key: cord-0894305-4loam1hd
authors: Phanish, Mysore; Ster, Irina Chis; Ghazanfar, Abbas; Cole, Nicholas; Quan, Virginia; Hull, Richard; Banerjee, Debasish
title: Systematic review and meta-analysis of COVID-19 and kidney transplant recipients, the South West London Kidney Transplant Network experience
date: 2020-12-19
journal: Kidney Int Rep
DOI: 10.1016/j.ekir.2020.12.013
sha: bb1d0b5f9ab9a376638ee56364da469dd66ae6f8
doc_id: 894305
cord_uid: 4loam1hd

INTRODUCTION: There is paucity of literature comparing outcomes of kidney transplant patients with Covid-19 to that of dialysis and waitlisted patients. This report describes our data, provides comparative analysis, together with meta-analysis of published studies and describes our protocols to restart the transplant programme. METHODS: Data were analysed on kidney transplant, dialysis and waitlisted patients tested positive for SARS-CoV-2 (naso-pharyngeal swab PCR) between March 1, 2020 and June 30, 2020 together with meta-analysis of 16 studies. RESULTS: 23/1494 kidney transplant patients tested positive for SARS-CoV-2 compared to 123/1278 haemodialysis patients (1.5% vs 9.6%, p<0.001), 12/253 waitlisted patients (1.5% vs 4.7%, p=0.002). 19 required hospital admission, 6 died and 13 developed AKI. Overall case fatality ratio was 26.1% compared to patients on haemodialysis (27.6%, p=0.99) and waitlisted patients (8.3%, p=0.38). Within our entire cohort, 0.4% of transplant patients died compared to 0.4% of waitlisted patients and 2.7% of haemodialysis patients. Patients who died were older [Alive (median 71years) vs. Dead (median 59years), p=0.01]. In meta-analysis of 16 studies, including ours, pooled case fatality ratio was 24% [95%CI (19%, 28%)]; AKI proportion in 10 studies was 50% [95%CI (45%, 56%)], with some evidence against no heterogeneity between studies (p=0.02). CONCLUSIONS: From our cohort of transplant patients, a significantly lower proportion of patients contracted COVID-19 compared to waitlisted and dialysis patients. The case fatality ratio was comparable to that of dialysis cohort and pooled case fatality ratio from meta-analysis of 16 studies. The pooled AKI ratio in the meta-analysis was similar to our experience.

SARS-CoV-2, the virus that causes COVID-19 continues to cause significant mortality and morbidity across the world as the pandemic evolves. As of 27 th July 2020, 300,111 people had tested positive for the virus in the UK and, of those tested positive, across all settings, 45,312 have died. The disease is primarily pulmonary but involvement of other organs, including the kidneys and heart during the course of illness is now well recognised. Kidney transplant recipients, due to their immunosuppressive burden and underlying co morbidities are thought to be at higher risk of acquiring the infection as well as developing severe disease requiring hospitalisation. We recently reported our initial experience of 7 renal transplant patients from 3 south London hospitals: 2 out of 7 patients were managed at home and one patient died 1 . All patients were managed with reduction of immunosuppression with no specific antiviral or anti-inflammatory therapies. In the same journal edition, Alberici et al published their early experience of 20 kidney transplant patients admitted with SARS CoV-2 pneumonia in which they described a 25% mortality in spite of additional treatment with various drugs that included Lopinavir/Ritonavir, Hydroxychloroquine, Dexamethasone and Tocilizumab 2 . Since these early reports, there have been several further reports of COVID-19 in kidney transplant patients describing overall case fatality ratios of 10-38% and 50-65% for patients requiring invasive ventilation. [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] We and others have advocated for immunosuppression reduction as a primary therapeutic strategy for hospitalised kidney transplant patients with COVID-19 pneumonia with cessation of antiproliferative agents (Mycophenolate mofetil/Azathioprine) and continuation of CNIs either at same or reduced dose depending on severity of disease along with continuation of corticosteroids [1] [2] [3] 6 .

In this report, we describe 23 kidney transplant patients tested positive for SARS CoV2 from two tertiary care renal centres from South London Renal transplant Network, UK. This includes follow up data on 5 patients described in our previous report. 

