Recurrence rates and outcomes of shoulder stabilization surgery in patients with seizures

Abstract

Background

Patients with seizure disorders face a heightened risk of shoulder instability, often presenting with complex pathology and greater bone loss than nonseizure patients. We hypothesize that seizure patients experience a higher rate of recurrent instability after primary shoulder stabilization as compared to nonseizure patients.

Methods

This single-institution retrospective study (2009-2023) compares patients with recurrent shoulder instability with and without a seizure history undergoing shoulder stabilization surgery. Procedures included Bankart repair, remplissage, Hill-Sachs bone grafting, Latarjet procedure, and distal tibia allograft.

Results

Among 573 patients, 83 had seizures (46 epileptic, 37 nonepileptic). The seizure group was significantly older (29.1 vs. 24.9 years; P < .001), had higher body mass index (28.5 vs. 26.8; P = .010), greater comorbidities (Charlson-Deyo Comorbidity Index, American Society of Anesthesiologists scores, smoking, and depression; P < .001), and more glenoid bone loss (19% vs. 13%; P = .002). The direction of instability in the seizure group was anterior in 78% of patients (vs. 72%). Seizure patients had a significantly lower range of motion preoperatively and postoperatively. Seizure patients had worse postoperative visual analog scale (3.3 vs. 1.6; P = .001), Single Assessment Numeric Evaluation (75 vs. 86; P = .002), American Shoulder and Elbow Surgeons (69 vs. 87; P < .001), and Shoulder Instability-Return to Sport after Injury (42 vs. 71; P = .007) scores. Recurrent instability was significantly higher in seizure patients (28% vs. 17%; P = .022). In seizure patients, arthroscopic soft tissue repair had a high recurrence rate (48%), significantly worse than open bone augmentation (13%; P = .004). Nonseizure patients had similar recurrence rates for soft tissue repair (19%) and bone augmentation (15%; P = .33). Recurrence rates were comparable between seizure and nonseizure groups after bone augmentation (P = .833) but significantly higher in seizure patients undergoing soft tissue repair (P < .001). Instability-free survival at 2, 5, and 10 years was lower in seizure patients (P = .039) and lower for soft tissue repair vs. bone augmentation within the seizure group (P = .0016).

Conclusion

Seizure patients with shoulder instability have more comorbidities, worse clinical outcomes, and greater bone loss than nonseizure patients. Soft tissue repair in seizure patients has a high recurrence rate (48%), while bone augmentation procedures show lower recurrence, even after postoperative seizures. Bone augmentation should be strongly considered for these patients.

Patients with seizure disorders have a heightened risk of shoulder instability.¹⁴ This instability is often recurrent, and repeat dislocation rates up to 61% have been reported.²⁹ Recurrent shoulder instability can lead to diminished quality of life and propagation of additional complications such as osteoarthritis, rotator cuff tear, and inability to return to sports.^13,23,25 An additional challenge in the epilepsy population is seizure control, yet often, surgery is pursued despite the seizure disorder not being fully controlled due to inherent medical limitations. Despite poor functional outcomes, the seizure disorder patient population remains understudied in the context of shoulder instability.

Surgical management of these patients is challenging, given they often present with complex pathology and high bone loss.³ In general, surgical options for shoulder instability are dependent on the direction of instability and can include soft tissue stabilization or bone augmentation. Soft tissue procedures include arthroscopic labral repair with capsulorrhaphy with or without the addition of remplissage. There is also the soft tissue option of an open Bankart repair or capsular shift. Bone augmentation procedures are more commonly performed open and can range from the Latarjet technique, Hill-Sachs bone grafting, or glenoid reconstruction with tibia allograft.

Although surgeons utilize a variety of these procedures within the seizure patient cohort, the most appropriate option is debated. In part, there remains scant quantifiable data comparing outcomes to the general nonseizure disorder patient population. Previous small retrospective studies have demonstrated a benefit of bone augmentation within the seizure population to reduce the risk of instability.^6,29 However, these studies lack a nonseizure control group, and direct outcome comparisons between cohorts remain understudied. There also remains a gap in the literature regarding demographic trends, functional status, and patient-reported outcomes (PROs) within the seizure disorder population.

