EFA’s Sixth Annual Medical Conference: Ending Endometriosis Starts at the Beginning
Pain mechanisms in endometriosis: understanding the neurobiology of chronic pain to enhance patient care
Sawsan (Suzie) As-Sanie, MD
(This intro was moved from the end of Dr. Rosen’s intro to the EFA Conference.)
It is now my pleasure to introduce our first speaker, Dr. Suzie As-Sanie. She is an Assistant Professor and Director of the Minimally Invasive Gynecology Fellowship in the Department of OBGYN and the Director of the University of Michigan’s Endometriosis Center. The title of her talk “The Pain Mechanism in Endometriosis: Understanding the Neurobiology of Chronic Pain to Enhance Patient Care”, Dr. As-Sanie.
Thank you so much to the course organizers, Dr. Seckin and Lone for inviting me to this tremendous organization and conference. It is my first time here and it is really wonderful to see so many people interested and dedicated to improving the care and quality of life for women with endometriosis.
I want to just give one qualifier. I have no conflicts of interest but I do want to qualify that I am not a neurobiologist or a neuroimaging specialist but this is the area of my research and I would not have been able to do this without my fantastic collaborators both at Mass General Hospital as well as the University of Michigan.
As you know, and I do not need to explain to this audience, endometriosis is one of the most enigmatic conditions in the field of gynecology despite its high prevalence and significant negative impact on the quality of life of women with endometriosis. There is still quite a bit that we do not know about it.
Logic and clinical intuition would tell you that a woman that has a normal pelvis without any anatomic lesions should be pain free and the woman here with stage four endometriosis should be the one that is experiencing the most pain. But any clinician that regularly treats and sees women with endometriosis knows that that is not necessarily true. There are a significant amount of women that have very advanced endometriosis with very little pain at all and many women that have significant pain without any pathologic findings.
While there is a lot we know about endometriosis and I am sure many of the lecturers have spoken to you yesterday as well as today about the underlying pathophysiology of endometriosis we know that it is likely that retrograde menstruation plays an important role. We know that implant proliferation, neovascularization are very significant factors in the underlying pathophysiology of endometriosis but to be quite honest we really do not understand why it is that these issues may or may not lead to pelvic pain in certain women.
The most simple and logical way would be to assume that endometriosis is a nociceptive pain condition and nociceptive pain is defined as a condition in which pain occurs as a result of injury or inflammation to the outside body. That injury or inflammation is perceived as travels through the spinal cord and up to the brain to give a perception of pain. It is quite logical that endometriosis should be able to do this because it has been well established that endometriosis involves direct infiltration of nerve fibers. There is increased inflammatory cytokines that stimulate those nerves and there is nerve cell proliferation. This model of nociceptive pain is very well understood. It is the underlying physiology of many acute pain issues such as acute surgical pain but we know that increasing evidence suggests that this model is really insufficient to explain the experience of chronic pain in women with endometriosis.
Why do I say this? Well first, as we talked about earlier, there is little if any correlation between the extent of disease and the severity of pain. Second, medical therapies that are directed specifically at the endometriosis lesions are non-specific and actually treat many other causes of pelvic pain. For example, the only FDA approved treatment for endometriosis, GnRH agonists are effective therapies for actually many other conditions including cyclic irritable bowel syndrome and pelvic pain without endometriosis.
Third, and most importantly, medical and surgical therapies are inadequate for many patients. When you look across multiple studies of both medical as well as surgical therapy there is up to a 30 percent non-response rate meaning women that have the same treatment as others report no improvement in their pain.
And finally, and very frustrating to patients even if the pain improves the frequency of recurrent pain is quite high following discontinuation of medical therapies as well as if you follow patients long enough after surgical therapies. The pain often recurs in the absence of recurrent pathology. It is quite well demonstrated that when you go back and do a second laparoscopy for instance in women that have recurrent pain we do not always see recurrent endometriosis lesions.
The objectives of this lecture are to briefly review the physiology of chronic pain, with a particular emphasis on central pain disorders and to highlight the differences between chronic and acute pain. Second, I would like to present briefly some evidence that endometriosis and chronic pelvic pain may be central pain disorders. Finally, I would like to suggest some clinical approaches to integrate the treatment of central sensitization to enhance patient care.
