Abstract
Aims
This study aimed to investigate the risk of postoperative complications in COVID-19-positive patients undergoing common orthopaedic procedures.
Methods
Using the National Surgical Quality Improvement Programme (NSQIP) database, patients who underwent common orthopaedic surgery procedures from 1 January to 31 December 2021 were extracted. Patient preoperative COVID-19 status, demographics, comorbidities, type of surgery, and postoperative complications were analyzed. Propensity score matching was conducted between COVID-19-positive and -negative patients. Multivariable regression was then performed to identify both patient and provider risk factors independently associated with the occurrence of 30-day postoperative adverse events.
Results
Of 194,121 included patients, 740 (0.38%) were identified to be COVID-19-positive. Comparison of comorbidities demonstrated that COVID-19-positive patients had higher rates of diabetes, heart failure, and pulmonary disease. After propensity matching and controlling for all preoperative variables, multivariable analysis found that COVID-19-positive patients were at increased risk of several postoperative complications, including: any adverse event, major adverse event, minor adverse event, death, venous thromboembolism, and pneumonia. COVID-19-positive patients undergoing hip/knee arthroplasty and trauma surgery were at increased risk of 30-day adverse events.
Conclusion
COVID-19-positive patients undergoing orthopaedic surgery had increased odds of many 30-day postoperative complications, with hip/knee arthroplasty and trauma surgery being the most high-risk procedures. These data reinforce prior literature demonstrating increased risk of venous thromboembolic events in the acute postoperative period. Clinicians caring for patients undergoing orthopaedic procedures should be mindful of these increased risks, and attempt to improve patient care during the ongoing global pandemic.
Cite this article: Bone Jt Open 2023;4(9):704–712.
Take home message
Preoperative COVID-19 positivity is a relatively rare occurrence among patients undergoing orthopaedic surgery in the NSQIP dataset.
COVID-19-positive patients undergoing orthopaedic surgery have higher rates of diabetes, heart failure, and pulmonary disease, as well as higher rates of several postoperative adverse events, including death, venous thromboembolism, and pneumonia.
COVID-19-positive patients undergoing hip/knee arthroplasty and trauma surgery were at increased risk of adverse events.
Introduction
COVID-19 continues to pose additional burdens on already strained healthcare systems worldwide.1,2 There is growing evidence demonstrating that COVID-19 is associated with an increased risk for morbidity and mortality in hospitalized patients, including respiratory failure, pneumonia, cardiac events, venous thromboembolism, and longer lengths of stay.3-11 Specifically in patients undergoing surgical and interventional procedures, COVID-19 positivity increases morbidity and mortality.3,4,12 The correlation between preoperative COVID-19 positivity and increased rates of perioperative complications has been demonstrated in various surgical specialties, including trauma, cardiothoracic, hepatobiliary, obstetric, and vascular.4,13 In the orthopaedic literature, some studies have investigated the relationship between COVID-19 positivity and perioperative complications in common orthopaedic injuries such as hip and ankle fractures.13-17 However, large-scale studies based on robust national data registries, investigating the relationship of COVID-19 positivity and complications in different orthopaedic procedures, are lacking.
Furthermore, while various patient comorbidities such as diabetes, cardiovascular disease (CVD), chronic obstructive pulmonary disease (COPD), and obesity have been associated with increased risk of suffering from morbidities and mortality associated with COVID-19-positive patients,18,19 the relationship of these comorbidities, COVID-19 positivity, and incidence of complications in the immediate postoperative period following orthopaedic procedures is not well studied.
An enhanced understanding of the relationship between COVID-19 positivity and perioperative complications across a wide spectrum of orthopaedic procedures allows for enhanced risk stratification, patient counselling and optimization, resource allocation, and care planning. This study aimed to understand the relationship between COVID-19-positivity in patients undergoing orthopaedic surgery and the risk of adverse outcomes using the National Surgical Quality Improvement Programme (NSQIP) 2021 database.
Methods
Data source
A retrospective cohort analysis was performed based on the NSQIP database on patients undergoing orthopaedic surgery. The date range evaluated was 1 January 2021 to 31 December 2021. The NSQIP database is a deidentified collection of patient records from over 680 participating hospitals across North America that reports on perioperative complications within a 30-day period recorded directly by the programme’s trained nurses.20,21 Hospital research ethics board approval was obtained for this study.
Study population
NSQIP data on surgical procedures are reported as Current Procedural Terminology (CPT) codes.22 Using CPT codes, patients who underwent the most common orthopaedic surgery procedures in the NSQIP database as previously reported by Molina et al23 were identified and extracted (Table I). Patient preoperative COVID-19 status was specified by a discrete preoperative COVID-19 diagnosis variable in the NSQIP database. Per database documentation, COVID-19 positivity was defined as an active diagnosis of COVID-19 within 14 days prior to the primary procedure which is confirmed by laboratory testing.24 Laboratory testing could be performed before or after the primary procedure to confirm the preoperative diagnosis.
Table I.
Incidence of preoperative COVID-19 positivity among patients undergoing the most common orthopaedic surgery procedures in the National Surgical Quality Improvement Programme database.
