Article Text
PDF
Abstract
A female patient with type 2 diabetes in her 50s presented to casualty with a 1-day history of red, painful right eye. Visual acuity (VA) bilaterally was 6/12, but a right anterior uveitis was noted, with hazy fundal view. She was discharged on topical steroid and mydriatic drops with a 2-day follow-up. VA remained unchanged, but she developed right proptosis, restricted eye movements, lid swelling, relative afferent pupillary defect and an intraocular pressure (IOP) of 39 mm Hg. She was admitted and treated with intravenous and intravitreal antibiotics, intravenous antifungals and IOP-lowering drugs. Blood tests showed raised inflammatory markers and an HbA1c of 127 mmol/mol. Over her admission, right eye vision deteriorated to no light perception. A B-scan ultrasound revealed panophthalmitis and a retinal abscess. All investigations looking for a source were negative. Inflammatory markers settled, but despite aggressive treatment, the panophthalmitis did not improve. She was discharged with a follow-up to consider enucleation.
- Ophthalmology
- Medical management
Statistics from Altmetric.com
Background
Endophthalmitis is a severe inflammatory process of the intraocular cavity and its contents, without involving the sclera or orbital contents. It is an ophthalmic emergency as inadequate or untimely management could lead to irreversible visual loss.
The cause is usually infective in nature, and depending on the mode of transmission, endophthalmitis can be either exogenous or endogenous. The more common exogenous endophthalmitis occurs due to direct inoculation of infective organisms from the environment, most commonly after ocular surgery or trauma. Comparatively, endogenous endophthalmitis occurs due to haematogenous spread of organisms from an infective focus.1 It is responsible for 2%–6% of endophthalmitis cases2 and is often associated with pre-existing medical comorbidities such as diabetes, HIV, systemic immunosuppressive treatment, intravenous drug abuse and malignancy. Interestingly, the right eye seems to be twice more commonly affected than the left due to more direct blood flow to the right carotid.3 The the most common infective foci reported for endogenous endophthalmitis are liver abscesses, followed by lung infection, endocarditis, soft tissue infection, urinary tract infection, meningitis, septic arthritis and orbital cellulitis.4
Panophthalmitis is a more severe inflammatory process involving all layers of the eye, extending to the orbit and periocular structures such as the eyelids. Similar to endophthalmitis, it presents with decreased vision, red eye, pain, photophobia, hypopyon and limited fundal view. However, due to the extent of inflammation, there is also usually proptosis, eyelid swelling, limitation of ocular movements and very high intraocular pressure (IOP). It is rapidly progressive and has a poor prognosis, with most cases deteriorating despite timely and effective management.5
We present a rare case of sterile endogenous panophthalmitis in a female patient in her late 50s, treated at a tertiary eye hospital.
Case presentation
A female patient in her late 50s presented to eye casualty with a 1-day history of right eye pain and redness, with occasional floaters. She denied fever or any systemic symptoms and had not travelled in the preceding 20 years. Her medical history included long-standing back pain, type 2 diabetes mellitus, hypertension, depression and a high body mass index. She had no past ophthalmic history. Her regular medications included insulin (insulin glargine and insulin aspart), venlafaxine and ramipril.
On examination, her best-corrected visual acuity (BCVA) was 6/12 bilaterally. There was a mild right ptosis. The right eye conjunctiva was hyperaemic, and there was significant inflammation in the right anterior chamber with 4+ cells, flare, a 1 mm hypopyon and fibrin. The IOP was 10 mm Hg on the right and 17 mm Hg on the left. While left fundal examination was normal, right fundal view was hazy due to the anterior chamber inflammation. A first episode of right-sided severe anterior uveitis was presumed, and full blood count, C reactive protein (CRP), erythrocyte sedimentation rate (ESR), urea and electrolytes and HbA1c were taken. A chest X-ray (CXR) was also performed. She was started on dexamethasone 0.1% preservative-free eye drops every hour along with cyclopentolate 1% eye drops three times a day in the right eye. She was then discharged with a 2-day follow-up, with a plan to monitor the treatment response and to examine the fundus when the view was clearer.
