Anesth Analg 2004;98:1798-1802
© 2004 International Anesthesia Research Society
doi: 10.1213/01.ANE.0000117145.50236.90
GENERAL ARTICLES
Brief Postoperative Delirium in Hip Fracture Patients Affects Functional Outcome at Three Months
Khwaja Zakriya, MD*,
Frederick E. Sieber, MD*,
Colleen Christmas, MD
,
James F. Wenz, Sr., MD
, and
Shawn Franckowiak, BS
*Department of Anesthesiology,
School of Medicine, Division of Geriatrics and Gerontology, and
Department of Orthopedics, Johns Hopkins Bayview Medical Center, Baltimore, Maryland
Address correspondence and reprint requests to Frederick E. Sieber, MD, Department of Anesthesiology, Johns Hopkins Bayview Medical Center, 4940 Eastern Ave., Baltimore, MD 21224. Address e-mail to fsieber{at}jhmi.edu
 |
Abstract
|
|---|
It is unclear how brief postoperative delirium (DEL) affects functional outcomes. In this study, we sought to determine if patients with brief postoperative DEL (<6-wk duration) have different living situations when compared with non-DEL patients after hip fracture repair. In a prospective study, patients admitted to the geriatric hip fracture service were assessed every postoperative day for the presence of DEL using the confusion assessment method (CAM) score. Patients were reassessed at 6 wk and 3 mo postoperatively for CAM score, current living situation, and activities of daily living. Group comparisons were tested after dividing patients into two groups: DEL (DEL; [+] CAM at any time during the postoperative period while in the hospital); no-DEL (no DEL; [–] CAM throughout the postoperative period while in the hospital). The study included 92 patients of whom 26 (28%) were CAM (+) after surgery. At 6 wk follow-up, n = 81; at 3 mo follow-up, n = 76. Eight patients died during the study. At 6 wk and 3 mo, a larger percentage of DEL patients were not living with a family member (27% versus 8% patients not living with a family member at 3 mo follow-up in DEL and no-DEL, respectively). There was no difference in activities of daily living by 3 mo. We conclude that brief postoperative DEL lasting <6 wk is a determining factor for poor long-term functional outcome after hip fracture repair, because it significantly impacts the ability to live independently.
IMPLICATIONS: Brief postoperative delirium lasting <6 wk is a determining factor for poor long-term functional outcome after hip fracture repair, because it significantly impacts the ability to live independently.
|
Introduction
|
|---|
Postoperative delirium (DEL) is a common problem in the elderly. In hip fracture patients, the incidence of perioperative DEL ranges from 28% to 41% (1,2). DEL at admission (i.e., before surgery) has been associated with poorer functioning in physical, cognitive, and affective domains (3) and is an important predictor of mortality (4). Persistent postoperative DEL is independently associated with poor functional recovery 1 mo after hip fracture even after adjusting for prefracture frailty (2). However, it is unclear how brief, acute postoperative DEL affects functional outcomes. The aim of this study was to determine if patients with brief postoperative DEL (<6 wk duration) have different living situations when compared with non-DEL patients after hip fracture repair. This prospective study was performed in the context of a multidisciplinary team approach to the treatment of hip fracture (5).
|
Methods
|
|---|
The IRB of Johns Hopkins Bayview Medical Center approved this study, and written informed consent was obtained from all study participants. At Johns Hopkins Bayview Medical Center, a hip fracture service operates as a multidisciplinary team consisting of members from the geriatric, orthopedic, anesthesia, and rehabilitation services. Ninety-six consecutive patients admitted to the hip fracture service were studied prospectively from March 1999 until February 2003. No hip fracture repairs are done electively. All of the hip fracture patients are operated on a semi-urgent basis. There was no age criterion for the study.
