|Year : 2003 | Volume
| Issue : 4 | Page : 335-340
Endophthalmitis caused by acinetobacter calcoaceticus.A profile
L Gopal, Anand A Ramaswamy, Hajib N Madhavan, Rajani R Battu, T Sharma, Mahesh P Shanmugam, Pramod S Bhende, M Bhende, D Ratra, Nitin S Shetty, Madhav K Rao
Vitreo Retinal Services, Sankara Nethralaya, Chennai, India
Vitreo Retinal Services, Sankara Nethralaya, Chennai
Source of Support: None, Conflict of Interest: None
Purpose: To report the clinical and microbiological profile of endophthalmitis caused by Acinetobacter calcoaceticus .
Methods: A retrospective study of case series of Acinetobacter calcoaceticus endophthalmitis. Outcome measures included ability to sterilise the eye, anatomical result (clear media and attached retina) and visual recovery (visual acuity > 6/60).
Results: Of the 20 cases studied, 10 were cases of postoperative endophthalmitis, 3 were posttraumatic, 6 were endogenous and one was bleb-related endophthalmitis. Specific features of interest observed were relative chronicity of presentation and absence of any obvious predisposing factor in endogenous endophthalmitis cases. All cases could be sterilised except one, which needed evisceration. Cases with postoperative endophthalmitis had better anatomical outcome (7/10 with attached retina and clear media) and visual outcome (4/10 regained vision > 6/18). Higher smear positivity was seen in vitreous samples (72.2%) compared to aqueous samples (37.5%). Culture positivity was higher from the vitreous cavity compared to aqueous. The organism was sensitive to ciprofloxacin in a high percentage (88.9%) of cases.
Conclusions: Visual recovery in Acinetobacter calcoaceticus endophthalmitis is modest. Ciprofloxacin is the antibiotic of choice
Keywords: Endophthalmitis, Acinetobacter calcoaceticus
|How to cite this article:|
Gopal L, Ramaswamy AA, Madhavan HN, Battu RR, Sharma T, Shanmugam MP, Bhende PS, Bhende M, Ratra D, Shetty NS, Rao MK. Endophthalmitis caused by acinetobacter calcoaceticus.A profile. Indian J Ophthalmol 2003;51:335-40
|How to cite this URL:|
Gopal L, Ramaswamy AA, Madhavan HN, Battu RR, Sharma T, Shanmugam MP, Bhende PS, Bhende M, Ratra D, Shetty NS, Rao MK. Endophthalmitis caused by acinetobacter calcoaceticus.A profile. Indian J Ophthalmol [serial online] 2003 [cited 2019 Oct 22];51:335-40. Available from: http://www.ijo.in/text.asp?2003/51/4/335/14651
Endophthalmitis is a devastating condition. The presentation, virulence of the infection and the treatment outcome depend to a certain extent upon the circumstances of infection and organismal load but to a great degree on the type of infecting organism. Acinetobacter calcoaceticus is a relatively rare cause of endophthalmitis. This is a gram-negative bacillus belonging to the family Neisseriae and is considered an ubiquitous saprophyte. No large series of endophthalmitis caused by this organism has been reported. In this article, we present a series of 20 eyes of 20 patients in whom this organism was identified as the cause of endophthalmitis. The pattern of clinical presentation and the bacteriological aspects of the infection are discussed.
| Materials and Methods|| |
The clinical and microbiological records at a tertiary eye care centre were reviewed retrospectively for a period of 10 years (1991-2001) and patients with a clinical diagnosis of endophthalmitis in whom Acinetobacter calcoaceticus was isolated from intraocular specimens were studied. A detailed proforma was completed for each chart, comprising details such as age, gender, eye involved, type of endophthalmitis (postoperative, posttraumatic, endogenous, and bleb related), type of presentation (acute or chronic), visual acuity at presentation, anterior segment and posterior segment features, intraocular pressure (IOP), interval between the event (surgery or trauma) and symptoms, interval between onset of symptoms and presentation to us, surgical treatment adopted, etc. For the purpose of this study, we labelled the infection 'acute' if the presentation was within 1-2 days of injury or surgery (where applicable), and when the manifestation was associated with a rapid turn of events characterised by pain and other signs of acute infection. Cases with long-standing problems with no pain were categorised as 'chronic'. The treatment approach and outcome of treatment with respect to the control of infection, status of retina and visual recovery were recorded. In addition, the microbiologic aspects were also recorded.
