Results of Database Studies in Spine Surgery Can Be Influenced by Missing Data

Incidence, Risk Factors, and Outcomes of Incidental Durotomy during Lumbar Spine Decompression with or without Fusion

STUDY DESIGN

Retrospective cohort study.

PURPOSE

The primary objective of this study was to determine the incidence and risk factors for incidental durotomies during lumbar decompression surgeries. In addition, we aimed to determine the changes in patient-reported outcome measures (PROMs) based on incidental durotomy status.

OVERVIEW OF LITERATURE

There is limited literature investigating the affect of incidental durotomy on patient reported outcome measures. While the majority of research does not suggest differences in complications, readmission, or revision rates, many studies rely on public databases, and their sensitivity and specificity for identifying incidental durotomies is unknown.

METHODS

Patients undergoing lumbar decompression with or without fusion at a single tertiary care center were grouped based on the presence of a durotomy. Multivariate analysis was performed for length of stay (LOS), hospital readmissions, and changes in PROMs. To identify surgical risk factors for durotomy, 3:1 propensity matching was performed using stepwise logistic regression. The sensitivity and specificity of the International Classification of Disease, 10th revision (ICD-10) codes (G96.11 and G97.41) were also assessed.

RESULTS

Of the 3,684 consecutive patients who underwent lumbar decompressions, 533 (14.5%) had durotomies, and a complete set of PROMs (preoperative and 1-year postoperative) were available for 737 patients (20.0%). Incidental durotomy was an independent predictor of increased LOS but not hospital readmission or worse PROMs. The durotomy repair method was not associated with hospital readmission or LOS. However, repair with collagen graft and suture predicted reduced improvement in Visual Analog Scale back (β =2.56, p=0.004). Independent risk factors for incidental durotomies included revisions (odds ratio [OR], 1.73; p<0.001), levels decompressed (OR, 1.11; p=0.005), and preoperative diagnosis of spondylolisthesis or thoracolumbar kyphosis. The sensitivity and specificity of ICD-10 codes were 5.4% and 99.9%, respectively, for identifying durotomies.

CONCLUSIONS

The durotomy rate for lumbar decompressions was 14.5%. No differences in outcomes were detected except for increased LOS. Database studies relying on ICD codes should be interpreted with caution due to the limited sensitivity in identifying incidental durotomies.

Injury Profiles of Police Recruits Undergoing Basic Physical Training: A Prospective Cohort Study

Purpose A lack of published epidemiological data among police recruits presents a major challenge when designing appropriate prevention programs to reduce injury burden. We aimed to report the injury epidemiology of Western Australian (WA) Police Force recruits and examine sex and age as injury risk factors. Methods Retrospective analyses were conducted of prospectively collected injury data from WA Police Force recruits between 2018-2021. Injury was defined as 'time-loss' and injury incidence rate per 1000 training days (Poisson exact 95% confidence intervals) was calculated. For each region and type of injury, the incidence, severity, and burden were calculated. The association between age, sex, and injury occurrence were assessed using Cox regression time-to-event analysis. Results A total of 1316 WA Police Force recruits were included, of whom 264 recruits sustained 304 injuries. Injury prevalence was 20.1% and the incidence rate was 2.00 (95%CI 1.78-2.24) injuries per 1000 training days. Lower limb injuries accounted for most of the injury burden. Ligament/ joint injuries had the highest injury tissue/pathology burden. The most common activity injuring recruits was physical training (31.8% of all injuries). Older age (Hazard Ratio = 1.5, 95%CI = 1.2 to 1.9, p = 0.002) and female sex (Hazard Ratio = 1.4, 95%CI = 1.3 to 1.6, p < 0.001) increased risk of injury. Conclusion Prevention programs targeting muscle/tendon and ligament/joint injuries to the lower limb and shoulder should be prioritised to reduce the WA Police Force injury burden. Injury prevention programs should also prioritise recruits who are over 30 years of age or of female sex, given they are a higher risk population.

Impact of Prolonged Operative Duration on Postoperative Symptomatic Venous Thromboembolic Events After Thoracolumbar Spine Surgery

OBJECTIVE

To determine the effect of operative duration on the rate of postoperative symptomatic venous thromboembolic (VTE) events in patients undergoing thoracolumbar spine fusion.

METHODS

We identified all thoracolumbar spine fusion patients between 2012 and 2021. Operative duration was defined as time from skin incision to skin closure. A 1:1 propensity match was conducted incorporating patient and surgical characteristics. Logistic regression was performed to assess predictors of postoperative symptomatic VTE events. A receiver operating characteristic curve was created to determine a cutoff time for increased likelihood of VTE.

RESULTS

We identified 101 patients with VTE and 1108 patients without VTE. Seventy-five patients with VTE were matched to 75 patients without VTE. Operative duration (339 vs. 262 minutes, P = 0.010) and length of stay (5.00 vs. 3.54 days, P = 0.008) were significantly longer in patients with a VTE event. Operative duration was an independent predictor of VTE on multivariate regression (odds ratio: 1.003, 95% confidence interval: 1.001-1.01, P = 0.021). For each additional hour of operative duration, the risk of VTE increased by 18%. A cutoff time of 218 minutes was identified (area under the curve [95% confidence interval] = 0.622 [0.533-0.712]) as an optimal predictor of increased risk for a VTE event.

CONCLUSIONS

Operative duration significantly predicted symptomatic VTE, especially after surgical time cutoff of 218 minutes. Each additional hour of operative duration was found to increase VTE risk by 18%. We also identify the impact of VTE on 90-day readmission rates, suggesting significantly higher costs and opportunity for hospital acquired conditions, in line with prior literature.

Impact of Missing Data on Identifying Risk Factors for Postoperative Complications in Hand Surgery

BACKGROUND

The mismanagement of missing data in large clinical databases may lead to inaccurate findings. The purpose of this study was to demonstrate the effects of missing data on hand surgery research findings using an analysis of postoperative morbidity in patients undergoing hospital-based hand surgery.

METHODS

The National Surgical Quality Improvement Program database was queried for patients undergoing common hand and upper extremity surgery between 2011 and 2016. Major and minor postoperative complications were identified. Demographics, comorbidity, and preoperative laboratory values were identified, and the percentage missing of each was tabulated. To demonstrate how missing data can alter analysis results, these variables were evaluated for an association with major complications using multivariable regression on 3 separate cohorts: (1) all patients; (2) all patients after exclusion of any patient entry with >10% of missing data; and (3) after removal of any patient entry with any missing data.

RESULTS

Groups 1, 2, and 3 had 48 370, 23 118, and 6280 patients, respectively. There were 14 variables associated with increased odds of major complications in group 1, yet only 10 and 9 variables for groups 2 and 3, respectively. Six variables were associated with increased major complications across all 3 groups, whereas only 1 was associated with decreased odds of major complications across all groups.

CONCLUSIONS

Filtering patient cohorts according to the amount of missing patient information affected analyses of predictors for major complications associated with hospital-based hand surgery. These findings highlight the importance of considering and addressing missing data in large database studies.

Trends in Single-Level Lumbar Fusions Over the Past Decade Using a National Database

OBJECTIVE

To compare rates of different fusion techniques using a nationwide database over the last decade and identify differences in complications and readmissions based on fusion technique.

METHODS

All elective, single-level lumbar fusions performed by orthopaedic surgeons from 2011 to 2020 were identified from the American College of Surgeons National Surgical Quality Improvement Program. Rates of lumbar fusion technique posterolateral decompression and fusion [PLDF], combined transforaminal lumbar interbody fusion and PLDF, anterior lumbar or lateral lumbar interbody fusion [ALIF/LLIF], and combined ALIF/LLIF and PLDF were recorded, and 30-day complications and readmissions were compared. Secondary analysis included multiple logistic regression to determine independent predictors of each outcome.

RESULTS

Inclusion criteria were met by 28,413 fusions: 8749 (30.8%) PLDFs, 11,973 (42.1%) transforaminal lumbar interbody fusions, 4769 (16.8%) ALIF/LLIFs, and 2922 (10.3%) combined ALIF/LLIF and PLDFs. The number of fusions increased over time with 1227 fusions performed in 2011 and 3958 fusions performed in 2019. Interbody fusions also increased over time with a subsequent decrease in PLDFs (39.0% in 2011, 25.2% in 2020). Patients were more likely to be discharged home over the course of the decade (85.4% in 2011, 95.0% in 2020). No difference was observed between the techniques regarding complications or readmissions. The modified 5-item frailty index was predictive of complications (odds ratio, 2.05; P = 0.001) and readmissions (odds ratio, 2.61; P < 0.001).

CONCLUSIONS

Lumbar fusions have continued to increase over the last decade with an increasing proportion of interbody fusions. Complications and readmissions appear to be driven by patient comorbidity and not fusion technique.

Impact of a same day admission project in reducing the preoperative bed occupancy demand in a pediatric inpatient hospital

Background

A same day admission approach was established for pediatric patients undergoing elective surgery owing to an increase in demand for bed availability and the need for medical, logistical, psychological, and fiscal improvements. This study aimed to assess the effectiveness of the same day admission approach for reducing demand for preoperative bed occupancy in pediatric inpatient units.

Method

Data on elective surgery patients considered for same day admission were prospectively collected in an Excel spreadsheet.

Results

Same day admission patients numbered 269 (25.87%; n = 1040), 461 (41.7%; n = 1104), 382 (38.67%; n = 998), and 560 (44.20%; n = 1267) in 2018, 2019, 2020, and 2021, respectively. Over the 4-year period between 2018 and 2021, pediatric orthopedic surgeries accounted for the majority of same day admissions (29.72%; n = 497), followed by ear, nose, and throat (21.30%; n = 356), general (16.99%; n = 284), plastic (14.53%; n = 243); urology (9.87%; n = 165); optometry and ophthalmology (3.77%; n = 63); neuro (2.51%; n = 42), and dental (1.31%; n = 22) surgeries. The total number of days of saved preoperative beds over the 4-year period was 1672 days (an average of 418 hospital days per year).

Conclusions

This study showed that same day admission approach should be implemented in pediatrics institutions to reduce hospital bed demand. The implementation of this initiative is widely variable between specialties due to interlinked medical, operational, and logistical factors.

Level of Evidence

III.

•

Same day admission is a model to reduce the average length of hospital stay for surgical patients and to reduce costs.

•

The study is confirming the feasibility of applying the same day admissions' project in the pediatric population.

•

This study compares the applicability of the same day admission’s project across pediatrics’ surgical subspecialties.

•

This study proves the possibility of applying same day admission's project in the local region.

Patient-Related Risk Factors Associated With Surgical Site Complications After Elective Hand Surgery

BACKGROUND

The purpose of this study was to identify independent risk factors associated with an increased rate of surgical site complications after elective hand surgery.

