Chronic venous disease (CVD) is a common medical condition that causes various pathologies, including varicosity, pain, swelling, edema, skin changes, and ulcerations. CVD includes varicose veins and superficial venous reflux conditions. Varicose veins, a common manifestation of venous incompetence in the lower limbs, are dilated, elongated, or tortuous superficial veins that are most often palpable subcutaneous veins with reversed blood flow (1). The incompetence of the deep, superficial, and/or perforating veins leads to increased venous pressure in the lower leg, resulting in skin changes, such as hyperpig-mentation and induration with eventual ulceration (2).

Wearing compression stockings is the primary treatment for symptomatic CVD. However, due to the inconveniences caused in ordinary life, patients often show relatively low treatment compliance. Moreover, CVD patients may be affected by several comorbidities, making their compliance with compressive therapy even worse. The most frequent comorbidities associated with CVD are diabetes mellitus, hypertension, heart failure, peripheral arterial disease, renal insufficiency, emphysema, chronic obstructive pulmonary disease, and malignant disease (3).

Therefore, there is an emerging need for alternative therapeutic strategies. While some studies have reported improvements in venous return in response to various methods in an asymptomatic group (4,5), the fact that these research were conducted with asymptomatic participants in a supine position makes it challenging to apply in clinical practice. In this study, we investigated the effect of diaphragmatic deep breathing (DB) on venous return in symptomatic patients.

Volunteers with venous symptoms participated in this prospective study after obtaining approval from the local institutional review board (C2011116 [566]). Thirteen patients (male:female=4:9) with CVD were recruited from our institution between August 2011 and September 2011. One patient dropped out of the study because of vasovagal syncope during the test, and 12 (four men and eight women) were included in this study.

The study included patients with CVD symptoms who could practice DB correctly and those not indicated for surgery or interventional procedures. Exclusion criteria were: (1) diagnosis of acute or chronic deep venous thrombosis; (2) current use of diuretics, cardiotonics, or vasodilators; (3) heart or kidney disease confirmed through patient history; and (4) current or future pregnancy plan.

Before evaluating its efficacy, we educated the participants on the proper method of DB and instructed them to practice it for 1 month for two reasons: 1) to prevent hyperventilation by exaggerated breathing during the test and allow natural breathing without awareness, and 2) to allow the influence of the respiratory cycle on venous return to persist as a natural course of daily life. During the inspiration phase, while performing DB, patients were instructed to inhale through the nasal cavity without cessation of inhalation and to distend the abdomen gradually. During expiration, the patients were asked to pull in the abdomen and breathe out as much as possible.

After obtaining informed consent from the volunteers, we evaluated the underlying pathologies, including venous reflux, deep vein occlusion, stenosis, or other anatomical abnormalities, using duplex ultrasound (ProSound a7, ALOKA Co., Ltd., Tokyo, Japan) with a 5∼13 MHz probe. Next, we measured the blood flow volume (ml/min) and maximal value of peak systolic velocity (PSV, cm/s) of the common femoral vein (CFV) and the venous flow velocity in both the supine and standing positions. Data averaged from three readings of PSV and blood flow volume were used for subsequent analyses. In addition, we performed radioisotope (RI) venography with a gamma camera (FORTE, Philips, Milpitas, CA, USA) to measure the wash-out time of ⁹⁹Technetium-macroaggregated albumin (⁹⁹Tc-MAA) in the standing position (Fig. 1).

Figure 1. Radioisotope venography. The time to reach the peak level of ⁹⁹Tc-MAA (T_max, min) and the half washout time of ⁹⁹Tc-MAA(T_1/2, min) was measured. ⁹⁹Tc-MAA: ⁹⁹Technetium-macroaggregated albu-min with normal breathing (A) and diaphragmatic deep breathing (B).

Following the measurements of blood flow volume, CFV velocity, and ⁹⁹Tc-MAA wash-out time, a professional trainer educated the patients to practice DB and instructed them to practice it for a month. Then, we re-assessed the blood flow volume, CFV velocity, time to reach the peak level of ⁹⁹Tc-MAA (t_max, min), and the half wash-out time of ⁹⁹Tc-MAA(t_1/2, min).

