Retrobulbar haemorrhage is a sight-threatening condition that can occur after orbital trauma. The aim of this study was to evaluate the frequency and outcome of orbital haemorrhages following orbital fractures in geriatric patients receiving anticoagulants. All patients aged 65 years or more suffering from orbital fractures between 2008 and 2009 were included in this study. The mechanism of trauma, underlying diseases, and medication were recorded. In case of a retrobulbar haemorrhage, surgical exploration, the elapsed time between the onset of haemorrhage symptoms and surgical treatment, and the outcome regarding visual acuity were documented. Sixty-eight orbital fractures occurred (31 males, 37 females, age 65–95 years) resulting in six (3%) orbital haemorrhages. Four cases were associated with initial orbital bleeding, two other patients developed orbital haemorrhage as a complication after surgical reconstruction. Anticoagulant therapy, but not aspirin, was associated with a significantly increased risk of retrobulbar haematoma ( p = 0.02). Two patients permanently lost vision, two partial recoveries and two total recoveries were observed. Patients receiving anticoagulants have a higher risk of orbital haemorrhage after orbital fracture and should be monitored closely. Any evidence of visual impairment should lead to further investigation and prompt treatment.
Craniofacial injuries occur in up to 75% of all accidents, with a male-to-female ratio of 2:1. Orbital fractures are common, and the orbit may be involved in nearly half of all facial fractures. An orbital haemorrhage occurs in only 0.6% of orbital fractures and thus represents a rare but severe traumatic complication of orbital fractures because of the risk of vision loss due to compression of the optic nerve. The risk of severe visual impairment in all facial fractures has been suggested to be 0.8–4.5%, increasing to as high as 30% in orbital blowout fractures.
Most injuries occur in childhood or adolescence and are typically caused by road traffic accidents. At ages over 65 years, the incidence of injuries of the head and face escalates due to an increased risk of falls. Demographic data from Western societies demonstrate an increasing life expectancy, resulting in a remarkable increase in the size of the population over 65 years of age.
Craniomaxillofacial surgeons are confronted with an increasing number of geriatric patients with craniofacial injuries that require special treatment and care due to age-associated diseases such as stroke or dementia, multimorbidity and multiple medications. In the elderly, the incidence of cardiovascular (40/1000) and cerebrovascular (14/1000) events is escalating. Hence, patients over 65 years of age often receive antiplatelet or anticoagulant drugs that result in impaired blood clotting. For patients receiving enoxaparin, orbital haemorrhage is a known complication of midfacial fractures. Warfarin increases the risk of major haemorrhage, as does aspirin, albeit to a minor degree. Both anticoagulant and antiplatelet therapies are likely to enhance the risk of traumatic haemorrhage.
The aim of the present study was to evaluate the incidence and outcome of retrobulbar haemorrhage in elderly patients with orbital fractures receiving oral anticoagulant or antiplatelet therapy.
Patients and methods
This interdisciplinary study from the departments of ophthalmology and maxillofacial surgery included all patients aged 65 years or older with radiologically proven orbital fractures during the years 2008 and 2009 and was approved by the local ethical committee (No. 3875-10).
Overall, 251 patients (68% males, 32% females) with isolated orbital fractures or facial fractures that affected the orbit were treated in the Departments of Oral and Maxillofacial Surgery and Ophthalmology at the Ruhr University Bochum. Of these 251 patients, 68 patients were 65 years old or older (range 65–98 years), including 31 (46%) males and 37 (54%) females. The following parameters were recorded: type of injury, date and mechanism of trauma, underlying diseases, medication and laboratory values related to coagulation (routine tests included platelet count, partial thromboplastin time (PTT) and international normalized ratio (INR)), clinical symptoms and occurrence of an orbital haemorrhage. Orbital fractures were diagnosed by computed tomography (CT) scans in all patients amended by magnetic resonance imaging (MRI) scans if patients complained of visual impairment. Orbital haemorrhage was defined by MRI and the presence of the haematoma during surgical evacuation. If a retrobulbar haemorrhage was diagnosed, the elapsed time between the appearance of the first symptoms of haemorrhage and surgical treatment and the outcome regarding visual acuity were documented.
