Preoperative Stratification Of Patients Research Articles

Machine learning model for prediction of permanent stoma after anterior resection of rectal cancer: A multicenter study

BackgroundThe conversion from a temporary to a permanent stoma (PS) following rectal cancer surgery significantly impacts the quality of life of patients. However, there is currently a lack of practical preoperative tools to predict PS formation. The purpose of this study is to establish a preoperative predictive model for PS using machine learning algorithms to guide clinical practice. MethodsIn this retrospective study, we analyzed clinical data from a total of 655 patients who underwent anterior resection for rectal cancer, with 552 patients from one medical center and 103 from another. Through machine learning algorithms, five predictive models were developed, and each was thoroughly evaluated for predictive performance. The model with superior predictive accuracy underwent additional validation using both an independent testing cohort and the external validation cohort. The Shapley Additive exPlanations (SHAP) approach was employed to elucidate the predictive factors influencing the model, providing an in-depth visual analysis of its decision-making process. ResultsEight variables were selected for the construction of the model. The support vector machine (SVM) model exhibited superior predictive performance in the training set, evidenced by an AUC of 0.854 (95 % CI:0.803–0.904). This performance was corroborated in both the testing set and external validation set, where the model demonstrated an AUC of 0.851 (95%CI:0.748–0.954) and 0.815 (95%CI:0.710–0.919), respectively, indicating its efficacy in identifying the PS. ConclusionsThe model(https://yangsu2023.shinyapps.io/psrisk/) indicated robust predictive performance in identifying PS after anterior resection for rectal cancer, potentially guiding surgeons in the preoperative stratification of patients, thus informing individualized treatment plans and improving patient outcomes.

NIMG-52. RADIOGENOMICS SIGNATURES IN KEY DRIVER GENES IN GLIOBLASTOMA EVALUATED WITH AND WITHOUT THE PRESENCE OF CO-OCCURRING MUTATIONS

Abstract PURPOSE Glioblastomas display significant heterogeneity on the molecular level, typically harboring several co-occurring mutations, which likely contributes to failure of molecularly targeted therapeutic approaches. Radiogenomics has emerged as a promising tool for in vivo characterization of this heterogeneity. We derive radiogenomic signatures of four mutations via machine learning (ML) analysis of multiparametric MRI (mpMRI) and evaluate them in the presence and absence of other co-occurring mutations. METHODS We identified a retrospective cohort of 359 IDH-wildtype glioblastoma patients, with available pre-operative mpMRI (T1, T1Gd, T2, T2-FLAIR) scans and targeted next generation sequencing (NGS) data. Radiomic features, including morphologic, histogram, texture, and Gabor wavelet descriptors, were extracted from the mpMRI. Multivariate predictive models were trained using cross-validated SVM with LASSO feature selection to predict mutation status in key driver genes, EGFR, PTEN, TP53, and NF1. ML models and spatial population atlases of genetic mutations were generated for stratification of the tumors (1) with co-occurring mutations versus wildtypes, (2) with exclusive mutations in each driver gene versus the tumors without any mutations in the pathways associated with these genes. RESULTS ML models yielded AUCs of 0.75 (95%CI:0.62-0.88) / 0.87 (95%CI:0.70-1) for co-occurring / exclusive EGFR mutations, 0.69 (95%CI:0.58-0.80) / 0.80 (95%CI:0.61-0.99) for co-occurring / exclusive PTEN mutations, and 0.77 (95%CI:0.65-0.88) / 0.86 (95%CI:0.69-1) for co-occurring / exclusive TP53 cases. Spatial atlases revealed a predisposition of left temporal lobe for NF1 and right frontotemporal region for TP53 in mutually exclusive tumors, which was not observed in the co-occurring mutation atlases. CONCLUSION Our results suggest the presence of distinct radiogenomic signatures of several glioblastoma mutations, which become even more pronounced when respective mutations do not co-occur with other mutations. These in vivo signatures can contribute to pre-operative stratification of patients for molecular targeted therapies, and potentially longitudinal monitoring of mutational changes during treatment.