Continuous variables were compared using the Mann- Whitney U test, while categorical variables were assessed using the c² or Fisher’s exact test, as appropriate; p<0.05 was considered significant. All statistical analyses were performed using IBM SPSS version 26.0.

Data from 12 participants, including four men and eight women (median age: 27 years, range 23∼32 years), were analyzed. None of the participants had an obstruction or pathological reflux in the lower extremity veins on duplex examination.

Table 1 shows the PSV and blood flow volume measured in supine and standing positions. In the supine position, PSV of CFV was significantly increased with DB compared to natural breathing (NB) on both sides (Right: 122.5 cm/sec vs. 27.5 cm/sec, Left: 81.4 cm/sec vs. 46.4 cm/sec, p=0.043, both). The blood flow volume measured in the supine position was higher after DB than after NB in the right CFV (NB: 655.0 ml/min and DB: 1098.0 ml/min, p=0.043). However, in the left CFV, the blood flow volume did not significantly increase after DB (NB: 450.7 ml/min and DB: 665.7 ml/min, p=0.138).

Table 1 . The median value of peak systolic velocity and blood flow volume of the CFV measured by duplex ultrasonography in the supine and standing position.

	NB	DB	p-value
Supine
Rt. PSV (cm/sec)	27.5	122.5	0.043
Lt. PSV (cm/sec)	46.4	81.4	0.043
Rt. flow volume (ml/min)	655.0	1098.0	0.043
Lt. flow volume (ml/min)	450.7	665.7	0.138
Standing
Rt. PSV (cm/sec)	15.0	42.0	0.105
Lt. PSV (cm/sec)	8.6	31.6	0.029
Rt. flow volume (ml/min)	220.7	310.4	0.573
Lt. flow volume (ml/min)	485.6	423.8	0.743

CFV: common femoral vein, PSV: peak systolic velocity flow, NB: normal breathing, DB: diaphragmatic breathing..

In the standing position, there was no significant increase in the PSV after DB compared to NB in the right CFV (NB: 15.0 cm/sec and DB: 42.0 cm/sec, p=0.105), while the PSV of left CFV significantly increased after DB (NB: 8.6 cm/sec and DB: 31.6 cm/sec, p=0.029). There was no significant increase in the flow volume on either side in the standing position (p=0.573 for the right CFV and p=0.743 for the left CFV).

The T_max and T_1/2 are summarized in Table 2. The T_max (right) was significantly different between the NB and DB conditions (1.15 min and 0.6 min, respectively, p=0.013), as was the measured T_max (left) (1.4 min and 0.7 min, between the NB and DB, respectively; p=0.016). The T_1/2 of ⁹⁹Tc-MAA (min) was lower in both CFV (right and left) after the DB education; T_1/2 (right) was 0.6 min and 0.5 min in NB and DB, respectively (p=0.007), while T_1/2 (left) was 0.8 min and 0.4 min in NB and DB, respectively (p=0.013).

Table 2 . The median value of time to reach the peak ⁹⁹Tc-MAA level and half washout time of ⁹⁹Tc-MAA measured by ⁹⁹Tc-MAA venography in standing position.

	NB	DB	p-value
Tmax Rt.	1.15	0.6	0.013
Tmax Lt.	1.4	0.7	0.016
T1/2 Rt.	0.6	0.5	0.007
T1/2 Lt.	0.8	0.4	0.013

Tmax: time to reach peak ⁹⁹Tc-MAA level (min), T_1/2: half washout time of ⁹⁹Tc-MAA (min), NB: normal breathing, DB: diaphragmatic breathing..

Venous return, the flow of blood back to the heart, is aided by (1) valves, (2) muscles (skeletal muscle), and (3) respiratory (thoracic) pumps. The one-way valves in the veins prevent the backflow of blood, thereby directing the blood flow back towards the heart during venous return. Skeletal muscles act as pumps in the limbs. When muscles are relaxed, closed valves prevent blood from flowing backward. However, when contracting muscles squeeze the veins, open valves allow blood to flow toward the heart. Finally, respiratory pumps modulate intrathoracic pressure. During inspiration at rest, the diaphragm descends, resulting in the lowering of intrathoracic pressure and the resultant expansion of the lungs. This lowered intrathoracic pressure is also transmitted across the walls of the right atrium, thereby promoting right atrial filling and widening of the right atrial transmural pressure (6). Inspiration causes a fall in intra-thoracic pressure and a simultaneous rise in intra-abdominal pressure due to the diaphragm’s descent (7). As we breathe in, the diaphragm flattens, pushing on internal organs, which, in turn, push on veins sending blood back to the heart.

