Dialysis parameters associated with SARS-CoV-2 infection and prognosis in end-stage kidney disease

Abstract

Background

The Covid-19 pandemic has affected patients with end-stage kidney disease (ESKD). Whether dialysis parameters have a prognostic value in ESKD patients with Covid-19 remains unclear.

Materials and Methods

We retrospectively evaluated clinical characteristics, blood pressure (BP) and dialysis parameters in ESKD patients undergoing maintenance outpatient hemodialysis, with (Covid-ESKD) and without (No-Covid-ESKD) Covid-19, at four Brazilian hemodialysis facilities. The Covid-ESKD (n = 107; 54% females; 60.8 ± 17.7 years) and No-Covid-ESKD (n = 107; 62% females; 58.4 ± 14.6 years) groups were matched by calendar time. The average BP and dialysis parameters were calculated during the pre-infection, acute infection, and post-infection periods. The main outcomes were Covid-19 hospitalization and all-cause mortality.

Results

Covid-ESKD patients had greater intradialytic and postdialysis systolic BP and lower predialysis weight, postdialysis weight, ultrafiltration rate, and interdialytic weight gain during acute-illness compared to 1-week-before-illness, while these changes were not observed in No-Covid-ESKD patients. After 286 days of follow-up (range, 276–591), there were 18 Covid-19-related hospitalizations and 28 deaths among Covid-ESKD patients. Multivariable logistic regression analysis showed that increases in predialysis systolic BP from 1-week-before-illness to acute-illness (OR, 95%CI = 1.06, 1.02–1.10; p = .004) and Covid-19 vaccination (OR, 95%CI = 0.16, 0.04–0.69; p = .014) were associated with hospitalization in Covid-ESKD patients. Multivariable Cox-regression analysis showed that Covid-19-related hospitalization (HR, 95%CI = 5.17, 2.07–12.96; p < .001) and age (HR, 95%CI = 1.05, 1.01–1.08; p = .008) were independent predictors of all-cause mortality in Covid-ESKD patients.

Conclusion

Acute Covid-19 illness is associated with variations in dialysis parameters of volume status in patients with ESKD. Furthermore, increases in predialysis BP during acute Covid-19 illness are associated with an adverse prognosis in Covid-ESKD patients.

KEY MESSAGES

1. Dialysis parameters were influenced by SARS-CoV-2 infection and may have prognostic value in patients with Covid-19.

2. Increases in blood pressure during acute Covid-19 illness and the lack of vaccination for Covid-19 were predictors of hospitalization for Covid-19.

3. Hospitalization for Covid-19 and age were independent risk factors for all-cause death.

Introduction

End-stage kidney disease (ESKD) patients constitute a vulnerable group with a higher risk of infection by SARS-CoV-2 due to immune dysfunction and an elevated comorbidity burden [1,2]. Among maintenance hemodialysis patients with Covid-19 (Covid-ESKD), the risk of death may increase by 21.1 times [3]. Consequently, the effect of the pandemic has been catastrophic in this population, leading to an unprecedented reduction in the number of patients [4, 5]. Respiratory symptoms are the leading cause of hospitalization among patients with Covid-19, while acute respiratory distress syndrome and acute cardiac injury constitute the main complications of SARS-CoV-2 infection [6].

In Covid-ESKD patients, male sex, higher age, poor blood pressure (BP) control, and higher Charlson comorbidity index have been associated with hospitalization and death [3, 7, 8]. Previous studies also demonstrated that fluid overload has prognostic impact in ESKD patients without Covid-19 (No-Covid-ESKD) [9–11]. Although fluid gains are expected during the interdialytic period in maintenance hemodialysis patients, greater fluid retention is associated with increased BP, hospitalization, and mortality [12–14]. Additionally, volume overload may also contribute to pulmonary dysfunction, including interstitial edema and airway obstruction [15, 16]. In Covid-ESKD patients, the tolerance to fluid gain might be reduced because the SARS-CoV-2 virus can cause lung inflammation with consequent alveolar exudation and myocardial dysfunction with consequent alveolar transudation [17, 18]. However, the association of dialysis parameters with SARS-CoV-2 infection and Covid-19 prognosis remains unclear; hence, this study aims to fill this gap.