As we talked about previously endometriosis is likely to have an ascending stimulus to the brain and we know when we study many other acute and chronic pain conditions there are very important areas of the brain that interpret the pain signal. The pain signal initiates, goes through the spinal cord up into the thalamus and there are other critical regions including the insula, the periaqueductal gray, the somatosensory cortex, the ACC and the prefrontal cortex. What we are beginning to understand is that this is not a one-way street. The brain has a remarkable ability to both facilitate or up-regulate that signal as well as to inhibit. There are many neurotransmitters and chemicals that are increasingly understood to facilitate this process. What we are beginning to understand is that the balance of where people feel pain like many other physiological processes has a volume control. There is a normal distribution across the population as to where each person lies in that pain sensitivity. These are likely influenced by many factors including genetics, including early life trauma and what we are beginning to understand is that in those patients that have this chronic, many chronic pain conditions they tend to lie at the end of the spectrum. They demonstrate that this concept of diffuse hyperalgesia or allodynia to stimulus that would not normally be perceived as pain.
What pain physiologists suggest is that in these patients that have chronic pain there is some sort of imbalance in the central nervous system such that there is either, or a combination of, increased facilitation or decreased inhibition that is associated with this state of chronic pain. For any given stimulus – on this graph here there is stimulus intensity so normal and pain free patients lie on this blue curve. For any given stimulus patients that have chronic pain experience a much higher pain intensity to the same stimulus. That idea of hyperalgesia is a central common finding in these patients that are considered to have central pain conditions. It is the concept that in these patients the volume control has been turned way up and might likely lead to the underlying physiology of chronic pain.
These patients that have what we typically term central pain conditions are typically characterized by having multi-focal pain so pain in more than one area of the body. They endorse neuropathic verbal descriptors of pain, such as burning, they have a higher current and lifetime history of pain and also multiple somatic symptoms. I think this was an important part of the discussion earlier in the conference of experiencing symptoms, not just pain but also fatigue, memory difficulty as well as sleep disturbance. Interestingly they are not just sensitive to nociceptive stimuli or stimuli that are perceived as pain but it has been demonstrated in some other pain conditions that they actually have increased sensitivity to multiple sensory stimuli including sound and light. There is a high rate of co-morbidities with other related pain conditions and in these patients opioids do not effectively or consistently reduce pain symptoms. These conditions are one and a half to two times more common in females. There is a very strong familial and genetic underpinning and they are often triggered or exacerbated by stressors.
When trying to understand the underlying physiology of pain we have come a long way using both experimental pain testing and most recently functional MRI to elucidate the underlying physiology of pain. What has been quite consistent there are very consistent areas that have previously been referred to as the pain network or the pain matrix that are important in pain sensation. They include the thalamus like we talked about, the periaqueductal gray and I just want to point that out because these are areas that have been found to be relevant in patients with endometriosis as well.
When we look broadly at the chronic pain literature there are some fairly consistent findings across many other chronic pain conditions in these patients that have these chronic pain conditions. The first is that they have increased sensitivity to pain and sensory stimuli so that left shifted bell curve that I described previously if you apply a stimulus these patients have a heightened response to that. When you look at brain function as well as brain volume there are also some fairly consistent findings. Patients have increased activation in areas of the pain network as well as increased connectivity between these regions within the pain network. The signaling between these regions is altered compared to healthy controls. When you look at actual anatomy the most consistent finding is decreased gray matter volume in the areas of this pain network.
So the questions that remain particularly unanswered in women with endometriosis associated chronic pelvic pain is do these women, similar to other chronic pain conditions, demonstrate altered sensitivity to experimental pain as a marker of central amplification? And what are the underlying central neural mechanisms? So what are the brain changes that we see in these women? And finally, can these measures of central pain amplification or can this knowledge be used to guide clinical care?
It has been actually quite some time, and this is a publication in 2005 by a group in Europe, Dr. Laursen B. Bajaj and Arendt-Nielsen, they are very well known pain researchers. I point this out because this is actually one of the earliest studies that has demonstrated that women that have endometriosis do have different pain sensitivity relative to healthy controls. What they showed here are these are healthy controls and these are people with various chronic pain conditions including women with endometriosis. This is one of the earliest studies that looked at what we would consider a visceral pain syndrome, a pain syndrome that affects the pelvic organs unlike fibromyalgia, low back pain, rheumatory arthritis, which are all somatic pain symptoms, pain syndromes in the muscles and joints that similar to those other pain syndromes these patients tend to be more pain sensitive. But there were several limitations to this study. It was a small sample size, only ten patients per group and they did not specifically look at different sub-categories of women with endometriosis.
