We conducted a national observational, retrospective, and multicenter study in which 30 Spanish hospitals participated with follow-up of patients with renal transplantation and who had included patients in the COVID registry of the SEN.

The study population was kidney transplant patients diagnosed with SARS-CoV2 infection from March 1, 2020, to December 31, 2020. We excluded patients diagnosed with coronavirus infection by serological methods, those who died during the study period, and patients with incomplete data in relation to the main objective of the study. Patients were followed for a period of 6 months.

Diagnosis of SARS-CoV2 infection was via polymerase chain reaction (PCR) technique of nasal or nasopharyngeal samples.

Patient inclusion covered two periods or waves of the pandemic in Spain; the first from February to July 2020 (N = 274, 44%) and the second from August to December 2020 (N = 341, 55.4%).

Demographic variables were collected (age, sex and race), clinical variables related to chronic kidney disease (CKD) and renal transplantation (etiology of CKD, number of renal transplantations, immunological risk, date of transplantation) as well as maintenance of immunosuppressive treatment, renal function (creatinine, glomerular filtration using the CKD-EPI equation in ml/min/1.73 m2, protein/creatinine index in urine of isolated sample in mg/g Cr) and anti-HLA antibodies prior to diagnosis of SARS-CoV2 infection.

Variables related to the management of immunosuppression after coronavirus infection, renal function (creatinine, glomerular filtration using the CKD-EPI equation in ml/min/1.73 m2 and protein/creatinine index in urine of isolated sample in mg/g Cr) at the diagnosis of the infection, at 7 and 15 days and at 2 and 6 months after the diagnosis of the disease, anti-HLA antibodies at 2 and 6 months after the diagnosis, as well as clinical variables of disease evolution (period of the pandemic, hospital or intensive care unit admission, days of admission, clinical manifestations, and antiviral treatment) were also collected.

Acute renal failure (ARF) was defined and classified according to the KDIGO clinical practice guideline of acute renal damage.12

The clinical and demographic data were obtained from the clinical charts of the patients and collected in the COVID registry of the S.E.N. Subsequently, they were included in an anonymized database on which the analysis was carried out.

The clinical research ethics committee of the Hospital Universitario La Paz (code PI-4332) approved this research as did the committees of each participating center.

The immunosuppression of patients was modified according to the clinical situation and the criteria of each center, there was no standard management protocol.

In general, immunosuppression was progressively reduced as the patient's clinical situation worsened and consisted, basically, of the reduction, suspension, or non-adjustment of immunosuppression:

• -

  In cases of SARS-CoV2 infection with mild or asymptomatic symptoms, immunosuppression was not modified, except in 4 centers where the adjuvant immunosuppressant (mycophenolic acid or m-TOR inhibitors) was reduced or discontinued.

• -

  -In cases of moderate infection with pneumonia and hypoxemia requiring oxygen therapy through nasal cannulas, the adjuvant immunosuppressant (mycophenolic acid or m-TOR inhibitors) was discontinued, the dose of calcineurin inhibitor was reduced, except in 3 centers in which both the adjuvant (mycophenolic acid or m-TOR inhibitors) and the main immunosuppressant were discontinued, and the dose of steroids was increased.

In cases of severe infection with pneumonia requiring mechanical ventilation, both the adjuvant immunosuppressant (mycophenolic acid or m-TOR inhibitors) and the main one were discontinued, and the dose of steroids was increased.

There was a percentage of patients in whom the initial suspension of the immunosuppressant did not materialize due to interactions of immunosuppressants with other drugs and who maintained their immunosuppressant levels in the blood during admission. This group of patients were considered, for the analysis of the management of immunosuppression, as patients in whom the medication was not adjusted or failed suspension.

Each center carried out treatment aimed at controlling the infectious disease according to its own protocols, which varied according to the scientific evidence available in each period of the pandemic.

In relation to antiviral treatment:

82.5% of patients (N = 226) received hydroxychloroquine during the first period of the pandemic while only one patient received it in the second period.