Materials and methods

Study design

This multicenter observational study retrospectively evaluated ESKD patients from four outpatient dialysis facilities in Recife, Brazil, from March 5, 2020 to August 6, 2022. The inclusion criteria were age >18 years old and on thrice-weekly dialysis for at least 3 months. We initially identified 114 ESKD patients with symptomatic and confirmed SARS-CoV-2 infection (Covid-ESKD) during the study period, but seven were excluded because they did not undergo hemodialysis when the Covid-19 diagnosis was made, leaving 107 Covid-ESKD patients for the analysis. Confirmed cases of Covid-19 were defined as a positive RT-PCR assay result from a nasopharyngeal swab. Each Covid-ESKD patient was arbitrarily matched with a No-Covid-ESKD patient at the same calendar time (1:1) (Supplementary Figure S1). The Institutional Review Board of the Oswaldo Cruz University Hospital/Pronto Socorro Cardiológico de Pernambuco (PROCAPE) Ethics Committee approved this study (CAAE 61782422.3.0000.5192) and waived the requirement for informed consent.

Variables

Baseline clinical variables were collected at the time of Covid-19 symptoms onset in Covid-ESKD participants and at the same calendar time in matched No-Covid-ESKD participants and comprised the following data: sex, age, body mass index, chronic kidney disease etiology, hemodialysis vintage, epoetin use, Covid-19 vaccination, history of hypertension, previous stroke, coronary heart disease, diabetes mellitus, heart failure, obesity, lung disease, HIV/AIDS, malignancy, and rheumatologic disease. These data were also used to create the Charlson Comorbidity Index Score [19]. The patients were considered vaccinated for Covid-19 if they received at least one dose of any Covid-19 vaccine.

Data on predialysis BP (one measurement obtained before dialysis), intradialytic BP (mean of all intradialytic BP measurements obtained with 30-min intervals), postdialysis BP (one measurement obtained after dialysis), ultrafiltration (UF) rate, interdialytic weight gain (IDWG), dry weight (DW), predialysis weight and postdialysis weight, were collected from all dialysis sessions of five different periods: 1) from 60 to 30 days before Covid-19 symptoms onset (2-months-before-illness); 2) from 7 days to the day before Covid-19 symptoms onset (1-week-before-illness); 3) from the day of Covid-19 symptoms onset to 14 days after (acute-illness); 4) from 30 to 60 days after the end of Covid-19 symptoms (1-month-after-illness); and 5) from 90 to 120 days after the end of Covid-19 symptoms (3-months-after-illness). The first three periods were also considered for the No-Covid-ESKD participants, who were matched to the Covid-ESKD participants during the same calendar time. The values of predialysis, intradialytic, and postdialysis BP, UF rate, IDWG, DW, predialysis weight, and postdialysis weight for each patient in the aforementioned periods were calculated as the mean of all corresponding measures performed during the time span of each period.

Trained dialysis technicians or nurses obtained BP measurements from sitting patients using an oscillometric BP monitor lodged in the dialysis machines (Fresenius 4008; Fresenius). Data on antihypertensive medication use and serum hemoglobin levels during the study period were also collected. The assistant nephrologist determined the patient’s DW based on a physical examination and the patient’s ability to tolerate weight. The signs and symptoms used to determine this tolerance were edema, cramps, need for saline solution during the hemodialysis session, or symptomatic hypotension [20]. Body bioimpedance spectroscopy (BCM – Body Composition Monitor IFU-PT 12 A-2018, Fresenius Medical Care LTDA) was used as a complementary method to adjust the DW when the nephrologist deemed necessary. IDWG was calculated as the patient’s weight at the beginning of each hemodialysis session (predialysis weight) minus the weight after each session (postdialysis weight) divided by DW, and expressed as a percentage of change. The UF rate was calculated as follows: (predialysis weight–postdialysis weight [mL])/delivered treatment time (h)/postdialysis weight (kg).

