Beruflich Dokumente
Kultur Dokumente
Abstract
Introduction: We performed this study to quantify resources required by mechanically ventilated patients with hypoxemia after
critical care transport (CCT) and to assess short-term clinical outcomes. Methods: We performed a retrospective review of
transports of patients with severe hypoxemic respiratory failure from referring hospitals to 3 tertiary care hospitals to assess the
outcomes including in-hospital mortality, ventilator days, intensive care unit length of stay (LOS), hospital LOS, disposition, and
reported neurologic status on hospital discharge as well as medical interventions specific to acute respiratory failure and critical care.
Results: Of 230 patients transported with hypoxemic respiratory failure, 152 survived to hospital discharge, for a mortality rate of
34.5%, despite a predicted mortality of 64% by Acute Physiology and Chronic Health Evaluation II (APACHE II) score. Twenty-five
percent of patients were treated with neuromuscular blockade, 10.1% received inhaled pulmonary vasodilators, and extracorporeal
membrane oxygenation was initiated in 2.6%. Conclusions: In this cohort with hypoxemic respiratory failure transported to tertiary
care facilities, patients had a mortality rate comparable to patients with acute respiratory distress syndrome treated with best practices
and a mortality rate lower than predicted based on APACHE-II score. The risks of CCT are outweighed by the benefits of transfer to a
tertiary care facility, and pretransport hypoxemia should not be used as an absolute contraindication to transport.
Keywords
complications, critical care, respiratory failure, hospital mortality, length of stay, mechanical ventilation, outcomes, resource
utilization
1
Division of Pulmonary, Critical Care and Sleep Medicine, Medical University
of South Carolina, Charleston, SC, USA
Introduction 2
Division of Emergency Medicine, Medical University of South Carolina,
Charleston, SC, USA
Over the last decade, studies have shown that transferring 3
Department of Emergency Medicine, Brigham and Women’s Hospital, Bos-
patients with acute respiratory distress syndrome (ARDS) to ton, MA, USA
tertiary care centers, especially those with extracorporeal mem- 4
Boston MedFlight, Bedford, MA, USA
5
brane oxygenation (ECMO) capabilities, improved survival as Division of Trauma and Acute Care Surgery, Department of Surgery, Brigham
compared to remaining in community hospitals.1,2 This finding and Women’s Hospital, Boston, MA, USA
6
Division of Vascular and Endovascular Surgery, Department of Surgery,
has increased interest in the development of ARDS centers,
Brigham and Women’s Hospital, Boston, MA, USA
similar to trauma or burn centers.3,4 However, patients with 7
Department of Emergency Medicine, Beth Israel Deaconess Medical Center,
hypoxemia specifically have been considered one of the most Boston, MA, USA
8
precarious patient populations to transport between facili- Department of Anesthesia Critical Care, Division of Critical Care, Beth Israel
ties,5-7 and mechanical ventilation has been shown to be an inde- Deaconess Medical Center, Boston, MA, USA
pendent risk factor for an adverse events during transport.8,9 Received August 2, 2015, and in revised form November 24, 2015.
Prior studies have demonstrated that patients on mechanical ven- Accepted for publication November 30, 2015.
tilation are more than twice as likely to have an adverse event in
transport as compared to those not on ventilation.9 Ventilation Corresponding Author:
Susan R. Wilcox, Division of Pulmonary, Critical Care and Sleep Medicine,
with an fraction of inhaled oxygen (FiO2) of greater than 50% has Division of Emergency Medicine, Medical University of South Carolina, 96
also been associated with the risk of respiratory deterioration Jonathan Lucas Street, Suite 812-CSB, Charleston, SC 29425-6300, USA.
after transfer.7 Clinical circumstances may warrant transport, Email: wilcoxsu@musc.edu
however, as the benefits of transferring to a tertiary care facility as a rehabilitation center or skilled nursing facility. The primary
may offset the risks of transporting a mechanically ventilated outcome was in-hospital mortality. Secondary outcomes were
patient with hypoxemia.5,10 In the United States, these high ventilator days, ICU LOS, hospital LOS, disposition, and
acuity patients are often transported by critical care transport reported neurologic status on hospital discharge. As this was a
(CCT) teams, generally composed of specially trained nurses retrospective review, without dedicated neurologic testing
and paramedics11-13 who provide intensive care procedures recorded on each patient, the physical examination the day of
beyond advanced life support capabilities, such as airway man- discharge as well as the discharge summary narrative was
agement, advanced ventilator support, invasive monitoring, and reviewed to determine patients’ neurologic status, noting
other high level interventions similar to the care a patient would whether the patient’s mental status at discharge was at the pre-
receive in an intensive care unit (ICU). morbid baseline or not, per the documentation of the discharging
Although prior studies have assessed outcomes associated provider. If the clinicians documented a ‘‘normal’’ or ‘‘baseline’’
with interfacility transport of critically ill patients,2,14-20 the neurologic examination, patients were documented as having
majority of prior studies involve heterogeneous populations achieved their premorbid baseline neurologic status.
