Postoperative management of antithrombotic medication in microvascular head and neck reconstruction: a comparative analysis of unfractionated and low-molecular-weight heparin

Microsurgical free-flap reconstruction following ablative surgery is a frequently used procedure for a large resection volume and maintains the functional outcome in patients with head and neck cancer [1]. The opportunity for free tissue transfer has many advantages over pedicled flaps if primary wound closure is not possible. A maintained volume leads to a better function of speech and swallowing [2, 3]. The outcome is dependent on the survival of the graft and proper integration into the adjacent tissue. The thrombosis-related reduced perfusion of the flap with consecutive necrosis of the tissue is a dreaded complication and a severe surgical emergency that necessitates immediate revision [1]. The majority of complications are attributed to thrombosis of the venous drainage [4]. In general, thromboses of the vascular pedicle are reported in up to 10% of micro anastomoses [5]. To prevent flap thrombosis, the use of anticoagulants appears to increase flap survival, which has been the objective of recent studies [6‐8]. Heparin, dextran, and aspirin are often used anticoagulants in the management of microvascular reconstruction. Unfractionated heparin (UFH) is applied intravenously and inhibits circulating thrombin. Thrombin converts fibrinogen to fibrin, which leads to the formation of cross-linked fibrin, promotes platelet aggregation, and activates blood clotting factors V and VIII [9]. Low-molecular-weight heparin (LMWH) has an inhibitory effect on activated factor X. However, anti-thrombin activity is less intense [9]. Observations in animal studies showed the ability of LMWH to reduce thrombosis formation in microvascular anastomoses [10]. Few clinical studies are suggesting an improvement in outcome [11]. However, the superiority of a therapy regimen is still not evident [9, 11, 12]. The purpose of this study was to distinguish the difference between flap survival and postoperative complications depending on the use of intravenous low-dose UFH and subcutaneous LMWH in patients undergoing free-flap reconstruction in the head and neck area, to provide a recommendation for clinical practice.

This is the first study comparing low-dose UFH and LMWH in patients with a free microvascular reconstruction of the head and neck. Based on the current data, we can assume comparable eligibility of low-dose, intravenously applied UFH, and subcutaneously applied LMWH. With an overall flap survival rate of 94.4%, the results presented are in line with those of the current literature [12, 15]. We could not find any significant difference regarding the flap survival rate (P = 0.280) and flap thrombosis (P = 0.767) in both groups (Table 2). Moreover, with an overall revision rate of 20.3%, the results of both therapy groups are also comparable (P = 0.912). Revascularization of the pedicle was archived in every case of revision surgery.

In most cases, thrombosis of the vascular pedicle is caused by an interruption of the venous outflow. A thrombus formation in the venous branch, a kink of the pedicle, or the pressure exerted by a hematoma formation are possible causes for venous congestion. Determined by experimental and clinical observations, arterial vessel occlusion occurs, preferably at the site of anastomosis, mostly dependent on fibrin formation [4]. Therefore, anticoagulative drugs seem to improve flap survival by reducing the probability of thrombus formation. Besides, the “ex vivo” irrigation of the transplant is recommended after flap raising [16]. There are several studies concerning the use of anticoagulants in the perioperative management of patients undergoing free microvascular tissue transplantation. However, the efficacy is not finally apparent [11].

Low-dose UFH is used in some reconstructive centers as standard medication for thrombosis prophylaxis. Chalian et al. administered 300 IU heparin/kilogram body weight/day by permanent infusion, confirming a significantly higher thrombosis rate using the external jugular vein compared to the internal jugular vein [17]. Overall, flap survival was 96%. Pohlenz et al. conducted a retrospective study referring to the anticoagulation with 200 IU heparin/kilogram body weight/day via infusion. Even with this lower dosage, a flap survival rate of 97.1% could be achieved [18]. However, the rate of postoperative medical, especially pulmonary and cardiac complications, was 11.4%. In a comparative study, Deutinger et al. examined the superiority of subcutaneous heparin in combination with dextran from intravenous administration of UFH at a dose of 500–800 IU/h as a continuous infusion over 24 h, concerning no significant impact on flap thrombosis (7.4% versus 6.5%; P = 0.790) [19]. This dosage of UFH is comparable to that of our study (500 IU/h).

In the regression analysis, we examined many different accompanying factors that could be correlated with flap failure. None showed a significant influence on flap survival. This is in line with an extensive study by Kroll et al. from the 1990s. On 854 patients, it was demonstrated that factors, such as age, smoking, and prior irradiation, in particular, do not affect flap survival [20]. Because venous thrombosis is the most common cause of vascular complications, many surgeons recommend anastomosing two veins [21‐23]. Interestingly, in regression analysis, we could not find any advantage of dual or triple venous anastomosis compared to a single-vein anastomosis (P = 0.904). The reason for this could be the use of the microvascular coupler in more than 80% of cases because it achieves advantages over the sutured anastomosis even in single-vein anastomosis [24‐26]. Despite the use of a microvascular coupler, postoperative thrombosis prophylaxis is indispensable.

