Decreasing Venous Thromboembolism With Anticoagulation in Plastic Surgery

Abstract

Introduction: Venous thromboembolism (VTE) is a devastating complication in aesthetic plastic surgery. Plastic surgeons are advised to approach VTE prevention proactively by modifying the perioperative risk profile of their patients. Herein, the authors review the evidence of pharmacologic prevention of VTE in plastic surgery, caution surgeons that blanket treatment based on Caprini score alone should be avoided, and advocate for case-by-case individualized therapy for every patient. Methods: A review of evidence for the pharmacologic prevention of VTE in plastic surgery was performed. Results/Conclusion: A proactive approach to VTE prevention can be achieved by employing both nonpharmacologic and pharmacologic evidence-based interventions. While the choice to use anticoagulation medications should be guided by calculation of a Caprini score for each patient, surgeons should continue to maintain their clinical judgment to prevent complications from over and under treatment.

Introduction

Venous thromboembolism (VTE) is a devastating and preventable complication that can occur after plastic surgery. Plastic surgeons are advised to take a proactive approach in mitigating the risk of VTE before surgery, and recommendations for risk identification, modification, and reduction have been published previously. ¹ In some patients, however, chemoprophylaxis with systemic anticoagulation is warranted due to a higher perceived risk of VTE. The 2005 Caprini score ² can be used for individualized VTE risk stratification in plastic surgery patients, and identifies inpatients with a favorable risk/benefit ratio for postoperative anticoagulation.^3–7 The score is supported by the American Society of Plastic Surgeons, ⁸ the American Association of Plastic Surgeons, ⁹ and the American College of Chest Physicians. ¹⁰ The 2005 Caprini score has been shown to predict symptomatic VTE risk in surgical populations and specific validation studies have been performed in plastic surgery, DIEP reconstruction, ¹¹ otolaryngology, ¹² general/vascular/urology, ¹³ and surgical patients as a whole. ¹⁴

Although the 2005 Caprini score is widely considered the gold standard scoring system for VTE prophylaxis, it stops short of offering objective recommendations, instead providing suggestions based on calculated score. Some have criticized this as it provides a somewhat vague guideline, with room for surgeon interpretation. ¹⁵ The variability in interpretation is important, as it contributes to both over- and undertreatment with systemic anticoagulants, which places patients at risk for complication—hematoma in overtreatment and VTE in undertreatment. The objective of the current manuscript is to review the evidence of anticoagulation in plastic surgery and suggest that plastic surgeons individualize treatments, guided by a 2005 Caprini score, but informed by their own clinical judgment.

Types of Systemic Anticoagulants in Plastic Surgery

Unfractionated heparins (UF) and low molecular weight heparin (LMWH) (Enoxaparin) both inhibit thrombin activation by increasing the activity of antithrombin III. Both are administered parenterally, with prophylactic doses typically administered by subcutaneous injection. Both UF and LMWHs have been extensively studied in the inpatient population, but fewer data are available for ambulatory or aesthetic patients. Direct factor Xa inhibitors (Rivaroxaban, Apixaban) inhibit factor Xa activity and prothrombinase activity. Unlike UF and LMWH, they are administered enterally which is more desirable for ambulatory patients. Several groups have published their experience with routine administration of oral Xa inhibitors after aesthetic surgery.^16–18 Indirect Xa inhibitors (Fondaparinux) bind to antithrombin III, potentiating its inhibition of factor Xa. Unlike direct Xa inhibitors, however, they do not have any effect on IIa. Of note, unlike enoxaparin, heparin, apixaban, and rivaroxaban, Fondaparinux is not indicated for DVT/PE treatment unless accompanied by Warfarin. Furthermore, it is eliminated in the urine so is contraindicated in patients with renal insufficiency.

Timing, Dose, and Duration of Anticoagulation

For surgeons choosing anticoagulation, the medication can be initiated before, during, or after a procedure. An AAPS 2016 meta-analysis did not support the use of preoperative anticoagulation in the overall plastic surgery population, as its utilization did not significantly reduce VTE or significantly increase bleeding. ¹⁹ A separate meta-analysis limited to anticoagulation in abdominoplasty, including 1596 patients in 10 trials, showed that preoperative anticoagulation significantly increased bleeding (7.4% vs 0.9%, P = .001), but postoperative anticoagulation did not (1.1% vs 0.9%, P = .774). ²⁰ Individual trials are similarly mixed, with one facelift study showing significant increased bleeding (16.2% vs 1.1%, P < .003) ²¹ and a breast reconstruction study ²² showing no significant increase (6.1% vs 3.3%, P = .42) with preoperative enoxaparin.

