Sage Journals: Discover world-class research

Abstract

Background

Dementia is widely recognized as a factor that influences the outcomes of total knee arthroplasty (TKA). However, recent studies have not fully elucidated the effects of dementia on perioperative medical and surgical complications, as well as in-hospital resource utilization. The objective of this study is to evaluate the specific impact of dementia on perioperative outcomes in elderly patients undergoing TKA.

Methods

Data on elderly patients undergoing TKA were extracted from the National Inpatient Sample database (2010-2019). Patients were categorized into a dementia group and a non-dementia group. Propensity score matching (PSM) was employed to control for differences in demographics, hospital characteristics, and comorbidities. Multivariate logistic regressions were conducted to evaluate perioperative outcomes.

Results

From 2010 to 2019, a group of 344,200 patients undergoing TKA was identified, among whom 6761 individuals (2.0%) were diagnosed with dementia. Following PSM, dementia remained an independent predictor for several perioperative outcomes, including urinary tract infection (aOR = 2.001, 95%CI = 1.561-2.564, P < 0.0001), pneumonia (aOR = 2.043, 95%CI = 1.244-3.355, P = 0.005), postoperative delirium (aOR = 5.854, 95%CI = 4.418-7.758, P < 0.0001), acute hemorrhagic anemia (aOR = 1.229, 95%CI = 1.128-1.340, P < 0.0001), prolonged hospital stay (aOR = 1.966, 95%CI = 1.785-2.165, P < 0.0001), and increased total healthcare charges (aOR = 1.122, 95%CI = 1.037-1.215, P = 0.004).

Conclusion

Dementia is independently associated with an increased risk of perioperative complications after TKA, underscoring the importance of tailored perioperative care for elderly patients with dementia.

Keywords

total knee arthroplasty dementia elderly patients national inpatient sample perioperative complications

Background

Total knee arthroplasty (TKA) is widely recognized as an effective intervention for end-stage knee osteoarthritis (KOA),¹ a degenerative disorder closely associated with advanced age.² As the population ages, the prevalence of KOA is expected to rise, leading to an increased demand for TKA.³ A previous study indicated that the demand for TKA in the United States is anticipated to surge by 673%, reaching approximately 3.48 millions procedures between 2005 and 2030.⁴ Despite high implant survival rates of 95-97% over 20 years,⁵ medications and comorbidities can significantly impact surgical outcomes and elevate the risk of complications,^6,7 highlighting the importance of identifying modifiable and non-modifiable risk factors.

A previous study showed that both osteoarthritis (OA) and dementia are conditions associated with aging, and there exists a correlation between OA and dementia.⁸ Dementia is a syndrome of acquired cognitive impairment, characterized by progressive decline in memory, language, and orientation, which impairs independent living and daily functioning.^9,10 The global population of individuals with dementia will rise from 57.4 millions in 2019 to 152.8 millions in 2050 due to population growth and aging.¹¹ Among those over 65 years, dementia prevalence is estimated at 7%, with higher rates observed in developed countries, reflecting longer life expectancies.¹⁰

In one study, patients with dementia experienced significant pain relief following joint replacement.¹² However, dementia was shown to be a risk factor for certain postoperative complications and increased mortality after surgical procedures.^13-17

In recent years, there has been a scarcity of large-scale studies examining whether elderly patients with dementia undergoing TKA face higher risks of medical and surgical complications, as well as increased in-hospital resource utilization. We hypothesized that dementia would increase the incidence of specific perioperative complications, prolong length of stay (LOS), and raise total charge (TOTCHG) in this population. Accordingly, we analyzed data from the National Inpatient Sample (NIS) database for the first time, comparing perioperative complications and in-hospital resource utilization between patients with and without dementia.

Methods and Materials

We used the STROBE cross-sectional checklist when writing our report.¹⁸

Study Design

This retrospective study was based on the NIS database and included patients over 65 years old who underwent TKA surgery for primary KOA between 2010 and 2019. According to whether they were diagnosed with dementia, eligible patients were stratified into a dementia group and a non-dementia group.

Data Source

Study data were obtained from the 2010-2019 NIS database, which is component of databases and software tools developed for the Healthcare Cost and Utilization Project (HCUP). As the largest publicly available, fully paid inpatient health care database in the United States, it contains 20.00% of all discharges in the United States, with approximately 8 millions acute admissions annually from 47 states and the District of Columbia.¹⁹ The database collects diagnoses and procedures coded according to the International Classification of Diseases, Ninth Revision (ICD-9) and 10th Revision (ICD-10), along with patients’ demographics, hospital characteristics, primary expected payer, LOS and TOTCHG.¹⁹

Study Cohort

Patients undergoing TKA were identified based on the principal procedure (PR1) using procedure codes from the International Classification of Diseases, Ninth Revision, Procedure Coding System (ICD-9-PCS: 81.54) and 10th Revision, Procedure Coding System (ICD-10-PCS: 0SRC, 0SRD).²⁰ Patients undergoing primary unilateral or bilateral TKA were eligible for inclusion. Patients were excluded if they had non-primary KOA (eg, secondary osteoarthritis or neoplasm of long bones of lower limb), were under 65 years old, had non-elective admissions, or had missing data for key variables (age, gender, insurance type, admission type, hospital bed size, teaching status, location, region, LOS, TOTCHG, or in-hospital mortality). To minimize confounding bias, 1:1 propensity score matching (PSM) was conducted. After matching, patients were categorized into a matched dementia group and a matched non-dementia group.

Baseline Data

We analyzed patients’ demographics, hospital characteristics and comorbidities within each group that could potentially affect perioperative outcomes. Patients’ demographics included age, age group, gender and race. Hospital characteristics encompassed insurance type, hospital bed size, teaching status, region and location. A total of 30 pre-existing comorbidities were assessed (Table 1) and the overall number of comorbidities was recorded. Of the 30 comorbidities, all were obtained from the database based on ICD-9 and ICD-10 codes.

Table 1.

Variables Used in Logistic Regression Analysis and Propensity Score Matching

Variables categories	Specific variables
Patients’ demographics	Age, age group (65-69 years, 70-79 years, ≥80 years), gender (male and female), Race (White, Black, Hispanic, Asian or Pacific Islander, Native American, Other)
Hospital characteristics	Type of insurance (medicare, medicaid, private insurance, self-pay, no charge, other), bed size of hospital (small, medium, large), teaching status of hospital (non-teaching, teaching), location of hospital (rural, urban), region of the hospital (northeast, Midwest or north central, south, west)
Comorbidities	Acquired immune deficiency syndrome, alcohol abuse, deficiency anemia, rheumatoid arthritis/collagen vascular diseases, chronic blood loss anemia, congestive heart failure, chronic pulmonary disease, coagulopathy, depression, diabetes (uncomplicated), diabetes (with chronic complications), drug abuse, hypertension, hypothyroidism, liver disease, lymphoma, fluid and electrolyte disorders, metastatic cancer, obesity, paralysis, peripheral vascular disorders, psychoses, pulmonary circulation disorders, renal failure, solid tumor without metastasis, peptic ulcer disease excluding bleeding, valvular disease, weight loss, drug-induced coagulopathy, and tobacco dependence
Other	Number of comorbidities (0, 1, 2, ≥3)

Outcomes

In this study, perioperative complications were selected as the primary outcomes, while LOS, TOTCHG, and in-hospital mortality were defined as secondary outcomes. The term “perioperative” was defined to encompass the duration from patient admission to discharge. Perioperative complications were identified using ICD-9 and ICD-10 codes (Table S1). Medical complications included acute renal failure, urinary tract infection, acute myocardial infarction, acute heart failure, deep vein thrombosis, pulmonary embolism, septicemia, postoperative shock, blood transfusion, thrombocytopenia, respiratory failure, pneumonia, postoperative delirium, acute hemorrhagic anemia and hemorrhage/seroma/hematoma. Surgical complications included surgical site infection, wound dehiscence, fracture and revision surgery. In-hospital resource utilization was assessed using LOS and TOTCHG, with values above the 75th percentile defined as prolonged LOS and higher TOTCHG.

Statistical Analysis

Continuous variables were compared using the unpaired t-test, and categorical variables using the Chi-square test. Multivariate logistic regressions were employed to assess the association between dementia and perioperative outcomes, with 95% confidence intervals (95%CI), unadjusted odds ratios (uOR), and adjusted odds ratios (aOR) reported. For the matched groups, two logistic regression models were constructed: Model 1 included dementia and propensity score provided by PSM, while Model 2 was adjusted for dementia, all patients’ demographics, 30 comorbidities, and the number of comorbidities (Table 1). Further, the Hosmer-Lemeshow test was employed to evaluate model goodness-of-fit. All tests were considered statistically significant at P < 0.05. All statistical analyses were conducted using IBM SPSS Statistics version 26.

1:1 PSM was conducted to align demographics, hospital characteristics, all pre-existing comorbidities and the number of comorbidities in both dementia and non-dementia groups (Table 1). The caliper value was set at 0.01. Standardized mean differences (SMD) for all variables were <0.10 after PSM, indicating satisfactory balance between the matched dementia and matched non-dementia groups.

Results

Study Cohort

Between 2010 and 2019, a total of 1 339 936 patients underwent TKA. After applying the eligibility criteria and performing PSM, the final analysis comprised 6726 matched patients with dementia and 6726 matched patients without dementia (Figure 1).

Figure 1.

Procedures for cohort inclusion and exclusion

Occurrence of Dementia in Elderly Patients Undergoing TKA

Finally, a total of 344 200 patients who underwent TKA were retrieved from the NIS database between 2010 and 2019. Overall, there were 6761 cases of dementia, resulting in an overall prevalence rate of 2.0% (6,761/344,200) (Figure 2). Importantly, our findings found a significant increase in the annual incidence of dementia among elderly inpatients undergoing TKA, particularly after 2011 (Figure 2).

Figure 2.

Annual incidences of dementia in patients undergoing total knee arthroplasty

Patients’ Demographics, Hospital Characteristics, and Comorbidities

Patients with dementia had a higher mean age (76 years) than those without dementia, with a greater proportion aged 70 years or older (P < 0.0001). The dementia group was predominantly male (P < 0.0001). Although both groups were primarily White, Hispanic and Native American patients were more frequently represented in the dementia group (P < 0.0001) (Table 2).

Table 2.

