days and showed a significant benefit in the combined endpoint, hospitalization or death at day 29 with 6.8% vs. 9.7% in the drug vs. placebo group. The most impressive result was the reduced death rate with 1 vs. 9 patients [51]. The EMA is currently evaluating the marketing authorization application; however, this medicine can already be used in the EU to treat COVID-19, after EMA’s CHMP completed its review. For Paxlovid™, a combination of PF-07321332 (nirmatrelvir) and ritonavir, EMA has started a rolling review process after a prospective study, showing a highly significant benefit in the combined endpoint of hospitalization or death at day 28, which was met by 1% vs. 7%. The drug comes in capsules and is taken BID for 5 days. It has been made available to the US population by the Biden administration and it can also be prescribed in Europe (positive CHMP assessment report and authorization expected). Convalescent plasma from patients with a recent SARS-CoV‑2 infection showed no effects in severe (late) COVID-19 ARDS. Specific antibodies against the SARS CoV‑2 spike protein showed significant beneficial effects; however, only in very early and mostly prehospital patients (preventive therapy). The combination of casirivimab and imdevimab (Ronapreve™) has been approved by the FDA as Regen-Cov™ and is now also approved by the EMA after positive studies [54]. Regdanvimab (Regkirona™) is also approved by the EMA. It was developed by the South Korean company Celltrion and is manufactured in Hungary. It was tested in a large prospective randomized study enrolling 1315 patients (80% European) and met its primary endpoint, rate of death, hospitalization or need of oxygen at day 28 with 3% vs. 11% in drug vs. placebo patients. It is easily available in Austria. It is given as single i.v. infusion of diluted drug at a dose of 40 mg/kg. Sotrovimab (Xevudy™) has also been approved by the EMA. It employed a combined primary endpoint of death or hospitalization after 28 days in a study with 1057 patients, where the endpoint was met by 1% vs. 6%. It is given intravenously for 5 days. Evusheld™, a combination of tixagevimab and cilgavimab, is currently in rolling review by the EMA and not yet available.

In conclusion, virustatic therapies are very attractive; however, they are not arguments against vaccination. From a practical point of view, oral drugs (molnupiravir and PF-07321332/ritonavir) are most important, because they are needed by non-hospitalized patients who have no or mild symptoms and have just been tested positive for COVID-19. All other drugs need intravenous infusions which is more difficult in the outpatient setting.

The German S3 guidelines [25] recommend that an oxygenation index (OI) <150 mm Hg represents a conditional indication for intubation and invasive ventilation and OI <100 mm Hg represents an urgent indication. We agree with this recommendation and provide an algorithm for the hospitalized patient (Fig. 4). The advantage of intubation is that by means of filters in the expiratory tube, there is no further contamination of the room with virus-laden aerosols; however, reports from China and the USA reported mortality rates of 60–80% for patients who were intubated and mechanically ventilated. The advantage of non-invasive ventilation is that the patient does not need deep sedation, which is associated with much lower risk of complications. The rationale to first try non-invasive ventilation before intubation is discussed in detail by Windisch et al. [73, 74]. The actual German data confirm the published mortality of 22% for patients with ventilatory support [25, 75]; however, the mortality depends very much on the setting in which mechanical ventilation is started (particularly outside the ICU), the clinical condition of the patient and the criteria for admission to the ICU. Finally, the indication for intubation is always based on clinical judgement (exhaustion, ventilatory drive, level of consciousness). It is very important to consider palliative care or terminal comfort care in individual patients, based on fragility, comorbidities or patients’ preferences. There have been positive experiences with inclusion of palliative care specialists in these difficult decisions [76].

In patients with progredient respiratory failure despite optimized mechanical ventilation, including prone position, it may be indicated to initiate ECMO. This avoids secondary damage to the lungs by high oxygen levels and high ventilatory forces; however, it is an unsolved question which patient is the right candidate for this treatment [77].

Even more challenging is the question who to consider for lung transplantation. There is a broad consensus that lung transplantation should only be considered in COVID-19 ARDS patients with irreversible lung damage, who fail to regenerate their native lungs despite several weeks of ECMO [78]. Criteria to select eligible ARDS patients for lung transplantation include: i) a negative virus status (virus culture or PCR), ii) no clinical/radiological improvement despite MV and/or ECMO support, iii) mono-organ failure, iv) absence of severe extrapulmonary comorbidities and v) a realistic potential for long-term recovery [79, 80]. The time period necessary to determine native lung recovery has been a moving target throughout the pandemic. Currently, a period of at least 6 weeks on ECMO is recommended before a patient can be listed for lung transplantation; however, a considerable number of COVID-19 ARDS patients develop recurrent bacterial superinfections/bacteremia, necrosis of large parts of the lungs, severe pulmonary or pleural hemorrhage or recurrent tension pneumothorax despite large bore drainage. All these clinical scenarios might warrant an earlier transplantation [81].

To date 22 COVID-19 ARDS patients have received a lung transplantation in Austria, nearly all had a favorable short-term outcome. The majority of these patients could be successfully discharged from hospital; however, data on long-term results are not yet available.

Klok et al. have published a post-COVID functional score (PCFS) that has been validated for post-COVID patients [137]. In grade 0, patients are free of symptoms, in grade 1 they have symptoms but can cope with all activities of daily life, in grade 2 they avoid some of these activities due to symptoms, in grade 3 they are unable to do some of these activities, and in grade 4 they are no longer able to cope with daily life and need regular care. This simple scale is suitable for both the initial assessment and the follow-up. For details see [137].

We recommend inviting patients with a PCFS ≥2 to undergo multidisciplinary diagnostics and eventually rehabilitation.

The WHO distinguishes 4 phases of rehabilitation:

Dyspnea on exertion and/or fatigue belong to the common complaints of long COVID patients. There are already some studies showing safety and remarkable efficacy of cardiopulmonary rehabilitation in this indication [138‐140], although randomized controlled studies are missing. Before starting a rehabilitation program, extensive diagnostics are necessary to exclude coronary heart disease, myocarditis, pulmonary embolism, and disturbances of lung function or gas exchange [141]. All these diseases necessitate special diagnostics and treatment and rehabilitation is secondary to that.

Patients with PCFS ≥2 who are able and interested, should be offered ambulatory rehabilitation to reconstitute their functional limitations.

Ambulatory rehabilitation includes:

If ambulatory therapy is not sufficient, multimodal or inpatient rehabilitation should be considered. This agrees with the German AWMF guidelines [142].

If the pneumologic, neurologic and cardiologic limitations are predominant, a specific rehabilitation should be prescribed.

For neurologic, psychotherapeutic, psychiatric and logopedic rehabilitation we refer to the recent S‑1 guidelines [5].