Does pain influence cognitive performance in patients with mild traumatic brain injury?

1 Introduction

Traumatic brain injury (TBI) has been defined as an “alteration in brain function, or other evidence of brain pathology, caused by an external force” [1]. TBI is a significant health concern with an annual estimated global incidence of approximately 50–60 million people [2]. The severity classification of TBI relies on presenting status, including the initial Glasgow Coma Scale score [3], duration of loss of consciousness, and post-traumatic amnesia [4]. Most TBIs (>90%) fall into the mild spectrum [5], commonly abbreviated as mild traumatic brain injury (mTBI). Individuals who suffer from an mTBI experience a wide range of acute symptoms, including physical, perceptual, cognitive, and emotional symptoms. In most cases, these symptoms are transient and gradually resolve within days or weeks after the injury [6]. However, for a significant minority of the mTBI population, some symptoms become persistent for months and even years [7,8,9]. Among the most frequent lingering symptoms is pain, especially headache [10,11].

Earlier, pain syndromes in this patient population were poorly characterized [12]. Bazarian et al. found that less than half of mTBI patients treated in emergency departments had any assessment and documentation of their pain [13]. In recent years, there has been growing attention to the issue of pain following TBI [14]. In a systematic review, Nampiaparampil reported a prevalence of chronic pain in TBI patients (n = 3,289) of approximately 52% [15]. Interestingly, the review also found that psychological disorders such as post-traumatic stress syndrome or depression were unrelated to chronic pain in these patients. Uomoto and Esselman described that mTBI patients reported significantly more headaches than patients with severe TBI [16]. Similar tendencies were found for neck/shoulder, back, and other pain syndromes [17].

Chronic pain defined as pain that persists or recurs for more than 3 months [18] has in general been associated with reduced cognitive function in several domains, including processing speed, executive functions, attention, and memory [19,20]. The level of pain seems to be related to the magnitude of cognitive impairment [21]. However, different underlying mechanisms contributing to the experience of pain may affect cognitive functions differently. Coppieters et al. [22] observed differences among women with whiplash-related chronic neck pain (WAD), idiopathic chronic neck pain, and women without neck pain. Both pain groups showed a lowered quality of life and cognitive functions compared to the healthy controls, but this decrease was more prominent in the WAD group. Specifically, the WAD group had poorer processing speed on an attention task (Trail Making Test [TMT]), memory functions, and general cognitive functions compared to patients with idiopathic neck pain without trauma. Furthermore, the authors found indications of central sensitization and interrelatedness among disability, cognitive deficits, and hyperalgesia in the WAD group. Other studies have found that patients diagnosed with localized chronic pain (e.g., injuries in the neck region) have both self-reported and verified explicit memory problems [23]. Studies have found that patients diagnosed with fibromyalgia exhibit deficits in working memory as well as impairments in both episodic and semantic memory [24].

Chronic pain and mTBI can often coexist and exhibit overlapping symptoms, making them challenging to distinguish based solely on reported symptoms. This means that an assessment of cognitive functions in mTBI patients that does not consider pain may misattribute deficits to brain injury rather than pain. However, the limited number of studies examining the relationship between TBI and chronic pain raises concerns [20]. Therefore, a holistic approach involving a multidisciplinary investigation of physiological, behavioral, cognitive, and emotional aspects is needed [25] as widespread pain assessed at 8 weeks is also a significant predictor for long-term (12-month) posttraumatic stress [26]. However, a study by McCracken and Iverson specifically examined the relationship between chronic pain and chronic cognitive impairment [27]. They found that symptoms overlapped and suggested that the presence of chronic pain should be considered in the interpretations of patients assessed after mTBI.

The current explorative study aimed to examine the frequency of self-reported musculoskeletal pain, and its relation to cognitive functions, measured with neuropsychological tests, in a consecutive sample of patients with mTBI. Additionally, the study aimed to investigate potential differences in how pain and mTBI may impact the pattern of cognitive performance.

2 Methods

This study has a cross-sectional design involving collaboration between two rehabilitation clinics for adult patients with acquired brain injuries.

3 Results

4 Discussion

The primary objective of this study was to investigate the prevalence of musculoskeletal pain in patients who experienced persistent cognitive complaints following an mTBI, in comparison to non-injured controls. Further, possible differences between patients and controls were examined with regard to how pain interacts with cognition. Chronic mTBI patients reported a remarkably high extent of headache and neck/shoulder pain, in the range of what patients with primarily acute or subacute pain report [31,40]. Higher pain level was associated with less severe injury, higher anxiety, and higher self-rated post-concussion symptoms.

