Computed Tomography Angiography in Microsurgery: Indications, Clinical Utility, and Pitfalls

METHODS

We obtained institutional review board approval and performed a retrospective analysis on all patients who underwent a preoperative CTA in preparation for microsurgical reconstruction after at a single tertiary care academic institution over a 5-year period (2001–2005). Patients who underwent CTA for any indication other than microsurgical planning were not included in the study. Patients obtained CTA routinely for cases of free fibula flap donor site evaluation; however, for other scenarios CTA was obtained on the basis of clinical judgment by the surgeon in which there was a concern for abnormal vascular anatomy. Computed tomographic angiography was obtained using a 16-row detector scanner (Lightspeed; GE Medical Systems, Waukesha, Wis). Our institution's previously published CTA protocol was followed in this study.^2,16

We determined the original plan for microsurgical reconstruction based on review of physicians' notes that also indicated when an alternative therapy was chosen subsequent to CTA findings. Abnormalities were identified upon review by both the radiologist and the surgeon. Abnormalities included vascular occlusion, stenosis, and atherosclerotic lesions but did not include normal anatomic variants. We evaluated defect type, choice of flap, vessel image quality, influence of CTA on operative planning, concordance with intraoperative findings, CTA-related complications, and flap-related complications. Intraoperative evaluation of blood vessels was considered the criterion standard in our analysis.

Standard statistical analysis was performed to calculate the following: sensitivity (number of abnormal CTA correlating with actual intraoperative abnormality divided by number of actual intraoperative abnormality); specificity (number of normal CTA correlating with normal intraoperative anatomy divided number of normal intraoperative anatomy); positive predictive value (number of abnormal CTA correlating with actual intraoperative abnormality divided by number of abnormal CTA); and negative predictive value (number of normal CTA correlating with normal intraoperative anatomy divided by number of normal CTA).

RESULTS

CASE REPORTS

Figure 1. CTA in reconstructive breast surgery. CTA imaging of thoracodorsal vessels to assess suitability of the vessels as recipients for a DIEP flap. The vessels were patent, but embedded in large amounts of surgical scar, with marked perivascular fibrosis. DIEP indicates deep inferior epigastric perforators.

Figure 2. CTA in reconstruction of the upper limb. CTA of the right upper extremity demonstrating occlusion of the ulnar artery in the distal forearm at the level of the fracture site. The hand is supplied through the radial artery and a complete palmar arch.

Figure 3. CTA in reconstruction of the lower limb. CTA of a traumatized right lower extremity demonstrating severe atherosclerotic disease as well as numerous focal occlusions of the anterior tibial and posterior tibial arteries distal to the trifurcation of the popliteal artery.

Figure 4. CTA of recipient site for head and neck reconstruction. CTA of head and neck vessels in a patient with a fractured mandibular reconstruction plate. There is total occlusion of the facial artery and significant soft tissue damage near the external carotid artery.

Case 5: CTA of donor site in head and neck reconstruction

A 41-year-old woman with ameloblastoma of the mandible presented for free fibula flap reconstruction. On examination, dorsalis pedis and posterior tibial pulses were weakly palpable. A preoperative CTA of the lower extremities demonstrated bilateral peroneus magnus (Fig 5). Therefore, the patient successfully underwent a free iliac crest flap instead.

Figure 5. CTA of donor site for head and neck reconstruction. CTA of bilateral lower extremities in a patient with ameloblastoma of the mandible demonstrating bilateral peroneus magnus.

DISCUSSION

Computed tomographic angiography has been lauded as an excellent alternative to traditional angiography because it is less invasive, involves less radiation exposure, and is considerably more cost-effective.¹¹ One of the main arguments against routine arteriography is the considerable rate of major complications, which has been reported to be as high as 7%.¹⁷ Numerous studies have demonstrated the safety of CTA, and our results further confirm this conclusion. We observed no CTA-related complications in our patient population.

The use of CTA has been documented in the vascular medicine literature, and a number of studies have highlighted its potential use in plastic surgery. Computed tomographic angiography has been used to identify zones of perfusion^18,19 and to aid in the intraoperative dissection of free abdominal flaps in breast reconstruction.²⁰ Preoperative CTA was found to be highly accurate in mapping the perforators of the deep inferior epigastric artery^21-23 and was associated with decreased operative times, flap complications, and donor site morbidity.^24,25

Our study shows that CTA is reliable in its ability to detect vascular anomalies that influence the planning of various microsurgical reconstructive procedures (Table 3). These results are similar to other studies in which CTA was performed for mapping of atherosclerotic lower limb vasculature.^26,27 Recent studies corroborate these findings for CTA used in plastic surgery applications for the anterolateral thigh flap harvest^14,28 and abdominal-based free flaps used for breast reconstruction.¹⁵

Preoperative vascular imaging at the free flap recipient site is often obtained when there has been significant trauma or prior extensive surgery. Routine imaging of the donor site is somewhat controversial. There is evidence that clinical examination of distal pulses in fibula flap evaluation is as accurate as traditional angiography while safer and more cost-effective.^29-31 Computed tomographic angiography, however, not only offers a volumetric display of the vasculature but also allows the surgeon to see the desired vascular anatomy in relationship to the surrounding soft tissues. This study is the first of sufficient size to document vascular abnormality rates detected by CTA in patients undergoing reconstructive surgery of all types.

