Abstract Disposable P2/N95 respirators are important in preventing nosocomial infections within the healthcare system. Healthcare workers who cannot shave for cultural, religious, or medical reasons, face restrictions in using tight fitted respirators. Alternative solutions, like powered air purifying respirators, do not provide source control and are not universally accepted in healthcare settings, potentially exacerbating staff shortages during periods of high demand. Over the past 18 months, The Royal Melbourne Hospital (RMH) has conducted a comprehensive evaluation of the under-respirator elastic band beard cover, also known as the Singh Thattha Technique. The protocol for implementing and evaluating the technique, has been published in a peer-reviewed open-access journal. Participants are required to pass three fit tests with the elastic band/respirator combination, exceeding standard fit testing practices. The trial results involving the initial 87 participants have also recently been published. Ninety-nine percent of participants successfully passed fit testing with the Industree Trident P2 respirator, while 78% met the protocol criteria using the 3M Aura 1870+. All 87 participants achieved successful fit testing on at least one of these respirators following our stringent protocol. This presentation will explore RMH’s methodology for assessing the elastic band beard cover, examine the test protocol, discuss findings, highlight limitations of the technique, and explore future implications. This work aligns with the conference theme, offering a transformative solution that empowers the healthcare workforce to be prepared for the future.

Objective: Many available facepiece filtering respirators contain ferromagnetic components, which may cause significant problems in the magnetic resonance imaging (MRI) environment. We conducted a randomized crossover trial to assess the effectiveness, usability, and comfort of 3 types of respirators, judged to be "conditionally MRI safe" with an aluminum nosepiece (Halyard 46727 duckbill-type respirators and Care Essentials MSK-002 bifold cup-type respirators) or "MRI safe" completely metal free (Eagle AG2200 semirigid cup-type respirators). Design and setting: We recruited 120 participants to undergo a quantitative fit test (QNFT) on each of the 3 respirators in a randomized order. Participants then completed a usability and comfort assessment of each respirator. Results: There were significant differences in the QNFT pass rates (51% for Halyard 46727, 73% for Care Essentials MSK-002, and 86% for Eagle AG2200, P < .001). The first-time fit test pass rate and overall fit factor were significantly higher for Eagle AG2200 compared with the other 2 respirators. Eagle AG2200 scored the lowest ratings in the ease of use and overall comfort. There were no significant differences in other modalities, including the seal rating, breathability, firmness, and overall assessment. Conclusions: Our study supports the utility of the Eagle AG2200 and Care Essentials MSK-002 respirators for healthcare professionals working in an MRI environment, based on their high QNFT pass rates and reasonably good overall usability and comfort scores. Eagle AG2200 is unique because of its metal-free construction. However, its comparatively lower usability and comfort ratings raise questions about practicality, which may be improved by greater user training.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

Objective: To evaluate a relatively new half-face-piece powered air-purifying respirator (PAPR) device called the HALO (CleanSpace). We assessed its communication performance, its degree of respiratory protection, and its usability and comfort level. Design and setting: This simulation study was conducted at the simulation center of the Royal Melbourne Hospital. Participants: In total, 8 voluntary healthcare workers participated in the study: 4 women and 4 men comprising 3 nursing staff and 5 medical staff. Methods: We performed the modified rhyme test, outlined by the National Institute for Occupational Safety and Health (NIOSH), for the communication assessment. We conducted quantitative fit test and simulated workplace protection factor studies to assess the degree of respiratory protection for participants at rest, during, and immediately after performing chest compression. We also invited the participants to complete a usability and comfort survey. Results: The HALO PAPR met the NIOSH minimum standard for speech intelligibility, which was significantly improved with the addition of wireless communication headsets. The HALO provided consistent and adequate level of respiratory protection at rest, during and after chest compression regardless of the device power mode. It was rated favorably for its usability and comfort. However, participants criticized doffing difficulty and perceived communication interference. Conclusions: The HALO device can be considered as an alternative to a filtering face-piece respirator. Thorough doffing training and mitigation planning to improve the device communication performance are recommended. Further research is required to examine its clinical outcomes and barriers that may potentially affect patient or healthcare worker safety.

