Covid-19 in Liver Transplant Recipients: An Italian Case-Control Double-Centre Study.

Background and Aims: Liver transplant (LT) recipients are vulnerable to SARS-CoV-2-infection (COVID-19), due to immunosuppression and comorbidities. This study aimed to evaluate the impact of COVID-19 on LT recipients compared to general population in the Campania region. Methods: In this case-control double-centre study, we enrolled all consecutive adult LT recipients with conrmed SARS-CoV-2-infection. Data were collected at diagnosis of COVID-19 and during follow-up and compared with the regional population. Results: Thirty LT recipients (3.28%) developed SARS-CoV-2-infection (76.6% male, median age 62.6 years). Sixteen (53.3%) were symptomatic. Common symptoms were fever, cough, fatigue, and anosmia. Twenty-ve (83.4%) were outpatients, 5 (16.6%) required hospitalization (6.6% admitted to Intensive Care Unit, 6.6% developed Acute Respiratory Distress Syndrome and 6.6% died). Immunosuppressors were in 3 (10%) patients. Incidence rate of COVID-19 was similar between LT patients and general population (3.28% vs 4.37%, p=0.1) with higher rate of symptoms in LT patients (46.67% vs 15.87%, p<0.00). At univariate analysis, hospitalization and case fatality rates were higher in LT patients compared to general population (16.67% vs 4.54%, p=0.001; and 6.67% vs 1.76%, p=0.04, respectively). At multivariable logistic regression analysis, LT patients with COVID-19 were more frequently symptomatic (OR 5.44 [95%CI 2.44-12.17], p<0.00), whereas hospitalization and death for COVID-19 were not signicatively associated with LT condition (p=0.72 and p=0.46, respectively) and were comparable with general population. Conclusions: LT is not a risk factor for acquiring COVID-19. Nonetheless, LT patients are more frequently symptomatic,


Introduction
The SARS-CoV-2 infection (COVID- 19) pandemic rapidly spread in the rst months of 2020, becoming a public health matter with several unforeseen challenges to health care systems worldwide and several unmet issues. To date, with 3.258.770 cumulative cases con rmed on January 16 th , 2021, Italy is among the most affected countries by the pandemic 1 . Currently, Campania region in Southern Italy is one of the most overwhelmed regions with a cumulative incidence of 5393.79 cases per 100000 inhabitants 2 .
At present, rm knowledge on disease evolution and optimal management of COVID-19 infection, in the setting of solid organ transplantation are lacking. Transplant recipients are considered as "clinically extremely vulnerable" sub-population for COVID-19 because of lifelong immunosuppressive therapy and their higher rate of underlying comorbidities 3 .
The available data on COVID-19 infection in liver transplant recipients are not conclusive and are limited to several case series 4-7 , 4 national cohorts from Italy, France, UK and Spain [8][9][10][11] and 3 international studies 12-14 characterized by marked heterogeneity. However, the lack of non-transplanted patients with COVID-19 as a control-group makes it challenging to assume the increased risk of severe outcomes in liver-transplanted patients. We collected data from our population of more than 900 liver transplanted patients who regularly attended the clinical follow-up program despite the pandemic environment. The current study aims at describing the incidence of SARS-CoV-2 infection, its clinical features at presentation and during disease course, early outcomes, and mortality in a cohort of liver transplanted patients compared to events occurring in the general population (non-transplanted subjects) in the geographical area of Campania Region in Southern Italy.

