Influence of psychological and somatosensory characteristics on treatment outcomes in adolescent chronic postsurgical and primary pain

Patients with chronic postsurgical pain and chronic primary pain have differences in biopsychosocial factors, somatosensory profiles, and treatment outcomes.

Abstract

Introduction:

Chronic postsurgical pain (CPSP) affects 20% to 30% of pediatric patients at 12 months after surgery. We hypothesized that differences in biopsychosocial and somatosensory profiles exist between patients with CPSP and those with chronic primary pain (CPP) and that these differences affect treatment outcomes.

Methods:

This cohort included prospectively collected, retrospectively analyzed data from adolescents treated at the Edwards Family Interdisciplinary Center for Complex Pain between May 2016 and May 2022. During clinic intake, patients completed the Functional Disability Inventory, Revised Child Anxiety and Depression Scale, and Pittsburgh Sleep Quality Index to assess for disability, anxiety, depression, and sleep quality, respectively. Baseline quantitative sensory testing/conditioned pain modulation (QST/CPM) was performed to characterize somatosensory functioning. The primary outcome was a score of 6 (“much better”) or 7 (“very much better”) on the Patient Global Impression of Change (PGIC) measure at treatment end.

Results:

Of 99 patients (14.8 ± 2.0 years, 90% female), 46 had CPSP and 53 had CPP. Compared to CPP, patients with CPSP had less baseline functional disability, coexisting depression, and social phobia and better sleep quality. After adjusting for psychosocial differences, patients with CPSP were more likely to score ≥6 on PGIC than patients with CPP (odds ratio = 3.0 [95% confidential interval: 1.3–7.7], P = 0.01). Patients with dynamic mechanical allodynia and/or temporal summation were less likely to score ≥6 on PGIC compared to patients without (odds ratio = 0.33 [95% confidential interval: 0.13–0.82], P = 0.02).

Conclusion:

Differences in biopsychosocial factors, somatosensory profiles, and treatment outcomes exist between patients with CPSP and CPP. The presence of centrally driven pain hypersensitivity on baseline QST/CPM was associated with worse recovery prognosis.

1. Introduction

Chronic postsurgical pain (CPSP) affects 20% to 30% of pediatric patients at 12 months after surgery and is a major barrier to recovery.²⁴ With up to 5 million children and adolescents undergoing surgery in the United States every year, the short- and long-term detrimental effects of this condition are prevalent, difficult to treat, and often result in substantial functional disabilities.^24,27 The early prediction, identification, and management of this pain is vital to avoiding long-term negative outcomes.^11,23,25

Although pediatric CPSP and chronic primary pain (CPP) are recognized as separate diagnoses within the ICD-11, both are characterized by substantial biopsychosocial burden. Adolescents with CPSP and CPP frequently report anxiety, depression, sleep disturbance, and school or activity limitations that intensify the impact of pain on daily functioning.²¹ Clinical management for both groups is also similar and typically involves multimodal strategies that integrate pharmacologic therapy, physiotherapy, and psychological interventions.⁵ However, despite these overlapping characteristics and treatment approaches, emerging evidence indicates that CPSP and CPP involve distinct underlying mechanisms.²⁶ Unlike CPSP, patients with CPP often present with a nociplastic pain mechanism, defined as pain that cannot be explained by tissue damage or a lesion to the somatosensory circuit.¹⁶ Available work suggests that CPP may be associated with greater psychosocial burden, worse functional disability, and more widespread sensory abnormalities than secondary pain conditions.²¹ In contrast, CPSP more often presents with localized nociceptive or neuropathic features, which may respond more favorably to mechanistically targeted interventions.^1,4,5,22 These observations support the hypothesis that treatment outcomes may differ between pediatric CPSP and CPP, yet systematic comparisons are lacking.

