- Email: [email protected]
The association of preoperative delirium with postoperative outcomes following hip surgery in the elderly
Journal of Clinical Anesthesia 60 (2020) 28–33
Contents lists available at ScienceDirect
Journal of Clinical Anesthesia journal homepage: www.elsevier.com/locate/jclinane
Original Contribution
The association of preoperative delirium with postoperative outcomes following hip surgery in the elderly
T
Shubham Agrawal (BS)a, Robby Turk (BS)a, Brittany N. Burton (MHS, MAS)a, ⁎ Jerry Ingrande (MD, MS)b, Rodney A. Gabriel (MD, MAS)b,c, a
School of Medicine, University of California San Diego, San Diego, CA, USA Department of Anesthesiology, Division of Regional Anesthesia and Acute Pain, University of California, San Diego, San Diego, CA, USA c Department of Medicine, Division of Biomedical Informatics, University of California, San Diego, San Diego, CA, USA b
A R T I C LE I N FO
A B S T R A C T
Keywords: Preoperative delirium Postoperative complications Hip surgery ACS NSQIP
Study objective: To determine the association of preoperative delirium with postoperative outcomes following hip surgery in the elderly. Design: Retrospective cohort study. Setting: Postoperative recovery. Patients: 8466 patients all of whom were 65 years of age or older undergoing surgical repair of a femoral fracture. Of the total population studied, 1075 had preoperative delirium. Of those with preoperative delirium, 746 were ASA class 3 or below and 327 were ASA class 4 or above. Of the 7391 patients without preoperative delirium, 5773 were ASA class 3 or below and 1605 were ASA class 4 or above. The remainder in each group was of unknown ASA class. Interventions: We used multivariable logistic regression to explore the association of preoperative delirium with 30-day postoperative outcomes. The odds ratio (OR) with associated 95% confidence interval (CI) was reported for each covariate. Measurements. Data was collected regarding the incidence of postoperative outcomes including: delirium, pulmonary complications, extended hospital stay, infection, renal complications, vascular complications, cardiac complications, transfusion necessity, readmission, and mortality. Main Results. After adjusting for potential confounders, the odds of postoperative delirium (OR 9.38, 95% CI 7.94–11.14), pulmonary complications (OR 1.83, 95% CI 1.4–2.36), extended hospital stay (OR 1.47, 95% CI 1.26–1.72), readmission (OR 1.27, 95% CI 1.01–1.59) and mortality (OR 1.92, 95% CI 1.54–2.39) were all significantly higher in patients with preoperative delirium compared to those without. Conclusions: After controlling for potential confounding variables, we showed that preoperative delirium was associated with postoperative delirium, pulmonary complications, extended hospital stay, hospital readmission, and mortality. Given the lack of studies on preoperative delirium and its postoperative outcomes, our data provides a strong starting point for further investigations as well as the development and implementation of targeted risk-reduction programs.
1. Introduction Delirium is a common complication in hospitalized patients, and occurs in roughly 61% of elderly patients with hip fracture [1]. The lifetime risk of hip fracture is 17.5% for women and 6% for men [2]. Mortality is as much as fourfold higher in patients who had non-operative treatment, thus surgery is indicated for most hip fracture
⁎
patients [3]. In the United States, femoral neck and intertrochanteric fractures occurred with increasing frequency as age progressed between 65 and 99 [4]. The number of persons aged > 70 years old is expected to double within the next three decades, which may lead to an increased volume of hip fractures and making associated complications an important area of interest. Age, dementia and preoperative delirium are significant risk factors
Corresponding author at: Department of Anesthesiology, University of California, San Diego, 9500 Gilman Drive, MC 0881, La Jolla, CA 92093-0881, USA. E-mail address: [email protected] (R.A. Gabriel).
https://doi.org/10.1016/j.jclinane.2019.08.015 Received 19 April 2019; Received in revised form 26 June 2019; Accepted 12 August 2019 0952-8180/ © 2019 Published by Elsevier Inc.
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
Fig. 1. Patient enrollment flow chart. ACS NSQIP = American College of Surgeons National Surgical Quality Improvement Program database.
participant user file. Cases were defined with Current Procedural Terminology (CPT) code of 27,236, 27,244, and 27,245. The study population was defined as patients ≥65 years of age who underwent hip fracture surgery in 2016. The outcomes included in the ACS NSQIP database were 30-day postoperative medical complications (delirium, pulmonary, infection, renal, vascular, and cardiac), ability to tolerate weight bearing on postoperative day 1, extended length of hospital stay that we defined as ≥75th percentile for the cohort (≥8 days), readmission; defined as readmission within 30 days of the principal procedure, and mortality; defined as death within 30 days of the principal procedure. The covariates we adjusted for included sex, body mass index, age, smoking status, steroid use, dyspnea, functional status, diabetes, chronic obstructive pulmonary disease, congestive heart failure, hypertension, kidney disease, bleeding disorder, dementia, bone medication, pressure sore, current procedural terminology code, American Society of Anesthesiologists (ASA) classification score, use of regional anesthesia, medical co-management, and care program.
for postoperative delirium [5–9]. Postoperative delirium in elderly patients undergoing hip surgery is associated with longer hospital stay, impaired functional and cognitive recovery, increased complications, increased institutionalization, higher costs and higher 6-month mortality [8]. Postoperative delirium has also been correlated with higher rates of discharge to and longer stay in inpatient facilities, hospital readmission, and hospital length of stay [6]. The association between postoperative delirium and clinical outcomes in elderly patients undergoing surgery for hip fracture has been well described in the literature [8,10]. However, a detailed understanding of the explicit association between preoperative delirium and postoperative outcomes in this population has yet to be thoroughly studied. To our knowledge, no study has analyzed the relationship between preoperative delirium and postoperative 30-day morbidity and mortality in the context of hip surgery for elderly patients with hip fractures. Thus, the primary endpoint of our study aims to use the National Surgical Quality Improvement Program (NSQIP) database to identify whether such an association exists. We hypothesized that preoperative delirium is an independent risk factor associated with postoperative 30-day morbidity and mortality among elderly hip fracture patients undergoing hip surgery.
2.3. Statistical analysis R (version 3.3.2) was the statistical computing software used to perform all statistical analysis discussed in our study. The tableone package and library in R was utilized to create the frequency tables of delirium cohorts. A univariable logistic regression was used to evaluate the association of preoperative delirium with certain demographic variables, preoperative characteristics, and postoperative complications. This was followed by a multivariable logistic regression to explore the association of preoperative delirium with our specified postoperative outcomes while controlling for demographic variables, preoperative medical comorbidities, surgery (open treatment of femoral fracture, peri-trochanteric femoral fracture with plate/screw type implant, and peri-trochanteric femoral fracture with intramedullary implant), and postoperative variables. A 2-tailed significance level was set at P < 0.05. The odds ratio (OR) with associated 95% confidence interval (CI) is reported for each covariate.
2. Methods 2.1. Data collection Data was collected from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) registry. This national prospective database contains demographic data (age, sex, race, ethnicity), medical history, and perioperative surgical data. The ACS NSQIP omits identifiers from the data sets as defined by the Health Insurance Portability and Accountability Act Privacy Rule. Patient health information is de-identified, and therefore the present study was exempt from approval by our institutional review board. 2.2. Study population
3. Results We merged the 2016 hip fracture procedure targeted patient user file collected from 117 United States hospitals with the ACS NSQIP
Of the 9390 hip surgery cases in the NSQIP database, a population 29
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
Table 1 Study characteristics following hip surgery in the elderly. Sample No. (%) Preoperative delirium
Demographics Male Body mass index < 18.5 ≥18.5 to < 25 ≥25 to < 30 ≥30 Unknown Age (mean (SD)) Age ≥65 to < 75 ≥75 to < 85 ≥85 Preoperative characteristics Active smoker Chronic steroid use Dyspnea Functional status Independent Partially dependent Completely dependent Unknown Diabetes mellitus COPD CHF HTN Kidney disease Bleeding disorder Dementia Bone protection medication Pressure sore Surgery Open treatment of femoral fracture Peri-trochanteric femoral fracture with plate/screw type implant Peri-trochanteric femoral fracture with intramedullary implant ASA class ≤3 – Severe disturbance ≥4 – Life threatening Unknown Regional anesthesia Medical co-management No Yes-co-management throughout stay Yes-partial co-management during stay Perioperative care program Postoperative complications Ability to tolerate weight bearing on postoperative day 1 Delirium Pulmonary Infectious Renal Vascular Cardiac Transfusion Hospital stay (median [IQR]) Extended hospital stay Readmission Mortality
Study Sample
No
Yes
P Value
8466
7391
1075
2414 (28.5)
2090 (28.3)
324 (30.1)
598 (7.1) 3489 (41.2) 2026 (23.9) 1101 (13.0) 1252 (14.8) 82.67 (7.24)
491 (6.6) 3016 (40.8) 1822 (24.7) 1012 (13.7) 1050 (14.2) 82.37 (7.33)
107 (10.0) 473 (44.0) 204 (19.0) 89 (8.3) 202 (18.8) 84.72 (6.18)
1443 (17.0) 2746 (32.4) 4277 (50.5)
1338 (18.1) 2448 (33.1) 3605 (48.8)
105 (9.8) 298 (27.7) 672 (62.5)
717 (8.5) 450 (5.3) 648 (7.7)
640 (8.7) 402 (5.4) 559 (7.6)
77 (7.2) 48 (4.5) 89 (8.3)
6457 (76.3) 1665 (19.7) 262 (3.1) 82 (1.0) 1488 (17.6) 909 (10.7) 344 (4.1) 5849 (69.1) 146 (1.7) 1522 (18.0) 2665 (31.5) 2493 (29.4) 287 (3.4)
5881 (79.6) 1257 (17.0) 197 (2.7) 56 (0.8) 1317 (17.8) 778 (10.5) 288 (3.9) 5117 (69.2) 126 (1.7) 1314 (17.8) 1880 (25.4) 2116 (28.6) 221 (3.0)
576 (53.6) 408 (38.0) 65 (6.0) 26 (2.4) 171 (15.9) 131 (12.2) 56 (5.2) 732 (68.1) 20 (1.9) 208 (19.3) 785 (73.0) 377 (35.1) 66 (6.1)
3046 (36.0) 1167 (13.8) 4253 (50.2)
2635 (35.7) 1030 (13.9) 3726 (50.4)
411 (38.2) 137 (12.7) 527 (49.0)
6519 (77.0) 1932 (22.8) 15 (0.2) 1364 (16.1)
5773 (78.1) 1605 (21.7) 13 (0.2) 1217 (16.5)
746 (69.4) 327 (30.4) 2 (0.2) 147 (13.7)
841 (9.9) 6222 (73.5) 1403 (16.6) 4646 (54.9)
779 (10.5) 5400 (73.1) 1212 (16.4) 4093 (55.4)
62 (5.8) 822 (76.5) 191 (17.8) 553 (51.4)
0.017
5616 (66.3) 2574 (30.4) 443 (5.2) 198 (2.3) 461 (5.4) 164 (1.9) 303 (3.6) 2399 (28.3) 5 [4, 8] 2309 (27.4) 695 (8.2) 590 (7.0)
5065 (68.5) 1703 (23.0) 343 (4.6) 160 (2.2) 413 (5.6) 139 (1.9) 266 (3.6) 2096 (28.4) 5 [4, 8] 1939 (26.3) 579 (7.8) 430 (5.8)
551 (51.3) 871 (81.0) 100 (9.3) 38 (3.5) 48 (4.5) 25 (2.3) 37 (3.4) 303 (28.2) 6 [4, 9] 370 (34.6) 116 (10.8) 160 (14.9)
< 0.001 < 0.001 < 0.001 0.008 0.149 0.384 0.864 0.935 < 0.001 < 0.001 0.001 < 0.001
0.22 < 0.001
< 0.001 < 0.001
0.112 0.209 0.445 < 0.001
0.135 0.112 0.051 0.471 0.81 0.226 < 0.001 < 0.001 < 0.001 0.216
< 0.001
0.023 < 0.001
Abbreviations = IQR (interquartile range), SD (standard deviation).
characteristics of the study population. Our population was predominantly female at 71.5% (n = 6052). Approximately half of the patients were 85 or older (4277, 50.5%). In terms of operative risk, there were 1932 (22.8%) with ASA class 4 or above. About one-half of our total population received an intramedullary implant for peri-trochanteric femoral fracture (4253, 50.2%), while 13.8% received a
of 8466 patients undergoing hip surgery for hip fracture was selected. (Fig. 1) Fig. 1 summarizes how the total sample size was broken down into each population group of interest. The 924 patients not included in our study were younger than 65 years old. The overall prevalence of preoperative delirium was 12.7% (n = 1075) Table 1 illustrates the overall distribution of population demographics and clinical
30
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
4. Discussion
plate/screw type implant (n = 1167). The remaining patients underwent an open treatment for femoral fracture (3046, 36.0%). In an age based comparison of patient populations with and without preoperative delirium, the prevalence of preoperative delirium was lower in the population aged 65 to under 75 (7.3% with, 92.7% without p < 0.001). However, as the population age increased, there was a significant increase in prevalence of preoperative delirium compared to no preoperative delirium past 85 years of age (15.7% with, 84.3% without, p < 0.001). The prevalence of dementia was significantly higher in the population with preoperative delirium (73%) compared to 25.4% prevalence of dementia in patients without preoperative delirium. The rate of readmission in patients with preoperative delirium was significantly higher than those patients without preoperative delirium. (10.8% with, 7.8 without, p = 0.001) Within their respective populations, it was found that 51.3% of patients with preoperative delirium were able to tolerate weight bearing on postoperative day 1 while 68.5% of patients without preoperative delirium were unable to tolerate weight bearing on postoperative day 1. Mortality was also noted to be significantly higher in patients with preoperative delirium (14.9% with, 5.8% without, p < 0.001) (Table 1). Results from the multivariable regression analysis of the data are summarized in Table 2. Each model was adjusted for sex, body mass index, age, smoking status, steroid use, dyspnea, functional status, diabetes, chronic obstructive pulmonary disease, congestive heart failure, hypertension, kidney disease, bleeding disorder, dementia, bone medication, pressure sore, CPT code, ASA class, regional anesthesia, medical co-management, and care program. Once adjusted, the odds of postoperative delirium (OR 9.38, 95% CI 7.94–11.14), pulmonary complications (OR 1.83, 95% CI 1.4–2.36), extended hospital stay (OR 1.47, 95% CI 1.26–1.72), readmission (OR 1.27, 95% CI 1.01–1.59) and mortality (OR 1.92, 95% CI 1.54–2.39) were all significantly higher in patients with preoperative delirium compared to those without preoperative delirium. The odds of tolerating weight bearing on postoperative day 1 was lower in patients with preoperative delirium (OR 0.66, 95% CI 0.57–0.76). After adjustments, the odds of postoperative infectious (OR 1.43, 95% CI 0.96–2.11), vascular (OR 1.49, 95% CI 0.91–2.35), renal (OR 0.85, 95% CI 0.61–1.17), and cardiac (OR 0.82, 95% CI 0.56–1.18) complications and receipt of postoperative transfusions (OR 0.9, 95% CI 0.77–1.06) were not significantly different among the delirium cohorts (Table 2).
