Scaphotrapeziotrapezoid Arthrodesis: A 10-Year Follow-up Study of Complications in 58 Wrists

Abstract

Background:

Scaphotrapeziotrapezoid (STT) arthrodesis is a procedure used for specific degenerative arthritis and instability patterns of the wrist. This study evaluates nonunion rate and risk factors for reoperation after STT arthrodesis in the Veterans Affairs Department patient population. The purpose of our study was to assess the long-term nonunion rate following STT arthrodesis and to identify factors associated with reoperation.

Methods:

The national Veterans Health Administration Corporate Data Warehouse and Current Procedural Terminology codes identified STT arthrodesis procedures from 1995 to 2016. Frequencies of total wrist arthrodesis (TWA) and secondary operations were determined. Univariate analyses provided odds ratios for risk factors associated with complications.

Results:

Fifty-eight STT arthrodeses were performed in 54 patients with a mean follow-up of 120 months. Kirschner wires (K-wires) were the most common fixation method (69%). Six wrists (10%) required secondary procedures: 5 TWAs and 1 revision STT arthrodesis. Four patients underwent additional procedures for nonunion (7%). Twenty-four patients required K-wire removal, 8 (14%) of these in the operating room, which were not included in regression analysis. Every increase in 1 year of age resulted in a 15% decrease in likelihood of reoperation (95% confidence interval: 0.77-0.93; P < .0001). Opioid use within 90 days before surgery (P = 1.00), positive smoking history (P = 1.00), race (P = .30), comorbidity count (P = .25), and body mass index (P = .19) were not associated with increased risk of reoperation.

Conclusions:

At a mean follow-up of 10 years, patients undergoing STT arthrodesis have a 10% risk of reoperation, and this risk decreases with older patient age. There was a symptomatic nonunion rate of 7%, similar to prior published rates. Patient demographics, comorbidity, smoking history, and opioid use did not appear to increase risk of reoperation.

Introduction

Scaphotrapeziotrapezoid (STT) arthrodesis has been used as a treatment for many types of wrist pathology. Scaphotrapeziotrapezoid arthrodesis can be indicated for STT arthritis,^1-3 carpal instability,^1,4,5 Kienböck disease,^1,6-11 and other wrist pathologies.^1,2,10,12-14 Despite its common use, the long-term effect is controversial. ¹⁵

In 1967, Peterson and Lipscomb ¹⁶ first described the use of STT fusion. Peer-reviewed literature related to STT arthrodesis is relatively rare. In 2003, Watson et al ¹⁷ reported outcomes in 800 STT fusions and found 36 (5%) patients developed nonunions, whereas 2% developed symptomatic radioscaphoid arthritis. Although a handful of other studies have confirmed similar findings,^6,10 multiple other reports have described higher rates of nonunion^18-23 or altered wrist kinematics after intercarpal arthrodesis.^24,25

We evaluated the long-term effect of STT fusion using the National Veterans Health Affairs (VHA) Corporate Data Warehouse (CDW). The purpose of this study was to evaluate the rate of secondary surgery performed after index STT arthrodesis and to identify factors associated with reoperation. Secondary aims included the assessment of long-term nonunion frequency. We hypothesized that perioperative tobacco exposure and more patient medical comorbidity would increase risk of complications and nonunion.

Materials and Methods

Our study was reviewed and approved by our institutional review board. The National VHA CDW is a widely used medical database containing clinical and demographic data for veterans using health care resources at Veterans Administration Hospitals throughout the United States. Current Procedural Terminology (CPT) codes (Supplemental Table S1) and the CDW database were used to identify all patients undergoing an intercarpal fusion with (25825) or without autograft (25820) between July 1,1995, and November 30, 2016.

Only adult patients aged ≥18 years were considered for inclusion. We excluded patients with intercarpal fusion other than STT arthrodesis (eg, capitolunate or 4-corner arthrodesis), those who underwent additional unrelated procedures at the same time as index STT arthrodesis, and where the clinical record lacked sufficient data to confirm the index procedure. Follow-up was estimated through patients actively receiving care through the Veterans Affairs (VA) Department throughout the duration of the follow-up period. Figure 1 is an attrition figure of patients who were included based on the exclusion criterion.

Figure 1.

Attrition of patients included in analysis based on study selection criteria.