Our current study aims to evaluate demographic trends, functional outcomes, and postoperative complications, including recurrent instability and other complications, in patients with a documented seizure history undergoing shoulder stabilization surgery as compared to patients without seizures and categorized by procedure type. We further stratify by procedure type to help clarify the role of bone augmentation and soft tissue stabilization procedures within the seizure disorder patient population. We hypothesize that patients with a seizure disorder will have higher rates of recurrent instability compared to patients without seizures, particularly among those treated with soft tissue procedures.

Instability due to seizures is uniquely challenging. The mechanical pathology is often more severe than in routine shoulder instability, and these patients frequently face complex medical and socioeconomic barriers that complicate care. Our data support a paradigm shift toward considering bone reconstructive procedures in seizure patients—even at lower thresholds of bone loss—to optimize recurrence-free outcomes.

Materials and Methods

Study design and population

Patients who underwent shoulder stabilization surgery with a primary indication of recurrent instability between July 2009 and November 2023 were identified through a query of our institution’s database using relevant International Classification of Diseases-ninth and 10th Edition codes 719.2 and 718.3 (indicating pathologic and recurrent dislocation of the shoulder, respectively) and Current Procedural Terminology codes 23455 (open capsulorrhaphy with labral repair (eg, Bankart procedure), 29806 (arthroscopic capsulorrhaphy with labral repair (eg, Bankart procedure), 23462 (Latarjet procedure), and 29807 (arthroscopic labral repair) following study approval by institutional review board. An initial retrospective chart review was conducted on all potential patients, excluding those with less than 6 months of postoperative follow-up. No age minimum was set; thus, adolescent patients were included. A thorough review was then conducted on the remaining eligible patients.

Seizure history was defined as the occurrence of at least one seizure prior to the index surgery, irrespective of etiology. Epilepsy was defined through neurologist diagnosis as determined through chart review. All patients in the seizure group, including those without a formal diagnosis of epilepsy, experienced at least one tonic-clonic seizure. Patients were then categorized into their respective epileptic and nonepileptic seizure groups for subgroup analyses. Recurrent shoulder instability was defined as any subsequent episode of instability following the initial dislocation, including recurrent subluxations and/or locked dislocations, regardless of whether a closed reduction was required. All patients had radiographic evidence consistent with a dislocation event. Direction of instability was noted as anterior, posterior, or multidirectional, all of which are included. Patients with multidirectional instability were only included if they had associated structural pathology, such as labral tears. Pure multidirectional instability cases treated solely with capsular shift, without concomitant labral or other structural lesions, were excluded to maintain cohort consistency.

Operative and postoperative management

Shoulder stabilization surgeries were performed by 7 fellowship-trained orthopedic shoulder surgeons at a single institution. The primary procedures included arthroscopic Bankart repair, capsulorrhaphy, remplissage, open Latarjet, distal tibial allograft (DTA), and Hill-Sachs bone grafting. All procedures were performed either arthroscopic or open, other than Latarjet, which was only performed open. The standard deltopectoral approach was used for all open procedures. Various techniques were employed to manage the subscapularis muscle based on the procedure performed, including tenotomy, lesser tuberosity osteotomy, peel, and split; however, the subscapularis was appropriately repaired in all cases. Patients began structured physical therapy on average four weeks postoperatively and followed a standardized rehabilitation protocol specific to their procedure.

The choice of procedure was influenced by various factors, primarily including bone loss. Soft tissue repairs were typically performed in seizure patients without significant bone loss, consistent with the understanding of outcomes at the time. However, as experience with these cases grew, the approach shifted toward bony reconstruction procedures, even in patients without substantial bone loss, due to observed instability recurrence patterns and evolving clinical judgment. Due to the retrospective design and imaging variability, consistent glenoid and humeral defect measurements were not always available. However, patients with on-track lesions and <10% glenoid bone loss typically underwent soft tissue repair, while those with off-track lesions or >10% glenoid bone loss underwent bone augmentation.