19.3% (N = 53) of patients received lopinavir/ritonavir in the first period and only 3 patients in the second period.

1.1% (N = 3) received remdesivir during the first period and increased to 5.9% (N = 20) during the second period.

In relation to anti-inflammatory treatment:

A 40.9% of patients (N = 112) received dexamethasone or methylprednisolone during the first period and 51.5% (N = 173) received it during the second period.

Interferon was used in 4% of patients (N = 11) in the first period and in a smaller proportion of patients in the second (N = 5, 1.5%).

Tocilizumab was used in 11% of patients (N = 30) in the first period and in 5.9% (N = 20) in the second period.

Anakinra was used in 11 patients, 8 during the first period and 3 patients in the second period.

Plasma from convalescents was used only in the second period of the pandemic in 7 patients (2.1%).

The results of the quantitative variables are expressed as mean ± standard deviation or as median. Comparison of means of continuous quantitative variables was performed using Student's t-test for paired and unpaired samples, as appropriate. If variables were not normally distributed, we used non-parametric tests: the Mann–Whitney U test to compare independent samples and in the case of paired samples, the Wilcoxon test was used to compare sets of 2 samples and the Friedman test when comparing more than 2 paired samples. The results of categorical variables are expressed with absolute and relative frequency. The statistical chi-square test or Fisher's exact test was used, as necessary, for the comparison of categorical variables. We treated p-value <0.01 as a statistically significant result. The statistical program SPSS version 20.0 was used for the analysis.

The predominant immunosuppressive (IS) treatment prior to COVID19 was triple therapy, being present in 68.3% (N = 420) of patients, followed by dual therapy (28.5%, N = 176) and tacrolimus monotherapy regimens in a very small proportion of patients (0.8%, N = 5).

The prednisone, tacrolimus and mycophenolic acid regimen was the majority (54.6%, N = 336), followed by the dual therapy regimen with tacrolimus and mycophenolic acid (10.2%, N = 63) and triple therapy with prednisone, tacrolimus, and m-TOR inhibitors (9.9%, N = 61). The rest of the guidelines were represented by a low proportion of patients, as shown in Fig. 1.

Fig. 1.

Immunosuppressive treatment prior to diagnosis of SARS-CoV2 infection of the 615 renal transplants included in the study. CSA: cyclosporine. i-mTOR: m-TOR inhibitors. MMF/mycophenolic acid Pred: prednisone.

(0.2MB).

After the diagnosis of SARS-CoV2 infection, treatment management basically consisted of the reduction, suspension, or non-adjustment of immunosuppression.

Fig. 2 shows the final management of each immunosuppressive drug, including in the graph the group of patients in whom the suspension of immunosuppression did not occur during admission as a result of the interactions of immunosuppressants with other drugs used (failed suspension) and the group of patients who received dexamethasone or pulses of methylprednisolone as part of the anti-inflammatory treatment of the disease.

Fig. 2.

Management of each immunosuppressive drug after diagnosis of SARS-CoV2 infection, as well as steroid treatment. Dexa: dexamethasone; iMTOR: m-TOR inhibitors; MMF: mycophenolic acid; MP: methylprednisolone.

(0.24MB).

Mycophenolic acid treatment was mostly discontinued: 73.8% (N = 332) of patients previously taking it (N = 450) discontinued treatment. Cyclosporine A was proportionally the drug that was most reduced, doing so in 36.7% (N = 11) of patients that were taking it before infection (N = 30). Tacrolimus was the least adjusted drug during infection remaining in 71.8% of patients (in 323 patients [54.8%] it was unchanged and in 66 patients [17%] the planned discontinuation failed). And m-TOR inhibitors were discontinued or not adjusted in the same proportion of patients (41.4% vs. 42.2% respectively, although in the group of patients without adjustment of the m-TOR inhibitor there was a high percentage of patients with failed suspension [24.5%]).

