Dr. Caroline Andrew and Dr. David Nathan join the show to discuss the literature pertaining to intraoperative administration of dexamethasone.
Dr. Caroline Andrew is an anesthesia resident at the Massachusetts General Hospital. Dr. David Nathan is the Director of the Diabetes Center at Massachusetts General Hospital and Professor of Medicine at Harvard Medical School.
This podcast was recorded as part of the Depth of Anesthesia podcast elective.
Thanks for listening! If you enjoy our content, leave a 5-star review on Apple Podcasts and share our content with your colleagues.
—
Follow us on Instagram @DepthofAnesthesia and on Twitter (X) @DepthAnesthesia for podcast and literature updates.
Email us at depthofanesthesia@gmail.com with episode ideas or if you’d like to join our team.
Music by Stephen Campbell, MD.
—
References
Tien M, Gan TJ, Dhakal I, White WD, Olufolabi AJ, Fink R, Mishriky BM, Lacassie HJ, Habib AS. The effect of anti-emetic doses of dexamethasone on postoperative blood glucose levels in non-diabetic and diabetic patients: a prospective randomised controlled study. Anaesthesia. 2016 Sep;71(9):1037-43. doi: 10.1111/anae.13544. PMID: 27523051.
Polderman JA, Farhang-Razi V, Van Dieren S, Kranke P, DeVries JH, Hollmann MW, Preckel B, Hermanides J. Adverse side effects of dexamethasone in surgical patients. Cochrane Database Syst Rev. 2018 Nov 23;11(11):CD011940. doi: 10.1002/14651858.CD011940.pub3. PMID: 30480776; PMCID: PMC6426282.
Andrew J. Toner, Vyhunthan Ganeshanathan, Matthew T. Chan, Kwok M. Ho, Tomas B. Corcoran; Safety of Perioperative Glucocorticoids in Elective Noncardiac Surgery: A Systematic Review and Meta-analysis. Anesthesiology 2017; 126:234–248 doi: https://doi.org/10.1097/ALN.0000000000001466
Corcoran TB, O’Loughlin E, Chan MTV, Ho KM. Perioperative Administration of Dexamethasone And blood Glucose concentrations in patients undergoing elective non-cardiac surgery – the randomised controlled PADDAG trial. Eur J Anaesthesiol. 2021 Sep 1;38(9):932-942. doi: 10.1097/EJA.0000000000001294. PMID: 32833858.
Pang, QY., Wang, JY., Liang, XL. et al. The safety of perioperative dexamethasone with antiemetic dosage in surgical patients with diabetes mellitus: a systematic review and meta-analysis. Perioper Med 12, 4 (2023). https://doi.org/10.1186/s13741-023-00293-4
De Oliveira GS Jr, Castro-Alves LJ, Ahmad S, Kendall MC, McCarthy RJ. Dexamethasone to prevent postoperative nausea and vomiting: an updated meta-analysis of randomized controlled trials. Anesth Analg. 2013 Jan;116(1):58-74. doi: 10.1213/ANE.0b013e31826f0a0a. Epub 2012 Dec 7. PMID: 23223115.
—
AI-Generated Transcript
David Hao
Welcome back to depth of anesthesia. This is a podcast that critically explores our clinical practices. I’m David Hao, and I’m an anesthesiologist at the Massachusetts General Hospital. Our first guest today is doctor Caroline Andrew. Doctor Andrew is an anesthesia resident in our program here at MGH, and she recently completed the depth of anesthesia podcast elective. This has been a highly requested episode topic. I’ve been getting outreach on Twitter.
I’ve been getting outreach by email, so I’m really looking forward to discussing some of our findings here. And as always, a sincere thank you to Dr.
Daniel Saddawi-Konefka, Dr. Keith Baker, and Dr. Seun Johnson-Akeju for their ongoing support of this elective, which has really propelled this podcast forward. Doctor Andrew, welcome to the show.
Caroline Andrew
Thank you so much for having me, doctor Hao.
David Hao
And our second guest is Dr. David Nathan. He’s the director of the Diabetes Center at Mass General and he’s a professor of medicine at Harvard Medical School. He’s internationally recognized for his clinical research. He explores innovative strategies for maintaining blood sugar levels within the normal range in both type 1 and type 2 diabetes, and his work also examines the immediate and lasting impacts of these treatments on diabetes related complications. I can think of no better guest for this episode. Doctor Nathan, welcome to the show.