Somatosensory characteristics, assessed by quantitative sensory testing and conditioned pain modulation (QST/CPM), may further differentiate these conditions. Quantitative sensory testing and conditioned pain modulation provides a structured approach to assessing peripheral and central pain processing.² Prior work has shown that QST/CPM findings, particularly those related to central sensitization, are associated with persistent pain trajectories.^17,19,21,30 However, no studies have directly compared QST/CPM between adolescents with CPSP and CPP or examined whether these somatosensory mechanisms predict treatment outcomes across diagnostic groups. Thus, it remains unclear how biopsychosocial differences between these 2 pediatric patient populations affect treatment approaches and overall long-term outcomes.^10,18

Therefore, the objective of this study was to characterize the psychosocial and somatosensory phenotypic profiles of adolescents with CPSP and CPP and to determine whether these profiles were associated with treatment outcomes at an interdisciplinary pediatric pain center. The primary outcome was to compare the Patient Global Impression of Change (PGIC) scores⁹ between the 2 groups. We hypothesized that patients with CPSP would have higher PGIC scores at the end of treatment suggesting greater responsiveness to targeted interventions. We also hypothesized that QST/CPM pain phenotypes indicating central sensitization would be associated with worse treatment response across both diagnostic groups.

2. Methods

This investigation is a retrospective, single-center cohort study that included data from patients treated at the Edwards Family Interdisciplinary Center for Complex Pain. Upon entering the clinic for treatment, patients either elected to or declined to be involved in pain-related research that would not affect their clinical treatment. Those who consented to enroll in the research arm had their clinical data prospectively collected and stored in a protected database for any future unspecified research. Before the initiation of study activities, the investigation was approved by the Research Ethics Board of the McGill University Health Center (2019-4887, 2019-4670). In accordance with the Research Ethics Board, written informed consent was obtained from parents or legal guardians and from patients aged 14 years and older, whereas verbal assent was obtained from patients younger than 14 years of age. Deidentified data containing no Protected Health Information were securely transferred to the Duke University Health System investigators for analysis. The Duke University Health System Institutional Review Board exempted the study from approval due to the retrospective and deidentified nature of the study. This manuscript adheres to the applicable STROBE guidelines.

2.1. Study environment

The Edwards Family Interdisciplinary Center for Complex Pain of the Montreal Children's Hospital is a comprehensive pain center, which treats approximately 500 children and adolescents per year with a variety of chronic pain conditions. The clinicians at the Edwards Family Interdisciplinary Center for Complex Pain accept and treat patients with chronic pain, defined as persistent or recurrent pain for at least 3 months. Over 90% of patients are referred to the Center by specialists.

2.2. Study cohort

The cohort included children and adolescents (<18 years of age) treated at the Edwards Family Interdisciplinary Center for Complex Pain from May 2016 to May 2022. Diagnoses of chronic postsurgical pain (CPSP) and chronic primary pain (CPP) were made clinically by pediatric pain physicians using the ICD-11 classification system developed by the International Association for the Study of Pain.^16,28 Patients were included in this analysis if they consented to research, had a complete work-up at their initial clinic visit including all assessments and sensory testing, and had a documented end of treatment date. Patients with ambiguous or co-occurring pain diagnoses were excluded to ensure diagnostic clarity for the current study.

2.3. Data collection

Center-specific procedures and the details on the collection of clinical outcome measures have been previously described in our prior work.^5,21 The use of comprehensive QST/CPM protocols before the initiation of treatment has been universally offered to and conducted for patients treated at the Center for Complex Pain as part of routine practice since 2017. After informed consent, data on clinical characteristics, thermal, mechanical, and vibrational QST and CPM evaluation,^2,5,12,21 baseline validated pain-related questionnaires and functional assessments,^{3,6–8,14,29} longitudinal pain scores using the 11-point numerical rating scale (NRS, 0–10), and multidisciplinary personalized treatment plans (medications prescribed, physiotherapy, psychology, interventional procedures) were prospectively collected and recorded in a protected database.^4,5 Treatment period was the time between the first clinic visit and end of treatment date. Patient data were deidentified before analysis.

2.4. Outcome measures

2.4.1. Primary outcome

The primary outcome of this investigation was a score of 6 or 7 on the PGIC, a validated 7-item self-report scale that assesses a patient's perception of overall improvement after treatment.⁹ Patient Global Impression of Change Scale responses range from 1 (“very much worse”) to 7 (“very much better”). For analytic purposes, we defined clinically meaningful improvement as a PGIC scores of 6 (“much better”) or 7 at the end of treatment, consistent with prior pediatric and adult chronic pain studies. All PGIC scores, including below this threshold, were included in analyses comparing patients who achieved clinically meaningful improvement (PGIC ≥6) with those who did not (PGIC <6). Differences in PGIC outcomes were examined by pain diagnosis (CPSP vs CPP), QST/CPM phenotype, and duration of pain at time of initial presentation.

