Impact of intraoperative use of vasopressors in lower extremity reconstruction: Single centre analysis of 437 free gracilis muscle and fasciocutaneous anterolateral thigh flaps

1 Introduction

Free tissue transfer has without doubt become a very reliable tool to meet a variety of reconstructive requirements. Complication rates on the microsurgical anastomosis have been reduced in the last decades to fewer than 6 percent and have led to success rates of up to 99 percent [11, 18]. Nonetheless flap losses represent a devastating event not only for the patient, but also for the involved medical team. Many variables can interfere with the success of a free tissue transfer and most of them can be influenced by improving the perioperative setting. Flap complications are inevitably connected with irregularities among the vascular system, more precisely with changes in blood-flow and flap circulation. Cardiac comorbidities, blood loss, extension of operation time, sedation and anaesthesia, may lead to hypotension peri- or postoperatively [17]. This may require the use of vasopressors that counteract the drop of blood pressure. However, the use of vasoconstrictive medication in order to increase blood pressure is still controversial.

Although there have been many ground-breaking studies in the field of perioperative management for free tissue transfer, a true standard on the use of vasopressors has yet not been either developed or accepted.

In 2013 Gooneratne et al. [8] published that 52 percent of Scandinavian anaesthesiologists still prefer using crystalloid infusion to increase blood pressure in the perioperative setting of a free flap in head and neck surgery. At the same time, 35 percent of the questioned collective claimed that crystalloids were contraindicated in this setting.

Animal models have shown that vasopressors are influencing blood flow and flap perfusion negatively because of their vasoconstrictive effects. These findings, although, could not be supported by numerous retrospective clinical studies [6, 12]. Some studies published questionnaires, which try to represent more a tendency of what microsurgeons’ own experiences with vasopressors are nowadays [21]. Other studies showed the effects of vasopressors either only on one recipient site (i.e. breast, head and neck, etc.), [19, 20] on only one kind of flaps (e.g. muscle flap, fasciocutaneous, etc.), [3, 20] or on one ethiology of defects (i.e. cancer patients) [7]. The few existing prospective studies are also limited by low numbers of patients [6]. In 2015 Motakef et al. published a systemic review on literature in order to establish strategies to improve the perioperative management for free tissue transfers [16]. Besides other interesting results, they showed that vasopressors do no harm and may even increase blood flow to the flap (evidence level 1b). Reviewed papers either included porcine models or free flaps for only head and neck reconstruction.

However, not all areas of the body are affected the same by vasopressors. Truncal, head and neck, and extremity all have different physiologic changes based on vasopressor administration.

To this date, the exact effect vasopressors exert in the setting of microsurgical tissue transfers remains debated. In order to further elucidate the impact of intraoperative vasopressor administration on postoperative complications, a collective of 437 free gracilis muscle- and fasciocutaneous anterolateral thigh (ALT) flaps was reviewed. The study is novel in its design as it is confined to only two flap types, all utilized for lower extremity reconstruction and performed at a single institution.

2 Patients and methods

Over a six-year period (2012–2017), a total of 425 patients received 437 free gracilis muscle or free fasciocutaneous ALT flaps for lower extremity reconstruction after trauma, infection, vascular disease, or malignancies. The medical files and patient data were retrospectively screened for patients’ demographics, perioperative details, flap survival, and complications. A complete preoperative dataset for every patient to be included was mandatory. Exclusion criteria contained insufficient documentation within the dataset, unclear status or time of the usage of vasopressors, and patient age under 18 years old. Taking the possible influence of preoperative medical conditions on postoperative outcomes into consideration, the preoperative medical status of each patient was assessed and classified according to the American Society of Anesthesiologists (ASA) Classification of Physical Status [15]. The anesthesiologist in charge determined this score for every patient preoperatively.

Patients received vasopressors (norepinephrine) as an intravenous bolus during surgery when blood pressure dropped more than 20 percent from the preoperative baseline and was balanced individually with fluid resuscitation. The cohort was then assigned to the group with intraoperative vasopressor usage (V) or to the group without the use of any intraoperative vasoactive medication (NV). Both groups where than compared regarding perioperative complications. Surgical complications were accounted for and divided into major (total flap loss, arterial or venous thrombosis, hematoma and partial flap loss of >10 percent) and minor complications (wound dehiscence, skin graft failure, wound infection, and partial flap loss of <10 percent).

All operations were performed by five different microsurgeons. Patients were transferred to and stayed sedated at intensive care unit only if necessary. In the majority of cases they were referred to the normal unit and monitored every hour for the first 24 hours. Monitoring was performed clinically with documentation of flap temperature, capillary refill, turgor and color, as well as with hand held doppler sonography. On day two and three, documentation intervals extended.

The study was performed in accordance with the ethical guidelines of Clinical Hemorheology and Microcirculation [1].

3 Results

The V group included 306 patients (210 males, 96 females) with a mean age of 54.42 years (range 18–88 years, SD 17.82 years) and received 318 free lower extremity reconstructions (138 gracilis muscle flaps and 180 ALT flaps). The NV group contained 119 patients (70 males, 49 females) with a mean age of 51.09 years (range 18–92 years, SD 18.66) who underwent a total of 119 free microsurgical procedures (71 gracilis muscle flaps and 48 ALT flaps). Patient demographics of both groups are presented in Table 1. Both groups were comparable regarding the prevalence of comorbidities including diabetes, obesity, and vascular disease (p > 0.05). Similarly, the distribution of preoperative ASA score and the smoking status correlated between the V and NV group, whereas in the NV group a significant higher prevalence of hypertension was observed (p = 0.037). Etiology of defect reconstruction was comparable in both groups.

Perioperative details (defect size, recipient vessels, and type of arterial or number of venous anastomosis) were comparable between the two groups, except mean OT time which was significantly shorter in the NV group (288 versus 331 minutes, p < 0.001). Details are summarized in Table 2.

The total percentage of major complications did not vary significantly between the groups (V: 21.38 (68/318) versus NV: 15.13 percent (18/119); p = 0.143). In detail, major complications included total flap losses of 5.35 (V: 17/318) versus 5.04 percent (NV: 6/119); p = 0.899), arterial thrombosis of 3.46 (V: 11/318)) versus 2.52 percent (NV: 3/119; p = 0.790), venous thrombosis of 11.01 (V: 35/318) versus 8.40 percent (NV:10/119; p = 0.366), and hematoma of 7.23 percent (V: 14/318) versus 1.68 percent (NV: 2/119; p = 0.398).

Also, minor complications did not differ significantly between the groups (p = 0.875). Complication rates are visualized in Table 3.

When further dividing the groups regarding flap type (gracilis versus ALT flaps), the complication rates did not differ significantly (p < 0.05). Details are shown in Table 4.

This study supports the findings of recent research groups on the safe use of vasopressors in the intraoperative setting of free tissue transfers in lower limb reconstruction. Free tissue transfers have become an irreplaceable tool in the treatment of large, full thickness defects with success rates of up to 95 percent in large volume centres. Although these operations can be regarded as standard procedures, free flaps often remain connected with long operation hours, long sedation times, possible high fluid losses, and sometimes require postoperative monitoring at intensive care units. All these factors may lead to hypotension, which occurs in 40 percent to 60 percent of patients undergoing general anaesthesia. Systemic hypotension negatively affects flap perfusion. Hypoperfusion of the flap may cause partial or full flap necrosis due to reduced inflow and consecutive pedicle thrombosis, which increases hospitalisation duration, costs and patients’ satisfaction [12].

Routinely systemic hypotension is treated with crystalloid solutions, or vasopressors such as norepinephrine or dobutamine. The use of the latter still remains controversial due to their vasoconstrictive effect and consequent potential of tissue ischemia. In 1997 Cordeiro et al. [4] and in 2007 Massey et al. [14] showed that vasoactive medication has negative influence on the outcomes of island or rotational flaps in pigs. The results of these animal models could not be support by numerous retrospective studies [10, 19]. One of the few prospective studies by Eley et al. shows the effect of different vasoconstrictive agents (Epinephrine, Norepinephrine, Dobutamine, and Dopexamine) on free flap skin blood-flow following resection of head and neck cancer [6]. They found out that both, dobutamine and norepinephrine, do even have beneficial influence on flap skin perfusion, further that norepinephrine is superior to the former. Chen et al. could show similar results for microsurgical breast reconstruction. These findings were also support by the works of Kelly et al. for head and neck and for lower and upper extremity defects [12].

Fang et al. published a paper on the effects of vasopressors on free flap outcomes with the largest volume so far. In 10 years, they performed 5671 free flaps for head and neck, trunk, breast and extremity defects of cancer patients. Not only could they proof that ephedrine, phenylephrine or calcium chloride have no negative influence on flap outcome, they also stated that there was a lower risk of venous congestion in those who were treated with vasopressors [6]. As mentioned by the author, their findings were limited to cancer patients and may not necessarily be transferable to the general patient population. Recently, Knackstedt et al. did a literature review and meta-analysis of outcomes in microsurgical reconstruction using vasopressors and concluded that vasopressor utilization does not directly result in increased complications. Furthermore, they were able to show that microsurgical procedures which received vasopressors had a statistically lower rate of take-back and failure rates. However, this review is limited to the paucity of data availability regarding perioperative details and characteristics [13].

Eltorai et al. examined the effect of vasopressors in 98 of a series of 333 consecutive patients undergoing deep inferior epigastric perforator (DIEP) flap breast reconstruction who received either ephedrine or phenylephrine. The authors were able to show a somewhat new tendency. Not only could they demonstrate that both vasoactive substances have no effect on flap complications, they also stated that the use of ephedrine to overcome perioperative hypotension shows a decreased risk of venous thrombosis [20]. Anker et al. were able to show similar results in a prospective trial analysing different standardized fluid restrictive perioperative hemodynamic management regimes with and without vasoactive substances [2]. Even though the collectives are limited to DIEP flaps, the findings still represent a strong impact on the existing aversion for the use of vasopressors. Critically speaking, it would be interesting to see whether these results are also reproducible for reconstruction on heart distant defects, such as on the lower extremity.

Therefore, we have scanned our patient collective of 437 free gracilis muscle and fasciocutaneous ALT flaps for lower limb reconstruction regardless of defect etiology and divided the cohort in two groups (V versus NV). The groups were comparable regarding patient demographics, except a higher prevalence of hypertension in the NV collective. Furthermore, mean OT time was significantly longer (by on average 43 minutes) in the V group, which is in accordance with the fact that longer administration of sedative medication can increase the risk for hypotension and subsequently promote the use of vasopressors. With regards to postoperative major and minor complication rates, this study supports the findings of recent literature and no significant differences regarding the outcomes in patient that received intraoperative vasopressors compared to the ones who did not, were observed. Similarly, in patients that required revision surgeries the presented data did not show lower rates of venous complications in the group treated with vasopressors, which stands in contrast to the findings of Fang et al. [7].

When further dividing the cohort according to the flap types into a gracilis- and an ALT-group in order to analyze the influence of vasopressor use in muscle- versus fasciocutaneous flaps, the complication rates did also not differ significantly (p < 0.05) between these subgroups.

Even though it is common consensus that vasoactive medication should be avoided, if possible, this study demonstrates that if blood pressure drops more than 20 percent from the preoperative baseline, vasopressors can be used safely without having negative influence on complication rates in free tissue lower extremity reconstruction. We therefore encourage our anaesthesiologists to overcome the antipathy against using vasoconstrictive medication in the setting of free tissue transfers. In our clinic fluid resuscitation is the first step of treating perioperative hypotension, which is fine as long as it doesn’t lead to fluid overload. Besides the systemic, negative effects of fluid overload with possible cardiopulmonary decompensation, it may lead to a higher rate of postoperative oedema. Especially in precisely moulded free tissue transfers, extensive swelling can cause pedicle compromise with subsequent flap necrosis. The necessary opening of stiches to release the pressure on the anastomosis implicates a secondary revision operation and extension of hospitalisation days. A rational balance between sufficient fluid resuscitation and a thoughtful use of vasopressors is needed to reduce perioperative hypotension. It is the surgeon’s responsibility to take part in the decision-making when it is time to integrate vasopressors during free flap surgery. As also shown in this study, the use of vasopressors is not the source of free flap complications. The majority of flap complications are caused by venous thrombosis, hematoma, spasm or kinking, representing issues that can be influenced by precise surgical skill, meticulous haemostasis and appropriate choice of vessels. Only by being able to use a bigger variety of different flaps, one can find the suitable flap with the right pedicle length and diameter for the respective defect. The flap’s thickness and size, the location and length of the pedicle or perforator need to be taken into consideration in the preoperative planning. Especially with perforator flaps, accurate preoperative perforator selection can save a lot of time and trouble in the operation room. With induction of anaesthesia and consequent systemic hypotension, it is more difficult to find a suitable perforator with the hand-held Doppler. Many centres go even further and use ultra sound perforator mapping or even CT-angiography on a routinely basis to find and mark the desired perforator preoperatively. Directly after the anastomosis we use Indocyanine Green (ICG) fluorescence angiography to evaluate the vascular integrity, which helps us to visualize problems either on a spastic artery or on insufficient outflow directly during the on-going operation. Care should also be taken, when fitting the free flap into the defect after the anastomosis. Tension and kinking must be avoided and therefore it sometimes makes sense to strengthen the pedicle’s position with the application of fibrin glue around it under direct vision. After the glue has dried, it will not move its position even when the flap is flipped over it. It also helps to overcome diffuse bleeding around the pedicle or the flap edges and therefore reduces hematoma formation. If the flap has been chosen wrongly and it is too thick for the defect, it needs to be thinned even though this might compromise overall flap perfusion. If the wound edges do not align properly and a thick flap is sutured in under tension, postoperative swelling will lead to more tension and subsequently to pressure on the pedicle. In our own experience, we do believe that complications during free flap reconstructions do occur mainly because of technical problems associated with pedicle tension, kinking, or twisting; intraoperative pedicle or perforator injury; and postoperative events such as thrombosis secondary to infection or pedicle compression.

The presented analysis has several limitations inherent to a retrospective review. Further, the study is limited by the fact that it is not a single-surgeon study and one might regard the variety of comorbidities or ethiologies as a bias.

However, the analysis of the presented cases supports the hypothesis that the use of vasopressors, only if a certain blood pressure baseline has been under run, can be regarded as save and should not be regarded as a reason for flap failure in lower extremity reconstruction, which has been shown in this collective.

The presented series re-evaluated the effects of vasoconstrictive agents in the setting of microsurgical lower extremity reconstruction. According to the findings, vasopressors may be used safely during free tissue transfers, while there exists no difference between muscle- or fascio-cutaneous flaps.

Conflict of interest

None.

Sources of funding

None.