Neutrophil to lymphocyte ratio and mortality in spinal epidural abscess

Neutrophil to Lymphocyte Ratio as an Indicator of Periprosthetic Joint Infection: A Retrospective Cohort Study

INTRODUCTION

Periprosthetic joint infection (PJI) after total joint arthroplasty (TJA) is a serious complication posing notable clinical implications for patients and substantial economic burdens. Neutrophil to lymphocyte ratio (NLR) is an emerging biomarker of inflammation, which may better predict PJI. The objective of this review was to evaluate NLR changes in patients with confirmed PJI, to compare NLR between an aseptic revision and a revision for PJI, and to establish whether an NLR of 2.45 is an appropriate cutoff for predicting infection.

METHODS

A retrospective review of patients who underwent revision TJA for PJI at a single center between January 1, 2005, and December 31, 2018, was performed and compared with an aseptic cohort who underwent aseptic revision TJA. NLR was calculated from complete blood counts performed at index surgery and at the time of revision surgery. Receiver operating characteristic curves were analyzed, along with sensitivity, specificity, and positive and negative likelihood ratios.

RESULTS

There were 89 patients included in each cohort. Mean NLR in patients who underwent revision for PJI was 2.85 (± 1.27) at the time of index surgery and 6.89 (± 6.64) at the time of revision surgery ( P = 0.017). Mean NLR in patients undergoing revision for PJI (6.89) was significantly higher than aseptic revisions (3.17; P < 0.001).

DISCUSSION

In patients who underwent revision surgery for PJI, NLR was markedly elevated at time of revision compared with the time of index surgery. Because it is a cost-effective and readily available test, these findings suggest that NLR may be a useful triage test in the diagnosis of PJI.

LEVEL OF EVIDENCE

Level III Diagnostic Study.

The utility of vertebral Hounsfield units as a prognostic indicator of adverse events following treatment of spinal epidural abscess

Background

Spinal epidural abscesses (SEAs) are a devastating condition with high levels of associated morbidity and mortality. Hounsfield units (HUs), a marker of radiodensity on CT scans, have previously been correlated with adverse events following spinal interventions. We evaluated whether HUs might also be associated with all-cause complications and/or mortality in this high-risk population.

Methods

This retrospective cohort study was carried out within an academic health system in the United States. Adults diagnosed with a SEA between 2006 and 2021 and who also had a CT scan characterizing their SEA within 6 months of diagnosis were considered. HUs were abstracted from the 4 vertebral bodies nearest to, but not including, the infected levels. Our primary outcome was the presence of composite 90-day complications and HUs represented the primary predictor. A multivariable logistic regression analysis was conducted adjusting for demographic and disease-specific confounders. In sensitivity testing, separate logistic regression analyses were conducted (1) in patients aged 65 and older and (2) with mortality as the primary outcome.

Results

Our cohort consisted of 399 patients. The overall incidence of 90-day complications was 61.2% (n=244), with a 7.8% (n=31) 90-day mortality rate. Those experiencing complications were more likely to have undergone surgery to treat their SEA (58.6% vs. 46.5%; p=.018) but otherwise the cohorts were similar. HUs were not associated with composite 90-day complications (Odds ratio [OR] 1.00 [95% CI 1.00-1.00]; p=.842). Similar findings were noted in sensitivity testing.

Conclusions

While HUs have previously been correlated with adverse events in certain clinical contexts, we found no evidence to suggest that HUs are associated with all-cause complications or mortality in patients with SEAs. Future research hoping to leverage 3-dimensional imaging as a prognostic measure in this patient population should focus on alternative targets.

Level of Evidence

Level III; Observational Cohort study.

The prognostic utility of neutrophil-lymphocyte ratio in spinal surgery: A systematic review and meta-analysis

PURPOSE

Neutrophil-lymphocyte ratio (NLR) is reportedly an effective prognostic tool across various medical and surgical fields, but its value in spinal surgery is unestablished. We aim to investigate the relationship between elevated baseline/postoperative NLR and patient outcomes in spinal surgery.

