Long-term outcomes of selective tibial neurotomy for spastic foot deformity: Initial clinical experience in Vietnam Authors: Nguyen Van Tuan, Le Dien Son, Le Nguyen Duy Khuong, Nguyen Van My Anh, Chu Tan Si, Tran Hoang Ngoc Anh, Françoise Lapierre Published: 2026 ABSTRACT IntroductionSpastic foot deformity after central nervous system lesions is clinically heterogeneous and often combines equinus, varus, claw toes, pain, and gait limitation. Selective tibial neurotomy (STN) is a peripheral neurosurgical option for focal lower-limb spasticity, but data from low- and middle-income settings remain limited.Research questionWhat are the long-term clinical and functional outcomes of STN for spastic foot deformity, and are these outcomes influenced by the extent of motor fascicle resection?Material and methodsWe conducted a retrospective study of 31 patients (36 feet) who underwent STN. Deformity, clonus, ankle dorsiflexion, pain, gait performance, orthotic comfort, and patient satisfaction were assessed preoperatively and at long-term follow-up. The relationship between fascicle resection extent and outcomes was analysed.ResultsSignificant improvements were observed in equinus, varus, and claw-toe deformities, ankle dorsiflexion, and clonus (all p < 0.001). Pain resolved in all symptomatic patients, and toe-tip skin lesions markedly decreased. Walking distance increased significantly, and 10-m walking time improved. Orthotic comfort and patient satisfaction were high. Resection of more than two-thirds of fascicles was associated with better clonus and varus control.Discussion and conclusionSTN provides sustained clinical and functional benefits for spastic foot deformity. Greater fascicle resection may enhance outcomes in selected patients. This technique is feasible and effective in a Vietnamese neurosurgical setting. Introduction Spastic foot deformity is a frequent and disabling consequence of lesions affecting the central nervous system. Beyond isolated equinus, patients commonly present with mixed patterns that include varus, claw toes, pain, toe-tip skin lesions, genu recurvatum, difficulty wearing shoes or orthoses, and impaired gait efficiency (Verdie et al., 2004; Bollens et al., 2011). The clinical problem is therefore broader than equinus alone and requires a treatment strategy that distinguishes dynamic spasticity from fixed soft-tissue contracture. Selective tibial neurotomy was first described by Stoffel in 1912 and later refined by Sindou, Mertens and Decq as a peripheral neurosurgical treatment for focal lower-limb spasticity (Stoffel, 1912; Sindou and Mertens, 1988; Decq et al., 2000; Decq, 2003). Long-term case series and later comparative studies have shown improvement in deformity, passive ankle dorsiflexion, pain, standing balance and gait, and randomized data suggest that the effect may be more durable than repeated botulinum toxin injections in appropriately selected patients (Buffenoir et al., 2004, 2013; Rousseaux et al., 2008; Bollens et al., 2013). Despite these results, the procedure has rarely been reported from low- and middle-income countries. In Vietnam, long-term access to repeated botulinum toxin injections may be limited by cost, whereas microsurgical treatment may offer a single-intervention alternative for selected patients. In addition, the optimal extent of motor fascicle resection remains incompletely defined. The aims of the present study were therefore to report the initial Vietnamese clinical experience with STN for spastic foot deformity, to describe the long-term clinical and functional outcomes, and to explore whether the amount of fascicle resection influenced the reduction of clonus and varus deformity. Materials and methods This study was a retrospective review of a consecutive clinical series of patients who underwent STN between 2006 and 2013 in Ho Chi Minh City. The archival thesis dataset describes procedures performed at 115 People's Hospital, Gia Dinh People's Hospital and Children's Hospital 2. Patients were eligible when they had a disabling focal spastic foot deformity after a central nervous system lesion, had completed a structured rehabilitation programme, and had reached a neurological plateau of at least 12 months before surgery. Clinical evaluation was performed preoperatively and at the longest available postoperative follow-up. Equinus, varus and claw-toe deformities were graded on a 3-point ordinal scale (0 = absent, 1 = mild, 2 = marked/symptomatic). Equinus severity incorporated heel contact during stance; varus and claw toes were graded according to the visible deformity and its functional impact. Knee recurvatum was graded on a similar 0–2 scale. Clonus of the triceps surae was recorded with the knee flexed and extended using a simple 0–4 clonus-based stretch-reflex scale adapted from Tardieu/Held-Tardieu methodology (Tardieu et al., 1954): 0 = none, 1 = mild, 2 = clonus that self-terminated, 3 = sustained clonus during fast stretch, and 4 = sustained clonus even during slow stretch. Passive ankle dorsiflexion was measured with the knee flexed and extended, with 0° representing the neutral 90° ankle position and negative values indicating plantarflexion. Functional assessment included standing and walking ease scores (0 = easy, 1 = possible only briefly or with marked difficulty, 2 = impossible), requirement for assistive devices, maximal walking distance, and usual-speed 10-m walking time with shoes/orthoses and barefoot. Orthotic comfort was recorded on a 0–10 numerical rating scale, where 0 represented inability to wear the device and 10 represented maximal comfort. Satisfaction at follow-up was scored from 1 (very dissatisfied) to 10 (very satisfied). All procedures were performed under general anaesthesia with the patient prone. Through an oblique popliteal-fossa incision, the tibial nerve and its motor branches were identified under the microscope. Intraoperative electrical stimulation and anatomical landmarks were used to identify branches to the soleus, gastrocnemius, tibialis posterior and long toe flexors, after which approximately 50–80% of the selected motor fascicles were sectioned according to the preoperative deformity pattern and the intraoperative response. The proximal soleus branch was addressed in every case. When a fixed component of deformity coexisted, associated orthopaedic procedures were added. Toe-flexor lengthening was used when clawing remained clinically important or when the long toe-flexor motor branch could not be safely isolated; peroneus brevis transfer was added for symptomatic severe varus; Achilles tendon lengthening was reserved for rare cases with persistent fixed equinus. In most cases the orthopaedic procedure was performed during the same session, although delayed toe-flexor lengthening could be added secondarily when residual clawing persisted. Descriptive statistics are reported as mean ± standard deviation, mean only when dispersion was not available in the archive, or number (%). Preoperative and postoperative paired outcomes were compared using Wilcoxon signed-rank or paired t tests, as appropriate in the original thesis. Exploratory univariable analyses examined whether more than two-thirds (>67%) fascicle resection in the soleus and tibialis posterior branches was associated with greater reduction of clonus or deformity. The study was approved by the coordinating institutional review board of 115 People's Hospital (No. 462/BVND115-NCKH), and written informed consent had been obtained from patients or legal guardians according to the archived study documentation. Results Thirty-one patients underwent 36 STN procedures. Baseline patient characteristics are summarized in Table 1. The cohort was predominantly male (28/31, 90.3%), with a mean age of 42.5 ± 15.3 years (range 4–69). Stroke was the main aetiology (21 patients, 67.7%), followed by traumatic brain injury (5, 16.1%), cerebral palsy (3, 9.7%), spinal cord injury (1, 3.2%) and another central cause (1, 3.2%). The mean interval between the neurological event and surgery was 37.9 months. Five patients had bilateral procedures, and all patients had completed rehabilitation before surgery.Table 1Baseline patient characteristics (n = 31 patients).Table 1CharacteristicValueAge, mean ± SD (range), years42.5 ± 15.3 (4–69)Male sex, n (%)28 (90.3)Adults/children, n28/3Stroke, n (%)21 (67.7)Traumatic brain injury, n (%)5 (16.1)Cerebral palsy, n (%)3 (9.7)Spinal cord injury, n (%)1 (3.2)Other aetiology, n (%)1 (3.2)Delay from neurological lesion to surgery, mean (range), months37.9 (12–96)Right/left/bilateral involvement, n15/11/5Prior oral baclofen, n (%)18 (58.1)Prior botulinum toxin injections, n (%)4 (12.9)No previous antispastic drug treatment, n (%)9 (29.0)Structured rehabilitation before surgery31/31 (100%)Values are reported at patient level unless otherwise specified. Table 1Baseline patient characteristics (n = 31 