Data collection

A retrospective review identified 2 cohorts: patients who had experienced at least one lifetime seizure prior to shoulder stabilization surgery (seizure cohort) and those without seizures (control cohort). Provider notes from preoperative visits were reviewed for seizure history and details of shoulder instability. Data collected included demographic characteristics, characteristics of instability including mode, duration, direction, history of prior shoulder dislocations, and shoulder function (eg, strength and range of motion [ROM]). ROM was assessed during follow-up visits via clinical examination. Additional data for the seizure cohort included seizure onset date, type of seizure disorder (epileptic vs. nonepileptic), and seizure recurrence. Information from operative reports included procedures performed, approach (open vs. arthroscopic), indications, rotator cuff status, glenoid bone loss, and Hill-Sachs engagement. Imaging reports were used to confirm rotator cuff status and glenoid bone loss if not mentioned in the operative reports. Anesthesia reports provided American Society of Anesthesiologists (ASA) scores and Charlson-Deyo Comorbidity Index (CCI) scores. Postoperative data were obtained from follow-up visit notes and included postoperative shoulder function (eg, strength and ROM), complications, and the need for revision surgery. PROs included visual analog scale pain score, Single Assessment Numeric Evaluation score, American Shoulder and Elbow Surgeons (ASES) score, and Shoulder Instability-Return to Sport after Injury score and were collected via phone interviews.

Statistical analyses

Kaplan-Meier survival curves were used to estimate survival probability, with a 95% confidence interval to assess failure rates. Differences in survival distributions were evaluated using log-rank tests. Continuous variables were analyzed with a Student's t-test or analysis of variance, while ordinal variables were compared using ordinal logistic regression or the Wilcoxon signed-rank test. A significance threshold of P < .05 was set for all two-sided tests. Results were presented as mean ± standard deviation. All statistical analyses were performed using Stata/SE v17.0 (StataCorp, College Station, TX, USA). This study was approved by the institutional review board of The Ohio State University, protocol #2024H0084.

Results

Demographics

A total of 573 patients (78% male) with recurrent shoulder instability underwent primary shoulder stabilization surgery. Of these, 83 had at least one seizure prior to the index procedure compared to 490 patients without seizures. The seizure group included 46 patients with epileptic seizures and 37 with nonepileptic seizures. Patients with seizures are on average older at 29.1 ± 8.8 years (vs. 24.9 ± 10.0 years; P < .001). Follow-up in our seizure cohort varied at 3.5 ± 4.0 years (range: 0.5 to 15.2 years). Patients with seizures had inferior physical health status prior to surgery as indicated by significantly higher ASA physical classification scores (2.1 ± 0.5 vs. 1.4 ± 0.6; P < .001) and CCI scores (0.3 ± 0.8 vs. 0.1 ± 0.4; P < .001). The seizure group also had significantly higher rates of smoking (38% vs. 10%; P < .001) and depression (55% vs. 17%; P < .001) compared to the control. The direction of instability in the seizure group was anterior in 79% of patients (vs. 72%). Average glenoid bone loss was significantly higher in the seizure group (18.7% vs. 12.6%; P = .002) (Table I).

Table I.

Comparative demographics.

	Seizure (n = 83)	No seizure (n = 490)	P value
Sex (M:F)	64:19	381:109	.89
Age (yr)	29.1 ± 8.8	24.9 ± 10.0	<.001
BMI (kg/m2)	28.5 ± 6.6	26.8 ± 5.4	.010
Follow-up (yr)	3.5 ± 4.0	3.2 ± 3.8	.62
Smoking rate (%)	38	10	<.001
Depression rate (%)	55	17	<.001
Glenoid bone loss (%)	18.7 ± 13.8	12.6 ± 9.7	.002
ASA	2.1 ± 0.5	1.4 ± 0.6	<.001
CCI	0.3 ± 0.8	0.1 ± 0.4	<.001

M, male; F, female; BMI, body mass index; ASA, American Society of Anesthesiologists; CCI, Charlson-Deyo Comorbidity Index.

