We performed a systematic review of knee ON cases after COVID-19 by collecting the main demographic, clinical and imaging findings. We have seen that post COVID-19 knee ON is a real complication, that can occur even in the absence of the use of CCS. However, it seems that low doses of CCS are able to trigger necrosis and that the latter appears prematurely compared with the literature data; this could be due to the COVID-effect on bone and vessels. SARS-CoV-1 and SARS-CoV-2 presented similarities regarding ON presentation. Cases of ON had been reported during SARS-CoV-1 infection [14, 15]. The aetiology of ON is not yet clear, but alterations of the microcirculation may be a trigger[16]. Both SARS-CoV-1 and SARS-CoV-2 enter cells via the angiotensin-converting enzyme 2 (ACE2) receptor which triggers the inflammatory response. ACE2 receptor is present in several tissues, including the musculoskeletal system for example smooth muscle, synovial tissue, cartilage and blood vessels [17]. The virus damages the endothelial pathway particularly represented at the vascular level and triggers an important inflammatory response, with leucocytic activation and aggregation [18, 19]. This strong inflammation of the vessels could predispose to thrombosis and both macro and micro infarctions. A recent study reported significant vascular alteration in lower limbs after Covid-19 [20]. The hypoxia and inflammatory response triggered by the virus activates several cytokines including CXCL10, IL-17, and TNF-a, which cause inhibition of osteoblasts and activation of osteoclasts initiating processes similar to those of osteoporosis [21–26]. CCS are often used to treat the most severe forms of Covid-19 to reduce the strong systemic inflammatory response [27–29]. Patients who had higher or longer doses of corticosteroids had an elevated risk of developing osteonecrosis [30, 31]. Post COVID-19 osteonecrosis has been already reported. Bagaria et al. [32] reported several sequelae after covid-19 including two cases of spontaneous osteonecrosis. Shetty et al. [33] reported cases of femoral head necrosis with and without the use of CCS. When CCS were used, ON had an early onset as usually reported [34]. Thankappan et al. [35] described an unusual case of atraumatic osteonecrosis of the proximal humerus metaphysis. The metaphyseal area is richly vascularized and necrosis in this area is quite rare. Similarly, Ghosh et al. [36] reported one rare case of vertebral bone marrow necrosis 1-month after COVID-19. The authors associated the complication primarily with the inflammatory and prothrombotic response associated with COVID-19. The post-mortem analysis did not isolate the virus in the knee joint environment in positive patients [37]. We described six cases of knee ON after COVID-19. Thannheimer et al. [38] presented a case of knee ON that occurred about one month after the diagnosis of COVID-19. The patient was seen for the first time in September 2020 with a clinical and radiographic picture of a medial meniscus posterior horn tear in the absence of bone edema, subchondral insufficiency fracture of the knee (SIFK), or ON. However, the chronological relationship between the onset of symptoms and the execution of the MRI is not documented. Therefore, we cannot exclude that the MRI was performed in the “window period”, as described by Nakamura [39], and that the osteonecrosis process had already begun, although not visible on the MR. In October 2020, the arthroscopy confirmed the meniscus tear, so a medial meniscectomy was performed. Sixteen days after the procedure the patient was positive for SARS-CoV-2. Several authors have previously hypothesized that the arthroscopic procedure could cause osteonecrosis, and called it post-operative or post arthroscopy ON [40, 41]. In the author's case report, the two possible causes of osteonecrosis were meniscal tear or arthroscopy. Malinowski et al. [11] found some distinctive features on the MRI of two patients with knee osteonecrosis after COVID-19. On the MRI, the bone edema was diffuse in the whole femoral condyle (mostly in STIR), soft tissue edema and there was no subchondral bone thickening, in contrast to classic SONK. Nonoperative treatment was successful in a few weeks. The authors referred to this form of ON as transient post-COVID-19 spontaneous osteonecrosis of the knee (PCT-SONK). Angulo-Ardoy et al. [12] described a case of post COVID-19 ON. However, the patient was affected by knee osteoarthritis and meniscal injury (severity not specified) and had performed knee surgery the previous year (not better specified). He had also taken CCS, Lopinavir and Ritonavir. He also performed three negative PCR on the nasopharyngeal tract. The diagnosis of COVID-19 was made only on the basis of the chest X-Ray. In this case, there are several risk factors for osteonecrosis, in particular drug therapy. The patients described by Agarwal et al. [13] presented a typical picture of avascular necrosis, probably associated with the use of CCS. The authors argued that COVID-19 makes one susceptible to necrosis, given that in their case avascular necrosis occurred with a low dose of corticosteroids and with an early onset. Our review showed that osteonecrosis arose at an average of 8 weeks after Covid infection and with an average dose of 471mg of prednisolone, thus prematurely and at lower doses compared to what is usually documented in the literature [42–45]. A cumulative MPS-equivalent dose of < 5000 mg for less than 10 days is reported as safe to prevent ON after Covid-19 but it may not be sufficient in the presence of other risk factors [46]. Shetty [33] proposed a set of recommendations for prevention, early diagnosis and treatment of ON in COVID‑19. In the authors’ opinion, is not possible to determine if COVID-19 infection is the cause of a further form of knee osteonecrosis with its own characteristics. The cause of ON is more likely to be multifactorial. Therefore, the SARS-CoV-2 infection may play a synergistic effect with other risk factors, such as the use of antiviral drugs and CCS, a predisposition of the patient, other risk factors such as previous traumas, previous surgery or coexisting intra-articular pathologies or comorbidities related to the patient.
Limitations
As limitations of our study, we analyzed case reports and case series with a low level of evidence. Secondly, the total number of cases of osteonecrosis in the literature is too few to understand the role of COVID‑19 in the cause of knee ON. In fact, it remains unclear to demonstrate if the post COVID‑19 osteonecrosis has specific symptoms or MRI features. Moreover, most patients with post COVID‑19 ON used CCS. This makes it even more difficult to isolate the role of COVID‑19 in the etiopathogenesis of the disease.