Clinical application of the “sellar barrier’s concept” for predicting intraoperative CSF leak in endoscopic endonasal surgery for pituitary adenomas with a machine learning analysis

Background

Recently, it was defined that the sellar barrier entity could be identified as a predictor of cerebrospinal fluid (CSF) intraoperative leakage. The aim of this study is to validate the application of the sellar barrier concept for predicting intraoperative CSF leak in endoscopic endonasal surgery for pituitary adenomas with a machine learning approach.

Methods

We conducted a prospective cohort study, from June 2019 to September 2020: data from 155 patients with pituitary subdiaphragmatic adenoma operated through endoscopic approach at the Division of Neurosurgery, Università degli Studi di Napoli “Federico II,” were included. Preoperative magnetic resonance images (MRI) and intraoperative findings were analyzed. After processing patient data, the experiment was conducted as a novelty detection problem, splitting outliers (i.e., patients with intraoperative fistula, n = 11/155) and inliers into separate datasets, the latter further separated into training (n = 115/144) and inlier test (n = 29/144) datasets. The machine learning analysis was performed using different novelty detection algorithms [isolation forest, local outlier factor, one-class support vector machine (oSVM)], whose performance was assessed separately and as an ensemble on the inlier and outlier test sets.

Results

According to the type of sellar barrier, patients were classified into two groups, i.e., strong and weak barrier; a third category of mixed barrier was defined when a case was neither weak nor strong. Significant differences between the three datasets were found for Knosp classification score (p = 0.0015), MRI barrier: strong (p = 1.405 × 10⁻⁶), MRI barrier: weak (p = 4.487 × 10⁻⁸), intraoperative barrier: strong (p = 2.788 × 10⁻⁷), and intraoperative barrier: weak (p = 2.191 × 10⁻¹⁰). We recorded 11 cases of intraoperative leakage that occurred in the majority of patients presenting a weak sellar barrier (p = 4.487 × 10⁻⁸) at preoperative MRI. Accuracy, sensitivity, and specificity for outlier detection were 0.70, 0.64, and 0.72 for IF; 0.85, 0.45, and 1.00 for LOF; 0.83, 0.64, and 0.90 for oSVM; and 0.83, 0.55, and 0.93 for the ensemble, respectively.

Conclusions

There is a true correlation between the type of sellar barrier at MRI and its in vivo features as observed during endoscopic endonasal surgery. The novelty detection models highlighted differences between patients who developed an intraoperative CSF leak and those who did not.

A Combined Radiomics and Machine Learning Approach to Overcome the Clinicoradiologic Paradox in Multiple Sclerosis

BACKGROUND AND PURPOSE

Conventional MR imaging explains only a fraction of the clinical outcome variance in multiple sclerosis. We aimed to evaluate machine learning models for disability prediction on the basis of radiomic, volumetric, and connectivity features derived from routine brain MR images.

MATERIALS AND METHODS

In this retrospective cross-sectional study, 3T brain MR imaging studies of patients with multiple sclerosis, including 3D T1-weighted and T2-weighted FLAIR sequences, were selected from 2 institutions. T1-weighted images were processed to obtain volume, connectivity score (inferred from the T2 lesion location), and texture features for an atlas-based set of GM regions. The site 1 cohort was randomly split into training (n = 400) and test (n = 100) sets, while the site 2 cohort (n = 104) constituted the external test set. After feature selection of clinicodemographic and MR imaging-derived variables, different machine learning algorithms predicting disability as measured with the Expanded Disability Status Scale were trained and cross-validated on the training cohort and evaluated on the test sets. The effect of different algorithms on model performance was tested using the 1-way repeated-measures ANOVA.

RESULTS

The selection procedure identified the 9 most informative variables, including age and secondary-progressive course and a subset of radiomic features extracted from the prefrontal cortex, subcortical GM, and cerebellum. The machine learning models predicted disability with high accuracy (r approaching 0.80) and excellent intra- and intersite generalizability (r ≥ 0.73). The machine learning algorithm had no relevant effect on the performance.

CONCLUSIONS

The multidimensional analysis of brain MR images, including radiomic features and clinicodemographic data, is highly informative of the clinical status of patients with multiple sclerosis, representing a promising approach to bridge the gap between conventional imaging and disability.

Persisting Embryonal Infundibular Recess in Morning Glory Syndrome: Clinical Report of a Novel Association

Morning glory syndrome is characterized by a congenital optic disc defect that resembles the eponymous flower. We present the MR imaging findings of 2 pediatric patients with morning glory disc anomaly and persisting embryonal infundibular recess, another rare malformative finding, a previously unreported association. Neuroradiologists should be aware of the possible presence of a persisting embryonal infundibular recess in patients with morning glory syndrome, to aid in the differential diagnosis including other pituitary malformations such as pituitary stalk duplication.

