Intraoperative fluid management in scoliosis
Myron Yaster MD and Mary Ellen McCann MD
One of the first clinical studies I ever performed as a young faculty investigator was a double blind prospective study to evaluate IV nitroglycerin v IV nitroprusside as a method of producing deliberate hypotension during scoliosis surgery.1 Why deliberate hypotension? And by the way “how low was low”? At the time, it was a commonly used technique to limit intra-operative blood loss. Although no longer or rarely used, the study did teach me many important lessons about clinical research design and just how difficult it is to study and measure intraoperative blood loss. In that study, we had to weigh sponges and OR linen and drapes, and measure how much irrigation fluid surgeons used during surgery and to compare that weight to the dry preoperative weight. What I learned was that measuring intra-operative blood loss was extremely difficult and highly inaccurate. Thus, from that point on in my career, when the junior surgical resident would ask at the end of a case “what was the fluid loss?” my best answer was either “minimal, modest, or I needed to put on galoshes.” But, a number was needed (“why”?). I fell back on something one of my teachers as a fellow taught me: in large blood loss cases add 500 mL to the total blood product in mLs administered during surgery as a rough estimate of blood loss.
To say the least, this was all very unsatisfying, and I suspect that this guesstimate is not any more accurate today than it was in the past. What do you do in your current practice? How do you calculate or estimate blood loss? Send me your responses and I will post in a Friday reader response.
In today’s PAAD a group of French investigators2 took another look at this issue. Myron Yaster MD
Original article
Pardessus, P, Loiselle, M, Silins, V, et al. The association between intraoperative fluid management and perioperative allogenic blood transfusion during adolescent idiopathic scoliosis surgery. Pediatr Anaesth. 2023; 33: 829-836. doi:10.1111/pan.14722
Blood loss is an expected feature of adolescent idiopathic scoliosis surgery. Over the past 50 years, multiple methods to decrease allogenic blood transfusion have been studied and utilized. Some have gone out of fashion (deliberate hypotension, deliberate hemodilution), others have become routine (antifibrinolytic therapy, cell saver, decreased threshold for transfusion, and preoperative hemoglobin optimization with erythropoietin). Surgical skill and technique is probably the most important element in blood loss control (“There’s no such thing as a bad fast surgeon”). The goal of the Pardesus et al. paper was “to investigate factors predicting allogenic red cell transfusion during the perioperative management of surgical correction of adolescent idiopathic scoliosis with a special interest in intraoperative crystalloids intake.”2
The authors describe a very detailed description of their standardized perioperative anesthetic technique as well as surgical technique. Some practices that stood out to us. All patients underwent 4 weeks of IV recombinant erythropoietin and iron therapy with hemoglobin concentration measurements at each of the weekly infusion sessions. Intraoperatively, all patients received tranexamic acid therapy (10 mg/kg bolus, 5 mg/kg/hour infusion) and cell saver autotransfusion. After orotracheal intubation and paralysis (atracurium), “the patient was placed in a prone position for surgery. Maintenance of anesthesia was performed using sevoflurane in a 50% mixture of O2/Air, and adjusted to maintain bispectral index (used in all patients) values between 40 and 60. Dexmedetomidine was then administered as an initial bolus of 0.4 mg.kg−1 over 10 min followed by an infusion of 0.04 mcg/kg/hour with sevoflurane used as the sole hypnotic agent. Intraoperative sufentanil boluses (0.2 μg.kg−1) were administered to maintain mean arterial pressure and heart rate within 20% of the preoperative values. Neuromonitoring including both somatosensory-evoked potentials and neurogenic motor-evoked potentials was used, with a recorded response in the spinal cord (allowing intraoperative muscle relaxation).”2 Additionally, following induction of anesthesia they often placed an esophageal Doppler probe to guide fluid management. Indexed stroke volume (iSV) was maximized using repeated 10 mL.kg−1 of 0.9% saline boluses. Finally, “any reduction in mean arterial pressure below 20% of the preoperative value was treated with repeated boluses (3 mg) of ephedrine.”2
Many elements of their protocol are uncommon in the U.S. For example, sevoflurane is often thought to interfere with neuromonitoring,3, 4 atracurium is a rarely used muscle relaxant, BIS monitoring has (or should be) been largely replaced by EEG monitoring and to be honest few groups use either, although BIS monitoring is commonly used in Boston.5 Doppler esophageal probes are rarely used and as we’ve discussed in previous PAADs hypotension is almost always treated with direct acting agonists like epinephrine, phenylephrine, or dopamine….almost never with ephedrine. Sufentanil was the intraoperative opioid. Really? Although used commonly in Boston, I (MY) wonder why? Do you use it in your practice? Because motor evoked potential monitoring is so frequently used, in Boston I (MEM) never use any muscle relaxant other than to facilitate intubation and we NEVER use dexmedetomidine because boluses interfere with MEPS. Finally, normal saline was the crystalloid used to maintain blood pressure and cardiac output. As we’ve discussed in many previous PAADs, normal saline should really be called abnormal saline because it has way too much chloride and can produce metabolic alkalosis. Much better to use either ringers lactate or a true balanced salt solution like Plasmalyte or Normasol.
OK, what did they find? They actually had a very successful protocol. Of the 250 patients they studied only 12 required a blood transfusion either during surgery (n=6) and over the 2 days postoperatively (n-2). When they looked at all of the possible variables that may have been associated with blood transfusion, only increased volume of intraoperative crystalloid administration was significant. We wonder if the increased crystalloid administration in the transfused patients was the result of blood loss rather than its cause. Finally, as our PAADs exec council member Dr Jim DiNardo always reminds us, this finding is an association and is not causation.
We’ve raised a lot of questions that we’d like you to think about and we’d love to hear your thoughts. Send to Myron (myasterster@gmail.com) who will post in a Friday reader response.
References
1. Yaster M, Simmons RS, Tolo VT, Pepple JM, Wetzel RC, Rogers MC. A comparison of nitroglycerin and nitroprusside for inducing hypotension in children: a double-blind study. Anesthesiology. 1986 1986;65(2):175-179. Not in File.
2. Pardessus P, Loiselle M, Silins V, et al. The association between intraoperative fluid management and perioperative allogenic blood transfusion during adolescent idiopathic scoliosis surgery. Pediatric Anesthesia. 2023;33(10):829-836. doi:https://doi.org/10.1111/pan.14722
3. Halsey MF, Myung KS, Ghag A, Vitale MG, Newton PO, de Kleuver M. Neurophysiological monitoring of spinal cord function during spinal deformity surgery: 2020 SRS neuromonitoring information statement. Spine Deform. Aug 2020;8(4):591-596. doi:10.1007/s43390-020-00140-2
4. Buhl LK, Bastos AB, Pollard RJ, Arle JE, Thomas GP, Song Y, Boone MD. Neurophysiologic Intraoperative Monitoring for Spine Surgery: A Practical Guide From Past to Present. J Intensive Care Med. Nov 2021;36(11):1237-1249. doi:10.1177/0885066620962453
5. Yuan I, Missett RM, Jones-Oguh S, et al. Implementation of an electroencephalogram-guided propofol anesthesia education program in an academic pediatric anesthesia practice. Paediatric anaesthesia. Jul 6 2022;doi:10.1111/pan.14520