Non-respiratory presentations of COVID-19, a clinical review

Patients with significant lab derangements had more severe disease and a greater need for critical care [,,,,, , , ]. Lymphopenia is often more pronounced in those requiring ICU level care [, , , , ]. In Singapore, Fan et al. noted that 28% of patients infected with COVID-19 had lymphopenia, with critical care patients demonstrating significantly decreased lymphocytes compared with those not requiring ICU treatment []. Furthermore, 4 of the 9 ICU patients also exhibited increased levels of LDH (median value of 1684 U/L) compared to the non-ICU patients where only 5 of 26 had moderately elevated LDH []. Huang et al. demonstrated that in addition to marked lymphopenia, patients needing intensive care had higher levels of d-dimer (level 2.4 mg/L [0.6–14.4]) on admission than those who did not require it (0.5 mg/L [0.3–0.8], p = 0.0042) and also higher prothrombin times 12.2 s [IQR 11.2–13.4] when compared with non-ICU patients (10.7 s [9.8–12.1], p = 0·012) []. In a study of 140 COVID-19 positive patients by Zhang et al., 58 patients who were considered to have “severe disease” demonstrated a 2-fold increase in d-dimer compared to those with mild disease []. Guan et al. also determined that a significant elevation in d-dimer was more pronounced in severe cases (59.6% vs. 43.2% in non-severe) [].

A meta-analysis examining the role of laboratory abnormalities in patients with severe COVID-19 vs. those with milder disease determined that the most predictive parameters of critical infection were lymphopenia (96.1% vs. 80.4%), thrombocytopenia (57.7% vs. 31.6%), increased LDH (58.1% vs. 37.2%) and elevated d-dimer (59.6% vs. 43.2%) []. Specifically, in Wu et al.'s retrospective analysis of the risk factors for disease progression to ARDS, they observed a statistically significant association between lymphopenia and the development of ARDS (P < 0.001) []. Several studies also established that higher CRP levels correlated with worse outcomes in COVID-19 such as ARDS, myocardial injury, and death [,,]. As such, tracking hematologic parameters is crucial in determining prognosis and management, particularly with regard to level of care and monitoring.

In addition to indicating the potential for more severe disease, laboratory abnormalities are also a predictor of mortality in infected patients [,]. In a case series of 138 patients with COVID-19, the mortality rate was about 4.3% (6 patients). Of those 6, 5 patients had persistent lab derangements including increased d-dimer and decreased lymphocyte counts []. Another retrospective cohort showed that elevated d-dimer was associated with a higher rate of in-hospital death with 81% of terminal cases exhibiting d-dimer >1 μg []. Tang et al. analyzed abnormal coagulation parameters in patients with SARS-CoV-2 pneumonia and determined that non-survivors had significantly higher d-dimer, fibrin degradation products (FDP), and prothrombin time on admission compared to survivors (P < 0.05) []. This study was further supported by Han et al. who also demonstrated that d-dimer and FDP were especially predictive of disease progression []. A recent report investigating the factors affecting 28-day mortality of patients with severe illness showed that elevated d-dimer, increased age, and prolonged PT were associated with a higher mortality []. Early recognition of these abnormal results will play a critical role in predicting disease severity and improving outcomes with earlier intervention and supportive therapy.

Coagulation parameters are often cited as indicators for worse prognosis in patients infected with COVID-19 [,,]. Compared to a healthy control, d-dimer, FDP, and fibrinogen levels are all increased in COVID-19 patients, while antithrombin (AT) levels are significantly reduced []. Emerging data suggests that deregulated thrombin generation and abnormal activation of the coagulation cascade can lead to the development of disseminated intravascular coagulation (DIC) and is associated with worsening pneumonia and mortality [,, , ]. DIC was a significantly more common finding in non-survivors (71.4%) vs. survivors (0.6%) []. Other manifestations of coagulopathy such as the development of antiphospholipid antibodies and subsequent thrombotic events were reported in 3 ICU patients infected with SARS-CoV-2 in Wuhan, China []. Due to coagulation abnormalities, COVID-19 patients are at a higher risk of VTE, especially those with pre-existing comorbidities []. Early monitoring of these parameters can help guide medical management such as the use of VTE prophylaxis and escalation of care.

