The present study described, for the first time, in a Brazilian university hospital, the results of effectiveness and safety indicators in RCA with a 4% sodium citrate solution in prolonged HD with the Hybrid Genius® System in critically ill patients with AKI.

In the present study, the outcomes of length of stay (30.7 days), general in-hospital mortality (84.9%), and in-hospital mortality in COVID-19 (93.1%) were high. The proportion of patients who achieved KDIGO stage 3 AKI was 67.4%. A prospective cohort carried out at a public university hospital in São Paulo in the first 90 days of the pandemic evaluated 101 patients hospitalized with COVID-19, 51.9% of which required an ICU. Of these critically ill patients, 77.3% developed AKI, and the third KDIGO stage was the most frequent (58.9%). Acute RRT was indicated in 61.5% of patients, and mortality was 65.4% [21].

In the present study, most patients (84.9%) had AKI on admission to the ICU, denoting a probable more severe clinical picture of the studied sample. This sample consisted of 29 patients with COVID-19, representing 59.18% of the total. Of these, 68.96% were receiving vasoactive medications, and 82.7% were on mechanical ventilation. Therefore, a higher occurrence of mortality is expected.

Ionic calcium increased slightly post-session (4.39–4.56 mEq/l) but remained in the normal range (4.5–5.3 mg/dl). The dialysis solution available at the service contains calcium, but calcium chloride infusion probably influenced these values. Calcium monitoring when using RCA is justified because this medication reduces iCa in the extracorporeal circuit (since it is an essential cofactor of the coagulation cascade). Up to 50% of the citrate-calcium complex is removed through the hemofilter during the first pass. Then, calcium gluconate infusion may be necessary to avoid a negative balance and compensate for calcium losses. However, not all protocols recommend this infusion [13].

Urea was above the reference value (19–43 mg/dl) in the pre-session and urea increased in post-session concentration (107–114 mg/dl). Among the possible limitations that justify it is the time of test collection, which did not occur shortly after the end of dialysis (recalling that it occurred during the COVID-19 pandemic). This limitation in this retrospective study is due to the fact that there is no research protocol. In this regard, it is also known that urea is a marker of nutritional performance. HD patients have a high prevalence of malnutrition [22]. In general, critically ill patients may have 20% muscle loss in the first 10 days of hospitalization [23].

Is important to note that the Hybrid Genius® System is a unique hemodialysis system that has the ability to keep the fresh and waste dialysate separated, despite both being stored on a unique reservoir. This is possible because of the dialysate tank, a glass container with a capacity to store 90 L of dialysis solution. Both dialysates, the used and the fresh, are kept in the same tank, but they do not mix. The difference in density and temperature is what keeps both liquids separated, were the wasted dialysate will remain on the bottom of the tank with the fresh one above it [6].

The main disadvantage of RCA use is metabolic and electrolyte disturbances (hypocalcemia, metabolic alkalosis, and hypernatremia) [24]. As shown in Table 2, the analysis of safety indicators showed that post-session sodium and bicarbonate concentrations remained within reference values. Dysnatremia occurred in 15% of sessions, and metabolic alkalosis was observed in 9.4% of those.

In the present study, the median flow of 4% sodium citrate infused in patients was 260 ml/h, with a citrate dose of 30 mmol/h. The median flow of the calcium chloride replacement solution was 22 ml/hour, with a calcium dose of 1.84 mmol/h. The occurrence of hypernatremia was evidenced in 1 session (1.8%). In the retrospective study by Wen et al., which evaluated 808 sessions with RCA with a 30% citrate solution, 1 HD session was interrupted due to sodium disturbance [11]. In this regard, the literature describes hypernatremia as a potential adverse event of RCA, with an uncommon occurrence when using a dialysate or replacement fluid with lower sodium concentrations or both [25, 26]. The infusion of concentrated sodium citrate resulted in sodium overload for the patient (420 mmol/l in a 4% solution and 3,060 mmol/l in a 30%), but this was not a significant problem in our study. Even using a dialysate with 135 mmol/l of sodium, dysnatremia was uncommon, hypernatremia being a rare event. In addition, the sodium content of other infusions and intravenous fluids, in addition to enteral feeding, must be considered to maintain the patient’s electrolyte balance [12].

Another potential adverse event in RCA is metabolic alkalosis. As previously described, the filter removes up to 50% of the citrate-calcium complexes. Those that enter the systemic circulation dissociate, and the Krebs cycle metabolizes citrate (mainly in liver cells). In the end, one citrate molecule produces energy and three bicarbonate molecules. Therefore, sodium citrate acts as both an anticoagulant and a buffer base. Thus, the acid-base balance of patients should be monitored since there is a risk of metabolic alkalosis [13, 14].

In the study by Wen et al., four interruptions of the dialysis session due to electrolyte and metabolic disturbances were observed, two of which were metabolic acidosis. This study reported no interruption due to alkalosis [11]. In the present study, the bicarbonate concentration increased after the session (25.5–25.85 mmol/l), remaining within the reference values ​​(22–28 mmol/l). However, metabolic alkalosis was observed in 5 sessions, and there was no detection of acidosis occurrence.

Regarding the effectiveness indicators shown in Table 3, system coagulation occurred in 17.3% of the HD sessions. A similar result was described by Schneider et al., which showed 19% coagulation with RCA using a 4% sodium citrate solution [25]. Wen et al. reported circuit coagulation in 38 sessions (5% of the total), yet, the RCA concentration was 30%. [11].

The second effectiveness indicator evaluated in the present study was the minimum completeness ≥ 80% of the prescribed HD time, which was 75.5% (Table 3). When assessing the sessions that reached 100% of the prescribed HD time, the result was 69.8%. A similar percentage appeared in the prospective study by Schneider et al., in which full dialysis time occurred in 73 of the 103 treatments (71% of sessions). However, the study occurred with 34 patients divided into 6 anticoagulation protocols. Only 11 sessions were performed with RCA using a 4% sodium citrate solution [25].

In ICUs and dialysis units, it is essential to measure dialysis performance to promote treatment effectiveness and patient safety [27]. For this, information supported by valid and reliable data is imperative when analyzing the treatment provided [28] and developing quality indicators [29].

Quality Indicators (QIs) represent methods for evaluating the performance of target areas of the health system. They seek to measure, monitor, analyze, and communicate the effectiveness of actions and services. Generally, QIs use the Donabedian triad, which proposes the evaluation of health care through Structure Elements (where they are provided), Process Elements (how they are provided), and Outcome Elements (the effects of care delivery). In the present study, QIs that are similar to the parameters of two domains of the referred method were analyzed: Process Element (minimum completeness of prescribed HD time) and Result Elements (dysnatremia and metabolic alkalosis and system coagulation).

One study sought to develop a method to assess how improving the quality of dialysis delivered to patients could impact clinical outcomes. Three categories of objective metrics were established for the filter, prescription, and fluid balance in order to perform quarterly reviews to drive staff training and measure care performance. A total of 184 critically ill patients on continuous HD from 2012 to 2017 were evaluated. The study concluded that implementing QIs can support the development of metrics to assess the performance of institutional standards of continuous HD, specifically in compliance with the proposed care [30].

Again, it is worth mentioning that no specific studies that addressed Quality Indicators in prolonged HD were found. Nor were they carried out in a hybrid system using RCA at a 4% concentration. Thus, the present study proposed to present data that could contribute to the discussion on the subject and provide insights that collaborate with new research in the area.