Surgical repair was performed using an infraorbital skin incision and a dissection towards the infraorbital rim. After visualizing the fracture, the herniated orbital contents were dissected gently and reduced into the orbit. To prevent postoperative herniation and entrapment, the fractured orbital floor was patched (PDS-Foil ® or Ethisorb-Patch ® ). Forced duction tests were performed at the end of surgery.
In case of retrobulbar haematoma, the same infraorbital approach was used. A blunt dissection into the retrobulbar region was performed, and the haematoma was removed. After drainage a silicon drain was inserted for 3 days. High-dose steroids and intravenous antibiotics were administered. Possible recurrence of the haematoma was monitored by MRI scans.
Following surgery, enoxaparin (40 mg) was administered once daily as thrombosis prophylaxis or two times daily as a substitute for phenprocoumon in patients undergoing anticoagulant treatment.
Ophthalmological examinations were performed immediately in case of any complaints regarding vision including diplopia, blurred vision, loss of sharpness of vision or visual loss. Preoperatively and on the day after the surgical treatment, all patients underwent a full ophthalmological examination including manifest refraction, Snellen best spectacle-corrected visual acuity (CDVA), anterior segment and posterior segment examination, measurement of the intraocular pressure (Goldman contact tonometry, Haag-Streit, Switzerland) and visual field testing (Octopus 101, Interzeag, Haag-Streit, Switzerland). The mean follow-up period was 14 months.
Statistical analysis was performed using Fisher’s exact test (SPSS 17 for Windows, SPSS Inc., Chicago, USA). Significance was assumed at p < 0.05.
Orbital fractures in the 68 patients aged 65 years or older were caused by falls in 95% of cases. Reasons for the falls varied widely: some patients had tripped or slipped, and other falls resulted from medical conditions such as syncope or vertigo. Two other orbital fractures resulted from road traffic accidents. One patient was injured while participating in sports ( Fig. 1 ). The use of an antiplatelet or anticoagulant treatment was reported by 32 (47%) of the geriatric patients, including aspirin in 20 patients, phenprocoumon in nine patients and clopidogrel in three patients ( Table 1 ).
|Medication||n||Orbital haemorrhage||Relative risk (%)|
Phenprocoumon was administered for atrial fibrillation in six cases, coronary artery disease in two cases, and pulmonary embolism in one case. Mean INR in this patient group was 2.11 (1.8–3.0).
A retrobulbar haematoma occurred in six (8.8%) of 68 patients; five women and one man were affected. Detailed information regarding age, gender, diagnosis, medication and INR is given in Table 2 . Three of these six patients were treated with phenprocoumon ( Fig. 2 ), two with aspirin and one patient with no medication that alters blood clotting. Anticoagulant therapy was associated with a significantly increased risk of traumatic retrobulbar haematoma ( p = 0.02). INR values in patients with a retrobulbar haematoma and phenprocoumon treatment did not differ significantly from INR values in patients taking phenprocoumon without developing a retrobulbar haematoma (1.97 vs. 2.18).
Aspirin showed no significant effect on the risk of traumatic orbital haemorrhage ( Table 1 ).
Blurred vision and impaired visual acuity were described as the first symptoms by all patients. A painfully tight orbit was observed. Immediate ophthalmological examination revealed a decrease in the mean visual acuity to 20/400 (no light perception to 20/160) in all patients’ eyes immediately after the appearance of the retrobulbar haematoma. MRI ( Fig. 3 ) verified the presence of an orbital haemorrhage; verification was followed by immediate surgical decompression of the orbit and the optic nerve. The time between onset of visual impairment and surgical decompression ranged from 51 min to 2:36 h ( Table 3 ).
|No.||Onset of haemorrhage||Duration of surgical fracture treatment||Latency onset – treatment||Outcome|
|3||8 Days after trauma, 8 days after surgery||85 min||2:26||Impaired vision|
|5||5 Days after trauma, 2 days after surgery||36 min||1:48||Vision loss|
|6||Immediate||n/a||2:06||Vision loss (able to distinguish light from dark)|