NASA employs some of the smartest and most dedicated professionals in the world.
Individually they are the cream of the crop in their various science, engineering, and
administrative disciplines. Their collective achievements have dramatically expanded
the boundaries of our knowledge of both our universe and our world. And yet in spite
of their individual dedication and collective brilliance, on February 1, 2003, seven
astronauts died—perhaps unnecessarily—when the Columbia Shuttle Mission STS-107
incinerated on reentry into the earth’s atmosphere. The reason? A key contributor to
the tragedy was a culture that “prevented effective communication of critical safety
information and stifled professional differences of opinion.”1 People’s inability to
confront risky topics allowed risks to go unaddressed—contributing to the loss of seven
lives.
Those who serve daily in U.S. hospitals could be described as similarly intelligent
and dedicated. In the aptly titled report To Err is Human2, the Institute of Medicine
acknowledges both the individual dedication and collective contributions of those who
give their all to improve and save the lives of patients in our healthcare institutions.
And yet, as the report points out, each year hundreds of thousands of patients are
brought to harm in the course of their healing because of fundamental problems in the
collective behavior of these caring professionals. These problems are not unlike those
that contributed to the loss of Columbia. For example, each year one in twenty in-patients
at hospitals will be given a wrong medication, 3.5 million will get an infection from
someone who didn’t wash his or her hands or take other appropriate precautions3, and
195,000 will die because of mistakes made while they’re in the hospital.4
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
Hospitals are responding aggressively to this crisis with new technologies, qualityimprovement systems, and methods of organizing. However, though the healthcare
community is taking needed action on a number of fronts, there is a deeper problem
that must be resolved before acceptable levels of improvement will be attainable.
As with NASA personnel, key problems that contribute to these tragic errors are
often known far in advance. And yet few people talk about them. Every day, many
healthcare workers stand next to colleagues and see them cut corners, make mistakes,
or demonstrate serious incompetence. But only a small percentage speak up and discuss
what they have seen—even though they’re standing only a few feet away. As a result,
problems go on for years—contributing to avoidable errors, high turnover, decreased
morale, and reduced productivity. Just as the unwitting behavior of well-intended
NASA personnel served to suppress key information that might have escalated risks,
many healthcare workers tend to act in ways that allow risks and problems to remain
unaddressed—sometimes for years.
A group of eight anesthesiologists agree a peer is dangerously incompetent, but
they don’t confront him. Instead, they go to great efforts to schedule surgeries for
the sickest babies at times when he is not on duty. This problem has persisted for
over five years. (Focus Group of Physicians)
A group of nurses describe a peer as careless and inattentive. Instead of
confronting her, they double check her work—sometimes running in to patient
rooms to retake a blood pressure or redo a safety check. They’ve “worked around”
this nurse’s weaknesses for over a year. The nurses resent her, but never talk to her
about their concerns. Nor do any of the doctors who also avoid and compensate
for her. (Focus Group of Nurses)
Past studies have indicated that more than 60 percent of medication errors are caused by
mistakes in interpersonal communication. The Joint Commission on Accreditation of
Healthcare Organizations suggests that communication is a top contributor to sentinel
events.5 This study builds on these findings by exploring the specific concerns people
have a hard time communicating that may contribute to avoidable errors and other
chronic problems in healthcare.
The study we report here suggests that there are seven crucial conversations that people
in healthcare frequently fail to hold that likely add to unacceptable error rates. The
nationwide study was conducted by VitalSmarts in partnership with the American
Association of Critical-Care Nurses. This study suggests that improvement in these seven
crucial conversations could not only contribute to significant reductions in errors, but
also to improvements in quality of care, reduction in nursing turnover, and marked
improvement in productivity.
In addition, we will offer healthcare leaders a simple method for measuring their current
performance in these seven crucial conversations, as well as an action plan for making
measurable improvement in this key competency.
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
Researchers conducted dozens of focus groups, interviews, and workplace observations,
and then collected survey data from more than 1,700 respondents, including 1,143
nurses, 106 physicians, 266 clinical-care staff, and 175 administrators during 2004. Their
research sites included thirteen urban, suburban, and rural hospitals from across the U.S.
