Evaluating Immediate and Short-Term Postoperative Clinical Outcomes of Patients Undergoing Ulnar Shortening for Ulnar Impaction Syndrome Using PROMIS

Abstract

Background  The early recovery trajectory of patients undergoing ulnar shortening for ulnar impaction syndrome using the Patient-Reported Outcomes Measurement Information System (PROMIS) is unknown.

Questions/Purposes  Using PROMIS Upper Extremity (UE), Physical Function (PF), Pain Interference (PI), and Depression, we asked (1) do patients undergoing operative management for ulnar impaction syndrome present at their preoperative visit with notable impairment?; (2) At immediate follow-up, do patients present with a clinically appreciable change in symptom severity?; and (3) At short-term follow-up, do patients present with a clinically appreciable change in symptom severity?

Materials and Methods  We identified patients from 01/2017 to 12/2019 at our institution undergoing ulnar shortening for ulnar impaction syndrome who completed all PROMIS domains at a preoperative visit and at least one postoperative time point (i.e., less than 4 weeks and/or greater than 12 weeks). Distribution- and anchor-based minimal clinically important difference estimates were used to evaluate clinically appreciable changes in symptoms over time.

Results  A total of 38 patients met our inclusion criteria. The average change in PROMIS UE, PF, PI, and Depression scores from preoperative to immediate postoperative follow-up were –3.8, –4.3, 3.2, and 0.5, respectively. However, by short-term follow-up, the average change in PROMIS UE, PF, PI, and Depression scores were 3.7, 3.2, –4.7, and –3.9, respectively.

Conclusions  Patients have worsening function at the immediate postoperative follow-up. By short-term postoperative follow-up, functional status and PI levels improve. Our findings can help hand surgeons provide evidence-based guidance on expected initial recovery following operative management for ulnar impaction syndrome.

Level of Evidence  This is a level II, prognostic study.

Ulnar impaction syndrome, a degenerative condition produced by excessive force across the ulnocarpal joint, is a common causes of ulnar-sided wrist pain. ¹ Nonoperative management is often attempted initially. However, when this approach fails, surgery is indicated.

To help set patient expectations and engage in improved clinical shared decision-making discussions, understanding the expected degree and pattern of recovery following operative management of ulnar impaction syndrome is of value. There is a growing body of literature examining the clinical outcomes of patients undergoing operative management for ulnar impaction syndrome, with many studies utilizing patient-reported outcome measures (PROMs) to capture the outcomes most important to patients. ^{2 3 4} Some of the PROMs utilized have included the Disabilities of the Arm, Shoulder, and Hand (DASH), Mayo Wrist Score, Gartland and Werley score, pain numeric rating sale, and visual analogue scale (VAS), among others. ^{2 3 4} However, to our knowledge, there is no study to date that utilizes Patient-Reported Outcomes Measurement Information System (PROMIS) questionnaires to evaluate clinical recovery following surgery for ulnar impaction syndrome.

Unlike legacy instruments, PROMIS is a general set of PROMs that allows for a more robust evaluation of a patient's overall health and well-being when the electronic versions of the questionnaires are used; in such cases, PROMIS utilizes item response theory as part of a computer adaptive test. ⁵ Further, PROMIS forms have been shown to perform similarly to or better than many of the legacy upper extremity(UE) PROMs. ^{6 7 8 9} However, to our knowledge, PROMIS questionnaires have not been robustly studied in patients with ulnar impaction syndrome. Nonetheless, there has been a growing movement to utilize PROMIS to monitor clinical outcomes in the setting of hand and upper conditions and surgery. ^{10 11} PROMIS domains are designed to follow a normal distribution based on a general US population and produce a t-score with a mean of 50 and an standard deviation (SD) of 10. ¹² Higher PROMIS Physical Function (PF) and Upper Extremity (UE) scores indicate improved functional status, while lower Pain Interference (PI) and Depression scores indicate less pain with activity and mental health distress. ¹²

In the present study, we aimed to answer three questions: 1) Do patients undergoing operative management for ulnar impaction syndrome present at their preoperative visit with notable impairment, as measured by PROMIS UE, PF, PI, and Depression?; 2) At immediate postoperative follow-up, do patients present with a clinically appreciable change in symptom severity (worsening or improvement), as measured by PROMIS domains?; and 3) At short-term postoperative follow-up, do patients present with a clinically appreciable change in symptom severity (worsening or improvement), as measured by PROMIS domains? We hypothesized that PROMIS questionnaires would be able to capture that patients present with notable clinical impairment preoperatively, worsen directly following operative management, and improve by short-term postoperative follow-up.

