Tourniquet Injuries, Implied Causality, Babies, and Bathwater

To the Editor:

In a recent analysis of selected retrospective data (only tourniquet times longer than 2 h), Horlocker et al. (¹) showed an association between increased tourniquet time and increased frequency of nerve injury. It is tempting to conclude from this association that prolonged compression and ischemia would result in nerve injury, and indeed this seems both reasonable and logical. However, the authors cite literature showing that nerve injury from the joint replacement itself is an order of magnitude more frequent than nerve injury from the tourniquet.

The orthopedic literature demonstrates an increased risk of nerve injury related to an increased complexity of the surgery. With a retrospective analysis, there is no way to separate the role of tourniquet time from the role of surgical complexity. More complex cases take longer. And what of the role of the surgeon? Less-skilled or less-experienced surgeons may take longer and may also have higher injury rates. We cannot therefore conclude that the tourniquet is responsible for the injury when tourniquet time is confounded by both surgical complexity and surgeon variability.

The authors conclude that the risk of sciatic nerve injury when prolonged tourniquet time is possible is a contraindication to performing a sciatic nerve block for knee replacement surgery. We do not agree. We do agree that it is important for us to be aware of the potential for neural blockade to obscure the diagnosis of nerve injury and thus delay intervention. Our practice is to preoperatively place a sciatic perineural catheter using saline. Postoperatively, the catheter is connected to a pump administering 0.05% ropivacaine at an initial infusion rate of 5 mL/h. This infusion is only begun when the patient (¹) complains of posterior knee or calf pain and (2) has full foot and ankle motion. This dilute concentration of ropivacaine and slow infusion rate generally produce a selective sensory block. We stop a sciatic infusion if foot weakness develops and closely observe the patient’s recovery (or lack thereof) to facilitate early intervention. In two cases, this approach has allowed us to identify surgically related sciatic injuries without any controversy about the role played by a sciatic nerve block.

Inferring a causal relationship from a correlation has led the authors to “throw out the baby with the bathwater.” First-rate postoperative analgesia should be available to all patients, albeit administered in a fashion that attempts to avoid and is respectful of the possibility of obscuring complications.

Bruce Ben-David, MD

Anna Uskova, MD

Department of Anesthesiology

UPMC Presbyterian-Shadyside Hospital

Pittsburgh, PA

[email protected]