Our analysis revealed associations between preoperative delirium and 30-day postoperative outcomes in elderly patients undergoing surgery for hip fracture. This cross-sectional analysis showed a 12.7% prevalence of preoperative delirium. After adjusting for demographics, comorbidities, and common postoperative hospital recovery complications, our multivariable regression analysis displayed decreased odds of the ability to tolerate weight bearing postoperative day 1 and increased odds of postoperative delirium, pulmonary complications, extended hospital stays, hospital readmission, and mortality in patients with preoperative delirium. There have been very few studies done on postoperative outcomes in elderly patients with preoperative delirium. It is estimated that the prevalence of delirium in elderly patients 65 and older stands between 11 and 42% in medical wards, with likely similar numbers in domestic homes, nursing homes, and long term care facilities [8,11]. The prevalence of dementia worldwide may reach over 81 million in elderly patients by 2040. An individual with dementia is up to 3 times more likely than a cognitively intact older adult to sustain a hip fracture [12]. Given the prevalence of both hip fractures and delirium within the elderly population, it is imperative we study the outcomes of treatment to provide improved care in terms of screening, diagnosis, and surgical optimization to minimize adverse effects. Our data suggest a positive correlation with preoperative delirium with the odds of postoperative delirium. In a retrospective study assessing preoperative risk factors for postoperative delirium in five hundred patients undergoing elective surgery, it was shown that preexisting cognitive impairment before surgery carried a relative risk of 3.1 for developing postoperative delirium. The same study also reported a relative risk of 4.1 of developing postoperative delirium in patients undergoing elective surgery who had a prior history of delirium [7,13,14]. A possible explanation as to why this relationship exists may revolve around the need for updated protocols for management of preoperative delirium differently than inpatient delirium [15–17]. When establishing these protocols, it is important to also address non-medical management such as correcting cognitive orientation, sleep deprivation, immobility, audio-visual impairment, and dehydration acutely before surgery [17,18]. It is also important to consider the duration of delirium as this can affect the continuation of effects postoperatively. As a result, another possible mechanism for this relationship could stem from preoperative delirium that has a duration lasting beyond the surgical procedure and thus presenting as postoperative delirium as well. It must also be considered that the stressors of surgery can also precipitate a state of postoperative delirium and care must be taken to separate this occurrence from patients who are already delirious before their operation. Given the incidence of delirium discussed above, optimization of risk factors for preoperative and postoperative delirium serves as an opportunity to greatly limit preoperative delirium and reduce surgical morbidity in elderly patients [17,19]. The return of functional ability in patients after surgery often serves as a metric for postoperative recovery [20,21]. Our study suggests a negative correlation between preoperative delirium and the ability to tolerate weight bearing on day after surgery. It is important to note, patients who could not tolerate weight bearing at baseline due to other primary medical concerns or who were bed-ridden were separated into their own population and not included for the following discussion. A potential explanation for this relationship may be due to the decreased cognitive ability of patients suffering from delirium. These patients often have difficulty understanding the limits of their body during recovery and as a result are more prone to injuring their surgical sites [5,12] . Great care must be taken with delirious patients, as adverse events can lead to prolonged recovery time and elevated costs of care for both the patient and the hospital [19,22]. The incidence of postoperative pulmonary complications in elderly
Table 2 The Association of Preoperative Delirium with Postoperative Outcomes following Hip Surgery in the Elderly.
Delirium Ability to tolerate weight bearing on postoperative day 1 Pulmonary Extended hospital stay Infection Renal Vascular Cardiac Transfusion Readmission Mortality
OR (95% CI)
P value
9.39(7.94–11.15) 0.66(0.57–0.76)
< 0.001 < 0.001
1.82(1.40–2.36) 1.47(1.26–1.72) 1.43(0.95–2.09) 0.85(0.61–1.16) 1.48(0.91–2.34) 0.81(0.55–1.17) 0.90(0.77–1.05) 1.27(1.01–1.60) 1.90(1.53–2.36)
< 0.001 < 0.001 0.077 0.322 0.100 0.275 0.196 0.040 < 0.001
Abbreviations = OR (odds ratio), CI (confidence interval). Each model adjusted for sex, body mass index, age, smoking status, steroid use, dyspnea, functional status, diabetes, COPD, CHF, HTN, kidney disease, bleeding disorder, dementia, bone medication, pressure sore, CPT code, ASA class, regional anesthesia, medical co-management, and care program. Extended Hospital Stay defined as greater than 75th percentile for the cohort. 31
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
In addition to the variable discussed above, preoperative delirium was also significantly associated with postoperative mortality in elderly patients who underwent surgery for hip fracture. It is widely accepted that surgical procedures often come with risks of infection, injury to local tissue, and many other complications that can increase the risk of postoperative mortality. As mortality is one of the most serious adverse outcomes of any surgery, it is imperative that these correlations continue to be studied to prevent their occurrence. There have been multiple studies establishing an association between delirium and postoperative mortality [17,35,36]. It is important to note however many of these studies look at postoperative delirium. One such study found an odds ratio of 2.22 of increased 30-day mortality in geriatric hip fracture patients experiencing postoperative delirium [6]. A probable reason for the correlation stems from the altered mental status and diminished functional status in patients with delirium. These factors can negatively influence recovery through improper self-care and health maintenance during the acute phase of recovery. Additional study in specifically preoperative delirium and its association with postoperative mortality is indicated as early identification can serve as a method of prevention of such adverse outcomes [6,37]. NSQIP has been used to evaluate outcomes across surgical disciplines and it is an excellent source for evaluating outcomes among elderly patients. However, there are limitations to our study, namely its retrospective nature. NSQIP does not include several important perioperative variables that would allow for a more robust analysis of anesthesia type with 30-day outcomes such as lesion morphology and location, surgeon's experience, hospital volume and location, intraoperative anesthetic agents delivered, and anesthesiologist's clinical experience. Additionally, NSQIP has program demands that may limit the participation of smaller community hospitals. Another limitation that comes with the use of the NSQIP database includes the heterogenous nature of screening and coding that comes with using ICD-9 codes. This can lead to over and underreporting of preoperative and postoperative delirium as well as inconsistent reporting of the many covariates and variables that come with such a study. There is a significant opportunity for the improvement of screening and recording methods so that future retrospective and prospective studies may analyze data to a higher degree of accuracy. After controlling for potential confounding variables, we showed that preoperative delirium was associated with several postoperative outcomes, namely postoperative delirium, ability to tolerate weight bearing on postoperative day 1, pulmonary complications, extended hospital stay, hospital readmission, and incidence of mortality. Taking into account such a wide array of complications, it can be suggested for future study to explore whether delaying surgery to fully optimize delirious patients may lead to significant improvements in postoperative outcomes. To our knowledge, this study is the first of its kind to explore preoperative delirium and 30-day postoperative outcomes in geriatric hip fracture patients undergoing surgery. From our literature review, there currently exists a multitude of data that has explored the associations of postoperative delirium with various postoperative outcomes. Given the lack of study of preoperative delirium and its postoperative outcomes, our data provides a strong starting point for further study as well as the development and implementation of targeted risk-reduction programs.
patients undergoing surgery has been reported to be as high as 40% [22,23]. Due to this large incidence, it is important we identify and optimize risk factors predisposing to these morbidities. Davies and Hussain et al. conducted a study on outcomes in patients undergoing non-cardiothoracic surgery and reported a “fair” association between patients with preoperative delirium and the incidence of postoperative pulmonary complications [9,24]. A potential mechanism for explaining this relationship stems from the increased risk of aspiration in patients suffering from acute delirium. These patients experience lethargy, agitation, and altered mental status during their episodes of delirium [25–27]. Combining these factors with the reduced respiratory drive and sedating effects of general anesthesia can reduce laryngeal protective reflexes and further increase the risk of aspiration, placing the patient at risk of developing postoperative pulmonary complications [28]. Perioperative care must be taken to avoid medications that can reduce respiratory drive and laryngeal protective reflexes when possible [29,30]. When discussing postoperative outcomes, an important metric commonly measured includes length of hospital stay. An extended length of stay directly affects both providers and consumers in terms of morbidity, mortality, quality of care, and healthcare costs [31,32]. This study analyzed the correlation between preoperative delirium and postoperative extended hospital stay in efforts to help elucidate factors that may be contributing to lengthening admissions. The data collected demonstrates an association, suggesting preoperative delirium has a direct influence on extended hospital stay. When consulting other studies on literature review, Thomas, Cameron et al. reported an average increase of thirteen days from expected hospital stay in their retrospective analysis of 133 hospitalized delirious patients. In another retrospective cohort study of 533 patients undergoing hip fracture surgery, it was found that patients with a diagnosis of delirium were hospitalized for a median of six more days than expected [8]. A proposed mechanism for such a correlation lies within the comorbidities faced by patients with delirium. These include likely greater functional dependency, reduced ability to ambulate, increased requirement of sedating medication, and poor nutritional state [21]. These findings are critical for both healthcare providers and consumers as they directly affect the level of care required as well as costs related to extended stay. It is crucial for future research trials to further evaluate these correlations to better serve this specialized population of patients. Another metric utilized for tracking postoperative outcomes of patients with preoperative delirium included studying incidence of readmission in both populations of patients. It is important to consider readmission in patients as readmission comes with increased morbidity, direct influences on overall patient outcomes, and increased costs and strains on overburdened hospital systems [6,33]. Many healthcare institutions also utilize readmission rates as a method of measuring the ability of their personnel [34]. A possible reason for such a correlation stems from the comorbidities that exist alongside patients with delirium, namely an altered mental status and loss of coordination. These factors place postoperative patients at an increased risk of exacerbating their injuries as well as prolonging recovery time. Our results are supported by other preceding literature that also reports increased readmission rates in delirious patients undergoing surgery [33]. Kates, Behrend et al. conducted a retrospective review of over 1000 patients and reported a 65% increase in the odds of 30-day readmission in delirious patients who underwent surgical procedures for hip fracture. Of the patients in this study, it was observed that 18.6% of readmissions were for surgical concerns and that delirium was a significant independent contributor to these admissions. These results align with the results observed in our study and our study goes further to link specifically preoperative delirium as another potential contributor. It is important to consider this correlation and use the data from similar studies to create a multi-disciplinary approach towards identifying patients at risk and providing care that prevents readmission to aid both the patient as well as hospitals.
