The Association of Benign Prostatic Hyperplasia on Postoperative Complications and Periprosthetic Joint Infections Following Total Shoulder Arthroplasty

Abstract

Background

A recently proposed risk factor for periprosthetic joint infections (PJI) in men is benign prostatic hyperplasia (BPH). The objective was to explore the association of BPH on 1) 90-day complications, 2) length of stay (LOS), 3) readmission rates, and 4) healthcare expenditures following total shoulder arthroplasty (TSA).

Methods

A retrospective query was performed using a nationwide claims database from January 2005 to March 2014 for male patients undergoing primary TSA. The study cohort included 5067 patients with BPH while 50,720 patients served as the comparison cohort. Logistic regression determined the association of BPH on complications and readmissions. A p value less than 0.001 was significant.

Results

Patients with BPH had higher incidence and odds (36.8 vs. 6.2%; OR: 2.73, p < 0.0001) of all ninety-day medical complications including PJIs (0.49 vs. 0.32%; OR: 1.54, p < 0.001). BPH patients had longer in-hospital LOS (3- vs. 2-days, p < 0.0001). Study group patients had higher 90-day episode of care reimbursements ($13,653 vs. $12,688), p < 0.0001).

Conclusions

BPH is associated with increased complications and healthcare expenditures following total shoulder arthroplasty. The investigation can be used to educate BPH patients of the possible adverse events which may occur within ninety-days following primary TSA for the treatment of glenohumeral osteoarthritis.

Introduction

 Despite increased utilization of total joint arthroplasty, complications in the form of periprosthetic joint infection (PJI) occur in 1–2% of patients.^1,2 Although shoulder PJI occurs less frequently than hip and knee arthroplasties, the morbidity and healthcare costs associated with this complication continue to be primary research efforts by shoulder surgeons.^1–9 Male gender has been demonstrated to be a risk for higher rates of periprosthetic joint infections compared to female counterparts after joint replacement of the hip, knee, and shoulder.^2,6,10–12 A recently proposed risk factor for prosthetic joint infection (PJI) in men is symptomatic benign prostatic hyperplasia (BPH).^13–15 Benign prostatic hyperplasia (BPH) is a nonmalignant proliferation of prostatic tissue that results in symptomatic lower urinary tract symptoms in men. ¹⁶ The lifetime prevalence of BPH in men was estimated to be about 26%, and influenced by increasing age. ¹⁶ Bladder outlet obstruction from BPH predisposes patients to urinary tract infections which hypothetically can seed a nonnative joint due to transient bacteremia. ¹⁷ The frequency of patients with BPH undergoing orthopaedic surgery procedures is unknown and further warrants study in this population as prior arthroplasty studies demonstrate somewhat mixed findings.

Yazdi et al. found a significant association of PJI after hip and knee arthroplasty in men with symptomatic BPH, with a nearly 3-fold increase (7.9% versus 2.8% in controls) in infection rate after lower extremity arthroplasty. ¹⁵ In contrast, Gu and colleagues found no significant difference of PJIs in patients with BPH following arthroplasty of the hip (1.54% vs 1.43%) and knee (1.99% vs 2.14%). ¹⁴ Periprosthetic joint infection (PJI) is among the leading causes of failure and reoperation in total shoulder arthroplasty, occurring in approximately 1% of patients. ¹² PJIs are projected to cost the healthcare system over $1.85 billion by 2030, therefore efforts to understand the risk factors and increased expenditures in male patients remain paramount. ¹⁸ Given that no studies have explored the association of BPH and increased medical complications including PJIs following total shoulder arthroplasty (TSA), further exploration is highly warranted.

In light of the lack of studies in TSA patients, the objective of this study was to determine the association of BPH on outcomes following total shoulder arthroplasty. Specifically, we investigated the impact of BPH on: 1) medical complications, 2) in-hospital length of stay (LOS), 3) readmission rates, and 4) healthcare expenditures. We hypothesized that male patients undergoing TSA with a history of BPH would have increased medical complications, PJIs, and healthcare expenditures compared to their counterparts.

