Full text of DOI:10.1007/BF03017346

844
General Anesthesia
Best evidence in anesthetic practice
Prevention: supplemental oxygen reduces the
incidence of surgical-wound infection
Article appraised
to staples removal, duration of postoperative hospital-
ization, proportion of ICU admission, and 15-day
mortality were secondary outcomes. Amounts of col-
lagen and protein deposition were measured in a sub-
group of patients (32 receiving 30% O2; 22 receiving
80% O2).
Grief R, Akça O, Horn E-P, Kurz A, Sessler DI, for the
Outcomes Research Group. Supplemental perioperative
oxygen to reduce the incidence of surgical-wound
infection. N Engl J Med 2000; 342: 161–7.
Structured abstract by Peter Choi, Department of
Anesthesia, McMaster University, Hamilton, Ontario,
Canada.
Question: Can supplemental administration of oxygen
during the perioperative period decrease the incidence
of postoperative wound infections in patients under-
going colorectal resection?
Design: Multicenter, randomized, double-blind,
controlled trial.
Setting: Three hospitals in Europe (Austria and
Germany).
Main results: Analysis was intention-to-treat.
Clinical characteristics, diagnoses, surgical procedures
and duration, hemodynamic values, use of anesthetics,
and types and duration of prophylactic antibiotic
adminstration were similar between both groups.
Overall incidence of surgical wound infections was 8%.
The 80% O2 group had fewer wound infections (5.2%)
compared to the 30% O2 group (11.2%) [absolute risk
reduction 6.0%; 95% confidence interval 1.2–10.8%;
number-needed-to-treat 17; P=0.01]. Similarly, the
mean ASEPSIS score was lower in the 80% O2 group
Patients: Five hundred patients between 18 to 80 yr
of age (mean age 57 yr, 44% women) undergoing elec-
tive open colorectal resection. Exclusion criteria were
minor colon surgery, recent history of fever or infec-
tion, serious malnutrition, and bowel obstruction.
Intervention: Two hundred fifty patients were allo-
cated to 30% oxygen / 70% nitrogen (30% O2 group);
250 patients were allocated to 80% oxygen / 20%
nitrogen (80% O2 group). Both received the assigned
concentrations during anesthesia until immediately
before extubation when oxygen was increased to
100%. After extubation, as soon as deemed safe by the
anesthesiologist, the oxygen concentration was
returned to the allocated levels for the first two hours
of recovery.
(3
7) compared to the 30% O2 group (5
9;
P=0.01). There were no statistically significant differ-
ences between groups for the other outcomes.
Conclusion: Administration of 80% supplemental
oxygen during colorectal resection and for two hours
afterwards halved the 15-day incidence of surgical-
wound infection.
Funding: U.S. National Institutes of Health, Fonds
zur Förderung der wissenschafflichen Forschung,
Joseph Drown Foundation, Anesthesia Patient Safety
Foundation, Max Kade Foundation, Erwin-
Schrödinger Foundation, Bürgermeister Fond der
Stadt Wien, Austrian National Bank Fund.
Correspondence: Dr. Sessler, Outcomes Research
Group, Department of Anesthesia, 374 Parnassus
Avenue, Third Floor, University of California – San
Francisco, San Francisco, CA 94143-0648, USA.
Email: sessler@anesthesia.ucsf.edu
Main outcomes: Fifteen-day incidence of surgical-
wound infection was the primary outcome. ASEPSIS
score, white-cell counts, time to first solid food, time
CAN J ANESTH 2001 / 48: 9 / pp 844–846