Outcomes

The main outcomes were Covid-19 hospitalization and all-cause deaths censored to November 1, 2022. The clinical status was assessed by medical chart review, and the start of follow-up was considered as the day of Covid-19 symptoms onset for each Covid-ESKD patient, and the same calendar time for a matched No-Covid-ESKD patient. Covid-19 hospitalization was considered when hospital records indicated that admission was associated with Covid-19 infection. Death and the cause of death were ascertained through a medical chart review.

Statistical analysis

Continuous variables with a normal distribution are presented as mean ± standard deviation, whereas those without a normal distribution are presented as median (25th and 75th percentiles). Categorical variables are presented as proportions. The following statistical tests were used to compare variables between the two groups: Student’s t-test for continuous variables with normal distribution, Mann–Whitney U test for continuous variables without normal distribution, and Chi-square test for categorical variables. One-way analysis of variance for repeated measures followed by the Holm-Sidak method for pairwise comparisons and the Friedman’s test were used to compare differences in the variables with normal and non-normal distributions, respectively, throughout the studied time points within the Covid-ESKD and No-Covid-ESKD groups. Multivariable logistic regression analysis was used to evaluate the variables independently associated with Covid-19 hospitalization in the Covid-ESKD group. Age, sex, and variables that were significantly different between hospitalized and non-hospitalized patients were included as independent variables in the multivariate logistic model. The cumulative mortality rate was calculated using the Kaplan–Meier method and comparisons between the curves were made using the log-rank test. Multivariable Cox regression analysis was used to evaluate the variables associated with death in Covid-ESKD patients. The adjusted Cox regression model included age, sex, and variables that were significantly different between patients who died and those who did not as independent variables. The proportional hazards assumption was tested for the model, and a violation was detected in the association between Covid-19 hospitalization and death. Therefore, we also assessed the relationship over more restricted time intervals by dividing the follow-up period into less than or more than 90 days. Statistical significance was set at p < 0.05. Analyses were performed using IBM SPSS Statistics 28.0.1 and Stata software V.14.2 (Stata Corp LP, College Station, Texas, USA).

Results

Characteristics of Covid-ESKD and No-Covid-ESKD participants

Covid-ESKD (n = 107; 62% males; age = 60.8 ± 17.7 years) and No-Covid-ESKD (n = 107; 54% males; age = 58.4 ± 14.6 years) patients had similar baseline characteristics, except for a lower Covid-19 vaccination rate in the Covid-ESKD group (59% vs. 72%; p = .044) (Table 1). Results of repeated measures analysis showed that Covid-ESKD patients had greater intradialytic and postdialysis systolic BP (SBP) and lower predialysis weight, postdialysis weight, UF rate, and IDWG during acute illness compared to 1-week-before-illness, while these changes were not observed in No-Covid-ESKD patients (Table 2). Furthermore, no significant differences were found in the prescription of antihypertensive medications before and after acute illness among surviving Covid-ESKD patients (Supplementary Table S1).

Table 1.

Baseline Characteristics of Covid-ESKD and No-Covid- ESKD patients.

Variables	Covid-ESKD (n = 107)	No-Covid-ESKD (n = 107)	p-value
Age, years	60.8 ± 17.7	58.4 ± 14.6	0.29
Male sex, %	61.7	54.2	0.27
Body mass index, kg/m2	25.1 ± 4.9	25.3 ± 5.5	0.78
Dialysis vintage, months	34 [12–72]	39 [11–73]	0.35
Chronic kidney disease etiology, %			0.74
Diabetes	31.8	31.8
Hypertension	18.7	12.1
Glomerulonephritis	15.9	13.1
APKD	2.8	6.5
Unknown	23.2	25.3
Others	7.5	11.2
Hypertension, %	86.0	93.5	0.07
Diabetes, %	42.1	43.0	0.89
Coronary Heart Disease, %	20.6	27.1	0.26
Heart Failure, %	19.6	21.5	0.74
Stroke, %	8.4	5.6	0.42
Obesity, %	8.4	16.8	0.06
Lung Disease, %	5.6	5.6	1.00
Charlson Comorbidity Index	4 [2-5]	4 [2-5]	0.74
Epoetin, %	87.9	85.0	0.55
Insulin, %	20.6	28	0.20
Antiplatelets, %	26.2	30.8	0.45
Statins, %	38.7	49.5	0.11
N° of antihypertensive medications	1 [0-2]	1 [1-2]	0.12
Antihypertensive medications, %	73.8	80.4	0.25
ACE inhibitors	1.9	6.5	0.09
Angiotensin receptor blockers, %	32.7	31.8	0.88
Calcium-channel blockers, %	26.2	31.8	0.37
Beta-blockers, %	44.9	46.7	0.78
Furosemide, %	13.1	10.3	0.52
Clonidine/Methyldopa, %	13.1	16.8	0.44
Hydralazine/Minoxidil, %	10.3	12.1	0.66
Covid-19 vaccine, %	58.9	72.0	0.044