Biomarker Risk Score Algorithm and Preoperative Stratification of Patients with Pancreatic Cystic Lesions

Pancreatic cysts are incidentally detected in up to 13% of patients undergoing radiographic imaging. Of the most frequently encountered types, mucin-producing (mucinous) pancreatic cystic lesions may develop into pancreatic cancer, while nonmucinous ones have little or no malignant potential. Accurate preoperative diagnosis is critical for optimal management, but has been difficult to achieve, resulting in unnecessary major surgery. Here, we aim to develop an algorithm based on biomarker risk scores to improve risk stratification. Patients undergoing surgery and/or surveillance for a pancreatic cystic lesion, with diagnostic imaging and banked pancreatic cyst fluid, were enrolled in the study after informed consent (n= 163 surgical, 67 surveillance). Cyst fluid biomarkers with high specificity for distinguishing nonmucinous from mucinous pancreatic cysts (vascular endothelial growth factor [VEGF], glucose, carcinoembryonic antigen [CEA], amylase, cytology, and DNA mutation) were selected. Biomarker risk scores were used to design an algorithm to predict preoperative diagnosis. Performance was tested using surgical (retrospective) and surveillance (prospective) cohorts. In the surgical cohort, the biomarker algorithm outperformed the preoperative clinical diagnosis in correctly predicting the final pathologic diagnosis (91% vs 73%; p < 0.000001). Specifically, nonmucinous serous cystic neoplasms (SCN) and mucinous cystic neoplasms (MCN) were correctly classified more frequently by the algorithm than clinical diagnosis (96% vs 30%; p < 0.000008 and 92% vs 69%; p= 0.04, respectively). In the surveillance cohort, the algorithm predicted a preoperative diagnosis with high confidence based on a high biomarker score and/or consistency with imaging from ≥1 follow-up visits. A biomarker risk score-based algorithm was able to correctly classify pancreatic cysts preoperatively. Importantly, this tool may improve initial and dynamic risk stratification, reducing overdiagnosis and underdiagnosis.

Prognostic Imaging Biomarkers for Pancreatic Ductal Adenocarcinoma Extracted from Structured Radiologist Reports

HomeRadiology: Imaging CancerVol. 2, No. 6 PreviousNext Research HighlightsFree AccessPrognostic Imaging Biomarkers for Pancreatic Ductal Adenocarcinoma Extracted from Structured Radiologist ReportsNikita ConsulNikita ConsulNikita ConsulPublished Online:Nov 27 2020https://doi.org/10.1148/rycan.2020209032MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked InEmail Take-Away Points■ Major Focus: Use of a structured reporting template for pancreatic ductal adenocarcinoma to extract potential prognostic and predictive imaging biomarkers.■ Key Result: Tumor-arterial contact, portal vein deformity, larger tumor size, and presence of venous collaterals were found to be associated with decreased overall survival in patients with pancreatic ductal adenocarcinoma.■ Impact: Routine use of the structured reporting template could allow for increased imaging data extraction and preoperative stratification of patients with pancreatic ductal adenocarcinoma based on imaging biomarkers.In this era of precision medicine, biomarkers are instrumental to prognosticate and predict survival in our patients with cancer. Such biomarkers could be advantageous in preoperative cases of pancreatic ductal adenocarcinoma, an aggressive epithelial cell malignancy associated with extremely poor prognosis. While CA-A-19 is a widely recognized serum marker for pancreatic ductal adenocarcinoma, levels of this protein do not have prognostic value alone (1). However, preoperative CT in conjunction with CA-A-19 levels and the physician’s subjective assessment have exhibited predictive value for 2-year survival (2). Consequently, many studies have evaluated the role for imaging in measuring preoperative tumor extent and posttherapy treatment response, recognizing that imaging biomarkers could be used for noninvasive prognostication.In a retrospective study of patients with pancreatic ductal adenocarcinoma who underwent preoperative CT imaging and curative intent surgical resection between 2006 and 2017, Dickinson et al describe multiple imaging features that were associated with decreased overall survival in patients who underwent surgery either with or without neoadjuvant therapy, including tumor contact with the common hepatic artery or other arteries, portal vein deformity, larger tumor size, and the presence of venous collaterals. These features were extracted from the structured American Pancreatic Association (APA)/Society of Abdominal Radiology (SAR) reporting template (3), and therefore the authors conclude that the use of this template may identify imaging features that predict survival. Furthermore, widespread use of the APA/SAR reporting template will equip future studies of imaging biomarkers for the preoperative stratification of patients with pancreatic ductal adenocarcinoma.Highlighted ArticleDickinson SM, McIntyre CA, Schilsky JB, et al. Preoperative CT predictors of survival in patients with pancreatic ductal adenocarcinoma undergoing curative intent surgery. Abdom Radiol 2020. doi: https://doi.org/10.1007/s00261-020-02726-w