MATERIALS AND METHODS

We performed a systematic search in PubMed, EMBASE, and SCOPUS databases for studies investigating the prognostic value of NLR in spine patients.Odds ratios (ORs) and hazard ratios (HRs) with 95% confidence intervals (CIs) were analysed on the RevMan 5.4 software. Where meta-analysis was not possible, we vote-counted the direction of the effect of elevated NLR. The GRADE framework for prognostic factor research was utilised to assess the certainty of the evidence for each outcome measure.

RESULTS

Five outcome measures (overall survival, mortality, disease-free survival, functional recovery and complications) were assessed across 16 studies involving 5471 patients. Elevated baseline NLR was associated with reduced overall survival (HR: 1.63, 95 % CI: 1.05 - 2.54) (GRADE: low) and worsened functional recovery (OR: 0.93, 95 % CI: 0.87 - 0.98) (GRADE: low). There was no association between baseline NLR and disease-free survival (HR: 2.42, 95 % CI: 0.49 - 11.83) (GRADE: very low) or mortality (OR: 1.39, 95 % CI: 0.41 - 4.75) (GRADE: very low). Elevated NLR levels measured on days 3-4 and days 6-7 postoperatively, but not NLR measured at baseline or on days 1-2 postoperatively, were associated with greater risks of complications (GRADE: low).

CONCLUSIONS

NLR is an objective tool with the potential to identify the patients that would benefit from surgery and facilitate shared decision-making.

New Inflammatory Marker Associated with Disease Activity in Gouty Arthritis: The Systemic Inflammatory Response Index

Background

The systemic immune-inflammatory index (SII) and systemic inflammatory response index (SIRI), as novel non-specific inflammatory markers, have recently drawn attention. At present, no studies have been conducted to investigate the value of SII and SIRI in gouty arthritis (GA), so we explored their possible association with GA disease activity.

Methods

The study enrolled 474 patients with acute gouty arthritis (AG), 399 patients with intercritical gouty arthritis (IG) and 194 healthy controls (HC). The differences in Monocyte-to-lymphocyte ratio (MLR), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), SII, and SIRI levels among different groups were assessed. The changes in the above indicators before and after treatment in the AG and IG groups were evaluated. Multivariate logistic regression analysis was assessed influencing factors for the acute gout attack. ROC curves were plotted to evaluate their diagnostic value for AG.

Results

Compared with the IG group, the erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), PLR, and incidence of hyperlipidemia in the AG group were significantly higher, and the duration of disease was significantly shorter (P < 0.05). The MLR, NLR, SII and SIRI in the AG group were significantly higher than those in the IG and HC groups (P < 0.05). Compared with baseline, decreased MLR, NLR, PLR, SII and SIRI were observed in the AG group after treatment (P < 0.05), while there was no significant difference in the IG group before and after treatment (P > 0.05). SIRI was positively correlated with ESR and CRP (P < 0.05). Multivariate logistic regression analysis result showed that duration of disease, hyperlipidemia, ESR, CRP, and SIRI were influencing factors of acute gout attack (P < 0.05). The AUC of ESR, CRP and SIRI on the diagnosis in AG were 0.664, 0.755, and 0.674, respectively.

Conclusion

SIRI may be used as a new inflammatory marker of disease activity with gouty arthritis.

Risk Factors, Outcomes, and Predictions of Extensively Drug-Resistant Acinetobacter baumannii Nosocomial Infections in Patients with Nervous System Diseases

Purpose

Acinetobacter baumannii has evolved to become a major pathogen of nosocomial infections, resulting in increased morbidity and mortality. This study aimed to investigate the risk factors, outcomes, and predictions of extensively drug-resistant (XDR)-A. baumannii nosocomial infections in patients with nervous system diseases (NSDs).

Methods

A retrospective study of patients infected with XDR-A. baumannii admitted to the Affiliated Hospital of Southwest Medical University (Luzhou, China) from January 2021 to December 2022 was conducted. Three multivariate regression models were used to assess the risk factors and predictive value for specific diagnostic and prognostic subgroups.

