The Association of Insurance Status and Complications After Carpal Tunnel Release

Abstract

Background:

Carpal tunnel release (CTR) is one of the most commonly performed procedures in hand surgery. Complications from surgery are a rare but significant patient dissatisfier. The purpose of this study was to determine whether insurance status is independently associated with complications after CTR.

Methods:

We retrospectively identified all patients undergoing CTR between 2008 and 2018 using the Indiana Network for Patient Care, a state-wide health information exchange, and built a database that included patient demographics and comorbidities. Patients were followed for 90 days to determine whether a postoperative complication occurred. To minimize dropout, only patients with 1 year of encounters after surgery were included.

Results:

Of the 26 151 patients who met inclusion criteria, 2662 (10.2%) had Medicare, 7027 (26.9%) had Medicaid, and 16 462 (62.9%) had commercial insurance. Compared with Medicare, Medicaid status (P < .001) and commercial insurance status (P < .001) were independently associated with postoperative CTR complications. The overall complication rate was 2.23%, with infection, wound breakdown, and complex regional pain syndrome being the most common complications. Younger age, alcohol use, diabetes mellitus, hypertension, and depression were also independently associated with complications.

Conclusions:

The incidence of complications after CTR is low. Insurance status, patient demographics, and medical comorbidities, however, should be evaluated preoperatively to appropriately risk stratify patients. Furthermore, surgeons can use these data to initiate preventive measures such as working to manage current comorbidities and lifestyle choices, and to optimize insurance coverage.

Keywords

hand anatomy wrist carpal tunnel syndrome nerve diagnosis evaluation research and health outcomes surgery specialty infection pain

Introduction

Carpal tunnel syndrome (CTS) affects 2.7% to 5.8% of the overall adult population^1,2 and as much as 7.8% of working adults in the United States.^3,4 The incidence of CTS in the United States has risen over time from 258 cases per 100 000 person-years between 1981 and 1985 to 424 cases per 100 000 person-years between 2001 and 2005.⁵ Correspondingly, there was a 33% increase in the number of carpal tunnel release (CTR) procedures performed from 1996 to 2006.⁶ As a result, CTS has been called the most expensive upper extremity musculoskeletal disorder, costing approximately $2 billion in direct costs per year.^3,7

Carpal tunnel release is a low-risk outpatient procedure that can be performed by either open or endoscopic techniques with comparable improvement of symptoms, return to functional status, time to return to work, and overall complication rates.^8-10 Intraoperative complication rates for CTR are rare, whereas postoperative complications, such as complex regional pain syndrome (CRPS), infection, and wound dehiscence, may occur.¹¹ Despite the low complication rate, affected patients may suffer from loss of work time and/or employment and a decreased quality of life.¹² As such, an evaluation of factors associated with CTR complications is warranted to optimize patient outcomes.

Health insurance status is one socioeconomic variable included in studies assessing surgical outcomes, including after hand and upper extremity surgery. For example, in a study by Devana et al¹³ evaluating complications after endoscopic or open CTR, they reported differential rates of median nerve injury in patients with different insurance types. Given the potential impact on outcomes and as a tool for surgical risk stratification, we sought to better understand the relationship between insurance type and postoperative complications following CTR. We hypothesize that insurance status impacts the incidence of CTR postoperative complications.

Materials and Methods

Data Collection

We used the Indiana Network for Patient Care (INPC) to build a comprehensive database with unique and deidentified patients, 18 years of age and older, who underwent surgical treatment for CTS via open or endoscopic CTR between 2008 and 2018. The INPC is a health information exchange that links health care data from disparate sources in real time through a global patient index. Sources cover the continuum of inpatient and outpatient health care for more than 14 million patients; therefore, it includes the vast majority of patients, encounters, and health care data in the state of Indiana and surrounding catchment areas. Data inputs are both codified and free text data from sources such as electronic medical records, insurance claims, operative records, public health databases, pharmacies, picture archiving and communication system, and a variety of other sources.^14-17 Therefore, data are directly input by providers at the point of care and validated through insurance claims and other mechanisms. For research purposes, the Regenstrief Institute provides access to the proprietary system and quality assurance of the data through a team of data analysts.