of critically ill patients.14-18 Only one prior study has evaluated Data regarding resource utilization were extracted from dis-
the outcomes of patients transported with severe acute respira- charge summaries and hospital medical records. Medical inter-
tory failure.2 Furthermore, resource utilization allocated to ventions specific to acute respiratory failure and hypoxemia
critically ill patients with hypoxemia after interfacility CCT were tabulated, including the use of neuromuscular blockade,
has not been well described in the literature. esophageal balloon placement, administration of pulmonary
To quantify the resources required by critically ill, mechani- vasodilators, and ECMO initiation. Other medical interven-
cally ventilated patients with hypoxemia after CCT and to assess tions not necessarily specific to respiratory failure that were
these patients’ short-term clinical outcomes, we performed a ret- performed in the ICU after transport were similarly extracted
rospective analysis of patients transported from community hos- and tabulated.
pitals to tertiary care centers with hypoxemic respiratory failure. Outcomes and resource utilization for patients who survived
We quantified clinical interventions performed in the ICU after to discharge were compared to patients who did not survive the
transport, including interventions specific to respiratory failure posttransport hospitalization. In addition, subgroup analyses of
as well as general intensive care procedures. Clinical outcomes outcomes and resource utilization were performed for patients
were determined, including ventilator days, ICU length of stay categorized by pretransport oxygenation based on pulse
(LOS), hospital LOS, disposition, and in-hospital mortality rate. oximetry.
We performed subgroup analyses of patients categorized by pre- Data were analyzed in a descriptive manner with patient
transport oxygen saturation (SpO2) in order to determine the out- demographics and clinical outcomes reported as means and
comes and resources used for patients with different levels of standard deviations. The number of interventions performed
pretransport hypoxemia. for all patients in the cohort was determined and is summarized
as unadjusted event rates and percentages.
After tabulating results in Microsoft Excel (Microsoft, Red-
Methods mond, Washington), data were exported to JMP Pro version
This study was approved by the internal review boards (IRBs) 11.0 (SAS Institute Inc, Cary, North Carolina) for statistical
of the 3 receiving hospitals, with all IRBs waiving the need for analyses. Comparisons between outcomes and resource utiliza-
informed consent. We performed a retrospective review of tion for survivors versus nonsurvivors were performed using
transports of mechanically ventilated patients with severe unmatched two-tailed student t tests with unequal variances.
hypoxemic respiratory failure from October 2009 to December For the subgroup analysis of patients with different pretransport
of 2012 from referring hospitals to 3 tertiary care hospitals. All SpO2, patients were categorized into one of the 3 subgroups based
decisions to transfer a patient were initiated by the physicians at on initial saturations of 100% to 90%, 89% to 80% for moderate
sending facilities. Transport records were queried for the terms hypoxemia, or <80% for severe hypoxemia.22,23 One-way analy-
‘‘ARDS,’’ ‘‘hypoxia,’’ or ‘‘hypoxemia.’’ Charts of patients sis of variance was used to compare outcomes and resource utili-
receiving an FiO2 of at least 50% were selected for inclusion, zation between these 3 subgroups. Pearson correlations were
and discharge summaries from the tertiary care hospital were calculated to assess for associations between mortality and trans-
reviewed for resource utilization and outcomes data. Full study port time between hospitals. For all statistical analyses, an a of
details have been described previously.21 <.05 was considered to be statistically significant.
Transport records were reviewed for demographic data and
pertinent comorbidities, including obesity, asthma, chronic
obstructive pulmonary disease, immunosuppression, and preg- Results
nancy as well as diagnosis. Outcomes data extracted from We identified 239 charts of patients transported with
receiving hospital records included in-hospital mortality, venti- hypoxemic respiratory failure from 51 community hospitals
lator days, ICU LOS in days, and hospital LOS in days. Dis- to 3 tertiary care centers. Of these, 230 charts had follow-
charge summaries were reviewed to determine if discharge up information available. Patients were high acuity, with
disposition was to home or to another health care facility such a mean Acute Physiology and Chronic Health Evaluation II
Table 3. Respiratory Interventions for All Patients. Table 4. Additional Interventions and Resources.