Study design and target population
This is a case-control double-centre study launched after the outbreak of COVID-19 in Italy on 9th of March 2020. The study was conducted in two regional referral hospital for liver transplanted patients (Cardarelli Hospital and Federico II Hospital) in accordance with the Declaration of Helsinki. The protocol was approved by the ethical board of the promoting centre (Federico II Hospital).
All consecutive adult liver transplanted (LT) patients with concomitant con rmed SARS-CoV-2 infection (with any symptom pro le or level of disease severity) were prospectively enrolled in this study at the time of diagnosis of COVID-19 and were followed-up until death or as of as January 16 th , 2021. Patients were excluded if they presented any of the following conditions: any recipient of more than 1 simultaneous organ transplant, SARS-CoV-2 infection not con rmed by a nasopharyngeal swab reverse-transcriptasepolymerase-chain-reaction (RT-PCR) test, hospitalisation status or mortality outcome unknown or unreported, or patient aged 18 years or younger at the time of COVID-19 diagnosis.
This collected data regarding COVID-19 infection in our transplanted population was compared with the Campania regional population extracted by National Health System COVID-19 dataset 15 . Our Region accounts for more than 7 million inhabitants and on January 16 th , 2021, 308.101 cumulative cases of COVD-19 were registered, with a cumulative incidence of 5393.79 cases per 100000 inhabitants 1 .
The same inclusion and exclusion criteria have been applied to the population of Campania region with SARS-CoV-2 con rmed infection.
Diagnostic procedure COVID-19 infection was con rmed by a RT-PCR assay of nasal and pharyngeal swab specimens, according to the WHO guidelines 16 . In particular, LT patients admitted to the hospital (because of hypoxaemia and/or radiological chest X-ray abnormalities or because of mild symptoms in subjects with signi cant comorbidities or who were over the age of 60 years) were tested onsite within the rst 24 hours, whereas outpatients with mild symptoms were tested at home or at 'drive-through' testing sites within the next 72 hours after contacting a dedicated phone number. For the general population, all symptomatic subjects had been tested at home or at 'drive-through' testing sites within the next 72 hours from onset of symptoms, while asymptomatic subjects were subjected to nasopharyngeal swab only on the occasion of targeted screening or if they were close contacts of a positive case as required by the ministerial guidelines for the surveillance and control of SARS-CoV-2 infection 16 .

Collection data
For LT patients, demographic data, comorbidities, information regarding indication to liver transplantation and underlying liver disease, clinical symptoms or signs and laboratory parameters as well as imaging features at the time of COVID-19 diagnosis were collected. Time from liver transplantation was classi ed as long term (more than 5 years from transplant) and short term (1 year or less from transplant). Modi cations of immunosuppression (reduction or withdrawal; in particular for patients on double immunosuppression, reduction was de ned as reduction of at least one immunosuppressive drug and/or discontinuation of one medication, while withdrawal was de ned as discontinuation of both drugs) therapy were registered as well as speci c drugs (oral hydroxychloroquine or azithromycin, antiviral therapy with lopinavir/ritonavir or remdesivir, steroids or tocilizumab or anti-thrombotic prophylaxis) or ventilation support prescribed for COVID-19.

Outcomes
The main outcome of the study was the analysis of the burden of COVID-19 infection in a cohort of liver transplanted patients in terms of incidence, type and severity of symptoms at presentation, hospitalisation rate and case fatality rate in comparison to non-transplanted general adult population.

Statistical analysis
Shapiro-Wilk test has been performed for testing normal distribution. Continuous variables were expressed as median and interquartile range (IQR), and qualitative variables as absolute frequency and percentage. Descriptive statistics of the study variables were performed using Kruskal-Wallis equality-ofpopulations rank test, Chi-square and Fisher's exact test statistics, as appropriate.
Multivariable logistic regression models were employed to assess the association between LT (dependent variable) and 5 variables (age, gender, symptoms, COVID19 hospitalization and death). In particular, a model has been adjusted for age, gender, COVID19 hospitalization and death to evaluate the impact of SARS-CoV-2 infection on LT (partial model). Afterwards the model was further adjusted for symptoms to evaluate the impact of SARS-CoV-2 disease on LT (Full model).
Statistical analyses were performed using STATA 15 statistical software (StataCorp, College Station, TX, USA).
The median time from LT to COVID-19 diagnosis was 12.8 (IQR 4-20) years and 22 patients (73.3%) had been transplanted more than 5 years before COVID-19 infection. The time from transplant to diagnosis was signi cantly shorter in asymptomatic (8.62 years) compared to symptomatic group (16.8 years) (p=0.03). No differences in LT indications, immunosuppressive regimens and comorbidities were found between asymptomatic and symptomatic groups.
Overall, 25 (83.4%) patients received outpatient care for COVID-19, while 5 (16.6%) subjects were hospitalized, with a median duration of hospital stay of 11 [IQR [10][11][12][13][14] days. Among hospitalized patients, 2 (6.6%) LT recipients were admitted to Intensive Care Unit (ICU), 2 (6.6%) developed Acute Respiratory Distress Syndrome (ARDS) and 2 (6.6%) died. General characteristics of hospitalized patients are summarised in Table 3. Three out of 5 hospitalized patients had > 60 years and HCC as indication for LT. All of them were long-term transplanted. Moreover, four out of 5 received CNI as single immunosuppressant agent and one of them received combination therapy (CNI + mTOR). Immunosuppressive therapy was modi ed after COVID-19 diagnosis in 3 patients. Two out of 5 patients required ICU admission, invasive ventilation and died for ARDS after a median time of 7 days. Two out of 30 LT patients died with a crude case fatality rate of 6.67% [95% CI 6.39-6.96]. The cumulative incidence rate of COVID-19 and the median recovery time did not differ between LT and non-LT patients (p=0.1 and p=0.34, respectively), whereas symptoms were more frequent in the LT population (46.67% vs 15.87%, p<0.00), see Figure 1. At univariate analysis, the hospitalization rate was signi cantly higher in LT recipients than in general population (16.67% vs 4.54%, p=0.001), but the median length of hospitalization was similar (p=0.31). Moreover, in the general population of Campania  Table 4).