Recurrent instability, recurrent seizures, and revisions

Recurrent instability was the primary indication for surgery in all patients. The type of procedure performed in each group can be found in Table II. The rate of recurrent instability following shoulder stabilization surgery was significantly higher in the seizure cohort (28% vs. 17%; P = .022). Notably, the recurrent instability rate was markedly higher in the seizure cohort after undergoing arthroscopic soft tissue repair as compared to seizure patients undergoing open augmentation, with 13 of 27 patients (48%) having recurrence in the soft tissue group compared to 4 of 30 patients (13%) in the open augmentation group (P = .004). Recurrent instability was comparable between procedure types in the nonseizure group, with 46 of 244 arthroscopic repairs (19%) and 19 of 128 open bone grafts (15%) having recurrence (P = .33). Seizure patients who underwent open bone augmentation had recurrence rates comparable to nonseizure patients (P = n.s.), while those who underwent arthroscopic soft tissue repair experienced significantly higher recurrence than their nonseizure counterparts (P < .001) (Table III).

Table II.

Procedure types.

Procedure type	Seizure (n = 83)	No seizure (n = 490)
Arthroscopic soft tissue repair (%)	27 (32%)	244 (50%)
Open bone augmentation (%)	30 (36%)	128 (26%)
Combination (%)	26 (31%)	118 (24%)

Combination = multiple procedures that cannot be confined to open bone augmentation or arthroscopic soft tissue repair.

Table III.

Comparative functional outcomes of range of motion.

	Seizure (n = 83)	No seizure (n = 490)	P value
ROM FE
Preoperative	143° ± 42°	159° ± 26°	<.001
Postoperative	147° ± 33°	165° ± 16°	<.001
ROM ER
Preoperative	56° ± 26°	64° ± 20°	<.001
Postoperative	49° ± 19°	64° ± 18°	<.001
ROM IR
Preoperative	T11	T11	.13
Postoperative	T11	T10	<.001

ROM, range of motion; FE, forward elevation; ER, external rotation; IR, internal rotation.

The recurrence-free rate was significantly lower at 2 years (73%), 5 years (61%), and 10 years (47%) in the seizure group vs. 83%, 72%, and 59%, respectively, in the non-seizure group (P = .039) (Fig. 1). The 2-year (seizure: 82% vs. 87%) and 5-year (seizure: 70% vs. 69%) revision rates were comparable between groups (P = n.s.). For seizure patients, the recurrence-free rate of 53% at 2 years, 38% at 5 years, and 20% at 10 years for soft tissue repair was significantly lower than open bone augmentation at 75% for 2, 5, and 10 years. In the nonseizure group, soft tissue repair resulted in recurrence-free rates of 81% at 2 years, 71% at 5 years, and 59% at 10 years, comparable to bone augmentation with 86%, 72%, and 64% at the respective time points (P = .0016) (Fig. 2).

Figure 1.

Instability recurrence-free survival Kaplan-Meier curve by patient type Seizure patients: 2-year 73%; 5-year 61%; 10-year 47% Nonseizure patients: 2-year 83%; 5-year 72%; 10-year 59%.

Figure 2.

Instability recurrence-free survival Kaplan-Meier curve by procedure classification: Seizure patients, soft tissue repair: 2-year 53%, 5-year 38%, 10-year 20%: Seizure patients, bone augmentation: 2-year 75%, 5-year 75%, 10-year 75%: Nonseizure patients, soft tissue repair: 2-year 81%, 5-year 71%, 10-year 59%: Nonseizure patients, bone augmentation: 2-year 85%, 5-year 72%, 10-year 64%.

