Does morphine cause more nausea than hydromorphone?

Welcome back to the Depth of Anesthesia. This is a podcast where we critically explore our clinical practices. Thank you to our listeners. We've had an overwhelmingly positive response to the first episode, and because this is a new podcast, we definitely rely on your word of mouth to share this. So if you enjoyed the show, leave us a five-star review on iTunes, and we can also be reached on depth of anesthesia. com.

You'll find up-to-date show notes there as well as references for all the studies that we talk about. You can also drop a comment there and get in touch with us. My name is David Hao and I'm an anesthesia resident at the Massachusetts General Hospital, and I'd like to welcome Dan Saddawi-Konefka back onto the show Hi, I'm uh Dan Saddawi-Konefka. As David said, I'm the program director here at MGH. So Dan, what are we talking about today?

Well, I think we're going to be talking about hydromorphone and morphine and I think you had a little case for us, right? I do. So the case today is a twenty three year old woman. Uh she's healthy and undergoing a general anesthetic. uh with an LMA. The surgery is laparoscopic management of an ovarian torsion, and during the case the resident selects hydromorphone for pain control.

Perfect. So if I were working with that resident, I would ask the resident, let's say you're the resident, so David, why is it that you picked hydromorphone? Well, I picked hydromorphone because I've heard that it causes less nausea compared to morphine in the post op period. Excellent. So we've uh just seen the practice habit.

I use hydromorphone. And then underlying that was this claim. that people can't always readily express, but uh David in the did in this case that he uses it because maybe it produces less nausea. So to our listeners out there, do you believe this claim? Do you believe that hydromorphone causes less nausea than morphine? And what's the level of evidence that supports your belief?

Do studies exist? Either correlation or randomized control trials? Or is it more of an expert consensus or extrapolation from other fields, other domains, other contexts? Dan, why do you think it's important that we ask these questions? Well, I think that uh it's to provide the best care for our patients.

Practice habits are very useful things, and and we're pretty good at detecting patterns, but We also are at risk for creating false patterns or making things up, having fixed false beliefs. And and to be honest with you, even if our beliefs happen to be true, I think we owe it to our patients to know when those beliefs are evidence based or not. Awesome. I I definitely agree with that, and I think that really speaks to the reason why we're doing this podcast. So for this uh claim that Morphine causes more nausea than hydromorphone.

Uh do we have evidence to support this claim? You're asking, is there any evidence that might suggest that morphine causes more nausea? Correct You know, there there's some, but it's from a different context and it's not randomized. We can talk about it though So this was in the chronic pain outpatient cancer setting. And uh so it's it's a little bit extrapolated, and also this isn't a randomized trial.

Right And when we start looking at studies that were done in more of the acute care setting, we will find a different result. So with those caveats, this is a study published in 2008 in Support Care Cancer. First author is WERS, W-I-R-Z. They looked at a hundred patients with cancer pain who were treated with hydromorphone or morphine, and compared a bunch of outcomes, including side effects like nausea and vomiting. How was this study set up?

Was this a randomized control trial? It's a great question, and if you read through it, some of the wording can be a little bit confusing. They say they used a randomized scheme. but it not that it's not actually randomized. What they did was they identified a cohort of patients in this outpatient center, and then for folks who are already on morphine, a computer randomly selected fifty of them. And for people who are already on hydromorphone, a computer randomly randomly selected 50 of them.

So you see it's not actually random assignments, and we're not going to be able to infer causality. We can't We can't necessarily say that it is the hydromorphone or the morphine that is causing the difference, and that's because There are other things that could have led to these patients being on morphine or hydromorphone in the first place. Right, and that's definitely the issue with studies that are not true randomization. So knowing that now, what does this study suggest? So they looked at nausea in two ways, uh, with two different scales.

So one was this European Organization for Research and Treatment of Cancer Scale. It's a validated scale that's used for cancer pain and side effects. And it ranges from 1 to 4. They also used a numerical rating scale that ranged from 0 to 10. Higher numbers in both cases, meaning worse side effects, worse nausea.

Sure. So they looked at the differences between the two groups. They found a significant difference on the numerical rating scale, with the morphine group experiencing more nausea. And that was statistically significant. But with the uh European Organization for Research and Treatment of Cancer Pain Scale, uh no difference in nausea symptoms.