Inpatient Anticoagulation for Plastic Surgery Patients

A 2016 AAPS meta-analysis showed routine anticoagulation significantly increased bleeding (OR 1.86, 95% CI [1.10-3.14]), but did not benefit general plastic surgery patients post general endotracheal anesthesia cases at a population level for VTE risk reduction (OR 0.74, 95% CI [0.47-1.17]). ²³ A surgery-wide meta-analysis similarly showed that routine anticoagulation significantly increased bleeding (OR 1.69, 95% CI [1.16-2.45], P = .006), and that only patients with Caprini 7-8 or >8 had a demonstrable benefit for VTE risk reduction. ²⁴ However, many hospital systems continue to mandate anticoagulation for all inpatients after surgery, and some surgeons even choose to routinely anticoagulate all aesthetic surgery patients.

The VTEPS study was a large, multicenter study run by the Plastic Surgery Foundation in 2008-2011, with the goal of examining the risks and benefits of routine anticoagulation in postoperative plastic surgery inpatients. The study accrued patients at four tertiary care facilities across the United States, including historic controls who received no enoxaparin, and intervention patients who routinely received enoxaparin 40 mg daily, begun within 24 h after surgery. In an analysis of 3334 patients, the study demonstrated that routine postoperative enoxaparin prophylaxis reduced VTE risk in patients with Caprini 7-8 (2.55% to 1.15%) and >8 (8.54% to 4.07%). Multivariable regression confirmed that receipt of enoxaparin was protective against 60-day symptomatic VTE in Caprini ≥7 patients (OR 0.39, 95% CI [0.16-0.97], P = .042), but not among all patients. ²⁵ The study showed a nonsignificant increase in postoperative bleeding (0.7%) when anticoagulation was routinely provided. Multivariable regression identified breast surgery, microvascular surgery, and postbariatric body contouring, but not postoperative enoxaparin, as independent predictors of 60-day bleeding. ²⁶

VTEPS demonstrated the utility of the 2005 Caprini score to identify postoperative surgical inpatients known to have significant VTE risk reduction with postoperative enoxaparin. This finding was confirmed in a larger meta-analysis encompassing 13 clinical trials in surgery (four were plastic surgery studies^27–30). The meta-analysis showed that for an overall surgical population, routine anticoagulation significantly reduced VTE risk in Caprini 7-8 (OR 0.60, 95% CI [0.37-0.97], P = .04) and Caprini > 8 (OR 0.41, 95% CI [0.26-0.65], P = .0002) patients, but not in lower risk (Caprini 3-4 and 5-6) groups. This is relevant because when provided to the general inpatient population, routine anticoagulation significantly increased bleeding (OR 1.69, 95% CI [1.16-2.45], P = .006). ³¹ Together, these studies support a targeted, as opposed to blanket, approach toward postoperative anticoagulation after inpatient plastic surgery.

