Predictive value of the ACS NSQIP calculator for head and neck reconstruction free tissue transfer

The efficacy of the National Surgical Quality Improvement Program surgical risk calculator in head and neck surgery: A meta-analysis

BACKGROUND

The National Surgical Quality Improvement Program surgical risk calculator (SRC) estimates the risk for postoperative complications. This meta-analysis assesses the efficacy of the SRC in the field of head and neck surgery.

METHODS

A systematic review identified studies comparing the SRC's predictions to observed outcomes following head and neck surgeries. Predictive accuracy was assessed using receiver operating characteristic curves (AUCs) and Brier scoring.

RESULTS

Nine studies totaling 1774 patients were included. The SRC underpredicted the risk of all outcomes (including any complication [observed (ob) = 35.9%, predicted (pr) = 21.8%] and serious complication [ob = 28.7%, pr = 17.0%]) except mortality (ob = 0.37%, pr = 1.55%). The observed length of stay was more than twice the predicted length (p < 0.02). Discrimination was acceptable for postoperative pneumonia (AUC = 0.778) and urinary tract infection (AUC = 0.782) only. Predictive accuracy was low for all outcomes (Brier scores ≥0.01) and comparable for patients with and without free-flap reconstructions.

CONCLUSION

The SRC is an ineffective instrument for predicting outcomes in head and neck surgery.

Factors Predicting Discharge Disposition Following Head and Neck Free Flap Reconstruction

OBJECTIVES

Patients undergoing head and neck free flap reconstruction (HNFFR) may have significant change to their baseline functional status requiring inpatient rehabilitation (IPR) after discharge. We sought to identify patient/procedure characteristics predictive of discharge destination.

METHODS

Patients undergoing elective HNFFR between July 2017 and July 2022 were reviewed for discharge destination. Those discharged to IPR versus home were compared across patient/procedure characteristics and physical/occupational therapy metrics. Significance was assessed via bivariate and multivariable analyses.

RESULTS

Of the 531 patients, 102 (19.2%) required IPR postoperatively. Patients discharged to IPR versus home were significantly older (70.1 [11.6] vs. 64.1 [13.1] years; p < 0.001) and more likely to lack family assistance (26.5% vs. 8.6%; p < 0.001), require baseline assistance for activities of daily living (ADLs) (31.4% vs. 9.8%; p < 0.001), have baseline cognitive dysfunction (15.7% vs. 6.1%; p = 0.001), were more likely to have neoplasm as the surgical indication for HNFFR (89.2% vs. 80.0%; p = 0.033) and more likely to have a tracheostomy postop (62.7% vs. 51.7%), and had a significantly longer length of stay (11.2 [8.0] vs. 6.8 [8.3] days; p < 0.001). There was no significant difference in gender, donor site, use of tube feeds, and use of assistive devices between the two groups. Following logistic regression, the strongest predictors of discharge to IPR include lack of family assistance (OR = 3.8; p < 0.001) and baseline assistance for ADLs (OR = 4.0, p < 0.001).

CONCLUSION

Certain patient factors predict the need for discharge to rehab after HNFFR. Perioperative identification of these factors may facilitate patient counseling and discharge planning with potential to reduce hospital length of stay and further optimize patient care.

LEVEL OF EVIDENCE

III Laryngoscope, 134:2721-2725, 2024.

Length of Stay Prediction Models for Oral Cancer Surgery: Machine Learning, Statistical and ACS-NSQIP

OBJECTIVE

Accurate prediction of hospital length of stay (LOS) following surgical management of oral cavity cancer (OCC) may be associated with improved patient counseling, hospital resource utilization and cost. The objective of this study was to compare the performance of statistical models, a machine learning (ML) model, and The American College of Surgeons National Surgical Quality Improvement Program's (ACS-NSQIP) calculator in predicting LOS following surgery for OCC.

MATERIALS AND METHODS

A retrospective multicenter database study was performed at two major academic head and neck cancer centers. Patients with OCC who underwent major free flap reconstructive surgery between January 2008 and June 2019 surgery were selected. Data were pooled and split into training and validation datasets. Statistical and ML models were developed, and performance was evaluated by comparing predicted and actual LOS using correlation coefficient values and percent accuracy.

