Cardiac Management in the ICU*
James G. Ramsay, MD
* From the Department of Anesthesiology, Emory University School of Medicine, Atlanta, GA.
Correspondence to: James G. Ramsay, MD, Department of Anesthesiology, Emory University Hospital, 1364 Clifton Rd, Atlanta, GA 30322; e-mail: james_ramsay{at}emory.org
 |
Abstract
|
|---|
Coronary artery disease (CAD) is common in the surgical population,
with up to 50% of postoperative deaths due to cardiac events. Most of
these events are ischemic, with some being exacerbations of underlying
congestive heart failure (CHF). Recent data indicate that acute
perioperative β-adrenergic blockade can reduce ischemia and ischemic
events. Postoperative monitoring should focus on myocardial ischemia,
with preparation for rapid treatment using IV therapy. A few studies
suggest that elderly patients with known CAD undergoing major
procedures might benefit from perioperative treatment guided by
information from a pulmonary artery catheter. Postoperative CHF, which
is likely to present early after surgery, may need aggressive
management with diuretics, vasodilators, and inotropic drugs.
Mechanical ventilation should be considered. When the patient develops
severe or refractory dysrhythmias, serum magnesium levels should be
supplemented and consideration given to IV use of amiodarone.
Postoperative hypertension is common and can precipitate ischemia, CHF,
and arrhythmias as well as cause bleeding. Newer IV drugs are arterial
specific and can lower BP in a smooth and predictable manner. All acute
cardiac disorders can be precipitated or exacerbated by inadequate pain
control, hypoxemia, and fluid or electrolyte
disorders.
|
Introduction
|
|---|
It
has been estimated that more than one half of postoperative deaths are
caused by cardiac events.1 Not surprisingly, a large
percentage of elective postoperative admissions to the ICU are for
"cardiac monitoring," and many unplanned ICU admissions are for
treatment of acute cardiovascular changes that occur intraoperatively.
For preexisting cardiac conditions, one of the most valuable aspects of
ICU admission is the ability to reinstate preoperative treatment
regimens in a timely fashion. Acute disorders are managed with IV
medications while the patient is intensively monitored.
Cardiac management is closely related to other aspects of postoperative
care in the ICU. Fluid and electrolyte status and oxygen content of
arterial blood, for example, are vitally important in maintaining
adequate cardiac function. Activation of the sympathoadrenal system by
pain and perturbations in other organ systems place major stress on the
heart. Therefore, assessment of cardiac function must be closely
coupled with a complete physical examination, history of prior status
and intraoperative events, and general assessment of blood chemistry,
oxygenation, and analgesic regimen. When an ICU admission is the result
of intraoperative events, direct communication between the
anesthesiologist and the ICU caregiver is essential. The following
discussion focuses on ischemic heart disease, with a brief
consideration of heart failure, dysrhythmias, and hypertension.
|
Coronary Artery Disease
|
|---|
In patients with known or suspected coronary artery disease (CAD),
perioperative ischemia is common and is most prevalent in the
postoperative period (Fig 1).2
This ischemia is associated with adverse cardiac outcomes during the
hospital admission and for up to 2 years later.34 Many
studies indicate that myocardial infarction (MI) occurs most frequently
in the first few days after noncardiac surgery; a recent study
documented the incidence of MI in patients with known CAD was 5.6%,
with almost half occurring on the day of surgery.5 In
vascular patients, the incidence of CAD may be as high as 70%, with
perioperative MI occurring at three times this rate.1
View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1.. Pattern of postoperative ST-segment depression in
patients undergoing noncardiac operations. For each postoperative day
(POD), the total minutes of ST-segment depression were summed and then
divided by the total duration of monitoring for all patients. The data
were further divided into three groups: episode durations with maximal
change from baseline –1 to –1.9 mm, –2 to –2.9 mm, and  3 mm.
Reprinted with permission.2
|
|
Diagnosis
Postoperative ischemia has been studied by the use of continuous
ECG monitoring. Modern ICU monitors have the capacity for continuous ST
segment trending, and at least one manufacturer of ambulatory ECG
monitors has incorporated an alarm triggered by ST segment changes.