These included a mix of teaching, general, and pediatric hospitals. Although this is a
modest sample, the findings fit together in a significant and compelling way.
The study identified the categories of conversations that are especially difficult and, at
the same time, especially essential for people in healthcare to master. The study showed
that the quality of these crucial conversations relates strongly with medical errors, patient
safety, quality of care, staff commitment, employee satisfaction, discretionary effort, and
turnover. We grouped these concerns into seven areas: Broken Rules, Mistakes, Lack of
Support, Incompetence, Poor Teamwork, Disrespect, and Micromanagement.
More than half of the healthcare workers surveyed in this study had occasionally
witnessed broken rules, mistakes, lack of support, incompetence, poor teamwork,
disrespect, and micromanagement. Many had seen some of their colleagues cutting
corners, making mistakes, and demonstrating serious incompetence. However, even
though they had these concerns, fewer than one in ten fully discussed their concerns
with the coworker. Furthermore, most healthcare workers neither believe it’s possible nor
even their responsibility to call attention to these issues.
About half of respondents say the concerns have persisted for a year or more. And
a significant number of those who have witnessed these persistent problems report
injurious consequences. For example, one in five physicians say they have seen harm
come to patients as a result of these concerns, and 23 percent of nurses say they are
considering leaving their units because of these concerns.
On the positive side, this study shows that healthcare workers who are confident in their
ability to raise these crucial concerns observe better patient outcomes, work harder, are
more satisfied, and are more committed to staying. About 10 percent of the healthcare
workers surveyed fall into this category. While additional confirming research is needed,
the implication is that if more healthcare workers could learn to do what this influential
10 percent seem to be able to do systematically, the result would be significantly fewer
errors, higher productivity, and lower turnover.
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
Participants were asked to indicate the percentage of their coworkers with whom they
had each of the seven crucial concerns. They had to indicate that at least 10 percent
of their coworkers were in violation to be categorized as “seeing the concern.” It is
important to point out that respondents consistently report that the vast majority
of healthcare workers do not exhibit the problems described below. And yet the vast
majority of healthcare workers do see some number who not only exhibit the problems,
but also continue to do so for long periods of time without being held accountable.
1.
Broken Rules: 84 percent of physicians and 62 percent of nurses and other clinicalcare providers see some number of their coworkers taking shortcuts that could be
dangerous to patients. This concern was focused on a relatively small number of
their colleagues. The median was 10 percent, meaning that they were comfortable
with 90 percent of their colleagues’ ability.
“A phlebotomist in a neonatal unit would slip on her gloves and immediately
tear the tip of the index finger off her glove, so she could feel the baby’s vein
better and wouldn’t miss. I talked to her about it twice. Finally I said, ‘If I
ever see you tear the finger out of another glove I will write you up for a willful
violation.’ Now she follows the rules.” (Nurse Manager)
2.
Mistakes: 92 percent of physicians and 65 percent of nurses and other clinical-care
providers work with some people who have trouble following directions; 88 percent
of physicians and 48 percent of nurses and other clinical-care providers see some
colleagues show poor clinical judgment when making assessments, doing triage,
diagnosing, suggesting treatment, or getting help. Again, these respondents are
pointing to a relatively small minority of their colleagues—the median was again just
10 percent.
“Some docs can make incorrect orders. We let it slide—especially if it is a
jerk . . . For example, one physician prescribed a drug that you should give three
times a day, but he said to give it twice a day. I let it go, because it was just a
pain pill. It wasn’t going to make the child any sicker.” (Pharmacist.)
3.
Lack of Support: 53 percent of nurses and other clinical-care providers report that
10 percent or more of their colleagues are reluctant to help, impatient, or refuse to
answer their questions. 83 percent have a teammate who complains when asked to
pitch in and help. On the positive side, 76 percent say that half or more of their
colleagues give them emotional support when they are down, and 64 percent say that
half or more of their colleagues pick up a share of their work when they need help.