Preoperative CT predictors of survival in patients with pancreatic ductal adenocarcinoma undergoing curative intent surgery.

To evaluate the associations between computed tomography (CT) imaging features extracted from the structured American Pancreatic Association (APA)/Society of Abdominal Radiology (SAR) template and overall survival in patients with resected pancreatic ductal adenocarcinoma (PDAC). This retrospective analysis included consecutive patients with PDAC who consented to genomic tumor testing and underwent preoperative imaging and curative intent surgical resection from December 2006 to July 2017. Two radiologists assessed preoperative CT imaging using the APA/SAR PDAC-reporting template. Univariable associations between overall survival and imaging variables were evaluated using Cox proportional hazards regression. The study included 168 patients (66years ± 11; 91 women). 126/168 patients (75%) received upfront surgical resection whereas 42/168 (25%) received neoadjuvant therapy prior to surgical resection. In the entire cohort, features associated with decreased overall survival were tumor arterial contact of any kind (hazard ratio (HR) 1.89, 95% CI 1.13-3.14, p = 0.020), tumor contact with the common hepatic artery (HR 2.33, 95% CI 1.35-4.04, p = 0.009), and portal vein deformity (HR 3.22, 95% CI 1.63-6.37, p = 0.003). In the upfront surgical group, larger tumor size was associated with decreased overall survival (HR 2.30, 95% CI 1.19-4.42, p = 0.013). In the neoadjuvant therapy group, the presence of venous collaterals was the only feature associated with decreased overall survival (HR 2.28, 95% CI 1.04-4.99, p = 0.042). The application of the APA/SAR pancreatic adenocarcinoma reporting template may identify predictors of survival that can aid in preoperative stratification of patients.

Image-Based Molecular Phenotyping of Pancreatic Ductal Adenocarcinoma.

To bridge the translational gap between recent discoveries of distinct molecular phenotypes of pancreatic cancer and tangible improvements in patient outcome, there is an urgent need to develop strategies and tools informing and improving the clinical decision process. Radiomics and machine learning approaches can offer non-invasive whole tumor analytics for clinical imaging data-based classification. The retrospective study assessed baseline computed tomography (CT) from 207 patients with proven pancreatic ductal adenocarcinoma (PDAC). Following expert level manual annotation, Pyradiomics was used for the extraction of 1474 radiomic features. The molecular tumor subtype was defined by immunohistochemical staining for KRT81 and HNF1a as quasi-mesenchymal (QM) vs. non-quasi-mesenchymal (non-QM). A Random Forest machine learning algorithm was developed to predict the molecular subtype from the radiomic features. The algorithm was then applied to an independent cohort of histopathologically unclassifiable tumors with distinct clinical outcomes. The classification algorithm achieved a sensitivity, specificity and ROC-AUC (area under the receiver operating characteristic curve) of 0.84 ± 0.05, 0.92 ± 0.01 and 0.93 ± 0.01, respectively. The median overall survival for predicted QM and non-QM tumors was 16.1 and 20.9 months, respectively, log-rank-test p = 0.02, harzard ratio (HR) 1.59. The application of the algorithm to histopathologically unclassifiable tumors revealed two groups with significantly different survival (8.9 and 39.8 months, log-rank-test p < 0.001, HR 4.33). The machine learning-based analysis of preoperative (CT) imaging allows the prediction of molecular PDAC subtypes highly relevant for patient survival, allowing advanced pre-operative patient stratification for precision medicine applications.

