Mohammad-Ali Shaikh, BS; Haneol S. Jeong, BA, BBA; Andrew Mastro, BS; Kathryn Davis, PhD; Jerzy Lysikowski, PhD; and Jeffrey M. Kenkel, MD.

Transcription

1 Research Analysis of the American Society of Anesthesiologists Physical Status Classification System and Caprini Risk Assessment Model in Predicting Venous Thromboembolic Outcomes in Plastic Surgery Patients Mohammad-Ali Shaikh, BS; Haneol S. Jeong, BA, BBA; Andrew Mastro, BS; Kathryn Davis, PhD; Jerzy Lysikowski, PhD; and Jeffrey M. Kenkel, MD Aesthetic Surgery Journal 2016, Vol 36(4) The American Society for Aesthetic Plastic Surgery, Inc. Reprints and permission: DOI: /asj/sjv198 Abstract Background: Venous thromboembolism (VTE) can be a fatal outcome of plastic surgery. Risk assessment models attempt to determine a patient s risk, yet few studies have compared different models in plastic surgery patients. Objective: The authors investigated preoperative ASA physical status and 2005 Caprini scores to determine which model was more predictive of VTE. Methods: A retrospective chart review examined 1801 patients undergoing contouring and reconstructive procedures from January 2008 to January Patients were grouped into risk tiers for ASA scores (1-2 = low, 3+ = high) with 2 cutoffs for Caprini scores (1-4 = low, 5+ high; 1-5 = low, 6+ = high), then re-stratified into 3 tiers using Caprini score cutoffs (1-4 = low, 5-8 = high, 9+ = highest; 1-5 = low, 6-8 = high, 9+ = highest). Median scores of VTE patients were compared to those without VTE. Odds ratio and chi-squared analyses were performed. Results: Of the 1598 patients included in the study, 1.50% developed VTE. Median ASA scores differed significantly between comparison groups but Caprini scores did not vary regardless of cutoff. When examining the 2-tiered Caprini scores, using low risk = 1-5 showed a significant relationship between risk tier and DVT development (P = ). Conclusion: The ASA system yielded the highest odds ratio of VTE development between low and high-risk patients. The Caprini model captured more patients with VTE in its high-risk category. Combining the two models for a more heuristic approach to preoperative care may identify patients at higher risk. Level of Evidence: 4 Accepted for publication September 10, 2015; online publish-ahead-of-print December 15, Venous thromboembolisms (VTE), encompassing deep vein thrombosis (DVT) and pulmonary embolism (PE), are morbid and potentially fatal complications that manifest in patients due to a variety of risk factors, among them major surgery, trauma, and prolonged periods of immobility. 1 In the United States, up to 600,000 people are admitted to hospitals annually due to DVT or PE and the number of deaths attributed to PE alone can range from 50,000 to 100, DVT incidence has been reported within multiple surgical sub-specialties, 3,4 with orthopedic surgery experiencing the highest percentage (50%-60%) without proper prophylaxis. Risk Messrs Shaikh, Jeong, and Mastro are medical students, Dr Davis is Co-director of Research, and Dr Kenkel is a Professor and Chairman, Department of Plastic Surgery, University of Texas Southwestern Medical Center, Dallas, TX; and Associate Editor of Aesthetic Surgery Journal. Dr Lysikowski is a Statistician, Office of Quality Improvement and Safety, University of Texas Southwestern Medical Center, Dallas TX. Corresponding Author: Dr Jeffrey M. Kenkel, Professor and Interim Chairman, Department of Plastic Surgery, University of Texas Southwestern Medical Center, 1801 Inwood Road, Dallas, TX , USA. jeffrey.kenkel@utsouthwestern.edu

2 498 Aesthetic Surgery Journal 36(4) Hip and knee arthroplasty cases see an even higher incidence of DVT at 80% when patients do not receive prophylaxis, and the number of fatal PE can reach 2%. 5 General surgery experiences a much lower, yet still alarming rate of DVT and fatal PE, which can range from 15%-30% and 0.2%-0.9%, respectively, if no preventative measures are taken. 