Patients With Down Syndrome and Total Hip and Total Knee Arthroplasty: Outcome Measures Show Increased Risk of Perioperative Complications

Abstract

Background:

Patients with Down syndrome (DS) are being considered for total joint arthroplasty. There is limited literature regarding outcomes of patients with DS after total hip arthroplasty (THA) and total knee arthroplasty (TKA).

Methods:

Data were abstracted from the 2010 to 2021 Q1 PearlDiver M151 database. THA and TKA osteoarthritis adult patients with and without DS were identified. Patients were matched 1:10. Ninety-day postoperative events and 2-year revision rates were compared.

Results:

For THA, 154 patients with DS were matched with 1,532 patients without DS. For TKA, 150 patients with DS were matched with 1,495 patients without DS. On multivariable logistic regression, THA and TKA patients with DS were at markedly greater odds of postoperative events including any adverse event, sepsis, minor adverse event, urinary tract infection (UTI), acute kidney injury (AKI), and pneumonia. For both THA and TKA, 2-year revision rates were not increased for those with DS.

Discussion:

This study represents the largest cohorts for matched patients with DS undergoing THA or TKA through 90 days postoperatively. For both procedures, DS patients were found to have greater risk of several adverse events, but not 2-year revisions. These findings may help guide perioperative risk assessment, patient/family counseling, and care pathways.

Down syndrome (DS), caused by trisomy 21,^1,2,3,4 was first described in 1866 by John Langdon Down.⁵ In the United States, there are more than 200,000 people living with DS,⁶ and the average lifespan of patients with DS has markedly increased over time from 25 years in 1983 to 49 years in 1997.⁷ Hip dysplasia is common in children with DS, particularly the presence of acetabular retroversion; this, in combination with ligamentous laxity, frequently results in hip instability and eventual arthrosis.^8,9 Ligamentous laxity also contributes to patellofemoral instability.¹⁰ As such, more DS patients are qualifying for total hip arthroplasty (THA) or total knee arthroplasty (TKA), but there is correspondingly limited or no literature regarding outcomes of these procedures in the DS population.

Patients with DS may experience a wide variety of medical comorbidities that can affect their health. These include, but are not limited to, congenital heart disease,¹¹ immunodeficiencies,¹² hematologic/oncologic disorders,^13-15 sleep disorders,¹⁶ thyroid abnormalities,¹⁷ neurodevelopmental disorders,¹⁸ dementia,¹⁹ autism,²⁰ celiac disease,²¹ and juvenile idiopathic arthritis.²²

Regarding THA for patients with DS, Gross et al²³ examined 21 patients from 1986 to 2008 and found that five patients had a revision of their THA and increased surgical challenges due to hip dysplasia, periarticular variations, and ligamentous laxity. Boylan et al²⁴ examined 241 THA patients with DS using the 1998 to 2010 National Inpatient Sample and found higher rates of pneumonia, urinary tract infection (UTI), and wound hemorrhage in patients with DS while hospitalized for their index procedure. Hernandez et al²⁵ examined 128 THA patients with DS from using the 2010 to 2018 PearlDiver database to evaluate a handful of 90-day postoperative events of which revision arthroplasty was the only markedly increased event.

In evaluation of TKA for patients with DS, no literature could be identified. This was determined to be a gap in the literature and thought to be potentially useful information to complement evolving understanding of outcomes for THA for patients with DS. To address limitations with existing literature regarding THA for patients with DS and the lack of literature on TKA for patients with DS, this study aimed to leverage a large, national, administrative data set to match patients with DS to controls. In this investigation, both perioperative outcomes and 2-year revision rates are assessed.

Methods

Data Source/Study Population

This study abstracted data from the 2010 to 2021 Q1 PearlDiver M151 database. PearlDiver is a large, national, multi-insurance, administrative claims data set. As PearlDiver outputs deidentified and aggregated data, our Institutional Review Board has determined studies using this data set exempt from review.

Patients undergoing THA and TKA were identified using Current Procedural Terminology (CPT) codes CPT-27130 and CPT-27447, respectively. Inclusion criteria included osteoarthritis diagnoses and age older than 18 years. Exclusion criteria include a preoperative indication of a traumatic, infectious, or oncologic diagnosis within 90 days before surgery.

Patients with DS were identified with International Classification of Diseases (ICD) codes ICD-9-D-7580 and ICD-10-D-Q909. THA and TKA patients with DS were matched 1:10 with patients without DS based on age, sex, and Elixhauser Comorbidity Index (ECI, a comorbidity index/score of overall comorbidity burden).

