REVIEW
Postoperative Complications in the Seriously Mentally Ill
A Systematic Review of the Literature
Laurel A. Copeland, PhD,* John E. Zeber, PhD,* Mary Jo Pugh, PhD,*
Eric M. Mortensen, MD, MSc,* Marcos I. Restrepo, MD, MSc,*
and Valerie A. Lawrence, MD, MSc*
Objective: To determine the knowledge base on clinical outcomes
of surgery among persons diagnosed with serious mental illness.
Background: Despite a burgeoning literature during the last 20
years regarding perioperative risk management, little is known about
intraoperative and postoperative complications among patients with
schizophrenia and other serious mental illnesses.
Methods: A systematic literature search of Medline (1966 August
2007) and review of studies was conducted. Eligible studies were of
any design with at least 10 patients diagnosed with serious mental
illness, reporting perioperative medical, surgical, or psychiatric
complications.
Results: The search identified 1367 potentially relevant publica-
tions; only 12 met eligibility criteria. Of 10 studies of patients with
schizophrenia, 9 had fewer than 100 patients, whereas one large
retrospective study reported higher rates of postoperative complica-
tions among 466 schizophrenia patients compared with 338,257
controls. These studies suggest that patients with schizophrenia,
compared with those without mental illness, may have higher pain
thresholds, higher rates of death and postoperative complications,
and differential outcomes (eg, confusion, ileus) by anesthetic tech-
nique. Two studies evaluated outcomes in patients with major
depressive disorder and found higher rates of postoperative delirium
and postoperative confusion. Both schizophrenia and depression
patients experienced more postoperative confusion or delirium when
From the *Veterans Affairs HSRD: South Texas Veterans Health Care
System (VERDICT), San Antonio, TX, University of Texas Health
Science Center, Department of Psychiatry, San Antonio, TX, and Uni-
versity of Texas Health Science Center, Department of Medicine, San
Antonio, TX.
Supported by the Department of Veterans Affairs, Veterans Health Admin-
istration, Health Services Research and Development Service, and by the
VERDICT field program at the South Texas Veterans Health Care
System, San Antonio, TX. Funded by a career development award
MRP-05-145 from the VA Health Services Research and Development
program (to L.A.C.) and by a grant from VHA VISN 17 to study
perioperative outcomes and safety in the seriously mentally ill elderly (to
J.E.Z.).
The views expressed in this article are those of the authors and do not
necessarily represent the views of the Department of Veterans Affairs.
Reprints: Laurel A. Copeland, PhD, South Texas Veterans Health Care
System-VERDICT Research, 7400 Merton Minter (11c6), San Antonio,
TX 78229-4404. E-mail: copelandl@uthscsa.edu.
Copyright 2008 by Lippincott Williams & Wilkins
ISSN: 0003-4932/08/24801-0031
DOI: 10.1097/SLA.0b013e3181724f25
Annals of Surgery Volume 248, Number 1, July 2008
psychiatric medications were discontinued preoperatively. We iden-
tified no studies of perioperative outcomes in patients with bipolar or
posttraumatic stress disorder.
Conclusions: There are few studies of perioperative outcomes in
patients with serious mental illness. Future research should assess
surgical risks among patients with serious psychiatric conditions
using rigorous methods and well-defined clinical outcomes.
(Ann Surg 2008;248: 3138)
P atients with schizophrenia and other serious psychiatric
conditions experience a significantly increased risk of
death from cardiovascular, gastrointestinal, and respiratory
illnesses.1 4 Yet, despite high rates of surgery in this patient
population5 and a burgeoning literature regarding periopera-
tive risk management in other patient populations, little
seems to be known regarding postoperative complications
and perioperative care of these at-risk patients.4 Clinical and
physiological data suggest that patients with serious mental
illness may be quite dissimilar, and may fare worse, com-
pared with surgical patients without serious mental illness.6,7
Perioperative care may be complicated by systemic effects of
psychotic disorders such as autonomic nerve dysfunction,
impaired thermoregulation, potential interactions between an-
tipsychotic medications, and anesthetic and analgesic agents,
and uncertainties about optimal anesthetic and analgesic reg-
imens.4 Individuals with schizophrenia have higher pain
thresholds compared with patients without mental illness,
more cognitive deficits, disorganized thinking, and other
functional difficulties. Thus they may have impaired ability to
recognize and communicate potentially important medical
symptoms and so may present at a later stage of disease,
resulting in higher risk treatment approaches.79 To improve
understanding of the challenges presented to surgeons by
patients with previously diagnosed mental illness, we con-
ducted a systematic literature review of the evidence base on
postoperative complications for patients with the serious
mental illnesses of schizophrenia, bipolar disorder, major
depressive disorder (MDD), and posttraumatic stress disorder
(PTSD).
31
Copeland et al
FIGURE 1. Flowchart of included articles on
perioperative risk management in patients with
serious mental illness.