A study of COVID-19 patients from the First Affiliated Hospital of Zhengshou University, Henan, China demonstrated that infected patients have an overall pro-thrombotic state due to platelet hyperactivity []. When SARS-CoV-2 and its spike protein bind directly to the platelet and ACE2 receptor, there is increased platelet activation and thrombus formation due to activation of the MAPK pathway. This facilitates the release of coagulation and inflammatory factors, leading to an overall pro-thrombotic and pro-inflammatory state []. The authors also suggest that the addition of ACE2 protein and anti-Spike neutralizing antibodies may be a therapeutic approach to avoid thrombotic events in these patients.

Myocardial injury is defined as an elevation in biomarkers such as cardiac troponin I. Acute myocardial injury was reported in some of the earliest cases of COVID-19 in Wuhan, China. An early 2019 study among 41 admitted hospital patients in Wuhan reported acute cardiac injury in 12% of the patients []. Furthermore, another study of 138 hospitalized patients in Wuhan reported that as high as 22% of those who required ICU care experienced acute myocardial injury []. Another retrospective, observational study of 52 critically ill patients in Wuhan, China reported cardiac injury in 23% of the patients []. The increased prevalence of cardiac injury among patients with COVID-19 could be explained by the significantly higher levels of cardiac troponin I in severely ill patients [].

A recent multi-hospital retrospective cohort of nearly 3000 patients demonstrated that myocardial injury was common among hospitalized COVID-19 patients (n = 985, 36%), and that those with a history of cardiovascular disease (CVD) were more likely to experience myocardial injury than those without []. Similarly, A study of 44,672 patients with COVID-19 demonstrated that a history of CVD was associated with a five-fold increase in case fatality rate compared to those without CVD (10.5% VS. 2.3%) [,].

A study by Lala, et al., showed that mild myocardial injury as evidenced by small increases in troponin was significantly associated with death (adjusted hazard ratio: 1.75; 95% CI: 1.37 to 2.24; p < 0.001), and that greater elevations correlated with a higher risk of mortality (adjusted HR: 3.03; 95% CI: 2.42 to 3.80; p < 0.001) []. Similarly, a single center retrospective study of 50 COVID-19 ICU patients in Turkey demonstrated that cardiac biomarkers including troponin I and NT-proBNP were higher in non-survivors compared to survivors [].

Myocarditis has also contributed to mortality associated with COVID-19. In a case series of 150 patients, researchers determined that 7% died from myocarditis with circulatory failure []. Clinically, diagnosing myocarditis can be challenging, especially when differentiating it from acute coronary syndrome. For this reason, echocardiogram evaluation is recommended, and myocarditis associated with COVID-19 will appear as global wall motion dysfunction without focal wall motion defects [,,]. Additionally, ECG abnormalities may occur as a result of myocardial inflammation, such as T wave inversion, PR and ST segment deviations []. Autopsy reports of COVID-19 patients have reported high viral loads, mononuclear cells, and lymphocytic infiltration as key players in mediating acute myocarditis [, , ].

A wide variety of arrythmias have been observed in patients with COVID-19. In a study of 137 patients in Hubei Province, 7.3% of patients reported heart palpitations []. In another report of 138 hospitalized patients, cardiac arrythmia was noted in 16.7%, and these arrythmias were more common in ICU patients []. In a much larger study of 700 patients, 9 cardiac arrests, 25 atrial fibrillation events, 9 clinically significant bradyarrhythmias, and 10 non-sustained ventricular tachycardias were reported []. Additionally, this study noted that ICU admissions were associated with incidents of atrial fibrillation among COVID-19 patients, indicating that these arrythmias can contribute to severity of illness []. Lastly, a retrospective case series published in Cureus discussed COVID-19 related prolonged QTc and transient bradycardia as manifestations of the illness [].

Many etiologies have been proposed to explain the occurrence of arrythmias among COVID-19 patients. Hypoxia, inflammatory damage, abnormal metabolism, neurohormonal stress, and response to medications are some of the likely etiologies [,,]. Furthermore, the effect of pro-inflammatory cytokines on the sinoatrial node could lead to the development of bradycardia in patients with COVID-19 [].

In a retrospective cohort study of COVID-19 patients in Wuhan, China, Zhou et al. reported heart failure in 23% of admitted patients []. This number was as high as 52% in those who did not survive the infection []. Another retrospective study demonstrated that heart failure and acute cardiac injury were more common in deceased patients, regardless of their cardiovascular history []. Specifically, this study reported heart failure in 24% of deceased patients, nearly half of whom did not have any history of cardiovascular disease or hypertension []. It is unclear whether the heart failure was due to a new cardiomyopathy or an exacerbation of existing undiagnosed left ventricular dysfunction, and the etiology is still being investigated [,]. Additionally, right heart failure can occur in patients with severe lung injury and ARDS [,]. As such, physicians should be cautious of underlying cardiac dysfunction when administering IV fluids [,].

