Endoscopic techniques for removal of intraorbital foreign bodies – a clinical series

Abstract

Background

Management of intraorbital foreign bodies (IOFBs) is challenging due to the confined space and proximity of neurovascular and ocular structures. The increasing adoption of endoscopic techniques has expanded minimally invasive treatment options. We report operative corridors, indications, and outcomes of various endoscopic approaches in a consecutive series.

Method

We analyzed 16 patients with IOFBs treated between 2015 and 2023 at a tertiary university center. All procedures were performed using exclusively endoscopic techniques. Eleven IOFBs were metallic and five organic. Surgical strategy was individualized according to IOFB location and included endoscopic transethmoidal, transmaxillary, transconjunctival, and retrocantal approaches. Neuronavigation was used in all cases with adjunctive fluoroscopy or ultrasonography when indicated.

Results

Nine IOFBs were intraconal and seven extraconal. Successful removal was achieved in 13 of 16 patients. All unsuccessful attempts involved small, non-ferromagnetic air-gun pellets located deep in the posterior orbit; these patients remain under observation without delayed inflammatory complications. No new postoperative visual deterioration or neurological deficits occurred. Transient periorbital edema was observed in all patients and resolved spontaneously. Cosmetic outcomes were satisfactory in all but one case with extensive initial soft-tissue loss. In three patients with presumed orbital tumors, removal of wooden fragments led to regression of inflammatory masses on follow-up.

Conclusions

Endoscopic techniques represent a valuable alternative to traditional open approaches, especially for deep-seated lesions. The optimal approach should prioritize the most direct path to the target, avoiding critical structures such as the optic nerve, extraocular muscles, ciliary nerves and vessels to minimize the risk of permanent deficits.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00701-026-06986-1.

Background

Surgical management of intraorbital foreign bodies (IOFBs) is technically demanding: operative corridors are narrow, orbital contents are highly mobile and the margin for globe, vascular and neural structures injury is small. As a result, initial observation is commonly chosen, particularly for well-tolerated inorganic materials—except for copper, iron and lead. Copper may trigger intense purulent inflammation, iron can cause siderosis bulbi; lead carries a risk of systemic toxicity [4, 11, 15]. In contrast, organic foreign bodies should be removed promptly due to their significant risk of infection and pseudotumor formation [4, 7, 9, 15].

Endoscopic approaches, such as endonasal or transconjunctival, extend access to regions of the orbit that are difficult to reach with classic open techniques and often provide a shorter, more direct trajectory to deeply seated IOFBs, with less soft-tissue disruption and no visible scarring [1, 18, 20].

With minimally disruptive corridors available, the threshold for surgical removal may be lower in selected patients, particularly when IOFB is symptomatic or carries inherent risks of infection and delayed intraorbital pseudotumor formation [5, 8, 16].

Since 2007 interdisciplinary collaboration between otolaryngologists and neurosurgeons at our tertiary-level university hospital has facilitated the execution of over 450 intraorbital surgical procedures. We also applied endoscopic corridors previously described by our group [18–20] to address the complex and challenging surgical management of IOFBs.

In this paper we report endoscopic operative corridors, indications, technical nuances, and outcomes in a consecutive interdisciplinary series of 16 IOFB removals.

Methods and materials

Between 2015 and 2023 our interdisciplinary team comprising neurosurgeons and otolaryngologists specializing in endoscopic skull base surgery treated 16 consecutive patients, aged 12 to 64 years, with IOFBs. All procedures were performed in a deferred non-emergent setting under general anaesthesia with patients positioned supine. For transnasal approaches permissive hypotension under arterial-line monitoring was used to reduce the mucosal bleeding and improve the endoscopic visualisation of the operative field.

Of the 16 cases, 11 involved metallic IOFBs and 5 involved organic IOFBs (Table 1). In three patients a preoperative diagnosis of an inflammatory pseudotumor was established but during surgery an associated organic foreign body was identified. Preoperative ophthalmological evaluation confirmed preserved visual acuity in 11 affected eyes, with only light perception retained in 5 cases. Ocular motility was impaired in 6 patients on the affected side. Two patients presented with confirmed eyeball perforation; enucleation was not pursued initially in either case. All surgical procedures were performed electively.

