Abstract
Importance
Given the current climate of outcomes-driven quality reporting, it is critical to appropriately risk stratify patients using standardized metrics.
Objective
To elucidate the risk associated with urgent surgery on complications and mortality after general surgical procedures.
Design, Setting, and Participants
This retrospective review used the American College of Surgeons National Surgery Quality Improvement Program database to capture all general surgery cases performed at 435 hospitals nationwide between January 1, 2013, and December 31, 2013. Data analysis was performed from November 11, 2015, to February 16, 2017.
Exposures
Any operations coded as both nonelective and nonemergency were designated into a novel category titled urgent.
Main Outcomes and Measures
The primary outcome was 30-day mortality; secondary outcomes included 30-day rates of complications, reoperation, and readmission in urgent cases compared with both elective and emergency cases.
Results
Of 173 643 patients undergoing general surgery (101 632 females and 72 011 males), 130 235 (75.0%) were categorized as elective, 22 592 (13.0%) as emergency, and 20 816 (12.0%) as nonelective and nonemergency. When controlling for standard American College of Surgeons National Surgery Quality Improvement Program preoperative risk factors, with elective surgery as the reference value, the 3 groups had significantly distinct odds ratios (ORs) of experiencing any complication (urgent surgery: OR, 1.38; 95% CI, 1.30-1.45; P < .001; and emergency surgery: OR, 1.65; 95% CI, 1.55-1.76; P < .001) and of mortality (urgent surgery: OR, 2.32; 95% CI, 2.00-2.68; P < .001; and emergency surgery: OR, 2.91; 95% CI, 2.48-3.41; P < .001). Surgical procedures performed urgently had a 12.3% rate of morbidity (n = 2560) and a 2.3% rate of mortality (n = 471).
Conclusions and Relevance
This study highlights the need for improved risk stratification on the basis of urgency because operations performed urgently have distinct rates of morbidity and mortality compared with procedures performed either electively or emergently. Because we tie quality outcomes to reimbursement, such a category should improve predictive models and more accurately reflect the quality and value of care provided by surgeons who do not have traditional elective practices.
This cohort study uses data from the American College of Surgeons NSQIP to examine the risk associated with complications and mortality after urgent surgery compared with elective and emergency surgery.
Key Points
Questions
Does surgery performed urgently have distinct rates of morbidity and mortality from that performed either electively or emergently?
Findings
In this nationwide cohort study of 173 643 patients who underwent general surgery, operations performed urgently had a 12.3% rate of morbidity and 2.3% rate of mortality, which were distinctly different from emergency and elective surgery.
Meaning
Because many quality metrics currently in use only distinguish emergency operations from nonemergency operations, the addition of an urgent category may improve predictive models and allow a more accurate determination of quality and value.
Introduction
Defining “quality” health care has become a major focus of the medical community, health care payers, and US Centers for Medicare & Medicaid Services (CMS). Postoperative outcome measures are increasingly being publicly reported, which has implications for hospital and physician reimbursement and reputation. In 2009, the CMS began publicly reporting 30-day rates of readmission for patients admitted to an acute care hospital for myocardial infarction, heart failure, and pneumonia. The program has since broadened to include several other outcome measures in both medical and surgical patients.
Medicare’s Physician Quality Reporting System is the largest US quality reporting program created by the CMS as part of the Affordable Care Act. In July 2007, the program began reporting measures of process quality and physiological patient outcomes and rewarding physicians who self-reported outcomes. However, in 2015 the program started imposing penalties in Medicare Part B reimbursement to physicians who fail to self-monitor and submit designated quality measures. Outcomes following many general surgical procedures, including colectomy, ventral hernia repair, appendectomy, and cholecystectomy, had previously been included, such as rates of anastomotic leak, surgical site infection, unplanned reoperation, and readmission, but are being excluded from the 2017 Medicare Value-Based Payment Modifier owing to inconsistent risk adjustment. Risk adjustment is necessary to account for the presence of comorbidities that put patients at higher risk of postoperative complications. Most risk adjustment models consider whether operations are performed emergently because the expected complication rate is substantially higher than that with elective surgery.