Our group in 2013 did a follow up study to this where we did a different type of sensory testing that is well validated in other literature. The main findings of this were published here and the main point is is that if you look at healthy controls are here this is the average pain sensitivity to an external stimuli. What we found is that all the subgroups that we looked at those with endometriosis and actually only painful periods, those with endometriosis and chronic pelvic pain, which we defined as having pain for at least 14 days of the month and those women who have chronic pelvic pain and no endometriosis showed altered and increased pain sensitivity defined as a less amount of pressure needed to apply to elicit the same amount of pain compared to healthy controls. But the endometriosis patients that did not have pelvic pains, so those women that were relatively pain free actually looked just like controls. That was the first study that looked at one, I think this important sub-categorization.
We also found in this particular study that endometriosis stage did not matter. The more endometriosis you had did not correlate with how pain sensitive the patient was. We also looked at whether or not the presence of other co-morbid pain conditions so we know and will talk about later that women with endometriosis do tend to have other pain conditions like irritable bowel syndrome, interstitial cystitis, etc. and that actually did not matter at all. In even those women which we had a fairly large sub-population that had no other demonstrable chronic pain condition were just as pain sensitive as the patients that had multiple pain conditions.
The next thing we did was a brain imaging study looking at regional gray matter volume in women that have pelvic pain both with and without endometriosis. As I alluded to before that what we have seen in other pain conditions is generally a decrease in gray matter volume in these key areas of the pain network. What we have found, so the top panel is 16 women with both endometriosis and chronic pelvic pain compared to healthy controls and the bottom panel is only six patients that had chronic pain but no endometriosis. The findings were fairly consistent that in these patients there is decreased gray matter volume similar to the pattern that we see in other chronic pain conditions. We saw that even in the six patients that had pain without endometriosis. And the amount of gray volume did correlate with patient’s perception of their daily pain as how they rated how unpleasant their pain was. The less volume a patient had, so the smaller the thalamus was the more those patients described their pain as being unpleasant. This regional gray matter volume was not associated with mood symptoms such as depression, anxiety and was also not associated with the use of hormonal contraceptives.
I think one of the most interesting things in this study, which actually was quite a surprise to use because my…hypothesis what I actually presumed was going to happen was that those women that have endometriosis but very little pain I assumed would look actually just like the controls. They did look like the controls in many ways in that they did not have a decrease in the gray matter volume in the thalamus and other areas of the pain matrix but they actually had an increase in the gray matter volume in the periaqueductal gray. Again, that is one of the key regions of pain sensory processing but it is the key region that we think is important in the inhibition of pain, meaning the more activity you have in the periaqueductal gray the less pain you experience because it is sort of a filter or a down regulation in pain signal. This was published in 2012. In these patients that gray matter volume was positively associated with experimental pain testing. So, the more gray matter volume you had the more pressure we needed to apply to elicit the same amount of pain. That was a clinical correlate that supported that underlying hypothesis.
Finally this is data that is in a manuscript that is currently under review. We next looked at whether these women with endometriosis and pelvic pain have any differences in brain chemistry as well as function. We know in other chronic pain conditions that the insula is a key region of pain and sensory processing and it is involved in both somatic as well as visceral sensory processing. It has been demonstrated in patients with fibromyalgia that they have elevated levels of excitatory neurotransmitters primarily glutamate in the insula as well as decreased levels of the primary inhibitory neurotransmitter GABA in the insula. What we wanted to do was see if we could demonstrate similar findings in women that have endometriosis associated pelvic pain and we did see similar but not exactly the same findings. What we found was that in the women who have both endometriosis and chronic pelvic pain, as well as the women that had no endometriosis but positive pelvic pain, compared to healthy controls they did have elevated levels of the primary excitatory neurotransmitter a combined signal of glutamate and glutamine in the anterior insula. There was no difference in neurotransmitter levels in the endometriosis patients without pelvic pain.
We then looked at how different regions of the brain spoke with the insula, this anterior insula region with methodology that is called connectivity. Again, I am not a neuroimaging expert but I will try to explain it to you in the way that I understand it, in very elementary terms. What has been consistently shown in other pain conditions is that changes in the connectivity between different regions have been shown in other pain conditions and we found the same thing in our chronic pelvic pain patients. We demonstrated that there was increased connectivity of the anterior insula to the medial prefrontal cortex which is an important region of the salience network, which is involved in self-referential thinking and your focus on internal systems in the body. What we found was that this altered connectivity. So this increased connectivity was correlated both with the excitatory neurotransmitter levels, the GLX and what I do not show here is that it is also correlated with pain intensity, the patient’s report of pain.
To summarize these findings: women with endometriosis-associated pelvic pain demonstrate increased pain sensitivity as well as altered brain chemistry and function. Women with endometriosis but without chronic pelvic pain do not show these differences in pain sensitivity or function but do have increased grey matter volume of the periaqueductal gray, that brain region associated with descending inhibition.
