It might sound like a good idea to ask patients to send photos postoperatively to evaluate surgical sites, but researchers of a new study raise concerns about the practice. They found that adding amateur images to patient symptom reports actually worsened surgeon accuracy and sensitivity in diagnosing surgical site infections.
“Effective widespread implementation of remote postoperative assessment with photography may require additional development of tools, participant training and mechanisms to verify image quality,” the authors conclude in the study published online October 24 in JAMA Surgery.
Researchers prospectively studied how wound photography impacts surgeon ability to remotely diagnose surgical site infections in real-world practice by comparing surgeons’ paired assessments of post-abdominal surgery case vignettes with and without wound photography. In all, 523 surgeons, including abdominal, colon and rectal, general, surgical oncology and vascular surgeons, completed an average 2.9 vignettes, yielding more than 1,500 total case assessments.
They found that having access to wound photography didn’t change how accurately surgeons diagnosed surgical site infections. While surgeons accurately diagnosed these infections in 57.1% of cases with symptom reports and no photography, they accurately diagnosed surgical site infections in 58.1% of cases with symptom reports and photographs.
Surgeons correctly diagnosed 58.4% of cases with surgical site infections when they had only symptom reports, versus 50.2% of cases when they had both the reports and photos. They correctly diagnosed cases in which there were no surgical site infections 56.1% of the time without the photographs and in 63.4% of cases with them.
Having access to the images increased surgeons’ confidence and the likelihood that they’d reassure patients rather than request an in-person patient assessment, regardless of whether they correctly identified a surgical site infection.
Surgeons were more likely to under-triage patients when vignettes included a wound photograph, regardless of whether they correctly diagnosed an infection, according to the study.
“Remote evaluation of patient-generated wound photographs may not accurately reflect the clinical state of surgical incisions,” the authors conclude.
In a commentary on the study, F. Thurston Drake, M.D., MPH, who works in the Department of Surgery, Boston Medical Center, Boston University School of Medicine, writes that surgeons might proceed with smartphone photographs, but should do so with caution.
Dr. Drake points out his concern with the study’s finding that the use of photographs seems to make surgeons more confident in their wrong answers.
“If a similar false confidence plays out in actual clinical decision making, delays in care may lead to progression of disease,” he writes.
Still, one can’t ignore that use of smartphone photographs to help manage postop concerns is reality, which highlights the need to determine how optimally to use photographs.
Patient-generated photographs seem to be suboptimal for making diagnoses but promising as triage tools, according to Dr. Drake.
“However, surgeons must be vigilant not to allow the impression formed from a patient-generated photograph to overshadow other data gleaned from history taking and knowledge of baseline risk,” he writes.