Interventions Number (%) SpO2 90% (n ¼ 184) SpO2 80%-89% (n ¼ 39) SpO2 <80% (n ¼ 5) F statistic P Value
information for clinicians at the sending facility and the CCT survival without severe disability, as this study demonstrates
team to better discuss the risks and benefits of transport, such that despite risks of transport, the benefits of tertiary care for
that families appreciate that even after transport to a tertiary hypoxemic respiratory failure on mortality are substantial.
care facility, individual outcomes may still be poor in critically Despite the high acuity of this population, this cohort of
ill, mechanically ventilated patients with hypoxemia. transferred patients overall had better than predicted out-
Despite the outcomes observed in this study, it is unlikely that comes. APACHE II has been shown to be a good predictor
transport directly contributed to poor outcomes. We found no sig- of outcomes in critically ill patients,24 especially patients with
nificant association between mode of transport, time en route, or respiratory failure.25 The mean APACHE II score in this
total transport time and outcomes. Previous work has shown that cohort was 28, with a predicted mortality of approximately
transport by a dedicated CCT team is associated with an improve- 64%,26 compared to the rate of 34% in our study. Addition-
ment in both the partial pressure of oxygen (PaO2) and the ratio of ally, patients with moderate ARDS, with a PaO2/FiO2 ratio
PaO2/FiO2.17,21 In assessing the oxygenation parameters of these of 100 to 200 mm Hg, when treated with best practices
transports, we previously reported that in these patients the mean including low tidal volume ventilation, have a mortality rate
change in PaO2/FiO2 ratio from the sending to the receiving hospi- of 32% (95% CI: 29-34).27,28 Our mortality rate was similar,
tal was an increase of 27.62 (confidence interval [CI] 15.84- despite the severity of disease in this group, reinforcing the
39.40, P ¼ .0003) and the mean change in PaO2 was an increase premise that transfer for optimal respiratory and tertiary care
of 27.85 torr (CI 17.49-38.22, P < .0001).21 Given improved oxy- can offer patients the lowest possible mortality rate.
genation after CCT, transport itself is unlikely to directly result in The patients in this cohort required a substantial amount of
high mortality rates, and the severity of patients’ disease is a more resources. This patient population had both long ICU and long
likely explanation for outcomes. hospital LOS, and survivors had substantially longer LOS than
Furthermore, the only randomized controlled trial involving nonsurvivors, demonstrating a survivor bias. Although many
transporting patients with hypoxemic respiratory failure, the of the interventions performed during the hospitalization are not
Conventional Ventilation or ECMO for Severe Adult Respira- unique to tertiary care centers, all require specialist expertise and
tory Failure (CESAR) Trial, published in 2009, found that dedicated equipment, such as echocardiograms, bronchoscopies,
patients randomized to be transferred to an ECMO center had and dialysis. Additionally, many of these interventions, espe-
a 36.7% 6 month mortality compared to 48.9% for those rando- cially respiratory interventions such as neuromuscular blockade
mized to care in their original institution.2 Notably, in this and inhaled pulmonary vasodilators, require considerable staff
study, 2 of the 90 patients randomized to transport for ECMO resources and require time and training from both nursing and
died in transfer. However, the authors still recommend transfer- respiratory care providers. The patterns and frequency of respira-
ring adult patients with severe but potentially reversible tory rescue therapy utilization in this cohort of patients trans-
respiratory failure to an ECMO center to significantly improve ferred between hospitals are similar to previous reports of
homogenous populations of patients who did not undergo inter- resource utilization is similar to undifferentiated cohorts of
facility transport, but rather, presented directly to the tertiary patients with ARDS who did not undergo interfacility transfer.
care facility.29 That rescue therapy utilization is similar, indicat- Given the small numbers of patients with severely hypoxemia
ing that interfacility transport does not appear to substantively in this study, conclusions regarding this subgroup of patients are
change clinical practice patterns and resource utilization in limited. While mortality rates and resource utilization are sub-
mechanically ventilated patients with hypoxemia. These obser- stantial in this population, the risks of CCT appear to be out-
vations can assist tertiary care anticipate resource allocation weighed by the benefits of transfer from a community hospital
needs when accepting such patients from community hospitals. to a tertiary care facility, and pretransport hypoxemia should not
To evaluate the effect of hypoxemia at the time of the CCT be used as an absolute contraindication to interfacility transport.
