Abstract
Background:
The incidence of revision total hip (rTHA) and knee (rTKA) arthroplasty continues to increase. Preoperative depression is known to influence outcomes following primary arthroplasty. Despite this, it remains unknown whether the same relationship exists for patients undergoing revision procedures. The purpose of this study, therefore, is to investigate this relationship.
Methods:
This is a retrospective cohort study. Patients undergoing rTHA and rTKA were identified from the Truven MarketScan database. Patients with a diagnosis of prosthetic joint infection were excluded. Two cohorts were created: those with preoperative depression and those without. We included patients who were enrolled in the database for 1 year preoperatively and postoperatively. Demographic and complication data were collected, and statistical analysis was then performed comparing complications between cohorts.
Results:
A total of 10,017 patients undergoing rTHA and 13,973 patients undergoing rTKA were included in this study. Of these, 1305 (13.1%) and 2012 (14.4%) had depression, respectively. Multivariate analysis found that, after rTHA, preoperative depression was associated with extended length of stay, nonhome discharge, 90-day readmission, 90-day emergency department visit, prosthetic joint infection, revision surgery, and increased costs (P < .001). Similarly, following rTKA, depression was associated with extended length of stay, nonhome discharge, 90-day readmission, 90-day emergency department visit, revision surgery, and increased costs (P < .001).
Conclusion:
Depression before revision total joint arthroplasty is common and is associated with increased risk of complication and increased healthcare resource utilization following both rTHA and rTKA. Further research will be needed to delineate to what degree this represents a modifiable risk factor.
Keywords: total hip, arthroplasty, depression, total knee, narcotic, revision
Total knee (TKA) and total hip arthroplasty (THA) are 2 of the most common procedures performed in medicine and their incidence continues to rise [1–3]. Not surprisingly, the incidence of revision TKA (rTKA) and THA (rTHA) is also increasing [4,5]. There is a need for ongoing identification of risk factors for complication following these procedures, especially in the setting of revision arthroplasty where the complication rates are known to be higher than those experienced in the primary setting [6]. One risk factor that has been repeatedly examined in primary THA and TKA is preoperative depression [7–20]. However, few studies to date have examined this risk factor in revision total joint arthroplasty.
This is an important relationship to understand as the incidence of these psychiatric conditions has been reported to be higher in arthroplasty patients than in the general population [7,11,13,14,16]. Data from the World Health Organization estimate that in 2015 the US prevalence of depression was 5.9% (over 17 million people) [21]. Again, however, it has been repeatedly demonstrated that primary arthroplasty patients have a much higher incidence of these conditions [7,11,14]. While there is some evidence that depressive symptoms decrease following total joint replacement, this is predicated on improvement in pain [9,10,13,22]. Therefore, it remains unknown what the prevalence of depression is in those with failed primary total joint arthroplasty and what the implications of this diagnosis are following rTKA or rTHA.
The purpose of this study is to determine the relationship between preoperative depression and postoperative complications, healthcare utilization, and inpatient costs following rTHA and rTKA. We hypothesized that depression would be associated with increased complications, healthcare utilization, and care costs when compared to those without depression.
Methods
Data Source
Patients for this study were identified and collected from the Truven MarketScan Commercial Claims and Encounters and Medicare Supplemental and Coordination of Benefit databases (Truven Health, Ann Arbor, MI). This is a commercial claims database that includes information on patients with private insurance and on patients with Medicare and private supplemental insurance. Since its inception in 1995, the database has amassed over 240 million patients. The primary strengths of the database include the ability to follow patients longitudinally if they remain enrolled in the database and the inclusion of information on patients from the inpatient, outpatient, and pharmaceutical arenas.
Included Patients
From 2009 to 2017, the database was queried for patients undergoing rTHA or rTKA. These patients were identified using Current Procedural Terminology (CPT) codes. Specifically, the following codes were used to identify patients undergoing rTKA: 27486 (rTKA, with or without allograft; 1 component) and 27487 (rTKA, with or without allograft; femoral and entire tibial component), and those undergoing rTHA: 27134 (rTHA, both components), 27137 (rTHA, acetabular component only), and 27138 (rTHA, femoral component only). To make for cleaner analysis, we chose to exclude patients with prosthetic joint infection (PJI). To do this, we excluded patients with the following CPT codes: 27488 (removal of prosthesis, including total knee prosthesis, methylmethacrylate with or without insertion of spacer, knee), 27090 (removal of hip prosthesis), and 27091 (removal of prosthesis, with or without spacer placement). We additionally queried International Classification of Diseases (ICD) codes and any patients with a preoperative diagnosis of PJI were then excluded (ICD-9 code 996.66; ICD-10 code T84.5). Additionally, in order to ensure that observed complications were associated with the revision procedure, we excluded patients who had their primary procedure within 90 days of their revision procedure. Last, patients without continuous enrollment in the database for 1 year preoperatively and postoperatively were excluded.