Several recent case reports have also highlighted the development of Takotsubo Syndrome (TTS) in those infected with COVID-19 [, , , ]. One report detailed the case of a patient who developed cardiogenic shock 16 days after infection despite normal initial troponin levels and LVEF []. Her bedside echo demonstrated the apical ballooning typical of TTS. Similarly, another case report discussed the case of a previously healthy 50 years old male who developed chest pain and signs of cardiogenic shock 8 days after the onset of symptoms. His echocardiogram showed akinesia of all of his basal segments, and he was ultimately diagnosed with inverted Takotsubo cardiomyopathy [].

Some studies have shown that GI symptoms can manifest before the onset of typical respiratory symptoms [,]. The first confirmed case of COVID-19 in the US presented with a two-day history of nausea without respiratory complaint. Days into the illness, the patient also complained of abdominal discomfort and loose stools []. Similarly, a recent report described a COVID-19 patient presenting with diarrhea, borborygmus, anorexia, and nausea in the absence of any respiratory symptoms []. Early research has suggested a correlation between GI symptoms and severity of illness. Henry et al. performed a pooled analysis of 10 different studies with a total sample of 1989 COVID-19 patients, 598 of whom (30.1%) were deemed as having “severe disease.” This study examined whether patients presenting with GI symptoms could be at an increased risk of critical illness and poor prognosis. The research highlighted a significant association between abdominal pain and illness severity. Furthermore, nausea and vomiting correlated with a marginally increased risk of severe COVID-19, while diarrhea was not reported to be associated with worse disease []. Another retrospective case-controlled study of 278 COVID-19 positive patients and 238 COVID-19 negative patients suggested that patients presenting with GI symptoms at time of testing were more likely to test positive for the virus []. In comparison, patients without GI symptoms were equally likely to test positive or negative for COVID-19 []. Lastly, in the past few months, two cases of paralytic ileus were reported among COVID-19 patients []. Histopathology of resected bowel specimen in these cases suggests a role for COVID-19-induced micro-thrombosis leading to GI perforation [].

In addition to research showing the potential correlation between GI symptoms and disease severity, several studies have determined that infected patients can shed viral particles in their stool [, , ]. One study of 42 COVID-19 positive patients demonstrated that about 67% of the patients had viral RNA present in their stool even in the absence of diarrhea or other GI symptoms. Interestingly, among this group, 64% of the patients continued to shed viral particles in fecal specimens even after the nasopharyngeal swabs turned negative []. This has been documented in both adult and pediatric patients. Xu et al. conducted an epidemiological and clinical study of 10 children with COVID-19 and found that 8 persistently tested positive in stool despite their nasopharyngeal swabs being negative [].

Fecal specimen testing is just as accurate in detecting COVID-19 as nasopharyngeal swabs [,]. Physiologically, the presence of viral particles in feces is plausible as there is a high level of viral receptor angiotensin converting enzyme 2 (ACE2) in the gastrointestinal tract []. While the current research is not definitive, studies have indicated that asymptomatic patients may shed COVID-19 viral particles in their stool [,, , , ]. Whether these particles are infectious and support the argument for possible the fecal-oral transmission of SARS-CoV-2 remains unclear.

Several studies have examined whether GI procedures such as endoscopies can safely be performed on COVID-19 positive patients. A study conducted in northern Italy identified 23 COVID-19 patients presenting with signs of upper GI bleeding, therefore necessitating urgent endoscopy []. The virus has been detected in biopsies of the esophagus, stomach, duodenum, and rectum []. Since endoscopes are in contact with mucus membrane and body fluids, it is possible that these instruments can be implicated in transmission the virus []. As such, most guidelines recommend the use of personal protective equipment (PPE) during endoscopic examination in order to prevent nosocomial outbreaks of COVID-19 [, , , ]. When the virus is highly suspected or confirmed, double gloves and N95 or FFP2/3 masks are indicated and the operative team should be properly trained to wear and remove PPE safely [].

