Intensive RRT for AKI: Dial Down Your Enthusiasm!

Introduction There is a widely held belief among nephrologists and intensivists that “more is better” as it pertains to dose of RRT in the management of AKI. Multicenter randomized controlled trials have established that dosing of RRT beyond continuous RRT (CRRT) effluent rates of 20–25 ml/kg per hour and intermittent hemodialysis (IHD) three times per week (single pool Kt/V urea of 1.3 per session) does not improve clinical outcomes. In this perspective, we strive to address and dispel the purported benefits of high intensity RRT in AKI while highlighting the detrimental consequences of higher doses of RRT. Lack of Benefit of Higher-Intensity RRT in Sepsis and AKI Rapid removal of bacterial endotoxins and proinflammatory cytokines from the circulation have been postulated to dampen the excessive, and consequently pathologic, immune response observed in some patients with severe sepsis. Although this is theoretically possible, data do not support using high-intensity RRT for any form of AKI. Hemofiltration is nonselective and results in the removal of anti-inflammatory cytokines. In addition, because it is not possible to determine exactly when an excessive proinflammatory response is occurring, dampening the immune response could predispose the critically ill patient to a life-threatening infection. Moreover, removal of proinflammatory cytokines is unlikely to be beneficial if endogenous production outpaces removal via RRT. A secondary analysis of the Veterans Affairs/National Institutes of Health Acute Renal Failure Trial Network (ATN) study found higher-intensity RRT did not consistently lower the concentration of proinflammatory cytokines relative to lower-intensity RRT (1). The evidence that even higher doses of RRT (effluent flow rates >35 ml/kg per hour) can lead to a decrease in the levels of proinflammatory cytokines has been mixed (2,3). A final consideration is that higher-intensity RRT increases the likelihood of inadequate drug dosing, specifically antimicrobial agents (4). In the Randomized Evaluation of Normal versus Augmented Level (RENAL) and ATN trials, no significant benefit (or harm) was detected with higher-intensity versus lower-intensity RRT in the subgroups of patients with sepsis (5,6). Trials that have evaluated clinical outcomes with the use of high-volume hemofiltration (i.e., >35 ml/kg per hour) have generally been small and underpowered to assess mortality, renal recovery, and other clinically relevant outcomes. No signal for benefit is evident on the basis of systematic reviews and meta-analyses of the limited data (N<200 overall) (4,7). As such, a higher dose of RRT than otherwise indicated for AKI should not be used for patients with septic AKI. Potential Harm with Higher-Intensity RRT? Arguments to prescribe higher doses of RRT for AKI than was done for the lower-intensity (“standard-dose”) arms of the ATN and RENAL trials are predicated on an assumption that higher-intensity RRT offers benefit, without causing any harm. However, current literature actually indicates a signal for the opposite. Both the ATN and RENAL studies found that hypophosphatemia occurred more often with higher-intensity than lower-intensity RRT (18% and 65% more frequently, respectively) (5,6). Hypophosphatemia is associated with diaphragmatic weakness and a longer duration of mechanical ventilation in patients receiving CRRT (8). Although not evidence of causality, a secondary analysis of the ATN study found that higher-intensity RRT was associated with a significantly lower rate of successful extubation relative to lower-intensity RRT (hazard ratio [HR]=0.67; 95% confidence interval [CI], 0.52 to 0.88) (1). Furthermore, participants randomized to higher-intensity RRT had significantly fewer ventilator-free days (2.07 versus 3.08 in the first 14 days; P<0.001). Baseline serum phosphate levels were found to be a significant effect modifier. Patients whose baseline serum phosphate was in the lowest tertile (1.2–4.3 mg/dl) were significantly less likely to be successfully extubated if randomized to higher-intensity versus lower-intensity RRT (HR=0.56; 95% CI, 0.36 to 0.87). Another important issue is whether intensity of RRT affects renal recovery after AKI. A secondary analysis of the ATN study showed higher-intensity RRT was associated with less daily urine output and a significantly greater unadjusted risk of a ≥50% decline in urine output (HR=1.29; 95% CI, 1.1 to 1.51) (9). A separate analysis of participants in the ATN study who were treated only with IHD found that those randomized to lower-intensity RRT (IHD three times per week) had 2.5 more RRT-free days through day 28 (95% CI, 0.27 to 4.79 days) than those randomized to higher-intensity RRT (IHD six times per week) (10). Overall, a meta-analysis (six trials; N=1926) found that participants randomized to relatively higher-intensity RRT spent more time requiring RRT and were significantly more likely to be RRT dependent at day 28 (HR=1.15, 95% CI, 1 to 1.33) (11). Overall, higher-intensity RRT appears to affect kidney recovery to RRT independence negatively in the short term. Few studies have specifically reported RRT dependence to day 90, but the available data do not suggest RRT intensity has a substantial effect in the long term (11). Higher-intensity RRT is more expensive in all aspects, whether prescribed as CRRT, prolonged intermittent RRT, or IHD. Cost of disposables such as filters and fluids, and labor costs for IHD nursing, will proportionally rise on the basis of higher effluent rates in CRRT or frequency of PIRRT and IHD. A 15%–25% increase in daily CRRT costs has been estimated to occur with the use of 35–45 ml/kg per hour doses versus 20 ml/kg per hour doses (12). If higher-intensity RRT leads to fewer RRT-free days as mentioned above, then this further amplifies the cost difference between lower-intensity and higher-intensity RRT.Figure 1.: RRT dose, clinical outcomes, and cost. IHD, intermittent hemodialysis.Appropriate Dosing of RRT in AKI The ATN trial compared continuous veno-venous hemodiafiltration effluent flow rates of 20 versus 35 ml/kg per hour (pre dilution), and the RENAL trial compared 25 versus 40 ml/kg per hour (post dilution). It should be noted that the achieved dose was lower than prescribed; the lower-intensity arms in the ATN and RENAL studies achieved effluent flow rates of 17.5 and 22 ml/kg per hour, respectively (Table 1). The results of the ATN and RENAL studies have been occasionally misinterpreted to recommend an effluent flow rate of 25–30 ml/kg per hour, so that if there is CRRT downtime due to circuit clotting or other reasons, patients will still receive 20–25 ml/kg per hour. Our recommendation is to follow the ATN and RENAL studies dosing strategy and target a dose of 20–25 ml/kg per hour, with exceptions for those patients with severe hyperkalemia, metabolic acidosis, or other reasons that would require very high effluent flow rates (Figure 1). It remains unclear what the “floor” for RRT dosing should be, but a retrospective study from Japan suggested that a dose as low as 14.3 ml/kg per hour may be as safe and effective as 20 ml/kg per hour (13). These data became extremely important during coronavirus disease 2019 surges when hospitals attempted to conserve valuable CRRT resources to treat as many patients as possible with safe and effective RRT. Decreasing the effluent dose from 25 to 20 ml/kg per hour confers a 20% reduction in the use of sterile fluids. With regards to IHD, three sessions with a Kt/V urea of 1.3 or a urea reduction ratio of 70% per session should provide adequate clearance for the majority of patients with AKI. Additional sessions may be required for ultrafiltration for volume overload, hyperkalemia, or acidosis. Table 1. - Target and achieved doses of CRRT in the ATN and RENAL studies Dose-Related Study Characteristics ATN Study RENAL Study CRRT mode Predilution CVVHDF Postdilution CVVHDF CRRT low-dose effluent target, ml/kg per h 20 25 CRRT high-dose effluent target, ml/kg per h 35 40 Achieved dose with high-dose CRRT, ml/kg per h 27.1 33.4 Achieved dose with low-dose CRRT, ml/kg per h a 17.5 22 CRRT, continuous RRT; ATN, Acute Renal Failure Trial Network; RENAL, Randomized Evaluation of Normal versus Augmented Level; CVVHDF, continuous veno-venous hemodiafiltration.aAchieved doses in the low-dose arms represent the currently established ‘floor’ for RTT dosing for acute kidney injury. In summary, we recommend that physicians “dial it down” when prescribing RRT in patients with AKI. In our opinion, prescribing higher effluent flow rates in order to achieve a dose of 20–25 ml/kg per hour is not necessary. The ATN and RENAL study protocols used a target (and not an achieved dose) of 20 and 25 ml/kg per hour in the lower-intensity (“standard-dose”) arms. The achieved dose was 85% and 88% of the target in those lower-intensity arms of the ATN and RENAL studies, respectively. In both of these large randomized controlled trials, higher-intensity RRT did not demonstrate any clinical benefit. Quality assurance for CRRT delivery can be focused on ensuring that achieved doses align with those achieved in the lower-dose arms of these two trials. Potential drawbacks to higher-intensity RRT include underdosing of antibiotics, hypophosphatemia and subsequent respiratory dysfunction, delay in renal recovery, and a higher cost. The minimum RRT intensity at which underdosing becomes harmful remains unknown. In this context, could there be clinical and economic benefits with even lower-intensity RRT than was administered to participants in the low-intensity arms of the ATN and RENAL trials? Adequately powered studies to assess whether lower doses can be safely implemented in AKI are necessary because conservation of resources will become even more vital in the future. Disclosures E.G. Clark reports being on the editorial board of the Canadian Journal of Kidney Health and Disease. A. Vijayan reports consultancy for Astute and NxStage; ownership interest in Outset (stock only); research funding from Astellas and Spectral; honoraria from NxStage; an advisory or leadership role for NxStage; and being a member of the National Kidney Foundation. Funding None.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.