Author contributions Shubham Agrawal, BS Contribution: this author helped with literature search, analysis of data, manuscript preparation, and review of manuscript. Robby Turk, BS Contribution: this author helped with literature search, analysis of 32
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
data, manuscript preparation, and review of manuscript.
[15] Aldecoa C, et al. European Society of Anaesthesiology evidence-based and consensus-based guideline on postoperative delirium. European Journal of Anaesthesiology (EJA) 2017;34(4):192. [16] Yang Y, et al. Risk factors for postoperative delirium following hip fracture repair in elderly patients: a systematic review and meta-analysis. Aging Clin Exp Res 2017;29(2):115–26. [17] Sieber FE, Barnett SR. Preventing postoperative complications in the elderly. Anesthesiol Clin 2011;29(1):83–97. [18] Inouye SK, et al. A multicomponent intervention to prevent delirium in hospitalized older patients. 2008. [19] Robles MJ, Formiga F, Vidán MT. Delirium prevention and treatment in elderly hip fracture. Medicina Clinica 2014;142(8):365–9. [20] Mazzola P, et al. Association between preoperative malnutrition and postoperative delirium after hip fracture surgery in older adults. J Am Geriatr Soc 2017;65(6):1222–8. [21] Mohanty S, et al. Optimal perioperative management of the geriatric patient: a best practices guideline from the American College of Surgeons NSQIP and the American Geriatrics Society. J Am Coll Surg 2016;222(5):930–47. [22] Ergina PL, Gold SL, Meakins JL. Perioperative care of the elderly patient. World J Surg 1993;17(2):192–8. [23] Gurunathan U, Marudhachalam K, Reynolds K. Preoperative delirium: not to be overlooked. Anesthesia & Analgesia 2016;122(5):1727. [24] Qaseem A, et al. Risk assessment for and strategies to reduce perioperative pulmonary complications for patients undergoing noncardiothoracic surgery: a guideline from the American College of Physicians. Ann Intern Med 2006;144(8):575. [25] Fong TG, Tulebaev SR, Inouye SK. Delirium in elderly adults: diagnosis, prevention and treatment. Nat Rev Neurol 2009;5(4):210–20. [26] Kukreja D, Günther U, Popp J. Delirium in the elderly: current problems with increasing geriatric age. Indian J Med Res 2015;142(6):655–62. [27] Neerland BE, et al. Perioperative hemodynamics and risk for delirium and new onset dementia in hip fracture patients; a prospective follow-up study. PLoS ONE 2017;12(7). [28] Lawrence VA, Cornell JE, Smetana GW. Strategies to reduce postoperative pulmonary complications after noncardiothoracic surgery: systematic review for the American College of Physicians. Ann Intern Med 2006;144(8):596. [29] Miskovic A, Lumb AB. Postoperative pulmonary complications. Br J Anaesth 2017;118(3):317–34. [30] Nason KS. Acute intraoperative pulmonary aspiration. Thorac Surg Clin 2015;25(3):301–7. [31] McGirt MJ, et al. Predictors of extended length of stay, discharge to inpatient rehab, and hospital readmission following elective lumbar spine surgery: introduction of the Carolina-Semmes grading scale. J Neurosurg Spine 2017;27(4):382–90. [32] Parikh, S.S. and F. Chung, Postoperative delirium in the elderly. Anesth Analg: p. 10. [33] Kates SL, et al. Hospital readmission after hip fracture. Arch Orthop Trauma Surg 2015;135(3):329–37. [34] Lagoo-Deenadayalan SA, Newell MA, Pofahl WE. Common perioperative complications in older patients. In: Rosenthal RA, Zenilman ME, Katlic MR, editors. Principles and practice of geriatric surgery. New York: New York, NY: Springer; 2011. p. 361–76. [35] Minden SL, et al. Predictors and outcomes of delirium. Gen Hosp Psychiatry 2005;27(3):209–14. [36] Thomas RI, Cameron DJ, Fahs MC. A prospective study of delirium and prolonged hospital stay: exploratory study. Arch Gen Psychiatry 1988;45(10):937–40. [37] Kim JY, et al. Risk factors and clinical outcomes of delirium in osteoporotic hip fractures. J Orthop Surg (Hong Kong) 2017;25(3). [p. 2309499017739485].
Brittany N. Burton, MHS, MAS Contribution: this author helped with literature search, analysis of data, manuscript preparation, and review of manuscript. Jerry Ingrande, MD, MS Contribution: this author helped with analysis of data, manuscript preparation, and review of manuscript. Rodney A. Gabriel, MD, MS Contribution: this author helped with literature search, data collection, study design, analysis of data, manuscript preparation, and review of manuscript. References [1] Gustafson Y, et al. Acute confusional states in elderly patients treated for femoral neck fracture. J Am Geriatr Soc 1988;36(6):525–30. [2] Melton LJ. Who has osteoporosis? A conflict between clinical and public health perspectives. Journal of Bone and Mineral Research: The Official Journal of the American Society for Bone and Mineral Research 2000;15(12):2309–14. [3] Tay E. Hip fractures in the elderly: operative versus nonoperative management. Singapore Med J 2016;57(4):178–81. [4] Karagas MR, et al. Heterogeneity of hip fracture: age, race, sex, and geographic patterns of femoral neck and trochanteric fractures among the US elderly. Am J Epidemiol 1996;143(7):677–82. [5] Adunsky A, et al. The unfavorable nature of preoperative delirium in elderly hip fractured patients. Arch Gerontol Geriatr 2003;36(1):67–74. [6] Arshi A, et al. Predictors and sequelae of postoperative delirium in geriatric hip fracture patients. Geriatric Orthopaedic Surgery & Rehabilitation 2018;9. [7] Litaker D, et al. Preoperative risk factors for postoperative delirium. Gen Hosp Psychiatry 2001;23(2):84–9. [8] Mosk CA, et al. Dementia and delirium, the outcomes in elderly hip fracture patients. Clin Interv Aging 2017;12:421–30. [9] Davies OJ, Husain T, Stephens RCM. Postoperative pulmonary complications following non-cardiothoracic surgery. BJA Education 2017;17(9):295–300. [10] Mantz J, Hemmings HC, Boddaert J. Case scenario: postoperative delirium in elderly surgical patients. Anesthesiology: The Journal of the American Society of Anesthesiologists 2010;112(1):189–95. [11] Juliebø V, et al. Risk factors for preoperative and postoperative delirium in elderly patients with hip fracture. J Am Geriatr Soc 2009;57(8):1354–61. [12] Friedman SM, et al. Dementia and hip fractures. Geriatric Orthopaedic Surgery & Rehabilitation 2010;1(2):52–62. [13] Denny DL, Lindseth G. Preoperative risk factors for Subsyndromal delirium in older adults who undergo joint replacement surgery. Orthop Nurs 2017;36(6):402–11. [14] Kassie GM, et al. Preoperative medication use and postoperative delirium: a systematic review. BMC Geriatr 2017;17.