patients).Table 1CharacteristicValueAge, mean ± SD (range), years42.5 ± 15.3 (4–69)Male sex, n (%)28 (90.3)Adults/children, n28/3Stroke, n (%)21 (67.7)Traumatic brain injury, n (%)5 (16.1)Cerebral palsy, n (%)3 (9.7)Spinal cord injury, n (%)1 (3.2)Other aetiology, n (%)1 (3.2)Delay from neurological lesion to surgery, mean (range), months37.9 (12–96)Right/left/bilateral involvement, n15/11/5Prior oral baclofen, n (%)18 (58.1)Prior botulinum toxin injections, n (%)4 (12.9)No previous antispastic drug treatment, n (%)9 (29.0)Structured rehabilitation before surgery31/31 (100%)Values are reported at patient level unless otherwise specified. Baseline patient characteristics (n = 31 patients). Values are reported at patient level unless otherwise specified. Preoperative clinical characteristics of the operated feet and baseline gait function are summarized in Table 2. Preoperatively, the deformity pattern extended beyond equinus alone. Equinus was present in 97.2% of operated feet, varus in 61.1%, and claw toes in 55.5%. Mean passive ankle dorsiflexion was 4.39° with the knee flexed and −3.67° with the knee extended. Clonus was frequent and severe, with mean scores of 2.81 with the knee flexed and 3.17 with the knee extended. Knee recurvatum was present in 61.1% of feet, toe-tip skin lesions in 44.4%, and pain during walking in 50.0%. Functionally, 48.4% of patients required an assistive device, mean walking distance with shoes was 445.5 m, and usual-speed 10-m walking time with shoes was 30.05 s.Table 2Preoperative clinical characteristics of the operated feet and gait function.Table 2VariableDenominatorValueEquinus present, n (%)36 feet35 (97.2)Varus present, n (%)36 feet22 (61.1)Claw toes present, n (%)36 feet20 (55.5)Passive ankle dorsiflexion with knee flexed, mean ± SD, °36 feet4.39 ± 3.14Passive ankle dorsiflexion with knee extended, mean ± SD, °36 feet−3.67 ± 3.67Clonus score with knee flexed, mean ± SD (0–4)36 feet2.81 ± 1.46Clonus score with knee extended, mean ± SD (0–4)36 feet3.17 ± 1.34Knee recurvatum present, n (%)36 feet22 (61.1)Toe-tip skin lesions, n (%)36 feet16 (44.4)Pain while walking, n (%)36 feet18 (50.0)Patients requiring an assistive device, n (%)31 patients15 (48.4)Walking distance with shoes, mean, m27 patients445.510-m walking time with shoes, usual speed, s31 patients30.05Orthotic comfort score, mean (0–10)8 patients6.1Orthotic comfort was available only in patients already using orthotic footwear before surgery. Table 2Preoperative clinical characteristics of the operated feet and gait function.Table 2VariableDenominatorValueEquinus present, n (%)36 feet35 (97.2)Varus present, n (%)36 feet22 (61.1)Claw toes present, n (%)36 feet20 (55.5)Passive ankle dorsiflexion with knee flexed, mean ± SD, °36 feet4.39 ± 3.14Passive ankle dorsiflexion with knee extended, mean ± SD, °36 feet−3.67 ± 3.67Clonus score with knee flexed, mean ± SD (0–4)36 feet2.81 ± 1.46Clonus score with knee extended, mean ± SD (0–4)36 feet3.17 ± 1.34Knee recurvatum present, n (%)36 feet22 (61.1)Toe-tip skin lesions, n (%)36 feet16 (44.4)Pain while walking, n (%)36 feet18 (50.0)Patients requiring an assistive device, n (%)31 patients15 (48.4)Walking distance with shoes, mean, m27 patients445.510-m walking time with shoes, usual speed, s31 patients30.05Orthotic comfort score, mean (0–10)8 patients6.1Orthotic comfort was available only in patients already using orthotic footwear before surgery. Preoperative clinical characteristics of the operated feet and gait function. Orthotic comfort was available only in patients already using orthotic footwear before surgery. Operative strategy and associated orthopaedic procedures are summarized in Table 3. Surgical treatment was individualised according to the deformity pattern. The proximal soleus motor branch was sectioned in all feet, and other branches were addressed according to clinical findings. Associated orthopaedic procedures were performed in 22 of 36 feet (61.1%), most commonly toe-flexor lengthening and peroneus brevis transfer. Fourteen feet underwent isolated STN without orthopaedic correction.Table 3Operative strategy, associated orthopaedic procedures and perioperative safety.Table 3ItemValuePrimary microsurgical strategySelective tibial neurotomy in all 36 operated feetMain motor branch always addressedProximal soleus branchLong toe-flexor motor branch selectively sectioned6 feet (18.2%)Isolated STN14 feet (38.9%)STN + toe-flexor lengthening12 feet (33.3%)STN + toe-flexor lengthening + Achilles tendon lengthening1 foot (2.8%)STN + peroneus brevis transfer3 feet (8.3%)STN + toe-flexor lengthening + peroneus brevis transfer6 feet (16.7%)Early complications3 procedures (8.3%)Complication profile1 superficial wound dehiscence; 1 reflex sympathetic dystrophy; 1 transient sensory disturbanceLate complications/deaths/re-operation for recurrence0/0/0Mean follow-up, months (range)38 (6–84)Associated orthopaedic procedures were usually performed during the same operative session when a fixed component of deformity was present. Table 3Operative strategy, associated orthopaedic procedures and perioperative safety.Table 3ItemValuePrimary microsurgical strategySelective tibial neurotomy in all 36 operated feetMain motor branch always addressedProximal soleus branchLong toe-flexor motor branch selectively sectioned6 feet (18.2%)Isolated STN14 feet (38.9%)STN + toe-flexor lengthening12 feet (33.3%)STN + toe-flexor lengthening + Achilles tendon lengthening1 foot (2.8%)STN + peroneus brevis transfer3 feet (8.3%)STN + toe-flexor lengthening + peroneus brevis transfer6 feet (16.7%)Early complications3 procedures (8.3%)Complication profile1 superficial wound dehiscence; 1 reflex sympathetic dystrophy; 1 transient sensory disturbanceLate complications/deaths/re-operation for recurrence0/0/0Mean follow-up, months (range)38 (6–84)Associated orthopaedic procedures were usually performed during the same operative session when a fixed component of deformity was present. Operative strategy, associated orthopaedic procedures and perioperative safety. Associated orthopaedic procedures were usually performed during the same operative session when a fixed component of deformity was present. Perioperative safety and follow-up are also summarized in Table 3. Mean clinical follow-up was 38 months (range 6–84). Early postoperative complications occurred in three procedures (8.3%): one superficial wound dehiscence, one reflex sympathetic dystrophy, and one transient sensory disturbance. All resolved with conservative treatment. No late complications, deaths or re-operations for recurrent disabling spasticity were documented during follow-up. Long-term structural and spasticity outcomes are shown in Table 4. Equinus score decreased from 1.53 ± 0.50 to 0.46 ± 0.51, varus from 1.74 ± 0.56 to 0.39 ± 0.49, and claw-toe score from 1.80 ± 0.33 to 0.50 ± 0.51 (all p < 0.001). Passive ankle dorsiflexion improved both with the knee flexed (4.39° to 7.58°; p < 0.001) and extended (−3.67° to +1.50°; p < 0.001). Clonus markedly decreased, with mean scores falling from 2.81 to 0.67 with the knee flexed and from 3.17 to 0.78 with the knee extended (both p < 0.001). Knee recurvatum also improved (0.64 to 0.47; p = 0.006). Toe-tip skin lesions decreased from 16 feet to 1 foot (p = 0.0002), and all 18 symptomatic patients reported relief of walking pain at final follow-up.Table 4Long-term structural and spasticity outcomes.Table 4OutcomePreoperativeLast follow-upp valueEquinus score (0–2)1.53 ± 0.500.46 ± 0.51<0.001Varus score (0–2)1.74 ± 0.560.39 ± 0.49<0.001Claw-toe score (0–2)1.80 ± 0.330.50 ± 0.51<0.001Passive dorsiflexion with knee flexed, °4.39 ± 3.147.58 ± 4.30<0.001Passive dorsiflexion with knee extended, °−3.67 ± 3.67+1.50 ± 4.16<0.001Clonus score with knee flexed (0–4)2.81 ± 1.460.67 ± 0.59<0.001Clonus score with knee extended (0–4)3.17 ± 1.340.78 ± 0.72<0.001Knee recurvatum score (0–2)0.640.470.006Toe-tip skin lesions, n (%)16/36 (44.4)1/36 (2.8)0.0002Pain while walking, n (%)18/36 (50.0)0/36 (0.0)All symptomatic patients improvedOrdinal outcomes were analysed with paired non-parametric tests; categorical outcomes were analysed with paired proportion tests when appropriate. Table 4Long-term structural and spasticity outcomes.Table 4OutcomePreoperativeLast follow-upp valueEquinus score (0–2)1.53 ± 0.500.46 ± 0.51<0.001Varus score (0–2)1.74 ± 0.560.39 ± 0.49<0.001Claw-toe score (0–2)1.80 ± 0.330.50 ± 0.51<0.001Passive dorsiflexion with knee flexed, °4.39 ± 3.147.58 ± 4.30<0.001Passive dorsiflexion with knee extended, °−3.67 ± 3.67+1.50 ± 4.16<0.001Clonus score with knee flexed (0–4)2.81 ± 1.460.67 ± 0.59<0.001Clonus score with knee extended (0–4)3.17 ± 1.340.78 ± 0.72<0.001Knee recurvatum score (0–2)0.640.470.006Toe-tip skin lesions, n (%)16/36 (44.4)1/36 (2.8)0.0002Pain while walking, n (%)18/36 (50.0)0/36 (0.0)All symptomatic patients improvedOrdinal outcomes were analysed with paired