Discharge to Inpatient Care Facility After Vertebroplasty/Kyphoplasty: Incidence, Risk Factors, and Postdischarge Outcomes

Vertebral compression fractures are a common clinical occurrence in elderly individuals with osteoporosis. No current evidence exists on risk factors and clinical impact of discharge to inpatient (IP) care facility after vertebral augmentation procedures. The 2012-2014 American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) database files were queried using Current Procedural Terminology codes for vertebroplasty (22520, 22521, and 22522) and kyphoplasty (22523, 22524, and 22525). Discharge to IP care facility included skilled-care facilities and IP rehabilitation units. A total of 2361 patients were included in the final cohort. Of 2361 patients, 1962 (83.1%) were discharged home and 399 (16.9%) were discharged to an IP care facility. Multivariate analysis identified age ≥65 years (P < 0.001), dependent preoperative functional health status (P < 0.001), hypertension (P= 0.001), preoperative transfusion (P= 0.043), IP admission status (P < 0.001), thoracolumbar procedure versus thoracic-only procedure (P=0.012), and length of stay >1 day (P < 0.001) to be significant predictors for a discharge to an IP care facility. In addition, discharge to an IP care facility was associated with a significant greater risk of 30-day mortality (P=0.001). No significant associations were found with between IP-care discharge destination and any 30-day complication, 30-day readmission, and 30-day reoperation. Discharge to IP care facilities after vertebroplasty/kyphoplasty is associated with a 3.6 times greater odds of mortality as compared with home discharge. Providers can use the risk profile data to better allow preoperative stratification of patients to ensure that discharge location is appropriate to a patient's need to minimize the risk of adverse outcomes.

Diagnostic Accuracy of Risk of Ovarian Malignancy Algorithm (ROMA) in Post-Menopausal Patients with Ovarian Mass.

To determine the diagnostic accuracy of ROMA in postmenopausal women with history of ovarian mass. Observational study. Dr. Ziauddin University Hospital, Karachi, from May 2014 to June 2015. Two hundred and sixty postmenopausal women of 40-65 years of age with ovarian masses, planned for surgery, were included in the study. Their samples were obtained preoperatively and analysed on Abbot Architect i1000 SR immunoassay analyser for quantitative estimation of tumor markers, i.e. HE4 and CA125. By combination of these two tumor markers, ROMA scores were calculated and studied after histopathological verification of masses. Total number of patients were 260, out of which 122 (46.9%) were diagnosed as having ovarian cancer, while 138 (53.0%) were diagnosed as benign condition. Median ROMA score levels in patients with malignant masses were 95.58 (IQR=44.4) as compared to 20.6 (IQR=14) in benign masses. ROMA had sensitivity 92.6% (CI=86.47-96.04), specificity 78.3% (CI=70.09-83.82), positive predictive value 79% (CI=70.87-84.29), negative predictive value 92.3% (CI=86.02-95.9) and positive likelihood ratio 4.26, while negative likelihood ratio 0.1. Diagnostic accuracy of ROMA was 85%, based on ROC curve analysis. ROMA had the highest sensitivity in detecting ovarian carcinoma. ROMA is a very useful diagnostic tool for the preoperative stratification of patients with ovarian masses showing 85% diagnostic accuracy. However, there is need of more studies with homogenous laboratory procedures for HE4 and CA125 assays as well as patients, selection criteria, so we can draw firm conclusion about utility of ROMA in clinical setups.

The preoperative stratification of patients based on renal scan data is unable to predict the functional outcome after partial nephrectomy.