The mTBI patients performed less well in the neuropsychological assessment than the non-injured controls. The most marked deficits were noted on tests measuring processing speed and working memory, but the neuropsychological test results were not related to pain among the patients. Pain was, however, associated with higher anxiety levels and higher reports of post-concussion symptoms. However, comparing self-reported data with cognitive test performance presents a general methodological challenge, as shown in a series of studies [41]. Intercorrelations among self-ratings may suggest distinctive response patterns, where some individuals tend to provide extreme estimates, while others tend to remain around the average. However, this does not necessarily imply significant differences in the actual symptoms experienced.

In controls, higher pain was associated with slower processing speed, but no associations were found between pain and anxiety and post-concussion symptoms. The association between pain and processing speed among the controls is in line with previous findings regarding pain and cognitive functions [20,42]. In the mTBI group, the absence of these associations may reflect a disrupted neurocognitive anticipatory pain network [43]. The pain modulatory systems in the brain are physiologically linked to the same regions in the brain stem which often are affected by a traumatic brain injury [15,44].

In the present study, 83% of patients with mTBI reported suffering from headaches, and 57–61% reported neck and shoulder pain. These results are in line with Uomoto and Esselman’s study [16], where the frequency of headache was 89%, neck/shoulder 51 and 45%, and other pain localizations 20%. The Mollayeva et al. study [25] also reported high frequencies of similar pain localizations in mTBI patients.

Previous epidemiological research has found that approximately 5% of individuals in the workforce have isolated neck and shoulder pain [45], which corresponds broadly to the 3–10% neck and shoulder pain we observed in the non-injured control group of our study. Other pain areas, such as the upper back, lower back, and legs, were not significantly different. In this study, we utilized the ÖMPSQ, a tool developed to measure musculoskeletal pain that measures the subjective experience of pain and psychological and social factors to predict the development of chronic pain [31]. The localized pain pattern, concentrated on the head, neck and shoulders, may be attributed to the specific nature of the injury and associated stressful events [17].

Higher pain was associated with a less severe injury. The inverse relationship between pain and severity of TBI has previously been demonstrated in studies, where mTBI patients report more headaches than patients with moderate and severe TBI [15,16,46]. Our study shows that this relationship also holds within the current mTBI group. According to one hypothesis regarding the differential reporting of pain, cognitive and executive impairments in moderate and severe TBI interfere with adequate self-perception and communication [15]. However, we found differences within the present mTBI sample where this explanation seems unlikely. Our tentative hypothesis is that selection bias instead can create this effect in mTBI samples. The largest group of the mTBI spectrum is the very mild mTBI. In these cases, high symptom reporting, including pain, may be required to motivate referrals, thereby aggregating higher pain and symptom reporting in clinical samples. The anxiety level was significantly higher in mTBI patients, in line with earlier findings [47]. Anxiety was related to pain but not to patients’ cognitive performance, as Weyer Jamora and colleagues have previously reported [48].

The patient’s primary cognitive complaint was the main reason for the referral to a neuropsychological assessment. Overall, the mTBI patients performed more poorly on neuropsychological tests, also after controlling for educational level. The tests measuring processing speed and working memory revealed the most significant deficits. This finding is in line with previous studies of clinical samples where moderate effects on cognition are usually observed [49] but also in line with patients suffering from chronic pain [42].

There are several theoretical models to explain cognitive dysfunction in chronic pain. One such model is the limited resource theory, which posits that pain signals disrupt attention and executive processing due to shared structures between pain and attention networks in the brain [50,51]. Another model, “the neuroplasticity theory” has been proposed as an explanation as studies have found pain-related changes in the brain. Studies on both structural and functional levels have found abnormalities in regions involved in attention and executive functioning. These regions include the anterior cingulate cortex (ACC) and prefrontal cortex, which exhibit abnormalities in grey matter volume in patients with chronic pain [52,53], as well as abnormal regional cerebral blood flow in the ACC, caudate nucleus, and thalamus [54].

