This cohort study investigates whether emergency case status is independently associated with venous thromboembolism compared with elective case status and tests the hypothesis that emergency cases would have a higher risk of venous thromboembolism.
Key Points
Question
Do patients undergoing emergency general surgery have a higher risk of venous thromboembolism than those undergoing elective surgery?
Findings
In this cohort study that included 604 537 adults, the rate of venous thromboembolism within 30 days was 1.9% in emergency general surgery and 0.8% in elective surgery, a statistically significant difference. On multivariable analysis, emergency general surgery was independently associated with venous thromboembolism.
Meaning
Compared with the elective surgery population, the emergency general surgery population had almost twice the risk for venous thromboembolism, and a more aggressive venous thromboembolism chemoprophylaxis regimen should be considered in these patients.
Abstract
Importance
Trauma patients have an increased risk of venous thromboembolism (VTE), partly because of greater inflammation. However, it is unknown if this association is present in patients who undergo emergency general surgery (EGS).
Objectives
To investigate whether emergency case status is independently associated with VTE compared with elective case status and to test the hypothesis that emergency cases would have a higher risk of VTE.
Design, Setting, and Participants
This retrospective cohort study used the American College of Surgeons National Surgical Quality Improvement Program database from January 1, 2005, to December 31, 2016, for all cholecystectomies, ventral hernia repairs (VHRs), and partial colectomies (PCs) to obtain a sample of commonly encountered emergency procedures that have elective counterparts. Emergency surgeries were then compared with elective surgeries. The dates of analysis were January 1 to 31, 2019.
Main Outcomes and Measures
The primary outcome was VTE at 30 days. A multivariable analysis controlling for age, sex, body mass index, bleeding disorder, disseminated cancer, laparoscopy approach, and surgery type was performed.
Results
There were 604 537 adults undergoing surgical procedures over 12 years (mean [SD] age, 55.3 [16.6] years; 61.4% women), including 285 847 cholecystectomies, 158 500 VHRs, and 160 190 PCs. The rate of VTE within 30 days was 1.9% for EGS and 0.8% for elective surgery, a statistically significant difference. Overall, 4607 patients (0.8%) had deep vein thrombosis, and 2648 patients (0.4%) had pulmonary embolism. A total of 6624 VTEs (1.1%) occurred in the cohort. As expected, when VTE risk was examined by surgery type, the risk increased with invasiveness (0.5% for cholecystectomy, 0.8% for VHR, and 2.4% for PC; P < .001). On multivariable analysis, EGS was independently associated with VTE (odds ratio [OR], 1.70; 95% CI, 1.61-1.79). Also associated with VTE were open surgery (OR, 3.38; 95% CI, 3.15-3.63) and PC (OR, 1.86; 95% CI, 1.73-1.99).
Conclusions and Relevance
In this cohort study, emergency surgery and increased invasiveness appeared to be independently associated with VTE compared with elective surgery. Further study on methods to improve VTE chemoprophylaxis is highly recommended for emergency and more extensive operations to reduce the risk of potentially lethal VTE.
Introduction
Seven percent of hospital admissions and 11% to 14% of operations performed are classified as emergency general surgery (EGS).1,2,3,4 However, these emergency operations account for 28% of postoperative complications and 47% to 53% of postoperative deaths.2,3 The disproportionate contribution of EGS to morbidity and mortality exemplifies the increased risk of this surgery type, and studies5,6 have shown that EGS is linked with a higher risk of surgical site infections (SSIs) and readmissions. The argument that patients undergoing EGS have more preoperative risk factors that account for such outcomes was challenged by Havens et al,7 who demonstrated that EGS is an independent risk factor for death. To our knowledge, this avenue of thought has yet to be explored regarding venous thromboembolism (VTE) and EGS.
In 2003, the American Public Health Association identified VTE as the most preventable cause of morbidity and mortality among hospitalized patients in the United States,8 and the yearly economic burden of this condition is estimated to be $7 billion to $10 billion.9 Venous thromboembolism encompasses deep vein thrombosis (DVT) and pulmonary embolism (PE). Although the incidence of VTE in the general population is 1 to 2 cases per 1000 persons,10 there is an exponential increase with age to a rate of 1 case per 100 persons by age 80 years.9 The morbidity and mortality associated with VTE events are substantial. The mortality rate for patients with PE is 20%, and 0.1% to 4% of patients with PE develop chronic thromboembolic pulmonary hypertension; 20% to 50% of patients with DVT develop postthrombotic syndrome; and 30% of patients with a VTE event have a recurrence within 10 years, most in the first year.9 Beckman et al10 reported that a DVT can adversely alter quality of life for up to 4 months after the event.
Therefore, it is recommended to use VTE chemoprophylaxis with short-acting reversible anticoagulants in the perioperative setting to prevent these complications from occurring.11 Specific surgery types and patients, including patients with bariatric surgery,12 cancer,13,14 and trauma,15,16 have increased dosing or higher frequency and duration of chemoprophylaxis given the higher risk of VTE because of stasis, hypercoagulability, and inflammation. Despite a known, well-studied association between these conditions and VTE events, little research has focused on VTE in the EGS population, which may also have these risk factors, especially inflammation. To date, there are no studies on the rate of VTE in EGS compared with the same operations performed electively.17 Using the American College of Surgeons National Surgical Quality Improvement Project (NSQIP) database, we aimed to compare the rates of VTE in the 3 most common operations that fall under the purview of both EGS and elective surgery, namely, cholecystectomies, ventral hernia repairs (VHRs), and partial colectomies (PCs). It was hypothesized that emergency cases would have a higher risk of VTE than their elective counterparts.