Data were collected on all kidney transplant recipients tested positive for SARS-CoV-2 between March 1, 2020 and June 30, 2020 (First wave of Covid19 in the UK) and followed until October 15, 2020. The data collected included demographics, clinical and laboratory parameters and outcomes. In addition, we collected data on dialysis patients that included all the patients on dialysis and those on the transplant waitlist. Data were collected as part of routine clinical processes and downloaded for the study from Electronic Patient Records. and stopped in severe cases where there was progressive clinical and radiological deterioration (n=8). Prednisolone dose was either unchanged (n=3) or increased (n=13) in all cases. Some of the patients were recruited in to Recovery trial (Randomised Evaluation of COVID-19 therapy, www.recoverytrial.net). As a part of this trial two patients received hydroxychloroquine and one received Dexamethasone. In addition, two patients received Tocilizumab. Out of 4 patients managed at home, one patient had his mycophenolate mofetil dose reduced by 50% with increase back to baseline dose after 2 weeks, remaining three patients were managed without any change to immunosuppression ( Table 2 ).

All patients who had Tacrolimus dose reduced had the dose progressively increased such that by 2 weeks post discharge the levels were in therapeutic range (5-8ng/mL).

Mycophenolate mofetil was re-introduced around 2-3 weeks post discharge provided patients were well with no fever or other symptoms of COVID-19 for at least 3 days and had a normal CRP.

Patient demographics, laboratory parameters and clinical outcomes are summarised in table 1. Age, admission to intensive care unit (ICU) and type of respiratory support required were the only variables significantly different in patients who died compared to those discharged home. Patients who died were significantly older (59.2±8.2 vs. 70.5±6.8 years; p=0.01 and required more ventilatory support (p=0.04). There were no significant differences between the two groups (living Vs died) with regards to co-morbidities, peak ferritin levels, C-reactive protein, baseline lymphocyte count or lowest lymphocyte count during admission (Table 1 ).

Duration of hospital stay, respiratory support, Acute kidney injury (AKI), Renal Replacement Therapy (RRT) and outcome of hospitalised patients (n=19) are described in table 2. 6/19 (31.57%) hospitalised patients died ( Table 2 ). Out of the total cohort of 1494 transplant patients, 6 patients died and this represents 0.4% of total cohort. Table 3 ). There was no difference in proportion of patients on maintenance steroids between the two groups (living Vs died) ( Table 1) .

Among 6 patients who died, one patient was white, 2 black, 2 South Asian, 1 other. The ethnicity was not significantly different between the two groups (Survived and Died) but this may be due to small sample size. All patients who died had hypertension and 4 had diabetes.

Respiratory support (see Tables 1 and 2 ): Out of 19 hospitalised patients, 3 were managed on high flow nasal cannula (HFNC) or non-invasive ventilation (NIV), 6 (31%) were intubated and ventilated; remaining 10 were managed with oxygen delivered through nasal cannula or venturi mask. Of the intubated and ventilated patients, 3 out of 6 patients (50%) died. Of the 3 other patients who were discharged home, two had a functioning graft and one remained on dialysis. The patient who was discharged on dialysis had poor graft function before he had COVID-19 and was on haemodialysis pre admission. He was ventilated for a prolonged period of 57 days. Tables 1 and 2 (Table 4 ). Case fatality ratio was 26.1% for transplant patients, 8.3% for waitlisted patients, 27.6% for haemodialysis patients and 75.0% for peritoneal dialysis patients. There was no statistically significant difference in case fatality ratio of transplant patients compared to waitlisted patients, haemodialysis patients and peritoneal dialysis patients.

( Table 4) .