Patients with acute disease processes who undergo general surgery are frequently admitted for medical optimization before surgery or a trial of nonoperative conservative management. This plan commonly occurs in the management of cholecystitis, adhesive small-bowel obstruction, and acute diverticulitis. Many of these patients will undergo surgical intervention later in their hospitalization. These urgent, albeit nonemergency, operations are performed following a period of nonoperative management. Complication rates and mortality for this substantial population of patients have not been well described in the surgical literature. The purpose of this study was to elucidate the risk associated with urgent surgery on 30-day complications and mortality after general surgical procedures. We hypothesized that patients who undergo urgent surgery will have a complication and mortality profile different from those who undergo either elective or emergency surgery, which would hold important implications for quality reporting and pay-for-performance reimbursement penalties.
Methods
Data Set
Patient data for this study were obtained from the 2013 American College of Surgeons National Quality Improvement Program Participant Use File (ACS NSQIP PUF). The ACS NSQIP PUF is a nationwide, comprehensive data set designed to improve the quality of surgical care. The 2013 ACS NSQIP PUF included data for patients who underwent surgery at 435 participating hospitals between January 1, 2013, and December 31, 2013. The ACS NSQIP PUF has been designated by the University of Virginia Institutional Review Board for Health Sciences Research as a public data set; as such, this study is considered exempt from formal institutional review board review.
Patients and Variables
Emergency surgery in the ACS NSQIP is captured as part of the American Society of Anesthesiologists (ASA) Physical Status classification system, which is designated by the surgeon or anesthesiologist immediately prior to each operation. Elective surgery is defined only as procedures performed on patients who are brought to a medical facility for a scheduled (elective) surgery on the day of their operation. It specifically does not include patients who are inpatients at an acute care hospital, are transferred from an emergency department or clinic, or undergo emergency or urgent surgery. As described in the ACS NSQIP protocols, surgical clinical reviewers are specifically instructed to code “urgent” operations as nonemergency and nonelective because these cases are not true emergencies. In this article, urgent status implies that an operation was coded as nonelective and nonemergency. Cases were excluded from analysis if either variable was not known or was omitted.
Variables were selected a priori. We included all preoperative variables used by the ACS NSQIP Risk Calculator to estimate a patient’s probability of morbidity and mortality. Variables included patient age, sex, functional status, ASA class, corticosteroid use, smoking status, body mass index class, presence of ascites, sepsis, ventilator dependence, disseminated cancer, type 1 or 2 diabetes, hypertension, congestive heart failure, chronic obstructive pulmonary disease, dialysis dependence, and acute renal failure. In addition to emergency surgery, which is included in the ACS NSQIP Risk Calculator, urgent status was also included in the present multivariate analysis. The primary end point was patient mortality within 30 days of surgery. Secondary end points examined included 30-day rates of complications, hospital readmissions, and unplanned reoperations.
Statistical Analysis
Statistical analysis was performed from November 11, 2015, to February 16, 2017. Comparisons were made as a function of the urgency with which an operation was performed. Unadjusted comparisons stratified by urgency classification were performed using Pearson χ2 test for categorical variables and the Mann-Whitney test for nonnormally distributed continuous data. Multivariate logistic regression analysis was performed, with mortality within 30 days of an index operation as the primary outcome. Secondary outcomes included the occurrence of any complication recorded in the ACS NSQIP PUF, need for reoperation, or readmission within 30 days. All tests were 2-sided, with P < .05 considered significant. Statistical analyses were performed with SAS software, version 9.3 (SAS Institute Inc).
