Deep Vein Thrombosis (DVT) and Abdominoplasty: A Holistic 8-Point Protocol-Based Approach to Prevent DVT

Abstract

Background

Plastic surgery as a specialty is afflicted with one of the highest incidence rates of thromboembolic events, with abdominoplasty procedures known to assimilate the greatest rates of deep vein thrombosis (DVT).

Objectives

The aim of this study was to develop a prophylactic protocol to reduce the rate of DVT occurrence postabdominoplasty.

Methods

Over a 7-year period 1078 abdominoplasty patients were enrolled onto a holistic 8-point prophylaxis protocol. For a 4-week period before the operation all patients were required to stop smoking, and to cease hormone replacement therapy and combined oral contraception. All patients were required to have a preoperative BMI of less than 40 kg/m². Participants were supplied with compression stockings, external pumping devices, and enoxaparin. Individuals with a history of DVT were also required to be 1-year treatment free prior to surgery. Furthermore, the protocol required postoperative ambulation of fit patients within 4 hours.

Results

Between 2008 and 2013, no incidence of DVT was recorded in all 1078 abdominoplasty surgery patients, indicating the potential for this protocol to lead to a significantly lower incidence than any previously published methodology. Previous studies of DVT incidence were reviewd to identify rates statistically significantly similar to our sample, thereby providing conservative incidence rate estimates.

Conclusions

This 8-point DVT prophylaxis protocol is the first noncriteria-based inclusive protocol aimed at preventing abdominoplasty-associated DVT. A holistic and procedure-specific approach to prophylaxis can drastically reduce the occurrence of DVT in abdominoplasty surgery.

With over 116,000 procedures performed annually in the United States, abdominoplasty has become one of the most popular and sought-after surgeries in the plastic and cosmetic field.¹ Despite its ever-increasing popularity and the advancement of techniques, abdominoplasty—as with any other surgery—has its complications, including infection, seroma, hematoma, thrombosis, embolism, scarring, and even death. Complication rates as high as 37% have been reported, with some studies reporting a 16% major complication rate.² One of the most serious and troubling complications for both surgeon and patient is deep vein thrombosis (DVT). With over 1 million patients tested, an estimated 250,000 cases of DVT are diagnosed per year in the United States alone.

Level of Evidence: 4

Plastic surgery as a speciality has a particularly high incidence of thromboembolic events.³ In their national survey Grazer and de Jong⁴ found that pulmonary embolism (PE) was the single biggest cause of mortality in patients undergoing large-volume liposuction. Reinisch et al⁹ reported that 0.35% of rhytidectomy patients developed DVT and 0.14% developed PE, with 1 fatality. Similar figures have also been reported for head and neck surgery, as well as for breast reconstruction surgery. Abdominoplasties have consistently been shown to accumulate the highest rates of DVT and PE in plastic surgery. Grazer and Goldwyn¹¹ reported a 1.2% incidence of DVT and 0.8% of PE in patients undergoing elective abdominoplasty. Similarly, van Uchelen et al⁶ reported a 1.4% incidence of DVT and PE in a series of 86 abdominoplasty patients.

DVT can be avoided through the implementation of specific prophylaxis conventions. Such prophylaxis encompasses graduated compression socks, sequential compression devices, and low-molecular-weight heparin (LMWH), to name a few. There are many guidelines, recommendations, and risk-assessment systems for venous thromboembolism (VTE) prophylaxis including the Davison-Caprini scale, the American College of Chest Physicians (ACCP) guidelines, and most recently the American Society of Plastic Surgeons (ASPS) evidence-based practice thromboembolism report. Despite this, information regarding DVT prophylaxis and specialist protocols in plastic surgery, and specifically abdominoplasty surgery, is limited. In this study we describe our experiences in implementing a self-designed DVT prophylaxis protocol for patients undergoing cosmetic abdominoplasty.

METHODS

To assess the effectiveness of our DVT protocol a prospective clinical study was performed involving 1078 female patients undergoing abdominoplasty surgery with flank lipocontouring between January 2008 and December 2013. All surgeries were performed in a private UK hospital, by the same surgical team. Combined abdominoplasty and lipocontour patients were selected due to their increased admission time, longer recovery, and higher risk of DVT compared with abdominoplasty-only patients, allowing us to extensively evaluate the efficiency of our protocol (Table 1). The protocol used in this study became hospital policy approved by the hospital’s Medical Advisory Board, and the design of this study followed the principles of the Declaration of Helsinki. Prior to participation in the study, written informed consent was acquired from all patients.

Table 1.

Eight-Point DVT Prophylaxis Protocol

Part 1

All patients taking HRT or COC must stop 4 weeks prior to surgery and only recommence 2 weeks postoperatively.

