Stable Long-term PROMIS Outcomes After Deep Peroneal Nerve Neurectomy for Midfoot Arthritis: A Longitudinal Study

Abstract

Background:

Midfoot arthritis can cause debilitating dorsal midfoot pain and functional limitation. Deep peroneal nerve (DPN) neurectomy is a minimally invasive surgical option that addresses pain through selective denervation. Although favorable midterm outcomes have been reported, long-term durability remains unclear. The purpose of this study was to evaluate changes in patient-reported physical function and pain interference between midterm and long-term follow-up after DPN neurectomy using validated Patient-Reported Outcomes Measurement Information System (PROMIS) instruments.

Methods:

A retrospective review identified patients who underwent isolated DPN neurectomy for midfoot arthritis between 2017 and 2021 and completed PROMIS Physical Function (PF) and Pain Interference (PI) questionnaires at midterm and long-term follow-up. Mean follow-up was 2.9 years (midterm) and 6.1 years (long-term). PROMIS scores were compared using paired t tests. Clinically meaningful improvement was defined using established minimal clinically important difference (MCID) thresholds (+5.88 for PF, −5.49 for PI).

Results:

The final cohort included 32 patients (81.2% female, mean age 68.4 ± 7.3 year). PF improved from 37.7 ± 6.2 to 39.0 ± 7.5 (Δ = +1.3, 95% CI [−0.7 to +3.4], P = .188), whereas PI decreased from 62.6 ± 8.8 to 59.2 ± 9.0 (Δ = −3.4, 95% CI [−6.3 to −0.5], P = .020). Clinically meaningful improvement was achieved by 7 of 32 (21.9%) for PF and 15 of 32 (46.9%) for PI. Five patients experienced clinically meaningful worsening, including patients with stroke, chronic pain syndrome, or contralateral foot and ankle pathology or surgery.

Conclusion:

DPN neurectomy was associated with stable longitudinal PROMIS outcomes through extended follow-up, with a nonsignificant mean PF change of +1.3 points and a statistically significant but sub-MCID mean PI decrease of −3.4 points over approximately 6 years. Outcomes were heterogeneous; fewer than half of patients achieved clinically meaningful improvement in pain interference, and a subset experienced worsening. These findings suggest that DPN neurectomy may provide durable symptom stability in appropriately selected patients with refractory dorsal midfoot pain, although clinically meaningful benefit at the individual level is not universal.

Level of Evidence:

Level III, comparative series.

Introduction

Midfoot osteoarthritis is a common cause of pain and activity limitation in older adults, affecting up to 12% of individuals over the age of 50 years.^1-3 Although many patients find relief with nonsurgical management, those with persistent, debilitating symptoms may require operative treatment. Midfoot arthrodesis remains the traditional surgical approach for advanced disease but carries notable morbidity and results in permanent loss of joint motion.^4-8

Deep peroneal nerve (DPN) neurectomy has emerged as a minimally invasive alternative for selected patients with symptomatic midfoot arthritis, particularly those seeking to avoid the prolonged recovery, non-weight-bearing requirements, and potential complications associated with midfoot arthrodesis.^9-12 Early work by Blacklidge et al ¹³ demonstrated promising preliminary outcomes following DPN neurectomy for painful midtarsal and tarsometatarsal arthritis, supporting the potential role of selective denervation for symptom management in appropriately selected patients. Prior work from our institution demonstrated high satisfaction and meaningful pain relief at 2-3 years following DPN neurectomy. ¹⁰ Kindred et al ¹¹ also demonstrated promising outcomes with patient satisfaction rates. However, durability of symptom relief remains unclear.

This study aimed to assess the long-term durability of DPN neurectomy using validated Patient-Reported Outcomes Measurement Information System (PROMIS) Physical Function (PF) and Pain Interference (PI) scores. We hypothesized that pain relief would be sustained and functional outcomes stable at long-term follow-up.

Methods

Following institutional review board approval, a retrospective review of patients who underwent isolated deep peroneal nerve (DPN) neurectomy for dorsal midfoot pain secondary to midfoot arthritis between January 2017 and December 2021 at a single tertiary care academic center was performed. All DPN neurectomy procedures were performed at a single institution by a single surgeon (G.G.S.).