Note. CPT = Current Procedural Terminology; CDW = Corporate Data Warehouse; FCA = 4-corner arthrodesis; STT = scaphotrapeziotrapezoid.

Chart review was performed to review relevant clinical documentation, including operative reports, brief operative reports, and clinical notes both preoperatively and postoperatively. Specific factors were extracted from a combination of chart review and CDW data, including variables such as age, body mass index (BMI), employment status, tobacco use, and opioid utilization within 90 days of surgery. Fifty-eight clinically relevant diagnoses were identified 1 year before STT arthrodesis using International Classification of Disease codes and included diagnoses such as infection, cancer, diabetes, and mental health disorders (Supplemental Table S2). These data were then quantified in the analysis by “count,” meaning the number of diagnoses (N = 11) documented per STT arthrodesis in the year before surgery.

The primary outcome was return to the operating room (OR) for any secondary surgery related to the index procedure. Patients undergoing Kirschner wire (K-wire) removal were noted, but this was not considered a secondary surgery given that it was considered a planned procedure with K-wire fixation, even if the patient returned to the OR for the removal of K-wire(s). This was done to avoid artificially inflating the rate of reoperation with “planned” procedures that did not deviate from standard of care.

Statistical Analysis

Exact logistic regression analyses were used to calculate the odds ratios and 95% confidence intervals (CIs) for reoperation, adjusting for age, BMI, race, smoking status, opioid use, comorbidity count, and procedure techniques. We chose to perform univariate analysis given the limited sample size. Microsoft SQL Server was used for data management and extraction, and SAS 9.4 was used to perform the statistical analyses.

Results

Patient Demographics

A total of 1861 patients were identified using CPT codes (25820 and 25825) and the CDW database from January 1, 1995, to November 30, 2016. Of these, 54 patients were confirmed to have undergone an isolated STT arthrodesis by chart review (Figure 1: database attrition) with a mean follow-up of 120 months (SD: ±76.5; range: 12-340 months) and a minimum of 12 months. Four patients had bilateral STT arthrodeses performed for a total of 58 STT procedures. Indications for procedure included STT arthrosis (49), rotatory subluxation (2), scapholunate dissociation (2), scapholunate instability (2), scapholunate tear (1), Kienböck disease (1), and unknown (1). The average patient age at index procedure was 56.0 ± 11.3 years. A total of 56 STT arthrodesis procedures were performed on male patients, and 2 were performed on female patients. The mean BMI was 29.1% ± 4.7%, and 75.9% of patient wrists had positive smoking history. Approximately 41% of patient wrists had documented handedness, and 88% of these underwent surgery on the dominant hand. A documented 19.0% of patients were using an opioid medication within the 90 days prior to their index STT procedure. Of the 20 patients with known preoperative employment (34%), 75% worked in a manual labor field. (Table 1)

Table 1.

Patient Demographic and Surgical Characteristics.

Patient demographics and surgical characteristics in 58 STT arthrodesis procedures
Variable	N a
Age, mean (±SD), y	56 (±11.3)
BMI, mean (±SD)	29.1 (±4.7)
Sex
Male	56
Female	2
Race
White	51
Nonwhite	7
Labor employment
Yes	15
No	5
Unknown	38
Handedness (surgery on dominant hand)
Dominant	21
Nondominant	3
Unknown	34
Smoking
Yes	44
No	5
Unknown	9
Total number of Clinical Classifications Software category, mean (±SD)	3.40 (±2.57)
Preoperative opioid use
Yes	11
No	47
Surgical implants b
K-wires	40
Screws	3
Staples	6
Plate	5
Other	1
Bone graft used
Yes	49
No	9
Type of graft
Scaphoid	2
Trapezoid	1
Iliac crest	1
Distal radius	45

Note. STT = scaphotrapeziotrapezoid; BMI = body mass index; K-wires = Kirschner wire.

^aN may be less than 58 if the data point was not available in chart review.

^bNot mutually exclusive. For data available, a small number of patients may have had hybrid fixation such as K-wires + screw fixation.

Procedure Details

Arthrodesis fixation methods consisted of: staples (6), plates (5), screws (3), and K-wires (40) (Table 1). The joint was decorticated in 44 (75.86%) wrists, and 1 operative report mentioned a posterior interosseous neurectomy was performed. Thirteen wrists (22.4%) underwent concurrent radial styloidectomy at the time of index procedure.