Table 1.

Characteristic of intraorbital foreign bodies (IOFB), surgical approaches and result of surgery

No	Type	Location	Approach	Result of surgery
1	round pellet	intraconal	eSC	-
2	hydrated wood	intraconal	eTEA	+
3	hydrated wood	extraconal	eTEA	+
4	metal shard	intraconal	eTEA	+
5	round pellet	intraconal	eTEA	-
6	hydrated wood	intraconal	eTEA	+
7	hydrated wood	extraconal	BeTMA	+
8	air-gun pellet	extraconal	ELO	+
9	metal circulating saw	extraconal	eLRC	+
10	metal rod	extraconal	eLRC	+
11	metal shard (hammer)	intraconal	lateral TC	+
12	air-gun pellet	extraconal	medial TC	+
13	metal shard	extraconal	lateral TC*	+
14	hydrated wood	intraconal	medial TC	+
15	metal shard	intraconal	medial TC	+
16	round pellet	intraconal	medial TC	-

eSC endoscopic superior eyelid crease approach, eTEA endoscopic transethmoidal approach, BeTMA biportal endoscopic transmaxillary approach, ELO endoscopic lateral orbitotomy, eLRC endoscopic lateral retrocantal, TC transconjunctival (endoscopic). + indicates complete removal of IOFB,—indicates insuccesful extractions (details in text). * the approach was transconjunctival via original IOFB entry wound

Procedures were performed using either a Karl Storz or an Olympus camera. Fixation of sheathed Hopkins 0°, 30° and 45° telescopes was achieved using either a mechanical or a pneumatic holder. A radiolucent head holder was universally utilized to ensure compatibility with the magnetic navigation system (Fusion, Medtronic, USA), which was used for transnasal approaches. Optical neuronavigation, based on preoperative CT and/or MRI imaging (S7, Medtronic, USA), was used for intraoperative localization of IOFBs in all other approaches, supplemented by a radiological C-arm for metallic IOFBs.

All transnasal procedures employed a standard skull base instrument set, supplemented by customized curettes bent intraoperatively to precisely target deeply situated foreign bodies. High-speed drills with extended sleeves were utilized for transnasal, transethmoidal and transmaxillary approaches, with the Medtronic Straightshot Microdebrider with angled burrs employed for access through the inferior orbital wall via transnasal antrostomy into the maxillary sinus, as detailed below. For transconjunctival approaches, ophthalmological instruments (Geuder AG, Germany) were utilized, with customized cotton sticks facilitating manipulation of orbital fat sheaths. During these non-transnasal procedures spatulas mounted on a Leyla self-retaining retractor maintained the operative corridor by retracting orbital contents. A miniature neodymium magnet was occasionally introduced through a conjunctival incision or nostril to extract ferromagnetic IOFBs.

Endoscopic approaches

Endoscopic transethmoidal approach (eTEA)

The endoscopic transethmoidal approach (eTEA) may be performed via a single-nostril or binostril (transseptal) technique to enhance visualization or expand the range of motion for surgical instruments. The approach starts with complete ethmoidectomy and sphenoidotomy combined with infundibulotomy, and a wide antrostomy to open maxillary sinus. This exposes the lamina papyracea which is then thinned with a high-speed drill and partially removed from the periorbit at a site determined by a navigated tool, typically a suction tube. When additional exposure is required the inferomedial orbital wall could also be drilled up to the infraorbital nerve (Fig. 1A). The periorbit was incised with a diamond blade and/or a sickled knife. The medial rectus muscle was then identified and—if necessary—pulled out to visualize the intraconal space. Then tissue preparation was carried out under neuronavigation guidance (supp. Video 1). Because the foreign body may move within the orbital fat adjunct localization via intraoperative C-arm x-ray or ultrasonography is often required. Hemostasis was obtained with a monopolar or bipolar coagulation or hemostatic agents. The periorbit was covered with a patch of Tachosil (Takeda) or Hemopatch (Baxter). Bony reconstruction was generally not performed.

Fig. 1.

