Patterns of Preoperative Laboratory Testing in Patients Undergoing Outpatient Plastic Surgery Procedures

Transcription

1 Research Patterns of Preoperative Laboratory Testing in Patients Undergoing Outpatient Plastic Surgery Procedures John P. Fischer, MD; Eric K. Shang, MD; Jonas A. Nelson, MD; Liza C. Wu, MD; Joseph M. Serletti, MD; and Stephen J. Kovach, MD Preoperative laboratory testing is commonplace and is often utilized at surgeon discretion without formal evidence-based input. 1-3 Although the cost of laboratory testing for an individual may be low and of little fiscal consequence, the aggregate cost of inefficient or potentially unnecessary preoperative laboratory testing in low-risk ambulatory surgery patients may be profound and approaches $30 billion per year. 4-6 The number of ambulatory surgery procedures per year in the United States is increasing and represents more than twothirds of the surgical procedures performed annually. 7 Unfortunately, recommendations for preoperative testing can vary widely, depending on the studies and reviews Aesthetic Surgery Journal 2014, Vol 34(1) The American Society for Aesthetic Plastic Surgery, Inc. Reprints and permission: journalspermissions.nav DOI: / X Abstract Background: Preoperative laboratory testing is commonplace in the clinical setting and is often utilized at surgeon discretion. We searched the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) data set to determine the impact of preoperative laboratory testing in ambulatory plastic surgery patients. Objective: The authors assess the utilization and predictive value of preoperative laboratory testing in outpatient plastic surgery procedures. Methods: Patients undergoing ambulatory plastic surgery were identified from the 2005 to 2010 NSQIP databases. Laboratory tests were categorized by group: hematologic, chemistry, coagulation, and liver function tests (LFT). We defined complications in 2 groups: major postoperative and wound complications. Multivariate analyses were used to identify patient characteristics associated with testing and to assess the ability of laboratory testing to predict postoperative complications. Results: A total of 5359 (62.0%) patients underwent testing; 881 (16.4%) tests were performed on the day of surgery. In patients with no defined NSQIP comorbidities, 59.4% underwent preoperative testing and had a significantly lower rate of abnormal findings (33.4% vs 25.3%, P <.0001). In multivariate analyses, testing was associated with older age, American Society of Anesthesiologists class >2, Hispanic or African American race, body contouring procedures, epidural or spinal procedures, and with diabetes, hypertension, and cancer. Major complications occurred in 0.34% of patients. Our analysis demonstrated that neither testing nor abnormal results were associated with postoperative complications, either major (P =.178) or wound (P =.150). Conclusions: We found no association between abnormal laboratory testing and postoperative morbidity. Preoperative testing in low-risk ambulatory plastic surgery patients may be costly and has limited direct clinical benefit. Keywords preoperative testing, complication, ambulatory surgery, laboratory, labs Accepted for publication April 25, cited. 4,8-10 Many authors have advocated against the use of routine preoperative laboratory testing, especially in asymptomatic and clinically normal patients undergoing elective, low-risk surgery, on the basis that it infrequently From the Department of Surgery, Division of Plastic Surgery, Hospital of the University of Pennsylvania, Philadelphia. Corresponding Author: Dr John P. Fischer, Division of Plastic Surgery, University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104, USA. John.Fischer2@uphs.upenn.edu

2 134 Aesthetic Surgery Journal 34(1) changes management. In fact, several randomized controlled trials featuring the elimination of preoperative testing in lowrisk patients found no difference in adverse events. 11,12 Despite this evidence, preoperative testing is still common in the ambulatory plastic surgery setting. In this study, we searched the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) data sets to determine the impact of preoperative laboratory testing in ambulatory plastic surgery patients. We examined all patients undergoing ambulatory plastic surgery and also a subgroup of patients with no NSQIP-quantified comorbidities, presuming that this group of patients had no indication for preoperative testing. We also identified factors associated with the use of preoperative laboratory testing and assessed the impact of obtaining preoperative laboratory results on adverse patient outcomes. Methods General We reviewed the 2005 to 2010 ACS-NSQIP databases, identifying encounters for outpatient plastic surgery procedures using defined NSQIP parameters. 