Individuals who are unable to be clean shaven for religious, medical or cultural reasons are unable to wear a filtering facepiece respirator (FFR), as the respirator cannot provide adequate protection against aerosol-transmissible diseases. There is currently a paucity of validated techniques to ensure the safe inclusion of bearded healthcare workers in the pandemic workforce. We propose to undertake a healthcare-based multi-modal evaluation study on the elastic band beard cover for FFR technique, examining the quantitative fit test (QNFT) results, usability and skill level of participants with repeated assessments over time. This is a prospective study conducted through the Respiratory Protection Program at the Royal Melbourne Hospital. Healthcare workers are invited to participate if they require respiratory protection and cannot shave for religious, cultural or medical reasons. An online education package on the use of respiratory protective equipment and the elastic band beard cover for FFR technique is provided. This is followed by a face-to-face session, where the participant will receive: one-on-one training; undergo a skill assessment on their donning, doffing and user seal check techniques; complete QNFTs and a usability survey. Participants will be invited to repeat the assessment within 3 months of the first session and at 12 months. This study involves multimodal and repeated assessments of an elastic band beard cover for FFRs. The findings of this study will provide information on: whether this simple technique can provide safe, consistent and effective respiratory protection; whether it will interfere with occupational activities; and whether it is comfortable and tolerable for the duration of wear. This is of significant importance to the health workforce around the world, who cannot shave but require access to respiratory protective equipment during the COVID-19 pandemic.

Background: The use of respiratory protection remains important in protecting health care workers from airborne pathogens such as viruses. Respirator supply is constantly changing with new models regularly becoming available. Health services should consider a broad range of factors when procuring respirators, including the results of quantitative fit testing in a representative sample of the workforce. Subjective comfort factors and compatibility with a variety of workplace tasks, such as suitability for staff use near magnetic resonance imaging (MRI) environments where relevant, should also be considered. This article compares the quantitative fit factors and user assessments for two styles of flat-fold cup respirators, Care Essentials (CE) MSK-002 P2 and BYD DE2322 N95. Methods: Quantitative fit tests (QNFT) were performed on 300 participants on each model of respirator in this randomised crossover trial. Participants then completed a questionnaire on their assessments of each respirator.Results: The Care Essentials MSK-002 had a significantly higher quantitative fit test pass rate than the BYD DE2322 (57% vs 18%, p < 0.001). There was no concordance between fit test pass rates for each model. Additionally, the Care Essentials MSK-002 achieved significantly higher scores on each of the responses in the subjective usability survey.Conclusion: It is recommended that the Care Essentials MSK-002 be made available for health care use due to higher QNFT pass rates, higher subjective usability assessment scores, plus its potential for use in MRI environments when compared to the BYD DE2322. 2022 The Author(s). Published by Elsevier B.V. on behalf of Australasian College for Infection Prevention and Control. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).

OBJECTIVES:To compare the performance of four N95 respirator types with respect to quantitative fit test pass rate and health care worker-rated usability and comfort.DESIGN, SETTING, PARTICIPANTS:Health care workers who participated in the respiratory protection program at the Royal Melbourne Hospital, 1 October 2020 - 31 May 2021. Participants underwent quantitative N95 respirator fit testing (at least three of four types: semi-rigid cup, flat-fold cup, duckbill, and three-panel flat-fold types), and were invited to complete an online usability and comfort assessment for respirators for which their fit test results were passes.MAIN OUTCOME MEASURES:Fit test pass rate, and user-rated overall comfort and assessment ratings (five-point Likert scales), by N95 respirator type.RESULTS:A total of 2161 health care workers underwent quantitative fit testing (women, 1586 [73.4%]; nurses, 1271 [58.8%]). The overall fit test pass rates were 65.0% for the semi-rigid cup respirators (1029/1583 tests), 32.4% for the flat-fold respirator (660/2035 tests), 59.2% for the duckbill respirators (2005/3387 tests), and 96.4% for the three-panel flat-fold respirator (1876/1946 tests). 378 health care workers completed the comfort and usability survey. Overall comfort and assessment ratings each differed by respirator group (P < 0.001); the median overall comfort (4; IQR, 3-4) and overall assessment values (4; IQR, 3-5) were highest for the three-panel flat-fold respirator and lowest for the semi-rigid cup respirators (comfort: 2 [IQR, 1-3]; assessment: 2 [IQR, 2-3]).CONCLUSIONS:The three-panel flat-fold N95 respirator outperformed the three alternative types with regard to fit test pass rate and user-rated comfort and usability. To maximise respiratory protection for health care workers, these factors should be considered when making respirator procurement decisions.

