Exploring a Computer Model to Predict Aggressive Pressure Sores

Researchers at Pitt’s School of Medicine have devised a computer model to help better understand, diagnose, and treat pressure ulcers related to spinal cord injury. 

In a report published online in PLOS Computational Biology, the Pitt team described results of virtual clinical trials showing that for effective treatment of the lesions, anti-inflammatory measures had to be applied well before the earliest clinical signs of ulcer formation.

Such computer modeling raises the specter that, in the future, a nurse could simply send a photo of a patient’s reddened skin to a doctor. The physician would use the computer model to predict whether the red area was likely to develop into a pressure sore and, if so, caregivers could provide quick and aggressive treatment.  

Pressure ulcers affect more than 2.5 million Americans annually, and patients who have movement-impairing spinal cord injuries are more vulnerable to developing them, said senior investigator Yoram Vodovotz, professor of surgery and director of the Center for Inflammation and Regenerative Modeling at Pitt’s School of Medicine.

“These lesions are thought to develop because immobility disrupts adequate oxygenation of tissues where the patient is lying down, followed by sudden resumption of blood flow when the patient is turned in bed to change positions,” Vodovotz said. “This is accompanied by an inflammatory response that sometimes leads to further tissue damage and breakdown of the skin.”

“Pressure ulcers are an unfortunately common complication after spinal cord injury, and cause discomfort and functional limitations,” said coauthor Gwendolyn A. Sowa, associate professor of physical medicine and rehabilitation, Pitt School of Medicine. “Improving the individual diagnosis and treatment of pressure ulcers has the potential to reduce the cost of care and improve quality of life for persons living with spinal cord injury.”

To address the complexity of the biologic pathways that create and respond to pressure sore development, the researchers designed a computational, or “in silico,” model of the process based on serial photographs of developing ulcers. The photos were from spinal cord-injured patients enrolled in studies at Pitt’s Rehabilitation Engineering Research Center on Spinal Cord Injury. Photos were taken when the ulcer was initially diagnosed, then three times per week in the acute stage, and once a week as it resolved.

Researchers then validated the model, finding that if they started with a single small round area over a virtual bony protuberance—and then altered factors such as inflammatory mediators and tissue oxygenation—they could recreate a variety of irregularly shaped ulcers that mimic what is seen in reality.

They also conducted two virtual trials of potential interventions, finding that anti-inflammatory interventions could not prevent ulcers unless applied very early in their development.

“Computational models like this one might one day be able to predict the clinical course of a disease or injury, as well as make it possible to do less expensive testing of experimental drugs and interventions to see whether they are worth pursuing with human trials,” Vodovotz speculated. “They hold great potential as a diagnostic and research tool.”

Other Pitt team researchers included Cordelia Ziraldo, Alexey Solovyev, Ana Allegretti, Shilpa Krishnan, and David Brienza.