A total of 87 patients underwent isolated DPN neurectomy during the study period. Twenty-six patients were excluded because PROMIS PF and PI questionnaires were unavailable. Of the remaining 61 patients, 25 were excluded because they did not complete both midterm PROMIS assessments (minimum 2-year follow-up) and long-term PROMIS assessments (minimum 4.5-year follow-up). Four additional patients who ultimately underwent subsequent midfoot arthrodesis during follow-up were excluded from longitudinal PROMIS analysis because postoperative PROMIS scores following fusion would reflect the combined effects of both arthrodesis and DPN neurectomy rather than isolated neurectomy outcomes. The final analytic cohort consisted of 32 patients.

Operative indications for DPN neurectomy, consistent with the technique described by Iturregui et al, ¹⁰ included persistent dorsal midfoot pain localized to the deep peroneal nerve distribution in the setting of clinically and radiographically confirmed symptomatic midfoot arthritis and/or deep peroneal nerve entrapment at the anterior tarsal tunnel. All patients had failed nonoperative management, including activity modification, orthotics, anti-inflammatory medications, and shoe modifications.^9,10 In the majority of patients, a preoperative diagnostic DPN block resulting in greater than 75% relief of dorsal midfoot pain was performed prior to surgical intervention; blocks were utilized at the discretion of the treating surgeon.

Because this was a retrospective study using longitudinal postoperative follow-up data, preoperative PROMIS scores were not consistently available and therefore were not included as part of the study inclusion criteria.

Patient-Reported Outcome Measures

Patient-reported outcomes were assessed using the PROMIS Physical Function (PF) and Pain Interference (PI) instruments, which have been validated in foot and ankle populations.^14,15 PROMIS scores were collected postoperatively at 2 intervals: midterm (>2 years) and long-term (>4.5 years) following DPN neurectomy.

Each PROMIS domain is scored on a standardized T-score metric (mean 50, SD 10). For PROMIS PF, higher scores reflect better physical function, whereas for PROMIS PI, higher scores indicate greater pain-related interference with daily activities. Thus, improvement is reflected by an increase in PF and a decrease in PI scores.

Clinically meaningful improvement was defined using previously published minimal clinically important difference (MCID) thresholds in foot and ankle patients, specifically for midfoot arthritis: +5.88 points for PF and −5.49 points for PI, based on Zona et al. ¹⁶

Operative Technique

The senior author performs the DPN neurectomy according to the technique described by Iturregui et al. ¹⁰

Statistical Analysis

Descriptive statistics were calculated for all demographic variables and PROMIS scores. Paired t tests were used to compare midterm and long-term PROMIS PF and PI scores within individuals. Statistical significance was defined as P < .05. Mean paired differences are reported with corresponding 95% CIs.

A priori power analysis was performed using MCID and SD values reported by Zona et al. For PROMIS PF, an MCID of 5.88 and SD of 9.06 produced an effect size (Cohen d = 0.65), requiring 21 subjects to achieve 80% power with α = .05. For PROMIS PI, an MCID of 5.49 and SD of 8.82 yielded an effect size of 0.62, requiring 23 subjects. ¹⁶ The final cohort of 32 patients exceeded these thresholds, indicating sufficient power to detect clinically meaningful changes in PROMIS scores.

Because PROMIS instruments measure global pain and functional domains, identifiable medical or orthopaedic events may influence scores independent of the operative extremity. Accordingly, an exploratory post hoc sensitivity analysis was performed to evaluate the potential influence of identifiable clinical events potentially influencing longitudinal PROMIS outcomes. Patients who demonstrated clinically meaningful worsening in PROMIS scores and also experienced identifiable medical or orthopaedic events during follow-up were excluded from this exploratory analysis (n = 5). Revision procedures directly related to the index neurectomy were retained, as these represent procedure-related outcomes rather than identifiable confounding events. The same paired t test framework was subsequently applied to this exploratory cohort.

All analyses were performed using Python (v3.11) with the pandas, numpy, and scipy libraries. Supplemental power analyses were conducted in R (v4.4.1).

Results

A total of 32 patients met inclusion criteria, with a mean age of 68.4 ± 7.3 years and average BMI of 29.6 ± 6.6. The majority were female (81.2%). Surgical laterality included 17 right-sided (53.1%), 11 left-sided (34.4%), and 4 bilateral (12.5%) procedures. The average time from surgery to midterm PROMIS assessment was 2.92 years, and 6.05 years for long-term follow-up (Table 1).

Table 1.

Demographic Data (N = 32).