All patients must stop smoking for 4 weeks prior to surgery

All patients must have a BMI under 40 kg/m2 prior to surgerya

All patients with a previous history of DVT must be treatment free for over 1 year

Part 2

Compressions socks fitted 1 hour preoperatively to all patients

Intermittent pneumatic compression devices (Flowtrons) applied to all patients throughout theater

20 mg of cleaxane to all patients with a BMI less than 30 kg/m2—1 dose pre- and postoperatively

40 mg of cleaxane to all patients with a BMI greater than 30 kg/m2—1 dose pre- and postoperatively

COC, combined oral contraceptive; DVT, deep vein thrombosis; HRT, hormone replacement therapy. ^aPatients with a BMI greater than 30 and/or patients who are unable to ambulate early have the intermittent pneumatic compression device fitted for an additional 12 hours.

Patients

All patients who followed the protocol were part of the study. Patients with a previous history of DVT were accepted after clearance by their general doctors after a vascular and pulmonary evaluation. In our hospital policy, patients with a BMI over 40 kg/m² are deferred for cosmetic surgery and referred to the bariatric surgery team. No other exclusion criteria were applied in the study. Patients were systematically reviewed at 7, 14, and 28-day intervals and whenever necessary for up to 1 year. DVT diagnosis was confirmed clinically, with anamnesis and physical examination. Patients with any symptoms were scheduled to be referred for an echo colour Doppler scan for diagnosis. Our aftercare scheme provides cover for up to 3 years and entitles patients to free revisions and/or treatments, provided they attend all mandatory appointments (which all patients in this study did).

Surgical Technique

All abdominoplasties performed involved a transverse lower abdominal incision and wide undermining of the skin and subcutaneous tissue to the costal margins and 2 cm below the xiphoid appendix. Tightening of the abdominal musculature with correction of rectus muscle diastasis with Prolene 1 sutures was then completed,⁵ followed by resection of redundant abdominal skin and subcutaneous tissue. All patients had quilting sutures placed according to the senior author’s technique. The umbilicus was then repositioned before final skin closure. Liposuction was employed in the lateral regions of the abdomen and hips.

Liposculpture was executed to improve the contouring result, but no liposculpture of the dorsum and/or abdominal flap or any other additional areas was performed in any of these cases.

All surgical procedures were completed within 2 hours, which by itself may contribute to preventing DVT. One size-14 vacuum-system drain was introduced through the pubic area in all 3 groups (Medinorm, Munich, Germany). Drains were subsequently removed postoperatively, providing that drainage was less than 30 mL in 24 hours. Drains were used in all groups according to hospital policy for abdominoplasties. All patients had compressive garments fitted in the operating room, and these remained in situ for 6 weeks. Hospital stay totalled 2 nights in all cases.

RESULTS

Among the 1078 abdominoplasty procedures, there was no record of any incidence of DVT, seroma, or hematoma, despite observations of 22 infections and 24 wound dehiscences (Table 2). The average follow-up was 13 weeks and all patients completed the mandatory 28-day appointment. Because no DVT was diagnosed, the sample gives rise to limited statistical inference. Nevertheless, there were 3 different approaches used to gain further understanding of the study’s results. First, the study presents the probability of not observing any DVT in 1078 abdominoplasty surgeries for different risk ratios of DVT. Second, the study compares results with the estimated lower confidence bounds of previously presented studies. Finally, the study also tests the risk difference between our protocol and other abdominoplasty samples from previous studies.

Table 2.

Demographic and Complications Information for All Patients

Number of patients	1078
Age range (years)	28-63
Age, mean [standard deviation] (years)	42.2 [8.28]
Sex	Female
BMI (kg/m2)	25-38
Liposculpture volume (total, mL)	250-600
Seroma	0 (%)
Hematoma	0 (0%)
Infections	24 (2.2%)
Wound dehiscence	22 (2%)

Although the study displays a zero risk of DVT, we cannot rule out the possibility that this is due only to sampling variation, and our protocol could in fact lead to some cases of DVT. This is why we decided to examine the levels of DVT risk that could be consistent with our findings. Figure 1 shows the probability of not observing any DVT incidences out of 1078 procedures for different levels of complication risk. For example, if the actual rate of DVT incidence is 0.05%, there is 60% probability that among 1078 randomly selected patients no DVT is observed; thus, this rate is compatible with our results. On the other hand, if the actual DVT rate with this protocol is 0.45%, there is only a 0.77% chance of not having any incidence rate among 1078 patients, which makes it unlikely that our protocol has such risk rates. Correspondingly, at a 5% significance level, the null hypothesis of any risk rates below 0.28% would not be rejected if 1078 procedures with zero incidence rate were observed.