6 Plastic surgery is not immune to thromboembolic disease. In a survey conducted by Broughton et al, 7 2.5% of responders reported at least 1 DVT/PE during rhytidectomy cases. During liposuction, 6.8% reported at least 1 thromboembolic event, and the highest incidence were found in combined abdominoplasty/liposuction procedures, where 13% of the cases were complicated by DVT/ PE. Another survey 8 estimated that about 19 out of every 100,000 (0.02%) liposuction procedures is fatal, with PE being attributable to about 23% of the deaths. Because mechanical and pharmacological agents can be used to prevent the occurrence of clots and prevent further clot formation after the initial thrombotic event, 9-11 it is paramount for plastic surgeons to take proper measures to minimize a patient s risk for a venous thromboembolic event after surgery. Other studies in plastic surgery have reported VTE when performing procedures such as abdominoplasty, 12 belt lipectomy, 13 and head and neck reconstruction. 14 Abdominoplasty VTE rates were reported to be 1.4% and belt lipectomy rates approached 9%. Chen et al compared oncologic reconstructions and reported that head and neck cases in particular had statistically significant higher rates of VTE (almost 1%) than surgeries on other regions. Plastic surgeons operate on a variety of regions on the body and must be aware of the varying risks associated with different procedures. Preoperative risk assessment is common in all forms of surgery. By using models, physicians are more aware of the likelihood that a patient may develop adverse outcomes and can take measures to help minimize risk. Multiple models exist to predict a variety of possible outcomes. Although initially developed as a means to classify patients based on their health, 15 the American Society of Anesthesiologist s Physical Status (ASA PS) grading system has been shown to be an effective preoperative risk assessment model for surgical patients. 16 Specifically, the higher the score a patient receives, the greater risk for perioperative complications. The ASA PS has also been shown to be efficacious in predicting venous thrombolic events after surgery. 17,18 The Caprini Risk Assessment Model (RAM) is a system created specifically to assess patient risk for thromboembolism after surgery. 19 In order to create a model more tailored to plastic surgery, a modified Caprini RAM was created 20 to help physicians comprehensively assess patient risk for DVT or PE. Despite these modifications, the original Caprini RAM is still more effective in predicting VTE risk in plastic surgery patients. 19 In order to determine the evaluate the ASA PS and 2005 Caprini RAM models and identify which is superior in its ability to predict DVT development, a retrospective chart review was conducted of patients who underwent complex reconstruction or body contouring procedures. The data were subsequently analyzed to determine whether the ASA PS or 2005 Caprini RAM score was more efficacious in predicting postoperative thrombus development in patients. METHODS After obtaining approval from the Institutional Review Board at UT Southwestern Medical Center (Dallas, Texas), a retrospective chart review was conducted in June The initial cohort included 1801 consecutive patient encounters for reconstructive and body contouring procedures from January 2008 to January 2012, who were identified using Current Procedural Terminology (CPT) codes. Procedures included flap-based procedures, removal of facial wrinkles, tissue excision, suction assisted lipectomy, breast prosthesis, and breast reconstruction. These procedures were chosen as they were recorded into the original Microsoft Excel (Microsoft, Redmond, Washington) database by the plastic surgery department. One or more of 19 faculty surgeons at the UT Southwestern Medical Center performed these surgeries at facilities associated with the university, including Parkland Memorial Hospital, Zale Lipshy University Hospital, St. Paul University Hospital, and the Outpatient Surgery Center (all located in Dallas, Texas). The original cohort of 1801 patients was cut down to 1598 due to inconsistencies in the medical records when reporting VTE within 30 days of the patient s procedure(s). The review of records revealed that 203 patients had missing information related to postoperative outcomes, thus warranting their exclusion. No other exclusion criteria were applied. The variables under investigation were the ASA PS score and 2005 Caprini Score assigned to each patient. Both scores were assigned preoperatively; the anesthesiologist assigned the ASA PS score and the attending surgeon assigned the Caprini Score based on the criteria used in the 2005 model. 