We used the INPC to identify all patients who underwent isolated CTR as defined by Current Procedural Terminology (CPT) codes and the International Classification of Diseases (ICD) 9th or 10th Edition codes: CPT 29848 or 64721, ICD-9 04.43, and ICD-10 01N50ZZ, 01N53ZZ, or 01N54ZZ. Among those who underwent CTR, only those with follow-up in an INPC participating institution within 1 year of the procedure and documented insurance status at the time of the procedure were included, which helped to minimize patient dropout. For patients who had more than 1 CTR, we studied only the first procedure. Patients with a diagnosis of 1 or more of the following complications, based on ICD-9 or ICD-10, within 90 days of surgery were identified: hand infection, neurovascular injury, retained foreign body, wound breakdown, and CRPS (Supplemental Table 1). The recurrence of CTS was not included. Comorbidities previously reported as risk factors for postoperative CTR complications were also identified and recorded. Only comorbidities active within the year before surgery were included in the study. Patients were divided into 3 discrete cohorts: Medicare insurance, Medicaid insurance, and commercial insurance. Patient data were stored using Indiana University cloud computing and analyzed using Stata/MP 16.1 (StataCorp LLC, College Station, Texas) in compliance with the Health Insurance Portability and Accountability Act of 1996.

Statistical Analysis

We performed univariate analysis to analyze the population sample, followed by χ² and Fisher exact tests to examine the bivariate relationships between variables. We then used backward stepwise multivariable logistic regression analysis to examine the relationship between patient variables and CTR complications. In the stepwise backward-selection estimation method, we used a P value of .25 as the threshold for removal from the final model. Factors included in the stepwise regression model comprised age, sex, alcohol use, diabetes mellitus (DM), hyperlipidemia, hypertension, arthritis, peripheral vascular disease, congestive heart failure, coronary artery disease, chronic kidney disease, depression, and coronary artery disease. Body mass index (BMI) data were reported in only 37.1% of patients, so was excluded from the multivariable analysis. Robust variance method was used to report standard errors and P values to account for uncertainty in estimates due to model misspecification, if any.^18-21 Adjusted odds ratios and 95% confidence intervals were calculated for each factor, with P < .05 considered statistically significant. Multicollinearity in the linear regression model was quantified via variance inflation factor (VIF), which indicates the strength of the collinearity and whether the degree of variances for each predictor in the multivariable model is inflated due to this collinearity. In general, a VIF >10 is considered to have a detrimental effect on the relationship between the predictors and the outcome.^22,23

Results

Overall, 26 151 patients met inclusion criteria: 2662 (10.2%) had Medicare, 7027 (26.9%) had Medicaid, and 16 462 (62.9%) had commercial insurance. Patient demographics varied across insurance cohorts. The average BMI was 32.4 ± 7.7 kg/m², and the Medicare cohort was the oldest (74.7 ± 7.0 years). The Medicaid cohort had the highest proportion of women (76.4%) and alcohol use (11.6%) (Table 1). The Medicaid cohort had the highest proportion of several comorbidities (Table 2). There was a statistically significant difference in comorbidities between insurance groups (P < .001) (Table 2).

Table 1.

Patient Demographics.

Patient Characteristic	Medicare (N = 2662)	Medicaid (N = 7027)	Commercial (N = 16 462)	Total (N = 26 151)	P value
Patient Characteristic	No. of patients (%)	No. of patients (%)	No. of patients (%)	No. of patients (%)	P value
Sex^a
Female	1614 (60.7)	5313 (76.4)	10 481 (63.7)	17 408 (66.8)	<.001
Male	1045 (39.3)	1638 (23.6)	5974 (36.3)	8657 (33.2)
Mean age ± SD	74.7 ± 7.0	48.6 ± 13.6	54.0 ± 13.5	54.6 ± 14.9	<.001
BMI,^a kg/m²					<.001
<18.5	15 (0.84)	44 (1.25)	33 (0.75)	92 (0.95)
18.5-25	319 (17.9)	547 (15.6)	691 (15.8)	1557 (16.1)
25-30	606 (33.9)	708 (20.2)	1217 (27.8)	2531 (26.2)
≥30	846 (47.4)	2211 (63.0)	2440 (55.7)	5497 (56.8)
Total	1786	3510	4381	9677 (100.0)
Mean BMI ± SD	30.5 ± 6.3	33.8 ± 8.7	32.0 ± 7.2	32.4 ± 7.7	<.001
Alcohol use	41 (1.54)	813 (11.6)	319 (1.94)	1173 (4.49)	<.001

Note. BMI = body mass index.