To expand on this topic, additional studies have found that mTBI might be a vulnerability factor for developing various pain-related conditions, even when controlling for related emotional symptoms [43]. Changes in periaqueductal grey activation following mTBI appear to modulate pain processing and tone and affect the degree of the perceived pain intensity. Furthermore, individual differences in pain vulnerability are associated with smaller amygdala size in the sub-acute stage for patients with lower back pain, as shown by Vachon-Presseau and colleagues [55,56]. Clinically, it can be challenging to sort out what causes what, nevertheless, the neuropsychological assessment reflects the patient’s actual impairments.

Our findings have clinical implications since both assessment and treatment of pain are often overlooked in the treatment of mTBI patients. Differences between patients and controls in our study regarding the relation between self-rated pain and cognition caution us against generalizing conclusions based on other chronic pain groups or healthy subjects to patients with mTBI. Pain seems to impair processing speed, but it does not exclude impaired cognitive performance in mTBI patients due to the injury itself or to other factors (psychosocial or emotional). Our results concerning neuropsychological assessment imply that the complex interactive effects of cognitive deficits, pain, and anxiety cannot be disentangled solely with neuropsychological instruments. There is thus a need for a more multidisciplinary approach with additional biological or neuroimaging markers. If the pain among mTBI patients is neglected, it might hamper the effect of cognitive rehabilitation. A multifactorial approach is, therefore, essential.

This study has limitations. First, the sample size of the mTBI group is relatively small, which may increase the risk of type II error. However, despite these limitations, a clear pattern emerged. Second, the broad time frame post-injury for inclusion in the study, ranging from 4 to 75 months, may have resulted in a heterogeneous sample. Future studies would benefit from a more restricted post-injury time frame as an inclusion criterion. Additionally, we did not employ Bonferroni adjustments for multiple comparisons, which can elevate the risk of type I error. While this was a deliberate choice to prevent significant findings to be obscured in our exploratory research, it necessitates a cautious interpretation of the significance levels reported. Future studies may seek to apply more stringent statistical corrections.

Moreover, while the participants in our study sought help for post-concussive symptoms that had persisted for months, even years post-injury, they do not represent all individuals who seek care immediately after an mTBI. This distinction is critical as it separates our sample from those who may experience different recovery trajectories, including those who never seek any subsequent care at all, apart from emergency department care. This selection biases our sample towards individuals with more chronic or long-term issues leading to their presentation in clinical settings. Consequently, while the findings are highly relevant for practitioners and reflect the patient profiles typically encountered in clinical practice, they may not generalize to the entire population of those who have sustained an mTBI. Prospective studies should aim to capture the full spectrum of mTBI experiences by including individuals with varied post-injury care-seeking behaviors.

A third limitation is that pain reports were solely based on symptom reports, which prevented further analysis of the pain’s specificity, onset, and its association with the event that caused the mTBI.

Another limitation of our study is that many patients were using pain medication, which could have affected their cognitive functioning. Subanalyses on the effects of these medications were not feasible due to the limited power of our study. Consequently, no definitive conclusions can be drawn regarding the impact of medication on the results.

It is also important to acknowledge an administrative oversight that resulted in the non-collection of three participants’ questionnaires, an error that was addressed in subsequent procedures to ensure data completeness. Despite this, the data analyzed did not contain missing responses.

This study also has strengths. First, the mTBI patients were consecutively enrolled. Thus, the findings reflect complaints of patients routinely seen in rehabilitation settings. Second, we used a broad measure of pain that encompassed current pain, average as well as pain in the last 3 months, and pain frequency. Third, the neuropsychological assessment was comprehensive and performed by experienced clinical neuropsychologists.

To conclude, mTBI patients with persistent cognitive symptoms endorse the prevalence of headache and neck and shoulder pains to a similar extent as subacute musculoskeletal pain patients. Still, in our study, mTBI patients did not report pain in other locations. Furthermore, the mTBI group exhibited significant cognitive impairment in the domains of processing speed and working memory. However, unlike controls, cognitive impairment in the mTBI group was not related to the reported pain level. In controls, there was a correlation between processing speed and pain level.

According to our results, musculoskeletal pain should be considered in rehabilitation settings where mTBI patients are treated. Additionally, given the shared underlying brain mechanisms involved in working memory, processing speed, pain, and anxiety, a combination of treatment approaches in mTBI or chronic pain rehabilitation is recommended. Examples of a treatment program might be a combination of intensive cognitive training [57], controlled aerobic exercise [58], and psychoeducative approaches [59].