Peripheral Nerve Stimulation Reduces Opioid Use and Facilitates Early Ambulation Following Fibular Flap Harvest: A Retrospective Study

Abstract

Background:

Fibular free flap (FFF) reconstruction is the gold standard for mandibular and maxillary defects, but donor-site pain remains challenging. Multimodal analgesia (MMA) relies on opioids, which risk dependence, systemic side effects, and delayed recovery. Regional nerve blocks reduce pain but have inconsistent efficacy owing to variable drug metabolism, potential effects on flap perfusion, and motor blockade. Temporary percutaneous peripheral nerve stimulation (TPNS) is a promising alternative that selectively modulates sensory pathways while preserving motor function.

Methods:

This retrospective, single-center study evaluated 26 patients who underwent FFF reconstruction between July 2021 and November 2025, divided into 2 cohorts: MMA alone and MMA with peroneal nerve TPNS. The primary outcome was opioid consumption, expressed as oral morphine equivalents. Secondary outcomes were postoperative pain scores, time to independent ambulation, and length of stay.

Results:

The TPNS group had significantly lower pain scores on postoperative days 2–5 (P < 0.05). Cumulative opioid use was clinically but not statistically lower (median oral morphine equivalents 11.25 versus 105 mg; P = 0.1), as was median time to independent ambulation (3.71 versus 4.76 d; P = 0.24).

Conclusions:

TPNS may be a safe, opioid-sparing adjunct that promotes earlier mobilization after FFF harvest. Further study is needed to justify prospective, multicenter trials validating efficacy, cost-effectiveness, and broader applicability in head and neck reconstructive surgery.

Takeaways.

Question: Can temporary percutaneous peripheral nerve stimulation reduce donor-site pain and opioid use after fibula free flap harvest while preserving motor function?

Findings: In this pilot retrospective study, peroneal nerve stimulation reduced postoperative pain by 44% and opioid use by 42%, with significant benefit on postoperative days 2–5.

Meaning: Temporary peroneal nerve stimulation may be an effective opioid-sparing adjunct that improves pain control without impairing ambulation after fibula free flap harvest.

INTRODUCTION

Reconstruction of maxillofacial defects with a fibular free flap (FFF) is a safe and functional solution for maxillofacial pathology, congenital anomalies, or trauma; however, the management of postoperative donor site pain is challenging. Multimodal analgesia (MMA) remains the standard in postoperative pain management,^1,2 but this option commonly employs opioids, putting patients at risk of opioid use disorder and negative systemic side effects, such as postoperative nausea, sedation, and respiratory depression.^3–5 Long-term sequelae reveal variable incidences of sensory and motor dysfunction, with neuropathic pain reported in multicenter cohorts and increased complications when reconstruction requires split-thickness skin grafting versus primary closure.^6–9

Recently, regional blocks using local anesthetics at the fibula flap donor site have gained traction in enhanced recovery after surgery (ERAS) protocols.^10,11 Henry et al and Le et al demonstrated opioid consumption reductions of 28.9% and 16%, respectively.^11,12 However, success has been variable and attributed to differences in patient metabolism, potential effects on flap perfusion (steal phenomenon), systemic toxicity risks, and delayed functional recovery due to motor blockade of the sciatic nerve.^13–15

In 2020, Ilfeld et al described percutaneous peripheral nerve stimulation (TPNS), referencing Melzack and Wall’s “gate control theory,” where neuromodulation of afferent nerve fibers prevents pain signal transmission and Deer’s reconditioning of the central nervous system through peripheral nerve stimulation.^16,17 This US Food & Drug Administration–cleared TPNS (maximum 60 d indwelling) demonstrated 65% pain reduction and 90% opioid reduction over the first 7 postoperative days when placed along the peroneal nerve.¹⁸

In the present study, the investigators applied TPNS as an adjunct to the local ERAS protocol for head and neck reconstruction using FFF. This study investigated the preoperative placement of a TPNS device adjacent to the ipsilateral sciatic nerve above the popliteal fossa to evaluate its effectiveness in limiting opioid consumption and enhancing functional gait restoration.