Objective: Discomfort and device-related pressure injury (DRPI) caused by N95 filtering facepiece respirators (FFRs) are common. The use of prophylactic hydrocolloid dressings is one of the strategies that may improve comfort and reduce DRPI. In this study, we investigated the impact of these dressings on N95 respirator fit. Methods: We performed a repeat quantitative fit testing through the Respiratory Protection Program on 134 healthcare workers (HCWs), who applied hydrocolloid dressings on the bridge of their nose under the N95 FFRs that they passed the initial fit test with, but reported discomfort with the FFR. Results: With the hydrocolloid dressings in place, the fit-test pass rate for the semirigid cup style (3M 1860) was 94% (108 of 115); for the the vertical flat-fold style (BYD), the pass rate was 85% (44 of 52); for the duckbill style (BSN medical ProShield and Halyard Fluidshield), the pass rate was 81% (87 of 108); and for the 3-panel flat-fold style (3M Aura) N95 FFRs, the pass rate was 100% (3 of 3). There was a statistically significant reduction in the overall fit factors for both the vertical flat-fold and duckbill type N95 respirators after the application of hydrocolloid dressings. Conclusions: Hydrocolloid dressings are likely to disturb the mask seal for nonrigid-style N95 FFRs, particularly the vertical flat-fold style and the duckbill style N95 FFRs. Given the risk of mask seal disturbance of N95 respirators as shown in this study, we advocate that any HCW requiring the use of prophylactic dressings should undergo repeat quantitative fit testing with the dressing in place prior to using the dressing and mask in combination.

AbstractObjective:Some manufacturers provide information on similar-fit model pairings of filtering facepiece respirators (FFRs), suggesting that fit-test outcome of one model helps predict the other. This guidance may be useful during crisis capacity when FFR supplies and/or fit tests are constrained. The purpose of this study was to compare quantitative fit-test (QNFT) results and concordance between a pair of similar-fit 3M Aura FFRs: the 3M 9320A+ and 3M 1870+.Methods:All participants completed online training and a QNFT with both respirators. The order of the respirator being examined first was randomly allocated. The outcomes included QNFT pass rate, concordance between the 2 models, overall and individual fit factors, and percentage of male and female participants who passed or failed the QNFT.Results:We recruited 1,000 participants (668 females and 332 males). The QNFT pass rate, overall fit factors, and individual fit factors were significantly higher for the 3M 9320A+ than the 3M 1870+ FFR. The concordance between the models was “fair” (κ coefficient, 0.38). Male participants who passed a QNFT with either of the FFRs had 96% chance of passing the QNFT for the alternate model. Female participants who passed the 3M 1870+ had 97% chance of passing the QNFT for the 3M 9320A+ model. However, ∼1 in 12 females who passed the QNFT for the 3M 9320A+ failed the QNFT for the 3M 1870+ FFR.Conclusions:Similar-fit paired FFR models may provide a consequentially different level of respiratory protection, especially for women. Our findings are important for FFR stockpiling and fit-testing strategies, especially during crisis capacity.