Ninety-day Postoperative Outcomes and 2-year Rates of Revision

After matching, 90-day postoperative outcomes were abstracted from the data set based on administrative coding. The 90-day postoperative events included severe adverse events (cardiac event, deep vein thrombosis [DVT], pulmonary embolism [PE], sepsis, and surgical site infection), and minor adverse events (acute kidney injury [AKI], pneumonia, transfusion, urinary tract infection, and wound complication). Any adverse events were identified and classified as an occurrence of a severe or minor adverse event.

Two-year revision rates were also determined for the matched cohorts. These were identified based on the CPT codes 27134, 27137, and 27138 for hips and 27486 and 27487 for knees.

Data Analysis

Matching was done with a ratio of 1:10 for patients with DS and patients without. The parameters for matching included age, sex, and ECI. Ninety-day postoperative events and 2-year revisions were compared between matched THA and TKA patients with and without DS using univariable (Chi-squared and Student t-tests, where appropriate) and multivariable logistic regression, controlling for age, sex, and ECI.

Data collection and statistical analysis were done with PearlDiver Bellwether and Excel version 16.63.1. For univariable analyses, P < 0.05 was considered significant. For multivariable analyses, a Bonferroni correction was applied, leading to significance being defined at P < 0.0036.

Results

Total Hip Arthroplasty

Initially, THA patients without DS (+THA−DS) were 492,551, and those with DS (+THA + DS) were 157. These populations differed regarding age, sex, and ECI (Table 1, left columns). After 10:1 matching, there were 1,532 and 154 patients and these differences were no longer significant (Table 1, right columns).

Table 1.

Characteristics of Unmatched and Matched (10:1) Cohort of Patients Undergoing THA With and Without DS From 2010 to 2021 Q1

Demographics	Unmatched			Matched
	+THA−DS	+THA+DS	P	+THA−DS	+THA+DS	P
Total	492,551	157		1532	154
Age (average ± SD)	63.7 ± 10.2	43.9 ± 13.5	<0.0001	44.7 ± 12.8	44.4 ± 13.2	0.8250
Sex
Female	274,081 (55.6%)	101 (64.3%)	0.0349	1000 (65.3%)	100 (64.9%)	1.0000
Male	218,470 (44.4%)	56 (35.7%)		532 (34.7%)	54 (35.1%)
ECI (average ± SD)	4.2 ± 3.3	4.8 ± 3.5	0.0217	4.7 ± 3.5	4.7 ± 3.5	0.9732

THA = total hip arthroplasty. Bold represents statistically significant data.

For univariable outcomes, matched THA patients with DS experienced significantly greater 90-day rates of any adverse events, severe adverse events, sepsis, minor events, AKI, pneumonia, and UTI (Table 2, left columns).

Table 2.

Univariable and Multivariable Analysis of 90-day Outcomes and Overall Revisions Among Matched Cohort of + THA + DS and +THA−DS From 2010 to 2021 Q1, Odds Ratios Represent Odds of 90-day Outcomes and Any Revision Surgery Among Patients Who + THA + DS Compared With + THA−DS Patients

Outcome Variables	Univariable			Multivariable (+THA + DS compared with + THA−DS)
	+THA−DS	+THA+DS	P	OR (95% CI)	P
90-day outcomes	1532	154
Any adverse event	309 (20.2%)	35 (22.7%)	<0.0001	2.4 (1.6-3.4)	<0.0001
Severe events	124 (8.1%)	22 (14.3%)	0.0141	1.9 (1.2-3.2)	0.0097
Cardiac event	14 (0.9%)	<11	0.9732	1.4 (0.2-5.5)	0.6340
DVT	43 (2.8%)	<11	0.9531	1.1 (0.4-2.7)	0.7725
PE	23 (1.5%)	<11	0.9316	1.3 (0.3-3.9)	0.6690
Sepsis	20 (1.3%)	13 (8.4%)	<0.0001	7.5 (3.5-15.7)	<0.0001
SSI	19 (1.2%)	<11	0.7148	1.6 (0.4-4.7)	0.4720
Minor events	240 (15.7%)	52 (33.8%)	<0.0001	3.0 (2.0-4.4)	<0.0001
AKI	33 (2.2%)	14 (9.1%)	<0.0001	5.3 (2.6-10.6)	<0.0001
Pneumonia	35 (2.3%)	25 (16.2%)	<0.0001	8.9 (5.1-15.6)	<0.0001
Transfusion	54 (3.5%)	<11	0.9929	1.1 (0.4-2.4)	0.8436
UTI	122 (8.0%)	24 (15.6%)	0.0022	2.2 (1.3-3.6)	0.0013
Wound complication	35 (2.3%)	<11	0.6118	0.5 (0.1-1.8)	0.4116
2-year revisions	58 (3.8%)	11 (7.1%)	0.0733	2.0 (1.0-3.7)	0.0479

