Allergic reaction and contrast-induced nephropathy are well-known adverse reactions associated with iodine contrast material. Additionally, we found and confirmed a new phenomenon: a transient decrease in the tracheal diameter and lung volume. The tracheal diameter and both lung volumes were decreased by 2.5% and 4.4%, respectively, in the arterial phase, and a more than 10%/20% decrease of tracheal diameter and both lung volumes was observed in 1.8%/0.5% and 18.1%/3.6% of patients, respectively. These phenomena were transient, and no significant effect was observed in the delayed phase.

There have been studies which reported factors affecting the diameter or airway volume. Yamada et al. reported that the ratio of inspiratory to expiratory airway volume in supine position is larger than the standing position in patients with chronic obstructive pulmonary disease (p < 0.001) [19]. In asthmatics, for whom the bronchoalveolar lavage fluid count tend to be high, the eosinophil count in the bronchoalveolar lavage fluid was negatively associated with the airway diameter (r =  − 0.7, p < 0.05) [23]. Additionally, while obesity is known to be associated with a higher incidence and prevalence of asthma [24], abdominal visceral fat area was negatively associated with airway diameter in asthmatics (r =  − 0.35, p = 0.01) [25]. Furthermore, the barometric pressure has an impact on the increased risk for hospital visit for asthma [26]. Our study is unique in that the association between contrast agent injection, which is known to be a risk factor for patients with poorly controlled asthma [9], and decrease in tracheal diameter was demonstrated. Whether this phenomenon can be a trigger for hypersensitivity reaction in patients with or without asthma requires further investigation.

Chest CT examination is sometimes performed at levels other than end-inspiratory level. One representative case is CT pulmonary angiography, which is sometimes scanned at mid-inspiratory level to reduce the risk of transient interruption of contrast [27]. However, chest CT image scanned at mid-inspiratory level is known to increase a risk of nondiagnostic lung images [27]. From our study, it was found that more than 10% or 20% decrease of both lung volumes was observed in 18.1% and 3.6% of patients, respectively. In daily clinical practice, it would be better to avoid scanning chest at the arterial phase when aimed to evaluate the lung at the end-inspiratory level.

In interstitial lung diseases, it has been reported that both lung volumes were 3345 mL, and the relative annual bilateral lung volume loss was reportedly 2.07% and 17.44% in patients without and with major adverse event at a 3-year follow-up, respectively [17]. In our study, it was found that both lung volumes decreased by an average of 4.4% for the arterial phase compared with the unenhanced CT, and more than 10% and 20% decrease was observed in 18.1% and 3.6% of patients, respectively. The effect of a transient decrease in both lung volumes in the arterial phase would not be negligible in the lung volume assessment of interstitial lung disease.

The ratio of the expiratory volume to inspiratory volume in the lower lobes has been reportedly smaller than those in the upper and middle lobes (41.1–41.7% vs. 51.9–65.4% [28] or 57.4–57.8% vs. 67.5–74.1% [19]). This would be attributed to the fact that the lower lobes are located near the diaphragm which play an important role in respiration. In the current study, the volume of the lower lobes was more largely decreased (93.1–94.1%) in the arterial phase compared with the upper or middle lobes (96.6–98.5%). This fact indicates that the lobe volume decrease in the arterial phase was caused by the movement of the diaphragm. In addition, not only the lung volume but also the tracheal diameter was decreased in the arterial phase, and significant correlation between TDAU and BLVAU was observed. Furthermore, we excluded patients in whom errors in segmentation of the lung or bronchus were seen. These facts suggest that the change in the arterial phase was caused by the respiration not merely by the segmentation performance variation caused by the CT attenuation of the pulmonary vessels.

The right-side injection was significantly associated with a lower BLVAU value (0.950) compared with the left-side injection (0.988) (p = 0.022). This may be associated with the fact that the length between the antecubital vein and the right side of the heart is shorter for the right antecubital vein. A transient dilatation of right side of the heart or transient chemical reaction caused by the contrast agent molecule may be possible reasons for this phenomenon; however, a detailed mechanism remains unknown.

This study has some limitations. First, the relationship between the hypersensitivity reaction secondary to the contrast agent and the tracheal diameter was not assessed in this study. Future study regarding this topic is warranted. Second, because a CT scan was performed during end-inspiration level, whether change could be observed even in resting breathing state or patients just experienced difficulty in deep inspiration remains unclear. Third, because patients included in this study did not necessarily have lung diseases, depiction of diseases was not assessed. Finally, all patients included in this study underwent vascular dynamic CT examinations. Future studies including patients with other conditions are warranted.

In conclusion, the tracheal diameter and both lung volumes were transiently decreased in the arterial phase of vascular dynamic CT examination compared to unenhanced CT. To reveal the onset and duration of this phenomenon, future studies including CT examinations other than vascular dynamic CT is needed. In addition, the mechanism of this phenomenon is needed to be investigated in future studies.