JMIR Publications

ABSTRACT

Background:

Monitoring surgical recovery has traditionally been confined to metrics measurable within the hospital and clinic setting. However, commercially available mobile sensors are now capable of extending measurements into a patient’s home. As these sensors were developed for non-medical applications their clinical role has yet to be established. The aim of this systematic review is to evaluate the relationship between data generated by mobile sensors and postoperative outcomes.

Objective:

The objective of this study is to describe the current utilization of mobile sensors in the perioperative setting, and the correlation between their data and clinical outcomes.

Methods:

A systematic search of EMBASE, Medline and Cochrane Library, from inception until April 2019, was performed to identify studies of surgical patients monitored with mobile sensors. Sensors were considered if they collected patient metrics such as step count, temperature or heart rate. Studies were included if patients underwent major surgery (≥ 1 inpatient postoperative day), were monitored using mobile sensors in the perioperative period and reported postoperative outcomes (i.e. complications and hospital readmission). For studies including step count, a pooled analysis of the step count per postoperative day was calculated for the complication and non-complication cohorts using mean and Random-effects RELM model. The GRADE tool was used to assess study quality.

Results:

From 2209 abstracts we identified 11 studies for review. Reviewed studies consisted of either prospective observational cohorts (N = 10) or randomized controlled trials (N = 1). Activity monitors were the most widely used sensors (N = 10), with an additional study measuring temperature, respiratory rate and heart rate (N = 1). Low step count was associated with worse postoperative outcomes. A median step count of less than around 1000 steps per postoperative day was associated with adverse surgical outcomes. Within the studies there was heterogeneity between the type of surgery and type of reported postoperative outcome.

Conclusions:

Despite significant heterogeneity in type of surgery and sensors, low step count was associated with worse postoperative outcomes across surgical specialties. Further studies and standardization are needed to assess the role of mobile sensors in postoperative care, but a threshold of approximately 1000 steps per postoperative day warrants further investigation.