Results

A total of 190 patients were included, of which 84 were diagnosed with NSDs and 80% of those were due to stroke. The overall rates of all-cause mortality for XDR-A. baumannii nosocomial infections and those in NSDs were 38.9% and 40.5%, respectively. Firstly, hypertension, indwelling gastric tube, tracheotomy, deep puncture, bladder irrigation, and pulmonary infections were independent risk factors for XDR-A. baumannii nosocomial infections in patients with NSDs. Moreover, pulmonary infections, the aspartate aminotransferase (AST) to alanine aminotransferase (ALT) ratio, and the neutrophil-to-lymphocyte ratio (NLR) were significantly associated with increased mortality rates in patients with nosocomial infections caused by XDR-A. baumannii. Thirdly, NLR and cardiovascular diseases accounted for a high risk of mortality for XDR-A. baumannii nosocomial infections in patients with NSDs. The area under the curves of results from each multivariate regression model were 0.827, 0.811, and 0.853, respectively.

Conclusion

This study reveals the risk factors of XDR-A. baumannii nosocomial infections in patients with NSDs, and proves their reliable predictive value. Early recognition of patients at high risk, sterilizing medical tools, and regular blood monitoring are all critical aspects for minimizing the nosocomial spread and mortality of A. baumannii infections.

External validation of a predictive algorithm for in-hospital and ninety-day mortality after spinal epidural abscess

BACKGROUND CONTEXT

Mortality in patients with spinal epidural abscess (SEA) remains high. Accurate prediction of patient-specific prognosis in SEA can improve patient counseling as well as guide management decisions. There are no externally validated studies predicting short-term mortality in patients with SEA.

PURPOSE

The purpose of this study was to externally validate the Skeletal Oncology Research Group (SORG) stochastic gradient boosting algorithm for prediction of in-hospital and 90-day postdischarge mortality in SEA.

STUDY DESIGN/SETTING

Retrospective, case-control study at a tertiary care academic medical center from 2003 to 2021.

PATIENT SAMPLE

Adult patients admitted for radiologically confirmed diagnosis of SEA who did not initiate treatment at an outside institution.

OUTCOME MEASURES

In-hospital and 90-day postdischarge mortality.

METHODS

We tested the SORG stochastic gradient boosting algorithm on an independent validation cohort. We assessed its performance with discrimination, calibration, decision curve analysis, and overall performance.

RESULTS

A total of 212 patients met inclusion criteria, with a short-term mortality rate of 10.4%. The area under the receiver operating characteristic curve (AUROC) of the SORG algorithm when tested on the full validation cohort was 0.82, the calibration intercept was -0.08, the calibration slope was 0.96, and the Brier score was 0.09.

CONCLUSIONS

With a contemporaneous and geographically distinct independent cohort, we report successful external validation of a machine learning algorithm for prediction of in-hospital and 90-day postdischarge mortality in SEA.

An Elevated Neutrophil-to-Lymphocyte Ratio Predicts Poor Prognosis in Patients with Liver Cancer after Interventional Treatments

This study is aimed at examining the prognostic value of blood neutrophil-to-lymphocyte ratio (NLR) in patients with hepatocellular carcinoma (HCC). Demographic and clinical data of 543 HCC patients treated with interventional therapies were retrospectively analyzed. Preoperative NLRs were determined and receiver operating characteristic (ROC) curves were plotted for survival time in patients with high (NLR ≥3.8) and low (NLR<3.8) NLR. The median overall survival (OS) was 1241 days after interventional therapies and was significantly reduced in the high NLR group when compared to the low NLR group. The median progression-free survival time (PFST) of patients was also significantly shorter in the high NLR group than in the low NLR group. Univariate analysis revealed that tumor type, therapy method, maximum tumor size (>3 mm), and NLR (>3.8) were risk factors for OST and PFST ( $P<0.05$ ). Multivariate analysis indicated that tumor type, maximum tumor diameter, therapy method, and NLR (>3.8) were independent risk factors for PFST ( $P<0.05$ ). Our results demonstrate that preoperative NLR has prognostic value for patients with HCC undergoing interventional therapies, and high NLR is an indication of poor prognosis.

Predictors of reoperation after surgery for spinal epidural abscess

BACKGROUND CONTEXT

Spinal epidural abscess is a rare but severe condition with high rates of postoperative adverse events.

PURPOSE

The objective of the study was to identify independent prognostic factors for reoperation using two datasets: an institutional and national database.