13 Institutional review board exemption was obtained for this study. Deidentified patient information is freely available to all institutional members who comply with the ACS-NSQIP Data Use Agreement. The Data Use Agreement implements the protections afforded by the Health Insurance Portability and Accountability Act (HIPAA) of The ACS-NSQIP and the hospitals participating in the ACS-NSQIP are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors of this study. The NSQIP data are collected by trained research nurses at various institutions using a systematic sampling of general and vascular operations performed in each participating institution. Results from audits completed to date reveal a disagreement rate of 1.8% for program variables. Each data set contains 240 HIPAA-compliant variables for each case encounter, including patient demographics, preoperative risk factors, baseline comorbidities, intraoperative variables, and 30-day postoperative morbidity and mortality. The list and definitions of variables collected in the database can be found at the ACS-NSQIP website ( Patients are contacted by letter or telephone survey after discharge to ensure a full 30-day follow-up period. Data were accessed on December 1, Cohort Selection The NSQIP Participant Use Files (PUF) include patients who underwent surgery at participating institutions between 2005 and Patients who underwent outpatient plastic surgery procedures were selected through defined NSQIP variables for surgspec and inout. We eliminated patients with age <18 years, incomplete data for sex or ethnicity, American Society of Anesthesio logists (ASA) physical status class 4 or 5, emergent operations, acute renal failure, impaired sensorium, ventilatory support, or sepsis. The final cohort included 8645 patients (Figure 1). Specific patient characteristics evaluated in this study included age, sex, race, obesity (body mass index [BMI] 30 kg/m 2 ), and presence of comorbidities. Procedurerelated variables included year of surgical procedure, type of anesthesia administered, and type of procedure. Procedure types were divided into 5 main groups: breast reconstruction, body contouring, and face, hand, and general reconstruction. These groupings, along with the 2 most common Current Procedural Terminology (CPT) codes in each categorized grouping, are summarized in Table 1. Laboratory Testing Preoperative laboratory testing was defined as testing that occurred within 30 days before a surgical procedure. The NSQIP database includes the following laboratory tests: hematocrit, white blood cell (WBC) count, platelet count, sodium, serum urea nitrogen (SUN), creatinine, partial thromboplastin time (PTT), prothrombin time (PT), international normalized ratio (INR), albumin, total bilirubin, aspartate aminotransferase (AST), and alkaline phosphatase. Preoperative tests were grouped by type. Hematology tests were defined to include hematocrit, WBC, and platelets. Chemistry tests were defined to include sodium, SUN, and creatinine. Coagulation tests were defined to include PTT, PT, and/or INR. Liver function test (LFT) panel was defined to include albumin, AST, total bilirubin, and alkaline phosphatase. We defined abnormal laboratory values based on previously published reports using the NSQIP data sets. 1 Normal ranges were defined as follows: hematocrit, 34% to 45%; WBC, 4000/mm 3 to /mm 3 ; and platelets, /mm 3 to /mm 3. Sodium levels from 135 to 145 mmol/l, SUN <23 mg/dl, and creatinine <1.04 mg/ dl. Coagulation tests were defined as normal if the following conditions were met: PTT <38 seconds, PT <14.7 seconds, and INR <1.5. LFT were defined as normal if the following conditions were met: albumin >3.5 g/dl, total bilirubin <1.1 mg/dl, AST <40 U/L, and alkaline phosphatase <122 U/L. Outcomes The primary outcome was to determine the incidence of patients who underwent preoperative laboratory testing prior to elective, ambulatory plastic surgery. This included a breakdown of test type and the incidence of abnormal test results. Of interest was the use of laboratory testing in patients with and without NSQIP-defined comorbidities. Additionally, we sought to characterize the use of sameday laboratory testing and its impact on management.