It’s clear that most people provide support. The problem is with a small minority
who don’t.
“Some people here are burnt out. They’ve lost the excitement or have some
personal issue in their life . . . People have to cover for them, pick up their slack.
People get mad at them, isolate them, don’t offer to help them, shy away from
them. If they need extra help, they don’t get it. They don’t call or ask for it.”
(Nurse)
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
4.
Incompetence: 81 percent of physicians and 53 percent of nurses and other clinicalcare providers have concerns about the competency of some nurse or other clinicalcare provider they work with; 68 percent of physicians and 34 percent of nurses and
other clinical-care providers have concerns about the competency of at least one
physician they work with.
“There is a cardiologist who everybody feels is incompetent. He makes himself
very accessible to general practitioners, so he gets a lot of referrals, but those of us
who have to work with him—the thoracic surgeons, the anesthesiologists, the other
cardiologists—would never put someone under his care.” (Physician)
5.
Poor Teamwork: 88 percent of nurses and other clinical-care providers have one or
more teammate who gossips or is part of a clique that divides the team. 55 percent
have a teammate who tries to look good at others’ expense.
“We have a nurse who is like your eccentric aunt—she’s a bully. She makes
unreasonable demands like, ‘I won’t take any more patients today.’ She gets
away with it. She’s a very good nurse, but she’s ornery and a bully. She doesn’t
do her fair share. It ticks all of us off. We’ve lost a couple of good nurses here
because they were sick of putting up with her and our supervisor won’t deal with
her.” (Nurse)
6.
Disrespect: 77 percent of nurses and other clinical-care providers work with some
who are condescending, insulting, or rude. 33 percent work with a few who are
verbally abusive—yell, shout, swear, or name call.
“A group of physicians went right into the patient’s room without gowns or
masks or gloves. This was a patient who was supposed to be in isolation. We
didn’t confront them because that cardio surgeon has a reputation. He belittles
nurses by saying things like, ‘Do they have any nurses on this unit who aren’t
stupid?’ If you question him, he starts yelling, and turns it into a war.” (Nurse)
7.
Micromanagement: 52 percent of nurses and other clinical-care providers work with
some number of people who abuse their authority—pull rank, bully, threaten, or
force their point of view on them.
“We have a charge nurse who . . . pages us to come to the desk so she can tell
us what to do . . . She will come into the room where we have a sick patient
and she’ll take over . . . She’ll say, ‘Do it because I say so.’ Sometimes when she
bosses me around I feel less inclined to correct her when she’s wrong about how to
treat the child. I’m sure I’ve gone along with something I shouldn’t have because
I resent her. But basically, I’ve started looking at other hospitals for a job.”
(Nurse)
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
Most healthcare respondents are happy in their careers and believe their organizations
do good work. And yet most respondents report that a number of their colleagues create
problems that are common, frequent, persistent, and dangerous. And, most important,
the data show these problems are rarely addressed.
The study focused in detail on three of the seven crucial conversations: incompetence,
poor teamwork, and disrespect. In these three areas the study mapped the frequency,
duration, and impacts of people’s concerns. It also measured whether and how these
concerns were addressed.
The survey asked about a variety of competency issues, ranging from “poor clinical
judgment” to “making decisions beyond their competency level” to “missing basic
skills.” Respondents indicated whether they had coworkers who are incompetent in these
areas. Next, respondents were asked to think of the coworker with the worst competency
problem, and to rate how often this person does something dangerous, how long the
problem has gone on, and how the person’s competency has impacted patient health and
safety.
The data in tables 1-a and 1-b reveal the scope of the problem. Most healthcare workers
have serious concerns about the competence of some of their coworkers. In fairness,
a person’s perceptions of another’s competence can sometimes be just a difference
of judgment—and nowhere more than in a field as complex and often ambiguous as
healthcare. And yet the prevalence of the perceptions, along with strong anecdotal data
from focus group interviews, suggest that real problems exist. Many cite a coworker who
does something dangerous as often as every month. Nearly half report the problem has
continued for a year or more. Some have witnessed the person causing harm to patients.