David Nathan
Well, thank you for that generous introduction. I’m delighted to join you.
David Hao
So our case today is that of a 65 year old female with a history of type 2 diabetes, hyperlipidemia, and hypertension who’s undergoing an elective laparoscopic hysterectomy for uterine fibroids. Her most recent hemoglobin a one c taken at her PCP’s office was 8.2. She undergoes an uneventful induction and intubation. And shortly thereafter, the resident draws up 6 milligrams of intravenous dexamethasone as part of their planned PONV prophylaxis regimen. The attending notices this and says, let’s actually avoid dexamethasone for this patient. She has a history of diabetes. Doctor Andrew, what are some of the claims in this case?
Caroline Andrew
So some of the claims we’re seeing in this case are, one, that dexamethasone in patients with diabetes may increase the risk of postoperative hyperglycemia and its associated complications. The second claim is that there is a hyperglycemic response to dexamethasone that is greater in patients with diabetes compared to in patients without diabetes.
David Hao
Here now are 2 questions for our listeners to think about. What is the level of your agreement with the claim that intraoperative dexamethasone for p ONV prophylaxis should be avoided in patients with diabetes? And what is the level of evidence for what you believe? Caroline, how’d you get interested in exploring the subject?
Caroline Andrew
So I’ve always been interested in understanding some of the claims we make in medicine and if these are grounded and reasonable evidence based. With regards to dexamethasone use in patients with diabetes, I’ve seen patients with diabetes receive intraoperative dexamethasone without any postoperative hyperglycemia. I’ve also seen the opposite, and I’ve been curious to better understand the evidence behind the pattern and the magnitude of the hyperglycemic response, specifically whether these differ in diabetic and nondiabetic patients.
David Hao
Yeah. I’ve certainly had some of those questions myself. Before we jump into the topic at hand, doctor Nathan, could you give our listeners some background on the effects of glucocorticoids in general?
David Nathan
Well, I think it’s well known to all of our listeners that glucocorticoids raise glucose levels. They increase gluconeogenesis in the liver. They are, of course, the quintessential stress hormone, and cause insulin resistance as well. Long term, they have effects on fat distribution, which also are diabetogenic, if you will. But here what we’re discussing really are the acute effects of dexamethasone given perioperatively. As the listeners I’m sure all know, it’s one of the recommended drugs to decrease postoperative nausea and vomiting. And I think what we’re gonna be looking at today are what are the potential adverse consequences of giving the decadron.
And along the way, I think we need to discuss how dexamethasone, decadron, differs from physiologic steroids.
David Hao
Thanks for that overview. As you mentioned, there are guidelines relating to use of dexamethasone in this context. As an example, the 4th consensus guidelines for the management of postoperative nausea and vomiting, this was published in ANA, recommends implementing multimodal POMV prophylaxis in patients with 1 or 2 risk factors. Institutionally, at MGH, we use dexamethasone, Zofran, and haloperidol as our first line intraoperative agents. I’ll also point out that PONV prophylaxis is increasingly tracked as a quality measure. The MPOG measure for PONV prophylaxis measures success as quote, at least 2 prophylactic pharmacologic antiemetic agents of different classes administered preoperatively or intraoperatively for patients with 1 or 2 risk factors. So suffice to say, I think this is a very pertinent topic and our listeners are looking forward to discussing some of the findings.
Caroline, can you tell us a bit about this first study, a prospective randomized controlled trial from 2016?
Caroline Andrew
Yes. Of course.