3 Fischer et al 135 Figure 1. Selection criteria used to obtain study cohort of ambulatory plastic surgery patients. ASA, American Society of Anesthesiologists; CPT, Current Procedural Terminology; NSQIP, National Surgical Quality Improvement Program. We defined complications into 2 groups: major postoperative events and wound complications. The incidence of any complication was defined as either a major or wound complication. Major postoperative complications were specifically defined according to previously published criteria based on reported NSQIP variables 1 : unplanned intubation, pulmonary embolism, stroke, coma for greater than 24 hours, renal failure requiring dialysis, myocardial infarction, cardiac arrest, sepsis, septic shock, blood transfusions, or death. Wound-related complications were similarly defined and included superficial and deep surgical site infections, organ space infections, and wound dehiscence. We also assessed whether patients required >24 hours admission. Statistical Analysis Descriptive statistics were obtained for the overall cohort, and the use of each type of preoperative testing was described. Patient-related and procedure characteristics were compared between those who underwent laboratory testing and those who did not. Pearson χ 2 tests were used to analyze categorical variables; unpaired Student t tests were employed for continuous variables. Comorbidities with an incidence less than 1% in both groups were not reported. Subgroup analysis was performed for the cohort of patients who underwent same-day preoperative laboratory testing and those with no NSQIP-defined comorbidities. Multivariate logistic regression analysis was used to determine factors predictive of the use of preoperative laboratory testing and the effect of preoperative laboratory testing on the incidence of postoperative complications. Logistic regression was performed using backward selection methods, with a cutoff of P <.10. All tests were 2-tailed, and statistical significance was defined as P <.05. All statistics were calculated using MATLAB algorithms (MathWorks, Natick, Massachusetts). Results Cohort Characteristics During the study period, a total of 8645 patients underwent ambulatory plastic surgery procedures. A total of

4 136 Aesthetic Surgery Journal 34(1) Table 1. Summary of the Most Common Procedures Identified in the Study Cohort From General Categories Table 2. Summary of Patient and Procedure Characteristics Among Those Who Received Preoperative Laboratory Testing Type of Procedure Specific Procedures (Associated CPT Codes) Number No Laboratory Testing (n = 3286) Laboratory Testing (n = 5359) P Value Breast reconstruction Breast reduction (19318) (62.0%) patients underwent testing, of whom 881 (16.4%) underwent same-day testing. The patients undergoing preoperative laboratory testing significantly differed from the patients without testing with regard to age (49.4 ± 15.2 vs 45.6 ± 15.1 years, P <.0001), ASA physical status >2 (20.3% vs 13.4%, P <.0001), Hispanic or African American race (4.4% vs 2.4%, P <.0001), body contouring procedures (8.4% vs 3.7%, P <.0001), epidural or spinal procedures (1.1% vs 0.2%, P <.0001), and in having at least 1 defined NSQIP comorbidity (44.8% vs 38.4%, P <.0001). The univariate analyses of other risk factors associated with laboratory testing are summarized in Tables 2 and 3. Preoperative Tests Implant-based reconstruction (19340, 19342, 19357) Body Panniculectomy/abdominoplasty (15830, 15847) The most commonly obtained tests were complete blood count (CBC) (57.4%), chemistry (44.3%), LFT (21.3%), and coagulation (20.5%). In all, 3811 patients underwent more than 1 type of testing: 2 tests in 17.9%, 3 tests in 15.0%, and all 4 tests in 11.2%. Abnormal testing was observed in 1970 (36.7%) with the following distribution: hematologic, 18.3%; chemistry, 22.4%; coagulation, 6.7%; and LFT, 26.9%. Most patients with abnormal testing had 1 abnormal type of test (79.8%), whereas 17.4% had 2 abnormal tests and 2.6% had 3 abnormal tests. Less than 1% of patients had all 4 types of tests abnormal (n = 5). Of the cohort of patients undergoing preoperative laboratory testing, 16.4% underwent same-day laboratory Suction-assisted lipectomy ( ) 4 Face Orbital fracture (21390) 54 Mandible fracture ( ) 42 Hand Tendon repair (26350, 26356, 26357, 26370, 26372, 26390, 26410, 26412, 26418, 26433) 213 Excision of ganglion wrist (25110, 25111) 173 General Full-thickness skin graft (15200, 15220, 15240, 1526) CPT, Current Procedural Terminology. 