21 The patient population was split into those who experienced a DVT/PE event within 30 days postoperatively and those who did not. The patient database did not explicitly state whether a patient suffered from DVT alone or if DVT was accompanied with PE, warranting the combination of the two conditions for analysis. Risk tiers were created from the ASA PS classification system and 2005 Caprini scores to compare patients who did and did not experience a VTE. A two-tiered model was created using cutoffs based on ASA and Caprini scores, dividing the patients into low- and high-risk groups. The lowand high-risk groups within the ASA PS classification system were defined as scores of 1-2 and 3-5, respectively. With the Caprini RAM, low risk was defined as 1-4 and

3 Shaikh et al 499 high risk was defined as 5+. An alternative two-tiered model was also used in an attempt to discover different trends in the data. The attenuated model defined low risk as 1-5 and high risk as 6+. Two three-tiered models were created to better analyze trends in Caprini scores. One model used the cutoffs 1-4 for low risk, 5-8 for high risk, and 9+ for highest risk. The second model defined low risk as 1-5 and used the same cutoff of 9 between high and highest risk. The median scores of each risk assessment model were compared between patients who were and were not diagnosed with VTE within 30 days postoperatively using Mann-Whitney U test. Analysis of the risk tier models was conducted using Chi-squared test on 2 2 and 2 3 contingency tables based on the aforementioned cutoffs for the various risk tiers. All patient data were stored in Excel. Odds ratios (OR) were calculated through Excel using standard formulas. Tests for comparing ASA/Caprini median scores and analyzing the contingency tables to determine significance, specificity, sensitivity, and other descriptive statistics were conducted using Graphpad Prism 6.00 (Graph Pad, San Diego, California). The cutoff used for statistical significance was P <.05. A high OR was the variable used to determine whether a model was more or less predictive of VTE. RESULTS Out of 1598 patients, 1290 (80.7%) were women, 172 of whom used oral contraceptive pills; 308 (19.3%) patients were men. The patient age ranged from years (mean, 49.9 years). The body mass index (BMI) ranged from kg/m 2 (mean, 28.2 kg/m 2 ) and 61.8% of the patients had a BMI greater than 25. The incidence of VTE was 1.5% (Table 1). All 24 patients who experienced a VTE were placed on sequential compression devices (SCDs) and only 5 patients did not receive pharmacologic prophylaxis. Pharmacologic prophylaxis for the other 19 patients involved usage of heparin, low molecular weight heparin (LMWH), and warfarin. The dosages were not fixed between patients, even if they received the same agent to prevent thrombosis. ASA scores ranged from 1-4 for patients who experienced DVT and 1-5 for those that did not experience DVT (Table 2). Caprini scores ranged from 2-10 for patients who experienced DVT and from 1-17 for those who did not suffer from DVT (Table 3). The median ASA score of the patients without DVT/PE diagnosis was 2 and the median score of patients who were diagnosed was 3. The median Caprini score was 5 for patients without VTE and 6 for those that did experience VTE. When comparing medians, The ASA score assigned to patients without VTE was statistically different than the score assigned to patients with VTE (P <.001); however, the difference between Caprini scores for patients in the 2 categories lacked significance (P =.095). ASA Scores Using ASA scores of 1-2 to define low risk and 3+ as the cutoff for high risk, patients were stratified into different risk categories (Table 4). A total of 1248 patients were categorized as low risk; of those patients, 11 suffered from VTE. Of the 350 patients who were categorized as high risk using the ASA model, 13 experienced VTE. Patients who were assigned into the ASA high-risk group (score of 3+) had a statistically significant chance of experiencing VTE complications postoperatively (P <.001, OR = 4.338). Caprini Scores The initial 2005 Caprini score cutoffs used to analyze VTE risk with a 2-tiered approach were 1-4 for low risk and 5+ for high risk. There were 620 low-risk patients who did not experience VTE, and 7 low-risk patients who