Materials and Methods

Our Institutional Review Board (IRB) approved the present study.

A retrospective analysis of our academic medical center's PROMIS data repository was undertaken. As part of routine clinical care as our institution, ¹³ all patients presenting to a hand clinic for evaluation are asked to complete PROMIS UE (v2.0), PF (v2.0), PI (v1.1), and Depression (v1.0) CATs.

We queried our database for all patients who underwent operative management for ulnar impaction syndrome with ulnar shortening (CPT Codes: 25360 or 25390) between March 2017 and December 2019. We excluded patients with additional procedures other than wrist arthroscopy. To be included in the study, patients were required to have a preoperative visit within 90 days of surgery at which all four PROMIS domains were fully completed. In addition, patients were required to have completed all four PROMIS domains postoperatively at a follow-up appointment less than 4 weeks after surgery and/or at a follow-up appointment greater than 12 weeks after surgery. All other patients were excluded.

In addition to PROMIS domain scores at each time point, the following patient characteristics were recorded to provide an overview of the sample: age (years), body mass index, sex (man or woman), self-reported race (white, black, or other/unknown), marital status (married, single, or other), smoking status (current, former, or never).

Statistical Analysis

Descriptive statistics were calculated and reported. Continuous variables were reported as means and SDs, while categorical variables were reported as counts and percentages. PROMIS UE, PF, PI, and Depression scores were compared between preoperative and immediate postoperative (average days since surgery: 16 days [SD: 14 days]) time points and preoperative and short-term postoperative (average days since surgery: 193 days [SD: 116 days]) time points. Differences in PROMIS scores between preoperative and each postoperative time points were reviewed compared with minimal clinically important difference (MCID) estimates. Using the standard distribution-based approach (i.e., one half of one SD of a given PROM), ¹⁴ the MCID estimates for PROMIS UE, PF, PI, and Depression in our patient sample were 3.7, 3.7, 4.2, 5.0, respectively. From the general hand literature, we used 2.1 and 1.7 as the anchor-based MCID estimates for PROMIS UE and PF, respectively. ¹⁵ However, to our knowledge, no anchor-based PROMIS PI or Depression MCID estimate exists among a general hand population or among patients undergoing ulnar shortening for ulnar impaction syndrome.

A power analysis for a two-tail paired t -test was performed and demonstrated that for a medium effect size (Cohen's d  = 0.5), α of 0.05, and power of 0.80, a total sample of 34 patients who had preoperative and follow-up PROMIS scores at each time point would be needed. Unfortunately, our sample did not meet this minimum for an appropriately powered analysis. Thus, because of the limited sample size, no comparative statistical tests (e.g., paired t -tests) were utilized. All descriptive statistics were calculated using Stata/SE 14.2 for Mac (StataCorp, College Station, TX).

Results

A total of 38 patients fit our inclusion criteria. Among those, 29 patients (76%) had complete PROMIS domain scores at a preoperative visit and at least the immediate follow-up visit, while 26 patients (68%) had complete PROMIS domain scores at a preoperative visit and at least the short-term follow-up visit. Across all patients, the average age was 50 years (SD: 17 years) and a majority of patients were women ( n  = 29 [76%]) ( Table 1 ). Over one-third of patients were former smokers ( n  = 13 [37%]) ( Table 1 ). The average time from preoperative clinic visit to surgery was 39 days (SD: 24 days) ( Table 1 ). The average time from surgery to immediate and short-term postoperative clinic visits were 16 days (SD: 14 days) and 193 days (SD: 116 days), respectively ( Table 1 ).

Table 1. Patient characteristics ( n  = 38) .

Characteristic	n (%) or mean (SD)
Age, years	50 (17)
BMI	30 (6)
Sex
Woman	29 (76)
Man	9 (24)
Race
White	26 (95)
Black	2 (6.3)
Marital status
Married	22 (58)
Single	10 (26)
Other	6 (16)
Smoking status
Current	1 (2.6)
Former	14 (37)
Never	23 (61)
Average time points
Preoperative visit to surgery	39 (25)
Surgery to immediate visit	16 (14)
Surgery to short-term visit	193 (116)

Abbreviation: SD, standard deviation.