33
Contents lists available at ScienceDirect
Journal of Clinical Anesthesia journal homepage: www.elsevier.com/locate/jclinane
Original Contribution
The association of preoperative delirium with postoperative outcomes following hip surgery in the elderly
T
Shubham Agrawal (BS)a, Robby Turk (BS)a, Brittany N. Burton (MHS, MAS)a, ⁎ Jerry Ingrande (MD, MS)b, Rodney A. Gabriel (MD, MAS)b,c, a
School of Medicine, University of California San Diego, San Diego, CA, USA Department of Anesthesiology, Division of Regional Anesthesia and Acute Pain, University of California, San Diego, San Diego, CA, USA c Department of Medicine, Division of Biomedical Informatics, University of California, San Diego, San Diego, CA, USA b
A R T I C LE I N FO
A B S T R A C T
Keywords: Preoperative delirium Postoperative complications Hip surgery ACS NSQIP
Study objective: To determine the association of preoperative delirium with postoperative outcomes following hip surgery in the elderly. Design: Retrospective cohort study. Setting: Postoperative recovery. Patients: 8466 patients all of whom were 65 years of age or older undergoing surgical repair of a femoral fracture. Of the total population studied, 1075 had preoperative delirium. Of those with preoperative delirium, 746 were ASA class 3 or below and 327 were ASA class 4 or above. Of the 7391 patients without preoperative delirium, 5773 were ASA class 3 or below and 1605 were ASA class 4 or above. The remainder in each group was of unknown ASA class. Interventions: We used multivariable logistic regression to explore the association of preoperative delirium with 30-day postoperative outcomes. The odds ratio (OR) with associated 95% confidence interval (CI) was reported for each covariate. Measurements. Data was collected regarding the incidence of postoperative outcomes including: delirium, pulmonary complications, extended hospital stay, infection, renal complications, vascular complications, cardiac complications, transfusion necessity, readmission, and mortality. Main Results. After adjusting for potential confounders, the odds of postoperative delirium (OR 9.38, 95% CI 7.94–11.14), pulmonary complications (OR 1.83, 95% CI 1.4–2.36), extended hospital stay (OR 1.47, 95% CI 1.26–1.72), readmission (OR 1.27, 95% CI 1.01–1.59) and mortality (OR 1.92, 95% CI 1.54–2.39) were all significantly higher in patients with preoperative delirium compared to those without. Conclusions: After controlling for potential confounding variables, we showed that preoperative delirium was associated with postoperative delirium, pulmonary complications, extended hospital stay, hospital readmission, and mortality. Given the lack of studies on preoperative delirium and its postoperative outcomes, our data provides a strong starting point for further investigations as well as the development and implementation of targeted risk-reduction programs.
1. Introduction Delirium is a common complication in hospitalized patients, and occurs in roughly 61% of elderly patients with hip fracture [1]. The lifetime risk of hip fracture is 17.5% for women and 6% for men [2]. Mortality is as much as fourfold higher in patients who had non-operative treatment, thus surgery is indicated for most hip fracture
⁎
patients [3]. In the United States, femoral neck and intertrochanteric fractures occurred with increasing frequency as age progressed between 65 and 99 [4]. The number of persons aged > 70 years old is expected to double within the next three decades, which may lead to an increased volume of hip fractures and making associated complications an important area of interest. Age, dementia and preoperative delirium are significant risk factors
Corresponding author at: Department of Anesthesiology, University of California, San Diego, 9500 Gilman Drive, MC 0881, La Jolla, CA 92093-0881, USA. E-mail address: [email protected] (R.A. Gabriel).
https://doi.org/10.1016/j.jclinane.2019.08.015 Received 19 April 2019; Received in revised form 26 June 2019; Accepted 12 August 2019 0952-8180/ © 2019 Published by Elsevier Inc.
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
Fig. 1. Patient enrollment flow chart. ACS NSQIP = American College of Surgeons National Surgical Quality Improvement Program database.
participant user file. Cases were defined with Current Procedural Terminology (CPT) code of 27,236, 27,244, and 27,245. The study population was defined as patients ≥65 years of age who underwent hip fracture surgery in 2016. The outcomes included in the ACS NSQIP database were 30-day postoperative medical complications (delirium, pulmonary, infection, renal, vascular, and cardiac), ability to tolerate weight bearing on postoperative day 1, extended length of hospital stay that we defined as ≥75th percentile for the cohort (≥8 days), readmission; defined as readmission within 30 days of the principal procedure, and mortality; defined as death within 30 days of the principal procedure. The covariates we adjusted for included sex, body mass index, age, smoking status, steroid use, dyspnea, functional status, diabetes, chronic obstructive pulmonary disease, congestive heart failure, hypertension, kidney disease, bleeding disorder, dementia, bone medication, pressure sore, current procedural terminology code, American Society of Anesthesiologists (ASA) classification score, use of regional anesthesia, medical co-management, and care program.
for postoperative delirium [5–9]. Postoperative delirium in elderly patients undergoing hip surgery is associated with longer hospital stay, impaired functional and cognitive recovery, increased complications, increased institutionalization, higher costs and higher 6-month mortality [8]. Postoperative delirium has also been correlated with higher rates of discharge to and longer stay in inpatient facilities, hospital readmission, and hospital length of stay [6]. The association between postoperative delirium and clinical outcomes in elderly patients undergoing surgery for hip fracture has been well described in the literature [8,10]. However, a detailed understanding of the explicit association between preoperative delirium and postoperative outcomes in this population has yet to be thoroughly studied. To our knowledge, no study has analyzed the relationship between preoperative delirium and postoperative 30-day morbidity and mortality in the context of hip surgery for elderly patients with hip fractures. Thus, the primary endpoint of our study aims to use the National Surgical Quality Improvement Program (NSQIP) database to identify whether such an association exists. We hypothesized that preoperative delirium is an independent risk factor associated with postoperative 30-day morbidity and mortality among elderly hip fracture patients undergoing hip surgery.
2.3. Statistical analysis R (version 3.3.2) was the statistical computing software used to perform all statistical analysis discussed in our study. The tableone package and library in R was utilized to create the frequency tables of delirium cohorts. A univariable logistic regression was used to evaluate the association of preoperative delirium with certain demographic variables, preoperative characteristics, and postoperative complications. This was followed by a multivariable logistic regression to explore the association of preoperative delirium with our specified postoperative outcomes while controlling for demographic variables, preoperative medical comorbidities, surgery (open treatment of femoral fracture, peri-trochanteric femoral fracture with plate/screw type implant, and peri-trochanteric femoral fracture with intramedullary implant), and postoperative variables. A 2-tailed significance level was set at P < 0.05. The odds ratio (OR) with associated 95% confidence interval (CI) is reported for each covariate.