team’s arrival at the sending facility, we compared outcomes
and resource utilization for patients with mild, moderate, and Author’s Note
severely hypoxemia. Unsurprisingly, patients with more severe The work contained herein was performed at Massachusetts General
hypoxemia were more likely to be treated with many respira- Hospital, Brigham and Women’s Hospital, and Beth Israel Deaconess
tory interventions. However, the outcomes were not statisti- Medical Center, Boston, MA, USA.
cally significantly different, with the sole exception of fewer
patients with normal or baseline neurologic status among Declaration of Conflicting Interests
patients with severe hypoxemia, although there were only 2 The author(s) declared no potential conflicts of interest with respect to
patients in who survived in the severe hypoxemia group, mak- the research, authorship, and/or publication of this article.
ing drawing conclusions in this group difficult.
Therefore, although transport to a tertiary care hospital does Funding
not exclude possible adverse outcomes such as death or morbid- The author(s) disclosed receipt of the following financial support for
ity, the benefits of tertiary care outweigh risks of transport for the research, authorship, and/or publication of this article: This study
critically ill, mechanically ventilated patients with hypoxemia. was supported by institutional funds without external grant support.
Transfer to a tertiary care center to receive evidence based,
resource-intensive care likely offers significant advantage over References
remaining at a community hospital with limited resources. 1. Noah MA, Peek GJ, Finney SJ, et al. Referral to an extracorpor-
eal membrane oxygenation center and mortality among patients
Limitations with severe 2009 influenza A (H1N1). JAMA. 2011;306(15):
1659-1668.
The limitations of this study are largely related to the nature of the 2. Peek GJ, Mugford M, Tiruvoipati R, et al. Efficacy and economic
retrospective design. Discharge records were reviewed thor- assessment of conventional ventilator support versus extracorpor-
oughly, however, some details of the hospitalization or resources eal membrane oxygenation for severe adult respiratory failure
used may have not been recorded. However, this would underes- (CESAR): a multicentre randomized controlled trial. Lancet.
timate the resources used, and therefore, the observation that this 2009;374(9698):1351-1363.
patient population requires substantial resources remains valid. 3. Adalja AA, Watson M, Waldhorn RE, Toner ES. A conceptual
The ability to assess for neurologic outcome was limited by nature approach to improving care in pandemics and beyond: Severe
of the retrospective review, as the patients did not necessarily lung injury centers. J Crit Care. 2013;28(3):318. e9-15.
undergo dedicated neurologic assessment prior to discharge. 4. Kahn JM, Linde-Zwirble WT, Wunsch H, et al. Potential value of
However, we used the physical examination documented at the regionalized intensive care for mechanically ventilated medical
time of discharge, as noted by the discharging provider, to record patients. Am J Respir Crit Care Med. 2008;177(3):285-291.
the neurologic assessment by the team providing care. As the 5. Warren J, Fromm RE Jr, Orr RA, Rotello LC, Horst HM.
decision to initiate ECMO was made on a case by case basis, the American College of Critical Care Medicine. Guidelines for the
generalizability of the ECMO findings may be limited. The sub- inter- and intrahospital transport of critically ill patients. Crit
groups of moderate and severe hypoxemia may have been under- Care Med. 2004;32(1):256-262.
powered to detect differences in interventions and outcomes. 6. Beckmann U, Gillies DM, Berenholtz SM, Wu AW, Pronovost P.
Additionally, these patients were transported by a highly trained Incidents relating to the intra-hospital transfer of critically ill
CCT team, and whether similar results could be obtained with patients. an analysis of the reports submitted to the australian inci-
other transport systems is unknown. dent monitoring study in intensive care. Intensive Care Med.
2004;30(8):1579-1585.
7. Marx G, Vangerow B, Hecker H, Leuwer M, Jankowski M,
Conclusion Piepenbrock S, Rueckoldt H. Predictors of respiratory function
In this group of patients transported to tertiary care facilities, deterioration after transfer of critically ill patients. Intensive Care
patients had a mortality rate comparable to patients having Med. 1998;24(11):1157-1162.
ARDS treated with best practices and a mortality rate lower than 8. Singh JM, MacDonald RD, Bronskill SE, Schull MJ. Incidence
predicted based on APACHE-II score. However, they had a sub- and predictors of critical events during urgent air-medical trans-
stantial use of resources at the receiving hospital, although port. CMAJ. 2009;181(9):579-584.