Results
A total of 173 643 general surgery cases met criteria for inclusion and were analyzed in this study. When operations were stratified according to urgency status, 130 235 (75.0%) were considered elective, 20 816 (12.0%) urgent, and 22 592 (13.0%) emergency. Demographic characteristics for all patients included are given in Table 1. Comparisons of preoperative variables and preoperative characteristics showed that patients who underwent elective, urgent, and emergency surgery differed for most variables with few exceptions. Table 2 provides the most common operations performed electively (laparoscopic cholecystectomy; 8420 cases [6.5%]), urgently (laparoscopic cholecystectomy; 4666 [22.4%]), and emergently (laparoscopic appendectomy; 10 305 [45.6%]). Length of stay before the operation was significantly longer for patients undergoing urgent operations (1 day [interquartile range, 0-3 days]) than for those undergoing elective surgery (0 days [interquartile range, 0-0 days]; P < .001) and those undergoing emergency surgery (0 days [interquartile range, 0-1 day]; P < .001).
Table 1. Univariate Analysis of Preoperative Patient Variables .
| Preoperative Variable | Surgery, No. (%) | P Value | ||
|---|---|---|---|---|
| Elective (n = 130 235) |
Urgent (n = 20 816)a |
Emergency (n = 22 592) |
||
| Age group, y | ||||
| <65 | 91 732 (70.4) | 14 281 (68.6) | 16 911 (74.9) | <.001 |
| 65-75 | 24 829 (19.1) | 3323 (16.0) | 2904 (12.9) | <.001 |
| 75-85 | 11 524 (8.9)b | 2432 (11.7) | 2085 (9.2)b | <.001 |
| ≥85 | 2150 (1.7) | 780 (3.8) | 692 (3.1) | <.001 |
| Female sex | 78 522 (60.3) | 11 371 (54.6) | 11 739 (52.0) | <.001 |
| BMI classification | ||||
| Underweight (<18.5) | 2834 (2.2) | 1407 (6.8) | 2335 (10.3) | <.001 |
| Normal weight (18.5-24.9) | 31 325 (24.1) | 5824 (28.0) | 6673 (29.5) | <.001 |
| Overweight (25.0-29.9) | 37 537 (28.8)b | 5930 (28.5)b | 6451 (28.6)b | .48 |
| Obese | ||||
| I (30.0-34.9) | 25 069 (19.3) | 3859 (18.5) | 3735 (16.5) | <.001 |
| II (35.0-39.9) | 14 422 (11.1) | 2025 (9.7) | 1793 (7.9) | <.001 |
| III (≥40.0) | 19 048 (14.6) | 1771 (8.5) | 1605 (7.1) | <.001 |
| Independent functional status | 128 475 (98.7) | 19 488 (93.6) | 21 546 (95.4) | <.001 |
| ASA classification | ||||
| Class 1 | 9461 (7.3) | 1681 (8.1) | 4329 (19.2) | <.001 |
| Class 2 | 63 870 (49.0) | 8306 (39.9)b | 9075 (40.2)b | <.001 |
| Class 3 | 53 475 (41.1) | 8887 (42.7) | 6314 (28.0) | <.001 |
| Class 4 | 3429 (2.6) | 1942 (9.3) | 2874 (12.7) | <.001 |
| Class 5 | 0 | 0 | 0 | NA |
| Chronic corticosteroid use | 5338 (4.1) | 1365 (6.6) | 1064 (4.7) | <.001 |
| Ascites | 314 (0.2) | 315 (1.5) | 462 (2.0) | <.001 |
| Presence of sepsis | ||||
| SIRS | 802 (0.6) | 2481 (11.9) | 3848 (17.0) | <.001 |
| Sepsis | 290 (0.2) | 1587 (7.6) | 3449 (15.3) | <.001 |
| Septic shock | 46 (0.04) | 163 (0.8) | 962 (4.3) | <.001 |
| Ventilator dependence | 35 (0.03) | 174 (0.8) | 473 (2.1) | <.001 |
| Disseminated cancer | 3843 (3.0) | 982 (4.7) | 589 (2.6) | <.001 |
| Diabetes | ||||
| Type 1 | 6979 (5.4)b | 1644 (7.9) | 1198 (5.3)b | <.001 |
| Type 2 | 12 419 (9.5) | 1727 (8.3) | 1390 (6.2) | <.001 |
| Hypertension | 56 683 (43.5) | 8702 (41.8) | 7498 (33.2) | <.001 |
| Cardiovascular event | 0 | 0 | 0 | NA |
| Congestive heart failure | 341 (0.3) | 346 (1.7) | 267 (1.2) | <.001 |
| Dyspnea | 8664 (6.7)b | 1335 (6.4)b | 1189 (5.3) | <.001 |
| Current smoker | 20 901 (16.1) | 4349 (20.9)b | 4627 (20.5)b | <.001 |
| COPD | 4284 (3.3) | 1169 (5.6) | 1055 (4.7) | <.001 |
| Hemodialysis dependent | 1344 (1.0) | 512 (2.5) | 353 (1.6) | <.001 |
| Acute renal failure | 158 (0.1) | 142 (0.7) | 288 (1.3) | <.001 |
Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); COPD, chronic obstructive pulmonary disease; NA, not applicable; SIRS, systemic inflammatory response syndrome.