So, yes, this first study by Tien et al is, as you mentioned, a prospective randomized controlled trial looking at about 80 patients, 40 patients who did not have diabetes, and 40 patients with a diagnosis of type 2 diabetes. Exclusion criteria for the study included things like pregnancy, patients with perioperative blood glucose levels greater than 200 milligrams per deciliter, and patients who were currently receiving steroid therapy. These patients were then randomly allocated to receive either 8 milligrams intravenous dexamethasone at the induction of the anesthesia or 4 milligrams ondansetron towards the end of the surgical procedure. The authors then measured the mean maximum blood glucose in the first 24 hours postoperatively and found that the mean maximum blood sugar was about a 164 milligrams per deciliter in nondiabetic patients receiving dexamethasone versus about a 140 milligrams per deciliter in nondiabetic patients receiving ondansetron. For patients with known type 2 diabetes, the average maximum blood glucose measured in the first 24 hours postoperatively was 252 milligrams per deciliter in the dexamethasone group versus a 194 milligrams per deciliter in the ondansetron group. The authors found that both of these findings were statistically significant. The authors also found that the maximum change in blood glucose levels from baseline preoperatively to 4 hours and 24 hours postoperatively was significantly greater in diabetic patients receiving dexamethasone compared with those receiving ondansetron.
So for patients with type 2 diabetes who received dexamethasone, the maximum blood glucose change from baseline to 24 hours was measured at a 126 milligrams per deciliter, while the maximum change is only 63 milligrams per deciliter in diabetic patients who received ondansetron. Inter interestingly, this degree of glucose variation was not statistically significant between the dexamethasone and ondansetron groups and nondiabetic patients.
David Hao
So we have a prospective RCT that suggests that the average maximum blood glucose within 24 hours was higher in both nondiabetic and diabetic patients receiving dexamethasone when compared to Zofran. So that’s in terms of the absolute value. In terms of the change from baseline, in diabetic patients, blood glucose increased significantly more at 4 and 24 hours in those treated with dexamethasone compared to Zofran, and this was not the case with nondiabetic patients. Caroline, you mentioned a key exclusion. Patients with preoperative blood glucose levels greater than around 200 milligrams per deciliter were excluded and none of the diabetic patients took their oral hypoglycemic drugs on the morning of surgery.
Doctor Nathan, any thoughts?
David Nathan
Well, I mean, so this is a well conducted randomized controlled trial. We do need to keep in mind that, of course, it was conducted ethically, and that, as Caroline mentioned, of course, those folks with diabetes, whose glucoses rose required more insulin. So it’s not as if we’re looking at this as a kind of a Odensitron versus Decadron and then nothing else. I mean, these patients are being managed clinically, of course, as appropriate in the clinical setting. And they are requiring more insulin. So this is kind of part of the wake up goal, of course, that anesthesiologists and those, clinicians who are managing the patients intraoperatively, that when you give decadron, you can expect that there’s gonna be a rise in glucose. If they had not given extra insulin, for example, to these patients, or the participants in this study, their glucose levels would have risen far more.
So what happens in the nondiabetics? Their glucose levels are going up also. This is the effects of of steroids, glucocorticoids in general, on both diabetic patients and nondiabetic is that it raises glucose levels. But with them, of course, their intact pancreases were managing just fine.
They made more insulin. Their glucose levels rose, but not into a level that would be considered clinically troublesome. So again, this is evidence that their own pancreasies are working. It’s not hardly anyone remembers it anymore, but there was actually a glucocorticoid challenge test for the diagnosis of diabetes in the 19 forties fifties. So they actually gave glucocorticoids to see how high glucose level would rise as a way of diagnosing diabetes. So what we’re looking at is this intraoperatively. The other issue that I think is worth mentioning is that there are very few patients in any of the studies we’re gonna review with type 1 diabetes.
So most of them have the epidemic form of diabetes type 2. Their pancreases are more or less intact, not functioning normally, but still functioning. If you take a type 1 diabetic patient who makes no insulin on his or her own, the effects of glucocorticoids in that setting are likely to be more dramatic.
David Hao
I think those are all very thoughtful points. And to your comment about managing elevated blood glucose levels, type 2 diabetics in this study receiving dexamethasone did require higher insulin doses over the 24 hour duration of the study and were more likely to require insulin insulin in the recovery area around 70% compared to 33% of those type 2 diabetics who got Sofran. They also describe in the manuscript that aside from the specified time points, quote, additional measurements were performed as is our normal standard of care, end quote, the measurements being blood glucose levels, and insulin was given according to an institutional sliding scale. So put simply, these patients were being actively managed for their hyperglycemia. We are somewhat fortunate with this body of literature that there are a number of systematic reviews which aggregate the findings for some of these smaller studies. And one such review is a Cochrane review published in 2018.