Methods

Database and cohort

A retrospective analysis of the 100% Parts A and B of the Medicare claims database from January 1^st, 2005 to March 31^st, 2014 was performed using the for-fee based PearlDiver (PearlDiver Technologies, Fort Wayne, Indiana, United States) platform. The database contains information of over 100 million patients from the Medicare claims and a private insurance database known as Mariner. The database allows researchers to identify and query cohorts of patients using an open-based syntax language using International Classification of Disease, Ninth Revision (ICD-9), ICD-10, Current Procedural Terminology (CPT), National Drug Codes, and Diagnostic Related Group codes. Information is then extracted as a comma separated value (.csv) spreadsheet for subsequent data analyses. The study was determined to be exempt from our institution’s institutional review board (IRB) review.

The Medicare database was initially queried for all patients who underwent primary TSA using ICD-9 procedural code 81.80. Those patients with a history of glenohumeral OA were identified using ICD-9 diagnostic codes 715.11, 715.21, 715.31, 715.91. Patients with a history of benign prostatic hyperplasia were identified using ICD-9 diagnostic code 600.01. These procedural and diagnostic codes were utilized as they have been used in previously published studies.^19–21 The “FIRST_INSTANCE” syntax command was utilized to ensure that patients with these procedures and diagnoses were counted for the first time, to prevent any overestimation on the association of BPH on the dependent variables measured within the study. Using Boolean command-syntax language of the PearlDiver database, the inclusion criteria for the study cohort was patients undergoing primary TSA with a concomitant diagnosis of BPH; whereas those patients without BPH served as the comparison cohort. As patients may enter and leave the Medicare database, patients with an active enrollment one year prior and one year following the procedure were included in the analyses. Exclusions from the study included those patients who were female, patients undergoing surgery for trauma, infections, or malignant etiologies as done in prior investigations.

Primary outcomes measured

Primary endpoints of the study were to compare medical complications, in-hospital LOS, readmission rates, and healthcare expenditures. Readmission rates and medical complications were those which occurred within the 90-day episode of care interval. Complications analyzed included: acute kidney injuries, cerebrovascular accidents, deep vein thromboses, ileus episodes, myocardial infarctions, pneumoniae, prosthetic joint infections, pulmonary emboli, respiratory failures, surgical site infections, transfusion of blood products, urinary tract infections, venous thromboemboli. For healthcare expenditures, day of surgery and total global ninety-day episode of care interval costs were analyzed using reimbursement data. Reimbursements were chosen instead of global costs as they are a more accurate predictor of what providers are paid from the insurance companies and has been used as a benchmark in previously published studies from the same database.^22,23

Statistical analyses

Statistical analyses were performed using the open programming language known as R (R, Foundation for Computational Statistics, Vienna, Austria). Baseline demographics of the two cohorts were analyzed using Pearson’s Chi-Square Analyses or Fischer’s Exact test for categorical variables. The PearlDiver database provides age as a categorical variable for patients less than the age of 64, 65 to 69, 70 to 74, 75 to 79, 80 to 84, and older than 85. Other baseline demographics were compared between the two cohorts and included prevalence of comorbid conditions which are found within the Elixhauser-Comorbidity Index (ECI). An overall mean ECI score was also calculated for each cohort. For continuous variables, Welch’s t-tests were used to assess for significance for LOS, healthcare expenditures, and mean ECI scores between the two cohorts. A multivariate binomial logistic regression model was performed to determine the association of BPH on medical complications and readmission rates. The model was adjusted for age, alcohol use disorder, chronic obstructive pulmonary disease (COPD), diabetes mellitus, general anxiety disorder, hyperlipidemia, hypertension, obesity – defined as a body mass index (BMI) greater than 30 kilograms per meter squared (kg/m2), and tobacco use. These comorbid conditions were entered into the regression model as studies have shown BPH to be associated with these comorbid conditions. To reduce the probability of a type I error, a Bonferroni-correction was performed, and a p-value less than 0.001 was considered to be statistically significant.