BEST EVIDENCE IN ANESTHETIC PRACTICE
845
Commentary by A. Denault, D. Fréchette
benefit be observed? On the other hand, the treat-
ment is so simple and inexpensive that one could
argue that the cost-effectiveness of such a strategy is
not prohibitive. In the era of evidence-based medi-
cine, the administration of 80% O2 for two hours for
all patients undergoing colorectal surgery can be con-
sidered level I evidence (large randomized trial with
clear-cut results).4 However, the universal application
of such a strategy cannot be generalized to all postop-
erative patients unless a higher oxygen concentration
is indicated.
Greif et al. demonstrated that the use of supplemental
oxygen reduced the frequency of surgical-wound
infections in colorectal surgery patients by half from
11.2% to 5.2%. In evaluating a randomized controlled
trial on therapy, validity and applicability of the results
to my population are the critical issues.1
In this single-blind randomized trial, the surgeons
evaluating the wounds were blinded to oxygen concen-
tration and all patients entering the trial were account-
ed for. Both groups were treated equally, apart from a
higher end-tidal isoflurane concentration in the 80% O2
group. This difference was small and clinically insignifi-
cant. Thirty- eight patients in the 30% O2 group
required higher FIO2 but the duration of exposure is
unknown. The definition of malnourished patients,
who were excluded, is unclear. This group is unlikely to
be immunocompetent and could be the one that bene-
fits most from an intervention aiming to reduce rates of
infection. The authors do not give a rationale for select-
ing two hours of 80% O2 as the intervention.
How large is the treatment effect? This was signifi-
cant for the 80% O2 group with an absolute risk reduc-
tion in surgical wound infection of 6%, which
represents a number-needed-to-treat of 17. However,
the confidence intervals for surgical wound infection
rates did overlap between the 30% O2 group (7.3% to
15.1%) and the 80% oxygen group (2.4% to 8.0%).
Patients with postoperative infections had longer
duration of hospitalization but the duration did not
differ between the study groups.
André Denault MD FRCPC
Denise Fréchette MD MPH
Montréal, Québec
References
1 Cook DJ, Levy MM. Evidence-based medicine. A tool
for enhancing critical care practice. Crit Care Clinics
1998; 14: 353–8..
2 Kotani N, Hashimoto H, Sessler DI, et al. Supplemental
intraoperative oxygen augments antimicrobial and
proinflammatory responses of alveolar macrophages.
Anesthesiology 2000; 93: 15–25.
3 Knight PR, Holm BA. The three components of hyper-
oxia (Editorial). Anesthesiology 2000; 93: 3–5.
4 Sackett DL. Rules of evidence and clinical recommen-
dations on the use of antithrombotic agents. Chest
1989; 95(suppl. 2): 2–4.
Commentary by Y. Skrobik
How can we explain that exposure to higher oxy-
gen concentration reduces the risk of infection but not
the duration of hospitalization? Safety was document-
ed in a subgroup of 30 patients receiving 80% O2.
There were no significant side effects such as atelecta-
sis. This would reassure clinicians about the safety of
such a treatment. Kotani et al. have shown that hyper-
oxia (100% FIO2) can prevent a decrease in antimicro-
bial therapy of alveolar macrophages and increase
pro-inflammatory cytokines.2 However, several studies
have demonstrated that hyperoxia can be harmful to
the lung and inactivate surfactant. This cellular dam-
age would not be seen necessarily with conventional
imaging studies such as chest radiograph or computed
tomography. Caution to the universal application of
high oxygen concentrations has been suggested by
Knight and Holm.3
Five percent of all surgical procedures, and 10 to 20%
of colorectal surgical interventions, are complicated by
wound infections. Predictable increases in morbidity
and cost ensue. Infectious outcome depends on the
invading organism’s load and virulence, and the ability
of the host’s tissue to protect itself. Tissue oxygenation
is felt to be significant in host defense because of neu-
trophil-activated oxidative killing, in which bactericidal
superoxide radicals play an important role.
The study by Grief et al. compares 30% FIO2 to
80% FIO2 during surgery and in the immediate post-
operative period in patients undergoing elective
colonic resection. The results indicate that administra-
tion of the higher oxygen concentration is associated
with a lower incidence of postoperative wound infec-
tion. In the subset of patients in whom tissue oxygen
concentration is measured (sc or intramuscularly),
interstitial pO2 levels are higher in the 80% O2 group.
This finding provides a physiologic explanation for the
improved outcome in patients given 80% O2.
Do the results apply to all postoperative patients?
On one hand, the study population had a non-sterile
surgical site that is different from the wound from an
elective orthopedic procedure in which surgery is per-
formed in a sterile environment. In the latter, would a
The authors randomized 500 patients. Severely mal-
nourished, or clinically infected patients, as well as those

846
CANADIAN JOURNAL OF ANESTHESIA
3 Grief R, Laciny S, Rapf B, Hickle RS, Sessler DI.
Supplemental oxygen reduces the incidence of postop-
erative nausea and vomiting. Anesthesiology 1999; 91:
1246–52.
4 Buggy D. Can anaesthetic management influence surgi-
cal wound-healing? Lancet 2000; 356: 355–7.
suffering from bowel obstruction, were excluded.
Chronic inflammatory disease and bowel cancer
patients, whose immune status could influence infec-
tious outcome, numbered equally in both groups, as
did factors such as age and the American Society of
Anesthesiologists physical status score. Risk factors for
wound infection, other than imminent colon surgery,
were thus evenly distributed in the two groups.
Perioperative management was carefully tailored to
allow maximal tissue perfusion pressures by generously
infusing iv crystalloid. Body temperature was fastidious-
ly maintained at 36.5°C, since the benefit of normoth-
ermia with regards to wound infections is known.1
Thirty-eight of 250 patients in the 30% O2 group and
only one in the 80% O2 group required perioperative
increase in oxygen administration. Hemodynamics were
similar in the two groups, implying perioperative insta-
bility was not masked in the 80% oxygen group.
Potential adverse effects of 80% O2 administration were
addressed. Frequency of postoperative atelectasis, care-
fully evaluated with roentgenograms and computed
tomography in a subgroup of patients, was not different
between the two groups.2 Unexpectedly, the 80% O2
group showed a marked (50%) decrease in postopera-
tive nausea and vomiting.3
Wound healing rates were no different between
groups. The authors carefully and objectively docu-
mented postoperative wound infection. Wound infec-
tion rates were significantly worse in the lower oxygen
group (11.2% in 30% O2 group vs 5.2% in the 80% O2
group; P=0.01). This translated into significant mor-
bidity (sepsis, multi- organ failure, and ICU admis-
sions) and cost.
This study shows a simple, safe, and inexpensive way
of minimizing wound infections. Benefits to the patient
extend to how they feel in the immediate postoperative
period. The paper’s message emphasizes the role of
anesthesiologists as active participants in the outcome
of the whole disease process,4 in addition to their role
as providers of perioperative stability and analgesia.
Yoanna Skrobik MD FRCPC
Montréal, Québec
References
1 Kurz A, Sessler DI, Lenhardt R. Perioperative nor-
mothermia to reduce the incidence of surgical-wound
infection and shorten hospitalization. N Engl J Med
1996; 334: 1209–15.
2 Akça O, Podolsky A, Eisenhuber E, et al. Comparable
postoperative pulmonary atelectasis in patients given
30% or 80% oxygen during two hours after colon
resection. Anesthesiology 1999; 91: 991–8.