Several studies have shown patient-centric factors can impact rate of enoxaparin metabolism. Specifically, patients with elevated total body surface area surgically injured ³² and those with elevated weight ³³ have more rapid enoxaparin metabolism. The general inpatient population, ³⁴ as well as microvascular surgery patients ³⁵ regularly receive inadequate prophylactic doses of enoxaparin. Pannucci and colleagues examined the patient-centric metabolism of enoxaparin 40 mg daily. In a general reconstructive surgery inpatient population, increased surface area of surgical injury and increased weight were independently associated with more rapid enoxaparin metabolism, indicated by anti-Factor Xa (aFXa) levels. The majority (56%) of patients had low aFXa with once daily dosing, and patients whose aFXa level was low were significantly more likely to have 90-day symptomatic VTE (10.2% vs 0%, P = .04). ³⁶ A second trial showed that most (90.4%) patients had in range aFXa with enoxaparin 40 mg twice daily, but this dose led to 27% being overanticoagulated with aFXa higher than goal for prophylaxis. Twice daily enoxaparin significantly decreases 90-day symptomatic VTE (0% vs 5.3%, P = .012) when compared to once daily at the population level, at a potential cost of increased 90-day clinically relevant bleeding (6.8% vs 3.2%, P = .25). ³⁷ The FIxed or Variable Enoxaparin (FIVE) trial, ³⁸ a double-blind randomized control trial, showed that weight-based enoxaparin at 0.5 mg/kg twice daily optimizes the pharmacokinetics of enoxaparin, when compared to fixed twice daily dosing. Specific to the FIVE trial, weight-based enoxaparin dosing did not necessarily improve clinical outcomes (bleeding or VTE). However, avoidance of inadequate anticoagulation (eg, achieving initial in range aFXa level) has been shown to be protective against 90-day symptomatic VTE in a surgery-wide meta-analysis. ³⁹ Together, these pharmacokinetic trials support the importance of twice daily enoxaparin dosing, and the potential for aFXa level monitoring in selected patients expected to have altered or more rapid enoxaparin metabolism.

Three studies have examined the role of extended duration anticoagulation after DIEP flap breast reconstruction, with mixed results. One study from Cornell/Columbia showed a low overall VTE rate (0.4%) with inpatient-only anticoagulation. ⁴⁰ Another comparative study from Indiana University ⁴¹ showed no significant difference for 120-day VTE for inpatient anticoagulation versus inpatient anticoagulation and postdischarge (14 days) enoxaparin (1.5% vs 2.8%, P = .6). A second comparative study from University of Colorado also compared inpatient to postdischarge (14 days) enoxaparin, showing significant decreased 90-day VTE with extended duration prophylaxis (3.7% vs 0%, P = .03). ⁴² The best available data is equivocal for the role of postdischarge prophylactic dose anticoagulation in DIEP breast reconstruction.

Outpatient Anticoagulation for Plastic Surgery Patients

Multiple groups have published on routine anticoagulation in the aesthetics population, using both injectable (heparin and enoxaparin) ⁴³ and oral (rivaroxaban and others)^44–46 medications. The aesthetics population is generally low risk for VTE—population-based data from CosmetAssure support an overall 30-day VTE rate of 0.09%, and risk stratification data supports that 96% to 98% of aesthetics patients ⁴⁷ are at lower VTE risk with Caprini ≤6. Two studies discussed below which routinely anticoagulated abdominoplasty patients had 98.6% ⁴⁸ and 90.8% ⁴⁹ Caprini ≤6 patients. A recent article by Restifo ⁵⁰ was the first to assess DVT risk assessment in an outpatient, private-practice abdominoplasty population. Interestingly, of 1370 abdominoplasties included in the study, only 7 had a Caprini score of ≥7. This cut-point has previously been shown to identify patients with a favorable risk/benefit to prophylactic anticoagulation.^51,52

The effectiveness of routine anticoagulation in aesthetics will be challenging to prove. Inpatient data supports a 50% reduction in symptomatic VTE risk with anticoagulation—VTEPS included over 3300 higher risk inpatients to identify this difference. The baseline rate of 30-day VTE in aesthetics is 0.09%, ⁵³ and one sample size calculation supports that over 50,000 patients would be required to adequately power a study to identify a VTE risk reduction of 50%. ⁵⁴

Two studies in aesthetics have examined the baseline risk level of rhinoplasty ⁵⁵ and breast augmentation ⁵⁶ patients. Among 412 consecutive rhinoplasty patients, 98% were at lower risk for VTE, with 2005 Caprini scores ≤6. Similarly, among 100 consecutive breast augmentation, 96% patients were at lower risk (Caprini ≤6). In Restifo's abdominoplasty study only 7 (0.5%) of patients had Caprini ≥7. For surgeons willing to generalize inpatient VTEPS data to the aesthetic population, existing data supports that only 2% to 4% of the aesthetic population merit consideration for prophylactic anticoagulation. ⁵⁷