Characteristic	Value
Age, y, mean (SD)	68.4 (7.3)
BMI, mean (SD)	29.6 (6.6)
Sex: female, n (%)	26 (81.2)
Laterality, n (%)
Right	17 (53.1)
Left	11 (34.4)
Bilateral	4 (12.5)
Midterm follow-up, y, mean	2.92
Long-term follow-up, y, mean	6.05

Abbreviation: BMI, body mass index.

PROMIS Outcomes

Mean PROMIS Physical Function (PF) scores improved from 37.7 ± 6.2 at midterm to 39.0 ± 7.5 at long-term (Δ = +1.3, 95% CI [−0.7 to +3.4], P = .188). PROMIS Pain Interference (PI) scores decreased from 62.6 ± 8.8 to 59.2 ± 9.0 (Δ = −3.4, 95% CI [−6.3 to −0.5], P = .020) (Figure 1, Table 2).

Figure 1.

Midterm and long-term PROMIS scores following DPN neurectomy. DPN, deep peroneal nerve; PF, Physical Function; PI, Pain Interference; PROMIS, Patient-Reported Outcomes Measurement Information System.

Table 2.

Longitudinal PROMIS Outcomes.

PROMIS Domain	Midterm Mean ± SD	Long-term Mean ± SD	Mean Change (Δ) a	95% CI	P Value b	Patients Achieving MCID, n/N (%)
Physical Function	37.7 ± 6.2	39.0 ± 7.5	+1.3	−0.7 to +3.4	.188	7/32 (21.9)
Pain Interference	62.6 ± 8.8	59.2 ± 9.0	−3.4	−6.3 to −0.5	.020	15/32 (46.9)

Abbreviations: PROMIS, Patient-Reported Outcomes Measurement Information System; Preop, preoperative; MCID, minimal clinically important difference

^aMean change (Δ) presented as long-term mean minus midterm mean.

^bSignificance was defined as P value less than 0.05.

Using MCID thresholds of +5.88 for PF and −5.49 for PI, clinically meaningful improvement was achieved by 7 patients (21.9%) for PF and 15 patients (46.9%) for PI. Clinically meaningful worsening occurred in 5 patients (15.6%), including 1 patient (3.1%) in PF alone, 2 patients (6.3%) in PI alone, and 2 patients (6.3%) in both domains.

Exploratory Sensitivity Analysis

Five patients (IDs 17, 24, 29, 30, and 31) met criteria for exclusion from the exploratory sensitivity analysis due to clinically meaningful worsening in PROMIS scores in the setting of identifiable medical or orthopaedic events potentially influencing longitudinal PROMIS outcomes. These events included ischemic stroke (n = 1), severe lateral ankle instability requiring Broström repair (n = 1), chronic pain syndrome (n = 1), and progression of contralateral foot or ankle arthritis or instability (n = 2).

Importantly, all primary analyses and study conclusions were based on the full cohort, and this exploratory sensitivity analysis was performed only to evaluate potential contributors to variability in longitudinal PROMIS outcomes.

In the remaining cohort (n = 27), PROMIS Physical Function improved from 37.0 ± 6.3 to 39.3 ± 7.4 (Δ = +2.30, 95% CI [ +0.5 to +4.1], P = .015), whereas PROMIS Pain Interference improved from 63.1 ± 9.3 to 57.1 ± 9.2 (Δ = −6.07, 95% CI [−8.8 to −3.3], P < .001). The observed change in PI exceeded the previously reported MCID threshold described by Zona et al ¹⁶ ; however, interpretation of MCID achievement is limited by the absence of preoperative PROMIS measurements and the potential influence of identifiable medical or psychosocial factors on longitudinal PROMIS scores. These exploratory findings suggest that medical or orthopaedic comorbidities may contribute to variability in longitudinal PROMIS outcomes within this medically complex patient population. Revision cases were intentionally retained in all analyses, as they represent true procedure-related outcomes rather than confounding events.

These exclusion criteria were not prespecified and were applied only within this exploratory post hoc sensitivity analysis. Primary outcome analyses were performed using the full study cohort.

Adverse Events and Revision Procedures

Two patients (6.3%) underwent revision procedures related to the index DPN neurectomy, including neuroma excision and neurolysis; these were retained in all analyses as true treatment failures. One additional patient developed chronic pain syndrome during follow-up. No perioperative infections, wound complications, toe deformities, or intraoperative adverse events were identified during the study period. Clinical events, complications and notable secondary procedures during the study period can be seen in Table 3.

