This study evaluates the acute effects of cervical manual therapy on retinal and choroidal microvascular perfusion using optical coherence tomography angiography (OCTA). In this prospective randomized sham-controlled trial, 100 patients with chronic mechanical neck pain were randomly assigned to receive either cervical high-velocity low-amplitude manipulation combined with Maitland mobilization (manual therapy group) or sham treatment. Bilateral OCTA examinations were performed at baseline and at 15, 30, 60, and 120 minutes after intervention. The primary outcome is superficial capillary plexus (SCP) foveal vessel density. Secondary outcomes include deep capillary plexus (DCP) vessel density, radial peripapillary capillary (RPC) density, subfoveal choroidal thickness, foveal avascular zone parameters, retinal nerve fiber layer thickness, ganglion cell complex thickness, and macular thickness. The study aims to determine whether cervical manual therapy produces a significant augmentation of retinal and choroidal perfusion compared with sham treatment, with peak effects expected at 60 minutes post-intervention.

Background and Rationale

Cervical musculoskeletal disorders, including mechanical neck pain and cervicogenic headache, affect an estimated 30-50% of the general population. Manual therapy encompassing spinal manipulation and mobilization techniques targeting the cervical spine has demonstrated efficacy in reducing pain intensity and improving functional outcomes. However, beyond its established biomechanical and neurophysiological effects, growing interest has emerged in elucidating the potential vascular and autonomic consequences of cervical manual therapy, particularly with respect to ocular circulation.

The eye is uniquely positioned as a window into systemic and regional vascular status. The retinal and choroidal vasculature are regulated by a complex interplay of local autoregulatory mechanisms, systemic blood pressure, intraocular pressure, and autonomic nervous system tone. The cervical sympathetic chain, lying in close anatomical proximity to the cervical vertebrae and the carotid artery, provides preganglionic and postganglionic sympathetic innervation to the ocular structures. Mechanical stimulation of the cervical spine may therefore modulate ocular perfusion through sympathetic pathways, alterations in vertebral artery blood flow, or changes in intracranial pressure dynamics.

Optical coherence tomography angiography (OCTA) has emerged as a non-invasive, high-resolution imaging modality capable of quantifying retinal and choroidal microvasculature with unprecedented detail. Quantitative metrics derived from OCTA, including vessel density, foveal avascular zone area, and choroidal thickness, provide sensitive and reproducible indices of microvascular status.

To date, the effects of cervical manual therapy on ocular circulation have been investigated only sporadically and with methodological limitations, including small sample sizes, absence of bilateral assessment, lack of temporal resolution, and absence of sham or no-treatment control arms. No prior study has employed OCTA to comprehensively evaluate the acute, time-resolved effects of cervical manual therapy across multiple retinal and choroidal vascular compartments in both eyes simultaneously using a sham-controlled design.

Objectives

The primary objective is to evaluate the acute effects of a single session of cervical manual therapy on retinal and choroidal microvascular perfusion as measured by OCTA in patients with chronic mechanical neck pain.

The secondary objectives are:

To assess changes in deep capillary plexus vessel density

To assess changes in radial peripapillary capillary density

To assess changes in subfoveal choroidal thickness

To assess changes in foveal avascular zone parameters

To assess changes in retinal nerve fiber layer thickness

To assess changes in ganglion cell complex thickness

To assess changes in macular thickness

To evaluate the temporal profile of vascular responses (at 15, 30, 60, and 120 minutes post-intervention)

To compare vascular responses between active manual therapy and sham treatment

Hypothesis

Cervical manual therapy will produce a significantly greater augmentation of retinal and choroidal perfusion compared with sham treatment, with peak between-group differences at 60 minutes post-intervention, followed by convergence toward baseline by 120 minutes.

Study Design

This is a prospective, single-center, parallel-group randomized sham-controlled trial. Participants are randomly assigned (1:1) to receive either cervical high-velocity low-amplitude manipulation combined with Maitland mobilization (manual therapy group) or sham treatment. Bilateral OCTA examinations are performed at baseline and at 15, 30, 60, and 120 minutes after intervention.

Setting

The study is conducted at SBU Van Training and Research Hospital, Van, Turkey.

Participant Timeline

Screening and enrollment: February 2026

Intervention and follow-up: Single session with 120-minute post-intervention observation

Final data collection: May 2026

Sample Size

Based on published OCTA test-retest data (intraday coefficient of variation ~2.5% for SCP foveal density; within-subject standard deviation ~1.2 percentage points), a minimum detectable between-group difference of ≥2 percentage points, α=0.05 (two-tailed), and 1-β=0.80, a minimum of 42 participants per group is required. The enrolled sample of 50 per group provides ≥85% power and accommodates an estimated 15% dropout rate.