The blood tests taken revealed raised inflammatory markers (CRP 181 mg/L, ESR 116 mm/hour), neutrophilia (12×109/L) and microcytic anaemia (haemoglobin 80 g/L). The HbA1c was 138 mmol/mol. CXR was normal. Two days later, when she returned for follow-up, the right BCVA had deteriorated to 6/60. The right upper lid was swollen and completely ptotic; the eye was proptosed and eye movements were restricted. There was a relative afferent pupillary defect (RAPD), conjunctival chemosis, corneal haze, a 3 mm hypopyon and an IOP of 39 mm Hg (14 mm Hg on the left) (figure 1). Once again, no view of the right fundus could be obtained. She was admitted with a working diagnosis of anterior uveitis complicated by orbital cellulitis. She was started on intravenous ceftriaxone 2 g two times per day, intravenous metronidazole 500 mg three times per day and intravenous fluconazole 200 mg once a day for possible Candida infection. Oral acetazolamide 250 mg two times per day was also started, along with topical dexamethasone every hour day and night and topical dorzolamide–timolol 2.0%–0.5% two times per day, both to the right eye. The cyclopentolate 1% was continued at three times a day to the right eye.
Appearance of the right eye, 2 days after initial presentation. The upper lid is swollen and completely ptotic (pulled up for the photo); the conjunctiva is hyperaemic and chemotic; the corneal is hazy, and there is a visible hypopyon.
Investigations
During admission, a number of investigations were performed. Blood cultures were taken, which were negative. An urgent CT scan of the head and orbit with contrast revealed a thickened right lacrimal gland and leptomeningeal–pachymeningeal enhancement at the anteroinferior right temporal lobe, posterior to the right orbit. There were no distinct features of orbital cellulitis or abscess formation. A right vitreous aspirate was done, yielding yellow vitreous, which was sent for investigation. Intravitreal ceftazidime 2 mg in 0.1 mL and vancomycin 1000 μg in 0.1 mL were administered. A B-scan ultrasound (US) showed extensive vitreous opacities, choroidal and posterior scleral thickening, and a retinal abscess in an otherwise flat retina. There was also marked fluid distention of Tenon’s space, orbital fat hyperdensity and muscle hypertrophy (figure 2). Contrast-enhanced MRI of the brain and orbit showed right preseptal soft tissue swelling, enhancement thickening of the sclera and anterior chamber, and mild inflammatory change in the retro-orbital fat (figure 3). These findings were all in keeping with a diagnosis of panophthalmitis and orbital cellulitis.
B-scan ultrasound of the right eye, performed during the early days of her admission. There are extensive vitreous opacities, choroidal and posterior scleral thickening, and a retinal abscess in an otherwise flat retina. There is also marked fluid distention of Tenon’s space, orbital fat hyperdensity and muscle hypertrophy.
MRI of the brain and orbit, showing right-sided proptosis, preseptal soft tissue swelling and mild inflammatory change in the retro-orbital fat. There is also enhancement and thickening of the sclera and enhancement of the anterior chamber of the right globe.
The pain slowly improved; however, 3 days into admission, the visual acuity (VA) dropped to no perception of light (NPL). The hypopyon had increased (4 mm height) and was accompanied by fibrin and anterior chamber blood clots. The proptosis, upper lid oedema, RAPD, conjunctival chemosis and raised IOP remained unchanged. Unfortunately, she became increasingly unwell, and despite antimicrobial treatment, inflammatory markers were rising (CRP 180 mg/L, white cell count 25×109/L, neutrophils 21.94×109/L and platelets 727×109/L). She was also noted to have relatively low blood pressure, with readings consistently below 130/80 mm Hg, despite her background of hypertension and the omission of her antihypertensives during admission. Furthermore, despite management, her blood glucose was consistently above 25 mmol/L. To complicate matters, cannulation was difficult with multiple unsuccessful US-guided attempts. Due to the rising risk of septic shock, she was transferred to the medical ward.