did. Out of the 971 high-risk patients, 17 experienced VTE (Table 5). There was no statistical significance when comparing VTE incidence between the low- and high-risk cutoffs, but the odds of experiencing DVT/PE did increase (P =.3087, OR = 1.578). When modifying the low- and high-risk cutoffs to 1-5 and 6+, a total of 956 and 642 patients were assigned to each tier, respectively. Ultimately, there were 10 cases of DVT/PE in the low-risk group and 14 cases in the high-risk group (Table 6). Despite altering the cutoffs for low and high risk, the incidence of VTE between the groups was not statistically significant (P =.0675). However, the odds increased even further when compared to the previous Caprini score cutoffs (odds ratio = 2.109). With a 3-tiered approach to VTE risk, the first Caprini score cutoffs examined were 1-4, 5-8, and 9+ for low, high, and highest risk. There were 627, 861, and 110 patients in each category, respectively (Table 7). There were 7 reports of VTE 30 days postoperatively in the low-risk group, 13 in the high-risk group, and 4 in the highest-risk group. There was no significance between groups when examining VTE incidence (P =.1008). However, the odds of experiencing VTE increased if a patient was in the high-risk group (odds ratio = 1.352) and increased even further if the patient was assigned to the highest-risk group (odds ratio = 3.257). The change in the odds ratio itself between high and highest risk could be determined to be through chance alone (P =.245). Altering the score cutoffs to 1-5 for low risk, 6-8 for high risk, and 9+ for highest risk yielded a statistically significant relationship between risk tiers and VTE development (P =.0266) (Table 8). The odds ratios increased using this new cutoff model as compared to the previously mentioned

4 500 Aesthetic Surgery Journal 36(4) Table 1. Patient Demographics and Statistics of Those Who Did and Did Not Have a VTE Without VTE % With VTE % P value Patient information (n = 1598) % % Female patients (1290) % % Contraceptive use (172) % % Mean age (years) Mean BMI (kg/m 2 ) BMI (kg/m 2 ) greater than % % Comorbidities including: Diabetes (%) % % Smoking (%) % % Hypertension (%) % % CAD, history of MI, or any other cardiovascular disease (%) COPD or any other pulmonary disease (%) % % % % Renal disease (%) % % Cancer (%) % % HIV/AIDS (%) % 1 4.2% History of DVT/PE (%) % 2 8.3% Major surgery within the past month (%) % 2 8.3% Mean operative time (hours) Patients where no prophylaxis administered Average ASA score Median ASA score Average 2005 Caprini score Median 2005 Caprini Table 2. Breakdown of ASA Scores for Those Who Did and Did Not Have a VTE ASA Score Patients with VTE Patients without VTE Total model (high-risk OR = 1.797, highest-risk OR = 3.476); however, this increase in OR between the two groups was also deemed insignificant (P =.246). It is worth noting that despite alterations in the 2005 Caprini score risk tier models, all odds ratios obtained for high risk in the case of the 2-tiered models (OR = and OR = 2.109) and for highest risk in the case of the 3-tiered models (OR = and 3.476), were smaller than the odds ratio obtained for high risk using the ASA score (OR = 4.338). Combining the standard ASA and Caprini models yielded a more sensitive and specific test for VTE than either test alone (Table 9). The sensitivity increased to and the specificity increased to

5 Shaikh et al 501 Table 3. Breakdown of Caprini Scores for Those Who Did and Did Not Have a VTE 2005 Caprini Score Patients with VTE Patients without VTE Total Table 6. Modified 2-Tier Caprini Scores Grouped into High- and Low-Risk Groups VTE No VTE Total High-risk Caprini (6+) Low-risk Caprini (1-5) P value Odds ratio (95% ( ) Table 4. ASA Scores Grouped into High- and Low-Risk Groups VTE No VTE Total High-risk ASA (3+) Low-risk ASA (1-2) P value * Odds ratio (95% Sensitivity (95% Specificity (95% ( ) ( ) ( ) Table 5. Caprini Scores Grouped into High- and Low-Risk Groups VTE No VTE Total High-risk Caprini (5+) Low-risk Caprini (1-4) P value Odds ratio (95% Sensitivity (95% Specificity (95% ( ) ( ) ( ) Table 7. 3-Tier Caprini Scores Grouped into Highest-, High-, and Low-Risk Groups Highest-risk Caprini (9+) High-risk Caprini (5-8) Low-risk Caprini (1-4) VTE No VTE Total Odds Ratio Sensitivity (95% ( ) P value of risk tiers and VTE Specificity (95% ( ) P value of odds ratios 0.245