238 Adjacent tissue transfer (14300, 14301) 91 Age, mean (SD), y 45.6 (15.1) 49.4 (15.2) <.0001 Sex Race Male 762 (23.2) 826 (15.4) <.0001 Female 2524 (77.8) 4533 (84.6) White 2871 (87.4) 4449 (83.0) <.0001 Black 921 (9.8) 648 (12.1) Hispanic 22 (0.7) 158 (2.9) Asian 57 (1.7) 82 (1.5) Other 15 (0.5) 22 (0.4) Length of stay, mean (SD), d Procedure type 0.14 (1.7) 0.24 (2.4).005 Breast 1869 (56.9) 3448 (64.3) <.0001 Body contouring 120 (3.7) 452 (8.4) Face 210 (6.4) 231 (4.3) Hand 777 (23.6) 704 (13.1) General reconstruction ASA classification 310 (9.4) 524 (9.8) Class (21.6) 899 (16.8) <.0001 Class (64.9) 3370 (62.9) Class (13.4) 1090 (20.3) Anesthesia type General 2938 (89.4) 4985 (93.0) <.0001 MAC 164 (5.0) 188 (3.5) Epidural/spinal 9 (0.2) 61 (1.1) Local/regional 171 (5.2) 122 (2.3) Values are presented as number (%) unless otherwise indicated. ASA, American Society of Anesthesiologists; MAC, monitored anesthesia care. testing, and 33.5% (n = 296) patients had an abnormal finding (Table 4). All of these patients went on to undergo surgery. In patients with no defined NSQIP comorbidities, 2957 (59.4%) still underwent preoperative testing and with a significantly lower rate of abnormal findings (25.3% vs

5 Fischer et al 137 Table 3. Summary of Comorbidities Among Patients Who Received Preoperative Laboratory Testing Table 4. Incidence of Abnormal Tests From Study Group Including Day of Surgery and Noncomorbid Patient Groups At least 1 NSQIP comorbidity No Laboratory Testing (n = 3286), No. (%) Laboratory Testing (n = 5359), No. (%) P Value 1264 (38.4) 2402 (44.8) <.0001 Overall cohort (n = 8645) Total, No. (%) Abnormal, No. (%) Any test 5359 (62.0) 1970 (36.7) At least 2 NSQIP comorbidities Obesity (BMI >30 kg/m 2 ) 80 (2.4) 395 (5.5) < (28.2) 1785 (33.5) <.0001 Smoking 682 (20.8) 849 (15.8) <.0001 Alcohol 78 (2.4) 95 (1.8).057 Diabetes 178 (5.4) 531 (10.4) <.0001 Hypertension 610 (18.6) 1573 (29.4) <.0001 Dyspnea 88 (2.6) 176 (3.3).122 Chronic obstructive pulmonary disease Previous percutaneous cardiac intervention 32 (1.0) 73 (1.4) (1.2) 93 (1.7).033 Bleeding disorders 23 (0.7) 74 (1.4).003 Prior transient ischemic attack 20 (0.6) 66 (1.2).004 Prior stroke 10 (0.3) 36 (0.7).022 Currently on steroids 30 (0.9) 68 (1.2).143 Chemotherapy (within 30 days) 32 (1.0) 95 (1.8).002 BMI, body mass index; NSQIP, National Surgical Quality Improvement Program. 36.7%, P <.0001). A total of 57.1% underwent hematology tests, 38.1% chemistry, 20.4% LFT, and 20.8% coagulation. Factors Associated With Testing Tables 5 and 6 summarize the multivariate regression analyses that assess independently associated factors related to preoperative testing. The analysis demonstrated that use of preoperative testing was associated with older age, ASA physical status >2, Hispanic or African American race, body contouring procedures, epidural or spinal procedures, and diabetes, hypertension, or cancer. With respect to race, Hispanics were more likely to undergo CBC, chemistry, LFT, and coagulation testing (Table 5). African Americans also more frequently underwent all types of testing except LFT. Patients undergoing body contouring procedures more frequently underwent all types of laboratory testing, whereas those undergoing hand Hematology 4966 (57.4) 1115 (22.4) Chemistry 3827 (44.3) 700 (18.3) LFT 1839 (21.3) 495 (26.9) Coagulation 1769 (20.5) 119 (6.7) Day of surgery tests (n = 881) Any test 881 (100.0) 296 (33.5) Hematology 698 (79.2) 156 (22.3) Chemistry 429 (48.7) 78 (18.2) LFT 67 (7.6) 17 (25.3) Coagulation 130 (14.8) 25 (19.2) No comorbidities (n = 4979) Any test 2957 (59.4) 748 (25.3) Hematology 2845 (57.1) 485 (17.0) Chemistry 1899 (38.1) 163 (8.6) LFT 1016 (20.4) 169 (16.6) Coagulation 1038 (20.8) 37 (3.6) LFT, liver function tests. or face surgery less frequently underwent all types of testing. Interestingly, patients receiving epidural or spinal anesthesia more frequently received all types of preoperative tests. Older patients also underwent all types of testing at higher rates. Diabetic patients more frequently underwent chemistry, patients with bleeding disorders more frequently