At this point, patients were separated into 2 cohorts: those with and those without depression. We identified depression using ICD-9 and ICD-10 diagnoses codes. This included the following codes: 3004, 30112, 3090, 3091, 311, 29682, 29620-6, 29630-6, F32, F33, F341, and F432 (Table 1). Patients were included if they had a documented history of depression within the year preceding surgery.
Table 1.
International Classification of Diseases (ICD) 9 and 10 Codes to Identify Depression.
| Code | Description |
|---|---|
| 3004 | Dysthymic disorder |
| 30112 | Chronic depressive personality disorder |
| 3090 | Adjustment disorder with depressed mood |
| 3091 | Prolonged depressive reaction |
| 311 | Depressive disorder, not elsewhere classified |
| 29682 | Atypical depressive disorder |
| 29620 | Major depressive affective disorder, single episode, unspecified |
| 29621 | Major depressive affective disorder, single episode, mild |
| 29622 | Major depressive affective disorder, single episode, moderate |
| 29623 | Major depressive affective disorder, single episode, severe, without mention of psychotic behavior |
| 29624 | Major depressive affective disorder, single episode, severe, specified as with psychotic behavior |
| 29625 | Major depressive affective disorder, single episode, in partial or unspecified remission |
| 29626 | Major depressive affective disorder, single episode, in full remission |
| 29630 | Major depressive affective disorder, recurrent episode, unspecified |
| 29631 | Major depressive affective disorder, recurrent episode, mild |
| 29632 | Major depressive affective disorder, recurrent episode, moderate |
| 29633 | Major depressive affective disorder, recurrent episode, severe, without mention of psychotic behavior |
| 29634 | Major depressive affective disorder, recurrent episode, severe, specified as with psychotic behavior |
| 29635 | Major depressive affective disorder, recurrent episode, in partial or unspecified remission |
| 29636 | Major depressive affective disorder, recurrent episode, in full remission |
| F32 | Major depressive disorder, single episode |
| F33 | Major depressive disorder, recurrent |
| F341 | Dysthymic disorder |
| F432 | Adjustment disorder with depressed mood |
Baseline Patient Information
The following information was collected for each included patient: age, sex, geographic region of treatment, surgery type (CPT code signifying 1 or 2 component revision), insurance type, and comorbidities. We collected and controlled for the following comorbidities: obesity, chronic kidney disease, alcohol use disorder, tobacco use, hypertension, coronary artery disease (CAD), congestive heart failure, rheumatic disease, and diabetes mellitus. Each of these parameters were compared between cohorts.
Complication, Healthcare Utilization, and Costs Data
Following finalization of patient inclusion, the database was queried for the following complication data: PJI, infection (PJI and superficial surgical site infection), wound complication, sepsis, and thromboembolic event. These were all collected for 90 days postoperatively. Repeat revision surgery was also collected, but for 1 year following surgery. The following healthcare utilization parameters were collected: extended length of stay (LOS; ≥4 days), nonhome discharge, 90-day readmission, 90-day all-cause emergency department (ED) visit, 90-day pain-related ED visit, and opioid prescription after 6 months postoperative. Additionally, we collected net payments for the inpatient stay for each patient and compared them between cohorts. These payment data represent the amount of money paid by insurance for the entire care episode (ie, the actual paid amount, not charges).
Statistical Analysis
Statistical analysis was conducted using SPSS (version 25, IBM Corporation, Armonk, NY). A P value of <.05 was selected as significant for this study. All analyses were conducted on rTKA and rTHA patients separately. First, baseline patient demographic and comorbid data were compared between cohorts using chi-square analysis. Chi-square analysis was then used to compare the rates at which our complication and healthcare utilization data occurred between cohorts. We then subsequently performed multivariate binomial logistic regression controlling for all variables listed in Table 2 to assess the independent association between preoperative depression and postoperative complication, healthcare utilization, and inpatient episode of care costs.
Table 2.
Demographics and Comorbidities of Opioid Use Groups.