MATERIALS AND METHODS

This study followed the World Medical Association Declaration of Helsinki and was approved by the institutional review board standards of the Naval Medical Center San Diego (NMCSD), San Diego, CA. The Clinical Investigation Department of NMCSD approved this study under the institutional review board protocol NMCSD.2023.0027. Written informed consent was obtained from all included patients regarding the reconstruction and placement of TPNS, if applicable. The percutaneous TPNS technique leverages ultrasound guidance principles already familiar to anesthesiologists trained in regional anesthesia blocks, with the local Regional Anesthesia Services team successfully placing all devices on the first attempt within an average of 1 hour in an outpatient setting.

Patients and Design

This was a single-site, retrospective study. All consecutive patients who were aged 18 years or older; able to walk (with or without aid); and scheduled for head and neck surgery, specifically reconstruction using an FFF, at NMCSD between July 2021 and November 2025 were reviewed for inclusion. The exclusion criteria were cognitive impairment preventing informed consent and patients who did not receive an FFF. It should be noted that patients with preoperative lower extremity vascular anomalies or motor and/or sensory dysfunction were excluded from fibular graft harvesting by the head and neck reconstruction team; hence, such impairments were not included in the exclusion criteria.

In this retrospective study, 26 adult patients who underwent FFF reconstruction of the head and neck were reviewed, and the first cohort (no-intervention group) was compared with the intervention group, who received preoperative placement of a single 100-Hz microlead TPNS device (SPR, Cleveland, OH). The device was used in conjunction with the NMCSD’s established head and neck ERAS protocol with specific modifications for FFF, where the TPNS device was used. (Table 1). The TPNS device was initially implemented as a quality improvement project for postoperative pain control, and as a result, the patients were automatically enrolled in the improved protocol without enrollment bias.

Table 1.

Summary of NMCSD Head and Neck ERAS Protocol for No Intervention Versus Intervention Cohort

No intervention cohort ERAS for pain management

Intervention cohort ERAS for pain management

MMA - Acetaminophen 650 mg q6h - Ibuprofen 600 mg q6h - Gabapentin 300 mg TID - Oxycodone 5 mg q6h PRN for pain >8/10

Regional anesthesia (FFF) - Preoperative placement of TPNS device with ultrasound guidance to sciatic nerve - Baseline anesthetic stimulation intensity level is recorded preoperatively - TPNS device is used in conjunction with MMA postoperatively based on preoperative impedance level and titrated by the patient as needed Multimodal analgesia - Acetaminophen 650 mg q6h - Ibuprofen 600 mg q6h - Gabapentin 300 mg TID - Oxycodone IR 5 mg q6h PRN for pain >8/10

PRN, pro re nata (as needed); TID, ter in die.

The primary outcome was the total opioid usage converted to oral morphine equivalent (OME).^2,19 Secondary outcomes included the number of days until independent ambulation without an assistive device (AD) and subjective pain scores (0–10) based on the visual analog Defense and Veterans Pain Rating Scale. Quantitative data from inpatient physical therapy sessions were used to assess gait quality.

Outcomes

The primary outcome was the total opioid usage converted to OME. The secondary outcomes were (1) postoperative pain based on subjective pain scores (0–10) according to the visual analog Defense and Veterans Pain Rating Scale, (2) total length of stay (LOS) of hospital admission, and (3) functional recovery.

Functional recovery, defined as the number of days until patients achieved independent ambulation without an AD, was determined through a structured review of quantitative physical therapy documentation. “Independent ambulation” was operationalized as ambulation without use of an AD (including front-wheel walker, rolling walker, or cane), as documented in physical therapy progress notes, reflecting the ability to ambulate safely without external support. Distance ambulated and specific functional milestones (eg, distance from the patient room to the nursing station, stair ambulation) were recorded when available in physical therapy documentation, but the primary outcome marker was achievement of any distance of ambulation without an AD. This definition reflects a clinically meaningful functional threshold for patient safety and discharge planning during the postoperative period.

Statistical Analysis

The sample size calculation was performed based on the expected two-point reduction in extremity Defense and Veterans Pain Rating Scale. Due to the lack of similar previous studies at the time of analysis, the assumed reduction of average pain level with TPNS was based on the reported average pain level of the no-intervention cohort that had undergone head and neck FFF surgery at NMCSD and the observed pain reduction scores reported by Ilfeld et al.¹⁸ Using Sigma Plot 12.3 software (Systat Software, San Jose) for nonparametric tests, 26 patients (12 in the control group and 14 in the intervention group) were required to detect a two-sided significance with 80% power and an alpha of 0.05 (subsequently, the limits of statistical significance were used in all analyses).