underwent coagulation testing, and those with known alcohol use more often underwent LFT (Table 6). Outcomes The use of preoperative testing was not associated with major postoperative complications (0.42% vs 0.21%, P =.178) or wound complications (2.1% vs 1.7%, P =.150) (Table 7). Multivariate logistic regression analysis analyzing specific laboratory tests demonstrated that neither the performance of preoperative testing nor the presence of abnormal results was associated with postoperative complications. Interestingly, patients who underwent preoperative laboratory testing were more likely to stay for >24 hours postoperatively (2.7% vs 1.5%, P <.001). Furthermore, abnormal preoperative hematologic (odds

6 138 Aesthetic Surgery Journal 34(1) Table 5. Summary of Patient and Procedural Characteristics Associated With Specific Preoperative Laboratory Tests OR (95% CI) Any Testing CBC Chemistry LFT Coagulation Race White Reference Black ( ) 1.45 ( ) 1.56 ( ) NS 1.28 ( ) Hispanic 5.8 ( ) 6.11 ( ) 1.58 ( ) 1.61 ( ) 4.36 ( ) Procedure Breast ratio [OR], 1.45; P <.001) and LFT (OR, 1.63; P <.001) results were associated with higher rates of >24-hour hospital admission (Table 8). Discussion Reference Body contour 2.14 ( ) 2.19 ( ) 1.28 ( ) 1.49 ( ) 1.41 ( ) Face 0.65 ( ) 0.59 ( ) NS NS NS Hand 0.55 ( ) 0.50 ( ) 0.65 ( ) 0.48 ( ) 0.38 ( ) LE reconstruction NS 0.81 ( ) NS 0.81 ( ) 0.71 ( ) Anesthesia General Reference Epidural/spinal 3.65 ( ) 3.52 ( ) 3.06 ( ) 0.30 ( ) 3.85 ( ) Local/regional 0.56 ( ) 0.61 ( ) 0.48 ( ) 0.61 ( ) NS MAC 0.75 ( ) 0.72 ( ) 0.62 ( ) 0.60 ( ) NS CBC, complete blood count; CI, confidence interval; LE, lower extremity; LFT, liver function tests; MAC, monitored anesthesia care; NS, not significant; OR, odds ratio. This study represents the first population-based assessment of preoperative laboratory use in ambulatory plastic surgery patients. Using a prospective, national registry of patient data compiled by the NSQIP over a 5-year period, we assessed 8645 plastic surgery patients undergoing lowrisk ambulatory procedures and determined that 62.0% of these patients underwent preoperative laboratory tests. Even in study patients with no defined comorbid conditions, a majority (59.4%) underwent preoperative testing. These summary statistics highlight the inefficient, non evidence-based utilization patterns of preoperative laboratory testing. Several recent well-designed studies, including metaanalyses and randomized controlled trials, have been performed assessing the utility of preoperative testing in low-risk ambulatory surgery patients, as well as those with stable underlying comorbid conditions. These studies have failed to demonstrate any clinical utility or predictive value to testing. 1,11,12,14-16 The primary indication for utilization of a test is its potential ability to alter clinical decision making and care. 17 Current recommendations for laboratory testing in the United States are derived from the 2002 ASA task force and are based on observational studies and expert opinions, 18 and this has recently been updated with the practice advisory published in Anesthesiology. 19 These expert opinions make it clear that there are no perfect and unambiguous or uniform recommendations in place due to a significant lack of robust literature on preanesthesia testing. The practice advisory suggests that a preoperative electrocardiogram be considered when there is underlying cardiopulmonary disease or the procedure has a potentially significant degree of invasiveness. 19 There is no current consensus about an age cutoff. Chest x-ray should be considered in patients with underlying cardiopulmonary disease, recent upper respiratory tract infections, and chronic lung diseases. Routine preoperative hemoglobin testing is not indicated unless there is significant procedural risk, liver disease, known anemia, or bleeding or hematologic disorder. Preoperative coagulation testing is best reserved for patients undergoing more invasive operations and in those with significant hepatic or renal dysfunction or in cases of known bleeding disorders. Last, routine serum chemistries are reserved for those with underlying hepatic or renal disease and in those with known endocrinopathies or those currently on medications that might alter normal values, such as those on diuretics.