While research suggests that advanced age, organ failure, and elevated d-dimer levels are indicators of poor prognosis in COVID-19 patients, a recent study highlights kidney dysfunction as a potential risk factor for mortality, as well [,]. A multi-center retrospective study of 193 COVID-19 patients (128 with non-severe disease, 65 severe) investigated the presence of kidney dysfunction and demonstrated that, on admission, 59% of patients had proteinuria, 44% had hematuria, 14% had elevated BUN, and 10% had elevated creatinine levels []. These lab derangements were found to be significantly worse in those with critical illness (including non-survivors), and 66% of the patients who developed an AKI (43/65) were considered to have severe disease []. Per Pei at el., 75.4% of patients with COVID-19 had an abnormal urine dipstick at initial presentation []. These findings have been reproduced by several other studies which have shown that development of an AKI is a common lab finding in COVID-19 and a feature of those with severe disease [,,,, , ]. Patients with elevated creatinine at admission were more likely to be admitted to the ICU [,, , ]. Patients who survive COVID-19-related AKI have been shown to be at an increased risk of developing progressive CKD after the initial infection []. The risk CKD is associated with the severity of the AKI and the presence or absence of tubular damage []. These results suggest that early recognition of kidney dysfunction in patients with SARS-CoV-2 is important for disposition and monitoring for potential decompensation and to reduce the progression to chronic kidney impairment.

In addition to being a predictor of severe disease, Li et al. demonstrated that patients who developed an AKI had a 5.3-fold increased risk of mortality compared to those without kidney abnormalities, illustrating the importance of renal dysfunction as a negative prognostic indicator for survival []. Chen et al. also determined that 25% of non-survivors had developed an AKI during hospitalization []. One study noted that the kidney was the third most commonly damaged organ after the lungs and heart in those who died due to COVID-19 []. A consecutive cohort study of 710 COVID-19 positive patients confirmed that markers of kidney dysfunction such as elevated BUN and creatinine were independent risk factors for in-hospital death even after adjusting for potential confounders []. Mortality for those who presented with evidence of renal failure on admission was 33.7% vs. 13.2% in those without kidney dysfunction []. Similarly, Pei et al. demonstrated that patients with renal involvement had higher overall mortality compared to those without (11.2% vs 1.2%) []. Patients with chronic kidney disease are particularly vulnerable to negative outcomes. A recent meta-analysis confirmed that those with pre-existing renal dysfunction appeared to have significantly higher pneumonia-related mortality than those without underlying renal disease and are at increased risk for severe COVID-19 infection [,].

When comparing SARS-CoV-2 with SARS-CoV, AKI was a much less common finding (6%) during the 2003 outbreak of SARS, but it was recognized as a significant indicator of mortality (92% of SARS patients with AKI died) []. Similarly, renal dysfunction was also associated with increased risk of death in patients infected with H1N1 []. Evidence of the prognostic implications of kidney injury in other serious respiratory viruses highlights the importance of recognizing dysfunction early in the disease course in order to improve morbidity and mortality. Early recognition and treatment of renal dysfunction may help to improve the prognosis of those infected with COVID-19.

Dermatologists have become increasingly involved in caring for COVID-19 patients due to shifting clinical responsibilities during the pandemic, thus sparking an interest in the possible cutaneous manifestations of SARS-CoV-2. Recalcati, an Italian dermatologist, looked at a cohort of 88 COVID-19 positive patients and determined that 18 (20.4%) developed skin symptoms []. The manifestations reported were erythematous rash (15.9%), generalized urticaria (3.41%) and chicken pox-like lesions (1.14%) []. Dermatologists in Rome identified 2 of 130 patients with COVID-19 who presented with isolated herpetiform lesions on their trunk during their inpatient stays []. A patient in Barcelona exhibited vesicular lesions on her back 8 days after COVID-19 diagnosis []. A multicenter case series of 22 patients in Italy described a varicella-like exanthem as a specific cutaneous manifestation of the virus []. Furthermore, the study showed that median time from onset of systemic symptoms (fever, fatigue, or cough) to presentation of the exanthem was 3 days with a median duration of 8 days []. 54.5% of patients had vesicular lesions on the trunk, and the lesions were scattered in the majority of cases (72.7%) []. 7 of the 22 patients underwent skin biopsy and demonstrated histology consistent with viral infection.

