Perioperative respiratory adverse events in T&A patients: Does the premed make a difference?
Myron Yaster MD and Melissa Brooks Peterson MD
Adenotonsillectomy (AT) is amongst the most common surgical procedures performed in children. And yet, despite the fact that millions of children undergo this surgical procedure yearly, there is little consensus on how to provide anesthesia, analgesia, and sedation perioperatively to these patients. Even the indications for the procedure – sleep disordered breathing and obstructive sleep apnea syndrome (OSAS) – are rarely confirmed diagnostically with sleep studies, which leaves the diagnos(es) and indication(s) for surgery suspect. I (Myron) have always said (not so facetiously): the primary indication for this surgery is insurance coverage and the surgeon’s OR schedule. I’ve often complained about this at SPA Quality and Safety committee meetings and privately to my friends on Wake Up Safe.
Today’s Pediatric Anesthesia Article of the Day by Shen et al.1 and its accompanying editorial by Dalesio and Kudchadkar2 discuss the effects of intranasal midazolam vs intranasal dexmedetomidine vs a control group of intranasal saline on perioperative respiratory adverse events (PRAEs). Although the results of this randomized controlled trial appear clear cut (that dexmedetomidine produced fewer complications), the actual intraoperative anesthetic plan(s) cloud the results and highlights that we still have more questions than answers.
I (MBP) will add that I am a staunch avoider of midazolam and always have been, so I review this paper with a high degree of bias against the use of midazolam as a premedication. I have routinely observed that children who get midazolam wake up effectively drunk (and often disoriented or angry) in the PACU. Adenotonsillectomy and eye surgery only exacerbates this phenomenon. In the (very rare) case that environmental control and non-pharmacologic tools cannot safely get a child into the operating room, and a premedication is really necessary, I will use either dexmedetomidine or fentanyl in lieu of midazolam. As I have routinely told many trainees and colleagues over the last decade of my work at Children’s Colorado: “Midazolam is a dirty, nasty drug and it should be abandoned in favor of cleaner, better alternatives. And, no one needs a drunk toddler in the PACU when it can be avoided.” I (MY) on the other hand, disagree with Mel on the use of midazolam perioperatively and based on personal experience and the landmark studies by Kain et al. believe it should be used almost ubiquitously, in the preop prep area to limit stormy inhalational inductions. We both agree that oral acetaminophen should ideally given to virtually all pediatric patients to limit post operative pain.
Myron Yaster MD and Melissa Brooks Peterson MD
Editorial
Nicholas M Dalesio, Sapna R Kudchadkar. Perioperative Respiratory Adverse Events After Pediatric Adenotonsillectomy-Evaluating the Role of Preoperative Pharmacologic Anxiolysis. JAMA Netw Open. 2022 Aug 1;5(8):e2225482. doi: 10.1001/jamanetworkopen.2022.25482. PMID: 35943746
Original article
Fangming Shen, Qin Zhang, Yahui Xu, Xinghe Wang, Jiayi Xia, Chao Chen, He Liu, Yueying Zhang. JAMA Netw Open. 2022 Aug 1;5(8):e2225473. doi: 10.1001/jamanetworkopen.2022.25473. Effect of Intranasal Dexmedetomidine or Midazolam for Premedication on the Occurrence of Respiratory Adverse Events in Children Undergoing Tonsillectomy and Adenoidectomy: A Randomized Clinical Trial. PMID: 35943745 PMCID: PMC9364121
Since the landmark studies by Zeev Kain and colleagues that began in the 1990s, oral midazolam has been the go-to premedication for anxiolysis +/- sedation in pediatric anesthetic practice.3-5 A challenger, intranasal dexmedetomidine, is increasingly becoming a popular alternative because of its analgesic, anti-emetic and anti-delirium effects.6 Shen et al. hypothesized that perioperative respiratory adverse events (PRAEs), manifested as minor adverse events (oxygen desaturation, airway obstruction, coughing, or wheezing) and major adverse events (laryngospasm and bronchospasm) would be less common in children premedicated with intranasal dexmedetomidine vs intranasal midazolam. The authors’ definitions of PRAE are the same as PRAE described by Von Ungern-Sternberg et al. in their series of outstanding papers, which can be considered the “gold standard” of PRAE definition for our specialty. They were buried in this paper in “Appendix 4” in “supplemental material 1,” so we have included that table below for your reference. In this study, apnea and respiratory depression were not included as a major PRAE. Intranasal saline was used as a control. Independent risk factors for PRAE include age 6 years and younger, upper respiratory tract infection (URTI), lung disease, obesity, obstructive sleep apnea (OSA), and passive smoking. Indeed, passive smoking was found in almost 50% of the patients included in this study and underscores the pervasiveness of smoking in China.
The results of this randomized, double blind, controlled trial are best visualized in the figure that was part of the supplemental figures included in the publication. Cutting to the chase: patients premedicated with intranasal dexmedetomidine had fewer complications than those treated with midazolam (or saline).