AKI = acute kidney injury, DVT = deep vein thrombosis, MI = myocardial infarction, PE = pulmonary embolism, SSI = surgical site infection, UTI = urinary tract infection

Using a Bonferroni correction, significance was set at P < 0.0036.

Bold represents statistically significant data.

For multivariable logistic regression outcomes (Table 2, right columns, and Figure 1), matched THA patients with DS experienced significantly greater 90-day odds of any adverse events (OR 2.4, P < 0.0001), sepsis (OR 7.5, P < 0.0001), minor adverse events (OR 3.0, P < 0.0001), AKI (OR 5.3, P < 0.0001), pneumonia (OR 8.9, P < 0.0001), and UTI (OR 2.2, P = 0.0013). Two-year revisions were not statistically significantly different between the groups.

Figure 1.

A forest plot of the multivariable logistic regression presented in Table 2 for the odds ratios of +THA/+DS, relative to + THA/−DS. The points in black are statistically significant, and the points in gray are not. The bold vertical line denotes the value of 1.

Total Knee Arthroplasty

Initially, TKA patients without DS (+TKA/−DS) were 1,152,719, and those with DS (+TKA/+DS) were 151. These populations differed regarding age and ECI (Table 3, left columns). After 10:1 matching, there were 1,495 and 150 patients and these differences were no longer significant (Table 3, right columns).

Table 3.

Characteristics of Unmatched and Matched (10:1) Cohort of Patients Undergoing TKA With and Without DS From 2010 to 2021 Q1

Demographics	Unmatched			Matched
	+TKA	+TKA	P	+TKA	+TKA	P
	−DS	+DS		−DS	+DS
Total	1,152,719	151		1495	150
Age (average ± SD)	65.8 ± 8.6	55.6 ± 12.0	<0.0001	56.1 ± 11.3	55.8 ± 11.5	0.7972
Sex			0.9411			1.0000
Male	724,723 (62.9%)	97 (64.2%)		56 (63.9%)	96 (64.0%)
Female	427,993 (37.1%)	54 (35.8%)		539 (36.1%)	54 (36.0%)
ECI (average ± SD)	4.2 ± 3.2	6.8 ± 4.2	<0.0001	6.9 ± 4.2	6.9 ± 4.2	0.9752

TKA = total knee arthroplasty. Bold represents statistically significant data.

For univariable outcomes, matched TKA patients with DS experienced significantly greater 90-day rates of any adverse events, severe adverse events, sepsis, minor events, AKI, pneumonia, and UTI (Table 4, left columns).

Table 4.

Univariable and Multivariable Analysis of 90-day Outcomes and Overall Revisions Among Matched Cohort of + TKA + DS and +TKA−DS From 2010 to 2021 Q1, Odds Ratios Represent Odds of 90-day Outcomes and Any Revision Surgery Among Patients Who + TKA + DS Compared With +TKA−DS Patients

Outcome Variables	Univariable			Multivariable (+TKA + DS Compared With + TKA−DS)
	+TKA−DS	+TKA+DS	P	OR (95% CI)	P
90-day outcomes	1495	150
Any adverse events	258 (17.3%)	50 (33.3%)	<0.0001	2.5 (1.7-3.7)	<0.0001
Severe events	126 (8.4%)	23 (15.3%)	0.0078	2.0 (1.2-3.2)	0.0053
Cardiac event	<11	<11	0.2035	3.0 (0.7-10.1)	0.0955
DVT	54 (3.6%)	<11	0.9895	1.1 (0.4-2.5)	0.8110
PE	34 (2.3%)	<11	1.0000	0.9 (0.2-2.5)	0.8321
Sepsis	22 (1.5%)	13 (8.7%)	<0.0001	7.1 (3.3-14.7)	<0.0001
SSI	22 (1.5%)	<11	0.8774	1.3 (0.3-4.0)	0.6355
Minor events	182 (12.2%)	45 (30.0%)	<0.0001	3.3 (2.2-4.9)	<0.0001
AKI	51 (3.4%)	15 (10.0%)	0.0002	3.4 (1.8-6.4)	0.0001
Pneumonia	38 (2.5%)	25 (16.7%)	<0.0001	8.3 (4.7-14.4)	<0.0001
Transfusion	29 (1.9%)	<11	0.7643	1.4 (0.4-3.6)	0.5457
UTI	58 (3.9%)	17 (11.3%)	<0.0001	3.3 (1.8-5.7)	<0.0001
Wound complication	43 (2.9%)	<11	0.9501	1.2 (0.4-2.7)	0.7610
2-year revisions	59 (3.9%)	<11	0.8350	1.2 (0.5-2.5)	0.6780