Throughout her medical admission, she was investigated for a source. Systemic examination was completely normal, with no peripheral stigmata of endocarditis, infective processes or other systemic disorders. Investigations looking for systemic diseases or autoimmune causes were negative, with a normal ACE (14 U/L) and antineutrophil cytoplasmic antibody test (myeloperoxidase antibodies <0.2 IU/L and proteinaise 3 antibodies <0.6 IU/L). In the search for an infective source, blood and urine cultures were repeatedly taken but were all negative, along with the culture and PCR analysis of the vitreous fluid taken on admission. HIV and QuantiFERON tests were also negative, along with syphilis antibodies, syphilis rapid plasma reagin and Treponema pallidum particle agglutination assay. An echocardiogram showed no obvious cardiac sources of septic emboli. Furthermore, a CT scan of the thorax abdomen and pelvis (CT-TAP) did not reveal any infective or malignant foci.
Seven days into admission, she started to systemically improve, with a down-trending CRP (65 mg/L). The right-sided hypopyon was now just a trace. The VA was still NPL, and the RAPD, right upper lid swelling, proptosis, eye chemosis and hyperaemia were all unchanged. The right eye IOP was now 20 mm Hg. Her blood glucose levels were optimised by the medical team through insulin modification. The fluconazole was stopped; her intravenous antibiotics were switched to oral co-amoxiclav 625 mg three times a day; and the topical dexamethasone was decreased to six times a day to the right eye.
No source had yet been identified and a beta-D-glycan test to assess for candidaemia or other fungal infections was performed, which ultimately came back negative. US B-scan was repeated, which showed no interval change. On day 14, she was transferred back to the eye ward; the acetazolamide (IOP was now 14 mm Hg in the right eye) and the antibiotics were stopped, and topical dexamethasone taper was started (decreased from six times a day to four times a day). By this time, her eyelid swelling had improved, and she was now able to open her eye. Otherwise, all the other clinical findings remained unchanged.
A positron emission tomography–CT (PET-CT) scan to look for any possible occult cause of the panophthalmitis did not show any septic or malignant sources. Throughout her admission, no fever was recorded, and the left eye remained unaffected.
Outcome and follow-up
A final diagnosis of right sterile endogenous panophthalmitis, most likely secondary to uncontrolled diabetes, was made, and she was referred to oculoplastics for consideration of evisceration or enucleation. She was discharged home to await follow-up on dexamethasone 0.1% preservative-free two times per day, dorzolamide–timolol 2%–0.5% two times per day and latanoprost once a day, all to the right eye. A uveitis follow-up for medication tapering was also arranged.
Discussion
This case is one of the first reported cases of sterile endogenous panophthalmitis.
As mentioned previously, endogenous endophthalmitis accounts for a very small proportion of endophthalmitis cases, with endogenous panophthalmitis accounting for an even lower proportion. Sterile endogenous panophthalmitis, on the other hand, is extremely rare, with no incidence available in the literature. The only other case of sterile endogenous panophthalmitis was reported by Puskher et al, which was found to be secondary to an intraocular medulloepithelioma.6
In most reported cases of endogenous endophthalmitis, a bacterial or fungal organism is found. In Europe and North America, Staphylococcus aureus and Streptococcus pneumoniae are the the most common bacterial causes, compared with Klebsiella pneumoniae in East Asia.4 With regard to fungal causes, Candida albicans and Aspergillus are the the most common.7 In our patient, no organism was cultured from vitreous fluid, and multiple blood cultures were negative. In one of the largest endogenous endophthalmitis systematic review, negative blood cultures were reported in 44% of cases, with 41% having negative vitreous samples. However, in all patients, a causative organism was identified from one of these samples or from diagnostic PCR.4 In our patient, investigations looking for a septic source or an underlying predisposition were all negative, including PCR analysis of a vitreous sample.
A rare, yet important, clinical entity that has to be kept in mind in cases of apparent sterile ocular inflammation is masquerade syndrome, secondary to an underlying occult disorder, usually neoplasm. In our case, all imaging done, including CT-TAP and PET-CT, did not show any sign of malignancy.