Baseline prednisone dose was increased in 50.9% of patients (N = 313) and of these a 66.5% also received dexamethasone or methylprednisolone pulses. In patients in whom the prednisone dose was not adjusted (N = 301, 49.1%), 26.8% (N = 81) received dexamethasone.

The levels of the immunosuppressants were monitored during admission in each patient and were lower, as expected, in patients in whom the dose was intentionally reduced compared to those who maintained their usual dose. The group of patients in whom the planned discontinuation of the immunosuppressive drug was unsuccessful had mean blood drug levels higher than those of patients in whom the medication was not adjusted (Table 2).

Immunosuppression was progressively introduced when the clinical situation of the patients allowed it, so that treatment with mycophenolic acid, tacrolimus, cyclosporin A or m-TOR inhibitors was restarted at 33, 17, 9 and 12 days on average, respectively. The usual prednisone dose was recovered at 36 days median (50.2 ± 48.8 days).

Six-point two percent of patients (N = 38) who, having discontinued some immunosuppressive drug (mainly mycophenolic acid) at the onset of the disease, did not recover from previous treatment once they clinically recovered from the infection.

Mean creatinine before COVID19 diagnosis, at diagnosis, at 7 and 15 days and at 2 and 6 months was 1.7 ± 0.8 mg/dl; 2.1 ± 1.2 mg/dl; 1.8 ± 1.1 mg/dl; 1.8 ± 1 mg/dl; 1.7 ± 0.9 mg/dl and 1.8 ± 1 mg/dl, respectively (p < 0.001).

Proteinuria was only available in most patients before diagnosis and at 2 and 6 months after diagnosis, with mean values of 217 ± 530 mg/g Cr, 220 ± 510 m g/g Cr and 178.1 ± 447 mg/g Cr, respectively (p = 0.64).

Fifty-seven percent of patients (N = 266) admitted to the hospital and 8.1% of patients (N = 12) followed on an outpatient basis presented with ARF. According to the ARF KDIGO classification, 78.8% (N = 219) were stage 1, 1.8% (N = 5) stage 2, and 16.5% (N = 46) stage 3. In 8 patients the stage of ARF could not be established. 8.6% of patients required renal replacement therapy (RRT) during the infectious disease process (N = 24).

Among admitted patients the most frequent cause of renal failure was functional impairment (70%, N = 186) followed by obstructive and immunological causes in 1.5% (N = 4) and 1.5% (N = 4), respectively; other causes were diagnosed in 4.1% (N = 11). In 22.9% (N = 61) of patients, no cause for deterioration was specified. Among outpatients ARF was of functional origin in all of them and all recovered their basel renal function.

There were 10 renal biopsies performed in the context of ARF and SARS-CoV2 infection. In 4 of them the histological diagnosis was of an immunological nature, in 2 of them acute tubular necrosis and in the other 4 the causes were toxicity by calcineurin inhibitors in 2 patients, cortical ischemia in one and chronic vascular lesions in another patient. There was no evidence of the presence of viral particles in any of the renal biopsies performed.

In the 4 patients with histological diagnosis of an immunological nature, T cell-mediated rejection was observed in 3 of them and in one of them had active antibody-mediated rejection. The patients who developed cell rejection had been transplanted for 15 days, one month and 4.7 months, which means an incidence of cell rejection in the first 6 months after transplantation of 6.9%; while the patient with humoral rejection had been transplanted for 7.2 years. Management of immunosuppression during SARS-CoV2 infection in these patients consisted of in discontinuation of mycophenolic acid in all 4 patients. Tacrolimus treatment was not discontinued in any of them (2 of them maintained the same dose of tacrolimus and the other 2 had their dose reduced). In relation to steroids, 3 of them had their dose increased. Only one of the patients with acute rejection presented with donor-specific anti-HLA antibodies within 2 months of COVID19 diagnosis. In the other 3 patients, in one of them the anti-HLA antibodies remained negative at 2 and 6 months after the diagnosis of COVID, in another patient the anti-HLA antibodies became negative after the treatment received, and the other patient maintained positive anti-HLA antibodies.