The presence of postoperative DEL was defined by the use of the confusion assessment method (CAM) score (6). The CAM score uses a diagnostic algorithm consisting of 4 features: 1) acute and fluctuating changes in mental status, 2) inattention, 3) disorganized or incoherent thinking, and 4) altered level of consciousness. A CAM score is considered to be positive ([+] CAM) if the patient displays the combination of features: (1, 2, and 3) or (1, 2, and 4) or (1, 2, 3, and 4). During the initial hospitalization, the attending geriatrician (CC) made patient rounds each day during the mid-morning. After the attending geriatrician finished rounds in the morning, the attending geriatrician, the patient’s primary nurse, and family were separately questioned by the investigator concerning each of the four features of the diagnostic algorithm. Feature 1 was determined to be present primarily by report from the primary nurse or family member. Features 2, 3, and 4 were determined to be present primarily by report from the attending geriatrician. Patients with DEL on hospital admission were eliminated from the study.
Outcomes were assessed during initial hospitalization, 6 wk and 3 mo after hospital discharge. Six weeks’ duration was selected based on the manner in which the orthopedic surgeons follow up their patients. We tried to make the study follow-up convenient for the patients to maximize our follow-up. The current practice of our orthopedic surgeons is to see patients postoperatively at 6 wk and 3 mo. Previous studies have also chosen arbitrary follow-up times such as 1 mo and 6 mo (1,2).
During initial hospitalization, the outcome variables analyzed included medical complications, length of stay (LOS) (days) in the intensive care unit (ICU) and hospital, and mortality.
Patients were assessed every postoperative day by the attending geriatrician (CC) for the presence of new postoperative medical complications. These included: pneumonia (pulmonary consolidation on chest radiograph), clinically diagnosed congestive heart failure, deep vein thrombosis, and myocardial infarction, atrial fibrillation (electrocardiogram criterion), additional fall in hospital, return to the operating room, or urinary tract infection (documentation of a positive urine culture). In addition, after hospital discharge, LOS in the rehabilitation unit was recorded.
At 6 wk and 3 mo after the initial date of hip fracture repair, the patient was examined and the patient and a family member were interviewed by a member of the research team in the orthopedic clinic. Best grip strength of 3 attempts using the dominant hand as measured by a hand-held dynamometer, pain levels (10-point Likert scale), ambulation status, and CAM score were determined. "Best grip strength" has been shown to be an excellent predictor of disability in the elderly (7). Hospital and/or emergency department admissions since initial hospital discharge were obtained. Current living situation as well as independence in activities of daily living were recorded. Because the purpose of this study was to determine if patients with acute postoperative DEL (<6 wk duration) have a different living situation when compared with non-DEL patients after hip fracture repair, patients who were CAM (+) at either 6 wk or 3 mo postoperatively were eliminated from the analysis.
The aim of this study was to determine if patients with brief postoperative DEL (<6 wk duration) have different living situations when compared with non-DEL patients after hip fracture repair. Sample size was calculated using data from Edlund et al. (1) who found a 28% incidence of DEL and a 28% and 67% requirement for institutionalization at 6 mo postoperatively in DEL and non-DEL patients, respectively. Setting 1-
at 95% and 1-ß at 80%, the sample size required to determine postoperative differences between DEL and non-DEL patients in the requirement for institutionalization at 6 mo is n = 72. Because we used a 3-mo follow-up in the current study, a slightly larger number was obtained to determine differences between groups. Thus, a sample size at 3 mo follow-up of n = 76 was targeted.
Patients were divided into two groups: DEL (DEL; [+] CAM at any time during the postoperative period while in the hospital); no-DEL (no DEL; [–] CAM throughout the postoperative period while in the hospital). Demographics and outcome variables were tested with either independent t-test or 
2 analysis using DEL or no-DEL as the grouping variable. A cut-off value of P < 0.05 was set to determine significance.
|
Results
|
|---|
Over the time period studied, 220 patients were admitted to the hospital for hip fracture. Forty-seven patients (21%) had preoperative DEL and 11 patients (5%) did not undergo surgery. Of the remaining possible subjects, 56% (n = 96) consented to participation in the study. Two patients were CAM (+) at 6 wk and 2 patients were CAM (+) at 3 mo. These four patients were eliminated from analysis. Ninety-two patients were studied of whom 26 (28%) became CAM (+) after surgery. Eight patients died during the study. The demographics and comorbidity of the population studied are shown in Table 1. DEL patients had a greater burden of systemic disease as assessed by ASA physical status. Most patients in both groups were able to ambulate independently before fracture. One patient in the DEL group and three patients in the no-DEL group had a previous hip fracture. In addition, DEL and no-DEL patients were of similar age (79 ± 10 and 78 ± 7 yr in DEL and no-DEL groups, respectively).