Caroline, can you introduce us to the study?
Caroline Andrew
Yes. So as, doctor Hao mentioned, this next study is a systematic review that looked at the adverse side effects of dexamethasone in all surgical patients, so not limited to patients with diabetes. The review included about 37 RCTs, and each study looked at surgical patients who received dex dexamethasone as an intervention and included a sufficient follow-up period to monitor for adverse effects. In total, about 25 100 participants received dexamethasone and 25 100 participants received some sort of control intervention and these are mostly, Zofran or 0.9% normal saline. The primary outcome as assessed in the review was postoperative wound or systemic infection, delayed wound healing, and the glycemic response. The authors found no statistically significant difference in postoperative wound difference in postoperative wound or systemic infection in the dexamethasone versus the control groups. They also found that dexamethasone did not increase the incidence of delayed wound healing.
David Hao
Caroline, before you head into the findings in terms of the glycemic response, I wanna take a moment to talk a little bit about what the authors did. It’s a bit confusing. So their primary outcome was glycemic response within 24 hours, and this was defined as the difference between preoperative and postoperative blood glucose. They mentioned that because the peak serum level of dexamethasone is achieved 2 to 12 hours after injection, the postoperative blood glucose level had to be measured at least 2 hours but within 24 hours after dexamethasone administration. So they looked at the change from baseline to 12 hours and at 24 hours postoperatively in patients without diabetes. And for patients with diabetes, they looked at the change from baseline to 24 hours postop. But when they actually reported the findings, findings, they reported on the response within 10 to 24 hours.
Hope that makes it a little bit less confusing. Alright, Caroline. What were their findings?
Caroline Andrew
For patients without, diabetes, the authors found that at 2 to 12 hours postoperatively, there was a statistically significant increase in blood glucose levels from baseline in the dexamethasone group compared to the control group. The mean increase in blood sugar in the dexamethasone group was about 13 milligrams per deciliter. At 24 hours postoperatively, there was also a statistically significant increase in blood glucose levels in the dexamethasone group with a mean change of about 21 milligrams per deciliter. In patients with diabetes, the change in blood glucose levels from baseline to 10 to 24 hours postoperatively was again statistically significant in the dexamethasone group with a mean increase of 32 milligrams per deciliter.
David Hao
So I think we’re getting a sense of the impact of dexamethasone perioperatively. Doctor Nathan, any thoughts about how we can put these numbers into context and and any commentary on how we should interpret the different time intervals that this Cochrane review selected for patients with and without diabetes?
David Nathan
I’m gonna put on my diabetes and endocrinology and maybe physiology hat here for a moment and just remind, our listeners that surgery is not a natural state. It represents a very, an acute stress state. And all of us, forget about whether we’re giving steroids intraoperatively, but everyone mounts a stress response. Unless you have Addison’s disease and your adrenals are shot or something. But otherwise everyone is making steroids during this period of time, which is why comparing this pharmacological administration of steroid needs to you need to keep in mind that in fact, we are all responding in a stress response during the procedure. So, the difference here is that decadron is a very non physiologic steroid. It’s used specifically because it’ll be given once a day, but it is a very steady step provides a very steady state of steroid levels versus what we do.
What we naturally do is, you know, we have hypothalamic stimuli from the suprachiasmatic nucleus, which then goes to the pituitary. ACTH gets secreted, which then goes to the adrenal glands. And normally we have a diurnal variation in steroid secretion. And in a 24 hour period, we used to think that people made about 20 milligrams of hydroxycortisone or cortisol in a 24 hour period. Now with better measurements, more specific measurements of steroids, it’s probably closer to 10 milligrams. Now during stress, assuming you have a normally functioning adrenal gland, and you know, what kind of stress? A sepsis, an MI, or anesthe, induction of anesthesia, you ramp up that steroid secretion dramatically to something like tenfold.
So instead of if you if you made that over a 24 hour period, instead of 10 milligrams, you might be making between 102 100 milligrams of cortisol in a 24 hour period. But nobody’s under that kind of stress for 24 hours. It just doesn’t happen. You know, stress waxes and wanes. When we give decadron, we’re giving a steroid that lasts for at least 24 and more like 36 hours with the physiologic effect going even several days after that. So now we’re giving a single intravenous administration. And at, let’s say, 8 milligrams, we’re giving them that one injection more steroids than you would make under usual stress in a 24 hour period ever.
