For two decades, the leading theory of phantom limb pain has been that this condition is caused by “maladaptive plasticity.” When the brain’s primary sensorimotor cortex no longer receives input from a missing body part, such as an amputated hand, signals from another body part, such as the lips, begin to take over that area. Although other phenomena might also be involved, that remapping has largely been thought to cause phantom pain. Tamar Makin from Oxford University, UK, and colleagues challenge that orthodoxy in a paper published March 5 in Nature Communications. They find that amputees who experience ample phantom pain have stronger rather than weaker cortical representations of the missing body part, and no evidence of remapping.
“This is truly significant work, which challenges previous views—close to axiomatic—that phantom limb pain is a marker of cortical reorganization,” said Peter Brugger, a phantom limb pain expert at University Hospital, Zurich, Switzerland, who was not involved in the research. It “investigated the cortical correlates of phantom limb pain in the area of the lost limb representation, and not, as all previous work did, changes in the representation of neighboring parts that are still present.” Given the findings, it’s not surprising, he said, that the success of treatments for phantom pain that are based on maladaptive plasticity has been limited.
Phantom pain is not trivial. An estimated 80 percent of amputees experience phantom pain, which in some cases can be “hugely debilitating,” and can continue for years, said Makin. Treatments, including drugs, neuromodulation, surgery, and other approaches are largely ineffective (for a review, see Knotkova et al., 2012).
“The support for the hypothesis of maladaptive plasticity has come primarily from non-invasive functional neuroimaging studies over the past two decades that have noted a shift in the cortical representation of the lip,” said Scott Frey, who studies the effects of amputation and limb paralysis on brain organization at the University of Missouri, Columbia, US, and who was not involved in the current study. “This finding has been interpreted as evidence that the face area is expanding into the former hand territory.” That remodeling is thought to be the source of phantom pain (for a review, see Flor et al., 2006).
But the theory never appeared quite right to Makin. That, she said, led her to conduct the current study. “It didn’t make sense that shifted cortical representation of the face towards the hand area, as has been documented, could trigger pain,” she said. Maladaptive plasticity predicts that “following loss of sensory input, the deprived hand area of the primary sensorimotor cortex becomes responsive to inputs from neighboring cortical areas, for example, the face area, thereby triggering phantom pain. The maladaptive plasticity model would therefore predict that representations of the missing hand should be reduced in the sensorimotor cortex of people who suffer from more pain, owing to greater remapping.”
But Makin and her co-investigators found the opposite: They found that greater pain in amputees correlated with greater preservation of representation of the hand within its cortical area, as seen in functional MRI (fMRI) scans. Additionally, fMRI scans taken while amputees and two-handed control subjects smacked their lips showed no difference, on average, in cortical activation in the hand area between the two groups, suggesting that cortical representation of the lips was not invading amputees’ missing-hand cortices. Nor was activation during lip movement associated with pain intensity among the amputees.
Experiments examining cortical activity during phantom hand movement further challenged the maladaptive plasticity theory. Interestingly, amputees can “move” their phantom limbs to varying degrees, meaning “they are capable of voluntarily inducing the sensation of moving the missing hand,” said Frey. That is different from imagining moving a limb—phantom or real—in terms of the sensation evoked, and the activity generated within the cortex.
In these experiments, the researchers compared cortical activity in amputees who moved their phantom hands with cortical activity evoked when two-handed control subjects moved their dominant or non-dominant hands. They found that the fMRI scans of phantom movement were similar to those of two-handed controls moving their non-dominant hands. In fact, when amputees moved their phantom hands, the scans showed that motor execution signals occurred within the cortex as if subjects were moving a real hand. This suggests that representation of the hand was preserved within the cortex in the amputees.
Moreover, contrary to the theory of maladaptive plasticity, the investigators found that greater chronic phantom pain correlated with the degree of activation within the hand cortex, and maintenance of its structural integrity. That finding raises the question of whether chronic phantom pain helps maintain hand cortical structure.
Furthermore, compared to amputees with less phantom pain, amputees with more phantom pain also exhibited less functional connectivity between the cortex for the amputated hand and the cortex for the intact hand, which the investigators suggested results from a lack of co-activation between the two hand cortices.
For his part, Frey criticizes the maladaptive plasticity theory and the studies that support it, noting among other things that one can’t tell whether cortical remodeling causes the pain, or whether pain drives the remodeling. But he said that same problem applies to Makin’s paper, too: Is cortical representation of the missing hand maintained by pain signals, as Makin suggests, or is it “somehow maladaptive, and driving these experiences?” Frey asked. Frey also suggested that “apparent differences between these findings and those of earlier studies might be, in part, attributable to differences in the fMRI tasks and/or the ways that pain perceptions have been quantified.” But he still praised the quality of the study, saying, “It raises important questions that will be important to replicate in additional samples and studies.”
Even Herta Flor of the University of Heidelberg, Germany, and a leading proponent of maladaptive plasticity, lauded the quality of Makin’s research.
Nonetheless, the problem of where in the body phantom pain arises remains unsolved, a problem that neural plasticity expert Michael Merzenich, University of California, San Francisco, US, finds “terribly frustrating…. Understanding the neurological basis of emergent pain qualia is key to understanding why pain emerges following massive deafferentiation [loss of incoming signals from the now-missing hand or limb] and/or whatever plastic adjustments do/do not follow it.”
Meanwhile, based on her current work, Makin is running a double-blind, placebo-controlled, Phase 1 clinical trial of transcranial direct current stimulation for phantom limb pain in hopes that the procedure will provide much-needed relief for patients.
David C. Holzman writes on science, medicine, energy, and the environment from Lexington, Massachusetts, US.