Clinical Study | Open Access
Vascular Waveform Analysis of Flap-Feeding Vessels Using Color Doppler Ultrasonography
We performed vascular waveform analysis of flap-feeding vessels using color Doppler ultrasonography and evaluated the blood flow in the flaps prior to surgery. Vascular waveform analysis was performed in 19 patients. The analyzed parameters included the vascular diameter, flow volume, flow velocity, resistance index, pulsatility index, and acceleration time. The arterial waveform was classified into 5 types based on the partially modified blood flow waveform classification reported by Hirai et al.; in particular, D-1a, D-1b, and D-2 were considered as normal waveforms. They were 4 patients which observed abnormal vascular waveform among 19 patients (D-4 : 1, D-3 : 1, and Poor detect : 2). The case which presented D-4 waveform changed the surgical procedure, and a favorable outcome was achieved. Muscle flap of the case which presented D-3 waveform was partially necrosed. The case which detected blood flow poorly was judged to be the vascular obstruction of the internal thoracic artery. In the evaluation of blood flow in flaps using color Doppler ultrasonography, determination of not only basic blood flow information, such as the vascular distribution and diameter and flow velocity, but also the flow volume, vascular resistance, and arterial waveform is essential to elucidate the hemodynamics of the flap.
For the evaluation of blood flow in flap-feeding vessels in the field of plastic and reconstructive surgery, various methods are used, such as angiography, MRA, and MDCT, and stable results are obtained; but these methods can be confirmed the presence and distribution of a flap-feeding vessels, and the details, such as the flow velocity and flow volume and vascular resistance, cannot be evaluated. Since color Doppler ultrasonography can be visualized noninvasively and easily, relative low cost of the microblood vessels has recently been used to identify flap-feeding vessels, particularly perforating arteries, and its usefulness has been reported [1–3]. We performed vascular waveform analysis of flap-feeding vessels using color Doppler ultrasonography and evaluated the blood flow in the flaps prior to surgery. In the present study, we introduce an overview of the tests and report on its usefulness.
2. Subjects and Methods
Vascular waveform analysis was performed in 19 patients who underwent reconstructive surgery with pedicle flap (9 patients males and 10 patients females; age was 3379; mean age was 59 years). Primal disease was 7 patients of mediastinitis, 7 patients of malignant tumor, 2 patients of esophagus skin fistula, 1 patient of scar contracture after burn, 1 patient of abdominal wall incisional herniatis, and 1 patient of vagina defect. Transplanted flap was 8 patients of latissimus dorsi musculocutaneous flap, 6 patients of rectus abdominis musculocutaneous flap, 3 patients of pectoralis major musculocutaneous flap, and 2 patients of tensor fascia lata musculocutaneous flap (Table 1).
|LDMC: lattisimus dorsi musculocutaneous flap; RAMC: rectus abdominis musculocutaneous flap; RAM: rectus abdominis muscle flap; PMMC: pectoralis major musculocutaneous flap; PMM: pectoralis major muscle flap; TFLMC: tensor fascia lata musculocutaneous flap; TDA: thoracodorsal artery; SEA: superior epigastric artery; DIEA: deep inferior epigastric artery; TAA: thoracoacrominal artery; LCFA: lateral circumflex femoral artery.|
No underlying severe disease, such as diabetes mellitus, atherosclerosis obliterans, or chronic renal disease, was observed in any patient.
The XARIO SSA-660A-770A (Toshiba) as an ultrasonic diagnosis system and a high frequency linear surface probe (7.5–10 MHz) was used.
Concerning items of vascular waveform analysis, the vascular diameter, flow volume (F-V), maximum flow velocity , minimum flow velocity , mean flow velocity , and resistance index (RI), pulsatility index (PI), and acceleration time (AT: time reaching a peak during systole) based on arterial waveforms were measured. F-V was calculated by multiplying the time integral value of by the cross-sectional area of the vessel assumed to be a circle. RI was calculated as and PI as . The vascular waveform was measured at a site of the flap-feeding vessels after branching within 3 cm. The incidence angle of ultrasound during vascular waveform analysis was 60° in principle.
Blood flow evaluation of flap feeding vessels was performed referring to the standard value of superior epigastric artery, deep inferior epigastric artery, subscapular artery, and thoracodorsal artery which we set up before  (Table 2).
D-1a, D-1b, and D-2 waveforms were regarded as normal; D-3 and D-4 were regarded as abnormal.
Moreover, venous blood flow was confirmed by visualize the running of the vessel and venous waveform.
This study was approved by the Ethics Committee of Toho University Omori Medical Center (number: 24026) and performed after obtaining comprehension from the patients with sufficient informed consent.