Patients’ Demographics and Hospital Characteristics

Characteristics	Study cohorts
Characteristics	Dementia	Whole non- dementia	P value	Matched non- dementia	P value	SMD^a after matching
Total (n = count)	6726	337,439		6726
Age (Mean ± SD^b)	76.01 ± 6.20	72.75 ± 5.76	<0.0001	76.08 ± 6.33	0.523	<0.0001
Age group (%)			<0.0001		0.834
65-69	17.8	35.4		17.4
70-79	52.5	50.4		52.8		<0.0001
≥80	29.7	14.2		29.8		<0.0001
Gender (%)			<0.0001		0.337	0.017
Male	42.1	37.5		42.9
Female	57.9	62.5		57.1
Race (%)			<0.0001		0.956
White	80.4	80.9		80.7
Black	5.3	5.9		5.0		0.012
Hispanic	6.4	5.2		6.3		0.002
Asian or Pacific Islander	1.7	1.7		1.7		<0.0001
Native American	0.5	0.3		0.6		<0.0001
Other	5.7	6.1		5.6		0.004
Type of insurance (%)			<0.0001		0.224
Medicare	91.0	85.7		91.1
Medicaid	1.1	0.7		1.0		0.009
Private insurance	6.6	11.9		7.0		<0.0001
Self-pay	0.2	0.3		0.1		0.018
No charge	0.0	0.0		0.0		<0.0001
Other	1.2	1.4		0.8		0.033
Bed size of hospital (%)			0.001		0.923
Small	29.0	31.1		28.7
Medium	28.0	27.5		28.2		<0.0001
Large	43.0	41.4		43.1		<0.0001
Teaching status of hospital (teaching %)	62.0	60.9	0.080	62.0	0.972	<0.0001
Location of hospital (urban, %)	89.2	89.8	0.128	89.2	1.000	<0.0001
Region of hospital (%)			0.006		0.865
Northeast	18.2	19.2		18.5
Midwest or North Central	25.7	24.9		25.1		0.012
South	38.0	36.7		38.4		<0.0001
West	18.2	19.3		18.0		0.005

^aSMD: standardized mean difference.

^bSD: standard deviation.

P values were calculated using the unpaired t-test for continuous variables and the Chi-square test for categorical variables; two-sided P < 0.05 was considered statistically significant.

Several hospital characteristics also differed significantly between the groups. Most patients were covered by Medicare, with Medicare and Medicaid more prevalent among dementia patients (P < 0.0001). Dementia patients were more frequently treated in hospitals with larger bed sizes (P = 0.001) and were more often hospitalized in the Midwest or North Central and Southern regions (P = 0.006) (Table 2).

More than three comorbidities were more common in dementia patients than in non-dementia patients (P < 0.0001). The prevalence of nearly all comorbidities was significantly higher in dementia patients (P < 0.05). In contrast, dementia patients had lower rates of diabetes (uncomplicated) and obesity (P < 0.0001) (Table 3).

Table 3.

Patients’ Pre-existing Comorbidities

Characteristics	Study cohorts
Characteristics	Dementia	Whole non- dementia	P value	Matched non- dementia	P value	SMD^a after Matching
Pre-existing comorbidity (%)
Acquired immune deficiency syndrome	0.0	0.0	1.000	0.0	1.000	<0.0001
Alcohol abuse	1.1	0.6	<0.0001	1.2	0.622	<0.0001
Deficiency anemias	3.6	2.6	<0.0001	3.8	0.524	<0.0001
Rheumatoid arthritis/collagen vascular diseases	4.1	3.5	0.006	4.4	0.306	<0.0001
Chronic blood loss anemia	0.8	0.6	0.068	0.7	0.311	0.017
Congestive heart failure	6.9	3.5	<0.0001	7.0	0.812	<0.0001
Chronic pulmonary disease	16.9	14.7	<0.0001	16.9	0.963	<0.0001
Coagulopathy	3.3	1.8	<0.0001	3.3	0.847	0.003
Depression	22.6	12.1	<0.0001	22.3	0.649	0.008
Diabetes (uncomplicated)	14.7	17.0	<0.0001	14.7	0.942	<0.0001
Diabetes (with chronic complications)	9.2	6.3	<0.0001	8.7	0.334	0.016
Drug abuse	0.4	0.3	0.032	0.6	0.391	<0.0001
Hypertension	75.4	73.3	<0.0001	75.6	0.703	<0.0001
Hypothyroidism	21.3	19.3	<0.0001	21.4	0.950	<0.0001
Liver disease	1.4	1.3	0.833	1.2	0.594	0.009
Lymphoma	0.3	0.2	0.224	0.3	0.516	0.011
Fluid and electrolyte disorders	8.6	6.5	<0.0001	9.0	0.377	<0.0001
Metastatic cancer	0.1	0.1	0.910	0.1	1.000	0.005
Obesity	18.3	25.6	<0.0001	18.0	0.687	0.007
Paralysis	0.3	0.1	<0.0001	0.3	0.763	0.005
Peripheral vascular disorders	4.5	2.9	<0.0001	4.2	0.399	0.014
Psychoses	3.6	1.1	<0.0001	3.2	0.218	0.020
Pulmonary circulation disorders	1.4	0.9	<0.0001	1.5	0.663	<0.0001
Renal failure	12.2	7.9	<0.0001	12.2	0.895	0.002
Solid tumor without metastasis	0.7	0.7	0.443	0.7	0.839	0.003
Peptic ulcer disease excluding bleeding	0.3	0.3	0.366	0.2	0.262	0.018
Valvular disease	6.7	4.3	<0.0001	6.8	0.891	<0.0001
Weight loss	0.6	0.2	<0.0001	0.6	0.913	<0.0001
Drug-induced coagulopathy	29.4	25.5	<0.0001	28.9	0.582	0.009
Tobacco dependence	27.8	27.8	0.957	28.2	0.659	<0.0001
Number of comorbidities (%)			<0.0001		0.925
0	5.5	7.9		5.5
1	15.8	19.1		15.8		<0.0001
2	23.1	24.9		22.7		0.011
≥3	55.6	48.1		56.0		<0.0001

^aSMD: standardized mean difference.