Results

Patient population

A total of 133,393 were identified. After the inclusion and exclusion criteria, the final query consisted of 55,787 patients who underwent TSA with (n = 5067) and without (n = 50,720) BPH. The demographics between the two cohorts differed with respect to age (p < 0.0001), and prevalence of comorbid conditions (Table 1). Specifically, study group patients had higher prevalence of all the comorbid conditions analyzed with an exception to morbid obesity (BMI 40 to 70 kg/m2; p = 0.791). Elixhauser-Comorbidity Index (ECI) was greater in the study group compared to control (6 vs 3, p < 0.0001)(Table 1).

Table 1.

Comparison of baseline demographic profiles of patients undergoing primary total shoulder arthroplasty for glenohumeral osteoarthritis With and without benign prostatic hyperplasia. BMI = Body Mass Index; ECI = Elixhauser-Comorbidity Index.

	Benign Prostate Hyperplasia		Control
Demographics	n	%	n	%	p-value
Age (Years)					<0.0001
<64	348	6.87	5177	10.21
65 to 69	1638	32.33	17,032	33.58
70 to 74	1424	28.10	13,172	25.97
75 to 79	1029	20.31	9414	18.56
80 to 84	499	9.85	4610	9.09
≥ 85	129	2.55	1315	2.59
Comorbidities
Alcohol Use Disorder	364	7.18	3460	6.82	<0.0001
Arrhythmia	2539	50.11	20,151	39.73	<0.0001
BMI 19 to 24 kg/m2	80	1.58	601	1.18	0.017
BMI 25 to 29 kg/m2	216	4.26	1408	2.78	<0.0001
BMI 30 to 39 kg/m2	685	13.52	5577	11.00	<0.0001
BMI 40 to 70 kg/m2	187	3.69	1915	3.78	0.791
Congestive Heart Failure	1436	28.34	11,000	21.69	<0.0001
Coagulopathy	935	18.45	7068	13.94	<0.0001
Depression	1467	28.95	11,071	21.83	<0.0001
Diabetes Mellitus	1958	38.64	17,891	35.27	<0.0001
Hypertension	4650	91.77	44,089	86.93	<0.0001
Hypothyroidism	1248	24.63	9942	19.60	<0.0001
Iron Deficiency Anemia	2671	52.71	19,818	39.07	<0.0001
Liver Failure	421	8.31	3292	6.49	<0.0001
Peptic Ulcer Disease	560	11.05	4202	8.28	<0.0001
Peripheral Vascular Disease	1878	37.06	13,698	27.01	<0.0001
Renal Failure	376	7.42	2504	4.94	<0.0001
Rheumatoid Arthritis	829	16.36	5756	11.35	<0.0001
Valvular Disorders	1562	30.83	11,504	22.68	<0.0001
Pathologic Weight Loss	351	6.93	2802		<0.0001
ECI	6		3		<0.0001

Ninety-Day medical complications

Patients with BPH were found to have significantly higher frequency and odds of developing adverse events within ninety-days following their surgical procedure (36.8 vs. 6.2%; OR: 2.73, 95%CI: 2.49 to 2.98, p < 0.0001). Study group patients were found to have higher rates of urinary tract infections (7.12 vs. 1.78%; OR: 4.03, 95%CI: 3.55 to 4.58, p < 0.0001), pneumoniae (3.00 vs. 1.21%; OR: 2.37, 95%CI: 1.97 to 2.84, p < 0.0001), acute renal failure (2.92 vs. 1.18%; OR: 2.33, 95%CI: 1.93 to 2.80, p < 0.0001), and PJI (0.49 vs. 0.32%; OR: 1.54, 95%CI: 1.01 to 2.36, p < 0.001), in addition to other adverse events (Table 2).

Table 2.

Comparison on the frequency of ninety-Day medical complications among study group and control cohorts undergoing primary total shoulder arthroplasty for glenohumeral osteoarthritis.

