Computed tomography angiography and microsurgical flaps for traumatic wounds: What is the added value?

1 Introduction

The reconstruction of traumatic wounds of limbs with free flap reconstruction is one of the most challenging in microsurgery because it has the highest rate of complication, especially microanastomosis thrombosis. Therefore, it is important in the preoperative planning of a microsurgical flap, in vascular evaluation of recipient arteries, with a thorough analysis of the extent of the lesion, previous vascular disease, and the perfusion status of the affected limb [1, 2]. Computed tomography angiography (CTA), a less invasive exam that allows a more accurate assessment of the initial vascular status in acute limb trauma [2–4], is currently replacing conventional angiography. However, there is no protocol that defines the cases that require a preoperative vascular assessment using CTA.

Few studies have evaluated whether preoperative angiographic vascular changes influence the outcomes of microsurgical flaps. Haddock et al. [5] showed that in cases where the posterior tibial artery is the recipient vessel, the rate of flap failure was higher despite the lesion of anterior tibial artery being the most common in CTA. Duymaz et al. [2] concluded that the presence of abnormalities in preoperative CTA was a risk factor for amputation (p <  0.05). Stranix et al. [1] evaluated flaps for lower limb and concluded that the presence of arterial lesions on preoperative CTA was associated with a higher rate of complications, a finding similar to Lee et al.’s [6], but this in the pediatric population.

The aim of the present study was to evaluate the occurrence of major complications (revision of the anastomosis, partial or total flap failure, and amputation) of microsurgical flaps in upper and lower limb trauma in two groups: with or without preoperative CTA. In the group in which CTA was performed, we describe the lesions found and statistical analysis of the findings in relation to the occurrence of complications. The main hypothesis of this study was that the rational use of preoperative CTA is a useful tool in planning microsurgical flaps.

2 Methods

A retrospective cohort study was performed in all consecutive patients who underwent limb reconstruction with microsurgical flap at the Department of Orthopaedics in University of Sao Paulo, between 2014 and 2020.

Inclusion criterion was patients with traumatic injuries of upper and lower limbs requiring microsurgical reconstruction with follow-up until flap healing. The exclusion criterion was absence of data in medical records on follow-up for at least one month after surgery. Variables of the patients’ demographic profile were analyzed, including age, weight, height, comorbidities, and type of trauma that occurred.

All patients were submitted to clinical examination with adequate perfusion and distal pulse in at least one main nutrient limb artery. The decision about to perform pre-operative CTA was made by the surgeon, through some considerations: clinical evaluation of the soft tissues and chronicity. The patients were then separated into two groups, according to pre-operative CTA: CTA+ and CTA –.

The presence of postoperative complications was evaluated, including major complications, classified as type III by Clavien–Dindo [7], indicating complications that required surgical reintervention (total or partial flap failure, take-back flap, infection, and amputation). Obesity was defined by the presence of body mass index (BMI) ≥30 kg/m² (WHO).

Other data evaluated were time elapsed between the trauma and microsurgical flap, type of flap, location of the lesion, type of arterial anastomosis performed (end-to-end versus end-to-side), recipient vessels of the vascular pedicle, indication of vein graft for arterial anastomosis, and flap ischemia time (time elapsed between section of the pedicle in the donor area and the release of arterial and venous clamps, with flap revascularization in the recipient area) in minutes.

All CTA were performed with a 128-multi-slice CT with iodine contrast, in two phases, with 3 mm of thickness.

In group CTA+, we also analyzed the following variables: number of patent arteries in preoperative CTA and description of the arterial injury.

Data were analyzed using SPSS 20.0. Qualitative data were analyzed using Pearson’s chi-square test or Fisher’s exact test, when the frequency was lower than expected. A p value <  0.05 was considered statistically significant. Level of Evidence: II (Retrospective cohort study).

3 Results

From 2014 to 2020, a total of 121 flaps were performed in 117 patients with complex limb defects requiring reconstruction. One patient was excluded due to lack of outpatient clinical follow-up. Of the 121 flaps, 84 (69.4%) were performed to reconstruct the lower limb and 37 (30.6%) the upper limb. Table 1.

Of the 121 flaps, 64 underwent preoperative CTA (CTA+) while 57 did not perform vascular investigation prior to surgery (CTA-). The indication of preoperative CTA was in cases when the surgeon wanted additional information of the exact anatomical place of artery occlusion or when clinical examination was not elusive.

Obesity was not a risk factor for complications in this study (p = 0.23), as well as the presence of comorbidities (p = 0.102).

When evaluating postoperative complications, type II of Clavien-Dindo, we observed a higher rate of complications in the CTA+ group. 45.3% of patients who underwent CTA had some type of postoperative complication, while in CTA–group only 26.3% had complications (p = 0.031; Odds Ratio = 2.32; Confidence interval = 1.1–5.0). There was also statistical significance between the patients who underwent preoperative CTA and total flap failure (p = 0.016; Odds Ratio = 5.71; Confidence interval = 1.2–26.9). But there was no difference between the two groups in partial flap failure (p = 0.122), take-back flap (p = 0.116), and amputation (p = 0.746).

When analyzing the time elapsed between the injury and the flap, we observed that most CTA+ patients were chronic cases (>  7 days), p = 0.001 (98.4% cases). Table 2.

We also observed a higher rate of complications in chronic cases, p = 0.034 (39.4% in chronic cases versus 8.3% in acute cases). Table 3.

In the 64 CTA+ patients, we were able to analyze the results and images of the exams for the presence of vascular lesions and which vessels were injured. Table 4.

The presence of at least one vascular lesion on CTA, regardless of whether the vessel was used or not for microanastomosis, was not a risk factor for major complications in general, p = 0.981 (45.5% versus 45.2%). Figure 1.

OPEN IN VIEWER

Fig. 1

CTA of a vascular injury of patient with anterior tibial artery lesion, the most common pattern of lower limb injury, observed in our study.

When assessing the number of patent arteries (1, 2 or 3) in lower limbs specifically or in conjunction with upper limbs, there was no statistical association with a higher rate of complications in general or specifically total flap failure and revision of the anastomosis. Table 5.

We recorded five amputations in our series. The CTA+ group had 4 amputations (3 due to infectious and only 1 due to thrombosis of the microanastomosis). In CTA-, only 1 amputation was recorded, due to infection approximately one month after the healed flap.

Approximately 1.35 million people die each year in traffic accidents, which represents an annual cost of approximately 3% of the gross domestic product of most countries [8]. Traffic accidents are responsible for 82% of traumatic cases with indication for microsurgical reconstruction in our hospital; these high-energy traumatic wounds have a higher risk of vascular damage to recipient vessels from microanastomosis. For this reason, CTA exams are indicated to optimize preoperative planning. However, there is still no protocol for indicating this exam and whether CTA is a useful tool for microsurgery.

As we consider the high costs of performing CTA, in our paper we proposed to study whether CTA provides better flap outcomes. We found that early vascular complications, including total flap failure, were more frequent in patients who underwent CTA (45.3%) than in those without (26.3%). However, CTA is not performed routinely for all cases that require skin coverage in our service, and in group CTA+ we observed a higher rate of complications, which can represent a selection bias as the most complex and chronic limb injuries may have CTA indicated more frequently.

The study demonstrated the influence of chronicity and complications: CTA was indicated in 57.8% of chronic cases (>  7 days) vs. 8.3% in acute cases (p = 0.001). We also found a higher rate of complications in chronic cases, p = 0.034 (39.4% versus 8.3%), which can be justified by the poor quality of recipient vessels and surrounding with fibrosis, and tendency to thrombosis due to undetected endothelial lesion. In the literature, the rate of chronic free flap failure [9] and infection rates [10] were also higher when compared with traumatic wounds covered before 7 days from trauma. Ninkovic et al. [11] defined a chronological classification for coverings with microsurgical flap in three categories, similar to simple wounds, which is the reference used in our study to define “acute <  7 days” and “chronic >  7 days.” Therefore, although some authors advocate serial debridement and coverage between 7 and 14 days of trauma [12], the advantages of early microsurgical flap are currently well established and include simultaneous bone and soft tissue treatment, less risk of infection, and shorter hospital stay and hospital cost [13]. Our study showed a higher rate of flap failure if performed after 7 days, which suggests a relationship between the time of exposure of wound and complications such as infection, flap failure, or amputation, emphasizing the importance of adequate coverage to obtain better functional and aesthetic results.

In evaluating patients with preoperative CTA in our study, the presence of lesion of at least one vessel on CTA was not a risk factor for major complications (total or partial flap failure, revision of anastomosis, and amputation), p = 0.981, with a rate of complications very similar between groups with or without arterial injury (45.5% vs 45.2%). This data contrasts with the findings of Stranix et al. [1], who evaluated 361 microsurgical flaps (only open fracture of the Gustilo III tibia) with preoperative CTA in 243 (67%) and found that the presence of any arterial injury, even when this artery was not used in anastomosis, was associated with a higher rate of complications. Lee et al. [6] evaluated a pediatric population with 53 flaps after trauma and found arterial injury on CTA in 35.8% of cases, also with a higher flap failure rate in this group than in patients without injury on CTA. Stranix et al. [1] also observed that when the injured artery was used as a recipient vessel, there was a statistically significant association with an increase in the percentage of complications, re-exploration of the anastomosis, and total or partial flap failure and concluded that the use of microanastomosis on an injured artery more than doubled the risk of complications. In our study, however, among the 64 flaps with preoperative angiographic evaluation, in 13% of cases a damaged vessel in CTA was used for microanastomosis proximally, with one total flap failure among them.

We have observed in the last decades a reduction in the number of primary amputations after complex traumatic injuries of the limbs. Secondary amputations usually occur after failure of microsurgical reconstruction, infections, and irreparable bone injuries; in our service, primary amputations in these injuries are less common. In our series, five amputations (4.1%) were necessary; of these the most common cause was infection in 80% of cases and only one after thrombosis. Duymaz et al. [2] observed 5.2% of amputations in their series of 76 patients, with at least one occluded artery in angiographic exams, concluding that abnormal preoperative CTA was a predictive factor for amputation.

There is not enough evidence in literature on the effectiveness of preoperative exams in surgical planning and postoperative outcomes and which imaging exam provides the best anatomical information of vessels, which could help to reduce postoperative complications after free flap reconstruction of limbs. CTA with multidetector system is a reliable angiographic technique [14] and has an acceptable rate of specificity and sensibility when compared with Doppler ultrasound and contrast enhanced magnetic resonance imaging (MRI) [15]. However, in our study we could not provide the usefulness of CTA in preventing microvascular complications that may lead to total flap loss; therefore Doppler ultrasound could be a good alternative and with reduced cost compared with CTA and contrast enhanced MRI. But Doppler ultrasound also has limitations as it is highly operator dependent and provides little additional information on the complete vascular map of the damaged limb.

The limitations of this study are the possibility of the groups not being homogeneous in the characteristics of the lesion, and indication for CTA being performed by the individual decision of each surgeon and could not be randomized.

Preoperative CTA can help in planning the recipient artery and the best location of microanastomosis at a level far from the injury zone. However, the anatomical intraoperative assessment of the vessel and arterial flow is even more important. The findings of this study reinforce the idea that more important than the quantity of available vessels in a traumatized limb requiring microsurgical reconstruction is the quality of the recipient vessel. Therefore, preoperative CTA must be performed with discretion and should be considered a tool for surgical planning in chronic cases. In these cases, the CTA can help in deciding on which recipient artery to perform microanastomosis, as well as its best location, avoiding anastomosis in the proximity of the lesion, as the good quality of the recipient vessel is a fundamental factor for the success of the microsurgical flap [16–19].

5 Conclusion

CTA should not be performed routinely, however, CTA may help in surgical planning, especially in complex cases of high-energy and chronic cases, as it provides information on the best recipient artery and the adequate level to perform microanastomosis, outside the lesion area.

Conflict of interest

The authors have no conflict of interest to report.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authors contribution

LSM and RBI: Wrote the paper and edited the manuscript.

RBI, LSM and RPR: Evaluation of medical record and Statistics.

FGBS, ABC, THW, MRR and RBI: Performed the surgeries.

THW, MRR, RMJ: Supervision.