2.4.2. Secondary outcomes

Secondary outcomes included differences between patients with CPSP and CPP in psychosocial functioning, sleep quality, neuropathic pain features, pain catastrophizing, family impact, sensory profiles on QST/CPM, and components of individualized multimodal treatment plans. The questionnaire measures, combined with the categorical QST/CPM phenotypes described below, form each patient's baseline biopsychosocial and somatosensory “profile” as used throughout this manuscript.

2.4.2.1. Patient and parental baseline questionnaires

The Functional Disability Inventory (FDI),²⁹ Revised Child Anxiety and Depression Scale (RCADS),⁷ and Pittsburgh Sleep Quality Index (PSQI)⁶ were completed to assess for degree of disability, depression and anxiety, and sleep quality, respectively. The FDI is a 15-item measure assessing perceived difficulty in performing everyday activities due to pain. Responses are rated on a 5-point Likert scale (0 = no trouble to 4 = impossible) with higher scores indicating greater disability. The RCADS is a 47-item self-report measure assessing symptoms of anxiety and mood disorders across 6 subscales including separation anxiety, social phobia, generalized anxiety, panic, obsessive–compulsive, and major depressive disorder. Scoring is determined by clinical cut-offs compared to peers based on age and gender with higher T-scores indicating greater severity. The PSQI is a 19-item questionnaire evaluating sleep quality over the past month, yielding a global sleep quality score (range 0-21) with higher scores indicating worse sleep quality. The Douleur Neuropathique 4 (DN4) questionnaire, a 10-item screening tool for sensory descriptors, was used to assess for the presence of neuropathic pain.³ A score ≥4 indicates the presence of neuropathic pain. Patients and parents completed the Pain Catastrophizing Scale for Children (PCS-C) and the Pain Catastrophizing Scale for Parents (PCS-P), respectively.^8,14 Both the PCS-C and PCS-P are 13-item scales assessing rumination, magnification, and helplessness related to pain. Responses are rated on a 5-point Likert scale (range 0-52) with ≥30 considered clinically significant. Parents also completed the Impact of Family Scale, a 15-item instrument evaluating the perceived burden that chronic childhood illness had on family functioning.³¹ Responses are rated on a 4-point Likert scale (range 15–60) with lower scores indicating greater negative impact. These measures were selected due to their strong psychometric properties and established use in pediatric chronic pain research.

2.4.2.2. Quantitative sensory testing/conditioned pain modulation

Thermal, mechanical, and vibrational QST and CPM were performed using standardized procedures consistent with pediatric protocols.² Mechanical QST was conducted on a control area (dominant hand) followed by the affected area (the most painful site reported by the patient). Thermal QST was conducted on the volar surface of both forearms. For each modality, raw scores were converted to age- and sex-adjusted z-scores of the control and affected areas.^2,19 A positive z-score reflects gain of function (hyperalgesia or allodynia), and a negative z-score reflects loss of function (sensory loss). The average z-score was calculated across parameters for each site. Conditioned pain modulation was assessed using a heterotopic noxious conditioning stimulus consistent with previously published pediatric methods.¹² Conditioned pain modulation efficiency was calculated as the change in pressure pain threshold at the test site during conditioning relative to baseline, with lower values indicating inefficient endogenous pain inhibition.

2.4.2.3. Phenotype classification

Pain phenotypes were derived using a deterministic approach adapted from prior work in pediatric chronic pain.^19–21 Each phenotype represented a categorical interpretation of underlying somatosensory mechanisms: phenotype A, “Central sensitization,” was defined by the presence of dynamic mechanical allodynia and/or temporal summation at either the control or pain site; phenotype B, “Peripheral sensitization,” was defined by reduced pressure pain thresholds at the pain site relative to age- and sex-adjusted norms; phenotype C, “Suboptimal/inefficient pain modulation,” was defined by suboptimal CPM response relative to normative pediatric data. Each phenotype was coded as present or absent for each patient. Phenotypes were not mutually exclusive, and patients could exhibit more than one phenotype or none, reflecting distinct combinations of pain processing profiles.⁵ These phenotypes were then used to stratify patients for analysis of PGIC outcomes.

