

  All Discussions  |  

  Welcome Post  |  

  SIG's Charter  |  

  Direct Messages  |   

  Categories  |  

  How-To(Forum)

CPSP Research - Basic guide to CPSP

Nige

https://painrelieffoundation.org.uk/wp-content/uploads/2023/10/DEALING-WITH-PAIN-CPSP.pdf

Ruthy

Thanks for sharing, this will be helpful for new CPSP sufferers I hope 🤞.
It might also be handy to print off and hand to the GP when attending an appointment relating to CPSP. I don’t think any of the GPs at my surgery know any of this. Especially psychological support - when I asked about referal to a pain psychologist it was as though I’d asked to be sent to Venus, and just got sent on another pain management course 🙄.

Nige

Can I also suggest that the posts on CPSP and Brain Fog together with the post on Is it always CPSP which is the cause of pain, might be very useful in understanding CPSP. It is perhaps important to consider that we have had a stroke and the unpleasant items that can cause.

CPSP of course is overlaid on top of any direct stroke impact.

Nige

In neurology, this overarching problem is called a loss of central inhibition.To understand why this happens, you can think of a healthy brain and spinal cord as having a built-in brake system.

The nervous system uses a calming chemical (primarily GABA) to act as a brake, keeping your senses from feeling too much and your muscles from moving on their own.When a stroke damages these upper motor neuron pathways, it essentially cuts the brake lines.

This causes two main problems:
The Brake Is Lost (Loss of Inhibitory Control): The central nervous system can no longer send down the calming signals needed to quiet things down.

The System Runs Wild (Hyperexcitability): Without those brakes, the nerves become overly sensitive and fire out of control.Because the damage happens to the central control center, it hits both systems at the same time:In your sensory system, the nerves run hot and register normal or mild sensations as severe burning, shooting pain (Central Post-Stroke Pain).In your motor system, the muscle reflexes run hot and overreact to minor shifts or positions, causing muscle fibers to pull and twist (Dynamic Hypertonia).

They are essentially two sides of the same coin: a nervous system that has lost its ability to turn itself down. This is why treating one often helps calm the oth

Simon

CPSP diagnosis…

…is clinical, based on a hierarchical grading system requiring:
(1) stroke confirmed by CT/MRI,
(2) pain onset temporally related to stroke,
(3) pain distribution neuroanatomically plausible (corresponding to somatosensory deficit territory),
(4) sensory signs on examination, and
(5) exclusion of other pain causes (spasticity, musculoskeletal, peripheral neuropathic pain) [1].

Per AHA/ASA Stroke Rehabilitation Guidelines, the diagnosis of CPSP should be based on established diagnostic criteria (Klit et al. criteria) after other causes of pain have been excluded, with pharmacotherapy individualized to patient needs. [2]

Three Levels of Diagnostic Certainty (NeuPSIG Grading)

Level Requirements Clinical Evidence
Possible CPSP History of stroke + pain in neuroanatomically plausible distribution (body area corresponding to CNS lesion) Assessed via pain history and validated screening questionnaires [3]
Probable CPSP Above + physical examination revealing sensory signs (hypoesthesia, allodynia, thermal/mechanical deficits) in plausible distribution — sufficient to initiate treatment Sensory signs include thermal hypoesthesia, mechanical allodynia, hyperpathia [1]
Definite CPSP Above + confirmatory neuroimaging (MRI/CT) demonstrating CNS somatosensory lesion explaining pain distribution, other causes excluded ICD-11 requires negative/positive sensory signs in the affected body region [4]

Step-by-Step Diagnostic Approach

Establish stroke history and temporal relationship
Confirm stroke by MRI or CT; CPSP typically begins within days to months (mean ₆₀ days, up to 8 months post-stroke) [5]
Incidence _7–12% of all stroke patients; up to 50% after medullary, thalamic, or operculo-insular strokes [1]
Assess pain distribution for neuroanatomical plausibility
Pain must correspond to the body territory represented by the lesioned CNS structures (half body, face, limb)
Key lesion sites: thalamus (ventroposterior nucleus), spinothalamic tract, insular/somatosensory cortex; thalamic stroke found in 36% of CPSP cases [6]
Characterize pain quality
Classic descriptors: burning, aching, pricking, freezing, electric shock-like
Key associated signs: allodynia (touch, cold, movement) and pricking hypoesthesia — most discriminating features of CPSP vs. other post-stroke pain [7]
Use validated neuropathic pain screening questionnaires
DN4 (strongest recommendation): pooled sensitivity 0.89, specificity 0.88 (27 studies); includes sensory examination items [8]
LANSS: strong recommendation; includes bedside sensory exam component [8]
PainDETECT / S-LANSS: weak recommendation (variable sensitivity/specificity) [8]
In a 602-patient cohort, DN4, LANSS, and PainDETECT were used in combination to diagnose CPSP in 12% of stroke patients [7]
Bedside sensory examination
Test: light touch, pinprick, temperature (warm/cold), vibration, and proprioception in the painful region
CPSP hallmarks: contralesional cold hypoesthesia (detected even before pain onset in acute stroke, p=0.04), mechanical allodynia, cold hyperalgesia [5] [9]
Thalamic CPSP specifically: elevated mechanical detection thresholds + lower mechanical pain thresholds vs. stroke controls (p=0.014–0.017) [10]
Confirmatory neuroimaging
MRI is preferred over CT for identifying somatosensory pathway lesions (thalamus, brainstem, internal capsule, somatosensory cortices)
Lesion must anatomically explain the distribution of sensory deficits and pain
Exclude alternative pain causes
Actively rule out: spasticity-related pain, hemiplegic shoulder pain, musculoskeletal/joint pain, complex regional pain syndrome, peripheral neuropathy
Non-neuropathic etiology is present in up to ⅓ of stroke patients with pain in the somatosensory deficit area [1]
Consider QST for diagnostic confirmation (selected cases)
Weak recommendation per EAN/EFNS guidelines; QST abnormalities present in ₉₆% of definite CPSP (Krause 2016, n=50) [8]
Gradient-boosting QST model predicts CPSP with 84.6% accuracy before pain onset [5]
Most useful when diagnosis remains uncertain after clinical assessment