And so the concerns are, you know, one is we’ve talked about already, does it raise glucose levels? And it does. Know modestly in nondiabetics because their pancreases respond. And in people with diabetes, since they have impaired pancreatic function, it raises it even more. The concern has been, and this has been well addressed in this meta analysis, that, you know, maybe it’s also causing problems that we see in people who have chronically elevated steroids, lower steroid levels, such as in Cushing’s disease. But keep in mind that even though the dechedron is a kind of a longish exposure, it does go away. This is not the same as a person with Cushing’s disease who makes elevated levels of steroids day in and day out, year after year.
And with Cushing’s disease, we know that people get thin skin, they have trouble with wound healing, and that was one of the major concerns. And what this meta analysis demonstrates, this Cochrane meta analysis, is that that’s not a concern. So one of the major safety issues that we might be concerned about I think, is dispelled, by this study, this set of studies, and Caroline description of it.
David Hao
Well said. And to some extent, we do focus a bit more on the glycemic impact of dexamethasone exposure in this episode. But wound healing and infection are significant considerations, and so I’m glad you touched on that.
David Nathan
If I can expand on, the issue about the timing, I must say that I got involved in this when I’m doing my clinical diabetes work in patient who’s been given intraoperative decodron. And then it’s like 2, 3, 4 days later, they’re now on the floor, and they’re requiring increased doses of insulin. So the effects of the decadron, although acutely, these studies all look over 24 hours, in fact there is an even a longer period of time when their glucose levels may be elevated. And this responsibility now passes from the anesthesiology staff, even from the surgeons in the ICU or the recovery room, now to the floor physician. So we have to just keep in mind that insulin requirements may still be transiently higher even for 48 hours, maybe for several days, after they got their decodron.
David Hao
I do think that’s very valuable context for us to understand, when we consider the care of a patient beyond the immediate perioperative environment. So just a year prior to the publication of this Cochrane review, a systematic review and meta analysis of 56 trials was published in Anesthesiology examining 3 specific safety concerns in non cardiac surgical patients. They looked specifically at possible increases in the risk of hyperglycemia, infection, and perioperative bleeding relating to perioperative glucocorticoid administration. In terms of their takeaways, similar to the Cochrane review, they found no difference in the incidence of any wound infection in patients who are randomized to perioperative glucocorticoid with the caveat that the collated trials lack sufficient surveillance and power to detect clinically important differences and complications such as wound infection. In terms of the glycemic response, the highest glucose value measured at a specific time intraoperatively or within the first 12 hours after surgery was greater in patients receiving glucocorticoids. There was a weighted mean difference of around 20 milligrams per deciliter. And when they excluded studies looking at diabetic patients, there was a smaller but still a statistically significant rise in blood glucose with a weighted mean difference of 14 milligrams per deciliter.
There have also been some interesting trials published subsequently, and one of these was published by Corcoran and colleagues in 2021, and they’d set out to answer some specific questions. 1st, they wondered, are perioperative blood glucose profiles different after the administration of a single dose of 4 milligrams of dexamethasone, 8 milligrams of dexamethasone, or placebo in both diabetic and nondiabetic patients? And second, is there an interaction between a patient’s preoperative hyperglycemic control and the effect of dexamethasone on maximal perioperative blood glucose concentrations? Certainly, 2 very interesting and, clinically relevant questions that have not been answered with the literature up to that point. Caroline, can you tell us about this article?
Caroline Andrew
Of course. So in this RCT, the authors looked at at about 300 patients scheduled for elective cases under general anesthesia. These cases needed to have at least a skin incision, be longer than 1 hour, and required at least one night in the hospital. The authors didn’t look at any cases that were maxillofacial, intracranial, cardiac, or obstetric. Of note, patients with hemoglobin a one c’s greater than 9.1 were excluded from the study. Of the 300 patients in the study, only about 20% had known type 1 or type 2 diabetes, and mostly, the patients included were type 2 diabetics. These participants were randomly allocated to receive either dexamethasone 4 milligrams, dexamethasone 8 milligrams, or a placebo control, which in this case was 0.9% normal saline.