Because endogenous panophthalmitis is extremely rare, there are no consensus guidelines about what treatment is most appropriate. Aggressive management based on the treatment of endogenous endophthalmitis should be employed, with some important considerations. In panophthalmitis, there is a much higher risk that any underlying infection could spread to the rest of the orbit and periorbital structures, causing orbital cellulitis. Furthermore, it can easily spread to nearby structures beyond the orbit, causing severe infections such as intracranial sepsis. Systemic spread leading to sepsis and shock is also a risk. Therefore, panophthalmitis is a serious infection that could potentially threaten both sight and life. Systemic and intravitreal antibiotics, with or without systemic antifungals, should be initiated very early on. Due to the high risk of spread, close monitoring is essential. Early multidisciplinary involvement is of the utmost importance, to ensure that the patient is investigated thoroughly for a septic source. In the very rare cases where no infective source is identified, such as in our case, aggressive management with antimicrobials is still highly recommended. One must remember that there could be an underlying infection that has not yet been diagnosed.
It is also important to remember that endophthalmitis can easily evolve into panophthalmitis. Therefore, in every case of endophthalmitis, it is important to perform a full ocular examination as a baseline, including a detailed assessment of the periorbital structures. The eye movements must be documented, alongside the eyelid position and function, degree of proptosis, and optic nerve function. This full assessment needs to be regularly updated so that if the periorbital structures become involved, it is picked up and managed early.
As mentioned previously, panophthalmitis cases need to be investigated early on to look for the septic source, with an aim to eliminate the focus of infection. 18F-fluorodeoxyglucose (FDG) PET-CT is emerging as a key player in the diagnostic work-up of infectious and inflammatory conditions. In infectious processes, inflammatory cells are recruited to the site of infection and large amounts of glucose are consumed. This means that infective foci are often visible on FDG PET-CT even before anatomical changes occur, giving it a good diagnostic role in unidentified metastatic infectious foci.8 9 Other uncommon investigations, such as the β-D-glucan test, which is useful to look for invasive fungal infections, need to be considered in these cases.10
In endophthalmitis, pars plana vitrectomy (PPV) can be considered in severe cases, as apart from removing infected material and endotoxins, it provides a diagnostic sample and allows better antibiotic diffusion in the vitreous cavity. However, its therapeutic use in panophthalmitis is limited. Since the infection is not restricted to the vitreous, doing a vitrectomy would leave behind a significant amount of infected tissue. Severe lid oedema and swelling of the orbital tissues might make establishing the three ports for PPV technically difficult. Furthermore, most of these patients would not be eligible for vitrectomy as they would be unwell with multiple comorbidities. Therefore, undergoing a vitrectomy in panophthalmitis is not recommended.
Enucleation is probably the most definitive surgical option in panophthalmitis, as it removes most of the infected material and endotoxins. Even in endophthalmitis, most patients ultimately require enucleation or evisceration.4 5 In fact, endogenous endophthalmitis has been shown to be an independent risk factor for enucleation or evisceration.11
Recently, a combination of multiple peribulbar and intravitreal injections of antibiotics and dexamethasone has been proposed as a possible treatment regime to prevent eyeball extraction in endogenous bacterial panophthalmitis.12 However, larger randomised control trials are required to assess the efficacy.
The use of intravitreal steroids in endogenous endophthalmitis is also controversial with multiple studies reporting conflicting results. In fact, the evidence for their use is limited, and these should be used with caution.13 14
In our patient, the blood glucose was severely uncontrolled, and it is likely that it contributed to the disease pathology. In fact, diabetes is the the most common risk factor for endogenous endophthalmitis, and animal studies show that it increases the blood–retina barrier permeability, creating a conductive environment for endogenous endophthalmitis.15 This highlights the importance of strict glucose control and early involvement of the ophthalmology team in patients with diabetes with ocular complaints.
It is evident that endogenous panophthalmitis presents a diagnostic and therapeutic challenge to clinicians. From this case, it is clear that early multidisciplinary involvement with aggressive investigation and treatment is essential. In cases were cultures are negative, we still advise clinicians to treat with systemic and intravitreal antimicrobials due to the risk of undiagnosed infections. Regular assessment and close monitoring is essential due to the risk of progression. Finally, a low threshold for invasive procedures such as enucleation is advised due to the high mortality rate of endogenous panophthalmitis.