Of patients with epilepsy, 93% were being medically treated for their seizures at the time of index surgery, yet 58% had seizure recurrence postoperatively. There were 15 patients who experienced postoperative seizures after the Latarjet procedure or DTA, with 1 developing recurrent instability resulting in subscapularis and hardware failure, leading to revision distal tibial allograft. Two patients had prolonged postoperative pain, and one patient's postoperative seizures resulted in bent screws and mild graft nonunion, while the other 11 had no complications. There were 12 patients who had postoperative seizures after isolated arthroscopic soft tissue repair, resulting in 8 cases of recurrent instability and 7 revisions, with revision procedures including open Bankart repair (n = 2), Latarjet (n = 2), proximal humerus ORIF (n = 1), remplissage (n = 1), and distal tibial allograft (n = 1). Revision rates were comparable between groups: 13% of seizure patients underwent revision on average 2.3 years after the index procedure compared to 11% of nonseizure patients at 3.3 years.

Functional outcomes and PROs

Patients with seizures had significantly lower ROM in nearly all measures compared to those without seizures. Measurements of preoperative and postoperative ROM can be found in Table IV.

Table IV.

Comparative patient-reported outcomes.

Patient-reported outcome	Seizure (n = 83)	No seizure (n = 490)	P value
VAS	3.3 ± 2.7	1.6 ± 2.2	.001
SANE	75 ± 16	86 ± 15	.002
ASES	69 ± 20	87 ± 16	<.001
SIRSI	42 ± 36	71 ± 25	.007

VAS, visual analog scale; SANE, Single Assessment Numeric Evaluation; ASES, American Shoulder and Elbow Surgeons; SIRSI, Shoulder Instability-Return to Sport after Injury score.

Seizure patients exhibited significantly worse PROs in all measures. Postoperative PROs can be found in Table V.

Table V.

Comparative recurrent instability rates by procedure classification.

Procedure type	Seizure cohort			Nonseizure cohort
	Patients	Instability recurrences	Recurrence rate	Patients	Instability recurrences	Recurrence rate	P value
Overall	83	23	28%	490	84	17%	.022
Soft tissue repair	27	13	48%	244	46	19%	<.001
Bone augmentation	30	4	13%	128	19	15%	.83

Bold indicates statistically significant P value.

Discussion

Stabilization surgery for recurrent anterior and posterior instability in seizure patients had significantly higher rates of postoperative instability recurrence compared to nonseizure patients. When stratified by procedure type, seizure patients had a significantly higher rate of recurrent instability after arthroscopic soft tissue repair (48%) versus open bone grafting (13%). Kaplan-Meier projections followed this trend, showing that only 59% of patients undergoing soft tissue repair remained stable at 10 years. Postoperative seizures occurred in both our soft tissue and bony augmentation cohorts, yet instability recurrence rates were significantly lower after bony augmentation compared to soft tissue repair, even in patients with recurrent seizures. Our study also found significantly more comorbidities, lower postoperative ROM, greater glenoid bone loss, and poorer PROs in seizure patients compared to nonseizure patients. Previous literature has discussed shoulder instability in seizure patients, often with relatively small cohorts, and most without a comparison group. This study, with one of the largest single-institution sample sizes to date of patients with recurrent shoulder instability and seizures, addresses the compelling differences in demographics, PROs, functional outcomes, recurrence rates, and revision surgeries between these cohorts.