2.4.2.4. Treatment plans

Multidisciplinary treatment decisions at our center were intentionally individualized based on each patient's diagnosis, biopsychosocial profile, and somatosensory findings from QST/CPM as previously described by Bruneau et al.^4,5 The treatment plans for our cohort of patients were assessed for differences in therapeutic approaches based on diagnosis.

2.5. Statistical analysis

Baseline cohort characteristics, clinical assessments, QST/CPM phenotypes, and treatment plans were summarized and stratified by pain diagnosis classification (CPSP vs CPP). Continuous variables were summarized as either median (IQR) or mean (SD), and categorical variables were summarized as frequency (%). Independent t-tests, Mann–Whitney U tests, Chi-square tests for independence, and Fisher exact tests were used to determine significant differences between patients with CPSP and CPP. PGIC, our primary outcome of interest, was also summarized and stratified by pain diagnosis classification and duration of pain at the initial presentation. In addition, univariate and multivariable regression models were run to determine the association between pain diagnosis and QST/CPM phenotype status with PGIC (≥6 vs <6). Patient global impression of change was analyzed via logistic regression, and results were reported as odds ratios with a 95% confidence interval. Our covariates of interest in the multivariable models were FDI and RCADS general anxiety scores recorded at the initial visit as they were identified a priori as potential confounders.

3. Results

There were 99 adolescents included in the analysis. Across the total cohort, the mean age at clinic entry was 14.9 (2.0) years, and 90% of patients were female. Over 67% of patients reported pain for >12 months at the initial clinic visit with an average pain score of 6/10 over the prior month. Only 13% of patients reported that their pain duration was between 3 and 6 months. The cohort included 46 patients with CPSP and 53 patients with CPP; baseline clinical characteristics stratified by pain diagnosis are shown in Table 1. Average pain scores, duration of pain, frequency of pain episodes, and duration of painful episodes were similar between groups. Median (IQR) duration of treatment was 318 (172–504) days for patients with CPP and 241 (131–363) days for patients with CPSP.

Table 1.

Clinical Characteristics of the cohort stratified by pain diagnosis.

Clinical characteristics	Overall N = 99	CPSP N = 46	CPP N = 53
Sex (% female)	88 (89.8)	39 (86.7)	49 (92.5)
Race (%)
Person of color	16 (16.2)	12 (26.1)	4 (7.5)
White/Caucasian	83 (83.8)	34 (73.9)	49 (92.5)
Age at entering date	14.9 (2.0)	15.2 (1.9)	14.6 (2.1)
Average score (month)	6.0 (1.8)	5.7 (1.8)	6.2 (1.7)
Duration of pain (%)
3–6 months	13 (13.3)	7 (15.6)	6 (11.3)
6–12 months	14 (14.3)	6 (13.3)	8 (15.1)
More than 12 months	66 (67.3)	28 (62.2)	38 (71.7)
Frequency of pain (%)
Once a day	84 (85.7)	40 (88.9)	44 (83.0)
Every 2nd day	11 (11.2)	3 (6.7)	8 (15.1)
Once a week	2 (2.0)	1 (2.2)	1 (1.9)
Once a month	1 (1.0)	1 (2.2)	0 (0.0)
Painful episodes (%)
Constant	77 (78.6)	34 (75.6)	43 (81.1)
Few seconds	3 (3.1)	3 (6.7)	0 (0.0)
Few minutes	8 (8.2)	4 (8.9)	4 (7.5)
Few hours	14 (14.3)	4 (8.9)	10 (18.9)

3.1. Global impression of change outcome

At the end of treatment, the proportion of patients reporting clinically meaningful improvement (PGIC ≥6) differed by pain diagnosis (Fig. 1). Patients with CPSP were more likely to achieve a PGIC ≥6 compared to patients with CPP (odds ratio [OR] = 3.0 [95% CI: 1.3–7.7], P = 0.013). Across both diagnostic groups, somatosensory phenotype was also associated with treatment outcome. Patients demonstrating dynamic mechanical allodynia and/or temporal summation on QST were less likely to achieve PGIC ≥6 than patients without these findings (OR = 0.33 [95% CI: 0.13–0.82], P = 0.018). Pain duration at initial presentation was additionally associated with treatment response. Among patients presenting with pain duration >12 months, 20 of 67 (30%) achieved PGIC ≥6, compared with 25 of 32 (83%) patients presenting with pain duration <12 months (OR = 8.4 [95% CI 3.1–22.5], P < 0.001). These findings were driven by patients with CPP, among whom pain duration >12 months at initial treatment was associated with lower likelihood of clinically meaningful improvement (P < 0.001). No significant association between pain duration and PGIC outcome was observed among patients with CPSP.

