Supine-induced Increase in Blood Flow Velocity in the Anterior Humeral Circumflex Artery Associates with Nocturnal Sitting-relief Shoulder Pain
2025 Volume 10 Article ID: 20250031
Details
2025 Volume 10 Article ID: 20250031
Objectives : Nocturnal pain is common in rotator cuff tears (RCTs). Among its patterns, a pain type relieved by sitting upright, referred to here as “sitting-relief pain,” appears to behave differently from other nocturnal pain types. Peak systolic velocity in the anterior humeral circumflex artery (PSV-AHCA) has been suggested as a marker, but findings have varied across settings. Exploring this vascular response may help connect patients’ subjective pain experiences with objective physiological data, potentially offering a bridge between symptoms and measurable signs. This study investigated the association between PSV-AHCA and nocturnal pain characteristics in patients with RCTs.
Methods : A cross-sectional study was conducted with 78 patients with RCTs. Participants were classified into three groups based on nocturnal pain characteristics: Sitting-Relief-Pain, Other-Pain, and Pain-Free. PSV-AHCA was measured in both sitting and supine positions using Doppler ultrasound. After confirming the absence of major confounders, comparisons were made across groups and positions.
Results : The Sitting-Relief-Pain group showed an increase in PSV-AHCA from sitting to supine (22.7 to 26.7 cm/s; d=0.712; P=0.001), whereas the Other-Pain and Pain-Free groups showed no significant changes. In the supine position, PSV-AHCA was higher in the Sitting-Relief-Pain group than in the Other-Pain (P=0.008) and Pain-Free (P=0.004) groups.
Conclusions : Sitting-relief pain may represent a specific pattern of nocturnal pain in RCTs, with a distinct vascular response to posture. Doppler ultrasound in physiotherapist-led assessment may assist in identifying this pattern and distinguishing it from other types of nocturnal pain.
Rotator cuff tears (RCTs) are a common shoulder disorder, with nocturnal pain being a major symptom for many patients.1,2,3) This pain severely disrupts patients’ quality of life and affects their ability to sleep, which often leads to increased reliance on pain management strategies.1) Previous studies have reported that 70% of patients experience nocturnal pain,4) and 89% suffer from sleep disturbances.1) Understanding the mechanisms responsible for nocturnal pain is important for improving clinical management.
Recent studies have increasingly used ultrasonography to explore the relationship between nocturnal pain and vascular changes.4,5,6,7,8) The anterior humeral circumflex artery (AHCA) has been identified as a key vessel reflecting synovial inflammation,6,7) and its peak systolic velocity (PSV) has been suggested as a potential indicator of nocturnal pain.4,5) However, the association between PSV in the AHCA (PSV-AHCA) and nocturnal pain remains controversial. Some studies have shown a correlation between increased PSV and nocturnal pain,4,5) whereas others have found no significant relationship.6,8) Notably, the former studies generally define nocturnal pain as pain that disturbs sleep causing the patient to want to sit up,4,5) whereas the latter studies often lack a clear definition of nocturnal pain.6,8) This discrepancy suggests that nocturnal pain may not be a uniform symptom and that its classification could influence study outcomes. Indeed, the authors of the latter studies themselves speculated that the absence of a clear definition of nocturnal pain might explain the discrepancy.6,8)
Despite increasing evidence associating vascular changes with nocturnal pain, prior studies have only measured PSV in a sitting position.4,5,6,7,8) Given that some patients experience pain in the supine position that improves upon sitting, assessing blood flow velocity in both positions is essential. Positional factors may influence vascular responses,9,10) and comparing PSV measurements in sitting and supine positions may provide novel insights into the pathophysiology of nocturnal pain in RCTs.
This study aimed to investigate the relationship between PSV-AHCA and nocturnal pain characteristics in patients with RCTs. Unlike previous studies, we specifically differentiated between (1) nocturnal pain that simply occurs from lying down and is relieved by sitting up and (2) other types of nocturnal pain, such as pain triggered by movement during sleep. We hypothesized that (1) patients whose nocturnal pain improves with sitting up would show higher PSV-AHCA than those with other types of nocturnal pain, and (2) in these patients, PSV would increase in the supine position relative to the sitting position, whereas no significant positional differences would be observed in other groups. By incorporating positional PSV measurements and refining the classification of nocturnal pain, this study aims to clarify the role of vascular changes in RCT-related nocturnal pain and contribute to more effective clinical management strategies.
This cross-sectional study was approved by the institutional ethics committee of Machida Orthopaedics (23–001). Written informed consent was obtained from all participants before the study, and the study was conducted in accordance with the Declaration of Helsinki.
