Anesthesia for pediatric MRI
Myron Yaster MD and Lynne Maxwell MD
Original article
Amy E Vinson, James Peyton, Anna Kordun, Steven J Staffa, Joseph Cravero. Trends in Pediatric MRI sedation/anesthesia at a tertiary medical center over time. Paediatr Anaesth. 2021 Sep;31(9):953-961. PMID: 34036674
Editorial
Jacob Karlsson, Gianluca Bertolizio. Anesthesia Service provision for MRI: Is shifting the technique enough? Paediatr Anaesth. 2021 Sep;31(9):916-917. PMID: 34409702
An urban legend: An elderly man presents to the ER with a chief complaint of “difficulty walking”. The intake resident sends the man down for a diagnostic MRI. On arrival, in moving from the stretcher to the MRI table, the technician notes that the patient is missing his prosthetic device…
I know I’m (Myron) not the only one, but don’t you ever wonder why so many MRIs are done? There are literally hundreds of thousands of MRIs performed on children yearly and the vast majority, usually for neuro, cardiac, and ortho, are done under general anesthesia. I mean really, how many of these MRIs, particularly neuro exams for developmental delay or autism, are necessary or provide findings that affect the diagnosis or alter the course of treatment? Further, as in the urban legend above: Does anyone ever examine patients anymore? Has physical diagnosis gone the way of the precordial stethoscope? Ok, I’ll stop ranting. However, it is important to realize that non-operating room anesthesia (NORA) now comprises at least 20% of all pediatric anesthetic cases and the vast majority of NORA cases are for diagnostic imaging. Calling out the need for these studies and the need for general anesthesia is not going to win me any friends in administration. Just think about it…a large chunk of our income and the income of the institutions in which we work comes from these studies. Saying that they may be unnecessary is a one-way ticket out of town.
Which brings us to this paper by Amy Vinson et al. from Boston Children’s Hospital and the accompanying editorial by Karlsson and Bertolizio. Over the last 20+ years general anesthesia has increasingly been used to provide the immobility required for MRI. Previous deep sedation techniques included pentobarbital, chloral hydrate, midazolam and dexmedetomidine. They all failed simply because they did not provide 100% efficiency. They sometimes worked and sometimes didn’t. “Failure”, as former NASA flight director Gene Kranz said during the Apollo 13 mission, is simply “not an option”. Scheduling delays because of movement, poor image quality, lack of predictability, and the need to reschedule when images were unsatisfactory led to the near universal use of general anesthesia in the pediatric population for these procedures. Surprisingly, there have been few large-scale studies documenting how to provide anesthesia for pediatric MRIs.
This single center study from the Boston Children’s Hospital looked at 24,000 anesthetics between 2013-19. The most significant trends noted were a decrease in “propofol-only” anesthetic techniques and an increase in propofol and dexmedetomidine combination techniques. Major adverse events were rare. A minority of the cases were done with vapor anesthetics alone or vapor PLUS propofol. The vapor anesthetics had a higher incidence of desaturation than the other techniques but the authors did not present enough detail to understand the time at which these desaturations occurred (induction, maintenance or emergence) or whether it was related to airway management attendant to a vapor anesthetic. Although supported in the accompanying editorial, dexmedetomidine alone given either intranasally (unpublished data mentioned in the editorial) or intravenously (multiple papers), has largely been abandoned by most of us, despite what the editorial said, because immobility was not guaranteed and as doses escalated resulted in prolonged recovery times. We look forward to the editorial authors’ institution publishing their data on the use of nasal dexmedetomidine for MRI.
Why did the anesthesiologists at Boston Children’s switch from primarily propofol to a combination of propofol and dexmedetomidine? The authors speculate that it was “because of growing concerns about the effect of anesthesia and sedation on the developing brain” AND it occurred around the time of the infamous 2016 FDA warning “about the repeated or lengthy use of general anesthetic or sedation drugs during surgeries or procedures in children younger than 3 years of age”. This absolutely makes no sense to us. Boston Children’s was the epicenter of the GAS study which demonstrated that the entire neuroapoptosis fear mongering was overblown and that there is no evidence that single anesthetics of a mean duration of one hour were associated with adverse neurocognitive outcomes. Furthermore, using 2 drugs when one does the job is a recipe for error (2 pumps when 1 will do only increases programming error).
One more thing. Vapor anesthetics were not used much either for IV placement or for the MRI itself. Although not discussed in the paper, why? If these children are going to be anesthetized why not place the IV under general anesthesia? Or if using a general anesthetic to place an IV why not continue it and control the airway with an LMA?
We’ve raised many more questions than this paper and the editorial addressed. A wider discussion either on the PAAD website or at a future SPA/ASA meeting are definitely warranted. What do you think?
Myron Yaster MD and Lynne Maxwell MD
PS: For our Jewish readers: Today is Yom Kippur, the day of atonement, we wish you an easy fast and a healthy and happy new year.