Figure 1.

Graph showing the probability of not observing any incidence of deep vein thrombosis out of 1078 procedures for different levels of complication risk.

To further assess our findings, results published by van Uchelen et al,⁶ Matarasso et al,⁷ Pontelli et al,⁸ Reinisch et al,⁹ Winocour et al,¹⁰ and Grazer and Goldwyn¹¹ were used to carry out additional statistical inference. The incidence rates, collected from surveys, range from 0.53%⁸ to 1.16%.⁶ Similarly, the sample size varies from 86⁶ to 25,478¹⁰ patients. Conversely, assuming that the average proportions are normally distributed, the corresponding 95% lower confidence bounds of these published DVT risk rates are 0.00%,⁶ 0.01%,⁷ 0.02%,⁸ 0.25%,⁹ 0.71%,¹⁰ and 0.93%,¹¹ indicating that the incidence rates reported by Reinisch et al,⁹ Winocour et al,¹⁰ and Grazer and Goldwyn¹¹ are significantly higher than zero, whereas the joint risk rates calculated by van Uchelen et al,⁶ Matarasso et al,⁷ and Pontelli et al⁸ are in line with our findings. The 95% lower confidence bounds of different studies are shown in Figure 2.

Figure 2.

Graph showing the 95% lower confidence bounds of reported deep vein thrombosis rate from 6 different studies.

Finally, the normality test was used to compare the proportions of DVT incidence in our sample and the survey results assuming the null hypothesis of equal DVT incidence rates. As shown above, there is a negligible possibility that our protocol results in larger incidence rates than previously examined procedures; therefore, a 1-tailed normal test was applied to compare risk differences. After calculating the overall risk ratio and its standard error of the 2 samples, 1-tailed P values of 39.68%,⁶ 28.44%,⁷ 20.52%,⁸ 3.91%,⁹ 0.23%,¹⁰ and 0.00%¹¹ were obtained. Therefore, at the 5% significance level, the hypothesis that the risk ratio of our protocol is the same as the risk rate calculated from the Reinisch et al,⁹ Winocour et al,¹⁰ and Grazer and Goldwyn¹¹ surveys can be rejected. However, the hypothesis that the DVT risk rate of our protocol is the same on average as those reported in the van Uchelen et al,⁶ Matarasso et al,⁷ and Pontelli et al⁸ surveys cannot be rejected. Figure 3 presents the P values and the corresponding DVT risk rates.

Figure 3.

Graph showing the P values and corresponding deep vein thrombosis risk rates in results published by van Uchelen et al,⁶ Matarasso et al,⁷ Pontelli et al,⁸ Reinisch et al,⁹ Winocour et al,¹⁰ and Grazer and Goldwyn.¹¹

Although our sample size of 1078 abdominoplasty procedures is large enough to perform the most commonly applied statistical analyses, because of the low risk ratio of DVT, only limited statistical inferences can be made. However, there is a clear indication that even if our protocol cannot eliminate all DVT occurrence in abdominoplasty, it can reduce its occurrence significantly compared with the risk ratios reported by Reinisch et al,⁹ Winocour et al,¹⁰ and Grazer and Goldwyn.¹¹ Assuming the most adverse scenario consistent with our finding, our protocol would result on average in a 0.025% incidence rate, which is still around a 30% reduction compared with the lowest incidence rate reported by Matarasso et al.⁷ To provide more accurate reduction rates, we hope to conduct our analysis on an increased sample size in the future.

DISCUSSION

Pathophysiology

The pathophysiology of DVT was first defined by Virchow at the turn of the 19th century. DVT pathogenesis was described to be a consequence of 3 main factors: hypercoagulability, hemodynamic changes, and endothelial vessel wall damage. As already discussed above, many studies have shown high incidences of DVT in surgical patients. It is thought that surgery generally increases blood stasis and endothelial wall damage, resulting in an increased predisposition to DVT formation. Venous stasis and pooling are also amplified due to prolonged operative times and the vasodilatory effect of general anesthesia. Patient immobilization in combination with inactive thigh and calf muscles are equally causative factors. In certain surgeries, such as abdominoplasty and breast augmentation, patients are sometimes placed in a flexed position, increasing the volume of blood accumulating in the lower limb vasculature. Hemodynamic stasis also results in a reduction in lower limb venous clearance, in turn reducing the exacerbation of clotting factors and increasing the probability of DVT development. Endothelial damage is the other factor propagated by surgery. Injury is induced as a result of surgical manipulation of the patient’s flaccid body; this can cause stretching, compression, and dilation of muscles and vasculature, resulting in minor intimal damage. Despite such injury appearing minimal, it is sufficient to stimulate the instigation of the clotting cascade.¹² We acknowledge that several etiologies—factor V Leiden deficiency, factor C deficiency, activated protein C resistance, protein S deficiency, etc—lead to thrombosis. Future studies should take also these in consideration, and omission of these etiologies can be considered one of the limitations of this paper.