7 Fischer et al 139 Table 6. Summary of Patient Characteristics Associated With Specific Preoperative Laboratory Tests OR (95% CI) Any Testing CBC Chemistry LFT Coagulation ASA classification ASA 2 NS NS 1.26 ( ) 1.51 ( ) NS ASA ( ) NS 1.65 ( ) 1.68 ( ) 1.47 ( ) Age (per year) ( ) ( ) ( ) ( ) ( ) Comorbidities Diabetes 1.23 ( ) NS 1.40 ( ) NS NS Hypertension 1.31 ( ) 1.14 ( ) 1.84 ( ) 1.16 ( ) NS Cancer 5.85 ( ) 4.05 ( ) NS 3.40 ( ) NS Smoking NS 0.88 ( ) NS NS NS Chemotherapy NS 1.51 ( ) 1.81 ( ) 1.71 ( ) NS Bleeding disorder NS NS NS NS 3.65 ( ) ETOH use NS NS NS 1.17 ( ) NS Obesity NS NS NS 0.87 ( ) 0.71 ( ) ASA, American Society of Anesthesiologists; CBC, complete blood count; CI, confidence interval; ETOH use, alcohol use; LFT, liver function tests; NS, not significant; OR, odds ratio. Table 7. Summary of Patient Characteristics Associated With Specific Preoperative Laboratory Tests Overall No. (%) Preoperative Testing No Preoperative Testing P Value Major complication 29 (0.34) 22 (0.42) 7 (0.21).178 Wound complication 169 (2.0) 114 (2.1) 55 (1.7).150 Any complication 187 (2.2) 127 (2.4) 60 (1.8).087 >24-Hour stay 192 (2.2) 144 (2.7) 48 (1.5).0001 There are also no consistent recommendations regarding the timing of preoperative testing, but if there have been no significant clinical changes in history or examination with 6 months, testing can be omitted. 1,19 In brief, the practice guidelines provided by the ASA task force are clear that routine preoperative testing is not advisable; rather, selective testing that might be used to guide or further optimize perioperative management is currently recommended. One of the important aspects of preoperative testing is how it affects decision making and management. Thus, while these data do not allow us to determine the rate at which abnormal results led to procedure cancellation, the fact that a significant number (36.7%) of patients proceeded to surgery despite abnormal tests suggests that these results may not alter management. Furthermore, the incidence of major complications in this study cohort was only 0.34%. Using multivariate logistic regression, neither the use of testing nor the occurrence of an abnormal laboratory value was shown to be of significant predictive clinical value, with respect to major complications or wound complications. In assessing which populations more frequently underwent testing, several notable trends were apparent. First, ethnic minorities were more frequently tested and those patients undergoing body contouring procedures (specifically abdominoplasty/panniculectomy) were more likely to undergo testing. Patients undergoing body contouring procedures may often have underlying laboratory abnormalities related to prior weight loss, which may account for the observed increase in laboratory testing. 20,21 Older patients were also more likely to be tested. Those patients with defined comorbidities, including diabetes and hypertension, were more likely to be tested, as well as those with cancer. Interestingly, patients not receiving general anesthesia were more often tested. Similar patterns were noted by Benarroch-Gampel et al 1 in their evaluation of the utility of preoperative testing in ambulatory hernia repairs. Subgroup analysis did reveal several significant findings. Patients who underwent coagulation testing and patients with abnormal hematologic or LFT tests were more often admitted for >24-hour hospital stay. Overall, patients who underwent any preoperative laboratory testing were more likely to be admitted for >24-hour hospital stay (2.7% vs

8 140 Aesthetic Surgery Journal 34(1) Table 8. Predictive Value of General and Specific Laboratory Testing With Respect to Complications and Admission Overall cohort (test vs no test) Wound Complications OR (95% CI) Major Complications 24-Hour Stay Hematology 1.47 ( ) 1.69 ( ) 1.08 ( ) Chemistry 0.81 ( ) 0.67 ( ) 1.16 ( ) Coagulation 0.78 ( ) 0.30 ( ) 1.65 ( ) Liver function test 0.75 ( ) 2.34 ( ) 1.35 ( ) Overall cohort (abnormal vs normal) Hematology 1.06 ( ) 1.05 ( ) 1.45 ( ) Chemistry 1.03 ( ) 2.05 ( ) 1.09 ( ) Coagulation 1.12 ( ) 1.34 ( ) 1.77 ( ) Liver function test 0.95 ( ) 1.29 ( ) 1.63 ( ) CI, confidence interval; OR, odds ratio. 