In the United States, the demand for both primary and revision total knee arthroplasty (TKA) is projected to dramatically increase during the next decade.1,2 Given the prevalence of the TKA procedures, it is crucial that orthopedic surgeons improve their understanding of the drivers of postoperative complications to minimize postoperative complications, optimize patient outcomes, and reduce the financial burden on the health care system. Recently, patient optimization protocols have been implemented to improve modifiable risk factors (eg, weight and diabetes management) in the perioperative period.3,4 Previous studies analyzing a variety of surgical procedures demonstrated an association between increased operative time and the development of surgical site infections.5–8 Furthermore, Procter et al9 noted an increased incidence of other infectious complications as well as an increase in postoperative length of stay with an increase in operative time. Currently, the relationship between operative time and postoperative complications following revision TKA is poorly understood.
Thus, this study used the American College of Surgeons National Surgical Quality Improvement Program (ACSNSQIP) database to (1) identify the incidence of specific postoperative complications following revision TKA and (2) determine the impact of operative time, as a continuous variable, on the development of specific postoperative complications.
Materials and Methods
Data were collected using the ACSNSQIP—a nationwide, multicenter database that provides risk-adjusted outcome measures to improve surgical care. Patients who had undergone revision TKA between 2007 and 2016 were identified using the following Current Procedural Terminology codes: 27486 and 27487. Individuals who had an infectious International Classification of Diseases code or did not have an operative time recorded were excluded from the study. Operative time, defined by the ACS-NSQIP as “the time from first incision to the end of the procedure,” was modeled as a continuous variable using 15-minute increments. To minimize potential transcription errors, cases less than 30 minutes or greater than 360 minutes were omitted. Additional patient information, including patient demographics and comorbidities, is presented in Table 1.
A total of 14,769 patients were included in this study. The mean operative time was 133 minutes (Figure 1). Among the selected cohort, the majority of patients were female (60.1%), were White (73.3%), received general anesthesia (61.0%), did not have diabetes (79.2%), had an American Society of Anesthesiologists score of 1 or 2 (95.3%), did not have dyspnea (92.7%), had an independent functional status (95.6%), did not smoke (88.1%), and were not obese (35.6%).
The 30-day postoperative complications were recorded and then subsequently grouped based on similar etiology. The complications and groups analyzed included wound complications (superficial surgical site wound infection, deep wound infection, organ or space infections, and wound dehiscence); pulmonary (pneumonia, reintubation, and failure to wean off ventilator for greater than 48 hours); cardiac (cardiac arrest and myocardial infarction); renal (renal insufficiency and renal failure); thromboembolic (pulmonary embolism, deep venous thrombosis, and stroke); sepsis (sepsis and sepsis shock); postoperative transfusion; urinary tract infection; death; extended hospital stay of greater than 7 days: and unplanned return to the operating room.
Statistical Analysis
The Statistical Package for the Social Sciences, version 22, software (IBM Corp) was used to perform multinomial multivariate analyses of patient demographics, comorbidities, and postoperative complications. Cofactors considered for multivariate analysis included collected demographics and comorbidities. For postoperative complications with a significant unadjusted odds ratio (OR), a multivariate analysis was conducted. P values and ORs with 95% CIs were recorded. P<.05 was used to determine statistical significance for unadjusted and adjusted analyses.
Results
Overall Complications
In total, there were 419 (2.8%) wound, 95 (0.6%) pulmonary, 61 (0.4%) cardiac, 45 (0.3%) renal, 184 (1.2%) thromboembolic, and 137 (0.9%) sepsis complications among patients in this cohort. In addition, 122 (0.8%) patients had a urinary tract infection. A postoperative transfusion was required by 1584 (10.7%) patients. Thirty-five patients (0.2%) died within the acute postoperative period. Finally, 664 (4.5%) patients had an extended hospital stay of greater than 7 days and 495 (3.4%) patients had an unplanned return to the operating room (Table 2).
Operative Time
On unadjusted analysis, every 15-minute increase in operative time led to a 1.028 times greater chance of developing a wound complication (95% CI, 1.004–1.053; P=.002), 1.177 times greater chance of requiring a postoperative transfusion (95% CI, 1.163–1.193; P<.001), 1.056 times greater chance of requiring an extended hospital stay (95% CI, 1.036–1.076; P<.001), and 1.036 times greater chance of requiring an unplanned return to the operating room (95% CI, 1.013–1.059; P=.002) (Table 3).