And yet only a small percentage discuss their concerns with the person.
The data show it is much tougher to confront a physician than to confront a nurse or other
clinical-care provider. Interestingly, the data also show physicians are about as unlikely to
confront nurses and other clinical-care providers as they are to confront physicians, even
though their clinical authority would seem to make it an easier discussion.
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
Nurses and Other Clinical Care Providers’ Concerns about Incompetence
53% are concerned about
a peer’s competence.
This peer does something dangerous at least
once a month.
27%
12% have spoken with this
peer and shared their full
concerns.
The problem with this peer has gone on for a
year or more.
48%
A patient has been harmed by this person’s
actions during the last year.
7%
34% are concerned about
a physician’s competence.
This physician does something dangerous at
least once a month.
19%
Less than 1% have spoken
with this physician and
shared their full concerns.
The problem with this physician has gone on
for a year or more.
54%
A patient has been harmed by this physician’s
actions during the last year.
8%
Table 1-a
Physicians’ Concerns about Incompetence
This person does something dangerous at least
once a month.
15%
The problem with this person has gone on for
a year or more.
46%
8% have spoken with this
person and shared their
full concerns.
A patient has been harmed by this person’s
actions during the last year.
9%
68% are concerned about
a physician’s competence.
This physician does something dangerous at
least once a month.
21%
Less than 1% have spoken
with this physician and
shared their full concerns.
The problem with this physician has gone on
for a year or more.
66%
A patient has been harmed by this physician’s
actions during the last year.
19%
81% are concerned
about a nurse’s or other
clinical-care provider’s
competence.
Table 1-b
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
The survey examined a variety of teamwork concerns, ranging from “gossiping” to
“making themselves look good at your expense” to “not doing their fair share of the
work.” Respondents indicated whether they had coworkers who demonstrated poor
teamwork in these areas. Next, respondents were asked to think of the coworker whose
poor teamwork has the most negative impact, and to rate how often this person does
something that undermines the team, how long the problem has gone on, and how the
person’s poor teamwork has impacted patient care and employee morale.
The data in Table 2 show a widespread problem. Three-quarters of the healthcare workers
surveyed are concerned about a teamwork issue, and more than two-thirds say this
problem has gone on for over a year. A smaller—yet significant—number (one-fifth) say
the teamwork issue is so severe they can’t trust that patients are getting the right level of
care, and even more are seriously considering leaving their jobs because of the teamwork
issue. And yet relatively few ever discuss their concerns with the person involved.
Nurses and Other Clinical Care Providers’ Concerns about Poor Teamwork
75% are concerned about
a peer’s poor teamwork.
16% have spoken with
this peer and shared their
full concerns.
This peer does something that undercuts the
team at least once a month.
61%
The problem with this peer has gone on for a
year or more.
69%
Because of this teamwork issue, the respondent
can’t trust that patients in their area are receiving
the right level of care.
22%
Because of this teamwork issue, the respondent
is seriously considering leaving the unit or the
hospital.
23%
Table 2
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
The survey asked about disrespectful and abusive behavior, ranging from “verbal abuse”
to “condescending, insulting, or rude” to “bullying and threatening.” Respondents
indicated whether they worked with people who were abusive toward them in these ways.
Next, respondents were asked to think of the person whose abuse has the most negative
impact, and to rate how often this person is disrespectful or abusive toward them, and
how long the problem has gone on.
The data in Table 3 show that three-quarters of the healthcare workers surveyed
experience some level of disrespect. For many, the treatment is frequent and longstanding. The correlations show that the more frequent the behavior and the longer it
has gone on, the greater the workers’ intent to quit their jobs. In fact, these correlations
are so strong (correlations where r>.1 are meaningful—here we find r=.424, which is
impressive) that disrespectful behavior is suggested to be a primary cause of people’s
desire to quit. Discussing their concerns with the person who is responsible for the abuse
is almost out of the question.
