Research Article: Explanatory models in neonatal intensive care: a tutorial

Date Published: December 20, 2018

Publisher: BioMed Central

Author(s): Willem W.L. van Meurs, Timothy A.J. Antonius.

http://doi.org/10.1186/s41077-018-0085-2

Abstract

Acute care providers intervening on fragile patients face many knowledge and information related challenges. Explanation based on causal chains of events has limitations when applied to complex physiological systems, and model-driven educational software may overwhelm the learner with information. We introduce a new concept and educational technology to facilitate understanding, reasoning, and communication in the clinical environment. The aim is to grasp complex physiology in a more intuitive way.

An EM is a representation of relevant physiologic processes that provides insight into the relationships between therapeutic interventions and monitored variables, and their dependency on incidents and pathologies. We systematically analyze types of information incorporated into models and displayed in simulations and consider their explanatory relevance.

A conceptual model (diagram) of the normal neonatal cardiorespiratory system is adapted to reflect TGA and implemented in animated, interactive software.

The use of this model is illustrated via the explanation to pediatric residents of the relationships between blood pressures, blood flow rates, ventilation, oxygen saturation, and oxygen distribution in a neonate with TGA. Learners explore clinical scenarios and effects of therapeutic interventions.

Explanatory models hold promise as mental models for clinical practice and could possibly play a role in clinical decision making in neonatal intensive care and beyond.

The software is freely available via the web addendum: https://www.dropbox.com/sh/ciufq5rqxgs9bkt/AAC7oKsvkEr73eYUJkx0pZ1Ya?dl=0

Partial Text

Acute care providers intervene on fragile patients with incomplete information on underlying physiology. They frequently do so under time pressure, and diagnosis and choice of therapeutic interventions may have far reaching consequences for the patient. Resources include their knowledge of physiology and pharmacology, clinical experience, patient specific information from clinical signs, monitoring instruments and additional exams, and protocols. Communication with other members of the healthcare team and consultation of information sources expand this basis for intervention. Acute care physicians thus face major knowledge and information related challenges:Complexity of underlying (patho)physiology.Incomplete information on underlying (patho)physiology.Integration of general knowledge and patient specific information.Effective communication with co-workers about the patient.

In its basic form, an explanation is a valid deductive argument, based on true antecedent conditions and applicable general laws, that has as its conclusion that the phenomenon to be explained occurred [4]. Deductive explanation frequently takes the form of a causal chain of events. This form of explanation is not well suited for situations in acute care because chains of events do not adequately reflect complex underlying physiologic processes involving multiple variables, interactions, and feedback loops.

The interactive class on complex clinical situations in neonatal intensive care is aimed at pediatric residents and interns. In this tutorial, we focus on explaining the underlying physiology and management options in a single complex condition, namely transposition of the great arteries (TGA) with ventricular septal defect (VSD), one of the more common complex congenital heart lesions in the newborn [7, 8]. In this defect, the vessels that carry blood to the lungs and to the body are improperly connected (Fig. 2). The pulmonary artery is connected to the left ventricle and the ascending aorta to the right ventricle. Oxygenated blood from the left ventricle returns to the lungs, and deoxygenated blood is carried to the tissues. Two connections (shunts) are required to make sure these two separate circulations can mix so oxygenated blood can be supplied to the body. The ductus arteriosus is a shunt which normally closes after birth. Affected babies are admitted to the ICU and given medication (prostaglandin E1) to make sure the ductus stays patent. Sometimes, a procedure called a balloon atrial septostomy is necessary to enlarge another shunt, the foramen ovale, to further improve mixing of oxygenated and deoxygenated blood. Surgical correction within 1–2 weeks after birth is often necessary to restore normal anatomy. Morbidity and mortality in this condition therefore depend on oxygen flows over these two shunts. Oxygen flow depends on blood flow and on oxygen content. Blood flow depends on systemic and pulmonary pressures and resistances, yet also affects the driving pressures. Partial pressures driving oxygen transport are in the same time affected by that transport, and by metabolism. A number of active control systems further complicate simple causal reasoning about the involved phenomena, making them a challenging to understand using traditional means.Fig. 2Schematic representation of a transposition of the great arteries. Pulmonary veins (PV), left atrium (LA), left ventricle (LV), ascending aorta (Aa), descending aorta (Ad), inferior vena cava (IVC), superior vena cava (SVC), right atrium (RA), right ventricle (RV), pulmonary arteries (PA), ductus arteriosus (DA), foramen ovale (FO)

There are several ways in which the described explanatory model can be used. Here, we describe a 2-h class for 6–10 pediatric residents. The goal is to explain the complex relationships between ventilation, blood pressures and flow rates, and oxygen distribution in a neonate with TGA. The faculty has two laptop computers at his or her disposition, both running the software.

Mechanistic, engineering approaches to modeling of naturally occurring systems have their limitations; our bodies are not made out of neatly separated interacting systems and subsystems, and discrete components. However, the analogies and traceable simplifications inherent to this approach facilitate the use of resulting models as explanatory models.

 

Source:

http://doi.org/10.1186/s41077-018-0085-2

 

0 0 vote
Article Rating
Subscribe
Notify of
guest
0 Comments
Inline Feedbacks
View all comments