So to summarize, it's sounds like there was some evidence for higher nausea with morphine on one of these scales, but that said this is correlation data and might not apply to the care of the intraoperative patient in our case. Right, because this is cancer patient uh sorry, cancer pain in an outpatient setting. So do we have any better data that might be more applicable? Yeah, definitely. There are studies that look at acute pain and there are randomized studies.

So there's one here published in IRS in 2008. It was done by Hung, H-U-N-G , and others And I want to read just a little bit of what they wrote in the background because I just love how they illustrate one of the issues with claims. So here in the abstract they say quote Despite clinical lore among healthcare providers, the treatment with hydromorphone results in improved pain control and fewer adverse side effects So I love that they just call out the claim. They call it clinical lore. I just really liked that.

It's like it was made for this podcast. So what do they find? So right, so it's a prospective randomized control trial. Uh actually it's a comparative effectiveness, isn't it? Because it's not really a control arm, it's hydromorphone versus morphine.

And it was double blinded, which is really nice. Neither the patients nor the providers knew what the drug was that uh folks were getting, and the folks who were assessing for nausea symptoms were also blinded. And they gave subjects PCAs after general and gynecologic surgery. And the PCAs either had morphine or hydromorphone. So tell me about the patients they included in this trial.

There were 50 patients and they were randomized. They were between the ages of 18 to 60. And they were all ASA Class 1 or 2 patients. They were undergoing general abdominal or pelvic surgery, and they had all uh uh there was a plan for PCA use in all of them postoperably. They did exclude patients who were on pain medication at baseline, patients who were morbidly obese, patients who had sleep apnea, and patients who had liver or kidney disease.

So they assessed nausea one hour after surgery and eight hours after surgery. One hour after surgery, eleven of the patients in the morphine group had nausea and 13 of the patients in the hydromorphone group. No statistical difference. And eight hours after surgery, 17 patients in the morphine group and 16 patients in the hydromorphone group had nausea. So no difference again So what do you think is the big takeaway from this paper?

So at least for this population, which is young, healthy folks undergoing general anesthesia and then using PCAs. There seems to be no clinical difference between hydromorphone and uh morphine. At least in terms of nausea. Now it is important to say that it's only 15 patients, 5 to 0. So if there were a small difference, we wouldn't necessarily pick it up in this study Right, so in that case we would probably need a a higher power study.

Absolutely. Uh either a higher power study or a meta-analysis. And I think we have one of those, right? Right, so this was the study in uh the British Journal of Anesthesia, right? That's right.

And it was published in 2011, it's by Felden and colleagues, F-E-L-D-E-N. And they looked at a number of studies between 1970 and 2009. And they looked at studies comparing hydromorphone and morphine both in the chronic and acute pain setting. So both of the studies that we just talked about were included in this meta-analysis. Their final data set included eleven studies.

Not all of them were randomized. Uh some of them were sort of like that first study that we looked at that used a randomizing scheme. But all of the acute uh pain ones were randomized. And uh there were five of them that actually looked at the incidence of nausea. Uh of those five, four used IV pain medicines and one of them used uh epidural pain medicines.

Gotcha. So why don't we quickly talk about those? Sure. One of them was the Hong trial. We we just talked about that one.

And that was the one 50 patients, no difference, in uh nausea. Let's take a look at the next one here. So two of them are in the emergency room setting. Both of them, the first author is Chang from Albert Einstein. The first one here is published in 2009. in the American Journal of Geriatric Pharmacotherapy.

So in this study they looked at adults over the age of sixty five with acute pain presenting to the emergency room and it had to be severe, they defined severe as the emergency room physician was planning to treat with opioids. Okay. And they were randomized to a single dose of either uh hydromorphone or morphine, and it was administered in a double blind fashion. Okay. The they used uh equal analgesic ratio between hydromorphone and morphine of uh one to six point six seven, which is pretty standard.

They looked at a number of outcomes, including nausea, and uh they assessed nausea at 10 minutes and two hours after the baseline. There were a total of 194 patients that were randomized to the two groups, and ultimately 93 in the hydromorphone group and 90 in the morphine group that were analyzed. There was uh no difference in the proportion of patients who received uh anti-ometics between the two groups. Um six patients, so six point five percent uh of the ones who were in the hydromorphone group got anti-ometics versus ten or eleven point one percent of those in the morphine group. And again, not statistically significantly different.