Definition and Classification of Nocturnal PainNocturnal pain was classified into three distinct groups: the Sitting-Relief-Pain group, the Other-Pain group, and the Pain-Free group. The Sitting-Relief-Pain group was defined by nocturnal pain that disturbs sleep causing the patient to want to sit up. This definition is consistent with a recent study4) and was determined through a careful interview process. The Other-Pain group was defined by nocturnal pain that disturbs sleep but does not occur simply by lying down. Instead, it arises because of specific actions, such as unexpected movements of the affected limb. The specific triggers of pain were identified through detailed questioning during the interview process. The Pain-Free group refers to participants who did not experience any nocturnal pain that disturbed sleep.
ParticipantsThe calculation of the appropriate sample size was based on the difference in PSV-AHCA between the groups with and without nocturnal pain, as shown in a previous study.4) Based on this, the required sample size for each group was calculated to be 22 (delta=12.3, standard deviation=16.0, alpha=0.05, power=0.80).4) Considering the possibility that up to 15% of participants might transition between groups because of the nature of the study design, we allowed for this potential group migration by increasing the sample size. As a result, the target sample size was set at 26 for each group to ensure that at least 22 participants would remain in each group for valid comparisons. Recruitment was conducted progressively, with participants added to each group until the target of 26 cases was reached. Once a group had reached 26 participants, recruitment for that group was closed, and recruitment continued for the other groups. To evaluate the test–retest reliability of these nocturnal pain classifications, follow-up interviews were conducted at least 3 days after the initial interview to detect any changes in the classification. These interviews were conducted by medical staff blinded to the ultrasound findings described below.
A total of 109 consecutive outpatients with RCTs who visited primary care clinics in Japan between March 2023 and October 2024 were initially considered for inclusion in the study. The inclusion criteria were: (1) a diagnosis of RCT by magnetic resonance imaging, (2) aged over 40 years, and (3) shoulder dysfunction lasting more than 2 months. The exclusion criteria were: (1) a history of previous shoulder surgery, (2) corticosteroid injection within the past 6 months, (3) unclear onset of nocturnal pain, (4) the presence of other conditions preventing sleep, and (5) patients from groups where recruitment had already closed. After applying the exclusion criteria (previous surgery, n=4; corticosteroid injection, n=0; unclear onset of nocturnal pain, n=11; other conditions preventing sleep, n=12; recruitment closed for relevant groups, n=4), 78 patients were included in the final analysis. Detailed information on these participants is shown in Fig. 1.
Flowchart of patient selection.
On the same day as the first nocturnal pain interview, PSV-AHCA was measured using ultrasonography (3–11 MHz linear probe, SONIMAGE HS2, Konica Minolta, Tokyo, Japan). This method demonstrates high reliability, as evidenced by our previous studies,5,7,8) as well as those of other research groups4,6) (intraobserver reliability, 0.983–0.996; interobserver reliability, 0.949–0.985). All examinations were conducted by a single examiner blinded to the results of the nocturnal pain interview. Participants were first asked to relax in a sitting position for 5 min, followed by PSV measurement in the sitting position. Following this, they were asked to lie down in a relaxed supine position for 5 min, after which PSV was measured again in the supine position. For the two groups experiencing nocturnal pain, we also recorded whether any pain occurred during the 5-min rest period in the supine position before PSV measurement. The testing procedure was as follows: the shoulder joint was abducted to 30 degrees and rotated to 0 degrees, the elbow joint was flexed to 90 degrees, and the hand was placed in a palm-up position. The bicipital groove and AHCA were first identified using transverse scanning, then the blood flow velocity in the AHCA was measured by longitudinal scanning. PSV was then calculated using pulse Doppler ultrasonography.
Assessment of Clinical ParametersAvailable clinical parameters included age, sex, duration of symptoms, diabetes, partial or full-thickness tear, the intensity of nocturnal pain (measured using the visual analog scale, 0=no pain; 100=worst pain imaginable), the Shoulder Pain and Disability Index (SPADI), and PSV-AHCA in the sitting and supine positions. RCTs were confirmed by magnetic resonance imaging and verified by a shoulder-specialized surgeon with more than 20 years of clinical experience. Partial-thickness and full-thickness tears were recorded; partial tears were not further subdivided by location. Full-thickness tears were graded according to Cofield et al.11) as small, medium, large, or massive. Furthermore, the use of analgesics was investigated as a potential confounding factor.