Unless otherwise stated, continuous values are presented as median and interquartile range (IQR), and counts with proportions as percentages (%). To test the normality of continuous variables, we used the Shapiro test, which indicated nonnormal distributions; thus, we reported the median (IQR) and compared groups using Mann-Whitney U tests. For our primary analysis, we considered data until the eighth postoperative day. We did not impute missing outcome data for our primary intent-to-treat analysis. We summarized the available sample sizes for each outcome at each time-point.

Conversion of opioid medications to OME followed standardized conversion factors derived from published guidelines¹⁹: oxycodone 1 mg = 1.5 mg OME, oral hydromorphone (Dilaudid) 1 mg = 5 mg OME, and tramadol 1 mg = 0.2 mg OME. Total OME was calculated as the cumulative daily dose summed over the 8-postoperative-day assessment window. To account for potential confounding by hospital LOS duration, sensitivity analyses explored whether differences in LOS between groups influenced OME comparisons; however, OME normalization to patient weight or specific LOS parameters was not performed in the primary analysis. Future investigations should consider weight-normalized dosing to account for interindividual pharmacokinetic variability.

To test for group differences in pain rating on each postoperative day, we used Mann-Whitney U tests to compare the average opioid consumption between the treatment and control groups. To assess whether there were differences in pain ratings over time, we fitted linear mixed-effects models with random intercepts to test the main effects of group, postoperative day, and a group-by-postoperative day interaction.

To examine differences in days to independent ambulation between the treatment and control groups, we used the Mann-Whitney U test. Data trimming was not performed. All authors approved the statistical analysis plan before beginning the analyses. All results were analyzed on an intention-to-treat basis.

RESULTS

Data were collected from July 2021 through November 2025, and all patients who had undergone reconstructive head and neck surgery using microvascular FFF were reviewed. A total of 26 patients were included in this retrospective study (Tables 2 and 3). The overall median age was 38 years (range, 18–76 y); 18 patients were male, and 8 were female. Preoperative baseline patient characteristics for both cohorts are shown in Tables 2 and 3. In the intervention cohort, the regional anesthesia team successfully placed the TPNS lead under moderate sedation within an average time of 1 hour in an outpatient setting 1 day before the FFF reconstruction date. None of the intervention group members experienced any postprocedural complications associated with TPNS placement.

Table 2.

Summary of FFF Patients Without Regional Anesthesia (No Intervention Cohort)

Sex	Age, y	Diagnosis	LOS
F	22	Ameloblastoma	7
M	23	Oral squamous cell carcinoma	10
F	26	Ameloblastoma	7
M	45	Maxillary fibrous dysplasia	7
M	21	Central giant cell granuloma	5
F	24	Trauma	7
M	66	Oral squamous cell carcinoma	9
M	32	Odontogenic keratocyst	7
M	28	Gunshot wound	8
M	26	Ameloblastoma	9
M	44	Odontogenic keratocyst	7
M	32	Ameloblastoma	7

F, female; M, male.

Table 3.

Summary of FFF Patients with Regional Anesthesia (Intervention Cohort)

Sex	Age, y	Diagnosis	LOS
F	65	Myxoma	13*
M	76	Oral squamous cell carcinoma	21*
M	60	Osteomyelitis	7
F	18	Ameloblastoma	7
M	27	Gunshot wound	7
F	44	Ameloblastoma	7
M	67	Parotid basal cell carcinoma	8
M	43	Osteomyelitis	4
F	30	Ameloblastoma	8
F	38	Osteomyelitis	6
M	22	Trauma	7
M	23	Central giant cell granuloma	4
M	76	Osteoradionecrosis of jaw	7
M	18	Ameloblastoma	6

^*LOS affected as a result of FFF postoperative complications, but not related to the TPNS device.

F, female; M, male.

According to our protocols, the patients were transferred postoperatively to the intensive care unit under anesthesia. As such, assessments of immediate outcomes commenced on the extubation date and ended at the discharge date.