These devices have not been validated for realtime use in ischemia
detection; clinicians usually rely on patients to complain of chest
discomfort or on "routine" sequential ECG recordings and enzyme
levels. In the study referred to above,5 only 17% of the
patients who developed an MI complained of chest pain; however, 56%
developed significant clinical findings (dysrhythmia, hypotension,
pulmonary edema). Frequent cardiac enzyme determinations (creatine
phosphokinase, myocardial band [CK-MB]) are costly and may not add to
the information gained from the 12-lead ECG in postoperative patients
with clinical findings.6 Troponin-t and troponin-i levels
appear to be more cardiac specific, peak at a similar time as CK-MB,
and stay elevated for days.7 Studies are currently
underway evaluating continuous 12-lead ECG monitoring in postoperative
patients, with the possibility that this might be done in a telemetry
or surgical floor setting.
Prevention
Prevention of myocardial ischemia is the main goal of ICU
management in this population. Recent reports of reduction in
postoperative ischemia and MI with the use of prophylactic β-blockade
(using atenolol) are compelling.34 These studies extend
to the surgical period the recognized benefit of β-blockade in
prevention of ischemia and reinfarction in nonsurgical
patients.89 Oral or IV β-blockade can be initiated the
morning of surgery and continued into the postoperative period. Other
therapy used in the treatment of ischemic heart disease either has not
shown benefit consistently (eg, prophylactic infusion of IV
nitroglycerin) or has not been adequately evaluated in the
perioperative setting (calcium blockers).
Another class of drugs that holds promise for the prevention of
perioperative ischemia is the 
2-adrenoreceptor agonists.
These drugs act centrally to reduce sympathetic nervous system output,
and are associated with mild sedation and reductions in heart rate and
BP. A single oral dose of clonidine preoperatively reduced the
incidence of perioperative ischemia in vascular
patients10; the newer drug mivazerol was found to
significantly reduce myocardial ischemia when given as an infusion for
72 h perioperatively.11
Reduction of the sympathoadrenal response to surgery by intraoperative
epidural anesthesia followed by postoperative epidural analgesia has
been associated with a reduction in perioperative cardiac morbidity and
mortality in some, but not all studies.1213 No study has
found a detrimental effect, and epidural analgesia is generally
superior to other methods of pain relief, leading to the conclusion
that wherever possible, this technique should be considered. In
addition to analgesia, two general measures that may have an impact on
the incidence of myocardial ischemia are body temperature and
hematocrit. Frank et al14 found that unintentional
hypothermia was associated with myocardial ischemia, and in patients
with CAD, maintenance of hematocrit that balances oxygen content with
optimal rheologic properties of blood (eg, 30%) is
appropriate.
Rao et al15 found that in patients with a prior MI,
postoperative management guided by the pulmonary artery catheter (PAC)
and aggressive treatment of hemodynamic abnormalities reduced the
incidence of MI and death compared with a retrospective cohort of
patients. One small prospective study in vascular patients
supports the concept of preoperative "optimization" of hemodynamics
according to measurements from the PAC, followed by intraoperative and
postoperative use of similar therapy.16 While a recent
well-publicized study failed to show a benefit with PAC use in the
medical setting, these findings may not apply to the surgical
population.17 Experienced clinicians can be incorrect in
estimating cardiac filling pressures and output from examination
alone18; therefore, the PAC may be useful, possibly in
conjunction with or as an alternative to echocardiography, as a
diagnostic tool. Another appropriate use is in the titration of
inotropic and vasodilator drugs (see below). While the pulmonary artery
pressures will often rise with acute ischemia, intraoperative use of
the PAC for detection of ischemia is both insensitive and
nonspecific.19 At the present time, the data supporting
PAC use in surgical populations are soft, and "routine" use is
probably not justifiable, even for cardiac procedures. Placement of a
PAC is most appropriate when there is a diagnostic question that cannot
be answered by echocardiography, or where ongoing treatment of
ventricular dysfunction can best be guided by information from the
catheter.