Demographic data for all patients were not available. Percentages for BMI pertain to patients for whom the information was available.

Table 2.

Patient Comorbidities.

Comorbidity	Medicare (N = 2662)	Medicaid (N = 7027)	Commercial (N = 16 462)	Total (N = 26 151)	P value
Comorbidity	No. (%)	No. (%)	No. (%)	No. (%)	P value
Diabetes mellitus	882 (33.1)	2976 (42.4)	3541 (21.5)	7399 (28.3)	<.001
Hyperlipidemia	2037 (76.5)	4329 (61.6)	8767 (53.3)	15 133 (57.9)	<.001
Hypertension	1975 (74.2)	4675 (66.5)	8601 (52.3)	15 251 (58.3)	<.001
Arthritis	341 (12.8)	951 (13.5)	1185 (7.20)	2477 (9.47)	<.001
Peripheral vascular disease	233 (8.75)	925 (13.2)	784 (4.76)	1942 (7.43)	<.001
Congestive heart failure	282 (10.6)	960 (13.7)	735 (4.46)	1977 (7.56)	<.001
Coronary artery disease	759 (28.5)	1708 (24.3)	2134 (13.0)	4601 (17.6)	<.001
Chronic kidney disease	332 (12.5)	623 (8.87)	535 (3.25)	1490 (5.70)	<.001
Liver disease	116 (4.36)	886 (12.6)	687 (4.14)	1689 (6.46)	<.001
Lung disease	1514 (56.9)	3419 (48.7)	3793 (23.0)	8726 (33.4)	<.001
Depression	489 (18.4)	4193 (59.7)	3462 (21.0)	8144 (31.1)	<.001
Hypercoagulability	13 (0.49)	56 (0.80)	49 (0.30)	118 (0.45)	<.001

Postoperatively, 11 (0.4%) Medicare patients, 303 (4.3%) Medicaid patients, and 268 (1.6%) commercial insurance patients had at least 1 complication (P < .001). The most common complications were infection (1.65%), wound breakdown (0.43%), and CRPS (0.21%). Cohort differences for each complication were statistically significant (P ≤ .001) (Table 3). There were 248 (3.35%) patients with DM and 334 (1.78%) non-DM patients who developed a complication (P < .001). For patients with DM, infection (2.53%) and wound breakdown (0.68%) were the most common complications (Table 4).

Table 3.

Complications After Carpal Tunnel Release.

Complication	Medicare (N = 2662)	Medicaid (N = 7027)	Commercial (N = 16 462)	Total (N = 26 151)	P value
Complication	No. (%)	No. (%)	No. (%)	No. (%)	P value
Infection	8 (0.30)	223 (3.17)	200 (1.21)	431 (1.65)	<.001
Wound breakdown	2 (0.08)	63 (0.90)	48 (0.29)	113 (0.43)	<.001
Foreign body	0	0	0	0
Vascular injury	0	0	0	0
Nerve injury	1 (0.04)	3 (0.04)	2 (0.01)	6 (0.02)	.212
CRPS	0	24 (0.34)	31 (0.19)	55 (0.21)	.001
Other	1 (0.04)	5 (0.07)	4 (0.02)	10 (0.04)	.202
Total unique patients^a	11 (0.41)	303 (4.31)	268 (1.63)	582 (2.23)	<.001

Note. CRPS = complex regional pain syndrome.

Some patients had multiple complications.

Table 4.

Complications After Carpal Tunnel Release in Patients With Diabetes Mellitus.

