Online Supplement to
What's New in Intensive Care – Ventilatory Support of Patients with Sepsis or Septic Shock Management in Resource–limited Settings
Ary Serpa Neto1,2,3, Marcus J. Schultz3,4,5, and Emir Festic6, for the ‘European Society of Intensive Care Medicine (ESICM) Global Intensive Care’ working group and the Mahidol–Oxford Research Unit (MORU) in Bangkok, Thailand
Hospital Israelita Albert Einstein, São Paulo, Brazil
1Medical Intensive Care Unit
Faculdade de Medicina do ABC; São Paulo, Brazil
2Program of Post-Graduation, Research and Innovation
Academic Medical Center, University of Amsterdam,
Amsterdam, The Netherlands
3Department of Intensive Care
4Laboratory of Experimental Intensive Care and Anesthesiology (L.E.I.C.A)
Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand
5Mahidol–Oxford Research Unit (MORU)
Mayo Clinic, Jacksonville, Florida, USA
6Pulmonary and Critical Care Medicine
Word count: 7,280 words (unrestricted)
Number of inserts: 4 tables (unrestricted)
Reference list: 83 references (unrestricted)
Correspondence:
Prof. Marcus J. Schultz, MD
Mahidol–Oxford Tropical Medicine Research Unit (MORU); Faculty of Tropical Medicine, Mahidol University
420/6 Rajvithi Road, Bangkok 10400, Thailand
Email:
Email other authors:
Ary Serpa Neto
Emir Festic
Group members of the ‘Sepsis in Resource–limited Settings’–guidelines group and the ‘Mahidol–Oxford Research Unit’ (MORU) in Bangkok, Thailand
(Heads) Arjen Dondorp (Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand & Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands &), Martin Dünser (Innsbruck Medical University, Innsbruck, Austria), and Marcus Schultz (Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands & Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand).
(Other group members, in alphabetic order) Neill K.J. Adhikari (Sunnybrook Health Sciences Centre & University of Toronto, Toronto, ON, Canada), Derek Angus (University of Pittsburgh, Pittsburgh, PA), Luciano Azevedo (Hospital Sirio–Libanes, Saõ Paulo, Brazil), Kwizera Arthur (Mulago National Referral Hospital, Kampala, Uganda), Timothy Baker (Karolinska Institute, Stockholm, Sweden), Ted Barnett (The Ochsner Medical Center, New Orleans, Louisiana), Chris Farmer (Mayo Clinic, Rochester, MI), Rashan Haniffa (Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand), Randeep Jawa (Stony Brook University Medical Center, Stony Brook, NY, USA), Niranjan Kissoon (British Columbia Children’s Hospital and University of British Columbia, Vancouver, Canada), Rakesh Lodha (All India Institute of Medical Science, Delhi, India), Ganbold Lundeg (Mongolian National University of Medical Sciences, Ulaanbaatar, Mongolia), Ignacio Martin Loeches (St. James's University Hospital, Dublin, Ireland), David Misango (Aga Khan University Hospital, Nairobi, Kenya), Mervyn Mer (Johannesburg Hospital and University of the Witwatersrand, Johannesburg, South Africa), Sanjib Mohanty (Ispat General Hospital, Rourkela, Sundargarh, Odisha, India), Srinivas Murthy (BC Children’s Hospital, University of British Columbia, Vancouver, Canada), Ndidiamaka Musa (Seattle Children’s Hospital and University of Washington, WA), Jane Nakibuuka (Mulago National referral and university teaching hospital, Kampala, Uganda), Mai Nguyen Thi Hoang (Oxford University Clinical Research Unit, Ho Chi Minh City, District 5, Vietnam), Binh Nguyen Thien (Trung Vuong Hospital, Ho Chi Minh City, Viet Nam), Rajyabardhan Pattnaik (Ispat General Hospital, Rourkela, Sundargarh, Odisha, India), Luigi Pisani (University of Bari Aldo Moro, Bari, Italy), Jason Phua (Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore), Jacobus Preller (Addenbrooke’s Hospital, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK), Pedro Povoa (Nova Medical School, CEDOC, New University of Lisbon, Lisbon, Portugal & Hospital de São Francisco Xavier, Centro Hospitalar de Lisboa Ocidental, Lisbon, Portugal), Suchitra Ranjit (Appolo’s Hospitals, Chennai, India), Jonarthan Thevanayagam (Mzuzu Central Hospital, Mzuzu, Malawi), and Louise Thwaites (Centre for Tropical Medicine and Global Health, Nuffield Department of Medicine, University of Oxford, UK).
