To the Editor: Patients with short bowel syndrome depend on total parenteral nutrition and are thus at increased risk of venous thrombosis. As available venous sites are lost through use, bowel transplantation becomes more urgent (1) but replacing blood becomes more difficult. Once the patient has no conventional transfusion sites, that patient is no longer considered to be a candidate for transplantation. After evaluating several patients who presented with this dilemma, we developed a plan for anesthesia using a combination of intraosseous, intraarterial, and surgically created venous access to facilitate organ transplantation in patients without conventional venous access. A 52-yr-old woman had a gastric bypass operation for weight reduction. The procedure was complicated by small bowel necrosis, requiring total removal of jejunum and ileum. The duodenum was anastomosed to the colon. After 20 yr of total parenteral nutrition treatment, venous access was limited to small caliber lumbar and trans-hepatic venous catheters inserted under radiological guidance. A computer tomographic scan demonstrated that the brachiocephalic vein, the superior vena cava, and the iliac veins were bilaterally occluded. The interventional radiology service at another center experienced in re-cannulating thrombosed central veins was unsuccessful in opening any veins, and the attempt was complicated by avulsion of the superior vena cava from the atrial appendage. The patient developed liver failure and was a candidate for multivisceral transplantation. We considered intraarterial and intraosseous transfusion with subsequent surgical creation of venous access because of the absence of traditional venous sites. The plan was presented to the Innovative Practices Committee of the University of Pittsburgh Medical Center and to the patient. After the committee's approval and repeated explanation of the potential complications, including the possibilities of direct arterial or bone injury, chemical, thrombotic, and embolic events, all agreed to the plan. By the time suitable donor organs became available, the only working venous access was a small-bore trans-hepatic catheter. We induced anesthesia through this with thiopental 250 mg, fentanyl 250 μg, and midazolam 5 mg. We facilitated tracheal intubation with rocuronium 50 mg and maintained general anesthesia with isoflurane in air/oxygen and additional fentanyl and lorazepam. We continuously assessed intravascular volume with transesophageal echocardiography. An intraosseous needle (Cook disposable 15.5 gauge intraosseous infusion needle; Cook Critical Care, Bloomington, IN) was inserted into the right tibia but flow was poor. This site would not have sustained rapid transfusion but the needle was left in place to prevent periosteal bleeding. We placed 3 arterial catheters percutaneously: a 20-gauge left radial arterial catheter (Arrow Radial Artery Catheterization Product No RA-04020; Arrow International, Reading PA), an 8.5F left femoral arterial catheter (Arrow Percutaneous Sheath Introducer Kit No AK-09801; Arrow International) and a 7F right femoral two-lumen catheter (Two-Lumen Central Venous Catheterization Kit MTO-17702; Arrow International). A continuous infusion for medication delivery was connected to the smaller lumen of the right femoral arterial line and delivered with a volumetric pump (Colleague 3 volumetric infusion pump; Baxter Healthcare, Deerfield, IL). Pressure transducers were connected to the remaining lumen of the right femoral and to the radial and left femoral arterial catheters. Blood and fluids were intermittently infused through both femoral arterial catheters using infusion pumps (Hemonetics Rapid Infusion System, REF 400; Hemonetics, Braintree, MA). Pulse oximeters were connected distal to each arterial cannulation site to monitor continued perfusion of the extremities. Laparotomy was complicated by adhesions from multiple prior operations. A catheter (Bard Hickman 12F Dual-Lumen CV catheter, Bard Access Systems, Salt Lake City, UT) was placed through the left abdominal wall and was threaded into the lumen of an inferior mesenteric vein. Another catheter (Bard Hickman 13.5F Dual-Lumen CV catheter, Bard Access Systems) was placed through the right abdominal wall and was threaded through an ovarian vein into the vena cava. Elastic bands secured the catheters and permitted automatic vein ligation on catheter removal. The existing trans-hepatic catheter was removed along with the liver. Donor liver, intestine, and pancreas were grafted en bloc. At the end of the operation the left femoral arterial catheter and the intraosseous needle were removed. The patient received more than 2 L of blood and colloid through intraarterial infusion before the establishment of venous access through the inferior mesenteric and ovarian veins. She subsequently returned to the operating room twice for abdominal exploration. Both transabdominal catheters were ultimately removed and the patient was discharged home with oral nutrition. Multivisceral transplantation is a major operation lasting many hours with large fluid requirements and the potential for massive bleeding. The alternative routes that we provided had the capacity for massive transfusion if necessary. Intraosseous infusion has been used for crystalloid resuscitation of hypovolemic patients in whom venous cannulation is difficult (2), more commonly in pediatric (3) than in adult patients. Accessing more proximal bone marrow spaces, including the pelvis and sternum (4), may have been more effective in providing intravascular volume infusion for our patient but would have compromised the surgical field. Intraarterial infusions are currently used in special circumstances (e.g., delivery of drug to targeted tissue) but their use in organ transplantation for patients with no venous access has not been reported. Some drugs, notably thiopental, can cause damage if delivered into an artery but others can be safely delivered. Several studies (5–7) have suggested that intra-arterial infusions may be better than IV delivery for resuscitation medications. Because the relative risks versus benefits of our approach were unknown, our consent process was extensive and carried out in several meetings. Careful planning and close cooperation between the anesthesiologist and the operating surgeon permitted successful multivisceral transplantation in this complex, compromised patient. Charles D. Boucek, MD Kareem Abu El Magd, MD Department of Anesthesiology University of Pittsburgh Medical Center Pittsburgh, PA [email protected]
No takes yet. Share an insight, caveat, or question.
Boucek et al. (2006) studied this question.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: