Intraflap Vascular Catheterization Method for Monitoring, Prevention, and Intervention of Thrombogenesis in Free-Flap Surgery

Abstract

Background 

Thrombosis at the anastomotic site is a significant problem in free tissue transfer with microvascular anastomosis. We report a newly developed intraflap vascular catheterization (IFVC) technique for monitoring hemodynamics, prevention of thrombogenesis, and transcatheter intervention of free-flap thrombosis.

Methods 

We performed a hospital-based, prospective study. Ninety-three patients underwent free tissue transfer by a single surgeon in a single hospital. In the IFVC group (n = 40), catheters were inserted into the arterial and venous branches of the flap main pedicle vessels near the anastomoses. The catheters were connected to the pressure monitor. A bolus injection of urokinase was administered every hour to the artery, and a continuous infusion of saline was initiated to the vein. The bolus injection of urokinase solution reached the arterial anastomosis by the retrograde flow. During the postoperative period, rapid injection of urokinase or saline was performed according to the pressure monitor. Intraflap vascular catheterization monitoring was performed postoperatively for 72 hours.

Results 

The overall flap survival rate in the IFVC group was 100% (40 of 40), whereas the overall flap survival rate in the non-IFVC group was 96% (51 of 53). In a subgroup analysis of lower extremity reconstruction, the flap survival rate was 100% (22 of 22) with no cases of reanastomosis requiring a return to the operation room in the IFVC group. By contrast, the flap survival rate was 92% (22 of 24), with 6 cases of reanastomosis requiring a return to the operation room in the non-IFVC group (P = 0.04).

Conclusions 

The IFVC method enables monitoring, prevention, and intervention of thrombi at anastomotic sites of the free flap. Intraflap vascular catheterization may increase free tissue transfer success rate, especially in high-risk cases, such as free-flap reconstruction after the lower extremity trauma or venous leg ulcer.

Thrombosis at the anastomotic site is a significant problem in free tissue transfer.^1,2 Various methods for detecting postoperative thrombosis have been reported. These methods include clinical monitoring,³ implantable Doppler probe,⁴ microdialysis,⁵ oximetry,^6,7 visible light spectroscopy,⁸ multispectral imaging,⁹ CO₂ monitoring,¹⁰ laser Doppler flowmetry,¹¹ fluorometry,¹² temperature measurement,¹³ glucose and lactate measurements,¹⁴ smartphone applications and telecommunications,¹⁵ photoplethysmography,¹⁶ handheld Doppler ultrasound,¹⁷ impedance plethysmography,¹⁸ sidestream dark-field imaging,¹⁹ nuclear medicine,²⁰ pH measurement,²¹ hydrogen clearance,²² and catheter insertion to the flap vessels.^23,24

Monitoring the flap through an inserted catheter is one of the most invasive methods. However, this approach provides direct and rapid information about the hemodynamic status of the flap. Another advantage of the catheter method is that it allows us to monitor, prevent, and treat thrombi. We developed a new method for free-flap monitoring called the intraflap vascular catheterization (IFVC) technique for artery and vein. In this method, catheters are inserted near the anastomosis. Intraflap vascular catheterization enables us to monitor the flap highly sensitively, prevent thrombosis, and treat small thrombi by injecting drugs from the catheters. We report our clinical experience using the IFVC method.

METHODS

Study Design, Setting, and Ethics

We performed a prospective hospital-based study. The study protocol was approved by the ethical committee of St Mary's Hospital (Approval Number 17-1708). Written consent was obtained from all patients. All studies were performed according to the guidelines of the Declaration of Helsinki.

Patients and Study Protocol

Between 2010 and 2018, 93 patients underwent free tissue transfer by a single surgeon at St Mary's Hospital. The non-IFVC group was 53 consecutive cases before starting the IFVC technique. The IFVC group was 40 consecutive cases after starting the IFVC technique. After starting the IFVC technique, the IFVC technique was applied to all cases that underwent free flap surgery.

Patient age, sex, cause of reconstructive surgery, reconstructed site, type of flap, number of reanastomoses requiring return back to the operation room, and flap survival were recorded. In the non-IFVC group, postoperative flap monitoring was performed by an experienced nurse every 1 hour for 72 hours postoperatively by monitoring color and capillary refilling of the flap. If there was suspicion about flap circulation, the surgeon was called, and reoperation was performed.

In the IFVC group, catheters inserted into the branches of the artery and vein of the flap near the anastomoses were used for monitoring, prevention, and intervention. The IFVC protocol follows.

From Flap Elevation to Catheter Placement

The branches of the artery and vein bifurcating the main pedicle near the planned microvascular anastomosis were marked and prepared to be longer than 10 mm in length while elevating the flap. The branches were prepared for later catheter insertion using microclips.

After the flap vessels were microscopically anastomosed to the recipient vessels, and before blood circulation was restored, the catheters were inserted into the prepared branch vessels of the flap. The intravessel catheters were 22- or 24-gauge intravenous catheters (BD Insyte IV Catheter; Becton, Dickinson and Company, NJ). After inserting the catheters into the vessels, the vessel wall was ligated with the catheter to prevent blood leakage. The catheters were connected to a pressure monitor using pressure transducers. The arterial catheter was connected via a 3-way stopcock to an infusion pump to administer 120 U/mL urokinase (Fig. 1). The venous catheter was connected via a 3-way stopcock to an infusion pump to administer 0.9% saline (Fig. 1). The base of each catheter was fixed tightly to the skin or wound bed (Fig. 2A). Zero calibrations were taken at the right atrium.

FIGURE 1: Schema of IFVC method. Catheters are inserted into the branches of the pedicle vessels of the flap. The branches in which the catheters are inserted are located near the anastomoses. The system allows monitoring, prevention, and intervention for thrombogenesis.

FIGURE 2: Photographs of the IFVC system. A, Catheters inserted in the flap artery (yellow clip) and vein (green clip). B, Hemodynamic monitoring. Continuous monitoring of systemic arterial pressure represented by the radial arterial pressure, flap arterial pressure, and flap venous pressure are shown.

Declamping of the Anastomotic Vessels

The vessels were then declamped, and circulation was restored at the flap. The pressure waveform was seen on the monitor (Fig. 2B). Intermittent rapid infusion of 3 mL of urokinase solution at a concentration of 120 U/mL every hour for the artery and continuous infusion of saline at 20 mL/h for the vein were started (Fig. 3). Regarding the artery, although the catheter is inserted at the flap side, a bolus injection of urokinase solution can reach the anastomosis site in a retrograde fashion. This phenomenon can be directly observed during the operation (see Video, Supplemental Digital Content 1, https://links.lww.com/SAP/A689, which demonstrates washing of arterial anastomosis by retrogradely injected urokinase solution).

FIGURE 3: Shema of catheter intervention using IFVC system. Artery: Bolus injection of urokinase solution can reach the anastomosis by retrograde flow and exert a thrombolytic effect. Vein: Bolus injection of saline can wash out the thrombus at the anastomosis.

Start of Observation Protocols

After declamping the recipient vessels, the observation protocols were initiated. The monitor alarm was set with a reference pressure for the artery and vein. The arterial reference pressure is defined as the flap's lowest normal arterial systolic pressure with no obstruction to blood flow. The arterial reference pressure was set at 50 mm Hg, which is close to the minimum pressure needed to secure the circulation in the flap.²⁵ The venous reference pressure is the highest limit of the normal venous mean pressure of the flap with no obstruction to blood flow. The venous reference pressure was 40 mm Hg, near the maximum pressure to secure flap venous return.²³ Because patients spend 72 hours in the supine position after surgery, the reference pressure was the same among the patients regardless of the location of the flap.

The wound was closed in standard fashion, taking care not to displace the catheters. The catheters were entirely buried in the wound when the anastomosis was located in deep space. Similarly, if regular monitoring through catheters could not be achieved because of kinking or other reasons when catheters were placed outside of the wound, the catheters were buried entirely in the wound.

Observation During the Postoperative Period

When the pressure monitor showed a decrease in flap arterial pressure or an increase in flap venous pressure despite normal flap color, catheter problems should be considered first. Possible catheter problems include catheter obstruction, catheter displacement, and catheter kinking. If a catheter problem was ruled out, 3 mL of urokinase at a concentration of 120 U/mL for the artery or 5 mL of 0.9% saline for the vein was rapidly injected (see Videos, Supplemental Digital Content 2 and 3. Supplemental Digital Content 2, https://links.lww.com/SAP/A690, demonstrates increasing flap arterial pressure after bolus injection of urokinase solution from the IFVC system. Supplemental Digital Content 3, https://links.lww.com/SAP/A691, demonstrates decreasing flap venous pressure after bolus injection of saline from the IFVC system). If pressure was restored, we maintained observation. However, if pressure was not recovered despite catheter intervention, a bolus injection of urokinase solution for the artery and saline for the vein were performed 5 times maximally. If pressure was not recovered to the normal range after 5 rapid injections, re-exploration was considered.

Duration of IFVC Monitoring and Removal of Catheters

Intraflap vascular catheterization was continued for 72 hours postoperatively. After a 72-hour monitoring period, branches in which catheters were inserted were ligated at the time of catheter removal. When catheters were placed in a superficial and easily accessible area, as in extremity reconstruction cases, the removal and ligation procedure was performed in the ward. On the other hand, when catheters were placed at deep space inside the body, as in breast reconstruction cases, the procedure was performed in the operation room.

Statistical Analysis

Continuous variables are presented as the mean ± standard deviation. Continuous variables were compared with Student t test under the condition that Levene test could assume equal variances. Mann-Whitney U test was used if Levene test could not assume equal variances. Categorical variables were compared with Fisher exact probability test. All P values quoted are 2-tailed. P values less than 0.05 were considered significant. Statistical analyses were conducted using SPSS software (Version 26; IBM, Armonk, NY).

RESULTS

Patient characteristics are shown in Table 1. There were no significant differences in age, sex, and surgery cause between non-IFVC and IFVC groups. There was a statistically significant difference in the recipient site. Compared with the non-IFVC group, the IFVC group had fewer upper extremity recipient sites, more breast recipient sites, and more trunk recipient sites (P < 0.001).

TABLE 1 - Patient Characteristics

	Non-IFVC Group (n = 53)	IFVC Group (n = 40)	P
Age, mean ± SD	47 ± 19	46 ± 20	0.74
Sex			0.45
Female	10	11
Male	43	29
Cause (cases)			0.167
Trauma	36	23
Osteomyelitis	8	6
Malignant tumor	3	9
Venous leg ulcer	2	0
Arteriovenous malformation	1	0
Scar	2	2
Hemifacial microsomia	1	0
Recipient site (cases)			<0.001
Head and neck	6	4
Breast	0	6
Trunk	0	3
Upper extremity	23	5
Lower extremity	24	22

The results of all reconstructive surgeries are shown in Table 2. The additional operation time for IFVC is approximately 30 to 60 minutes. There were no cases of catheter-related hemorrhagic complications. There were no cases of systemic or local hemorrhagic side effects due to urokinase or saline injection. There was a statistically significant difference in the types of flaps used in the non-IFVC group versus the IFVC group. In the IFVC group, there were more deep inferior epigastric perforator flaps, fewer superficial circumflex iliac artery flaps, fewer anteromedial thigh flaps, and fewer anterolateral thigh flaps compared with the non-IFVC group (P = 0.001). The number of cases requiring reanastomosis of the vessels and return to surgery was significantly lower in the IFVC group compared with the non-IFVC group (P = 0.01). Reanastomosis was performed on 2 arteries and 7 veins in the non-IFVC group, whereas there were no cases in the IFVC group. Six cases were lower extremity reconstruction after trauma, and 3 cases were upper extremity reconstruction after trauma. Two cases resulted in total flap loss: one case was arterial thrombus of ALT flap for lower extremity reconstruction after trauma, and the other case was venous thrombus of AMT flap for lower extremity reconstruction after venous leg ulcer. Flap survival rate was 96% (51 of 53 cases) versus 100% (40 of 40 cases) in the non-IFVC group versus IFVC group, respectively (P = 0.50).

TABLE 2 - Results of All Reconstructive Surgeries

	Non-IFVC Group (n = 53)	IFVC Group (n = 40)	P
Type of free flap (cases)			0.001
LD	0	1
TAP	0	1
VRAM	0	1
DIEP	0	6
DCIA	1	0
SCIA	4	0
TFL	1	2
ALT	38	28
AMT	7	1
MT	2	0
Reanastomosis* (cases)			0.01
Artery	2	0
Vein	7	0
Flap survival (cases)	51 (96%)	40 (100%)	0.50

*Reanastomosis with return to the operation room.

ALT, anterolateral thigh; AMT, anteromedial thigh; DIEP, deep inferior epigastric perforator; DCIA, deep circumflex iliac artery; LD, latissimus dorsi; MT, medial thigh; SCIA, superficial iliac perforator artery; TAP, thoracoacrominal perforator; TFL, tensor fascia lata; VRAM, vertical rectus abdominis muscle.

In lower extremity reconstruction subgroup analysis, there were no significant differences in age, sex, cause of surgery, and type of flap between the non-IFVC group and IFVC group (Table 3). The number of cases requiring reanastomosis with a return to the operation room was significantly lower in the IFVC group than in the non-IFVC group (P = 0.04). There were 2 cases of arterial reanastomosis and 4 cases of venous anastomosis in the non-IFVC group, whereas there were no cases in the IFVC group. Flap survival rate was 92% (22 of 24) versus 100% (22 of 22) in the non-IFVC group versus IFVC group, respectively (P = 0.49).

TABLE 3 - Results of Lower Extremity Reconstruction

	Non-IFVC Group (n = 24)	IFVC Group (n = 22)	P
Age, mean ± SD	43 ± 20	44 ± 23	0.83
Sex			1.00
Female	2	2
Male	22	20
Cause (cases)			0.69
Trauma	15	17
Osteomyelitis	4	4
Malignant tumor	2	0
Venous leg ulcer	2	0
Scar	1	1
Type of free flap			0.07
SCIA	2	0
TFL	1	2
ALT	14	19
AMT	6	1
MT	1	0
Reanastomosis* (cases)			0.04
Artery	2	0
Vein	4	0
Flap survival (cases)	22 (92%)	22 (100%)	0.49

*Reanastomosis with return to the operation room.

ALT, anterolateral thigh; AMT, anteromedial thigh; MT, medial thigh; SCIA, superficial iliac perforator artery; TFL, tensor fascia lata.

Regarding the frequency of monitor alarms in the IFVC group, there were few alarms for both arterial and venous monitors until 24 hours after surgery. From 24 to 72 hours after surgery, the frequency of monitoring alarm increased compared with that in the period up to 24 hours after surgery. Bolus injection to the artery was performed in 5 cases, whereas bolus injection into the vein was performed in 7 cases. In 1 case requiring venous bolus injection, re-exploration was performed because of continuously increased venous pressure despite bolus injection. In that case, during the re-exploration, the hematoma was found around the anastomoses. After removal of the hematoma, venous pressure was successfully decreased, and reanastomosis was not necessary. Other than that case, all bolus injections, either for artery or vein, were successful.

Removal of catheters in the ward was performed in 14 cases, whereas removal of catheters in the operation room was performed in 26 cases. There were no complications related to catheter removals, such as bleeding, difficulty in removal, or rupture of anastomoses.

DISCUSSION

We herein report our experience using the IFVC method for free-flap transfer in 40 consecutive cases. Intraflap vascular catheterization allowed continuous real-time monitoring of arterial and venous pressure of the flap, prevention of thrombogenesis, and transcatheter intervention of thrombogenesis. As a result, we could significantly reduce the need for reoperation for reanastomosis. We also achieved a flap survival rate of 100% in 40 consecutive cases of IFVC, including 22 cases of lower extremity reconstruction.

Various methods have been reported for monitoring free tissue transfer.^1,2 Most reports have focused on minimally invasive methods. Thus, few reports are available regarding IFVC because of its invasiveness. Hudson et al²⁶ reported using a catheter inserted into a side branch of the vein in the free flap. Through the catheter, heparin was infused continuously to the venous anastomotic site for thrombus prevention. However, they did not use the catheter for venous pressure monitoring. Only 3 reports of clinical application of intraflap vessel catheter monitoring were found.^23–25 Matsumine et al²⁴ reported continuous venous pressure monitoring and heparin infusion using a catheter inserted into the pedicle vein stump opposite of the venous anastomosis, side branch of the main pedicle vein, or subcutaneous vein of the flap for 72 hours. Two reports from Sakurai et al^23,25 are the only ones to show intraflap vessel pressure monitoring of both the artery and vein. They use catheters inserted into the distal ends of the pedicle vessels in the transferred tissues. Heparin was continuously infused through the venous catheter to the venous anastomotic site to prevent thrombogenesis. Venous and arterial pressure was continuously measured through the catheter for 72 hours postoperatively. In those three reports, catheters were inserted far from the anastomosis. By contrast, in our method, the catheters were inserted in the branches near the anastomosis. Placing catheters near the anastomosis makes it possible to monitor, prevent, and treat thrombi directly at the anastomosis site. The direct intervention of the arterial anastomosis site can be achieved by rapid injection of urokinase solution from the branch of the flap artery in a retrograde fashion (see Videos, Supplemental Digital Content 1 and 2, https://links.lww.com/SAP/A689 and https://links.lww.com/SAP/A690). The pressure monitoring showed immediate resolution of increased venous pressure, indicating thrombogenesis at the anastomotic site by bolus injection of 5 mL of saline (see Video, Supplemental Digital Content 3, https://links.lww.com/SAP/A691).

There are several reports about pharmacological agents used for thrombus prevention or thrombolysis in free tissue transfer in humans or animals.^26–34 One of the aims of IFVC is early thrombolysis, so thrombolytic agents such as urokinase or recombinant tissue plasminogen activator (rt-PA) are suitable. Urokinase was selected because it is prohibited to use rt-PA in a patient who has just undergone surgery. Another reason is that urokinase is less expensive than rt-PA. As for the safety of urokinase usage, urokinase was used as an intermittent rapid infusion of 360 U every hour, which means 8640 U/d in the IFVC protocol. On the other hand, urokinase doses approved for other diseases in our country are as follows: for cerebral thrombosis, 60,000 U of intravenous injection once a day for 7 days; for occlusion in peripheral arteries or veins, 240,000 U of intravenous injection once a day for 7 days; and for coronary artery, thrombosis in acute myocardial infarction 960,000 U of intravenous injection once. We think that the urokinase dose in the IFVC protocol is safe at least in systemic effect because it was much lower than the approved urokinase dose for other diseases. However, careful observation is needed because urokinase is administered at the surgical site in our protocol. There may be a possibility that the local concentration of urokinase is high enough to cause hemorrhage in the surgical site. The appropriate dose of urokinase for safe and effective flap salvage has yet to be established.

In the vein, we used continuous saline infusion. With surgical microscopic observation during the operation, we can directly observe that venous blood passing through the anastomosis is diluted by infused saline. We think that the dilution effect is important to avoid thrombus formation especially in veins, in which blood flow velocity is low.

The number of reanastomoses requiring repeat surgery was significantly reduced in the IFVC group compared with the non-IFVC group. We think that there are 2 explanations for this difference. First, the microsurgical skills of the surgeon are improved, with surgery in the IFVC group taking place later than in the non-IFVC group. Second, the effects of monitoring, prevention and intervention on thrombogenesis by IFVC during the postoperative period contributed to the difference.

Lower extremity reconstruction using free flap is more difficult than reconstruction of other body sites.³⁵ There are several challenges in the free-flap reconstruction of the lower extremity, including the limited number of candidates for recipient vessels, high venous pressure, thick venous wall, atherosclerosis, severe scar formation, and potential endothelial damage. The demand for early detection, prevention, and intervention of thrombus is higher in the free-flap reconstruction of the lower extremity than in free-flap reconstruction of other sites of the body. The subgroup analysis limited to lower extremity reconstruction in our study showed a flap survival rate of 100% (22 of 22) with no cases of reanastomosis with a return back to the operating room. Intraflap vascular catheterization seems to be an option for reducing the failure rate of free-flap lower extremity reconstruction. Although IFVC has the disadvantage of requiring a large amount of resources, we believe that the advantages outweigh the disadvantages in lower extremity reconstruction for trauma or venous leg ulcer.

The disadvantages of IFVC are increased operation time, risk of hemorrhage due to catheter insertion or urokinase use, and need for catheter removal. The additional operation time for IFVC is less than 60 minutes. We think that this additional time matches the value added by the high success rate of free-flap surgery with reduced possibility of reanastomosis, especially in lower extremity reconstruction. There were no cases of systemic or local hemorrhagic complications due to the IFVC method in our study. Catheter fixation is the most critical problem with IFVC. Fixing the catheter stably without leakage is essential to the success of the system. In some situations, the catheters must be embedded in the wound to be stably fixed. In these situations, the wound has to be reopened for catheter removal. This problem is one of the significant disadvantages of IFVC. Another limitation of the IFVC method is that it is difficult to quantify the amount of urokinase that can reach anastomosis. In addition, although we could find appropriate branches in all cases of the IFVC group in our experience, there is a possibility of the absence of appropriate branches to insert catheters near the anastomoses. The catheters cannot be inserted into branches less than 0.7 mm in diameter because the 24-gauge intravenous catheter with 0.7 mm in outer diameter is the smallest of the commercially available catheters.

We think that there is a possibility that IFVC can be applied to difficult reconstruction cases and research on flap physiology. One possible, but still not practiced, method of reconstruction with IFVC is continuous draining of the venous blood from the catheter when there is no appropriate recipient vein. In this way, angiogenesis from the surrounding tissue to the flap can be waited for. Using IFVC, we can constantly measure the flap venous pressure, reflecting a degree of progression of capillary networks that serve as venous outflow channels between the flap and the recipient tissue. Draining the venous blood from the catheter is more hygienic than phlebotomy from the surface of the skin or using medical leeches, and with appropriate processing, it may be possible to return the withdrawn blood to the patient, reducing the risk of transfusion.

Our study has several limitations. The study was based on a single surgeon experience in a single center with a small sample size. The IFVC group and the non-IFVC group were not decided by randomization. Consecutive cases before starting the IFVC technique were used as a historical control group. There are differences in the types of flaps used between the 2 groups. There can also be an improvement in the surgeon's skill because the IFVC technique was performed later than the non-IFVC group. There were no differences in flap survival rates between the 2 groups. The small sample size and many biases make it difficult to offer a definitive conclusion.

CONCLUSIONS

The IFVC method enables monitoring, prevention, and intervention for thrombi at anastomotic sites of the artery and vein at a free flap. Intraflap vascular catheterization may increase free tissue transfer success rate, especially in high-risk cases, such as free-flap reconstruction after the lower extremity trauma or venous leg ulcer.