RESULTS

Totally 837 patients were selected with mean patient age being 62.5 ± 11.7 [SD] years and 67% being male. The ML model demonstrated the best accuracy (validation correlation 0.48, 4-day accuracy 70%), compared with the statistical models: multivariate analysis (0.45, 67%) and least absolute shrinkage and selection operator (0.42, 70%). All were superior to the ACS-NSQIP calculator's performance (0.23, 59%).

CONCLUSION

We developed statistical and ML models that predicted LOS following major free flap reconstructive surgery for OCC. Our models demonstrated superior predictive performance to the ACS-NSQIP calculator. The ML model identified several novel predictors of LOS. These models must be validated in other institutions before being used in clinical practice.

LEVEL OF EVIDENCE

Level 3 Laryngoscope, 2024.

Emerging Concepts Impacting Head and Neck Cancer Surgery Morbidity

All treatment modalities for head and neck cancer carry with them a risk of adverse events. Head and neck surgeons are faced with significant challenges to minimize associated morbidity and manage its sequelae. Recognizing situations in which a surgical complication is an adverse event inherent to the procedure can alleviate the psychologic impact a complication might have on the treatment team and minimize external and internal pressures. Focusing on the complications that can be effectively modified, future complications can be avoided. Also, some surgical morbidities may not be preventable, necessitating the option to reconsider whether the incidents should be labeled toxic reactions rather than a complication. This discussion highlights some of the areas in which additional research is needed to achieve the goal of minimizing the impact of surgical morbidity.

Evaluating the Inaccuracy of the National Surgical Quality Improvement Project Surgical Risk Calculator in Plastic Surgery: A Meta-analysis of Short-Term Predicted Complications

BACKGROUND

Preoperative surgical risk assessment is a major component of clinical decision making. The ability to provide accurate, individualized risk estimates has become critical because of growing emphasis on quality metrics benchmarks. The American College of Surgeons National Surgical Quality Improvement Project (NSQIP) Surgical Risk Calculator (SRC) was designed to quantify patient-specific risk across various surgeries. Its applicability to plastic surgery is unclear, however, with multiple studies reporting inaccuracies among certain patient populations. This study uses meta-analysis to evaluate the NSQIP SRC's ability to predict complications among patients having plastic surgery.

METHODS

OVID MEDLINE and PubMed were searched for all studies evaluating the predictive accuracy of the NSQIP SRC in plastic surgery, including oncologic reconstruction, ventral hernia repair, and body contouring. Only studies directly comparing SCR predicted to observed complication rates were included. The primary measure of SRC prediction accuracy, area under the curve (AUC), was assessed for each complication via DerSimonian and Laird random-effects analytic model. The I2 statistic, indicating heterogeneity, was judged low (I2 < 50%) or borderline/unacceptably high (I2 > 50%). All analyses were conducted in StataSE 16.1 (StataCorp LP, College Station, Tex).

RESULTS

Ten of the 296 studies screened met criteria for inclusion (2416 patients). Studies were classified as follows: (head and neck: n = 5, breast: n = 1, extremity: n = 1), open ventral hernia repair (n = 2), and panniculectomy (n = 1). Predictive accuracy was poor for medical and surgical complications (medical: pulmonary AUC = 0.67 [0.48-0.87], cardiac AUC = 0.66 [0.20-0.99], venous thromboembolism AUC = 0.55 [0.47-0.63]), (surgical: surgical site infection AUC = 0.55 [0.46-0.63], reoperation AUC = 0.54 [0.49-0.58], serious complication AUC = 0.58 [0.43-0.73], and any complication AUC = 0.60 [0.57-0.64]). Although mortality was accurately predicted in 2 studies (AUC = 0.87 [0.54-0.99]), heterogeneity was high with I2 = 68%. Otherwise, heterogeneity was minimal (I2 = 0%) or acceptably low (I2 < 50%) for all other outcomes.

CONCLUSIONS

The NSQIP Universal SRC, aimed at offering individualized quantifiable risk estimates for surgical complications, consistently demonstrated poor risk discrimination in this plastic surgery-focused meta-analysis. The limitations of the SRC are perhaps most pronounced where complex, multidisciplinary reconstructions are needed. Future efforts should identify targets for improving SRC reliability to better counsel patients in the perioperative setting and guide appropriate healthcare resource allocation.