| Demographic | Revision THA |
Revision TKA |
||||
|---|---|---|---|---|---|---|
| No Depression | Depression | P Value | No Depression | Depression | P Value | |
| Age-group | ||||||
| Total number of included patients Age-group | 8712 (86.9%) | 1305 (13.1%) | 11,961 (85.6%) | 2012 (14.4%) | ||
| <55 | 1802 (20.7%) | 309 (23.7%) | <.001 | 2036 (17.0%) | 487 (24.2%) | <.001 |
| 55-64 | 3025 (34.7%) | 546 (41.8%) | 4780 (40.0%) | 957 (47.6%) | ||
| 65-74 | 1704 (19.6%) | 234 (17.9%) | 2823 (23.6%) | 387 (19.2%) | ||
| 75-84 | 1666 (19.1%) | 162 (12.4%) | 1994 (16.7%) | 154 (7.7%) | ||
| 85+ | 515 (5.9%) | 54 (4.1%) | 328 (2.7%) | 27 (1.3%) | ||
| Sex | ||||||
| Male | 4113 (47.2%) | 399 (30.6%) | <.001 | 5069 (42.4%) | 468 (23.3%) | <.001 |
| Female | 4599 (52.8%) | 906 (69.4%) | 6892 (57.6%) | 1544 (76.7%) | ||
| Region | ||||||
| Northeast | 1703 (19.5%) | 240 (18.4%) | .739 | 2012 (16.8%) | 295 (14.7%) | .002 |
| Midwest | 2606 (29.9%) | 380 (29.1%) | 4095 (34.2%) | 698 (34.7%) | ||
| South | 2946 (33.8%) | 456 (34.9%) | 4236 (35.4%) | 686 (34.1%) | ||
| West | 1411 (16.2%) | 221 (16.9%) | 1553 (13.0%) | 319 (15.9%) | ||
| Surgery type (component replaced) | ||||||
| Revision THA, both components | 5080 (58.3%) | 719 (55.1%) | <.001 | n.a. | n.a. | .557 |
| Revision THA, acetabulum | 2293 (26.3%) | 353 (27.0%) | n.a. | n.a. | ||
| Revision THA, femur | 1339 (15.4%) | 233 (17.9%) | n.a. | n.a. | ||
| Revision TKA, 1 component | n.a. | n.a. | 4774 (39.9%) | 817 (40.6%) | ||
| Revision TKA, both components | n.a. | n.a. | 7187 (60.1%) | 1195 (59.4%) | ||
| Insurance | ||||||
| Comprehensive | 2035 (24.0%) | 233 (18.3%) | <.001 | 2901 (24.9%) | 363 (18.6%) | <.001 |
| Preferred provider organization | 4526 (53.4%) | 715 (56.2%) | 6008 (51.5%) | 1087 (55.7%) | ||
| Health maintenance organization | 864 (10.2%) | 157 (12.3%) | 1168 (10.0%) | 213 (10.9%) | ||
| Point-of-service | 511 (6.0%) | 79 (6.2%) | 814 (7.0%) | 129 (6.6%) | ||
| High deductible health plans | 541 (6.4%) | 88 (6.9%) | 764 (6.6%) | 158 (8.1%) | ||
| Comorbidities | ||||||
| Obesity | 828 (9.5%) | 199 (15.2%) | <.001 | 1797 (15.0%) | 530 (26.3%) | <.001 |
| Chronic kidney disease | 537 (6.2%) | 71 (5.4%) | .307 | 651 (5.4%) | 123 (6.1%) | .224 |
| Alcohol use disorders | 82 (0.9%) | 46 (3.5%) | <.001 | 57 (0.5%) | 47 (2.3%) | <.001 |
| Tobacco use | 514 (5.9%) | 176 (13.5%) | <.001 | 469 (3.9%) | 172 (8.5%) | <.001 |
| Hypertension | 5139 (59.0%) | 800 (61.3%) | .112 | 7982 (66.7%) | 1367 (67.9%) | .286 |
| Coronary artery disease | 1520 (17.4%) | 195 (14.9%) | .025 | 2177 (18.2%) | 325 (16.2%) | .027 |
| Congestive heart failure | 511 (5.9%) | 100 (7.7%) | .011 | 690 (5.8%) | 109 (5.4%) | .530 |
| Rheumatic disease | 514 (5.9%) | 80 (6.1%) | .742 | 649 (5.4%) | 148 (7.4%) | .001 |
| Diabetes | 1489 (17.1%) | 217 (16.6%) | .678 | 3172 (26.5%) | 512 (25.4%) | .313 |
THA, total hip arthroplasty; TKA, total knee arthroplasty; n.a., not applicable.
Institutional Review Board Approval
This study was institutional review board approved by our institution.
Results
Demographics and Baseline Patient Information
In total, 10,017 patients undergoing rTHA were identified in the database. This included 8712 patients (86.9%) without and 1305 (13.1%) with a diagnosis of depression. Patients with depression were more likely to be young, female, obese, tobacco users, have alcohol use disorder, have congestive heart failure, and be undergoing a single-component revision (P ≤ .011). Depressed patients were less likely to have CAD (P = .025). There were no differences between nondepressed and depressed patients with regard to geographic region, chronic kidney disease, hypertension, rheumatic disease, or diabetes (P > .05).
Similarly, there were 13,973 patients identified undergoing rTKA. This consisted of 11,961 patients (85.6%) who were not depressed and 2012 (14.4%) who were depressed. The depressed patients in this group were more likely to be young, female, from the West and less likely from the Northeast. There were also differences in comorbidity profiles where depressed patients were more likely to be obese, have alcohol use disorder, be tobacco users, and have rheumatic disease. Depressed patients were less likely to have CAD (Table 2).