9. Singh JM, Macdonald RD, Ahghari M. Critical events during land- 20. Strauch U, Bergmans DCJJ, Winkens B, Roekaerts PMHJ. Short-
based interfacility transport. Ann Emerg Med. 2014;64(1):9-15.e2. term outcomes and mortality after interhospital intensive care
10. Whiteley S, Macartney I, Mark J, Barratt H, Binks R. On behalf of transportation: an observational prospective cohort study of 368
the Council of the Intensive Care Society. Guidelines for the consecutive transports with a mobile intensive care unit. BMJ
Transport of the Critically Ill Adult. 3rd edition. London: Inten- Open. 2015:5(4):3006801.
sive Care Society; 2011. 21. Wilcox SR, Saia MS, Waden H, et al. Improved oxygenation after
11. Roeder JR. Flight team configuration of an air medical serivce. transport in patients with hypoxemic respiratory failure. Air Med
Top Emerg Med. 1994;16(4):66-72. J. 2015;34(6):369-376.
12. Gebremichael M, Borg U, Habashi NM, Cottingham C, Cunsolo 22. Mort TC. Emergency tracheal intubation: complications associ-
L, McCunn M, . . . Reynolds HN. Interhospital transport of the ated with repeated laryngoscopic attempts. Anesth Analg. 2004;
extremely ill patient: the mobile intensive care unit. Crit Care 99(2):607-613.
Med. 2000;28(1):79-85. 23. Mort TC, Waberski BH, Clive J. Extending the preoxygenation
13. Kashani KB, Farmer JC. The support of severe respiratory failure period from 4 to 8 mins in critically ill patients undergoing emer-
beyond the hospital and during transportation. Curr Opin Crit gency intubation. Crit Care Med. 2009;37(1):68-71.
Care. 2006;12(1):43-49. 24. Beck DH, Taylor BL, Millar B, Smith GB. Prediction of outcome
14. Fan E, MacDonald RD, Adhikari NK, Scales DC, Wax RS, Stewart from intensive care: a prospective cohort study comparing Acute
TE, . . . Ferguson ND. Outcomes of interfacility critical care adult Physiology and Chronic Health Evaluation II and III prognostic
patient transport: a systemic review. Crit Care. 2006;10(1):R6. systems in a United Kingdom intensive care unit. Crit Care Med.
15. Barillo DJ, Dickerson EE, Cioffi WG, Mozingo DW, Pruitt BA Jr. 1997;25(1):9-15.
Pressure-controlled ventilation for the long-range aeromedical 25. Del Bufalo C1, Morelli A, Bassein L, et al. Severity scores in
transport of patients with burns. J Burn Care Rehabil. 1997; respiratory intensive care: APACHE II predicted mortality better
18(3):200-205. than SAPS II. Respir Care. 1995;40(10):1042-1047.
16. Remond C, Jimeno MT, Dubouloz F. Mesures du CO2 expire 26. Kane SP. Acute Physiology and Chronic Health Evaluation
en transport extrahospitalier: Interets et limites. Jeur. 1998;4: (APACHE II) Calculator. Web site. ClinCalc: //clincalc.com/
179-186. IcuMortality/APACHEII.aspx. Published October 26, 2011.
17. Uusaro A, Parviainen I, Takula J, Ruokonen E. Safe long-distance Updated November 20, 2014. Accessed July 25, 2015.
interhospital ground transfer of critically ill patients with acute 27. ARDS Definition Task Force, Ranieri VM, Rubenfeld GD, et al.
severe unstable respiratory and circulatory failure. Intensive Care Acute respiratory distress syndrome: the Berlin Definition. JAMA.
Med. 2002;28(8):1122-1125. 2012;307(23):2526-2533.
18. Veldman A, Diefenbach M, Fischer D, Benton A, Bloch R. Long- 28. Khandelwal N, Hough CL, Bansal A, Veenstra DL, Treggiari
distance transport of ventilated patients: advantages and limita- MM. Long-term survival in patients with severe acute respiratory
tions of air medical repatriation on commercial airlines. Air Med distress syndrome and rescue therapies for refractory hypoxemia.
J. 2004;23(2):24-28. Crit Care Med. 2014;42(7):1610-1618.
19. Orf J, Thomas SH, Wedel SK. Ventilation rates in intubated head 29. Walkey AJ, Wiener RS. Utilization patterns and patient outcomes
injury patients undergoing helicopter EMS (HEMS) transport associated with use of rescue therapies in acute lung injury. Crit
[abstract] Crit Care Med. 2000;28:A208. Care Med. 2011;39(6):1322-1328.