Prophylactic hydration to protect renal function from intravascular iodinated contrast material in patients at high risk of contrast-induced nephropathy

Nijssen E. The Lancet; published online February 20, 2017

Clinical Question

  • In high-risk patients receiving contrast medium, does prophylactic hydration with normal saline compared to no prophylaxis impact the rate of contrast-induced nephropathy?


  • Prospective non-inferiority trial
  • Randomised 1:1 assignment
  • Computer generated randomisation
  • Minimisation stratification to assist allocation concealment and balance subgroups
  • Blinding of laboratory staff processing creatinine samples but not of patients or clinicians administering the treatment
  • Informed consent by participants
  • Cost-effectiveness examined
  • Pre-specified subgroups were analysed: Diabetes, eGFR<45 vs eGFR>45, route (IA vs IV), procedure type (interventional vs diagnostic)


  • Mastricht University Medical Centre (Netherlands)
  • June 2014 – July 2016


  • Inclusion:
    • All consecutive patients aged 18 years or older, referred for an elective procedure requiring intravascular contrast material WITH EITHER
      • eGFR 45–60 ml/min/m2 + diabetes OR 2 of >75 years; anaemia (HCT <0.39 L/L for men, < 0.36 L/L for women); CVS disease; NSAIDs; diuretics OR
      • eGFR 30–45 ml/min/m2 OR multiple myeloma OR Waldenstrom’s macroglobulinaemia with small chain proteinuria
  • Exclusion:
    • eGFR < 30
    • RRT
    • emergency procedures
  • 328 allocated to hydration group (32 subsequently excluded as no creatinine available at day 2–6). 332 to no prophylactic group (25 subsequently excluded)
  • Baseline characteristics evenly matched between groups


  • Hydration Group:
    • 3–4 ml/kg/hr of normal saline for 4 hours before and after contrast OR
    • 1 ml/kg/hr of normal saline for 12 hours before and after contrast


  • No hydration


  • Primary outcome:
    • Incidence of contrast-induced nephropathy (defined as proportion of patients with an increase in serum creatinine by more than 25% or 44 µmol/L within 2–6 days of contrast exposure)
      • 2.7% (8/296 in hydration group) vs 2.6% (8/307 in no hydration group).
      • Difference -0.1% (95% CI -2.25% to 2.06%; p=0.47)
    • Cost-effectiveness of hydration vs no hydration
      • Mean total cost €1455 vs €792
      • Difference -€663 (95% CI -€1234 to -€191)
  • Secondary outcome:
    • Mean change in creatinine from baseline at 2–6 days and 26–35 days
      • Mean 2–6 day change in Cr was 0.31 µmol/L in the hydration group and 1.3 µmol/L in the no hydration group (p=0.40)
      • Mean 26–35 day change in Cr was 1.44 µmol/L in the hydration group and 1.39 µmol/L in the no hydration group
    • Major adverse events: mortality, RRT, ICU admission, sequelae of fluid administration… all non-significant differences except that 4% of the hydration group developed heart failure vs 0% in the no hydration group P=0.0001
  • Subgroups: no different in CIN in any of the prespecified subgroups

Authors’ Conclusions

  • No hydration was non-inferior to hydration in the prevention of contrast induced nephropathy. Hydration is more expensive and can cause problems in itself.


  • Appropriate randomised, allocation concealment and blinding to minimise bias
  • At risk population tested


  • Sample size reduced midway through study however unlikely to impact the validity of the results given the predefined non-inferiority margin.
  • Primary outcome a surrogate outcome and not powered to detect differences in a clinically meaningful outcomes which would need a much larger sample size. However, protracted adverse effects of contrast administration seem very rare so surrogate is a reasonable option.
  • Not applicable to Intensive Care patients as these patients excluded. A trial of STEMI patients receiving contrast for coronary angiography showed a much higher incidence of CIN generally and superiority of those hydrated vs those without hydration perhaps related to haemodynamic instability, nephrotoxics and higher contrast volumes.
  • Single centre (but similar protocols used in other centres)
  • Single blinding but a robust laboratory endpoint measured
  • Modified intention to treat analysis (patients were excluded if creatinine unavailable on Day 2-6. These patients did not have differing baseline characteristics to the analyzed sample.)

The Bottom Line

  • Hydration prior to contrast administration may not be necessary in all patients previously thought at risk to develop CIN.
  • However, this trial does not specifically answer the question for my ICU patient. At this stage, intravenous hydration in this population should continue especially if the patient has haemodynamic instability, nephrotoxic drugs and high contrast doses. Patients with GFR < 30 should still receive peri-procedure hydration. With patients at risk of pulmonary oedema, hydration should be used judiciously.

External Links


Summary author: Celia Bradford
Summary date: 5 March 2017
Peer-review editor: Duncan Chambler


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.