Table 3.

Clinical Events, Complications and Secondary Procedures During Long-Term Follow-up.

Patient ID	PROMIS Worsening	Event Type	Description	Related to Index Procedure?
14	No	Reoperation	Second/third toe osteotomies, neuroma excision, hallux excision	Yes
16	No	Reoperation	Revision DPN neuroma excision and SPN neurolysis	Yes
17	Yes	Contralateral arthritis	Contralateral midfoot arthritis requiring partial excision	No
24	Yes	Contralateral operation	Contralateral ankle instability with Broström repair	No
29	Yes	System event	Stroke affecting lower extremity function	No
30	Yes	New diagnosis	Chronic pain syndrome	No
31	Yes	Contralateral arthritis	Severe contralateral foot arthritis	No

Abbreviations: DPN, deep peroneal nerve; ORIF, open reduction internal fixation; PROMIS, Patient-Reported Outcomes Measurement Information System; SPN, superficial peroneal nerve.

Four patients (11.1%) ultimately underwent subsequent midfoot arthrodesis during the follow-up period, representing clinically important treatment failures. These patients were excluded from longitudinal PROMIS analysis because postoperative PROMIS scores following fusion would reflect the combined effects of both arthrodesis and DPN neurectomy rather than isolated neurectomy outcomes.

Discussion

This study represents the first report of the longest follow-up of PROMIS outcomes following deep peroneal neurectomy (DPN) in this patient population to date. These findings expand upon prior case series evaluating DPN using PROMIS outcome measures by providing longitudinal follow-up beyond 6 years and further characterizing the durability of symptom relief over time. ¹⁷ The availability of paired midterm and long-term PROMIS scores makes this cohort unique and offers new insight into the durability of symptom relief after DPN neurectomy for dorsal midfoot pain.

In the full cohort, PROMIS Pain Interference (PI) scores remained stable, with a mean decrease of −3.4 points that did not reach the MCID threshold of −5.49, between midterm and long-term follow-up, while Physical Function (PF) scores remained generally preserved over time, consistent with the expected physiologic effect of a denervation procedure aimed at reducing pain rather than altering structural mechanics. Importantly, PF did not decline over time, and a subset of patients demonstrated clinically meaningful functional gains, suggesting that pain reduction may permit higher activity levels in appropriately selected individuals. A subset of patients (5/32, 15.6%) demonstrated clinically meaningful worsening in one or both PROMIS domains over time, further emphasizing the heterogeneous nature of patient responses following DPN neurectomy.

Because the present study compared postoperative midterm and long-term PROMIS measurements rather than preoperative vs postoperative outcomes, it was uniquely positioned to evaluate the durability of symptom relief over time following DPN neurectomy. Although preoperative PROMIS scores would provide additional insight into the magnitude of initial postoperative improvement, the current longitudinal design demonstrates that PROMIS PF and PI scores remained generally stable at extended follow-up. These findings suggest that DPN neurectomy may provide sustained symptom relief and functional benefit rather than only short-term improvement in appropriately selected patients.

These results build on the findings by Iturregui et al, ¹⁰ who reported high satisfaction and pain relief at 2-3 years using nonstandardized surveys. These findings build on prior reports describing favorable patient satisfaction and pain relief following DPN neurectomy and suggest that symptom relief may remain stable over extended follow-up in selected patients.

Compared with midfoot arthrodesis, which remains the traditional surgical treatment for advanced midfoot arthritis, DPN neurectomy avoids the morbidity associated with joint fusion.^5-8 Prior studies have reported conversion to arthrodesis in up to 9% of patients at midterm follow-up following denervation procedures.^10,11 In the present study, 4 patients (11.1%) ultimately underwent subsequent midfoot arthrodesis during long-term follow-up. These patients were excluded from longitudinal PROMIS analysis because postoperative PROMIS scores following fusion would reflect the combined effects of arthrodesis and DPN neurectomy rather than isolated neurectomy outcomes. The observed 11% conversion rate at long-term follow up aligns with prior literature and suggests that most patients do not require subsequent fusion.^10,11 Including fusion cases could have introduced heterogeneity and confounded the interpretation of PROMIS scores, which would then reflect outcomes from both procedures. Nevertheless, exclusion of these patients may bias the cohort toward more favorable outcomes and should be considered when interpreting the study findings.