METHODS
We conducted a comprehensive literature search of
Medline from 1966 through August 15, 2007, using the
following criteria:
1. MeSH terms: schizophrenia; depressive disorder, major; de-
pression; bipolar disorder; stress disorders, posttraumatic;
affective disorders; psychotic; or mood disorders excluding
cyclothymic disorder, postpartum depression, seasonal affec-
tive disorder; and surgical procedures, operative.
2. Text word terms: preoperative care or intraoperative care
or postoperative care or intraoperative complications or
postoperative complications.
3. Linkage strategy: 1 and 2, limited to publications in
English.
We also searched the bibliographies of relevant review
articles for additional eligible publications. We required stud-
ies report primary data. Additional exclusion criteria were as
follows: (1) studies of psychiatric conditions developing after
or as a result of an operation (eg, PTSD after abortion, major
depression after cardiac, or transplant operations); (2) studies
with unclear preoperative diagnosis of a serious mental ill-
ness; (3) studies reporting only physiologic outcomes (eg,
interleukin levels, cortisol, epinephrine, intraoperative car-
diac regulation) without clinical outcomes (eg, 30-day mor-
tality, postoperative myocardial infarction, pneumonia, delir-
ium); (4) case series with fewer than 10 patients; (5) studies
reporting only long-term outcomes or health services usage;
and (6) studies in pediatric populations, stereotactic surgery,
or leucotomy for psychiatric conditions. In reporting results,
we have calculated group sizes and P values (Fisher exact
test, 2-tailed) where the authors did not report them. Although
some studies report nonclinical outcomes such as physiologic
parameters or health services use, our synthesis is limited to
the clinical outcomes.
32
Annals of Surgery Volume 248, Number 1, July 2008
RESULTS
Search Results
Overall, the search identified 1367 potentially relevant
publications, 337 regarding schizophrenia, and 1030 regard-
ing other serious mental illness of MDD, bipolar disorder, or
PTSD (Fig. 1). After review of title and abstracts, 44 publi-
cations (schizophrenia) and 34 publications (other serious
mental illness) were potentially eligible and reviewed in full.
Of these, 10 met eligibility criteria for schizophrenia,6,7,10 17
and 2 met eligibility criteria for other serious mental ill-
ness.18,19 We found no eligible publications regarding peri-
operative complications in patients with bipolar disorder or
PTSD. Abstracted studies are summarized below, organized
by study design within type of serious mental illness.
Perioperative Outcomes in Patients with
Schizophrenia
Case Series (2)
One early study described a series of abdominal surgery
patients with schizophrenia.10 Eight patients had emergency
operations. Eleven of 12 patients died within 11 days of
surgery, a 93% mortality rate that contrasted with the overall
3.1% mortality rate among 968 patients without schizophre-
nia observed at the same facility (P 0.001). Causes of death
were reported as paralytic ileus with progression to circula-
tory collapse, unexpected cardiac arrest, and respiratory
depression or aspiration. The authors speculated that the high
mortality rate may have been due to late presentation of surgical
disease and to adverse effects of chronic use of phenothiazines
(eg, arrhythmias, inhibition of intestinal peristalsis, or depression
of hypothalamus-adrenal function) (Table 1).
A second case series report described 14 surgery pa-
tients with schizophrenia on chronic antipsychotic medica-
tions who received ketamine with droperidol and fentanyl for
2008 Lippincott Williams & Wilkins
Annals of Surgery Volume 248, Number 1, July 2008 Postoperative Complications in the Seriously Mentally Ill

TABLE 1. Literature Reviewed on Perioperative Outcomes Among Seriously Mentally Ill Patients
Study Design, Follow-up Serious
Study Period, Question Mental Illness No. Patients Results Conclusion
Matsuki et al10 Case series; abdominal Schizophrenia 12 Death (P 0.001): In this small case series,
surgery patients postoperative mortality
was much greater in
11 d Schizophrenia 11/12
patients with,
Mortality Controls est. 30/968 compared to those
without, schizophrenia
possibly due to late
presentation and
conventional
antipsychotic agents.
Ishihara et al11 Case series; various Schizophrenia 14 Death: In this small case series,
surgeries postoperative mortality
30 d Schizophrenia 2/14 was high but ketamine
appeared safe.
Mortality and safety of Psychosis worse post-op 0/14
ketamine
Kudoh et al12 Cohort; orthopedic surgery Schizophrenia 50 with VAS Pain Scores (mean SD) Among schizophrenia
patients schizophrenia, 2 h post-op (P 0.05): patients, there was no
25 controls advantage to either
Schizophrenia, pentazocine 4.0 1.7
analgesic regimen.
Schizophrenia, pentazocine 3.9 1.6 Compared with
haloperidol controls, patients with
Control, pentazocine 5.0 1.6 schizophrenia had less
5 h post-op (P 0.05): perceived pain and less
3d Schizophrenia, pentazocine 3.8 1.5 analgesic consumption.