Statistical AnalysesMultivariate analysis was conducted to examine the factors affecting PSV-AHCA among the groups. Independent variables included factors associated with nocturnal pain and/or synovitis in patients with RCTs, including diabetes,8,12) tear type,13) nocturnal pain intensity,8) and SPADI,14) as well as nocturnal pain classification. The normal distribution of PSV was tested with the Shapiro–Wilk test.
After controlling for significant independent variables from the multivariate analysis, an analysis of variance (ANOVA) was conducted to examine PSV in AHCA. Tukey’s test was used for post hoc pairwise comparisons, and paired t-tests were used to assess differences between the sitting and supine positions. The effect size (Cohen’s d) was also calculated.15)
A statistical significance level of 5% was adopted for all analyses. All statistical procedures were performed with R version 3.4.1 (R Foundation for Statistical Computing).
Table 1 shows the clinical characteristics of all 78 participants, as well as those of the three groups, each consisting of 26 participants. The overall mean age was 68.4 ± 8.5 years, with 44 (56%) men and 34 (44%) women. Mean PSV-AHCA was 19.4 ± 7.7 cm/s in the sitting position and 22.7 ± 7.2 cm/s in the supine position. None of the participants had used non-steroidal anti-inflammatory drugs within 24 h of PSV measurement.
| Characteristic | Total(n=78) | Sitting-Relief-Pain group(n=26) | Other-Pain group(n=26) | Pain-Free group (n=26) |
| Age, years | 68.4 ± 8.5 | 68.6 ± 8.3 | 68.1 ± 8.4 | 68.6 ± 8.7 |
| Female sex | 34 (44%) | 10 (38%) | 14 (54%) | 10 (38%) |
| Duration of symptoms, months | 5 [2–7.8] | 5.5 [4–7.3] | 4 [2–13] | 4.8 [2–7.8] |
| Diabetes | 13 (17%) | 5 (19%) | 6 (23%) | 2 (8%) |
| Partial tear | 19 | 4 | 8 | 7 |
| Full-thickness tear | 59 | 22 | 18 | 19 |
| Cofield | ||||
| Small (≤1 cm) | 13 | 4 | 4 | 5 |
| Medium (1–3 cm) | 12 | 5 | 4 | 3 |
| Large (3–5 cm) | 20 | 7 | 6 | 7 |
| Massive (>5 cm) | 14 | 6 | 4 | 4 |
| Nocturnal pain intensity (VAS) | 30.2 ± 29.2 | 51.1 ± 32.3 | 39.4 ± 8.8 | 0 |
| SPADI | 31.8 ± 20.2 | 28.1 ± 16.0 | 31.0 ± 21.2 | 36.3 ± 21.9 |
| PSV in AHCA, cm/s | ||||
| Sitting | 19.4 ± 7.7 | 22.7 ± 6.2 | 18.1 ± 9.0 | 17.4 ± 6.3 |
| Supine | 22.7 ± 7.2 | 26.7 ± 7.5 | 20.9 ± 6.6 | 20.5 ± 5.8 |
Data are presented as mean ± standard deviation, number (percentage), median [interquartile range], or number.
No changes were observed in the nocturnal pain classification between the initial and follow-up interviews. In the Sitting-Relief-Pain group (n=26), four participants (15%) experienced shoulder pain similar to their usual nocturnal pain after 5 min of rest in the supine position but reported being pain-free within 10 min of sitting up. In contrast, no participants in the Other-Pain group reported pain onset during the 5-min supine rest period. The causes of nocturnal pain in the Other-Pain group were pain from unexpected limb movements (n=12), turning over (n=7), and sleeping on the affected side (n=7).
Table 2 shows the results of the multivariate analyses for PSV-AHCA in both the sitting and supine positions. Nocturnal pain classification was the only significant factor identified in both analyses (sitting, P=0.006; supine position, P=0.008). Diabetes, tear type, pain intensity, and SPADI scores were not significant predictors of PSV.
| Factor | Sitting | Supine | |||
| t value | P value | t value | P value | ||
| Diabetes | 0.660 | 0.512 | 0.582 | 0.562 | |
| Tear type | 0.641 | 0.524 | 0.731 | 0.468 | |
| Pain intensity (VAS) | 0.578 | 0.566 | 0.067 | 0.947 | |
| SPADI | 0.301 | 0.764 | 0.488 | 0.627 | |
| Nocturnal pain classification | 2.818 | 0.006** | 2.712 | 0.008** | |
Nocturnal pain classification includes the Sitting-Relief-Pain group, the Other-Pain group, and the Pain-Free group.