2. Methods 2.1. Data collection Data was collected from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) registry. This national prospective database contains demographic data (age, sex, race, ethnicity), medical history, and perioperative surgical data. The ACS NSQIP omits identifiers from the data sets as defined by the Health Insurance Portability and Accountability Act Privacy Rule. Patient health information is de-identified, and therefore the present study was exempt from approval by our institutional review board. 2.2. Study population
3. Results We merged the 2016 hip fracture procedure targeted patient user file collected from 117 United States hospitals with the ACS NSQIP
Of the 9390 hip surgery cases in the NSQIP database, a population 29
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
Table 1 Study characteristics following hip surgery in the elderly. Sample No. (%) Preoperative delirium
Demographics Male Body mass index < 18.5 ≥18.5 to < 25 ≥25 to < 30 ≥30 Unknown Age (mean (SD)) Age ≥65 to < 75 ≥75 to < 85 ≥85 Preoperative characteristics Active smoker Chronic steroid use Dyspnea Functional status Independent Partially dependent Completely dependent Unknown Diabetes mellitus COPD CHF HTN Kidney disease Bleeding disorder Dementia Bone protection medication Pressure sore Surgery Open treatment of femoral fracture Peri-trochanteric femoral fracture with plate/screw type implant Peri-trochanteric femoral fracture with intramedullary implant ASA class ≤3 – Severe disturbance ≥4 – Life threatening Unknown Regional anesthesia Medical co-management No Yes-co-management throughout stay Yes-partial co-management during stay Perioperative care program Postoperative complications Ability to tolerate weight bearing on postoperative day 1 Delirium Pulmonary Infectious Renal Vascular Cardiac Transfusion Hospital stay (median [IQR]) Extended hospital stay Readmission Mortality
Study Sample
No
Yes
P Value
8466
7391
1075
2414 (28.5)
2090 (28.3)
324 (30.1)
598 (7.1) 3489 (41.2) 2026 (23.9) 1101 (13.0) 1252 (14.8) 82.67 (7.24)
491 (6.6) 3016 (40.8) 1822 (24.7) 1012 (13.7) 1050 (14.2) 82.37 (7.33)
107 (10.0) 473 (44.0) 204 (19.0) 89 (8.3) 202 (18.8) 84.72 (6.18)
1443 (17.0) 2746 (32.4) 4277 (50.5)
1338 (18.1) 2448 (33.1) 3605 (48.8)
105 (9.8) 298 (27.7) 672 (62.5)
717 (8.5) 450 (5.3) 648 (7.7)
640 (8.7) 402 (5.4) 559 (7.6)
77 (7.2) 48 (4.5) 89 (8.3)
6457 (76.3) 1665 (19.7) 262 (3.1) 82 (1.0) 1488 (17.6) 909 (10.7) 344 (4.1) 5849 (69.1) 146 (1.7) 1522 (18.0) 2665 (31.5) 2493 (29.4) 287 (3.4)
5881 (79.6) 1257 (17.0) 197 (2.7) 56 (0.8) 1317 (17.8) 778 (10.5) 288 (3.9) 5117 (69.2) 126 (1.7) 1314 (17.8) 1880 (25.4) 2116 (28.6) 221 (3.0)
576 (53.6) 408 (38.0) 65 (6.0) 26 (2.4) 171 (15.9) 131 (12.2) 56 (5.2) 732 (68.1) 20 (1.9) 208 (19.3) 785 (73.0) 377 (35.1) 66 (6.1)
3046 (36.0) 1167 (13.8) 4253 (50.2)
2635 (35.7) 1030 (13.9) 3726 (50.4)
411 (38.2) 137 (12.7) 527 (49.0)
6519 (77.0) 1932 (22.8) 15 (0.2) 1364 (16.1)
5773 (78.1) 1605 (21.7) 13 (0.2) 1217 (16.5)
746 (69.4) 327 (30.4) 2 (0.2) 147 (13.7)
841 (9.9) 6222 (73.5) 1403 (16.6) 4646 (54.9)
779 (10.5) 5400 (73.1) 1212 (16.4) 4093 (55.4)
62 (5.8) 822 (76.5) 191 (17.8) 553 (51.4)
0.017
5616 (66.3) 2574 (30.4) 443 (5.2) 198 (2.3) 461 (5.4) 164 (1.9) 303 (3.6) 2399 (28.3) 5 [4, 8] 2309 (27.4) 695 (8.2) 590 (7.0)
5065 (68.5) 1703 (23.0) 343 (4.6) 160 (2.2) 413 (5.6) 139 (1.9) 266 (3.6) 2096 (28.4) 5 [4, 8] 1939 (26.3) 579 (7.8) 430 (5.8)
551 (51.3) 871 (81.0) 100 (9.3) 38 (3.5) 48 (4.5) 25 (2.3) 37 (3.4) 303 (28.2) 6 [4, 9] 370 (34.6) 116 (10.8) 160 (14.9)
< 0.001 < 0.001 < 0.001 0.008 0.149 0.384 0.864 0.935 < 0.001 < 0.001 0.001 < 0.001
0.22 < 0.001
< 0.001 < 0.001
0.112 0.209 0.445 < 0.001
0.135 0.112 0.051 0.471 0.81 0.226 < 0.001 < 0.001 < 0.001 0.216
< 0.001
0.023 < 0.001
Abbreviations = IQR (interquartile range), SD (standard deviation).
characteristics of the study population. Our population was predominantly female at 71.5% (n = 6052). Approximately half of the patients were 85 or older (4277, 50.5%). In terms of operative risk, there were 1932 (22.8%) with ASA class 4 or above. About one-half of our total population received an intramedullary implant for peri-trochanteric femoral fracture (4253, 50.2%), while 13.8% received a
of 8466 patients undergoing hip surgery for hip fracture was selected. (Fig. 1) Fig. 1 summarizes how the total sample size was broken down into each population group of interest. The 924 patients not included in our study were younger than 65 years old. The overall prevalence of preoperative delirium was 12.7% (n = 1075) Table 1 illustrates the overall distribution of population demographics and clinical
30
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
4. Discussion
plate/screw type implant (n = 1167). The remaining patients underwent an open treatment for femoral fracture (3046, 36.0%). In an age based comparison of patient populations with and without preoperative delirium, the prevalence of preoperative delirium was lower in the population aged 65 to under 75 (7.3% with, 92.7% without p < 0.001). However, as the population age increased, there was a significant increase in prevalence of preoperative delirium compared to no preoperative delirium past 85 years of age (15.7% with, 84.3% without, p < 0.001). The prevalence of dementia was significantly higher in the population with preoperative delirium (73%) compared to 25.4% prevalence of dementia in patients without preoperative delirium. The rate of readmission in patients with preoperative delirium was significantly higher than those patients without preoperative delirium. (10.8% with, 7.8 without, p = 0.001) Within their respective populations, it was found that 51.3% of patients with preoperative delirium were able to tolerate weight bearing on postoperative day 1 while 68.5% of patients without preoperative delirium were unable to tolerate weight bearing on postoperative day 1. Mortality was also noted to be significantly higher in patients with preoperative delirium (14.9% with, 5.8% without, p < 0.001) (Table 1). Results from the multivariable regression analysis of the data are summarized in Table 2. Each model was adjusted for sex, body mass index, age, smoking status, steroid use, dyspnea, functional status, diabetes, chronic obstructive pulmonary disease, congestive heart failure, hypertension, kidney disease, bleeding disorder, dementia, bone medication, pressure sore, CPT code, ASA class, regional anesthesia, medical co-management, and care program. Once adjusted, the odds of postoperative delirium (OR 9.38, 95% CI 7.94–11.14), pulmonary complications (OR 1.83, 95% CI 1.4–2.36), extended hospital stay (OR 1.47, 95% CI 1.26–1.72), readmission (OR 1.27, 95% CI 1.01–1.59) and mortality (OR 1.92, 95% CI 1.54–2.39) were all significantly higher in patients with preoperative delirium compared to those without preoperative delirium. The odds of tolerating weight bearing on postoperative day 1 was lower in patients with preoperative delirium (OR 0.66, 95% CI 0.57–0.76). After adjustments, the odds of postoperative infectious (OR 1.43, 95% CI 0.96–2.11), vascular (OR 1.49, 95% CI 0.91–2.35), renal (OR 0.85, 95% CI 0.61–1.17), and cardiac (OR 0.82, 95% CI 0.56–1.18) complications and receipt of postoperative transfusions (OR 0.9, 95% CI 0.77–1.06) were not significantly different among the delirium cohorts (Table 2).