The average preoperative PROMIS UE, PF, PI, and Depression scores were 34.6 (SD: 7.4), 40.9 (SD: 7.4), 58.9 (SD: 8.5), and 48.1 (SD: 10.0), respectively. The average immediate postoperative follow-up PROMIS UE, PF, PI, and Depression scores were 30.8 (SD: 7.3), 36.6 (SD: 7.5), 62.1 (SD: 7.7), and 48.6 (SD: 9.8), respectively. The average short-term postoperative follow-up PROMIS UE, PF, PI, and Depression scores were 38.3 (SD: 9.8), 44.1 (SD: 7.3), 54.2 (SD: 8.2), and 44.2 (SD: 9.3), respectively. The recovery trajectory, as measured by PROMIS UE, PF, PI, and Depression, can be visualized in Figs. 1–4 , respectively.

Fig. 1.

An illustration of the recovery trajectory, as measured by PROMIS UE, of patients undergoing ulnar shortening for ulnar impaction syndrome. Error bars represent the standard deviation of PROMIS UE at each time point. PROMIS UE, Patient-Reported Outcomes Measurement Information System Upper Extremity.

Fig. 2.

An illustration of the recovery trajectory, as measured by PROMIS PF, of patients undergoing ulnar shortening for ulnar impaction syndrome. Error bars represent the standard deviation of PROMIS PF at each time point. PROMIS PF, Patient-Reported Outcomes Measurement Information System Physical Function.

Fig. 3.

An illustration of the recovery trajectory, as measured by PROMIS PI, of patients undergoing ulnar shortening for ulnar impaction syndrome. Error bars represent the standard deviation of PROMIS PI at each time point. PROMIS PI, Patient-Reported Outcomes Measurement Information System Pain Interference.

Fig. 4.

An illustration of the recovery trajectory, as measured by PROMIS Depression, of patients undergoing ulnar shortening for ulnar impaction syndrome. Error bars represent the standard deviation of PROMIS Depression at each time point. PROMIS, Patient-Reported Outcomes Measurement Information System.

The average change in PROMIS UE, PF, PI, and Depression scores from preoperative to immediate postoperative follow-up were –3.8, –4.3, 3.2, and 0.5, respectively. PROMIS UE and PF demonstrated clinically appreciable worsening of functional symptoms when considering either distribution- or anchor-based MCID estimates. Neither PROMIS PI nor depression demonstrated a clinically appreciable change from the preoperative time point to the immediate postoperative follow-up clinic visit.

The average change in PROMIS UE, PF, PI, and Depression scores from preoperative to short-term postoperative follow-up were 3.7, 3.2, –4.7, and –3.9, respectively. Using anchor-based MCID estimates, both PROMIS UE and PF demonstrated clinically appreciable symptom improvement at short-term postoperative follow-up. When using distribution-based MCID estimates, PROMIS UE and PI demonstrated clinically appreciable symptom improvement at short-term postoperative follow-up. PROMIS Depression did not demonstrate a clinically appreciable change from the preoperative time point to the short-term postoperative follow-up clinic visit.

Discussion

When symptoms from ulnar impaction persist in spite of nonoperative treatment, surgery becomes the intervention of choice. A better understanding of the symptom state patients undergoing operative management present with preoperatively, as well as the trajectory of expected postoperative recovery, is helpful in setting patient expectations and improving shared clinical decision-making. Further, such insights can help alert hand surgeons when patients are deviating from expected recovery. Importantly, while the postoperative recovery trajectory found in this study may provide important for hand surgeons to consider when providing care for such patients, it is also vital that our findings are viewed with the understanding of our limited sample size. Nonetheless, the use of robust, validated general PROMs (e.g., PROMIS) that allow for an understanding of a patient's overall health in the setting of their UE condition and treatment may be ideal to best capture this important information in a more patient-centric way. In the present study, we found that patients with ulnar impaction syndrome undergoing operative management present with notable functional and pain impairment, which worsens over the first few weeks postoperatively; however, by 6 months, on average, patients report clinically appreciable improvement in UE function and pain.

Our study shows that patients with ulnar impaction syndrome have noteworthy functional and pain impairment prior to undergoing operative management. Preoperative PROMIS UE scores suggest that patients seeking treatment have UE functional limitations 1.5 SDs worse than the general population. In addition, preoperative PROMIS PF and PI scores demonstrate that such patients also have general functional and pain impairment about one SD worse than the general population. We believe our findings make sense, as only those who continue to have substantial functional and pain limitations will progress to operative management.

