The n e w e ng l a n d j o u r na l of m e dic i n e
ern California shows strong association with COVID-19 inci- neighbourhood level wastewater surveillance and subtyping of
dence. mSystems 2021;​6(5):​e0082921.
3. Hughes B, Duong D, White BJ, et al. Respiratory Syncytial
Virus (RSV) RNA in wastewater settled solids reflects RSV clini-
cal positivity rates. Environ Sci Technol Lett 2022;​9:​173-8.
4. Mercier E, D’Aoust PM, Thakali O, et al. Municipal and
Early Active Mobilization
during Mechanical Ventilation in the ICU
To the Editor: Previous investigations have
shown that oversedation and immobilization of
mechanically ventilated patients are associated
with or result in greater length of hospital stay,
decreased survival, and greater incidence of ac-
quired dementia.1-3 The evidence-based ABCDEF
(analgesia monitoring and management, breaks
daily from both sedation and ventilator use,
choice of sedation [avoiding benzodiazepines],
delirium monitoring and management, early
mobility, and family engagement) safety bundle
helps us rehumanize care and establish higher
adherence to key steps; E (for early mobility)
yields dose–response results of earlier discharge
and greater survival.4 The TEAM (Treatment of
Mechanically Ventilated Adults with Early Activ-
ity and Mobilization) trial (Nov. 10 issue)5 is a
landmark achievement.
There was, however, a narrow separation be-
tween the intervention group and the control
group, with only a 12-minute mean difference
per day in mobilization owing to improvements
in the control group as compared with earlier
studies3 and an unintended upper limit of mobil-
ity imposed by heavy sedation in patients in the
intervention group (Figure S4 in the Supplemen-
tary Appendix of the article, available with the
full text of the article at NEJM.org). Data regard-
ing levels of consciousness may help explain
upstream limitations on the intervention group
that were caused by sedation practices. Figure S7
shows similarly high numbers of patients in
both groups unable to make sustained eye con-
tact (indicating a lack of separation between
groups), as measured by a score of −2 or lower
on the Richmond Agitation and Sedation Scale
(RASS), which ranges from −5 (unresponsive) to
4 (combative); 97.8% of patients in the interven-
tion group were receiving continuous sedative
572 n engl j med 388;6  nejm.org  February 9, 2023
The New England Journal of Medicine
Downloaded from nejm.org by TIAGO XAVIER on March 21, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
an influenza virus outbreak. Sci Rep 2022;​12:​15777.
5. Li Y, Zhao H, Wilkins K, Hughes C, Damon IK. Real-time PCR
assays for the specific detection of monkeypox virus West African
and Congo Basin strain DNA. J Virol Methods 2010;​169:​223-7.
DOI: 10.1056/NEJMc2213882
infusions at enrollment. Can the authors provide
data regarding the ongoing frequency of sedative
infusions and the daily mean doses of major
sedatives according to trial group?
E. Wesley Ely, M.D., M.P.H.
Tennessee Valley Geriatric Research Education and Clinical
Center, Veterans Affairs Hospital
Nashville, TN
wes​.­ely@​­vumc​.­org
Dr. Ely reports receiving grants from the National Institutes
of Health and the Department of Veterans Affairs. No other
potential conflict of interest relevant to this letter was reported.
1. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a
paired sedation and ventilator weaning protocol for mechani-
cally ventilated patients in intensive care (Awakening and
Breathing Controlled trial): a randomised controlled trial. Lan-
cet 2008;​371:​126-34.
2. Pandharipande PP, Girard TD, Jackson JC, et al. Long-term
cognitive impairment after critical illness. N Engl J Med 2013;​
369:​1306-16.
3. Schweickert WD, Pohlman MC, Pohlman AS, et al. Early
physical and occupational therapy in mechanically ventilated,
critically ill patients: a randomised controlled trial. Lancet 2009;​
373:​1874-82.
4. Pun BT, Balas MC, Barnes-Daly MA, et al. Caring for criti-
cally ill patients with the ABCDEF bundle: results of the ICU
Liberation Collaborative in over 15,000 adults. Crit Care Med
2019;​47:​3-14.
5. The TEAM Study Investigators and the ANZICS Clinical Trials
Group. Early active mobilization during mechanical ventilation
in the ICU. N Engl J Med 2022;​387:​1747-58.