Depression, Postoperative Complications, Healthcare Utilization, and Costs Following rTHA
Outcomes and complication data were collected for patients undergoing rTHA. First, these were compared between cohorts using univariate analysis. This revealed that the following were higher in the depression group when compared to those without depression: infection, PJI, sepsis, revision surgery, extended LOS, nonhome discharge, 90-day readmission, and all-cause and pain-related ED visits (P ≤ .005). There were no differences in rates of wound complications, thromboembolic events, or prolonged opioid use between the 2 groups (P > .05). Univariate analysis also revealed that depressed patients had higher inpatient hospital costs when compared to those without preoperative depression (Table 3).
Table 3.
Univariate Analysis of Complications.
| Complication | Revision THA |
Revision TKA |
||||
|---|---|---|---|---|---|---|
| No Depression | Depression | P Value | No Depression | Depression | P Value | |
| Extended LOS (≥4 d) | 2560 (29.4%) | 433 (33.2%) | .005 | 2327 (19.5%) | 422 (21.0%) | .113 |
| Nonhome discharge | 2029 (24.2%) | 383 (30.4%) | <.001 | 2117 (18.5%) | 414 (21.5%) | .002 |
| 90-d Readmission | 969 (11.1%) | 220 (16.9%) | <.001 | 950 (7.9%) | 244 (12.1%) | <.001 |
| ED visit | 1456 (16.7%) | 331 (25.4%) | <.001 | 1574 (13.2%) | 391 (19.4%) | <.001 |
| Pain-related ED visit | 77 (0.9%) | 28 (2.1%) | <.001 | 127 (1.1%) | 52 (2.6%) | <.001 |
| Infection | 547 (6.3%) | 117 (9.0%) | <.001 | 729 (6.1%) | 140 (7.0%) | .138 |
| Prosthetic joint infection | 286 (3.3%) | 71 (5.4%) | <.001 | 332 (2.8%) | 68 (3.4%) | .113 |
| Wound complication | 327 (3.8%) | 62 (4.8%) | .082 | 392 (3.3%) | 85 (4.2%) | .030 |
| Sepsis | 83 (1.0%) | 29 (2.2%) | <.001 | 95 (0.8%) | 25 (1.2%) | .045 |
| Thromboembolic event | 480 (5.5%) | 85 (6.5%) | .143 | 636 (5.3%) | 112 (5.6%) | .646 |
| Revision surgerya | 837 (9.6%) | 178 (13.6%) | <.001 | 829 (6.9%) | 174 (8.6%) | .006 |
| Opioid prescription after 6 mo | 712 (8.2%) | 110 (8.4%) | .753 | 1152 (9.6%) | 181 (9.0%) | .369 |
| Net payments for admission (USD)b | $23,033 ± 278 | $26,466 ± 703 | <.001 | $21,556 ± 229 | $25,666 ± 516 | <.001 |
THA, total hip arthroplasty; TKA, total knee arthroplasty; LOS, length of stay; ED, emergency department; USD, US dollars.
Revision surgery and opioid overdose displayed with 1-y follow-up, all other complications displayed for 90 d postoperative follow-up.
Displayed as mean ± standard error of the mean.
Subsequently, multivariate logistic regression was performed to control for confounding variables. This revealed that even when controlling for baseline patient demographic and comorbid data, patients with depression had significantly higher odds of incurring extended LOS (≥4 days; odds ratio [OR], 1.28; 95% confidence interval [CI], 1.12-1.47; P < .001), nonhome discharge (OR, 1.60; 95% CI, 1.38-1.85; P < .001), 90-day readmission (OR, 1.37; 95% CI, 1.16-1.63; P < .001), all-cause ED visit (OR, 1.55; 95% CI, 1.34-1.78; P < .001), pain-related ED visit (OR, 2.1; 95% CI, 1.33-3.30; P =.001), infection (OR, 1.34; 95% CI, 1.07-1.66; P =.010), PJI (OR, 1.52; 95% CI, 1.15-2.01; P =.004), sepsis (OR, 1.86; 95% CI, 1.17-2.95; P =.008), and revision surgery (OR, 1.36; 95% CI, 1.14-1.63; P =.001). Additionally, multivariable logistic regression found that $2660 (95% CI, 1120-3999; P < .001) of increased costs was attributable to a preoperative diagnosis of depression. No significant differences were identified between depression and wound complication, thromboembolic events, or prolonged opioid prescriptions (P > .146; Table 4).
Table 4.
Adjusted Risks of Complications.