There was one in-hospital death in the DEL group. The incidence of in-hospital medical complications was similar in the 2 groups except for new onset atrial fibrillation (Table 2). Urinary tract infection was the most common complication. A larger percentage of DEL patients went to the ICU postoperatively (46% versus 20% of patients in DEL and no-DEL groups, respectively). ICU LOS and hospital LOS were longer in the DEL group (5 ± 4 versus 2 ± 1 ICU days; 8 ± 6 versus 5 ± 2 hospital days, in DEL and no-DEL groups, respectively). Rehabilitation LOS did not differ between groups (15 ± 8 versus 14 ± 7 rehabilitation days in DEL and no-DEL groups, respectively).
Eighty-one patients (19 of 26 DEL and 63 of 66 no-DEL patients) returned for follow-up examination 6 wk after fracture repair. Three patients died between hospital discharge and 6 wk follow-up (2 DEL and 1 no-DEL). All patients were CAM (–) who presented for 6 wk follow-up. Best grip strength and pain levels were similar between groups (20 ± 10 versus 24 ± 9 kg and 2 ± 3 versus 3 ± 3 pain level in DEL and no-DEL groups, respectively). A larger percentage of DEL patients were not living with a family member (32% versus 10% in DEL and no-DEL groups, respectively; P = 0.017). Activities of daily living were similar between groups except for ability to take public transportation (Table 3).
Seventy-six patients (15 of 26 DEL and 61 of 66 no-DEL patients) returned for study at 3 mo after fracture. Four deaths occurred between the 6-wk and 3-mo follow-up (2 deaths in each group). No patient was CAM (+) at 3 mo follow-up. Best grip strength and pain levels were similar between groups (28 ± 8 versus 24 ± 9 kg and 2 ± 3 versus 3 ± 3 pain level in DEL and no-DEL groups, respectively). In addition, best grip strength and pain levels at 3 mo were comparable to those observed at 6 wk. At 3 mo after hospital discharge, a larger percentage of DEL patients were not living with a family member (27% versus 8% patients not living with a family member in DEL and no-DEL groups, respectively; P = 0.047). At 3 mo after hospital discharge, activities of daily living did not differ between groups (Table 3).
|
Discussion
|
|---|
This study suggests that DEL resolving in six weeks or less negatively impacts living situation after hip fracture. Brief DEL also has pronounced effects on short-term outcomes. In particular, brief DEL is associated with more in-hospital medical complications, and longer ICU and hospital LOS.
DEL is clearly associated with medical comorbidity (2,3,8). Postoperative DEL is well known to be associated with increased mortality (4), a more frequent incidence of medical complications (9), and prolonged hospital stay (5). The current study shows that DEL that resolves in less than six weeks is associated with longer hospital and ICU LOS. In addition, postoperative DEL in this study was associated with increased medical morbidity (new onset atrial fibrillation), even though these outcomes were rare in this group of patients.
In long-term follow-up studies, DEL has been associated with a decreased ability to live independently (1) and persistent DEL is associated with poorer long-term functional outcomes (2). The current study shows that by three months after hospital discharge patients exhibiting postoperative DEL were more likely not to be living with a family member. This suggests that both brief and persistent DEL may have significant effects on long-term functional recovery or living situation.
Several series examining postoperative DEL in orthopedic surgical patients have reported that DEL has an onset of approximately 24 ± 21 hours after surgery in most patients (10). The duration of DEL is generally <48 (mean 42 ± 43) hours (10), and usually resolves within 1 week (11). However, in a study of patients with hip fracture, Marcantonio et al. (2) reported that persistent DEL is common, with 6% of patients still confused at 6 months postoperatively. Our study demonstrated similar rates of postoperative DEL in a population of patients with acute hip fracture to those previously reported (1,2). In addition, four patients were eliminated from the study because of persistent DEL at either six weeks or three months. Our incidence of persistent DEL was 4% and is comparable to that of Marcantonio et al. (2).
