Neurovascular Reactivity in Smokers and Nonsmokers Measured by High-Speed MR Flow Mapping During Volitional Apnea
Felix W Wehrli1, Yongxia Zhou1, Zachary B Rodgers1, and Michael C Langham1

1Department of Radiology, University of Pennsylvania, Philadelphia, PA, United States


Smoking is well known to cause vasoconstriction as a result of the formation of reactive oxygen species, which reduce nitric oxide availability. Vasomotor reactivity can be measured in terms of the change in cerebral blood flow in response to a hypercapnic challenge. Here we measured the change in superior sagittal sinus flow at 2-second temporal resolution during breath-hold (a predominantly hypercapnic stimulus) and computed a breath-hold index (BHI) as the slope of the flow velocity-time curve in 20 nonsmokers and 13 chronic smokers. The data suggest reduced BHI in the smoking group (0.252±0.097 vs. 0.306±0.098 cm/s2, p<0.07) indicative of dysregulation of vascular reactivity.

Background and Motivation

Endothelial cells produce a variety of vasodilators, chief among which is nitric oxide (NO), but also vasoconstrictors such as endothelin. Reactive oxygen species (ROS) such as the superoxide radical anion (O2-.), occurring in cigarette smoke, are well known to dysregulate the normal vasodilatory response of the endothelium to increased demand for blood flow (1). Endothelial dysfunction is believed to be the earliest initiator of atherogenesis followed by the well-known sequelae leading to neurovascular and heart disease (2).


Vascular reactivity was assessed in a subset of patients (20 nonsmokers and 13 smokers) drawn from a larger study designed to evaluate the effect of smoking and age on a variety of vascular measures in 169 subjects without symptomatic cardiovascular disease conducted at 3T field strength in (3). Subjects underwent an apnea paradigm consisting of a single 30-second normal breathing baseline period followed by three repeated blocks of a 30-second breath-holds, each separated by a 90-second normal breathing recovery period. Blood flow velocity in the superior sagittal sinus (SSS) during the 6-min apnea paradigm was quantified using a non-gated phase-contrast BRISK acquisition scheme with BRISK k-space sampling and view sharing (4) achieving a temporal resolution of 2s. Sequence parameters: flip angle 15°, bandwidth 347 Hz/pixel, TE/TR 6.5/31.25 ms, VENC 50 cm/s, FOV 220×220×5 mm3, resolution 1.15×1.15×5 mm3. The data were evaluated in terms of the slope of the velocity versus time curve, expressed as cm/s2, referred to as BHI slope.

Results, Discussion and Conclusions

Preliminary data of SSS blood flow velocity time-course during an apneic challenge are shown in Figs. 1 and 2. Sample SSS velocity images are displayed in Fig. 1 while Fig. 2 shows velocity versus time plots for three successive BH cycles. Fig. 3 shows box plots comparing smokers to nonsmokers in terms of the BHI slope. The data suggest reduced neurovascular reactivity in smokers (0.252±0.097 vs. 0.306±0.098 cm/s2, p=0.07). The effect does not appear to be driven by age since when stratified according to age, nonsmokers were indistinguishable from their smoking peers. Due to the large intra-group spread the method may be more appropriate to follow subjects during treatment. Nevertheless, the results suggest that the BHI may be a useful supplementary test to evaluate subjects with presymptomatic vascular disease and may be equally applicable to other subject groups at risk of developing neurovascular disease such as those with hypertension or hypercholesterolemia. Lastly, a BHI scan can easily be appended to a standard MRI protocol as the data take only minutes to acquire. The current pilot study is not adequately powered and corroboration of the results in larger subject groups is indicated.


NIH K25HL111422

NIH R01HL122754

NIH R01HL109545


1. Messner et al. ATBV 2014;34(3):509-515.

2. Celemajer et al. J Am Coll Cardiol 1994;24(6):1468-1474.

3. Langham et al. J Cardiov Magn Reson 2015;17(1):19.

4. Rodgers et al. JCBFM 2013; 33(10):1514.


Figure 1 Sample velocity maps during apnea, plotted in 4s increments.

Figure 2 Mean blood flow velocity time-course in the SSS. Shaded areas correspond to period of volitional apnea. Dashed red lines are best fits of velocity during the apnea paradigm yielding BHI slope.

Figure 3 Comparison between nonsmokers (NS) and smokers (S) in terms of the slope of the SSS velocity versus time curve obtained via a breath-hold challenge (see Figure 1)

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)