Our study investigated the therapeutic applicability of DB as an intrinsically powerful pump for enhancing venous return. Theoretically, expansion of the abdominal cavity using the appropriate DB method during inspiration might compensate for increased intra-abdominal pressure. However, the effect of DB during inspiration was not significant. In contrast, the increase in venous flow volume and PSV with DB during expiration was evident compared to that with NB. During DB, the duration of each respiratory cycle was much longer than that during NB. A longer expiratory time with DB could have lengthened the time to decrease intra-abdominal pressure, increasing venous return. The effective decrease in intra-abdominal pressure may generate a larger pressure gradient between the infra- and supra- diaphragmatic parts of the inferior vena cava (IVC), “pulling” the blood towards the right atrium and increasing venous return.

The major finding of our study was that DB increased venous return of the lower extremities, which was measured using RI venography in the standing position. This result is supported by previous reports (4,8). Kwon et al. (4) studied the effects of ankle exercise combined with DB on the blood flow velocity in the CFV in the supine position. The mean peak blood flow velocity in the CFV was highest in the participant group that combined ankle exercise with DB. Byeon et al. (8) compared the superior vena cava (SVC) and IVC in patients who had practiced regular DB for over 2 years (mean duration of training was 9.6 years) with the control group. The diameter of the IVC and SVC were measured using transthoracic echocardiography in the supine position. The IVC of participants who practice abdominal breathing had a greater degree of collapse than that of control, suggesting that DB exercise can positively affect venous return via the IVC. The results of our study can be interpreted in the same context. However, previous studies were only performed in the supine position. Considering that CVD symptoms present themselves in the standing position, our study was significant because we examined patients in the supine and standing positions. In our study, not every finding supports the beneficial effect of DB on venous return. The difference in the flow volume in the standing position was not striking, and the flow volume of the left CFV decreased even after DB. In contrast, the T_max and T_1/2 measured by RI venography were improved after DB in the standing position. This difference might result from unclear effects and the small sample size. The difference might have resulted from the different characteristics of both examinations; ultrasound was measured only in the CFV, and RI venography was the sum of all flows in the observed area. The respiration system is a delicate system involved in gas exchange and the body’s balance of acids and bases. To use it therapeutically and not to cause undesirable effects, further research on the exact ratio of expiration and inspiration and breathing methods is required.

The major strength of our study is that we used RI venography, an objective index of venous flow, as the examination tool. Duplex ultrasound has several advantages in diagnosis, treatment guidance, and follow-up, including low cost, portability, noninvasiveness, and safety (9). Similar to contrast venography, this tool provides a roadmap of vein anatomy and essential hemodynamic information (10,11). Although duplex ultrasound is the diagnostic test of choice for diagnosing and evaluating venous disease, RI veno-graphy was used simultaneously to increase reliability as an objective test. RI venography, while technically simple, is a reliable test for examining vessels with no blood flow, collateral blood flow, various types of adjacent blood flow, or other asymmetric blood flow insufficiency that suggests venous insufficiency (12).

A major limitation of our study is the small sample size. We believe that this might have resulted in the statistical insignificance of the results of the left CFV in the supine position. Although there seemed to be a difference in the blood flow volume between the NB and DB conditions, this effect did not reach statistical significance. Recruiting patients with CVD symptoms is difficult, given that using compression stockings is regarded as the standard method to improve the symptoms of CVD. Thus, attempting DB before using compression stockings is not advisable. In addition, since ⁹⁹Tc-MAA venography is not a mandatory tool for diagnosing CVD, it was difficult to examine every patient because the test requires intravenous injection. Another key limitation is that the follow-up time was short and symptomatic improvement could not be measured. Further studies are needed to examine the long-term efficacy of this approach.

The time-to-peak and wash-out times of CFV in the standing position increased after DB. This result implies that DB can benefit venous return and improve symptoms experienced by patients with CVD. A well-organized practice program that recruits a larger sample is required.