The incidence of DVT in plastic surgery is highly variable. Procedures such as TRAM breast reconstruction report DVT rates of 1.3% in comparison to 0.3% in head and neck surgery.^13,14 Abdominoplasty surgery in particular reports one of the highest incidences of DVT in plastic surgery. Neaman et al² analyzed abdominoplasty-associated complications and discovered that 0.5% of their cohort experienced DVT. Similarly, Grazer and Goldwyn¹¹ reported a 1.2% incidence rate of DVT. Combined abdominoplasty procedures have been shown to have even higher rates of DVT, up to 5%.¹⁵ To understand the scale of the problem Grazer and de Jong conducted a census survey of just under half a million abdominoplasty and lipoplasty procedures. They found that mortality rates from such surgery were in the region of 1 in 5000, with nearly a quarter of all deaths resulting from PE following DVT.⁴

So why do abdominoplasty procedures carry such a high risk of DVT? Body contouring procedures involve a number of factors that can increase the risk of developing a thromboembolic event. First of all, it is thought that intraoperative positioning leads to venous stasis and disrupts lower-extremity venous return during surgery. Young and Watson¹² describe how vessel injury from the disruption of superficial and pelvic veins can contribute to thrombus formation. In addition, plication of the diastasis recti can lead to a decrease in venous return. Gray et al¹⁶ note that postoperative abdominoplasty patients struggle to mobilize more than other plastic surgery patients do and must deal with relatively long operative times, further increasing the risk of DVT. It should also be mentioned that individuals undergoing abdominoplasty surgery have a tendency to possess weight management issues, a further risk factor for DVT formation.

Risk Factors and Prophylaxis

There are many risk factors that predispose abdominoplasty and surgical patients to DVT formation. A summary of these risk factors can be found in Table 3. The aim of prophylaxis is to reduce or eliminate as many of these factors as possible in order to prevent DVT occurrence.

Table 3.

Main Risk Factors for DVT

Pregnancy

Age greater than 40 years

Major surgery

Smoking

Immobility

Cancer

Obesity

Inherited or acquired thrombophilia

Estrogen-containing oral contraceptives

History of venous thromboembolism

Selective estrogen receptor modulators

Hormone replacement therapy

DVT incidence has long been associated with a wide variety of risk factors. It is clear, therefore, that before any surgery can take place a thorough history and examination must be performed in order to assess the patient’s risk of developing DVT. Identifying risk factors enables us to optimize patients preoperatively and assess their relative suitability for the procedure in hand. In our study we concentrated on addressing several variables in order to reduce the chance or prevent the occurrence of DVT.

Smoking

Our protocol required a 4-week preoperative smoking cessation period. Smoking has detrimental effects on health and the human body.¹⁷ Adverse effects include increased risk of stroke, heart disease, and a variety of different cancers. Unsurprisingly, smoking also increases the probability of developing DVT. Smoking is thought to cause such consequences by increasing activation of the coagulation cascade. A recent case-control study compared DVT incidence in 3989 smokers and 4900 nonsmokers.¹⁸ The authors found the relative risk of DVT was 1.42 (95% CI: 1.28-1.58) in current smokers and 1.23 (95% CI: 1.10-1.37) in ex-smokers, compared with those who had never smoked. Furthermore, those who smoked the most and for the longest periods of time had a relative risk of 4.30 (95% CI: 2.95-7.14). This study demonstrated not only the strong association between smoking and DVT but also the dose-dependent and reversible association of smoking and DVT risk.¹⁹ In their review of DVT in plastic surgery, Young and Watson also highlighted the significance of smoking, stating that smoking is not an unimportant risk factor.¹²

Little direct evidence exists assessing the relation between DVT and smoking in abdominoplasty. Neamen et al² found no significant association between smoking and DVT in their 205-patient abdominoplasty review. However, hematoma incidence was significantly correlated with smoking, an important consideration in DVT chemoprophylaxis. The Caprini model, the ASPS, and the ACCP do not include smoking cessation in their guidelines or recommendations.^20-22 Smoking in general should be discouraged in aesthetic surgery not just due to DVT, but also because of its detrimental effects on wound healing and because it increases the risks of incurring other complications, such as seroma and infection.