METHODS

This study is a retrospective review of all patients who underwent elective hand, wrist, forearm, and elbow surgery over a 10-year period at a single institution. Electronic medical records were reviewed, and information regarding patient demographics, past medical and social history, perioperative laboratory values, procedures performed, and surgical complications was collected. Surgical site complications included surgical site infections, seromas or hematomas, and delayed wound healing or wound dehiscence. A univariate analysis was then performed to identify potential risk factors, which were then included in a multivariate regression analysis.

RESULTS

A total of 3261 patients who underwent elective hand surgery and met the above inclusion and exclusion criteria were included in this study. The mean age was 57 years, with 65% female and 35% male patients. The overall surgical complication rate was 2.2%. Univariate analysis of patient factors identified male sex; number of procedures >1; history of drug, alcohol, or smoking use; American Society of Anesthesiologists (ASA) class III and IV; and serum albumin <3.5 mg/dL to be significantly associated with complications. However, multivariate regression analysis identified that only ASA class III and IV (odds ratio = 3.27) was significantly associated with surgical complications.

CONCLUSIONS

Patients classified as ASA class III or IV were identified to be at a significantly increased risk of complications following elective hand surgery. Health factors which triage patients into these 2 groups may represent potentially modifiable factors to mitigate perioperative risk in the elective hand surgery population.

Increased risk of postoperative wound complications among obesity classes II & III after ALIF in 10-year ACS-NSQIP analysis of 10,934 cases

BACKGROUND CONTEXT

Anterior lumbar interbody fusion (ALIF) procedures for lumbar spine disease have been increasing amid a growing obese patient population with limited studies available focusing exclusively on risk-factors for post-operative ALIF complications.

PURPOSE

The objective of this study was to compare 30-day post-operative complications among different obesity World Health Organization classes according to body mass index (BMI) in comparison to non-obese patients who underwent an ALIF procedure.

STUDY DESIGN/SETTING

Retrospective cohort study of the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) from 2009 to 2019.

PATIENT SAMPLE

A total of 10,934 patients undergoing an ALIF.

OUTCOME MEASURES

Primary outcome measures include 30 day cardiac, pulmonary, urinary, infectious, and wound complications. Secondary outcomes included rates of blood transfusion, reintubation, deep vein thrombosis, pulmonary embolism, 30-day return to the operating room (OR), and 30 day mortality.

METHODS

Patients were identified by use of the current procedural terminology codes 22558 and 22585 from 2009 to 2019. Patients were divided into the following groups: non-obese (BMI 18.5-29.9 kg/m²), Obese I (BMI 30-34.9 kg/m²), Obese II (BMI 35-39.9 kg/m²), and Obese III (BMI ≥40 kg/m²). Age, gender, race, American Society of Anesthesiologists status, smoking status, hypertension requiring medication, steroid used, chronic obstructive pulmonary disease, history of a bleeding disorder, and diabetes was identified as risk factors after a univariate analysis conducted for demographic variables and pre-operative comorbidities. A multivariate logistic regression analysis was then performed to adjust for these preoperative risk factors and compare obesity classes I-III to non-obese patients.

RESULTS

Obesity classes II and III had a significant odds ratio (OR) for superficial infection (OR:2.7, 95%CI(1.7-4.5); OR:2.8, 95%CI(1.5-5.2) respectively), organ space infection (OR:3.8, 95%CI(1.6-7.4); OR:3.2, 95%CI(1.1-9.9) respectively), wound disruption (OR:2.8, 95%CI(1.1-7.4); OR:4.6, 95%CI(1.6-13.6) respectively), and total wound complication (OR:2.6, 95%CI(1.8-3.9); OR:3.4, 95%CI(2.2-5.4) respectively) following a multivariate logistic regression analysis.

CONCLUSIONS

Risk for post-operative wound complications following an ALIF were found to be significantly higher for obesity classes II-III in comparison to non-obese patients. These findings can further support the use of additional wound care in the perioperative setting for certain levels of obesity.

Association of Anesthesia Type with Postoperative Outcome and Complications in Patients Undergoing Revision Total Knee Arthroplasty

Revision total knee arthroplasty (TKA) is an increasingly common procedure and is effective in treating knee osteoarthritis, but it has higher complication rates than primary TKA. Anesthetic choice poses perioperative risk that has been extensively studied in primary TKA, showing favorable results for regional anesthesia compared with general anesthesia. The impact of anesthetic choice in revision TKAs is not well studied. A retrospective cohort study was conducted using the American College of Surgeons National Surgical Quality Improvement Program database. Patients who underwent revision TKAs between 2014 and 2017 were divided into three anesthesia cohorts: (1) general anesthesia, (2) regional anesthesia, and (3) combined general-regional anesthesia. Univariate and multivariate analyses were used to analyze patient characteristics and 30-day postoperative outcomes. Bonferroni correction was applied for post hoc analysis. In total, 8,820 patients were identified. Of whom, 3,192 patients underwent general anesthesia, 3,474 patients underwent regional anesthesia, and 2,154 patients underwent combined anesthesia. After multivariate analyses, regional anesthesia was associated with decreased odds for any complication (p = 0.008), perioperative blood transfusion (p < 0.001), and extended length of stay (p < 0.001) compared with general anesthesia. In addition, regional anesthesia was associated with decreased odds for perioperative blood transfusion (p < 0.001) and extended length of stay (p = 0.006) compared with combined anesthesia. However, following multivariate analysis, regional anesthesia was not associated with decreased odds of wound, pulmonary, renal, urinary tract, thromboembolic, and cardiac complications, and was not associated with return to operating room, extended length of stay, minor and major complications, and mortality. Retrospective analysis of a large surgical database suggests that patients receiving general anesthesia have increased likelihood for developing adverse postoperative outcomes relative to patients receiving regional anesthesia. Prospective and controlled trials should be conducted to verify these findings.

Age as a Risk Factor for Complications Following Anterior Cervical Discectomy and Fusion: Analysis From the Michigan Spine Surgery Improvement Collaborative (MSSIC)

STUDY DESIGN

Retrospective analysis of prospectively collected registry data using multivariable analyses of imputed data.

OBJECTIVE

We sought to demonstrate that age would not be associated with complications in patients undergoing anterior cervical discectomy and fusion (ACDF).

SUMMARY OF BACKGROUND DATA

Elderly patients (≥70 yrs) undergoing ACDF are considered a higher risk for complications. However, conclusive evidence is lacking. The Michigan Spine Surgery Improvement Collaborative (MSSIC) is a quality improvement collaborative with 30 hospitals across Michigan.

METHODS

The study included all patients who had 1 to 4 level ACDF (September 2015-August 2019) for 90-day complications. Major and minor complications were defined using a validated classification. Multiple imputations were used to generate complete covariate datasets. Generalized estimating equation model was used to identify associations with complications using the whole cohort and elderly subgroup analyses. Bonferroni correction was used.

RESULTS

Nine thousand one hundred thirty five patients (11.1% ≥ 70 yrs and 88.9% <70 yrs) with 2266 complications were analyzed. Comparing elderly versus non-elderly, the elderly had a significantly higher rate of any complications (31.5% vs. 24.0%, P < 0.001) and major complications (14.1% vs. 7.0%, P < 0.001). On multivariable analysis, age was not independently associated with any complication. POD#0 ambulation and preop independent ambulation were independently associated with significantly decreased odds of any complication. In the elderly, independent preoperative ambulation was protective for any complication (odds ratio [OR] 0.53, 0.39-0.73 95% confidence interval [CI]), especially major complications (OR 0.41, 0.27-0.61 95% CI).

CONCLUSION

Age was not an independent risk factor for complications in patients that underwent ACDF. In the elderly, independent preoperative ambulation was especially protective for major complications.Level of Evidence: 3.

Early Failures After Lumbar Discectomy Surgery: An Analysis of 62 690 Patients

STUDY DESIGN

Retrospective cohort study.

OBJECTIVE

To determine the rate of early failures (readmission or reoperation for new or recurrent pain/neurological symptoms) within 30 days after lumbar discectomy and identify associated risk factors.

METHODS

A retrospective cohort study was conducted of patients undergoing lumbar discectomy in the National Surgical Quality Improvement Program database between 2013 and 2017. Rates of readmission for new or recurrent symptoms or reoperation for revision discectomy or fusion within 30 days postoperatively were measured and correlated with risk factors.

RESULTS

In total 62 690 patients were identified; overall rate of readmission within 30 days was 3.3%, including 1.2% for pain or neurological symptoms. Populations at increased risk of readmission were those with 3 or more levels of treatment (2.0%, odds ratio [OR] 2.8%, P < .01), age >70 years (1.8%, OR 1.6, P < .01), class 3 obesity (1.5%, OR 1.4, P = .04), and female gender (1.4%, OR 1.2, P = .02). The overall rate of reoperation within 30 days was 2.2%, including 1.2% for revision decompression or lumbar fusion surgery. Populations at increased risk of reoperation were revision discectomies (1.4%, OR 1.7, P < .01) and females (1.1%, OR 1.4, P < 0.01). Extraforaminal discectomies were associated with lower rates of readmission (0.7%, OR 0.6, P = 0.02) and reoperation (0.4%, OR 0.4, P = .01).

CONCLUSIONS

Early failures after lumbar discectomy surgery are rare. However, certain subpopulations are associated with increased rates of early failure: obesity, multilevel surgery, females, and revision discectomies.

An Assessment of Nonoperative Management Strategies in a Herniated Lumbar Disc Population: Successes Versus Failures

STUDY DESIGN

Retrospective cohort study.

OBJECTIVE

To compare the utilization of conservative treatments in patients with lumbar intervertebral disc herniations who were successfully managed nonoperatively versus patients who failed conservative therapies and elected to undergo surgery (microdiscectomy).

METHODS

Clinical records from adult patients with an initial herniated lumbar disc between 2007 and 2017 were selected from a large insurance database. Patients were divided into 2 cohorts: patients treated successfully with nonoperative therapies and patients that failed conservative management and opted for microdiscectomy surgery. Nonoperative treatments utilized by the 2 groups were collected over a 2-year surveillance window. "Utilization" was defined by cost billed to patients, prescriptions written, and number of units disbursed.

RESULTS

A total of 277 941 patients with lumbar intervertebral disc herniations were included. Of these, 269 713 (97.0%) were successfully managed with nonoperative treatments, while 8228 (3.0%) failed maximal nonoperative therapy (MNT) and underwent a lumbar microdiscectomy. MNT failures occurred more frequently in males (3.7%), and patients with a history of lumbar epidural steroid injections (4.5%) or preoperative opioid use (3.6%). In a logistic multivariate regression analysis, male sex and utilization of opioids were independent predictors of conservative management failure. Furthermore, a cost analysis indicated that patients who failed nonoperative treatments billed for nearly double ($1718/patient) compared to patients who were successfully treated ($906/patient).