Seizures are relatively common, with up to 8% of people experiencing at least one in their lifetime, and 1.7% of adults reporting either active or resolved epilepsy.^7,19 The incidence of shoulder dislocation during seizure is around 0.6%, though the actual incidence is likely higher due to undetected dislocation events.^1,9 Recurrent instability after the first dislocation can occur in up to 61% of patients.²⁹ Shoulder stabilization procedures primarily aim to restore shoulder stability, though some function may be lost. The poorer functional outcomes observed in our seizure cohort are likely multifactorial, influenced by worse baseline measurements, higher rates of depression, smoking, and preoperative comorbidities. Up to 75% of nonseizure patients can expect to regain their baseline ROM, with nearly all recovering baseline strength following shoulder stabilization.⁵ In a study of 44 patients with epilepsy and posterior shoulder dislocations, 52% reported ROM deficits from baseline, though 96% considered their mobility adequate.³⁴ Our data reflect this functional difference, with our seizure cohort having significantly lower ROM in nearly all directions (forward elevation, external rotation, internal rotation) than those without seizures. Depression, prevalent in epilepsy (up to 23%), is linked to worse functional outcomes and lower return-to-sport rates poststabilization.^11,12,18,32 Smoking, known to impair healing in rotator cuff repair, may also impact recovery for shoulder stabilization procedures but requires further investigation.^2,4 While not extensively studied in shoulder instability, patients with more comorbidities have shown poorer shoulder function at baseline, as evidenced in a study of 1,817 shoulder arthroplasties.²² Higher preoperative comorbidity, reflected by higher smoking, depression, CCI, and ASA in our seizure cohort, may partly explain the observed baseline deficits in ROM and lower postoperative measurements.

PROs play a vital role in assessing success following shoulder instability procedures, as they capture the patient's perspective on functional recovery, pain, and quality of life. However, the literature on PROs in this population is limited, and no consensus exists on the most effective tool for assessing this group. Agha et al¹ report a postoperative ASES score of 67.3 in 25 patients with epilepsy undergoing shoulder stabilization, yet there was no difference found when compared to nonseizure patients. We observed significantly worse scores in our seizure patients across all PRO measures when compared to nonseizure patients, including Single Assessment Numeric Evaluation, ASES, visual analog scale, and Shoulder Instability-Return to Sport after Injury. These findings, along with poor ROM measurements, indicate that individuals with seizure disorders not only experience worse objective functional outcomes but also report higher pain and lower perception of their shoulder's functionality. This underscores the need for tailored assessment to address both physical and subjective challenges in this complex patient population.

The optimal surgical approach for recurrent anterior or posterior shoulder instability in patients with seizures—whether soft tissue repair or bony reconstruction—remains a topic of extensive debate. This debate is influenced by discrepant bone loss and instability recurrence rates compared to shoulder stabilization in patients without seizure history.^3,14 Seizure patients have been found to have higher bone loss when compared to those without seizures, but seizure history alone is not a predictor of undergoing open bone grafting procedures.¹ When bone loss causes significant glenoid defects, recurrence rates can be as high as 67% following arthroscopic repair.³³ Current literature indicates that bone grafting or a Latarjet procedure is often considered when bone loss exceeds 13.5%,³⁵ whereas soft tissue repair is preferred for less bone loss.¹⁵ It is important to note that it has been suggested that seizure patients may not do well with the Latarjet procedure, as postoperative seizures can potentially disrupt the coracoid transfer through fracture, nonunion, or malunion of the graft. This concern should be addressed however possible, though postoperative seizures are often unavoidable. There is a role of interdisciplinary cooperation with neurologists for preoperative and postoperative seizure control optimization, as well as with anesthesiologists during the perioperative period. Patient-specific factors (ie, concomitant seizures) and surgeon-specific experience or preferences may take precedence over a standardized bone loss percentage or postoperative seizure risk when determining the appropriate procedure. In this study, soft tissue repairs were initially performed in seizure patients without significant bone loss, in line with the understanding of outcomes at that time. As experience with these cases grew and instability recurrence patterns were observed, the approach shifted toward bony augmentation procedures, even in patients without substantial bone loss, reflecting evolving clinical judgment in light of these outcomes.

