Crit Care Clin 19 (2003) 33 – 53

Critical care in orthopedic and spine surgery

Daniel Nazon, MDa,*, Glen Abergel, MDa,
Carlo M. Hatem, MDb
a
Department of Critical Care Medicine, Montefiore Medical Center,
Albert Einstein College of Medicine, 111 East 210th Street, Bronx, NY 10467, USA
b
Department of Critical Care/Pulmonary Medicine, Montefiore Medical Center,
Albert Einstein College of Medicine, 111 East 210th Street, Bronx, NY 10467, USA

Orthopedic and spine operations are being performed at an increased rate and
account for a significant part of the health care budget [1]. Familiarity with the
perioperative aspects of these procedures will help improve outcomes, reduce the
length of stay (LOS) in the intensive care unit (ICU), and limit health care costs.
The need for ICU care can often be predicted based on patient’s comorbidities
and operative considerations. Unplanned admissions can also arise from prob-
lems that occur in the operative or recovery period. This article reviews those
aspects of critical care related to elective orthopedic and spine surgery.

Hip
An estimated 200,000 total hip replacement (THR) operations are performed
each year in the United Sates for management of painful arthritic hips [2]. Newer
anesthetic techniques, better perioperative care, and improved prophylaxis for
deep vein thrombosis (DVT) have contributed to a threefold decrease in
morbidity and mortality.

Preoperative considerations
Data from a National Institutes of Health (NIH) survey estimate the overall
risk of death after THR to be less than 1% [3]. The most important determinant
of mortality is age. Patients between 66 and 69 years of age have a perioperative
mortality of 0.34%, compared with 3.75% in the age group above 85 years.
Women usually have a lower mortality rate. Total hip replacement for hip fracture

* Corresponding author.
E-mail address: dnazon@aol.com (D. Nazon).

0749-0704/03/$ – see front matter D 2003, Elsevier Science (USA). All rights reserved.
PII: S 0 7 4 9 - 0 7 0 4 ( 0 2 ) 0 0 0 5 2 - 0
34 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

carries a greater mortality risk (6.2%) than THR for arthritis or revision
(0.95%) [4].
Baseline pulmonary function correlates with the risk of postoperative pul-
monary complications. Patients with chronic obstructive pulmonary disease
(COPD) or abnormal pulmonary function tests have a 70% pulmonary complica-
tion rate [5]. Age has an independent effect on lung function manifested by a
progressive reduction in forced expiratory volume in 1 second (FEV1) and loss of
elasticity. Arterial blood gases and spirometry are recommended preoperatively
for smokers and patients with known lung disease.
Goldman [6,7] describes nine variables that predispose a patient to cardiac
complications. These variables are (1) S3 gallop or jugular venous distension, (2)
myocardial infarction in past 6 months; (3) premature ventricular contractions
(more than 5/minute); (4) rhythm other than sinus on preoperative electrocardio-
gram (EKG); (5) age over 70 years; (6) emergency surgery; (7) intraperitoneal,
intrathoracic, or aortic surgery; (8) suspected critical aortic stenosis; and (9) poor
general medical condition. Patients over age 40 years with cardiovascular risk
factors should have a resting EKG before surgery. If ischemic changes are noted,
a stress test is required. Because patients undergoing THR are usually unable to
exercise, a thallium-dipyridamole scan is an alternative. The decision to proceed
to a cardiac catheterization should be made by a cardiologist.
Other important risk factors include obesity, smoking, and malnutrition. In an
Australian cohort, up to 71% of patients undergoing THR had one or more
serious medical problems preceding surgery [8]. Useful prophylactic measures
are smoking cessation, weight reduction, breathing exercises, and correction of
nutritional deficiencies.

Anesthesia considerations
General and regional anesthesia affects the pulmonary and cardiovascular
systems in different ways. General anesthesia decreases lung compliance,
depresses cough reflex, increases pulmonary secretions, and alters mucociliary
clearance. Preexisting pulmonary disease magnifies these effects. The duration of
general anesthesia also plays a role in the incidence of postoperative respiratory
problems. Procedures lasting longer than 3 hours have a threefold higher rate of
pulmonary complications [9]. On the other hand, regional anesthesia has been
shown to reduce postoperative pulmonary morbidity. A randomized trial com-
paring general and regional anesthesia in elderly patients undergoing THR
showed a significant drop in the postoperative PaO2 in the general anesthesia
group [10].
Regional anesthesia compared with general anesthesia reduces the incidence
of thromboembolic events, probably because of reduced venous stasis [11,12],
and is preferred in patients with difficult airways, such as concomitant unstable
rheumatoid spine.
General anesthesia tends to depress cardiac contractility, whereas regional
anesthesia has minimal effects on the myocardium. Regional anesthesia lowers
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 35

peripheral vascular resistance, however. Few studies have compared anesthetic

techniques and cardiac complications after THR. One retrospective study found
fewer cardiac complications after spinal anesthesia [13]. In another prospective
study, there were no cardiac complications in the regional anesthesia group,
compared with a high rate of postoperative hypoxemia and one death from
myocardial infarction after general anesthesia [10].
Peripheral vasodilatation that occurs during regional anesthesia leads to
hypotension with a concomitant reduction in blood loss. A mean arterial blood
pressure of 60 mm Hg was found to be optimal, maximizing beneficial effects
while avoiding neurologic injury and tissue ischemia [14]. Deliberate hypoten-
sion requires invasive cardiac monitoring and may be contraindicated in patients
with ischemic or valvular disease [7].

Cardiopulmonary complications
Pulmonary complications are among the most common problems encountered
in the perioperative period. Atelectasis is common after general anesthesia. Its
incidence has been estimated to be between 2% and 7% [15 –17].
Postoperative pneumonia occurs in as many as 10% of elderly patients [18].
Colonization of the upper respiratory tract, particularly after protracted ventila-
tory support, occurs in nearly 100% of these cases. Initial empiric broad-spectrum
antibiotic coverage can be readjusted based on culture results.
Acute respiratory distress syndrome (ARDS) characterized by intrapulmonary
shunting, mismatch of the ventilation/perfusion ratio (V/Q), and decreased lung
compliance with subsequent respiratory insufficiency has been described after
major orthopedic operations. Numerous causes have been implicated, including
sepsis and fat embolism syndrome (FES). Mechanical ventilation with the use of
positive end-expiratory pressure (PEEP) may be necessary to improve gas
exchange [19]. Limiting the tidal volume to 5 to 7 mL/kg reduces mortality
from ARDS [20]. Efforts should be made to identify and treat confounding
problems such as sepsis, multiorgan failure, and cardiac failure [19,21].
Cardiac complications after THR occur at a rate of 2% to 10% [18]. The
incidence of ischemic heart disease, including angina pectoris and myocardial
infarction, is between 0.59% and 1.37% after THR [16,22]. Myocardial infarction
following noncardiac surgery is associated with a mortality rate as high as 70%
[23]. Chest pain is reported in 50% of these patients and may be masked by the
administration of narcotics or sedatives. Postoperative EKG changes of unknown
significance occur in 20% of patients [24]. Elevation of the creatinine phospha-
tase isoenzyme (CPK-MB) should not be attributed to the trauma of hip surgery.
Troponin assays are sensitive markers for cardiac ischemia [25].
Congestive heart failure (CHF) is common in patients with preexisting cardiac
disease who receive large amounts of fluids and blood products. Finally, cardiac
arrhythmias are common reasons for ICU care. The incidence of supraventricular
tachycardia is estimated at 4.8% after THR [26] and is more likely to occur in
older patients with a previous cardiac or pulmonary history.
36 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

Pulmonary embolism
Among patients admitted to the hospital, those with multiple traumas or
undergoing elective major joint replacement are at greatest risk for DVT [27].
The risk of thromboembolism increases with age, the extent and duration of
surgery, the type of anesthesia, the duration of immobilization, and the severity of
comorbidities [28,29]. Up to 50% of patients with lower extremity fractures
develop DVT, with 10% progressing to pulmonary embolism (PE), and 2% dying
from massive PE [30]. Of those confirmed cases, 11% die within the first hour,
and 8% will suffer a fatal PE despite adequate anticoagulation [31]. Because of
the lack of uniformity among studies, the actual incidence of DVT and PE after
THR remains controversial. A recent study evaluated the risk of DVT after THR
to be between 10% and 25%, with a fatal PE rate of 0.15% [32]. Haake and
Berkman [33] reported a cumulative incidence of fatal PE of 0.06% to 6.7%
compared with 3.6% to 12.9% following hip fractures. The true incidence of
thromboembolism after THR is probably underestimated because of limitations in
diagnostic tools.
Prophylaxis against thromboembolism in orthopedic surgery is paramount.
The Sixth American College of Chest Physicians Conference on Thromboem-
bolism has made detailed recommendations for prophylaxis after joint replace-
ment [34]. For THR, low molecular weight heparin (LMWH) or adjusted-dose
warfarin, with a target international normalized ratio (INR) of 2.5, should be
started preoperatively or immediately after surgery. Other prophylactic regimens
have been advocated.
The test most commonly used to diagnose DVT is duplex ultrasonography. It
can be safely performed at the bedside and has almost 100% sensitivity and 99%
specificity for proximal DVT [35].
Suspicion of PE must remain high in patients undergoing THR [36]. A
ventilation/perfusion scan is less likely to be useful in patients with respiratory
complications. A spiral computed tomographic (CT) scan with intravenous
contrast infusion is a useful alternative. It has an estimated sensitivity of 70%
and a specificity of 81 to 100%. Spiral CT has limited ability to detect peripheral
emboli [37]. Pulmonary angiogram remains the criterion standard, but it is greatly
underutilized [38]. Guidelines on the treatment of thromboembolism have been
published elsewhere [39].
Most patients with pulmonary embolism do not need admission to ICU unless
they are experiencing cardiopulmonary instability requiring monitoring, respi-
ratory support, or hemodynamic augmentation. Patients with shock after PE are at
greatest risk of death. Thrombolytic therapy has been advocated in this setting. It
has been proven superior to heparin in the resolution of PE as assessed by
pulmonary angiogram, lung scan, and hemodynamic derangement [40]. It is
associated with a significant risk of bleeding, however, and might be contra-
indicated in the immediate postoperative period. Finally, patients with PE and
contraindications to anticoagulation should be considered for an inferior vena
cava filter [41].
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 37

Fat embolism
Fat embolism is a well-recognized complication of orthopedic surgery [42 –
44]. Fat embolism syndrome is a triad syndrome, and fat embolism is a
subclinical form of FES. A review of the literature reveals a wide variation on
the prevalence of FES. The occurrence of fat embolism after major trauma with
long bone involvement has been estimated to be 90% [45,46]. The prevalence of
FES in these trauma cases is believed to be underestimated. The operations most
commonly associated with FES are THR, total knee replacement (TKR), and
intramedullary nailing of the shaft [47,48]. Although not reported in the literature,
the actual incidence of FES with these operations is believed to be high.
There are two theories for the pathogenesis of FES: a mechanical theory and a
biochemical theory [49]. The mechanical theory proposes that bone marrow
contents enter the pulmonary vasculature through the venous system. Intra-
operatively, this process could result from pressurization and manipulation of
long bone intramedullary canals [50]. Small fat droplets ranging in size from 7 to
10 microns in diameter may circulate through the pulmonary capillaries and into
the systemic circulation, terminating in the brain or other organs. Another mode
of entering the systemic vasculature is through a patent foramen ovale [51]. Up to
10% of the adult population has a patent foramen ovale [52].
The biochemical theory of FES suggests that circulating free fatty acids
directly affect the pneumocytes. The source of free fatty acids may be multi-
factorial. C-reactive protein, which may cause chylomicrons to coalesce and form
fat globules, is elevated after trauma and sepsis. Catecholamines promote the
release of free fatty acids and are also elevated after trauma [53]. The two
mechanisms are believed to act synergistically [45]. The presence of fat through-
out the pulmonary vasculature may lead to pulmonary hypertension, right heart
strain, hypotension, hypoxia, and significant cardiopulmonary compromise.
Patients who are vulnerable to clinical compromise are those with reduced
cardiopulmonary reserve, pathologic fractures, and a right-to-left shunt [54].
Surgical techniques may have a significant effect on the severity of fat emboliza-
tion [42]. For example, careful intramedullary lavage before placement of the
prosthesis will decrease the amount of debris available for embolization [55].
Fat embolism syndrome is a diagnosis of exclusion that depends on the
clinician’s index of suspicion. Gurd [56] described major and minor features for
diagnosing FES. The major features are respiratory insufficiency, cerebral signs,
and petechial rash; the minor features are pyrexia, tachycardia, retinal involve-
ment, jaundice, renal involvement, and fat macroglobulinemia. The diagnosis of
FES requires at least one major and four minor criteria. Signs and symptoms
usually become apparent within 24 to 48 hours after insult; however, they may
appear immediately.
The classic clinical triad for FES is pulmonary insufficiency, neurologic
derangement, and petechial rash [57,58]. Pulmonary insufficiency is always
present is and usually the first manifestation of the syndrome. Neurologic
derangements occur in up to 86% of patients with FES; they manifest as headache,
38 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

lethargy, irritability, delirium, stupor, convulsions, and coma [59]. Preservation of

normal muscle tone and reflexes suggests a favorable prognosis. Nonreactive
pupils and depressed deep tendon reflexes that do not resolve after correction of
hypoxemia are suggestive of ischemia and microscopic hemorrhage in association
with emboli in the brain [11]. Petechiae occur in 20% to 50% of cases; these
lesions represent fat globules lodged in dermal capillaries. They are usually seen
on the head, neck, anterior chest, axilla, oral mucous membrane, and conjunctivae.
Although thrombocytopenia is common in FES, petechiae may be present in
the absence of thrombocytopenia; these lesions usually resolve within 5 to 7 days
[58,60]. Bronchoalveolar lavage seems to be unreliable in the detection of fat
embolism because normal, control patients have fat-containing alveolar macro-
phages ranging from 5% to 95% in alveolar samples. Pulmonary artery catheter-
ization with frozen section of aspirated blood may be useful when the cause of
respiratory failure is equivocal. The overall mortality rate is high (5% to 33%)
despite adequate resuscitation. The cause of death is usually related to the severity
of the respiratory problem. Long-term morbidity associated with FES is caused by
cerebral complications [59]. Treatment for FES is supportive, consisting primarily
of mechanical ventilation for respiratory compromise and hemodynamic stabi-
lization. There are no data supporting the use of heparin or corticosteroids in FES.

Cement-related issues
Complications occurring during cemented arthroplasty have been reported
since the early 1970s. In early studies, the incidence of polymethylmethacrylate
(PMMA)-related hypotension was between 33% and 100%. More recent studies
have shown a lower incidence. One study noted acute hypotension in 4.8% of
cemented THR compared with 0% after insertion of noncemented stems [61].
The cause of cement-related hypotension is still controversial. Initially
methylmethacrylate monomers were implicated as a causative factor. Intravenous
administration of methylmethacrylate in animal studies produces profound
peripheral vasodilation. Anaphylatoxin release (complement fractions C3a and
C5a), prostaglandins [62], and cement-related myocardial depression [63] have
also been suggested as other possible mechanisms.
Finally, abundant experimental evidence now implicates pulmonary emboliza-
tion of marrow contents (air, fat, and bone marrow) as the cause of cement-related
hypotension. These emboli contribute to hypotension through a mechanical effect
and the initiation of a chemical cascade [64]. Intraoperative transesophageal
echocardiography (TEE) performed during cemented THR demonstrates a snow
flurry of echogenic material in the right atrium after component insertion. This
phenomenon was attenuated by placement of a venting hole in the shaft of the
femur, thus preventing the rise of pressure in the medullary space [65].
Clinical changes during cemented THR occur after 30 minutes [66]. The most
common clinical finding is a transient decrease in arterial oxygen tension.
Pulmonary shunting has been calculated to be 28%, a finding that persists for
48 hours [67]. Venting of the femoral shaft to prevent intramedullary pressure and
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 39