Complication	Diabetes (N = 7399)	No diabetes (N = 18 752)	P value
Complication	No. (%)	No. (%)	P value
Infection	187 (2.53)	244 (1.30)	<.001
Wound breakdown	50 (0.68)	63 (0.34)	<.001
Foreign body	0	0
Vascular injury	0	0
Nerve injury	1 (0.01)	5 (0.03)	.458
CRPS	17 (0.23)	38 (0.20)	.666
Other	6 (0.08)	4 (0.02)	.036
Total unique patients^a	248 (3.35)	334 (1.78)	<.001

Note. CRPS = complex regional pain syndrome.

Some patients had multiple complications.

In multivariable analysis, those with Medicaid (P < .001) and commercial insurance (P < .001) were more likely than Medicare patients to sustain a postoperative CTR complication. There was no significant association between sex and postoperative complications. Younger age (P < .001) and alcohol use (P = .008) were associated with increased odds for complications. Diabetes mellitus (P < .001), hypertension (P = .018), and depression (P < .001) were the only comorbidities independently associated with an increased odds for postoperative complications (Table 5).

Table 5.

Multivariable Logistic Regression for Any Complication After Carpal Tunnel Release.

Variable	Adjusted OR(95% CI)	P-value
Insurance type
Medicare	Reference
Medicaid	5.30 (2.81-10.00)	< .001
Commercial	3.08 (1.66-5.74)	< .001
Patient demographics
Age	0.98 (0.97-0.99)	< .001
Sex
Male	Reference
Female	0.88 (0.73-1.06)	.167
Alcohol use	1.48 (1.11-1.97)	.008
Medical comorbidities
Diabetes mellitus	1.44 (1.18-1.75)	< .001
Hypertension	1.29 (1.05-1.60)	.018
Peripheral vascular disease	1.30 (0.98-1.72)	.066
Congestive heart failure	1.30 (0.99-1.70)	.063
Liver disease	1.25 (0.96-1.62)	.092
Depression	1.44 (1.18-1.76)	< .001

Note. The model considers only comorbidities with a value of P < .25 in stepwise analysis. OR = odds ratio; CI = confidence interval.

Discussion

In this study, we found that insurance status independently conferred an increased odds for complications after CTR. Our findings are consistent with other studies showing that Medicaid patients have higher rates of postoperative complications.^24-26 In those studies, the phenomenon was partly attributed to potential biases related to insurance status.²⁵ Compared with commercial insurance, Medicaid typically reimburses less, and a smaller cohort of physicians accept it for coverage.²⁷ As such, insurance status alone can have implications for surgeon access (eg, a smaller pool of qualified providers) and can lead to delays before treatment.

It is also possible that insurance status may be, in some cases, a relative proxy for patient-related lifestyle choices. In our study, the Medicaid cohort had the highest proportion of patients with most of the assessed comorbidities and had a higher average BMI and rate of alcohol consumption. The higher rate of comorbidities may be related to decreased health literacy, language barriers, and nutritional deficiencies that can predispose to poorer health outcomes. Insurance status may also be indirectly related to postoperative adherence to recommendations. Patients with less generous insurance coverage may be unable to afford a prescribed antibiotic or undergo a full course of occupational therapy. These physician- and patient-related considerations may partially explain the increased incidence of infection and wound breakdown in the Medicaid cohort and how insurance status may be independently associated with an increased odds for complications after CTR.

The overall rate of complications in this study was relatively low (2.23%) and was similar to previous estimates of 2.5% after CTR.²⁸ Although 1.65% of all patients in our study developed an infection, only 0.30% of Medicare patients developed a post-CTR surgical site infection. This rate was similar to a study in 454 987 Medicare patients that showed a postoperative infection rate of 0.32%.²⁹ The incidence of infection was nearly double in patients with DM. This finding may support the use of perioperative antibiotic prophylaxis in this cohort. Previous studies have reported post-CTR CRPS rates of between 2% and 8%,^12,30-32 which is significantly higher than our finding of 0.21%; it is important to note that, however, some of the randomized controlled studies used comparatively smaller sample sizes of often less than 200 patients. Wound breakdown/dehiscence rates in our study (0.43%) were similar to previous estimates in the literature of 0.2%.²⁸ Finally, neurovascular injury rates in our study (0.02%) were similar to those reported in the literature (0.10%-0.30%).¹²