LMWH is currently the most commonly used postoperative thrombosis prophylaxis. It is a derivative of unfractionated heparin produced by the determinative hydrolysis of standard heparin into short polysaccharide fragments. These molecules have the same inhibitory effect on the active factor X, but a weaker anti-thrombin activity. The half-life is approximately 4 h [9]. A retrospective analysis by Reiter et al. showed favorable results with a flap survival rate of 97.1% using subcutaneously applied LMWH (20 mg or 40 mg) [12]. A drawback is a lack of information about medical complications and the absence of a control group. Zhou et al. conducted a randomized, prospective trial, determining no significant difference in flap failure (P = 0.615) between patients with aspirin and dextran, patients with LMWH, and those who did not receive any antithrombotic medication at all [27]. However, it was remarkable that patients with LMWH significantly more often developed hematomas in need of revision (P = 0.032). An et al. showed similar results in a cohort of 101 patients who did not receive anticoagulative therapy after microvascular flap reconstruction. The comparison group included 33 patients receiving UFH (2.500 IU in the operation room and postoperative infusion of 200 IU/h for 7 days). The authors could not find any significant differences in the rate of flap necrosis (15.84% versus 18.18%; P = 0.75) and hematoma formation (13.86% versus 15.15%; P = 0.66) [28]. However, there is no mentioning of information on the type of pedicle thrombosis, the number of revision operations, and medical complications, such as pulmonary artery embolism.

The development of a hematoma, which can lead to secondary thrombosis due to compression of the vascular pedicle, is a feared complication. In our study cohort, we observed a hematoma rate of 4.6% and 3.9% in patients who received either UFH or LMW (P = 0.792; Table 2). These results are also consistent with those of other studies [18]. Besides, Kroll et al. found a hematoma rate of 5.3% in patients who did not receive anticoagulation medication undergoing microvascular free-flap transplantation [29]. Furthermore, there was no cause-and-effect relationship between the use of anticoagulants and flap loss or prevention under the application of low-dose UFH (100–400 IU/h).

We can confirm that the low-dose application of UFH in a dosage of 500 IU/h has no significant impact on coagulation parameters, but showed a tendency for increased APTT (e.g., POD1–3, 42.3 ± 12.6 s) compared to the LMWH group (38.6 ± 10.3 s, P = 0.077; Table 3). Regarding the coagulation parameters, we could not find significant differences between patients with and without postoperative bleeding in both prophylaxis groups of our study cohort (Table 4). Concerning postoperative medical complications, we found one patient in the UFH group, receiving pulmonary artery embolism (Table 2). Although this was a single incident, there might be a significant result with an increase in the number of cases. Furthermore, there was no case of deep vein thrombosis.

The presented results need to be interpreted, considering the retrospective character. However, our cohort included every free-flap reconstruction that fulfilled the inclusion criteria at our department between 2004 and 2017. Thus, two groups of patients with a similar distribution of relevant oncological, medical, and surgical parameters were obtained, which reduces the inevitable selection bias. Therefore, a comparison of postoperative thrombosis prophylaxis was possible. Furthermore, there was no negative control group that could prove the superiority of anticoagulative therapy over patients without postoperative thrombosis prophylaxis. In our opinion, however, this comparison is questionable in a randomized, prospective study, since a malignant disease is always a risk factor for the development of thrombosis. Moreover, oncological tumor resections with microvascular reconstructions are associated with long operating times and represent a sole risk factor. Even if there is no clear statement on the perioperative risk of thrombosis in head and neck patients due to a lack of randomized clinical studies, perioperative thrombosis prophylaxis is recommended according to the German guideline [30]. Determining the effect of anticoagulant medication on flap survival is always critical as the outcome depends on multiple factors. The choice of the appropriate graft, a precise surgical technique, and rigorous postoperative monitoring are the keys to a high success rate in microvascular surgery.

Even if we could not determine a significant superiority of one of the anticoagulants, in general, we recommend postoperative thrombosis prophylaxis. According to the recommendations of the current guideline, priority should be given to LMWH over UFH as LMWH has improved pharmacological properties, a lower risk of side effects, better bioavailability with a longer half-life, and reasonable practicability for once-daily administration [29]. However, it should be noted that therapy monitoring and the detection of overdose by determining the APTT are not possible when applying LMWH. In patients with a high risk of postoperative bleeding, we prefer UFH because it is easier to control due to the possibility of titration, a shorter half-life compared to LMWH, and the possibility of antagonism with protamine. Mainly due to the high risk of hematoma development in the neck, this is of importance since such a complication can cause thrombosis of the vascular pedicle. Patients with pedicel thrombosis also receive UFH via infusion, usually within the APTT range of 50–60 s. In conclusion, we can summarize that the choice of thrombosis prophylaxis is still an individual case decision in high-risk patients as the reconstruction of complex three-dimensional defects of the upper aerodigestive tract using free microvascular grafts is still challenging.

Based on the present data, we cannot demonstrate the superiority of low-dose UFH or LMWH concerning flap survival and the occurrence of postoperative complications. The results after the application of UFH or LMWH are consistent with those reported in the current literature. More important key factors seem to be the surgical technique and careful postoperative monitoring for at least 5 days, as well as the decision to perform immediate surgical revision in case of any sign of thrombosis, which is crucial for flap survival.