STUDY DESIGN/SETTING

Retrospective Review.

PATIENT SAMPLE

Database 1: Review of five medical centers from 1993 to 2016. Database 2: The National Surgical Quality Improvement Program (NSQIP) was queried between 2012 and 2016.

OUTCOME MEASURES

Thirty-day and ninety-day reoperation rate.

METHODS

Two independent datasets were reviewed to identify patients with spinal epidural abscesses undergoing spinal surgery. Multivariate analyses were used to determine independent prognostic factors for reoperation while including factors identified in bivariate analyses.

RESULTS

Overall, 642 patients underwent surgery for a spinal epidural abscess in the institutional cohort, with a 90-day unplanned reoperation rate of 19.9%. In the NSQIP database, 951 patients were identified with a 30-day unplanned reoperation rate of 12.3%. On multivariate analysis in the NSQIP database, cervical spine abscess was the only factor that reached significance for 30-day reoperation (OR=1.71, 95% CI=1.11-2.63, p=.02, Area under the curve (AUC)=0.61). On multivariate analysis in the institutional cohort, independent prognostic factors for 30-day reoperation were: preoperative urinary incontinence, ventral location of abscess relative to thecal sac, cervical abscess, preoperative wound infection, and leukocytosis (AUC=0.65). Ninety-day reoperation rate also found hypoalbuminemia as a significant predictor (AUC=0.66).

CONCLUSION

Six novel independent prognostic factors were identified for 90-day reoperation after surgery for a spinal epidural abscess. The multivariable analysis fairly predicts reoperation, indicating that there may be additional factors that need to be uncovered in future studies. The risk factors delineated in this study through the use of two large cohorts of spinal epidural abscess patients can be used to improve preoperative risk stratification and patient management.

Synovial Fluid and Serum Neutrophil-to-Lymphocyte Ratio: Novel Biomarkers for the Diagnosis and Prognosis of Native Septic Arthritis in Adults

BACKGROUND

Septic arthritis (SA) is a musculoskeletal emergency for which prompt diagnosis and treatment are critical. However, traditional diagnostic criteria of a synovial fluid (SF) white blood-cell count (WBC) of >50,000 cells/mm 3 or >90% polymorphonuclear leukocytes (%PMN) are not particularly sensitive or specific for the diagnosis of SA. Furthermore, prognostic markers are lacking. The purposes of this study were to assess the discriminative ability of the SF neutrophil-to-lymphocyte ratio (NLR) in the diagnosis of SA and of the serum NLR in the prognosis of SA.

METHODS

A multi-institution, retrospective study of 598 patients with native shoulder, hip, or knee SA in 2000 to 2018 was conducted. SF-NLR was calculated from the arthrocentesis cell count with differential. Receiver operating characteristic curves were analyzed, and the optimal threshold of SF-NLR for SA diagnosis was determined using the Youden index. Results were compared with traditional SF diagnostic criteria. Similar analyses assessed the association of serum NLR with 90-day treatment failure and mortality for the subset of patients with confirmed hip or knee SA and with serum complete blood-cell counts with differentials (n = 235). Results were compared with traditional serum prognostic markers (WBC, C-reactive protein [CRP], and erythrocyte sedimentation rate [ESR]).

RESULTS

The SF-NLR (area under the receiver operating characteristic curve [AUC], 0.85 [95% confidence interval (CI), 0.82 to 0.88]) was significantly more accurate for an SA diagnosis than SF-WBC (AUC, 0.80 [95% CI, 0.76 to 0.83]; p = 0.002) and SF-%PMN (AUC, 0.81 [95% CI, 0.77 to 0.84]; p = 0.01). The optimal threshold of SF-NLR was 25 (78% sensitivity and 81% specificity), compared with >50,000 cells/mm 3 for SF-WBC (56% sensitivity and 80% specificity) and >90% for SF-%PMN (65% sensitivity and 78% specificity). Elevated serum NLR was independently associated with 90-day treatment failure (odds ratio [OR], 7.04 [95% CI, 3.78 to 13.14]; p < 0.001) and mortality (OR, 7.33 [95% CI, 2.00 to 26.92]; p = 0.003); elevated serum WBC and CRP were also associated with treatment failure, and WBC, CRP, and ESR were not associated with mortality.