Student’s t-test, Mann-Whitney U test, and chi-square test were used to compare continuous variables with normal distribution, continuous variables without normal distribution, and categorical variables, respectively.

APKD – Autosomal dominant polycystic kidney disease; ACE – angiotensin-converting enzyme; ESKD –end-stage kidney disease.

Table 2.

Repeated measures analyses of dialysis parameters and antihypertensive medications in Covid-ESKD and No-Covid-ESKD patients.

	Covid-ESKD (n = 107)				No-Covid-ESKD (n = 107)
Variables	2-months-before-illness	1-week-before-illness	acute-illness	p-value*	2-months-before-illness	1-week-before-illness	acute-illness	p-value*
Blood pressure (mmHg)
Predialysis SBP	147.8 ± 21.9	147.3 ± 25.8	149.7 ± 22.9	0.21	149.6 ± 19.9	149.7 ± 21.5	151.7 ± 21.7	0.22
Predialysis DBP	74.8 ± 11.3	75.4 ± 13.9	75.2 ± 12.5	0.78	75.3 ± 12.8	76.0 ± 13.1	75.7 ± 14.6	0.73
Intradialytic SBP	142.2 ± 20.1†	142.6 ± 19.9†	147.9 ± 19.3	<0.001	147.0 ± 18.7	146.0 ± 21.5	147.9 ± 21.7	0.25
Intradialytic DBP	73.4 ± 10.9	73.7 ± 11.4	75.2 ± 11.3	0.05	74.3 ± 10.8	74.5 ± 12.4	74.4 ± 12.4	0.97
Postdialysis SBP	142.5 ± 19.8†	141.9 ± 20.7†	147.8 ± 19.4	<0.001	146.0 ± 16.7	147.0 ± 20.2	147.3 ± 19.7	0.54
Postdialysis DBP	73.6 ± 10.6†	74.7 ± 11.6	76.4 ± 12.0	0.018	75.7 ± 8.8	76.3 ± 10.7	75.8 ± 10.7	0.70
UF Rate. mL/kg/h	7.2 ± 3.1†	7.2 ± 3.8†	6.5 ± 4.0	0.015	7.6 ± 3.3	7.9 ± 3.6	7.8 ± 3.6	0.30
Dry weight. kg	68.9 ± 17.7	68.9 ± 17.7	68.8 ± 17.8	0.67	70.1 ± 17.3	70.1 ± 17.4	70.1 ± 17.5	0.73
Predialysis weight. kg	71.7 ± 18.6†	71.6 ± 18.7†	71.0 ± 18.5	<0.001	72.9 ± 18.1	72.9 ± 18.3	73.0 ± 18.2	0.85
Postdialysis weight. kg	69.7 ± 18.3	69.6 ± 18.4†	69.3 ± 18.3	0.036	70.7 ± 17.7	70.7 ± 17.8	70.8 ± 17.7	0.70
IDWG. %	2.9 ± 1.4†	3.0 ± 1.8†	2.5 ± 1.7	0.006	3.14 ± 1.41	3.2 ± 1.7	3.1 ± 1.6	0.88
N° of antihypertensive medications	1 [0-2]	1 [0-2]	1 [0-2]	0.89	1 [1-2]	1 [1-2]	1 [1-2]	0.93
Hemoglobin. g/dL	10.9 ± 1.7	11.1 ± 1.8	11.0 ± 1.7	0.30	11.4 ± 1.7	11.5 ± 1.6	11.5 ± 1.6	0.73

^*One-way analysis of variance for repeated measures and the Friedman’s test were used to compare variables with normal and non-normal distributions within each group.