Schematic coronal representation of intraorbital compartments suitable for specific endoscopic approaches. orange – lateral approaches (ELO, eLRC) for extraconal space (n = 3); yellow – BeTMA for extraconal space (n = 1); dark green – eTEA for extraconal space (n = 1); light green – eTEA for intraconal space (n = 5); dark blue – intraconal space; light blue – extraconal space; grey intrupted line – optic nerve; red intrupted line – extraocular muscles; pink– infraorbital nerve (lateral limitation of EEA approach)

Biportal endoscopic transmaxillary approach (BeTMA)

The biportal endoscopic transmaxillary approach (BeTMA) was employed in one case to address an IOFB embedded in the inferior orbit. The first port was established via a transnasal corridor, involving infundibulotomy followed by wide antrostomy to access the maxillary sinus medially. The second port was created by drilling through the anterior wall of the maxillary sinus via the canine fossa, forming a mini Caldwell-Luc opening. This biportal approach facilitated simultaneous introduction of multiple instruments from different angles, including the telescope, suction tube and other dissection tools (supp. Video 2).

Endoscopic transconjunctival approach

To access foreign bodies located in the anterior orbit, adjacent to the surface of the globe a conjunctival – tenon flap was created using micro-scissors with a curvilinear incision, approximately 150 degrees in circumference adjacent to the iris. The flap was gently elevated and secured with stay sutures to maintain exposure. The adjacent rectus muscles (superior, inferior and medial) were identified using a strabismus hook and a suture was passed beneath each rectus muscle near its insertion to the globe. Both ends of the suture were secured with a clamp, facilitating controlled mobilization of the eyeball by gentle traction on this provisional construct. Under continuous guidance from a neuronavigation system (and/or intraoperative fluoroscopy, when indicated), an endoscopic corridor was established targeting the IOFB. Orbital fat was carefully retracted using cotton-tipped applicators to maintain a clear operative field. In select cases, a fixed miniature spatula was employed for additional retraction.

This approach is also well-suited for IOFBs located lateral to the globe. Following the conjunctival incision, the lateral rectus muscle is isolated with a muscle hook and secured with stay sutures. The telescope is then introduced into the extraconal retrobulbar space lateral to the globe (supp. Video 3).

Endoscopic lateral retrocantal approach (eLRC)

Endoscopic lateral retrocantal (eLRC) represents a minimally invasive modification of the microsurgical technique designed to access IOFBs located peripherally and extraconally. This technique provides extended access to the entire lateral orbit and orbital apex while preserving the structural integrity of the eyelid. The conjunctival incision is performed posterior to the lateral canthal ligament to avoid compromising this structure. Dissection is maintained in the plane between the orbital wall and periosteum, thereby sparing the lacrimal gland and displacing it away from the orbital wall with its blood supply preserved. A periorbital incision is made at the location of the IOFB, as determined by neuronavigation. Instruments are introduced under endoscopic visualization, guided by navigation, fluoroscopy and/or ultrasound, as indicated. Although eLRC is routinely employed at our center for tumor resection or biopsy, it was utilized only twice in this clinical series for debridement of the lateral orbit in a patient injured by a disintegrating circular saw (supp. Video 4) and removal of metal rod.

Results

Approach selection was based on IOFB compartment (intraconal vs extraconal) and quadrant (Fig. 1 and Fig. 2, Table 1). In one case (#13) the foreign body was removed via entry wound.

Fig. 2.

Schematic sagittal representation of intraorbital compartments suitable for specific endoscopic approaches. No IFOB was noted in deep located zones (stripes) but technically these are achievable by eTCA or eSC. yellow – eTC (n = 2) or eSC (n = 1) for extraconal space, orange – eTC for intraconal space (n = 4), blue circle – globe, black interrupted lines – optic nerve, red intrupted line – extraocular muscles

Penetrating globe injury

In four patients, the globe was penetrated by the intraorbital foreign body (IOFB) along its trajectory. Two of these patients had previously undergone ophthalmological management that included vitrectomy and retained useful vision. In such cases, an endoscopic approach is preferred because it eliminates the need for direct spatula compression on the globe, thereby minimizing the risk of further compromising residual visual function.