Treatment
When myocardial ischemia or infarction is suspected in the
postoperative period, short-term therapy is similar to that in the
medical setting, with the exception of heparinization or thrombolytic
agents. While the latter will be contraindicated, depending on the
nature of the surgical procedure and the time between surgery and the
ischemia, cautious heparinization may be possible. Short-term therapy
with IV agents to control abnormal hemodynamics (especially
tachycardia) should be initiated, along with specific anti-ischemic
therapy such as IV nitroglycerin. Surgical pain, anxiety, and
fluid/hemoglobin deficits must be treated. Table 1
lists some therapies useful in this setting. Intervention in the
cardiac catheterization laboratory should be considered, as should use
of an intra-aortic balloon pump, in cases of severe refractory
ischemia.
|
Congestive Heart Failure
|
|---|
It has long been recognized that congestive heart failure (CHF) is
a major risk factor in the development of perioperative cardiac
morbidity and mortality20; one clinical series suggested
that 95% of perioperative acute CHF occurs within 1 h of the end
of surgery.21 In patients with little or no cardiac
reserve, or where there is a new cardiac event, activation of the
sympathetic nervous system with awakening and pain may put intolerable
stress on the left ventricle. Resumption of spontaneous breathing can
unmask inadequate cardiac function that was concealed by positive
pressure ventilation.22
Diagnosis
The presence of unusually poor oxygenation associated with dyspnea
or the appearance of a new dysrhythmia should alert the clinician to
the diagnosis, which is confirmed by the chest radiograph. Review of
the intraoperative course with the anesthesiologist may identify
precipitating events or conditions, such as the patient being in the
head-down position for an extended period. Nonanesthesiologists often
remark on the large volumes of fluid given intraoperatively; most
patients undergoing major or lengthy procedures require this fluid, and
many require additional fluid resuscitation postoperatively. While
fluid overload must be part of the differential diagnosis, underlying
heart disease or a new ischemic event should be investigated.
Echocardiography can provide diagnostic information and help guide the
therapy.
Prevention
Identification of heart failure in the preoperative period should
result in cancellation of elective surgery. Medical treatment to
optimize the cardiac status should reduce the risk of perioperative
CHF, and patients should continue to receive their medications through
the perioperative period. This is one group of patients in whom use of
a PAC to help guide perioperative fluid management may be indicated,
although this has not been evaluated in a prospective study.
Prophylactic inotropic infusions are probably not indicated in patients
with histories of heart failure; however, use of PAC will help the
clinician know when such intervention is indicated. As in CAD, close
attention to electrolyte status, hematocrit, oxygenation, and
analgesia will reduce the stress on the heart, and the patient must be
closely observed during withdrawal of positive pressure ventilation.
Hypertension and tachycardia should be prevented (adequate analgesia,
resumption of preoperative therapy) and rapidly treated when they
occur.
Treatment
As in the medical setting, treatment of acute postoperative heart
failure includes oxygen, diuretics, preload, and afterload reduction
where possible, and positive intropic agents. There should be a low
threshold for intubation to remove the work of breathing and provide
the preload and afterload reduction associated with positive
intrathoracic pressure.23 Pulmonary congestion can be
relieved rapidly by the venodilating effects of IV nitroglycerin and
loop diuretics.24 If there is any delay in response to
treatment or if hypotension develops, insertion of a PAC should be
considered to guide administration of potent vasoactive drugs and
inotropes. Whereas for long-term medical management of CHF,
angiotensin-converting enzyme inhibitors are of proven benefit, IV
enalaprilat is difficult to use in the acute setting (slow onset and
offset, unpredictable effect on BP). Similarly, digoxin has a slow
onset of action and a very modest inotropic effect. Other vasodilators
and inotropic drugs are much easier to use in the ICU. In chronic heart
failure, there is a relative insensitivity or "down regulation" of
the β-receptor,25 and use of a phosphodiesterase-3
inhibitor such as milrinone should be considered.26 Table 2
summarizes some IV therapies for acute postoperative CHF.
|
Cardiac Dysrhythmias
|
|---|
The occurrence of a new cardiac dysrhythmia in the postoperative
period is most commonly due to electrolyte disturbances and/or the
increased sympathetic nervous system activity, although myocardial
ischemia or CHF must be considered. Supraventricular tachydysrhythmias
and ventricular extrasystoles are common, and many can be controlled by
β-adrenergic blockade and correction of electrolyte disorders.
