Forearm Volume Changes Estimated by Photo-Plethysmography During an Original Candlestick/Prayer Maneuver in Patients With Suspected Thoracic Outlet Syndrome

Abstract

Objective: Hemodynamic investigations in thoracic outlet syndrome (TOS) remain difficult, even in trained hands. Results are generally reported as either presence or absence of venous compression. In fact, in patients with suspected TOS but without chronic venous occlusion, the forearm volume changes may result from various combinations of forearm position from heart level, arterial inflow, and/or venous outflow positional impairment. Design: Cross sectional, retrospective, single center study, accessible on Clinicaltrial.gov under reference NCT04376177. Material: We used venous photo-plethysmography (V-PPG) in 151 patients with suspected TOS. The subjects elevated their arms to the "candlestick" (Ca) position for 30 s and then kept their arm elevated in front of the body for an additional 15 s ("prayer" position; Pra). This CA-Pra procedure was repeated three times by each patient with recording of both arms. Method: We classified V-PPG recordings using an automatic clustering method. Result: The blinded clustering classification of 893 V-PPG recordings (13 missing files) resulted in four out of seven clusters, allowing the classification of more than 99% of the available recordings. Each cluster included 65.73, 6.16, 17.13, and 10.8% of the recordings, respectively. Conclusion: Venous hemodynamic profiles in TOS are not only either normal or abnormal. With V-PPG, four clusters were observed to be consistent with, and assumed to result from, the four possible associations of presence/absence of arterial inflow/venous outflow positional impairment: (1) normal response (maximal emptying in Ca and Pra), (2) isolated inflow impairment (emptying in Ca and filling in Pra due to post-ischemic vasodilation), (3) isolated venous outflow impairment (emptying then filling in Ca due to arterial inflow and emptying in Pra), and (4) simultaneous inflow/outflow impairment (emptying in Ca but no filling due to concomitant inflow impairment and further emptying in Pra).

Keywords: classification; hemodynamic response; movement; peripheral vessel; plethysmography; thoracic outlet syndrome; upper extremities.

Figure 1

Procedures of arm positioning used in the present study.

Figure 2

Bland–Altman representation of the absolute peak values observed on each arm for the 151 patients studied during photo-plethysmography recordings.

Figure 3

Representation of the different steps of the clustering analysis. As shown, the analysis resulted in seven clusters, four of which allowed the classification of more than 99% of recordings and appeared of clinical interest. For these four clusters, the representation is the interval 25° and 75° centile of the different values (gray zone); the other three are mean values.

Figure 4

Relevance of the four clusters (α, β, δ, and ε) obtained as assessed by the observation of the Silhouette coefficient on the records.

Figure 5

Schematic representations of the venous photo-plethysmography pattern interpretation, depending on the presence of inflow and/or outflow impairment (emptying is an upward change, while filling is a downward change). Differences from the assumed normal pattern (blue dot points) are underlined by hash bars and arrows. Isolated inflow impairment during the candlestick phase is assumed to result in post-ischemic vasodilation. The green figure is expected to result from arrested inflow and outflow during the candlestick that are normalized during the prayer position. Outflow impairment with persistent inflow results in forearm filling, followed by emptying at the prayer position (yellow figure). The lowering was not standardized and might be different in each pattern and thus presented as dash lines.