Hormone replacement therapy and combined oral contraceptive

Hormone replacement therapy (HRT) and the combined oral contraceptive (COC) pill have conclusively been shown to increase the likelihood of DVT formation by 2- to 4-fold. One randomized control trial conducted in the early part of the 21st century involved the administration of 2 mg estradiol plus 1 mg norethisterone acetate to 71 patients, while 69 patients received a placebo.²³ The results showed a significant difference between the 2 patient subgroups, with 10.7% of HRT patients and 2.3% of placebo patients developing DVT. Many other studies have come to similar conclusions.²⁴ Interestingly, Li et al²⁵ demonstrated only a slight increase in DVT events when they analyzed gynecologic procedures, and did not find strong evidence to support stopping COC before surgery. COC, in particular, has been shown to have a greater effect on DVT development than estrogen-only HRT. Despite compelling evidence, a recent survey on British plastic surgeons found that only 50% believed HRT was a risk factor for DVT.²⁶ In addition, only 25% and 54% of surgeons disbarred patients from HRT and COC, respectively, prior to surgery. The ASPS does recommend preoperative cessation of HRT, but only for patients with a risk-assessment model (RAM) score of 7 or more, the highest-risk patients. Our protocol involves complete abstinence from HRT and COC for 4 weeks preoperatively and 2 weeks postoperatively for all patients.

BMI and Previous DVT

BMI is yet another factor that has been shown to influence DVT in a variety of fields. Shiota et al²⁷ assessed the effects of BMI in 843 Japanese ovarian cancer patients. They found that a BMI greater than 25 kg/m² was statistically associated with a higher incidence of DVT. Similarly, a 5-year review of the CosmetAssure database revealed that obese abdominoplasty and breast augmentation patients with a BMI greater than 30 kg/m² had a 21% higher incidence of DVT and PE.²⁸ Patronella et al²⁹ also investigated BMI and DVT in a variety of plastic surgery procedures involving over 3800 patients. Their results indicated that patients who acquire DVT tend to have higher BMIs (P = 0.017). Such findings support our notion of limiting BMI to a maximum of 40 kg/m² as well as administering patients of higher BMIs greater doses of chemoprophylaxis. The ASPS recommends weight management, again, only for the highest-risk patients, with no real maximum or safe BMI target recommendation. The Davison-Caprini model and the ACCP guidelines are even more vague, listing obesity as just a risk factor with no optimization recommendations. In addition to BMI, many multidisciplinary fields have found that a history of DVT also predisposes individuals to future DVT incidence. Our protocol requires all patients with a DVT history to be treatment free for 1 year prior to surgery. This not only helps prevent recurrence of DVT but also minimizes the chance of hematoma from previous DVT therapy.

Early Ambulation

Early ambulation is one of the simplest and oldest DVT prophylaxis measures in plastic surgery. A significant number of studies have included early ambulation in their prophylaxis regimen.^12,30,31 Some authors, such as Somogyi et al, ³² recommend ambulation from as early as 1 hour postoperatively for day case surgery. Moreover, the ACCP, the ASPS and the Davison-Caprini RAM all recommend early and aggressive ambulation for all patients. We recommend ambulation within 4 hours of surgery. Extra time before ambulation is allowed as a reflection of the stress inflicted on the body by abdominoplasty in comparison to other procedures, such as breast augmentation.

Intermittent Pneumatic Compression Devices and Compression Stockings

Intermittent pneumatic compression (IPC) devices are effective in relieving venous stasis by actively pumping blood, stimulating fibrinolytic activity in the veins by reducing PMA1 and increasing the release of tissue plasminogen activator, which is needed to activate plasmin from plasminogen.^12,33 A meta-analysis of 15 studies involving over 2200 patients found that IPC reduces the risk of DVT by 60% (relative risk: 0.40; 95% CI: 0.29-0.56; P < 0.001).³⁴ A randomized controlled trial conducted by McMaster University also found that IPC alone could achieve a DVT rate of 5.3% in abdominal surgery patients.³⁵ Currently the ASPS recommends IPC or mechanical prophylaxis for patients with a Caprini score of 3 or more. The modified Davison-Caprini RAM recommends IPC for all low-risk patients undergoing general anesthesia for more than 1 hour. Given that both guidelines categorize abdominoplasty as higher-risk surgery, IPC is recommended and should be considered in all abdominoplasty cases.

However, the ACCP has recently downgraded IPC, favoring chemoprophylaxis for lower-risk patients. IPC is only recommended when combined with chemoprophylaxis for high- and highest-risk patients. This seems strange given the ample evidence demonstrating the effectiveness of IPC in preventing DVT, without risking hematoma. All our patients receive IPC during surgery in addition to compression stockings 1 hour preoperatively. Patients with BMIs over 30 kg/m² have IPC use maintained for an additional 12 hours postoperatively. We believe that all patients should be given IPC, because it is a simple and effective means of reducing DVT incidence with no side effects. The timing of administration is up for debate. The Caprini-Davison RAM recommends application of IPC before, during, and after surgery, whereas we only use IPC during theatre, and postoperatively in higher-BMI patients.