Of the 58 STT fusions performed, 6 wrists (10.3%) required return to the OR for a secondary procedure other than K-wire removal. Three patients (3.5%) underwent multiple (≥2) additional surgeries after index STT, although the regression model only used the first outcome to avoid artificially inflating the reoperation rate. A total of 86.2% of wrists did not require additional surgery. Four patients experienced nonunion after index STT arthrodesis (7%). Five patients returned to the OR for conversion to total wrist arthrodesis (TWA)—3 for STT nonunion and 2 for persistent pain. One patient underwent revision STT arthrodesis due to continued pain and instability (Table 2). One patient who experienced nonunion did not elect to have further surgery. The average time between the index STT arthrodesis and conversion to TWA was 82.4 ± 115.78 months. Figure 2 demonstrates the 5-year survivorship of index STT fusion in our patient population. Ninety-two percent (95% CI: 81%-97%) of patients had 5-year survivorship free of reoperation following their original STT fusion. Of patients who underwent reoperation, 88% had known history of tobacco usage.

Table 2.

Secondary Surgery and Procedures Performed in Patients Following Index STT Arthrodesis.

Reoperation and removal of hardware following STT arthrodesis a
Secondary procedure	No. of outcomes	Percentage
Total wrist fusion	5	8.6
Revision STT arthrodesis	1	1.7
Removal of K-wires in the operating room a	8	13.8
Total	14	24.1

Note. STT = scaphotrapeziotrapezoid; K-wires = Kirschner wire.

^aTwenty-four patients required removal of K-wire fixation following index STT arthrodesis. Eight patients returned to the operating room for this procedure, and 16 were removed in clinic. Kirschner wire removal was not considered an outcome in the exact logistic regression analyses.

Figure 2.

Kaplan-Meier survival curve for STT arthrodesis.

Note. STT = scaphotrapeziotrapezoid.

Important to note are the 24 K-wire removal procedures performed. While 16 of them were performed in the office, 8 wrists (13.8%) required return to the OR for this procedure (Table 2). The overall rate of return to the OR was therefore 24%, when K-wire removals were included.

Risk Factors for Reoperation

Exact univariate logistic regression was performed on the 58 STT wrists with the outcomes of interest (Table 3). Age at the time of index STT arthrodesis statistically influenced risk of reoperation. With every increase in 1 year of age, there was a 15% decrease in likelihood of reoperation (95% CI: 0.76-0.93, P < .0001). There was a 19% rate of opioid pain medication utilization prior to surgery, although this did not increase the risk of reoperation (odds ratio = 0.84, CI = 0.016-8.87, P = 1.00). Positive smoking history (odds ratio = 1.65, CI = 0.16-84.81, P = 1.00), race (odds ratio = 0.22, CI = 0.02-3.02, P = .30), comorbidity profile (odds ratio = 0.78, CI = 0.49-1.15, P = .25), and BMI (odds ratio = 1.14, CI = 0.94-1.41, P = .19) were not associated with increased risk of secondary surgery.

Table 3.

Univariate Exact Logistic Regression.

Variable	Odds ratio	LCL	UCL	P value
Age at STT Arthrodesis	0.85	0.76	0.93	<.0001
BMI	1.14	0.94	1.41	.19
Race	0.22	0.02	3.02	.30
Smoking	1.65	0.16	84.81	1.00
Opioid use	0.84	0.016	8.87	1.00
Comorbidity count	0.78	0.49	1.15	.25
Bone graft utilization	0.91	0.084	48.31	1.00
Joint decortication	0.56	0.069	6.91	.83

Note. LCL = lower bound of 95% confidence interval; UCL = upper bound of 95% confidence interval; STT = scaphotrapeziotrapezoid; BMI = body mass index.

Discussion

The purpose of this study was to identify risk factors for reoperation after STT arthrodesis. The main finding of this study was that increasing age appeared to decrease the rate of unplanned reoperation after STT arthrodesis in a nationwide cohort with minimum 1-year and mean 10-year follow-up. In addition, we note a 7% nonunion rate and 10% unplanned reoperation rate after STT arthrodesis. These findings are relatively consistent with the peer-reviewed literature to date, where nonunion rates after STT arthrodesis vary.^{14,19,21,23,26-29} Prior research has demonstrated a nonunion rate range from 2.5% ²⁸ to 33% found in the 1997 meta-analysis by Larsen et al. ³⁰ The STT arthrodesis indicated for Kienböck disease has shown rates of nonunion from 4% ¹⁴ to 15%. ²⁷

Other documented complications after index STT arthrodesis include the need for secondary styloidectomy ²⁷ and surgical revision. ¹⁹ Our study supports this prior finding of requiring additional procedures after index STT arthrodesis—with 8.6% of patients being converted to TWA and 1.7% requiring revision STT arthrodesis. There is a documented need for secondary styloidectomy in prior studies ²⁷ and high rates of radiostyloid impingement in the long-term follow-up of original patients in the study by Rogers and Watson, ³¹ which has led to the recommendation of concomitant radial styloid resection at the index procedure.^18,28,32,33 Interestingly, we did not note any secondary operations for radial styloidectomy in this cohort. The 22% rate of radial styloidectomy seen in our data underlines that this accompanies index STT fusion in approximately a quarter of patients in this database.

Of the risk factors for reoperation assessed in our models, only age at the time of index STT arthrodesis statistically influenced risk of reoperation, with decreasing likelihood for reoperation as patients aged. Reasons for this finding remain speculative. Risk for reoperation was not significantly influenced by the presence of comorbidities, race, tobacco use, or preoperative opioid use. Research dedicated to the impact of tobacco products and its influence on STT arthrodesis is not, to our knowledge, available in the literature. It seems reasonable to consider tobacco use a modifiable risk factor until this topic is better understood, given the influence of tobacco on nonunion rates after 4-corner arthrodesis. ³⁴

It is pertinent to consider other management options for STT osteoarthritis (OA) given the reoperation rate associated with STT arthrodesis. Alternative procedures have been described, including excisional arthroplasty, ³⁵ partial resection of scaphoid, ³⁶ conservative treatment (splints), ²⁶ fibrous arthroplasty, ²⁶ silicone arthroplasty, ²⁶ and trapezial replacement arthroplasty. ²⁶ Garcia-Elias ³⁵ demonstrated that excisional arthroplasty can be used to treat STT arthrosis with less complications seen than STT fusion. Finally, ligament reconstruction and interposition with and without trapeziectomy has been proposed as a reasonable option for both isolated STT arthritis and STT pathology combined with thumb trapeziometacarpal OA.^37,38

Limitations

There are several relevant limitations of our study that need mention. Our results have the potential for selection bias inherent in the retrospective nature. This and other limitations inherent in the retrospective design are a challenge for all similarly designed studies evaluating the effectiveness of STT arthrodesis. A multicenter randomized controlled clinical trial comparing the results of STT fusion with other treatment modalities would be very enlightening, although this is challenging given the relative infrequency of the procedure. ³⁹ In addition, we did not have access to radiographs, patient activity levels, or clinical outcomes such as postoperative satisfaction or pain. Similarly, surgeon preference or skill, and postoperative protocols were varied or unknown. Due to the sample size, univariate analyses were performed. The results may have been different if we had been able to control all covariates.

An additional limitation of our study is that our cohort was primarily composed of white male patients. Although this reflects the patient demographics of the VA, it may limit the ability to generalize the results to other sexes and ethnicities. It is unknown to what degree our patients were lost to follow-up or sought care outside the VA system for additional surgery, although veteran contact with the VHA remained consistent given the availability of our follow-up data. It is difficult to know to what degree this limitation could have altered rates of TWA, secondary surgeries, or other complications.

A major strength of our study is that it is one of the largest studies focused on reoperation after STT arthrodesis with an average of 10-year follow-up.^{3,14,19,21,23,26-29} Prior research included vastly different follow-up periods, from 2 years ¹⁹ up to just over 5 years. ²⁹ Our research included a 10-year follow-up, and patients were required to have a minimum of 12 months of follow-up, which provides the longest follow-up status after STT arthrodesis to our knowledge to date.

Finally, we emphasize the rate of overall return to the OR, whether for removal of hardware or revision surgery, and importance in counseling patients regarding this consideration. This is important because return to the OR may subject the patient to additional anesthetic or incurred costs not found in the clinical setting.

Conclusion

Our study demonstrates that reoperation after STT fusion is relatively common, up to and including conversion to TWA. Surgeons should take these results into consideration when deciding the utility of STT fusion for historically indicated patients. It is not clear whether tobacco use, preoperative opioid use, elevated BMI, or comorbidities increase the risk of secondary surgery after index STT arthrodesis.