6 502 Aesthetic Surgery Journal 36(4) Table 8. Modified 3-Tier Caprini Scores Grouped into Highest-, High-, and Low-Risk Groups Highest-risk Caprini (9+) High-risk Caprini (6-8) Low-risk Caprini (1-5) DISCUSSION VTE No VTE Total Odds Ratio P value of risk tiers and VTE P value of odds ratios * A prior study has suggested that venous thromboembolism causes the largest number of preventable hospital-related deaths. 22 Despite this, there have been reports indicating a lack of adequate preemptive thrombus prevention among hospitalized patients. The ENDORSE study determined that about 50% of their patients were at risk for VTE. Despite the large volume of patients at risk, only half of the at risk patients received the prophylaxis recommended by ACCP guidelines. 23 Another study restricted to face lift cases reported that 59% of surgeons administered no form of thromboprophylaxis. 24 A more recent study published in 2007 noted a slightly higher, yet still underwhelming percentage of plastic surgeons who used prophylaxis all the time: 48.7% when performing face lifts, 43.7% for liposuction, and 60.8% for combined procedures. 7 Given that up to one-third of patients diagnosed with DVT can suffer from asymptomatic PE, 25 identifying patients at risk for clotting prior to surgery can help reduce the physical and economic burden of venous thromboembolic disease. Treating a VTE event and managing the secondary morbid outcomes can cost patients thousands of dollars Hospital costs can also be cut substantially if proper measures are taken to preempt thrombus development. 26,28 Different models have been used to predict VTE development, 17-19,29,30 and some have been tailored specifically to plastic surgery, among them the Davison-Caprini RAM. 20 Despite the prevalence of these models, few studies have compared models to each other. One study comparing the Caprini RAM with 2 other VTE prediction models found Caprini scores more useful in predicting thrombus-related risks in both non-surgical and surgical patients. 31 The relative lack of studies comparing different RAMs was the impetus for our investigation. In our patient population, the median ASA score assigned to patients without VTE was 2, whereas the median Table 9. Comparison of the ASA and Caprini Sensitivities and Specificities and a Combined Test Utilizing Both of These Methods ASA Sensitivity (95% ( ) Specificity (95% ( ) Caprini Sensitivity (95% ( ) Specificity (95% ( ) Combined test using ASA & Caprini Sensitivity Specificity score assigned to patients who did experience DVT/PE was 3. After analysis, the scores assigned were statistically different (P <.001) between groups, indicating that anesthesiologists typically assigned a higher score preoperatively to the patients who eventually went on to experience clot related complications. This contradicts prior literature, which has reported that ASA scores are subjective and are 32, 33 not consistently assigned. One reason our results reflected ASA scores for higher DVT/PE patients might be the design of the ASA physical status classification system. An ASA score of 2 is assigned for moderate systemic disturbance whereas a score of 3 is reserved for severe systemic disturbance. 15 Despite the subjective nature of these descriptions, the words moderate and severe can influence the way an anesthesiologist assigns scores. Each word has a very specific connotation, and the results indicate that perhaps the physicians effectively limit a score of 3+ to patients in very advanced disease states as intended by score criteria. This notion is further supported by a study stating that more experienced anesthesiologists have a tendency to use lower ranks when assigning ASA scores. 34 This might also explain why the high-risk ASA patients had such a high rate of DVT development (P <.001, OR = 4.338). It is technically possible that an otherwise healthy individual who has had 2 prior VTEs be assigned an ASA of 1 or 2 but would be at very high risk of a subsequent VTE. A score of 3+ might have been reserved for patients who were severely ill and thus more prone to develop complications postoperatively. Examples of conditions potentially warranting an ASA 3, 4, or 5 include trauma, cancer, respiratory disease, stroke, severe diabetes, age, and other conditions that are also risk factors for VTE development. 1,35 The Caprini RAM was designed to assign point values to specifically listed risk factors to quantify a patient s risk for DVT or PE. 19,21 In this cohort, the average patient was