INTRODUCTION:When performing quantitative fit testing (QNFT) on filtering facepiece respirators using an ambient aerosol technique, a twin sampling tube is connected between the condensation nuclei count machine and the probed respirator. To achieve high quality and repeatable QNFT results, robust sampling tube stabilization is required. METHODS:In this prospective randomized crossover study, conducted in December 2021 to February 2022, we compared the commonly used hand-hold technique with the manufacturer-recommended lanyard technique in stabilizing the sampling tube during QNFT on a Halyard N95 respirator. Outcomes included QNFT pass rates, overall and individual fit factors, and concordance between the two techniques. RESULTS:A total of 228 out of 316 participants (72.2%) passed the QNFT with the hand-hold technique, compared to the lanyard technique (166/316, 52%, P < .001). The most significant drop in the fit factors with the lanyard technique occurred during head movement side-to-side and up-and-down. The concordance between the 2 techniques was fair (Kappa coefficient = 0.39). CONCLUSION:Our study demonstrates that the method of sampling tube stabilization during QNFT has a significant impact on fit test pass rates, with a potential for false negative fit tests due to inadequate tube stabilization. Further research is required to examine the generalizability of these results to other respirators and fit testing apparatuses.

N95 particulate respirator masks are currently recommended for all healthcare workers who care for patients with suspected or confirmed coronavirus disease (COVID-19) when performing aerosol-generating procedures. The protection provided by N95 particulate respirator masks is dependent on the filter's efficiency and seal quality. In this prospective randomised crossover study, we conducted the user seal check and the quantitative fit test on two readily available duckbill models of N95 masks, the Halyard Fluidshield® N95 (Halyard, Alpharetta, GA, USA) and the BSN Medical ProShield® N-95 (BSN Medical, Mount Waverley, Victoria) particulate respirator masks. We recruited a total of 96 anaesthetic staff, of whom 26% were of South-East Asian ethnicity. We found that both types of masks provided reasonably high fit test pass rates among our participants and there was no significant difference between the two brands (77% for the Fluidshield and 65% for the ProShield, P = 0.916). Ninety-two percent of the participants could find at least one well-fitted mask among these two types of masks. We also demonstrated that the user seal check had low accuracy and low concordance (kappa coefficient of 0.16 for the Fluidshield and 0.08 for the ProShield) when compared to the quantitative fit test, and hence was not a reliable method to test seal quality.

Editor—The cardio-renal benefits1Zinman B. Wanner C. Lachin J. et al.Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes.N Engl J Med. 2015; 373: 2117-2128Crossref PubMed Scopus (6533) Google Scholar of sodium–glucose cotransporter-2 inhibitors (SGLT2i) in people with diabetes mellitus have led to their widespread use. SGLT2i predispose to perioperative diabetic ketoacidosis (DKA),2Chacko B. Whitley M. Beckmann U. Murray K. Rowley M. Postoperative euglycaemic diabetic ketoacidosis associated with sodium-glucose cotransporter-2 inhibitors (gliflozins): a report of two cases and review of the literature.Anaesth Intensive Care. 2018; 46: 215-219Crossref PubMed Scopus (18) Google Scholar including euglycaemic DKA. To minimise perioperative DKA risk, current guidelines recommend withholding SGLT2i for 48 h before surgery (where fasting and anaesthesia are required)3Burke K. Schumacher C. Harpe S. SGLT2 inhibitors: a systematic review of diabetic ketoacidosis and related risk factors in the primary literature.Pharmacotherapy. 