Authors then stratified the patients by their diabetes status. The primary outcome was perioperative blood glucose profiles during the interoperative and postoperative periods up to 24 hours after the end of surgery. The secondary outcome was the interaction between preoperative hemoglobin a one c’s and the dose of dexamethasone on maximum perioperative blood glucose concentrations. So the authors actually found that neither 4 milligrams of dexamethasone nor 8 milligrams of dexamethasone had a statistically significant effect on the maximal blood glucose concentrations within 24 hours of surgery in both diabetic or nondiabetic patients. For the secondary outcome, the authors used a linear regression model to assess the effect of preoperative hemoglobin a one c on postoperative blood glucose levels and found that the maximum blood glucose concentrations within 24 hours correlated positively with preoperative hemoglobin a one c. So based on this linear regression model, the authors found that the maximum blood glucose concentrations were 34 milligrams per deciliter, 30 milligrams per deciliter, and 72 milligrams per deciliter higher per 1% increment in baseline hemoglobin a one c after receiving placebo, 4 milligrams, and 8 milligrams dexamethasone, respectively. And this is a lot of numbers to kinda wrap our heads around, but, overall, the authors are suggesting that in patients with high baseline hemoglobin a one c’s, only 8 milligrams of defamexasone increases maximum blood glucose substantially compared with placebo or 4 milligrams of dexamethasone.
David Hao
Super interesting findings. Doctor Nathan, I was wondering if you had any thoughts before I take a moment to try to summarize this trial.
David Nathan
Well, you know, I think what has just been brought out is that the higher your glucose levels going into the surgery, the more likely it is that you’re gonna have higher blood glucose levels during post operatively and potentially during the surgery. And that they may be a need to be addressed as we mentioned earlier with higher doses of insulin, perioperatively. It’s worth pointing out as Caroline did that patients with a one c levels greater than 9 were excluded as as has been done in many of these studies, that were analyzed in the meta analysis, for example. And I think in this study, if I remember correctly, the median A1c in the diabetic group was 6 point 8, which is, you know, great. If all of our patients had a one c’s at around 6.8, we’d be very happy and the incidence of complications would be far lower. So it’s it’s kind of interesting when you select for for patients who have better glucose control going into the surgery, maybe not surprisingly, they have less of an adverse effect on glycemia. And as Caroline noted, the higher your A1c, the greater the impact on the glucose levels.
The other point I is worth bringing out is that we haven’t talked much about the dose of decodram that’s given. I did mention that 8 milligrams is basically higher translates into a higher amount of glucocorticoids than any human being would ever make in a 24 hour period. The studies that have compared different doses of, glucocorticoids to prevent the postoperative nausea and vomiting have not really been so clear in terms of whether you need 4 or 8 or 12 or 20. But certainly the more you give, the higher the glucose levels. And the more persistent the adverse effect on glucose levels. You know, I only, I got drawn into this during a conference with the, our anesthesia department, about a year ago or so. And I, of course, I didn’t understand why it was always 4 or 8 or 12 or whatever.
And of course the vials come in 4 milligram vials. So instead of giving 6, you give 2 vials and give 8. It’s not clear to me that the rationale for doing that in terms of preventing the nausea, vomiting, decreasing pain is so well established. But what Caroline has just talked about, I think here, is that with 4 or 8 milligrams of decadron and a high A1c, glucose levels are sure to rise, for some period of time postoperatively.
David Hao
And to your point, there are different indications certainly for perioperative administration of dexamethasone. A meta analysis from 2013 in ANA indicated that 4 to 5 milligram doses were comparable to 8 to 10 milligrams for the reduction of PONV. But dexamethasone doses closer to 0.1mgs per kg, perhaps more effective for purposes of reduction in postoperative pain and opioid consumption. A few additional points about this trial, the trial was terminated earlier than planned, and this was due to competition for enrolling similar patients in the perioperative administration of dexamethasone and infection trial. And so this reduced its overall statistical power. The study, similar many other studies, did exclude patients with poorly controlled diabetes. In this case, that was a hemoglobin a one c greater than 9.