embolization of marrow contents has resulted in lower drops in arterial oxygen

tension [68]. Significant hypotension occurs less commonly, in about 5% of
cases. Cardiac arrest has also been reported [69].
Those at risk are elderly patients with malignancy, cardiac, and pulmonary
disease [66]. Insertion of a long-stem femoral component in an unoperated
canal is also a predisposing factor. Modern techniques such as high-volume,
pulsatile lavage of the femoral canal with combined pressurized cement
insertion have been shown to eliminate significant reduction in arterial oxygen
tension as well as pulmonary shunt [55]. Maintenance of arterial oxygenation
and adequate volume replacement are essential during surgery. Patients at risk
benefit from invasive monitoring and vasopressors at the time of cement
insertion. Consideration should also be given to placement of a venting hole
into the femoral component.
Noncemented hip arthroplasty was developed in response to evidence that
cement promotes bone loosening and prosthetic failure. Although the incidence
of cardiac complications is lower with noncemented stems, cemented hip
arthroplasty remains the standard in elderly patients. Noncemented prosthesis
is reserved to patients under the age of 60 years or those undergoing surgical
revision. In many centers, hybrid THR is now the preferred technique. It consists
of a cemented femoral stem with a noncemented acetabular cup [70].

Local complications
Complications related to wound healing range from prolonged wound drain-
age to perhaps the most dreaded complication after THR, deep wound infection.
Factors that predispose to poor wound healing should be identified preopera-
tively: rheumatoid arthritis, chronic steroid administration, diabetes mellitus,
advanced age, and malnutrition [71].
Prophylactic antibiotics, meticulous surgical techniques, and control of the
intraoperative environment are critical in the prevention of infections. The
infection rate after THR is 5.8% when THR is performed in a regular operating
room without antibiotic coverage, 1.3% for a regular operating room with
preventive antibiotics, 0.7% for a laminar flow room without antibiotics, and
0.6% for a laminar flow room with antibiotics [72].
Early deep wound sepsis can be defined as deep infection of the surgical
wound as well as the prosthetic components. Supportive care in the ICU may be
indicated, in the setting of fulminant sepsis or septic shock, but the treatment
remains essentially surgical. The most widely advocated plan of care consists of
aggressive débridement of the wound, retention of the prosthetic components,
and antibiotics targeted at gram-positive organisms. This approach had a 71%
success rate in one series [73]. Failure of treatment will necessitate more
extensive procedures, such two-stage exchange arthroplasty.
Fortunately, neurovascular injuries are uncommon, but they are certainly
devastating for the patient and the surgeon. The incidence of vascular injury is
quite low, in the range of 0.1% to 0.2% [74]. The most serious injuries affect the
40 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

iliac and femoral vessels and require immediate surgical correction. Delayed
injury to the iliac vessels has been reported as a result of implant migration [75].
The incidence of nerve palsy after primary THR ranges from 0% to 3% and from
2.9% to 7.6% after revision. Most of those injuries are to the sciatic nerve [76]. The
causes are diverse, including direct trauma, ischemia, injury from the heat of
the cement, compression by a hematoma, or dislocation of the prosthesis [77]. The
prognosis is variable, with at least some recovery unless the nerve was transected.

Other complications
Gastrointestinal complications range from 1.2% to 4.6% following THR
[16,22]. The most common is postoperative paralytic ileus. Gastrointestinal
bleeding can also complicate the postoperative course. Patients at risk for gastro-
intestinal bleeding after THR include those with a history of peptic ulcer disease
and the use of nonsteroidal anti-inflammatory drugs (NSAIDS) [78]. Perioperative
management consists of administration of histamine 2 (H2) receptor antagonists,
proton-pump inhibitors, and cessation of NSAIDS. Postoperative management
will require ICU admission, fluid resuscitation, antiulcer therapy, and endoscopy.
Electrolyte imbalances, most commonly hyponatremia, can complicate the
postoperative course. After THR, the incidence of hyponatremia is 4.4% [79].
Patients may be asymptomatic or exhibit symptoms ranging from lethargy to
coma. Hyponatremia occurs in the setting of low total body sodium and water (in
patients taking diuretics) or administration of hypotonic fluids after surgery [80].
Rapid correction of severe hyponatremia carries the risk of cerebral edema.
One of the most common problems after THR is urinary retention requiring
bladder catheterization. Its incidence varies from 1% to 35% [81]. Urinary
retention predisposing to urinary tract infection after THR can have serious
consequences. Patients requiring bladder catheterization have a 6.2% incidence of
deep hip infections [82]. Most authors recommend insertion of a short-term
indwelling catheter after surgery.

Knee
It is estimated that about 150,000 TKRs are performed yearly in the United
States [83]. Major complications include pulmonary embolism, myocardial
infarction, and infections. Careful preoperative evaluation is critical. The type
of anesthesia is critical, with epidural anesthesia the preferred choice. Intra-
operative monitoring includes the use of a Swan-Ganz catheter in selected
unilateral and in most bilateral cases.

Thromboembolism
Deep vein thrombosis after TKR occurs in approximately 50% of unilateral
cases and as many as 75% of bilateral cases when no prophylaxis is used [84,85].
Although most authors agree that DVT occurs mainly in calf veins and is probably
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 41

of little consequence, it should be regarded as a harbinger of proximal clots with

serious consequences. Hass et al [86] found that 50% of patients had thrombosis of
the calf veins and 8% had thrombosis of the proximal veins. Of those, 4.5% had
documented positive lung scans. Symptomatic PE occurred in 1.1%. Prophylaxis
for DVT remains a prudent decision after TKR. Commonly used prophylaxis
includes LMWH, warfarin, heparin, aspirin, and compression devices.

Fat embolism
Fat embolism may be unrecognized as the cause of transient postoperative
confusion or hypoxia mimicking PE. The exact prevalence after TKR has not
been reported in the literature. In a review, Monto et al [87] reported 19 cases
with 9 deaths, mostly associated with long-stem cemented prostheses.
The use of a pneumatic tourniquet does not protect against fat embolism.
During tourniquet deflation, the absorption of fat and surgical debris into the
medullary sinusoids might result in a sudden embolic load to the lungs. Trans-
esophageal echocardiography has demonstrated echogenic material in the right
heart chambers in most patients following tourniquet deflation [88,89]. For-
tunately, clinical manifestations are rare [90].

Local complications
Tourniquets are used to control blood loss during lower extremity procedures.
Inappropriate usage has been associated with both neurovascular and tissue
damage. Tissues directly beneath are subject to ischemia and to compressive and
shearing forces. To minimize complications, the lowest pressure that prevents
capillary blood flow should be used, around 250 mm Hg in the lower extremities.
Tourniquet time should not exceed 2 hours. If longer use is necessary, the
tourniquet should be deflated for 10 minutes before reinflation.
Arterial complications are rare [91,92]. In 1987, Rush et al [93] reported 13
cases of arterial complications. The authors suggested that thrombosis was caused
by local pressure from the tourniquet and recommended not using a tourniquet in
the presence of proximal artery calcifications and poor peripheral pulses. If there
is any doubt, a vascular surgeon should be consulted preoperatively [93].
Among neurologic complications, peroneal nerve palsy remains infrequent but
worrisome. Mont et al [94] found a cumulative prevalence of 0.58% in the
literature. Recovery is usually excellent when the palsy is initially incomplete [95].
Finally, wound drainage and delayed wound healing are of concern because
they can mask a deep infection. Some authors advocate open débridement if
drainage continues after 5 days of appropriate therapy.

Spine
Spinal surgery is widely performed for a variety of indications including pain,
degenerative disease, deformity, and fracture. The great variations in the types of
42 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

procedures performed make comparison of outcomes difficult. Procedures can be

simple or involve fusion with or without instrumentation. The rates of fusion for
similar indications vary widely among centers and specialties [96,97]. The specific
approach to the spine can be anterior or posterior, and much debate remains con-
cerning the merits of each approach. Some patients undergo procedures involving
both approaches, either in the same anesthesia session (combined procedures) or a
few days apart but during the same hospitalization (staged procedure).