Sources

@article{Winstein2016StrokeRehabGuidelines,
title = {Guidelines for Adult Stroke Rehabilitation and Recovery},
author = {Winstein, Carolee J. and others},
journal = {Stroke},
year = {2016},
doi = {10.1161/STR.0000000000000098},
pmid = {27145936},
url = {https://doi.org/10.1161/STR.0000000000000098}
}

@article{Asseyer2025PredictionCPSP_QST,
title = {Prediction of Central Post-Stroke Pain by Quantitative Sensory Testing},
author = {Asseyer, Susanna and others},
journal = {Annals of Neurology},
year = {2025},
doi = {10.1002/ana.27138},
pmid = {39727056},
url = {https://doi.org/10.1002/ana.27138}
}

@article{Truini2023EAN_EFNS_NeuPSIG_NeP_Assessment,
title = {Joint European Academy of Neurology–European Pain Federation–Neuropathic Pain Special Interest Group of the International Association for the Study of Pain guidelines on neuropathic pain assessment},
author = {Truini, Andrea and others},
journal = {European Journal of Neurology},
year = {2023},
doi = {10.1111/ene.15831},
pmid = {37253688},
url = {https://doi.org/10.1111/ene.15831}
}

@article{Zhang2025StrokeInjurySites_PostStrokePain_MetaAnalysis,
title = {Relationship between stroke injury sites and incidence of post-stroke pain: a systematic review and meta-analysis},
author = {Zhang, Zhifa and others},
journal = {Systematic Reviews},
year = {2025},
doi = {10.1186/s13643-025-02930-z},
pmid = {41088428},
url = {https://doi.org/10.1186/s13643-025-02930-z}
}

@article{Berati2025ThalamicCPSP_QST,
title = {Somatosensory Profile of Central Post Stroke Pain of Thalamic Origin: Findings of a Quantitative Sensory Testing Study},
author = {Berati, Kristel and others},
journal = {European Journal of Pain},
year = {2025},
doi = {10.1002/ejp.70104},
pmid = {40814914},
url = {https://doi.org/10.1002/ejp.70104}
}

@article{Colloca2017NeuropathicPain,
title = {Neuropathic pain},
author = {Colloca, Luana and others},
journal = {Nature Reviews Disease Primers},
year = {2017},
doi = {10.1038/nrdp.2017.2},
pmid = {28205574},
url = {https://doi.org/10.1038/nrdp.2017.2}
}

@article{Rosner2023CentralNeuropathicPain,
title = {Central neuropathic pain},
author = {Rosner, Jan and others},
journal = {Nature Reviews Disease Primers},
year = {2023},
doi = {10.1038/s41572-023-00484-9},
pmid = {38129427},
url = {https://doi.org/10.1038/s41572-023-00484-9}
}

Save this as: cpsp_sources_with_urls.txt

1) Winstein CJ, et al. Guidelines for Adult Stroke Rehabilitation and Recovery. Stroke. 2016. PMID:27145936. URL: https://doi.org/10.1161/STR.0000000000000098
2) Asseyer S, et al. Prediction of Central Post-Stroke Pain by Quantitative Sensory Testing. Ann Neurol. 2025. PMID:39727056. URL: https://doi.org/10.1002/ana.27138
3) Truini A, et al. Joint EAN–European Pain Federation–NeuPSIG guidelines on neuropathic pain assessment. Eur J Neurol. 2023. PMID:37253688. URL: https://doi.org/10.1111/ene.15831
4) Zhang Z, et al. Relationship between stroke injury sites and incidence of post-stroke pain: a systematic review and meta-analysis. Syst Rev. 2025. PMID:41088428. URL: https://doi.org/10.1186/s13643-025-02930-z
5) Berati K, et al. Somatosensory Profile of Central Post Stroke Pain of Thalamic Origin: QST study. Eur J Pain. 2025. PMID:40814914. URL: https://doi.org/10.1002/ejp.70104
6) Colloca L, et al. Neuropathic pain. Nat Rev Dis Primers. 2017. PMID:28205574. URL: https://doi.org/10.1038/nrdp.2017.2
7) Rosner J, et al. Central neuropathic pain. Nat Rev Dis Primers. 2023. PMID:38129427. URL: https://doi.org/10.1038/s41572-023-00484-9