	BPH	Control	OR	95%CI	p-value
Urinary Tract Infection	7.12	1.78	4.03	3.55 – 4.58	<0.0001
Pneumoniae	3.00	1.21	2.37	1.97 – 2.84	<0.0001
Ileus	0.55	0.22	2.34	1.51 – 3.51	<0.0001
Acute Renal Failure	2.92	1.18	2.33	1.93 – 2.80	<0.0001
Transfusions	1.89	0.04	2.25	1.38 – 3.53	0.0006
Myocardial Infarctions	0.73	0.30	2.17	1.49 – 3.09	<0.0001
Respiratory Failure	0.26	0.12	2.02	1.06 – 3.58	0.02
Deep Vein Thromboses	6.61	0.12	1.71	1.28 – 2.25	0.0001
Venous Thromboemboli	8.31	0.14	1.62	1.24 – 2.08	0.0001
Cerebrovascular Accidents	1.11	0.64	1.58	1.14 – 2.03	<0.0001
PJIs	0.49	0.32	1.54	1.01 – 2.36	<0.001
Surgical Site Infections	0.280	0.08	3.60	1.95 – 6.63	0.0001
Pulmonary Emboli	3.51	0.05	1.27	0.80 – 1.91	0.268
Total	36.78	6.20	2.73	2.49 – 2.98	<0.0001

BPH = Benign Prostate Hyperplasia; PJIs = Peri-Prosthetic Joint Infections; OR = Odds-Ratio; 95%CI = 95% Confidence Interval.

* = Adjusted for Age, Geographic Region, Alcohol Use Disorder, Chronic Obstructive Pulmonary Disease, General Anxiety Disorder, Diabetes Mellitus, Hyperlipidemia, Hypertension, Obesity, and Tobacco Use.

In-Hospital LOS and readmission rates

Patients undergoing total shoulder arthroplasty with BPH were found to have longer in-hospital LOS (3-days vs. 2-days, p < 0.0001) compared to controls. Additionally, study group patients were found to have similar incidence and odds of readmission rates within 90 days of surgery (6.49% vs. 6.43%, OR: 0.96, 95% CI: 0.85 to 1.09, p = 0.587) compared to the control population.

Healthcare expenditures

When assessing healthcare expenditures, BPH patients were found to have significantly higher day of surgery ($11,664.55 vs. $11,135.75, p < 0.0001) and total global 90-day episode of care reimbursement rates $13,652.71 vs. $12,688.37, p < 0.0001) compared to the control cohort following TSA.

Discussion

Shoulder arthroplasty procedures have exponentially increased over the last few decades, with the proposed revision burden following a similar trajectory.^24–29 Despite an abundance of literature detailing the risk factors for complication following TSA,^6,30–37 no studies to date have explored the association of BPH history on postoperative outcomes. To address a gap in the existing literature, we attempted to determine the associated comorbidities in patients with BPH undergoing TSA and the rates of postoperative complications in comparison to controls. Using a nationwide database, this study found that patients undergoing TSA with a history of BPH have a higher comorbidity burden (ECI 6 vs ECI 3), significantly higher 90 day medical complications (including PJIs), longer lengths of stay, and increased healthcare associated costs.

Perhaps the most clinically important finding of our study was the increased rate of medical complications including PJIs in patients with a history of BPH within 90 days of surgery. Gender, specifically male patients undergoing arthroplasty, has been reported to be a increased risk for PJI compared to females. It is postulated that sex hormones may modulate and interact with the immune system differently between genders, however more study is needed to confirm these findings.^38,39 Groups have proposed that underlying BPH may be a potential contributor due to a higher systemic bioburden from postop urinary retention, catheterization, and UTIs.^14,15 Yazdi et al. demonstrated in all total joint arthroplasties of the hip and knee, BPH was associated with a nearly 3-fold increase in PJI rate (7.9% versus 2.8% in controls). ¹⁵ In the present study, we noted a increase in 90-day PJI rate (0.49% vs 0.32%) for patients with a history of BPH. The 90-day PJI rates of the BPH cohort in our study (0.49%) were midway between 90-day (0.23%) and 1 year (0.75%) rates of all TSA patients in another study. ⁴⁰ Because of the economic and morbidity ramifications of PJI, optimization of perioperative voiding dynamics has recently become an area of study in the lower extremity arthroplasty population. Moverman et al. performed a break-even economic model to determine the viability of routine voiding optimization in BPH patients undergoing lower extremity TJA. ¹³ They found that medical intervention was economically justified whereas surgical intervention was less recommended. Surgical intervention was only a consideration when factoring in the long-term costs of PJI. Because this area of study has not been undertaken in upper extremity arthroplasty, further economic analyses should be performed to understand the cost benefit of urinary optimization in preventing postoperative complications in TSA patients with BPH. Moving forward, it is imperative to understand the reason for higher rates of PJI in patients with symptomatic BPH. Longer term prospective studies should be done to understand the underlying cause and incidence of PJI in male patients undergoing TSA.