Group members of the subgroup ‘Ventilatory Support’
(Heads) Ary Serpa Neto (Hospital Israelita Albert Einstein, São Paulo, Brazil), Marcus J. Schultz (Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands & Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand), and Emir Festic (Mayo Clinic, Jacksonville, Florida, USA)
(Other subgroup members, in alphabetic order) Neill K.J. Adhikari (Sunnybrook Health Sciences Centre, Toronto, ON, Canada), Arjen Dondorp (Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand & Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands), Rajyabardhan Pattnaik (Ispat General Hospital, Rourkela, Sundargarh, Odisha, India), Luigi Pisani (University of Bari Aldo Moro, Bari, Italy), Pedro Povoa (Nova Medical School, CEDOC, New University of Lisbon, Lisbon, Portugal & Hospital de São Francisco Xavier, Centro Hospitalar de Lisboa Ocidental, Lisbon, Portugal) and Ignacio Martin Loeches (St. James's University Hospital, Dublin, Ireland) and Louise Thwaites (Centre for Tropical Medicine and Global Health, Nuffield Department of Medicine, University of Oxford, UK).
Guideline development
Selection of group members
The selection of the group members was based on interest in specific aspects of sepsis and hands–on experience in intensive care units (ICUs) in resource–limited settings. Marcus Schultz, Martin Dünser and Arjen Dondorp contacted potential team members through email and in person at the ‘27th Annual Congress of the European Society of Intensive Care Medicine’ in Barcelona, Spain from 27 September – 1 October 2014, and the ‘35th International Symposium on Intensive Care and Emergency Medicine’ in Brussels, Belgium from 17 March – 20 March 2015, and created nine subgroups assigned to nine areas in sepsis management, i.e., ‘diagnosis of sepsis’, ‘levels of ICU’, ‘organizational aspects’, ‘infection control’, ‘hemodynamic support’, ‘ventilatory support’, ‘sedation, renal failure, prophylaxes, glucose control and feeding’, ‘tropical aspects’ and ‘pediatric aspects’. Additional team members were appointed by the group heads to address content needs for the development process. Several group members had experience in ‘Grading of Recommendations Assessment, Development and Evaluation’ (GRADE) process and use of the GRADEpro Guideline Development Tool [1].
Selection of subgroup heads and subgroup ‘Ventilatory Support’
Ary Serpa Neto, Marcus J. Schultz, and Emir Festic were appointed as the group heads; Neill Adhikari, Arjen Dondorp, Rajyabardhan Pattnaik, Luigi Pisani, Pedro Povoa, Ignacio Martin Loeches and Louise Thwaites were assigned to this subgroup based on their specific expertise and interest in ventilatory support.
Meetings
An initial Internet subgroup heads–meeting was held to establish the procedures for literature review and drafting of tables for evidence analysis. The subgroup heads continued work via the Internet. Several meetings occurred at major international meetings, teleconferences and electronic–based discussions among the subgroup heads and with members of other subgroups.
In the first meetings ten clearly defined question regarding ventilatory support were formulated, using the GRADEpro Guideline Development Tool [1]. These were reviewed for content and clarity by the subgroup members. After approval by the subgroup members, the subgroup heads split up, each one to seek for evidence for recommendations regarding three or four of the specific questions posed, seeking help from the subgroup members in identifying relevant publications where necessary. During this process, two questions were combined, so the subgroup heads were finally left with nine questions. The subgroup heads summarized the evidence in the online supplement, and formulated the recommendations after three interactive telephone conferences. These were communicated among the subgroup members. After their approval, the subgroup heads summarized the evidence in a report, which was sent for approval by all members of the other groups.
In the review process one reviewer suggested one extra question. This question was merged with the last question (see below), and processed in the same way as the other questions.
Search techniques
The search for literature followed the same techniques as described for the development of the Surviving Sepsis Campaign–guidelines [2]. In case a question was identical to one in those guidelines, the subgroup members only searched for additional articles, specifically (new) investigations or metaanalyses related to the questions, in a minimum of one general database (i.e., MEDLINE, EMBASE) and the Cochrane Libraries. Furthermore, the subgroup members identified investigations from LMICs.
Grading of Recommendations
The subgroup members also followed the principles of the GRADE process as described for the development of the Surviving Sepsis Campaign–guidelines [2]. In short, GRADE classifies quality of evidence as high (grade A), moderate (grade B), low (grade C), or very low (grade D) and recommendations as strong (grade 1) or weak (grade 2). The factors influencing this classification are presented in the online supplement Table 1. Different from the grading of recommendations in the Surviving Sepsis Campaign–guidelines [2], the subgroup members paid extensive attention to several other factors as used before, but now focusing on resource–limited settings, i.e., availability and feasibility in resource–limited ICUs, affordability for low–resource ICUs, and last but not least its safety in resource–limited ICUs. A strong recommendation was worded as ‘we recommend’ and a weak recommendation as ‘we suggest’. A number of recommendations could remain ‘ungraded’ (UG), when, in the opinion of the subgroup members, such recommendations were not conducive for the process described above.
Reporting
The report was edited for style and form by Marcus Schultz, with final approval by subgroup heads and then by the entire ‘Sepsis in Resource–limited Settings’–guidelines group. A final document was submitted to ‘Intensive Care Medicine’ for potential publication as a 1,000–word ‘What is new in …’–article.