Patient’s perspective
I still remember how this all started. That week I felt really light headed and I thought it was because I had changed my blood pressure tablets. On Tuesday I noticed my eye was swelling up and it was hurting; my daughter told me that my eye was red. I went to casualty, and I was initially prescribed two different drops. Over the following days I realised that my vision was getting worse, but I continued putting the drops in. When I went back 2 days later I was admitted to the eye ward. It was a bit of a shock for me to notice that my sight had gone. I had different treatments, intravenous antibiotics, numerous blood tests. My thoughts were all over the place. It was a very challenging and scary time for me. I was thinking about what was going to happen to me and how I was going to manage. But my main concern was my other eye. And I really wanted to know what was going on; what had happened, and whether I had done something wrong medically. I kept questioning myself about whether I had eaten something wrong or whether I’m allergic to something. There was not a moment, especially at night where I did not question myself about something regarding my eyes. I kept covering my eyes every day to see whether the vision was going to come back. And the fear that was there, it was like somebody had died; it was like a bereavement. It is still a bit scary now, as even though I am grateful that I have still got the left eye, I am concerned that something will happen to it.
The treatment was quite challenging. The hourly drops were mentally traumatic and the eye injection was insanely painful; the pain was worse than labour. Different scans and tests were done, and nobody could come up with an answer as to what was the root of the problem. This made me question everything. It also shows that doctors do not know everything. The doctors did everything, and they tried their best. But at the end of the day, doctors are human.
I think the scariest one was when I went to the theatre, they were trying to put a line in, and it would not go in. It was surreal as I was having electric shocks in my hand and could not cope after that. I was also worrying about not being able to get blood tests done to in order for me to have the correct antibiotics. It was a very scary, challenging time. I must admit that month I felt like I was on another planet, not in hospital.
During the time I spent at hospital I kept thinking about my children. I was missing stressing about them. I knew they were alright, but I still missed them and my cat. My colleagues at work were worried about me, and I was worried about work. Financial worries…everything just came to my mind, especially at night - I was just getting depressed. It did not help my anxiety, and I found that I had become more worried, more frightened, more scared. But still grateful, that I have got the one eye and that I’m still alive.
It was nice when I was told I was going home at first, but then I became frightened. I was afraid to leave, because I felt safe in hospital. But then I had to see how I was going to adapt to life and to try to be as normal as possible, get back to work and see how I get on. I just had to be confident and build myself back up.
At the moment, I’m not too bad during the day. But it is scary at night when I’m walking home, as I have to constantly turn around on my right side to check who’s behind me. In fact, I am tending to take the taxi from wherever I am at night since I feel safer.
It’s scary that I now suddenly have just one functioning eye. But there is nothing I can do about it. I have to just try and be positive, go forward with my life and try to do as much as I can to look after my right eye.
I just want to say, that out of everything in your body, your eyes are the most precious gems. I think sometimes we take things for granted. You really have to look after your eyes, because when you cannot see, it’s frightening. You can replace arms and limbs, probably lungs. You can live without one kidney, you can replace the heart, but you cannot replace your eyes. And it’s very important to be appreciative of what you can see, and be grateful.
I’m just grateful for the doctors and nurses that helped me, it’s the scariest thing I have ever been through, I hope I do not have to go through anything like this again.
Learning points
Endogenous panophthalmitis is rapidly progressive, and is a risk to both sight and life. Regular examination and close monitoring are essential due to the risk of progression and high mortality rate, even in cases of endophthalmitis (due to the risk of progressing to panophthalmitis).
Early multidisciplinary involvement with aggressive investigation and treatment is crucial. In cases were cultures are negative, we still advise clinicians to treat with systemic and intravitreal antimicrobials, due to the risk of undiagnosed infections.
Uncontrolled diabetes is more dangerous than commonly perceived and early involvement of the ophthalmology team in diabetic patients with ocular complaints is of utmost importance.
A low threshold for invasive procedures such as enucleation is advised due to the high mortality rate of endogenous panophthalmitis.
Ethics statements
Patient consent for publication
References
Footnotes
Twitter @MAzzopardi_
Contributors MA and YJC were involved in the direct clinical care of the patient. MA and BN were responsible for drafting the manuscript and conducting the literature search. YJC was responsible for the critical review of manuscript. All authors were involved in the submission process and final approval.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.