Pain perception, analgesic Schizophrenia, pentazocine 3.7 1.5
consumption, haloperidol
pentazocine versus Control, pentazocine 5.1 1.9
pentazocine Total pentazocine (mg, mean SD)
haloperidol in patients through POD 3 (P 0.05):
with schizophrenia
Schizophrenia, pentazocine 15.6 24.7
Schizophrenia, pentazocine 21.6 26.7
haloperidol
Control, pentazocine 38.4 30.6
Kudoh et al14 Cohort; orthopedic surgery Schizophrenia 50 with Postoperative delirium & confusion Schizophrenia patients
patients schizophrenia, (P 0.01): had a higher rate of
3d 35 controls Schizophrenia 14/50 postoperative
confusion compared
Incidence of postoperative Controls 2/35
with controls.
delirium or confusion
Daumit et al6 Cohort Schizophrenia 466 with OR (95% CI) for schizophrenia: Schizophrenia was
schizophrenia, Respiratory failure: associated with
338,257 increased rates of
Incidence 2.1 (1.43.1)
controls respiratory failure,
Length of stay ICU admission 9.9 (3.925.5) DVT/PE, and sepsis.
Postoperative Death 8.9 (3.026.0) Patients with
complications, mortality DVT or PE: schizophrenia and
Incidence 2.0 (1.23.3) respiratory failure had
higher rates of ICU
ICU admission 1.0 (0.24.2) admission and death
Death 2.1 (0.49.6) than controls with the
Sepsis: same complications.
Incidence 2.3 (1.53.5) Patients with
schizophrenia and
ICU admission 3.0 (0.99.2)
sepsis had higher rates
Death 7.1 (1.632.5) of death than control
Hemorrhage or hematoma: patients with sepsis.
Incidence 1.1 (0.71.8)
ICU Admission 0.5 (0.11.9)
Death 4.6 (0.826.9)
(Continued)

2008 Lippincott Williams & Wilkins 33

Copeland et al Annals of Surgery Volume 248, Number 1, July 2008

TABLE 1. (Continued)
Study Design, Follow-up Serious
Study Period, Question Mental Illness No. Patients Results Conclusion
Cooke et al7 Cohort; appendectomy Schizophrenia 55 Incision left open 24 (44%) Schizophrenia patients
patients Perforated appendix 36 (66%) seemed to present at
late stage of disease
Length of stay Gangrenous appendix 9 (16%)
and to have high rates
Outcomes of acute No perforation/gangrene 10 (18%) of complications,
appendicitis Complications 31 (56%) death, and disruptive
Disruptive behavior 24 (44%) behavior
Death 2 (4%)
Kudoh et al13 RCT; abdominal surgery Schizophrenia 46 VAS Pain Score (mean SD): Among patients with
patients 8 h post-op (P 0.05): schizophrenia, there
was no advantage to
Buprenorphine 36 12.8
either analgesic
Epidural bupivacaine 25.4 13.2 regimen. Compared
5d 24 h post-op (P 0.05): with controls, patients
Postoperative Buprenorphine 31.7 10.7 with schizophrenia had
buprenorphine versus Epidural bupivacaine 20.5 9.4 less perceived pain and
epidural analgesia less analgesic
Hours to passage of flatus (P 0.05): consumption.
Buprenorphine 105.3 33
Epidural bupivacaine 86.5 30.1
Hours to passage of feces (P 0.05):
Buprenorphine 114.5 28.2
Epidural bupivacaine 97.2 29.4
Kudoh et al15 RCT; orthopedic surgery Schizophrenia 76 Psychosis or confusion through POD 2 Intravenous anesthesia
patients (P 0.05): with ketamine plus
Sevoflurane NO2 20/37 propofol was
fentanyl associated with less
Ketamine propofol 11/37 early (but not late)
fentanyl postoperative psychosis
or confusion than
Psychosis or confusion through POD 5 inhaled anesthesia with
(P 0.24): sevoflurane and nitrous
10 d Sevoflurane NO2 22/37 oxide.
fentanyl
Postoperative confusion Ketamine propofol 17/37
and regimen for fentanyl
maintenance of
anesthesia
Kudoh, Takase RCT; abdominal surgery Schizophrenia 70 with Postoperative confusion or psychosis Among patients with
et al.16 patients schizophrenia through POD 2 (P 0.01 for combined schizophrenia, rates of
(randomized), schizophrenia groups vs. controls): confusion or psychosis
35 controls Schizophrenia, epidural 7/33 did not differ by type
(not bupivacaine of analgesia.
randomized) Schizophrenia patients
Schizophrenia, 10/33
had higher rates of
buprenorphine
confusion or psychosis
Controls, buprenorphine 1/35 than controls.