Validation of a risk prediction calculator in Australian patients undergoing head and neck microsurgery reconstruction

BACKGROUND

The American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) surgical risk calculator (SRC) is an open access calculator predicting patients' risk of postoperative complications. This study aims to assess the validity of the SRC in patients undergoing microsurgical free flap reconstruction at an Australian tertiary referral centre.

METHODS

This is a retrospective cohort study of 200 consecutive patients treated up to November 2020. SRC-predicted rates of postoperative complications and hospital length of stay (LOS) were compared to those observed for the ablative and reconstructive components of the procedure. The performance of the SRC was assessed using Brier scores, area under the receiver operating characteristic (ROC) curve (AUC), and the Hosmer-Lemeshow test.

RESULTS

For both ablative and reconstructive components, the SRC discriminates well for pneumonia and urinary tract infection, and it is calibrated well for readmission and sepsis, but it does not discriminate and calibrate well for any single outcome. SRC-predicted hospital LOS and actual LOS did not correlate well for the reconstructive component, but they correlated strongly for the ablative component.

CONCLUSIONS

The SRC is a poor predictor of postoperative complication rates and hospital LOS in patients undergoing head and neck microsurgical reconstruction.

Validation of the Emergency Surgery Score (ESS) in a Greek patient population: a prospective bi-institutional cohort study

Purpose

The Emergency Surgery Score (ESS) is a reliable point-based score that predicts mortality and morbidity in emergency surgery patients. However, it has been validated only in the U.S. patients. We aimed to prospectively validate ESS in a Greek patient population.

Methods

All patients who underwent an emergent laparotomy were prospectively included over a 15-month period. A systematic chart review was performed to collect relevant preoperative, intraoperative, and postoperative variables based on which the ESS was calculated for each patient. The relationship between ESS and 30-day mortality, morbidity (i.e., the occurrence of at least one complication), and the need for intensive care unit (ICU) admission was evaluated and compared between the Greek and U.S. patients using the c-statistics methodology. The study was registered on "Research Registry" with the unique identifying number 5901.

Results

A total of 214 patients (102 Greek) were included. The mean age was 64 years, 44% were female, and the median ESS was 7. The most common indication for surgery was hollow viscus perforation (25%). The ESS reliably and incrementally predicted mortality (c-statistics = 0.79 [95% CI 0.67–0.90] and 0.83 [95% CI 0.74–0.92]), morbidity (c-statistics = 0.83 [95% CI 0.76–0.91] and 0.79 [95% CI 0.69–0.88]), and ICU admission (c-statistics = 0.88 [95% CI 0.81–0.96] and 0.84 [95% CI 0.77–0.91]) in both Greek and U.S. patients.

Conclusion

The correlation between the ESS and the surgical outcomes was statistically significant in both Greek and U.S. patients undergoing emergency laparotomy. ESS could prove globally useful for preoperative patient counseling and quality-of-care benchmarking.

Machine Learning Demonstrates High Accuracy for Disease Diagnosis and Prognosis in Plastic Surgery

INTRODUCTION

Machine learning (ML) is a set of models and methods that can detect patterns in vast amounts of data and use this information to perform various kinds of decision-making under uncertain conditions. This review explores the current role of this technology in plastic surgery by outlining the applications in clinical practice, diagnostic and prognostic accuracies, and proposed future direction for clinical applications and research.

METHODS

EMBASE, MEDLINE, CENTRAL and ClinicalTrials.gov were searched from 1990 to 2020. Any clinical studies (including case reports) which present the diagnostic and prognostic accuracies of machine learning models in the clinical setting of plastic surgery were included. Data collected were clinical indication, model utilised, reported accuracies, and comparison with clinical evaluation.

RESULTS

The database identified 1181 articles, of which 51 articles were included in this review. The clinical utility of these algorithms was to assist clinicians in diagnosis prediction (n=22), outcome prediction (n=21) and pre-operative planning (n=8). The mean accuracy is 88.80%, 86.11% and 80.28% respectively. The most commonly used models were neural networks (n=31), support vector machines (n=13), decision trees/random forests (n=10) and logistic regression (n=9).