Intubation considerations
In patients undergoing cervical spine surgery, intubation often poses a
challenge. Those with degenerative disease of the spine usually have limited
extension of the neck. Patients with rheumatoid arthritis might have various
forms of atlantoaxial subluxation, temporomandibular ankylosis reducing the
mouth opening, and chronic cricoarytenoiditis that predisposes to acute airway
obstruction during intubation and extubation. Direct laryngoscopy should be
avoided with known posterior and vertical atlantoaxial subluxations, whereas
anterior, lateral, and rotatory subluxations can tolerate direct laryngoscopy after
stabilization of the neck [98].
Awake intubation of a sedated patient is preferred, because intubation of the
unconscious patient predisposes to greater risk of hypoxic injury. Nasotracheal
intubation is usually easier than oral intubation, because the nasopharynx,
oropharynx, and glottis are commonly in the same axis [99]. Nasotracheal
intubation, however, is associated with a high risk of sinusitis in patients
requiring prolonged periods of ventilatory support [100]. Fiberoptic intubation
is preferred to a blind approach: Analysis of 128 cases undergoing posterior
cervical spine fusion showed that the frequency of postoperative respiratory
complications and reintubations was significantly lower in patients intubated with
fiberoptic assistance [101].

Determinants of need for critical care

Most patients undergoing spinal surgery do not require intensive care. Harris
et al [102] attempted to identify clinical factors associated with unexpected
critical care management and prolonged hospitalization after elective cervical
spine surgery. They reviewed their center’s experience with 109 cases performed
between 1995 and 1999. Their data suggest that multilevel decompression,
preexisting myelopathy, pulmonary disease, cardiovascular disease, hypertension,
and diabetes are separately linked to the unanticipated need for critical care and
longer hospitalization. In a similar study, Nahtomi-Schick et al [103] attempted to
determine if intraoperative fluid management in spine surgery can predict ICU
LOS. They divided their patients into three groups: those who did not require
ICU care, those requiring 1 day of ICU care, and those requiring more than 1 day.
Significant variations in the type and complexity of the operations limited the
comparison between the study groups. Patients with longer ICU LOS, however,
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 43

required more transfusion with blood products and fluid replacement with
crystalloids intraoperatively. The authors concluded that ICU admission can be
predicted based on the type of surgery and amount of crystalloid administration.

Risk factors for complications

Complications of spine surgery can be linked to procedure- and patient-related
factors. In two separate, large cohort studies based on hospital discharge records
in the State of Washington, Deyo et al [104,105] demonstrated a significantly
higher complication rate when fusion was performed than for surgery without
fusion. This increased rate of complications resulted in 2.2 times greater nursing
home placement, 1.5 times higher hospital charges, and 2.0 times greater 6-week
mortality. Also, they found that operations for conditions other than herniated
disc were associated with greater use of resources and more complications,
particularly when arthrodesis was performed.
In a retrospective review, Fujita et al [106] compared the complication rates
between cervical and thoracolumbar spine fusion in patients older than 60 years.
They found that mean blood loss, operative time, and hospital LOS were greater
with thoracolumbar procedures than with cervical procedures. In a separate study,
McDonnell et al [107] reviewed the hospital records of 447 patients undergoing
anterior spinal fusion, excluding cervical operations. They found that a thoracic
approach takes less time to perform, and more blood is lost when a lumbar
approach is performed, but neither approach is significantly superior in terms of
complications. The same is true when thoracic and lumbar approaches are
compared with thoracolumbar approaches. Multivariate analysis also showed a
greater risk of major complications with a combined procedure than with staged
procedures and anterior procedures. The number of levels involved in the
procedure increases complication risks. More extensive procedures are associated
with higher blood volume loss, more tissue edema, and greater cardiovascular
stress. Data comparing anterior and posterior approaches to the spine are limited.
The most important patient-related factors are age and preexisting comorbid-
ities. In a review of 1223 anterior spinal fusions of the thoracic and lumbar spine
in adults, Faciszewski et al [108] found an increased risk of complications for
patients over age 60 years, women, and patients with multiple preexisting health
problems. Their finding that women are at an increased risk of complications has
not been duplicated in any subsequent study. In fact, larger cohort studies suggest
that women undergoing these operations tend to be older and more likely to
survive, although they require nursing home care more often [105]. The age
factor has been well established in many studies, as has the effect of the number
of comorbid conditions. Some studies suggest a linear correlation between the
number of comorbid conditions and complication rates, whereas others found an
increased risk only when two or more comorbidities were present. The most
frequently cited comorbid conditions include cardiovascular disease, pulmonary
disease, hypertension, and diabetes. Preexisting myelopathy in the operated
patient carries a higher risk of complications.
44 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

Major and minor complications

Major complications are more likely to be associated with longer hospital stay
and need for intensive care. These complications include pneumonia, respiratory
failure, prolonged postoperative intubation, cardiac arrest, shock, arrhythmia,
congestive heart failure, stroke, delirium, prolonged use or need for reinsertion of
chest tube, upper gastrointestinal bleeding, wound infection requiring débride-
ment, diffuse intravascular coagulopathy, thromboembolic disease, injury to great
vessels, monoplegia, and ureteral obstruction.
The syndrome of inappropriate secretion of antidiuretic hormone (SIADH)
reportedly occurs in 5% of patients undergoing spinal fusion. It is more likely
to occur in patients with spinal deformity and large intraoperative blood loss
[109]. Therapy includes fluid restriction while maintaining appropriate intra-
vascular status.
After spinal surgery, DVT and thromboembolic disease occur at a very low
rate, estimated at 0.3%, but can contribute disproportionately to mortality [110].
The risk increases with LOS and immobilization. No anticoagulation is required
perioperatively unless there is a history of thromboembolic disease. Appropriate
preventive measures include use of compressive stockings and sequential
compression devices. Other minor complications include transient hypoxia, ileus,
necrosis of edge of wound, need for multiple transfusions, dural tear, transient
confusion, dysethesia, atrial fibrillation, cystitis, urticaria, impotence, and ret-
rograde ejaculation.
Intraoperative complications are usually related to the specific approach and
the structures involved in the dissection field. These complications are often
addressed in the operating room. The most serious is a laceration of the great
vessels, such as carotid, aorta, or vena cava.
Intraoperative positioning is associated with specific complications. The
sitting position increases the risk of air embolism, hypotension, and spinal
hypoperfusion. The prone position may raise intraabdominal pressure, leading
to ventilatory impairment and increased bleeding from epidural veins, and
predispose to an epidural hematoma [99].

Infections
Patients undergoing spinal fusion are at particularly increased risk for
infection. Conventional lumbar discectomy with antibiotic prophylaxis has a
low infection risk, less than 1% [111]. Microdiscectomies may have a higher
infection rate [112,113]. Percutaneous lumbar discectomy might have a slightly
lower rate; however, the numbers reported are limited [114]. The incidence of
infection increases with fusion and instrumentation. Fusion has been demon-
strated to increase the infectious risk of discectomy from 1% to 3% to 6% to 8%
[115]. This increase is related to larger dissection fields, greater tissue damage,
longer operative time, and more significant disruption of tissue vascularity. The
addition of implants also raises infection rates proportionately to the increase in
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 45

operative time. These rates vary with specific operations and implants. Implants
rarely act as a source of infection but more often constitute a safe harbor for
bacteria. In general, cervical spine operations carry a slightly lower risk of
infection than lumbar or thoracic spine operations [116].
Specific patient-related risk factors have been determined as a cause of
increased infection risk. These risk factors are more easily addressed when the
operation is nonemergent. Diabetics have a higher incidence of postoperative
complications, two thirds of which are infectious in nature. In a study of 1548
patients in a surgical ICU, Van den Berghe et al [117] demonstrated that intensive
insulin therapy to maintain blood glucose at or below 110 mg/dL reduced the
mortality rate from 8.0% with conventional treatment to 4.6%. The greatest
reduction in mortality involved deaths from multiple-organ failure with a proven
septic focus. Intensive insulin therapy also reduced the overall in-hospital
mortality rate by 34% and bloodstream infections by 46%. Patients were less
likely to require prolonged mechanical ventilation and intensive care.
Other contributing factors are tobacco smoking [118], malnutrition, obesity,
and cachexia. In two separate reviews, the majority of patients with spine
infections were considered malnourished. Indicators include weight loss of
more than 10 pounds, serum albumin less than 3.4 g/dL, total lymphocyte
count less than 1500 cells/mL, skin test anergy, and arm circumference less
than 80% of normal [119]. Rheumatoid arthritis, chronic steroid therapy,
previous infections, coexisting infections (pneumonia, urinary colonization,
ulcers, and so forth), and neoplasm are also known to increase the rate of
infectious complications. Emerging data indicate that nasal carriage of Staphy-
lococcus aureus is a significant risk factor for septicemia, and that eradication
with topical antibiotics could reduce nosocomial pneumonia, wound infection,
and sepsis [120].
Longer perioperative hospitalization leads to increased rates of wound infec-
tion. The rate of infection doubles with each additional week of stay. Most patients
admitted to an ICU are colonized by hospital pathogens within 2 weeks of
admission. The problem is compounded by antibiotic administration that increases
rates of resistance. The use of surgical adjuncts such as methyl methacrylate, bone
substitutes, and hemostatic agents may contribute to infection rates.

Neurologic complications
Neurologic complications occur in 0.3% of patients undergoing anterior spine
decompression and fusion. These complications usually manifest as myelopathy,
radiculopathy, or myeloradiculopathy. Frequently, the offending process is
intrusion of graft or instrumentation material into the spinal canal. In kyphosis/
scoliosis cases, factors that increase the risk of postoperative neurologic compli-
cations include congenital deformities, large curves, kyphosis, postradiation
deformity, intraoperative correction exceeding the preoperative bending correc-
tion, passage of sublaminar wires in the thoracic and lumbar spine, and surgeon’s
lack of expertise [121].
46 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

In the recovery room and during the immediate postoperative period, checks
should be performed regularly at short intervals to detect neurologic deteriora-
tion. When such a finding is noticed, immediate investigation should be
performed to determine the cause and reversibility of the process. Initial steps
in management include cervical spine radiograms to check for proper vertebral
alignment and, when available, MR imaging to detect extrinsic spinal cord
compression by bone, intramedullary swelling, or hematoma. When a signifi-
cant clot is suspected, the patient should be returned immediately to the
operating room. When no compression is identified, immobilization measures
such as cervical collar should be used, and spinal cord perfusion should be
maintained by normotension and intravenous fluid administration. Based on
data from spinal cord injury, the use of intravenous methylprednisolone is
justified at a bolus dose of 30 mg/kg over 1 hour followed by 5.4 mg/kg over
the next 24 hours [122].
Delayed neurologic injury can occur. Such injury has been described in
patients with Down syndrome undergoing upper cervical spine fusion for
atlantoaxial instability [123]. Delayed neurologic injury has also been described
after instrumentation of narrow segments of the spine, presumably related to
edema compromising blood supply [124].

Graft- and instrument-related complications

Graft extrusion and instrumentation failure can occur in the perioperative
period. Posterior extrusion of the graft can result in myelopathy and needs
surgical intervention. Anterior extrusion is more likely to be clinically significant
at a cervical level because of the potential for tracheal obstruction, dysphagia,
neurologic injury, and kyphosis. The incidence of graft extrusion is between 1%
and 13%. Good surgical technique and anterior plating can prevent it.
Instrument failure occurs at a rate of 5% to 17%. Known risk factors include
age greater than 70 years, low bone mineral density, revision operations, and long
plates [121].

Site-specific complications
After cervical fusion, carotid injury, upper airway edema, recurrent laryngeal
nerve injury, and vertebral artery laceration can occur. Recurrent laryngeal nerve
injury is less likely with a left-sided approach. Horner-Bernard syndrome caused
by injury to the stellate ganglion has been described after high dissection
involving T1 and T2 levels [108]. Thoracic and thoracolumbar approaches
usually require chest tube placement. Pneumothorax, pleural effusion, and
persistent pain syndrome are common in this setting. Thoracolumbar and lumbar
operations can cause abdominal hernia, retroperitoneal lymphocele, ureteral
laceration, lumbar plexus injury, postsympathectomy neuralgia, and femoral
nerve palsy.
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 47

Donor site-related complications

When autogenous bone grafting is performed, the donor site should be
evaluated for signs of infection, bleeding, and neurologic injury, particularly
when the iliac crest is harvested. Arrington et al [125] reported an incidence of
5.8% of major complications and 10% of minor complications at the donor site.
Major complications included vascular injuries, deep hematomas, neurologic
deficits, deep infection, iliac wing fracture, and abdominal herniation. Fracture of
the iliac crest occurs more often in women and is usually treated with
conservative non –weight-bearing measures. Pelvic instability can also occur
after posterior iliac crest harvest and is thought to be caused by injury to the
superoposterior sacroiliac ligaments. Multiparous women might be at an
increased risk of this complication.

Special considerations in scoliosis surgery

Scoliosis is a progressive disease in which severity is related to the angle of the
curvature. A curvature less than 60° is unlikely to be associated with pulmonary
dysfunction. The earliest manifestations consist of a restrictive ventilatory pattern
with reduction in the vital capacity. Gas exchange is also affected by ventilation/
perfusion mismatch, alveolar hypoventilation, increased dead space ventilation,
and widened A-a gradient. This impairment of gas exchange results in pulmonary
hypertension and hypercapnea, eventually leading to respiratory failure. Patients
with scoliosis have respiratory muscle weakness leading to inadequate cough,
limited ability to handle secretions, and greater predisposition to infection.
Surgery is usually performed to limit the progression of the disease and prevent
complications. There is some debate about the long-term effect of surgery on lung
function and gas exchange. There is evidence, however, that lung function might
deteriorate acutely during surgery and the postoperative period.
Complications can arise after rapid correction of extreme deformities because
of interference with the circulation to the spine leading to anterior spinal artery
syndrome. When a thoracotomy is performed, a left-sided approach is preferred,
because it is easier to handle the thick-walled aorta than the relatively vulnerable
vena cava on the right. Deliberate hypotension is a technique used during
scoliosis surgery to reduce blood loss; it is associated with greater risk of spinal
cord ischemia. Agents used to achieve hypotension include nitroprusside, which
is associated with methemoglobenemia. Anesthesia in scoliosis patients could
cause hyperkalemia, cardiac arrhythmia, and malignant hypertension [126,127].
The intensivist should be alert to these complications.
Preoperative respiratory impairment predicts the need for postoperative me-
chanical ventilation. Some patients might require a tracheostomy. Patients in whom
a thoracotomy or a thoracoabdominal approach is used have a significantly higher
number of pulmonary complications. Scoliosis patients are quite sensitive
to respiratory suppressants, and care should be taken to avoid heavy seda-
tion perioperatively.
48 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

Another common complication of scoliosis surgery is the development of a

paralytic ileus. Therefore a nasogastric tube is usually inserted and left in place
until bowel function is recovered.

Other aspects of postoperative care

Some medications should be avoided in the perioperative periods including
NSAIDs, especially ketorolac, because it has been shown to inhibit and delay
bone formation, resulting in a fivefold increase in the rate of pseudoarthrosis in
instrumented spinal fusion [128]. Also, many agents used for the treatment of
rheumatoid arthritis should be avoided. Methotrexate is known to inhibit growth
of rapidly regenerating tissue by inhibiting DNA synthesis, and thus its use
should be discontinued in the perioperative period [129].

Summary
Complications of orthopedic and spine operations can be life threatening.
Proper patient selection, careful planning of patient care, and prophylactic
measures are important determinants of a successful outcome. After elective
orthopedic surgery such as total joint replacement, the intensivist should be aware
of potential systemic complications common to any major surgical intervention
(pneumonia, pulmonary embolism, sepsis, myocardial infarction) and also of
procedure-specific problems (cement-related cardiac events, fat embolism) and
local complications (neurovascular injuries). Patients undergoing spine proce-
dures should have close neurologic monitoring for immediate and delayed deficits.

Acknowledgment
The authors are indebted to Dr. Diane Maggiore for her review of the manuscript.

References
[1] Taylor VM, Deyo RA, Cherkin DC, et al. Low back pain hospitalization. Recent United States
trends and regional variations. Spine 1994;19(11):1207 – 12.
[2] Praemer A, Furner S, Rice DP. Medical implants and major joint procedures. In: Musculoske-
letal conditions in the United States. Park Ridge (IL): American Academy of Orthopaedic
Surgeons; 1992. p. 611 – 26.
[3] NIH Consensus Conference. Total hip-joint replacement in the United States. JAMA
1982;248(15):1817 – 21.
[4] Whittle J, Steinberg EP, Anderson GF, et al. Mortality after elective total hip arthroplasty in
elderly Americans. Clin Orthop 1993;295:119 – 26.
[5] Stein M, Koota GM, Simon M, et al. Pulmonary evaluation in surgical patients. JAMA
1962;181:765 – 70.
[6] Goldman L. Cardiac risks and complications of noncardiac surgery. Ann Intern Med 1983;
98(4):504 – 13.
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 49

[7] Weitz HH, Goldman L. Noncardiac surgery in the patient with heart disease. Med Clin North
Am 1987;71(3):413 – 32.
[8] Morand EF, Littlejohn GO. Medical problems in joint replacement patients. A retrospective
study of 243 total hip arthroplasties. Med J Aust 1990;152(8):408 – 13.
[9] Latimer RG, Dickman M, Day WC, et al. Ventilatory patterns and pulmonary complications
after upper abdominal surgery determined by preoperative and postoperative computerized
spirometry and blood gas analysis. Am J Surg 1971;122(5):622 – 32.
[10] Hole A, Terjesen T, Breivik H. Epidural versus general anesthesia for total hip arthroplasty in
elderly patients. Acta Anaesthsiol Scand 1980;24(4):279 – 87.
[11] Davis FM, Laurenson VG, Gillispie WJ, et al. Deep vein thrombosis after total hip replacement. A
comparison between spinal and general anesthesia. J Bone Joint Surg Br 1989;71(2):181 – 5.
[12] Lieberman JR, Huo MM, Hanway J, et al. The prevalence of deep venous thrombosis with
hypotensive epidural anesthesia. J Bone Joint Surg Am 1994;76(3):341 – 8.
[13] Sculco TP, Ranawat C. The use of spinal anesthesia for total hip replacement arthroplasty.
J Bone Joint Surg Am 1975;57(2):173 – 7.
[14] Sharrock NE, Mineo R, Urquhart B. Haemodynamic effects and outcome analysis of hypo-
tensive extradural anesthesia in controlled hypertensive patients undergoing total hip arthro-
plasty. Br J Anaesth 1991;67(1):17 – 25.
[15] Daniel WW, Coventry MB, Miller WE. Pulmonary complications following total hip arthro-
plasty with Charnley prosthesis as revealed by chest radiographs. J Bone Joint Surg Am
1972;54(2):282 – 3.
[16] Eftekhar NS, Kiernan HA, Stinchfield FE. Systemic and local complications following low
friction arthroplasty of the hip joint: a study of 800 consecutive cases. Arch Surg 1976;111(2):
150 – 5.
[17] Shaw JA, Chung R. Febrile response after knee and hip arthroplasty. Clin Orthop 1999;
367:181 – 9.
[18] Sheppeard H, Cleak DK, Ward DF, et al. A review of early mortality and morbidity in elderly
patients following Charnley total hip replacement. Arch Orthop Trauma Surg 1980;97(4):
243 – 8.
[19] Wellman JJ, Smith BA. Adult respiratory distress syndrome in the surgical patient. In: Lubin
MF, Walker HK, Smith RB, editors. Medical management of the surgical patient. Boston:
Butterworth; 1988. p. 175 – 86.
[20] The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as
compared with traditional tidal volumes for acute lung injury and the acute respiratory distress
syndrome. N Engl J Med 2000;342(18):1301 – 8.
[21] Bell RC, Coalson JJ, Smith JD, et al. Multiple organ system failure and infection in the adult
respiratory distress syndrome. Ann Intern Med 1983;99(3):293 – 8.
[22] Ling RSM. Systemic and miscellaneous complications. In: Ling RSM, editor. Complications of
total hip replacement. Edinburgh: Churchill Livingstone; 1984. p. 201 – 12.
[23] Adams III JE, Sicard GA, Allen BT, et al. Diagnosis of perioperative myocardial infarction with
measurement of cardiac troponin I. N Engl J Med 1994;330(10):670 – 4.
[24] Ashton CM. Perioperative myocardial infarction with non-cardiac surgery. Am J Med Sci
1994;308(1):41 – 8.
[25] Wukich DK, Callaghan JJ, Graeber GM, et al. Cardiac isoenzyme values after total joint
arthroplasty. Clin Orthop 1989;242:232 – 40.
[26] Kahn RL, Hargelt MG, Urquhart B, et al. Supraventricular tachyarrhythmias during total joint
arthroplasty: incidence and risk. Clin Orthop 1993;296:265 – 9.
[27] Geers WH, Code KI, Jay RM, et al. A prospective study of venous thromboembolism after
major trauma. N Engl J Med 1994;331(24):1601 – 6.
[28] Crandon AJ, Peel VR, Anderson JA, et al. Postoperative deep vein thrombosis. Identifying high
risk patients. BMJ 1980;281(6236):343 – 4.
[29] Sikorski JM, Hampson WG, Staddon GE. The natural history and aetiology of deep vein
thrombosis after total hip replacement. J Bone Joint Surg Br 1981;63-B(2):171 – 7.
50 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

[30] DeSanctis RW, Harris WH, Salzman EW, et al. Prevention of venous thromboembolism follow-
ing total hip replacement. JAMA 1972;220(10):1319 – 22.
[31] Rosenow III EC, Osmundson PJ, Brown ML. Pulmonary embolism. Mayo Clin Proc 1981;
56(3):161 – 78.
[32] Pierson JL, Tavel ME. Thromboembolic prophylaxis in total joint replacement. Chest 2001;
120(1):302 – 4.
[33] Haake D, Berkman S. Venous thromboembolic disease after hip surgery. Clin Orthop 1989;
242:212 – 31.
[34] Dalen JE, Hirsh J, Guyatt GH, et al. Sixth ACCP Consensus Conference on Antithrombotic
Therapy. Chest 2001;119(Suppl):140 – 75.
[35] Lensing AW, Prandoni P, Brandjes D, et al. Detection of deep vein thrombosis by real time
B-mode ultrasonography. N Engl J Med 1989;320(6):342 – 5.
[36] King MB, Harmon KR. Unusual forms of pulmonary embolism. Clin Chest Med 1994;15(3):
561 – 80.
[37] Rathbun SW, Raskob GE, Whitsett TL. Sensitivity and specificity of helical computed tomog-
raphy in the diagnosis of pulmonary embolism: a systematic review. Ann Intern Med 2000;
132(3):227 – 32.
[38] Stein PD, Athanasoulis C, Alavi A, et al. Complication and validity of pulmonary angiography
in acute pulmonary embolism. Circulation 1992;85(2):462 – 8.
[39] Ginsberg JS. Management of venous thromboembolism. N Engl J Med 1996;335(24):
1816 – 28.
[40] Arcasoy SM, Kreit JW. Thrombolytic therapy of pulmonary embolism: a comprehensive review
of current evidence. Chest 1999;115(6):1695 – 707.
[41] Kanter B, Moser KM. The Greenfield vena cava filter. Chest 1988;93(1):170 – 5.
[42] Heine TA, Halambeck BL, Mark JB. Fatal pulmonary fat embolism in the early postoperative
period. Anesthesiology 1998;89(6):1589 – 91.
[43] Peltier LF, Collins JA, Evarts CM, et al. A panel by correspondence: fat embolism. Arch Surg
1974;109(1):12 – 6.
[44] Woo R, Minster GJ, Fitzgerald Jr RH, et al. Pulmonary fat embolism in revision hip arthro-
plasty. Clin Orthop 1995;319:41 – 53.
[45] Fabian TC. Unraveling the fat embolism syndrome. N Engl J Med 1993;329(13):961 – 3.
[46] Palmovic V, McCarroll JR. Fat embolism in trauma. Arch Pathol 1965;80(6):630 – 5.
[47] Duis HJ, Nijsten MWN, Klasen JH, et al. Fat embolism in patients with an isolated fracture of
the femoral shaft. J Trauma 1988;28(3):383 – 90.
[48] Fabian TC, Hoots AV, Stanford DS, et al. Fat embolism syndrome: prospective evaluation in 92
fracture patients. Crit Care Med 1990;18(1):42 – 6.
[49] Hulman G. Pathogenesis of non-traumatic fat embolism. Lancet 1988;1(8599):1366 – 7.
[50] Gresham GA, Kuczynski A, Rosborough D. Fatal fat embolism following replacement arthro-
plasty for transcervical fractures of femur. BMJ 1971;2(762):617 – 9.
[51] Sulek CA, Davies LK, Enneking FK, et al. Cerebral microembolism diagnosed by transcranial
Doppler during total knee arthroplasty: correlation with transesophageal echocardiography.
Anesthesiology 1999;91(3):672 – 6.
[52] Pell AC, Hughes D, Keating J, et al. Brief report: fulminating fat embolism syndrome
caused by paradoxical embolism through a patent foramen ovale. N Engl J Med 1993;
329(13):926 – 9.
[53] Baker PL, Pazell JA, Peltier LF. Free fatty acids, catecholamines, and arterial hypoxia in
patients with fat embolism. J Trauma 1971;11(12):1026 – 30.
[54] Christie J, Robinson CM, Pell ACH, et al. Transcardiac echocardiography during invasive
intramedullary procedures. J Bone Joint Surg Br 1995;77(3):450 – 5.
[55] Sherman RM, Byrick RJ, Kay JC, et al. The role of lavage in preventing hemodynamic and
blood-gas changes during cemented arthroplasty. J Bone Joint Surg Am 1983;65(4):500 – 6.
[56] Gurd AR. Fat embolism: an aid to diagnosis. J Bone Joint Surg Br 1970;52(4):732 – 7.
[57] Johnson MJ, Lucas GL. Fat embolism syndrome. Orthopedics 1996;19(1):41 – 8.
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 51

[58] Spengler DM, Costenbader M, Bailey R. Fat embolism syndrome following total hip arthro-
plasty. Clin Orthop 1976;121:105 – 7.
[59] Jacobson DM, Terrence CF, Reinmuth OM. The neurologic manifestations of fat embolism.
Neurology 1986;36(6):847 – 51.
[60] Shaw M, Mandell BF. Perioperative management of selected problems in patients with rheu-
matic diseases. Rheum Dis Clin North Am 1999;25(3):623 – 38.
[61] Mohler CG, Collis DK. Early complications and their management. In: Callaghan JJ, Rosen-
berg AG, Rubash HE, editors. The adult hip. Philadelphia: Lippincott-Raven Publishers; 1998.
p. 1125 – 47.
[62] Bengtson A, Larsson M, Gammer W, et al. Anaphylatoxin release in association with methyl-
methacrylate fixation of hip prostheses. J Bone Joint Surg Am 1987;69(1):46 – 9.
[63] Mir GN, Lawrence WH, Autian J. Toxicological and pharmacological actions of meth-
acrylate monomers. I. Effects on isolated, perfused rabbit heart. J Pharm Sci 1975;62(5):
778 – 82.
[64] Byrick RJ, Kay JC, Mullen JBM. Pulmonary marrow embolism: a dog model simulating dual
component cemented arthroplasty. Can J Anesth 1987;34(4):336 – 42.
[65] Lafont ND, Kostucki WM, Marchand MN, et al. Embolism detected by transesophageal echo-
cardiography during hip arthroplasty. Can J Anaesth 1994;41(9):850 – 3.
[66] Patterson BM, Lieberman JR, Salvati EA. Intraoperative complications during total hip arthro-
plasty. Orthopedics 1995;18(11):1089 – 95.
[67] Ries MD, Lynch F, Rauscher LA, et al. Pulmonary function during and after total hip replace-
ment. Findings in patients who have insertion of a femoral component with and without cement.
J Bone Joint Surg Am 1993;75(4):581 – 7.
[68] Kallos T. Impaired arterial oxygenation associated with use of bone cement in the femoral shaft.
Anesthesiology 1975;42(2):210 – 5.
[69] Duncal JAT. Intraoperative collapse or death related to use of acrylic bone cement in hip
surgery. Anesthesia 1989;44(2):149 – 53.
[70] Siopack JS, Jergesen HE. Total hip arthroplasty. West J Med 1995;162(3):243 – 9.
[71] Nasser S. Prevention and treatment of sepsis in total hip replacement surgery. Orthop Clin
North Am 1992;23(2):265 – 77.
[72] Nelson JP, Glassburn AR, Talbott RD, et al. The effect of previous surgery, operating room
environment and preventive antibiotics on postoperative infection following total hip arthro-
plasty. Clin Orthop 1980;147:167 – 9.
[73] Tsukyama DT, Estrada R, Gustilo RB. Infection after total hip arthroplasty. A study of the
treatment of one hundred and six infections. J Bone Joint Surg Am 1996;78(4):512 – 23.
[74] Nachbur B, Meyer RP, Verkkala K, et al. The mechanisms of severe arterial injury in surgery of
the hip joint. Clin Orthop 1979;141:122 – 33.
[75] Reiley MA, Bond D, Branick RJ, et al. Vascular complications following total hip arthroplasty.
A review of the literature and a report of two cases. Clin Orthop 1984;186:23 – 8.
[76] Edwards BN, Tullos HS, Noble PC. Contributory factors and etiology of sciatic nerve palsy in
total hip arthroplasty. Clin Orthop 1987;218:136 – 41.
[77] Solheim LF, Hagen R. Femoral and sciatic neuropathies after total hip arthroplasty. Acta Orthop
Scand 1980;51(3):531 – 4.
[78] Soll A. Gastric, duodenal, and stress ulcer. In: Sleisenger MH, Fordtran JS, editors. Gastro-
intestinal disease. 5th edition. Philadelphia: WB Sanders; 1993. p. 580 – 652.
[79] Chung HM, Kluge R, Schrier RW, et al. Postoperative hyponatremia. A prospective study. Arch
Intern Med 1986;146(2):333 – 6.
[80] Hornick P, Allan P. Acute hyponatremia following total hip replacement. Br J Clin Pract
1990;44(12):776 – 7.
[81] Redfern TR, Machin DG, Parsons KF, et al. Urinary retention in men after total hip arthroplasty.
J Bone Joint Surg Am 1986;68(9):1435 – 8.
[82] Wroblewski BM, del Sel HJ. Urethral instrumentation and deep sepsis in total hip replacement.
Clin Orthop 1980;146:209 – 13.
52 D. Nazon et al / Crit Care Clin 19 (2003) 33–53