Methods
Data Source
In this retrospective cohort study, the NSQIP database was queried from January 1, 2005, to December 31, 2016, for 3 commonly performed procedures with both emergency and elective counterparts in general surgery, namely, cholecystectomies, VHRs, and PCs. These procedures were also chosen to include increasing levels of invasiveness and to stratify the perceived level of VTE risk. Both laparoscopic and open variants of cholecystectomy, VHR, and PC were included. The dates of analysis were January 1 to 31, 2019. The NSQIP data collection and preoperative and intraoperative risk factors and outcomes for the procedures have been described in detail in other studies.18,19,20,21,22 To date, more than 800 hospitals participate in NSQIP data collection,23 and more than 250 variables are recorded for each participant, with additional variables available for colorectal, pancreatic, and hysterectomy procedures. Patients undergoing the 3 procedures were identified by the following Current Procedural Terminology codes: laparoscopic (codes 47562, 47563, and 47564) and open cholecystectomy (codes 47600, 47605, and 47610), laparoscopic (codes 49654, 49655, 49656, and 49657) and open VHR (codes 49560, 49561, 49565, and 49566), and laparoscopic (codes 44188, 44204, 44205, and 44206) and open PC (codes 44140, 44141, 44143, 44144, and 44160). Before data analysis, an institutional review board at UT Southwestern exempted the study from review and informed consent requirements given the use of public deidentified data.
Study Design
The 3 procedures included in the study were selected to provide a breadth of cases commonly performed by both general surgeons and acute care surgeons that had emergency and elective counterparts. We also sought to evaluate the consequences of emergency physiology and inflammatory etiologies on the same surgery type. After identification of patients, cases were coded for laparoscopic approach and by surgery type. If a patient had more than 1 of the 3 procedures, the patient was coded for the more invasive and potentially complicated procedure (the order of greatest to least severity was PC, VHR, and cholecystectomy). The primary dependent variable was elective status as coded in the NSQIP, and the primary outcome was VTE at 30 days. A multivariable analysis controlling for age, sex, body mass index (BMI [calculated as weight in kilograms divided by height in meters squared]), bleeding disorder, disseminated cancer, laparoscopy approach, and surgery type was performed.
NSQIP Outcome Measures
The NSQIP variables are recorded by dedicated abstractors specially trained in surgical outcomes and data collection and include preoperative demographics, comorbidities, laboratory values, operative and postoperative details, 30-day complications and mortality, and disposition. The primary outcome of interest in this study was VTE, which was an aggregate of the component variables of DVT and PE. Secondary outcomes of interest were the time to diagnosis of DVT and PE, reoperation, 30-day readmission, hospital length of stay, and 30-day mortality. The NSQIP does not provide hospital location or region; therefore, hospital clustering methods cannot be used to mitigate bias implicit by regional or local clusters of practice and trends, thereby prohibiting comparison of similar practice areas.
More than 25 complication variables are used in the NSQIP; however, they are not categorized or grouped in a manner to allow for appropriate statistical analysis or reporting. Therefore, complications were grouped using a strategy similar to that used in prior work (general complications and major complications).22 The SSIs were an aggregate of wound and SSI complications. Complications were grouped according to severity: SSIs involved the surgical site, general complications involved less severe medical diagnoses, and major complications involved potentially severe and possibly lethal diagnoses. A similar method has been used in previous studies.22,24,25,26
In addition, a modified version of the Charlson Comorbidity Index (CCI) was applied to the NSQIP data because there is no inherent risk-scoring system present within the NSQIP.27 The CCI is a scoring system that has been extensively validated for surgery patients.28,29,30 Age group and major comorbidities are ranked from 1 (lowest) to 6 (highest) points. Points are then summed to provide a total patient index.31 Previous studies32,33 have verified that the CCI adapted to large administrative databases is sensitive in stratifying mortality, and the CCI has been used with the NSQIP database by 2 other research groups.34,35 The specific approach used in the present study has been reported previously.22
Statistical Analysis
All data were analyzed using SAS, version 9.3 (SAS Institute Inc). Patients with missing data were maintained in the analysis if elective status, the required surgery types, and VTE information were included; no imputation analysis was used. Descriptive statistics were reported as means with corresponding SDs for continuous variables and percentages for categorical variables. Global univariate analyses were performed between patients based on emergency status and elective status. Categorical variables were evaluated using Pearson χ2 test and Fisher exact test where appropriate. Continuous variables and ordinal variables were evaluated using Wilcoxon rank sum test, Mann-Whitney test, and Kruskal-Wallis test. Multivariable logistic regression was performed to evaluate the independent association of emergency surgery with VTE controlling for the following key confounding variables that were established a priori: age, sex, BMI, CCI, independent functional status, tobacco use, history of bleeding disorder or disseminated cancer, and surgery type. Odds ratios (ORs) with corresponding 95% CIs were used to report the results of the multivariable regression models. Statistical significance was set at P < .05, and all reported P values are 2-tailed.
Results
Overall Findings
During the 12-year study period, 604 537 adults (mean [SD] age, 55.3 [16.6] years; 61.4% women) undergoing surgical procedures with the defined Current Procedural Terminology codes were identified in the NSQIP. The patients had 256 726 laparoscopic and 37 311 open cholecystectomies, 33 630 laparoscopic and 128 513 open VHRs, and 62 366 laparoscopic and 98 944 open PCs. These procedures were categorized into the highest level of invasiveness if 2 or more surgery types were performed on the same patient, resulting in 285 847 cholecystectomies, 158 500 VHRs, and 160 190 PCs.
More than half (57.3%) of all operations were performed laparoscopically. Most patients were in their 50s (mean [SD] age, 55.3 [16.6] years), women (61.4%), and overweight (mean [SD] BMI, 30.9 [7.9]) and had few comorbidities (mean [SD] CCI, 0.5 [1.3]).
For the entire population, the rate of VTE within 30 days was low (1.1%), including rates of 0.8% for DVT and 0.4% for PE. The mean (SD) time to diagnosis of DVT was 11.4 (7.8) days. The mean (SD) time to diagnosis of PE was shorter at 10.3 (7.9) days. The aggregate SSI rate was only 4.1% at 30 days. The rate of general complications in this population was 27.0%, the rate of major complications was 6.8%, and the 30-day readmission rate was 7.0%. Reoperation occurred in 3.1% of patients. Death within 30 days of operation occurred in 8443 patients (1.4%).