7 Shaikh et al 503 assigned a baseline score of 4 (Age: = 1 point, BMI: > 25 = 1 point, Major surgery: >45 minute = 2 points). A score of 5 or greater is the cutoff for high risk and necessitates use of both mechanoprophylaxis and chemoprophylaxis as per ACCP guidelines. 21,36 However, the average patient is already on the cusp of the high-risk cutoff. This slight inflation of scores was an important reason that both 2-tiered models failed to note significance when assessing the relationship between Caprini tiers and VTE incidence. Despite this, it was established that high-risk categorization in either model corresponded to an increase in a patient s odds of DVT/PE, supporting the findings initially elicited by other studies. 19,21,37 The current ASA system of low risk = 0-2, high risk 3+ is well-suited as a screening tool for identifying those patients at risk for DVT with an OR of If these patients subsequently have a Caprini score of 5+, they were at a much greater risk of DVTs. Thus a screening test would be seeing if ASA > 2, and then if Caprini was >4, then these patients were at a statistically significant risk of getting DVT. Extra care should be given to patients who fall under these categories in the future and doctors need to ensure that they take proper precautions in these patients. Three-tiered models were created to determine whether a clearer relationship could be established given the large spread of Caprini Scores. Evidence suggests that at a Caprini score of 9+, a patient s risk for VTE increases, potentially up to 11%. 37 Using the altered 1-5 low risk, 6-8 high risk, and 9+ highest-risk cutoffs, a statistically significant relationship between risk tier and VTE development was discovered (P <.0266) that was not seen using the cutoff 1-4 for low risk (P =.1008). This was likely due to the large number of patients who received a score of 5. In concordance with Pannucci et al, this study supported the observation that a higher Caprini score leads to increasing risk for thrombus complications postoperatively. 37 This was reflected by the increasing odds ratios when examining the two 3-tiered models. The 3-tiered model was designed to further stratify patients and determine a better relationship between scoring and VTE development. The usage of a 2-tiered model mistakenly implies that all patients at high risk have similar odds of developing clots postoperatively. One potential limitation of the Caprini RAM is its incomplete nature. Although the original Caprini RAM was modified for application in plastic surgery, 20 it has been noted that the model has previously neglected to address risk factors for DVT that surgeons must account for. 38 Furthermore, 2 patients with the same Caprini score may not have the same physiological risk for DVT. The model states that major surgeries over 45 minutes warrant 2 points, but it fails to factor in risks for VTE unique to certain procedures. Physiologically surgeons understand that there is a difference between a 4-hour rhytidectomy and a 4-hour abdominoplasty with liposuction. One study showed that almost half of the patients who experienced DVT or PE underwent an abdominoplasty. 39 Furthermore, when combining abdominoplasty with other intra-abdominal procedures, the incidence of VTE increases. 39,40 The reasons for this could be multifactorial and may relate to surgical time, immobilization, changes in intra-abdominal pressures, and physiological changes among many etiologies. While it receives greater emphasis in the Caprini RAM, both Caprini and ASA risk assessment models rely exclusively upon the mere presence of certain characteristics while failing to comprehensively delineate the context of those characteristics. Further studies using multivariate regression might be able to examine particular variables, including age, BMI, operative time, specific procedure type, and type and duration of prophylactic regimen to determine degree at which they are factored into each risk assessment model. The biggest limitation of this study was sample size, as only 24 patients in a 