So what we have is a multicenter, triple blinded, randomized controlled trial that found that administering a single dose of dexamethasone, either 4 or 8 milligrams at the start of anesthesia, did not have a significant effect on maximal blood glucose concentrations within 24 hours in either well controlled diabetic or nondiabetic patients as compared to placebo. But what they did find is that maximum perioperative blood glucose concentrations in patients with diabetes were indeed related to baseline hemoglobin a one c values in a concentration dependent fashion only after administration of 8 milligrams of dexamethasone as opposed to 4. Caroline, I believe our last study that we’re gonna be talking about today is a very recent systematic review and meta analysis that synthesizes the literature up to June of 2022 and, fortunately, includes both the TN and Corcoran studies. What can you tell us about this manuscript?
Caroline Andrew
Yes. So this systematic review by Peng et al, it included 16 studies, and their primary outcome of interest in the study was glucose response after intraoperative dexamethasone, which included glucose changes after dexamethasone administration and peak glucose levels. And all the patients in this study were either patients with known type 1 or type 2 diabetes, but, again, there were more patients with type 2 diabetes included. Secondary outcomes included insulin requirements postoperatively, wound healing, and infection. The authors found that compared with control, dexamethasone increased glucose levels in diabetic patients intraoperatively at the end of surgery on postoperative day 1 and on postoperative day 2. The increases in blood glucose compared to placebo were relatively small with an average increase of about 8 milligrams per deciliter intraoperatively, about 15 milligrams per deciliter at the end of surgery, 19.5 milligrams per deciliter on postoperative day 1, and 9 milligrams per deciliter on postoperative day 2. The study did show that blood glucose reached its peak level within 24 hours after surgery in both dexamethasone and control groups, which was interesting, and it did decrease thereafter.
The meta analysis from all the studies also showed that dexamethasone increased peak glucose levels compared with the control group by about 36 milligrams per deciliter in a 24 hour period. Within the study, they looked at 2 RCTs and 2 cohort studies and, really, only those 4 studies looked at wound healing and infection, and the pooled results found that there was no between group difference in wound infection or dehiscence in patients with diabetes who received dexamethasone versus a control. Lastly, only one RCT in 3 cohort studies within the review commented on insulin requirements postoperatively, And, overall, they could not find any difference in insulin requirements between the dexamethasone and control group. But, again, they they weren’t able to include a lot of studies that looked at this.
David Hao
And to your point about the limitations, the authors describe their results as low to moderate evidence from the RCTs and very low to low evidence from the cohort studies. As might be expected, the dosages of dexamethasone, the types of surgeries included, they were not standardized so there’s tons of heterogeneity. And just with everything else that we’ve talked about thus far, the RCTs often did not include patients with poorly controlled diabetes. So it’s very difficult to extrapolate those findings, beyond the reasonably well controlled population. Doctor Nathan, any thoughts about this, closing review?
David Nathan
Well, no. I think it adds to our confidence that dexamethasone treatment intraoperatively is not going to increase wound dehiscence or infections. There are really a number of studies that have now endorsed that. Interestingly, I must say that in these studies that include an arm with decadron and then the arm that doesn’t get the decadron gets other anti emetic therapy, they don’t often point out whether the rates of nausea and vomiting have actually been reduced with dexamethasone. That was said in the Corcoran study where if I remember correctly it was about a 22% reduction. So the patients would get whatever therapy. They could get the Zofran.
They would get what other antiemetics were used routinely, But it was plus or minus decodron. And the decodron really does have a significant and I would say nominally a pretty substantial effect on reducing nausea and vomiting. So it’s doing what it’s supposed to do. I think that this study again reassures us that they’re not gonna get dehiscence or increased wound infections. This is after all only a modest period of time when they have high cortisol levels. So to endocrinologists, that’s not surprising. And then the other thing, the remaining topic, which we focused on, Caroline is focused on in in great detail, is, of course, the glycemia.
And it’s gonna go up. But again, all of these studies are to some extent confounded because at the same time as they’re doing the study they are giving insulin as needed, in the study, which diminishes any potential differences in glucose. And as we’ve pointed out before, you know, once they’re out of the OR and once they’re out of the PACU, then that’s when we get called in, the diabetes folks, to try to, again, bring the glucose levels back into control in the subsequent days weeks ahead.