Successful transnasal endoscopic removal was achieved in one patient with a posteriorly located IOFB. In the second case (IOFB adjacent to the globe and optic nerve), transconjunctival extraction was aborted after minimal globe compression elicited pupil dilation and deformity secondary to ciliary nerve irritation within the optic nerve sheath. The patient is under observation without evidence of inflammatory pseudotumor formation.

Two patients presented initially with acute penetrating globe injuries. First one with sustained gunshot wound (round bullet) resulting in profound visual loss and severe globe deformation. Extensive soft-tissue destruction precluded safe intraoperative localization of the metallic fragment. Planned vitrectomy was not performed because the patient was lost to follow-up. The second patient suffered a hammer-chisel injury in which a metallic fragment entered the globe inferiorly, just beneath the iris and exited posteriorly near the optic nerve insertion. The fragment, lodged in the posterior exit wound, was successfully extracted via a transconjunctival approach. The posterior scleral laceration was repaired with TachoSil, whereas the minimal anterior entry wound required no additional intervention.

Unsuccessful extraction

Notably, all three cases in which the IOFB could not be identified intraoperatively – and therefore remained in situ – involved non-ferromagnetic air-gun round pellets. The first was the previously described patient with associated globe penetration. The pellet was located posterior to the globe and was approached transconjunctivally but could not be visualized or safely retrieved. The second patient had a pellet situated intraconally at the orbital apex, between the optic nerve and the medial rectus muscle, with preserved vision. A transnasal endoscopic approach was attempted; however, the foreign body could not be adequately visualized, and further manipulation was avoided because of the high risk of optic nerve injury and permanent visual loss. The third patient was a child with an air-gun round pellet located intraconally in the superolateral orbital compartment. A superior eyelid crease approach was performed with intraoperative X-ray guidance, but the pellet could not be located despite extensive exploration. All three patients have been managed conservatively with close follow-up and have shown no evidence of delayed inflammatory complications to date.

Foreign bodies found within the extirpated inflammatory tumors

Endoscopy also enables simultaneous diagnostic and therapeutic intervention, allowing an initial diagnostic biopsy to be immediately converted into an extended/debulking procedure.

In three patients referred with a presumed orbital tumor, intraoperative extended biopsy unexpectedly revealed retained organic foreign bodies (wood) despite the patients’ strong denial of any prior trauma. In each case, surgery was confined to removal of the foreign material, with no attempt at gross-total resection of the presumed neoplastic/inflammatory mass. Follow-up imaging demonstrated marked and satisfactory regression of the residual lesion in all three patients.

Morbidity and complications

No procedure-related morbidity was observed beyond the primary injury. No vision-threatening complications or new neurological deficits occurred. All patients developed some degree of periorbital and eyelid edema related to the surgical approach and the use of self-retaining retractors or spatulas. This edema resolved spontaneously and uneventfully within 5—10 days. Cosmetic outcomes were assessed as satisfactory to good by all patients except one. In this case (rotary saw injury), the wound healed uneventfully; however, a persistent cosmetic defect remained due to the extensive soft-tissue loss at presentation.

Discussion

IOFBs that become embedded within the orbit pose a formidable surgical challenge. In our series we prioritized the most direct corridor that preserved a straight-line trajectory to the IOFB and minimized manipulation of the vital intraorbital structure. In practice, selection among the surgical approaches, techniques and instruments depends heavily on the operating surgeon’s knowledge, technical proficiency and cumulative experience. Equally important is the surgical team’s flexibility and readiness to transition effortlessly between conventional microsurgical and endoscopic methods as dictated by intraoperative situation.

The preferred surgical corridors also varies considerably across specialties. Traditionally, neurosurgeons have favoured transcranial approaches such as the Housepian transorbital-roof orbitotomy [12] or the classic Krönlein–Reese–Berke lateral orbitotomy [3] with temporary osteotomy of the zygomatic arch and lateral orbital rim. Over the past two decades the widespread adoption of endoscopic skull-base surgery by combined neurosurgery–otolaryngology teams has led to the development of several minimally invasive endoscopic transorbital and transnasal corridors.

Ophthalmologists and oculoplastic surgeons, by contrast, tend to employ familiar extraocular or anterior approaches, including microsurgical transconjunctival, precaruncular, lateral retrocantal, superior eyelid crease and preseptal trans-eyelid techniques.