| CPT code | Surgery type | Procedure | Patients, n | COVID-19 patients, n (% of COVID-19 cases) |
|---|---|---|---|---|
| 194,121 (100) | 740 (0.38) | |||
| 27447 | Arthroplasty | TKA | 58,277 (30.02) | 73 (9.86) |
| 27130 | Arthroplasty | THA | 40,978 (21.11) | 78 (10.54) |
| 27245 | Trauma | Treatment of intertrochanteric, peritrochanteric, or subtrochanteric femoral fracture, with intramedullary implant | 12,670 (6.53) | 224 (30.27) |
| 63047 | Spine | Laminectomy, single vertebral segment, lumbar | 9,722 (5.01) | 20 (2.70) |
| 63030 | Spine | Laminotomy (hemilaminectomy), with decompression of nerve root, lumbar | 8,554 (4.41) | 19 (2.57) |
| 27236 | Trauma | Open treatment of femoral fracture, proximal end, neck, internal fixation or prosthetic arthroplasty | 8,306 (4.28) | 134 (18.11) |
| 29881 | Sports | Arthroscopy knee (chondroplasty), with meniscectomy (medial or lateral) | 7,125 (3.67) | 9 (1.22) |
| 29827 | Shoulder | Shoulder arthroscopy, rotator cuff repair | 6,879 (3.54) | 7 (0.95) |
| 23472 | Shoulder | Total shoulder arthroplasty | 5,804 (2.99) | 12 (1.62) |
| 22612 | Spine | Arthrodesis lumbar | 4,389 (2.26) | 7 (0.95) |
| 29888 | Sports | Arthroscopically aided ACL | 3,577 (1.84) | 2 (0.27) |
| 27487 | Arthroplasty | Revision of TKA, both femoral and tibial components | 3,097 (1.60) | 5 (0.68) |
| 27814 | Trauma | Open treatment of bimalleolar ankle fracture, including internal fixation | 2,604 (1.34) | 23 (3.11) |
| 27125 | Arthroplasty | Hemiarthroplasty, hip, partial (eg, femoral stem prosthesis, bipolar arthroplasty) | 2,576 (1.33) | 46 (6.22) |
| 22630 | Spine | Arthrodesis, posterior interbody technique, including laminectomy and/or discectomy to prepare interspace, single interspace; lumbar | 2,241 (1.15) | 4 (0.54) |
| 27134 | Arthroplasty | Revision of THA | 2,205 (1.14) | 14 (1.89) |
| 29880 | Sports | Arthroscopy, knee (chondroplasty), with meniscectomy (medial or lateral) | 2,187 (1.13) | 4 (0.54) |
| 27446 | Sports | Arthroplasty, knee, condyle and plateau; medial or lateral compartment | 1,793 (0.92) | 1 (0.14) |
| 27792 | Trauma | ORIF of distal fibular fracture (lateral malleolus), including internal fixation | 1,615 (0.83) | 16 (2.16) |
| 27486 | Arthroplasty | Revision of TKA, with or without allograft; 1 component | 1,548 (0.80) | 5 (0.68) |
| 27244 | Trauma | ORIF intertrochanteric femur fracture with plate and screw | 1,471 (0.76) | 26 (3.51) |
| 29806 | Shoulder | Arthroscopy, shoulder, surgical; capsulorrhaphy | 1,454 (0.75) | 4 (0.54) |
| 29877 | Sports | Arthroscopy knee, débridement/shaving of articular cartilage (chondroplasty) | 1,074 (0.55) | 5 (0.68) |
| 29807 | Shoulder | Arthroscopy, shoulder, surgical; repair of slap lesion | 781 (0.40) | 0 (0) |
| 23412 | Shoulder | Repair of ruptured musculotendinous cuff (e.g. rotator cuff), open; chronic | 700 (0.36) | 0 (0) |
| 29826 | Shoulder | Arthroscopy, shoulder, surgical; decompression of subacromial space with partial acromioplasty, with coracoacromial ligament (i.e. arch) release | 675 (0.35) | 1 (0.14) |
| 22554 | Spine | Arthrodesis, anterior interbody technique, including minimal discectomy to prepare interspace, cervical below C2 | 537 (0.28) | 0 (0) |
| 23470 | Shoulder | Arthroplasty, glenohumeral joint; hemiarthroplasty | 273 (0.14) | 1 (0.14) |
| 63075 | Spine | Discectomy, anterior, cervical, single interspace | 269 (0.14) | 0 (0) |
-
ACL, anterior cruciate ligament; CPT, Current Procedural Terminology; ORIF, open reduction and internal fixation; THA, total hip arthroplasty; TKA, total knee arthroplasty.
Patient data
Baseline demographic data on patient age, sex, BMI, race, American Society of Anesthesiologists (ASA) grade,25 and type of surgery were extracted. Using International Classification of Diseases (ICD)-10 codes, patient comorbidities were also extracted. Investigated comorbidities included diabetes mellitus (DM), congestive heart failure (CHF), hypertension (HTN), severe COPD, dependence on ventilators, chronic steroid use, and history of smoking.
Additionally, 30-day postoperative adverse events were identified based on ICD-10 codes. Perioperative outcomes were classified into major, minor, and any adverse events. Major adverse events included: deep surgical site infection, mechanical ventilation > 48 hours, unplanned intubation, acute renal failure, sepsis, venous thromboembolism, stroke, cardiac arrest/myocardial infarction (MI), return to operating theatre, and death. Minor adverse events included: superficial surgical site infection, wound dehiscence, pneumonia, urinary tract infection, and Clostridium difficile infections. Any adverse event was tabulated when a patient experienced one of more of the predefined major or minor adverse events. In addition, data on other perioperative outcomes, such as blood transfusion, length of stay (≥ 3 days), and discharge to non-home facilities, were extracted.
Statistical analysis
NSQIP data files from 2021 were analyzed in a master file using SPSS version 28 (2021; IBM, USA). Patient demographics, comorbidities, and adverse events data were compared between COVID-19-positive and COVID-19-negative patients undergoing all-type orthopaedic surgery. Chi-squared tests were used to compare demographics and postoperative adverse events. Independent-samples t-tests were used to compare continuous variables (age and BMI). COVID-19-positive patients were then compared to a 1:10 matched COVID-19-negative cohort, and propensity score matching was conducted based on age, sex, and all examined comorbidities (listed above). Matching was performed using R package “Matchit” package.26 Multivariable regression was then performed to identify association of COVID-19 positivity with 30-day adverse events. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated for each specific adverse event, adverse event group, and surgery type (Table I for procedure type classifications). Using univariate analysis, complications and adverse events occurring within the 30-day postoperative period were compared. These are expressed as percentages of the cohorts, summarized in Table II.
Table II.