Primary Outcome

Opioid Consumption

Figure 1 and Table 4 depict the total opioid dose by the control and intervention groups across the 8 postoperative days in OME. Although the differences were not statistically significant for other days, the intervention group generally had a lower total opioid dose during the first four postoperative days. We further assessed whether there were differences in the cumulative opioid doses. Although the differences were not significant by group, the intervention group had a lower cumulative opioid dose than the control group [mean (IQR) = 91.25 (75.58) mg versus 39.86 (62.47) mg; P = 0.20; Table 5].

Fig. 1.

Effects of peripheral nerve stimulation on opioid consumption.

Table 4.

Description of Opioid Dose (mg) by Postoperative Day

Postoperative Day	Control, n	Intervention, n	Control, Mean (SD)	Intervention, Mean (SD)	SMD	Control, Median (IQR)	Intervention, Median (IQR)	P
1	8	7	23.75 (13.36)	16.50 (7.99)	0.20	18.75 (17.50)	16.00 (8.50)	0.48
2	10	5	32.25 (23.99)	16.50 (9.78)	0.22	30.00 (20.62)	15.00 (15.00)	0.15
3	7	3	30.36 (9.73)	20.00 (18.87)	0.40	30.00 (8.75)	22.50 (18.75)	0.49
4	8	5	16.25 (10.69)	15.00 (12.99)	0.04	11.25 (16.88)	7.50 (7.50)	0.87
5	7	3	12.86 (9.40)	27.50 (4.33)	-0.63	7.50 (7.50)	30.00 (3.75)	0.06
6	6	4	15.00 (11.62)	24.38 (11.25)	-0.35	11.25 (7.50)	22.50 (16.88)	0.18
7	5	3	12.00 (6.71)	12.50 (4.33)	-0.04	7.50 (7.50)	15.00 (3.75)	0.87
8	0	1	—	7.50 (NA)	—	—	7.50 (0.00)	—

SMD, standardized mean difference.

Table 5.

Description of Opioid Dose (OME) Sum Over 8 Postoperative Days

Control, Mean (SD)	Intervention, Mean (SD)	SMD	Control, Median (IQR)	Intervention, Median (IQR)	P
91.25 (75.58)	39.86 (62.47)	0.20	105.00 (133.75)	11.25 (47.12)	0.10

SMD, standardized mean difference.

Secondary Outcomes

Postoperative Pain

Figure 2 and Table 6 depict the reported average pain levels in the control and intervention groups across the eight postoperative days using the Defense and Veterans Pain Rating Scale. The pain-level data points for that specific postoperative day were preferentially screened for lower extremity-specific pain levels, if specified. The reported pain levels on postoperative days 2 through 5 and day 8 were significantly lower in the intervention group than in the control group (P < 0.03; Table 6). This persisted on days 1, 6, and 7, although the difference was not statistically significant (P > 0.08). This is consistent with linear mixed-effects models, which suggest that the intervention group relative to controls had a faster decline in reported pain levels over time, although this difference was not statistically significant (β_{group by postoperative day} = −0.13; 95% CI −0.40, 0.14).

Fig. 2.

Effects of peripheral nerve stimulation on pain levels.

Table 6.

Description of Pain Levels by Postoperative Day

Postoperative Day	Control, n	Intervention, n	Control, Mean (SD)	Intervention, Mean (SD)	SMD	Control, Median (IQR)	Intervention, Median (IQR)	P
1	11	11	3.34 (2.50)	2.74 (1.97)	0.08	3.33 (3.90)	2.40 (2.86)	0.57
2	12	14	3.78 (2.34)	1.33 (1.57)	0.31	3.80 (3.46)	0.55 (2.62)	0.01
3	12	14	3.91 (2.19)	0.84 (1.10)	0.42	4.67 (2.82)	0.30 (1.50)	0.00
4	12	14	3.53 (1.84)	0.95 (1.12)	0.42	3.50 (2.71)	0.54 (1.58)	0.00
5	12	13	3.45 (2.12)	0.71 (0.89)	0.39	3.35 (2.58)	0.33 (1.20)	0.00
6	9	12	2.64 (2.25)	1.45 (1.66)	0.18	2.75 (3.00)	0.58 (3.00)	0.23
7	10	9	3.21 (2.76)	1.22 (2.12)	0.24	3.25 (5.05)	0.00 (1.00)	0.08
8	4	4	2.58 (2.36)	0.00 (0.00)	0.63	1.67 (2.25)	0.00 (0.00)	0.02

SMD, standardized mean difference.