CONCLUSION

Our results suggest that the majority of patients are successfully managed nonoperatively. However, in the subset of patients that fail conservative management, male sex and prior opioid use appear to be independent predictors of treatment failure.

Big data registries in spine surgery research: the lurking dangers

Spine surgery research has improved considerably over the last few decades. Its' most recent growth is in large part due to the mounting increase in studies conducted using national databases and registries. With easy access to a large number of patients, the benefit of these registries has become evident. However, as with any research, this type of data must be used responsibly with the appropriate strengths and limitations kept in mind. Inappropriate use of these registries continues to be a growing concern as potentially false or inaccurate conclusions can adversely impact clinical practice. It is, therefore, the author and the readers' responsibility to acknowledge and understand the limitations of this type of data. Knowledge of methodological requirements in the use and analyses of registry data is essential to ensuring quality evidence with proper interpretation.

Age as a Predictor for Complications and Patient-reported Outcomes in Multilevel Transforaminal Lumbar Interbody Fusions: Analyses From the Michigan Spine Surgery Improvement Collaborative (MSSIC)

STUDY DESIGN

Retrospective review of a multi-institutional data registry.

OBJECTIVE

The authors sought to determine the association between age and complications & patient-reported outcomes (PRO) in patients undergoing multilevel transforaminal interbody lumbar fusion (MTLIF).

SUMMARY OF BACKGROUND DATA

Elderly patients undergoing MTLIF are considered high risk. However, data on complications and PRO are lacking. Additionally, safety of multilevel lumbar fusion in the elderly remains uncertain.

METHODS

Patients ≥50-year-old who underwent MTLIF for degenerative lumbar spine conditions were analyzed. Ninety-day complications and PROs (baseline, 90-d, 1-y, 2-y) were queried using the MSSIC database. PROs were measured by back & leg visual analog scale (VAS), Patient-reported Outcomes Measurement Information System (PROMIS), EuroQol-5D (EQ-5D), and North American Spine Society (NASS) Patient Satisfaction Index. Univariate analyses were used to compare among elderly and complication cohorts. Generalized estimating equation (GEE) was used to identify predictors of complications and PROs.

RESULTS

A total of 3120 patients analyzed with 961 (31%) ≥ 70-y-o and 2159 (69%) between 50-69. A higher proportion of elderly experienced postoperative complications (P = .003) including urinary retention (P = <.001) and urinary tract infection (P = .002). Multivariate analysis demonstrated that age was not independently associated with complications. Number of operative levels was associated with any (P = .001) and minor (P = .002) complication. Incurring a complication was independently associated with worse leg VAS and PROMIS scores (P = <.001). Preoperative independent ambulation was independently associated with improved PROMIS, and EQ5D (P = <.001). Within the elderly, preoperative independent ambulation and lower BMI were associated with improved PROMIS (P = <.001). Complications had no significant effect on PROs in the elderly.

CONCLUSIONS

Age was not associated with complications nor predictive of functional outcomes in patients who underwent MTLIF. Age alone, therefore, may not be an appropriate surrogate for risk. Furthermore, baseline preoperative independent ambulation was associated with better clinical outcomes and should be considered during preoperative surgical counseling.Level of Evidence: 3.

Substantial Loss to Follow-Up and Missing Data in National Arthroscopy Registries: A Systematic Review

PURPOSE

To report follow-up methodologies, compliance, and existing strategies for handling missing data in national arthroscopy registries collecting patient-reported outcome measures (PROMs).

METHODS

Annual reports, EMBASE, and MEDLINE were queried following the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines to identify national arthroscopy registries reporting follow-up with a validated PROM and sample size greater than 500. Extracted data included weighted compliance in peer-reviewed publications, cumulative compliance throughout the time span of data collection, and missing-data methodologies.

RESULTS

Nine national arthroscopy registries currently collect PROMs, with cumulative rates of follow-up ranging from less than 10% to more than 70%. We identified 36 publications from 5 national registries reporting hip and knee arthroscopies. The weighted mean compliance with PROMs in national registry publications was 56% at 0.5 years, 44% to 59% at 1 year, 40% to 61% at 2 years, 35% to 54% at 5 years, and 40% at 10 years. A missing-data analysis was reported or referenced in 58% of publications.

CONCLUSIONS

In national arthroscopy registries, compliance with 2-year PROMs does not meet traditional follow-up thresholds of 60% or 80% and reporting of missing-data methodologies is inconsistent.

LEVEL OF EVIDENCE

Level IV, systematic review of Level II through IV studies.

Perioperative Outcomes in 17,947 Patients Undergoing 2-Level Anterior Cervical Discectomy and Fusion Versus 1-Level Anterior Cervical Corpectomy for Treatment of Cervical Degenerative Conditions: A Propensity Score Matched National Surgical Quality Improvement Program Analysis

Objective

To compare the perioperative morbidity of 2-level anterior cervical discectomy and fusion (ACDF) with that of 1-level anterior cervical corpectomy and fusion (ACCF) for the treatment of cervical degenerative conditions.

Methods

A retrospective study of the 2005–2016 National Surgical Quality Improvement Program database for patients undergoing 2-level ACDF and 1-level ACCF was performed. Patient data included: age, sex, body mass index (BMI), functional status, and American Society of Anesthesiologists (ASA) physical status (PS) classification. Hospital data included: operative time and length of hospital stay (LOS). Thirty-day outcome data included: any, serious, and minor adverse events, return to the operating room, readmission, and mortality. After propensity matching for age, sex, ASA PS classification, functional status, and BMI, multivariate logistic regression analysis was used to compare outcomes between the 2 propensity-matched subcohorts. Finally, multivariate logistic regression that additionally controlled for operative time was performed to compare the 2 propensity-matched subcohorts.

Results

A total of 17,497 cases were identified, with 90.20% undergoing 2-level ACDF and 9.80% undergoing 1-level ACCF. Patients undergoing 2-level ACDF were younger, more likely to be female, had higher functional status, and had shorter operative time and LOS (p < 0.001). After propensity score matching, cases undergoing 1-level ACCF had a statistically significant higher rate of serious adverse events (p = 0.005). This difference was no longer significant after controlling for operative time.

Conclusion

While there was noted to be additional morbidity in 1-level ACCF cases relative to 2-level ACDF cases, the lack of difference once controlling for the surgical time supports using the procedure that best accomplishes the surgical objectives.

Comparing short-term AIS post-operative complications between ACS-NSQIP and a surgeon study group

STUDY DESIGN

Prospective cohort review.

OBJECTIVE

To compare two AIS databases to determine if a performance improvement-based surgeon group has different outcomes compared to a national database. The American College of Surgeon's National Quality Improvement Program (ACS-NSQIP) and a surgeon study group (SG) collect prospective data on AIS surgery outcomes. NSQIP offers open enrollment to all institutions, and SG membership is limited to 15 high-volume institutions, with a major initiative to improve surgeon performance. While both provide important outcome benchmarks, they may reflect outcomes that are not relatable nationwide.

METHODS

The ASC-NSQIP Pediatric Spine Fusion and SG database were queried for AIS 30- and 90-day complication data for 2014 and 2015. Prospective enrollment and a dedicated site coordinator with rigorous data quality assurance protocols existed for both registries. Outcomes were compared between groups with respect to superficial and deep surgical site infections (SSI), neurologic injury, readmission, and reoperation.

RESULTS

There were a total of 2927 AIS patients included in the ASC-NSQIP data and 721 in the SG database. Total complication rate was 9.4% NSQIP and 3.6% SG. At 90 days, there were fewer surgical site infections reported by SG than ASC-NSQIP (0.6% vs. 1.6%, p = 0.03). Similarly, there were less spinal cord injuries (0.8% vs 1.5%, p = 0.006), 30-day readmissions (0.8% vs. 2.6%, p = 0.002), and 30-day reoperations (0.6% vs. 1.7%, p = 0.02) in the SG cohort.

CONCLUSIONS

Comparison of these two data sets suggests a range of complications and readmission rates, with the SG demonstrating lower values. These results are likely multi-factorial with the performance improvement initiative of the SG playing a role. Understanding the rate and ultimate risk factors for readmission and complications from big data sources has the potential to further drive quality improvement.

LEVEL OF EVIDENCE

III.

Policies Restricting Overlapping Surgeries Negatively Impact Access to Care, Clinical Efficiency and Hospital Revenue: A Forecasting Model for Surgical Scheduling

OBJECTIVE

To model the financial impact of policies governing the scheduling of overlapping surgeries, and to identify optimal solutions that maximize operating efficiency that satisfy the fiduciary duty to patients.

BACKGROUND

Hospitals depend on procedural revenue to maintain financial health as the recent pandemic has revealed. Proposed policies governing the scheduling of overlapping surgeries may dramatically impact hospital revenue. To date, the potential financial impact has not been modeled.

METHODS

A linear forecasting model based on a logic matrix decision tree enabled an analysis of surgeon productivity annualized over a fiscal year. The model applies procedural and operational variables to policy constraints limiting surgical scheduling. Model outputs included case and financial metrics modeled over 1000-surgeon-year simulations. Case metrics included annual case volume, case mix, operating room (OR) utilization, surgeon utilization, idle time and staff overtime hours. Financial outputs included annual revenue, expenses and contribution margin.

RESULTS

The model was validated against surgical data. Case and financial metrics decreased as a function of increasingly restrictive scheduling scenarios, with the greatest contribution margin loses ($1,650,000 per surgeon-year) realized with the introduction of policies mandating that a second patient could not enter the OR until the critical portion of the first surgery was completed. We identify an optimal scheduling scenario that maximizes surgeon efficiency, minimizes OR idle time and revenue loses, and satisfies ethical obligations to patients.

CONCLUSIONS

Hospitals may expect significant financial loses with the introduction of policies restricting OR scheduling. We identify an optimal solution that maximizes efficiency while satisfying ethical duty to patients. This forecast is immediately relevant to any hospital system that depends upon procedural revenue.

Preoperative High, as well as Low, Platelet Counts Correlate With Adverse Outcomes After Elective Total Hip Arthroplasty

Preoperative laboratory studies are often obtained as part of the workup for surgeries such as total hip arthroplasty (THA). An increasing need exists to be able to identify patients at risk for adverse outcomes. Thus, metrics that correlate with postoperative adverse events and readmissions are increasingly important to optimize patient care. The implications of varying abnormal platelet counts, especially on the high end of the spectrum, have yet to be assessed in large, multicenter patient populations. This study aims to risk stratify THA patients with varying preoperative platelet counts to address these questions. The purposes of this study were to (1) evaluate cutoffs for normal versus abnormal platelet counts for patients undergoing THA by using postoperative complications data and (2) assess the correlation of such values with readmission data using the National Surgical Quality Improvement Program database.