2017; 37: 187-194Crossref PubMed Scopus (103) Google Scholar,4Goldenberg R. Berard L. Cheng A. et al.SGLT2 inhibitor-associated diabetic ketoacidosis: clinical review and recommendations for prevention and diagnosis.Clin Ther. 2016; 38: 2654-2664Abstract Full Text Full Text PDF PubMed Scopus (140) Google Scholar; otherwise, postponement should be considered.5Kerridge R. Whyte I. Prior F. Luu J. Story A. The good, the bad, and the ugly: sodium-glucose cotransporter-2 inhibitors (gliflozins) and perioperative diabetes.Anaesth Intensive Care. 2018; 46: 155-158Crossref PubMed Scopus (6) Google Scholar Further research is required to understand the optimal management approach, where SGLT2i has been administered within 48 h before surgery. We report metabolic outcomes in a case series of patients treated with SGLT2i immediately before emergency or elective surgery. The series was assembled from cases where SGLT2i were used in the 48 h before surgery during 2018–20. DKA was defined as the presence of ketosis (capillary beta-hydroxybutyrate [BOHB] >0.6 mM) and acidosis (blood pH <7.30: mild 7.25–7.29; moderate 7.00–7.24; severe <7.00).6Kitabchi A.E. Umpierrez G.E. Miles J.M. Fisher J.N. Hyperglycemic crises in adult patients with diabetes.Diabetes Care. 2009; 32: 1335-1343Crossref PubMed Scopus (1019) Google Scholar Ketone screening was performed in the perioperative period if a patient was identified to have used SGLT2i, or if SGLT2i use was not recognised in the perioperative period, then ketones were measured if the patient became unwell with acidosis in the postoperative period. Fisher's exact test statistical analysis was performed, and results should be regarded as hypothesis generating. The study was approved by the Human Research and Ethics Committee of Melbourne Health (QA2019038). We identified 23 patients who underwent surgery despite SGLT2i administration (empagliflozin: n=15; dapagliflozin: n=8) in the 48 h prior. Of these, 17 were male and six were female with median age 69 yr (range: 46–79 yr). Median Charlson Comorbidity Index was 5.0 (inter-quartile range [IQR]: 4.5–6.0). There were 21 with type 2 diabetes (T2D), two with type 1 diabetes (T1D) with off-label SGLT2i use, seven were insulin requiring pre-admission, and median glycosylated haemoglobin was 7.8% (IQR: 7.0–8.7%) or 62 mmol mol−1 (IQR: 53–72). Seventeen patients underwent emergency surgery; six patients who underwent elective surgery had inadvertently not withheld their SGLT2i, as recommended. The SGLT2i dose was last administered a median 24 h before surgery (IQR: 24–29), median preoperative fasting time was 12 h (IQR: 9–17), and median procedure duration was 2.5 h (IQR: 1.8–4.0). Seven patients underwent high-risk cardiac or vascular surgery, 13 intermediate-risk and two low-risk surgery (American College of Cardiology/American Heart Association classification),7Fleisher L.A. Beckman J.A. Brown K.A. et al.ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines.Circulation. 2007; 116: 1971-1996Crossref PubMed Scopus (551) Google Scholar and 15 patients were managed in the ICU after surgery. Overall, 20 of 23 patients (87%) developed ketosis, with 13 (57%) having ketosis without acidosis and seven (30%) having DKA (mild, n=5; moderate, n=2; severe, n=0) (Table 1). Three patients with DKA were euglycaemic at the time of peak ketosis. Median initial BOHB was 2.1 mM (IQR: 0.7–3.6) and median peak BOHB was 2.7 mM (IQR: 1.2–4.6). For patients with ketosis without acidosis, the median ketosis duration was 16 h (IQR: 12–42), whereas in those with DKA it was 10 h (IQR: 8–75). Two patients recognised before surgery as having received SGLT2i, and therefore at risk for DKA, were managed expectantly with i.v. insulin and glucose, and did not develop ketosis. Five patients were detected with ketosis before surgery (initial BOHB range: 0.7–3.2 mM; median: 2.4 mM), eight during surgery (initial BOHB range: 0.7–4.8 mM; median: 2.3 mM), and seven after surgery (initial BOHB range: 0.2–6.5 mM; median: 4.0 mM). In patients recognised with ketosis before or during surgery, 15% (two of 13) developed DKA, whereas in those recognised with ketosis after surgery, 71% (five of seven) developed DKA (P=0.02). None of the five