| Complication | Revision THA |
Revision TKA |
||
|---|---|---|---|---|
| Odds Ratio | P Value | Odds Ratio | P Value | |
| Extended LOS (≥4 d) | 1.28 (1.12-1.47) | <.001 | 1.17 (1.03-1.32) | .016 |
| Nonhome discharge | 1.60 (1.38-1.85) | <.001 | 1.44 (1.26-1.64) | <.001 |
| 90-d Readmission | 1.37 (1.16-1.63) | <.001 | 1.43 (1.22-1.68) | <.001 |
| ED visit | 1.55 (1.34-1.78) | <.001 | 1.52 (1.34-1.74) | <.001 |
| Pain-related ED visit | 2.10 (1.33-3.30) | .001 | 2.06 (1.46-2.91) | <.001 |
| Infection | 1.34 (1.07-1.66) | .010 | 1.13 (0.93-1.38) | .228 |
| Prosthetic joint infection | 1.52 (1.15-2.01) | .004 | 1.17 (0.88-1.56) | .270 |
| Wound complication | 1.09 (0.81-1.46) | .560 | 1.20 (0.93-1.56) | .161 |
| Sepsis | 1.86 (1.17-2.95) | .008 | 1.75 (1.10-2.77) | .018 |
| Thromboembolic event | 1.20 (0.94-1.54) | .146 | 1.06 (0.85-1.31) | .613 |
| Revision surgerya | 1.36 (1.14-1.63) | .001 | 1.20 (1.01-1.44) | .046 |
| Opioid prescription after 6 mo | 1.04 (0.83-1.30) | .749 | 0.93 (0.78-1.10) | .421 |
| Net payments for admission (USD)b | +$2660 (1120-3999) | <.001 | +$3051 (1916-4186) | <.001 |
Bold values are those that are statistically significant.
THA, total hip arthroplasty; TKA, total knee arthroplasty; LOS, length of stay; ED, emergency department; USD, US dollars.
Revision surgery and opioid overdose displayed with 1-y follow-up, all other complications displayed for 90-d postoperative follow-up.
Beta value of linear regression presents dollar increase in total hospital payments associated with a diagnosis of depression.
Depression, Postoperative Complications, Healthcare Utilization, and Costs Following rTKA
Complication data were also collected and compared between cohorts for those patients undergoing rTKA. When compared to the nondepressed cohort, patients with depression had significantly higher rates of the following on univariate analysis: wound complications, sepsis, revision surgery, nonhome discharge, 90-day readmissions, and 90-day all-cause and pain-related ED visits (P < .05). There were no significant differences identified in rates of extended LOS, infection, PJI, thromboembolic events, or prolonged opioid use (P ≥ .113; Table 3).
Next, a multivariate logistic regression model was used to determine whether depression was an independent risk factor for these findings. Even after controlling for patient information and comorbidity data, depressed patients had significantly increased odds of the following: extended LOS (≥4 days; OR, 1.17; 95% CI, 1.03-1.32; P =.016), nonhome discharge (OR, 1.44; 95% CI, 1.26-1.64; P < .001), 90-day readmission (OR, 1.43; 95% CI, 1.22-1.68; P < .001), all-cause ED visit (OR, 1.52; 95% CI, 1.34-1.74; P < .001), pain-related ED visit (OR, 2.06; 95% CI, 1.46-2.91; P =.001), sepsis (OR, 1.75; 95% CI, 1.1-2.77; P =.018), and revision surgery (OR, 1.20; 95% CI, 1.01-1.44; P =.046). Depression was found to have no impact on infection, PJI, wound complications, thromboembolic events, or prolonged (>6 months) opioid prescriptions (P ≥ .161). However, compared to those who were not depressed, depressed patients were found to have increased hospital costs of $3051 (95% CI, 1916-4186; P < .001), even after controlling for baseline confounders (Table 4).
Discussion
rTHA and rTKA procedures are becoming more common [1,4,5]. However, the complication rates after revision procedures continue to be much higher than those reported following primary arthroplasty procedures [6]. Taken together, there is a continued need for identification of risk factors for complication and increased healthcare utilization. One such risk factor that has garnered significant attention in primary arthroplasty patients but has been largely ignored in the revision setting is a preoperative diagnosis of depression.
In primary total joint arthroplasty, a preoperative diagnosis of depression has been associated with increased medical complications [7,11,12,18,23], PJI [18,24,25], worse subjective improvement in pain [20,26], decreased satisfaction [8], readmission [15], postoperative transfusion [12], nonhome discharge [12,27], LOS [16,19], and increased costs [23,27]. The results of the current investigation, which investigated revision procedures, is largely in agreement with the above prior findings in primary arthroplasty patients. In the rTHA patient, we found that a preoperative diagnosis of depression was associated with increased healthcare utilization, increased infection, increased revision rates, and increased hospital costs. In rTKA patients, we found that depression was associated with the same outcomes with the exception of increased surgical site infection and PJI.
An interesting finding of this study is that patients undergoing revision arthroplasty are frequently depressed. We found that 13.1% of rTHA and 14.4% of rTKA patients had a diagnosis of preoperative depression in the year preceding surgery. This is much higher than the national prevalence of 5.9% [28], and higher than the prevalence found in similar studies in the primary arthroplasty population (10%-11%) [7,11]. A component of this is likely explained by selection bias as patients with depression are known to have less improvement in pain following surgery and higher rates of revision [8,20,26,29]. While the prevalence of depression is still likely underestimated in this study, our cohort was more likely to be young females, similar to the demographics found in prior investigations [7,11,28].