On adjusted multivariate analysis, every 15-minute increase in operative time was associated with wound complications (OR, 1.023; 95% CI, 1.012–1.049; P=.020), postoperative blood transfusion (OR, 1.169; 95% CI, 1.152–1.185; P<.001), and extended hospital stay (OR, 1.060; 95% CI, 1.037–1.083; P<.001) (Table 3). Operative time was not independently associated with an unplanned return to the operating room (Table 3).
Discussion
A variety of studies have demonstrated a relationship between operative time and various postoperative complications in the primary setting.8–12 However, there remains a paucity of studies regarding operative time and postoperative complications in the revision TKA literature. Recently, Garbarino et al13 noted the impact of operative time on length of stay in revision TKA but failed to comment on the influence of operative time on the development of postoperative complications. In the current study, the authors examined 14,769 revision TKAs to determine that a 15-minute increase in operative time increased the likelihood of wound complications, postoperative blood transfusions, and extended hospital stay.
The authors found a small but significant increase in the risk for wound infection, with an OR of 1.02 for each 15-minute increase in operative time. Similar findings for primary TKA and total hip arthroplasty have noted the association between prolonged operative time and wound complication.6,7,10 With both sepsis and deep surgical site infections associated with longer operative times, a reduction in revision TKA operative times could reduce infection risk after these complex procedures.
In some respects, operative time can be viewed as a marker for surgical complexity. Although there are estimated time frames, patients are not the same. To that end, an increased risk of postoperative transfusion and extended hospital stay are likely considerations with an increase in operative time. The current results agree with this notion, showcasing significant increases in risk for both complications for every 15-minute increase in operative time. Operative time as a surrogate marker for surgical complexity can also be suggestive of an increased necessity to return to the operating room. The current results further concur with this, as the risk for unplanned return to the operating room increased every 15 minutes.
Finally, the authors did not find increased operative time correlating with thromboembolic complications. These results differ from prior literature showing a direct association between increased operative time and increased risk for a venous thromboembolic event.14 Previous studies have revealed that early release of the tourniquet in primary TKAs can significantly decrease the incidence of deep venous thrombosis, and that certain nonstandard prophylaxis treatments such as the concurrent use of fish oil and aspirin may significantly lower risks for thromboembolic events.15,16
There were several notable limitations to this study. The NSQIP database is unable to account for factors that may influence a prolonged operative time, such as differences in surgical complexity, surgeon experience and volume, hospital volume, and specific operating room personnel (attending vs resident physicians). Further, the differences in postoperative management between surgeons were not accounted for. All of these factors could have produced variation in postoperative status and the development of subsequent postoperative complications. Specifically, the NSQIP database does not document the use of a tourniquet or tranexamic acid, both of which may influence the need for postoperative blood transfusions. Despite these limitations, the authors attempted to limit the impact of most confounders through the exclusion of septic procedures and the control of numerous demographic variables and comorbidities. Further, the NSQIP database collects only 30-day postoperative outcomes, making it difficult to determine the long-term effects of operative time. The NSQIP database also collects only general postoperative outcomes and does not record data on specific TKA outcomes such as range of motion and function. Finally, this was a retrospective study of a large national database. Prospective studies should be performed to establish causality. Despite these limitations, the sample size of 14,769 revision TKAs lends significant power to the conclusions of this study. Furthermore, the size of the cohort allowed for the analysis of operative time in a continuous fashion, broken down into 15-minute increments.
Conclusion
This study evaluated the effect of increased operative time, for 14,769 revision TKA patients, on multiple postoperative complications. Given the increase in revision TKAs, orthopedic surgeons must consider operative time as a predictor for postoperative complications.