A critical comparison of surgical approaches reveals that open craniotomy provides excellent exposure of the superior orbital compartment. However, it remains one of the most invasive and traumatising options, carrying substantial risks related to frontal lobe retraction, breach of the paranasal sinuses (with consequent sinusitis or mucocele formation) and cerebrospinal fluid leakage. Additional potential sequelae—although not exclusive to craniotomy—include loss of vision, impaired ocular motility with ptosis, pupillary dysfunction and a prominent, externally visible scar.

In our experience the superior orbital compartment can be accessed via a superior eyelid crease incision (SC) or a transconjunctival approach (TC), both followed by microsurgical dissection. These techniques are highly effective for anterior and mid-orbital lesions; however, they become increasingly challenging when the target lies in the superior apical orbit owing to restricted instrument angulation and limited depth of exposure.

The transconjunctival approach remains truly minimally invasive and facilitates rapid exposure, requiring only incision of the conjunctiva and Tenon capsule. To manage the orbital fat we prefer the use of cotton-tipped applicator, but also a malleable fixed retractors can be useful.

As with all orbital procedures, meticulous care must be taken to avoid injury to the ciliary ganglion, ciliary arteries and short/long ciliary nerves. Excessive manipulation of these structures can readily provoke mydriasis or a tonic (Adie-like) pupil. We strongly recommend the frequent monitoring of pupil size, shape and reactivity during the procedure. The surgeon must be ready to immediately withdraw of the instruments from the surgical field and wait until the shape or size of the pupil normalize. In that situation the surgery can be continued but with a greater care and delicacy during manipulations in the orbital compartment.

Although conjunctival incisions typically heal without visible scarring, postoperative retraction of the conjunctival flap may lead to granuloma formation or epithelial inclusion cysts [26]. Other rare reported complications of the transconjunctival (TC) approach include eyelid distortion, ectropion, entropion and haematoma formation [13]. In our practice we always try to adapt the conjunctiva with as few surgical stitches as possible to achieve complete coverage of the globe with the conjunctiva.

The superior eyelid crease approach is particularly well-suited for lesions in the superior, superomedial and superolateral orbit; however, it is considerably more labour-intensive than the transconjunctival approach. The skin incision is placed in the natural supratarsal skin crease, followed by dissection through the orbicularis oculi muscle in a preseptal plane down to the orbital septum and periosteum. Subperiosteal dissection along the orbital roof then permits instrument passage as far as the orbital apex. We prefer this approach especially to access to the deep intraconal or extraconal contents of the superior orbit. Great care must be taken during the approach to avoid injury to the levator palpebrae superioris aponeurosis which may result in postoperative upper eyelid ptosis [14].

Lateral-to-optic-nerve foreign bodies have traditionally been accessed via the classic Krönlein lateral orbitotomy, first described in 1889. This technique and its modifications remain in widespread use and continue to provide the most versatile exposure of the lateral orbital compartment. The standard Krönlein procedure typically requires lateral canthotomy and temporary removal of the lateral orbital rim, which can result in cosmetic sequelae such as canthal malposition, eyelid deformity, or visible scarring.

Our minimally invasive endoscopic modification – endoscopic lateral orbitotomy (ELO) or endoscopic lateral retrocantal orbitotomy (eLRC) – allows complete exploration of the entire lateral orbit, including the apex, while preserving the integrity of the lateral orbital rim. Reconstruction of the lateral orbital wall is unnecessary owing to its small size. Additionally, the temporalis muscle is not disturbed [18].

For foreign bodies confined to the anterior half of the orbit, microsurgical lateral retrocantal (LRC), transconjunctival (TC), or their endoscopic variants are often preferable. However, neurosurgeons are generally less familiar with LRC and TC approaches; in that situation we recommend the cooperation with an oculoplastic surgeon at least for the first few cases.

The classic Caldwell–Luc anterior maxillotomy, or the more extensive Denker approach, provides access to the inferior orbital compartment by removing the anterior wall of the maxillary sinus. Potential complications of these trans-antral corridors include chronic infraorbital nerve neuropathy manifesting as facial pain, numbness, dental hypoesthesia and persistent swelling [17]. Several authors emphasise that even the widest Caldwell–Luc bony window offers limited instrument angulation and working space for effective intraorbital manipulation and foreign-body retrieval [2].