David Hao
And to your point about insulin, the only RCT that was included that commented on this issue was the Tien et al manuscript that we reviewed at the start of the show. As a reminder, they reported that more type 2 diabetic patients given dexamethasone required insulin in the recovery area compared to those given Zofran. There was no difference in the number of patients requiring insulin over 24 hours, but more insulin was administered in those who were given dexamethasone. There’s only one study that looked at insulin requirements on post op day 1. This was a cohort study, which showed no between group differences, but the quality of this evidence was very low. So to summarize, this is the only systematic review and meta analysis to date that has examined the safety of dexamethasone in patients specifically with diabetes. Dexamethasone increased glucose intraoperatively and up to postop date 2, but the overall level of evidence for the glucose response was very low to low.
And, of course, there were a number of limitations pertaining to the differences in types of doses and types of surgeries. Caroline, I think that wraps up our last study. Would you mind sharing some of your takeaways with our listeners?
Caroline Andrew
Yes. I would love to.
So I think overall, the evidence does suggest that you will see an increase in postoperative blood glucose in patients with and without diabetes who receive either 4 or 8 milligrams of dexamethasone intraoperatively. And this increase in blood glucose is likely higher in patients with diabetes. I think it remains unclear if this temporary increase in postoperative blood glucose is significant since they were not able to demonstrate an increased rate of infection or delayed wound healing. But, again, they are mostly looking at, postoperative complications that affect the surgical community and, again, not what doctor Nathan often deals with, days later on the floor. And one thing I do wanna note for the listeners that we have talked about a couple of times this podcast is that pretty much all of these studies excluded patients with poorly controlled diabetes, which is one thing to consider when extrapolating these results to our surgical patients who come in with higher hemoglobin a one c’s.
David Hao
Appreciate those thoughts. And and as you were talking, another thought crossed my mind is we also don’t have any data on patients who repeatedly go back to the OR. And I can imagine if they’re having repeat procedures time after time that that could potentially look a little bit different if they’re getting PONV prophylaxis with dexamethasone every other day. And, doctor Nathan, I do wanna finish the episode with you and and to just get your overall thoughts on both the body of literature that we’ve discussed and, as well as any additional insights you have from your own clinical experience in managing these patients postoperatively.
David Nathan
Well, first, I want to thank and congratulate Caroline in summarizing what is now a large body of literature. I mean, you know, when we’re reviewing 3 meta analyses, we know that there are lots of trials that are going into them. And she just did a splendid job. I must say that my initial, you know, as irritating as diabetes specialists find it to kind of manage these patients several days after, they come out of the operating room. I must say that I am increasingly impressed that this is the right therapy to use. And that the consequences of some short term, relatively short term hyperglycemia, But no increased risk for dehiscence, for wound infections, for all the things that people were concerned about initially. But that the benefit they have in terms of reducing, PONV makes these drugs important, tools, I think, to use.
And just that we have to be alert that they are gonna raise glycemia, as Caroline has pointed out, and that we need to deal with that. But fortunately, it’s really a short term, relatively short term effect, which we can deal with.
David Hao
All right. Well, I want to thank you both for taking time to join the show. I think this is gonna be an incredibly useful episode, not only because we’ve managed to cover a lot of the primary literature, but we’ve also been very fortunate to have doctor Nathan share some of his clinical insights, from somebody who actively takes care of and manages these patients, through their hospital courses. So I wanna thank, first of all, Caroline for, all the work she’s done in preparing this episode, doing the research, and then to doctor Nathan for taking the time to join us and and sharing his clinical insights as well as his, interpretation of some of these studies.
Caroline Andrew
Thank you so much for having us.
David Nathan
My pleasure.
David Hao
Thank you to all our listeners for tuning in today. I know it’s been a long minute since our last episode. I’ve been quite a bit busy with all sorts of different things. I had a very interesting experience at a friend’s birthday party where I actually met some colleagues who listened to the podcast. Those interactions are always kinda surreal for me, but they’re definitely a very nice reminder to both myself and members of our team that our work here is interesting and hopefully helpful. If you have questions or ideas for future episodes or wanna help us in any way, you can reach out to us through our social media platforms, platforms, including Instagram and Twitter, or you can email us at depthofanesthesia@gmail.com. As always, stay hungry and keep asking questions.
Podcast: Play in new window | Download