Nurses and Other Clinical Care Providers’ Concerns about Disrespect and Abuse
77% are concerned about
disrespect they experience.
7% have spoken with this
peer and shared their full
concerns.
This person is disrespectful or abusive
toward them in at least a quarter of their
interactions.
28%
The behavior has gone on for a year or
more.
44%
Correlation between the frequency of
mistreatment and intent to quit their job.
r = .424, p <
.001
Correlation between the duration of abuse
and intent to quit their job.
r = .190, p <
.001
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
Why Don’t People Speak Up and Share Their Full Concerns?
The obvious reason is that confronting people is difficult. In fact, most respondents
to the survey indicated it was between difficult and impossible to confront people in
these crucial situations. People’s lack of ability, belief that it is “not their job,” and
low confidence that it will do any good to have the conversation are the three primary
obstacles to direct communication.
When the Concern Is…
Percentage Saying It Is Difficult to Impossible to
Confront the Person
Incompetence
56% of Physicians
72% of Nurses and other Clinical-Care Providers
Poor Teamwork
78% of Nurses and other Clinical-Care Providers
Disrespect or Abuse
59% of Nurses and other Clinical-Care Providers
Table 4
Other obstacles include time and fear of retaliation. The survey asked respondents
to indicate the reasons they didn’t confront people when they had these important
concerns. The reasons they selected were similar for each concern, with the most
common reasons being: “There wasn’t a time or opportunity,” “It’s not my role,” “I’ve
seen them get angry,” and “I thought they would retaliate.” People don’t want to make
others angry or undercut their working relationships, so they leave difficult discussions
to others or to another time, and never get back to the person.
However, some people don’t remain silent about the problems they see. They talk about
them with others. Depending on the nature of the problem, a quarter to half of the
respondents discussed the problem with coworkers or with the person’s manager. In
interviews, participants suggested that the purpose for discussing these problems with
coworkers is not to solve problems. Instead, it’s to work around them, warn others about
them, and blow off steam. The comments below, taken from focus groups, illustrate
these workarounds, warnings, and venting sessions.
“We all know who I’m talking about. She has bad habits, or is missing good
ones. She gets busy and leaves the rails down on an infant bed or the door open
on an incubator. We all check on her patients just to make sure about things.”
(Nurse)
“People give you the word. A nurse will call from surgery and say, ‘He’s in a
mood.’ If something goes wrong in surgery, he’ll come in yelling at people. You
are just waiting for your turn.” (Nurse)
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
“She can’t be trusted with cases. She can give meds, but she won’t ever get it.
She’s been there for seven months. This nurse would do fine in a doctor’s office,
but won’t make it in the hospital. The other nurses all agree.” (Nurse)
Most respondents also say going to the person’s manager creates problems.
“I’m embarrassed. I saw a nurse cutting corners and instead of talking to
her I talked to her boss. Here’s the situation. I used to be this nurse’s boss, but
now she’s training me and we are peers. I should have gone to her, but I was
concerned about our relationship and I went to her boss instead. It was a bad
move.” (Nurse Practitioner)
The data suggest that going to the person’s manager is, indeed, a bad move. Although
managers are somewhat more likely than employees to confront the person and fully
discuss the problem, they are still very unlikely to do so. Taking a concern to a manager
was often a dead end.
Percentage of NonSupervisory Employees Who
Confront the Person
Competence of a Nurse 3%
or other Clinical-Care
Provider
Less than 1%
Competence of a
Physician
5%
Poor Teamwork
When the Concern Is
Disrespect or Abuse
2%
Percentage of
Supervisors Who
Confront the Person
16%
Less than 1%
9%
5%
Table 5
People Who Do Step Up to these Crucial Conversations
Within each hospital there is a fascinating minority, 5–15 percent of healthcare workers,
depending on the issue, who step up to these crucial conversations. They work in the
same units or departments as the 85–95 percent of their coworkers who don’t feel able
to speak up. Are they crazy? Are they destroyed by the unsafe environment? No; these
people prove that it’s possible to discuss serious concerns in almost any environment
and succeed.