DOI: 10.1056/NEJMc2216086
To the Editor: The TEAM investigators found
that maximally aggressive escalation of mobili-
zation did not enhance effectiveness as com-
pared with standard early mobilization in mod-
ern intensive care unit (ICU) practice. However,
the potential benefits of early mobilization might
have been diluted by the inclusion of trial par-
ticipants who were too ill to initiate mobilization
early or, alternatively, too well and would have
adequately recovered regardless of the timing of
mobilization. The ICU Mobility Scale (IMS),
 Correspondence

which rates the level of mobilization from 0 (no the control groups in the two trials were similar.1
mobilization) to 10 (independent walking), was The minimization of sedation was an impor-
used to quantify activity but is not linear. In our tant component of the intervention in the TEAM
experience, the momentum to be overcome when trial; however, some patients could not be safely
a patient is moved out of bed to a chair (i.e., pro- de-sedated to facilitate mobilization on each
ceeding from IMS level 0 to 2) — not to mention trial day. Light sedation was achieved in most
the additional personnel and close monitoring patients; 66% (487 of 741 patients) had a highest
that are required — is greater than that in help- RASS score that was in the desired range (−2 to
ing a standing patient to march in place (pro- 1) within the first 2 trial days, and 79% (584 of
ceeding from IMS level 4 to 6).1 A subgroup analy- 741 patients) had a highest RASS score that was
sis of ventilator-free days as compared with in the desired range within the first 3 trial days.
ICU-free days in the early mobilization group We chose to focus on documenting the sedation
and the control group, with the analysis limited levels that were reached and did not collect data
to those trial participants whose IMS level by day on the frequency of use of continuous sedative
3 was 0 to 1 (i.e., not out of bed), may be infor- infusions or daily doses of sedatives necessary to
mative. Patients who were “too well” and had establish those sedation levels. We agree that
higher IMS levels would be excluded, which depth of sedation may be a determinant of treat-
might point to a group more likely to benefit ment response and are undertaking exploratory
from additional early out-of-bed mobilization. analyses to assess whether there is heterogeneity
Jonathan C.H. Cheung, M.B., Ch.B., M.Clin.US. of treatment response according to the depth of
Prince of Wales Hospital sedation at baseline.
Hong Kong, China Our data do not support the assertion that
jonathanchunheicheung@​­gmail​.­com the lack of benefit reflected patients who were
Laptin Ho, M.B., Ch.B., M.Res. either too sick or too well. We found no evidence
North District Hospital of heterogeneity of treatment response accord-
Hong Kong, China ing to illness severity. A postrandomization com-
Yu‑Yeung Yip, M.B., Ch.B. parison of outcomes among patients in the in-
Prince of Wales Hospital tervention group and those in the control group
Hong Kong, China who did not have an IMS level of more than 1 by
No potential conflict of interest relevant to this letter was day 3 may be subject to bias.2 Patients in the
reported.
intervention group who did not have mobility
1. Ho L, Tsang JHC, Cheung E, et al. Improving mobility in the above an IMS level of 1 despite intensive efforts
intensive care unit with a protocolized, early mobilization pro- to mobilize may have differed systematically
gram: observations of a single center before-and-after the imple-
mentation of a multidisciplinary program. Acute Crit Care 2022;​ from patients in the control group who were not
37:​286-94. mobilized. Specific reasons that such patients
DOI: 10.1056/NEJMc2216086 were not mobilized that were likely to affect
their outcomes included physiologic instability,
The authors reply: We submit that a 12-minute end-of-life care, and neurologic issues, as shown
mean between-group difference in active exercise in Figure S4.
per day in the ICU phase of care is a clinically
Carol L. Hodgson, Ph.D.
relevant difference, particularly when one con-
Michael Bailey, Ph.D.
siders that the mobilization time in the interven-
Monash University
tion group was more than double that in the Melbourne, VIC, Australia
usual-care group and that key mobilization mile- carol​.­hodgson@​­monash​.­edu
stones were reached 1 to 2 days sooner in the
intervention group than in the usual-care group. Paul J. Young, M.B., Ch.B., Ph.D.
Improvements in usual-care mobilization prac- Medical Research Institute of New Zealand
tice may not explain why our findings appear to Wellington, New Zealand
be at variance with a previous trial that suggested for the TEAM Study Investigators
benefits from early mobilization, because the time Since publication of their article, the authors report no fur-
from baseline to key mobilization milestones in ther potential conflict of interest.

n engl j med 388;6  nejm.org  February 9, 2023 573

The New England Journal of Medicine
Downloaded from nejm.org by TIAGO XAVIER on March 21, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l
1. Schweickert WD, Pohlman MC, Pohlman AS, et al. Early
physical and occupational therapy in mechanically ventilated,
critically ill patients: a randomised controlled trial. Lancet 2009;​
373:​1874-82.