P values were calculated using the Chi-square test; two-sided P < 0.05 was considered statistically significant.

After 1:1 PSM, all baseline differences, including demographics, hospital characteristics, and comorbidities, were eliminated, resulting in two well-balanced matched groups (SMD <0.1) (Tables 2 and 3).

Perioperative Complications

Before matching, most complications occurred more frequently in dementia patients than in non-dementia patients (P < 0.05), with acute hemorrhagic anemia, postoperative delirium, and blood transfusion being the most common. After 1:1 PSM, most complication rates were comparable between groups, with dementia patients showing a significantly higher risk only for urinary tract infection (2.8% vs 1.5%, P < 0.0001), pneumonia (0.7% vs 0.4%, P = 0.009), postoperative delirium (4.8% vs 0.9%, P < 0.0001), and acute hemorrhagic anemia (21.7% vs 18.5%, P < 0.0001) (Table 4).

Table 4.

Perioperative Complications of Total Knee Arthroplasty

Perioperative complications	Study cohorts
Perioperative complications	Dementia	Whole non-dementia	P value	Matched non-dementia	P value
Medical complication (%)
Acute renal failure	3.3	2.1	<0.0001	2.9	0.233
Urinary tract infection	2.8	1.2	<0.0001	1.5	<0.0001
Acute myocardial infarction	0.3	0.1	<0.0001	0.2	0.095
Acute heart failure	0.3	0.2	0.002	0.6	0.054
Deep vein thrombosis	0.4	0.2	0.015	0.4	1.000
Pulmonary embolism	0.3	0.3	0.239	0.3	0.881
Septicemia	0.2	0.1	<0.0001	0.1	0.317
Postoperative shock	0.0	0.0	0.662	0.0	1.000
Blood transfusion	4.5	2.9	<0.0001	3.9	0.086
Thrombocytopenia	2.9	1.4	<0.0001	2.8	0.603
Respiratory failure	1.2	0.6	<0.0001	1.0	0.508
Pneumonia	0.7	0.3	<0.0001	0.4	0.009
Postoperative delirium	4.8	0.5	<0.0001	0.9	<0.0001
Acute hemorrhagic anemia	21.7	16.8	<0.0001	18.5	<0.0001
Hemorrhage/seroma/hematoma	0.2	0.1	0.138	0.2	0.695
Surgical complication (%)
Surgical site infection	0.1	0.0	0.389	0.0	0.371
Wound dehiscence	0.1	0.1	0.056	0.0	0.132
Fracture	0.1	0.1	0.889	0.1	1.000
Revision surgery	0.0	0.0	1.000	0.0	1.000

P values were calculated using the Chi-square test; two-sided P < 0.05 was considered statistically significant.

In the logistic regression analysis of the unmatched groups, dementia was associated with a higher risk of several perioperative complications, whereas most other complications did not differ significantly from the non-dementia group (Figure 3).

Figure 3.

Unadjusted odds ratios of perioperative outcomes for the dementia and the whole non-dementia groups. uOR: unadjusted odds ratio, 95%CI: 95% confidence interval, LOS: length of stay, TOTCHG: total charge. uOR (squares) with 95%CI (horizontal lines) were estimated by multivariable logistic regression; two-sided P < 0.05 indicates significance

Model 1 logistic regression analysis after PSM revealed dementia statistically significantly increased the risk of urinary tract infection (aOR = 1.959, 95%CI = 1.533-2.504, P < 0.0001), pneumonia (aOR = 1.888, 95%CI = 1.160-3.071, P = 0.010), postoperative delirium (aOR = 5.698, 95%CI = 4.306-7.541, P < 0.0001), and acute hemorrhagic anemia (aOR = 1.218, 95%CI = 1.119-1.326, P < 0.0001) (Figure 4). Model diagnostics showed adequate fit for some outcomes (P > 0.05), but suboptimal fit was observed for pneumonia (P = 0.047), postoperative delirium (P = 0.007), and acute hemorrhagic anemia (P < 0.0001) (Table S2).

Figure 4.

Adjusted odds ratios of perioperative outcomes for the dementia and the matched non-dementia groups (Model 1). aOR: adjusted odds ratio, 95%CI: 95% confidence interval, LOS: length of stay, TOTCHG: total charge. aOR (squares) with 95%CI (horizontal lines) were estimated by multivariable logistic regression; two-sided P < 0.05 indicates significance

After further adjustment for patient demographics and comorbidities, the associations observed in Model 1 remained consistent. Dementia was a statistically significantly risk factor for urinary tract infection (aOR = 2.001, 95%CI = 1.561-2.564, P < 0.0001), pneumonia (aOR = 2.043, 95%CI = 1.244-3.355, P = 0.005), postoperative delirium (aOR = 5.854, 95%CI = 4.418-7.758, P < 0.0001), and acute hemorrhagic anemia (aOR = 1.229, 95%CI = 1.128-1.340, P < 0.0001) (Figure 5). Model diagnostics indicated improved goodness-of-fit compared with Model 1 (P > 0.05) (Table S2).

Figure 5.