Geographic Variations and Trends in Primary and Revision Knee and Total Hip Arthroplasties in the United States

Background:

Over 1 million joint arthroplasties are performed annually in the United States. Ideally, as devices and surgical techniques improve, the number of revision arthroplasties relative to primary arthroplasties should decrease. To our knowledge, this is the first study to evaluate state-by-state disparities in the ratio of revision to primary knee arthroplasty (unicompartmental and total) and total hip arthroplasty (THA).

Methods:

The National Inpatient Sample was used to identify patients who had undergone primary or revision knee arthroplasty or primary or revision THA from 2001 to 2011. Demographic characteristics, surgical rates, and revision ratios (the number of revision procedures divided by the number of primary procedures) were determined for the United States as a whole and by state.

Results:

During the study window, 47 states were sampled. For knee arthroplasty, 1,251,484 patients were identified: 91% underwent primary procedures and 9% underwent revision procedures. Compared with the primary knee arthroplasty cohort, the revision knee arthroplasty cohort had a younger mean age, had more male patients, and had more chronic conditions and longer hospitalizations (p < 0.001 for each). Over the years studied, the mean age of patients who had undergone primary knee arthroplasty decreased 1.8 years (p < 0.0001) and the mean age of those who had undergone revision knee arthroplasties decreased 2.4 years (p < 0.0001). The national revision ratio remained unchanged at around 0.1 (p = 0.8792). However, there was a 2.2-fold variation in revision ratio by state (revision ratio state range, 0.065 to 0.141). For THA, 614,638 patients were identified: 85% underwent primary procedures and 15% underwent revision procedures. Compared with the primary THA cohort, the revision THA cohort had an older mean age, had fewer male patients, and had more chronic conditions and longer hospitalizations (p < 0.001 for each). Over the years studied, the mean age of patients who had undergone primary THA decreased 1.5 years (p = 0.0016), whereas patients who had undergone revision had no significant age trend (p = 1.0000). Unlike for knee arthroplasty, the national THA revision ratio trended downward (0.24 evolved to 0.18, p = 0.0016), and there was a 2.1-fold variation in the revision ratio by state (revision ratio state range, 0.119 to 0.248).

Conclusions:

This study found significant variability in state-by-state revision ratios. It also found that the national revision ratio stayed relatively steady for knee arthroplasty but was decreasing for THA, and that patients who had undergone revision knee arthroplasty were getting younger, whereas patients who had undergone revision THA were not. These discrepancies suggest divergent histories for primary knee arthroplasty and THA and warrant further detailed evaluation.

Level of Evidence:

Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

Missing Data in the National Surgical Quality Improvement Program Database: How Does It Affect the Identification of Risk Factors for Shoulder Surgery Complications?

PURPOSE

The main purpose of this study was to establish whether different approaches to handling missingness affect the determination of risk factors associated with 30-day postoperative major and minor complications. A secondary purpose was to determine the frequency of missingness in the National Surgical Quality Improvement Program (NSQIP) records of patients who underwent shoulder surgery.

METHODS

We queried the American College of Surgeons NSQIP database using Current Procedural Terminology codes to identify patients who underwent shoulder surgery from 2011 to 2016 (n = 61,963). Data on major and minor postoperative complications were extracted. We also extracted data on patient characteristics, comorbidities, American Society of Anesthesiologists classifications, and preoperative laboratory values. We calculated the percentages of missingness for each variable. Each variable was then evaluated for associations with major and minor complications by using multivariable regression and 4 methods of handling missingness (involving imputation or exclusion, depending on the completeness of the data set). For 10 variables, the method using no exclusion or imputation produced higher odds of major complications compared with imputation. For 5 variables, the method using no exclusion or imputation produced higher odds of minor complications compared with imputation.

RESULTS

Only 6.5% of all patients had no missing data (n = 4,042), whereas 44% had <10% missingness (n = 27,165). Fewer variables were associated with both major and minor complications after shoulder surgery when patient records with missing data were excluded from analysis.

CONCLUSIONS

Different methods of handling missingness produced different odds ratios for some variables when determining risk factors for complications after shoulder surgery.

LEVEL OF EVIDENCE

III, Case control study.

Development of a Risk Prediction Model With Improved Clinical Utility in Elective Cervical and Lumbar Spine Surgery

STUDY DESIGN

Retrospective cohort.

OBJECTIVE

We present a universal model of risk prediction for patients undergoing elective cervical and lumbar spine surgery.

SUMMARY OF BACKGROUND DATA

Previous studies illustrate predictive risk models as possible tools to identify individuals at increased risk for postoperative complications and high resource utilization following spine surgery. Many are specific to one condition or procedure, cumbersome to calculate, or include subjective variables limiting applicability and utility.

METHODS

A retrospective cohort of 177,928 spine surgeries (lumbar (L) Ln = 129,800; cervical (C) Cn = 48,128) was constructed from the 2012 to 2016 American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP) database. Cases were identified by Current Procedural Terminology (CPT) codes for cervical fusion, lumbar fusion, and lumbar decompression laminectomy. Significant preoperative risk factors for postoperative complications were identified and included in logistic regression. Sum of odds ratios from each factor was used to develop the Universal Spine Surgery (USS) score. Model performance was assessed using receiver-operating characteristic (ROC) curves and tested on 20% of the total sample.

RESULTS

Eighteen risk factors were identified, including sixteen found to be significant outcomes predictors. At least one complication was present among 11.1% of patients, the most common of which included bleeding requiring transfusion (4.86%), surgical site infection (1.54%), and urinary tract infection (1.08%). Complication rate increased as a function of the model score and ROC area under the curve analyses demonstrated fair predictive accuracy (lumbar = 0.741; cervical = 0.776). There were no significant deviations between score development and testing datasets.

CONCLUSION

We present the Universal Spine Surgery score as a robust, easily administered, and cross-validated instrument to quickly identify spine surgery candidates at increased risk for postoperative complications and high resource utilization without need for algorithmic software. This may serve as a useful adjunct in preoperative patient counseling and perioperative resource allocation.

LEVEL OF EVIDENCE

3.

Risk Factors Associated With 90-Day Readmissions After Degenerative Lumbar Fusion: An Examination of the Michigan Spine Surgery Improvement Collaborative (MSSIC) Registry

BACKGROUND

Most studies have evaluated 30-d readmissions after lumbar fusion surgery. Evaluation of the 90-d period, however, allows a more comprehensive assessment of factors associated with readmission.

OBJECTIVE

To assess the reasons and risk factors for 90-d readmissions after lumbar fusion surgery.

METHODS

The Michigan Spine Surgery Improvement Collaborative (MSSIC) registry is a prospective, multicenter, and spine-specific database of patients surgically treated for degenerative disease. MSSIC data were retrospectively analyzed for causes of readmission, and independent risk factors impacting readmission were found by multivariate logistic regression.

RESULTS

Of 10 204 patients who underwent lumbar fusion, 915 (9.0%) were readmitted within 90 d, most commonly for pain (17%), surgical site infection (16%), and radicular symptoms (10%). Risk factors associated with increased likelihood of readmission were other race (odds ratio [OR] 1.81, confidence interval [CI] 1.22-2.69), coronary artery disease (OR 1.57, CI 1.25-1.96), ≥4 fused levels (OR 1.41, CI 1.06-1.88), diabetes (OR 1.34, CI 1.10-1.63), and surgery length (OR 1.09, CI 1.03-1.16). Factors associated with decreased risk were discharge to home (OR 0.63, CI 0.51-0.78), private insurance (OR 0.79, CI 0.65-0.97), ambulation same day of surgery (OR 0.81, CI 0.67-0.97), and spondylolisthesis diagnosis (OR 0.82, CI 0.68-0.97). Of those readmitted, 385 (42.1%) patients underwent another surgery.

CONCLUSION

Ninety-day readmission occurred in 9.0% of patients, mainly for pain, wound infection, and radicular symptoms. Increased focus on postoperative pain may decrease readmissions. Among factors impacting the likelihood of 90-d readmission, early postoperative ambulation may be most easily modifiable. Optimization of preexisting medical conditions could also potentially decrease readmission risk.

High, As Well As Low, Preoperative Platelet Counts Correlate With Adverse Outcomes After Elective Posterior Lumbar Surgery

STUDY DESIGN

Retrospective cohort study of prospectively collected data.

OBJECTIVE

Assess correlation between preoperative platelet counts and postoperative adverse events after elective posterior lumbar surgery procedures.

SUMMARY OF BACKGROUND DATA

Preoperative low platelet counts have been correlated with adverse outcomes after posterior lumbar surgery. Nonetheless, the effect of varying platelet counts has not been studied in detail for a large patient population, especially on the high end of the platelet spectrum.

METHODS

Patients who underwent elective posterior lumbar surgery were identified in the 2011 to 2016 National Surgical Quality Improvement Program database. Preoperative platelet counts were considered relative to 30-day perioperative adverse outcomes. Patients were classified into platelet categories based on determining upper and lower bounds on when the adverse outcomes crossed a relative risk of 1.5. Univariate and multivariate analyses compared 30-day postoperative complications, readmissions, operative time, and hospital length of stay between those with low, normal, and high platelet counts.

RESULTS

In total, 137,709 posterior lumbar surgery patients were identified. Using the relative risk threshold of 1.5 for the occurrence of any adverse event, patients were divided into abnormally low (≤140,000/mL) and abnormally high (≥447,000/mL) platelet cohorts. The abnormally low and high platelet groups were associated with higher rates of any, major, minor adverse events, transfusion, and longer hospital length of stay. Furthermore, the abnormally low platelet counts were associated with a higher risk of readmissions.

CONCLUSION

The data-based cut-offs for abnormally high and low platelet counts closely mirrored those found in literature. Based on these definitions, abnormally high and low preoperative platelet counts were associated with adverse outcomes after elective posterior lumbar surgery. These findings facilitate risk stratification and suggest targeted consideration for patients with high, as well as low, preoperative platelet counts.

LEVEL OF EVIDENCE

3.

Is Delayed Time to Surgery Associated with Increased Short-term Complications in Patients with Pathologic Hip Fractures?

BACKGROUND

Delayed time to surgery of at least 2 days after hospital arrival is well known to be associated with increased complications after standard hip fracture surgery; whether this association is present for pathologic hip fractures, however, is unknown.

QUESTIONS/PURPOSES

(1) After controlling for differences in patient characteristics, is delayed time to surgery (at least 2 days) for patients with pathologic hip fractures independently associated with increased complications compared with early surgery (fewer than 2 days)? (2) What preoperative factors are independently associated with major complications and mortality after surgery for pathologic hip fractures?