Hospital LOS Following FFF Reconstruction

As demonstrated in Tables 2 and 3, the overall mean LOS in the control and intervention groups was 7.5 versus 8 days, respectively. Two patients in the intervention cohort experienced extended hospital stays attributed to postoperative complications unrelated to the TPNS device: 1 patient (age 65 y, myxoma diagnosis) experienced FFF failure requiring graft removal and debridement (LOS = 13 d), and 1 patient (age 76 y, oral squamous cell carcinoma) developed temporary hypoxia with encephalopathy secondary to central sleep apnea triggered by olanzapine administration for delirium management (LOS = 21 d). When excluding these complications-driven outliers, the intervention group mean LOS decreased from 8 to 6.5 days, representing a 1-day reduction compared to controls (7.5 d).

Days to Independent Ambulation

Figure 3 depicts the days to independent ambulation in the control and intervention groups. Although not statistically significant, the intervention group had lower median days to ambulation than the control group [median (IQR) = 3.71 (2.50) versus 4.76 (3.50) d; P = 0.24]. Notably, one patient in the TPNS cohort achieved independent ambulation as early as postoperative day 1, whereas 1 patient progressed to independent stair climbing by postoperative day 4. The TPNS cohort generally demonstrated earlier achievement of gait independence and reduced reliance on AD at discharge compared to controls, and this functional distinction parallels the observed reduction in pain and suggests qualitative improvements in gait confidence and motor capacity.

Fig. 3.

Days to ambulation by group.

DISCUSSION

This retrospective cohort study evaluated TPNS of the peroneal nerve for postoperative donor site pain management following FFF reconstruction. Although TPNS has demonstrated efficacy in other acute postoperative contexts, its application in head and neck reconstructive surgery represents a novel approach warranting systematic evaluation.

Clinical Significance and Mechanisms

The clinical significance of optimizing fibular donor-site pain management is underscored by a substantial body of literature documenting long-term neuropathic sequelae. A recent multicenter analysis identified neuropathic pain as a persistent problem following fibula flap harvest,^6,9 with an increased incidence when donor site reconstruction requires split-thickness skin grafting versus primary closure.⁷ These complications emphasize the importance of acute pain management strategies that preserve early mobilization through selective sensory neuromodulation without compromising motor function. By preserving early mobilization, TPNS may reduce acute pain-driven functional limitations and potentially influence longer-term neuropathic outcome trajectories—hypotheses warranting prospective investigations.

The primary outcome demonstrated that opioid consumption was consistently lower in the TPNS cohort during the first four postoperative days, showing clinically significant reductions in cumulative OME and pain levels by 42% and 44%, respectively. The demonstrated 89% reduction in median OME sum (11.25 versus 105 mg) represents a clinically meaningful opioid-sparing effect aligned with the ERAS principles. The secondary outcomes showed statistically significant pain reductions on postoperative days 2–5, the peak of postoperative inflammation.

The mechanisms underlying TPNS-mediated analgesia extend beyond the opioid-sparing effects. Gate control theory posits that selective stimulation of large-diameter sensory A-beta fibers gates pain transmission by competing with nociceptive C-fiber signaling, reducing pain perception without affecting motor function. The critical advantage of TPNS over regional anesthetic blocks lies in selective sensory neuromodulation—TPNS preserves ankle dorsiflexion and plantarflexion essential for ambulation without motor blockade. Consequently, patients report lower pain levels, express greater confidence in mobilization, and demonstrate increased participation in physical therapy. The observed trend toward earlier ambulation (3.71 versus 4.76 d; P = 0.24) supports this mechanistic framework, as patients in the TPNS cohort were less reliant on AD at discharge.

Interaction Between Donor- and Recipient-site Pain

An underexplored aspect is the interaction between donor- and recipient-site pain. Patients with significant recipient-site pathology (eg, osteomyelitis requiring extensive resection) demonstrated higher overall postoperative pain scores and greater opioid consumption despite effective donor site analgesia, suggesting that recipient-site pain may overwhelm donor site analgesic benefits. Conversely, effective donor-site pain control may reduce the overall pain burden and improve the willingness of patients to participate in rehabilitation. Future prospective studies with anatomically specific pain assessments are needed to elucidate these relationships and inform patient selection.