patients recognised with ketosis before surgery who received i.v. insulin and glucose infusions early (before or during surgery) developed DKA.Table 1Perioperative outcomes of patients with preoperative SGLT2 inhibitor use.PatientOperationPreoperative SGLT2i timing (h)∗Duration between last administered SGLT2i dose and commencement of surgery.Preoperative fasting time (h)Operation duration (h)Initial BOHB (mM)Peak BOHB (mM)Glucose at peak BOHB (mM)Lowest pHHCO3– (mM)Timing of ketosis detectionTiming of insulinNo ketosisEmergency surgery1Coronary artery bypass graft2485.00.40.45.97.3123NilIntra i.v.2Laparoscopic cholecystectomy24242.50.10.418.47.3627NilNil3Coronary artery bypass graft24106.00.40.613.47.3023NilIntra i.v.Ketosis without acidosisEmergency surgery4Laparotomy bowel resection22244.51.03.210.87.3521PrePre i.v.5Toe amputations24121.03.13.112.77.3721PrePre i.v.6Hip hemiarthroplasty2481.54.04.08.07.3319PostPost s.c.7Dental extraction24241.02.52.512.47.3022IntraIntra i.v.8Laminectomy2482.51.31.313.77.3326IntraIntra i.v.9Quadriceps tendon repair30161.53.14.4157.3124IntraPost i.v.10Coronary artery bypass graft34113.52.12.78.47.3222IntraIntra i.v.11Percutaneous cholecystostomy36360.52.42.917.97.3220PrePre i.v.12Anterior cervical discectomy/fusion43432.53.24.87.97.3222PrePre i.v.Elective surgery13Sternal plate removal881.00.21.38.4NA23PostPost s.c.14Robotic radical prostatectomy1284.00.70.76.27.4523PreIntra i.v.15Bilateral above-knee amputation26182.04.85.46.97.3220IntraPost i.v.16Atrial fibrillation radiofrequency ablation28123.01.52.65.67.3124IntraPost s.c.Diabetic ketoacidosisEmergency surgery17Coronary artery bypass graft2455.04.84.89.07.2822PostPost i.v.18Laminectomy24123.50.71.17.57.2920IntraIntra i.v.19Craniotomy24122.56.56.511.57.2617PostPost i.v.20Laparotomy bowel resection30165.04.34.811.27.2818IntraIntra i.v.21Endovascular clot retrieval31163.01.81.814.27.2323PostPost i.v.Elective surgery22Coronary artery bypass graft1144.04.16.412.07.2014PostPost i.v.23Laparoscopic hysterectomy and bilateral salpingo-oophorectomy5122.00.60.97.37.2722PostPost i.v.BOHB, beta-hydroxybutyrate; HCO3–, bicarbonate; Intra, intraoperative; NA, not available; Post, postoperative; Pre, preoperative; SGLT2i, sodium–glucose cotransporter-2 inhibitor.∗ Duration between last administered SGLT2i dose and commencement of surgery. Open table in a new tab BOHB, beta-hydroxybutyrate; HCO3–, bicarbonate; Intra, intraoperative; NA, not available; Post, postoperative; Pre, preoperative; SGLT2i, sodium–glucose cotransporter-2 inhibitor. Awareness of SGLT2i-associated perioperative DKA is increasing; yet, formal investigation of metabolic outcomes in patients with diabetes administered SGLT2i immediately before surgery is limited. Our case series identified that ketosis was common (87% of cases); however, less than half of those with ketosis progressed to DKA (30% of cases overall) and none developed severe DKA. Patients who experienced ketosis without acidosis likely represent cases where earlier detection and intervention prevented progression to DKA. The older population in our series reflects SGLT2i use in older comorbid T2D populations, compared with traditional DKA cohorts of younger patients with T1D.6Kitabchi A.E. Umpierrez G.E. Miles J.M. Fisher J.N. Hyperglycemic crises in adult patients with diabetes.Diabetes Care. 2009; 32: 1335-1343Crossref PubMed Scopus (1019) Google Scholar The duration of preoperative fasting seen in our series is not expected to significantly raise BOHB levels. In weight loss studies, ketosis with BOHB levels of 1.0 mM is seen after multiple weeks on ketogenic diets.8Chearskul S. Delbridge E. Shulkes A. Proietto J. Kriketos A. Effect of weight loss and ketosis on postprandial cholecystokinin and free fatty acid concentrations.Am J Clin Nutr. 2008; 87: 1238-1246Crossref PubMed Scopus (72) Google Scholar Of interest, four patients experienced ketosis duration >48 h, and additional risk factors were evident. Patient 9 experienced a 6 h delay from recognition of ketosis to