Clinical Trials in Perioperative Medicine


Key points

  • Because most improvements in perioperative care are incremental, large numbers of patients need to be studied to have adequate statistical power to detect a clinically important difference.

  • Large, pragmatic, multicenter, randomized trials are more reliable because they provide an unbiased evaluation of specific interventions delivered across a range of healthcare settings.

  • Large pragmatic trials therefore test for effectiveness in routine clinical practice; their results are more generalizable.

  • Examples of practice-improving clinical trials are outlined.

The Importance of Clinical Trials

Advances in anesthesia, surgery, and other perioperative practices over recent decades have led to amazing improvements in patient experience and outcomes in the days, weeks, and months after surgery. How and why do these improvements come about? The history of medicine has mostly been founded on accruing experience, often as trial and error, whereby obviously harmful therapies are discarded and more effective therapies are eventually taken up by clinicians. Most of these changes, however, happen too slowly or there is too much inexplicable variation in care, such that optimal care is not delivered to many patients.

Our patients are older and sicker, and increasingly complex surgical procedures are being done more often. The risks of adverse perioperative outcomes are greater. We need to reliably identify effective treatments to optimize care.

Training and experience are essential to good practice, but quality improvement demands evaluation and research. Genuine improvements in care, and the reasons for these, can be difficult to identify. Audit is important but accurate and complete data collection of all relevant variables, especially patient outcomes, can often miss important information. Single-center studies may not accrue enough cases, and their results may not apply to other centers. Large observational (nonrandomized) studies can collect extensive data from many centers, but these are prone to numerous biases that can result in misleading conclusions. This perpetuates uncertainty and helps explain why variations in practice are so common.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here