Broader Applicability and Future Directions

Although TPNS could theoretically be extended to other free flap procedures based on gate control theory mechanisms, soft tissue-only free flaps without significant osseous harvest may not derive equivalent benefits. Future prospective studies should investigate TPNS application across diverse free flap types. Additionally, integration of intraoperative neurophysiological monitoring with preoperative TPNS placement merits investigation, as baseline intraoperative neurophysiological monitoring data could inform TPNS lead positioning and provide objective poststimulation nerve function assessment.

An important consideration is the potential application of TPNS in recipient-site analgesia. In head and neck reconstruction involving the maxilla or mandible, surgical neurectomy often necessitates same-day nerve reconstruction, resulting in immediate postoperative recipient-site anesthesia that limits traditional sensory-based pain management. However, in non–head and neck reconstructions or procedures with delayed nerve reconstruction, preoperative or intraoperative TPNS placement at the recipient-site nerves may represent a complementary analgesic strategy, warranting future investigation.

Limitations

The retrospective design and small sample size (n = 12–14 per group) significantly limited the statistical power and causal inference. Prospective, multicenter, randomized controlled trials with adequate sample size calculations (estimated n = 50–100 per group) are essential to validate these findings. Critical confounders that were not systematically controlled included patient age, comorbidities, body weight, donor-site closure method, baseline neurological status, and preoperative chronic pain or opioid use history. Pain scores were reviewed to identify donor site–specific pain; however, not all postoperative datasets included anatomical distinction, resulting in potential confounding by recipient site pain. Additionally, opioid dosing was not normalized to individual patient weight or adjusted for hospital LOS variation.

Although TPNS is US Food & Drug Administration–cleared for acute postoperative pain, its widespread use for routine acute postsurgical pain remains limited by the cost relative to typical postoperative analgesic techniques and lack of widespread insurance reimbursement. However, potential reduction of healthcare resource utilization, not limited to LOS but also including follow-up visits or readmission related to pain or additional pain treatment, could theoretically offset device costs through reduced hospital admission expenses.

CONCLUSIONS

This study represents the first evaluation of TPNS for pain management at FFF donor sites. The findings demonstrate clinically meaningful reductions in postoperative pain (44%) and opioid consumption (42%), with a statistically significant pain reduction on postoperative days 2–5. Patients who received TPNS achieved earlier functional gait independence, reflecting preserved motor function and reduced pain-related limitations.

Selective sensory neuromodulation achieved through TPNS addresses the critical limitations of regional anesthetic approaches, including inconsistent efficacy, potential vascular compromise, and motor blockade. As the medical community intensifies its focus on perioperative opioid reduction, alternative analgesic strategies such as TPNS represent valuable adjuncts to multimodal protocols.

Future investigations should include prospective, multicenter, randomized controlled trials with adequate sample sizes, stratification by relevant variables (reconstruction type, closure method, and patient characteristics), detailed patient-reported outcomes distinguishing donor versus recipient site pain, cost-effectiveness analyses, and assessment of long-term neuropathic outcomes. In summary, TPNS shows promise as a safe, effective, and potentially cost-neutral analgesic strategy for fibula free-flap reconstruction, warranting rigorous prospective validation.

DISCLOSURES

Dr. Scherschel received consultation and travel fee from Pacira Biosciences. Dr Scherschel is a military service member or employee of the US Government. This work was prepared as part of our official duties. Title 17, USC §105 provides that copyright protection under this title is not available for any work of the US Government. Title 17, USC§101 defines a US Government work as work prepared by a military service member or employee of the US Government as part of that person’s official duties. Report No. 25-80 was supported by the Defense Health Agency and Naval Health Research Center Research Hub under work unit no. 62436. The views expressed in this work are those of the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense nor the US Government. The authors have no financial interest to declare in relation to the content of this article.

ETHICAL APPROVAL

The study protocol was approved by the NMCSD institutional review board in compliance with all applicable federal regulations governing the protection of human subjects. Research data were derived from approved NMCSD institutional review board protocol number NMCSD.2023.0027.