I am a
Home I AM A Search Login

Chronic Pain or Current Pain: What Do Changes in Default Mode Network Functional Connectivity Actually Reflect?

Commonly seen changes in brain connectivity among fibromyalgia patients may more closely reflect these individuals’ current clinical pain state rather than their pain condition

by Megan McPhee Christensen


12 November 2020


PRF News

DMNFeatured

Commonly seen changes in brain connectivity among fibromyalgia patients may more closely reflect these individuals’ current clinical pain state rather than their pain condition

The brains of patients with different chronic pain conditions, including fibromyalgia, often show differences in resting-state functional connectivity, compared to pain-free controls. Researchers have often interpreted this to mean that having a chronic pain condition reorganizes the brain. But more recent brain imaging studies have suggested that altered connectivity could instead result from pain that patients are currently experiencing during a brain scan (pain state) and may not necessarily reflect the severity of their chronic pain condition (pain trait).

 

New work from co-first authors Marta Čeko, University of Colorado Boulder, US, and Eleni Frangos, National Institutes of Health, Bethesda, US, and colleagues now confirms that hypothesis in fibromyalgia patients by examining connectivity between the insula and default mode network, but also adds new detail and insight.

 

“This paper highlights and again corroborates the decade worth of research showing that greater resting functional brain connectivity between the default mode network and insular cortex is associated with greater clinical pain at the time of the MRI scan,” said Vitaly Napadow, Massachusetts General Hospital, Harvard Medical School, Boston, US, who was not involved in the study. The default mode network (DMN) refers to a network of brain regions that are active while a person is not engaged with the outside world, while the insula is a well-known pain-related brain region.

 

However, the researchers take it a step further. By combining two existing case-control datasets – one where patients with fibromyalgia had pain during fMRI scanning and one where they did not – the investigators had a unique opportunity to explore the isolated effect of pain state versus pain trait on the brain. The team demonstrates that while current pain was associated with DMN-insula connectivity changes, just having fibromyalgia without current pain was not. Further, when patients had current pain, connectivity changes were not related to clinical features (pain duration, anxiety, and depression), but when patients were pain free, DMN-insular connectivity was in fact associated with pain duration. In short, by removing the confounding influence of current pain, the investigators could reveal features of fibromyalgia itself.

 

This study was published August 1, 2020, in NeuroImage.

 

What do default mode network connectivity changes mean?

About a decade ago, Napadow and colleagues were the first to identify an association between DMN-insular connectivity and clinical pain intensity reported immediately prior to MRI scanning in fibromyalgia patients (Napadow, et al., 2010). They have since replicated that finding both in fibromyalgia (Napadow et al., 2012) and in chronic low back pain (Kim et al., 2019 and see PRF related journal club discussion here). Other research groups have also demonstrated group-level alterations in DMN connectivity among patients with fibromyalgia and other chronic pain conditions, compared to pain-free controls, though associations with pain at the time of scanning are less consistently reported. Nevertheless, DMN connectivity clearly is affected by pain, though that should be unsurprising.

 

“When we talk about the DMN, it is what is supposed to be active when you’re lying there doing nothing, but if you are in pain, you’re not doing nothing,” Čeko explained.

 

The question that had remained, though, was whether changes in DMN connectivity related to clinical pain intensity are actually reflective of the severity of an individual’s chronic pain condition. If they were, these changes could be used as biomarkers of chronic pain for diagnostic or interventional purposes.

 

However, the authors suspected this might not be the case – that current pain state could instead represent a random source of variation, potentially poorly related to disease severity, and thus could lead to a misinterpretation of functional connectivity findings. The authors further speculated that, if current pain intensity is not a disease-relevant measure, and if observed changes in DMN-insular connectivity are just a reflection of current pain state, then this functional connectivity finding may simply be a laborious way of obtaining a pain rating – something that is potentially useful in specific populations with limited ability to communicate pain, but not when communication is operational.

 

“There is the diagnosis of chronic pain – the trait of chronic pain, the condition. But this does not have to be the same as the clinically experienced pain that is there right now. We wanted to differentiate the resting-state connectivity of this current pain and understand whether it actually has any biological, long-term clinical significance to the patient,” said Čeko.