The strength of this study is that it is prospective and examines a relatively homogeneous surgical population. In addition, thorough follow-up was obtained on each patient allowing an accurate presentation of patient outcomes. DEL was diagnosed using a standardized screening tool, the CAM score, which has been shown to have excellent sensitivity and specificity (6), so it is unlikely we have underdiagnosed the condition in the current study.
Our study has several shortcomings. First, there was selection bias. Only patients who consented to participation in the study were included, and it is possible that there were even more frequent rates of DEL in the group that declined investigation. The number of patients available for follow-up decreased from 92 to 81 to 76 at 3 months. The patients lost to follow-up may represent individuals with continued changes in their living situation. Second, we may have missed some DEL at the six-week and three-month follow-up, because the CAM at these time points was applied to only a brief observation time (while the patient was in the office for follow-up). Third, severity of DEL was not tested, nor the exact length of time the DEL lasted. Previous studies suggest that persistent DEL at 1 month is associated with poorer outcome (2). It is possible that both severity and duration of DEL are important variables in determining outcome.
We conclude that even brief postoperative DEL is associated with poorer long-term functional outcome as evidenced by differences in living situation by three months. Brief postoperative DEL also significantly lengthens hospital and ICU LOS and is associated with increased medical comorbidity and complications.
|
Acknowledgments
|
|---|
Supported by John A. Hartford Foundation Grant 97214-G and the Johns Hopkins Geriatrics/Nursing Fund.
|
References
|
|---|
- Edlund A, Lundstrom M, Lundstrom G, et al. Clinical profile of delirium in patients treated for femoral neck fractures. Dement Geriatr Cogn Disord 1999; 10: 325–9.[ISI][Medline]
- Marcantonio ER, Flacker JM, Michaels M, et al. Delirium is independently associated with poor functional recovery after hip fracture. J Am Geriatr Soc 2000; 48: 618–24.[ISI][Medline]
- Dolan MM, Hawkes WG, Zimmerman SI, et al. Delirium on hospital admission in aged hip fracture patients: prediction of mortality and 2-year functional outcomes. J Gerontol A Biol Sci Med Sci 2000; 55: M527–34.[Abstract/Free Full Text]
- Magaziner J, Simonsick EM, Kashner TM, et al. Survival experience of aged hip fracture patients. Am J Public Health 1989; 79: 274–8.[Abstract/Free Full Text]
- Parikh SS, Chung F. Postoperative delirium in the elderly. Anesth Analg 1995; 80: 1223–32.[Abstract]
- Inouye SK, van Dyck CH, Alessi CA, et al. Clarifying confusion: the confusion assessment method—a new method for detection of delirium. Ann Intern Med 1990; 113: 941–8.[ISI][Medline]
- Giampaoli S, Ferrucci L, Cecchi F, et al. Hand-grip strength predicts incident disability in non-disabled older men. Age Ageing 1999; 28: 283–8.[Abstract/Free Full Text]
- O’Keeffe ST, Lavan JN. Predicting delirium in elderly patients: development and validation of a risk-stratification model. Age Ageing 1996; 25: 317–21.[Abstract/Free Full Text]
- Marcantonio ER, Goldman L, Mangione CM, et al. A clinical prediction rule for delirium after elective noncardiac surgery. JAMA 1994; 271: 134–9.[Abstract]
- Duppils GS, Wikblad K. Acute confusional states in patients undergoing hip surgery: a prospective observation study. Gerontology 2000; 46: 36–43.[ISI][Medline]
- Williams-Russo P, Urquhart BL, Sharrock NE, et al. Post-operative delirium: predictors and prognosis in elderly orthopedic patients. J Am Geriatr Soc 1992; 40: 759–67.[ISI][Medline]
Accepted for publication December 18, 2003.
Top
Abstract
Introduction