5d
Postoperative analgesic
regimen and confusion
Kudoh et al17 RCT; minor surgery Schizophrenia 101 Postoperative confusion through POD 4 Preoperative
patients (P 0.05): discontinuation of
4d Antipsychotic continued 7/49 antipsychotics was
associated with more
Preoperative continuation Antipsychotic discontinued 16/52
postoperative
of antipsychotic drugs
confusion.
and postoperative
confusion
(Continued)

34 2008 Lippincott Williams & Wilkins

Annals of Surgery Volume 248, Number 1, July 2008 Postoperative Complications in the Seriously Mentally Ill

TABLE 1. (Continued)
Study Design, Follow-up Serious
Study Period, Question Mental Illness No. Patients Results Conclusion
Kudoh et al18 RCT-cohort; orthopedic Major 80 with MDD Postoperative confusion (P 0.01 for In patients with MDD,
surgery patients depressive (randomized), combined MDD vs. Controls): induction with
disorder 40 controls propofol and fentanyl,
(not MDD, propofol 5/40 compared with
randomized) fentanyl induction propofol alone, was
possibly associated
3d MDD, propofol induction 13/40 with less postoperative
Anesthetic induction Control, propofol 1/40 confusion. Patients
technique and fentanyl induction with MDD had higher
postoperative confusion rates of confusion than
controls.
Kudoh et al19 RCT-cohort; orthopedic Major 80 Postoperative confusion (P 0.05): Preoperative
surgery patients depressive MDD, antidepressant 5/40 discontinuation of
disorder continued antidepressants was
associated with more
MDD, antidepressant 12/40
postoperative
discontinued
confusion and
3d Control, no MDD, no 1/40 worsened depressive
antidepressant symptoms.
Preoperative continuation Depressive symptoms worsened (P 0.04):
of antidepressant MDD, antidepressant 2/40
medication and continued
postoperative confusion MDD, antidepressant 8/40
discontinued
CI indicates confidence interval; DVT or PE, deep venous thrombosis or pulmonary embolism; MDD, major depressive disorder; NO2, nitrous oxide; OR, odds ratio; POD,
postoperative day; post-op, postoperative; RCT, randomized controlled trial; SD, standard deviation; VAS, visual analog scale.

various operations.11 The only significant complications in
the first postoperative week were one case each of delayed
recovery from anesthesia (90 minutes) and supraventricular
tachycardia associated with cannulation of the internal jugu-
lar vein in a patient with Wolf-Parkinson-White syndrome.
Over 30 days of follow-up, 2 patients died of complications
of their underlying disease; 1 death was due to sepsis, the
other to multiorgan failure. The study concluded that the
anesthesia was safe for schizophrenia patients.
Cohort Studies (4)
One small study found pain perception thresholds were
significantly higher and postoperative pain scores signifi-
cantly lower for 50 schizophrenia patients compared with 25
controls.12 A second small study reported that patients with
schizophrenia had significantly more postoperative confusion
(14/50 (28%) schizophrenia patients vs. 2/35 (6%) controls,
P 0.01).15 There were no significant differences between
patients with and without postoperative confusion on postop-
erative pain scores, duration of operation and anesthesia, and
mean estimated blood loss, and total dose of fentanyl.
A third small cohort study evaluated postoperative
complications in veterans with schizophrenia undergoing
appendectomy in Veterans Health Administration (VHA)
hospitals from 1995 to 1999.7 A report from the VHA
National Surgical Quality Improvement Project (NSQIP) on
appendectomy provided comparison data. Of 55 patients, 9
(16%) had 1 postoperative complication and 22 (40%) expe-
rienced 2 or more complications, and 2 patients died. Com-
plications included postoperative ileus (13 patients), altered
mental status,13 wound infection,8 pneumonia,6 sepsis,6
wound dehiscence,4 and failure to extubate by the second
postoperative day.2 In addition, many patients (number not
reported) presented with symptoms or signs of psychiatric
disorder, including lack of insight, confusion, hallucinations,
and paranoia after surgery; a psychiatric consult was re-
quested for 40% of patients. Several factors suggested that
these patients with schizophrenia presented at a later stage of
disease than is typical. The incision was left open to heal by
secondary intention in 24 patients (44%), and there were
36 perforated (65%) and 9 gangrenous appendices (16%).
Overall morbidity and mortality rates were 56% and 4%,
compared with 16% and 1.8%, respectively, in the NSQIP
report.
The largest study was a retrospective cohort study that
compared outcomes for 1746 patients with schizophrenia and
732,158 patients without schizophrenia hospitalized for medical
or surgical conditions.6 Surgical patients comprised 26% of
those with schizophrenia (calculated n 466), and 46% of those
without schizophrenia (calculated n 338,257). Surgical pa-
tients with schizophrenia had significantly higher rates of post-
operative respiratory failure (adjusted odds ratio (OR) 2.1, 95%
confidence interval (CI) 1.4 3.1), postoperative deep-vein
thrombosis or pulmonary embolism (OR 2.0, CI 1.23.3), and
sepsis (OR 2.3, CI 1.53.5) compared with patients without
schizophrenia in models adjusting for demographics, system
factors, and medical versus surgical patient status.
Randomized Clinical Trials (4)
We identified only 4 randomized clinical trials (RCT),
all with relatively small sample sizes and conducted by one
research group, Kudoh et al.13,1517 These trials examined the

2008 Lippincott Williams & Wilkins 35

Copeland et al
effects of anesthesia protocols and medication discontinua-
tion on the perioperative clinical outcomes of postoperative
ileus, duration of ileus, and postoperative pain, psychosis and
confusion or delirium.