CONCLUSIONS

This study provides compelling data on the superior diagnostic and prognostic ability of serum NLR and SF-NLR for SA compared with current clinical standards. Given that this biomarker requires no additional cost or time to return than current laboratory tests already being performed, pending validation, it can readily be used to aid clinicians in the diagnosis and prognostication of SA.

LEVEL OF EVIDENCE

Diagnostic Level IV . See Instructions for Authors for a complete description of levels of evidence.

C-reactive Protein-to-albumin Ratio in Spinal Epidural Abscess: Association with Post-treatment Complications

INTRODUCTION

Spinal epidural abscess (SEA) is a complex medical condition with high morbidity and healthcare costs. Clinical presentation and laboratory data may have prognostic value in forecasting morbidity and mortality. C-reactive protein-to-albumin ratio (CAR) demonstrates promise for the prediction of adverse events in multiple orthopaedic and nonorthopaedic surgical conditions. We investigated the relationship between CAR and outcomes after treatment of SEA.

METHODS

We retrospectively evaluated adult patients treated within a single healthcare system for a diagnosis of SEA (2005 to 2017). Laboratory and clinical data included age at diagnosis, sex, race, body mass index, smoking status, history of intravenous drug use, Charlson Comorbidity Index, and CAR. The primary outcome was the occurrence of any complication; mortality and readmissions were considered secondarily. We used logistic regression to determine the association between baseline CAR and outcomes, adjusting for confounders.

RESULTS

We included 362 patients with a 90-day mortality rate of 13.3% and a 90-day complication rate of 47.8%. A reduced complication rate was observed in the lowest decile of CAR values compared with the remaining 90% of patients, a threshold value of 2.5 (27.0% versus 50.2%; odds ratio [OR] 2.66, 95% confidence interval [CI] 1.22 to 5.81). CAR values in the highest two deciles experienced significantly increased odds of complications compared with the lowest decile (80th: OR 3.44; 95% CI 1.25 to 9.42; 90th: OR 3.28; 95% CI 1.19 to 9.04).

DISCUSSION

We found elevated CAR to be associated with an increased likelihood of major morbidity in SEA. We suggest using a CAR value of 2.5 as a threshold for enhanced surveillance and recognizing patients with values above 73.7 as being at exceptional risk of morbidity.

LEVEL OF EVIDENCE

Level III observational cohort study.

Systemic immune-inflammation index is associated with disease activity in patients with ankylosing spondylitis

Background

The systemic immune‐inflammation index (SII) is a recently developed indicator for systemic inflammatory response. We aimed to explore the association between SII and disease activity in patients with ankylosing spondylitis (AS).

Methods

This retrospective study included 136 patients with AS and 63 healthy controls. Patients were divided into two groups according to Bath Ankylosing Spondylitis Disease Activity Index (BASDAI); active group (n = 60) and remission group (n = 76). Clinical, laboratory, and demographic characteristics were recorded. Spearman's correlation analysis was used to determine correlations of SII with C‐reactive protein (CRP) level, erythrocyte sedimentation rate (ESR), and BASDAI in AS patients. Binary logistic regression analysis was used to assess risk factors for AS disease activity. Receiver operating characteristic curve analysis was used to evaluate the diagnostic value of SII and the above variables for the active group compared with the remission group.

Results

Systemic immune‐inflammation index levels were higher in AS patients than in healthy controls (p < 0.001). SII levels were higher in the active group than in the remission group (p < 0.001). For patients with AS, SII correlated positively with CRP (r_s = 0.483, p < 0.001), ESR (r_s = 0.374, p < 0.001), and BASDAI (r_s = 0.667, p < 0.001). SII (OR = 1.009, 95% CI = 1.006–1.012, p < 0.001) was an independent risk factor affecting AS disease activity. The specificity and sensitivity of SII using a cutoff value of 513.2 were 83.33% and 86.84%, respectively, for the active group.

Conclusion

Systemic immune‐inflammation index was increased in AS. SII may be a novel indicator for monitoring AS disease activity.

