Friday, 24 October 2014

Simvastatin in the Acute Respiratory Distress Syndrome (HARP-2)


Paper Link

McAuley D et al for the HARP-2 investigators and the Irish Critical Care Trials Group
New England Journal of Medicine (2014); published online 30th September 2014



At a glance

  • Intervention - 80mg oral simvistatin given to patients with ARDS of any cause
  • Randomised - Yes
  • Controlled - Yes
  • Placebo used - Yes
  • Blinded - Yes
  • Population / Sample size calculation - 540 patients from 40 hospitals 
  • Groups same at randomisation - Yes
  • Groups treated the same after randomisation - It appears so
  • Intention to treat analysis - Yes
  • Follow-up - Almost complete to 28 days
  • Significant primary endpoint - No difference was found in ventilator free days at day 28
  • Harms analysed - Serious adverse events were reported, with no difference between the groups


Much interest has been shown in recent years in the potential anti-inflammatory benefits of the statins, and how this might be applied to critically ill patients.  Inhibition of the HMG-CoA reductase enzyme can result in dampening of a number of putative mechanisms involved in the development of the systemic inflammatory response syndrome.  This action has a general anti-inflammatory effect through multiple mechanisms rather than targeting a single component of the inflammatory pathway.

Acute respiratory distress syndrome is a disease process that is related at least in part to uncontrolled inflammatory pathways.  This inflammation leads to damage to the alveolar architecture with accumulation of protein-rich fluid and immune cells in the alveoli and pulmonary interstitium.  

Additionally, ARDS may then contribute to ongoing pro-inflammatory effects by the release of mediators into the system circulation, resulting in persistent systemic inflammatory response syndrome and multi-organ dysfunction.

The use of statins to mute the pro-inflammatory state has previously been tested in laboratory settings, in animal studies and in small preliminary trials in humans with encouraging results.  Circumstantial evidence of clinical effect has been seen in studies that have demonstrated improved outcomes for patients in whom these drugs were continued during their ICU admission.  Preadmission use of statins is also associated with improved outcomes for critically ill in large database observational studies.  These findings taken together gave researchers great optimism that statins might prove to be the magic bullet for sepsis and ARDS management.

This study sought to detect any signal that statins may reduce the impact of ARDS on meaningful outcomes such as mortality, ventilation duration and end-organ dysfunction.

Trial Details

540 ventilated patients within 48 hours of onset of ARDS (defined with standard criteria and a PaO2/FiO2 <300) were included, from 40 hospitals across the UK and Ireland, and randomised to receive enteral simvistatin or placebo.  The treatment was continued for up to 28 days, and was stopped at ICU discharge or for other clinically relevant reasons (such as futile treatment or the development of rhabdomyolysis).

The sample size was selected to detect a change in ventilation free days of 2.6 days - previous studies have suggested that the baseline for this group of patients would be 12.7 ventilator free days in the first 28 days.

The patients appear to be essentially the same in each group prior to the intervention, and to have been treated in a similar fashion other than the treatment in question.  Patients were in general treated along internationally recognised ventilation guidelines, though the average tidal volume of 8.1ml/kg was slightly higher than the authors would have liked.  

The authors suggest that the sample includes a broad representation of causes of ARDS, with sepsis present in 75%, pneumonia in 58%, and other causes such as smoke inhalation, pancreatitis, thoracic and non-thoracic trauma, and aspiration all included.  The average P/F ratio in the study was around 125, along with the average APACHE 2 score of 19, suggest the patients were reasonably sick (compared with 15.4 in ARISE for instance).

Outcomes

Patients were given the drug or placebo for an average of 10.5 days during the study, with the most common reasons for cessation being discharge from ICU or death.  

The primary outcome measure examined by the study was ventilator free days at day 28, a relatively common endpoint for ventilation trials.  No significant difference was noted between the groups (12.6 in the intervention arm and 11.5 in the control arm)

Other outcomes of note included organ dysfunction free days at day 28, the daily SOFA score, and mortality, although the authors acknowledge the study was not powered for this endpoint.  None of these outcomes, nor a number of other predefined measures, were significantly different between the groups.

No difference was noted in terms of serious adverse effects between the two groups.

Summary

So what does this all mean? 

In short, there is no evidence that simvistatin has any significant clinical impacts in patients with moderate to severe ARDS across a broad range of aetiologies.  

The results are consistent with the 745-patient Statin for Acutely Injured Lungs (SAILS) trial published earlier in 2014, conducted by the ARDS-Net group, which used rosuvastatin in patients with ARDS related to sepsis, and a trial of simvistatin in patients with ventilator associated pneumonia (STATIN-VAP).  Furthermore, a phase 2 trial conducted in Australia and New Zealand studying its role in sepsis also showed no benefit (ANZ-STATiNs)

Ventilator-free days is a composite endpoint and is in effect a surrogate marker, but never-the-less, there is little if any signal of benefit in this group at this stage.