**P <0.01.
Table 3 shows the results of the ANOVA, which included nocturnal pain classification and position as the independent variables. Both main effects, nocturnal pain classification (P <0.001) and position (P=0.043), as well as the interaction term “nocturnal pain classification * position” (P=0.001), were statistically significant.
| Df | SS | MS | F value | P value | |
| Nocturnal pain classification | 2 | 616.3 | 308.1 | 8.11 | <0.001*** |
| Position | 1 | 161.8 | 161.8 | 4.26 | 0.043* |
| Nocturnal pain classification * position | 2 | 555.2 | 277.6 | 7.31 | 0.001** |
| Residuals | 72 | 2736.2 |
Df, degrees of freedom; SS, sum of squares; MS, mean square (SS divided by Df); F = (MS of the factor)/ (MS of the residual).
Nocturnal pain classification includes the Sitting-Relief-Pain group, the Other-Pain group, and the Pain-Free group. Position includes Sitting position and Supine position. Nocturnal pain classification * position is presented as an interaction of nocturnal pain classification and position.
*P <0.05, **P <0.01, ***P <0.001.
Table 4 and Fig. 2 show comparisons of PSV-AHCA across groups and positions. No significant factors affecting PSV were identified, and the comparison was made without covariate adjustment. In the sitting position, the Sitting-Relief-Pain group had significantly higher PSV than the Pain-Free group (P=0.029). In the supine position, the Sitting-Relief-Pain group had significantly higher PSV than the Other-Pain group (P=0.008) and the Pain-Free group (P=0.004). In addition, the Sitting-Relief-Pain group showed significantly higher PSV in the supine position than in the sitting position (P=0.001), whereas no significant differences were found between positions for the Other-Pain (P=0.202) or Pain-Free (P=0.090) groups.
| PSV in AHCA | Comparison of groups | |||||||
| Sitting-Relief-Pain | Other-Pain | Pain-Free | Difference (95% CI) | Effect size d | P value | |||
| Sitting | 22.7 ± 6.2 | 18.1 ± 9.0 | 17.4 ± 6.3 | SRP vs. OP | 4.67 (−0.26 to 9.60) | 0.595 | 0.068 | |
| SRP vs. PF | 5.38 (0.45 to 10.3) | 0.848 | 0.029* | |||||
| OP vs. PF | 0.72 (−4.21 to 5.65) | 0.090 | 0.936 | |||||
| Supine | 26.7 ± 7.5 | 20.9 ± 6.6 | 20.5 ± 5.8 | SRP vs. OP | 5.77 (1.27 to 10.3) | 0.821 | 0.008** | |
| SRP vs. PF | 6.14 (1.64 to 10.6) | 0.925 | 0.004** | |||||
| OP vs. PF | 0.37 (−4.12 to 4.88) | 0.064 | 0.978 | |||||
| Comparison of position | Sitting vs. supine | Sitting vs. supine | Sitting vs. supine | |||||
| Difference (95% CI) | 3.95 (1.68 to 6.21) | 2.85 (−1.62 to 7.31) | 3.18 (−0.53 to 6.90) | |||||
| Effect size d | 0.712 | 0.355 | 0.562 | |||||
| P value | 0.001** | 0.202 | 0.090 | |||||
PSV given as mean ± standard deviation (cm/s).
SRP, Sitting-Relief-Pain group; OP, Other-Pain group; PF Pain-Free group; 95% CI; 95% confidence interval.
*P <0.05, **P <0.01.
Comparison of PSV in the AHCA in the sitting and supine positions. Closed circles, Sitting-Relief-Pain group; open circles, Other-Pain group; open squares, Pain-Free group. Comparison of groups, *P <0.05, **P <0.01; comparison of positions, ††P <0.01.
The most important findings of this study are as follows: (1) the Sitting-Relief-Pain group showed higher PSV-AHCA than the Other-Pain group and the Pain-Free group, and (2) only the Sitting-Relief-Pain group showed an increase in PSV-AHCA when transitioning from the sitting to the supine position. These results support our hypotheses and suggest that PSV-AHCA may play an important role in specific types of nocturnal pain.
First, the Sitting-Relief-Pain group showed higher PSV-AHCA than the other groups. Previous studies linking nocturnal pain to PSV-AHCA have observed this in cases of sitting-relief pain.4,5) In contrast, studies that found no correlation often lacked a clear definition of nocturnal pain.6,8) These reports support our current findings suggesting that PSV could be an important marker of specific nocturnal pain. Furthermore, several studies have linked such nocturnal pain to synovitis. A histologic study found a direct association between PSV-AHCA and inflammatory biomarkers,6) whereas a macro-observation study suggested that PSV more accurately reflects synovitis severity than Doppler activity.7)
In contrast, the Other-Pain group, which included pain from unexpected limb movements or rolling over, showed lower PSV, like the Pain-Free group. This suggests that the pain mechanisms in the Other-Pain group differ from those associated with synovitis. It is plausible that these actions cause a transient elevation in subacromial pressure, thereby eliciting impingement-like pain. This observation aligns with prior data on position-dependent pressure16) and our finding that a supine-induced PSV rise was absent in this group.
