Clinical Prediction Models for Prognosis of Colorectal Liver Metastases: A Comprehensive Review of Regression-Based and Machine Learning Models

Colorectal liver metastasis (CRLM) is a disease entity that warrants special attention due to its high frequency and potential curability. Identification of "high-risk" patients is increasingly popular for risk stratification and personalization of the management pathway. Traditional regression-based methods have been used to derive prediction models for these patients, and lately, focus has shifted to artificial intelligence-based models, with employment of variable supervised and unsupervised techniques. Multiple endpoints, like overall survival (OS), disease-free survival (DFS) and development or recurrence of postoperative complications have all been used as outcomes in these studies. This review provides an extensive overview of available clinical prediction models focusing on the prognosis of CRLM and highlights the different predictor types incorporated in each model. An overview of the modelling strategies and the outcomes chosen is provided. Specific patient and treatment characteristics included in the models are discussed in detail. Model development and validation methods are presented and critically appraised, and model performance is assessed within a proposed framework.

Goodbye Hartmann trial: a prospective, international, multicenter, observational study on the current use of a surgical procedure developed a century ago

BACKGROUND

Literature suggests colonic resection and primary anastomosis (RPA) instead of Hartmann's procedure (HP) for the treatment of left-sided colonic emergencies. We aim to evaluate the surgical options globally used to treat patients with acute left-sided colonic emergencies and the factors that leading to the choice of treatment, comparing HP and RPA.

METHODS

This is a prospective, international, multicenter, observational study registered on ClinicalTrials.gov. A total 1215 patients with left-sided colonic emergencies who required surgery were included from 204 centers during the period of March 1, 2020, to May 31, 2020. with a 1-year follow-up.

RESULTS

564 patients (43.1%) were females. The mean age was 65.9 ± 15.6 years. HP was performed in 697 (57.3%) patients and RPA in 384 (31.6%) cases. Complicated acute diverticulitis was the most common cause of left-sided colonic emergencies (40.2%), followed by colorectal malignancy (36.6%). Severe complications (Clavien-Dindo ≥ 3b) were higher in the HP group (P < 0.001). 30-day mortality was higher in HP patients (13.7%), especially in case of bowel perforation and diffused peritonitis. 1-year follow-up showed no differences on ostomy reversal rate between HP and RPA. (P = 0.127). A backward likelihood logistic regression model showed that RPA was preferred in younger patients, having low ASA score (≤ 3), in case of large bowel obstruction, absence of colonic ischemia, longer time from admission to surgery, operating early at the day working hours, by a surgeon who performed more than 50 colorectal resections.

CONCLUSIONS

After 100 years since the first Hartmann's procedure, HP remains the most common treatment for left-sided colorectal emergencies. Treatment's choice depends on patient characteristics, the time of surgery and the experience of the surgeon. RPA should be considered as the gold standard for surgery, with HP being an exception.

Development and internal validation of a clinical prediction model for serious complications after emergency laparotomy

PURPOSE

Emergency laparotomy (EL) is a common operation with high risk for postoperative complications, thereby requiring accurate risk stratification to manage vulnerable patients optimally. We developed and internally validated a predictive model of serious complications after EL.

METHODS

Data for eleven carefully selected candidate predictors of 30-day postoperative complications (Clavien-Dindo grade >  = 3) were extracted from the HELAS cohort of EL patients in 11 centres in Greece and Cyprus. Logistic regression with Least Absolute Shrinkage and Selection Operator (LASSO) was applied for model development. Discrimination and calibration measures were estimated and clinical utility was explored with decision curve analysis (DCA). Reproducibility and heterogeneity were examined with Bootstrap-based internal validation and Internal-External Cross-Validation. The American College of Surgeons National Surgical Quality Improvement Program's (ACS-NSQIP) model was applied to the same cohort to establish a benchmark for the new model.

RESULTS

From data on 633 eligible patients (175 complication events), the SErious complications After Laparotomy (SEAL) model was developed with 6 predictors (preoperative albumin, blood urea nitrogen, American Society of Anaesthesiology score, sepsis or septic shock, dependent functional status, and ascites). SEAL had good discriminative ability (optimism-corrected c-statistic: 0.80, 95% confidence interval [CI] 0.79-0.81), calibration (optimism-corrected calibration slope: 1.01, 95% CI 0.99-1.03) and overall fit (scaled Brier score: 25.1%, 95% CI 24.1-26.1%). SEAL compared favourably with ACS-NSQIP in all metrics, including DCA across multiple risk thresholds.

CONCLUSION

SEAL is a simple and promising model for individualized risk predictions of serious complications after EL. Future external validations should appraise SEAL's transportability across diverse settings.