Interventions

Manual therapy group: A single, standardized session consisting of cervical HVLA thrust manipulation targeting C1-C2 and C3-C4 bilaterally, supplemented by sustained Maitland grade III-IV mobilization of the mid-cervical spine. Total session duration: 15-20 minutes. Audible cavitation is documented as a quality indicator.

Sham control group: Light, non-therapeutic manual contact at the same anatomical sites without joint movement, traction, or thrust (<5 N contact force). Session duration identical to active intervention.

Both groups remain seated throughout the observation window to control for postural effects on choroidal thickness.

Outcome Measures

Primary outcome: Superficial capillary plexus foveal vessel density (percentage points)

Secondary outcomes:

Deep capillary plexus foveal and parafoveal vessel density Radial peripapillary capillary optic disc density Subfoveal choroidal thickness (μm) Foveal avascular zone area, perimeter, and circularity Retinal nerve fiber layer global thickness Ganglion cell complex thickness Macular center thickness Signal strength index Data Collection OCTA imaging is performed with a swept-source device (DRI OCT Triton, Topcon, Tokyo, Japan) at T0, T15, T30, T60, and T120 minutes post-intervention by the same blinded technician. Protocols include 6×6 mm foveal-centered and 4.5×4.5 mm optic disc-centered scans. Only images with SSI ≥7 are included.

To mitigate repeated-scan artifacts, the same automated segmentation algorithm is applied uniformly across all time points, and all segmentations are manually verified by a masked grader. Participants undergo a 2-minute dark adaptation prior to each scan, and the built-in eye-tracking system is activated. Scan registration across time points is performed using the device's automated follow-up mode.

Statistical Analysis Statistical analyses use Python 3.11 (SciPy 1.11, pandas 2.1, pingouin 0.5.4, statsmodels 0.14). Normality is assessed using the Shapiro-Wilk test. Within-group temporal changes are analyzed using the Friedman test; post-hoc Wilcoxon signed-rank tests with Bonferroni correction are applied. Between-group differences are analyzed using linear mixed-effects models with random intercept for subject and random slope for time, including fixed effects for group, time, and group×time interaction. Benjamini-Hochberg FDR correction is applied across all secondary outcomes.

Analysis is performed on both intention-to-treat (ITT) and per-protocol (PP) basis.

Adverse Event Monitoring All participants are monitored for adverse events throughout the study period (from randomization through 24 hours post-intervention). Participants are asked at T15, T30, T60, T120, and via telephone at 24 hours about the occurrence of: neck pain worsening, dizziness, headache, nausea, visual symptoms (blurred vision, floaters, photopsia), or any other unusual symptom. All events are recorded and graded for severity and relationship to intervention.

Inclusion Criteria:

• Age 18-70 years
• Chronic mechanical neck pain lasting ≥3 months according to IASP criteria
• Baseline VAS pain score ≥3
• Absence of cervical radiculopathy or myelopathy
• Ability to provide informed consent and comply with study procedures
• Adequate ocular image quality for OCTA assessment

Exclusion Criteria:

• Previous cervical spine surgery or major cervical trauma
• Vertebrobasilar insufficiency or other contraindications to cervical manual therapy
• Active inflammatory rheumatic disease or systemic connective tissue disorder
• Known retinal or optic nerve disease that may affect OCTA measurements (glaucoma, diabetic retinopathy, age-related macular degeneration, optic neuritis, etc.)
• Diabetes mellitus with end-organ complications or uncontrolled systemic hypertension (>160/100 mmHg)
• Refractive error > ±6.0 D spherical equivalent or astigmatism >3.0 D
• Intraocular pressure >21 mmHg
• Previous ocular surgery, ocular trauma, or retinal laser treatment
• Media opacity preventing adequate OCTA image acquisition
• Pregnancy or lactation
• Participation in another interventional clinical study within the previous 30 days

All individual participant data (IPD) collected during the trial, including demographic characteristics, OCTA measurements (superficial and deep capillary plexus vessel density, radial peripapillary capillary density, subfoveal choroidal thickness, foveal avascular zone parameters, retinal nerve fiber layer thickness, ganglion cell complex thickness, and macular center thickness) collected at baseline and at 15, 30, 60, and 120 minutes post-intervention, de-identified.

Data are available immediately upon publication (June 21, 2026) and will remain accessible indefinitely in the Zenodo repository.

The datasets are publicly accessible without restrictions. Anyone can access the de-identified individual participant data and supporting documents via the Zenodo repository without requiring additional permissions or data use agreements. The analysis scripts are available from the corresponding author upon reasonable request.

The treating physiotherapist could not be blinded. Participants were blinded to group allocation. OCTA acquisition was performed by a technician blinded to group assignment. Image analysis was performed by an investigator masked to both group allocation and temporal sequence using randomized alphanumeric file codes.