Urgent surgery is defined in the Patients and Variables subsection of the Methods section.
Denotes no significant difference among identified groups on multiple comparison (P > .05).
Table 2. Most Common Operations for Each Category of Urgency.
| Surgical Procedure | No. (%) |
|---|---|
| Elective surgery (n = 130 235) | |
| Laparoscopic cholecystectomy | 8420 (6.5) |
| Inguinal hernia repair | 6291 (4.8) |
| Sleeve gastrectomy | 6232 (4.8) |
| Roux-en-Y gastric bypass | 5453 (4.2) |
| Partial mastectomy | 4860 (3.7) |
| Urgent surgery (n = 20 816)a | |
| Laparoscopic cholecystectomy | 4666 (22.4) |
| Laparoscopic appendectomy | 3675 (17.7) |
| Partial colectomy | 2829 (13.6) |
| Small-bowel resection | 679 (3.3) |
| Lysis of adhesions | 571 (2.7) |
| Emergency surgery (n = 22 592) | |
| Laparoscopic appendectomy | 10 305 (45.6) |
| Laparoscopic cholecystectomy | 1388 (6.1) |
| Small-bowel resection | 1093 (4.8) |
| Partial colectomy | 1454 (6.4) |
| Lysis of adhesions | 555 (2.5) |
Urgent surgery is defined in the Patients and Variables subsection of the Methods section.
Unadjusted 30-day patient outcomes examined included the occurrence of any complication recorded in the ACS NSQIP PUF, mortality, unplanned reoperation, and readmission (Table 3). Surgical procedures performed urgently had a 12.3% rate of morbidity (n = 2560) and a 2.3% rate of mortality (n = 471) compared with a 13.8% rate of morbidity (n = 3114) and 3.7% rate of mortality for emergency surgery and a 6.7% rate of morbidity (n = 8718) and 0.4% rate of mortality (n = 516) for elective surgery. Multivariate logistic regression analysis was performed to determine the independent contribution of urgency status on 30-day mortality and postoperative morbidity (Table 4). When we controlled for standard ACS NSQIP preoperative risk factors, with elective surgery as the reference value, the 3 groups had significantly distinct odds ratios (ORs) of experiencing any complication (urgent surgery: OR, 1.38; 95% CI, 1.30-1.45; P < .001; and emergency surgery: OR, 1.65; 95% CI, 1.55-1.76; P < .001) and of mortality (urgent surgery: OR, 2.32; 95% CI, 2.00-2.68; P < .001; and emergency surgery: OR, 2.91; 95% CI, 2.48-3.41; P < .001).
Table 3. Unadjusted Incidence of 30-Day Postoperative Outcomes Stratified by Urgency Status.
| 30-d Outcome | Urgency Status, No. (%) | P Value | ||
|---|---|---|---|---|
| Elective (n = 130 235) |
Urgent (n = 20 816)a |
Emergency (n = 22 592) |
||
| Mortality | 516 (0.4) | 471 (2.3) | 846 (3.7) | <.001 |
| Morbidity | 8718 (6.7) | 2560 (12.3) | 3114 (13.8) | <.001 |
| Hospital readmission | 4818 (3.7) | 1140 (5.5) | 1435 (6.4) | <.001 |
| Unplanned reoperation | 1146 (0.9) | 322 (1.6) | 314 (1.4) | <.001 |
Urgent surgery is defined in the Patients and Variables subsection of the Methods section.