Demographics by Emergency Status
Patient demographics and comorbidities by emergency status are listed in Table 1. Patients with emergency status were on average older, were more likely to be men, had a lower BMI, and had more comorbidities than patients with elective status. Patients undergoing EGS had a decreased risk of bleeding disorder and more risk factors for VTE, including prior cerebrovascular accident, history of disseminated cancer, and chemotherapy or radiotherapy in the last 30 days.
Table 1. Patient Characteristics by Emergency Status.
| Variable | No. (%) | P value | |
|---|---|---|---|
| Emergency surgery (n = 183 099) | Elective surgery (n = 421 438) | ||
| Age, mean (SD), y | 56.2 (18.0) | 55.0 (15.9) | <.001 |
| Male | 78 857 (43.1) | 161 149 (38.2) | <.001 |
| BMI, mean (SD) | 30.7 (8.4) | 31.1 (7.7) | <.001 |
| Charlson Comorbidity Index, mean (SD) | 0.6 (1.5) | 0.4 (1.2) | <.001 |
| Systemic | |||
| Independent functional status | 171 622 (93.7) | 414 172 (98.3) | <.001 |
| Tobacco use | 36 872 (20.1) | 75 715 (18.0) | <.001 |
| Alcohol use, No./total No. (%)a | 458/16 168 (2.8) | 750/40 947 (1.8) | <.001 |
| >10% Body mass loss in last 6 mo | 6535 (3.6) | 7029 (1.7) | <.001 |
| Endocrine | |||
| Corticosteroid use | 10 209 (5.6) | 15 622 (3.7) | <.001 |
| Diabetes | 30 097 (16.4) | 64 401 (15.3) | <.001 |
| Cardiovascular | |||
| Hypertension | 81 429 (44.5) | 183 961 (43.7) | <.001 |
| Congestive heart failure | 3435 (1.9) | 1816 (0.4) | <.001 |
| MI in last 6 mo, No./total No. (%)a | 135/16 166 (0.8) | 79/40 907 (0.2) | <.001 |
| Prior CVA, No./total No. (%)a | 366/16 166 (2.3) | 430/40 907 (1.1) | <.001 |
| History of cardiac surgery, No./total No. (%)a | 800/16 166 (4.9) | 1482/40 907 (3.6) | <.001 |
| Pulmonary | |||
| COPD | 11 268 (6.2) | 17 019 (4.0) | <.001 |
| Exertional dyspnea | 9580 (5.2) | 24 062 (5.7) | <.001 |
| Kidney | |||
| Kidney failure | 3254 (1.8) | 2523 (0.6) | <.001 |
| Acute kidney failure | 2090 (1.1) | 466 (0.1) | <.001 |
| Hematological/oncological | |||
| History of bleeding disorder | 1040 (0.6) | 9415 (2.2) | <.001 |
| History of disseminated cancer | 8253 (4.5) | 13 375 (3.2) | <.001 |
| Chemotherapy in last 30 d, No./total No. (%)a | 356/16 166 (2.2) | 745/40 907 (1.8) | .003 |
| Radiotherapy in last 30 d, No./total No. (%)a | 109/16 166 (0.7) | 205/40 653 (0.5) | .01 |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident; MI, myocardial infarction.
Information not available on all patients; percentages were calculated based on data available.
Operative Details by Emergency Status
The number of individual procedures and operative characteristics by emergency status are listed in Table 2. The American Society of Anesthesiologists’ physical status classification level, which is used to describe patients’ medical comorbidities before anesthesia (with level I indicating a healthy patient and level V indicating a patient who is not expected to survive without the surgical procedure), was statistically significantly higher in the patients with emergency status, with emergency cases having higher rates of classification levels III, IV, and V than elective cases. Rates of laparoscopy use were higher among the patients with elective status vs those with emergency status (58.8% vs 53.7%, P < .001). For all surgery types, there was a much higher rate of elective case mix, with a higher rate of laparoscopy use for each stratum of procedure among the patients with elective status. Wound class was much more likely to be contaminated or dirty among the patients with emergency status, who were 7.9 times more likely to have dirty wounds than the patients with elective status.
Table 2. Operative Details by Emergency Statusa.
| Variable | No. (%) | |
|---|---|---|
| Emergency surgeryb | Elective surgeryc | |
| ASA classification level | ||
| I | 11 321 (6.2) | 30 844 (7.3) |
| II | 72 170 (39.4) | 210 594 (50.0) |
| III | 76 587 (41.8) | 167 774 (39.8) |
| IV | 21 317 (11.6) | 11 667 (2.8) |
| V | 1400 (0.8) | 67 (0) |
| Preoperative WBC count, mean (SD), /μL | 10 200 (5400) | 7600 (2900) |
| Preoperative creatinine level, mean (SD), mg/dL | 1.0 (0.9) | 0.9 (0.6) |
| Inpatient | 165 036 (90.1) | 205 891 (48.9) |
| Laparoscopy | 98 369 (53.7) | 247 862 (58.8) |
| Cholecystectomy (n = 285 847) | 97 969 (34.3) | 187 878 (65.7) |
| Laparoscopic (n = 256 726) | 87 664 (34.1) | 169 062 (65.9) |
| Open (n = 37 311) | 13 121 (35.2) | 24 190 (64.8) |
| Ventral hernia repair (n = 158 500) | 21 069 (13.3) | 137 431 (86.7) |
| Laparoscopic (n = 33 630) | 2049 (6.1) | 31 581 (93.9) |
| Open (n = 128 513) | 20 354 (15.8) | 108 159 (84.2) |
| Partial colectomy (n = 160 190) | 64 061 (40.0) | 96 129 (60.0) |
| Laparoscopic (n = 62 366) | 10 702 (17.2) | 51 664 (82.8) |
| Open (n = 98 944) | 53 675 (54.2) | 45 319 (45.8) |
| Wound class | ||
| Clean | 15 428 (8.4) | 125 031 (29.7) |
| Clean contaminated | 83 632 (45.7) | 250 106 (59.3) |
| Contaminated | 49 308 (26.9) | 36 060 (8.6) |
| Dirty or infected | 34 731 (19.0) | 10 240 (2.4) |
| Operative time, mean (SD), min | 108.5 (77.3) | 113.6 (6.4) |
Abbreviations: ASA, American Society of Anesthesiologists; WBC, white blood cell.