There is evidence that bone grafting procedures can reduce recurrence rates more effectively than soft tissue repairs in patients with seizures. For instance, in a study of 49 shoulders in 33 epileptic patients, Thangarajah and Lambert³⁰ found that skeletal stabilization significantly reduced recurrence rates when compared to soft tissue repair. Hutchinson et al¹⁷ similarly reported better outcomes in 14 patients with seizures and recurrent anterior instability treated with bony reconstruction. Despite these reported advantages, some studies find the contrary. A recent systematic review by Rai et al²⁶ found that the Latarjet procedure carries over three times the risk of failure in seizure patients when compared to nonseizure patients, especially so in patients with postoperative seizures. In Raiss et al's²⁷ study of 12 epileptic patients with recurrent shoulder instability, high recurrence rates (36%) and patient dissatisfaction (45%) were reported following the Latarjet procedure. Meuffels et al^24,28 and Struck et al^24,28 also observed similar challenges with posterior bone blocks. However, our study results differ from these findings, as we report an insignificant difference in instability recurrence between seizure and nonseizure groups following open bone grafting procedures.

Soft tissue repairs, while less invasive, can be associated with higher failure rates in seizure patients. Bühler and Gerber⁶ found a 30% recurrence rate following soft tissue repair in 34 shoulders with seizure-induced instability. Similarly, Thangarajah and Lambert²⁹ reported a 71% recurrence rate following soft tissue repair in patients with epilepsy, compared to only 38% for bony reconstructions. In a study of 27 patients with seizures undergoing arthroscopic Bankart repair with remplissage, Guity and Eraghi¹⁶ found excellent functional outcomes and a recurrence rate of 17%. Overall, data regarding instability recurrence between soft tissue repair and bone augmentation are heterogenous, further exemplifying the uniqueness of this study population. Our study's findings favor open bone grafting, as soft tissue repair alone resulted in a nearly four-fold higher rate of recurrent instability. Recurrence-free rates at 2, 5, and 10 years were significantly lower for seizure patients undergoing soft tissue repair compared to bony augmentation, whereas in nonseizure patients, recurrence-free rates were similar between procedures. Notably, seizure patients who underwent bone augmentation had recurrence-free rates comparable to nonseizure patients undergoing these procedures, while soft tissue repair showed a significantly higher instability recurrence rate and lower recurrence-free rate only in seizure patients. Given the elevated risk of recurrent instability in seizure patients, we recommend prioritizing stability with bony reconstruction, particularly in cases with significant bone loss, to enhance long-term stability.

Recurrent seizures following shoulder stabilization procedures pose a significant risk for patients with seizure disorders, often leading to redislocations and the need for revision surgery. Traditional guidance suggests delaying surgery until a patient has been seizure-free for six months, but this is not always feasible. Many patients never experience a 6-month seizure-free period, and for quality-of-life reasons, surgery often becomes necessary despite the risks. While every attempt should be made to coordinate with neurology for the optimal medical management, postoperative seizures remain a significant concern, requiring careful risk-benefit assessment. Particularly, seizure recurrence following open bone grafting procedures carries risks such as graft or hardware failure, especially when early in the postoperative course. Thon et al³¹ reported that among 9 patients with seizure disorders, 4 experienced seizure recurrence after Latarjet (44%), with 3 of those requiring revision surgery. Erşen et al¹⁰ found that 3 of 9 (33%) patients with epilepsy experienced recurrent seizures following their initial stabilization procedure, with 1 requiring revision surgery. They noted that earlier seizure recurrence was more likely to result in failure, likely due to immature graft union.¹⁰ In a study of 12 patients with epilepsy (14 shoulders) who underwent the Latarjet procedure for recurrent seizures and anterior shoulder dislocations, Raiss et al^26,27 found that 6 shoulders (43%) experienced redislocation due to a postoperative seizure. Of these, 5 required revision surgery. Rai et al²⁶ found that 33% of seizure patients had a postoperative seizure after the Latarjet procedure, and 67% of these patients subsequently had a recurrent instability rate. The mainstay to prevent both recurrent instability and subsequent revision is the prevention of postoperative seizures, though efforts may be futile in cases of refractory epilepsy. Ninety-three percent of patients with a seizure disorder were actively managed by a neurologist preoperatively. In all cases, we coordinated with the treating neurologist to optimize medical therapy. However, due to the nature of these conditions and socioeconomic barriers, strict seizure-free intervals were not always achievable. In some patients, seizures persisted despite advanced interventions such as brain stimulation or neurosurgical procedures. When instability significantly impaired quality of life and nonoperative management failed, surgery proceeded despite imperfect seizure control. Our data support considering bone augmentation even if a patient is at risk of recurrent seizures.