Table 4. Risk-Adjusted Odds Ratios for Factors Associated With Postoperative Mortality and Morbidity.
| Variable | 30-d Mortalitya | Morbidityb | ||||
|---|---|---|---|---|---|---|
| Wald χ2 | Odds Ratio (95% CI) | P Value | Wald χ2 | Odds Ratio (95% CI) | P Value | |
| Urgency | 189.00 | NA | <.001 | 271.35 | NA | <.001 |
| Elective | NA | 1 [Reference] | NA | NA | 1 [Reference] | NA |
| Urgentc | 124.57 | 2.32 (2.00-2.68) | <.001 | 128.26 | 1.38 (1.30-1.45) | <.001 |
| Emergency | 172.72 | 2.91 (2.48-3.41) | <.001 | 243.26 | 1.65 (1.55-1.76) | <.001 |
| Age group, y | 190.56 | NA | <.001 | 116.73 | NA | <.001 |
| <65 | NA | 1 [Reference] | NA | NA | 1 [Reference] | NA |
| 65-75 | 72.08 | 1.76 (1.54-2.01) | <.001 | 74.70 | 1.23 (1.17-1.30) | <.001 |
| 75-85 | 135.51 | 2.31 (2.01-2.66) | <.001 | 68.98 | 1.29 (1.21-1.37) | <.001 |
| ≥85 | 122.82 | 3.06 (2.51-3.73) | <.001 | 28.68 | 1.34 (1.20-1.49) | <.001 |
| Female sex | 8.09 | 0.86 (0.78-0.96) | .004 | 25.76 | 0.91 (0.87-0.92) | <.001 |
| BMI classification | 74.35 | NA | <.001 | 38.34 | NA | <.001 |
| Underweight (<18.5) | 25.37 | 1.59 (1.33-1.90) | <.001 | 3.08 | 1.08 (0.99-1.19) | .08 |
| Normal weight (18.5-24.9) | NA | 1 [Reference] | NA | NA | 1 [Reference] | NA |
| Overweight (25.0-29.9) | 5.22 | 0.86 (0.75-0.98) | .02 | 0.22 | 0.99 (0.94-1.04) | .64 |
| Obese | ||||||
| I (30.0-34.9) | 10.78 | 0.77 (0.65-0.90) | .001 | 4.03 | 1.06 (1.00-1.12) | .04 |
| II (35.0-39.9) | 13.27 | 0.67 (0.55-0.83) | <.001 | 21.95 | 1.176 (1.10-1.26) | <.001 |
| III (≥40.0) | 10.88 | 0.71 (0.58-0.87) | .001 | 0.78 | 0.97 (0.91-1.04) | .38 |
| Independent functional status | 56.75 | 0.57 (0.50-0.66) | <.001 | 112.79 | 0.63 (0.58-0.69) | <.001 |
| ASA classification | 595.24 | NA | <.001 | 1116.32 | NA | <.001 |
| 1 | NA | 1 [Reference] | NA | NA | 1 [Reference] | NA |
| 2 | 9.22 | 5.96 (1.88-18.85) | .002 | 161.95 | 2.34 (2.05-2.67) | <.001 |
| 3 | 35.36 | 31.85 (10.18-99.66) | <.001 | 423.17 | 4.04 (3.53-4.61) | <.001 |
| 4 | 58.80 | 87.79 (27.97-275.52) | <.001 | 632.75 | 6.44 (5.57-7.45) | <.001 |
| Wound classification | 21.63 | NA | <.001 | 687.18 | NA | <.001 |
| Clean | NA | 1 [Reference] | NA | NA | 1 [Reference] | NA |
| Clean-contaminated | 14.53 | 1.37 (1.17-1.62) | .001 | 533.02 | 1.89 (1.79-1.99) | <.001 |