SI conversation factors: To convert creatinine level to micromoles per liter, multiply by 88.4; to convert WBC count to ×109/L, multiply by 0.001.
Totals are not summative given overlap in surgery type and possible conversions with dual codes; rates for procedures were calculated using the number of procedures as a denominator. P < .001 for all comparisons.
Unless otherwise specified, n = 183 099 emergency procedures.
Unless otherwise specified, n = 421 438 elective procedures.
VTE Outcomes by Emergency Status
The rate of VTE within 30 days was 1.9% in EGS and 0.8% in elective surgery, a statistically significant difference. Overall, 4607 patients (0.8%) had DVT, and 2648 patients (0.4%) had PE. As expected, when VTE risk was examined by surgery type, the risk increased with invasiveness (0.5% for cholecystectomy, 0.8% for VHR, and 2.4% for PC; P < .001).
The rates of VTE by surgery type and laparoscopy approach are listed in Table 3. The VTE rates were higher for open procedures compared with laparoscopic procedures within each stratum of procedure. The rate of VTE within 30 days for open cholecystectomy was 8.3 times higher than that for laparoscopic cholecystectomy, with comparable values 2.5 times higher in VHR and 2.9 times higher in PC. Open PC had the highest rate of VTE within 30 days, with 3.2% having DVT or PE; conversely, laparoscopic cholecystectomy had the lowest rate of VTE within 30 days at 0.3%. A total of 6624 VTEs (1.1%) occurred in the cohort.
Table 3. Venous Thromboembolism (VTE) by Surgery Typea.
| Variable | No. (%) | |
|---|---|---|
| No VTE | VTE | |
| Total laparoscopic (n = 346 231) | 344 795 (99.6) | 1436 (0.4) |
| Total open (n = 258 306) | 253 118 (98.0) | 5188 (2.0) |
| Cholecystectomy (n = 285 847) | 284 410 (99.5) | 1437 (0.5) |
| Laparoscopic (n = 256 726) | 256 044 (99.7) | 682 (0.3) |
| Open (n = 37 311) | 36 375 (97.5) | 936 (2.5) |
| Ventral hernia repair (n = 158 500) | 157 164 (99.2) | 1336 (0.8) |
| Laparoscopic (n = 33 630) | 33 499 (99.6) | 131 (0.4) |
| Open (n = 128 513) | 127 185 (99.0) | 1328 (1.0) |
| Partial colectomy (n = 160 190) | 156 339 (97.6) | 3851 (2.4) |
| Laparoscopic (n = 62 366) | 61 671 (98.9) | 695 (1.1) |
| Open (n = 98 994) | 95 817 (96.8) | 3177 (3.2) |
Totals are not summative given overlap in surgery type and possible conversions with dual codes. P < .001 for all comparisons.
Patient outcomes by emergency status are listed in Table 4. The rate of VTE within 30 days was 240% higher in the patients with emergency status vs those with elective status (1.9% vs 0.8%, P < .001). The rates of DVT and PE were higher in patients with emergency status, and the time to diagnosis of DVT was shorter in these patients by almost 1.5 days. However, the time to diagnosis of PE was similar in both groups. Similarly, in patients with emergency status, the rates of SSI were 1.5 times higher, general complications were 1.9 times higher, and major complications were 3.8 times higher than the rates of their elective counterparts. Reoperation and readmission rates were also increased among patients with emergency status compared with those with elective status. In addition, 30-day mortality was 9 times higher in patients with emergency (3.6%) vs elective status (0.4%) (P < .001).
Table 4. Patient Outcomes by Emergency Status.
| Variable | No. (%) | P value | |
|---|---|---|---|
| Emergency surgery (n = 183 099) | Elective surgery (n = 421 438) | ||
| VTE | 3443 (1.9) | 3181 (0.8) | <.001 |
| DVT or thrombophlebitis | 2562 (1.4) | 2045 (0.5) | <.001 |
| PE | 1190 (0.6) | 1458 (0.3) | <.001 |
| Time to diagnosis of DVT, mean (SD), d | 10.7 (7.8) | 12.1 (7.8) | <.001 |
| Time to diagnosis of PE, mean (SD), d | 10.4 (7.9) | 10.2 (7.8) | .63 |
| Surgical site infections | 9758 (5.3) | 15 121 (3.6) | <.001 |
| General complications No./total No. (%)a | 37 068/91 983 (40.3) | 39 960/193 368 (20.7) | <.001 |
| Major complications | 25 812 (14.1) | 15 517 (3.7) | <.001 |
| Reoperation | 8394 (4.6) | 10 379 (2.5) | <.001 |
| 30-d Readmission | 16 015 (8.7) | 25 407 (6.0) | <.001 |
| Hospital length of stay, mean (SD), d | 2.4 (7.1) | 2.9 (5.8) | <.001 |
| 30-d Mortality | 6601 (3.6) | 1842 (0.4) | <.001 |
Abbreviations: DVT, deep vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism.
Information not available on all patients; percentages were calculated based on data available.
Multivariable Analysis by Emergency Status
On multivariable analysis, EGS was independently associated with VTE (OR, 1.70; 95% CI, 1.61-1.79). Also associated with VTE were open surgery (OR, 3.38; 95% CI, 3.15-3.63) and PC (OR, 1.86; 95% CI, 1.73-1.99).