References
- Fingar KR, Stocks C, Weiss AJ, Steiner CA. Most frequent operating room procedures performed in US hospitals. Agency for Healthcare Research and Quality. Accessed June 13, 2017. http://www.ncbi.nlm.nih.gov/pubmed/25695123
- Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007;89(4):780–785. doi:10.2106/00004623-200704000-00012 [CrossRef] PMID:17403800
- Holt JB, Miller BJ, Callaghan JJ, Clark CR, Willenborg MD, Noiseux NO. Minimizing blood transfusion in total hip and knee arthroplasty through a multimodal approach. J Arthroplasty. 2016;31(2):378–382. doi:10.1016/j.arth.2015.08.025 [CrossRef] PMID:26391927
- Golladay GJ, Satpathy J, Jiranek WA. Patient optimization—strategies that work: malnutrition. J Arthroplasty. 2016;31(8):1631–1634. doi:10.1016/j.arth.2016.03.027 [CrossRef] PMID:27118349
- Bohl DD, Ondeck NT, Darrith B, Hannon CP, Fillingham YA, Della Valle CJ. Impact of operative time on adverse events following primary total joint arthroplasty. J Arthroplasty. 2018;33(7):2256–2262.e4. doi:10.1016/j.arth.2018.02.037 [CrossRef] PMID:29551302
- Catanzarite T, Saha S, Pilecki MA, Kim JY, Milad MP. Longer operative time during benign laparoscopic and robotic hysterectomy with increased 30 day perioperative complications. J Minim Invasive Gynecol. 2015;22(6):1049–1058. doi:10.1016/j.jmig.2015.05.022 [CrossRef] PMID:26070725
- Naranje S, Lendway L, Mehle S, Gioe TJ. Does operative time affect infection rate in primary total knee arthroplasty?Clin Orthop Relat Res. 2015;473(1):64–69. doi:10.1007/s11999-014-3628-4 [CrossRef] PMID:24740318
- Cheng H, Chen BPH, Soleas IM, Ferko NC, Cameron CG, Hinoul P. Prolonged operative duration increases risk of surgical site infections: a systematic review. Surg Infect (Larchmt). 2017;18(6):722–735. doi:10.1089/sur.2017.089 [CrossRef] PMID:28832271
- Procter LD, Davenport DL, Bernard AC, Zwischenberger JB. General surgical operative duration is associated with increased risk-adjusted infectious complication rates and length of hospital stay. J Am Coll Surg. 2010;210(1):60–5.e1, 2. doi:10.1016/j.jamcollsurg.2009.09.034 [CrossRef] PMID:20123333
- Dicks KV, Baker AW, Durkin MJ, et al. Short operative duration and surgical site infection risk in hip and knee arthroplasty procedures. Infect Control Hosp Epidemiol. 2015;36(12):1431–1436. doi:10.1017/ice.2015.222 [CrossRef] PMID:26391277
- Ong KL, Lau E, Manley M, Kurtz SM. Effect of procedure duration on total hip arthroplasty and total knee arthroplasty survivorship in the United States Medicare population. J Arthroplasty. 2008;23(6)(suppl 1):127–132. doi:10.1016/j.arth.2008.04.022 [CrossRef] PMID:18555641
- Ou L, Chen J, Hillman K, et al. The impact of post-operative sepsis on mortality after hospital discharge among elective surgical patients: a population-based cohort study. Crit Care. 2017;21(1):34. doi:10.1186/s13054-016-1596-7 [CrossRef] PMID:28219408
- Garbarino LJ, Gold PA, Sodhi N, et al. The effect of operative time on in-hospital length of stay in revision total knee arthroplasty. Ann Transl Med. 2019;7(4):66. doi:10.21037/atm.2019.01.54 [CrossRef] PMID:30963061
- Kim JYS, Khavanin N, Rambachan A, et al. Surgical duration and risk of venous thromboembolism. JAMA Surg. 2015;150(2):110–117. doi:10.1001/jamasurg.2014.1841 [CrossRef] PMID:25472485
- Bonutti PM, Sodhi N, Patel YH, et al. Novel venous thromboembolic disease (VTED) prophylaxis for total knee arthroplasty: aspirin and fish oil. Ann Transl Med. 2017;5(S3)(suppl 3):S30. doi:10.21037/atm.2017.11.22 [CrossRef] PMID:29299477
- Zan P, Mol MO, Yao JJ, et al. Release of the tourniquet immediately after the implantation of the components reduces the incidence of deep vein thrombosis after primary total knee arthroplasty. Bone Joint Res. 2017;6(9):535–541. doi:10.1302/2046-3758.69.BJR-2017-0149.R2 [CrossRef] PMID:28899856