Figure 1.

Patient Global Impression of Change score stratified by pain diagnosis. The Patient Global Impression of Change (PGIC) is a 7-item self-report questionnaire that reflects the patient's belief about the efficacy of treatment.

3.2. Baseline psychosocial characteristics of pain

At baseline, patients with CPP reported greater pain-related functional disability than patients with CPSP, as measured by higher FDI scores (22 [17–27] vs 16 [10–23], P < 0.02). In addition, patients with CPP reported greater sleep disturbances and poorer sleep quality, as measured by higher PSQI scores (10 [7–12] vs 8 [5–11], P = 0.04). Scores on the DN4 questionnaire, Impact of Family Scale, and patient and parent-reported PCS were similar between groups.

Results of the RCADS are shown in Figure 2. Patients with CPP reported a higher burden of social phobia (44 [37–52] vs 32 [26–40], P < 0.001) and depression (58 [48–69] vs 51 [40–57], P = 0.010) compared with patients with CPSP. The remainder of the RCADS subscale scores, including separation anxiety, general anxiety, panic, and obsessive–compulsive symptoms, were comparable across groups.

Figure 2.

The Revised Child Anxiety and Depression Scale (RCADS) stratified by pain diagnosis. The Revised Child Anxiety and Depression Scale (RCADS) is a 47-item, youth self-report questionnaire with the following subscales: separation anxiety disorder, social phobia, generalized anxiety disorder, panic disorder, obsessive–compulsive disorder, and low mood (major depressive disorder). Subscale results from RCADS questionnaires administered at the time of clinic entry are plotted by primary pain diagnosis—CPSP or CPP. *P < 0.05, **P < 0.01.

3.3. Baseline quantitative sensory testing/conditioned pain modulation

Quantitative sensory testing/conditioned pain modulation results are summarized in Table 2. Patients with CPP demonstrated lower pressure pain thresholds (P = 0.007) and greater dynamic mechanical allodynia at their control site (P = 0.02) compared with patients with CPSP. Patients with CPP also exhibited greater sensory loss at the primary pain site, as reflected by vibrational detection threshold measures (P = 0.03).

Table 2.

Quantitative sensory testing/conditioned pain modulation stratified by pain diagnosis.

	Chronic postsurgical pain (CPSP)	Chronic primary pain (CPP)	P
MDT (g), mean ± SD
Affected site (z-score, log)	1.17 (1.79)	0.53 (1.53)	0.06
Control site (raw)	0.10 (0.15)	0.09 (0.15)	0.27
DMA (NRS 0–10), mean ± SD
Affected site (z-score, scale)	1.57 (1.52)	1.00 (1.43)	0.06
Control (raw)	0.03 (0.13)	0.09 (0.22)	0.02
VDT (x/8), mean ± SD
Affected (z-score)	−0.46 (1.07)	−1.05 (1.44)	0.03
Control (raw)	6.84 (0.96)	6.83 (0.86)	0.78
PPT (Newtons), mean ± SD
Affected (z-score, log)	−1.11 (1.25)	−0.92 (1.49)	0.51
Control (raw)	32.20 (12.17)	26.37 (15.79)	0.007
TSP (NRS −10 to +10), mean ± SD	0.33 (2.11)	−0.20 (1.57)	0.16

MDT, mechanical detection threshold; DMA, dynamic mechanical allodynia; VDT, vibration detection threshold; PPT, pressure pain threshold; TSP, temporal summation of pain.

Pain phenotypes derived from QST/CPM are shown in Figure 3A (CPP) and B (CPSP). The presence of central sensitization (phenotype A) was higher among patients with CPP than CPSP (63% vs 43%, P = 0.037). There were no differences between groups in the presence of peripheral sensitization, inefficient conditioned pain modulation, nor all 3 phenotypes. Two patients in each diagnostic group demonstrated normal QST/CPM findings.