Spinal Epidural Abscess: Diagnosis, Management, and Outcomes

An infection of the spinal epidural space, spinal epidural abscess (SEA) is a potentially devastating entity that is rising in incidence. Its insidious presentation, variable progression, and potential for precipitous neurologic decline make diagnosis and management of SEA challenging. Prompt diagnosis is key because treatment delay can lead to paralysis or death. Owing to the nonspecific symptoms and signs of SEA, misdiagnosis is alarmingly common. Risk factor assessment to determine the need for definitive MRI reduces diagnostic delays compared with relying on clinical or laboratory findings alone. Although decompression has long been considered the benchmark for SEA, considerable risk associated with spinal surgery is noted in an older cohort with multiple comorbidities. Nonoperative management may represent an alternative in select cases. Failure of nonoperative management is a feared outcome associated with motor deterioration and poor clinical outcomes. Recent studies have identified independent predictors of failure and residual neurologic dysfunction, recurrence, and mortality. Importantly, these studies provide tools that generate probabilities of these outcomes. Future directions of investigation should include external validation of existing algorithms through multi-institutional collaboration, prospective trials, and incorporation of powerful predictive statistics such as machine learning methods.

The Prognostic Value of Laboratory Markers and Ambulatory Function at Presentation for Post-Treatment Morbidity and Mortality Following Epidural Abscess

STUDY DESIGN

Retrospective cohort study.

OBJECTIVE

To develop a comprehensive understanding of the prognostic value of laboratory markers on morbidity and mortality following epidural abscess.

SUMMARY OF BACKGROUND DATA

Spinal epidural abscess is a serious medical condition with high rates of morbidity. The value of laboratory data in forecasting morbidity and mortality after epidural abscess remains underexplored.

METHODS

We obtained clinical data on patients treated for epidural abscess at two academic centers from 2005 to 2017. Our primary outcome was the development of one or more complications within 90-days of presentation, with mortality a secondary measure. Primary predictors included serum albumin, serum creatinine, platelet-lymphocyte ratio, and ambulatory status at presentation. We used multivariable logistic regression techniques to adjust for confounders. The most parsimonious set of variables influencing both complications and mortality were considered to be clinically significant. These were then examined individually and in combination to assess for synergy along with model-discrimination and calibration. We performed internal validation with a bootstrap procedure using sampling with replacement.

RESULTS

We included 449 patients in this analysis. Complications were encountered in 164 cases (37%). Mortality within 1-year occurred in 39 patients (9%). Regression testing determined that serum albumin, serum creatinine, and ambulatory status at presentation were clinically important predictors of outcome, with albumin more than 3.5 g/dL, creatinine less than or equal to 1.2 mg/dL, and independent ambulatory function at presentation considered favorable characteristics. Patients with no favorable findings had increased likelihood of 90-day complications (odds ratio [OR] 5.43; 95% confidence intervals [CI] 1.98, 14.93) and 1-year mortality (OR 8.94; 95% CI 2.03, 39.37). Those with one favorable characteristic had greater odds of complications (OR 4.00; 95% CI 2.05, 7.81) and mortality (OR 5.71; 95% CI 1.60, 20.43).

CONCLUSION

We developed a nomogram incorporating clinical and laboratory values to prognosticate outcomes after treatment for epidural abscess. The results can be used in shared-decision making and counseling.

LEVEL OF EVIDENCE

3.

Multiple preoperative biomarkers are associated with incidence of surgical site infection following surgeries of ankle fractures

The aim of the study was to investigate the epidemiologic characteristics of surgical site infection (SSI) following surgeries of ankle fractures. This was a retrospective study. Patients who underwent surgeries for ankle fractures in our hospital between January 2016 and June 2019 were included. Inpatient medical records were inquired for data collection, including demographics, comorbidities, injury-related data, laboratory biomarkers, and confirmation of the SSI cases. Univariate analyses and multivariate logistic regression analyses were used to identify the independent risk factors. Among the 1532 patients, 45 had a postoperative SSI, indicating the incidence rate of 2.9%. About 18% of SSIs were identified after discharge. Twenty percent of SSIs were caused by mixed bacteria, and 39% were caused by drug-resistant bacteria. In the final multivariate model, 7 factors including 5 biomarkers were identified to be independently associated with SSI: gender (male vs female, OR, 2.69; 95% CI, 1.33-4.76), perioperative blood transfusion (OR. 3.02; 95% CI, 1.30-7.04), albumin <35 g/L (OR, 2.87; 95% CI, 1.31-6.31), lower high-density lipoprotein cholesterol (HDL-C) (OR, 2.34; 95% CI, 1.19-4.60), haemoglobin (OR, 2.16; 95% CI, 1.03-4.67), elevated alanine aminotransferase (OR, 2.09; 95% CI, 1.10-3.95) and neutrophile/lymphocyte rate (NLR, OR, 3.45; 95% CI, 1.33-6.74). These epidemiologic data on SSI may help counsel patients about the risk of SSI, individualised assessment of the risk factors, and accordingly the risk stratification.