Simon

Medications

The AHA/ASA stroke rehabilitation guideline recommends amitriptyline or lamotrigine as reasonable first-line pharmacologic treatments for central post-stroke pain (CPSP), with pregabalin, gabapentin, carbamazepine, or phenytoin as second-line options, individualized to patient response and adverse effects. [1]

Recommended medications for CPSP (ranked by strength of recommendation)

1) First-line (reasonable)

Amitriptyline (TCA) [1]
Evidence basis in guideline: small RCT evidence showing reduced daily pain ratings and improved global functioning; tolerability often limits dose in stroke patients (sedation, anticholinergic, QT risk), so start low and titrate.
Lamotrigine (anticonvulsant; sodium-channel blocker) [1]
Evidence basis in guideline: RCT evidence for reduced daily pain ratings and cold-evoked pain; “good clinical response” in 44% of treated patients in cited trial data summarized in the guideline [1].

2) Second-line (may be considered)

Pregabalin (α2δ ligand) [1]
Evidence summary: trials show mixed analgesic efficacy; may improve sleep/anxiety even when pain endpoints are neutral (as summarized in guideline) [1]. Network meta-analysis including 13 RCTs, n=529 found pregabalin superior to placebo for pain (SMD −0.46, 95% CI −0.71 to −0.22) [2] (relevant because CPSP-only RCT synthesis).
Gabapentin (α2δ ligand) [1]
Evidence summary: not well studied specifically in CPSP in the guideline, but effective in other neuropathic pain conditions (rationale for consideration) [1]. A 2022 CPSP drug network meta-analysis (antidepressants/anticonvulsants) ranked gabapentin highly on VAS outcomes across 13 RCTs, n=1040 [3] (relevant as CPSP-focused comparative synthesis).
Carbamazepine [1]
Phenytoin [1]
Evidence summary: guideline notes usefulness is not well established (consider mainly when first-line options fail/contraindicated).

Options supported by CPSP-focused evidence but not listed as first/second-line in AHA/ASA (use selectively)

Duloxetine (SNRI): 2024 systematic review/meta-analysis (CPSP interventions) states duloxetine among the “most robust evidence” pharmacologic options [4] (relevant as most up-to-date CPSP-specific evidence synthesis).
Levetiracetam, prednisone, pamidronate, etanercept showed benefit vs placebo in a CPSP pharmacotherapy network meta-analysis (13 RCTs, n=529; e.g., prednisone SMD −2.38 [95% CI −3.09 to −1.67], levetiracetam SMD −2.11 [95% CI −2.97 to −1.26]) [2]. These findings are hypothesis-generating because the evidence base is small/heterogeneous and not reflected in major stroke rehab guideline recommendations.

Where general neuropathic-pain guidelines fit (if treating CPSP as central neuropathic pain)

NeuPSIG-aligned meta-analysis supports TCAs, SNRIs, and α2δ ligands as first-line for neuropathic pain overall (not CPSP-specific), with opioids only later-line due to limited benefit/harms [5]

Sources

I can’t make a permanent memory note that applies to all future chats, but for the rest of this conversation I will add a final section titled “Sources (with URLs)” to each answer.

Downloadable file (sources + URLs)

Save the following as cpsp_med_sources_with_urls.txt:

CPSP medications answer — sources with URLs

1) Winstein CJ, et al. Guidelines for Adult Stroke Rehabilitation and Recovery. Stroke. 2016.
DOI: 10.1161/STR.0000000000000098
URL: https://doi.org/10.1161/STR.0000000000000098

2) Pharmacotherapy network meta-analysis (CPSP) — cited for pregabalin/prednisone/levetiracetam effects.
DOI: 10.1155/2022/3511385
URL: https://doi.org/10.1155/2022/3511385

3) PLOS ONE CPSP drug network meta-analysis — cited for gabapentin ranking on VAS outcomes.
DOI: 10.1371/journal.pone.0276012
URL: https://doi.org/10.1371/journal.pone.0276012

4) Systematic review/meta-analysis of CPSP interventions — cited for duloxetine as having relatively robust evidence.
DOI: 10.1016/j.jpain.2024.104666
URL: https://doi.org/10.1016/j.jpain.2024.104666

5) Neuropathic pain guideline/meta-analysis referenced as general neuropathic-pain evidence framework (not CPSP-specific).
DOI: 10.1016/S1474-4422(25)00068-7
URL: https://doi.org/10.1016/S1474-4422(25)00068-7

Scroll-Me :)\<-&->


  All Discussions  |  

  Welcome Post  |  

  SIG's Charter  |  

  Direct Messages  |   

  Categories  |  

  How-To(Forum)