| Contaminated | 17.15 | 1.50 (1.24-1.82) | <.001 | 465.30 | 2.14 (2.00-2.30) | <.001 |
| Dirty or infected | 16.98 | 1.51 (1.24-1.83) | <.001 | 426.19 | 2.29 (2.12-2.48) | <.001 |
| Diabetes | 14.24 | NA | <.001 | 21.49 | NA | <.001 |
| None | NA | 1 [Reference] | NA | NA | 1 [Reference] | NA |
| Type 2 | 8.18 | 0.78 (0.66-0.93) | .004 | 0.32 | 1.02 (0.96-1.08) | .57 |
| Type 1 | 3.61 | 1.16 (0.99-1.35) | .06 | 21.42 | 1.17 (1.10-1.25) | <.001 |
| Sepsis | 208.87 | NA | <.001 | 196.70 | NA | <.001 |
| None | NA | 1 [Reference] | NA | NA | 1 [Reference] | NA |
| SIRS | 60.95 | 2.01 (1.68-2.39) | <.001 | 73.18 | 1.43 (1.31-1.55) | <.001 |
| Sepsis | 57.83 | 2.00 (1.67-2.39) | <.001 | 55.91 | 1.41 (1.29-1.54) | <.001 |
| Septic shock | 190.25 | 4.45 (3.60-5.50) | <.001 | 134.02 | 2.44 (2.10-2.84) | <.001 |
| Chronic corticosteroid use | 12.29 | 1.33 (1.13-1.55) | <.001 | 114.05 | 1.45 (1.36-1.56) | <.001 |
| Ascites | 82.36 | 2.71 (2.19-3.37) | <.001 | 35.28 | 1.59 (1.37-1.86) | <.001 |
| Ventilator dependence | 10.38 | 1.45 (1.16-1.81) | .001 | 3.81 | 0.83 (0.68-1.00) | .05 |
| Disseminated cancer | 257.59 | 3.32 (2.870-3.85) | <.001 | 41.43 | 1.28 (1.19-1.38) | <.001 |
| Hypertension | 24.41 | 1.35 (1.20-1.52) | <.001 | 23.08 | 1.11 (1.06-1.16) | <.001 |
| Congestive heart failure | 8.51 | 1.43 (1.12-1.81) | .003 | 4.89 | 1.20 (1.02-1.42) | .03 |
| Dyspnea | 14.88 | 1.33 (1.15-1.54) | <.001 | 13.58 | 1.13 (1.06-1.21) | <.001 |
| Current smoker | 1.46 | 0.92 (0.80-1.05) | .23 | 112.34 | 1.29 (1.23-1.35) | <.001 |
| COPD | 19.89 | 1.41 (1.21-1.64) | <.001 | 52.48 | 1.32 (1.23-1.42) | <.001 |
| Hemodialysis dependent | 17.27 | 1.57 (1.27-1.94) | <.001 | 1.23 | 0.92 (0.80-1.06) | .27 |
| Acute renal failure | 10.02 | 1.55 (1.18-2.03) | .002 | 47.42 | 2.02 (1.65-2.47) | <.001 |
| Relative value unit | 100.51 | 1.02 (1.02-1.03) | <.001 | 3482.98 | 1.05 (1.05-1.05) | <.001 |
Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); COPD, chronic obstructive pulmonary disease; NA, not applicable; SIRS, systemic inflammatory response syndrome.
For 30-day mortality, the C statistic is 0.93; Hosmer-Lemeshow test, 0.002; and Brier score, 0.01.
For morbidity, the C statistic is 0.79; Hosmer-Lemeshow test, <0.0001; and Brier score, 0.07.
Urgent surgery is defined in the Patients and Variables subsection of the Methods section.