 

Pain state versus pain trait

To address those issues in their new study, Čeko and colleagues included data from two prior case-control samples of fibromyalgia patients who reported having had the condition for more than a year with at least moderate levels of average daily pain but, critically, either did report pain during scanning (16 fibromyalgia patients and 16 controls) or did not (27 patients and 27 controls).

 

In patients who had pain during scanning, increased functional connectivity was observed between the DMN and anterior regions of both the right and left anterior insula when compared to matched controls. (Increased connectivity with the dorsolateral prefrontal cortex was also observed, but this finding did not hold up after correction for differences in gray matter volume). Further, the magnitude of increased DMN connectivity to the left mid-insular cortex was positively related to self-reported pain ratings during scanning, consistent with Napadow and colleagues’ prior work, but was unrelated to disease factors such as anxiety, depression, and pain duration.

 

In contrast, using the same methodological approach, the researchers saw no differences in DMN connectivity in fibromyalgia patients without pain during scanning compared to matched controls. However, when analysis was performed without statistical corrections, a subtle increase in DMN-insular connectivity was present in these fibromyalgia patients, and the magnitude of this increase was positively associated with pain duration.

 

Taken together, the results showed that current pain state had a clear and expected relation to enhanced DMN-insular connectivity, but changes related to the fibromyalgia pain trait (pain duration) were only observed in the absence of current pain.

 

This study therefore highlights an important distinction between pain state and pain trait. This topic has recently come into the spotlight and merits greater consideration, as per a review paper from Karen Davis, University of Toronto, Canada, and Joshua Cheng, Stony Brook University School of Medicine, US, which was published last year in PAIN Reports (Davis and Cheng, 2019).

 

Looking ahead to future imaging studies

Interestingly, the current study authors told PRF that when they first submitted their article for publication, a reviewer commented that their sample could not really be said to include fibromyalgia patients if those patients did not have pain when in the fMRI scanner; the reviewer was under the presumption that chronic pain patients must always have pain.

 

Such a misunderstanding may arise from the often-cited definition of chronic pain as pain lasting longer than three months. But pain is not necessarily a continuous experience. There are many examples, especially in chronic musculoskeletal pain conditions, where pain is mostly movement-evoked and rarely constant. In these cases, where the ultimate source of pain is also often less clear, it may be especially important during pain studies to differentiate between pain trait and pain state, and to control for the latter to gain better insight into features of the underlying condition.

 

So how does the field move forward from here with regard to future imaging studies?

 

“I think that we need a more comprehensive assessment of what goes into the condition of chronic pain. We usually ask once, when people come in [to the lab], about pain in the last week, and then we ask them about their pain right now in the scanner, but that’s just not enough,” said Čeko.

 

Some researchers use pain diaries over longer durations to capture variation in pain intensity over time, along with various questionnaires to capture pain-related interference with daily activities. However, Čeko would like to see more standardized ways of accurately capturing the variation and impact of an individual’s pain over time.

 

Researchers still have work to do in teasing out confounding factors that may be present in functional connectivity brain imaging studies to ensure that observed differences really are due to the chronic pain condition under investigation. Čeko is optimistic that a better understanding of this issue will ultimately make for better treatment decisions.

 

“If we knew that changes in functional connectivity were actually a biological trait of the condition, then we could target some of those connections with treatment. For example, if we observed increased connectivity between the DMN and the insula, and we know that some psychological techniques like mindfulness-based meditation can reduce that hyperconnectivity, then we could confidently apply this treatment to specifically target those connections.”

 

“Hopefully this objective metric [functional connectivity] can eventually be used in clinical care – to choose appropriate therapy and improve clinical outcomes,” Napadow agreed.

 

In the meantime, Čeko cautioned her fellow researchers to continue to pay attention to the state-versus-trait conundrum.

 

“Remember, if you don’t know that the observed functional connectivity differences are just due to state-related transient features, then you’ll waste a lot of time trying to fix something that’s not even broken.”

 

Megan McPhee Christensen is a PhD fellow at the Center for Neuroplasticity and Pain (CNAP), Aalborg University, Denmark.

 

Image credit: Čekoet al. Default mode network changes in fibromyalgia patients are largely dependent on current clinical pain. NeuroImage. 2020 Aug 1;216:116877.

Share This