^†p < 0.05 compared to acute-illness period using the Holm-Sidak method.

SBP, systolic blood pressure; DBP, diastolic blood pressure; UF, ultrafiltration; IDWG, interdialytic weight gain.

Covid-19 hospitalization

Among the 107 Covid-ESKD patients, 18 (17%) were hospitalized due to Covid-19. Notably, all Covid-ESKD patients who were hospitalized reported experiencing dyspnea before hospitalization. Covid-ESKD patients who were hospitalized were older (68.7 ± 15.7 vs. 59.2 ± 17.7 years; p = .037) , less likely to be vaccinated for Covid-19 (27.8 vs. 65.2%; p = .003), had a higher prevalence of heart failure (38.9 vs. 15.7%, p = .024), and were more likely to have lower UF rate (4.6 ± 3.3 vs. 6.9 ± 4.0 mL/Kg/h; p = .025) and a trend toward lower IDWG (1.9 ± 1.6 vs. 2.7 ± 1.7; p = .08) during acute-illness than those who did not require hospitalization (Supplementary Table S2). When comparing the dialysis parameters of volume status between acute-illness and 1-week-before-illness, hospitalized patients had higher increases in predialysis SBP (p = .002) and intradialytic SBP (p = .048) from 1-week-before-illness to acute-illness than those who were not hospitalized (Supplementary Table S3).

Further results of unadjusted logistic regression analysis in Covid-ESKD patients showed that Covid-19 hospitalization was inversely associated with UF rate (odds ratio, 95%CI = 0.82, 0.69–0.98; p = .026) at acute-illness, and directly associated with the difference in predialysis SBP (ΔpredialysisSBP) (odds ratio, 95%CI = 1.05, 1.01–1.08; p = .004) between acute-illness and 1-week-before-illness (Figure 1).

Figure 1.

Univariate logistic regression between hospitalization due to Covid-19 and relevant hemodialysis variables among Covid-ESKD patients.

Multivariable logistic regression analysis including age, sex, heart failure, Covid-19 vaccination, UF rate at acute-illness and ΔpredialysisSBP between acute-illness and 1-week-before-illness demonstrated that only ΔpredialysisSBP between acute-illness and 1-week-before-illness (odds ratio, 95%CI = 1.06, 1.02–0.10; p = .004) and Covid-19 vaccination (odds ratio, 95%CI = 0.16, 0.04–0.69; p = .014) were independently associated with Covid-19 hospitalization (Supplementary Table S4).

All-cause death

During a median follow-up of 286 days (range, 276–591), there were 28 (26%) and 4 (4%) deaths among Covid-ESKD and No-Covid-ESKD patients, respectively, which resulted in an unadjusted mortality hazard ratio of 7.55 (95%CI, 2.65–21.54) (p < .001). As observed in the Kaplan–Meier curves (Figure 2), some of the excessive deaths in Covid-ESKD patients occurred in the first 90 days of follow-up, even though the mortality curves of the studied groups continued to progressively diverge afterwards. Covid-ESKD patients died due to Covid-19 infection (n = 12), cardiovascular diseases (n = 5), infection/sepsis (n = 7), fatal bleeding (n = 3), and cancer (n = 1), whereas non-Covid-ESKD patients died due to cardiovascular diseases (n = 3) and sepsis (n = 1).

Figure 2.

Kaplan-Meier curves for all-cause mortality.

Covid-ESKD patients who died were more likely to be older (71.4 ± 13.5 vs. 57.0 ± 13.5 years; p < .001), had higher risk of heart failure (35.7 vs. 17.9%; p = .013), previous stroke (21.4 vs. 3.8%; p = .004), lung disease (14.3 vs. 2.5%; p = .020) and higher Charlson Comorbidity Index (5 [3–6] vs. 3 [2–4]; p = .002), to use statins (60.7 vs. 30.4%; p = .005) and beta-blockers (64.3 vs. 38.0%; p = .016), to be hospitalized due to Covid-19 (42.9 vs. 7.6%; p < .001), and to have lower Covid-19 vaccination rate (35.7 vs. 67.1%; p = .004) and UF rate at acute-illness (4.7 ± 2.9 vs. 7.1 ± 4.2 mL/Kg/h; p = .006) than those who survived (Supplementary Table S5). Conversely, there were no differences in the dialysis parameters of volume status between 1-week-before-illness and acute-illness between Covid-ESKD patients who died or survived (Supplementary Table S6).