© 2005 VitalSmarts, L.C. All Rights Reserved. VitalSmarts is a trademark and Crucial Conversations is a registered trademark of VitalSmarts, L.C.
The significant correlations in Table 6 show that people who are confident in their
ability to have crucial conversations achieve positive outcomes for their patients, for the
hospital, and for themselves. This is counterintuitive. Most of those who don’t speak up
believe that to do so would lead to disaster. The opposite seems to be the case for this
critical minority of interpersonally skilled individuals. Again, the correlations of from .2
up to .465 suggest that these peoples’ ability to deal with tough interpersonal challenges
is highly related to all of the outcomes described. The “p 3 years
Strabismus surgery
History of POV/relative with PONV
Consider
Patient preferences
Fear of PONV
Frequency of
PONV causing
headaches/migraine
Costeffectiveness
Reducing baseline risks
Avoidance/minimization of:
Nitrous oxide
Volatile anesthetics
Post-op opioids
Patient risk
Medium
Low
Wait and See
Pick 1 or 2 Interventions
Dexamethasone
Propofol
Anesthesia
Regional
Anesthesia
Droperidol†
Haloperidol
5-HT3
antagonist
Non
Pharmacological:
Acupuncture
High
>2 Interventions/
Multimodal Approach
Portfolio of
prophylaxis
and treatment
strategies
Scopolamine
Perphenazine
Dimenhydrinate
Treatment Options
• If prophylaxis fails or was not received: use
antiemetic from different class than prophylactic
drug
• Readminister only if >6 hours after PACU;
• Do not readminister dexamethasone or
scopolamine
NK-1 receptor
antagonists
Propofol subhypnotic
dose infusion or
Propofol in PACU
(rescue only)
† Use droperidol in
children only if other
therapy has failed and
patient is being
admitted to hospital;
Haloperidol for adults
only
Figure 4. Algorithm for management of postoperative nausea and vomiting. PONV = postoperative nausea and vomiting.
(5-HT3) receptor antagonists (ondansetron, dolasetron,
granisetron, tropisetron, ramosetron, and palonosetron),
neurokinin-1 (NK-1) receptor antagonists (aprepitant,
casopitant, and rolapitant), corticosteroids (dexamethasone and methylprednisolone), butyrophenones (droperidol and haloperidol), antihistamines (dimenhydrinate and
meclizine), and anticholinergics (transdermal scopolamine
[TDS]). While PONV prevention is recommended in a subset of patients, current evidence does not support giving
January 2014
Volume 118
Number 1
prophylactic antiemetics to all patients who undergo surgical procedures. However, with more inexpensive generics
becoming available, properly conducted cost-effectiveness
(C/E) studies need to be done to support the more universal use of prophylactic antiemetics. Ondansetron 4 mg,
droperidol 1.25 mg, and dexamethasone 4 mg were equally
effective, and each independently reduced PONV risk by
approximately 25%.47 The recommended doses and timing
of these drugs are listed in Table 3. Recommendations given
www.anesthesia-analgesia.org
91
E SPECIAL ARTICLE
are evidence based, and not all the drugs have a Food and
Drug Administration (FDA) indication for PONV.
vomiting and decrease nausea for patients receiving fentanyl patient-controlled analgesia (PCA).102
5-HT3 Receptor Antagonists
Palonosetron
Palonosetron is a second generation 5HT3 receptor antagonist with a half-life of 40 hours.103,104 The most effective dose
is 0.075 mg IV approved for 24 hours.105,106 Palonosetron
0.075 mg is more effective than granisetron 1 mg96 and
ondansetron 4 mg82 in preventing PONV.