Specific etiology and treatment of dysrhythmias is beyond the scope of
this brief summary; only two relatively new treatments of dysrhythmias
will be reviewed. The reader is referred to the excellent algorithms
for acute dysrhythmias published by the American Heart
Association.27
Magnesium
In recent years, it has become apparent that major fluid shifts
and losses are associated with hypomagnesemia, magnesium loss occurs
with diuretic use, many postoperative patients are hypomagnesemic, and
there is at least a functional deficit of magnesium associated with
acute MI.2829 Both atrial and ventricular dysrhythmias
can be precipitated by hypomagnesemia, and many can be treated by
magnesium administration; infarct size can be limited by administration
of magnesium during or immediately after reperfusion, and survival
after MI appears to be improved if magnesium levels are supplemented
(although one large trial has failed to support this finding). This ion
should be measured in any patient with dysrhythmias, and in most
patients, consideration should be given to routine supplementation.
Amiodarone
This complex antidysrhythmic drug has moved into the forefront for
treatment of both supraventricular and ventricular tachydysrhythmias
refractory to first-line drugs. Most studies have examined its use in
the treatment of malignant ventricular rhythms, and amiodarone compares
favorably to virtually all other drugs in terms of efficacy, safety,
and adverse effect profile.30 While current algorithms
place this drug as an alternative or next step to bretylium for acute
refractory ventricular dysrhythmias, many practitioners are now turning
directly to amiodarone if lidocaine is ineffective. Similarly,
amiodarone appears to be at least comparable to quinidine and
procainamide in terminating and preventing atrial dysrhythmias. An
excellent review of this drug in the perioperative setting suggests
that its role will dramatically expand as safety studies in the
surgical population become available.31 One undesirable
feature of amiodarone is the need for a prolonged loading dose. Another
precaution is the potential for ARDS, which has been observed in some
patients after major pulmonary resections.
|
Hypertension
|
|---|
Acute postoperative hypertension secondary to high sympathoadrenal
tone is common, especially in patients with preexisting hypertension.
Hypertension may cause excessive surgical bleeding (especially after
cardiovascular procedures), and may precipitate myocardial ischemia,
cardiac dysfunction, and/or pulmonary congestion. Analgesia and
adequate gas exchange should be assured, and other potential causes of
hypertension (eg, bladder or gastric distention) should be
evaluated before specific treatment of hypertension is initiated.
Cold, volume-depleted patients often are hypertensive until they warm
to above 36.5°C, at which time they vasodilate and the hypovolemia
becomes apparent. Intermittent doses of vasodilating drugs produce
unpredictable responses, both in terms of effect and duration; the
nitrodilators (nitroprusside and nitroglycerin) are potent venodilators
and can cause major swings in BP due to their preload reducing effect.
β-Adrenergic-blocking drugs may be effective, and considering the
cause of most postoperative hypertension, they should be the drug of
first choice. Newer drugs that are arterial specific (nicardipine,
fenoldopam) are now available for continuous infusion and provide a
real benefit in achieving smooth control of the BP. These latter drugs
should be considered if β-blockers are either not tolerated or
ineffective. Table 3
summarizes some IV therapies for acute postoperative hypertension.
Nicardipine
Nicardipine is a dihydropyridine calcium channel-blocking drug
with a similar structure to nifedipine. Unlike the latter drug,
however, it is relatively vascular specific with little effect on the
myocardium and capacitance vessels, and it is water
soluble.32 Postoperatively, in a direct comparison to
nitroprusside, nicardipine was found to control the BP more rapidly,
with fewer changes in dose and low incidence of need to stop the drug
treatment.33 In addition, calcium-blocking drugs have
antispasmodic, anti-inflammatory, and cytoprotective
effects.34
Fenoldopam
Fenoldopam is a selective dopamine1-agonist that is
approved for the treatment of hypertension. Postsynaptic
dopamine1-receptors are found in multiple vascular beds
where their stimulation causes arterial vasodilation, and also in the
distal renal tubule where sodium excretion is modulated. Infusion of
this drug causes a lowering of systemic BP in association with an
increase in renal blood flow and sodium excretion.3536 In
comparison to nitroprusside, cardiac filling pressures and cardiac
output are better maintained with fenoldopam owing to a lack of
pulmonary and systemic venodilation. It also appears that
intrapulmonary shunt is not affected with this drug. Although there are
extensive laboratory data, relatively few clinical trials have been
conducted in surgical patients. There is, however, great interest in a
drug that can maintain or improve renal perfusion and sodium excretion
and not affect intrapulmonary shunt fraction, while lowering BP with
efficacy comparable to that of nitroprusside.
