Recent global burden of disease surveys tell us that pain and depression are two of the most common and pervasive issues in modern society. The fact that these two are also highly comorbid hardly generates surprise. Living with chronic pain often means restricted mobility, reduced capacity to work and general impairment of quality of life, in addition to the general unpleasantness of the pain, so mood will inevitably suffer.
Less intuitive, perhaps, is the suggestion that depression might cause a change in pain, rather than result from it. One theory is that a neurochemical imbalance of monoamines (such as serotonin), which are involved in both mood regulation and the processing of pain processing, could cause both mood disturbance and sensory hypersensitivity, and increase the number and intensity of painful complaints. Patients with depression often report more somatic complaints, such as abdominal pain, back pain and headache (amongst others) compared to those without depression. The usual problems of ascribing causality from clinical data have led researchers to use experimental pain paradigms to compare depressed and non-depressed individuals in their response to noxious stimuli (ice cold water, heat, lasers and similarly fun things). Results from these studies, however, have been startling in their inconsistency. While some have found increased pain in those with depression consistent with clinical data, even more studies have found decreased pain. So which is it – does depression putatively exacerbate or diminish pain?
We conducted a meta-analysis of these studies to try to better understand reasons for the variation in findings. Our search identified 32 relevant experimental studies consisting of a total of 1,317 participants (depressed=641, healthy controls=676). The average age of the aggregated depressed sample was 39 years (74% female), and was 38 years (71% female) for the control sample. Patients were mostly diagnosed with Major Depressive Disorder and pain was assessed using several outcome measures during both anti-depressant medicated and unmedicated states. Pain stimuli included heat, cold, electrical and ischemic (e.g. blood flow restriction via tourniquet followed by handgrip exercises to produce deep muscle pain).
What did we discover? As expected, findings were remarkably varied and much of this variation remains hard to explain. However, one key finding was that the link between depression and pain may be influenced by the type of pain stimulus used. For ischemic inductions, depressed people reported significantly more pain compared to controls (reduced pain threshold and tolerance). For exteroceptive skin stimulation (e.g. heat, cold, electrical), depressed people tended to demonstrate less pain (or no difference). We also found a moderately analgesic effect of anti-depressant medication. Full results are published in the Journal of Pain and the clinical implications of the findings will be considered in the JoP’s December edition of Journal Club.
Explaining this apparent stimulus effect is difficult given that the available data does not allow direct insights into underlying mechanisms. One speculative interpretation that seems to fit neatly is provided by Lautenbacher and Krieg’s global theory of pain processing in psychiatric disorders. In a nutshell, this theory suggests that neurotransmitter dysfunction in depression: (1) globally diminishes processing of all sensory input, and (2) disrupts pain pathways that produce ‘natural’ analgesia that occur as a defence against pain that is sustained or interoceptive in origin (e.g. muscle or joint pain). This could explain why pain might be generally diminished in depression – except when pain is evoked through nociceptors innervating deep structures (e.g. muscle or joint pain, including that activated by ischemic pain inductions) which might fail to benefit from this natural analgesia. The theory also offers a nice explanatory framework that might contribute to our understanding of not only why findings seem to differ within the experimental literature, but also across clinical (involving interoceptive pain) and experimental (usually involving exteroceptive pain) studies.
It is worth restating that this interpretation of the results is speculative. There may be several alternative influences – ischemic pain inductions may have been longer, more painful, or different in some other fundamentally important unidentified way. Furthermore, although evidence for ischemic pain differences in particular seems convincing, it is impossible to be absolutely certain of a stimulus effect. While meta-analysis is an extremely powerful technique, chance results or unreliable findings from individual studies are still propagated into the wider aggregated data (albeit with reduced influence). Nevertheless, it is interesting to note that the explanatory theory predates (therefore predicted?) the vast majority of the included empirical studies and a broadly similar pattern of results is suggested by close inspection of a subsequent independent study whose publication date prohibited its inclusion in the review.
Overall, these findings do support a relationship between depression and pain. But they also indicate that this relationship is complex and requires further research. The fact that pain and depression are leading causes of years lived with disability and are commonly comorbid in clinics across the world suggests this research is warranted.
About Trevor Thompson
Trevor Thompson is a senior lecturer in the Psychology department at the University of Greenwich, UK. He has previously worked as a consultant statistician in the area of patient-reported outcomes in the US. He enjoys research examining the neurophysiological foundations of pain and exploring the potential for technology in pain management. But hates mime artists.
Reference
Trevor Thompson, Christoph U. Correll, Katy Gallop, Davy Vancampfort, Brendon Stubbs (2016) Is Pain Perception Altered in People With Depression? A Systematic Review and Meta-Analysis of Experimental Pain Research. J Pain Volume 17 (12); 1257–1272.