Room Air Quality Monitoring Misses Surgical Field Exposure: Rethinking Ventilation Standards for Surgical Site Infection (SSI) Prevention

Authors:
Justin Benton, Divya Kewalramani, Lily Wushanley, Oluwaseun Adeyemi, Alexander Wurtz, Yana Chaudharu, Les Barta, Joelle Getrajman, Leonardo Calderone, Keith Kaye, Brian Buckley, Gediminas Mainelis, Philip Barie, Mayur Narayan

Body of Abstract:
Background: Intraoperative airborne particulate matter (PM), of which electrosurgical smoke is a major constituent, serves as the primary vector for exogenous microbial delivery to surgical incisions. Yet current gold-standard ventilation strategies show inconsistent effectiveness in reducing surgical site infections (SSIs). These strategies assume operating room (OR) air quality (measured at entry points for ventilation compliance) reflects surgical field exposure (where pathogens contact open incisions). This assumption remains untested. We hypothesized that electrosurgical smoke exposure varies by measurement location.

Methods: In an OR with active laminar airflow ventilation (≥20 air changes/h, high efficiency particulate air filtration,) we performed electrocautery activations on draped cadaver arms under aseptic technique. Synchronized PM sensors (1-Hz resolution, 0.1–10.0 μm) were positioned field-proximal (L1: 1m from surgical site, 1m height, 4m from door) and at room-entry (L2: 1m from door, 1m height, 3m from surgical site). Following 10-minute baselines, electrocautery (without smoke evacuation) was activated in repeated 5-second ON/180-second OFF cycles on cadaver tissue. We compared simultaneous sensor measurements using paired Wilcoxon tests for magnitude and timing differences, Cohen’s kappa for extreme event agreement (>50 μg/m³), and Spearman correlation for magnitude-dependent spatial discordance (α=0.05).

Results: Across 151 paired cycles, peak concentrations differed between locations (median difference 3.0 μg/m³, Interquartile range [IQR] 0.5–8.0 μg/m³; p<0.0001), with L1 measuring higher in 75% of cycles. Peak timing varied (median 9 sec, range: -17.5-69.5 seconds; p=0.0009), with L2 peaks occurring first in 40% of cycles and later in 60% of cycles. Extreme PM spikes (>50 μg/m³) showed poor spatial agreement (κ=0.13): only 2/12 extreme events was detected by both sensors simultaneously, with 10 events recorded at L1 while L2 remained at baseline (<2 μg/m³). Spatial discordance increased systematically with event magnitude (ρ=0.889, p<0.0001), with the largest events showing up to 198-fold differences (593μg/m³ vs 3μg/m³) in peak PM2.5 between L1 and L2. Conclusions: Current OR ventilation monitoring standards demonstrate a fundamental measurement problem because the extreme PM exposures most likely to deliver infectious microbial inocula are highly localized and spatially heterogeneous, with substantially higher PM exposure detected at a field-proximal location. As a result, room-level standards risk normalizing metrics that are uncoupled from biologically relevant exposure. SSI prevention efforts should shift from costly architectural proxies (ventilation standards) to field-proximal PM monitoring that would support targeted smoke evacuation when exposure thresholds are exceeded.