There are multiple relevant limitations to this study. Race data were not recorded for many patients. It has been documented in vascular surgery that nonwhite Hispanic and black patients are more likely to have adverse outcomes and more likely to be enrolled in Medicaid than white non-Hispanics.²⁴ All data were derived from an administrative database, so there are additional concerns for nonreporting bias and inaccurate data abstraction, which can confound results. Also, all data in this study came from a single state, which may limit generalization across the United States. We attempted to address selection bias by only using patients with documented postoperative follow-up. There was no concern for collinearity in this study as the VIF never exceeded 10 in any of the variables studied. Despite these limitations, our findings remain clinically relevant, and surgeons can use these data for considerations in CTR to optimize health care outcomes.

Conclusions

Complications after CTR are rare, yet insurance status, patient age, alcohol use, DM, hypertension, and depression are independently associated with postoperative complications. Surgeons can use these findings to appropriately stratify patients preoperatively and to initiate preventive measures such as working to manage current comorbidities and lifestyle choices, and to optimize insurance coverage.

Supplemental Material

sj-pdf-1-han-10.1177_1558944721990818 – Supplemental material for The Association of Insurance Status and Complications After Carpal Tunnel Release

Supplemental material, sj-pdf-1-han-10.1177_1558944721990818 for The Association of Insurance Status and Complications After Carpal Tunnel Release by Phoebus Sun Cao, Scott N. Loewenstein, Lava R. Timsina and Joshua M. Adkinson in HAND

Footnotes

Acknowledgements

The authors would like to acknowledge Dr. Benjamin Lancaster for his assistance in preparing this manuscript.

Ethical Approval

This study was approved by our institutional review board.

Statement of Human and Animal Rights

This article does not contain any studies with human or animal subjects.

Statement of Informed Consent

Informed consent was obtained when necessary.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Phoebus Sun Cao

Scott N. Loewenstein

Supplemental material is available in the online version of the article.

References

Atroshi

Gummesson

Johnsson

, et al. Prevalence of carpal tunnel syndrome in a general population. JAMA. 1999;282(2):153-158.

Latinovic

Gulliford

Hughes

RAC

. Incidence of common compressive neuropathies in primary care. J Neurol Neurosurg Psychiatry. 2006;77(2):263-265.

Dale

Harris-Adamson

Rempel

, et al. Prevalence and incidence of carpal tunnel syndrome in US working populations: pooled analysis of six prospective studies. Scand J Work Environ Health. 2013;39(5):495-505.

Newington

Harris

Walker-Bone

Carpal tunnel syndrome and work. Best Pract Res Clin Rheumatol. 2015;29(3):440-453.

Gelfman

Melton

III Yawn

, et al. Long-term trends in carpal tunnel syndrome. Neurology. 2009;72(1):33-41.

Fajardo

Kim

Szabo

RM.

Incidence of carpal tunnel release: trends and implications within the United States ambulatory care setting. J Hand Surg Am. 2012;37(8):1599-1605.

Falkiner

Myers

When exactly can carpal tunnel syndrome be considered work-related. ANZ J Surg. 2002;72(3):204-209.

Brown

Gelberman

Seiler

, et al. Carpal tunnel release. A prospective, randomized assessment of open and endoscopic methods. J Bone Joint Surg Am. 1993;75(9):1265-1275.

Atroshi

Hofer

Larsson

, et al. Open compared with 2-portal endoscopic carpal tunnel release: a 5-year follow-up of a randomized controlled trial. J Hand Surg Am. 2009;34(2):266-272.

10.

Zuo

Zhou

Wang

, et al. Endoscopic versus open carpal tunnel release for idiopathic carpal tunnel syndrome: a meta-analysis of randomized controlled trials. J Orthop Surg Res. 2015;10:12.

11.

Loewenstein

Duquette

Adkinson

JM.