Figure 3.

Proportional 3-set Euler diagram by pain phenotypes. Pain phenotypes represented include (A) allodynia and/or temporal summation (“central sensitization”), (B) pain pressure sensitivity (“peripheral sensitization”), (C) suboptimal or inefficient conditioned pain modulation (CPM), and (D) = normal QST/CPM. (A) Chronic primary pain diagram: central sensitization = 63%; peripheral sensitization = 72%; suboptimal or inefficient CPM = 71%; all 3 phenotypes = 30%; normal QST/CPM = 4%. (B) Chronic postsurgical pain diagram: central sensitization = 43%; peripheral sensitization = 86%; suboptimal or inefficient CPM = 61%; all 3 phenotypes = 15%; normal QST/CPM = 4%. CPP had a higher prevalence of central sensitization than CPSP (P = 0.037).

3.4. Treatment plans

Treatment plans at the initial clinic visit were individualized based on pain diagnosis, pain distribution (multisite or local), functional and behavioral assessments, and QST/CPM findings, as described in our prior work.⁵ Pharmacologic therapy was common, utilized in 89% of patients and at similar rates between groups. Psychology services were more frequently prescribed in patients with CPP than CPSP (91% vs 54%, P < 0.001) as was physiotherapy (96% vs 80%, P = 0.02). In contrast, interventional treatment with regional anesthesia was more commonly utilized in patients with CPSP than CPP (80% vs 11%, P < 0.001). Regional anesthesia techniques included fascial plane blocks (eg, rectus sheath, transverse abdominal plane block, erector spinae block), peripheral nerve blocks (eg, brachial plexus, femoral, saphenous, sciatic, genicular), sympathetic blocks (eg, lumbar sympathetic, ganglion impar), and scar infiltration. Among the 37 patients who had a regional anesthetic, 28 (75%) reported noticeable improvement in their pain at the end of treatment.

4. Discussion

The purpose of this investigation was to examine treatment outcomes in pediatric patients with CPSP and CPP and determine whether diagnosis-specific biopsychosocial and somatosensory profiles may be useful prognostic tools to better predict clinical recovery. After adjusting for biopsychosocial differences between the groups, we found that patients with CPSP had greater improvement in their pain symptoms at the end of the treatment period compared to those with CPP. However, regardless of a primary vs secondary pain diagnosis, patients with QST/CPM findings consistent with central sensitization at the initial clinic visit were less likely to achieve clinically meaningful recovery at the end of treatment.

In this cohort, patients diagnosed with CPSP and CPP had similar pain intensity, duration, frequency, quality, and descriptors as well as similar levels of both patient and parental catastrophizing. Nonetheless, we observed differences between the baseline functional and psychological assessments in patients with CPSP vs CPP, specifically greater disability, social phobia, depression, and poorer sleep quality in patients with CPP. This suggests that despite reporting similar pain levels, symptoms, and duration on clinic entry, patients with CPP experienced more daily limitations and pain interference before treatment. In addition, there were notable differences observed in the QST/CPM between pain diagnosis. Patients with CPP demonstrated dynamic mechanical allodynia and lower pressure pain thresholds at the control site, suggesting more widespread pain sensitivity, along with greater sensory loss at the primary pain site, which may reflect altered sensory processing rather than localized nerve injury.

As we have described in our prior work, all treatment plans were individualized, and all patients had access to comprehensive interdisciplinary care.^4,5 Although pharmacologic therapy was common, the selection of specific medications, interventional procedures, and psychological therapies was tailored to each patient's diagnosis, biopsychosocial profile, and somatosensory findings on QST/CPM.^4,5 Thus, there were considerable differences in treatment strategies between patients with CPSP and CPP, particularly the use of interventional procedures and psychotherapy. Nonetheless, even within this mechanistically informed and personalized approach, patients with CPP demonstrated poorer global improvement than those with CPSP.