Chemoprophylaxis

Currently, the ACCP recommends chemoprophylaxis for all major surgery patients and for moderate- and high-risk patients. However, with regards to plastic surgery the guidelines are vague in defining what constitutes major and minor surgery. Major surgery, according to the ACCP, includes orthopedic and gynecologic surgery; given the high-risk status of abdominoplasty we can only assume that it too falls under the category of major surgery. The Davison-Caprini model recommends chemoprophylaxis for all abdominoplasty patients with 3 or more risk factors with normal risk of bleeding. The ASPS recommendations follow suit, advocating chemoprophylaxis for patients with 3 or more risk factors, including abdominoplasty cases.

Our choice of chemoprophylaxis was enoxaparin (Clexane), a LMWH. Nearly all major studies on DVT prophylaxis advocate enoxaparin use.²¹ However, timing and dosage is still disputed. The ASPS recommends postoperative LMWH with no indication of dose. The Davison-Caprini model recommends administration of enoxaparin 12 hours postoperatively at a dose of 40 mg. The range of drug doses recommended and used in major studies varies from 30 mg up to 60 mg.²¹ The timing of administration is equally variable, with some administering chemoprophylaxis up to 8 hours preoperatively, and others not until 12 hours postoperatively. All studies, except that by Hatef et al,³⁰ demonstrated that enoxaparin significantly reduced DVT incidence without increasing hematoma rates. Hatef et al’s anomalous results may be due to the fact that the authors administered enoxaparin twice a day postoperatively in comparison to only once daily in other studies. The protocol used in this study prescribes enoxaparin immediately preoperatively and 24 hours postoperatively. The dosage varies depending on BMI; those with a BMI of less than 30 kg/m² receive 20 mg, whereas those greater than 30 kg/m² receive 40 mg. We believe that this is an appropriate dose to offset DVT incidence, as well as avoiding hematoma.

Concerns with Prophylaxis

One of the main concerns regarding prophylaxis, in particular chemoprophylaxis, is the risk of hematoma. It has been demonstrated in head, neck, and orthopedic surgery that chemoprophylaxis increases hematoma rates.^36-38 With regards to abdominoplasty, Hatef et al³⁰ found that 7.3% of patients given enoxaparin developed hematoma compared with 0.5% of those without (P < 0.001). Dini et al³⁹ conducted the only randomized double-blind study assessing the safety of chemoprophylaxis in abdominoplasty reported to date. They administered rivaroxaban 10 mg daily for 10 days postoperatively to one group of patients, while a control group received a placebo. The results were shocking: of the 27 rivaroxaban patients operated on, 8 developed large hematomas, 6 of which required drainage. This complication rate was so high that the study was halted and no further patients received rivaroxaban. The authors concluded that rivaroxaban was the principle cause of the high incidence of hematoma.

This should not discourage surgeons from administering chemoprophylaxis, but rather stresses the importance of drug potency and suitability. In the case of abdominoplasty, it appears that enoxaparin is the most suitable treatment. With careful patient selection criteria and drug administration, hematoma can be avoided. Optimizing patients through appropriate measures, such as weight management and smoking cessation, not only helps prevent DVT but also works to decrease the risk of hematoma.

Another point of concern for surgeons may be the overall cost of such a protocol.⁴⁰ However, as can be seen Table 4, the cost of implementing each of our recommendations is relatively low—a small price compared to a patient with DVT.

Table 4.

Cost of the Prophylaxis Measures Implemented in this Study

Factor	Cost
Compression socks	£3.40
Intermittent pneumatic compression device	£12.00
Clexane 2 × 20 mg	£4.54
Clexane 2 × 40 mg	£6.05
Total	£25.99

Current Guidelines and Recommendations

A comparison of main existing guidelines and protocols to the protocol implemented in this study is given in Table 5.

Table 5.

A Comparison of the Major Guidelines and Protocols to the Protocol Implemented in This Study

Factor	ACCP	ASPS	Davison-Caprini	This study
HRT or COC cessation	—	—	—	✓
Smoking cessation	—	—	—	✓
BMI management	—	—	—	✓
Past history of DVT	—	—	—	✓
Compression stockings	—	—	—	✓
IPC	✓	✓	✓	✓
Chemoprophylaxis	✓	✓	✓	✓
Ambulation	✓	✓	✓	✓

✓ = recommended for all abdominoplasty patients; — = not recommended or mentioned. ACCP, American College of Chest Physicians; ASPS, American Society of Plastic Surgeons; COC, combined oral contraceptive; DVT, deep vein thrombosis; HRT, hormone replacement therapy; IPC, intermittent pneumatic compression.