Comparison of postoperative adverse events between COVID-19-positive (+) and the 1:10 propensity score matched COVID-19-negative (-) cohort. Data are expressed as number (%).
| Variable | COVID-19 (+) cohort | COVID-19 (-) cohort | p-value | 1:10 propensity matched COVID-19 (-) cohort | p-value |
|---|---|---|---|---|---|
| Patients, n | 740 (9.09) | 193,381 (99.62) | 7,400 (90.91) | ||
| Any adverse event | 155 (20.95) | 11,871 (6.14) | < 0.001 | 771 (10.42) | < 0.001 |
| Major adverse event | 112 (15.14) | 7,611 (3.94) | < 0.001 | 537 (7.26) | < 0.001 |
| Deep SSI | 4 (0.54) | 330 (0.17) | 0.015 | 24 (0.32) | 0.338 |
| Mechanical ventilation > 48 hrs | 11 (1.49) | 219 (0.11) | < 0.001 | 23 (0.31) | < 0.001 |
| Unplanned reintubation | 19 (2.57) | 448 (0.23) | < 0.001 | 45 (0.61) | < 0.001 |
| Acute renal failure | 4 (0.54) | 118 (0.06) | < 0.001 | 9 (0.12) | 0.007 |
| Sepsis/septic shock | 32 (4.32) | 1,078 (0.56) | < 0.001 | 71 (0.96) | < 0.001 |
| Venous thromboembolism | 18 (2.43) | 1,692 (0.87) | < 0.001 | 95 (1.28) | 0.011 |
| Stroke | 6 (0.81) | 291 (0.15) | < 0.001 | 27 (0.36) | 0.069 |
| Cardiac arrest/MI | 19 (2.57) | 1,108 (0.57) | < 0.001 | 96 (1.30) | 0.005 |
| Return to operating theatre | 23 (3.11) | 2,964 (1.53) | < 0.001 | 133 (1.80) | 0.013 |
| Death | 49 (6.62) | 1,611 (0.83) | < 0.001 | 193 (2.61) | < 0.001 |
| Minor adverse event | 84 (11.35) | 5,876 (3.04) | < 0.001 | 339 (4.58) | < 0.001 |
| Superficial SSI | 9 (1.22) | 2,040 (1.05) | 0.668 | 85 (1.15) | 0.870 |
| Wound dehiscence | 1 (0.14) | 454 (0.23) | 0.576 | 12 (0.16) | 0.861 |
| Pneumonia | 50 (6.76) | 1,362 (0.70) | < 0.001 | 103 (1.39) | < 0.001 |
| Urinary tract infection | 27 (3.65) | 2,060 (1.07) | < 0.001 | 144 (1.95) | 0.002 |
| C. difficile infection | 1 (0.14) | 232 (0.12) | 0.905 | 15 (0.20) | 0.692 |
| Readmission | 62 (8.38) | 6,658 (3.44) | < 0.001 | 396 (5.35) | < 0.001 |
| Blood transfusion | 98 (13.24) | 7,841 (4.05) | < 0.001 | 695 (9.39) | < 0.001 |
| Length of stay > 3 days | 498 (67.30) | 48,076 (24.86) | < 0.001 | 2,996 (40.49) | < 0.001 |
| Discharge to non-home facility | 336 (45.41) | 26,486 (13.70) | < 0.001 | 2,156 (29.14) | < 0.001 |
-
MI, myocardial infarction; SSI, surgical site infection.
Results
Patient population
Of 194,121 included patients who underwent orthopaedic surgery in 2021, 740 (0.38%) were identified as COVID-19-positive. Table I demonstrates the incidence of preoperative COVID-19 positivity among patients undergoing the most common orthopaedic surgery procedures in the NSQIP database. Trauma patients undergoing hip hemiarthroplasty had the highest rate of COVID-19 positivity at 1.79% (46/2,576). Among the included patients undergoing orthopaedic surgery, 108,902 (56.1%) underwent hip/knee arthroplasty surgery, 27,089 (14.0%) trauma surgery, 25,762 (13.3%) spine surgery, 15,777 (8.1%) sports surgery, and 16,591 (8.6%) shoulder surgery.
COVID-19 patient demographic data and comorbidities
Compared to COVID-19-negative patients, COVID-19-positive patients were older (mean age 71.1 years (standard deviation (SD)) 16.5 vs 64.7 years (SD 14.8); p < 0.001), had lower mean BMI (26.5 kg/m2 (SD 9.0) vs 29.9 (SD 8.6); p < 0.001), were from different race groups, more commonly white, (p < 0.001) and had higher ASA grades (p < 0.001) (Table III). Incidence of patient comorbidities are demonstrated in Table IV. Comparison of comorbidities demonstrated that COVID-19-positive patients had higher rates of DM (9.2% vs 4.4%; p < 0.001), CHF (10.0% vs 3.0%; p < 0.001), history of severe COPD (10.27% vs 4.20%; p < 0.001), ventilator dependency (0.3% vs 0.0%; p < 0.001), and had higher rates of chronic steroid use (6.5% vs 4.2%; p = 0.002).
Table III.
Comparison of basic demographics between COVID-19-positive (+) and -negative (-) patients.
