|
|
||||||||
Ann Thorac Surg 2005;79:1004-1009
© 2005 The Society of Thoracic Surgeons
ld
zeli, MDa
uzhan Özyurtkan, MDa
rel, MDa
cu, MDb
lu, PhDc
a Department of Thoracic Surgery, Marmara University Hospital, Istanbul, Turkey
b Department of Psychiatry, Marmara University Hospital, Istanbul, Turkey
c Department of Biostatistics, Marmara University Hospital, Istanbul, Turkey
Accepted for publication June 4, 2004.
* Address reprint requests to Dr Yüksel, Department of Thoracic Surgery, Marmara University Hospital, PK 97, 34718 Ac
badem, Istanbul, Turkey (E-mail: myuksel{at}marmara.edu.tr).
| Abstract |
|---|
|
|
|---|
METHODS: All patients (n = 432) who underwent thoracotomy or sternotomy from 1996 to 2003 were analyzed retrospectively. The diagnosis of postoperative delirium was based on Diagnostic and Statistical Manual of Mental Disorders- IV criteria.
RESULTS: Postoperative delirium developed in 23 patients (5.32%) between postoperative days 2 to 12 (mean, 4.4 ± 2.6 days). There were 15 males and 8 females, with a mean age of 59.4 years (24 to 77 years). The delirium group was older (59.4 ± 14.6 vs 51.3 ± 15.5 years, p < 0.01) and had a longer operation time than the nondelirious group (5.34 ± 1.58 vs 4.38 ± 1.6 hours, p = 0.005). Morbidity and mortality rates were not significantly different between the two groups (56.5% vs 47.1%; 13.0% vs 3.66%, respectively). Univariate analysis showed that the older age, markedly abnormal postoperative levels of sodium, potassium, or glucose, sleep deprivation, operation time, and diabetes mellitus were risk factors (p < 0.05). According to multivariate analyses, four factors were selected as predictive risk factors: (1) markedly abnormal postoperative levels of sodium, potassium, or glucose (p = 0.038); (2) sleep deprivation (p = 0.05); (3) age (p = 0.033); and (4) operation time (p = 0.041).
CONCLUSIONS: Postoperative delirium may cause higher morbidity and mortality rates after thoracic surgery. Close postoperative follow-up and early identification of predisposing factors such as older age, sleep deprivation, abnormal postoperative levels of sodium, potassium, or glucose, and longer operation time can prevent occurrence of postoperative delirium.
| Introduction |
|---|
|
|
|---|
| Patients and Methods |
|---|
|
|
|---|
|
A psychiatric consultation was requested once delirium symptoms were first noted after the operation. A diagnosis was made based on the Diagnosis and Statistical Manual of Mental Disorders (DSM-IV) [5]. Antipsychotics, haloperidol, risperidone, and quetapine have been used in treating delusions, paranoia, and perceptual disturbances. Benzodiazepines were used in alcohol and sedative withdrawal syndromes.
The preoperative and postoperative assessments were performed accordingly [3, 6]. The preoperative evaluation included a medical history highlighting the presence of either chronic or acute illnesses like pulmonary, cardiac, renal, and liver disease, presence of diabetes mellitus (DM), alcohol abuse, previous psychological or neurologic diseases, and history of delirium. The definition of alcohol abuse was made based on DSM-IV [5] that describes alcohol abusers as those who drink despite recurrent social, interpersonal, and legal problems as a result of alcohol use. We adopted the approach described by Marcantonio and colleagues [3] for abnormal preoperative chemistry values for postoperative serum electrolyte panels. Abnormal postoperative results were considered present when postoperative serum potassium levels were less than 3.0 meq/L or greater than 6 meq/L, serum sodium was less than 130 meq/L or greater than 150 meq/L, or blood glucose values were less than 60 mg/dL or greater than 300 mg/dL. Respiratory insufficiency was defined as arterial hypoxemia PO2 less than 55 mm Hg on room air and hypercarbia PCO2 greater than 45 mm Hg [7]. In order to determine the effects of duration of operation in development of POD, the length of operation time, which was defined as the period from induction of anesthesia to extubation, was measured. Infection was diagnosed when white blood cell count was greater than 12,000/mm3 and postoperative fever was greater than 37.0°C [6]. The other preoperative and postoperative risk factors that were analyzed are listed in Table 2.