ABSTRACTIntroduction:eGFR-categories are used to predict functional outcome after partial nephrectomy (PN); no study categorized patients according to preoperative renal scan (RS) data. Aim of the study was to evaluate if stratification of patients according to RS is a reliable method to predict minor/major loss of renal function after PN.Materials and Methods:We considered patients who underwent PN and RS pre-/post-PN for T1 tumor in our Institution (2007-2017). Demographics, perioperative and specifically functional data were analysed. On the basis of the baseline Split Renal Function (SRF), patients were stratified into risk-categories: 1) baseline operated-kidney SRF range 45-55%; 2) baseline operated-kidney SRF <45%. Risk categories were analysed with postoperative functional outcome: postoperative operated-kidney SRF decrease below 90% of baseline was considered significant loss of function. Contingency tables and univariate/multivariate regression were analysed looking for independent factors of postoperative functional impairment.Results:224 patients were analysed, 125 (55.8%) maintained >90% of their baseline function. Worse probability of maintaining ≥90 baseline renal function was found in patients with Charlson's Comorbidity Index (CCI≥3) (p=0.004) and patients with PADUA score ≥8 (p=0.023). After stratification by baseline renal function, ischemia was the only independent factor: no effect on patients with poorer baseline renal function. Patients with baseline SRF 45-55% who did not experience ischemia had the highest probability to maintain ≥90% baseline SRF (p=0.028). Ischemia >25 minutes was detrimental (p=0.017).Conclusions:Stratification of patients by SRF before PN is not a reliable predictor of renal functional outcome. Ischemia seems to scarcely influence patients with poorer renal function.

Diagnostic usefulness of the Risk of Ovarian Malignancy Algorithm using the electrochemiluminescence immunoassay for HE4 and the chemiluminescence microparticle immunoassay for CA125.

The present study aimed to investigate the usefulness of the Risk of Ovarian Malignancy Algorithm (ROMA) in the preoperative stratification of patients with ovarian tumors using a novel combination of laboratory tests. The study group (n=619) consisted of 354 premenopausal and 265 postmenopausal patients. The levels of carbohydrate antigen 125 (CA125) and human epididymis protein 4 (HE4) were determined, and ROMA calculations were performed for each pre- and postmenopausal patient. HE4 levels were determined using an electrochemiluminescence immunoassay, while CA125 levels were determined by a chemiluminescence microparticle immunoassay. A contingency table was applied to calculate the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Receiver operating characteristic curves were also constructed, and areas under the curves (AUCs) were compared between the marker determinations and ROMA algorithms. In terms of distinguishing between ovarian cancer and benign disease, the sensitivity of ROMA was 88.3%, specificity was 88.2%, PPV was 75.3% and NPV was 94.9% among all patients. The respective parameters were 71.1, 90.1, 48.2 and 91.1% in premenopausal patients and 93.6, 82.9, 86.6 and 91.6% in postmenopausal patients. The AUC value for the ROMA algorithm was 0.926 for the ovarian cancer vs. benign groups in all patients, 0.813 in premenopausal patients and 0.939 in postmenopausal patients. The respective AUC values were 0.911, 0.879 and 0.934 for CA125; and 0.879, 0.783 and 0.889 for HE4. In this combination, the ROMA algorithm is characterized by an extremely high sensitivity of prediction of ovarian cancer in women with pelvic masses, and may constitute a precise tool with which to support the qualification of patients to appropriate surgical procedures. The ROMA may be useful in diagnosing ovarian endometrial changes in young patients.

Preoperative risk factors for conversion from laparoscopic to open cholecystectomy: A validated risk score derived from a prospective U.K. database of over 8000 patients

Aim: Laparoscopic cholecystectomy (LC) is a common procedure, and several factors increase the risk of open conversion, thereby increasing operating time and hospital stay. Preoperative stratification of patients at risk of conversion has potential benefits, including improved consent and identifying appropriate training cases. The aim of this study was to develop a validated risk score for conversion. Methods: Preoperative patient and disease-related variables were identified from a prospective LC database (CholeS) of 8820 patients, divided into main and validation sets. Preoperative predictors of conversion were identified by multivariable binary logistic regression. A risk score was developed and validated using a forward stepwise approach. Results: 297 procedures (3.4%) were converted. The risk score was derived from six significant predictors: age (p = 0.005), sex (p<0.001), diagnosis (p<0.001), ASA (p<0.001), thick-walled gallbladder (p = 0.04) and CBD diameter (p = 0.004). Testing the score on the validation set yielded an AUROC = 0.766 (p < 0.001), and a score >6 identified high-risk patients (7.1% v. 1.2%) [See Figure 1]. Conclusion: This validated risk score allows preoperative identification of patients at six-fold increased risk of conversion to open cholecystectomy. High-risk patients should be managed by experienced laparoscopic surgeons, and it may be appropriate for very high-risk patients to be treated in a specialist centre.