The relationship between depression and arthroplasty outcomes, however, is complex. It is well known that patients in chronic pain are at risk of developing psychological manifestations of their pain, including depression and anxiety [30]. Additionally, there is some evidence to suggest that anxiety and depressive symptoms may be decreased following TKA and that perhaps some components of these preoperative psychological symptoms are manifestations of pain [9,10,13,22]. The beneficial effects that TKA has on mental health can persist for years [17]. Therefore, patients with depression should not be denied surgery based on their elevated risk as their depressive symptoms may improve postoperatively.
However, we did find that patients who have depression undergoing revision arthroplasty procedures are at elevated risk. We found that rates of infection, sepsis, and specifically PJI were higher in the depressed cohort than the nondepressed controls. This was significant, however, only after rTHA. This is similar to findings in primary arthroplasty patients [18]. While the connection between the two may seem elusive, there is actually a fairly well-established connection between postoperative infection and depressed state [31]. There is evidence that psychological stress in humans induces an inflammatory state via release of proinflammatory cytokines [31]. This preoccupation of the immune system may make patients more susceptible to infection, especially when coupled with the proinflammatory postsurgical state. Additionally, depression may lead to a shift if in T-cell phenotype, leading to further susceptibility [32,33]. As a result, an association of depression with postoperative infection has been observed in orthopedics as well as other fields [24,25,34–36].
We also found significantly increased healthcare utilization among patients with depression following both rTKA and rTHA. While studies on depression in the primary arthroplasty literature have had inconsistent results regarding hospital LOS [7], the majority of studies agree with our findings that patients with depression are at higher risk of extended LOS [16,19]. This is not surprising given difficulty with postoperative pain control [26,29] in these patients and given that LOS is longer even in general ward patients with depression [37]. Additionally, our findings of increased ED visits and readmissions are likely a result of our increased rates of complications. The same is true of revision surgery given that depressed patients had 52% increased odds of PJI following rTHA. These outcomes culminated in increased net payments of >$2600 dollars for depressed patients following rTHA and rTKA. This increased cost is similar to prior investigations [23].
Last, despite the fact that Singh and Lewallen found that depression predicts moderate to severe pain following rTKA [26] and rTHA [29], our results demonstrated that a preoperative diagnosis of depression is not associated with prolonged (>6 months) opioid use. This is in contrast to results previously reported in primary arthroplasty [38]. The reason for this discrepancy may be explained by our private insurance patient population or by our modern cohort (2009-2017) during which the opioid epidemic has become a focus of the medical community. This is especially true given that only 8.2%-9.6% of our patients remained on opioid 6 months after their procedures, a number on par with those reported previously for primary procedures at 1 year [39]. While patients often receive opioids from other providers, in the Truven database, these prescriptions would also be captured.
There are multiple limitations to the present study and the results of the study must be interpreted with these in mind. First, similar to prior studies [7,11], we are subject to complete and accurate coding for the identification of depressed patients. Prior studies using mental health questionnaires preoperatively have found higher rates of psychological distress [13,14,16]. This is likely explained by prior literature showing diagnostic recognition of psychiatric conditions is poor [40]. Therefore, it is likely that a subset of patients with mild depressive symptoms are not captured by this study. Additionally, for the purposes of this investigation, we excluded patients undergoing revision for PJI. This methodological decision is clearly made for cleaner, easier to interpret analysis but does limit the generalizability of our results. This is especially true as infection represents the indication for 15% of rTHA [41] and 25% of rTKA [42]. Along the same lines, only insured patients were included in this analysis. This may further limit the generalizability to uninsured patients or those with Medicaid. Last, given the retrospective nature of this study, causality is unable to be determined. Despite these limitations, which are inherent to analysis of large databases, the Truven MarketScan database represents a strength of this study. This database allowed for longitudinal analysis of a large number of patients, permitting the identification of differences between groups, even in rare outcomes of interest.
In conclusion, patients with preoperative depression who are undergoing rTHA or rTKA are at higher odds of incurring several postoperative complications and having higher healthcare utilization and are associated with increased net hospital payments. These findings should be used to counsel patients and further work will be necessary to discern to what degree depression is a modifiable risk factor.
Acknowledgments
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2019.11.025.
Footnotes
Investigation was performed at Emory University, Atlanta, GA.
IRB statement: This study was IRB exempt given the use of a national database with no identifiers.