This present study found that patients with BPH had longer in-hospital LOS (3-days vs. 2-days, p < 0.0001) with similar rates of 90-day readmissions (6.49% vs. 6.43%) compared to the control population. Further, long term prospective studies are needed to understand the influence of BPH history on shoulder arthroplasty reoperation, readmission, and LOS. One facet that has not been evaluated is the healthcare associated costs in patients with a history of BPH undergoing joint arthroplasty. The healthcare expenditures in the current study reflect an increase in patients with a history of BPH both on the day of surgery, and within the 90-day episode of care interval. Understanding where insurance and payer reimbursements are being allocated from a surgeon and hospital standpoint remain complex issues. Casp et al. reported trends and variation in hospital charges and payments compared with surgeon charges and payments for TSA in Medicare patients. ²² They found that hospital institutions were reimbursed at disproportionately higher rates relative to surgeons despite stable patient complexity and a decreasing LOS. These findings are important in the context of our study as upper extremity arthroplasty surgeons aim to provide quality and cost effective care while minimizing patient and financial risks.

There are some limitations to this retrospective study. The study relies on accurate coding for proper patient identification and procedural and diagnostic coding, however a presumed small amount of coding errors occur with the use of administrative databases. Patient charts cannot be directly reviewed for verification, thus the scope of the investigation is limited to information captured in billing codes. Additionally, analysis of this study was done using a single insurance database and the results may not be generalizable and a true representation of the association of BPH following total shoulder arthroplasty. Since the database does not track patients based on insurance plan but instead on provider network, longitudinal followup may be missed. Furthermore, there may be an underestimation of patients with symptomatic BPH due to a lack of clinical diagnosis, therefore influencing the strength of association of BPH on the dependent variables of interest Patients with BPH were shown to have an increased comorbidity burden, therefore the increased complications in the study group is subject to potential confounding. Despite controlling for the common comorbidities that are associated with BPH within the logistic regression model, it is possible the increased complication rate is explained by other factors. Lastly, there is a lack of information regarding postoperative catheterization, which may influence the results seen in the study. Chronic intermittent catheterization and indwelling urinary catheterization have been demonstrated to be a risk factor for PJI, and to what extent this plays a role in postoperative complications as it relates to BPH is an area of further research. There are other factors that potentially impact readmission and complication rates after total shoulder arthroplasty that are unable to be characterized within the database, including operative time, surgical approach, hospital volume, and anesthetic technique. Some of the medications (5-alpha reductase inhibitors −5ARIs) used as long term treatment for BPH have been hypothesized to be associated with an increased risk of VTE and could confound some of the results seen in our study. Furthermore, while the large sample size is an advantage to the study, a formal power analysis was not done due to the retrospective nature of the study. Despite these limitations, this is the first study to use a large patient population to analyze the association between BPH on multiple outcomes following total shoulder arthroplasty.

Conclusion

Patients with an BPH are at a significant risk for medical complications, PJIs, and healthcare expenditures following TSA within 90 days of surgery. In-hospital LOS was significantly longer in BPH patients, despite 90-day readmission rates similar to controls. The current investigation can be used by healthcare professionals to educate patients who have BPH on the potential outcomes following their TSA procedure. Future investigations should stratify the severity of BPH on postoperative outcomes following these orthopaedic procedures to identify at which threshold is it safe and cost effective to undergo primary TSA for the treatment of glenohumeral osteoarthritis. Furthermore, longer term follow-up is also recommended as this would allow to identify additional adverse events which may have been overlooked as in this investigation.