Adjusted odds ratios of perioperative outcomes for the dementia and the matched non-dementia groups (Model 2). aOR: adjusted odds ratio, 95%CI: 95% confidence interval, LOS: length of stay, TOTCHG: total charge. aOR (squares) with 95%CI (horizontal lines) were estimated by multivariable logistic regression; two-sided P < 0.05 indicates significance

In-Hospital Resource Utilization

As expected, dementia patients generally had a longer median LOS (3 days) than both the whole and matched non-dementia groups (P < 0.0001). Dementia was also associated with a higher median TOTCHG ($63 402.95), with an average increase exceeding $3000. In-hospital mortality after TKA was low in both groups, with no significant difference observed (Table 5).

Table 5.

In-Hospital Resource Utilization of Total Knee Arthroplasty

Resource consumption	Study cohorts
Resource consumption	Dementia	Whole non - dementia	P value	Matched non- dementia	P value
LOS^a (Days ± SD^c)	3.03 ± 2.07	2.40 ± 1.62	<0.0001	2.54 ± 1.62	<0.0001
TOTCHG^b (USDs ± SD)	63,402.95 ± 37,790.67	60,113.11 ± 35,675.66	<0.0001	60,341.97 ± 32,961.80	<0.0001
Death	0.1	0.0	0.084	0.0	0.724

^aLOS: length of stay.

^bTOTCHG: total charge.

^cSD: standard deviation.

P values were calculated using the unpaired t-test for continuous variables and the Chi-square test for categorical variables; two-sided P < 0.05 was considered statistically significant.

In the analysis of the unmatched groups, dementia was associated with prolonged LOS and higher TOTCHG (Figure 3). Model 1 showed similar associations (LOS: aOR = 1.890, 95%CI = 1.721-2.076, P < 0.0001; TOTCHG: aOR = 1.115, 95%CI = 1.031-1.206, P = 0.006) (Figure 4). After adjusting for covariates, Model 2 confirmed these findings (LOS: aOR = 1.966, 95%CI = 1.785-2.165, P < 0.0001; TOTCHG: aOR = 1.122, 95%CI = 1.037-1.215, P = 0.004) (Figure 5) and demonstrated good model fit, supporting the robustness of the results.

Discussion

This study conducted a large-scale analysis of elderly patients with dementia undergoing TKA, focusing on health economic aspects. This study is the first in recent years to examine the outcomes of TKA in elderly patients with dementia from various aspects utilizing the NIS database. In this study, we observed that dementia was independently associated with an increased risk of various perioperative complications and greater utilization of healthcare resources. Our research encompasses a wide range and diverse set of patient information; thus, our findings may still be informative for other similar healthcare settings.

According to our research, between 2010 and 2019, the incidence of dementia ranged from 1.3% to 2.3%, with a slight increase after 2011, likely related to the NIA-AA diagnostic criteria.²¹ Additionally, although the transition from ICD-9 to ICD-10 in 2015 introduced additional codes,²² the annual incidence remained stable, suggesting that diagnostic or coding changes are unlikely to have influenced our findings.

Adjusted multivariate analysis of both models after PSM revealed that dementia significantly increased the risk of urinary tract infection and pneumonia after TKA. Bail K et al compared dementia and non-dementia inpatients and demonstrated that dementia patients had higher rates of urinary tract infections and pneumonia.²³ Wu YM et al found that the rate of pneumonia and urinary tract infection was significantly increased in dementia patients following major surgery.²⁴ Clinically, urinary incontinence and low bladder function may predispose patients to catheter-related urinary tract infections,²⁵ while swallowing dysfunction increases the risk of aspiration pneumonia.^26,27 Awareness of these vulnerabilities allows clinicians to implement targeted preventive measures, such as minimizing catheter use and duration, and early assessment and management of swallowing difficulties.^26,28

Our study identified dementia as a significant risk factor for postoperative delirium, consistent with findings from previous research. For example, Lih Wang et al reported that among patients undergoing TKA, a history of dementia was a risk factor linked with postoperative delirium.²⁹ Likewise, Kyu Sung Chung et al showed that a history of dementia was identified as a significant independent risk factor and the strongest independent predictor of delirium.³⁰ Dementia likely increases brain vulnerability, making patients more susceptible to delirium when exposed to surgical stress and other predisposing factors, potentially through systemic inflammation, neuroinflammation, and neuronal damage.⁹ Further research is warranted to clarify the underlying mechanisms and to develop effective strategies for preventing postoperative delirium.

In our study, even after eliminating differences in coagulopathy, deficiency anemia, and chronic blood loss anemia among groups through PSM, dementia still showed a statistically significant correlation with acute hemorrhagic anemia. Previous studies have reported that several dementia subtypes, including Alzheimer’s disease, vascular dementia, Parkinson’s disease dementia, and Lewy body dementia, are associated with thrombocytopenia and platelet functional impairment.^31-35 Collectively, these alterations may compromise hemostasis and increase susceptibility to perioperative bleeding in the setting of total joint arthroplasty.^36-38 Although evidence is limited, dementia-related platelet dysfunction may contribute to acute hemorrhagic anemia, highlighting the need for greater perioperative awareness and further research.

Several of our findings are consistent with studies conducted in other surgical populations. Studies in total hip arthroplasty, hip fracture, and spinal surgeries have shown that dementia increases the risks of postoperative pneumonia, urinary tract infection, and delirium.^39-42 Notably, evidence from cardiovascular surgery further supports these results, demonstrating that patients with dementia are at an elevated risk for postoperative pneumonia and delirium.⁴³ These parallels strengthen the generalizability of our findings and emphasize the need for tailored perioperative strategies across different surgical fields, beyond TKA.