Less invasive inferior approaches – such as the subciliary or transconjunctival preseptal (inferior fornix) – remain the standard for anteriorly located IOFBs. Subciliary (inferior eyelid) incisions, however, carry a risk of postoperative lower-eyelid malposition, including lid retraction with scleral show, rounding of the lateral canthal angle and ectropion [24]. When a transconjunctival inferior approach is employed, meticulous care must be taken to protect the cornea from abrasion during conjunctival retraction.

A further minimally invasive option is the transnasal endoscopic approach to the inferior orbit using a bimanual or four-handed technique. To facilitate this, we have developed a biportal strategy. We previously described two variants of a combined endoscopic transmaxillary approach (eTMA) that allow complete coronal-plane access to the entire orbital floor and inferior retrobulbar compartment [19].

In the first variant, instruments are introduced via a limited trephination in the canine fossa combined with a middle meatal antrostomy. The second variant employs a binostril trans-septal antrostomy, providing considerably greater surgical freedom and enabling a true four-handed technique. This configuration permits instruments to be inserted from multiple angles, thereby reaching the entire inferior retrobulbar space. The deliberately small anterior maxillotomy reduces the risk of dental or stomatological complications.

For medially located IOFBs, particularly those in the retrobulbar space, a transnasal endoscopic approach is highly suitable. This corridor provides sufficient exposure of the medial, superomedial and inferomedial orbital compartments, including the orbital apex and optic canal [1, 18]. Although it leaves no external scars, the intranasal dissection is relatively extensive, typically requiring middle meatal antrostomy, complete ethmoidectomy, wide sphenoidotomy and partial or complete middle turbinate resection. A trans-septal window is often added to facilitate bimanual or four-handed instrumentation. A practical consideration after transnasal surgeries is frequent postoperative laryngological follow-up to prevent the formation of the nasal synechiae—adhesions made of fibrous scar tissue between damaged mucosal surface.

We chose endoscopic transethmoidal approach for medially and retrobulbar-located IOFBs. Less invasive alternatives—precaruncular, transconjunctival and superior lid crease approaches—are generally preferable only for anteriorly situated foreign bodies. The endoscopic modification of the transconjunctival approach proved highly successful in our patients, yielding excellent functional and cosmetic outcomes with minimal morbidity.

Accurate identification of IOFBs remains challenging. The wide variation in material density (e.g., glass, plastic, wood, iron, stone, aluminium) combined with trauma-related orbital edema and haemorrhage frequently leads to misdiagnosis or underestimation of the injury extent. Even with contemporary multimodality imaging (CT, MRI, ultrasonography), radiologists often struggle to detect and correctly characterise organic foreign material, particularly wood, which may initially appear isodense to air or soft tissue and later mimic an abscess or inflammatory mass [23]. Computed tomography (CT) remains the gold-standard imaging modality, reliably detecting most IOFBs while remaining safe in the presence of metallic fragments. When wooden foreign bodies are suspected, MRI is recommended because hydrated (“green”) wood may appear isodense to air or soft tissue on CT, leading to misdiagnosis [16, 22]. Ultrasonography can be particularly valuable in such cases when CT and MRI findings are equivocal.

A separate issue is the use of intraoperative imaging. As a neurosurgical center, we rely primarily on neuronavigation, as it is the most intuitive tool in our daily practice. We currently use intraoperative ultrasound for aspiration of orbital abscesses and consider it an effective method, mainly because abscesses are readily visualized and the procedure can be performed percutaneously [21]. Reports on ultrasound-guided orbital surgery are increasing, particularly for needle biopsy of orbital tumors and percutaneous extraction of lesions with Hartmann forceps under ultrasound guidance [6, 10, 25]. Ultrasound has a clear advantage over neuronavigation because it provides real-time imaging. Displacement of a foreign body within the orbital fat compartment can rapidly make neuronavigation unreliable. With ongoing probe miniaturization, improved ultrasound resolution, and integration with neuronavigation systems—as currently seen, for example, in glioma surgery—ultrasound is likely to become an increasingly important component of the orbital surgeon’s armamentarium.