METHODS

A retrospective study was conducted using the American College of Surgeons National Surgical Quality Improvement Program database of pathologic hip fractures (including femoral neck, trochanteric, and subtrochanteric fractures) from 2007 to 2017. This database was chosen over other databases given the high-quality preoperative medical history and postoperative complication (including readmissions, reoperations, and mortality) data collected from patient medical records through the thirtieth postoperative day. Patients were identified using Common Procedural Terminology codes for hip fracture treatment (THA, hemiarthroplasty, proximal femur replacement, intramedullary nail, and plate and screw fixation) with associated operative diagnoses for pathologic fractures as identified with International Classification of Diseases codes. A total of 2627 patients with pathologic hip fractures were included in this study; 65% (1714) had surgery within 2 days and 35% (913) had surgery after that time. Patient demographics, hospitalization information, and 30-day postoperative complications were recorded. Differences in characteristics between patients who underwent surgery in the early and delayed time periods were assessed with chi-square tests for categorical variables and t-tests for continuous variables. Delayed-surgery patients were more medically complex at the time of admission than early-surgery patients, including having higher American Society of Anesthesiologists classification (mean ± SD 3.18 ± 0.61 versus 2.94 ± 0.60; p < 0.001) and prevalence of advanced, "disseminated" cancer (53% versus 39%; p < 0.001). Propensity-adjusted multivariable logistic regression analyses were performed to assess the effect of delayed time to surgery alone on the various outcome measures. Additional independent risk factors for major complications and mortality were identified using backwards stepwise regressions.

RESULTS

After controlling for baseline factors, the only outcome associated with delayed surgery was extended postoperative length of stay (odds ratio 1.94 [95% CI 1.62 to 2.33]; p < 0.001). Delayed surgery was not associated with any postoperative complications, including major complications (OR 1.23 [95% CI 0.94 to 1.6]; p = 0.13), pulmonary complications (OR 1.24 [95% CI 0.83 to 1.86]; p = 0.29), and mortality (OR 1.26 [95% CI 0.91 to 1.76]; p = 0.16). Histories of chronic obstructive pulmonary disease (OR 2.48), congestive heart failure (OR 2.64), and disseminated cancer (OR 1.68) were associated with an increased risk of major complications, while dependent functional status (OR 2.27), advanced American Society of Anesthesiologists class (IV+ versus I-II, OR 4.81), and disseminated cancer were associated with an increased risk of mortality (OR 2.2; p ≤ 0.002 for all).

CONCLUSIONS

After controlling for baseline patient factors, delayed time to surgery was not independently associated with increased 30-day complications after surgical treatment of pathologic hip fractures. These results are in contrast to the traditional dogma for standard hip fractures that surgery within 2 days of hospital arrival is associated with reduced complications. Although surgery should not be delayed needlessly, if the surgeon feels that additional time could benefit the patient, the results of this study suggest surgeons should not expedite surgery because of the risk of surgical delay observed for standard hip fractures.

LEVEL OF EVIDENCE

Level III, therapeutic study.

Safety and Effectiveness of Antifibrinolytics in Posterior Scoliosis Surgery for Adolescent Idiopathic Scoliosis: An Analysis of the NSQIP-Pediatric Database

STUDY DESIGN

This was a retrospective study of prospectively collected data.

OBJECTIVE

To utilize a large national database with prospectively collected data [National Surgical Quality Improvement Program Pediatric (NSQIP-Pediatric)] to study the safety and effectiveness of antifibrinolytic use during multilevel posterior spinal fusion (PSF) for adolescent idiopathic scoliosis (AIS).

SUMMARY OF BACKGROUND DATA

There is currently a lack of consensus and evidence regarding the safety and effectiveness of antifibrinolytic use for pediatric patients undergoing corrective surgery for AIS.

MATERIALS AND METHODS

Patients who underwent multilevel PSF for AIS in the 2016 NSQIP-Pediatric database were identified. Preoperative and procedural characteristics were compared between patients who received antifibrinolytics versus those who did not. Multivariate regressions were used to compare perioperative transfusion rates and postoperative outcomes, such as rate of return to the operating room, 30-day readmission, and intensive care unit and hospital length of stay between the 2 treatment groups.

RESULTS

This study included 975 patients who received antifibrinolytics and 223 patients who did not. Patients who received these agents tended to have more levels fused, osteotomies performed, and longer operative times. After controlling for these variances, there were no statistical differences in rate and volume of transfusion, rate of return to the operating room, 30-day readmission, 30-day postoperative complications, or intensive care unit or hospital length of stay between the 2 treatment groups.

CONCLUSIONS

This study did not demonstrate transfusion reduction in the group that received antifibrinolytics. This finding may be, in part, secondary to nonoptimized or nonstandardized protocols for antifibrinolytic use in pediatric deformity surgery or the inability to adequately control for selection bias, as those with greater surgical invasiveness may be more likely to receive antifibrinolytics. Nonetheless, using antifibrinolytics in this population appears safe and not associated with increased perioperative complications.

LEVEL OF EVIDENCE

Level III.

Ambulation on Postoperative Day #0 Is Associated With Decreased Morbidity and Adverse Events After Elective Lumbar Spine Surgery: Analysis From the Michigan Spine Surgery Improvement Collaborative (MSSIC)

BACKGROUND

While consistently recommended, the significance of early ambulation after surgery has not been definitively studied.

OBJECTIVE

To identify the relationship between ambulation on the day of surgery (postoperative day (POD)#0) and 90-d adverse events after lumbar surgery.

METHODS

The Michigan Spine Surgery Improvement Collaborative (MSSIC) is a prospective multicenter registry of spine surgery patients. As part of routine postoperative care, patients either ambulated on POD#0 or did not. The 90-d adverse events of length of stay (LOS), urinary retention (UR), urinary tract infection (UTI), ileus, readmission, surgical site infection (SSI), pulmonary embolism/deep vein thrombosis (PE/DVT), and disposition to a rehab facility were measured.

RESULTS

A total of 23 295 lumbar surgery patients were analyzed. POD#0 ambulation was associated with decreased LOS (relative LOS 0.83, P &lt; .001), rehab discharge (odds ratio [OR] 0.52, P &lt; .001), 30-d (OR 0.85, P = .044) and 90-d (OR 0.86, P = .014) readmission, UR (OR 0.73, P = 10), UTI (OR 73, P = .001), and ileus (OR 0.52, P &lt; .001) for all patients. Significant improvements in LOS, rehab discharge, readmission, UR, UTI, and ileus were observed in subset analysis of single-level decompressions (4698 pts), multilevel decompressions (4079 pts), single-level fusions (4846 pts), and multilevel fusions (4413 pts). No change in rate of SSI or DVT/PE was observed for patients who ambulated POD#0.

CONCLUSION

POD#0 ambulation is associated with a significantly decreased risk for several key adverse events after lumbar spine surgery. Decreasing the incidence of these outcomes would be associated with significant cost savings. As ambulation POD#0 is a modifiable factor in any patient's postoperative care following most spine surgery, it should be encouraged and incorporated into spine-related, enhanced-recovery-after-surgery programs.

The Association of Preoperative Opioid Usage With Patient-Reported Outcomes, Adverse Events, and Return to Work After Lumbar Fusion: Analysis From the Michigan Spine Surgery Improvement Collaborative (MSSIC)

BACKGROUND

It is important to delineate the relationship between opioid use and spine surgery outcomes.

OBJECTIVE

To determine the association between preoperative opioid usage and postoperative adverse events, patient satisfaction, return to work, and improvement in Oswestry Disability Index (ODI) in patients undergoing lumbar fusion procedures by using 2-yr data from a prospective spine registry.

METHODS

Preoperative opioid chronicity from 8693 lumbar fusion patients was defined as opioid-naïve (no usage), new users (&lt;6 wk), short-term users (6 wk-3 mo), intermediate-term users (3-6 mo), and chronic users (&gt;6 mo). Multivariate generalized estimating equation models were constructed.

RESULTS

All comparisons were to opioid-naïve patients. Chronic opioid users showed less satisfaction with their procedure at 90 d (Relative Risk (RR) 0.95, P = .001), 1 yr (RR 0.89, P = .001), and 2 yr (RR 0.89, P = .005). New opioid users were more likely to show improvement in ODI at 90 d (RR 1.25, P &lt; .001), 1 yr (RR 1.17, P &lt; .001), and 2 yr (RR 1.19, P = .002). Short-term opioid users were more likely to show ODI improvement at 90 d (RR 1.25, P &lt; .001). Chronic opioid users were less likely to show ODI improvement at 90 d (RR 0.90, P = .004), 1 yr (RR 0.85, P &lt; .001), and 2 yr (RR 0.80, P = .003). Chronic opioid users were less likely to return to work at 90 d (RR 0.80, P &lt; .001).

CONCLUSION

In lumbar fusion patients and when compared to opioid-naïve patients, new opioid users were more likely and chronic opioid users less likely to have improved ODI scores 2 yr after surgery. Chronic opioid users are less likely to be satisfied with their procedure 2 yr after surgery and less likely to return to work at 90 d. Preoperative opioid counseling is advised.

Reliability of the 6-minute walking test smartphone application

OBJECTIVE

Objective functional measures such as the 6-minute walking test (6WT) are increasingly applied to evaluate patients with degenerative diseases of the lumbar spine before and after (surgical) treatment. However, the traditional 6WT is cumbersome to apply, as it requires specialized in-hospital infrastructure and personnel. The authors set out to compare 6-minute walking distance (6WD) measurements obtained with a newly developed smartphone application (app) and those obtained with the gold-standard distance wheel (DW).

METHODS

The authors developed a free iOS- and Android-based smartphone app that allows patients to measure the 6WD in their home environment using global positioning system (GPS) coordinates. In a laboratory setting, the authors obtained 6WD measurements over a range of smartphone models, testing environments, and walking patterns and speeds. The main outcome was the relative measurement error (rME; in percent of 6WD), with |rME| < 7.5% defined as reliable. The intraclass correlation coefficient (ICC) for agreement between app- and DW-based 6WD was calculated.

RESULTS

Measurements (n = 406) were reliable with all smartphone types in neighborhood, nature, and city environments (without high buildings), as well as with unspecified, straight, continuous, and stop-and-go walking patterns (ICC = 0.97, 95% CI 0.97-0.98, p < 0.001). Measurements were unreliable indoors, in city areas with high buildings, and for predominantly rectangular walking courses. Walking speed had an influence on the ME, with worse accuracy (2% higher rME) for every kilometer per hour slower walking pace (95% CI 1.4%-2.5%, p < 0.001). Mathematical adjustment of the app-based 6WD for velocity-dependent error mitigated the rME (p < 0.011), attenuated velocity dependence (p = 0.362), and had a positive effect on accuracy (ICC = 0.98, 95% CI 0.98-0.99, p < 0.001).

CONCLUSIONS

The new, free, spine-specific 6WT smartphone app measures the 6WD conveniently by using GPS coordinates, empowering patients to independently determine their functional status before and after (surgical) treatment. Measurements of 6WD obtained for the target population under the recommended circumstances are highly reliable.