The question, and I should qualify because it makes it seem like I am just a pain physiologist but I am actually a surgeon. I see patients. I see patients with chronic pelvic pain and endometriosis and the question that I throw out is why do we need to care? Does this really affect our clinical practice and I would say that even if you are not a person or physician or clinician that is going to directly treat the chronic pain in your patients you really need to care because you might do more harm than good. This knowledge really helps us guide clinical treatment decisions because even if you are not going to treat the pain you need to make a decision as to whether or not you are going to do something like surgery in a patient that might not respond to that because surgery is obviously not without risk.
This is what I would propose, and this is where the evidence really begins to dwindle away and we are really just extrapolating from other pain conditions. The first clinical implication is that chronic pain conditions cluster. Other chronic pain syndromes are far more common in women that have chronic pelvic pain and endometriosis. And we really need to recognize that. In the far left column is the incidence of these various chronic pain conditions in the general population and what you can see is that in women that have both pelvic pain and/or endometriosis the incidence of other comorbid pain conditions is elevated, elevated two to four times.
This is also true and I know there is a big focus, an excellent focus, in this conference as well as prior years is that this is true even in the young women that we see. We did a retrospective study on adolescence that presented to the University of Michigan with endometriosis. These were all women who had their endometriosis diagnosed prior to the age of 21 and the mean age of diagnosis was about 18 years and there were 131 women. What we demonstrated in a systematic evaluation of the patient that even adolescents that have endometriosis and pelvic pain have far more chronic pain conditions than the general population, shockingly more so. Fifty-eight percent of these adolescents had more than one pain condition. Twenty-seven percent had more than two pain conditions and the prevalence of each individual pain condition, like IBS 25 percent, chronic headaches almost 20 percent. Even fibromyalgia, which is thought to occur in one to two percent of the general population, not well described in adolescents, was diagnosed in these young girls as high as seven percent. Comorbid mood disorders are also quite common, 48 percent of these adolescents had more than one mood condition, either depression or anxiety. Of course this is a tertiary care referral clinic so these are all young women that were referred to a specialty clinic for the care of adolescents in pelvic pain so it is not likely generalizable to the general population but it does speak to the burden that even these young women suffer.
The next point that I would like to make is that patients with central pain changes probably respond differently to therapy. They are probably less likely to respond to peripherally directed therapy. What I mean by that is that they are less likely to respond to therapies that are directed specifically at the endometriosis lesions, such as hormone suppression and surgery. We have actually recently published a paper, not specifically on endometriosis, that these patients experience both more acute post-surgical pain as well as chronic pain after surgery. These patients are more difficult to take care of, more difficult to manage their pain after surgery and that is important for us to recognize as surgeons.
There are many ways, and I do not mean to imply that there is only one way to get to the system. This is a diagram from an excellent review by Katy Vincent in human reproduction last year that summarized the various ways that patients can reach a chronic pain condition. There is peripheral stimuli, this is specifically in endometriosis, peripheral stimuli that is likely to be very relevant. There are alterations in hypothalamic pituitary adrenal access. We know there is increased inflammation. There are alterations in the immune system, prior surgical procedures likely affect this and of course, psychological and cognitive state features affect this and then this idea of central sensitization.
While there is really no evidence to say this I looked at one of the lectures that was given here a couple of years ago and I have not really found anything more recent logic would say that we really need to treat early to prevent transition from acute to chronic pain. And whether whatever part in this pathway that we can break in an earlier time we would assume should at least in some part be helpful to avoid having so many women in a chronic pain state that becomes so much more difficult to treat.
What I would propose to clinicians is first pelvic pain is not just endometriosis. Endometriosis is what we are all here to talk about but as we discussed there are a lot of conditions in the pelvis that can cause chronic pain both in the urological system, in the gastrointestinal system as well as the musculoskeletal system. Whether or not you decide to treat these other conditions is fine but it is important that when patients come with endometriosis many of them think endometriosis is the sole cause of their pain. It is really important to do a comprehensive evaluation because these other conditions are highly comorbid in these patients. The central nervous system is a big and important player but there are also other conditions that we need to treat outside of the pelvis because ultimately our goal is to improve a patient’s quality of life. That is their end goal and our job as well. If they have other conditions like fibromyalgia or chronic fatigue or migraines, these all need to be addressed, not necessarily by you but by a provider in a multi-disciplinary fashion.