The ACCP releases one of the most important prophylaxis and treatment guidelines for DVT every 4 years. The ninth and most recent edition has excluded any plastic surgery–specific prophylaxis protocol. The ACCP has published specialty- and procedure-specific guidelines based upon orthopedic and nonorthopedic cases. As is the case in all previous editions, experimental data on plastic surgery VTE are not taken into account. This makes following the most comprehensive and respected VTE guidelines extremely difficult for plastic surgeons.

Traditionally the ACCP has advocated a risk-stratification model for guidelines and recommendations. These models involve the use of predisposing risk factors and risks associated with current illness as the main indicators for prophylaxis. These factors are then used to advocate an appropriate prophylaxis measure based upon a composite risk estimate. This form of prophylaxis protocol is referred to as an RAM.³²

Many international RAM protocols exist, with the Caprini scoring method being the most recognized and implemented.⁴¹ First produced in the early part of the 2000s, this stratification system was designed for all surgical patients and specialties and was quickly endorsed and integrated into ACCP guidelines. The protocol is based primarily on predisposing patient risk factors with little regard given to procedure-specific risks. Patients are classified into 3 risk groups (low, moderate, and high), with each group subsequently assigned to a corresponding prophylaxis regimen. Prophylaxis would range from ambulation only, in the low-risk group, to graded compression stockings with IPC and chemoprophylaxis in the high-risk group.

In 2004 the Caprini system was adapted by Davison et al for plastic surgery to encompass risk factors and prescribe prophylaxis deemed more suitable for cosmetic surgery patients. This revised model was subsequently shown to be an accurate validator in predicting DVT incidence.^42-44 Pannucci et al⁴² found that 1 in 9 plastic surgery patients with a Caprini-Davison score greater than 8 experienced a DVT event. In addition, it was also discovered that patients with a Caprini score greater than 8 were significantly more likely to develop DVT when compared to patients with scores of 3 to 4, 5 to 6, or 7 to 8. A Chinese RAM study also showed that the Caprini-Davison system is more effective in identifying high-risk patients than other RAM models such as the Kucher and Padua Prediction Score.⁴³

Despite showing strong evidence to support the validity of RAM in identifying and predicting high VTE risk patients, such protocols are not effective in eliminating DVT. A RAM protocol developed by the University of Michigan found that 118 patients out of 8000 developed DVT following its implementation.⁴⁴ Such results do not come as a surprise when the highest level of recommended prophylaxis only includes compression devices and LMWH. This highlights a critical issue, namely, that RAM and the recommended prophylaxis do not necessarily work in harmony to effectively prevent DVT.

Most of these RAMs are designed for entire surgical units, encompassing a wide variety of specialties. These RAMs do take into account individual patient profiles but fail to fully quantify and appreciate the individual risks encountered with each specialty and procedure. Different procedures and specialities have different risks. For example, the incidence of DVT in orthopedic surgery has been reported to be as high as 40% to 60%, whereas for general surgery the DVT incidence is around 2%.^45,46 Therefore, the risk of developing DVT following an orthopedic procedure is generally higher than that following general surgery. This infers the necessity for a varying degree of precaution; the orthopedic patient should receive a more aggressive prophylaxis regimen than the general surgery patient.

In response, the ACCP made a significant change to guidelines and recommendations in 2008. The College began to advocate procedure and specialty-specific risk assessment and prophylaxis guidelines in preference to personal risk factor models. It stated that RAM systems are cumbersome, hard to keep track of, not likely to be used on a widespread basis, and lack prospective validation.⁴⁷ A Milwaukee-based study concluded that compliance with RAM and prophylaxis protocols was less than 100% in a study on over 2400 surgical patients. The study found that of the 89 patients diagnosed with DVT, 12% did not receive the appropriate DVT prophylaxis.⁴⁸ Caprini et al also produced a revised RAM which now factored in procedure-specific risk. Risk was now classified according to (1) exposing risk factors—the procedural risk, and (2) predisposing risk factors—the patient’s personal risk.⁴⁹

Studies assessing DVT prophylaxis in abdominoplasty surgery are scarce. Reish et al⁵⁰ conducted a retrospective analysis on 105 body contour procedures performed between 2008 and 2010. They implemented a DVT protocol based on ACCP guidelines and the modified Caprini-Davison RAM. This system factored both personal and procedural risk. All patients were given pneumatic compression devices. Moderate-risk patients additionally received a low dose of unfractionated heparin preoperatively, whereas high- and highest-risk patients received a high dose of unfractionated heparin pre- and postoperatively (the latter for a longer period of time). The study recorded no cases of VTE and only 1 case of hematoma. Despite achieving very impressive results, the small number of participants raises questions about the validity of this study.