At immediate postoperative follow-up (i.e., average of about 2 weeks following surgery), our findings suggest that patients have greater functional limitations than they do preoperatively. We believe these results are to be expected, as patients treated operatively are typically immobilized for 2 to 4 weeks. Further, while PROMIS PI scores worsened (i.e., increased), it did not do so to an appreciable level according to MCID estimates. Prior research in the UE literature suggests that PROMIS PI can be considered as a proxy for ability to cope with pain. ¹⁶ Thus, the lack of clinically measurable worsening of pain may be secondary to patient expectations that surgery will lead to pain postoperatively; therefore, they are able to better prepare themselves for it. We believe that hand surgeons can help prepare patients with ulnar impaction syndrome undergoing operative management for their expected early postoperative recovery by recognizing that function may be more limited than it was preoperatively (perhaps secondary to casting) and that it may also be painful, though not substantially worse than preoperatively, if at all.

At short-term postoperative follow-up (i.e., average of over 6 months following surgery), we found that patients appear to demonstrate improved functional status and pain levels. These findings are positive markers that, on average, patients undergoing operative management for ulnar impaction syndrome do improve from operative intervention at an average of 6 months postoperatively. Prior research has also demonstrated the clinical outcomes benefits, as measured by legacy PROMs, of ulnar shortening. For example, in a study of 28 patients undergoing ulnar shortening for ulnar impaction syndrome, the mean DASH score improved from 40 preoperatively to 26, ¹⁷ an improvement greater than the MCID estimate for DASH (i.e., 10). ¹⁸ However, follow-up evaluation in this study occurred at an average of 29 months postoperatively. ¹⁷ Our work suggests that patients can expect an appreciable clinical improvement in symptoms at an average of 6 months. Based on our findings using PROMIS in conjunction with a prior study utilizing DASH, ¹⁷ we believe that patients can be informed that they can expect to demonstrate improved symptoms by about 3 to 6 months postoperatively, and that this improvement tends to last long-term. By sharing this information with patients preoperatively, patients can better plan for the most appropriate time to pursue operative treatment and understand what to expect following surgery. Further, these findings can help hand surgeons better understand when patients are deviating from the expected postoperative recovery pattern.

This study has limitations. First, our sample comes from a single academic medical center. Thus, the generalizability of our findings is unknown. However, we believe our findings mirror the anecdotal evidence shared among hand surgeons. Second, we have only evaluated the results of PROMs from patients over time. We did not evaluate time to radiographic healing, wrist motion, or grip strength, and these can also affect recovery trajectory. Additionally, we do not have return to work or activity data. However, we believe that PROMs provide insight into elements of recovery that do matter a great deal to patients. Indeed, it is the information from PROMs that we believe is most valuable in improving shared clinical decision-making discussions and setting patient expectations preoperatively. Third, our small sample size limits our ability to conduct more advanced statistics. Further, our sample size is consistent with previously published literature on the topic ^{3 4} or larger. ² Fourth, although all patients had ulnar shortening, the surgical techniques used were not exactly the same. While some ulnar shortening procedures were in the diaphyseal region, others involved the distal metaphyseal region. Additionally, most, but not all, surgeries had simultaneous wrist arthroscopy. More research on the topic with larger sample sizes across multiple centers would beneficial. Fifth, while all patients in our sample had a complete set of preoperative PROMIS questionnaires, not all patients had complete sets of PROMIS questionnaires completed at both follow-up time points. Indeed, only 26 of the 38 patients, for example, had complete follow-up PROMIS scores at over 6 months postoperatively. If anything, we believe that this would suggest our findings underestimate the improvement patients appreciate at short-term follow-up because it is likely that patients doing well would be the ones to not return for clinic visits and complete the PROMIS questionnaires. Lastly, to our knowledge, there is no anchor-based MCID estimate for each of the studied PROMIS domains in patients undergoing operative treatment for ulnar impaction syndrome. In lieu of this, we utilized anchor-based PROMIS UE and PF MCID estimates from a general hand population ¹⁵ in addition to MCID estimates determined using the distribution-based method with our current patient sample. We believe these are appropriate MCID estimates but understand that our findings may differ slightly pending the MCID estimates used. Despite these limitations, we believe our work helps to provide previously uncaptured insight into the recovery pattern of this specific patient population using a set of PROMs growing in popularity.

Overall, patients with ulnar impaction syndrome undergoing operative management present with notable functional and pain impairment, worsen functionally in the immediate postoperative period, and ultimately improve functionally and pain-wise by short-term postoperative follow-up. While these findings may be expected, our work helps build upon the literature to date on this patient population by providing more insight into the recovery timeframe. Further, the present study uses PROMIS—a set of general PROMs that are growing in popularity within hand and UE surgery. ^{10 19 20} As the call to increase the use of PROMs in daily clinical practice grows, ²¹ we believe our work helps provide important detail on this unique patient population. However, future work is warranted with a larger sample size to confirm these findings and determine continued patterns of recovery.

Funding Statement

Funding None.