References
- [1].Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007;89:780–5. [DOI] [PubMed] [Google Scholar]
- [2].Sloan M, Premkumar A, Sheth NP. Projected volume of primary total joint arthroplasty in the U.S., 2014 to 2030. J Bone Joint Surg Am 2018;100:1455–60. [DOI] [PubMed] [Google Scholar]
- [3].Maradit Kremers H, Larson DR, Crowson CS, Kremers WK, Washington RE, Steiner CA, et al. Prevalence of total hip and knee replacement in the United States. J Bone Joint Surg Am 2015;97:1386–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Rajaee SS, Campbell JC, Mirocha J, Paiement GD. Increasing burden of total hip arthroplasty revisions in patients between 45 and 64 years of age. J Bone Joint Surg Am 2018;100:449–58. [DOI] [PubMed] [Google Scholar]
- [5].Kurtz SM, Lau E, Ong K, Zhao K, Kelly M, Bozic KJ. Future young patient demand for primary and revision joint replacement: national projections from 2010 to 2030. Clin Orthop Relat Res 2009;467:2606–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Liodakis E, Bergeron SG, Zukor DJ, Huk OL, Epure LM, Antoniou J. Perioperative complications and length of stay after revision total hip and knee arthroplasties: an analysis of the NSQIP database. J Arthroplasty 2015;30:1868–71. [DOI] [PubMed] [Google Scholar]
- [7].Pan X, Wang J, Lin Z, Dai W, Shi Z. Depression and anxiety are risk factors for postoperative pain-related symptoms and complications in patients undergoing primary total knee arthroplasty in the United States. J Arthroplasty 2019;34:2337–46. [DOI] [PubMed] [Google Scholar]
- [8].Ali A, Lindstrand A, Sundberg M, Flivik G. Preoperative anxiety and depression correlate with dissatisfaction after total knee arthroplasty: a prospective longitudinal cohort study of 186 patients, with 4-year follow-up. J Arthroplasty 2017;32:767–70. [DOI] [PubMed] [Google Scholar]
- [9].Badura-Brzoza K, Zajac P, Kasperska-Zajac A, Brzoza Z, Matysiakiewicz J, Piegza M, et al. Anxiety and depression and their influence on the quality of life after total hip replacement: preliminary report. Int J Psychiatry Clin Pract 2008;12:280–4. [DOI] [PubMed] [Google Scholar]
- [10].Blackburn J, Qureshi A, Amirfeyz R, Bannister G. Does preoperative anxiety and depression predict satisfaction after total knee replacement? Knee 2012;19:522–4. [DOI] [PubMed] [Google Scholar]
- [11].Browne JA, Sandberg BF, D’Apuzzo MR, Novicoff WM. Depression is associated with early postoperative outcomes following total joint arthroplasty: a nationwide database study. J Arthroplasty 2014;29:481–3. [DOI] [PubMed] [Google Scholar]
- [12].Buller LT, Best MJ, Klika AK, Barsoum WK. 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;30:165–70. [DOI] [PubMed] [Google Scholar]
- [13].Duivenvoorden T, Vissers MM, Verhaar JA, Busschbach JJ, Gosens T, Bloem RM, et al. Anxiety and depressive symptoms before and after total hip and knee arthroplasty: a prospective multicentre study. Osteoarthr Cartil 2013;21:1834–40. [DOI] [PubMed] [Google Scholar]
- [14].Ellis HB, Howard KJ, Khaleel MA, Bucholz R. Effect of psychopathology on patient-perceived outcomes of total knee arthroplasty within an indigent population. J Bone Joint Surg Am 2012;94:e84. [DOI] [PubMed] [Google Scholar]
- [15].Gold HT, Slover JD, Joo L, Bosco J, Iorio R, Oh C. Association of depression with 90-day hospital readmission after total joint arthroplasty. J Arthroplasty 2016;31:2385–8. [DOI] [PubMed] [Google Scholar]
- [16].Halawi MJ, Chiu D, Gronbeck C, Savoy L, Williams VJ, Cote MP. Psychological distress independently predicts prolonged hospitalization after primary total hip and knee arthroplasty. J Arthroplasty 2019;34:1598–601. [DOI] [PubMed] [Google Scholar]
- [17].Jones AR, Al-Naseer S, Bodger O, James ETR, Davies AP. Does pre-operative anxiety and/or depression affect patient outcome after primary knee replacement arthroplasty? Knee 2018;25:1238–46. [DOI] [PubMed] [Google Scholar]
- [18].Klement MR, Nickel BT, Penrose CT, Bala A, Green CL, Wellman SS, et al. Psychiatric disorders increase complication rate after primary total knee arthroplasty. Knee 2016;23:883–6. [DOI] [PubMed] [Google Scholar]
- [19].March MK, Harmer AR, Dennis S. Does psychological health influence hospital length of stay following total knee arthroplasty? A systematic review. Arch Phys Med Rehabil 2018;99:2583–94. [DOI] [PubMed] [Google Scholar]
- [20].Singh JA, Lewallen DG. Depression in primary TKA and higher medical comorbidities in revision TKA are associated with suboptimal subjective improvement in knee function. BMC Musculoskelet Disord 2014;15:127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Organization GWH. Depression and other common mental disorders: global health estimates In: Licence: CC BY-NC-SA 3.0 IGO. World Health Organization; 2017. [Google Scholar]