Patient Demographics and Comorbidities (N=14,769)
| Factor | No. |
|---|
| Demographics | |
| Sexa | |
| Male | 5887 (39.9%) |
| Female | 8876 (60.1%) |
| Agea | |
| ≤50 y | 1020 (6.9%) |
| 51–60 y | 3352 (22.7%) |
| 61–70 y | 5179 (35.1%) |
| 71–80 y | 3776 (25.6%) |
| 80+ y | 1391 (9.4%) |
| BMIa | |
| 18.5–30 kg/m2 | 5256 (35.6%) |
| 31–34.9 kg/m2 | 4107 (27.8%) |
| 35–39.9 kg/m2 | 2772 (18.8%) |
| ≥40 kg/m2 | 2492 (16.9%) |
| Racea | |
| White | 10,821 (73.3%) |
| Black | 1753 (11.9%) |
| Hispanic | 689 (4.7%) |
| American Indian or Alaska Native | 76 (0.5%) |
| Asian | 140 (0.9%) |
| Native Hawaiian or Pacific Islander | 54 (0.4%) |
| Anesthesiaa | |
| General | 9011 (61.0%) |
| Regional | 4203 (28.5%) |
| MAC/IV sedation | 1012 (6.9%) |
| Diabetes mellitus status | |
| No diabetes mellitus | 11,689 (79.2%) |
| Non–insulin-dependent diabetes mellitus | 2089 (14.1%) |
| Insulin-dependent diabetes mellitus | 991 (6.7%) |
| ASAa | |
| 1 or 2 | 14,079 (95.3%) |
| 3 or 4 | 489 (3.3%) |
| Dyspnea | |
| No dyspnea | 13,690 (92.7%) |
| Moderate exertion | 1036 (7.0%) |
| At rest | 43 (0.3%) |
| Functional status preoperativea | |
| Independent | 14,125 (95.6%) |
| Partially dependent | 509 (3.4%) |
| Totally dependent | 28 (0.2%) |
| Smoking | 1764 (11.9%) |
| Comorbidities | |
| COPD | 787 (5.3%) |
| CHF | 75 (0.5%) |
| Hypertension | 9840 (66.6%) |
| Renal failure | 7 (0.0%) |
| Dialysis | 49 (0.3%) |
| Weight loss | 41 (0.3%) |
| Bleeding disorder | 592 (4.0%) |
| Steroid use | 676 (4.6%) |
| Preoperative transfusion | 70 (0.5%) |
Postoperative Complications of Patients Undergoing Revision TKA (N= 14,769)
| Complication | No. |
|---|
| Superficial surgical site wound | 126 (0.9%) |
| Deep wound infection | 100 (0.7%) |
| Organ or space infection | 162 (1.1%) |
| Wound dehiscence | 66 (0.4%) |
| Pneumonia | 78 (0.5%) |
| Pulmonary embolism | 60 (0.4%) |
| Failure to wean off ventilator for >48 h | 17 (0.1%) |
| Renal insufficiency | 35 (0.2%) |
| Renal failure | 11 (0.1%) |
| Urinary tract infection | 122 (0.8%) |
| Stroke | 12 (0.1%) |
| Cardiac arrest | 21 (0.1%) |
| MI | 45 (0.3%) |
| Postoperative transfusion | 1584 (10.7%) |
| DVT | 123 (0.8%) |
| Sepsis | 120 (0.8%) |
| Septic shock | 21 (0.1%) |
| Death | 35 (0.2%) |
| Extended stay (>7 d) | 664 (4.5%) |
| Return to operating room | 495 (3.4%) |
| Reoperation | 28 (0.2%) |
| Wound complications | 419 (2.8%) |
| Pulmonary complications | 95 (0.6%) |
| Cardiac complications | 61 (0.4%) |
| Renal complications | 45 (0.3%) |
| Thromboembolic complications | 184 (1.2%) |
| Sepsis complications | 137 (0.9%) |
Unadjusted and Adjusted Odds Ratios for Operative Time
| Complication | Continuous at 15-minute intervals |
|---|
|
|---|
| Unadjusted OR | Adjusted OR |
|---|
|
|
|---|
| P | OR | 95% CI | P | OR | 95% CI |
|---|
| Wound complications | .002 | 1.028 | 1.004–1.053 | .020 | 1.023 | 1.012–1.049 |
| Pulmonary complications | .157 | 1.036 | 0.986–1.089 | | | |
| Cardiac complications | .372 | 1.029 | 0.967–1.094 | | | |
| Renal complications | .700 | 1.015 | 0.941–1.095 | | | |
| Thromboembolic complications | .121 | 1.029 | 0.992–1.067 | | | |
| Sepsis complications | .313 | 1.023 | 0.979–1.070 | | | |
| Postoperative transfusion | <.001 | 1.177 | 1.163–1.193 | <.001 | 1.169 | 1.152–1.185 |
| Urinary tract infection | .546 | 1.014 | 0.969–1.061 | | | |
| Death | .378 | 1.037 | 0.956–1.126 | | | |
| Extended stay (>7 d) | <.001 | 1.056 | 1.036–1.076 | <.001 | 1.060 | 1.037–1.083 |
| Return to operating room | .002 | 1.036 | 1.013–1.059 | .162 | 1.018 | 0.993–1.043 |