|
2 test, and Fishers exact test of these factors were performed. Continuous data are presented as mean ± SD unless otherwise noted. To identify factors independently related to the development of delirium, we also performed forward multivariate stepwise (conditional) logistic regression. A p value of less than or equal to 0.05 was considered significant. Approval for this study was granted by the Institutional Review Board of the Marmara University School of Medicine. | Results |
|---|
|
|
|---|
The length of operation time was significantly longer in the delirium subset than in the nondelirium subset (5.34 ± 1.58 hours vs 4.38 ± 1.6 hours, p = 0.005). However, no significant differences in the mean intensive care unit (ICU) stay (2.95 vs 4.47 days) and length of hospital stay (11.65 vs 9.8 days) were observed.
One or more complications were observed in 56.5% of the delirious patients and in 47.1% of the nondelirious patients (Table 3). The most common complication (17.3%) was air leak and wound infection (13%) in the POD group, whereas it was atelectasis (9.2%) with poor clearance of pulmonary secretions necessitating bronchoscopy in the non-POD group. The mortality rate was 13% in the POD group while it was 3.66% in the non-POD group (p = 0.06). Although the differences were not significant, more complications were observed in the POD group.
|
|
| Comment |
|---|
|
|
|---|
Dyer and colleagues [2] reported that the overall incidence of POD was 36.8%, with a range of 0% to 73.5%. The primary reasons for this wide variation were the underdiagnosis of quiet delirium and inconsistent application of numerous diagnostic methods (DSM-III, revised edition, DSM-IV, Confusion Assessment Method, Cognitive status, Mini-Mental State Examination) [11, 12]. The incidence of postoperative delirium has been reported to be 3% to 47% in cardiac surgery [13], 23.8% to 41% in orthopedic surgery [14, 15], 17% in gastrointestinal surgery [16], 13.8% to 17% in head and neck surgery [6], and 36.2% in vascular surgery [17, 18]. Despite evidence of a high risk for developing delirium, reliable studies on risk factors and prediction of postoperative delirium are rare in general thoracic surgery. Of the studies examining postoperative delirium, other than the study of Aakerlund and Rosenberg [4], only four studies included noncardiac thoracic surgery patients in a large cohort of postoperative delirium after noncardiac surgery [3, 9, 12, 19]. The incidence of delirium in our study was 5.32%, in the lower range in comparison with previous reports [3, 4, 12]. Its incidence has been reported to be 21% by Aakerlund and Rosenberg [4], 16% by Marcantonio and colleagues [3], 7% after thoracotomy and 13.6% after thoracoscopic lung resection by Lynch and colleagues [12]. Our relatively low incidence of POD may be because our patients were younger than those of the reported series.
In this study, the mean duration of POD was 4.4 days. On the other hand, the onset of delirium has been reported as mean 2.5 to 3.4 days after vascular surgery [11, 18], 3.7 days after hip surgery [14], and 3 days after thoracic surgery [4]. It may be speculated that the onset of postoperative delirium in thoracic surgery patients may begin later than other types of operations.
Various hypotheses have been proposed to explain the pathogenesis of POD [19]. One suggests that when the oxidative metabolism of the brain decreases, the levels of neurotransmitters within the brain, such as acetylcholine, decline and cause mental dysfunction. Medications, with anticholinergic activity frequently can cause confusion in the elderly. Recently, Herrmann and colleagues [20] suggested that increased levels of postoperative serum concentrations of S-100 ß (glial) and neuron-specific enolase, which are neurobiochemical markers of brain damage, might have a high predictive value in the development of POD after cardiac surgery. Similarly, blood concentrations of protein S-100 ß increase after abdominal surgery and seem to be related to POD [21]. On the other hand, Shigeta and colleagues [22] proposed that decreased melatonin secretion after surgery triggers sleep disturbances in elderly patients, which in turn cause delirium. Our study focused on predictors of delirium, therefore we cannot assess mechanisms of delirium from our data.