Artificial Intelligence in Surgical Risk Prediction

Risk prediction and stratification of short-term and long-term postoperative outcomes are growing in importance and scope of application in everyday clinical practice [...].

Prospective multicenter external validation of postoperative mortality prediction tools in patients undergoing emergency laparotomy

BACKGROUND

Accurate preoperative risk assessment in emergency laparotomy (EL) is valuable for informed decision making and rational use of resources. Available risk prediction tools have not been validated adequately across diverse health care settings. Herein, we report a comparative external validation of four widely cited prognostic models.

METHODS

A multicenter cohort was prospectively composed of consecutive patients undergoing EL in 11 Greek hospitals from January 2020 to May 2021 using the National Emergency Laparotomy Audit (NELA) inclusion criteria. Thirty-day mortality risk predictions were calculated using the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP), NELA, Portsmouth Physiological and Operative Severity Score for the Enumeration of Mortality and Morbidity (P-POSSUM), and Predictive Optimal Trees in Emergency Surgery Risk tools. Surgeons' assessment of postoperative mortality using predefined cutoffs was recorded, and a surgeon-adjusted ACS-NSQIP prediction was calculated when the original model's prediction was relatively low. Predictive performances were compared using scaled Brier scores, discrimination and calibration measures and plots, and decision curve analysis. Heterogeneity across hospitals was assessed by random-effects meta-analysis.

RESULTS

A total of 631 patients were included, and 30-day mortality was 16.3%. The ACS-NSQIP and its surgeon-adjusted version had the highest scaled Brier scores. All models presented high discriminative ability, with concordance statistics ranging from 0.79 for P-POSSUM to 0.85 for NELA. However, except the surgeon-adjusted ACS-NSQIP (Hosmer-Lemeshow test, p = 0.742), all other models were poorly calibrated ( p < 0.001). Decision curve analysis revealed superior clinical utility of the ACS-NSQIP. Following recalibrations, predictive accuracy improved for all models, but ACS-NSQIP retained the lead. Between-hospital heterogeneity was minimum for the ACS-NSQIP model and maximum for P-POSSUM.

CONCLUSION

The ACS-NSQIP tool was most accurate for mortality predictions after EL in a broad external validation cohort, demonstrating utility for facilitating preoperative risk management in the Greek health care system. Subjective surgeon assessments of patient prognosis may optimize ACS-NSQIP predictions.

LEVEL OF EVIDENCE

Diagnostic Test/Criteria; Level II.

The Hellenic Emergency Laparotomy Study (HELAS): A Prospective Multicentre Study on the Outcomes of Emergency Laparotomy in Greece

BACKGROUND

Emergency laparotomy (EL) is accompanied by high post-operative morbidity and mortality which varies significantly between countries and populations. The aim of this study is to report outcomes of emergency laparotomy in Greece and to compare them with the results of the National Emergency Laparotomy Audit (NELA).

METHODS

This is a multicentre prospective cohort study undertaken between 01.2019 and 05.2020 including consecutive patients subjected to EL in 11 Greek hospitals. EL was defined according to NELA criteria. Demographics, clinical variables, and post-operative outcomes were prospectively registered in an online database. Multivariable logistic regression analysis was used to identify independent predictors of post-operative mortality.

RESULTS

There were 633 patients, 53.9% males, ASA class III/IV 43.6%, older than 65 years 58.6%. The most common operations were small bowel resection (20.5%), peptic ulcer repair (12.0%), adhesiolysis (11.8%) and Hartmann's procedure (11.5%). 30-day post-operative mortality reached 16.3% and serious complications occurred in 10.9%. Factors associated with post-operative mortality were increasing age and ASA class, dependent functional status, ascites, severe sepsis, septic shock, and diabetes. HELAS cohort showed similarities with NELA patients in terms of demographics and preoperative risk. Post-operative utilisation of ICU was significantly lower in the Greek cohort (25.8% vs 56.8%) whereas 30-day post-operative mortality was significantly higher (16.3% vs 8.7%).

CONCLUSION

In this study, Greek patients experienced markedly worse mortality after emergency laparotomy compared with their British counterparts. This can be at least partly explained by underutilisation of critical care by surgical patients who are at high risk for death.

Complications of modern pancreaticoduodenectomy: A systematic review and meta-analysis

BACKGROUND

In the past decades, the perioperative management of patients undergoing pancreaticoduodenectomy (PD) has undergone major changes worldwide. This review aimed to systematically determine the burden of complications of PD performed in the last 10 years.

DATA SOURCES

A systematic review was conducted in PubMed for randomized controlled trials and observational studies reporting postoperative complications in at least 100 PDs from January 2010 to April 2020. Risk of bias was assessed using the Cochrane RoB2 tool for randomized studies and the methodological index for non-randomized studies (MINORS). Pooled complication rates were estimated using random-effects meta-analysis. Heterogeneity was investigated by subgroup analysis and meta-regression.