We performed meta-analyses of 15 published studies and our data to derive a pooled estimate of case fatality ratio (of hospitalised patients) and AKI in kidney transplant patients who tested positive for SARS-CoV-2, including a recent publication of TANGO international consortium. 2, 3, 5, [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] The total number of hospitalised patients included in these studies J o u r n a l P r e -p r o o f were 871. The pooled case fatality ratio was 24% (95% CI 19%, 28%). The variability in the effect size attributable to between study heterogeneity was moderate (I 2 =51.5%) consistent with some evidence against the null hypothesis stating no heterogeneity between studies (p = 0.01) (Fig 1A) . The Montefiore 2 study 18 , the third most influential in this analysis exhibited a case fatality ratio of 38% [95%CI (29%, 48%)], well above the pooled estimate of 24%

[95%CI (19%, 28%)]. Excluding this study, the I 2 drops to 34.3% with a p-value=0.09 indicating consistency with the magnitude of I 2 and with the null hypothesis of not much heterogeneity between studies. The pooled case fatality ratio in this analysis was 22%

[95%CI (18%, 27%)] very close to our first analysis. Given the size and hence the precision of estimate in Montefiore 2 study (n=111) we opted to include all studies and provide evidence that pooled case fatality ratio of hospitalised kidney transplant patients with Covid-19 is

The analyses of AKI included 10 studies which reported AKI. The pooled proportion of AKI was 50% [95%CI (45%, 56%)]. There was no evidence to suggest heterogeneity between the studies, p = 0.27 and therefore, the data estimates that 50% [95%CI (45%, 56%)] of kidney transplant patients with Covid19 develop AKI (Fig 2A) . We also separately analysed pooled proportion of severe AKI (Stage3 AKI or those requiring RRT). This analysis showed pooled proportion of stage 3 AKI of 18% [95%CI (12%, 25%)] ( Fig 2B) . However, there was evidence for presence of significant heterogeneity between studies P=<0.001. Re analysis after removal of Bologna study showed that high stage3 AKI percentage of 45% yielded results with pooled stage3/RRT requiring AKI estimate of 16% [95%CI (10%, 22%)] although the evidence against no heterogeneity between studies remained significant (p=0.02) with I 2 = 56.96% (Fig 2C) . A small proportion of our overall transplant patient cohort got COVID-19 (1.5%) compared to 9.6% haemodialysis patients, 4.7% peritoneal dialysis patients and 4.7% of waitlisted patients with an in-hospital case fatality ratio of 31.57%. A German multicentre study 10, 021 patients with COVID-19 admitted to 920 hospitals showed overall mortality of 22% and 53% mortality in those requiring mechanical ventilation 20 . An UK study of 20,133 patients admitted to 208 hospitals demonstrated an overall mortality of 26% 21 compared to pooled mortality of 24% in our meta-analyses. Our overall case fatality ratio of 26% is very similar to 27% mortality shown in UK data in renal transplant recipients (NHSBT weekly Covid19 reports). Transplant patients had comparable case-fatality ratios to that of haemodialysis patients. The case fatality ratio of waitlisted patients was lower compared to transplant patients but this difference did not achieve statistical significance. Waitlisted patients tend to be younger with fewer co morbidities compared to some of the older transplant patients and this may largely explain this difference. Older age was associated with poor prognosis We observed high percentage of Acute Kidney Injury (AKI) in these patients (68%) and 6 (31.5%) patients with stage 3 AKI. In comparison, the pooled proportion of AKI was 50% and that of stage 3AKI/AKI reporting RRT was 16%-18%. There was significant heterogeneity in studies reporting AKI as some reported all stages and some only reporting patients requiring RRT. We analysed these separately and included stage3/RRT requirement in one group as this indicates severe AKI. However, it must be noted that studies that reported patients requiring RRT only would have excluded patients with stage 3 AKI not needing RRT and J o u r n a l P r e -p r o o f therefore, true number of stage3 AKI in these studies is likely to be higher than reported.

Reassuringly, the AKI recovered in majority of the cases. Two patients who remained dialysis dependent had poor baseline kidney function (eGFR<20ml/min/1.73m 2 ) and their kidney function deteriorated further during their hospital stay. It is possible that these patients developed rejection upon immunosuppression reduction but we did not have transplant kidney biopsy results to prove this. On clinical grounds, we felt that treating COVID-19 pneumonia was the priority and therefore, felt that risk/benefit ratio did not favour doing a biopsy. During this period, we found an AKI risk of 26% in hospitalised patients with COVID- Comparisons were performed between aggregated proportions. Only p-values lower than 0.05/12=0.0042 are considered significant (shown in green) due to Bonferroni correction for multiple comparisons.