Because PROMIS instruments measure global health domains rather than site-specific symptoms, scores may be influenced by medical, orthopaedic, or psychosocial conditions, particularly over extended follow-up intervals in older or medically complex patients. Accordingly, an exploratory post hoc sensitivity analysis was performed to evaluate the potential influence of identifiable clinical events on longitudinal PROMIS outcomes. In this exploratory cohort, exclusion of patients who experienced clinically meaningful worsening in the setting of identifiable medical or orthopaedic events was associated with larger observed improvements in both PROMIS PI and PF scores. These findings suggest that identifiable comorbidities may contribute to variability in longitudinal PROMIS outcomes within this patient population. Importantly, the primary study conclusions were based on analyses of the full cohort, and the exploratory sensitivity analysis should be interpreted cautiously given its post hoc nature. Revision cases were intentionally retained in all analyses, as they represent true procedure-related outcomes rather than identifiable confounding events.

The medically complex profile of this cohort underscores the importance of low-morbidity surgical options. DPN neurectomy is minimally invasive, avoids joint fusion, and typically allows rapid recovery.^9,10,12 This makes it a particularly valuable option for older or medically frail patients for whom midfoot arthrodesis may carry greater risk and postoperative morbidity.^5-8 Notably, no cases of toe deformity were identified at long-term follow-up. Given theoretical concerns that denervating the deep peroneal nerve may affect intrinsic toe extensors, our findings provide reassuring evidence that DPN neurectomy does not predispose patients to deformity even at more than 6 years postoperatively.

This study has several limitations, including its retrospective design, modest sample size, and lack of preoperative PROMIS scores. As preoperative PROMIS measurements were not available, this limits the interpretation of interval PROMIS changes and prevents direct assessment of perioperative improvement attributable to DPN neurectomy. Future prospective investigations incorporating preoperative PROMIS assessments would allow for more robust evaluation of clinically meaningful improvement following DPN neurectomy.

Additionally, PROMIS instruments assess global pain and functional domains rather than site-specific symptoms and therefore may be influenced by medical, orthopaedic, or psychosocial conditions occurring during long-term follow-up. These factors may either negatively or positively affect PROMIS scores independent of the operative extremity and introduce variability into longitudinal outcome assessment. To better contextualize this limitation, an exploratory post hoc sensitivity analysis was performed to evaluate the potential influence of identifiable events on longitudinal PROMIS outcomes.

An additional limitation relates to exclusion of patients who ultimately underwent conversion to midfoot arthrodesis. These patients likely represent treatment failures, and exclusion of these cases may bias the cohort toward more favorable outcomes. However, postoperative PROMIS values following fusion would reflect the combined effects of DPN neurectomy and arthrodesis rather than isolated outcomes attributable to DPN neurectomy alone. Therefore, these patients were excluded to maintain focus on the durability of isolated DPN neurectomy. Future analyses incorporating pre-conversion PROMIS scores for these patients, where available, would provide a more complete representation of the full operative cohort and should be considered in prospective investigations.

Radiographic data, long-term patient satisfaction, and comparative outcomes relative to primary midfoot arthrodesis were not assessed in the present study and represent important areas for future investigation. A noninferiority design may help determine whether comparable long-term outcomes can be achieved with reduced morbidity. Nevertheless, this study offers one of the most comprehensive PROMIS-based evaluation of long-term outcomes following DPN neurectomy and provides important evidence regarding the durability, safety, and clinical relevance of this procedure. As denervation techniques gain popularity in foot and ankle surgery, longitudinal data such as ours will be increasingly important for patient counseling and shared decision-making

Conclusion

Deep peroneal neurectomy was associated with stable longitudinal PROMIS outcomes through extended follow-up, with a nonsignificant mean Physical Function change of +1.3 points and a statistically significant but sub-MCID mean Pain Interference decrease of −3.4 points over approximately 6 years. Neither reached the predefined MCID thresholds of +5.88 and −5.49, respectively. Patient responses were heterogeneous: fewer than a quarter of patients achieved clinically meaningful improvement in physical function (21.9%) and fewer than half did so for pain interference (46.9%), whereas a subset demonstrated worsening longitudinal outcomes. These findings suggest that DPN neurectomy may provide durable symptom stability in appropriately selected patients with refractory dorsal midfoot pain, while highlighting the importance of careful patient selection, preoperative counseling regarding realistic outcome expectations, and further prospective investigation incorporating preoperative outcome measures and comparative cohorts.