The second important finding was that only the Sitting-Relief-Pain group showed an increase in PSV-AHCA when transitioning from the sitting position to the supine position. This suggests that the Sitting-Relief-Pain group is particularly sensitive to positional changes. When transitioning to a supine position, the upper limbs are relatively lowered, potentially altering blood flow dynamics,9,10) which could explain the observed PSV increase. This finding is novel, because previous studies have measured PSV only in the sitting position.4,5,6,7,8) Given that nocturnal pain typically occurs in the supine position, measuring PSV in this position provides valuable insight into the mechanisms underlying nocturnal pain.
This finding raises the question of how supine-induced PSV changes may contribute to shoulder pain. Although our cross-sectional data cannot establish mechanisms, we speculate that synovial inflammation may increase peripheral sensitivity17,18,19,20) such that posture-related hemodynamic changes in the supine position are sufficient to provoke pain in the Sitting-Relief-Pain group.21,22) This remains a hypothesis and requires dedicated mechanistic studies.
Our findings highlight the importance of distinguishing different types of nocturnal pain rather than treating it as a single entity. Clinically, detailed assessment of pain characteristics is crucial for appropriate management. Similarly, in research, failing to classify nocturnal pain subtypes may introduce bias, because study outcomes could be influenced by the proportional distribution of different pain types within a given cohort. Future studies should consider stratifying patients based on nocturnal pain characteristics to enhance the accuracy and applicability of their findings.
Our study has important clinical implications. In patients with nocturnal pain, a detailed assessment of pain onset, combined with measurements of PSV-AHCA, can help guide appropriate pain management strategies. If nocturnal pain is classified as Sitting-Relief-Pain with elevated PSV, synovitis is likely the cause. In such cases, anti-inflammatory interventions, such as corticosteroid injections,23) along with guidance to sleep in a slightly elevated position, would be beneficial. In addition, because this type of nocturnal pain correlates with a decrease in PSV over time,5) it may also provide a useful tool for monitoring pain progression.
This study has several strengths. First, it relied on noninvasive testing, thereby reducing patient burden and increasing clinical applicability. Although previous studies have examined synovitis by arthroscopic observation and its correlation with clinical outcomes,24,25,26) our approach allows broader coverage of patient populations that do not require surgery. Second, the study employed a strictly designed sample size to minimize Type I errors. Third, the participants were consecutive cases from primary care settings, thereby reducing selection bias and making the findings highly representative of the disease population. Fourth, we excluded younger patients with RCT, those with acute or transient symptoms, and those unable to clearly recognize the timing and context of their nocturnal pain. These criteria helped minimize variability caused by differences in pathophysiology27) and eliminated transient symptoms, thereby maintaining the homogeneity and interpretability of the findings. Finally, multivariate analysis helped control for potential confounders, showing that factors such as diabetes and tear type do not significantly affect PSV, which is valuable for clinicians explaining PSV to patients without adjustments for these factors.
Despite these strengths, our study has several limitations. First, because it was conducted at a primary care facility, the severity of patient cases may differ from those in other research settings,28,29) and therefore, generalization of our findings should be made with caution. Second, we only measured ultrasonographic findings and did not assess factors related to central sensitization or psychosocial factors.30,31) Future studies incorporating a broader range of factors contributing to nocturnal pain will be needed. Finally, the mechanisms proposed for nocturnal pain in both the Sitting-Relief-Pain and Other-Pain groups should be regarded as hypothetical; mechanistic validation (biomarkers or experimental provocation) was beyond the scope of this study and warrants future research.
The findings of this study demonstrated that the Sitting-Relief-Pain group had significantly higher PSV-AHCA than the Other-Pain group and the Pain-Free group, and PSV increased when the patient transitioned from a sitting position to a supine position. Our results suggest that PSV-AHCA serves as a key marker for nocturnal pain that worsens in the supine position and is relieved by sitting up. This distinction highlights the importance of assessing nocturnal pain patterns in clinical practice. By incorporating PSV measurements and considering the characteristics of nocturnal pain, clinicians may better understand its underlying mechanisms and develop more-targeted treatment strategies for patients with RCTs.
The authors thank the participants for their contribution to this study.
The authors declare no conflict of interest.