We report a total abnormality rate of 51%. This number at first blush may seem high, but one must consider our particular patient population. It is important to note that the vast majority of patients receiving free flaps who also received a CTA were considered a high-risk group with complex reconstructive issues, comorbidities, and concerns for possible abnormal vascular anatomy. Hence, our study includes a selected population, which, in part, explains the high incidence of vascular anomaly findings on CTA. Nonetheless, an abnormality rate of 51% during preoperative CTA is large enough to warrant attention. Most abnormality rates reported in the plastic surgery literature using traditional angiography or clinical pulse examination are considerably lower. These differences may be attributed to the better spatial resolution of CTA and particularly to its ability to delineate soft tissue injuries that cannot be detected with traditional angiography.

Vascular abnormalities detected as a result of preoperative CTA caused changes in operative plans in 45% of patients (Table 2). Changes included use of alternate recipient vessels, use of pedicled flaps instead of free flaps, or harvest of free flaps from the contralateral side. Furthermore, surgeons were able to work around known soft tissue injuries and decide on the location of the anastomosis using patency data from CTA as well as direct vessel visualization. Of note, obtaining CTA was considerably less expensive than traditional angiography with an average cost at our institution of $1140 as compared with $3900 for traditional angiography.¹¹

Our study identified some limitations of CTA, especially in the presence of external metal fixators (Table 3). Although CTA provides the ability to rotate images 3-dimensionally, artifact created by metallic hardware can make it difficult to reliably delineate the vascular anatomy. However, current improvements in CTA technology are underway to suppress the metal-induced aberrations. Another drawback of CTA is that it requires at least as much contrast load as traditional angiography, which limits its use in patients with impaired renal function. Radiation is also a concern with any additional CT scan.³² Currently, MRA is under investigation to provide imaging without radiation, but studies are small and have mixed results for plastic surgery applications.^33,34 In addition, any patient with a pacemaker or noncompatible metal implant would be eliminated from MRA evaluation.

Our study shows that CTA is a safe and reliable tool for delineating the vascular anatomy prior to microsurgical reconstruction (Table 3). Preoperative CTA may find its greatest utility in situations where there is concern or ambiguity about the vascular anatomy. In our series, 51% of the scans revealed some vascular abnormality. The operative plan was modified on the basis of CTA results in 45% of all patients (88% of patients with abnormal CTA findings). However, not all anatomic sites had the same rate of CTA abnormalities. Abnormality rates in the breast and chest wall were less than 33%, in the lower extremity between 33% and 66%, and in the upper extremity greater than 66% (Table 2). These differences can be explained in part by the etiology of the disease process. For example, CTA of the upper extremity was mainly performed in trauma cases in which there was a 55% incidence of abnormal findings as compared with infection (45%) or cancer (43%).

The degree of CTA influence on operative planning varied considerably with the type of surgery performed. Reconstructions were more likely to be influenced by CTA after trauma or cancer than after infection (Table 2).

Our results serve as a basis upon which recommendations for preoperative CTA can be made. The use of CTA should be considered for cases of microsurgical reconstruction where the vascular anatomy may be in question. This was most often the case for traumatic or infectious injuries of the upper extremity, for fibula flap donor sites, and injuries of the lower extremities. The plastic surgeon is advised to have a low threshold for obtaining a CTA in those cases. However, for cases where the vascular anatomy has shown to be fairly consistent or where there is no trauma and no potential cause for vascular distortion, the utility of CTA may decrease. This was the case for breast or chest reconstructions, for which preoperative CTA is not recommended unless extensive surgery, scarring, or prior radiation is present. Further studies are warranted to evaluate the cost-effectiveness of routine CTA for all cases of reconstruction.

CONCLUSIONS

Computed tomographic angiography is a safe and reliable method of preoperatively assessing vessels used in microsurgical reconstruction. When an abnormality in vascular anatomy is detected by CTA, the surgeon is advised to consider altering the operative plan accordingly. Choosing alternate vessels to use as recipients or changing the method of reconstruction may be necessary to avoid vascular complications. Computed tomographic angiography may not be able to fully delineate the vascular anatomy when there is significant metallic artifact present such as an external fixation device, but technological advancements are improving CTA resolution in these circumstances.