The pooled case fatality ratio was 24% (95% CI 19%-28%). There was moderate heterogeneity between the studies (I 2 =51.5% (variation in effect size attributable to heterogeneity), Heterogeneity chi 2 = 30.90 (d.f. = 15), p = 0.01). The New York Montefiore2 study, the third most influential in this analysis exhibited a case fatality ratio of 38% [95%CI (29%, 48%)], well above the pooled estimate of 24% [95%CI (19%-28%)]. (Fig 1A) We then analysed 14 studies excluding this study and with this analysis the I 2 drops to 34.3% with a p-value=0.09 consistent with the null hypothesis of not much heterogeneity between studies. The pooled case fatality ratio in this analysis was 22% [95%CI (18%, 27%)]. (Fig 1B) The pooled proportion of AKI was 50% [95%CI (45%, 56%)]. There was no significant heterogeneity between the studies, chi 2 = 11.02 (d.f. = 9), p = 0.27; I 2 (variation in effect size (ES) attributable to heterogeneity) = 18.37%. Therefore, the pooled proportion of AKI is 50%

[95% CI (45%, 56%)].

The pooled proportion of severe AKI (Stage3/requiring RRT) was 18% [95%CI (12%, 25%)] ( Fig 2B) . However, there was a significant heterogeneity I 2 = 66.27%, P=<0.001. Re analysis after removal of Bologna study that showed high stage3 AKI percentage of 45% yielded results with pooled stage3/RRT requiring AKI estimate of 16% [95%CI (10%, 22%)] but the heterogeneity, although improved, remained significant with I 2 = 56.96%, P=<0.02 ( Fig 2C) . Therefore, it appears from these analyses that pooled proportion of severe AKI is 16-18%. Table 3 J o u r n a l P r e -p r o o f

GROUP COMPARISONS (as above in 4A) Pooled case-fatality ratio Figure 1B 

COVID-19 infection in kidney transplant recipients

A single center observational study of the clinical characteristics and short-term outcome of 20 kidney transplant patients admitted for SARS-CoV2 pneumonia

Covid-19 and Kidney Transplantation

COVID-19 and kidney transplantation: an Italian Survey and Consensus

COVID-19 in kidney transplant recipients

COVID-19: implications for immunosuppression in kidney disease and transplantation

Immunological risk stratification and tailored minimisation of immunosuppression in renal transplant recipients

Preliminary data on outcomes of SARS-CoV-2 infection in a Spanish single center cohort of kidney recipients

Kidney allograft recipients, immunosuppression, and coronavirus disease-2019: a report of consecutive cases from a New York City transplant center

Early Description of Coronavirus 2019 Disease in Kidney Transplant Recipients in New York

Coronavirus Disease 2019 Pneumonia in Immunosuppressed Renal Transplant Recipients: A Summary of 10 Confirmed Cases in Wuhan, China

COVID-19 and Kidney Transplantation: Results from the TANGO International Transplant Consortium

COVID-19 Infection in Kidney Transplant Recipients: Disease Incidence and Clinical Outcomes

COVID-19 Outcomes in Patients Waitlisted for Kidney Transplantation and Kidney Transplant Recipients

COVID-19 among kidney-transplant recipients requiring hospitalization: preliminary data and outcomes from a single-center in Brazil

Outcomes of COVID-19-positive kidney transplant recipients: A single-center experience

COVID-19 in solid organ transplant recipients: no difference in survival compared to general population

COVID-19 infection in kidney transplant recipients at the epicenter of pandemics

An initial report from the French SOT COVID Registry suggests high mortality due to Covid-19 in recipients of kidney transplants

We thank all the members of transplant team involved in taking care of transplant patients and planning COVID-19 specific pathways.

Prof Banerjee has received research grant and speaker fees from AstraZeneca, Pfizer and Vifor Pharma, research grant from kidney research UK. Other authors have nothing to disclose.

The PRISMA checklist document is available at KI Reports website.J o u r n a l P r e -p r o o f