In the first RCT of schizophrenia patients having partial
or total colectomy for malignancy, 46 patients were random-
ized to either postoperative epidural analgesia with bupiva-
caine or intravenous buprenorphine by continuous infusion
for 72 hours and then with as-needed intravenous opioids and
diclofenac by rectal suppository.13 There were no differences
in mean duration of anesthesia, mean blood loss or duration
or potency of chronic antipsychotic medications, or fentanyl
consumption between the 2 groups. Patients receiving intra-
venous opioids had higher incidence of ileus (11/23 vs. 5/23;
P 0.06) and significantly longer duration of postoperative
ileus. Patients receiving buprenorphine also had higher post-
operative pain scores, compared with those receiving epidural
analgesia, at 8 hours and 24 hours after surgery; there were no
significant differences in pain scores for postoperative days 2
through 4.
A second trial investigated whether inhaled anesthesia
with fentanyl versus intravenous anesthesia with ketamine,
propofol, and fentanyl was associated with differential rates
of postoperative psychosis or confusion, using DSM-III cri-
teria to assess psychosis and the Confusion Assessment
Method (CAM)20 for postoperative confusion. There were no
significant differences in duration of surgery or anesthesia, or
duration of chronic antipsychotic medications, or mean blood
loss. Patients receiving inhaled anesthesia had higher rates of
postoperative psychosis or confusion compared with those
receiving intravenous anesthesia (20/37 patients (54%) vs.
11/37 patients (30%), P 0.05).15
The third RCT investigated whether intra- and postop-
erative epidural analgesia with bupivacaine affected the inci-
dence of postoperative confusion compared with analgesia
with intravenous buprenorphine.16 There were no significant
differences in mean duration of surgery and anesthesia, mean
blood loss, total fentanyl consumption, or postoperative pain
scores through 3 days after surgery. There was also no
difference in the incidence of postoperative confusion among
patients receiving epidural analgesia compared with those
receiving intravenous opioids (7/33 (21%) vs. 100/33 (30%),
P 0.57).
In the fourth trial, 101 schizophrenia patients undergo-
ing minor surgery were randomized to continue or discon-
tinue antipsychotic medications 72 hours before surgery.17
Patients were assessed daily for confusion with the CAM20
until 10 days after surgery. Postoperative confusion occurred
significantly less often in patients continuing antipsychotic
medications compared with those who discontinued medica-
tions (7/49 (14%) vs. 16/52 (31%), P 0.048). Serious
confusion with hallucinations occurred in 1 patient in the
group continued on medications and in 6 patients in the group
whose medications were discontinued (P 0.11). There were
no differences in rates of intraoperative hypotension (systolic
blood pressure 70 mm Hg at induction, 8 and 7 patients) or
arrhythmias (3 and 4 patients).
36
Annals of Surgery Volume 248, Number 1, July 2008
Perioperative Outcomes in Patients With Other
Serious Mental Illnesses
Although only 10 articles met inclusion criteria for
clinical surgery outcomes among schizophrenia patients, far
fewer focused on patients with other serious mental illnesses.
We identified only 2 relevant publications in patients with
MDD. We found no relevant studies of patients with bipolar
disorder or PTSD. Both eligible studies of patients with MDD
were combined RCT-cohort studies in which patients with
MDD were randomized to treatment condition and then
compared with a randomly selected cohort of patients without
depression; the controls were not randomized to treatment
condition.
In the first RCT-cohort study, 80 patients with MDD
were randomized to receive (N 40), or not receive (N
40) fentanyl intraoperatively; the 40 patients without MDD
also received the same dose of fentanyl intraoperatively.18
For all 3 groups, postoperative analgesia included a single
dose of intravenous buprenorphine and then diclofenac by
rectal suppository. Patients were evaluated with the CAM for
confusion for the first 3 days after surgery. There were no
significant differences in mean duration of surgery or anes-
thesia, and mean blood loss among the 3 groups. Patients with
MDD who received fentanyl developed more postoperative
confusion than patients with MDD who did not receive
intraoperative fentanyl (13/40 (33%) vs. 5/40 (13%), P
0.06). Confusion occurred less often in control subjects (1/40)
compared with the MDD group who received fentanyl
(P 0.001). The higher incidence of confusion in those with
MDD not receiving fentanyl was associated with higher
levels of plasma cortisol at 15 and 60 minutes after skin
incision, compared with those with MDD who received
fentanyl. It is unclear why the authors described the anes-
thetic regimens in detail and clearly stated that one group of
patients with MDD did not receive fentanyl intraoperatively
but then reported no difference in total fentanyl consumption
among all 3 groups (376.2 (SD 30.7), 358.2 (SD 25.4), and
365.8 (SD 20.4) g, respectively).