Discussion
COVID-19 caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has been associated with high mortality worldwide. Identifying patients at high risk of infection or those likely to have a severe clinical course is mandatory to guide treatment decisions and infection prevention strategies. In Campania region (Southern Italy) on the 16th of January 2021 the infection rate of SARS-CoV-2 and its case fatality rate have been of 5.39% and 3.15%, respectively 1 . In this scenario, the liver transplant population represents a priori a vulnerable cohort of patients at increased risk of infections and poor outcome due to chronic immunosuppression, high rates of comorbidities and advanced age. Whether they are at particularly high risk for critical COVID-19 in real clinical practice still requires further evidence. Herein we report comprehensive data on an Italian regional cohort of LT recipients with con rmed COVID-19 infection, analysing the clinical characteristics, the management and outcome of these patients, and providing a comparison group using patients COVID-19 information of the general population of the same geographical area. Within our study, multivariable analysis of 30 LT patients showed that, despite more symptomatic, they are not at increased risk of COVID-19 infection and of worse outcome in terms of hospitalization and mortality compared to non-transplanted subjects.
The cumulative incidence of SARS-CoV-2 infection in our LT cohort was 3.28%, comparable to the incidence of the general population (4.37%), highlighting that LT patients do not necessarily have a higher risk of infection. This rst result was in line with data reported by Bhoori et al. 4 showing an incidence rate of 3% in long term transplanted patients, and with data by Mocchegiani et al. 17 showing an incidence rate of COVID-19 infection in LT patients (0.87%) similar to the general population of Marche Region (0.44%).
Conversely, our results contrast with Trapani et al. 8 that reported an incidence rate in LT patents three times higher than that estimated for the Italian population.
In terms of clinical presentation, in our cohort of LT patients, almost 50% showed a symptomatic course, a percentage decisively higher than that observed in the control group (46.67% vs 15.87%). These observations are distinct from the data reported by Colmenero et al. 11 showing a symptom rate of 93% in LT subjects. No other studies published until now reported this information, since most of them were focused only on symptomatic cohort of LT patients 5,8-10,12−14 . Moreover, the clinical presentation of COVID-19 infection in LT recipients is similar to that observed in non-transplanted population, with fever, fatigue, anosmia and cough being the four more common symptoms. These ndings are sustained also by Dumortier et al. 9 and Becchetti et al. 14 , both showing that fever and cough were the most common symptoms in LT population, while anosmia was present only in 10% of patients.
Based on our results, the management at home of the majority of COVID-19 positive LT patients seems to be possible with good outcomes and limiting the need of hospitalization (in our cohort only 16.6%). In our experience, we decide for hospitalisation on a single case-basis according to a close monitoring of clinical conditions (indication to hospitalisation in case of worsening of respiratory symptoms with hypoxaemia and/or radiological chest X-ray abnormalities or because of mild symptoms in subjects with signi cant comorbidities or who were over the age of 60 years). For the group of patients managed at home, teleconsultation or phone call surveillance until disease resolution was offered 18 . Our Hepatology Unit recognised the pandemic challenges early and implemented several preventative measures for minimizing in-person visits. Teleconsultations, like a fully implemented telemedicine service, partially restructured for the COVID-19 pandemic with video-consultations were prioritized, as well as, extended helpline hours and smart working from home to increase remote patient monitoring and mobile health care. Concomitantly, these measures reduced the risk due to contact with medical staff 18 .
It has been described that more than 50% of LT patients develop severe forms of COVID-19 disease 19 .
Fortunately, our study does not support this observation. Despite well-known risk factors for poor outcome of COVID-19 infection such as chronic immunosuppression, high rates of comorbidities and advanced age, COVID-19 infection in our LT patients showed a clinical course not more severe than that observed in general population as also reported by Becchetti et al 14 . This result is probably related to several factors: the younger median age in our LT cohort (62.6 vs > 65 years) compared to other studies 11,14 , the higher median time from LT to COVID-19 diagnosis (12.8 vs < 10 years) 11,12,14 and the lower prevalence of comorbidities (50% vs 79% of LT patients with one comorbidity) 12 compared to other cohorts.