non-parametric tests; categorical outcomes were analysed with paired proportion tests when appropriate. Long-term structural and spasticity outcomes. Ordinal outcomes were analysed with paired non-parametric tests; categorical outcomes were analysed with paired proportion tests when appropriate. Functional outcomes are summarized in Table 5. Mean walking distance with shoes increased from 445.5 m to 872.3 m (p = 0.002). Standing and walking ease scores improved both with shoes (0.65 to 0.27 and 1.02 to 0.54, respectively) and barefoot (0.91 to 0.31 and 1.17 to 0.64, respectively). Usual-speed 10-m walking time improved from 30.05 s to 24.75 s with shoes (p = 0.004) and from 36.14 s to 29.25 s barefoot (p = 0.008). Orthotic comfort increased from 6.1 to 7.9/10 (p = 0.0005) among preoperative orthotic users, and the mean satisfaction score at follow-up was 7.8/10. Overall, 93% of patients were judged to have achieved 85–100% of their predefined goals.Table 5Functional outcomes after selective tibial neurotomy.Table 5OutcomePreoperativeLast follow-upp valueOrthotic comfort score (0–10; n = 8)6.17.90.0005Walking distance with shoes, m (n = 27)445.5872.30.002Standing score with shoes (0–2)0.650.270.005Walking ease score with shoes (0–2)1.020.54<0.000110-m walking time with shoes, usual speed, s30.0524.750.004Standing score barefoot (0–2)0.910.310.0002Walking ease score barefoot (0–2)1.170.64<0.000110-m walking time barefoot, usual speed, s36.1429.250.008Patients achieving 85–100% of predefined goals–93%–Satisfaction score at last follow-up (1–10)–7.8–For standing and walking scales, 0 = easy, 1 = possible only briefly or with marked difficulty, and 2 = impossible. Table 5Functional outcomes after selective tibial neurotomy.Table 5OutcomePreoperativeLast follow-upp valueOrthotic comfort score (0–10; n = 8)6.17.90.0005Walking distance with shoes, m (n = 27)445.5872.30.002Standing score with shoes (0–2)0.650.270.005Walking ease score with shoes (0–2)1.020.54<0.000110-m walking time with shoes, usual speed, s30.0524.750.004Standing score barefoot (0–2)0.910.310.0002Walking ease score barefoot (0–2)1.170.64<0.000110-m walking time barefoot, usual speed, s36.1429.250.008Patients achieving 85–100% of predefined goals–93%–Satisfaction score at last follow-up (1–10)–7.8–For standing and walking scales, 0 = easy, 1 = possible only briefly or with marked difficulty, and 2 = impossible. Functional outcomes after selective tibial neurotomy. For standing and walking scales, 0 = easy, 1 = possible only briefly or with marked difficulty, and 2 = impossible. Exploratory associations between fascicle resection extent and outcome are shown in Table 6. Resection of more than two-thirds of fascicles in the soleus branch was associated with greater reduction in clonus (p = 0.03). More extensive resection of the soleus and tibialis posterior branches was also associated with better correction of varus deformity (p = 0.01 and p = 0.002, respectively), whereas no significant association was demonstrated for equinus correction.Table 6Exploratory association between fascicle resection extent and surgical outcome.Table 6Outcome analysedMotor branchComparisonFindingpGreater clonus reductionSoleus branch>67% vs ≤ 67% fascicle resectionMore extensive resection was associated with greater reduction in clonus0.03Greater equinus correctionSoleus branch>67% vs ≤ 67% fascicle resectionNo significant association0.20Greater equinus correctionTibialis posterior branch>67% vs ≤ 67% fascicle resectionNo significant association0.59Greater varus correctionSoleus branch>67% vs ≤ 67% fascicle resectionMore extensive resection was associated with greater reduction in varus0.01Greater varus correctionTibialis posterior branch>67% vs ≤ 67% fascicle resectionMore extensive resection was associated with greater reduction in varus0.002High clonus reduction was defined in the thesis dataset as a decrease of 3–4 grades; high deformity correction was defined as a decrease of 2 grades. Table 6Exploratory association between fascicle resection extent and surgical outcome.Table 6Outcome analysedMotor branchComparisonFindingpGreater clonus reductionSoleus branch>67% vs ≤ 67% fascicle resectionMore extensive resection was associated with greater reduction in clonus0.03Greater equinus correctionSoleus branch>67% vs ≤ 67% fascicle resectionNo significant association0.20Greater equinus correctionTibialis posterior branch>67% vs ≤ 67% fascicle resectionNo