Use of Transexamic Acid in Plastic Surgery

Transexamic acid (TXA) is an antifibrinolytic agent that reversibly inhibits conversion of plasminogen to plasmin. It has a terminal half-life of 2 h when administered IV and 10 h when oral. The time to peak of a single dose is 2.5 h (range 1-5) and the recommended dose for patients with normal renal function is 10 mg/kg for IV and 1300 mg for oral. Despite aesthetic plastic surgery generally being considered a low blood loss surgery, the use of perioperative TXA has become increasingly common in plastic surgery to prevent hematoma. Additionally, some argue that it decreases bruising, lowers hemoglobin drop, reduces drain output, and shortens operative time.^58–64 In liposuction, TXA in tumescent solution has been shown to decrease blood loss by 60%. ⁶⁵ In facelift surgery, IV TXA was shown to reduce bruising and postoperative collections. ⁶⁶ In breast reduction surgery IV use was shown to decrease hematomas significantly.^67,68

Many studies advocate for use of TXA based on its safety profile. A meta-analysis of 216 studies including over 125,000 patients undergoing surgery with TXA noted no increased risk of any VTE across all medical specialties. ⁶⁹ Similarly, a meta-analysis of aesthetic plastic surgery patients treated with TXA noted that no RCTs reported increases in perioperative VTE complications with TXA use. ⁷⁰

Indiscriminate use of TXA, however, must be cautioned. Obvious absolute contraindications to IV TXA use are subarachnoid hemorrhage, active intravascular clotting and specific hypersensitivity to TXA. However, the concomitant use of combined hormonal contraceptives and estrogen derivatives are also considered a relative contraindication, as they are thought to enhance the thrombogenic effects of the TXA. Surgeons must also consider the side effect profile of TXA. Minor side effects include visual effects, dizziness, headache, back pain, abdominal pain, muscle cramps, fatigue, and diarrhea. Major side effects include wound healing complications, seizures, and most importantly, VTE. A retrospective, propensity score matched cohort study of trauma patients by Myers, for instance, showed that in 189 matched pairs, TXA was associated with greater than threefold increase in VTE (adjusted odds ratio, 3.26; 95% CI [1.3-9.1]; P = .02). ⁷¹ A recent single surgeon retrospective review of abdominoplasty patients also demonstrated an increased risk of DVT. ⁷²

Despite these studies, TXA is still considered safe in plastic surgery, but the decision to use TXA must be evaluated after consideration of patient-specific (BMI and OCP use) and procedure-specific (length and type of surgery) risk profiles.

An Argument for Individualized Anticoagulant Regimens

While a 2005 Caprini score should be calculated for each patient undergoing aesthetic surgery, it should serve as a “jumping off” point for surgeons to discuss patient-centric VTE risk with their patients, and then tailor an individualized anticoagulation regiment. The score should be used in addition to, not instead of, clinical judgment.

A variety of important patient and procedural risk factors exist that are unaccounted for in the Caprini score. The type of anesthesia (general vs local), procedure length (1 h vs 9 h), the use (including route and dose) of tranexamic acid, the type of OCP/HRT, level of obesity (BMI 26 vs 46), the nature of the procedure (plication vs no plication, abdominoplasty vs blepharoplasty, single procedure vs multiple) and the perceived mobility status of the patient (a 30 vs 60 year old abdominoplasty patient) are all examples of risk factors not incorporated into the Caprini score. Thus, while the score is known to predict baseline VTE risk among plastic surgery inpatients who receive no anticoagulation ⁷³ and to identify inpatients with favorable risk/benefit to anticoagulation^74,75 there are potentials for improvement via future research.

Limitations

The validity of the evidence presented in this review is limited by the quality of the study designs available for analysis. VTE is a rare event for aesthetic plastic surgery patients. As such, the RCT design mentioned in some of the studies included in this review would only be able to detect meaningful differences at an extremely large sample sizes. In contrast, a case-control study design^76,77 could similarly identify a cause-effect relationship, but can often produce statistically and clinically meaningful results with a smaller sample size. Future work should consider RCT and case-control designs when considering the role of anticoagulation for VTE prevention in aesthetic surgery.

Conclusion

For some patients, prophylactic anticoagulation after plastic surgery procedures can reduce risk for VTE. The authors suggest that the 2005 Caprini score should identify patients who may have a favorable risk/benefit to prophylactic postoperative anticoagulation after plastic surgery. However, decisions to provide prophylactic dose anticoagulation are not binary based on Caprini score alone and must always be individualized to the patient, including consideration of patient and procedural risk factors.