| Variable | Total | COVID-19 (+) | COVID-19 (-) | p-value | 1:10 propensity matched COVID-19 (-) cohort | p-value |
|---|---|---|---|---|---|---|
| Patients, n (%) | 194,121 (100.00) | 740 (0.38) | 193,381 (99.62) | 7,400 (90.91) | ||
| Mean age, yrs (SD) | 64.74 (14.81) | 71.12 (16.48) | 64.71 (14.80) | < 0.001 | 71.02 (13.71) | 0.874 |
| Male sex, n (%) | 87,425 (45.04) | 301 (40.68) | 87,124 (45.05) | 0.057 | 2,927 (39.55) | 0.798 |
| Mean BMI, kg/m2 (SD) | 29.91 (8.58) | 26.51 (9.00) | 29.92 (8.58) | < 0.001 | 26.21 (10.85) | 0.393 |
| Race, n (%) | < 0.001 | < 0.001 | ||||
| White | 127,488 (65.67) | 580 (78.38) | 126,908 (65.63) | 5,803 (78.42) | ||
| Black or African-American | 15,672 (8.07) | 43 (5.81) | 15,629 (8.08) | 387 (5.23) | ||
| Other | 8,194 (4.22) | 37 (5.00) | 8,157 (4.22) | 172 (2.32) | ||
| Unknown/Not reported | 42,767 (22.03) | 80 (10.81) | 42,687 (22.07) | 1,038 (14.03) | ||
| ASA grade, n (%) | < 0.001 | 0.537 | ||||
| I | 8,841 (4.55) | 16 (2.16) | 8,825 (4.56) | 120 (1.62) | ||
| II | 84,344 (43.45) | 160 (21.62) | 84,184 (43.53) | 1,725 (23.31) | ||
| III | 91,971 (47.38) | 435 (58.78) | 91,536 (47.33) | 4,261 (57.58) | ||
| IV+ | 8,557 (4.41) | 129 (17.43) | 8,428 (4.36) | 1,284 (17.35) | ||
| Case acuity, n (%) | < 0.001 | < 0.001 | ||||
| Elective | 171,365 (88.28) | 427 (57.70) | 170,938 (88.39) | 5,723 (77.34) | ||
| Urgent | 8,730 (4.50) | 116 (15.68) | 8,614 (4.45) | 610 (8.24) | ||
| Emergent | 14,026 (7.23) | 197 (26.62) | 13,829 (7.15) | 1,067 (14.42) | ||
| Surgery type, n (%) | < 0.001 | < 0.001 | ||||
| Hip/knee arthroplasty | 108,902 (56.10) | 221 (29.86) | 108,681 (56.20) | 3,562 (48.14) | ||
| Trauma | 27,089 (13.95) | 423 (57.16) | 26,666 (13.79) | 2,142 (28.95) | ||
| Spine | 25,762 (13.27) | 50 (6.76) | 25,712 (13.30) | 863 (11.66) | ||
| Sports | 15,777 (8.13) | 21 (2.84) | 15,756 (8.15) | 323 (4.36) | ||
| Shoulder | 16,591 (8.55) | 25 (3.38) | 16,566 (8.57) | 510 (6.89) |
-
ASA, American Society of Anesthesiologists; SD, standard deviation.
Table IV.
Comparison of comorbidities of interest between COVID-19-positive (+) and -negative (-) patients. Data are presented as number (%).
| Variable | Total | COVID-19 (+) | COVID-19 (-) | p-value | 1:10 propensity matched COVID-19 (-) cohort | p-value |
|---|---|---|---|---|---|---|
| Patients, n | 194,121 (100.00) | 740 (0.38) | 193,381 (99.62) | 7,400 (90.91) | ||
| Smoking history | 21,427 (11.04) | 76 (10.27) | 21,351 (11.04) | 0.504 | 792 (10.70) | 0.716 |
| Diabetes | < 0.001 | 0.643 | ||||
| Insulin | 8,615 (4.44) | 68 (9.19) | 8,547 (4.42) | 644 (8.70) | ||
| Non-insulin | 22,665 (11.68) | 84 (11.35) | 22,581 (11.68) | 922 (12.46) | ||
| No | 162,841 (83.89) | 588 (79.46) | 162,253 (83.90) | 5,834 (78.84) | ||
| Congestive heart failure | 5,927 (3.05) | 74 (10.00) | 5,853 (3.03) | < 0.001 | 782 (10.57) | 0.631 |
| Hypertension | 106,362 (54.79) | 412 (55.68) | 105,950 (54.79) | 0.628 | 4,045 (54.66) | 0.597 |
| Ventilator-dependent | 49 (0.03) | 2 (0.27) | 47 (0.02) | < 0.001 | 16 (0.22) | 0.765 |
| History of severe COPD | 8,201 (4.22) | 76 (10.27) | 8,125 (4.20) | < 0.001 | 825 (11.15) | 0.468 |
| Chronic steroid use | 8,191 (4.22) | 48 (6.49) | 8,143 (4.21) | 0.002 | 498 (6.73) | 0.801 |
-
COPD, chronic obstructive pulmonary disease.
Propensity score matching to create a 1:10 COVID-19-negative cohort was constructed based on demographics and comorbidities of the COVID-19-positive cohort. After propensity matching, there were no statistically significant differences between most demographics and all comorbidities between the COVID-19-positive and -negative groups (Table III, Table IV). The demographics that were significantly different despite propensity score matching were patient race, and type of surgery.
Adverse events
Using univariate analysis, complications and adverse events occurring within the 30-day postoperative period were compared (Table II). Based on univariate analyses of the full groups, COVID-19-positive patients had higher rates of any adverse events (21.0% vs 6.1%; p < 0.001), major adverse events (15.1% vs 3.9%; p = 0.013), minor adverse events (11.4% vs 3.0%; p < 0.001), readmission (8.4% vs 3.4%; p < 0.001), blood transfusion (13.2% vs 4.1%; p < 0.001), longer lengths of stay (67.3% vs 24.9%; p < 0.001), and discharge to non-home facilities (45.4% vs 13.7%; p < 0.001).
More specifically, COVID-19-positive patients had higher rates of complications, including mechanical ventilation (1.5% vs 0.1%; p < 0.001), unplanned intubation (2.6% vs 0.2%; p < 0.001), sepsis (4.3% vs 0.6%; p < 0.001), venous thromboembolism (2.4% vs 0.9%; p < 0.001), pneumonia (2.4% vs 0.7%; p < 0.001), and death (6.6% vs 0.8%; p < 0.001) (Table II). COVID-19-positive patients undergoing hip/knee arthroplasty and trauma surgery had significantly higher rates of overall adverse events compared to COVID-19-negative patients (Table V).
Table V.
Comparison of postoperative adverse events between COVID-19-positive (+) and -negative (-) patients based on surgery type.
| Surgery type | COVID-19 (+) cohort | COVID-19 (-) cohort | p-value | 1:10 propensity matched COVID-19 (-) cohort | p-value |
|---|---|---|---|---|---|
| Hip/knee arthroplasty, n (%) | 35 (15.84) | 5,414 (4.98) | < 0.001 | 251 (7.05) | < 0.001 |
| Trauma, n (%) | 114 (26.95) | 4,100 (15.38) | < 0.001 | 421 (19.65) | < 0.001 |
| Spine, n (%) | 5 (10.00) | 1,665 (6.48) | 0.312 | 75 (8.69) | 0.750 |
| Sports, n (%) | 0 (0) | 334 (2.12) | 0.500 | 9 (2.79) | 0.438 |
| Shoulder, n (%) | 1 (4.00) | 358 (2.16) | 0.528 | 15 (2.94) | 0.762 |
After propensity score matching and controlling for all preoperative variables, multivariable analysis found that COVID-19-positive patients were at increased risk of 30-day postoperative complications, including: any adverse events (OR 2.3 (95% CI 1.9 to 2.8); p < 0.001), major adverse events (OR 2.3 (95% CI 1.8 to 2.9); p < 0.001), and minor adverse events (OR 2.6 (95% CI 2.0 to 3.3); p < 0.001). Specifically, risk of death (OR 2.6 (95% CI 1.9 to 3.7); p < 0.001), venous thromboembolism (OR 1.8 (95% CI 1.1 to 3.0); p < 0.001), and pneumonia (OR 4.91 (95% CI 3.4 to 7.0); p < 0.001) were also increased. COVID-19-positive patients undergoing total hip/knee arthroplasty (OR 2.6 (95% CI 1.7 to 3.9); p < 0.001) and trauma surgery (OR 1.8 (95% CI 1.4 to 2.3); p < 0.001) were at increased risk of 30-day adverse events. With regard to other postoperative outcomes, COVID-19-positive patients were at increased risk for: longer lengths of stay (OR 5.2 (95% CI 4.2 to 6.4); p < 0.001), discharge to non-home facilities (OR 2.5 (95% CI 2.0 to 3.0); p < 0.001), and readmission (OR 1.5 (95% CI 1.1 to 2.0); p = 0.004) (Figure 1).
Fig. 1
Association of COVID-19 positivity with 30-day adverse events: multivariate analysis (controlling for age, sex, race, BMI, and American Society of Anesthesiologists grade). CI, confidence interval; OR, odds ratio.
Discussion
Since the start of the COVID-19 pandemic, hospitals, medical institutions, and healthcare systems have experienced unprecedented levels of stress: the American Hospital Association estimates a financial impact of $50.7 billion per month in lost revenue for America’s hospitals and healthcare systems.2
From a surgical perspective, early evidence from 2020 indicated that COVID-19 positivity was associated with increased rates of perioperative morbidity and mortality.3 Furthermore, due to the pandemic, in 2020 healthcare facilities suspended elective and non-urgent surgical procedures, leading to massive backlogs and loss of projected revenue.27 In 2021, as elective and non-urgent procedures resumed, the complications associated with COVID-19 positivity during the immediate perioperative period were further documented. With the recent release of 2021 data from NSQIP, we were able to assess the impact of COVID-19 positivity on perioperative adverse outcomes. We conducted a matched cohort analysis to identify patient risk factors independently associated with the occurrence of 30-day postoperative adverse events. We demonstrated that COVID-19-positive patients undergoing orthopaedic surgery had increased odds of many 30-day postoperative complications, with hip/knee arthroplasty and trauma surgery being the most high-risk procedures.
The rate of COVID-19 positivity in this cohort was found to be 0.38%. In the literature, reported rates of COVID-19 positivity have varied widely, from approximately 0.5% to 15% depending on patient cohort, timing since the start of the pandemic, variant surges, and geographical location.15,28-34 In this analysis, COVID-19 positivity was defined based on lab-confirmed diagnosis. While these rates may be underestimating the true COVID-19 positivity rate among patients undergoing orthopaedic surgery, considering the small number of potential missed cases, it would be unlikely to alter the findings of our study.
At baseline, we demonstrated that COVID-19-positive patients were significantly older, had lower BMIs, and higher ASA grades. When comparing patient comorbidities, we identified that COVID-19-positive patients had higher rates of DM, CHF, and severe COPD, were more likely to be ventilator-dependent, and had higher rates of chronic steroid use. The largest difference between rates of comorbidities between COVID-19-positive and -negative patients was found to be for history of CHF and severe COPD. These findings regarding comorbidities are consistent with previous reports.5,15,35-37
After propensity matching and controlling for all preoperative variables, multivariable analysis found that COVID-19-positive patients were at increased risk of several postoperative complications, including any adverse events, major adverse events, minor adverse events, and death. Our findings are consistent with multiple previous studies investigating COVID-19 positivity in perioperative patients. Large multicentre studies have demonstrated increased rates of pulmonary complications and 30-day mortality in COVID-19-positive patients undergoing a variety of surgical procedures.4,38 However, our findings are in contrast to a recently published population-based cohort study of 71,144 Canadian patients undergoing major non-cardiac surgery, in which COVID-19 positivity was not associated with death, major adverse cardiovascular events, or rehospitalization following elective major non-cardiac surgery.34 These variations may be partly explained by the quality of the database and retrospective nature of both analyses.
Based on the multivariable analysis, we also demonstrated that COVID-19-positive patients undergoing orthopaedic procedures have almost a two-fold increase in odds of venous thromboembolism and a five-fold increase in odds of perioperative pneumonia. Numerous studies have established COVID-19 as a risk factor for deep vein thrombosis, pulmonary embolism, and bleeding, due to the endothelial dysfunction and the subsequent procoagulative state caused by endotheliitis.10,39-41 Increased risk of symptomatic perioperative pneumonia associated with COVID-19 positivity is rather expected, either caused by progression of the upper respiratory viral infection, worsening of a pre-existing pneumonia, or development of a secondary superimposed bacterial pneumonia.42
In this analysis, we also demonstrated that among COVID-19-positive cases, a higher proportion were deemed urgent or emergent trauma. This is likely due to the preoperative screening and isolation protocols implemented by many hospitals prior to elective surgery. Moreover, it has been demonstrated that the rate of morbidity and mortality in COVID-19-positive patients are higher after emergency surgery compared with elective surgery, possibly due to the nature related to the associated acute, emergent diagnosis.4 The COVIDSurg Collaborative conducted a multicentre prospective cohort study on patients undergoing emergent and elective surgery, demonstrating a four-fold increased odds of 30-day mortality in COVID-19-positive patients compared to those without infection at the time of surgery.43
To better risk-stratify the orthopaedic surgical population, patients were further stratified by type of orthopaedic procedure; among the included patients, the most common procedures were hip and knee arthroplasty followed by trauma, spine, sports, and finally, shoulder surgery. We demonstrated an increased risk of 30-day adverse events in COVID-19-positive patients undergoing hip and knee arthroplasty and trauma surgery. While there appears to be consensus on increased risk of perioperative complications in trauma surgery in COVID-19-positive patients,16,27,44,45 our results with regard to joint arthroplasty are in contrast to some recent reports arguing COVID-19 positivity does not significantly increase risk of respiratory, infectious, cardiac, and thromboembolic complications.46,47 These differences may again be explained by the quality and size of studied databases and retrospective nature of analyses.
