A Randomized Double-Blind Controlled Trial of Microstream Capnography to Reduce Oxygen Desaturation During Gastrointestinal Procedures with Conscious Sedation

Abstract

Oxygen desaturation is a relatively late sign of ventilatory compromise. ASGE monitoring guidelines for all GI procedures call for continuous pulse oximetry and visual assessment during conscious sedation (CS), which may not detect apnea until O2 desaturation has occurred. Microstream capnography may provide an “early warning system” by generating real-time waveforms of respiration (capnograms) in non-intubated patients. Aim: To determine whether intervention based on capnography indications of apnea reduces O2 desaturation during CS. Methods: We performed an IRB-approved double-blind randomized controlled trial of ASA Class I/II children undergoing GI procedures with midazolam + fentanyl CS, 2L supplemental O2, standard patient monitoring, and routine care. Endoscopy staff was blinded to additional continuous capnography monitoring (Philips M4 with Microstream® CO2). Study arm was determined by opening a pre-numbered sequence of envelopes containing randomly permuted assignments. In the intervention arm, a trained independent monitor (IM) signaled to the endoscopy staff by raised hand if capnograms indicated apnea for >15 sec. In the control arm, an IM signaled if capnograms indicated apnea for >60 sec. Upon IM signals, endoscopy RNs instructed patients to breathe deeply and/or touched their backs. We hypothesized that patients randomized to the intervention arm would have fewer episodes of minor O2 desaturation (O2 sats <95% for 5 sec) than patients in the control arm. Results: 163 participants (89 male; mean age 13 yr, range 6mo-19 yr; 86% ASA I) underwent 174 procedures (80% EGD, 13% colonoscopy, 7% both). Intervention and control patients did not differ significantly in sex, age, ASA class, procedures undergone, baseline O2 sats, or doses of midazolam and fentanyl. Endoscopy RNs documented poor ventilation in 2.7% of study patients and no apnea. Capnography indicated disordered ventilation during 56% of all procedures and apnea in 24%. Patients in the intervention arm were significantly less likely to have an intraprocedural episode of O2 desaturation than those in the control arm (11% vs 24%, p < 0.03). Conclusions: The results of this controlled effectiveness trial support the routine use of microstream capnography to improve detection of apnea and reduce O2 desaturation during CS for GI procedures. Supported by AHRQ (K08 HS1-367502), a Risk Management Foundation Patient Safety Grant and Children's Hospital Boston GCRC (M01-RR02172). Oxygen desaturation is a relatively late sign of ventilatory compromise. ASGE monitoring guidelines for all GI procedures call for continuous pulse oximetry and visual assessment during conscious sedation (CS), which may not detect apnea until O2 desaturation has occurred. Microstream capnography may provide an “early warning system” by generating real-time waveforms of respiration (capnograms) in non-intubated patients. Aim: To determine whether intervention based on capnography indications of apnea reduces O2 desaturation during CS. Methods: We performed an IRB-approved double-blind randomized controlled trial of ASA Class I/II children undergoing GI procedures with midazolam + fentanyl CS, 2L supplemental O2, standard patient monitoring, and routine care. Endoscopy staff was blinded to additional continuous capnography monitoring (Philips M4 with Microstream® CO2). Study arm was determined by opening a pre-numbered sequence of envelopes containing randomly permuted assignments. In the intervention arm, a trained independent monitor (IM) signaled to the endoscopy staff by raised hand if capnograms indicated apnea for >15 sec. In the control arm, an IM signaled if capnograms indicated apnea for >60 sec. Upon IM signals, endoscopy RNs instructed patients to breathe deeply and/or touched their backs. We hypothesized that patients randomized to the intervention arm would have fewer episodes of minor O2 desaturation (O2 sats <95% for 5 sec) than patients in the control arm. Results: 163 participants (89 male; mean age 13 yr, range 6mo-19 yr; 86% ASA I) underwent 174 procedures (80% EGD, 13% colonoscopy, 7% both). Intervention and control patients did not differ significantly in sex, age, ASA class, procedures undergone, baseline O2 sats, or doses of midazolam and fentanyl. Endoscopy RNs documented poor ventilation in 2.7% of study patients and no apnea. Capnography indicated disordered ventilation during 56% of all procedures and apnea in 24%. Patients in the intervention arm were significantly less likely to have an intraprocedural episode of O2 desaturation than those in the control arm (11% vs 24%, p < 0.03). Conclusions: The results of this controlled effectiveness trial support the routine use of microstream capnography to improve detection of apnea and reduce O2 desaturation during CS for GI procedures. Supported by AHRQ (K08 HS1-367502), a Risk Management Foundation Patient Safety Grant and Children's Hospital Boston GCRC (M01-RR02172).