|
Conclusions
|
|---|
This brief overview has highlighted some current trends in the
short-term cardiovascular treatment of postoperative patients. As CAD
is the most prevalent underlying cardiovascular disease in surgical
patients, and as it is associated with considerable morbidity and
mortality in this setting, a major focus of postoperative care must be
on monitoring for ischemia and on therapy designed to reduce or prevent
ischemia and MI. While less common, acute CHF in the postoperative
setting is associated with mortality and may be the consequence of
underlying CAD. Perioperative invasive monitoring in high-risk patients
may reduce adverse outcome; however, studies supporting this practice
are not compelling. Newer drugs to treat perioperative dysrhythmias,
reduced cardiac performance, and arterial hypertension may provide the
clinician with valuable tools to prevent and treat adverse cardiac
outcomes.
|
References
|
|---|
- Mangano, DT (1990) Perioperative cardiac morbidity. Anesthesiology 72,153-184[ISI][Medline]
- Mangano, DT, Wong, MG, London, MJ, et al Perioperative myocardial ischemia in patients undergoing noncardiac surgery: II. Incidence and severity during the first week after surgery. J Am Coll Cardiol 1991;17,851-857[Abstract]
- Mangano, DT, Layug, EL, Wallace, A, et al Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. N Engl J Med 1996;335,1713-1720[Abstract/Free Full Text]
- Wallace, A, Layug, B, Tateo, I, et al Prophylactic atenolol reduces postoperative myocardial ischemia. Anesthesiology 1998;88,7-17[CrossRef][ISI][Medline]
- Badner, NH, Knill, RL, Brown, JE, et al Myocardial infarction after noncardiac surgery. Anesthesiology 1998;88,572-578[CrossRef][ISI][Medline]
- Chedrawy, E, Hall, R, Nedelcu, V Postoperative elevation of creatine kinase (CK-MB): does it contribute to diagnosis of myocardial infarction? Can J Anaesth 1997;44,843-848[Abstract/Free Full Text]
- Mangano, DT Beyond CK-MB: biochemical markers for perioperative myocardial infarction. Anesthesiology 1994;81,1317-1320[CrossRef][ISI][Medline]
- First International Study of Infarct Survival Collaborative Group.. Mechanisms for the early mortality reduction produced by beta-blockade started early in acute myocardial infarction: ISIS-1. Lancet 1988;1,921-923[CrossRef][Medline]
- Pepine, CJ, Cohn, PF, Deedwania, PC, et al Effects of treatment on outcome in mildly symptomatic patients with ischemia during daily life: the Atenolol Silent Ischemia Study (ASIST). Circulation 1994;90,762-768[Abstract/Free Full Text]
- Stuhmeier, KD, Mainzer, B, Cierpka, J, et al Small, oral dose of clonidine reduces the incidence of intraoperative myocardial ischemia in patients having vascular surgery. Anesthesiology 1996;85,706-712[CrossRef][ISI][Medline]
- McSPI-Europe Research Group. Perioperative sympatholysis: beneficial effects of the alpha-2 adrenoreceptor agonist mivazerol on hemodynamic stability and myocardial ischemia. Anesthesiology 1997;86,346-363[CrossRef][ISI][Medline]
- Tuman, KJ, McCarthy, RJ, March, RJ, et al Effects of epidural anesthesia and analgesia on coagulation and outcome after major vascular surgery. Anesth Analg 1991;73,696-704[Abstract/Free Full Text]
- Yeager, MP, Glass, DD, Neff, RK, et al Epidural anesthesia and analgesia in high-risk patients. Anesthesiology 1987;66,729-736[ISI][Medline]
- Frank, SM, Beattie, C, Christopherson, R, et al Intentional hypothermia is associated with postoperative myocardial ischemia: the Study of Perioperative Ischemia (SPI) Study Group. Anesthesiology 1993;78,468-476[ISI][Medline]
- Rao, TK, Jacobs, KH, El Etr, AA Reinfarction following anesthesia in patients with myocardial infarction. Anesthesiology 1983;59,449-505