Ondansetron
Most of the available research on the 5-HT3 receptor antagonists involves ondansetron, which has greater antivomiting
than antinausea effects. Ondansetron is the “gold standard”
compared with other antiemetics. It has a recommended
dose of 4 mg, a NNT of approximately 6 for prevention
of vomiting (0–24 hours), and a NNT of approximately 7
for prevention of nausea.74 The effect of the ondansetron
8 mg oral disintegrating table is equivalent to the 4 mg
IV dose.75,76 Ondansetron is as effective as other 5-HT3s74
including ramosetron 0.3 mg.77 It is also as effective as dexamethasone47 and haloperidol 1 mg IV,78–80 with no difference
in effect on the QTc interval.81 However, it is less effective
than aprepitant81 for reducing emesis and palonosetron for
the incidence of PONV.82
Dolasetron
Prospective RCTs show a prophylactic dose of 12.5 mg
dolasetron effectively prevents PONV.83–85 That prophylactic dose is as effective as ondansetron 4 mg.86,87 Other data
show dolasetron is more effective than droperidol in preventing PONV after surgery for prognathism.88 A study
by Janicki et al.89 found granisetron is more effective in
preventing PONV than dolasetron. These differences may
be due to duplication of the CYP2D6 allele causing ultrarapid metabolism of dolasetron. In December 2010, the FDA
announced that IV dolasetron should no longer be used
for chemotherapy-induced nausea and vomiting in adults
and children because of concerns of QT prolongation and
torsade de pointes.90 At present, dolasetron is no longer
marketed in the United States but may be available in other
countries.
Granisetron
Granisetron, 0.35 to 3 mg IV (5–20 mcg/kg), is as effective as other first generation 5HT3 receptor antagonists.91–93
Granisetron, 3 mg IV, is also as effective as dexamethasone
8 mg, and the combination is better than either drug alone.94
Similarly, granisetron 1 mg plus cyclizine 40 mg is more
effective than granisetron 1 mg or cyclizine 50 mg alone.95
However, compared with palonosetron 0.075 mg, granisetron 2.5 mg is as effective at 3 hours and 3 to 24 hours but
less effective at 24 to 48 hours.96
Tropisetron
Tropisetron 2 mg IV is effective for PONV prophylaxis.97 It
is as effective as ondansetron, granisetron,98 and droperidol
and more effective than metoclopramide.99 The combination of tropisetron plus dexamethasone is more effective
than either drug alone.100 Tropisetron is not approved in the
United States.
Ramosetron
Ramosetron is not approved in the United States but available in other parts of the world. It is more effective with
IV versus PO dosing (1–24 hours postoperatively).101
Ramosetron 0.3 mg IV is the most effective dose to prevent
92
www.anesthesia-analgesia.org
Timing of Administration
Ondansetron, dolasetron, granisetron, and tropisetron are
most effective in the prophylaxis of PONV when given
at the end of surgery,85,107–110 although some data on dolasetron suggest timing may have little effect on efficacy.111
Palonosetron is typically given at the start of surgery.105,106
Adverse Events
The 5-HT3 receptor antagonists have a favorable side effect
profile, and while generally considered equally safe, all
except palonosetron affect the QTc interval. In June 2012, the
U.S. FDA recommended the dose of ondansetron for chemotherapy-induced nausea and vomiting should not exceed 16
mg in a single dose because of risks of QT prolongation. In
December 2012, the FDA notified that the 32 mg single IV
dose will no longer be marketed.112 However, there was no
change in the recommended dose of ondansetron 4 mg to
prevent PONV.90 The number-needed-to-harm (NNH) with
a single dose of ondansetron is 36 for headache, 31 for elevated liver enzymes, and 23 for constipation.54
NK-1 Receptor Antagonists
Aprepitant
Aprepitant is an NK-1 receptor antagonist with a 40-hour
half-life. In 2 large RCTs, aprepitant (40 and 80 mg per os)
was similar to ondansetron in achieving complete response
(no vomiting and no use of rescue antiemetic) for 24 hours
after surgery. However, aprepitant was significantly more
effective than ondansetron for preventing vomiting at 24
and 48 hours after surgery and in reducing nausea severity
in the first 48 hours after surgery.81,113 It also has a greater
antiemetic effect compared with ondansetron. When used in
combination, aprepitant 40 mg per os, plus dexamethasone,
is more effective than ondansetron plus dexamethasone in
preventing POV in patients undergoing craniotomy.114 A
dose-ranging study for gynecologic laparotomy patients
found a 80 mg per os dose of aprepitant is the most appropriate dose and is more effective than a 40 mg dose.115 The
clinical experience with the use of aprepitant is still limited,
and its role in routine prophylaxis is not established.116
Casopitant
A Phase 3 study of casopitant shows the combination of
casopitant, 50 to 150 mg per os, plus ondansetron 4 mg, is
more effective than ondansetron alone.117,118 Casopitant has
not been approved for use.