Preoperative patient stratification results in low rates of occult uterine malignancy in women undergoing uterine surgery and morcellation

Preoperative patient stratification results in low rates of occult uterine malignancy in women undergoing uterine surgery and morcellation

Charlson comorbidity index: an additional prognostic parameter for preoperative glioblastoma patient stratification.

Dismal glioblastoma (GB) patient outcome calls for the elucidation of further reliable predictors of prognosis. Established "biomarkers," age and functional status, employed in today's patient stratification have limits in fingerprinting this heterogeneous tumor entity. We aimed at ascertaining additional prognostic factors that may facilitate patient stratification for surgery. A retrospective review of 233 consecutive adult patients operated on for newly diagnosed GB at a single tertiary institution over a 5-year period (2006-2011) was conducted. Modern defined outcome associating factors recorded included demographics (preoperative age, gender, signs, symptoms, comorbidity status quantified by the Charlson comorbidity index (CCI), functional status computed by the Karnofsky performance scale (KPS)), tumor characteristics (size, location, isocitrate dehydrogenase mutation, and O-6-methylguanine-DNA methyltransferase promoter methylation status), and treatment parameters (volumetrically quantified extent of resection and adjuvant therapy). Survival analysis was performed by the Kaplan-Maier method. Influence of variables was evaluated using log-rank test. Median neuroradiographic evidence of tumor progression was 6 months after surgery (range 0-72). The median overall survival was 9.5 months (range 0-72). Age > 65 years, KPS ≤ 70, and CCI > 3 were significantly associated with both poor OS (each p < 0.0001) and PFS (p < 0.0001, p < 0.001 and p < 0.002), respectively. Also, patients older than 65 years significantly had a CCI > 3 (p < 0.0001). Our data evidence that aside established prognostic parameters (age and KPS) for GB patient outcome, the CCI additionally significantly impacts outcome and may be employed for preoperative patient stratification.

Preoperative Diagnostic Performance of ROMA (Risk of Ovarian Malignancy Algorithm) in Relation to Etiopathogenesis of Epithelial Ovarian Tumors

Background: We studied the usefulness of ROMA for preoperative stratification of patients in relation to the menopausal status, etiopathogenesis of epithelial ovarian tumors and FIGO stage. Material and methods: The study group (n=214) consisted of 116 premenopausal and 98 postmenopausal patients, including 83 with ovarian cancer and 131 with benign lesions. CA125 and HE4 were determined in each pre- and postmenopausal patient. ROC analysis was done to calculate the sensitivity, specificity, PPV, and NPV and a contingency table was applied to assess the usefulness of ROMA. Results: ROC analysis identified AUC (area under curve) as the most valuable component of ROMA (0.921) in the study group with respect to CA125 (0.919) and HE4 (0.855). Sensitivity was highest for CA125 (90.4% for the whole group, 85.7% for premenopausal and 91.7% for postmenopausal patients). Specificity was highest for ROMA with cutoff points determined by us (95.4% for the whole group, 96.8% for premenopausal and 91.7% for postmenopausal patients). AUC, sensitivity, specificity, PPV, and NPV calculated from a contingency table demonstrated the superiority of ROMA with our cutoff points in type II cancers (0.979, 93.3%, 95.4%, 87.5%, 97.7%) and advanced cancers (0.980, 95.1%, 95.4%, 90.7%, 97.7%), compared with type I (0.851, 76.3%, 95.4%, 82.9%, 93.3%) not advanced (0.754, 59.1%, 95.4%, 68.4%, 93.3%) cancers. Conclusions: ROMA is actually a useful diagnostic method for preoperative stratification of patients with a pelvic mass. It performs better in type II and more advanced ovarian cancers. However, its sensitivity, specificity, PPV, and NPV values in type I ovarian cancer make ROMA useful in this group of cancer patients as well.