It was 4 patients which observed abnormal vascular waveform among 19 patients which performed vascular waveform analysis of flap-feeding vessels using color Doppler ultrasonography prior to surgery (D-4: 1 patient, D-3: 1 patient, and Poor detect: 2 patients). An abnormal D-4 waveform was noted in a case in which superior epigastric arterial waveform analysis was performed before surgery for sternal osteomyelitis (Patient 1). The was 15.6 cm/sec, being slower than the standard; RI was 0.34, and PI was 0.50, being lower than the standard. Since the waveform of internal thoracic artery was a normal waveform, a part of internal thoracic artery was damaged on the occasion of sternum debridement, and a possibility of angiostenosis was suspected (Figure 2). Therefore the surgical procedure was changed to one using latissimus dorsi musculocutaneous flap, and a favorable outcome was achieved. The case which presented D-3 waveform planned reconstruction by rectus abdominis muscle flap to mediastinitis and vascular waveform analysis of the superior epigastric artery was performed. However, the was 15.0 cm/sec, being lower than the standard, and the muscle flap was partially necrosed (Figure 3). The case which detected blood flow poorly planned reconstruction by rectus abdominis musculocutaneous flap to mediastinitis and vascular waveform analysis of the superior epigastric artery was performed. However blood flow detection of the internal thoracic artery to the superior epigastric artery was poor. The blood vessel which flows into rectus abdominis muscle was confirmed in the direction of the outside; we performed blood flow evaluation by vascular waveform analysis. Although the vessel diameter was slightly as thin as about 1 mm, the vascular waveform was normal as D-2 waveform. Therefore we judged that the flap transplantation was possible and performed reconstruction using rectus abdominis musculocutaneous flap which was planned. The flap was partially necrosed after the operation. It was guessed to be because the tension of the flap distal part was strong in order to cover the defect of the chest upper part. Patient 17 was judged to be the vascular obstruction of the internal thoracic artery and was changed into reconstruction using omentum and pectoralis major muscle flap. Other 15 patients were valued with sufficient blood flow volume and flow velocity, and the vascular waveform was also normal waveform and the flap completely survived (Table 3).
|TDA: thoracodorsal artery; SEA: superior epigastric artery; DIEA: deep inferior epigastric artery; TAA: thoracoacrominal artery; LCFA: lateral circumflex femoral artery.|
Pedicled rectus abdominis musculocutaneous flaps supplied by the superior epigastric artery and vein were transplanted in 4 patients. The waveform type was D-1b in 1 patient and D-2 in 2 patients and D-3 in 1 patient. One of 2 flaps prepared with D-2 waveform vessels was completely survived, but the other flap was partially necrosed. One flap prepared with D-3 waveform vessels was partially necrosed.
Pedicled rectus abdominis musculocutaneous flaps supplied by the deep inferior epigastric artery and vein were transplanted in 2 patients. The waveform type was D-1b in 1 patient and D-2 in 1 patient, and all flaps were completely survived.
Pedicled latissimus dorsi musculocutaneous flaps supplied by the thoracodorsal artery and vein were transplanted in 8 patients. The waveform type was D-1a in 3 and D-1b in 3 and D-2 in 2 patients, and all flaps were completely survived.
Pedicled pectralis major musculocutaneous flaps supplied by the thoracoacrominal artery and vein were transplanted in 3 patients. The waveform type was D-1b in 3 patients, and all flaps were completely survived.
Pedicled tensor fascia lata musculocutaneous flaps supplied by the lateral circumflex femoral artery and vein were transplanted in 2 patients. The waveform type was D-1b in 1 and D-2 in 1 patient, and all flaps were completely survived (Table 4).
|PN: partial necrosis; S: survive.|
All surgeries were performed by the same operator.
To plan reconstruction with a vascularized flap, it is necessary to investigate the hemodynamics of feeding vessels before surgery in consideration of individual differences, vascular mutation, angiostenosis, and vascular obstruction. For the evaluation of blood flow in flap-feeding vessels in the field of plastic and reconstructive surgery, various methods are used, such as Doppler probe, angiography, MRA, and MDCT [6, 7].
Doppler probe has many advantages, such as simple operation, low cost, and short time of inspection. On the other hand, since the depth of the detected blood vessel cannot be judged, there is a risk of detecting a blood vessel deeper or shallower than the target blood vessel by mistake . Although angiography is suitable for blood-flow evaluation of medium sized artery, it has problems, such as high invasion, radioactive exposure, and a high cost. Moreover, the special technique of image processing method is needed for imaging a thin artery like perforating vessels. Since MRA and MDCT have high contrast resolution, not only blood vessels and bones can be visualized, but also muscles and the surrounding soft tissue. Moreover, a perforating vessel less than 1 mm  also can be visualized. Thus, although highly precise image is obtained by low invasion, condition setting is necessary in order to enable the imaging of perforating vessels, such as to squeeze the particular vessel and to reduce the slice thickness of the image. In addition, the available facilities are limited because equipment is expensive.