There are a few problems with this study. The details of the intra-and post-operative management are laid out only in the supplementary text and tables, perhaps because it was published in a general medical journal and not an anesthesiology journal. Many of the findings of this study may be explained or tempered by how the anesthetic was conducted. As highlighted by the editorial by Dalesio and Kudchadkar2, “Children with OSAS are at increased risk for major PRAEs when undergoing adenotonsillectomy (AT) at a rate 5 times higher than that in children without OSAS. Furthermore, it is well documented that patients with severe OSAS have higher sensitivity to opioids, and it is recommended that opioid dosing be reduced by half in these patients to mitigate respiratory depression. Although Shen et al.1 note different frequencies of OSAS between groups, the severity of disease was not accounted for and the method for diagnosing OSAS was not standardized. No patient underwent polysomnography, the criterion standard diagnostic test to identify the presence and severity of OSAS. Instead, patients were designated as having OSAS according to a clinical diagnosis made by the otolaryngologist. Clinical diagnosis of OSAS has poor accuracy, with rates ranging from 30% to 85%.7 Therefore, the randomized groups may have represented an unequal distribution of patients with severe OSAS because the patient population was not appropriately screened, diagnosed, and controlled for in the analysis”.2
What was the actual anesthetic technique in this study? About 90% of patients underwent an IV induction, which consisted of 1 mcg/kg dexmedetomidine, etomidate 0.3 mg/kg, fentanyl 1-2 mcg/kg and cisatracurium 0.15 mg/kg. In those who underwent a mask induction, once an IV was placed, ancillary drugs were given (and were not specified). Following intubation, maintenance consisted of sevoflurane, propofol 2-4 mg/kg/hour, and remifentanil 10-18 mcg/kg/hour. At the conclusion of surgery, patients were extubated in the PACU and reversal of neuromuscular blockade with neostigmine+atropine was based on clinical signs, with no standardization and no monitoring. Pain in the PACU was treated with fentanyl 1-2 mcg/kg. Thus, patients in both groups received dexmedetomidine either as the premed intranasally and/or IV intraoperatively at induction. Given that every patient got dexmedetomidine, just at different total doses, and some patients got midazolam in addition to the low- or higher-dose dexmedtomidine, the study groups needed to have been defined that way. Though it is “hidden” in the supplemental materials, this is a critical weakenss of this study and should have been addressed in the discussion. As an additional concern with resepect to the pharmacology design of this study, all patients were paralyzed but neuromuscular blockade was not monitored and not all patients were reversed. Could some of the reported PRAEs been the result of residual neuromuscular blockade? –Yes! , and this is another major weakness of this paper. As readers of the PAAD know, I (MY) have been on a crusade urging the use of quantitative neuromuscular blockade monitoring whenever a paralytic is used in the OR.
Perioperative pain control was also problematic in this study. Weak analgesics with anti-pyretic effects like acetaminophen and ibuprofen were not used but both fentanyl and remifentanil were. The use of remifentanil in this population makes no sense to us, as the rapidity of its offset is almost certainly a cause of PACU agitation, pain, and delirium, which pain scores will not differentiate. Finally, PACU extubation remains uncommon in the U.S., which begs the question if the results may have been different if extubation occurred in the OR.? Would the dexmedetomidine group have delayed emergence, wakeup and extubation?
Despite the clear issues with this paper in its design and methodology, it does raise an interesting finding of a suggestion of increased PRAE with the use of midazolam as a premedication. As many papers the PAAD reviews, this (somewhat problematic) study should spur us as a community to design and perform multi-center trials of the perioperative management of adenotonsillectomy.
References
1. Shen F, Zhang Q, Xu Y, Wang X, Xia J, Chen C, Liu H, Zhang Y: Effect of Intranasal Dexmedetomidine or Midazolam for Premedication on the Occurrence of Respiratory Adverse Events in Children Undergoing Tonsillectomy and Adenoidectomy: A Randomized Clinical Trial. JAMA Netw Open 2022; 5: e2225473
2. Dalesio NM, Kudchadkar SR: Perioperative Respiratory Adverse Events After Pediatric Adenotonsillectomy-Evaluating the Role of Preoperative Pharmacologic Anxiolysis. JAMA Netw Open 2022; 5: e2225482
3. Kain ZN, Caldwell-Andrews AA, Maranets I, McClain B, Gaal D, Mayes LC, Feng R, Zhang H: Preoperative anxiety and emergence delirium and postoperative maladaptive behaviors. Anesth Analg 2004; 99: 1648-1654
4. Kain ZN, Mayes LC, Wang SM, Caramico LA, Krivutza DM, Hofstadter MB: Parental presence and a sedative premedicant for children undergoing surgery: a hierarchical study. Anesthesiology 2000; 92: 939-46
5. Kain ZN, Mayes LC, Wang SM, Caramico LA, Hofstadter MB: Parental presence during induction of anesthesia versus sedative premedication: which intervention is more effective? Anesthesiology 1998; 89: 1147-56; discussion 9A-10A
6. Jun JH, Kim KN, Kim JY, Song SM: The effects of intranasal dexmedetomidine premedication in children: a systematic review and meta-analysis. Can J Anaesth 2017; 64: 947-961
7. Goldstein NA, Stefanov DG, Graw-Panzer KD, Fahmy SA, Fishkin S, Jackson A, Sarhis JS, Weedon J: Validation of a clinical assessment score for pediatric sleep-disordered breathing. Laryngoscope 2012; 122: 2096-104