Several case reports have also described the potential skin manifestations of COVID-19. A cluster of eight children in the U.K. presented with features similar to atypical Kawasaki including the archetypal skin rash appearance []. Following this report, additional cases with similar presentation were observed across the globe. Henry et al. outlined the case of a 27-year-old female who presented with pruritic disseminated erythematous plaques without cough or fever. Two days later she tested positive for COVID-19 after the onset of chest pain and fever []. Another case report discussed a 28-year-old female who developed confluent “erythematous-yellowish” papules that progressed to pruritic, hardened plaques 13 days after testing positive for COVID-19 []. Joob and Wiwanitkit detailed the case of a petechial eruption and thrombocytopenia initially thought to be Dengue fever, but ultimately proven to be COVID-19 after the development of respiratory symptoms []. Amatore et al. discussed a 39-year-old male who complained of fever and exhibited “erythematous and edematous non-pruritic annular fixed plaques involving the upper limbs, chest, neck, abdomen and palms, sparing the face and mucous membranes” without cough or dyspnea []. The patient was tested for and diagnosed with COVID-19 after reporting exposure to a family member with the virus. Similarly, Van Damme et al. examined cases of disseminated urticaria in two febrile patients who later developed respiratory symptoms and tested positive for COVID-19 [].

Dermatologists have been challenged with differentiating between infectious and allergic etiologies of rashes associated with COVID-19, since they are both clinically and histologically similar. Many of the medication combinations currently under investigation for the treatment of the virus may lead to drug eruptions []. In a study of 140 COVID-19 patients by Zhang et al., drug hypersensitivity (11.4%) and urticaria (1.4%) were the most prevalent cutaneous symptoms associated with the virus []. Jimenez-Cauhe et al. described a patient with “erythemato-purpuric, millimetric, coalescing macules, located in flexural regions” that developed 3 days into treatment with hydroxychloroquine and lopinavir/ritonavir []. It was unclear whether these lesions were a manifestation of SARS-CoV-2 or an adverse reaction to the medications; however, there are no additional reports of dermatological symptoms in patients treated with this combination therapy []. Another report described sterile pustules, similar to the typical findings of acute generalized exanthematous pustulosis (AGEP), as a cutaneous manifestation of COVID-19 []. Although AGEP is classically associated with drugs, it is possible that COVID-19 can predispose certain patients to develop AGEP-like cutaneous eruptions as a late-onset manifestation of the infection.

In addition to potential drug hypersensitivity or direct viral infection of the skin, Manalo et al. hypothesized that underlying DIC and microthrombi may contribute to cutaneous symptoms []. The study described two cases of unilateral transient livedo reticularis in non-critically ill COVID-19 positive patients. Similarly, a retrospective study of 7 critically ill patients in Wuhan, China exhibited significant limb ischemia with plantar plaques and acral cyanosis as dermatologic manifestations of their underlying hypercoagulable state [].

While the cutaneous manifestations of COVID-19 can vary, early identification of unusual lesions in those without a known trigger is crucial to limiting the spread of COVID-19.

The SARS-CoV-2 virus is known to cause severe respiratory distress through direct invasion into the lung parenchyma as evidenced by the destructive pattern seen on imaging of COVID-19 positive patients. However, some patients with evidence of significant lung damage and severe hypoxic do not develop tachypnea []. Scientists have hypothesized that there is an abnormal response of the peripheral afferent fibers in the lungs and airways that stimulate respiration []. Direct entry of the virus into brain tissue, notably the brainstem, may also result in the loss of involuntary control of breathing [,].

SARS-CoV-2 is believed to enter the nervous system via hematogenous spread, directly through the cribriform plate, or through retrograde neuronal synapses from the olfactory bulb and vagal afferents [,,,]. Upon entering the nervous system, it can manifest as a viral encephalitis, cerebrovascular disease, or peripheral nerve symptoms [,,].

A large retrospective study of 221 patients with COVID-19 at the Union hospital in Wuhan found that 5% of patients presented with acute ischemic stroke, one patient developed cerebral venous sinus thrombosis (CVST), and one had cerebral hemorrhage [,]. Those with cerebrovascular disease were significantly older (71·6 ± 15·7 years vs 52·1 ± 15·3 years; p < 0·05) and had cardiovascular risk factors. Lab evaluation of this cohort determined that they were more likely to have elevated CRP and d-dimer, lymphopenia, thrombocytopenia, and uremia [, , ].

Cerebral hemorrhage is thought to be a consequence of the virus binding to ACE-2 receptors on endothelium, contributing to break down of the blood brain barrier [,]. Ischemic changes and CVST are likely secondary to a hypercoagulable and pro-inflammatory state, further supported by elevation in CRP and d-dimer in these patients [,,]. Numerous other studies summarized in Table 2 discuss presentations of cerebrovascular disease in COVID-19 patients [,,].