On the other hand, color Doppler ultrasonography can be visualized noninvasively and easily, relative low cost the micro blood vessels compared to these tests, it has recently been used to identify flap-feeding vessels, particularly perforating arteries, and its usefulness has been reported [1–3]. Recently, the detectability of low-speed blood flow employing the Doppler method has markedly improved with advancement in ultrasonographic diagnostic devices, and the accuracy has also markedly improved. In particular, the introduction of the power Doppler method, which is excellent for detecting low velocity blood flow, has allowed the visualization of peripheral vessels, which was conventionally difficult, and its application has expanded to many departments.
Regarding vascular waveform analysis of peripheral blood vessels, analysis is applied to diagnose lower limb occlusive arterial diseases. Hirai et al.  applied it for the screening of pelvic and lower limb occlusive arterial diseases, and classified the vascular waveforms into 4 types: D-1 to D-4 (Figure 4). Baba et al.  performed vascular waveform analysis of severely ischemic lower limbs and observed that whether the waveform type is D-1 or not could be determined following the waveform classification reported by Hirai et al.; but classification of the other types was difficult in many cases. They classified waveforms based on the presence or absence of diastolic components and direction (antegrade/retrograde) into the following 3 types: Type A to C (Figure 5). However, some arterial waveforms were difficult to classify employing the classifications reported by Hirai et al. and Baba et al. in our patients: systolic waves rose steeply but were not followed by reflux components in diastole, a notch was present between the systolic and diastolic waves, and anterograde diastolic waves were observed. No reflux component was noted in diastolic waves in many cases, even though the vascular resistance was high and systolic waves showed a favorable steep rise in flap-feeding arteries, because these are peripheral arteries with a smaller diameter than those of the distal external iliac artery and proximal region of the popliteal artery involved in the waveform classification reported by Hirai et al. Thus, we considered it inappropriate to classify all these waveforms as D-2 and classified them as D-1b when a notch was noted between systolic and diastolic waves and designated the original D-1 as D-1a (Figure 1).
Hirai et al. regarded the D-1 type as a normal waveform with no significant stenosis on the proximal side and suggested the presence of a stenosis lesion on the proximal side in D-2, D-3, and D-4 waveforms. However, these are applied to arterial waveforms in the pelvis and thighs but difficult to apply to peripheral arteries unless some modification is made. In peripheral arteries, diastolic waveforms are lost because peripheral vascular resistance is high, and is nearly 0 in many waveforms. Since these waveforms may also be normal in peripheral arteries, we regarded D-1a, D-1b, and D-2 as normal waveforms in the classification.
As parameters for the quantification of pulsative waveforms, based on , , and , the A/B ratio calculated as , RI as , PI as , and AT have been used. High and low values of A/B ratio, RI, and PI indicate high and low peripheral vascular resistances, respectively. The A/B ratio and RI are simply calculated from and , and the reproducibility is superior because values with relatively small errors are used, showing that these are practical parameters. However, the A/B ratio reaches infinity when comes close to 0, and the RI can be calculated even when peripheral vascular resistance is markedly high and no diastolic blood flow component is observed in waveforms, but the value becomes RI = 1 regardless of the waveform shape. On the other hand, the PI value has no problem because of the use of even when no diastolic blood flow component is observed, but the calculation is not simple because the mean flow rate is calculated by tracing blood flow waveforms. Moreover, its calculation accuracy is inferior to that of RI due to technical errors. AT is an index reflecting changes in the time axis. Its calculation is simple, and an element in the time axis direction is added, for which improvement of the accuracy of blood flow evaluation is expected. In this study, we used RI, PI, and AT as indices for quantification of the pulsating waveforms of blood flow.
We performed vascular waveform analysis of the flap-feeding vessels of about 40 patients and reported the established reference values based on measurement values of vascular diameter, F-V, RI, PI, and AT  (Table 2). Now it is applied as an index of the blood flow in the flaps prior to surgery. Although there is some difference by the kind of flap-feeding vessels about the standard value for judged whether the transplantation of the skin flap is safe; vascular diameter ≧ 1 mm; mL/min; cm/sec; ; and msec.
On a comparison of the results of vascular waveform analysis in actual cases of flap transplantation with the established standard values, the vascular waveform were normal (D-1a or D-1b or D-2) in cases which achieved complete survival, the diameter, FV, and , were higher than the established values, indicating favorable blood flow, and the RI, PI, and AT were within the standard value ranges.
On the other hand, an abnormal D-4 waveform was noted in a case in which superior epigastric arterial waveform analysis was performed before surgery for sternal osteomyelitis. The was 15.6 cm/sec, being slower than the standard; RI was 0.34, and PI was 0.50, being lower than the standard (Figure 2). The surgical procedure was changed to one using lattisimus dorsi musculocutaneous flap, and it was possible to perform a safe operation. In the case in which the pedicled flap containing the superior epigastric artery was partially necrosed, the waveform was abnormal D-3, and the was 15.0 cm/sec, being lower than the standard (Figure 3).