4-year period developed thrombusrelated complications. Further, the sample population for Parkland Memorial Hospital has a large number of comorbidities and is not reflective of the plastic surgery population in general unless the practice is very specialized. A multi-institutional study would be extremely beneficial in gathering data to better grasp how patients are categorized in various RAMs and how that categorization relates to VTE. Investigations with a larger patient population and utilizing other VTE RAMs would be advantageous when investigating which can most reliably predict DVT or PE postoperatively and to see how well each risk tier correlates to a similar risk tier in another model. Ultimately, this study can contribute to the existing literature by encouraging further investigation into DVT prophylaxis to reduce morbidity and mortality in plastic surgery, and potentially other fields of surgery as well. The extra step in screening can help with taking precautions in patients at risk for clots. CONCLUSION Both the ASA PS classification system and Caprini RAM can predict VTE development to some extent, yet there are some limitations on both systems, chief among them being vagueness and lack of context. Limitations aside, the ASA scoring system showed a profound increase in the odds of clot development when grouped into the high-risk vs. low-risk category, yet the 2-tiered Caprini RAMs concentrated a higher percentage of VTE in the high-risk category as compared to ASA high risk. Using the Caprini RAM scores to administer prophylaxis is the standard, but taking extra precautions with patients assigned an ASA score of 3 or above may further minimize VTE development. Ultimately, VTE pathogenesis is multietiological and using a checkbox system with numbers to categorize risk fails to

8 504 Aesthetic Surgery Journal 36(4) adequately factor in all of the potential causes of clot development. A more contextual approach used in conjunction with the risk assessment models may better address the patient s needs preoperatively so that the chances of postoperative complications can be minimized. Disclosures Dr Davis receives grants from ConvaTec (Skillman, NJ), Innovative Therapies, Inc. (Pompano Beach, FL), Unilever (Englewood Cliffs, NJ), Andrew Technologies (Tustin, CA), and Kensey Nash (Exton, PA). Dr Kenkel is an investigator for Allergan (Irvine, CA) and Solta (Hayward, CA), and is on the advisory boards of Kythera (Calabasas, CA) and Ulthera (Mesa, AZ). The other authors have nothing to disclose. Funding The authors received no financial support for the research, authorship, and publication of this article. REFERENCES 1. Anderson FA Jr., Spencer FA. Risk factors for venous thromboembolism. Circulation. 2003;107(23 suppl 1): I9-I Silver D. An overview of venous thromboembolism prophylaxis. Am J Surg. 1991;161(4): Geerts WH, Heit JA, Clagett GP, et al. Prevention of venous thromboembolism. Chest. 2001;119(suppl 1):132s-175s. 4. Gutt CN, Oniu T, Wolkener F, Mehrabi A, Mistry S, Buchler MW. Prophylaxis and treatment of deep vein thrombosis in general surgery. Am J Surg. 2005;189(1): Senaran H, Acaroglu E, Ozdemir HM, Atilla B. Enoxaparin and heparin comparison of deep vein thrombosis prophylaxis in total hip replacement patients. Arch Orthop Trauma Surg. 2006;126(1): Agnelli G. Prevention of venous thromboembolism in surgical patients. Circulation. 2004;110(24 suppl 1):Iv4-Iv Broughton G II, Rios JL, Rohrich RJ, Brown SA. Deep venous thrombosis prophylaxis practice and treatment strategies among plastic surgeons: survey results. Plast Reconstr Surg. 2007;119(1): Grazer FM, de Jong RH. Fatal outcomes from liposuction: census survey of cosmetic surgeons. Plast Reconstr Surg. 2000;105(1): ; discussion Kearon C. Duration of venous thromboembolism prophylaxis after surgery. Chest. 2003;124(suppl 6):386s-392s. 10. Ramzi DW, Leeper KV. DVT and pulmonary embolism: part II. treatment and prevention. Am Fam Physician. 2004;69(12): Goldhaber SZ, Fanikos J. Cardiology patient pages. prevention of deep vein thrombosis and pulmonary embolism. Circulation. 