The composition of the foreign body is of crucial importance. All organic IOFBs must be removed expeditiously owing to the high risk of severe orbital inflammation, cellulitis, or abscess formation [4, 7, 9, 15]. Among inorganic materials, most are relatively well tolerated; however, copper-containing fragments provoke intense suppurative inflammation, iron may cause siderosis bulbi and lead carries the risk of systemic toxicity [4, 11, 15]. Metallic IOFBs composed of steel alloys are especially concerning because of potential siderosis and the lifelong contraindication to MRI.

As our results show small-calibre air-gun pellets (typically 4.5 mm BBs or similar spherical projectiles from air rifles/pistols) pose a particularly challenging subgroup of foreign bodies. Their retrieval is frequently difficult or impossible for several reasons: minute size, non-ferromagnetic composition (often copper-coated lead, pure copper, or even plastic/polymer) and lack of intrinsic contrast against surrounding orbital soft tissues. Intraoperatively, pellets may remain invisible within orbital fat tissue despite use of navigation, fluoroscopy and magnetic extraction.

In our experience the anteriorly located, easily palpable foreign bodies should always be removed. In contrast, there is a paucity of evidence-based guidelines for posteriorly (retrobulbar) located IOFBs. Management decisions hinge on balancing the risks of surgical intervention against the likelihood of late complications. Our consensus holds that well-tolerated, chemically inert metallic IOFBs in the posterior orbit—without evidence of inflammation, infection, optic neuropathy, or functional deficit—may be safely managed with close observation and regular follow-up imaging.

Most retained intraorbital metallic foreign bodies carry a favourable prognosis and do not adversely affect visual acuity when left in situ. In the series by Fulcher et al., 40 patients underwent surgical management of IOFBs: 22 metallic inorganic, 5 non-metallic inorganic and 13 organic. Six patients with posteriorly located inorganic foreign bodies were managed conservatively, with the authors proposing a treatment protocol advocating observation unless significant orbital complications arise [9].

A retrospective review by Finkelstein et al. over seven years identified 27 patients with exclusively metallic foreign bodies; 80% of posteriorly positioned IOFBs were left in place without intervention and remained clinically quiescent [8]. Similarly, Ho et al. reported a 6-year retrospective series of 43 patients with retained intraorbital metallic foreign bodies (duration of retention 6 months to 8 years), 37 of which were posterior to the globe. Among the 20 cases without initial ocular penetration, only two developed secondary complications requiring surgery. Late complications from retained metallic IOFBs were absent in 95% of patients with an intact globe [11]. In our series we also do not observe any late complications from retained metallic IOFBs.

Based on our experience we advocate prompt removal of all organic foreign bodies. For metallic IOFBs we favour removal when lesion is symptomatic, non-inert, MRI-incompatible or readily accessible via low morbidity corridor; otherwise observation is reasonable, particularly in the posterior orbit within orbital apex.

Conclusion

The endoscopic techniques represent a valuable alternative to traditional open approaches, especially for deep-seated lesions. The optimal approach should prioritize the most direct and shortest path to the target, avoiding critical vital structures such as the optic nerve, extraocular muscles, ciliary nerves and vascular elements to minimize the risk of permanent deficits.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors contributions

All authors contributed to the study. Conceptualization of the study, surgical strategy of the approaches and execution was performed by Tomasz Lyson, Andrzej Sieskiewicz and Robert Chrzanowski. Data collection and analysis were performed by Robert Chrzanowski, Magdalena Rybaczek, Tomasz Lyson and Andrzej Sieskiewicz. The first draft of the manuscript was written by Robert Chrzanowski, Magdalena Rybaczek, and Zenon Mariak. All authors commented on previous versions of the manuscript. Zenon Mariak and Tomasz Lyson supervised the project. The figures was designed and created by Robert Chrzanowski. All authors read and approved the final manuscript.

Funding

No funding was received to assist with the preparation of this study and manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Human ethics and consent to participate

Human Ethics and Consent to Participate declarations: not applicable. This study is a retrospective analysis of existing data of standard surgical treatment and contains no patient-identifiable information; therefore, separate local bioethics committee approval was not applicable.

Competing interests

The authors declare no competing interests.