Comorbid psychological disease, and this is a lecture or a series of lectures, all on its own we know they are highly comorbid. The vast majority of the literature suggests that these happen after the chronic pain, it is not the depression and anxiety that caused the pain but these are a consequence of the chronic pain. It is quite clear that without treating this comorbid psychological disease it is very hard to treat the pain.
Finally, many patients with chronic pain conditions are very maladaptive. Pain cognitive features like catastrophizing and these are also cognitive behaviors that we know make pain very difficult to treat. These all need to be addressed.
What I would say is begin with gold standard treatment therapies for contributing factors. In no way I am saying that you should not use hormonal treatments, that you should not use surgery, these are the first line treatments that we know work for many patients. But also look for other pain conditions. Many of these patients, for example, have significant musculoskeletal pain whether it is a cause or a consequence or a parallel factor, we do not really know. But referring to physical therapy is a huge component of our practice for these patients and can often single handedly be one of the most important factors that help patients.
When these standard treatments fail then re-consider other diagnoses, re-evaluate comorbid psychosocial variables and I would argue and maybe this would be volatile in this room, do not do that tenth laparoscopy for the patient that has stage one endometriosis. It did not work the fifth time, it did not work the sixth time, it is not going to work the eighth, ninth and tenth time. I think we have all seen those patients.
It is important to recognize that abnormalities in pain processing are common and might be an underlying problem. When you suspect that, whether it is because they have failed therapies or whether it is obvious from their clinical care consider adding a centrally acting therapy when standard therapies work. There are a lot of pharmacologic therapies that have demonstrated efficacy in central pain conditions. One little pearl I would tell you when you look at the range of antidepressants that work for chronic pain it is the ones that have a balance in the reuptake of both the serotonin and norepinephrine inhibitors that are the most helpful. The pure SSRIs do not actually tend to work that well in central pain conditions and if you use it for mood and anxiety that is totally fine. That is what they are used for. But if you want to have an analgesic benefit independent of the benefit of mood the mixed reuptake inhibitors work.
There are many antiepileptics with analgesic properties including Neurontin and Pregabalin. All of these medicines unfortunately have significant side effects, they are not necessarily our first line treatment but these can be helpful in many of these patients. The key issue with any of them is that you can probably minimize the side effects by starting out at really baby doses, doses that we know probably do not have an analgesic effect and then gradually titrating them up.
Again, this is a topic of a whole lecture in and of itself but there is significant evidence to support, and I saw a poster back there on acupuncture, that non-pharmacological therapies can also be very effective; exercise, physical therapy, cognitive behavioral therapy, acupuncture and sleep therapy. These are all important in patients that suffer from chronic pain.
The only thing I would like to end on is one size just does not fit all. All of these women are individual people. They probably all got to their state of chronic pain through different pathways and it is our job as clinicians that want to improve their quality of life to consider that. These are actually real patients that I have seen within the last year so if you have tunnel vision and you are really just looking at the pelvis you very well might have that one patient where that is really the only problem. The Lupron, the OCPs, the surgery are likely to be helpful and the patient is going to be very satisfied and will do well. But we all see these patients as well, this is a real patient that filled this out, if you are only focusing at the pelvis it is not likely that the simple things that we do for those endometriosis lesions are going to improve her quality of life. And that is why it is really important for us to take a step back and look at these patients as a whole.
I will end with one of my favorite Harry Potter quotes. For those of you who know me I am a huge Harry Potter fan. I used to call a lot of my lectures sort of with the introduction of “is it all in our head” and I never meant that to be derogatory because really the point is that of course it is happening inside our head because that is where pain happens. But why on earth that should mean it is not real? Thank you.
Tamer Seckin, MD: Very, very beautiful presentation. One of the best I have heard. Just curious, the dependency on pain medication, how much of that is a problem in what you see in that population that you deal with. You did not touch that.
Sawsan As-Sanie, MD: I know. That is a big can of worms, opioids, yes.
Tamer Seckin, MD: How much of a difficulty does that cause for you?
Sawsan As-Sanie, MD: I think every clinician needs to do what they are comfortable with. Opioids have obviously a very high addiction potential. There is very clear literature that opioids are probably not helpful in central pain problems but they are helpful in acute pain problems and peripheral pain problems. Hopefully what you took away is that these women lie in a spectrum, some of them are probably purely peripheral and some of them are probably purely central and the majority of them are probably somewhere in between. There is probably a role for them but as any clinician that prescribes opioids should be aware of is that they can only be used as a mechanism to improve their quality of life, that they are more active, that they are doing the things that they need to do, that there are limits and there are checks.