In this study, dementia was independently associated with longer hospital stays and higher healthcare costs. This finding aligns with earlier research showing that patients with dementia often experience prolonged hospitalization following TKA.⁴⁴ Additionally, prior studies have demonstrated that postoperative complications prolong hospital stays, increase resource utilization, and elevate costs.^45,46 Our results similarly indicate that patients with dementia tend to have more comorbidities and are more prone to complications, which likely contribute to extended hospitalization and greater overall costs.

Limitations

This study has several limitations. First, it did not distinguish between dementia subtypes or perform subgroup analyses to assess their differential impact on perioperative outcomes of TKA. Future studies should incorporate subtype-specific analyses to facilitate more tailored preventive strategies for perioperative complications in patients with various forms of dementia.

Second, the NIS database lacks detailed data on certain assessments of preoperative health conditions and laboratory parameters such as the preoperative functional status and serum albumin, as well as post-discharge health status and long-term outcomes. Therefore, our analysis may have overlooked some potential confounding factors and could not include long-term postoperative outcomes such as chronic pain, long-term survival, or readmission rates. Previous studies have reported higher readmission and mortality rates among patients with dementia after TKA.^13,47 To more fully elucidate the impact of dementia on TKA outcomes, particularly over the long term, a more detailed evaluation of patients’ perioperative health status is needed, and prospective longitudinal studies with extended follow-up are warranted.

Third, our results rely on the accuracy and completeness of International Classification of Diseases (ICD) coding. Dementia and TKA may be underdiagnosed or misclassified, as ICD codes may not fully capture all cases or may be subject to coding errors. Such misclassification could potentially lead to underestimation or overestimation of the true association between dementia and perioperative complications.

Conclusion

As dementia becomes more prevalent and the demand for TKA continues to rise, understanding its impact on perioperative outcomes is increasingly critical. Using the NIS database, this study demonstrates that dementia in elderly TKA patients is significantly associated with higher perioperative complications and increased healthcare resource utilization. These findings highlight the need for tailored perioperative care to improve surgical safety in elderly patients with dementia.

Supplemental Material

Supplemental Material - Impact of Dementia on Perioperative Complications in Elderly People Undergoing Total Knee Arthroplasty: A Retrospective National Inpatient Sample Database Study

Supplemenral Material for Impact of Dementia on Perioperative Complications in Elderly People Undergoing Total Knee Arthroplasty: A Retrospective National Inpatient Sample Database Study by Yusong Zhang, Linjia Li, Yuanwei Wu, Jian Wang, Yuxin Zhong, Qinfeng Yang, and Jiwei Luo in Geriatric Orthopaedic Surgery & Rehabilitation

Footnotes

ORCID iDs

Linjia Li

Yuanwei Wu

Jian Wang

Qinfeng Yang

Ethical Considerations

Not applicable. This article does not contain any studies conducted by any of the authors on human participants or animals. Additionally, this observational study was granted exempt status by the institutional review board because of its use of deidentified publicly available data. Therefore, there is no need to grant permission in the ethical approval and consent to participate sections. All methods were carried out following relevant guidelines and regulations.

Consent for Publication

All participants involved in this study have provided informed consent for the publication of the findings, including any personal data, images, or videos that may be included in the manuscript.

Author Contributions

All authors contributed to the thinking and design of the study. Yusong Zhang, Linjia Li, and Yuanwei Wu contributed equally to this work. Material preparation, data collection, and analysis were performed by Yusong Zhang, Linjia Li, and Yuanwei Wu. The first draft of the manuscript was written by Yusong Zhang and Linjia Li, with revisions, guidance, and feedback by Jian Wang, Yuxin Zhong, Qinfeng Yang, and Jiwei Luo. All authors commented on earlier versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data were obtained from the National Inpatient Sample database of the Healthcare Cost and Utilization Project database funded by the Agency for Healthcare Research and Quality and related information is available at

Supplemental Material

Supplemental material for this article is available online.

Appendix

References

Carr

Robertsson

Graves

, et al. Knee replacement. Lancet. 2012;379(9823):1331-1340. doi:10.1016/S0140-6736(11)60752-6

Vina

Kwoh

. Epidemiology of osteoarthritis: literature update. Curr Opin Rheumatol. 2018;30(2):160-167. doi:10.1097/BOR.0000000000000479

Singh

. Epidemiology of knee and hip arthroplasty: a systematic review. Open Orthop J. 2011;5:80-85. doi:10.2174/1874325001105010080

Kurtz

Ong

Lau

Mowat

Halpern

. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007;89(4):780-785. doi:10.2106/JBJS.F.00222

Liu

Varady

Antonelli

Thornhill

Chen

. Similar 20-year survivorship for single and bilateral total knee arthroplasty. Knee. 2022;35:16-24. doi:10.1016/j.knee.2022.02.001

Hustedt

Goltzer

Bohl

Fraser

Lara

Spangehl

. Calculating the cost and risk of comorbidities in total joint arthroplasty in the United States. J Arthroplasty. 2017;32(2):355-361.e1. doi:10.1016/j.arth.2016.07.025

Zarling

Sikora-Klak

Bergum

Markel

. How Do preoperative medications influence outcomes after total joint arthroplasty? J Arthroplasty. 2017;32(9S):S259-S262. doi:10.1016/j.arth.2017.04.031

Weber

Mak

Berenbaum

, et al. Association between osteoarthritis and increased risk of dementia: a systemic review and meta-analysis. Medicine (Baltim). 2019;98(10):e14355. doi:10.1097/MD.0000000000014355

Fong

Inouye

. The inter-relationship between delirium and dementia: the importance of delirium prevention. Nat Rev Neurol. 2022;18(10):579-596. doi:10.1038/s41582-022-00698-7

10.