Use of Patient Health Questionnaire-2 scoring to predict patient satisfaction and return to work up to 1 year after lumbar fusion: a 2-year analysis from the Michigan Spine Surgery Improvement Collaborative

OBJECTIVE

The Michigan Spine Surgery Improvement Collaborative (MSSIC) is a prospective, longitudinal, multicenter, quality-improvement collaborative. Using MSSIC, the authors sought to identify the relationship between a positive Patient Health Questionnaire-2 (PHQ-2) screening, which is predictive of depression, and patient satisfaction, return to work, and achieving Oswestry Disability Index (ODI) minimal clinically important difference (MCID) scores up to 2 years after lumbar fusion.

METHODS

Data from a total of 8585 lumbar fusion patients were analyzed. Patient satisfaction was measured by the North American Spine Society patient satisfaction index. A positive PHQ-2 score is one that is ≥ 3, which has an 82.9% sensitivity and 90.0% specificity in detecting major depressive disorder. Generalized estimating equation models were constructed; variables tested include age, sex, race, past medical history, severity of surgery, and preoperative opioid usage.

RESULTS

Multivariate analysis was performed. Patients with a positive PHQ-2 score (i.e., ≥ 3) were less likely to be satisfied after lumbar fusion at 90 days (relative risk [RR] 0.93, p < 0.001), 1 year (RR 0.92, p = 0.001), and 2 years (RR 0.92, p = 0.028). A positive PHQ-2 score was also associated with decreased likelihood of returning to work at 90 days (RR 0.76, p < 0.001), 1 year (RR 0.85, p = 0.001), and 2 years (RR 0.82, p = 0.031). A positive PHQ-2 score was predictive of failure to achieve an ODI MCID at 90 days (RR 1.07, p = 0.005) but not at 1 year or 2 years after lumbar fusion.

CONCLUSIONS

A multivariate analysis based on information from a large, multicenter, prospective database on lumbar fusion patients was performed. The authors found that a positive score (≥ 3) on the PHQ-2, which is a simple and accurate screening tool for depression, predicts an inability to return to work and worse satisfaction up to 2 years after lumbar fusion. Depression is a treatable condition, and so in the same way that patients are medically optimized before surgery to decrease postoperative morbidity, perhaps patients should have preoperative psychiatric optimization to improve postoperative functional outcomes.

Trends in the surgical treatment of pathological proximal femur fractures in the United States

BACKGROUND AND OBJECTIVES

Large scale data on the treatment of pathologic proximal femur fractures (PPFFs) are lacking. The purpose of this study was to evaluate trends in patient demographics, complication rates, and relative utilization rates of various techniques associated with PPFFs.

METHODS

The American College of Surgeons-National Surgical Quality Improvement Program (NSQIP) database was queried for PPFFs from 2009 to 2017. Patient demographics, 30-day complications, and utilization rates were recorded. Trends in these variables were determined over the study period.

RESULTS

Most patient demographics did not change during the study period. There were no trends toward decreasing rates of major complications (P = .82), reoperations (P = .65), non-home discharges (P = .17), readmissions (P = .07), or deaths (P = .75); transfusion rates significantly decreased (P < .001). Rates of hemiarthroplasty decreased (P = .03) and rates of intramedullary nailing increased (P = .001).

DISCUSSION

Despite advances in cancer therapeutics, the average PPFF patient has not significantly changed over the past decade. Similarly, most short-term outcomes after PPFF surgery have not improved, demonstrating a need for improved perioperative protocols. Finally, rates of IMN fixation are increasing while rates of HA are falling at NSQIP hospitals. Given that orthopedic oncologists favor endoprosthetic reconstruction in most cases, there may be a need for increased communication between orthopedic oncologists and other members of the orthopedic community treating PPFFs.

Abnormally High, as Well as Low, Preoperative Platelet Counts Correlate With Adverse Outcomes and Readmissions After Elective Total Knee Arthroplasty

BACKGROUND

Laboratory studies are routinely performed as a part of the preoperative workup for a total knee arthroplasty (TKA). The ramifications of abnormal preoperative platelet counts remain uncharacterized in large, multicenter patient populations.

METHODS

Patients who underwent elective primary TKA were identified in the 2011-2015 National Surgical Quality Improvement Program database. Risk of 30-day postoperative complications was calculated as a function of preoperative platelet counts. Patients were characterized as having a normal platelet count, abnormally low platelet count, and abnormally high platelet count based on relative risk calculations. Univariate and multivariate analyses were performed to associate abnormal platelet counts with patient demographics, operative variables, 30-day postoperative complications, and readmissions.

RESULTS

In total, 140,073 patients who underwent elective TKA were identified. Using the relative risk threshold of 1.5 for any adverse event, abnormally low and abnormally high platelet count thresholds were set at ≤116,000/mL and ≥492,000/mL, respectively. Multivariate analyses revealed low platelet counts to be associated with higher rates of any, major, and minor adverse events and longer length of stay. Analogously, high platelet counts were associated with higher rates of any and minor adverse events and longer length of stay.

CONCLUSION

The present study employed a large patient sample size and showed that elective TKA patients with abnormally high, as well as low, platelet counts are at increased risk of postoperative adverse outcomes. Focused attention needs to be paid to TKA patients with preoperative abnormal platelet counts for optimization and postoperative care.

LEVEL OF EVIDENCE

Level III, retrospective comparative study.

Reporting Methodology of Neurosurgical Studies Utilizing the American College of Surgeons-National Surgical Quality Improvement Program Database: A Systematic Review and Critical Appraisal

BACKGROUND

Use of large databases such as the American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) has become increasingly common in neurosurgical research.

OBJECTIVE

To perform a critical appraisal and evaluation of the methodological reporting for studies in neurosurgical literature that utilize the ACS-NSQIP database.

METHODS

We queried Ovid MEDLINE, EMBASE, and PubMed databases for all neurosurgical studies utilizing the ACS-NSQIP. We assessed each study according to number of criteria fulfilled with respect to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement, REporting of studies Conducted using Observational Routinely-collected Health Data (RECORD) Statement, and Journal of American Medical Association–Surgical Section (JAMA-Surgery) Checklist. A separate analysis was conducted among papers published in core and noncore journals in neurosurgery according to Bradford's law.

RESULTS

A total of 117 studies were included. Median (interquartile range [IQR]) scores for number of fulfilled criteria for STROBE Statement, RECORD Statement, and JAMA-Surgery Checklist were 20 (IQR:19-21), 9 (IQR:8-9), and 6 (IQR:5-6), respectively. For STROBE Statement, RECORD Statement, and JAMA-Surgery Checklist, item 9 (potential sources of bias), item 13 (supplemental information), and item 9 (missing data/sensitivity analysis) had the highest number of studies with no fulfillment among all studies (56, 68, 50%), respectively. When comparing core journals vs noncore journals, no significant difference was found (STROBE, P = .94; RECORD, P = .24; JAMA-Surgery checklist, P = .60).

CONCLUSION

While we observed an overall satisfactory reporting of methodology, most studies lacked mention of potential sources of bias, data cleaning methods, supplemental information, and external validity. Given the pervasive role of national databases and registries for research and health care policy, the surgical community needs to ensure the credibility and quality of such studies that ultimately aim to improve the value of surgical care delivery to patients.

Patient Demographic and Surgical Factors that Affect Completion of Patient-Reported Outcomes 90 Days and 1 Year After Spine Surgery: Analysis from the Michigan Spine Surgery Improvement Collaborative (MSSIC)

BACKGROUND

The Michigan Spine Surgery Improvement Collaborative is a statewide multicenter quality improvement registry. Because missing data can affect registry results, we used MSSIC to find demographic and surgical characteristics that affect the completion of patient-reported outcomes (PROs) at 90 days and 1 year.

METHODS

A total of 24,404 patients who had lumbar surgery (17,813 patients) or cervical surgery (6591 patients) were included. Multivariate logistic regression models of patient disease were constructed to identify risk factors for failure to complete scheduled PRO surveys.

RESULTS

Patients ≥65 years old and female patients were both more likely to respond at 90 days and 1 year. Increasing education was associated with greater response rate at 90 days and 1 year. Whites and African Americans had no differences in response rates. Calling provided the highest response rate at 90 days and 1 year. For cervical spine patients, only discharge to rehabilitation increased completion rates, at 90 days but not 1 year. For lumbar spine patients, spondylolisthesis or stenosis (vs. herniated disc) had a greater response rate at 1 year. Patients with leg (vs. back) pain had a greater response only at 1 year. Patients with multilevel surgery had an increased response at 1 year. Patients who underwent fusion were more likely to respond at 90 days, but not 1 year. Discharge to rehabilitation increased response at 90 days and 1 year.

CONCLUSIONS

A multivariate analysis from a multicenter prospective database identified surgical factors that affect PRO follow-up, up to 1 year. This information can be helpful for imputing missing PRO data and could be used to strengthen data derived from large prospective databases.

30- and 90-Day Unplanned Readmission Rates, Causes, and Risk Factors After Cervical Fusion: A Single-Institution Analysis

STUDY DESIGN

Retrospective cohort study OBJECTIVE.: To study 30- and 90-day readmission rates, causes, and risk factors after anterior cervical discectomy and fusion (ACDF) and posterior cervical fusion (PCF).

SUMMARY OF BACKGROUND DATA

Existing data on readmission after cervical fusion is majorly derived from national databases. Given their inherent limitations in accuracy, follow-up available, and missing data, we intend to add to literature from our institutional analysis.

METHODS

Patients who underwent ACDF and PCF for degenerative cervical pathology in 2013 and 2014 were identified for the study. Comprehensive chart review was performed to record demographics and clinical patient profile. Hospital readmission within 30 and 90 days was identified, and the causes and management were recorded. Binary logistic regression analysis was done to study risk factors for readmission. ACDF and PCF were studied separately.

RESULTS

Our analysis included a total of 549 patients, stratified as 389 ACDFs and 160 PCFs. The 30- and 90-day unplanned readmission rate was 5.1% and 7.7% after ACDF. These rates were 11.2% and 16.9% after PCF. The most common cause of readmission was systemic infection and sepsis after ACDF and PCF (31.4% and 25.8% of readmitted, respectively), followed by pulmonary complications after ACDF (14.3% of readmitted) and wound complications after PCF (19.4% of readmitted). Predictors of readmission after ACDF included heart failure, history of malignancy, history of deep vein thrombosis/pulmonary embolism, and any intraoperative complication. In the PCF cohort, history of ischemic heart disease, increasing number of fusion levels and longer length of stay were independently predictive.

CONCLUSION

The rates, causes, and risk factors of readmission after ACDF and PCF have been identified. There is variation in published data regarding the incidence and risk factors for readmission after cervical fusion; however, majority of readmissions occur due to medical complications and systemic infection.

LEVEL OF EVIDENCE

3.

Safety of Outpatient Single-level Cervical Total Disc Replacement: A Propensity-Matched Multi-institutional Study

STUDY DESIGN

A retrospective cohort comparison study.