Finally, with regard to other postoperative outcomes, we demonstrated that COVID-19-positive patients were at increased risk for longer lengths of stay, discharge to non-home facilities, and readmission. Our findings are consistent with another large population-based study on hip fracture patients, which demonstrated a two-fold increase in length of stay in COVID-19-positive patients.17,48 Notably, multiple studies and a systematic review compared lengths of stay in hip fracture patients before and after the pandemic, and demonstrated reduced lengths of stay, likely due to an attempt to prevent nosocomial viral transmission to the patient or other hospitalized patients.49 A previous study compared postoperative outcomes of hip fracture patients before and after the pandemic, and found no differences in rates of 30-day hospital readmission and a decrease in rates of patients placed in skilled nursing facilities. The decreased placement of patients in nursing facilities may be attributed to the lack of nursing resources, together with the burden of the pandemic on nurses and healthcare providers.50
Strengths of this study include its large sample size, use of propensity score matching to compare COVID-19-positive versus negative patients across a wide breadth of orthopaedic procedures, and use of a highly validated dataset across hundreds of participating hospitals across the USA and Canada. This is the first study to assess perioperative complications associated with COVID-19 positivity using the NSQIP database in orthopaedic surgery. NSQIP is a nationally validated, risk-adjusted, outcomes-based registry designed to measure and improve the quality of surgical care.51 However, this study is not without limitations, and as such the results should be interpreted accordingly. Limitations of analysis using the NSQIP registry include the fact that this database only tracks 30-day morbidity and mortality after a surgical procedure.52 Furthermore, the NSQIP registry also does not provide data on severity of patient comorbidities or initial indications for surgeries.52
To conclude, in this matched cohort analysis, we demonstrated that COVID-19 positivity during orthopaedic surgery increases odds of major and minor 30-day postoperative complications, with hip/knee arthroplasty and trauma surgery being the highest-risk procedures. Furthermore, these data reinforce prior literature demonstrating increased risk of mortality and venous thromboembolic events in the acute postoperative period. Clinicians caring for patients undergoing orthopaedic procedures should be mindful of these increased risks and attempt to improve patient care during the ongoing global pandemic.
References
1. Hamid H , Abid Z , Amir A , Rehman TU , Akram W , Mehboob T . Current burden on healthcare systems in low- and middle-income countries: recommendations for emergency care of COVID-19 . Drugs Ther Perspect . 2020 ; 36 ( 10 ): 466 – 468 . Crossref PubMed Google Scholar
2. Kaye AD , Okeagu CN , Pham AD , et al. Economic impact of COVID-19 pandemic on healthcare facilities and systems: international perspectives . Best Pract Res Clin Anaesthesiol . 2021 ; 35 ( 3 ): 293 – 306 . Crossref PubMed Google Scholar
3. Knisely A , Zhou ZN , Wu J , et al. Perioperative morbidity and mortality of patients with COVID-19 who undergo urgent and emergent surgical procedures . Ann Surg . 2021 ; 273 ( 1 ): 34 – 40 . Crossref PubMed Google Scholar
4. COVIDSurg Collaborative . Mortality and pulmonary complications in patients undergoing surgery with perioperative SARS-CoV-2 infection: an international cohort study . Lancet . 2020 ; 396 ( 10243 ): 27 – 38 . Crossref PubMed Google Scholar
5. Suleyman G , Fadel RA , Malette KM , et al. Clinical characteristics and morbidity associated with Coronavirus disease 2019 . JAMA Netw open . 2020 . Crossref PubMed Google Scholar
6. Casanova J , Pissarra D , Costa R , Salgueiro E , Pinho P . Cardiothoracic surgery during the Covid-19 pandemic: perioperative care, safety, and surgical results . J Card Surg . 2020 ; 35 ( 10 ): 2605 – 2610 . Crossref PubMed Google Scholar
7. Kiyatkin ME , Levine SP , Kimura A , et al. Increased incidence of post-operative respiratory failure in patients with pre-operative SARS-CoV-2 infection . J Clin Anesth . 2021 ; 74 : 110409 . Crossref PubMed Google Scholar
8. Raisi-Estabragh Z , Cooper J , Salih A , et al. Cardiovascular disease and mortality sequelae of COVID-19 in the UK biobank . Heart . 2022 ; 109 ( 2 ): 119 – 126 . Crossref PubMed Google Scholar
9. Zeleke AJ , Moscato S , Miglio R , Chiari L . Length of stay analysis of COVID-19 hospitalizations using a count regression model and quantile regression: a study in Bologna, Italy . Int J Environ Res Public Health . 2022 ; 19 ( 4 ): 2224 . Crossref PubMed Google Scholar
10. Katsoularis I , Fonseca-Rodríguez O , Farrington P , et al. Risks of deep vein thrombosis, pulmonary embolism, and bleeding after COVID-19: nationwide self-controlled cases series and matched cohort study . BMJ . 2022 ; 377 : e069590 . Crossref PubMed Google Scholar
11. Green G , Abbott S , Vyrides Y , Afzal I , Kader D , Radha S . The impact of the COVID-19 pandemic on the length of stay following total hip and knee arthroplasty in a high volume elective orthopaedic unit . Bone Jt Open . 2021 ; 2 ( 8 ): 655 – 660 . Crossref PubMed Google Scholar