2004;110(16):e445-e van Uchelen JH, Werker PM, Kon M. Complications of abdominoplasty in 86 patients. Plast Reconstr Surg. 2001;107(7): Aly AS, Cram AE, Chao M, Pang J, McKeon M. Belt lipectomy for circumferential truncal excess: the University of Iowa experience. Plast Reconstr Surg. 2003;111(1): Chen CM, Disa JJ, Cordeiro PG, Pusic AL, McCarthy CM, Mehrara BJ. The incidence of venous thromboembolism after oncologic head and neck reconstruction. Ann Plast Surg. 2008;60(5): Saklad M. Grading of patients for surgical procedures. Anesthesiology. 1941;2(3): Owens WD, Felts JA, Spitznagel EL Jr. ASA physical status classifications: a study of consistency of ratings. Anesthesiology. 1978;49(4): Tiret L, Hatton F, Desmonts JM, Vourc h G. Prediction of outcome of anaesthesia in patients over 40 years: a multifactorial risk index. Stat Med. 1988;7(9): Wolters U, Wolf T, Stutzer H, Schroder T. ASA classification and perioperative variables as predictors of postoperative outcome. Br J Anaesth. 1996;77(2): Caprini JA, Arcelus JI, Hasty JH, Tamhane AC, Fabrega F. Clinical assessment of venous thromboembolic risk in surgical patients. Semin Thromb Hemost. 1991;17(suppl 3): Davison SP, Venturi ML, Attinger CE, Baker SB, Spear SL. Prevention of venous thromboembolism in the plastic surgery patient. Plast Reconstr Surg. 2004;114(3):43e-51e. 21. Caprini JA. Thrombosis risk assessment as a guide to quality patient care. Dis Mon. 2005;51(2-3): Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008(suppl 6);133:381s-453s. 23. Cohen AT, Tapson VF, Bergmann JF, et al. Venous thromboembolism risk and prophylaxis in the acute hospital care setting (ENDORSE study): a multinational crosssectional study. Lancet. 2008;371(9610): Reinisch JF, Bresnick SD, Walker JW, Rosso RF. Deep venous thrombosis and pulmonary embolus after face lift: a study of incidence and prophylaxis. Plast Reconstr Surg. 2001;107(6): ; discussion Tzoran I, Saharov G, Brenner B, et al. Silent pulmonary embolism in patients with proximal deep vein thrombosis in the lower limbs. JThrombHaemost. 2012;10(4): Pineo G, Lin J, Stern L, Subrahmanian T, Annemans L. Economic impact of enoxaparin versus unfractionated heparin for venous thromboembolism prophylaxis in patients with acute ischemic stroke: a hospital perspective of the PREVAIL trial. J Hosp Med. 2012;7(3): Bullano MF, Willey V, Hauch O, Wygant G, Spyropoulos ACHoffman L. Longitudinal evaluation of health plan cost per venous thromboembolism or bleed event in patients with a prior venous thromboembolism event during hospitalization. J Manag Care Pharm. 2005;11(8): Avorn J, Winkelmayer WC. Comparing the costs, risks, and benefits of competing strategies for the primary prevention of venous thromboembolism. Circulation. 2004; 110(24 suppl 1):Iv25-Iv Barbar S, Noventa F, Rossetto V, et al. A risk assessment model for the identification of hospitalized medical patients

9 Shaikh et al 505 at risk for venous thromboembolism: the Padua Prediction Score. JThrombHaemost. 2010;8(11): Wells PS, Anderson DR, Bormanis J, et al. Value of assessment of pretest probability of deep-vein thrombosis in clinical management. Lancet. 1997;350(9094): Zhou HX, Peng LQ, Yan Y, et al. Validation of the Caprini risk assessment model in Chinese hospitalized patients with venous thromboembolism. Thromb Res. 2012;130 (5): Aronson WL, McAuliffe MS, Miller K. Variability in the American society of anesthesiologists physical status classification scale. AANA J. 2003;71(4): Haynes SR, Lawler PG. An assessment of the consistency of ASA physical status classification allocation. Anaesthesia. 1995;50(3): Bernard PA, Makin CE, Hongying D, Ballard HO. Variability of ASA physical status class assignment among pediatric sedation practitioners. Int J Adolesc Med Health. 2009;21(2): Martinelli I. Risk factors in venous thromboembolism. Thromb Haemost. 2001;86(1): Guidelines ACoCPE-BCP. Antithrombotic therapy and prevention of thrombosis, 9th ed. Chest. 2012;141 (2_suppl). 37. Pannucci CJ, Bailey SH, Dreszer G, et al. Validation of the Caprini risk assessment model in plastic and reconstructive surgery patients. JAmCollSurg. 2011;212(1): Hatef DA, Kenkel JM, Nguyen MQ, et al. Thromboembolic risk assessment and the efficacy of enoxaparin prophylaxis in excisional body contouring surgery. Plast Reconstr Surg. 2008;122(1): Spring MA, Gutowski KA. Venous thromboembolism in plastic surgery patients: survey results of plastic surgeons. Aesthet Surg J. 2006;26(5): Hsu P, Basu CB, Venturi M, Davison S. Venous thromboembolism prophylaxis. Semin Plast Surg. 2006;20(4):