Propofol infusion and auditory evoked potentials

The effects of propofol on auditory evoked potentials (brainstem and middle latency responses) were recorded in six patients. Two different infusion rates were used, 54 and 108 micrograms/kg/minute. Effects on brainstem responses were not found. Regression of amplitude and latency of middle latency auditory potentials were dose related (p less than 0.01).

Evidence for a differential interaction of buprenorphine with opiate receptor subtypes controlling prolactin secretion

We studied the effects of various doses of the opiate derivative buprenorphine on serum prolactin levels and whether these effects could be counteracted by pretreatment with the opiate receptor blocker naloxone. The administration of increasing doses of buprenorphine exerted a dual effect on serum prolactin levels. At low doses (3, 10 and 30 micrograms/kg) this agent increased serum prolactin levels. This effect disappeared with increasing doses (100 and 300 micrograms/kg), and at the highest doses (1000 and 3000 micrograms/kg) the levels of serum prolactin decreased. Naloxone (30 mg/kg) decreased serum prolactin levels and reversed both the stimulatory and the inhibitory action of buprenorphine. These data are compatible with the hypothesis that buprenorphine could interfere with two different, but inter-dependent receptors: at low doses the oripavine derivative could act at one receptor site to cause an increase of serum prolactin, whereas at higher doses it could interact with a second site of lower affinity that is responsible for the inhibition of prolactin secretion. When buprenorphine (at high doses) activates the lower affinity site, the interaction with this receptor counteracts and reverses the effects of the high affinity site. On the basis of this hypothesis, naloxone should block both receptors.

Photooxidation of epinephrine sensitized by methylene blue as an assay for the evaluation of alpha-mercaptopropionylglycine as a free radical scavenger

Free radicals, which are instable intermediates of some biochemical reactions, are produced everywhere in the living matter. They may induce cell membrane lipid peroxidation, with non reversible letal effects. Alpha-mercaptopropionylglycine capability of scavenging free radicals has been tested. H2O2 has been used as a free radical donor under actinic light, epinephrine oxidation to adrenochrome as a revealing system and methylene blue as sensitizing agent. alpha-MPG at 6 mM concentration showed the maximal inhibition of free radical formation in our system. Higher concentrations were unable to produce further inhibition.

[Preparation of a polyvalent antiserum specific for human muscle proteins. I. Muscle extraction: a macromethod and a micromethod]

Protein metabolism may be upset dramatically by several pathological condition such as starvation, malnutrition, surgical stress. Serum proteins have been investigated in many works but very few studies exist about muscle and organ proteins. This article describes two methods for extraction of muscle water soluble proteins: a macromethod that uses some hundred grams of tissue, and a micromethod that starts from a little muscle biopsia. Extracts have been tested for protein content by electrophoresis on cellulose acetate and isoelectrofocusing on polyacrylamide gel and have shown a rich variety of protein fractions.

[Preparation of a polyvalent antiserum specific for human muscle proteins. II. Animal immunization and purification of the antiserum]

A new preparation technique to raise rabbit antisera against human water-soluble muscle proteins is presented. Immunization, blood collection and further purification of antibody fraction in serum are described in detail. The present method is fast and easy to perform; chromatography and dialysis are not required and antisera so produced lend themselves to immunoelectrophoretic evaluation of human water-soluble muscle proteins.

An improved immunoelectrophoretic methodology for the evaluation of protein caloric malnutrition

An improved technique of two dimensional immunoelectrophoresis of serum proteins on gelatinized cellulose acetate for routine evaluation of protein nutritional state is described. The method, which allows the estimation of more than 30 protein fractions, proves extremely useful in monitoring the effectiveness of total parenteral nutrition and protein metabolic conditions. Sera samples from healthy subjects and cachectical neoplastic patients have been examined and preliminary results are reported.

Effects of althesin and ketamine on resting and stress stimulated adrenocortical activity in rats

The effects were studied of two i.v. anaesthetics (Althesin and ketamine) on resting and activated adrenocortical function in the rat. Small doses of Althesin and ketamine comparable to those required to produce induction of anaesthesia in man evoked adrenocortical stimulation, but the stressing effects of ketamine were more prolonged. During deep anaesthesia with either anaesthetic there was an inhibition of pituitary adrenal activation induced by handling and surgery. However, during recovery, ketamine was associated with intense and prolonged adrenocortical stimulation. The stress of laparotomy in control rats and in rats pretreated with the anaesthetics caused a depletion of noradrenaline in the hypothalamus; this suggests that both anaesthetics inhibit adrenocortical activation by mechanisms different from adrenergic inhibition.

Partial cardio-pulmonary bypass in peripheral vascular surgery: case reports

Two cases of aneurysmatic vascular lesions (a large cranio-encephalic angiodysplasia with arterio-venous communications and a post-traumatic aneurysm of the vertebral artery) are reported; both were of a very difficult surgical approach and were successfully treated with the use of extra-corporeal circulation. The partial cardiopulmonary bypass, with femoro-femoral cannulation, allowed complete recovery of the lost blood, controlled hypotension, and a certain degree of hypothermia. These factors were decisive for the good outcome of the operations.