Oxygen-Saturation Targets for Adults
Receiving Mechanical Ventilation
To the Editor: Semler and colleagues (Nov. 10
issue)1 found no difference in outcomes with the
use of lower (90%), intermediate (94%), and
higher (98%) targets for oxygen saturation as
measured by pulse oximetry (Spo2) in critically ill
patients receiving mechanical ventilation. In-
creased rates of occult hypoxemia, wherein pulse
oximetry is falsely reassuring for a given partial
pressure of arterial oxygen, are well described
among Black patients.2,3 Because only 15% of the
patients in the trial conducted by Semler et al.
were Black, it is not clear whether the results
would be generalizable to populations with a
higher proportion of Black patients or other pa-
tients with dark skin pigment. Accordingly, a
more conservative oxygen target may risk in-
creased rates of unrecognized hypoxemia and be
potentially dangerous in Black patients.
Amira Mohamed, M.D.
Montefiore Medical Center
Bronx, NY
ammohamed@​­montefiore​.­org
Jen‑Ting Chen, M.D.
University of California, San Francisco
San Francisco, CA
Ari Moskowitz, M.D., M.P.H.
Montefiore Medical Center
Bronx, NY
No potential conflict of interest relevant to this letter was
reported.
1. Semler MW, Casey JD, Lloyd BD, et al. Oxygen-saturation
targets for critically ill adults receiving mechanical ventilation.
N Engl J Med 2022;​387:​1759-69.
2. Sjoding MW, Dickson RP, Iwashyna TJ, Gay SE, Valley TS.
Racial bias in pulse oximetry measurement. N Engl J Med 2020;​
383:​2477-8.
3. Wong AI, Charpignon M, Kim H, et al. Analysis of discrep-
ancies between pulse oximetry and arterial oxygen saturation
measurements by race and ethnicity and association with organ
dysfunction and mortality. JAMA Netw Open 2021;​ 4(11):​
e2131674.
DOI: 10.1056/NEJMc2216088
574 n engl j med 388;6  nejm.org  February 9, 2023
The New England Journal of Medicine
Downloaded from nejm.org by TIAGO XAVIER on March 21, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
2. Yusuf S, Wittes J, Probstfield J, Tyroler HA. Analysis and in-
terpretation of treatment effects in subgroups of patients in ran-
domized clinical trials. JAMA 1991;​266:​93-8.
DOI: 10.1056/NEJMc2216086
The authors and a colleague reply: Mo-
hamed and colleagues correctly note that Spo2 is
less accurate in patients with darkly pigmented
skin.1 In our recent study, among patients with
normal Spo2 values, Black patients were more
likely than White patients to have both hypox-
emia (3.5% vs. 1.1%) and hyperoxemia ( 4.7% vs.
2.4%).2 We and others have called for the devel-
opment of oximeters that do not give racially bi-
ased results.2 Until such oximeters are available,
it is difficult to know whether patients with
darkly pigmented skin would have better out-
comes with a higher Spo2 target. Because previ-
ous trials of oxygen-saturation targets have not
reported race,3,4 our trial provides some of the
only available data. Among 392 self-identified
Black patients in our trial, the number of ventila-
tor-free days did not differ between the lower-
target group and the higher-target group (odds
ratio, 0.80; 95% confidence interval, 0.51 to
1.24). Data from one trial are insufficient to an-
swer this important question. Future trials of
oxygen-saturation targets should deliberately re-
cruit persons from historically excluded groups,
including Black, Latinx, Indigenous, and other
groups, so that the full range of skin pigmenta-
tions is represented.
Matthew W. Semler, M.D.
Consuelo H. Wilkins, M.D.,
Todd W. Rice, M.D.
Vanderbilt University Medical Center
Nashville, TN
matthew​.­w​.­semler@​­vumc​.­org
Dr. Wilkins reports no potential conflict of interest relevant
to this letter. Since publication of their article, Drs. Semler and
Rice report no further potential conflict of interest.
1. Sjoding MW, Dickson RP, Iwashyna TJ, Gay SE, Valley TS.
Racial bias in pulse oximetry measurement. N Engl J Med 2020;​
383:​2477-8.
2. Seitz KP, Wang L, Casey JD, et al. Pulse oximetry and race in
critically ill adults. Crit Care Explor 2022;​4(9):​e0758.