In the second RCT-cohort study, 80 patients with MDD
were randomized to perioperative continuation or discontin-
uation of antidepressants 72 hours before surgery and com-
pared with 40 controls.19 Antidepressants were restarted on
the first postoperative day. There were no significant differ-
ences in mean duration of anesthesia and surgery, mean blood
loss, or total fentanyl consumption between the 3 groups; the
reported values for these variables are almost identical to
those in the study reviewed just above. Preoperative discon-
tinuation of antidepressants was associated with increased
postoperative depression symptoms (8/40 (20%) vs. 2/40
(5%), P 0.04) and delirium or confusion (12/40 (30%) vs.
4/40 (13%), P 0.05) but not with postoperative pain levels,
intraoperative hypotension (systolic blood pressure 70 mm Hg
at induction, 2 and 3 patients in the MDD groups), or arrhyth-
mias (ventricular ectopy, 2 patients in each MDD group).
DISCUSSION
The results of our systematic literature review reveal a
very limited knowledge base regarding postoperative clinical
2008 Lippincott Williams & Wilkins
Annals of Surgery Volume 248, Number 1, July 2008
complications and perioperative risk management among
patients previously diagnosed with serious mental illness.
The available evidence suggests that patients with schizo-
phrenia undergoing elective and emergency operations may
have more advanced surgical disease at presentation (eg,
perforated appendix), higher postoperative complication rates
(eg, cardiovascular events, venous thromboembolism, or re-
spiratory failure), deteriorating psychiatric status if psycho-
tropic medications are discontinued preoperatively, and in-
creased postoperative mortality. Perioperative anesthesia and
analgesia may need to be tailored for patients with schizo-
phrenia to prevent complications but little is known regarding
the best anesthetic and analgesic regimens. The published
information also confirms clinical perceptions that schizo-
phrenia patients seem to have higher pain thresholds and less
postoperative pain. Reduced pain perception may impair or
delay patients ability to report symptoms of postoperative
infection.
Information about perioperative care and outcomes
among patients with serious mental illness other than schizo-
phrenia is even sparser. We found no studies of postoperative
complications in patients with bipolar disorder or PTSD.
Only 2 studies of patients with MDD undergoing surgery met
our eligibility criteria.18,19 In one, intraoperative fentanyl was
associated with higher rates of postoperative confusion; in the
other, preoperative discontinuation of antidepressant media-
tions was associated with higher rates of postoperative delir-
ium or confusion and worsened depression. However, the
studies, as published, raise questions regarding the accuracy
of the data and whether the same patients were in each study
and thus received 2 interventions without a clear factorial
design.
The current implications for clinicians include1 psychi-
atric medications should be continued, if possible, during the
perioperative period, and2 much more research is needed
regarding anesthetic and analgesic care and other issues of
perioperative risk management to reduce complications. Be-
cause there is evidence of increased postoperative complica-
tions in patients with serious mental illness, and given schizo-
phrenia patients risk for cardiac irregularities,21 obesity,22
and respiratory disorders,23,24 it also seems advisable to3
involve psychiatric consultation services in perioperative
management of these patients.
Perioperative pharmacological management of aging
patients with SMI is challenging. In general, older patients
with SMI are at greater risk of receiving drugs that should be
avoided in the elderly;25 thus perioperative risk management
may further complicate prescribing decisions. Many older
patients with SMI also have an array of physical comorbidi-
ties and thus receive numerous medications.26 Treatment of
common disease states and their concomitant risk factors (eg,
hyperlipidemia, hypertension) often requires 5 or more med-
ications,2,27,28 which may interact with both the psychiatric
medications and postoperative pain medications. For in-
stance, Micromedex indicates that fentanyl has major drug
interactions with antihypertensives such as nifedipine, amlo-
dipine, and atenolol that may result in severe hypotension.
Oxycodone, a common oral pain medication, and sertraline, a
2008 Lippincott Williams & Wilkins
Postoperative Complications in the Seriously Mentally Ill
medication commonly used to treat depression, have a
major interaction that may result in increased risk of
tachycardia, hyperthermia, myoclonus, and mental status
changes all of which complicate the acute care and recov-
ery of patients with SMI.
Moreover, a number of patients with SMI have comor-
bid substance abuse disorders.29,30 For those with substance
abuse disorders in remission, clinicians are presented with the
challenge of managing postoperative pain without exacerbat-
ing or reactivating addiction. Physicians may under-prescribe
opiates to avoid aggravating the addiction, but this may lead
to poor pain control, triggering relapse. This should be of
particular concern for patients with affective disorders, for
whom pain may be heightened,31,32 rather than for patients
with schizophrenia for whom pain may be less. For patients
with an active substance abuse disorder, postoperative care is
complicated by dealing with detoxification.33
Increased rates of delirium and confusion in the post-
operative period have been reported in patients with schizo-
phrenia. In surgical, medical, and critically ill patients, delir-
ium is associated with higher mortality, longer hospital stay,
and impairment at hospital discharge.34 Some studies have
described a benefit of using antipsychotic medication in
critically ill patients, suggesting that appropriate recognition
and treatment of delirium and confusion is associated with
better outcomes.17,35 These data suggest that the development
of confusion, delirium, or cognitive disorders in postoperative
patients may have important consequences in health care
utilization and patient outcomes, and further studies in pa-
tients with SMI are needed.
To our knowledge, no studies have adequately assessed
the epidemiology of surgical care and outcomes for patients
with schizophrenia and other serious mental illnesses. Al-
though very large, with more than one million major surgical
cases, the Veterans Affairs National Surgical Quality Im-
provement Program (NSQIP)36 does not collect data on
preoperative psychiatric disease, nor does its recently intro-
duced counterpart in the private sector through the Amer-
ican College of Surgeons (ACS NSQIP).37 Thus, these 2
rich prospective databases cannot be used to study the
epidemiology of perioperative care and outcomes in pa-
tients with serious mental illness, although it has been
suggested that they be modified to gather psychiatric
comorbidity in the future.7
Patients with schizophrenia or other serious mental
illness comprise a vulnerable population that relies heavily on
publicly funded healthcare. Most, including many veterans of
the Vietnam era, are from the aging Baby Boomer generation,
entering a phase of life marked by increasing frequency of
surgery. As noted in the Introduction section, these patients
have high rates of cardiovascular, gastrointestinal, and respi-
ratory illness, yet few studies of their surgical treatment were
found. Their deficits in self-care and engagement in un-
healthy lifestyle behaviors are well documented,29,38 42 sug-
gesting that adverse sequelae of surgery may be especially
problematic for this group. As this large cohort of patients
moves into late life, clinicians must deal increasingly with
potentially large numbers of difficult-to-manage patients with
37
Copeland et al
serious medical disease profiles whereas policy-makers jug-
gle scarce public dollars and medical resources. Documenta-
tion of SMI patients comorbid disease burden and postop-
erative complications such as respiratory failure, sepsis, and
deathin covariable-adjusted analyses is sorely needed.
REFERENCES
1. Mortensen PB, Juel K. Mortality and causes of death in first admitted
schizophrenic patients. Br J Psychiatry. 1993;163:183189.
2. Brown S, Inskip H, Barraclough B. Causes of the excess mortality of
schizophrenia. Br J Psychiatry. 2000;177:212217.
3. Curkendall SM, Mo J, Glasser DB, et al. Cardiovascular disease in
patients with schizophrenia in Saskatchewan, Canada. J Clin Psychiatry.
2004;65:715720.
4. Kudoh A. Perioperative management for chronic schizophrenic patients.
Anesth Analg. 2005;101:18671872.
5. Copeland LA, Zeber JE, Mortensen EM, et al. Surgery rates among
Veterans with Schizophrenia. International congress on Schizophrenia
research. Colorado Springs, CO, 2007.
6. Daumit GL, Pronovost PJ, Anthony CB, et al. Adverse events during
medical and surgical hospitalizations for persons with schizophrenia.
Arch Gen Psychiatry. 2006;63:267272.
7. Cooke BK, Magas LT, Virgo KS, et al. Appendectomy for appendicitis
in patients with schizophrenia. Am J Surg. 2007;193:41 48.
8. Goldman LS. Medical illness in patients with schizophrenia. J Clin
Psychiatry. 1999;60(suppl 21):S10 S15.
9. Kemp R, Hayward P, Applewhaite G, et al. Compliance therapy in
psychotic patients: randomised controlled trial. BMJ. 1996;312:345
349.
10. Matsuki A, Oyama T, Izai S, et al. Excessive mortality in schizophrenic
patients on chronic phenothiazines treatment. Agressologie. 1972;13:
407 418.
11. Ishihara H, Kudo H, Murakawa T, et al. Uneventful total intravenous
anaesthesia with ketamine for schizophrenic surgical patients. Eur J
Anaesthesiol. 1997;14:4751.
12. Kudoh A, Ishihara H, Matsuki A. Current perception thresholds and
postoperative pain in schizophrenic patients. Reg Anesth Pain Med.
2000;25:475 479.
13. Kudoh A, Katagai H, Takazawa T. Effect of epidural analgesia on
postoperative paralytic ileus in chronic schizophrenia. Reg Anesth Pain
Med. 2001;26:456 460.
14. Kudoh A, Takahira Y, Katagai H, et al. Schizophrenic patients who
develop postoperative confusion have an increased norepinephrine and
cortisol response to surgery. Neuropsychobiology. 2002;46:712.
15. Kudoh A, Katagai H, Takazawa T. Anesthesia with ketamine, propofol,
and fentanyl decreases the frequency of postoperative psychosis emer-
gence and confusion in schizophrenic patients. J Clin Anesth. 2002;14:
107110.
16. Kudoh A, Takase H, Takahira Y, et al. Postoperative confusion in
schizophrenic patients is affected by interleukin-6. J Clin Anesth. 2003;
15:455 462.
17. Kudoh A, Katagai H, Takase H, et al. Effect of preoperative discontin-
uation of antipsychotics in schizophrenic patients on outcome during and
after anaesthesia. Eur J Anaesthesiol. 2004;21:414 416.
18. Kudoh A, Takahira Y, Katagai H, et al. Cortisol response to surgery and
postoperative confusion in depressed patients under general anesthesia
with fentanyl. Neuropsychobiology. 2002;46:2226.