Results of multivariable Cox-regression analysis including age, sex, heart failure, stroke, lung disease, Charlson Comorbidity Index, hospitalization due to Covid-19, Covid-19 vaccination and UF rate at acute-illness showed that only Covid-19 hospitalization (hazard ratio, 95%CI = 5.17, 2.07–12.96; p < .001) and age (hazard ratio, 95%CI = 1.05, 1.01–1.08; p = .008) were independently associated with all-cause mortality in Covid-ESKD patients (Supplementary Table S7). However, violation of the proportional hazard assumption (p < .001) was observed for the association between Covid-19 hospitalization and death, such that the risk associated with Covid-19 hospitalization was greater during the early follow-up period. Indeed, adjusted Cox-regression analyses showed that Covid-19 hospitalization was associated with a higher risk of death up to 90 days of follow-up (hazard ratio, 95%CI = 15.53, 4.04–59.62; p < .001) but not afterwards (hazard ratio, 95%CI = 0.26, 0.03–2.22; p = .22).

To further summarize the results, the combined predictors of Covid-19 hospitalization and all-cause death are shown in a single direct acyclic graph (Figure 3).

Figure 3.

Directed acyclic graph (DAG) showing the predictors of the studied outcomes.

ΔpredialysisSBP – difference in predialysis systolic blood pressure between acute-illness and 1-week-before-illness.

Discussion

The present study evaluated the clinical characteristics and dialysis parameters of volume status associated with SARS-CoV-2 infection and Covid-19 prognosis in a population undergoing maintenance hemodialysis and provided three major results. First, Covid-ESKD patients had increased intradialytic and postdialysis SBP and decreased predialysis weight, postdialysis weight, UF rate, and IDWG during acute illness compared to 1-week-before-acute-illness. Second, among Covid-ESKD patients, the lack of SARS-CoV-2 vaccination and larger increases in predialysis SBP during acute illness compared to 1-week-before-acute-illness were independently associated with hospitalization due to Covid-19. Third, hospitalization due to Covid-19 and age were independent predictors of all-cause mortality in Covid-ESKD patients. In general, the present findings indicate that dialysis parameters of volume status are influenced by SARS-CoV-2 infection and may have a prognostic value in patients with Covid-19.

One major finding of the present study was that acute Covid-19 illness was associated with variations in dialysis parameters of volume status compared with 1-week-before-illness. Although Covid-ESKD patients did not show a significant change in predialysis SBP during the acute illness, they showed an increase in intradialytic and postdialysis SBP coupled with a decrease in the UF rate during that period. The absence of the expected reduction in intradialytic and post-dialysis SBP, along with a decreased UF rate indicate that Covid-19 patients still have too much fluid in their body. In addition, their pre- and post-dialysis weights and IDWG decreased, whereas their DW remained unchanged during acute illness. These findings suggest that Covid-ESKD patients might have experienced an overestimation of DW during acute illness due to total weight loss that did not translate into an equivalent DW adjustment. Previous data have shown that patients with acute Covid-19 may experience weight reduction [21, 22]. Furthermore, among maintenance hemodialysis patients, those who do not gain as much weight as usual between dialysis sessions are considered more likely to have an undetected SARS-CoV-2 infection [23]. Therefore, weight change seems to be a common manifestation of SARS-CoV-2 infection among patients undergoing maintenance hemodialysis and should be systematically evaluated.