12. Cataldo P , Verdugo FJ , Bonta C , et al. Consequences of COVID-19 pandemic on myocardial infarction reperfusion therapy and prognosis . Rev Med Chil . 2021 ; 149 ( 5 ): 672 – 681 . Crossref PubMed Google Scholar
13. Doglietto F , Vezzoli M , Gheza F , et al. Factors associated with surgical mortality and complications among patients with and without Coronavirus disease 2019 (COVID-19) in Italy . JAMA Surg . 2020 ; 155 ( 8 ): 691 – 702 . Crossref PubMed Google Scholar
14. Galivanche AR , Mercier MR , Schneble CA , et al. Clinical characteristics and perioperative complication profiles of COVID-19-positive patients undergoing hip fracture surgery . J Am Acad Orthop Surg Glob Res Rev . 2021 ; 5 ( 10 ): e21.00104 . Crossref PubMed Google Scholar
15. Mercier MR , Galivanche AR , Brand JP , et al. COVID-positive ankle fracture patients are at increased odds of perioperative surgical complications following open reduction internal fixation surgery . PLoS One . 2021 ; 16 ( 12 ): e0262115 . Crossref PubMed Google Scholar
16. Al-Humadi SM , Tantone R , Nazemi AK , et al. Outcomes of orthopaedic trauma surgery in COVID-19 positive patients . OTA Int . 2021 ; 4 ( 2 ): e129 . Crossref PubMed Google Scholar
17. Zhong H , Poeran J , Liu J , Wilson LA , Memtsoudis SG . Hip fracture characteristics and outcomes during COVID-19: a large retrospective national database review . Br J Anaesth . 2021 ; 127 ( 1 ): 15 – 22 . Crossref PubMed Google Scholar
18. Rastad H , Karim H , Ejtahed H-S , et al. Risk and predictors of in-hospital mortality from COVID-19 in patients with diabetes and cardiovascular disease . Diabetol Metab Syndr . 2020 ; 12 ( 1 ): 57 . Crossref PubMed Google Scholar
19. Singh R , Rathore SS , Khan H , et al. Association of obesity With COVID-19 severity and mortality: an updated systemic review, meta-analysis, and meta-regression . Front Endocrinol . ; 13 : 780872 . n.d. Crossref PubMed Google Scholar
20. Chouairi F , Mercier MR , Persing JS , Gabrick KS , Clune J , Alperovich M . National patterns in surgical management of syndactyly: a Review of 956 Cases . Hand (N Y) . 2020 ; 15 ( 5 ): 666 – 673 . Crossref PubMed Google Scholar
21. Jawad MU , Delman CM , Campbell ST , et al. Traumatic proximal femoral fractures during COVID-19 pandemic in the US: an ACS NSQIP® analysis . J Clin Med . 2022 ; 11 ( 22 ): 6778 . Crossref PubMed Google Scholar
22. No authors listed . CPT (Current Procedural Terminology) . American Medical Association . 2023 . https://www.ama-assn.org/amaone/cpt-current-procedural-terminology ( date last accessed 8 August 2023 ). Google Scholar
23. Molina CS , Thakore RV , Blumer A , Obremskey WT , Sethi MK . Use of the national surgical quality improvement program in orthopaedic surgery . Clin Orthop Relat Res . 2015 ; 473 ( 5 ): 1574 – 1581 . Crossref PubMed Google Scholar
24. No authors listed . ACS NSQIP 2021 PUF User Guide , American College of Surgeons . 2022 . https://www.facs.org/media/tjcd1biq/nsqip_puf_userguide_2021_20221102120632.pdf ( date last accessed 6 July 2023 ). Google Scholar
25. Saklad M . Grading of patients for surgical procedures . Anesthesiology . 1941 ; 2 ( 3 ): 281 – 284 . Crossref Google Scholar
26. Ho D , Imai K , King G , MatchIt SEA . Nonparametric preprocessing for parametric causal inference . J Stat Softw . 2011 ; 42 ( 8 SE-Articles ): 1 – 28 . Crossref Google Scholar
27. Hanreich C , Boettner F , DeNegre S , Jungwirth-Weinberger A , Jerabek S . The impact of the COVID-19 associated shutdown on orthopedic patient care . Arch Orthop Trauma Surg . 2023 ; 143 ( 6 ): 2885 – 2892 . Crossref PubMed Google Scholar
28. Xu X-W , Wu X-X , Jiang X-G , et al. Clinical findings in a group of patients infected with the 2019 novel coronavirus (SARS-Cov-2) outside of Wuhan, China: retrospective case series . BMJ . 2020 ; 368 : m606 . Crossref PubMed Google Scholar
29. Sood N , Simon P , Ebner P , et al. Seroprevalence of SARS-CoV-2-specific antibodies among adults in Los Angeles County, California, on april 10-11, 2020 . JAMA . 2020 ; 323 ( 23 ): 2425 – 2427 . Crossref PubMed Google Scholar
30. Eckerle I , Meyer B . SARS-CoV-2 seroprevalence in COVID-19 hotspots . Lancet . 2020 ; 396 ( 10250 ): 514 – 515 . Crossref PubMed Google Scholar
31. Pollán M , Pérez-Gómez B , Pastor-Barriuso R , et al. Prevalence of SARS-CoV-2 in Spain (ENE-COVID): a nationwide, population-based seroepidemiological study . Lancet . 2020 ; 396 ( 10250 ): 535 – 544 . Crossref PubMed Google Scholar
32. Stringhini S , Wisniak A , Piumatti G , et al. Seroprevalence of anti-SARS-CoV-2 IgG antibodies in Geneva, Switzerland (SEROCoV-POP): a population-based study . Lancet . 2020 ; 396 ( 10247 ): 313 – 319 . Crossref PubMed Google Scholar
33. Poon RW-S , Chan BP-C , Chan W-M , et al. SARS-CoV-2 IgG seropositivity after the severe Omicron wave of COVID-19 in Hong Kong . Emerg Microbes Infect . 2022 ; 11 ( 1 ): 2116 – 2119 . Crossref PubMed Google Scholar
34. Quinn KL , Huang A , Bell CM , et al. Complications following elective major noncardiac surgery among patients with prior SARS-CoV-2 infection . JAMA Netw Open . 2022 ; 5 ( 12 ): e2247341 . Crossref PubMed Google Scholar
35. Xie J , Tong Z , Guan X , Du B , Qiu H . Clinical characteristics of patients who died of coronavirus disease 2019 in China . JAMA Netw Open . 2020 ; 3 ( 4 ): e205619 . Crossref PubMed Google Scholar