[83] Praemer A, Furner S, Rice DP. Medical implants and major joint procedures. In: Musculo-
skeletal conditions in the United States. Park Ridge (IL): American Academy of Orthopaedic
Surgeons; 1992. p. 125 – 42.
[84] Lotke PA. Thrombophlebitis in knee arthroplasty. In: Scott WN, editor. The knee. St Louis: CV
Mosby; 1994. p. 1217 – 25.
[85] Parmet JL, Horrow JC, Pharo G, et al. The incidence of venous emboli during extramedullary
guided total knee arthroplasty. Anesth Analg 1995;81(4):757 – 62.
[86] Haas SB, Insall JN, Scuderi GR, et al. Pneumatic sequential compression boots compared with
aspirin prophylaxis of deep vein thrombosis after total knee arthroplasty. J Bone Joint Surg Am
1990;72(1):27 – 31.
[87] Monto RR, Garcia J, Callaghan JJ. Fatal fat embolism following total condylar knee arthro-
plasty. J Arthroplasty 1990;5(4):291 – 9.
[88] Berman AT, Parmet JL, Harding SP, et al. Emboli observed with the use of transeosophageal
echocardiography immediately after tourniquet release during total knee arthroplasty with
cement. J Bone Joint Surg 1998;80(3):389 – 96.
[89] Parmet JL, Horrow JC, Singer R, et al. Echogenic emboli upon tourniquet release during total
knee arthroplasty: pulmonary hemodynamic changes and embolic composition. Anesth Analg
1994;79(5):940 – 5.
[90] Orsini EC, Richards RR, Mullen JM. Fatal fat embolism during cemented total knee arthro-
plasty: a case report. Can J Surg 1986;29(5):385 – 6.
[91] Kumar SN, Chapman JA, Rawling I. Vascular injuries in total knee arthroplasty. A review of
the problem with special reference to the possible effects of the tourniquet. J Arthroplasty 1998;
13(2):211 – 6.
[92] Rand JA. Vascular complications of total knee arthroplasty: report of three cases. J Arthroplasty
1987;2(2):89 – 93.
[93] Rush JH, Vidovich JD, Johnson MA. Arterial complications of total knee replacement. The
Australian experience. J Bone Joint Surg Br 1987;69(3):400 – 2.
[94] Mont MA, Dellon AL, Chen F, et al. The operative treatment of peroneal nerve palsy. J Bone
Joint Surg Am 1996;78(6):863 – 9.
[95] Asp JPL, Rand JA. Peroneal nerve palsy after total knee arthroplasty. Clin Orthop 1990;
261:233 – 7.
[96] Deyo RA, Cherkin D, Conrad D, et al. Cost, controversy, crisis: low back pain and the health of
the public. Annu Rev Pub Health 1991;12:141 – 56.
[97] Dunsker SB. Lumbar spine stabilization. Indications. Clin Neurosurg 1990;36:147 – 58.
[98] Macarthur A, Kleiman S. Rheumatoid cervical joint disease – a challenge to the anaesthetist.
Can J Anaesth 1993;40(2):154 – 9.
[99] Concepcion M. Anesthesia for orthopedic surgery. In: Longnecker DE, Tinker JH, Morgan Jr
GE, editors. Principles and practice of anesthesiology. 2nd edition. St. Louis: Mosby; 1998. p.
2113 – 37.
[100] Holzapfel L, Chastang C, Demingeon G, et al. A randomized study assessing the systematic
search for maxillary sinusitis in nasotracheally mechanically ventilated patients. Influence of
nosocomial maxillary sinusitis on the occurrence of ventilator-associated pneumonia. Am J
Respir Crit Care Med 1999;159(3):693 – 4.
[101] Wattenmaker I, Concepcion M, Hibberd P, et al. Upper airway obstruction and perioperative
management of the airway in patients managed with posterior operations on the cervical spine
for rheumatoid arthritis. J Bone Joint Surg Am 1994;76(3):360 – 5.
[102] Harris OA, Runnels JB, Matz PG. Clinical factors associated with unexpected critical care
management and prolonged hospitalization after elective cervical spine surgery. Crit Care Med
2001;29(10):898 – 902.
[103] Nahtomi-Shick O, Kostuik JP, Winters BD, et al. Does intraoperative fluid management in
spine surgery predict intensive care length of stay? J Clin Anesth 2001;13(3):208 – 12.
[104] Deyo RA, Cherkin DC, Loeser JD, et al. Morbidity and mortality in association with operations
on the lumbar spine. J Bone Joint Surg Am 1992;74A(4):536 – 43.
 D. Nazon et al / Crit Care Clin 19 (2003) 33–53 53

[105] Deyo RA, Ciol MA, Cherkin DC, et al. A cohort study of complications, reoperations, and
resource use in the Medicare population. Spine 1993;18(11):1463 – 70.
[106] Fujita T, Kostuik JP, Huckell CB, et al. Complications of spinal fusion in adult patients more
than 60 years of age. Orthop Clin North Am 1998;29(4):669 – 78.
[107] McDonnell MF, Glassman SD, Dimar Jr II, et al. Perioperative complications of anterior
procedures on the spine. J Bone Joint Surg Am 1996;78(6):839 – 47.
[108] Faciszewski T, Winter RB, Lonstein JE, et al. The surgical and medical perioperative compli-
cations of anterior spinal fusion surgery in the thoracic and lumbar spine in adults. A review of
1223 procedures. Spine 1995;20(14):1592 – 9.
[109] Elster AD. Hyponatremia after spinal fusion caused by inappropriate secretion of antidiuretic
hormone (SIADH). Clin Orthop 1985;194:136 – 41.
[110] Rokito SE, Schwartz MC, Neuwirth MG. Deep vein thrombosis after major reconstructive
spinal surgery. Spine 1996;21(7):853 – 8.
[111] Ford LT. Postoperative infection of lumbar intervertebral disc space. South Med J 1976;
69(11):1477 – 81.
[112] Leung PC. Complications in the first 40 cases of microdiscectomy. J Spinal Dis 1988;1(4):
306 – 10.
[113] Stolke D, Solimann WP, Seifert V. Intra and post operative complications in lumbar disc
surgery. Spine 1989;14(1):56 – 9.
[114] Bonaldi G, Belloni G, Prosetti D, et al. Percutaneous discectomy using Onik’s method: 3 years
experience. Neuroradiology 1991;33(6):516 – 9.
[115] Horwitz NH, Curtin JA. Prophylactic antibiotics and wound infections following laminectomy
for lumbar disc herniation: a retrospective study. J Neurosurg 1975;43(6):727 – 31.
[116] Bertalanffy H, Eggert HR. Complications of anterior cervical discectomy without fusion in 450
consecutive patients. Acta Neurochir 1989;99(1 – 2):41 – 50.
[117] Van Den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill
patients. N Engl J Med 2001;345(19):1359 – 67.
[118] Silcox III DH, Daftari T, Boden SD, et al. The effect of nicotine on spinal fusion. Spine
1995;20(14):1549 – 53.
[119] Jensen TE, Jensen TG, Smith TK, et al. Nutrition in orthopaedic surgery. J Bone Joint Surg Am
1982;64(9):1263 – 72.
[120] Campbell W, Hendrix E, Schwalbe R, et al. Head-injured patients who are nasal carriers of
Staphylococcus aureus are at a high risk for Staphylococcus aureus pneumonia. Crit Care Med
1999;27(4):798 – 801.
[121] Brown CA, Eismont FJ. Complications in spinal fusion. Orthop Clin North Am 1998;29(4):
679 – 99.
[122] Slacman M. Complications of cervical spine surgery. Crit Care Med 2001;29(10):2027 – 8.
[123] Segal LS, Drummond DS, Zanotti RM. Complications of posterior arthrodesis of the cervical
spine in patients who have Down syndrome. J Bone Joint Surg Am 1991;73(10):1547 – 54.
[124] Johnston II CE, Happel LT, Norris R, et al. Delayed paraplegia complicating sublaminar
segmental spinal instrumentation. J Bone Joint Surg Am 1986;68(4):556 – 63.
[125] Arrington ED, Smith WJ, Chambers HG, et al. Complications of iliac crest bone graft harvest-
ing. Clin Orthop 1996;329:300 – 9.
[126] Britt BA, Kalow W. Malignant hyperthermia: a statistical review. Can Anaesth Soc J 1970;
17(4):293 – 315.
[127] Relton JE, Creighton RE, Johnston AE, et al. Hyperpyrexia in association with general anaes-
thesia in children. Can Anaesth Soc J 1966;13(5):419 – 24.
[128] Glassman SD, Rose SM, Dimar JR. The role of post-operative nonsteroidal anti-inflammatory
administration on spinal fusion. Spine 1998;23(7):834 – 8.
[129] Brighton CT, Fisher Jr JR, Levine SE, et al. The biochemical pathways mediating the prolif-
erative response of bone cells to chemical stimuli. J Bone Joint Surg Am 1996;78(9):1337 – 47.