The incidence of nausea and the incidence of vomiting were similarly not different between the two groups. So in this geriatric population in the emergency room setting We have no evidence to suggest that hydromorphone or morphine is better in terms of a side effect profile. At least for nausea. And uh this is single dose. Right.

So Chang did another study. This is also in the emergency room. This one is published in 2006 in the Annals of Emergency Medicine. Uh this is patients aged 21 to 65, again presenting with acute pain and severe enough that they wanted to treat it with IV narcotics. Same randomization scheme, both uh double blinded and uh randomized to a single dose of morphine or hydromorphone.

Secondary outcomes were looked at five minutes after baseline and two hours after baseline and they baseline and they did include uh nausea and vomiting. In this study, they analyzed a total of 97 patients in the morphine group and 94 patients in the hydromorphone group. So what do they find? So if you looked at the incidence of adverse effects uh after giving the drugs, nausea was present in 36% of the hydromorphone group and 32% of the morphine group No difference. Vomiting was 19% of the hydromorphone group and 15% of the morphine group.

Again, no statistical difference. No difference either in the use of RESTO anti -emetics. So with Chang's follow-up study, we have evidence yet again to suggest that there's no difference between hydromorphone and morphine in the acute setting But now this extends to patients in the twenty-one to sixty-five age range. That's exactly right. So here's another study comparing hydromorphone and uh morphine uh in acute care, uh acute pain setting.

This is uh for oral mucositis after bone marrow transplants, which is pretty severely painful. This is published by CODA, C-O-D-A, and uh colleagues in Payne in 1997. And they randomized patients actually to three groups. The third group, which we won't be talking about, is the Suefentanyl group. And they began collecting data on day one of PCA use, and they continued until the end of opioid therapy, which could be on the order of a few weeks.

Um on each study day patients were asked to rate their pain and their nausea and some other things, and they used a visual analog scale. uh which was a hundred millimeters long and they just had to mark where their symptoms were along the scale. And again not surprisingly there was no difference between the morphine and the hydromorphone group The uh mean values on this scale were 17.33 versus 23.05, uh slightly a note lower in the morphine group They also looked at uh the reasons for not being able to continue in the trial, that they had to discontinue their narcotic because of side effects. Uh owing to nausea, no one in the morphine group dropped out, but two people in the hydromorphone group did drop out.

Again, though, those differences are not significant between groups. Sure. So again, we now have a third study uh now in patients who are being treated for oral mucositis getting Hydromorphone versus morphine PCAs and again no difference in nausea. That's right. So in addition to that post op pa uh study that we talked about before, the one by Hong that showed no difference.

That was the fifty patient one. That's exactly right, yeah. In addition to that, the meta-analysis by Felden included one additional study It's a little bit different because they're looking at uh epidural morphine versus hydromorphone. This is your favorite one, right? No, no, no, that's coming soon.

Okay, I can't wait. Uh that's that one was that was actually published after the meta-analysis. Oh, so it wasn't included. Right, right. So um in this one that was included, there these are all kids. uh getting epidural hydromorphone versus morphine.

They're ASA one or twos uh age three to nineteen and they're all getting elective orthopedic surgeries and they're randomized to one of three groups in a again a double fly blinded fashion. It was either a morphine group, a uh fentanyl group, or a hydromorphone group. And all of the narcotics in the study were given postoperatively via the epidural And not surprisingly, again, there was no significant difference between the groups. Nausea and vomiting in the morphine group was 25 % And in the hydromorphone group was 10%, no statistical difference between the two groups. So how many patients did they include in each group?

Yeah, so it for analysis it was thirty patients in each of the morphine and the hydromorphone groups. Okay So to summarize this one, uh when we compare epidural morphine versus hydromorphone for postop pain control in children, uh there was again no incidence, uh increased incidence of uh nausea in one group compared to another. Right, right. And I think it would probably be harder to make that claim at large based on this study, just because it's fairly small. Right.

But it certainly is consistent with what we're seeing again and again in each of these other acute pain studies. Yeah, I'd agree with that. Okay, so what did the meta-analysis show? Because as we know, sometimes when you combine studies, there might be different results that you can tease out of the data. Right.