AKI = acute kidney injury, TKA = total knee arthroplasty, UTI = urinary tract infection

Using a Bonferroni correction, significance was set at P < 0.0036.

Bold represents statistically significant data.

For multivariable logistic regression outcomes (Table 2, right columns, and Figure 2), matched TKA patients with DS experienced significantly greater 90-day odds of any adverse events (2.5, P < 0.0001), sepsis (7.1, P < 0.0001), minor events (3.3, P < 0.0001), AKI (3.4, P < 0.0001), pneumonia (8.3, P < 0.0001), and UTI (3.3, P < 0.0001). Two-year revisions were not statistically significantly different between the groups.

Figure 2.

A forest plot of the multivariable logistic regression presented in Table 4 for the odds ratios of +TKA/+DS, relative to +TKA/−DS. The points in black are statistically significant, and the points in gray are not. The bold vertical line denotes the value of 1.

Discussion

This study represents the largest overall cohort identified for patients with Down syndrome undergoing THA and TKA. Using a national claims database, this study examined 154 THA patients with DS and 150 TKA patients with DS. Furthermore, ninety-day postoperative adverse events and 2-year revision rates were also assessed and compared with matched populations without DS.

On multivariable logistic regression, after both THA and TKA, patients with DS had a markedly higher odds of having any adverse event, sepsis, minor adverse event, AKI, pneumonia, and UTI. These trends are consistent with previously published studies examining THA in patients with DS.²⁴ Three of these individual adverse outcomes are infection-related (sepsis, pneumonia, and UTI) and may align with previous studies suggesting patients with DS experience cellular and humoral immune deficiency, particularly poor chemotaxis and suboptimal antibody responses.^26-28 In addition, patients with DS may also have increased incidence of certain airway^29,30 and genitourinary anomalies (such as renal hypoplasia, hydroureteronephrosis, and uteropelvic junction obstruction).^31,32

The other adverse outcome that was found to be of greater odds for patients with DS after both THA and TKA was AKI. This may be attributed to the genitourinary anomalies^31,32 noted above. Finally, the aggregated minor and any adverse events noted above seem to be driven by the individual adverse outcomes noted above (including AKI, UTI, and pneumonia).

Specific pertinent negatives identified in the study include lack of increased odds of SSI for those with DS. Given that patients with DS can have immunodeficiencies,¹² it can be helpful for surgeons to know that was not found to be associated with increased odds for SSI. Furthermore, the lack of increased odds of 2-year revisions after TKA and THA has not been examined in the literature previously with a large cohort of DS patients and matched cohorts. This finding suggests the surgical outcomes for DS patients seeking THA are equivalent at 2 years to those observed in the non-DS matched cohort, and this should provide support and reassurance to orthopaedic surgeons, patients, and families.

This study does have limitations. Inherent to any study that utilizes administrative databases are potential inaccuracies in coding. In addition, as this is a retrospective study, causality is not able to be directly assessed. Finally, condition-specific outcome measures, patient-reported outcomes, socioeconomic status, surgical approach, and anesthesia were not able to be assessed. However, with the use of this large database, this study represents the largest total cohort of patients with DS undergoing total joint arthroplasty matched to control subjects. Moreover, it provides the first data set concerning the outcomes of TKA procedures in the DS patient population.

Conclusion

In summary, this study represents the largest cohorts ever reported for matched patients with DS undergoing THA or TKA through 90 days postoperatively. For both procedures, DS patients were found to have greater risk of a number of postoperative adverse events, but not 2-year revisions. Understanding these findings may be helpful for preoperative risk assessment, patient and family counseling, surgical planning, and the evolution of targeted care pathways to benefit DS patients seeking equitable access to arthroplasty care.