In hospitalized patients, renal function was recovered in 93.2% of patients with ARF (N = 205), in 2.3% it was partially recovered, and 3.6% (N = 8) did not recover from acute kidney damage.

The average hospital stay of patients with ARF was significantly higher than that of patients without ARF (22.1 ± 22.3 days vs. 14.2 ± 12.1 days, p < 0.001). Patients who developed ARF had significantly higher mean creatinine values before diagnosis of coronavirus infection, at diagnosis and at 2 and 6 months post-infection than patients who did not develop acute kidney damage (before: 2 ± 0.9 vs. 1.5 ± 0.5; at diagnosis: 2.7 ± 1.3 vs. 1.6 ± 0.8; at 2 months: 2.1 ± 1.1 vs. 1.5 ± 0.6; at 6 months: 2.1 ± 1 vs. 1.5 ± 0.7, p < 0.001, respectively). Likewise, during the infection process, patients admitted with ARF presented a mean maximum value of creatinine significantly higher than that of patients admitted without ARF (3.1 ± 1.5 vs. 1.6 ± 0.8, p < 0.001), with the mean peak creatinine values according to the etiology of ARF being the following: 3.1 ± 1.4 in functional cases, 4.1 ± 3.1 in obstructive cases, 6.1 ± 3.1 in immunological cases, 5.1 ± 2.1 in other causes and 3.5 ± 1.2 in cases of ARF without specified cause. In the case of patients followed as outpatients, the mean maximum creatinine during infection was 2.4 ± 1 in patients with ARF and 1.7 ± 1.2 in patients who did not develop ARF, but these differences were not statistically significant (p = 0.13). The group of patients who required RRT during infection had a significantly higher baseline creatinine before the diagnosis of COVID19 than patients who did not require RRT (4.04 ± 1.9 vs. 2.6 ± 1.2, p < 0.001).

Twelve patients (1.9%) lost renal graft and 11 of them had presented with an ARF during admission. The causes of graft loss were: graft CKD progression in 5 patients, vascular in 2 patients, immunological in one patient, and other causes in 4 patients. The mean baseline creatinine of these 12 patients was 3.8 ± 2 mg/dl (minimum 1.19 mg/dl–maximum 6.99 mg/dL). Both the group of patients who required TRS-CHK and the group of patients who lost the graft during the course of the infectious disease already had advanced renal function before diagnosis, which could influence these results.

Renal function was also analyzed in the group of patients on tacrolimus receiving antiviral treatment with lopinavir/ritonavir (N = 46). Of these patients, only 36 had data on immunosuppression levels during admission. A tacrolimus trough level ≥10 ng/ml was considered as a cutoff point for high tacrolimus levels. Eighteen patients treated with lopinavir/ritonavir presented high levels of tacrolimus and of these 77.8% (N = 14) presented ARF with maximum creatinine values during admission of 2.2 ± 1.1 mg/dl, while the other 18 patients presented tacrolimus levels <10 ng/mL, representing 83.3% (N = 15) of ARF patients with maximum creatinine figures of 2.8 ± 1.4 mg/dl (there was no significant difference between maximum creatinine concentrations in the two groups).

Fifty-six point nine percent of patients (N = 350) had anti-HLA antibody monitoring before and after SARS-CoV2 infection and in 94% of these patients (N = 329) presented no changes in anti-HLA antibodies after infection.

Twenty-one patients (6%) tested positive for anti-HLA after infection, of which 9 (2.6%) patients had donor-specific anti-HLA antibodies. Of these patients, 7 had discontinued an immunosuppressive drug during infection (mycophenolic acid was discontinued in 5 of them and tacrolimus discontinued in 2), one patient had discontinued the 2 immunosuppressive drugs and the other patient had not changed in immunosuppression.