Yeah, more power. So when they combined all these studies, nothing. Uh they they look at something called uh uh an effect size of the uh the overall impact. An effect size, the common one people use, it's called Cohen 's D. Gotcha And uh it's uh what it is you you look at the effect size, the mean um for one group compared to the mean for the other group.

You subtract that difference And then to normalize it, you divide by the pooled standard deviation. So just to give you some real numbers, if one group's incidents were 25% and the other groups incidents were 30%. The difference there is 5%. The overall standard deviation is 20%. then we would say, well, five divided by twenty, the effect size is. 25. Now are there cutoffs for what's considered small, medium, and large for this Value?

Yeah. Uh standardly accepted cutoffs are 0.2 for small, 0.5 for medium, and 0.8 for large effect size. Okay. So let me tell you what they found for nausea between the two groups The effect size was 0.097 with no statistical significance. For vomiting, the effect size was. 0.175 again with no statistical significance.

So here we have in meta-analysis for acute pain with a fair number of patients, not thousands, but hundreds uh no difference between hydromorphone and morphine. And for our listeners out there, we'll definitely put some information about this uh Cohen 's D statistic on our show notes just to help clarify this concept It's time, right? Your favorite study. I just thought this was so cute. So this is this is a little trial.

It was published in 2012 in Pain Medicine They looked at 96 patients, and these were all kids, uh ages 8 to 21, and they were randomized to receive either morphine or hydromorphone as a PCA These were all postoperative patients, and this was another double-blinded randomized study. So the providers didn't know and the patients didn't know. The only one who knew was a pharmacist, uh the who could reveal it if there were intolerable side effects and you wanted to the and the physician wanted to change narcotic to the other type. Gotcha. So they could withdraw the patients.

There were actually a total of 11 patients that were withdrawn from the study because of side effects, and I'll talk about that in a little bit. The effect on nausea and vomiting though, no different than the studies we've looked at before. 74% reported nausea across both groups 43% had some vomiting across both groups. The needs for anti-emetics was the same, 67 versus 65% between the two groups. No statistical difference. So Dan, why do you like this trial so much?

Well I just really liked the power of uh habits or the powers of uh people's um impression of what things uh how they work and what they do be and it's because of the patients that were withdrawn. So two patients were withdrawn from the study at a time of day when uh pharmacist wasn't available to say what drug they were on And both of those patients, the physicians changed the PCA to hydromorphone. Okay. So they felt like it was morphine and they were concerned that it was causing the the pride effects. Yeah.

Both of those patients that they changed to hydromorphone had already been on hydromorphone. That's too good. Yeah. So uh they actually the authors write, um in our practice, quote, the opinion that morphine is more likely to cause side effects is shown by the fact that in the absence of information as to the drug's identity Uncomfortable patients were changed, quote, uh to hydromorphone PCA, even though they were already receiving it. And and I think David this is this is the power of habits and claims and they're they're often so immutable in the face of pretty good evidence.

This is I think why that your podcast is so important that we that we critically examine our claims. Right. So would you agree that on the basis of all these studies we've reviewed, the meta-analysis Everything so far in the acute pain setting seems to suggest that there's no difference in terms of nausea and vomiting between morphine and hydromorphone. You know that's sort of what the data would suggest, but I think instead of us concluding, I'd put that to our listeners. So what do you guys think?

That original claim that morphine causes more side effects in terms of nausea than hydromorphone. What's your current belief? And what's the level of evidence that that belief is based on? Awesome. Thanks, Dan.

Thank you. Hope you all enjoyed listening Thanks again for joining us this week. And if you did enjoy the show, please uh remember to leave us a five-star review on iTunes. This really helps get uh depth of anesthesia out there for for people looking for anesthesia podcast to listen to And uh again, we are reachable at depth of anesthesia. com as well as depth of anesthesia at gmail. com. So if anything uh came up during the show that you either had questions about or comments, you can reach out to us either on the website or by email and we'll do our best to get back to you.

If it's something interesting that we want to share with the rest of the listeners, we'll definitely mention it on future episodes. Thanks again for being a part of this. Uh you know this wouldn't be possible without our listeners. And uh stay hungry and keep asking questions.