CONCLUSIONS

ML has demonstrated high accuracies in diagnosis and prognostication of burn patients, congenital or acquired facial deformities, and in cosmetic surgery. There are no studies comparing ML to clinician's performance. Future research can be enhanced using larger datasets or utilising data augmentation, employing novel deep learning models, and applying these to other subspecialties of plastic surgery.

Can the Emergency Surgery Score (ESS) predict outcomes in emergency general surgery patients with missing data elements? A nationwide analysis

BACKGROUND

The Emergency Surgery Score (ESS) is an accurate mortality risk calculator for emergency general surgery (EGS). We sought to assess whether ESS can accurately predict 30-day morbidity, mortality, and requirement for postoperative Intensive Care Unit (ICU) care in patients with missing data variables.

METHODS

All EGS patients with one or more missing ESS variables in the 2007-2015 ACS-NSQIP database were included. ESS was calculated assuming that a missing variable is normal (i.e. no additional ESS points). The correlation between ESS and morbidity, mortality, and postoperative ICU level of care was assessed using the c-statistics methodology.

RESULTS

Out of a total of 4,456,809 patients, 359,849 were EGS, and of those 256,278 (71.2%) patients had at least one ESS variable missing. ESS correlated extremely well with mortality (c-statistic = 0.94) and postoperative requirement of ICU care (c-statistic = 0.91) and well with morbidity (c-statistic = 0.77).

CONCLUSION

ESS performs well in predicting outcomes in EGS patients even when one or more data elements are missing and remains a useful bedside tool for counseling EGS patients and for benchmarking the quality of EGS care.

Short-Term Morbidity and Predictors of Adverse Events Following Esthesioneuroblastoma Surgery

INTRODUCTION

The short-term adverse events and predictors of morbidity in surgical resection of esthesioneuroblastoma (ENB) are largely unknown, and investigating these variables can help direct planning for at-risk patients.

METHODS

The 2005-2017 National Surgical Quality Improvement Program database was queried to identify patients with a diagnosis of ENB undergoing skull base surgery for tumor resection. Information regarding demographics, patient morbidity score, pre-operative and intra-operative data, and post-operative outcomes were extracted. Cox proportional hazard analysis was utilized to assess complication and readmission/reoperation rates.

RESULTS

A total of 95 patients undergoing skull base surgery for resection of ENB were included. Mean age, BMI, operation time, and post-operative length of stay (LOS) of the cohort were 53.6 ± 16.2 years, 29.1 ± 6.5, 392.0 ± 204.6 minutes, and 5.8 ± 4.6 days, respectively. In total, 31 patients (32.6%) experienced at least one 30-day adverse event, which included blood transfusion intra-operatively or within 72 hours from the operation (22.1%), readmission (10.7%), intubation >48 hours (7.4%), reintubation (4.2%), organ or space infection (4.2%), reoperation (4.0%), superficial or deep surgical site infection (2.1%), sepsis (2.1%), pulmonary embolism (1.1%), and myocardial infarction (1.1%). Patients who experienced at least one adverse event had significantly higher operation time (486.8 ± 230.4 vs. 347.5 ± 176.2 minutes, p = 0.002), LOS (9.2 ± 5.6 days vs. 4.2 ± 3.0, p < 0.001), and lower hematocrit (37.3 ± 5.9 vs. 41.2 ± 3.8, p < 0.001) and albumin levels (3.8 ± 0.6 vs. 4.2 ± 0.3, p = 0.009). Patients with a higher American Society of Anesthesiologists (ASA) score (HR = 2.39; p = 0.047) or longer operation time (HR = 1.004; p = 0.001) had a significantly higher risk for experiencing adverse events. Obesity was not associated with different intra- or post-operative outcomes, but older patients had shorter operations (p = 0.002) and LOS (p = 0.0014).

CONCLUSION

Longer operation time and lower pre-operative hematocrit and albumin levels may all increase complication rates in ENB resection. Patients with high ASA score or more advanced age may have different short-term outcomes.