The multivariable analysis results were adjusted for confounders (Table 5). Emergency cases had an almost 2-fold increased odds of VTE compared with elective cases (OR, 1.70; 95% CI, 1.61-1.79). Other risk factors identified for VTE were increasing age, male sex, BMI, CCI per point increase, dependent functional status, and history of bleeding disorder. Procedures with increased invasiveness were associated with higher VTE risk: patients undergoing open surgery had approximately 3 times the risk for VTE compared with patients undergoing laparoscopic surgery, and patients undergoing PC had almost twice the risk for VTE compared with patients undergoing cholecystectomy.
Table 5. Multivariable Analysis Adjusted Odds for Venous Thromboembolisma.
| Variable | OR PE (95% CI) | P value |
|---|---|---|
| Emergency status | 1.70 (1.61-1.79) | <.001 |
| Age, y | ||
| 18-44 | 1 [Reference] | <.001 |
| 45-54 | 1.55 (1.40-1.72) | |
| 55-64 | 1.88 (1.71-2.07) | |
| 65-74 | 2.37 (2.16-2.61) | |
| 75-84 | 2.70 (2.44-2.99) | |
| ≥85 | 2.77 (2.42-3.18) | |
| Male compared with female | 1.15 (1.09-1.21) | <.001 |
| BMI | ||
| Normal, 18 to 25 | 1 [Reference] | <.001 |
| Underweight, <18 | 1.04 (0.98-1.13) | |
| Overweight, 26 to ≤30 | 1.04 (0.97-1.11) | |
| Obese, 31 to ≤40 | 1.29 (1.20-1.40) | |
| Morbid obesity, 41 to ≤50 | 1.60 (1.44-1.77) | |
| Superobesity, 51 to <60 | 2.04 (1.72-2.43) | |
| Hyperobesity, ≥60 | 1.72 (1.26-2.35) | |
| Charlson Comorbidity Index, per point increase | 1.08 (1.04-1.12) | <.001 |
| Tobacco use | 0.99 (0.92-1.05) | .68 |
| Diabetes | 0.94 (0.87-1.01) | .11 |
| Functional status, dependent vs independent | 1.30 (1.06-1.59) | .26 |
| Disseminated cancer | 1.25 (0.99-1.57) | .05 |
| History of bleeding disorder | 1.28 (1.17-1.39) | <.001 |
| Open surgery, compared with laparoscopy | 3.38 (3.15-3.63) | <.001 |
| Surgery type, compared with cholecystectomy | ||
| Ventral hernia repair | 0.76 (0.70-0.83) | <.001 |
| Partial colectomy | 1.86 (1.73-1.99) | <.001 |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); OR, odds ratio; PE, parameter estimate.
The multivariable analysis results were adjusted for confounders, including emergency status, increasing age, male sex, BMI, Charlson Comorbidity Index per point increase, dependent functional status, and history of bleeding disorder.
Discussion
To our knowledge, this study is the first to demonstrate that EGS is independently associated with VTE, almost doubling a patient’s risk compared with elective surgery. The risk of VTE also increases in proportion to the invasiveness of the procedure, with laparoscopic procedures having lower rates of VTE than open procedures and PCs having higher rates of VTE than cholecystectomies. These findings should be a call to action for surgeons and hospitals to promote research and quality improvement processes aimed at patients undergoing EGS in an effort to prevent and mitigate VTE.
The overall rate of VTE within 30 days of EGS was 1.9%, which is similar to percentages found by McCoy et al2 (2.4%) and by Sakran et al36 (1.3%). The independent risk factors for VTE identified on multivariable regression analysis in the present study were similar to those found by Beckman et al10 and by Sakran et al,36 including age, obesity, and immobility or dependent functional status. We also showed that male sex and history of bleeding disorder were independent risk factors for VTE. Sakran et al36 found emergency colectomies to be an independent risk factor for VTE complications, which we also observed with PCs in our study.
The time to diagnosis of PE was shorter by approximately 1 day compared with the time to diagnosis of DVT. The time to diagnosis of DVT was shorter in emergency cases vs elective cases, but the time to diagnosis of PE was almost the same in both groups. Similar results were reported by Sakran et al.36 This finding can likely be attributed to the overt clinical signs of a PE, which make diagnosis much more obvious compared with the more subtle signs of a developing DVT, as well as an increased index of suspicion regarding potential DVT among acute care surgeons compared with their general surgery counterparts. In addition, acute care surgeons may be more cognizant of the lack of optimal preoperative and postoperative dosing of VTE chemoprophylaxis or of the contributions of missed dosing of VTE chemoprophylaxis to increased risk of VTE given the emergency nature of surgical procedures. Therefore, as others have demonstrated,37 linking the processes surrounding VTE chemoprophylaxis and adherence to VTE prevention measures is vital in understanding and benchmarking the resultant outcomes.
