We describe the case of a 74-year-old man who was scheduled for an urgent AKA. The patient was urgently admitted from home after his home nurse noticed a dehiscence of his prior right below-the-knee amputation (BKA) stump. He received his BKA surgery 1 month prior for critical limb ischemia due to vascular disease which was complicated by poor wound healing. During this prior admission for a BKA, he also received a right-sided common femoral endarterectomy procedure. His past medical history is significant for end-stage renal disease status post-transplantation, aortic stenosis status post-transcatheter aortic valve replacement, and atrial fibrillation for which he is on apixaban, heart failure with reduced ejection fraction, and peripheral vascular disease. Due to the concern for wound dehiscence and infection noted by his home nurse, he was admitted urgently, started on broad-spectrum antibiotics, and scheduled for an AKA. Given the likelihood of surgery, his apixaban was held on admission, and he was transitioned to a heparin drip. His heparin drip was titrated to a goal-activated partial thromboplastin time (aPTT) of 60–80 s (reference range 20.9–30.9).

On the morning of surgery, his heparin drip was stopped 4 h prior to surgery, and an aPTT was measured preoperatively which returned normal. Given the need for anticoagulation after the surgery and concerns for the difficulty of placement given the patient body habitus with a BMI of 41, neuraxial was avoided and the anesthesia team planned to proceed with a peripheral nerve block. At the request of the surgeons, there would be no catheter placement. In preparation for the femoral nerve block, it was noted that the site of his prior right common femoral endarterectomy appeared erythematous and concerning for soft tissue infection. As a result, the femoral nerve block was aborted. Given these limitations, it was decided to place a right-sided erector spinae plane block in the lumbar region. The patient was then positioned on his left side, and a preliminary scan was completed at the third lumbar space (L3). This level was chosen for two reasons. First, we expected spread one to two levels above and below along and second at this level we encountered the best ultrasound image quality. Midline, iliac crest, and the right transverse process were identified at L3. Preparation of the field with sterile preparation was completed. A 10-mm 21-gauge needle was inserted under ultrasound guidance. Gentle contact was made with the transverse process. Aspiration was negative, and 30 ml of 0.2% ropivacaine was injected under direct visualization. The local anesthetic spread was noted below the erector spinae muscle.

Intraoperatively, the patient received general anesthesia. Induction medications included 50 mg of propofol, 100 μg of fentanyl, 50 mg of rocuronium, and 60 ml of lidocaine. General anesthesia was maintained with sevoflurane between 1.3 and 2.0%, he received 2 g of cefazolin for antibiotics, and he received a total of 1 ml of hydromorphone IV in incremental doses over the last hour of the case. In addition, he received 4 mg ondansetron and 96 mg of sugammadex at the end of the case. He was extubated and taken to the post-anesthesia care unit without complications.

While in the PACU, the patient’s pain was monitored using a Numerical Rating Scale from 0 to 10 (0 = no pain, 10 = worst imaginable pain). His first three documented pain score levels were zero and over the next 3 h increased to a maximum pain score of 4 for which he received 1 g of acetaminophen and 5 mg of oxycodone. These are commonly provided pain medications for moderate pain in our PACU. He recovered appropriately and was transferred to the ward. The patient reported pain scores of 0 to 4 for 11 h postoperatively. On hour 12, his surgical site pain increased to 7, requiring the use of 0.4 mg hydromorphone IV and 5 mg of oxycodone. We hypothesize this change in pain occurred as the nerve block wore off. On postoperative day 1, the patient was interviewed about his experience with the nerve block. Consistent with the pain scores documented, he reported adequate pain control for about 10–12 h and then an abrupt change as the nerve block wore off.