Discussion
This study found that general surgical procedures performed urgently had a 12.3% rate of morbidity and a 2.3% rate of mortality, which are rates distinctly different from both emergency (morbidity, 13.8%; mortality, 3.7%) and elective (morbidity, 6.7%; mortality, 0.4%) surgery. To our knowledge, no prior study has elucidated rates of morbidity and mortality associated with urgent general surgery. This finding reveals important insight that the urgency of surgery is not binary, but rather that there are 3 categories, each with a distinct morbidity and mortality profile, and that surgical urgency can help predict postoperative complications independent of a patient’s preoperative comorbidities. Ultimately, urgency is a crucial consideration when performing patient risk stratification, reporting surgical outcomes, and establishing benchmarks for quality and performance under the Affordable Care Act.
The goal of pay-for-performance models is to account for quality of care and clinical outcomes in the compensation of physicians, rather than to solely reimburse for services provided. The findings in this study hold important implications for pay-for performance reimbursement, value-based programs, and surgical outcome reporting. Currently, risk stratification for the ACS NSQIP and the Physician Quality Reporting System considers preoperative comorbidities, functional status, ASA classification, and whether surgery was performed emergently. These programs do not account for urgent, nonelective operations that are not determined to be emergencies by the surgeon or anesthesiologist at the time of surgery. In this study, we found that these urgent operations are associated with substantially higher rates of complications and mortality when compared with elective surgery. Surgeons who commonly operate on an urgent basis, including many acute care and emergency general surgeons, are at risk of being penalized by the CMS in Medicare Part B value-based reimbursement. These surgeons may even unfairly be labeled as poor performers by current outcome reporting guidelines.
Several prior investigators have demonstrated increased rates of complications and mortality after emergency surgery. Mortality following emergency gastrointestinal surgery has been found to be as much as 5 times greater than for elective gastrointestinal surgery. This difference has previously been attributed to patient comorbidities and physiological derangements associated with acute disease processes. However, Havens et al recently demonstrated that increased morbidity and mortality following emergency general surgery is independent of preoperative comorbidities and physiological status. Our multivariate analysis demonstrated similar results: complication rates were higher in patients who underwent urgent surgery compared with those who underwent elective surgery, independent of other preoperative risk factors. Consistent with prior studies, we found that patients with preoperative comorbidities, obesity, rising ASA classification, age greater than 65 years, and contaminated surgical wounds are at an increased risk of morbidity and mortality. Current smokers were not found to be at greater risk of mortality following surgery, but smoking was predictive of postoperative complications.
Patients undergoing urgent surgery had the highest rates in our cohort of congestive heart failure, chronic obstructive pulmonary disease, diabetes, preoperative hemodialysis, chronic corticosteroid use, and disseminated cancer. Coupled with the fact that patients undergoing urgent operations have a longer preoperative length of stay, this finding suggests that patients with serious preoperative comorbidities who present with acute surgical issues may be having operative care delayed, presumably for medical optimization or following a failed trial of nonoperative management. A delay in surgical intervention could be contributing to the observed increase in morbidity and mortality. For instance, cholecystitis initially treated with antibiotics and medical optimization can progress, leading to perforation, or a small-bowel obstruction that is initially managed nonoperatively may develop ischemia. Delay in surgery for both of these conditions may result in worse pathologic conditions and the need for a more extensive operation. The prevalence of systemic inflammatory response syndrome, sepsis, and septic shock were greatest in patients undergoing emergency surgery, which would tip the risk to benefit ratio toward more expeditious surgery.
Limitations
This study has some notable limitations. First, these analyses are limited by the retrospective study design. Although the ACS NSQIP is a large national data set, its data are limited to participating hospitals, which are primarily academic medical centers. Next, surgical urgency was determined based on variables recorded in the ACS NSQIP, which is subject to anesthesiologist and surgeon discretion at the time of an operation, as well as to documentation error. Cardiac risk factors, including the presence of coronary artery disease, and history of cardiac events were not recorded in the 2013 ACS NSQIP PUF. We sought to identify trends among a large population of patients, so all general surgical procedures were included for this study; therefore, there is a lack of granularity that may make generalization of these findings to specific procedures difficult.