Our analysis also revealed that Covid-ESKD patients who were hospitalized had a lower UF rate during acute illness along with greater increases in SBP from 1-week-before-illness to acute-illness compared to those who were not hospitalized. Importantly, predialysis SBP variation from 1-week-before-illness to acute-illness was an independent and robust predictor of hospitalization. Considering that BP is a parameter usually associated with volume overload in hemodialysis patients and that UF rate is usually defined before dialysis onset based on DW estimation, these findings suggest that a subclinical state of hypervolemia not identified by the nephrologist’s clinical judgment during acute Covid-19 illness might have contributed to the hospitalization of patients [20]. In this regard, there is robust evidence that hypervolemia is associated with worse outcomes among No-Covid-ESKD patients [12, 14]. Additionally, Covid-19 patients frequently exhibit intense pulmonary inflammation and may develop pulmonary interstitial edema even under small increases in intravascular volume [17, 24]. In agreement with these assumptions, we found that all Covid-ESKD patients who were hospitalized reported experiencing dyspnea before hospitalization. Conversely, it could be argued that the lower UF rate in hospitalized patients might be a result of hemodynamic instability and hypotension during dialysis due to SARS-CoV-2 infection. However, this assumption seems improbable because Covid-ESKD hospitalized patients had an elevated average SBP (>140 mmHg) throughout the intradialytic and postdialysis periods during acute illness. Taken together, these findings have several clinical implications. First, they suggested that ESKD patients who experienced marked increases in predialysis BP during acute illness in comparison with 1-week-before-illness might be hypervolemic and at a higher risk of adverse evolution and, therefore, would require further DW and UF rate adjustments. Second, they underscored how challenging it is to accurately estimate DW among patients undergoing maintenance hemodialysis under acute disease scenarios. In this context, the use of alternative strategies, rather than the usual approach currently recommended to estimate DW in clinical practice, may be necessary to provide adequate fluid removal in this population [25, 26].

Some of the results of this study require further comments. First, Covid-19 vaccination was associated with an 84% reduction in Covid-19 hospitalizations in the COVID-ESKD group. This finding is similar to that reported by Ashby et al. who showed a 75% lower risk of Covid-19 hospitalization after vaccination in alternative ESKD patients, underscoring the vaccine’s protective effect against severe Covid-19 disease [27]. Second, Covid-19 hospitalization and age were independently associated with all-cause death, thus confirming that these variables are significant predictors of adverse prognosis in Covid-19 disease [28–30]. Third, we observed sustained increased mortality after the acute phase of Covid-19 in the Covid-ESKD group relative to the No-Covid-ESKD group, which reproduces data obtained in ESKD and national cohorts and reinforces the long-term deleterious effects of SARS-CoV-2 infection [31, 32]. Notably, most deaths not directly related to Covid-19 in the Covid-ESKD group were due to cardiovascular diseases and non-Covid infections, which is consistent with previous studies [31]. Overall, the similarities of our results with data reported in the literature strengthen the validity of our findings.

This study has some limitations. First, this was an observational retrospective study; therefore, the reported associations may not be causal. Second, the relatively small sample size may have limited the ability to detect further relevant associations. Third, volemic status was not evaluated using complementary methods during the acute phase of the disease. Fourth, the inclusion of controls was arbitrary, which might have potentially led to inclusion bias. However, to the best of our knowledge, this is the first study that comprehensively evaluated the BP behavior and dialysis parameters of volume status in Covid-ESKD patients from several months before up to several months after the onset of acute Covid-19 illness.

Conclusions

In conclusion, increased BP during acute Covid-19 illness and lack of vaccination for Covid-19 are predictors of hospitalization for Covid-19 among Covid-ESKD patients, while hospitalization and age are risk factors for all-cause death. These data might help identify high-risk dialysis patients who would require closer surveillance during acute Covid-19 illness.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Authors contributions

Rodrigo Bezerra: Conceptualization, data curation, investigation, methodology, formal analysis, resources, validation, writing, review, and editing.

Audes D.M. Feitosa, Odilson M. Silvestre, Miguel M. Fernandes-Silva, Roberto B. Amazonas, Flavio Teles, Cibele Isaac Saad, Jose Luiz Lima Filho, Andrei C. Sposito: Visualization, data analysis and interpretation, writing – review & editing.

Wilson Nadruz Jr: Conceptualization, data curation, formal analysis, methodology, project administration, resources, visualization, writing – original draft, writing – review, and editing.