- [22].Tarakji BA, Wynkoop AT, Srivastava AK, O’Connor EG, Atkinson TS. Improvement in depression and physical health following total joint arthroplasty. J Arthroplasty 2018;33:2423–7. [DOI] [PubMed] [Google Scholar]
- [23].Rasouli MR, Menendez ME, Sayadipour A, Purtill JJ, Parvizi J. Direct cost and complications associated with total joint arthroplasty in patients with preoperative anxiety and depression. J Arthroplasty 2016;31:533–6. [DOI] [PubMed] [Google Scholar]
- [24].Bozic KJ, Lau E, Kurtz S, Ong K, Berry DJ. Patient-related risk factors for postoperative mortality and periprosthetic joint infection in Medicare patients undergoing TKA. Clin Orthop Relat Res 2012;470:130–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Bozic KJ, Lau E, Kurtz S, Ong K, Rubash H, Vail TP, et al. Patient-related risk factors for periprosthetic joint infection and postoperative mortality following total hip arthroplasty in Medicare patients. J Bone Joint Surg Am 2012;94: 794–800. [DOI] [PubMed] [Google Scholar]
- [26].Singh JA, Lewallen DG. Medical and psychological comorbidity predicts poor pain outcomes after total knee arthroplasty. Rheumatology (Oxford) 2013;52:916–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Stundner O, Kirksey M, Chiu YL, Mazumdar M, Poultsides L, Gerner P, et al. Demographics and perioperative outcome in patients with depression and anxiety undergoing total joint arthroplasty: a population-based study. Psychosomatics 2013;54:149–57. [DOI] [PubMed] [Google Scholar]
- [28].Brody DJPL, Hughes JP. Prevalence of depression among adults aged 20 and over: United States, 2013-2016 In: Statistics NCfH. Hyattsville, MD: NCHS Data Brief, no 303; 2018. [PubMed] [Google Scholar]
- [29].Singh JA, Lewallen D. Age, gender, obesity, and depression are associated with patient-related pain and function outcome after revision total hip arthroplasty. Clin Rheumatol 2009;28:1419–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Merskey H Psychological aspects of pain. Curr Med Res Opin 1974;2:515–20. [DOI] [PubMed] [Google Scholar]
- [31].Ghoneim MM, O’Hara MW. Depression and postoperative complications: an overview. BMC Surg 2016;16:1–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Elenkov IJ. Systemic stress-induced Th2 shift and its clinical implications. Int Rev Neurobiol 2002;52:163–86. [DOI] [PubMed] [Google Scholar]
- [33].Elenkov IJ, Chrousos GP. Stress hormones, Th1/Th2 patterns, pro/anti-inflammatory cytokines and susceptibility to disease. Trends Endocrinol Metab 1999;10:359–68. [DOI] [PubMed] [Google Scholar]
- [34].Doering LV, Cross R, Vredevoe D, Martinez-Maza O, Cowan MJ. Infection, depression, and immunity in women after coronary artery bypass: a pilot study of cognitive behavioral therapy. Altern Ther Health Med 2007;13:18–21. [PubMed] [Google Scholar]
- [35].Gordon RJ, Weinberg AD, Pagani FD, Slaughter MS, Pappas PS, Naka Y, et al. Prospective, multicenter study of ventricular assist device infections. Circulation 2013;127:691–702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [36].Chang SM, Parney IF, McDermott M, Barker FG 2nd, Schmidt MH, Huang W, et al. Perioperative complications and neurological outcomes of first and second craniotomies among patients enrolled in the Glioma Outcome Project. J Neurosurg 2003;98:1175–81. [DOI] [PubMed] [Google Scholar]
- [37].Bressi SK, Marcus SC, Solomon PL. The impact of psychiatric comorbidity on general hospital length of stay. Psychiatr Q 2006;77:203–9. [DOI] [PubMed] [Google Scholar]
- [38].Namba RS, Singh A, Paxton EW, Inacio MCS. Patient factors associated with prolonged postoperative opioid use after total knee arthroplasty. J Arthroplasty 2018;33:2449–54. [DOI] [PubMed] [Google Scholar]
- [39].Cook DJ, Kaskovich SW, Pirkle SC, Mica MAC, Shi LL, Lee MJ. Benchmarks of duration and magnitude of opioid consumption after total hip and knee arthroplasty: a database analysis of 69,368 patients. J Arthroplasty 2019;34:638–44. [DOI] [PubMed] [Google Scholar]
- [40].Wancata J, Windhaber J, Bach M, Meise U. Recognition of psychiatric disorders in nonpsychiatric hospital wards. J Psychosom Res 2000;48:149–55. [DOI] [PubMed] [Google Scholar]
- [41].Bozic KJ, Kurtz SM, Lau E, Ong K, Vail TP, Berry DJ. The epidemiology of revision total hip arthroplasty in the United States. J Bone Joint Surg Am 2009;91:128–33. [DOI] [PubMed] [Google Scholar]
- [42].Bozic KJ, Kurtz SM, Lau E, Ong K, Chiu V, Vail TP, et al. The epidemiology of revision total knee arthroplasty in the United States. Clin Orthop Relat Res 2010;468:45–51. [DOI] [PMC free article] [PubMed] [Google Scholar]