Combined carpal tunnel release and palmar fasciectomy for Dupuytren’s contracture does not increase the risk for complex regional pain syndrome. Plast Reconstr Surg. 2018;142(5):1251-1257.

12.

Santosa

Chung

Waljee

JF.

Complications of compressive neuropathy: prevention and management strategies. Hand Clin. 2015;31(2):139-149.

13.

Devana

Jensen

Yamaguchi

, et al. Trends and complications in open versus endoscopic carpal tunnel release in private payer and Medicare patient populations. Hand (N Y). 2019;14(4):455-461.

14.

Overhage

Dexter

Perkins

, et al. A randomized, controlled trial of clinical information shared from another institution. Ann Emerg Med. 2002;39(1):14-23.

15.

Biondich

Grannis

SJ.

The Indiana Network for Patient Care: an integrated clinical information system informed by over thirty years of experience. J Public Health Manag Pract. 2004; suppl:S81-S86.

16.

The Indiana Network for Patient Care. A working local health information infrastructure. Health Aff (Millwood). 2005;24(5):1214-1220.

17.

Dixon

Whipple

Lajiness

, et al. Utilizing an integrated infrastructure for outcomes research: a systematic review. Health Info Libr J. 2016;33(1):7-32.

18.

Voorman

Lumley

McKnight

, et al. Behavior of QQ-plots and genomic control in studies of gene-environment interaction. PLoS ONE. 2011;6(5):e19416.

19.

Rao

Province

MA.

A framework for interpreting type I error rates from a product-term model of interaction applied to quantitative traits. Genet Epidemiol. 2016;40(2):144-153.

20.

Nsengimana

Barrett

JH.

Power, validity, bias and robustness of family-based association analysis methods in the presence of linkage for late onset diseases. Ann Hum Genet. 2008;72(pt 6):793-800.

21.

Rosenblum

van der Laan

MJ.

Using regression models to analyze randomized trials: asymptotically valid hypothesis tests despite incorrectly specified models. Biometrics. 2009;65(3):937-945.

22.

Kutner

Nachtsheim

Neter

, et al. Applied Linear Statistical Models. 5th ed. Boston: McGraw-Hill Irwin; 2005.

23.

Yoo

Mayberry

Bae

, et al. A study of effects of multicollinearity in the multivariable analysis. Int J Appl Sci Technol. 2014;4(5):9-19.

24.

Giacovelli

Egorova

Nowygrod

, et al. Insurance status predicts access to care and outcomes of vascular disease. J Vasc Surg. 2008;48(4):905-911.

25.

LaPar

Bhamidipati

Mery

, et al. Primary payer status affects mortality for major surgical operations. Ann Surg. 2010;252(3):544-50; discussion 550-541.

26.

Kelz

Gimotty

Polsky

, et al. Morbidity and mortality of colorectal carcinoma surgery differs by insurance status. Cancer. 2004;101(10):2187-2194.

27.

Skaggs

Lehmann

Rice

, et al. Access to orthopaedic care for children with medicaid versus private insurance: results of a national survey. J Pediatr Orthop. 2006;26(3):400-404.

28.

Lipira

Sood

Tatman

, et al. Complications within 30 days of hand surgery: an analysis of 10,646 patients. J Hand Surg Am. 2015;40(9):1852-1859.e1853.

29.

Werner

Teran

Deal

DN.

Patient-related risk factors for infection following open carpal tunnel release: an analysis of over 450,000 Medicare patients. J Hand Surg Am. 2018;43(3):214-219.

30.

da Costa

de Oliveira

Fernandes

, et al. Incidence of regional pain syndrome after carpal tunnel release. Is there a correlation with the anesthetic technique? Rev Bras Anestesiol. 2011;61(4):425-433.

31.

Mertz

Trunzter

, et al. National trends in the diagnosis of CRPS after open and endoscopic carpal tunnel release. J Wrist Surg. 2019;8(3):209-214.

32.

Shinya

Lanzetta

Conolly

WB.

Risk and complications in endoscopic carpal tunnel release. J Hand Surg Br. 1995;20(2):222-227.

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

0.17 MB