The NSQIP does not record the type of VTE chemoprophylaxis, dosing, or adherence to standard dosing regimens. These quality metrics are vital to understand the process measures behind VTE chemoprophylaxis use and dosing, with prior studies showing that missed doses of enoxaparin sodium are associated with more than a 5-times increased risk of VTE in trauma and general surgery patients.38 Further evidence has shown a high correlation between missed doses and DVT, with only 47% of patients with DVT in 1 study11 receiving defect-free VTE chemoprophylaxis regimens. However, in that study, 86% of patients with DVT were prescribed optimal VTE chemoprophylaxis and still had a VTE, indicating that additional adherence or dosing parameters are required to further decrease the risk of VTE in high-risk patients like the EGS population. A possible mechanism is ensuring that the VTE chemoprophylaxis is in an appropriate range by anti–factor Xa testing, which has proven efficacious in trauma patients previously.39
Although missed dosing of VTE chemoprophylaxis contributes substantially to an increased risk of VTE, greater inflammation resulting from the illness that precipitated EGS likely also is associated with a hypercoagulable state in the EGS population. The fact that open surgery had the highest risk of VTE seems to support this hypothesis because inflammation and tissue disruption would be greater in that setting than in laparoscopic surgery. This finding may be similar to the changes seen after trauma resulting from increased expression of tissue factor, likely associated with the high rates of VTE seen in EGS populations. Patients undergoing EGS and those with trauma are initially seen under emergency circumstances and have little to no chance for preoperative risk modification. Therefore, in addition to the need for better metrics to capture VTE chemoprophylaxis adherence, an emphasis on aggressive postoperative VTE chemoprophylaxis and treatment pathways should also be pursued. Black et al40 and Starr et al41 have demonstrated that implementing an early, risk-stratified VTE chemoprophylaxis algorithm, guided by anti–factor Xa levels, can substantially reduce the risk of VTE among patients with traumatic injuries. Such an approach for patients undergoing EGS is likely to lead to comparable outcomes. At our centers, similar VTE protocols are being enacted for patients undergoing EGS to increase frequency and dosing of VTE chemoprophylaxis among this high-risk population.
After the hospital stay, the risk of VTE continues to be high, with more than 30% of VTEs in the EGS population occurring after admission and contributing to readmission, signaling that outpatient VTE chemoprophylaxis needs further study for high-risk patients undergoing procedures, such as bowel or open surgery.42 Implementation of clinical decision-making tools in the electronic medical record is another important method to increase adherence to and early initiation of VTE chemoprophylaxis and has been found to increase the odds of correct ordering by 2.35 times.43 However, a multidisciplinary approach is needed to decrease missed chemoprophylaxis dosing, initiate measures for identifying VTE chemoprophylactic levels, adjust to higher chemoprophylaxis dosing or other medications, identify patients for earlier VTE chemoprophylaxis, and mitigate VTE when it does occur.44 A group approach and multiple points of attack are required to change the culture of how an institution combats VTE risk. Acute care surgeons should be leading the charge to adopt these and other measures to combat VTE in patients undergoing EGS.
Limitations
This study has several limitations inherent to database research. The NSQIP database is limited to participating institutions and represents only a sample of operations performed; therefore, the NSQIP designation of an emergency procedure may be limited. Also, given the large number of patients, clinically insignificant findings were able to reach statistical significance. Furthermore, missing data in the NSQIP may have been missing not at random; however, complete data were available for the primary outcomes of interest. The NSQIP does not provide information on region, setting (urban, rural, or suburban), or facility type (academic, tertiary, or community) or specific hospital location to be able to cluster hospital outcomes into similar comparator groups. If it had, these data would have added to the value of the analysis because specific practice locations or types may have had much higher rates of VTE and could be targeted for higher-yield VTE reduction quality improvement processes. The rate of screening for DVT is also not reported in the NSQIP and could be high or low in some practices; therefore, VTE may be underreported in some centers vs others. In addition, there are no data in the NSQIP on VTE chemoprophylaxis adherence, dosing, or the type of anticoagulants used. The lack of appropriate VTE chemoprophylaxis might be a high risk factor for VTE and is likely a confounding variable with emergency status (emergency cases are less likely to receive preoperative and postoperative VTE chemoprophylaxis if bleeding or hemodynamic instability is present), which could lead to a type I error (ie, assuming a correlation where there is none). Improved adherence to VTE prevention measures should be pursued for best patient care; however, we are unable to make any inference about this recommendation in the present study given the lack of these data in the NSQIP.
Conclusions
Through multivariable regression analysis of the NSQIP database, EGS was found to be an independent risk factor for VTE, with an OR of 1.70. These data could be used to identify high-risk patient groups that could benefit from further study and quality improvement processes to reduce VTE risk. Our data should serve as a call to action for acute care surgeons to further study the EGS population, implement evidence-based VTE prevention guidelines at their centers, and expand the scope of VTE process measures being recorded in registries like the NSQIP both locally and nationally.
References
- 1.Gale SC, Shafi S, Dombrovskiy VY, Arumugam D, Crystal JS. The public health burden of emergency general surgery in the United States: a 10-year analysis of the Nationwide Inpatient Sample: 2001 to 2010. J Trauma Acute Care Surg. 2014;77(2):202-208. doi: 10.1097/TA.0000000000000362 [DOI] [PubMed] [Google Scholar]
- 2.McCoy CC, Englum BR, Keenan JE, Vaslef SN, Shapiro ML, Scarborough JE. Impact of specific postoperative complications on the outcomes of emergency general surgery patients. J Trauma Acute Care Surg. 2015;78(5):912-918. doi: 10.1097/TA.0000000000000611 [DOI] [PubMed] [Google Scholar]
- 3.Smith M, Hussain A, Xiao J, et al. The importance of improving the quality of emergency surgery for a regional quality collaborative. Ann Surg. 2013;257(4):596-602. doi: 10.1097/SLA.0b013e3182863750 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Shafi S, Aboutanos MB, Agarwal S Jr, et al. ; AAST Committee on Severity Assessment and Patient Outcomes . Emergency general surgery: definition and estimated burden of disease. J Trauma Acute Care Surg. 2013;74(4):1092-1097. doi: 10.1097/TA.0b013e31827e1bc7 [DOI] [PubMed] [Google Scholar]
- 5.Isik O, Kaya E, Dundar HZ, Sarkut P. Surgical site infection: re-assessment of the risk factors. Chirurgia (Bucur). 2015;110(5):457-461. [PubMed] [Google Scholar]
- 6.Muthuvel G, Tevis SE, Liepert AE, Agarwal SK, Kennedy GD. A composite index for predicting readmission following emergency general surgery. J Trauma Acute Care Surg. 2014;76(6):1467-1472. doi: 10.1097/TA.0000000000000223 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Havens JM, Peetz AB, Do WS, et al. The excess morbidity and mortality of emergency general surgery. J Trauma Acute Care Surg. 2015;78(2):306-311. doi: 10.1097/TA.0000000000000517 [DOI] [PubMed] [Google Scholar]
- 8.Thrombosis D-V. Advancing Awareness to Protect Patient Lives. American Public Health Association; 2003. White Paper: Public Health Leadership Conference on Deep-Vein Thrombosis; February 26, 2003; Washington, DC.