19. Kudoh A, Katagai H, Takazawa T. Antidepressant treatment for chronic
depressed patients should not be discontinued prior to anesthesia. Can J
Anaesth. 2002;49:132136.
20. Inouye SK, van Dyck CH, Alessi CA, et al. Clarifying confusion: the
confusion assessment method. A new method for detection of delirium.
Ann Intern Med. 1990;113:941948.
21. Bar KJ, Boettger MK, Koschke M, et al. Non-linear complexity mea-
38
Annals of Surgery Volume 248, Number 1, July 2008
sures of heart rate variability in acute schizophrenia. Clin Neurophysiol.
2007;118:2009 2015.
22. Tirupati S, Chua LE. Obesity and metabolic syndrome in a psychiatric
rehabilitation service. Aust N Z J Psychiatry. 2007;41:606 610.
23. Copeland LA, Mortensen EM, Zeber JE, et al. Pulmonary disease among
inpatient decedents: Impact of schizophrenia. Prog Neuropsychophar-
macol Biol Psychiatry. 2007;31:720 726.
24. Sokal J, Messias E, Dickerson FB, et al. Comorbidity of medical
illnesses among adults with serious mental illness who are receiving
community psychiatric services. J Nerv Ment Dis. 2004;192:421 427.
25. Pugh MJ, Fincke BG, Bierman AS, et al. Potentially inappropriate
prescribing in elderly veterans: are we using the wrong drug, wrong
dose, or wrong duration? J Am Geriatr Soc. 2005;53:12821289.
26. Osby U, Correia N, Brandt L, et al. Mortality and causes of death in
schizophrenia in Stockholm county, Sweden. Schizophr Res. 2000;45:
2128.
27. Holt RI, Bushe C, Citrome L. Diabetes and schizophrenia 2005: are we
any closer to understanding the link? J Psychopharmacol. 2005, 19
(6 suppl):S56 S65.
28. Boyd CM, Darer J, Boult C, et al. Clinical practice guidelines and
quality of care for older patients with multiple comorbid diseases:
implications for pay for performance. JAMA. 2005;294:716 724.
29. Lambert TJ, Velakoulis D, Pantelis C. Medical comorbidity in schizo-
phrenia. Med. J. Aust 2003;178(suppl):S67S70.
30. Rounsaville BJ. DSM-V research agenda: substance abuse/psychosis
comorbidity. Schizophr Bull. 2007;33:947952.
31. Arnow BA, Hunkeler EM, Blasey CM, et al. Comorbid depression,
chronic pain, and disability in primary care. Psychosom Med. 2006;68:
262268.
32. Johnson KM, Bradley KA, Bush K, et al. Frequency of mastalgia among
women veterans. Association with psychiatric conditions and unex-
plained pain syndromes. J Gen Intern Med. 2006;21(suppl 3):S70 S75.
33. Zeigler PP. Treatment of pain in the patient with a substance use
disorder. Psychiatr Times. 2007;24:1.
34. Ely EW, Shintani A, Truman B, et al. Delirium as a predictor of
mortality in mechanically ventilated patients in the intensive care unit.
JAMA. 2004;291:17531762.
35. Milbrandt EB, Kersten A, Kong L, et al. Haloperidol use is associated
with lower hospital mortality in mechanically ventilated patients. Crit
Care Med. 2005;33:226 229, discussion 263265.
36. Khuri SF. Safety, quality, and the National Surgical Quality Improve-
ment Program. Am Surg. 2006;72:994 998; discussion 10211030,
11331148.
37. Fink AS, Campbell DA Jr, Mentzer RM Jr, et al. The National Surgical
Quality Improvement Program in non-veterans administration hospitals:
initial demonstration of feasibility. Ann Surg. 236:344 353, 2002;
discussion 353354.
38. Dixon L, Weiden P, Delahanty J, et al. Prevalence and correlates of
diabetes in national schizophrenia samples. Schizophr Bull. 2000;26:
903912.
39. Tarone RE, Chu KC. Age-period-cohort analyses of breast-, ovarian-,
endometrial- and cervical-cancer mortality rates for Caucasian women in
the USA. J. Epidemiol Biostat. 2000;5:221231.
40. Daumit GL, Pratt LA, Crum RM, et al. Characteristics of primary care
visits for individuals with severe mental illness in a national sample. Gen
Hosp Psychiatry. 2002;24:391395.
41. Gfroerer JC, Penne MA, Pemberton MR. The aging Baby Boom Cohort
and future prevalence of substance abuse. In: Korper SP, Council CL,
eds. Substance Use by Older Adults: Estimates of Future Impact on the
Treatment System. Rockville, MD: Substance Abuse and Mental Health
Services Administration, Office of Applied Studies, 2002. Available at:
http://www.oas.samhsa.gov/Aging/chap5.htm.
42. ADA. Consensus development conference on antipsychotic drugs and
obesity and diabetes. Diabet Care. 2004;27:596 601.
2008 Lippincott Williams & Wilkins