OBJECTIVE

The aim of this study was to investigate the perioperative adverse event profile of cervical total disc replacement (CTDR) performed as an outpatient relative to inpatient procedure.

SUMMARY OF BACKGROUND DATA

Recent reimbursement changes and a push for safe reductions in hospital stay have resulted in increased interest in performing CTDRs in the outpatient setting. However, there has been a paucity of studies investigating the safety of outpatient CTDR procedures, despite increasing frequency.

METHODS

Patients who underwent single-level CTDR were identified in the 2005 to 2016 National Surgical Quality Improvement Program database. Outpatient versus inpatient procedure status was defined by length of stay, with outpatient being less than 1 day. Patient baseline characteristics and comorbidities were compared between the two groups. Propensity score matched comparisons were then performed for 30-day perioperative complications and readmissions between the two cohorts. In addition, perioperative outcomes of outpatient single-level CTDR versus matched outpatient single-level anterior cervical discectomy and fusion (ACDF) cases were compared.

RESULTS

In total, 373 outpatient and 1612 inpatient single-level CTDR procedures were identified. After propensity score matching was performed to control for potential confounders, statistical analysis revealed no significant difference in perioperative complications between outpatient versus matched inpatient CTDR. Notably, the rate of readmissions was not different between the two groups. In addition, there was no difference in rates of perioperative adverse events between outpatient single-level CTDR versus matched outpatient single-level ACDF.

CONCLUSION

The perioperative outcomes evaluated in the current study support the conclusion that, for appropriately selected patients, single-level CTDR can be safely performed in the outpatient setting without increased rates of 30-day perioperative complications or readmissions compared with inpatient CTDR or outpatient single-level ACDF.

LEVEL OF EVIDENCE

3.

Development and validation of risk-adjustment models for elective, single-level posterior lumbar spinal fusions

Background

There is a paucity of literature examining the development and subsequent validation of risk-adjustment models that inform the trade-off between adequate risk-adjustment and data collection burden. We aimed to evaluate patient risk stratification by surgeons with the development and validation of risk-adjustment models for elective, single-level, posterior lumbar spinal fusions (PLSFs).

Methods

Patients undergoing PLSF from 2011-2014 were identified in the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP). The derivation cohort included patients from 2011-2013, while the validation cohort included patients from 2014. Outcomes of interest were severe adverse events (SAEs) and unplanned readmission. Bivariate analysis of risk factors followed by a stepwise logistic regression model was used. Limited risk-adjustment models were created and analyzed by sequentially adding variables until the full model was reached.

Results

A total of 7,192 and 4,182 patients were included in our derivation and validation cohorts, respectively. Full model performance was similar for the derivation and validation cohorts in both 30-day SAEs (C-statistic =0.66 vs. 0.69) and 30-day unplanned readmission (C-statistic =0.62 vs. 0.65). All models demonstrated good calibration and fit (P≥0.58). Intraoperative variables, laboratory values, and comorbid conditions explained >75% of the variation in 30-day SAEs; ASA class, laboratory values, and comorbid conditions accounted for >80% of model risk prediction for 30-day unplanned readmission. Four variables for the 30-day SAE models (age, gender, ASA ≥3, operative time) and 3 variables for the 30-day unplanned readmission models (age, ASA ≥3, operative time) were sufficient to achieve a C-statistic within four percentage points of the full model.

Conclusions

Risk-adjustment models for PLSF demonstrated acceptable calibration and discrimination using variables commonly found in health records and demonstrated only a limited set of variables were required to achieve an appropriate level of risk prediction.

Pediatric obesity is associated with short-term risks after pelvic osteotomy

The risk factors for increased perioperative morbidity following pediatric pelvic osteotomies are poorly understood. The purpose of this study was to characterize differences in adverse events, operative time, length of stay, and readmission following pelvic osteotomy for obese and nonobese patients. A retrospective cohort study was carried out using the National Surgical Quality Improvement Program Pediatric database to identify patients that underwent pelvic osteotomy with or without femoral osteotomy. Obesity was found to be an independent risk factor for blood transfusion (relative risk: 1.4, P=0.007) and readmission (relative risk: 2.3, P=0.032) within 30 days. These data can facilitate patient counseling and informed decision-making when planning for surgical correction of hip dysplasia.

Unlike Native Hip Fractures, Delay to Periprosthetic Hip Fracture Stabilization Does Not Significantly Affect Most Short-Term Perioperative Outcomes

BACKGROUND

The incidence of periprosthetic hip fractures is increasing due to higher numbers of total hip arthroplasties being performed. Unlike native hip fractures, the effect of time to surgery of periprosthetic hip fractures is not well established. This study evaluates the effect of time to surgery on perioperative complications for patients with periprosthetic hip fractures.

METHODS

Patients who underwent surgery for periprosthetic hip fracture were identified in the 2005-2016 National Surgical Quality Improvement Program database and stratified into 2 groups: <2 and ≥2 days from hospital admission to surgery. Multivariate regressions were used to compare risk for perioperative complications between the 2 groups. Independent risk factors for postoperative serious adverse events were characterized.

RESULTS

In total, 409 (<2 days from admission to surgery) and 272 (≥2 days from admission to surgery) patients were identified. Multivariate analysis revealed only higher risk of extended postoperative stay for patients who had delays of ≥2 days to surgery compared to those who had <2 days from admission to surgery. Independent risk factors for serious adverse events included increasing age, dependent preoperative functional status, and preoperative congestive heart failure, but not time to surgery.

CONCLUSION

Unlike for native hip fractures, time to surgery for periprosthetic hip fractures does not appear to affect most 30-day perioperative complications. However, it is worth noting that this study was unable to control for all potential confounders and therefore the results may not be generalizable to all types of periprosthetic hip fractures.

Adverse events and their risk factors 90 days after cervical spine surgery: analysis from the Michigan Spine Surgery Improvement Collaborative

OBJECTIVE

The Michigan Spine Surgery Improvement Collaborative (MSSIC) is a statewide, multicenter quality improvement initiative. Using MSSIC data, the authors sought to identify 90-day adverse events and their associated risk factors (RFs) after cervical spine surgery.

METHODS

A total of 8236 cervical spine surgery cases were analyzed. Multivariable generalized estimating equation regression models were constructed to identify RFs for adverse events; variables tested included age, sex, diabetes mellitus, disc herniation, foraminal stenosis, central stenosis, American Society of Anesthesiologists Physical Classification System (ASA) class > II, myelopathy, private insurance, anterior versus posterior approach, revision procedures, number of surgical levels, length of procedure, blood loss, preoperative ambulation, ambulation day of surgery, length of hospital stay, and discharge disposition.

RESULTS

Ninety days after cervical spine surgery, adverse events identified included radicular findings (11.6%), readmission (7.7%), dysphagia requiring dietary modification (feeding tube or nothing by mouth [NPO]) (6.4%), urinary retention (4.7%), urinary tract infection (2.2%), surgical site hematoma (1.1%), surgical site infection (0.9%), deep vein thrombosis (0.7%), pulmonary embolism (0.5%), neurogenic bowel/bladder (0.4%), myelopathy (0.4%), myocardial infarction (0.4%), wound dehiscence (0.2%), claudication (0.2%), and ileus (0.2%). RFs for dysphagia included anterior approach (p < 0.001), fusion procedures (p = 0.030), multiple-level surgery when considering anterior procedures only (p = 0.037), and surgery duration (p = 0.002). RFs for readmission included ASA class > II (p < 0.001), while preoperative ambulation (p = 0.001) and private insurance (p < 0.001) were protective. RFs for urinary retention included increasing age (p < 0.001) and male sex (p < 0.001), while anterior-approach surgery (p < 0.001), preoperative ambulation (p = 0.001), and ambulation day of surgery (p = 0.001) were protective. Preoperative ambulation (p = 0.010) and anterior approach (p = 0.002) were protective of radicular findings.

CONCLUSIONS

A multivariate analysis from a large, multicenter, prospective database identified the common adverse events after cervical spine surgery, along with their associated RFs. This information can lead to more informed surgeons and patients. The authors found that early mobilization after cervical spine surgery has the potential to significantly decrease adverse events.

Is Discharge Within a Day of Total Knee Arthroplasty Safe in the Octogenarian Population?

BACKGROUND

Reduced hospital stay programs for total knee arthroplasty (TKA) are being implemented in order to increase patient satisfaction and reduce healthcare costs. Although elderly patients are often included in these pathways, there have been limited data on whether older patients can safely be discharged within a day after TKA. The purpose of this study is to compare perioperative complications following primary TKA with ≤1 day in the hospital in patients aged ≥80 compared to <80 years old in the National Surgical Quality Improvement Program database.

METHODS

Patients who underwent primary TKA with hospital length of stay ≤1 day were identified in the 2005-2016 National Surgical Quality Improvement Program database. These patients were separated into 2 age groups: <80 and ≥80 years old. Preoperative and procedural characteristics were compared. Multivariate regressions were used to compare risk for perioperative adverse events and readmission. Independent risk factors for serious adverse events following such TKAs were identified.

RESULTS

In total, 17,191 (<80 year olds) and 1005 (≥80 year olds) cases were identified. Of these patients, 1750 cases were discharged the same day. Multivariate analysis revealed only higher risk for 30-day readmission and nonhome discharge in ≥80 compared to <80 year olds. Notably, the octogenarians had a significantly higher rate of nonsurgical site-related readmissions. Independent risk factors for serious adverse events include only American Society of Anesthesiologists score ≥3 and not patient age.

CONCLUSION

These data suggest that, although octogenarians can safely be discharged in ≤1 day, greater postdischarge care may be warranted to reduce the rate of nonsurgical site-related readmissions.

Machine Learning to Predict Delays in Adjuvant Radiation following Surgery for Head and Neck Cancer

OBJECTIVE

To apply a novel methodology with machine learning (ML) to a large national cancer registry to help identify patients who are high risk for delayed adjuvant radiation.

STUDY DESIGN

Observational cohort study.

SETTING

National Cancer Database (NCDB).

SUBJECTS AND METHODS

A total of 76,573 patients were identified from the NCDB who had invasive head and neck cancer and underwent surgery, followed by radiation. The model was constructed from 80% of the patient data and subsequently evaluated and scored with the remaining 20%. Permutation feature importance analysis was used to understand the weighted model construction.

RESULTS

A total of 76,573 patients met inclusion and exclusion criteria. Our ML model was able to predict whether patients would start adjuvant therapy beyond 50 days after surgery with an overall accuracy of 64.41% and a precision of 58.5%. The 2 most important variables used to build the model were treating facility and urban versus rural demographics.

CONCLUSION

Statistics can provide inferences within an overall system, while ML is a novel methodology that can make predictions. We can identify patients who are "high risk" for delayed radiation using information from >75,000 patient experiences, which has the potential for a direct impact on clinical care. Our inability to achieve greater accuracy is due to limitations of the data captured by the NCDB, and we need to continue to identify new variables that are correlated with delayed radiation therapy. ML will prove to be a valuable clinical tool in years to come, but its utility is limited by available data.