Gale

Acar

Daffner

. Dementia. Am J Med. 2018;131(10):1161-1169. doi:10.1016/j.amjmed.2018.01.022

11.

GBD 2019 . Dementia forecasting collaborators. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the global burden of disease study 2019. Lancet Public Health. 2022;7(2):e105-e125. doi:10.1016/S2468-2667(21)00249-8

12.

Ritter

Harty

. Total joint replacement in patients with dementia syndromes: a report of thirteen cases. Orthopedics. 2004;27(5):516-517. doi:10.3928/0147-7447-20040501-21

13.

Johnson

Brennan

King

Turcotte

MacDonald

. Comparison of postoperative outcomes of patients undergoing total hip and total knee arthroplasty following a diagnosis of dementia: a TriNetX database study. Arthroplast Today. 2024;27:101359. doi:10.1016/j.artd.2024.101359

14.

Hernandez

Cunningham

Jiranek

Bolognesi

Seyler

. Total knee arthroplasty in patients with dementia. J Knee Surg. 2022;35(1):26-31. doi:10.1055/s-0040-1712086

15.

Bozic

Lau

Kurtz

Ong

Berry

. Patient-related risk factors for postoperative mortality and periprosthetic joint infection in medicare patients undergoing TKA. Clin Orthop Relat Res. 2012;470(1):130-137. doi:10.1007/s11999-011-2043-3

16.

Buller

Best

Klika

Barsoum

. The influence of psychiatric comorbidity on perioperative outcomes following primary total hip and knee arthroplasty; a 17-year analysis of the national hospital discharge survey database. J Arthroplasty. 2015;30(2):165-170. doi:10.1016/j.arth.2014.08.034

17.

Memtsoudis

Della Valle

Besculides

Esposito

Koulouvaris

Salvati

. Risk factors for perioperative mortality after lower extremity arthroplasty: a population-based study of 6,901,324 patient discharges. J Arthroplasty. 2010;25(1):19-26. doi:10.1016/j.arth.2008.11.010

18.

von

Altman

Egger

, et al. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Int J Surg. 2014;12(12):1495-1499. doi:10.1016/j.ijsu.2014.07.013

19.

Meier

Pollock

Kurtz

Lau

. State and government administrative databases: Medicare, national inpatient sample (NIS), and state inpatient databases (SID) programs. J Bone Joint Surg Am. 2022;104(Suppl 3):4-8. doi:10.2106/JBJS.22.00620

20.

Huang

Wang

Yang

Xie

Chen

. Incidence and risk factors of blood transfusion after total knee arthroplasty: a retrospective nationwide inpatient sample database study. Heliyon. 2024;10(14):e34406. doi:10.1016/j.heliyon.2024.e34406

21.

McKhann

Knopman

Chertkow

, et al. The diagnosis of dementia due to alzheimer’s disease: recommendations from the national institute on aging-alzheimer’s association workgroups on diagnostic guidelines for alzheimer’s disease. Alzheimers Dement. 2011;7(3):263-269. doi:10.1016/j.jalz.2011.03.005

22.

Utter

Cox

Atolagbe

Owens

Romano

. Conversion of the agency for healthcare research and quality’s quality indicators from ICD-9-CM to ICD-10-CM/PCS: the process, results, and implications for users. Health Serv Res. 2018;53(5):3704-3727. doi:10.1111/1475-6773.12981

23.

Bail

Berry

Grealish

, et al. Potentially preventable complications of urinary tract infections, pressure areas, pneumonia, and delirium in hospitalised dementia patients: retrospective cohort study. BMJ Open. 2013;3(6):e002770. doi:10.1136/bmjopen-2013-002770

24.

Kuo

, et al. Preexisting dementia is associated with increased risks of mortality and morbidity following major surgery: a nationwide propensity score matching study. Int J Environ Res Public Health. 2020;17(22):8431. doi:10.3390/ijerph17228431

25.

Lai

Kolanko

, et al. Population incidence and associated mortality of urinary tract infection in people living with dementia. J Infect. 2024;88(6):106167. doi:10.1016/j.jinf.2024.106167

26.

Wilkinson

Codipilly

Wilfahrt

. Dysphagia: evaluation and collaborative management. Am Fam Physician. 2021;103(2):97-106.

27.

Banda

Chu

Chen

, et al. Prevalence of oropharyngeal dysphagia and risk of pneumonia, malnutrition, and mortality in adults aged 60 years and older: a meta-analysis. Gerontology. 2022;68(8):841-853. doi:10.1159/000520326

28.

Chenoweth

Gould

Saint

. Diagnosis, management, and prevention of catheter-associated urinary tract infections. Infect Dis Clin North Am. 2014;28(1):105-119. doi:10.1016/j.idc.2013.09.002

29.

Wang

Seok

Kim

Lee

. The risk factors of postoperative delirium after total knee arthroplasty. J Knee Surg. 2017;30(6):600-605. doi:10.1055/s-0036-1593872

30.

Chung

Lee

Park

Choi

. Risk factors of delirium in patients undergoing total knee arthroplasty. Arch Gerontol Geriatr. 2015;60(3):443-447. doi:10.1016/j.archger.2015.01.021

31.