Hatef et al³⁰ also looked into DVT prophylaxis in abdominoplasty, employing the modified Caprini-Davison system on a cohort of over 350 patients. It was established that 5.28% of patients developed DVT despite strict adherence to the recommended protocol. Interestingly the authors found that BMI and the use of HRT was significantly associated with DVT incidence. It was also noted that no DVT occurred in any patient administered with chemoprophylaxis, but that 90.1% of hematoma cases were in this particular patient subgroup. This highlights the dilemma of administering chemoprophylaxis: it can prevent DVT but also has the potential to cause hematoma. Similar studies on abdominoplasty and RAM have achieved results ranging from 0.4% to 1.89% DVT incidence.^29,30,32 Despite all these studies showing a significant correlation between risk stratification and DVT incidence, none were able to eliminate DVT.

Rohrich et al⁵¹ describe the only non-RAM-based, all-inclusive prophylaxis protocol in body contour surgery. In their series of 151 patients, all individuals were treated with calf pneumatics and pre- and postoperative chemoprophylaxis, and were also encouraged to engage in early ambulation. The study recorded 2 incidences of DVT. We believe that to completely eliminate DVT more robust preoperative measures must be put in place, as demonstrated in this study.

The American Society of Plastic Surgeons (ASPS) has also recently published VTE guidelines,²¹ having developed an integrated RAM and procedure-specific protocol based on the 2005 Caprini model. The assessment criteria are initially divided by “inpatient” and “outpatient” populations. The former is recommended to complete a Caprini-style RAM, whereas the latter should only be considered following assessment. Prophylaxis is then based upon the RAM score and whether an individual undergoes elective surgery or any of the following procedures: body contouring, abdominoplasty, breast reconstruction, lower-extremity procedures, and head/neck cancer procedures.

A recent paper investigated the prevalence of DVT in plastic surgery, alongside the prophylaxis of choice by 1729 members of The Aesthetic Society.⁵² Regarding prophylaxis, over 90% of respondents claimed to use a patient risk-stratification assessment tool, with the majority using the Caprini model. However, there was still a high prevalence of DVT—and 50% of respondents reported having had a patient who had developed DVT. This could be due to nonadherence by 39% of surgeons to the prophylaxis protocol, or other risk factors (eg, smoking and BMI) that could accumulate to increase the chance of having DVT.

As mentioned previously, DVT guidelines must factor in procedure-specific risk and variance when setting out prophylaxis recommendations. However, equally important—and never before mentioned—is the recognition of the difference between elective and emergency surgery. This is a critical point which may seem obvious, but has been disregarded up until now. Elective procedures have the great advantage of time in comparison to emergency cases. Time allows for full preoperative patient optimization, a precaution unavailable to the emergency patient. The ASPS, however, only recommends full optimization in patients with a Caprini score of greater than 7.

Keyes et al⁵³ recently confirmed that a BMI over 40 kg/m² increases the risk of DVT and also demonstrated that 95.5% of the DVT cases were presented in patients with Caprini risk scores of 2 to 8. Most of these patients are not recommended to have chemoprophylaxis according to current guidelines.

Plastic surgery, and in particular cosmetic surgery, is nearly always performed as an elective procedure. Time is readily available to ensure a patient has the best chance possible to avoid DVT. As such, we believe that every single elective abdominoplasty patient should be prescribed the prophylaxis measures described in the first part of our guidelines. We believe this to be of critical importance in high VTE risk surgeries. Previous guidelines that only advocate mechanical and chemoprophylaxis have failed to eliminate DVT. The same is true for prophylaxis protocols that ignore the importance of procedural risk factors. In this study, we employed a holistic protocol specific to abdominoplasty procedures. We managed to achieve a 0% DVT incidence rate in arguably the most DVT-prone surgery in plastic and reconstructive surgery.

CONCLUSIONS

Our new 8-point DVT prophylaxis protocol focuses on patient- and procedure-specific optimization to prevent DVT in elective surgery. Not a single case of DVT was recorded over the 6-year period of this study. In addition, no incident of hematoma was found either. Therefore, it is clear that a holistic, preparatory, and procedure-specific approach to DVT prophylaxis can reduce the occurrence of this complication in abdominoplasty surgery.

Disclosures

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding

The authors received no financial support for the research, authorship, and publication of this article.