Prognostic value of laboratory values in older patients with cervical spine fractures

OBJECTIVES

To understand the prognostic value of laboratory markers at presentation on post-treatment survival of patients 50 and older following cervical spine fracture.

PATIENTS AND METHODS

We obtained clinical data on patients 50 and older treated for cervical spine fracture in a single healthcare system (2006-2016). Our primary outcome consisted of 1-year mortality, with mortality within 3-months of presentation considered secondarily. Our primary predictors included serum glucose, serum creatinine, platelet-lymphocyte ratio (PLR) and neutrophil-lymphocyte ratio (NLR) at presentation. We used multivariable logistic regression to adjust for confounding from sociodemographic and clinical characteristics. Point estimates and 95 % confidence intervals (CI) from the final model were refined using Bayesian regression techniques.

RESULTS

We included 1781 patients in this analysis, with an average age of 75.3 (SD 12.0). The mortality rate at 3-months was 12 % and 17 % at 1-year. In multivariable testing, neither elevated PLR or NLR were significant predictors of 1-year mortality. Elevated serum creatinine was associated with increased mortality at 1-year (OR 1.89; 95 % CI 1.30, 2.74), as was hyperglycemia (OR 1.50; 95 % CI 1.06, 2.13). Elevated serum creatinine remained influential (OR 1.64; 95 % CI 1.06, 2.54) on mortality at 3-months.

CONCLUSIONS

This is the first study to evaluate laboratory values at presentation in conjunction with survival following cervical fractures. The results can be used to help forecast natural history and in expectation management. They may also help formulate treatment plans, especially when the need for surgical intervention is not clearly defined.

Development of machine learning algorithms for prediction of mortality in spinal epidural abscess

BACKGROUND CONTEXT

In-hospital and short-term mortality in patients with spinal epidural abscess (SEA) remains unacceptably high despite diagnostic and therapeutic advancements. Forecasting this potentially avoidable consequence at the time of admission could improve patient management and counseling. Few studies exist to meet this need, and none have explored methodologies such as machine learning.

PURPOSE

The purpose of this study was to develop machine learning algorithms for prediction of in-hospital and 90-day postdischarge mortality in SEA.

STUDY DESIGN/SETTING

Retrospective, case-control study at two academic medical centers and three community hospitals from 1993 to 2016.

PATIENTS SAMPLE

Adult patients with an inpatient admission for radiologically confirmed diagnosis of SEA.

OUTCOME MEASURES

In-hospital and 90-day postdischarge mortality.

METHODS

Five machine learning algorithms (elastic-net penalized logistic regression, random forest, stochastic gradient boosting, neural network, and support vector machine) were developed and assessed by discrimination, calibration, overall performance, and decision curve analysis.

RESULTS

Overall, 1,053 SEA patients were identified in the study, with 134 (12.7%) experiencing in-hospital or 90-day postdischarge mortality. The stochastic gradient boosting model achieved the best performance across discrimination, c-statistic=0.89, calibration, and decision curve analysis. The variables used for prediction of 90-day mortality, ranked by importance, were age, albumin, platelet count, neutrophil to lymphocyte ratio, hemodialysis, active malignancy, and diabetes. The final algorithm was incorporated into a web application available here: https://sorg-apps.shinyapps.io/seamortality/.

CONCLUSIONS

Machine learning algorithms show promise on internal validation for prediction of 90-day mortality in SEA. Future studies are needed to externally validate these algorithms in independent populations.