Predictors for moderate to severe acute postoperative pain after total hip and knee replacement

The ability to identify and focus care to patients at higher risk of moderate to severe postoperative pain should improve analgesia and patient satisfaction, and may affect reimbursement. We undertook this multi-centre cross-sectional study to identify preoperative risk factors for moderate to severe pain after total hip (THR) and knee (TKR) replacement. A total of 897 patients were identified from electronic medical records. Preoperative information and anaesthetic technique was gained by retrospective chart review. The primary outcomes were moderate to severe pain (pain score ≥ 4/10) at rest and with activity on postoperative day one. Logistic regression was performed to identify predictors for moderate to severe pain. Moderate to severe pain was reported by 20 % at rest and 33 % with activity. Predictors for pain at rest were female gender (OR 1.10 with 95 % CI 1.01-1.20), younger age (0.96, 0.94-0.99), increased BMI (1.02, 1.01-1.03), TKR vs. THR (3.21, 2.73-3.78), increased severity of preoperative pain at the surgical site (1.15, 1.03-1.30), preoperative use of opioids (1.63, 1.32-2.01), and general anaesthesia (8.51, 2.13-33.98). Predictors for pain with activity were TKR vs. THR (1.42, 1.28-1.57), increased severity of preoperative pain at the surgical site (1.11, 1.04-1.19), general anaesthesia (9.02, 3.68-22.07), preoperative use of anti-convulsants (1.78, 1.32-2.40) and anti-depressants (1.50, 1.08-2.80), and prior surgery at the surgical site (1.28, 1.05-1.57). Our findings provide clinical guidance for preoperative stratification of patients for more intensive management potentially including education, nursing staffing, and referral to specialised pain management.

Blood transfusion after on-pump coronary artery bypass grafting: focus on modifiable risk factors

Perioperative transfusions are known to increase morbidity and mortality after coronary artery bypass grafting (CABG). The aims of the study were (1) to identify the clinical profile of the patient subset at highest risk from transfusion and (2) to disclose causative relationship and dose-dependency of transfusion on hospital mortality. A prospective observational design was employed on a cohort of 1047 consecutive patients (median age 63.2 ± 9.3, 18.8% female, 30.6% diabetics, 31.9% urgent/emergent, 15.3% with low preoperative left ventricular ejection fraction (LVEF)) who underwent on-pump isolated CABG between January 2004 and December 2007. Univariate and multivariate regression analysis and post-hoc risk stratification, by means of propensity scoring and binary segmentation, were adopted. The following independent risk factors were identified: age, body surface area (BSA), preoperative glomerular filtration rate, preoperative haemoglobin, surgical priority, length of cardiopulmonary bypass, intraoperative haemodilution and early postoperative blood loss. The patient population was stratified in quintiles of transfusional risk, by means of propensity scoring. As to modifiable risk factors, patients in the highest quintiles of risk were those with BSA ( < 1.73, preoperative haemoglobin < 12 g/dl, intraoperative haemoglobin < 8.0 g/dl and those undergoing cardiopulmonary bypass > 90'). Binary segmentation was performed to avoid any association between red cell transfusion and worse outcomes being causative and dose-dependent. A dose-dependent pattern was disclosed, with patients receiving > 5 units being at highest risk. High exposure to blood transfusions may be prevented by preoperative patient stratification and by the close tailoring of management strategies on planning and implementing surgical timing, as well as by cardiopulmonary bypass technique.

The pre-operative stratification of patients with colorectal liver metastases: Computed tomography arterial portography (CTAP) has no added value