Surgical repair of perforated peptic ulcers: laparoscopic versus open approach

INTRODUCTION

Perforated peptic ulcers are a surgical emergency that can be repaired using either laparoscopic surgery (LS) or open surgery (OS). No consensus has been reached on the comparative outcomes and safety of each approach.

METHODS

Using the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database, we conducted a 12-year retrospective review (2005-2016) and identified 6260 adult patients who underwent either LS (n = 616) or OS (n = 5644) to repair perforated peptic ulcers. To mitigate selection bias and adjust for the inherent heterogeneity between groups, we used propensity-score matching with a case (LS):control (OS) ratio of 1:3. We then compared intraoperative outcomes such as operative time, and 30-day postoperative outcomes including infectious and non-infectious complications, and mortality.

RESULTS

Propensity-score matching created a total of 2462 matched pairs (616 in the LS group, 1846 in the OS group). Univariate analysis demonstrated successful matching of patient characteristics and baseline clinical variables. We found that OS was associated with a shorter operative time (67.0 ± 28.6 min, OS versus 86.9 ± 57.5 min, LS; P < 0.001) but a longer hospital stay (8.6 ± 6.2 days, OS versus 7.8 ± 5.9 days, LS; P = 0.001). LS was associated with a lower rate of superficial surgical site infections (1.5%, LS versus 4.2%, OS; P = 0.032), wound dehiscence (0.3%, LS versus 1.6%, OS; P = 0.030), and mortality (3.2%, LS versus 5.4%, OS; P = 0.009).

CONCLUSION

Fewer than 10% of patients with perforated peptic ulcers underwent LS, which was associated with reduced length of stay, lower rate of superficial surgical site infections, wound dehiscence, and mortality. Given our results, a greater emphasis should be provided to a minimally invasive approach for the surgical repair of perforated peptic ulcers.

Very Early Versus Early Readmissions in General Surgery Patients

BACKGROUND

Hospital readmission rates are an important quality metric. A readmission very soon after discharge may be related to a different cause than readmissions that occur later in the first 30 d.

MATERIALS AND METHODS

The National Surgical Quality Improvement Program data sets from 2014 to 2015 were used to identify patients undergoing general surgery procedures. Demographics, comorbidities, and morbidity were analyzed. Stepwise regression was used to determine statistical predictors for any readmission. The final model variables were a combination of selected clinical variables and statistically significant variables. Multinomial logistic regression was then used with these variables to develop models for "very early" (days 0-3 after discharge) and "early" (days 4-30) readmissions.

RESULTS

A total of 744,492 patients were included with 5.9% readmitted within 30 d and 1.5% readmitted within 3 d of discharge (26.1% of all readmissions). Significant risk factors for any readmission included ≥3 comorbidities, major surgery (operative time >1 h, length of stay greater >2 d), and American Society of Anesthesiologists class ≥3. When examining "very early" readmissions, the greatest risk factor was experiencing a severe complication (≥Grade III) before discharge.

CONCLUSIONS

Readmissions within 3 d of discharge constitute a large portion of all 30 d readmissions. The greatest risk factor for "very early" readmission was a severe complication before discharge. Better understanding of the reason for this association is needed to develop effective prevention strategies.

Missing data treatments matter: an analysis of multiple imputation for anterior cervical discectomy and fusion procedures

BACKGROUND CONTEXT

The presence of missing data is a limitation of large datasets, including the National Surgical Quality Improvement Program (NSQIP). In addressing this issue, most studies use complete case analysis, which excludes cases with missing data, thus potentially introducing selection bias. Multiple imputation, a statistically rigorous approach that approximates missing data and preserves sample size, may be an improvement over complete case analysis.

PURPOSE

The present study aims to evaluate the impact of using multiple imputation in comparison with complete case analysis for assessing the associations between preoperative laboratory values and adverse outcomes following anterior cervical discectomy and fusion (ACDF) procedures.

STUDY DESIGN/SETTING

This is a retrospective review of prospectively collected data.

PATIENT SAMPLE

Patients undergoing one-level ACDF were identified in NSQIP 2012-2015.

OUTCOME MEASURES

Perioperative adverse outcome variables assessed included the occurrence of any adverse event, severe adverse events, and hospital readmission.

METHODS

Missing preoperative albumin and hematocrit values were handled using complete case analysis and multiple imputation. These preoperative laboratory levels were then tested for associations with 30-day postoperative outcomes using logistic regression.

RESULTS

A total of 11,999 patients were included. Of this cohort, 63.5% of patients had missing preoperative albumin and 9.9% had missing preoperative hematocrit. When using complete case analysis, only 4,311 patients were studied. The removed patients were significantly younger, healthier, of a common body mass index, and male. Logistic regression analysis failed to identify either preoperative hypoalbuminemia or preoperative anemia as significantly associated with adverse outcomes. When employing multiple imputation, all 11,999 patients were included. Preoperative hypoalbuminemia was significantly associated with the occurrence of any adverse event and severe adverse events. Preoperative anemia was significantly associated with the occurrence of any adverse event, severe adverse events, and hospital readmission.

CONCLUSIONS

Multiple imputation is a rigorous statistical procedure that is being increasingly used to address missing values in large datasets. Using this technique for ACDF avoided the loss of cases that may have affected the representativeness and power of the study and led to different results than complete case analysis. Multiple imputation should be considered for future spine studies.

Evaluating the effect of growing patient numbers and changing data elements in the National Surgical Quality Improvement Program (NSQIP) database over the years: a study of posterior lumbar fusion outcomes

BACKGROUND CONTEXT

The use of national databases in spinal surgery outcomes research is increasing. A number of variables collected by the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) changed between 2010 and 2011, coinciding with a rapid increase in the number of patients included per year. However, there has been limited study evaluating the effect that these changes may have on the results of outcomes studies.

PURPOSE

The present study aimed to investigate the influence of changing data elements and growth of the NSQIP database on results of lumbar fusion outcomes studies.

STUDY DESIGN/SETTING

This is a retrospective cohort study of prospectively collected data.

PATIENT SAMPLE

The NSQIP database was retrospectively queried to identify 19,755 patients who underwent elective posterior lumbar fusion surgery with or without interbody fusion between 2005 and 2014. Patients were split into two groups based on year of surgery: 2,802 from 2005 to 2010 and 16,953 from 2011 to 2014.

OUTCOME MEASURES

The occurrence of adverse events after discharge from the hospital, within postoperative day 30, was determined.

METHODS

Preoperative characteristics and 30-day perioperative outcomes were compared between the era groups using bivariate analysis. To illustrate the effect of such changing data elements, the association between age and postoperative outcomes in the era groups was analyzed using multivariate Poisson regression. The present study had no funding sources, and there were no study-related conflicts of interest for any authors.

RESULTS

There were significant differences between the era groups for a variety of preoperative characteristics. Postoperative events such blood transfusion and deep vein thrombosis were also significantly different between the era groups. For the 2005-2010 cohort, age was significantly associated with septic shock by multivariate analysis. For the 2011-2014 cohort, age was significantly associated with septic shock, urinary tract infection, blood transfusion, myocardial infarction, and extended length of stay.

CONCLUSIONS

The NSQIP database has undergone substantial changes between 2005 and 2014. These changes may contribute to different results in analyses, such as the association between age and postoperative outcomes, when using older versus newer data. Conclusions from early studies using this database may warrant reconsideration.

Proliferation and Contributions of National Database Studies in Otolaryngology Literature Published in the United States: 2005-2016

OBJECTIVE

Health registries and discharge-level databases are powerful tools. Commonly used data sets include the Nationwide Inpatient Sample (NIS); Surveillance, Epidemiology, and End Results Program (SEER); National Cancer Database (NCDB); and American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP). This study investigated the frequency with which these resources are being used and categorized their contributions to literature.

DESIGN

A literature review from 2005 to 2016 for papers utilizing the aforementioned databases and publishing in The Laryngoscope, JAMA-Otolaryngology, Head and Neck, Otolaryngology-Head and Neck Surgery, and International Forum of Allergy and Rhinology was conducted. Results were categorized based on the contribution(s) of the paper. The incidence rate of database publications was calculated for each year along with the 95% confidence intervals using a Poisson distribution.

RESULTS

Three hundred ten studies were identified. Seventy percent report descriptive findings, and 65% report outcomes/survival. Approximately 18% made clinical recommendations. In 2005, the incidence rate of database publications was 3 per 1000 journal publications (95% CI, 1-9) and remained relatively stable until 2008. From 2010 onward, there was a persistent increase in publications, culminating in the highest incidence rate in 2016 of 26 database publications per 1000 journal publications (95% CI, 20-32).

CONCLUSIONS

There was a nearly 10-fold increase in database publications in 2016 compared to 2005. The majority provide descriptive data and outcomes measures. The role of these studies warrants further investigation.

Systematic Changes in the National Surgical Quality Improvement Program Database Over the Years Can Affect Comorbidity Indices Such as the Modified Frailty Index and Modified Charlson Comorbidity Index for Lumbar Fusion Studies

STUDY DESIGN

Retrospective cohort study of prospectively collected data.

OBJECTIVE

The aim of this study was to investigate the influence of changes in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database over the years on the calculation of the modified Frailty Index (mFI) and the modified Charlson Comorbidity Index (mCCI) for posterior lumbar fusion studies.

SUMMARY OF BACKGROUND DATA

Multiple studies have utilized the mFI and/or mCCI and showed them to be predictors of adverse postoperative outcomes. However, changes in the NSQIP database have resulted in definition changes and/or missing data for many of the variables included in these indices. No studies have assessed the influence of different methods of treating missing values when calculating these indices on such studies.

METHODS

Elective posterior lumbar fusions were identified in NSQIP from 2005 to 2014. The mFI was calculated for each patient using three methods: treating conditions for which data was missing as not present, dropping patients with missing values, and normalizing by dividing the raw score by the number of variables collected. The mCCI was calculated by the first two of these methods. Mean American Society of Anesthesiologists (ASA) scores used for comparison.

RESULTS

In total, 19,755 patients were identified. Mean ASA score increased between 2005 and 2014 from 2.27 to 2.50 (+10.1%). For each of the methods of data handling noted above, mean mFI over the years studied increased by 33.3%, could not be calculated, and increased by 183.3%, respectively. Mean mCCI increased by 31.2% and could not be calculated respectively.

CONCLUSION

Systematic changes in the NSQIP database have resulted in missing data for many of the variables included in the mFI and the mCCI and may affect studies utilizing these indices. These changes can be understood in the context of ASA trends, and raise questions regarding the use of these indices with data available in later NSQIP years.

LEVEL OF EVIDENCE

3.