Revision rates for repeat shoulder stabilization can vary widely based on the type of primary procedure and are inconsistently reported in the literature.²¹ Comorbid seizures further add to the complexity, as there is limited research on this specific population. While seizure patients exhibited a higher recurrence rate (48%), the revision surgery rates remained comparable between groups, suggesting that recurrence may be more prevalent but not necessarily lead to the need for more frequent revisions. Our revision rates—13% in patients with seizures and 11% in those without—are comparable to previously reported rates after primary shoulder stabilization. Similarly, a study of 1,298 primary shoulder stabilizations found a revision rate of 13.7%, with no difference between 1,273 nonseizure and 25 seizure patients.¹ Failure of bone block procedures in seizure patients can leave few or no viable revision options. Although some patients in our cohort underwent Latarjet procedures, our practice has shifted away from using this technique in seizure patients due to risks of early graft failure caused by coracoid displacement during tonic-clonic seizures. We prefer distal tibial allograft and recommend avoiding titanium and cannulated screws. When subscapularis takedown is needed, we coordinate with physical medicine & rehabilitation to administer botulinum toxin injections to major shoulder muscle groups 2–3 weeks preoperatively to mitigate violent muscle firing during seizures. In patients where the subscapularis must be preserved, we advocate for an arthroscopic DTA approach; this is a relatively new technique and was not performed during the majority of the study period, but we now find it very promising as it allows subscapularis preservation.

This study has several notable limitations. The retrospective case-control design introduces challenges, particularly regarding data collection and potential biases. Consequently, we were unable to gather preoperative PROs, as they were not previously recorded. The reliance on existing records creates variability in data quality and completeness, potentially affecting the accuracy and reliability of our findings. Although not a direct limitation of our methodology, we acknowledge that this study included only operatively treated patients, introducing a significant selection bias, particularly in the epileptic population. Previous literature suggests that only 50-60% of seizure patients with shoulder instability undergo surgical treatment, and outcomes for nonoperatively treated patients remain unclear.^8,20 While shoulder instability has generally demonstrated positive outcomes across various conditions, it is crucial to recognize that underlying pathologies can significantly influence postoperative results. Bone loss was not matched between cohorts, as doing so would have resulted in an underpowered analysis. Moreover, this study does not account for certain complex aspects of epilepsy and mechanisms of seizures that could benefit from a neurologic specialist's input. Seizure mechanisms can be variable and are not accounted for in this study. We did not separately analyze patients with epilepsy vs. nonepileptic seizures due to limited sample size, though we acknowledge these groups may differ in recurrence risk due to factors like seizure frequency, medication effects, and bone health. Conducted at a single institution, the research may also reflect variability in surgical techniques, potentially impacting outcome consistency.

Conclusion

Seizure patients with shoulder instability who undergo shoulder stabilization procedures have higher rates of postoperative recurrent instability when compared to patients without seizures. Instability recurrence-free rates are significantly lower for seizure patients undergoing soft tissue repair, with only 20% remaining recurrence-free at 10 years, compared to 75% for those treated with bony augmentation. Bone loss was greater in the seizure group, highlighting the need for bone augmentation to reduce recurrence rates. Additionally, patients with seizures tend to have higher rates of comorbidities, worse functional outcomes, and poorer PROs compared to nonseizure patients. While soft tissue repair may suffice in cases with minimal bone loss, our data support prioritizing bone augmentation in seizure patients, even in those who have a risk of recurrent seizures.

Disclaimers:

Funding: No funding was disclosed by the authors.

Conflicts of interest: The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.