Conflicts of interest
No members of the nine subgroup represented industry, and there was no industry input into guidelines development. No member of the nine subgroup received honoraria for any role in the guideline development process. Each member provided a standard COI–form, to be uploaded through the GRADEpro Guideline Development Tool website. Of 30 members, none reported conflicts of interest.
Questions
Nine clearly defined questions regarding ventilatory support were formulated:
(1) Is it essential to diagnose acute respiratory distress syndrome (ARDS) in patients with sepsis or septic shock in resource–limited ICUs?
(2) Should a semi–recumbent position be used in patients with sepsis or septic shock resource–limited ICUs?
(3) Should non–invasive ventilation (NIV) be used in patients with sepsis or septic shock with acute respiratory failure in resource–limited ICUs?
(4) Should spontaneous breathing trials be used to recognize extubation readiness in patients with sepsis or septic shock in resource–limited ICUs?
(5) Should low tidal volumes always be used, i.e., irrespective of the presence of ARDS in patients with sepsis or septic shock in resource–limited ICUs?
(6) Should a minimal level of PEEP always be used, and should higher levels of PEEP be preferred in patients with sepsis or septic shock with ARDS in resource–limited ICUs?
(7) Should low oxygen fractions with high levels of PEEP be preferred over high oxygen fractions with low levels of PEEP in patients with sepsis or septic shock with hypoxemia in resource–limited ICUs?
(8) Should recruitment maneuvers be used in patients with sepsis or septic shock with refractory hypoxemia in resource–limited ICUs?
(9) Should assisted ventilation be preferred over controlled ventilation and should a ‘volume controlled’ mode be preferred over a ‘pressure controlled’ mode in patients with sepsis or septic shock in resource–limited ICUs?
Detailed results
Question 1: Is it essential to diagnose acute respiratory distress syndrome (ARDS) in patients with sepsis or septic shock in resource–limited ICUs?
Rationale: ARDS is a combined clinical and radiographic diagnosis, which per the latest consensus definition [3] requires presence of acute bilateral chest infiltrates (onset within less than 1 week), not primarily caused by hydrostatic pulmonary edema, associated with hypoxemia based on PaO2 to FiO2 ratio (P/F) of less than 300 mm Hg and requiring at least 5 cm H2O of PEEP. Therefore, in order to diagnose ARDS, performance of a chest radiograph (CXR) and arterial blood gas analysis (ABG) is necessary. However, the availability (or the lack of) of CXR and ABG in resource–limited settings might preclude their timely performance and utility in diagnosing septic patient with ARDS. It is not clear whether the mere diagnosis of ARDS would impact care and/or outcomes of patients with sepsis or septic shock with acute respiratory failure. There is a growing body of evidence suggesting that the diagnostic utility of the ultrasound exam of the lung compares well with that of the CXR [4-8]. When coupled with the performance of point–of–care echocardiogram, the diagnosis of non–cardiogenic pulmonary edema can be achieved in acutely hypoxemic patients, including those with sepsis [9-11].
Search results: The MEDLINE search did not result in any studies from either resource–limited ICUs or resource–rich ICUs that directly answered the question of interest. We discuss several pertinent studies from resource–rich ICUs and one from a resource–limited setting.
Evidence: The search based on the key terms identified one metaanalysis done by the Cochrane collaboration group [12] that included only two randomized controlled trials (RCTs), one of which was done on children less than 60 months old. Another RCT with limited applicability was done on 2,018 emergency room and walk–in clinic patients in Texas and investigated whether performance CXR had any effect on clinical course and management [13]. The majority of the patients were ambulatory without CXR abnormalities and there was no overall difference in the course and management based on whether the CXR was performed or not. However, in the subgroup of patients with a pulmonary infiltrate, performance of CXR resulted in management strategies, which reduced clinical symptoms and duration of illness.
Low oxygen saturation to FiO2 ratio (S/F) within 6 hours of presentation to emergency room was found to predict early development of ARDS [14]. A secondary analysis of the ARDS Network trial of a lower tidal volume ventilator strategy showed that S/F correlated with P/F in patients with ARDS [15]. Another study suggests that ARDS patients diagnosed by S/F have very similar clinical characteristics and outcomes compared with patients diagnosed by P/F [16]. A retrospective study from Brazil showed that a low S/F at ICU admission was associated with increased risk of death in patients with severe sepsis or septic shock [17]. However, diagnosis of ARDS or its severity categorization based on S/F alone can be difficult given a wide variability in S/F relative to P/F observed in the studies on the topic [14-21]. In addition to oxygenation impairment, bilateral infiltrates on CXR at the time of hospital presentation have also been shown to predict development of early ARDS [22].
Importantly, a metaanalysis using individual patient data from trials that enrolled patients without ARDS at the onset of mechanical ventilation showed that the use of low tidal volume was associated with significantly shorter duration of mechanical ventilation [23], thus patients without ARDS probably benefit from this life–saving intervention demonstrated in patients with ARDS. This means that it may not be essential to diagnose ARDS in patients who need ventilation because of acute respiratory failure.