Our analysis revealed associations between preoperative delirium and 30-day postoperative outcomes in elderly patients undergoing surgery for hip fracture. This cross-sectional analysis showed a 12.7% prevalence of preoperative delirium. After adjusting for demographics, comorbidities, and common postoperative hospital recovery complications, our multivariable regression analysis displayed decreased odds of the ability to tolerate weight bearing postoperative day 1 and increased odds of postoperative delirium, pulmonary complications, extended hospital stays, hospital readmission, and mortality in patients with preoperative delirium. There have been very few studies done on postoperative outcomes in elderly patients with preoperative delirium. It is estimated that the prevalence of delirium in elderly patients 65 and older stands between 11 and 42% in medical wards, with likely similar numbers in domestic homes, nursing homes, and long term care facilities [8,11]. The prevalence of dementia worldwide may reach over 81 million in elderly patients by 2040. An individual with dementia is up to 3 times more likely than a cognitively intact older adult to sustain a hip fracture [12]. Given the prevalence of both hip fractures and delirium within the elderly population, it is imperative we study the outcomes of treatment to provide improved care in terms of screening, diagnosis, and surgical optimization to minimize adverse effects. Our data suggest a positive correlation with preoperative delirium with the odds of postoperative delirium. In a retrospective study assessing preoperative risk factors for postoperative delirium in five hundred patients undergoing elective surgery, it was shown that preexisting cognitive impairment before surgery carried a relative risk of 3.1 for developing postoperative delirium. The same study also reported a relative risk of 4.1 of developing postoperative delirium in patients undergoing elective surgery who had a prior history of delirium [7,13,14]. A possible explanation as to why this relationship exists may revolve around the need for updated protocols for management of preoperative delirium differently than inpatient delirium [15–17]. When establishing these protocols, it is important to also address non-medical management such as correcting cognitive orientation, sleep deprivation, immobility, audio-visual impairment, and dehydration acutely before surgery [17,18]. It is also important to consider the duration of delirium as this can affect the continuation of effects postoperatively. As a result, another possible mechanism for this relationship could stem from preoperative delirium that has a duration lasting beyond the surgical procedure and thus presenting as postoperative delirium as well. It must also be considered that the stressors of surgery can also precipitate a state of postoperative delirium and care must be taken to separate this occurrence from patients who are already delirious before their operation. Given the incidence of delirium discussed above, optimization of risk factors for preoperative and postoperative delirium serves as an opportunity to greatly limit preoperative delirium and reduce surgical morbidity in elderly patients [17,19]. The return of functional ability in patients after surgery often serves as a metric for postoperative recovery [20,21]. Our study suggests a negative correlation between preoperative delirium and the ability to tolerate weight bearing on day after surgery. It is important to note, patients who could not tolerate weight bearing at baseline due to other primary medical concerns or who were bed-ridden were separated into their own population and not included for the following discussion. A potential explanation for this relationship may be due to the decreased cognitive ability of patients suffering from delirium. These patients often have difficulty understanding the limits of their body during recovery and as a result are more prone to injuring their surgical sites [5,12] . Great care must be taken with delirious patients, as adverse events can lead to prolonged recovery time and elevated costs of care for both the patient and the hospital [19,22]. The incidence of postoperative pulmonary complications in elderly
Table 2 The Association of Preoperative Delirium with Postoperative Outcomes following Hip Surgery in the Elderly.
Delirium Ability to tolerate weight bearing on postoperative day 1 Pulmonary Extended hospital stay Infection Renal Vascular Cardiac Transfusion Readmission Mortality
OR (95% CI)
P value
9.39(7.94–11.15) 0.66(0.57–0.76)
< 0.001 < 0.001
1.82(1.40–2.36) 1.47(1.26–1.72) 1.43(0.95–2.09) 0.85(0.61–1.16) 1.48(0.91–2.34) 0.81(0.55–1.17) 0.90(0.77–1.05) 1.27(1.01–1.60) 1.90(1.53–2.36)
< 0.001 < 0.001 0.077 0.322 0.100 0.275 0.196 0.040 < 0.001
Abbreviations = OR (odds ratio), CI (confidence interval). Each model adjusted for sex, body mass index, age, smoking status, steroid use, dyspnea, functional status, diabetes, COPD, CHF, HTN, kidney disease, bleeding disorder, dementia, bone medication, pressure sore, CPT code, ASA class, regional anesthesia, medical co-management, and care program. Extended Hospital Stay defined as greater than 75th percentile for the cohort. 31
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
In addition to the variable discussed above, preoperative delirium was also significantly associated with postoperative mortality in elderly patients who underwent surgery for hip fracture. It is widely accepted that surgical procedures often come with risks of infection, injury to local tissue, and many other complications that can increase the risk of postoperative mortality. As mortality is one of the most serious adverse outcomes of any surgery, it is imperative that these correlations continue to be studied to prevent their occurrence. There have been multiple studies establishing an association between delirium and postoperative mortality [17,35,36]. It is important to note however many of these studies look at postoperative delirium. One such study found an odds ratio of 2.22 of increased 30-day mortality in geriatric hip fracture patients experiencing postoperative delirium [6]. A probable reason for the correlation stems from the altered mental status and diminished functional status in patients with delirium. These factors can negatively influence recovery through improper self-care and health maintenance during the acute phase of recovery. Additional study in specifically preoperative delirium and its association with postoperative mortality is indicated as early identification can serve as a method of prevention of such adverse outcomes [6,37]. NSQIP has been used to evaluate outcomes across surgical disciplines and it is an excellent source for evaluating outcomes among elderly patients. However, there are limitations to our study, namely its retrospective nature. NSQIP does not include several important perioperative variables that would allow for a more robust analysis of anesthesia type with 30-day outcomes such as lesion morphology and location, surgeon's experience, hospital volume and location, intraoperative anesthetic agents delivered, and anesthesiologist's clinical experience. Additionally, NSQIP has program demands that may limit the participation of smaller community hospitals. Another limitation that comes with the use of the NSQIP database includes the heterogenous nature of screening and coding that comes with using ICD-9 codes. This can lead to over and underreporting of preoperative and postoperative delirium as well as inconsistent reporting of the many covariates and variables that come with such a study. There is a significant opportunity for the improvement of screening and recording methods so that future retrospective and prospective studies may analyze data to a higher degree of accuracy. After controlling for potential confounding variables, we showed that preoperative delirium was associated with several postoperative outcomes, namely postoperative delirium, ability to tolerate weight bearing on postoperative day 1, pulmonary complications, extended hospital stay, hospital readmission, and incidence of mortality. Taking into account such a wide array of complications, it can be suggested for future study to explore whether delaying surgery to fully optimize delirious patients may lead to significant improvements in postoperative outcomes. To our knowledge, this study is the first of its kind to explore preoperative delirium and 30-day postoperative outcomes in geriatric hip fracture patients undergoing surgery. From our literature review, there currently exists a multitude of data that has explored the associations of postoperative delirium with various postoperative outcomes. Given the lack of study of preoperative delirium and its postoperative outcomes, our data provides a strong starting point for further study as well as the development and implementation of targeted risk-reduction programs.
patients undergoing surgery has been reported to be as high as 40% [22,23]. Due to this large incidence, it is important we identify and optimize risk factors predisposing to these morbidities. Davies and Hussain et al. conducted a study on outcomes in patients undergoing non-cardiothoracic surgery and reported a “fair” association between patients with preoperative delirium and the incidence of postoperative pulmonary complications [9,24]. A potential mechanism for explaining this relationship stems from the increased risk of aspiration in patients suffering from acute delirium. These patients experience lethargy, agitation, and altered mental status during their episodes of delirium [25–27]. Combining these factors with the reduced respiratory drive and sedating effects of general anesthesia can reduce laryngeal protective reflexes and further increase the risk of aspiration, placing the patient at risk of developing postoperative pulmonary complications [28]. Perioperative care must be taken to avoid medications that can reduce respiratory drive and laryngeal protective reflexes when possible [29,30]. When discussing postoperative outcomes, an important metric commonly measured includes length of hospital stay. An extended length of stay directly affects both providers and consumers in terms of morbidity, mortality, quality of care, and healthcare costs [31,32]. This study analyzed the correlation between preoperative delirium and postoperative extended hospital stay in efforts to help elucidate factors that may be contributing to lengthening admissions. The data collected demonstrates an association, suggesting preoperative delirium has a direct influence on extended hospital stay. When consulting other studies on literature review, Thomas, Cameron et al. reported an average increase of thirteen days from expected hospital stay in their retrospective analysis of 133 hospitalized delirious patients. In another retrospective cohort study of 533 patients undergoing hip fracture surgery, it was found that patients with a diagnosis of delirium were hospitalized for a median of six more days than expected [8]. A proposed mechanism for such a correlation lies within the comorbidities faced by patients with delirium. These include likely greater functional dependency, reduced ability to ambulate, increased requirement of sedating medication, and poor nutritional state [21]. These findings are critical for both healthcare providers and consumers as they directly affect the level of care required as well as costs related to extended stay. It is crucial for future research trials to further evaluate these correlations to better serve this specialized population of patients. Another metric utilized for tracking postoperative outcomes of patients with preoperative delirium included studying incidence of readmission in both populations of patients. It is important to consider readmission in patients as readmission comes with increased morbidity, direct influences on overall patient outcomes, and increased costs and strains on overburdened hospital systems [6,33]. Many healthcare institutions also utilize readmission rates as a method of measuring the ability of their personnel [34]. A possible reason for such a correlation stems from the comorbidities that exist alongside patients with delirium, namely an altered mental status and loss of coordination. These factors place postoperative patients at an increased risk of exacerbating their injuries as well as prolonging recovery time. Our results are supported by other preceding literature that also reports increased readmission rates in delirious patients undergoing surgery [33]. Kates, Behrend et al. conducted a retrospective review of over 1000 patients and reported a 65% increase in the odds of 30-day readmission in delirious patients who underwent surgical procedures for hip fracture. Of the patients in this study, it was observed that 18.6% of readmissions were for surgical concerns and that delirium was a significant independent contributor to these admissions. These results align with the results observed in our study and our study goes further to link specifically preoperative delirium as another potential contributor. It is important to consider this correlation and use the data from similar studies to create a multi-disciplinary approach towards identifying patients at risk and providing care that prevents readmission to aid both the patient as well as hospitals.