Conclusions
At a time when reimbursement is contingent on value-based outcomes reporting and performance, it is imperative to ensure that appropriate risk adjustment is performed. We have identified operative urgency as a key consideration for patient risk stratification. If this issue is not recognized, quality outcome reporting and value-based reimbursement will continue to incentivize operating on an elective basis and will make surgeons more reluctant to operate on patients who urgently require care.
References
- 1.Tsai TC, Orav EJ, Jha AK. Patient satisfaction and quality of surgical care in US hospitals. Ann Surg. 2015;261(1):2-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Tsai TC, Joynt KE, Orav EJ, Gawande AA, Jha AK. Variation in surgical-readmission rates and quality of hospital care. N Engl J Med. 2013;369(12):1134-1142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Dharmarajan K, Hsieh AF, Lin Z, et al. . Hospital readmission performance and patterns of readmission: retrospective cohort study of Medicare admissions. BMJ. 2013;347:f6571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Centers for Medicare & Medicaid Services Eligible professionals. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/PQRS/downloads/EligibleProfessionals.pdf. Accessed January 20, 2016.
- 5.Centers for Medicare & Medicaid Services Medicare’s Physician Quality Reporting System (PQRS): quality measurement and beneficiary attribution. https://www.cms.gov/mmrr/Downloads/MMRR2014_004_02_a04.pdf. Accessed January 2, 2016. [DOI] [PMC free article] [PubMed]
- 6.Wiler JL, Granovsky M, Cantrill SV, Newell R, Venkatesh AK, Schuur JD. Physician Quality Reporting System program updates and the impact on emergency medicine practice. West J Emerg Med. 2016;17(2):229-237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Centers for Medicare & Medicaid Services Detailed methodology for the 2017 Value Modifier and the 2015 Quality and Resource Use Report. https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeedbackProgram/Downloads/Detailed-Methodology-for-the-2017-Value-Modifier-and-2015-Quality-and-Resource-Use-Report-.pdf. Accessed February 6, 2017.
- 8.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] [PMC free article] [PubMed] [Google Scholar]
- 9.Sheetz KH, Waits SA, Krell RW, Campbell DA Jr, Englesbe MJ, Ghaferi AA. Improving mortality following emergent surgery in older patients requires focus on complication rescue. Ann Surg. 2013;258(4):614-617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ingraham AM, Cohen ME, Bilimoria KY, et al. . Comparison of 30-day outcomes after emergency general surgery procedures: potential for targeted improvement. Surgery. 2010;148(2):217-238. [DOI] [PubMed] [Google Scholar]
- 11.Bilimoria KY, Cohen ME, Merkow RP, et al. . Comparison of outlier identification methods in hospital surgical quality improvement programs. J Gastrointest Surg. 2010;14(10):1600-1607. [DOI] [PubMed] [Google Scholar]
- 12.Ingraham AM, Cohen ME, Bilimoria KY, et al. . Comparison of hospital performance in nonemergency versus emergency colorectal operations at 142 hospitals. J Am Coll Surg. 2010;210(2):155-165. [DOI] [PubMed] [Google Scholar]
- 13.American College of Surgeons National Surgical Quality Improvement Program. User guide for the 2013 ACS NSQIP participant use data file (PUF). https://www.facs.org/~/media/files/quality programs/nsqip/2013_acs_nsqip_puf_user_guide.ashx. Published November 2014. Accessed April 6, 2017.
- 14.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] [PubMed] [Google Scholar]
- 15.Kwan TL, Lai F, Lam CM, et al. . Population-based information on emergency colorectal surgery and evaluation on effect of operative volume on mortality. World J Surg. 2008;32(9):2077-2082. [DOI] [PubMed] [Google Scholar]
- 16.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] [PubMed] [Google Scholar]
- 17.Sørensen LT, Malaki A, Wille-Jørgensen P, et al. . Risk factors for mortality and postoperative complications after gastrointestinal surgery. J Gastrointest Surg. 2007;11(7):903-910. [DOI] [PubMed] [Google Scholar]