36. Richardson S , Hirsch JS , Narasimhan M , et al. Presenting characteristics, comorbidities, and outcomes among 5700 patients hospitalized with COVID-19 in the New York City area . JAMA . 2020 ; 323 ( 20 ): 2052 – 2059 . Crossref PubMed Google Scholar
37. Ejaz H , Alsrhani A , Zafar A , et al. COVID-19 and comorbidities: deleterious impact on infected patients . J Infect Public Health . 2020 ; 13 ( 12 ): 1833 – 1839 . Crossref PubMed Google Scholar
38. COVIDSurg Collaborative . Outcomes and their state-level variation in patients undergoing surgery with perioperative SARS-CoV-2 infection in the USA: a prospective multicenter study . Ann Surg . 2022 ; 275 ( 2 ): 247 – 251 . Crossref PubMed Google Scholar
39. Teuwen LA , Geldhof V , Pasut A , Carmeliet P . COVID-19: the vasculature unleashed . Nat Rev Immunol . 2020 ; 20 ( 7 ): 389 – 391 . Crossref PubMed Google Scholar
40. Varga Z , Flammer AJ , Steiger P , et al. Endothelial cell infection and endotheliitis in COVID-19 . Lancet . 2020 ; 395 ( 10234 ): 1417 – 1418 . Crossref PubMed Google Scholar
41. Ackermann M , Verleden SE , Kuehnel M , et al. Pulmonary vascular endothelialitis, thrombosis . N Engl J Med . 2020 . Crossref PubMed Google Scholar
42. De Bruyn A , Verellen S , Bruckers L , et al. Secondary infection in COVID-19 critically ill patients: a retrospective single-center evaluation . BMC Infect Dis . 2022 ; 22 ( 1 ): 207 . Crossref PubMed Google Scholar
43. Timing of surgery following SARS-Cov-2 infection: an international prospective cohort study . Anaesthesia England . 2021 ; 76 ( 6 ): 748 – 758 . Crossref PubMed Google Scholar
44. Crozier-Shaw G , Hughes AJ , Conlon B , Sheehan E , Merghani K . Hip fracture care during Covid-19: a regional trauma centre’s experience . Ir J Med Sci . 2021 ; 190 ( 4 ): 1275 – 1280 . Crossref PubMed Google Scholar
45. COVIDSurg Collaborative . Mortality and pulmonary complications in emergency general surgery patients with COVID-19: a large international multicenter study . J Trauma Acute Care Surg . 2022 ; 93 ( 1 ): 59 – 65 . Crossref PubMed Google Scholar
46. Lung BE , Taka TM , Donnelly M , et al. Prior diagnosis of COVID has no increased complications in total joint arthroplasty . Cureus . 2022 ; 14 ( 8 ): e27974 . Crossref PubMed Google Scholar
47. Rosas S , Pollock DC , Roche MW , et al. Patients with previous covid-19 infection can safely undergo primary total joint arthroplasty . J Arthroplasty . 2023 ; 38 ( 4 ): 649 – 654 . Crossref PubMed Google Scholar
48. Yawar B , Salmon J , McSorley A , et al. Impact of COVID-19 pandemic on the length of hospital stay in hip fracture patients: a single centre study . Cureus . 2022 ; 14 ( 7 ): e27328 . Crossref PubMed Google Scholar
49. Tayyebi H , Hasanikhah M , Heidarikhoo M , Fakoor S , Aminian A . Length of hospital stay and mortality of hip fracture surgery in patients with Coronavirus disease 2019 (COVID-19) infection: a systematic review and meta-analysis . Curr Orthop Pract . 2022 ; 33 ( 2 ): 172 – 177 . Crossref PubMed Google Scholar
50. Helal A , Botros D , Qureshi F , et al. Effects of the COVID-19 pandemic on hip fracture volume, disposition, and readmission rates . Proc (Bayl Univ Med Cent) . 2022 ; 35 ( 4 ): 444 – 446 . Crossref PubMed Google Scholar
51. Khuri SF , Daley J , Henderson W , et al. The department of veterans affairs’ NSQIP: the first national, validated, outcome-based, risk-adjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care. National VA surgical quality improvement program . Ann Surg . 1998 ; 228 ( 4 ): 491 – 507 . Crossref PubMed Google Scholar
52. Alluri RK , Leland H , Heckmann N . Surgical research using national databases . Ann Transl Med . 2016 ; 4 ( 20 ): 393 . Crossref PubMed Google Scholar
Author contributions
M. R. Mercier: Conceptualization, Methodology, Data curation, Writing – original draft, Project administration.
R. Koucheki: Methodology, Formal analysis, Investigation, Writing – original draft, Visualization.
J. R. Lex: Methodology, Validation, Writing – review & editing.
A. Khoshbin: Writing – review & editing, Conceptualization, Supervision.
S. S. Park: Writing – review & editing, Validation, Supervision.
T. R. Daniels: Supervision, Writing – review & editing.
M. M. Halai: Project administration, Data curation, Writing – review & editing, Supervision.
Funding statement
The authors received no financial or material support for the research, authorship, and/or publication of this article.
ICMJE COI statement
T. R. Daniels reports grants and personal fees from Integra, grants and personal fees from Stryker, personal fees from Journal of Bone and Joint Surgery (American Volume), and grants and personal fees from Wright Medical Technology, outside the submitted work. A. Khoshbin reports consultant fees from Smith & Nephew. The remaining authors have no conflicts of interest to report.
Data sharing
The data for this study are publicly available at https://www.facs.org/quality-programs/data-and-registries/acs-nsqip/
Ethical review statement
Local institutional ethics review board approval was obtained (REB#: 18-246).
Open access funding
The authors confirm that the open access fee for this article was self-funded.
Follow M. Robert. Mercier @MichaelRMercier
Follow R. Koucheki @RobKoucheki
Follow J. R. Lex @DrJRLex
© 2023 Author(s) et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/