Shinitzky

Deckmann

Kessler

Sirota

Rabbs

Elizur

. Platelet autoantibodies in dementia and schizophrenia. Possible implication for mental disorders. Ann N Y Acad Sci. 1991;621:205-217. doi:10.1111/j.1749-6632.1991.tb16980.x

32.

Liang

Jin

Wang

. Mean platelet volume and platelet distribution width in vascular dementia and alzheimer’s disease. Platelets. 2014;25(6):433-438. doi:10.3109/09537104.2013.831064

33.

Beura

Yadav

Panigrahi

Singh

. Unveiling the mechanism of platelet dysfunction in parkinson’s disease: the effect of 6-hydroxydopamine on human blood platelets. Parkinsonism Relat Disord. 2023;112:105453. doi:10.1016/j.parkreldis.2023.105453

34.

Jiang

Zhong

Huang

Bian

Huang

. Monocytohigh-density lipoprotein ratio has a high predictive value for the diagnosis of multiple system atrophy and the differentiation from parkinson’s disease. Front Aging Neurosci. 2022;14:1035437. doi:10.3389/fnagi.2022.1035437

35.

Arias

Boigues

Ispierto

, et al.

Platelet and immune cell crosstalk in dementia with lewy bodies: is there a unique inflammatory profile?

J Neuroimmunol. 2025;406:578668. doi:10.1016/j.jneuroim.2025.578668

36.

Sang

Roest

de Laat

de Groot

Huskens

. Interplay between platelets and coagulation. Blood Rev. 2021;46:100733. doi:10.1016/j.blre.2020.100733

37.

Ghadimi

Levy

Welsby

. Perioperative management of the bleeding patient. Br J Anaesth. 2016;117(suppl 3):iii18-iii30. doi:10.1093/bja/aew358

38.

Park

Rasouli

Mortazavi

SMJ

Tokarski

Maltenfort

Parvizi

. Predictors of perioperative blood loss in total joint arthroplasty. J Bone Joint Surg Am. 2013;95(19):1777-1783. doi:10.2106/JBJS.L.01335

39.

Morioka

Moriwaki

Tomio

Fushimi

Ogata

. Dementia and patient outcomes after hip surgery in older patients: a retrospective observational study using nationwide administrative data in Japan. PLoS One. 2021;16(4):e0249364. doi:10.1371/journal.pone.0249364

40.

Ahluwalia

Lugo

Gordon

, et al. The association of dementia on perioperative complications following primary total hip arthroplasty for femoral neck fractures. Eur J Orthop Surg Traumatol. 2023;33(4):971-976. doi:10.1007/s00590-022-03236-9

41.

Hernandez

Cunningham

Jiranek

Bolognesi

Seyler

. Total hip arthroplasty in patients with dementia. J Arthroplasty. 2020;35(6):1667-1670.e2. doi:10.1016/j.arth.2020.01.070

42.

Ashraf

Karnati

Tao

, et al. Does cognition status affect spinal surgery outcomes: a systematic review and meta-analysis. Spine J. 2025;S1529-9430(25):29-48. doi:10.1016/j.spinee.2025.07.020

43.

Strategy of cardiovascular surgery for patients with dementia as evaluated by mini-mental state examination - PubMed. 2025. https://pubmed.ncbi.nlm.nih.gov/30259878/

44.

Luan Erfe

Boehme

Erfe

Brovman

Bader

Urman

. Postoperative outcomes in primary total knee arthroplasty patients with preexisting cognitive impairment: a systematic review. Geriatr Orthop Surg Rehabil. 2018;9:2151459318816482. doi:10.1177/2151459318816482

45.

Oliveira

FMD

Costa

LAV

Bastos

AMPde A

Paião

Ferretti

Lenza

. Assessment of the risk factors related to the length of hospital stay and postoperative complications in patients undergoing primary total knee arthroplasty. Rev Bras Ortop (Sao Paulo). 2023;58(3):435-442. doi:10.1055/s-0042-1753534

46.

Dimick

Chen

Taheri

Henderson

Khuri

Campbell

. Hospital costs associated with surgical complications: a report from the private-sector national surgical quality improvement program. J Am Coll Surg. 2004;199(4):531-537. doi:10.1016/j.jamcollsurg.2004.05.276

47.

Jämsen

Peltola

Puolakka

Eskelinen

Lehto

MUK

. Surgical outcomes of hip and knee arthroplasties for primary osteoarthritis in patients with alzheimer’s disease: a nationwide registry-based case-controlled study. Bone Joint J. 2015;97-B(5):654-661. doi:10.1302/0301-620X.97B5.34382

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB

Impact of Dementia on Perioperative Complications in Elderly Patients Undergoing Total Knee Arthroplasty: A Retrospective National Inpatient Sample Database Study

Abstract

Background

Methods

Results

Conclusion

Keywords

Background

Methods and Materials

Study Design

Data Source

Study Cohort

Baseline Data

Outcomes

Statistical Analysis

Results

Study Cohort

Occurrence of Dementia in Elderly Patients Undergoing TKA

Patients’ Demographics, Hospital Characteristics, and Comorbidities

Perioperative Complications

In-Hospital Resource Utilization

Discussion

Limitations

Conclusion

Supplemental Material

Supplemental Material - Impact of Dementia on Perioperative Complications in Elderly People Undergoing Total Knee Arthroplasty: A Retrospective National Inpatient Sample Database Study

Footnotes

ORCID iDs

Ethical Considerations

Consent for Publication

Author Contributions

Funding

Declaration of Conflicting Interests

Data Availability Statement

Supplemental Material

Appendix

References

Supplementary Material