As has been demonstrated in several studies [3, 19, 23], various factors have separately influenced the development of POD; however, several factors might simultaneously effect the development of delirium. The factors that influence the development of POD are divided into preoperative, intraoperative, and postoperative [3, 4, 12, 19, 23]. The preoperative factors are identified as older age, male gender, alcohol consumption, mental and physical disorders, and medications. Type of operation (eg, cardiac, orthopedic, ophthalmologic), longer duration of surgery, use of anesthetic drugs with anticholinergic and antimuscarinic activities, and complications during surgery (eg, hypotension, hyperventilation, embolism, hypoxemia, greater intraoperative blood loss) are determined as intraoperative factors. Since all thoracic surgery operations are performed under general anesthesia, we did not examine the role of intraoperative anesthetic technique. However, type of anesthesia (regional vs general) and intraoperative hemodynamic complications have not been associated with delirium [12]. Since noncardiac thoracic surgery puts older patients at risk for impaired respiratory effort due to age-related respiratory changes and incision pain, which may lead to hypoxemia, hypercapnia, and pneumonia, it has been proposed as a factor associated with increased risk for delirium [3, 4, 9, 12, 19]. Since inadequate oxygenation was proposed for the development of POD in patients undergoing thoracotomy for pulmonary malignancy, supplementary oxygen has been advised for the treatment of delirium [4]. Hypocarbia, sepsis, more postoperative blood transfusions and postoperative hematocrit less than 30%, sensory deprivation or overload, and electrolyte or metabolic problems have also been reported as postoperative factors that influence the development of POD. Our findings are consistent with other studies [3, 12, 19, 23]; that patients who are old, have abnormal postoperative levels of sodium, potassium, or glucose, a presence of sleep deprivation, and longer duration of surgery, are at high risk of developing POD.
In our study, operation time was significantly longer in the POD group than in the non-POD group. As the definition implies, this period includes instrumentation and catheterization by the anesthesia team and the surgical procedure. The use of staplers is limited in our institution during a regular resection, therefore the average operation times are acceptable for such a training institution.
Although delirium appeared to be strongly associated with alcohol abuse [3], no alcohol abuse was related to delirium; possibly due to the small number of these patients in our sample. However, patients with alcohol abuse were more common in the POD group than in the non-POD group (13.0% vs 5.6%, respectively, not significant).
Primary prevention is more effective than intervention and supportive measures after the manifestation of POD [14]. The prediction score, which is based on a cross-validation analysis, has been suggested to quantify an individual patients risk for POD [18]. Although nonmodifiable patient clinical characteristics, such as age and preexisting central nervous system disease, place patients at higher risk for POD, a multimanagement approach may prevent delirium. It includes eliminating all medications that can precipitate delirium, optimizing the patients fluid status, aggressively treating pain, promoting early ambulation, and ensuring a familiar, tranquil postoperative care setting [19, 24]. If it is possible, using restraints, intravenous lines, and Foley catheters should be avoided. In our study group, the patients with postoperative delirium were not restrained and if a patient developed POD the Foley catheter was immediately removed; if otherwise there was no contraindication. In about 16% of patients no cause for delirium may be found and then efforts should focus on supportive measures [24].
Effective pain management after a thoracotomy is essential to reduce pulmonary complications. An epidural or intrathecal infusion of opioids, with or without local anesthetic, is effective with parenteral opioids. Patient controlled analgesia should be used whenever possible. Our policy is to place an epidural catheter preoperatively or to use intravenous PCA. Because the anticholinergic activity of meperidine may increase the risk of delirium [3], we avoid its use especially in older adults.