1.5%, P =.0001). It is likely that these patients were not admitted secondary to their abnormal laboratory finding but because of underlying comorbid conditions necessitating further observation rather than immediate discharge. Unfortunately, due to the nature of this data set, we are unable to determine the reason or rationale for the admission. These findings show that laboratory testing was associated in some instances with rescheduling from outpatient to inpatient, as reflected by a stay >24 hours. Testing was not predictive of complications, nor was abnormal findings. This finding has been confirmed by several other large studies. 16,22,23 Additionally, in patients who underwent same-day testing with abnormal findings, the surgery proceeded as planned, further confirming the limited scope of clinical utility of testing in low-risk ambulatory patients. In the preoperative workup, a comprehensive history and physical examination are paramount, with selective laboratory tests helping to complete the picture of preoperative risk. The data suggest a role of selective, evidencebased preoperative laboratory testing in an effort to minimize risk of patient morbidity and to better reduce cost and resource allocation. 24 Although preoperative laboratory tests may be perceived as a vehicle for assessing overall fitness for anesthesia and surgery, this assumption is likely not true in the low-risk ambulatory surgery patient. 11 As demonstrated in this study, the overuse of testing is likely an inferior substitute for a comprehensive history and physical examination in the low-risk ambulatory patient, further adding to the existing data on this topic. 7 Such overused, unnecessary testing is costly, inconvenient, and of limited clinical utility. 3,7 Several steps must occur before effecting change that reaches beyond the scope of this study. First and foremost, practitioners must recognize that laboratory testing is not indicated in the vast majority of elective, ambulatory surgery patients. However, this is only part of the solution. Looking beyond the patient, practitioners often order laboratory tests for a variety of reasons. Recent data suggest several commonly cited reasons for obtaining a preoperative laboratory workup: fear of litigation, risk for cancellation of case, practice tradition, lack of understanding of guidelines, and belief that other practitioners also would want testing. 2 Clearly, many of these reasons are not evidence based. A comprehensive understanding of the reasons for obtaining testing is needed to optimally effect change and conformance with evidence-based practices. As this occurs, strategies that incorporate evidence-based decision making and include interdisciplinary collaboration with anesthesiologists and primary care physicians may be an opportunity to improve care and cut costs with appropriate preoperative workup and laboratory utilization. 23,25,26 This study is not without limitations. First, the ACS- NSQIP includes only 30-day outcomes and thus may not capture the true picture of postoperative complications or long-term outcomes. Additionally, this study suffers from a significant selection bias, as patients who underwent testing had a higher rate of comorbidities. The greatest limitation of this study is that it lacks the ability to determine the thought process or rationale for testing. Clinical evaluation of a patient and the gestalt of performance status may represent a variable that cannot be captured and may be used in the experienced clinicians decision tree. Furthermore, the occurrence of an intraoperative complication cannot be directly assessed from this data set, which further limits the clinical utility of the outcomes. Also, the lack of direct cost data or individual institutional criteria for defining a laboratory abnormality may somewhat limit the generalizability of these data. Finally, there is a strong selection bias for hospital-based, reconstructive procedures using the NSQIP data sets, and for this reason, these data may not directly translate into practice for the ambulatory, outpatient surgery center patient. It is worth noting, however, that these data in relatively more complex patients undergoing a reconstructive procedure show limited predictive value for preoperative testing; thus, in lower risk ambulatory patients, there would likely be even less benefit derived from testing. Overall, this large population-based assessment of preoperative laboratory use patterns in low-risk ambulatory plastic surgery patients highlights the overuse of testing and its clinical inefficiency as a reliable indicator of perioperative morbidity. These data can be used to effect change in practice patterns of laboratory testing with the ultimate goal of improving patient care, experience, and minimizing extraneous costs. Conclusions This study highlights the overuse of preoperative laboratory testing and the inability of either preoperative laboratory testing or abnormal values to be of significant predictive value