Conversely, even if the role of comorbidities as relevant risk factors for COVID-19 mortality has been demonstrated in the general population 20 and despite being highly prevalent among LT patients, in the present study no speci c comorbidity, nor a combination of comorbidity resulted independently associated with worse clinical outcome. This agrees with the results reported by Belli et al 12 .
The management of immunosuppression in LT patients with COVID-19 is a matter of concern in scienti c communities with a large debate on the opportunity to decrease or swich drugs. In our cohort, the majority of patients with COVID-19 were kept on their usual immunosuppressive regimen without need of modi cations [drugs modi cations only for 3 patients, all hospitalised for severe COVID-19 disease]. On the other hand, in 50% of our LT patients who required hospitalisation, immunosuppressive treatment was modi ed, according to the experience reported in literature showing a need to decrease immunosuppression in severe and critical conditions 4,7,9,14 . Nonetheless it is relevant to underline that therapy for COVID-19 differs across centres and countries and varied overtime with the increasing knowledge acquired. In our experience, the most used drug was acetaminophen, with a lower use for steroids and antibiotics in comparison to other published studies 7,11,14 . Finally, regarding the positive role of tacrolimus, as antivirals, in the treatment of COVID-19 disease as suggested by some reports, since 90% of our patients were treated with CNIs, it was not possible to assess any speci c effect of this drug.
The hospitalization rate in this study was 16.6%, a rate extremely lower than the rates reported in other March to May -7,11,12,14 -and not later than September 2020 (8)].
The overall case fatality rate of SARS-CoV-2 infection in our LT cohort was 6.6%, signi cantly higher than that observed in regional general population (1.76%), but decisively lower than the mortality rates reported by literature ranging from 10-27.3% 4,7,8,10−14 and by a recent review reporting a cumulative mortality rate of 20% 21 . To analyse these results, it is necessary to consider that the case fatality rate in Campania was always lower than in the other Italian regions (1.54% vs 3.99%), probably due to the presence of a younger population. Conversely, the mortality rate in hospitalized patients resulted higher (40%) and in line with reported ndings showing a rate ranging from 17-29% 7,9,14 .
This study has some strengths. First of all, we provided a large control group of non-transplanted subjects living in the same geographical area according to the epidemiology of the COVID-19 infection; this permits a better characterization of the impact of COVID-19 in patients with LT and strengthens the argument that these patients are not at higher risk of adverse COVID-19 outcomes. Second, our study is a prospective study and not a survey, with prospective acquisition of data collected in 2 centers with high levels of expertise in hepatology and infectious diseases. Third, we can give a picture of the real incidence of COVID-19 infection in LT recipients because, during the study period we contacted or visited in-person all the LT patients followed-up at our centers to be sure they were not infected or have suspicious of infection until the end of the study (16th of January 2021). Finally, this is the rst study providing data on LT patients with asymptomatic COVID-19 infection.
Nonetheless some potential limitations are also to be acknowledged. Firstly, the small sample size (30/915 LT patients), even if picturing the epidemiological situation of a well-de ned geographical area, can limit the interpretation of the results. Second, some information relating to the comorbidities of COVID-19 positive subjects is not routinely collected in the NHS-ISS dataset. For this reason, the characterization of the two populations appears unbalanced with a lack of clinical information relating to non-LT subjects.

Conclusions
We can assert that LT patients do not appear to be more prone to COVID-19 infection and, although they more often exhibit symptoms, they are not at increased risk of adverse outcomes in terms of hospitalization or mortality when compared to the general population. The authors believe that, although this population is more fragile, early diagnosis, careful and personalized home management in patients accustomed to the responsible relationship with the healthcare professional have a positive in uence on the outcomes. Abbreviations SARS-CoV-2 infection (COVID-19); liver transplanted (LT); reverse-transcriptase-polymerase-chainreaction (RT-PCR); cumulative incidence (CI); interquartile range (IQR); hepatocellular carcinoma (HCC); hepatitis B virus (HBV); hepatitis delta virus (HDV); calcineurin inhibitors (CNI); tacrolimus (TAC);