Table 2 Reproducibility of database studies in arthroplasty: a comparative analysis of original and reanalyzed data from ACS-NSQIP database

1

Duchman et al., 2015

Journal of Bone and Joint Surgery

Current smokers have an increased risk of wound complications and both current and former smokers have an increased risk of total complications following total hip or total knee arthroplasty. Increasing pack-year history of smoking resulted in increasing total complication risk.

Current smokers have an increased risk of wound complications but did not have an increased risk of total complications following total hip or total knee arthroplasty. Former smokers have a reduced risk of total complications. There is no significant association between increasing pack-year history of smoking and total complications.

2

Liodakis et al., 2015

The Journal of Arthroplasty

Patients who underwent revision hip arthroplasty (RHA) had more perioperative complications, longer operative time, more blood transfusions, and longer hospital stay compared to those who underwent revision knee arthroplasty (RKA). The strongest modifiable risk factor for major complications and prolonged hospital stay after RHA and RKA was low preoperative hematocrit.

The reanalyzed data findings and conclusions were comparable to the original manuscript.

3

Keswani et al., 2015

The Journal of Arthroplasty

Multivariate analysis revealed risk factors for readmission after revision hip and knee arthroplasty including male sex, pulmonary disease, severe adverse event before discharge, stroke, cardiac disease, and American Society of Anesthesiologists (ASA) class 3 or 4. Independent predictors of extended length of stay included infection or fracture etiology relative to mechanical loosening etiology, dependent functional status, BMI ≥ 40 kg/m², history of smoking, diabetes, cardiac disease, stroke, bleeding disorders, wound class 3 or 4, and ASA class 3 or 4.

Pulmonary disease is not a significant predictor for readmission after revision hip and knee arthroplasty. Fracture etiology relative to mechanical loosening etiology, history of smoking, and bleeding disorders were not significant predictors for extended length of stay in the revision knee arthroplasty group. Results were otherwise similar to the original manuscript.

4

Courtney et al., 2016

The Journal of Arthroplasty

Patients age ≥ 70 years, those with a cardiac history, smoking history, malnutrition, or diabetes have greater risk for readmission and complications after total joint arthroplasty (TJA). Outpatient TJA alone did not increase risk of readmission or reoperation, and it was a negative independent risk factor for postoperative complications.

Cardiac history is not a significant risk factor associated with readmission and complications after TJA. Results were otherwise similar to the original manuscript.

5

Sher et al., 2016

The Journal of Arthroplasty

Patients discharged within 24 h after total joint arthroplasty (TJA) were more likely to be younger, male sex, American Society of Anesthesiologists (ASA) class 1 or 2, and less likely to be taking steroids or have comorbidities. Multivariate analysis revealed independent predictors for adverse events or readmission including age ≥ 80 years, smoking, bleeding disorders, ASA class 3 or 4, and experiencing severe adverse events (SAE) prior to discharge.

Smoking, ASA class 3 or 4, and experiencing SAE prior to discharge were not significant predictors for adverse events or readmission after TJA. Results were otherwise similar to the original manuscript.

6

Bedard et al., 2018

The Journal of Arthroplasty

Multivariate analysis showed that current smokers have an increased risk of deep infection and reoperation after revision total hip arthroplasty (THA). Smoking status additionally had no effect on wound complications after revision THA.

Current smokers did not have an increased risk of deep infection or reoperation after revision THA. Results were otherwise similar to the original manuscript.

7

Bedard et al., 2018

The Journal of Arthroplasty

Multivariate analysis showed that current smokers have an increased risk of any wound complication and deep infection after revision total knee arthroplasty (TKA). Smoking status additionally had no effect on reoperation after revision TKA.

Current smokers have an increased risk of any wound complication; however, current smokers did not have an increased risk of deep infection after revision TKA. Results were otherwise similar to the original manuscript.

8

Sahota et al., 2018

The Journal of Arthroplasty

Smokers in the combined total hip and knee arthroplasty cohort had higher rates of readmission and deep surgical site infection compared to non-smokers. Smokers in the combined cohort were also more likely to have a surgical complication compared to non-smokers.

The reanalyzed data findings and conclusions were similar to the original manuscript.

9

Johnson et al., 2019

The Journal of Arthroplasty

Increasing age, obesity, smoking, diabetes, chronic obstructive pulmonary disease (COPD), hypertension, bleeding disorders, corticosteroid use, and dependent functional status conferred an increased risk of discharge after 24 h following total knee arthroplasty (TKA). Male gender, spinal anesthesia, and monitored anesthesia care were protective against length of stay greater than 24 h.

The reanalyzed data findings and conclusions were similar to the original manuscript.

10

Agrawal et al., 2021

The Journal of Arthroplasty

Multivariate analysis revealed smokers have an increase in pulmonary and infectious complications and longer hospital stays compared to non-smokers.

Smokers do not have an increase in pulmonary complications compared to non-smokers. Results were otherwise similar to the original manuscript.

11

Heckmann et al., 2021

Orthopedics

Risk of total complications or thrombotic events is not accentuated in smokers who underwent total joint arthroplasty. Regardless of pack-year exposure, smokers have increased risk of readmission and wound complications.

The reanalyzed data findings and conclusions were similar to the original manuscript.