9 Fischer et al 141 with regard to postoperative morbidity. The inefficient use of preoperative laboratory testing in low-risk ambulatory patients represents a potential area for evidence-based interventions to reduce costs. Interestingly, testing and abnormal findings were associated with >24-hour stays. In conclusion, in low-risk ambulatory plastic surgery patients, preoperative laboratory testing is costly, associated with limited clinical benefit, and may be eliminated with significant cost savings. 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. References 1. Benarroch-Gampel J, Sheffield K, Duncan C, et al. Preoperative laboratory testing in patients undergoing elective, low-risk ambulatory surgery. Ann Surg. 2012;256(3): Brown S, Brown J. Why do physicians order unnecessary preoperative tests? A qualitative study. Fam Med. 2011;43(5): Kumar A, Srivastava U. Role of routine laboratory investigations in preoperative evaluation. J Anaesthesiol Clin Pharmacol. 2011;27(2): Fischer S. Cost-effective preoperative evaluation and testing. Chest. 1999;115(5)(suppl):96S-100S. 5. Vogt A, Henson L. Unindicated preoperative testing: ASA physical status and financial implications. J Clin Anesth. 1997;9(6): Marcello P, Roberts P. Routine preoperative studies: which studies in which patients? Surg Clin North Am. 1996;76(1): Richman D. Ambulatory surgery: how much testing do we need? Anesthesiol Clin. 2010;28(2): Munro J, Booth A, Nicholl J. Routine preoperative testing: a systematic review of the evidence. Health Technol Assess. 1997;1(12):i-iv, Smetana G, Macpherson D. The case against routine preoperative laboratory testing. Med Clin North Am. 2003;87(1): Velanovich V. Preoperative laboratory screening based on age, gender, and concomitant medical diseases. Surgery. 1994;115(1): Chung F, Yuan H, Yin L, et al. Elimination of preoperative testing in ambulatory surgery. Anesth Analg. 2009;108(2): Schein O, Katz J, Bass E, et al. The value of routine preoperative medical testing before cataract surgery: Study of Medical Testing for Cataract Surgery. N Engl J Med. 2000;342(3): American College of Surgeon National Surgical Quality Improvement Program. pant-usedata-file/. Accessed October 1, Keay L, Lindsley K, Tielsch J, et al. Routine preoperative medical testing for cataract surgery. Cochrane Database Syst Rev. 2012;3:CD Flamm M, Fritsch G, Seer J, et al. Non-adherence to guidelines for preoperative testing in a secondary care hospital in Austria: the economic impact of unnecessary and double testing. Eur J Anaesthesiol. 2011;28(12): Narr B, Warner M, Schroeder D, et al. Outcomes of patients with no laboratory assessment before anesthesia and a surgical procedure. Mayo Clin Proc. 1997;72(6): Hepner D. The role of testing in the preoperative evaluation. Cleve Clin J Med. 2009;76(suppl 4):S22-S Practice advisory for preanesthesia evaluation: a report by the American Society of Anesthesiologists Task Force on Preanesthesia Evaluation. Anesthesiology. 2002;96(2): Apfelbaum JL, Connis RT, Nickinovich DG, et al. Practice advisory for preanesthesia evaluation: an updated report by the American Society of Anesthesiologists Task Force on Preanesthesia Evaluation. Anesthesiology 2012:116: Naghshineh N, O Brien Coon D, McTigue K, et al. Nutritional assessment of bariatric surgery patients presenting for plastic surgery: a prospective analysis. Plast Reconstr Surg. 2010;126(2): Michaels JT, Coon D, Rubin JP. Complications in postbariatric body contouring: strategies for assessment and prevention. Plast Reconstr Surg. 2011;127(3): Narr B, Hansen T, Warner M. Preoperative laboratory screening in healthy Mayo patients: cost-effective elimination of tests and unchanged outcomes. Mayo Clin Proc. 1991;66(2): Macpherson D. Preoperative laboratory testing: should any tests be routine before surgery? Med Clin North Am. 1993;77(2): Charpak Y, Blery C, Chastang C, et al. Usefulness of selectively ordered preoperative tests. Med Care. 1988;26(2): Finegan B, Rashiq S, McAlister F, et al. Selective ordering of preoperative investigations by anesthesiologists reduces the number and cost of tests. Can J Anaesth. 2005;52(6): Delahunt B, Turnbull P. How cost effective are routine preoperative investigations? N Z Med J. 1980;92(673):