The case which detected blood flow poorly, the vascular obstruction of the internal thoracic artery, was able to be judged prior to surgery, and it was possible to perform safe operation by changing into the reconstruction using omentum and pectoralis major muscle flap. The above clinical findings reflected the usefulness of the established standards.
The following advantages can be mentioned in the color Doppler ultrasonography. (1) It can be visualized the vessels at the same time as the fascia and muscle, adipose tissue, bone, it is possible to understand the detailed running of flap-feeding vessels. (2) We can select the skin flap with good blood flow more by measuring vessel diameter, flow volume, and flow velocity. (3) In the free flap transplant patients, postoperative patency confirmation of vascular anastomosis is easy; it can be observed noninvasively and frequent times. (4) Therefore, it is useful for early detection of vascular occlusion such as postoperative thrombosis. (5) Quantitative observation of blood flow velocity of the anastomotic vessels after surgery is possible. However, technical skills, such as the skill of acquiring vascular images and experience in device operation are necessary. To perform vascular waveform analysis by visualizing the vascular distribution in detail and measuring the vascular diameter, F-V, RI, PI, and AT, collaboration with technologists is necessary, which will take time to conduct tests to some extent. To identify the distribution of a flap-feeding vessels and detailed hemodynamics, that is, for preoperative evaluation to perform safe surgery, it may be necessary for an operator familiar with vascular anatomy to attend to the tests and collaboration with technologists for the operation and setting of the ultrasonography system.
As preoperative blood flow assessment of the flap-feeding vessels, we performed vascular waveform analysis using color Doppler ultrasonography. Based on arterial waveform of the flap-feeding vessels, F-V, RI, PI, and AT were measured and classified vascular waveforms into 5 types partially modifying the blood flow waveform classification of Hirai et al. Although skills are necessary for the manipulation and setting of the ultrasonography system, and collaboration with clinical technologists is required, it is important to evaluate not only basic blood flow information of the vascular distribution, vascular diameter, and flow velocity, but also the flow volume, vascular resistance, and arterial waveform in analyzing the hemodynamics of the vascularised flaps and essential in performing Doppler ultrasonography.
Conflict of Interests
The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in this paper.
The authors are grateful to Mr. Yakuwa, technologist at the Clinical Physiological Laboratory, Toho University Omori Medical Center, for his cooperation with this study.
- G. G. Hallock, “Evaluation of fasciocutaneous perforators using color duplex imaging,” Plastic and Reconstructive Surgery, vol. 94, no. 5, pp. 644–651, 1994.
- P. N. Blondeel, G. Beyens, R. Verhaeghe et al., “Doppler flowmetry in the planning of perforator flaps,” British Journal of Plastic Surgery, vol. 51, no. 3, pp. 202–209, 1998.
- A. Ogino, Y. Maruyama, K. Onishi, K. Kanda, and T. Muro, “Preoperative color Doppler assessment for identifying vascular of skin flaps,” Journal of Japan Society of Plastic and Reconstructive Surgery, vol. 30, no. 2, pp. 45–51, 2010.
- A. Ogino, K. Onishi, and Y. Maruyama, “Vascular waveform analysis of vascularized flaps using color Doppler ultrasonography,” the Japan Society for Simulation Surgery, vol. 19, no. 3, pp. 1–11, 2011.
- T. Hirai, H. Ohishi, K. Kichikawa, H. Yoshimura, and H. Uchida, “Ultrasonographic screening for arterial occlusive disease in the pelvis and lower extremities,” Radiation Medicine—Medical Imaging and Radiation Oncology, vol. 16, no. 6, pp. 411–416, 1998.
- G. I. Taylor, M. Doyle, and G. McCarten, “The doppler probe for planning flaps: anatomical study and clinical applications,” British Journal of Plastic Surgery, vol. 43, no. 1, pp. 1–16, 1990.
- J. Pauchot, S. Aubry, A. Kastler, O. Laurent, B. Kastler, and Y. Tropet, “Preoperative imaging for deep inferior epigastric perforator flaps: a comparative study of computed tomographic angiography and magnetic resonance angiography,” European Journal of Plastic Surgery, vol. 35, pp. 795–801, 2012.
- R. Baba, K. Minowa, A. Kawamoto, M. Katayama, N. Handa, and Y. Kaneko, “Usefullness of the doppler wave form analysis in doppler ultrasonography: screening for critical limb ischemia,” Journal of Japanese College of Angiology, vol. 48, no. 2, pp. 203–211, 2008.
Copyright © 2014 Akihiro Ogino and Kiyoshi Onishi. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.