- 9.Di Nisio M, van Es N, Büller HR. Deep vein thrombosis and pulmonary embolism. Lancet. 2016;388(10063):3060-3073. doi: 10.1016/S0140-6736(16)30514-1 [DOI] [PubMed] [Google Scholar]
- 10.Beckman MG, Hooper WC, Critchley SE, Ortel TL. Venous thromboembolism: a public health concern. Am J Prev Med. 2010;38(4)(suppl):S495-S501. doi: 10.1016/j.amepre.2009.12.017 [DOI] [PubMed] [Google Scholar]
- 11.Haut ER, Lau BD, Kraus PS, et al. Preventability of hospital-acquired venous thromboembolism. JAMA Surg. 2015;150(9):912-915. doi: 10.1001/jamasurg.2015.1340 [DOI] [PubMed] [Google Scholar]
- 12.Thereaux J, Lesuffleur T, Czernichow S, et al. To what extent does posthospital discharge chemoprophylaxis prevent venous thromboembolism after bariatric surgery? results from a nationwide cohort of more than 110,000 patients. Ann Surg. 2018;267(4):727-733. doi: 10.1097/SLA.0000000000002285 [DOI] [PubMed] [Google Scholar]
- 13.Trinh VQ, Karakiewicz PI, Sammon J, et al. Venous thromboembolism after major cancer surgery: temporal trends and patterns of care. JAMA Surg. 2014;149(1):43-49. doi: 10.1001/jamasurg.2013.3172 [DOI] [PubMed] [Google Scholar]
- 14.Ivatury SJ, Holubar SD. Just do it: changing our culture to embrace optimal perioperative venous thromboembolism chemoprophylaxis for abdominal and pelvic cancer patients. Ann Surg Oncol. 2016;23(5):1420-1421. doi: 10.1245/s10434-016-5130-6 [DOI] [PubMed] [Google Scholar]
- 15.Shaikh S, Boneva D, Hai S, McKenney M, Elkbuli A. Venous thromboembolism chemoprophylaxis regimens in obese trauma and surgery patients: a systematic review. J Trauma Acute Care Surg. Published online October 31, 2019. [DOI] [PubMed] [Google Scholar]
- 16.Karcutskie CA, Meizoso JP, Ray JJ, et al. Association of mechanism of injury with risk for venous thromboembolism after trauma. JAMA Surg. 2017;152(1):35-40. doi: 10.1001/jamasurg.2016.3116 [DOI] [PubMed] [Google Scholar]
- 17.Murphy PB, Vogt KN, Lau BD, et al. Venous thromboembolism prevention in emergency general surgery: a review. JAMA Surg. 2018;153(5):479-486. doi: 10.1001/jamasurg.2018.0015 [DOI] [PubMed] [Google Scholar]
- 18.Advani V, Ahad S, Gonczy C, Markwell S, Hassan I. Does resident involvement effect surgical times and complication rates during laparoscopic appendectomy for uncomplicated appendicitis? an analysis of 16,849 cases from the ACS-NSQIP. Am J Surg. 2012;203(3):347-351. doi: 10.1016/j.amjsurg.2011.08.015 [DOI] [PubMed] [Google Scholar]
- 19.Choi JJ, Palaniappa NC, Dallas KB, Rudich TB, Colon MJ, Divino CM. Use of mesh during ventral hernia repair in clean-contaminated and contaminated cases: outcomes of 33,832 cases. Ann Surg. 2012;255(1):176-180. doi: 10.1097/SLA.0b013e31822518e6 [DOI] [PubMed] [Google Scholar]
- 20.Kern SQ, Lustik MB, McMann LP, Thibault GP, Sterbis JR. Comparison of outcomes after minimally invasive versus open partial nephrectomy with respect to trainee involvement utilizing the American College of Surgeons National Surgical Quality Improvement Program. J Endourol. 2014;28(1):40-47. doi: 10.1089/end.2013.0051 [DOI] [PubMed] [Google Scholar]
- 21.Mason RJ, Moazzez A, Sohn HJ, Berne TV, Katkhouda N. Laparoscopic versus open anterior abdominal wall hernia repair: 30-day morbidity and mortality using the ACS-NSQIP database. Ann Surg. 2011;254(4):641-652. doi: 10.1097/SLA.0b013e31823009e6 [DOI] [PubMed] [Google Scholar]
- 22.Ross SW, Oommen B, Kim M, et al. A little slower, but just as good: postgraduate year resident versus attending outcomes in laparoscopic ventral hernia repair. Surg Endosc. 2014;28(11):3092-3100. doi: 10.1007/s00464-014-3586-7 [DOI] [PubMed] [Google Scholar]
- 23.American College of Surgeons. ACS NSQIP frequently asked questions. Accessed March 13, 2020. https://www.facs.org/quality-programs/acs-nsqip/faq
- 24.Ross SW, Oommen B, Huntington C, et al. National outcomes for open ventral hernia repair techniques in complex abdominal wall reconstruction. Am Surg. 2015;81(8):778-785. [PubMed] [Google Scholar]
- 25.Ross SW, Oommen B, Wormer BA, et al. National outcomes of laparoscopic Heller myotomy: operative complications and risk factors for adverse events. Surg Endosc. 2015;29(11):3097-3105. doi: 10.1007/s00464-014-4054-0 [DOI] [PubMed] [Google Scholar]
- 26.Ross SW, Seshadri R, Walters AL, et al. Mortality in hepatectomy: Model for End-stage Liver Disease as a predictor of death using the National Surgical Quality Improvement Program database. Surgery. 2016;159(3):777-792. doi: 10.1016/j.surg.2015.08.021 [DOI] [PubMed] [Google Scholar]
- 27.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-383. doi: 10.1016/0021-9681(87)90171-8 [DOI] [PubMed] [Google Scholar]