RESULTS

A total of 20 randomized and 49 observational studies reporting 63 229 PDs were reviewed. Mean MINORS score showed a high risk of bias in non-randomized studies, while one quarter of the randomized studies were assessed to have high risk of bias. Pooled incidences of 30-day mortality, overall complications and serious complications were 1.7% (95% CI: 0.9%-2.9%; I² = 95.4%), 54.7% (95% CI: 46.4%-62.8%; I² = 99.4%) and 25.5% (95% CI: 21.8%-29.4%; I²= 92.9%), respectively. Clinically-relevant postoperative pancreatic fistula risk was 14.3% (95% CI: 12.4%-16.3%; I² = 92.0%) and mean length of stay was 14.8 days (95% CI: 13.6-16.1; I² = 99.3%). Meta-regression partially attributed the observed heterogeneity to the country of origin of the study, the study design and the American Society of Anesthesiologists class.

CONCLUSIONS

Pooled complication rates estimated in this study may be used to counsel patients scheduled to undergo a PD and to set benchmarks against which centers can audit their practice. However, cautious interpretation is necessary due to substantial heterogeneity.

Vomiting-induced costal cartilage fracture: a case report

The use of ultrasonography as a first line imaging test in cases of possible costal cartilage fracture can be pivotal. In this case report, we present the case of a patient with a suspected atraumatic vomiting-induced costal cartilage fracture. The costal cartilage fracture was non-displaced and incomplete, thus not visible in a Computed Tomography scan. When Ultrasound imaging was employed at the area of tenderness, soft tissue edema and hematoma around the cartilage were visualized. High level of suspicion for a cartilage fracture in this case revealed a subtle osseous injury.

Maximizing patient benefit through a reversed pathway from specialist to generalist: the case of chronic pain

Polymyalgia Rheumatica (PMR) is a syndrome characterized by chronic pain and/or stiffness in the neck, shoulders or upper arms and hips. It affects adult patients usually over 50 years old and is treated with low-dose oral corticosteroids. In this case, a 68-year-old female with a history of PMR, diagnosed by a specialist sporadically seen in the past, presented to a primary care physician due to herpes zoster (HZ) infection. Thorough history taking, along with a careful review of previous laboratory results, raised serious doubts concerning her diagnosis (PMR). Because the patient described diffuse pain throughout her body, sleep disturbances and a depressed emotional state, fibromyalgia was suspected instead and appropriate treatment was given. The patient remained free of symptoms and corticosteroids for almost a year. Information from this case may help to point out that PMR is a disorder that can be easily confused with other chronic pain conditions with similar manifestations, especially when the initial diagnosis is sped up in terms of consultation depth and care continuity. Under certain circumstances, primary care can lead to improved clinical outcomes.

A case of incidental infantile gallbladder adenomyomatosis: an unusual US finding of uncertain clinical significance

Adenomyomatosis of the gallbladder is defined as hypertrophy of the gallbladder mucosal epithelium that invaginates into a thickened muscularis propria, leading to the formation of intramural diverticula. It is typically considered a benign condition most commonly affecting adults and, rarely, children. In this case report, we present a case of gallbladder adenomyomatosis in a 3-month-old infant. The diagnosis was made incidentally on ultrasound examination of the abdomen, in an otherwise asymptomatic child with no laboratory test abnormalities and no underlying disease. The purpose of this case report is to make infantile adenomyomatosis and its implications known to clinicians, as the literature on this topic is limited. To the best of our knowledge, this is the youngest reported case of adenomyomatosis in the pediatric population.

An FDA phase I clinical trial of quinacrine sterilization (QS)

OBJECTIVE

To review the significance of a United States Food and Drug Administration (FDA) approved Phase I clinical trial of a new use for an old drug, quinacrine. To discover whether ultrasound may have utility in quinacrine sterilization (QS).

METHOD

This clinical trial began on 16 September 2000 at the Women's and Children's Hospital of Buffalo (WCHOB) in Buffalo, New York. Ten patients volunteered to have QS. These subjects were carefully followed with regularly scheduled examinations, including extensive laboratory blood tests. In addition, each patient had a trans-abdominal ultrasound examination six weeks or later past the date of the second insertion of quinacrine. The trial was completed on 30 April 2003.

RESULTS

Laboratory results fell within normal limits, thus providing additional evidence to affirm the lack of toxic effects of QS. With ultrasound, we were able to see scars in both oviducts on all of our patients. One patient with a small scar as seen on ultrasound became pregnant.

CONCLUSION

QS was found to be safe and effective. Ultrasound holds the promise of reducing the failure rate.