Author contributions Shubham Agrawal, BS Contribution: this author helped with literature search, analysis of data, manuscript preparation, and review of manuscript. Robby Turk, BS Contribution: this author helped with literature search, analysis of 32
Journal of Clinical Anesthesia 60 (2020) 28–33
S. Agrawal, et al.
data, manuscript preparation, and review of manuscript.
[15] Aldecoa C, et al. European Society of Anaesthesiology evidence-based and consensus-based guideline on postoperative delirium. European Journal of Anaesthesiology (EJA) 2017;34(4):192. [16] Yang Y, et al. Risk factors for postoperative delirium following hip fracture repair in elderly patients: a systematic review and meta-analysis. Aging Clin Exp Res 2017;29(2):115–26. [17] Sieber FE, Barnett SR. Preventing postoperative complications in the elderly. Anesthesiol Clin 2011;29(1):83–97. [18] Inouye SK, et al. A multicomponent intervention to prevent delirium in hospitalized older patients. 2008. [19] Robles MJ, Formiga F, Vidán MT. Delirium prevention and treatment in elderly hip fracture. Medicina Clinica 2014;142(8):365–9. [20] Mazzola P, et al. Association between preoperative malnutrition and postoperative delirium after hip fracture surgery in older adults. J Am Geriatr Soc 2017;65(6):1222–8. [21] Mohanty S, et al. Optimal perioperative management of the geriatric patient: a best practices guideline from the American College of Surgeons NSQIP and the American Geriatrics Society. J Am Coll Surg 2016;222(5):930–47. [22] Ergina PL, Gold SL, Meakins JL. Perioperative care of the elderly patient. World J Surg 1993;17(2):192–8. [23] Gurunathan U, Marudhachalam K, Reynolds K. Preoperative delirium: not to be overlooked. Anesthesia & Analgesia 2016;122(5):1727. [24] Qaseem A, et al. Risk assessment for and strategies to reduce perioperative pulmonary complications for patients undergoing noncardiothoracic surgery: a guideline from the American College of Physicians. Ann Intern Med 2006;144(8):575. [25] Fong TG, Tulebaev SR, Inouye SK. Delirium in elderly adults: diagnosis, prevention and treatment. Nat Rev Neurol 2009;5(4):210–20. [26] Kukreja D, Günther U, Popp J. Delirium in the elderly: current problems with increasing geriatric age. Indian J Med Res 2015;142(6):655–62. [27] Neerland BE, et al. Perioperative hemodynamics and risk for delirium and new onset dementia in hip fracture patients; a prospective follow-up study. PLoS ONE 2017;12(7). [28] Lawrence VA, Cornell JE, Smetana GW. Strategies to reduce postoperative pulmonary complications after noncardiothoracic surgery: systematic review for the American College of Physicians. Ann Intern Med 2006;144(8):596. [29] Miskovic A, Lumb AB. Postoperative pulmonary complications. Br J Anaesth 2017;118(3):317–34. [30] Nason KS. Acute intraoperative pulmonary aspiration. Thorac Surg Clin 2015;25(3):301–7. [31] McGirt MJ, et al. Predictors of extended length of stay, discharge to inpatient rehab, and hospital readmission following elective lumbar spine surgery: introduction of the Carolina-Semmes grading scale. J Neurosurg Spine 2017;27(4):382–90. [32] Parikh, S.S. and F. Chung, Postoperative delirium in the elderly. Anesth Analg: p. 10. [33] Kates SL, et al. Hospital readmission after hip fracture. Arch Orthop Trauma Surg 2015;135(3):329–37. [34] Lagoo-Deenadayalan SA, Newell MA, Pofahl WE. Common perioperative complications in older patients. In: Rosenthal RA, Zenilman ME, Katlic MR, editors. Principles and practice of geriatric surgery. New York: New York, NY: Springer; 2011. p. 361–76. [35] Minden SL, et al. Predictors and outcomes of delirium. Gen Hosp Psychiatry 2005;27(3):209–14. [36] Thomas RI, Cameron DJ, Fahs MC. A prospective study of delirium and prolonged hospital stay: exploratory study. Arch Gen Psychiatry 1988;45(10):937–40. [37] Kim JY, et al. Risk factors and clinical outcomes of delirium in osteoporotic hip fractures. J Orthop Surg (Hong Kong) 2017;25(3). [p. 2309499017739485].
Brittany N. Burton, MHS, MAS Contribution: this author helped with literature search, analysis of data, manuscript preparation, and review of manuscript. Jerry Ingrande, MD, MS Contribution: this author helped with analysis of data, manuscript preparation, and review of manuscript. Rodney A. Gabriel, MD, MS Contribution: this author helped with literature search, data collection, study design, analysis of data, manuscript preparation, and review of manuscript. References [1] Gustafson Y, et al. Acute confusional states in elderly patients treated for femoral neck fracture. J Am Geriatr Soc 1988;36(6):525–30. [2] Melton LJ. Who has osteoporosis? A conflict between clinical and public health perspectives. Journal of Bone and Mineral Research: The Official Journal of the American Society for Bone and Mineral Research 2000;15(12):2309–14. [3] Tay E. Hip fractures in the elderly: operative versus nonoperative management. Singapore Med J 2016;57(4):178–81. [4] Karagas MR, et al. Heterogeneity of hip fracture: age, race, sex, and geographic patterns of femoral neck and trochanteric fractures among the US elderly. Am J Epidemiol 1996;143(7):677–82. [5] Adunsky A, et al. The unfavorable nature of preoperative delirium in elderly hip fractured patients. Arch Gerontol Geriatr 2003;36(1):67–74. [6] Arshi A, et al. Predictors and sequelae of postoperative delirium in geriatric hip fracture patients. Geriatric Orthopaedic Surgery & Rehabilitation 2018;9. [7] Litaker D, et al. Preoperative risk factors for postoperative delirium. Gen Hosp Psychiatry 2001;23(2):84–9. [8] Mosk CA, et al. Dementia and delirium, the outcomes in elderly hip fracture patients. Clin Interv Aging 2017;12:421–30. [9] Davies OJ, Husain T, Stephens RCM. Postoperative pulmonary complications following non-cardiothoracic surgery. BJA Education 2017;17(9):295–300. [10] Mantz J, Hemmings HC, Boddaert J. Case scenario: postoperative delirium in elderly surgical patients. Anesthesiology: The Journal of the American Society of Anesthesiologists 2010;112(1):189–95. [11] Juliebø V, et al. Risk factors for preoperative and postoperative delirium in elderly patients with hip fracture. J Am Geriatr Soc 2009;57(8):1354–61. [12] Friedman SM, et al. Dementia and hip fractures. Geriatric Orthopaedic Surgery & Rehabilitation 2010;1(2):52–62. [13] Denny DL, Lindseth G. Preoperative risk factors for Subsyndromal delirium in older adults who undergo joint replacement surgery. Orthop Nurs 2017;36(6):402–11. [14] Kassie GM, et al. Preoperative medication use and postoperative delirium: a systematic review. BMC Geriatr 2017;17.
33