Evidence regarding the optimal dose of dialysis

This has been asked about in Question 12 from the first paper of 2010 and Question 5 from the first paper of 2003.

Specifically, " Briefly outline the relationship between dose of dialysis and outcome"

The 2010 question makes reference to the following "relevant studies":

• Ronco (2000)
• Bouman (2002)
• Saudan (2006)
• Tolwani (2008)
• VA/NIH (2008)
• RENAL study (2009)

Since then, there have been a few more:

• Vesconi (2009) - an observational study
• Van Wert (2010) - a meta-analysis
• Zhang et al (2012)
• IVOIRE study (2013) (sepsis)
• Clark (2014) ( sepsis)

In summary:

It seems beyond a certain dose, renal replacement therapy removes as many useful molecules as it does toxins, and the benefit from escalating the dose deteriorates.

The evidence

In detail:

• Ronco (2000):
  • 425 patients, single centre RCT
  • Compared 20ml/kg/hr, 35ml/kg/hr, 45ml/kg/hr
  • Observed survival at 15 days after cessation of dialysis
  • Survival in the three groups was 41%, 57%, and 58%
  • Conclusion: "We recommend that ultrafiltration...should reach at least 35 mL/kg/hr"

• Bouman (2002)
  • 106 patients, two-centre RCT
  • Compared 19ml/kg/hr performed later with 20ml/kg/hr performed earlier, and 48ml/kg/hr performed earlier.
  • Observed survival at 28 days
  • For the three grous, 28 day survival was 75%, 68.8% and 74.3%
  • Conclusion: "survival ... and recovery of renal function were not improved using high ultrafiltrate volumes"

• Saudan (2006)
  • 206 patients, single centre RCT
  • CVVH (25ml/kg/hr) compared with CVVHDF (42ml/kg/hr)
  • Observed survival at 28 and 90 days
  • Survival at 90 days was 34% for CVVH and 59% for CVVHDF
  • Conclusion: "increasing the dialysis dose especially for low molecular weight solutes confers a better survival"

• Tolwani (2008)
  • 200 patients, single-centre RCT
  • 20ml/kg/hr compared with 35ml/kg/hr
  • Observed in-ICU survival or 30 day survival, whichever happened first
  • Survival for low-dose CRRT patients was 56%, vs. 49% for the high-dose group
  • Conclusion: "a difference in patient survival or renal recovery was not detected"

• VA/NIH (2008)
  • 1124 patients, massive multicentre RCT
  • Compared 35ml/kg/hr and 20ml/kg/hr
  • Observed 60 day survival, as well as many other parameters
  • Survival for low-dose CRRT was 51.5% and 53.6% for high-dose group
  • Conclusion:"Intensive renal support in critically ill patients with acute kidney injury did not decrease mortality, improve recovery of kidney function, or reduce the rate of nonrenal organ failure"

• RENAL study (2009)
  • 1508 patients, massive multicentre RCT
  • Compared 25ml/kg/hr to 40ml/kg/hr
  • Observed 90 day survival
  • Survival in both groups was the same: 44.7%
  • Conclusion: " treatment with higher-intensity continuous renal-replacement therapy did not reduce mortality at 90 days"

So, though small trials occasionally found a mortality benefit, large well-designed studies were unable to detect one. At this stage, the issue could be said to have been put to rest. Subsequent investigators have made attempts to find subtle non-survival benefits, or to investigate higher dialysis doses in patients with sepsis. More negative trials ensued.

• Vesconi (2009)
  • 15,200 patients - prospective multicentre observational study.
  • Groups were separated by CRRT dose, according to whether they received less or more than 35ml/kg/hr.
  • Outcome measures included ICU mortality, ICU length of stay and duration of mechanical ventilation.
  • Higher CRRT dose patient mortality was 60.8% vs. 52.5% for the low-dose group
  • However, the higher CRRT dose was associated with shorter ICU stay and more rapid extubation among the survivors.
  • Conclusion: "no evidence for a survival benefit afforded by higher dose RRT"

• Van Wert (2010) - a meta-analysis.
  • 12 trials met inclusion criteria (3999 patients).
  • Mortality was only one among numerous parameters assessed.
  • Meta-analyses found no effect of high-dose renal replacement therapy on mortality.
  • Conclusion: "High-dose renal replacement therapy in acute kidney injury does not improve patient survival or recovery of renal function overall or in important patient subgroups, including those with sepsis."

• Zhang et al (2012)
  • 141 patients with sepsis and AKI; single centre RCT.
  • Compared super-high CRRT doses: 50ml/kg/hr vs 85ml/kg/hr.
  • Observed survival at 28, 60 and 90 days.
  • No difference between the groups was found at any timeframe.
  • Conclusion: "increasing the intensity of renal replacement therapy... had no effect on survival"

• IVOIRE study (2013)
  • 140 patients with septic shock and AKI, multicentre RCT.
  • Compared 35ml/kg/hr vs 70ml/kg/hr (HVHF, High Volume HaemoFiltration).
  • Primary endpoint was 28 day mortality.
  • Mortality was 40.8% vs 37.9%.
  • Conclusion: "HVHF... cannot be recommended for treatment of septic shock complicated
    by AKI"

• Clark (2014) - a meta-analysis specific for CRRT intensity in sepsis
  • 4 trials (470 patients) met inclusion criteria.
  • Comparion of standard intensity to high intensity (over 50ml/kg/hr)
  • 28 day mortality was the main outcome measure.
  • Pooled analysis did not show any meaningful difference.
  • Conclusion: "Insufficient evidence exists of a therapeutic benefit".
  • Furthermore, "further trials should focus on alternative extracorporeal therapies as an adjuvant therapy for septic AKI rather than HVHF". Essentially, the authors are saying that this is a dead end, and we should think about something else.