Pharmacologic treatment is essential to prevent injury or to allow further evaluation. A high-potency antipsychotic, haloperidol (doses of 0.5 to 2.0 mg), is the most commonly used drug [19]. It is useful in treating delusions, paranoia, and perceptual disturbances. Benzodiazepines are the drug of choice in alcohol and sedative withdrawal syndromes. Recently, successful use of melatonin has been reported in treating severe POD that is unresponsive to antipsychotics or benzodiazepines [25]. Since most of our POD patients presented with agitation and hallucinations, and they also required rapid pharmacologic intervention, haloperidol was used as the initial treatment in our cohort of patients.
Aizawa and colleagues [26] have suggested "the delirium-free protocol treatment" (DFP) for controlling POD in elderly patients after general surgery. By using intramuscular injection of diazepam and a continuous intravenous infusion of flunitrazepam and pethidine, the incidence of POD was reported significantly lower in the DFP group than that of the non-DFP group (5% vs 35%, p = 0.023).
The prognosis of delirium is related to the prompt recognition and the appropriate management of the underlying cause(s). Studies clearly show that patients who develop delirium have increased rates of major complications, longer lengths of stay, and higher rates of discharge to long-term care or rehabilitative facilities than patients who did not develop this complication [3, 18]. During the index admission, mortality for patients with delirium ranges from 10% to 26%, then increases with the length of follow-up [24]. At 1 year after admission mortality is 38%, and at 5 years is 51% [27]. But in this study, we could not find any significant relationship between the development of POD and postoperative complications. The small number and quite younger ages of our delirious patients compared to those of reported series may explain why we did not observe more complications in this study. However, more than half of the patients with POD presented at least one major complication and mortality occurred in 13% of them. In this study, postoperative delirium did not lead to a significant increase in ICU stay but did lead, although not significant, to an increase in length of hospital stay. This conflicting result is due to the fact that the mean onset of delirium in our study was 4.4 days and our patients who developed POD had already been moved from ICU to the surgical ward in a mean of 2.95 days. Therefore we did not observe more patients with POD in the ICU unit. On the other hand, our patients with POD did not need more ICU stay; it is also our policy to promote early ambulation and to ensure a tranquil postoperative care setting for patients who undergo a thoracic surgery operation. In order to achieve this, we try to transfer our patients as soon as possible when they no longer need ICU monitoring. In addition, another possible reason for a mean of 9.8 days of hospital stay in the non-POD group may be due to our cultural habit for a longer hospital stay for patients who undergo a major thoracic surgery. A large-scale international study in older patients found 25.8% had postoperative cognitive dysfunction after major abdominal or orthopedic surgery, and 9.9% still had cognitive impairment three months after surgery [28].
Our results have led us to develop a prospective trial designed to identify risk factors for the development of POD within a population of thoracic surgery patients. The study was designed to predict the individual risk for developing POD at the end of surgery based on preoperative data and intraoperative course.
In conclusion, delirium may cause more postoperative complications in thoracic surgery patients and its variable presentation may mask the diagnosis and its cause. To prevent the development of postoperative delirium, intensive care should be provided for patients who are older and present with postoperative sleep deprivation, and who also have longer operation time and markedly abnormal postoperative levels of sodium, potassium, or glucose. Therefore, we recommend detecting and eliminating these risk factors during the preoperative and postoperative periods to prevent the occurrence of POD.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Koster, F. G.J. Oosterveld, A. G. Hensens, A. Wijma, and J. van der Palen Delirium After Cardiac Surgery and Predictive Validity of a Risk Checklist. Ann. Thorac. Surg., December 1, 2008; 86(6): 1883 - 1887. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. L. Fricchione, S. H. Nejad, J. A. Esses, T. J. Cummings Jr., J. Querques, N. H. Cassem, and G. B. Murray Postoperative Delirium Am J Psychiatry, July 1, 2008; 165(7): 803 - 812. [Full Text] [PDF] |
||||
![]() |
E. Anton, J. Marti, M. M. Kwatra, H. Bohner, F. Schneider, and S. K. Inouye Delirium in older persons. N. Engl. J. Med., June 8, 2006; 354(23): 2509 - 2511. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |