Intensive Care Med (2013) 39:1207–1210

DOI 10.1007/s00134-013-2905-0 WHAT’S NEW IN INTENSIVE CA RE

Jesús Villar
Robert M. Kacmarek
What is new in refractory hypoxemia?

ARDS for providing adequate oxygenation and carbon

Received: 18 February 2013
Accepted: 16 March 2013 dioxide elimination, while the underlying disease process
Published online: 11 April 2013 is being treated to restore normal pulmonary function.
Ó Springer-Verlag Berlin Heidelberg and ESICM 2013 However, some ARDS patients develop refractory hyp-
oxemia and MV is unable to sustain adequate gas
J. Villar exchange despite the use of high inspired oxygen con-
CIBER de Enfermedades Respiratorias, centrations (FiO2) and high levels of positive end-
Instituto de Salud Carlos III, Madrid, Spain expiratory pressure (PEEP) [2].
There are a number of adjunctive and alternative
J. Villar techniques to MV (that are in use or under evaluation
Research Unit, Hospital Universitario Dr. Negrı́n,
Las Palmas de Gran Canaria, Spain worldwide) for improving oxygenation in ARDS patients
with refractory hypoxemia. These techniques include
J. Villar recruitment manoeuvres (RM), extracorporeal membrane
Keenan Research Center at the Li Ka Shing Knowledge Institute oxygenation (ECMO), high frequency oscillatory venti-
of St. Michael’s Hospital, Toronto, Canada lation (HFOV), prone positioning, and pharmacologic
agents. Today, refractory hypoxemia is rare and an
R. M. Kacmarek infrequent cause of death (\15 % of ARDS deaths) [3].
Department of Respiratory Care, Massachusetts General Hospital,
Boston, MA, USA However, there is no standard definition for refractory
hypoxemia in terms of a predetermined PaO2 value under
R. M. Kacmarek a specific FiO2 and PEEP level for a specific period of
Department of Anesthesia, Harvard University, time. In most reports, it has been defined as a PaO2
Boston, MA, USA \70 mmHg on a FiO2 0.8–1.0 and PEEP[10 cmH2O for
more than 12–24 h.
J. Villar ())
Multidisciplinary Organ Dysfunction Evaluation Research
Network, Hospital Universitario Dr. Negrin,
Barranco de la Ballena, s/n-4th floor, South wing,
35010 Las Palmas de Gran Canaria, Spain Recruitment manoeuvres
e-mail: [email protected]
Tel.: ?34-92-8449413 RMs are intended to reopen collapsed alveoli and to
Fax: ?34-92-8449813 attenuate the injurious effects of the repetitive opening
and closing of alveolar units. As a result, oxygenation and
lung mechanics are improved. In general, a RM applies a
transient intentional pressure higher than that applied
Acute respiratory distress syndrome (ARDS) is an during a normal breath either intermittently (for 2–3 min)
inflammatory disease process of the lungs characterized or sustained (up to about 40 s) [4]. A RM may also
by severe hypoxemia, reduced lung compliance, and influence ventilation by reducing PaCO2. The amount of
bilateral radiographic infiltrates [1]. Mechanical ventila- potentially recruitable lung tissue seems to correlate with
tion (MV) is the standard life support technique during the severity of ARDS.
1208

Much controversy exists over the sustained benefit of a all ECMO patients were treated in the same center and the
RM. Some have noted no sustained benefit [5, 6]; others ECMO center did not treat patients randomised to the
have shown sustained and marked improvement in oxy- conventional management group. Additional prospective
genation and lung mechanics [4, 7, 8]. The difference ECMO trials for ARDS are underway (ClinicalTrails.gov
seems to be based on the selection of PEEP post-RM. NCT01470703) which compare ECMO to a modern
Where improvement has been sustained a decremental protocolized lung protective MV strategy.
PEEP trial has been used to identify the minimal PEEP
level sustaining the benefit of the RM [4, 7, 8]. In a meta-
analysis of seven clinical trials involving 1,170 patients
with ARDS, there was no significant difference in sur- HFOV
vival between groups receiving an ‘‘open-lung’’
ventilatory strategy that included RM and groups given Some clinicians have proposed that HFOV is an ideal
standard ventilatory care [9]. However, most of those mode of ventilation for ARDS patients as it is the natural
trials had major methodological limitations. We are culmination of low tidal volume ventilation [17]. HFOV
waiting for the results of a recent trial evaluating the should theoretically be an ideal mode to ventilate patients
application of RMs early in the course of patients with with severe lung damage since it achieves gas exchange
persistent ARDS (ClinicalTrials.gov NCT00431158). by delivering very small tidal volumes that are typically
1–3 ml/kg (often less than the anatomic dead space) at
frequencies ranging from 3 to 10 Hz around a relatively
constant mean airway pressure. However, there is no
ECMO evidence that conventional MV with low tidal volumes,
high PEEP and limited plateau pressures is more harmful
During the last few years, there have been considerable that HFOV. Two recent RCTs comparing HFOV to pro-
advances in extracorporeal life support technology and tective MV demonstrated that HFVO does not reduce [18]
widespread use in patients with ARDS [10]. To provide or may increase [19] in-hospital mortality. One of the
gas exchange during ECMO, a portion of the cardiac major factors explaining the lack of beneficial effects with
output must go through the ECMO circuit via the femoral, HFOV was the higher hemodynamic instability in patients
saphenous or jugular veins. During ECMO, CO2 is allocated in the HFOV arms. These two trials are con-
removed by the extracorporeal circuit with conventional vincing examples that it is hard to put theory into practice.
MV maintained at low ventilatory rates, high PEEP levels One could conclude that HFOV, as applied in these trials,
and with tidal volumes to maintain a plateau pressure is ineffective and perhaps worse than current conventional
below 30 cm H2O. Some investigators believe that there MV practice.
exists a role for ECMO in patients with single organ
system failure who are deemed to have potentially
reversible pulmonary dysfunction when all other con-
ventional modalities have failed [11]. In the past, the Prone positioning
highly specialized equipment and knowledge required to
provide ECMO made this technique available only in Changes in posture can have profound effects on the
specific medical centers. Today, ECMO is safer, cheaper, pulmonary function of patients with severe respiratory
and simpler [12]. Both circuit and patient can be cared for failure. By tradition, patients with respiratory failure are
by appropriately trained personnel [13]. Promising results cared for supine. During acute respiratory failure, a
in recent reports during the H1N1 pandemic [14] have led reduction of functional residual capacity results in supine
to renewed interest in ECMO for managing severe ARDS. hypoxemia regardless of age. The proposed mechanisms
However, a recent French matched pairs analysis of H1N1 by which prone positioning improves oxygenation in
patient who received ECMO and those receiving con- ARDS include an increase in functional residual capacity,
ventional lung protective MV demonstrated no survival a change in regional diaphragm motion, redistribution of
benefit [15]. perfusion to better ventilated lung units, redistribution of
A recent muticenter trial, referred to as the CESAR ventilation to better perfused lung units, and improved
trial, assessed the effectiveness of ECMO in 180 patients secretion clearance [20]. In general, prone positioning can
with severe ARDS [16]. Survival at 6 months or absence be performed safely if staff is appropriately trained.
of severe disability was achieved in 63 % of the ECMO Although there are sufficient data to conclude that
patients, compared with 47 % of the control group oxygenation frequently improves when ARDS patients
(p = 0.03). However, there are a number of major con- are turned prone, prone positioning is still not widely
cerns with this study since 30 % of patients in the control implemented. Systematic reviews and meta-analyses [21]
group were not ventilated with a lung protective strategy, in patients with ARDS showed that in general prone
 1209

positioning does not reduce mortality or duration of MV treatment for ARDS patients, no RCT has demonstrated
despite improved oxygenation and a decreased risk of any outcome benefit. The most recent systematic review
pneumonia. However, a most recent meta-analysis [22] and meta-analysis of 12 RCTs including a total of 1,237
revealed that in patients with severe ARDS, as defined by patients with severe ARDS found that, as a whole, nitric
a PaO2/FiO2 ratio \100 mmHg, prone positioning was oxide is associated with no effect on duration of venti-
able to significantly reduce all-cause mortality. Although lation, does not confer mortality benefits, and may cause
no recommendations can be offered on the optimum harm [25].
timing or duration of prone positioning, extended prone
positioning seems to be most beneficial when maintained
18–20 h daily. A recent trial (ClinicalTrials.gov NCT
00527813) examining the effects of prone positioning in Implications for clinical practice
patients with persistent and severe ARDS (as defined by a
PaO2/FiO2 \150 mmHg on FiO2 C0.5 with PEEP C5 In conclusion, HFVO has proven to be deleterious.
cmH2O for 12–24 h) has been completed [23]. They Recruitment maneuvers do improve oxygenation and
randomized 474 patients to prone positioning for 16 h per pulmonary mechanics but have not demonstrated
day or supine. The 28-day mortality was 32.8 % in the improved mortality. ECMO has been shown to be bene-
supine group and 16 % in the prone group (p \ 0.0001), a ficial in the sole poorly designed recent RCT. Prone
difference that persisted at 90 days after randomization. ventilation has been shown to reduce mortality in patients
with severe ARDS in one RCT and in several meta-
analysis. The major issues with these techniques are: (1)
they must be applied early in ARDS and correctly, (2)
Pharmacologic agents they must be applied to the appropriate patient, and (3)
clinicians need to define when, how and in whom we can
Anti-inflammatory agents and vasodilators (including optimally use these techniques.
prostaglandins, ibuprofen, nitric oxide, and cortocoster-
oids) have been tried as prophylaxis or treatment of Acknowledgments Supported in part by Instituto de Salud Carlos
ARDS patients. None of them has shown any major III, Spain (#10/0393 and CB06/06/1088).
benefit on outcome in large randomized control trials, Conflicts of interest On behalf of all authors, the corresponding
although improvements in oxygenation have been author states that there is no conflict of interest in relation to this
observed with some of these agents [24]. Despite the fact manuscript.
that many clinicians consider nitric oxide a useful rescue

References
1. Villar J (2011) What is the acute 5. Brower RG, Lanken PN, MacIntyre N, 8. Borges J, Okamoto VN, Matos GFJ,
respiratory distress syndrome? Respir Matthay MA, Morris A, Ancukiewicz Caramez MPR, Arantes PR, Barros F,
Care 56:1539–1545 M, Schoenfeld D, Thompson BT, Souza CE, Victorino JA, Kacmarek
2. Villar J, Kacmarek RM (2012) The National, Heart, Lung, and Blood RM, Barbas CSV, Carvalho CRR,
American-European consensus Institute ARDS Clinical Trials Network Amato MPB (2006) Reversibility of
conference definition of the acute (2004) Higher versus lower positive lung collapse and hypoxemia in early
respiratory distress syndrome is dead, end-expiratory pressures in patients acute respiratory distress syndrome. Am
long live positive end-expiratory with the acute respiratory distress J Respir Crit Care Med 174:268–278
pressure. Med Intensiva 36:571–575 syndrome. N Engl J Med 351:327–336 9. Hodgson C, Keating JL, Holland AE,
3. Villar J, Blanco J, Añón JM, Santos- 6. Meade MO, Cook DJ, Griffith LE, Davies AR, Smirneos L, Bradley SJ,
Bouza A, Blanch L, Ambrós A, Gandı́a Hand LE, Lapinsky SE, Stewart TE, Tuxen D (2009) Recruitment
F, Carriedo D, Mosteiro F, Basaldúa S, Killian KJ, Slutsky AS, Guyatt GH manoeuvres for adults with acute lung
Fernández RL, Kacmarek RM, ALIEN (2008) A study of the physiologic injury receiving mechanical ventilation.
Network (2011) The ALIEN study: responses to a lung recruitment Cochrane Database Syst Rev
incidence and outcome of acute maneuver in acute lung injury and acute 2:CD006667
respiratory distress syndrome in the era respiratory distress syndrome. Respir 10. Morris A (2012) Exciting new ECMO
of lung protective ventilation. Intensive Care 53:1441–1447 technology awaits compelling scientific
Care Med 37:1932–1941 [Erratum, 7. Tugrul S, Ozkan A, Perihan OE, Sibel I, evidence for widespread use in adults
Intensive Care Med 37:1942] Figen E, Lutfi T, Kutay A, Cakar N with respiratory failure. Intensive Care
4. Girgis K, Hamed H, Khater Y, (2003) Effects of sustained inflation and Med 38:186–188
Kacmarek RM (2006) A decremental postinflation positive end-expiratory
PEEP trial identifies the PEEP level that pressure in acute respiratory distress
maintains oxygenation after lung syndrome: focusing on pulmonary and
recruitment. Respir Care 51:1132–1139 extrapulmonary forms. Critical Care
Med 31:738–746
1210

11. Madershahian N, Wittwer T, Strauch J, 16. Peek GJ, Mugford M, Tiruvoipati R, 21. Abroug F, Ouanes-Besbes L, Elastrous
Franke UF, Wippermann J, Kaluza M, Wilson A, Allen E, Thalanany MM, S, Brochard L (2008) The effect of
Wahlers T (2007) Application of Hibbert CL, Truesdale A, Clemens F, prone positioning in acute respiratory
ECMO in multitrauma patients with Cooper N, Firmin RK, Elbourne D, distress syndrome or acute lung injury:
ARDS as rescue therapy. J Card Surg CESAR trial collaboration (2009) a meta-analysis. Areas of uncertainty
22:180–184 Efficacy and economic assessment of and recommendations for research.
12. Gariboldi V, Grisoli D, Tarmiz A, conventional ventilatory support versus Intensive Care Med 34:1002–1011
Jaussaud N, Chalvignac V, Kerbaul F, extracorporeal membrane oxygenation 22. Sud S, Friedrich JO, Taccone P, Polli F,
Collart F (2010) Mobile extracorporeal for severe adult respiratory failure Adhikari NK, Latini R, Pesenti A,
membrane oxygenation unit expands (CESAR): a multicentre randomised Guérin C, Mancebo J, Curley MA,
cardiac assist surgical programs. Ann controlled trial. Lancet 374:1351–1363 Fernandez R, Chan MC, Beuret P,
Thorac Surg 90:1548–1552 [Erratum, Lancet 374:1330] Voggenreiter G, Sud M, Tognoni G,
13. MacLaren G, Combes A, Bartlett RH 17. Ferguson ND, Villar J, Slutsky AS Gattinoni L (2010) Prone ventilation
(2012) Contemporary extracorporeal (2007) Understanding high-frequency reduces mortality in patients with acute
membrane oxygenation for adult oscillation: lessons from the animal respiratory failure and severe
respiratory failure: life support in the kingdom. Intensive Care Med hypoxemia: systematic review and
new era. Intensive Care Med 33:1316–1318 meta-analysis. Intensive Care Med
38:210–220 18. Young D, Lamb S, Shah S, MacKenzie 36:585–599
14. Noah MA, Peek GJ, Finney SJ, Griffiths I, Tunnicliffe W, Lall R, Rowan K, 23. Guerin C, Reignier J, Beuret P, Gacouin
MJ, Harrison DA, Grieve R, Sadique Cuthbertson BH, For the OSCAR Study A, Ayzac L, Boulain T, Mercier E,
MZ, Sekhon JS, McAuley DF, Firmin Group (2013) High-frequency Badet M, Mercat A, Desachy A, Clavel
RK, Harvey C, Cordingley JJ, Price S, oscillation for acute respiratory distress M, Robert R, Jaber S, Rosselli S,
Vuylsteke A, Jenkins DP, Noble DW, syndrome. N Engl J Med 368:806–813 Mancebo J, Sirodot M, Bayle F,
Bloomfield R, Walsh TS, Perkins GD, 19. Ferguson ND, Cook DJ, Guyatt GH, Richard JC (2012) Effect du decubitus
Menon D, Taylor BL, Rowan KM Mehta S, Hand L, Austin P, Zhou Q, ventral sur la survie des maladies avec
(2011) Referral to an extracorporeal Matte A, Walter S, Lamontagne F, SDRA confirmé et severe. Essai
membrane oxygenation center and Granton JT, Arabi YM, Arroliga AC, prospectif randomise, contrôlé,
mortality among patients with severe Stewart TE, Slutsky AS, Meade M, For multicentrique. Réanimation 22:S64–
2009 influenza A (H1N1). JAMA the OSCILLATE trial investigators and S67 [Abstract SO096]
306:1659–1668 the Canadian Critical Care Trials Group 24. Raghavendran K, Pryhuber GS, Chess
15. Pham T, Combes A, Rozé H, Chevret S, (2013) High frequency oscillation in PR, Davidson BA, Knight PR, Notter
Mercat A, Roch A, Mourvillier B, Ara- early acute respiratory distress RH (2008) Pharmacotherapy of acute
somohano C, Bastien O, Zogheib E, syndrome. N Engl J Med 368:795–805 lung injury and acute respiratory
Clavel M, Constan A, Richard JCM, 20. Lamm WJ, Graham MM, Albert RK distress syndrome. Curr Med Chem
Brun-Buisson C, Brochard L, For the (1994) Mechanisms by which the prone 15:1911–1924
REVA Research Network (2013) position improves oxygenation in acute 25. Adhikari NK, Burns KEA, Friedrich
Extracorporeal membrane oxygenation lung injury. Am J Respir Crit Care Med JO, Granton JT, Cook DJ (2007) Effect
for pandemic influenza A (H1N1)- 150:184–193 of nitric oxide on oxygenation and
induced acute respiratory distress mortality in acute lung injury:
syndrome. Am J Respir Crit Care Med systematic review and meta-analysis.
187:276–285 BMJ 334:779