Aims The purpose of this study was to determine if Computed Tomography Arterial Portography (CTAP) has additional value to Contrast Enhanced helical CT (CE-CT) in selecting patients for hepatic surgery or Isolated Hepatic Perfusion/systemic chemotherapy. Materials and methods Forty-one patients were included. All CT's were performed in the normal pre-operative work-up of patients with liver metastases in our regular clinical setting and reviewed blinded by a radiologist. For CE-CT and CTAP the number, size (largest diameter) and location of all suspected malignant liver lesions were recorded. The favourable treatment option was determined based on the results of CE-CT and CTAP independently. The therapeutic decision based on CE-CT and CTAP was compared with the definite treatment. For all patients with recorded findings during surgery, consisting of intra-operative ultrasound, liver palpation and histology a standard of reference for lesion detection was available. For these patients detection rates and the fraction of false positive lesions were calculated. Results Twenty-seven patients were treated with hepatic resection and/or RFA. Fourteen patients were treated with chemotherapy, 4 with Isolated Hepatic Perfusion (IHP) and 10 with systemic therapy. Based on the findings on CE-CT 31 patients were classified as surgical candidates and 10 as non-surgical patients. Based on the findings on CTAP, surgery should be the treatment of choice in 29 patients and 12 patients were classified non-surgical. CE-CT and CTAP disagreed in two cases (4.9%). Seventy-four metastases were identified at surgery and pathologically proved. CE-CT and CTAP showed 53 (.72) and 66 (.89) metastases, respectively. Conclusion Despite a significantly higher detection rate for hepatic metastases, CTAP has no added value in the therapeutic stratification in candidates for resection of hepatic metastases of colorectal cancer.

The diagnostic value of nipple discharge cytology in 618 consecutive patients

Aim Preoperative stratification of patients presenting with nipple discharge (ND) according to malignancy risk has proven difficult. Nevertheless, cytological examination is considered to be a diagnostic aid. The aim of this study was to determine its complementary value in clinical decision-making in patients presenting with ND. Methods We retrospectively collected data on macroscopic ND colour, ND cytology, physical examination, mammography, ultrasound and fine-needle aspiration cytology results. On ND cytology, benign diagnoses were considered negative, whereas suspicious and malignant diagnoses were considered positive for malignancy. Results From 1992 to 2006, 618 patients had an ND smear, of those 163 patients had a biopsy. Sensitivity and specificity were 16.7% and 66.1%, respectively. These values were lower when ND was bloody than when ND was non-bloody ( p = 0.66 and p < 0.05 for sensitivity and specificity, respectively). When macroscopically defining bloody ND as positive and non-bloody ND as negative, macroscopic ND colour examination had a remarkably higher sensitivity (60.6 vs. 18.2%, p < 0.001) and only a slightly lower specificity (53.6 vs. 65.0%, p = 0.07) when compared to cytological ND examination. Only 1 malignant lesion was designated positive solely by ND cytology (unique sensitivity (95% CI), 2.8% (0.0–8.4%)) and 3 lesions were correctly classified as negative by ND cytology (unique specificity (95% CI), 1.6%, 0.0–3.7%)). Conclusion Nipple discharge cytology has little complementary diagnostic value. Therefore, its routine use for detection of ND-related breast pathology should be reconsidered carefully. Nipple discharge cytology may redirect patient management well in some cases, but it may confuse work-up in the majority.

Preoperative C-reactive protein levels predict 9-month mortality after coronary artery bypass grafting surgery for the treatment of left main coronary artery stenosis

Preprocedural levels of C-reactive protein predict mid-term mortality after percutaneous coronary intervention for the treatment of unprotected left main coronary artery stenosis. However, there are no data regarding the impact of C-reactive protein on mid-term mortality in patients with unprotected left main coronary artery stenosis treated with coronary artery bypass graft. The predictive value of preoperative C-reactive protein levels, leukocyte counts, and fibrinogen levels were evaluated in a series of 108 patients who underwent coronary artery bypass graft surgery at our Institution from 1st January 2002 to 31st April 2005. Patients were divided in two groups: Group 1 included patients with C-reactive protein levels in quartiles IV (C-reactive protein levels > or =1.22mg/dl) and Group 2 included patients with C-reactive protein levels in quartiles I+II+III. At 9-month follow-up the rate of mortality was 25.9% in Group 1 and 4.9% in Group 2 (hazard ratio=5.86, 95% confidence intervals=1.71-20.03; p=0.005). In all patients who had cardiac mortality, C-reactive protein levels were >0.5mg/dl. In the multivariate analysis age >75 years, peripheral vascular disease and C-reactive protein quartiles were the only independent predictors of mortality. Elevated preoperative levels of C-reactive protein indicate an increased risk of death after coronary artery bypass graft surgery for the treatment of unprotected left main coronary artery stenosis. Inflammatory risk assessment in patients with unprotected left main coronary artery stenosis provides incremental prognostic value for adequate preoperative patient stratification.