significant association0.59Greater varus correctionSoleus branch>67% vs ≤ 67% fascicle resectionMore extensive resection was associated with greater reduction in varus0.01Greater varus correctionTibialis posterior branch>67% vs ≤ 67% fascicle resectionMore extensive resection was associated with greater reduction in varus0.002High clonus reduction was defined in the thesis dataset as a decrease of 3–4 grades; high deformity correction was defined as a decrease of 2 grades. Exploratory association between fascicle resection extent and surgical outcome. High clonus reduction was defined in the thesis dataset as a decrease of 3–4 grades; high deformity correction was defined as a decrease of 2 grades. Discussion This series shows that STN was associated with durable clinical and functional improvement in Vietnamese patients with spastic foot deformity after central nervous system injury. The gains were not limited to equinus correction; varus, claw toes, clonus, pain, skin integrity, orthotic tolerance and walking performance all improved over a mean follow-up of more than 3 years. These findings are broadly consistent with the classic French and later multicentre series that established STN as a valid treatment for distal lower-limb spasticity (Sindou and Mertens, 1988; Decq et al., 2000; Decq, 2003; Buffenoir et al., 2004, 2013; Rousseaux et al., 2008; Bollens et al., 2013). An important message of the present cohort is that the clinical phenotype was broader than “spastic equinus foot” alone. More than half of the operated feet also had varus and claw-toe deformities, which justifies reframing the study around spastic foot deformity rather than isolated equinus. This also explains why combined orthopaedic procedures were required in 61.1% of feet. In our practice, STN addressed the dynamic component of spasticity, whereas tendon or transfer procedures were added when a fixed structural component remained clinically relevant. The postoperative results should therefore be interpreted as the outcomes of an integrated treatment strategy whenever fixed deformity coexisted, not as the isolated effect of neurotomy alone in every case. The most original aspect of the study is the analysis of fascicle resection extent. Although many surgeons section between one-half and four-fifths of the selected motor branch, clinical data relating the exact proportion of fascicles resected to outcome remain sparse. In the present series, more than two-thirds resection of the soleus branch was associated with greater clonus reduction, and more than two-thirds resection of the soleus and tibialis posterior branches was associated with better varus correction. These findings support the concept that, within safe surgical limits and under intraoperative stimulation, sufficiently extensive motor fascicle resection may optimise control of the most disabling dynamic components of the deformity. The study also has practical relevance for resource-constrained settings. Several patients in the series had previously received oral baclofen or botulinum toxin, but the benefits had been incomplete or temporary. In contrast, STN offered a single microsurgical intervention with sustained improvement. For a health system in which repeated botulinum toxin treatment can be difficult to maintain, this is an important consideration. The main limitations are the retrospective design, the modest sample size, heterogeneity of underlying neurological aetiologies, and the frequent use of associated orthopaedic procedures. Electrophysiological measurements were not systematically available. In addition, some variables were recorded with pragmatic clinical scales developed for routine care rather than formal external validation. Nevertheless, the consistency of improvement across several domains and the relatively long follow-up strengthen the clinical relevance of the findings. Conclusion Selective tibial neurotomy, used alone or combined with orthopaedic correction when fixed deformity coexisted, was associated with long-term improvement in spastic foot deformity, gait, pain and orthotic tolerance, with a low complication rate. This initial Vietnamese experience supports STN as a feasible and durable option for focal lower-limb spasticity and suggests that resection of more than two-thirds of selected motor fascicles may optimise reduction of clonus and varus deformity. Declaration of competing interest We have no conflicts of interest to declare.