- 28.Birim O, Maat AP, Kappetein AP, van Meerbeeck JP, Damhuis RA, Bogers AJ. Validation of the Charlson comorbidity index in patients with operated primary non-small cell lung cancer. Eur J Cardiothorac Surg. 2003;23(1):30-34. doi: 10.1016/S1010-7940(02)00721-2 [DOI] [PubMed] [Google Scholar]
- 29.Abdollah F, Sun M, Schmitges J, et al. Development and validation of a reference table for prediction of postoperative mortality rate in patients treated with radical cystectomy: a population-based study. Ann Surg Oncol. 2012;19(1):309-317. doi: 10.1245/s10434-011-1852-7 [DOI] [PubMed] [Google Scholar]
- 30.Schroeder RA, Marroquin CE, Bute BP, Khuri S, Henderson WG, Kuo PC. Predictive indices of morbidity and mortality after liver resection. Ann Surg. 2006;243(3):373-379. doi: 10.1097/01.sla.0000201483.95911.08 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47(11):1245-1251. doi: 10.1016/0895-4356(94)90129-5 [DOI] [PubMed] [Google Scholar]
- 32.Sundararajan V, Henderson T, Perry C, Muggivan A, Quan H, Ghali WA. New ICD-10 version of the Charlson comorbidity index predicted in-hospital mortality. J Clin Epidemiol. 2004;57(12):1288-1294. doi: 10.1016/j.jclinepi.2004.03.012 [DOI] [PubMed] [Google Scholar]
- 33.D’Hoore W, Bouckaert A, Tilquin C. Practical considerations on the use of the Charlson comorbidity index with administrative data bases. J Clin Epidemiol. 1996;49(12):1429-1433. doi: 10.1016/S0895-4356(96)00271-5 [DOI] [PubMed] [Google Scholar]
- 34.Ehlert BA, Durham CA, Parker FM, Bogey WM, Powell CS, Stoner MC. Impact of operative indication and surgical complexity on outcomes after thoracic endovascular aortic repair at National Surgical Quality Improvement Program Centers. J Vasc Surg. 2011;54(6):1629-1636. doi: 10.1016/j.jvs.2011.05.116 [DOI] [PubMed] [Google Scholar]
- 35.Bergman S, Martelli V, Monette M, et al. Identification of quality of care deficiencies in elderly surgical patients by measuring adherence to process-based quality indicators. J Am Coll Surg. 2013;217(5):858-866. doi: 10.1016/j.jamcollsurg.2013.07.387 [DOI] [PubMed] [Google Scholar]
- 36.Sakran JV, Ezzeddine H, Haut ER, et al. Prolonged operating room time in emergency general surgery is associated with venous thromboembolic complications. Am J Surg. 2019;218(5):836-841. doi: 10.1016/j.amjsurg.2019.04.022 [DOI] [PubMed] [Google Scholar]
- 37.Aboagye JK, Lau BD, Schneider EB, Streiff MB, Haut ER. Linking processes and outcomes: a key strategy to prevent and report harm from venous thromboembolism in surgical patients. JAMA Surg. 2013;148(3):299-300. doi: 10.1001/jamasurg.2013.1400 [DOI] [PubMed] [Google Scholar]
- 38.Louis SG, Sato M, Geraci T, et al. Correlation of missed doses of enoxaparin with increased incidence of deep vein thrombosis in trauma and general surgery patients. JAMA Surg. 2014;149(4):365-370. doi: 10.1001/jamasurg.2013.3963 [DOI] [PubMed] [Google Scholar]
- 39.Ko A, Harada MY, Barmparas G, et al. Association between enoxaparin dosage adjusted by anti–factor Xa trough level and clinically evident venous thromboembolism after trauma. JAMA Surg. 2016;151(11):1006-1013. doi: 10.1001/jamasurg.2016.1662 [DOI] [PubMed] [Google Scholar]
- 40.Black SR, Howard JT, Chin PC, Starr AJ. Toward a more robust prediction of pulmonary embolism in trauma patients: a risk assessment model based on 38,000 patients. J Orthop Trauma. 2016;30(4):200-207. doi: 10.1097/BOT.0000000000000484 [DOI] [PubMed] [Google Scholar]
- 41.Starr AJ, Shirley Z, Sutphin PD, et al. Significant reduction of pulmonary embolism in orthopaedic trauma patients. J Orthop Trauma. 2019;33(2):78-81. doi: 10.1097/BOT.0000000000001346 [DOI] [PubMed] [Google Scholar]
- 42.DeWane MP, Davis KA, Schuster KM, Maung AA, Becher RD. Venous thromboembolism–related readmission in emergency general surgery patients: a role for prophylaxis on discharge? J Am Coll Surg. 2018;226(6):1072-1077.e3. doi: 10.1016/j.jamcollsurg.2018.03.021 [DOI] [PubMed] [Google Scholar]
- 43.Borab ZM, Lanni MA, Tecce MG, Pannucci CJ, Fischer JP. Use of computerized clinical decision support systems to prevent venous thromboembolism in surgical patients: a systematic review and meta-analysis. JAMA Surg. 2017;152(7):638-645. doi: 10.1001/jamasurg.2017.0131 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Lau BD, Haut ER. Computerized clinical decision support systems for prevention of venous thromboembolism: why can’t my electronic health record be more like Netflix, Amazon, Google, and Apple? JAMA Surg. 2017;152(7):646-647. doi: 10.1001/jamasurg.2017.0107 [DOI] [PubMed] [Google Scholar]