Notably, most patients included in this study were referred to the clinic late in their pain course and largely experienced constant symptoms. The severity and chronification of pain reported in this cohort is an important consideration in the interpretation of our results. We found that patients with CPP greater than 12 months at the time of clinic entry were significantly less likely to progress to clinically meaningful recovery. This was in contrast to patients with CPSP, in which duration of pain before treatment was not a significant predictor of recovery. This suggests that although the early treatment of CPSP is critical to improve function and reduce suffering, it may be possible for those with long-term pain from CPSP to be managed successfully and recover from this condition.

Most patients with CPSP in this cohort received regional anesthesia as part of their interdisciplinary treatment plan for localized pain. In adult surgical populations, studies have shown that perioperative regional anesthesia may be associated with a lower risk of developing CPSP.^1,22 In contrast, evidence supporting the use of regional anesthesia as a modality to treat CPSP and other chronic pain conditions is limited, particularly in pediatric populations. In the present study, regional anesthesia was used in conjunction with pharmacologic and psychological therapies and was observed to be a successful approach for most patients with CPSP. Prospective investigations are needed to examine long-term analgesic and functional outcomes of regional anesthesia use in pediatric patients with CPSP and other localized chronic pain conditions.

Finally, although the primary purpose of this study was to more formally differentiate and characterize pediatric CPSP and CPP as distinct syndromes, we also found that regardless of pain diagnosis, patients with centrally driven pain hypersensitivity on QST/CPM were less likely to achieve meaningful clinical recovery. This observation is aligned with adult studies that have demonstrated a predictive relationship between baseline QST and follow-up musculoskeletal pain irrespective of the underlying diagnosis (ie, osteoarthritis, rheumatoid arthritis, low back pain).¹³ Taken together, this suggests that the mechanism of pain, not the diagnosis, may be more crucial in the chronification of pain. In the present study, our results highlight the importance of baseline QST/CPM for pediatric patients with chronic pain conditions as these evaluations may provide an opportunity for early risk stratification, improved prognostication, and targeted interventions. To that end, a brief QST/CPM protocol was recently published that demonstrated feasibility, tolerability, and reliability in 2 cohorts of children and adolescents with acute postsurgical and post-traumatic pain.¹⁵ This protocol included a pressure task, temporal summation of pain, heat task in conjunction with CPM, and cold pressor tolerance and may potentially be considered as a standardized approach for use in CPSP evaluation.

Interpretation of our findings is subject to a number of limitations. This is a retrospective study with data from a single site, and thus, results may not be generalizable. Future prospective, multicenter studies that use a standardized QST/CPM protocol are needed to determine relationships between pain mechanisms and additional clinical outcomes. Although we used a mechanistically informed approach to tailor and individualize treatment strategies within a comprehensive biopsychosocial framework, we acknowledge that the variation in treatment represents an important limitation in this retrospective analysis. Furthermore, the clinic's standardized intake questionnaire used a categorical format, which did not allow us to analyze pain duration as a continuous variable. Actual duration of pain is likely to influence clinical trajectories, and the use of a continuous measure may provide more nuanced insight. As most patients had pain greater than 12 months at clinic entry, additional studies that conduct biopsychosocial assessments and develop somatosensory profiles at precise time points are needed to further compare pain chronification and pain phenotypes in pediatric patients with CPSP and CPP. In addition, the use of regional anesthesia as treatment approach was not standardized in that the treating clinician determined the optimal technique. Developing formalized protocols to study specific nerve block techniques to manage localized chronic pain will be needed to evaluate the clinical effect of these interventional procedures on outcomes. Finally, we were unable to evaluate the use of regional analgesia at the time of the index surgery. These perioperative data were not consistently available in the retrospective dataset, and the CPSP subgroup was not sufficiently powered to explore the preventive role of regional anesthesia on CPSP development. Such analyses would require a different study design and a substantially larger prospective cohort focused on perioperative predictors of chronic postsurgical pain.

In summary, pediatric CPSP and CPP represent distinct syndromes along the spectrum of complex pain conditions, characterized by different biopsychosocial and somatosensory profiles. Across diagnoses, patients who demonstrated QST/CPM results consistent with a centrally driven pain mechanism were less likely to achieve clinically meaningful improvement, even within an individualized intradisciplinary treatment framework. These findings highlight the potential value of incorporating QST/CPM into clinical assessment to better identify patients at risk for poorer outcomes and to guide mechanism-informed treatment planning within the biopsychosocial framework.

Disclosures

The authors have no conflict of interest to declare.