- Berlauk, JF, Abrams, JH, Gilmour, IJ, et al Preoperative optimization of cardiovascular hemodynamics improves outcome in peripheral vascular surgery. Ann Surg 1991;214,289-299[ISI][Medline]
- Connors, AF, Speroff, T, Dawson, NV, et al The effectiveness of right heart catheterization in the initial care of critically ill patients. JAMA 1996;276,889-897[Abstract]
- Connors, AF, McCaffree, DR, Gray, BA Evaluation of right-heart catheterization in the critically ill patient without acute myocardial infarction. N Engl J Med 1983;308,263-267[ISI][Medline]
- Haggmark, S, Hohner, P, Ostman, M, et al Comparison of hemodynamic, electrocardiographic, mechanical and metabolic indicators of intraoperative myocardial ischemia in vascular surgical patients with coronary artery disease. Anesthesiology 1989;70,19-25[ISI][Medline]
- Goldman, L, Caldera, DL, Nussbaum, SR, et al Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297,845-850[Abstract]
- Cooperman, LH, Price, HL Pulmonary edema in the operative and postoperative period: a review of 40 cases. Ann Surg 1970;172,883-891[ISI][Medline]
- Lemaire, F, Teboul, JL, Cinotti, L, et al Acute left ventricular dysfunction during unsuccessful weaning from positive pressure ventilation. Anesthesiology 1988;69,171-179[ISI][Medline]
- Grace, MP, Greenbaum, DM Cardiac performance in response to PEEP in patients with cardiac dysfunction. Crit Care Med 1982;10,358-360[ISI][Medline]
- Raftery, EB Hemodynamic effects of diuretics in heart failure. Br Heart J 1994;72,S44-S47[CrossRef]
- Bristow, MR, Ginsberg, R, Minobe, W, et al Decreased catecholamine sensitivity and beta-adrenergic receptor density in failing human hearts. N Engl J Med 1982;307,205-211[Abstract]
- Monrad, ES, Baim, DS, Smith, HS, et al Effects of milrinone on coronary hemodynamics and myocardial energetics in patients with congestive heart failure. Circulation 1985;71,972-979[Abstract/Free Full Text]
- Cummins RO, ed. Textbook of advanced cardiac life support. Dallas, TX: American Heart Association, 1994
- Chernow, B, Bamberger, S, Stoiko, M, et al Hypomagnesemia in patients in postoperative intensive care. Chest 1989;95,391-397[Abstract/Free Full Text]
- Arsenian, MA Magnesium and cardiovascular disease. Prog Cardiovasc Dis 1993;35,271-310[CrossRef][ISI][Medline]
- Kowey, PR, Levine, JH, Herre, JM, et al Randomized, double-blind comparison of intravenous amiodarone and bretylium in the treatment of patients with recurrent hemodynamically destabilizing ventricular tachycardia or fibrillation. Circulation 1995;92,3255-3263[Abstract/Free Full Text]
- Balser, JR The rational use of intravenous amiodarone in the perioperative period. Anesthesiology 1997;86,974-987[CrossRef][ISI][Medline]
- Ogawa, T, Sekiguchi, T, Ishii, T, et al Acute effects of intravenous nicardipine on hemodynamics and cardiac function in patients with a healed myocardial infarction and no evidence of congestive heart failure. Am J Cardiol 1991;68,301[CrossRef][ISI][Medline]
- Halpern, NA, Goldberg, M, Neely, C, et al Postoperative hypertension: a multicenter, prospective, randomized comparison between intravenous nicardipine and sodium nitroprusside. Crit Care Med 1992;20,1637-1643[ISI][Medline]
- Levy, JH The ideal agent for perioperative hypertension and potential cytoprotective effects. Acta Anaesthesiol Scand 1993;37(suppl 99),20-25[ISI][Medline]
- Murphy, MB, McCoy, CE, Weber, RR, et al Augmentation of renal blood flow and sodium excretion in hypertensive patients during blood pressure reduction by intravenous administration of the dopamine1 agonist fenoldopam. Circulation 1987;76,1312-1318[Abstract/Free Full Text]
- Hill, AJ, Feneck, RO, Walesby, RK A comparison of fenoldopam and nitroprusside in the control of hypertension following coronary artery surgery. J Cardiothorac Vasc Anesth 1993;7,279-284[CrossRef][Medline]
TOP
Abstract
Introduction