treatment initiation. Patient 12 fasted for 43 h before surgery (because of operating theatre delays). Patient 17 inadvertently self-administered double the maximum recommended dose of SGLT2i. Patient 22 was administered SGLT2i 1 h before surgery. Our definition of DKA was based upon British9Association of British Clinical DiabetologistsJoint British diabetes societies (JBDS) for inpatient care group. The management of diabetic ketoacidosis (DKA) in adults.Second edition. September 2013http://www.diabetologists-abcd.org.uk/JBDS/JBDS.htmGoogle Scholar and American6Kitabchi A.E. Umpierrez G.E. Miles J.M. Fisher J.N. Hyperglycemic crises in adult patients with diabetes.Diabetes Care. 2009; 32: 1335-1343Crossref PubMed Scopus (1019) Google Scholar guidelines utilising pH and bicarbonate, although based on base excess, some patients may have had contributory respiratory acidosis (Supplementary Table 1). We identified several risk factors for protracted ketoacidosis, including delayed perioperative recognition of SGLT2i use, administration of SGLT2i close to surgery, inadvertent preoperative administration of higher-than-recommended SGLT2i doses, prolonged fasting, and high-risk surgery (e.g. cardiac surgery). Raising awareness for all clinicians involved in perioperative care regarding risk factors for SGLT2i-associated DKA is important, as risk mitigation strategies can prevent or minimise severity of DKA. These include minimising ongoing SGLT2i exposure, minimising fasting, regular 2–4 hourly blood ketone monitoring, and early intervention with i.v. insulin and glucose. Early involvement of diabetes specialists to assist with management is recommended. In patients presenting for elective surgery who have not withheld SGLT2i, risk–benefit analyses to determine pros and cons of deferring surgery are recommended. If patients must undergo emergency or elective surgery after recent SGLT2i administration, risk mitigation strategies may prevent DKA. Although further research is required to determine optimal therapeutic strategies, our experience suggests that early recognition of SGLT2i use, early detection of ketosis, and early initiation of i.v. insulin and glucose are likely to minimise the risk of perioperative DKA. The authors declare that they have no conflicts of interest. The following is the Supplementary data to this article: Supplementary material is available at British Journal of Anaesthesia online. Download .docx (.02 MB) Help with docx files Multimedia component 1

Purpose of review The purpose of this article is to review complications and unplanned hospital admissions in patients presenting for ambulatory procedures requiring anesthesia care in the gastrointestinal endoscopy, bronchoscopy, and radiology suites. Recent findings The range of ambulatory diagnostic and therapeutic procedures being undertaken in the gastrointestinal endoscopy, bronchoscopy, and radiology suites is expanding rapidly. Recent observational studies in gastrointestinal endoscopy confirm low incidences of complications and unplanned admissions. Deep propofol-based sedation is associated with more complications than lighter sedation. Older patients suffer more complications but obstructive sleep apnea does not appear to increase risk. Sedation improves patient comfort during bronchoscopy. Propofol-based sedation is associated with fewer complications than benzodiazepine-based sedation, but all combinations are associated with high patient satisfaction. Obesity and obstructive sleep apnea are not associated with worse outcomes in bronchoscopy patients. Sedation is increasingly required for interventions in the radiology suite. When patients are involved in choosing sedation depth, there is a trend to lighter sedation and high patient satisfaction. Summary Sedation and anesthesia are required for the increasing number of increasingly complex procedures being undertaken outside the operating suite. Large randomized trials are required to define the optimum sedation drugs, sedation depth and sedation provider.