Rolapitant
Rolapitant has a 180-hour half-life and better PONV prophylaxis than placebo. A clinical trial by Gan et al.119 showed
no difference between groups receiving oral rolapitant and
ANESTHESIA & ANALGESIA
Consensus Guidelines for the Management of PONV
ondansetron 4 mg IV at 24 hours, but more patients experienced no emesis with the rolapitant 70 and 200 mg doses
at 72 and 120 hours, respectively. Rolapitant has not been
approved for use.
Corticosteroids
Dexamethasone
The corticosteroid dexamethasone effectively prevents nausea and vomiting in postoperative patients.120,121 A prophylactic dose of 4 to 5 mg IV for patients at increased risk for
PONV is recommended after anesthesia induction rather
than at the end of surgery.121 For PONV prophylaxis, the
efficacy of dexamethasone 4 mg IV is similar to ondansetron
4 mg IV and droperidol 1.25 mg IV.47 More recent studies
increasingly use the higher dose of dexamethasone 8 mg IV
rather than the minimum effective dose of 4 to 5 mg.122–126
Preoperative dexamethasone 8 mg enhances the postdischarge quality of recovery in addition to reducing nausea,
pain, and fatigue.127 Dexamethasone also has dose-dependent effects on quality of recovery. At 24 hours, patients
receiving dexamethasone 0.1 vs 0.05 mg/kg required less
opioid and reported less nausea, sore throat, muscle pain,
and difficulty falling asleep.128 A meta-analysis evaluating
the dose-dependent analgesic effects of perioperative dexamethasone found that doses >0.1 mg/kg are an effective
adjunct in multimodal strategies to reduce postoperative
pain and opioid consumption.129,130 With these additional
benefits of pain relief and better quality of recovery, a prophylactic dose of dexamethasone 0.1 mg/kg or 8 mg in
adults may be considered though further confirmation is
needed for this larger dose.
Data on safety of perioperative dexamethasone are
inconclusive. In most studies, a single dose of perioperative dexamethasone does not appear to increase the risk
of wound infection.120,129 However, a recent study reported
that intraoperative dexamethasone 4 to 8 mg may confer an
increased risk of postoperative infection.131 Weighing the
risk-benefit ratio, a recent editorial suggests a single dose of
dexamethasone 4 to 8 mg is safe when used for PONV prophylaxis.132 In addition, recent studies showed significant
increases in blood glucose that occur 6 to 12 hours postoperatively in normal subjects,133,134 those with impaired glucose tolerance,134 and type 2 diabetic135 and obese134 surgical
patients who receive dexamethasone 8 mg. In view of this
evidence, use of dexamethasone in labile diabetic patients is
relatively contraindicated.
Methylprednisolone
Methylprednisolone 40 mg IV is effective for the prevention
of late PONV.136,137 There is no evidence to suggest that the
adverse effect of methylprednisolone is any different from
dexamethasone.
Butyrophenones
Droperidol
Prophylactic doses of droperidol 0.625 to 1.25 mg IV are
effective for the prevention of PONV.138–140 The efficacy of
droperidol is similar to ondansetron for PONV prophylaxis,
with an NNT of approximately 5 for prevention of nausea
and vomiting (0–24 hours).140 Droperidol is most effective
when administered at the end of surgery.140 For PONV
January 2014
Volume 118
Number 1
prevention, droperidol is superior to metoclopramide doses
of
Purchase answer to see full
attachment