The microbiological techniques used for isolation of the organism from the intraocular specimen were as described previously. Uncontaminated vitreous was aspirated by a syringe connected to the suction port of the vitreous cutter at the beginning of the vitrectomy. A sterile disposable needle was fixed to the syringe. Aqueous humour samples were collected aseptically in a tuberculin syringe with a 30G needle. After collection of the specimen, the air in the syringe was expelled carefully and the needle was stuck into a sterile rubber bung and sent to laboratory immediately. Three to four smears were made from each specimen using the cytospin (cytospin2, Shandon, UK). The smears were stained by Gram's method for bacteria and KOH- Calcoflour method for fungus. In two eyes, the inoculation was done only in liquid media - brain heart infusion broth (BHIB) and Robertson's cooked meat broth (RCM broth) since these surgeries were performed late in the night. In all the rest, inoculation was done in BHIB, RCM broth, MacConkey's medium (MAC), Blood agar (BA), and Chocolate agar (CA). The identification of Acinetobacter calcoaceticus was done by standard methods. It was identified by its characteristic well grown, grey-white colonies 2-3mm in size 18-24 hours after incubation. It was further identified as Acinetobacter calcoaceticus based on the following characteristics. These were gram negative, catalase positive, oxidase negative, non-motile, glucose non-fermenting bacilli that grew on MacConkey agar, showed negative fermentation reactions with mannitol, fructose, maltose, and sucrose and negative reactions with Simmons citrate agar, Christensen urea, nitrate reduction and indole. The antibiotic susceptibility patterns were assessed by the Kirby-Bauer disc diffusion technique.
| Results|| |
In this retrospective study, there were 20 eyes of 20 patients (15 males and 5 females). The average age of the subjects was 42.8 years (range 7-77 years) and the right eye was affected in 8 patients. The circumstances leading to the occurrence of endophthalmitis have been presented in [Table - 1]. The clinical details have been presented in [Table - 2]. Most frequently endophthalmitis occurred following intra ocular surgery, mostly after cataract surgery. Diabetes was present in 5 of 10 postoperative endophthalmitis patients. Out of the six cases of endogenous endophthalmitis, one manifested itself as bilateral neuro retinitis to the referral doctor. When this case presented to us, one eye was phthisical and hence intraocular specimens of only one eye were processed. Apparently, none of the cases with endogenous endophthalmitis patients had any identifiable predisposition such as HIV infection, intravenous drug use, parenteral administration of fluids or symptoms indicative of sepsis in the body such as urinary tract infection etc. The fellow eye was totally normal in 5 cases of endogenous endophthalmitis.
Although 11 of 20 eyes had acute onset of the symptoms, the subsequent course of the disease had been one of chronicity. Four of the 6 cases of endogenous endophthalmitis were mistaken as cases of pan uveitis, while one was diagnosed as neuro retinitis. These were treated with corticosteroids before they were referred to us. The postoperative cases were all treated as sterile reactions for several days before presentation to us. The mean interval between the surgery and the onset of symptoms in postoperative cases was 4.1 days (range 1-12 days). In contrast, all the posttraumatic cases had symptoms within 24 hours of injury, although this could have been due to the injury rather than the infection itself. The mean interval between symptoms and presentation was 50.4 days (median, 55 days) in the postoperative group, 51 days (median, 40 days) in the endogenous group and 32 days in the posttraumatic group. The agent of injury in the posttraumatic cases was different in the three cases, namely, broomstick, metal, and stone.
Ultrasonography was done in all the cases and revealed intra vitreal echoes commensurate with vitreous exudates. The retina was attached on ultrasonography in all but one case of endogenous endophthalmitis.
The details of management are given in [Table - 2]. Only one eye (a case of post vitrectomy endophthalmitis) could be sterilised with intra vitreal antibiotics alone while the rest needed vitrectomy. All the cases of posttraumatic endophthalmitis developed rapid onset retinal detachment post operatively due to the presence of necrotic retina. In 2 eyes, this led to a loss of light perception and severe rubeosis, while the third patient was not willing to undergo further surgery.
One eye with endogenous endophthalmitis required evisceration after a therapeutic keratoplasty with vitrectomy failed to sterilise the eye. None of the eyes that had retinal detachment or developed retinal detachment post vitrectomy, were considered operable due to the severity of retinal necrosis. The only case of bleb related endophthalmitis had a large retinal abscess in the posterior pole.
The follow-up period was a mean of 16.35 months, ranging from a few days to 84 months. All the five eyes with a follow-up of less than one month were cases that had inoperable retinal detachments with sterilised eyes and thus did not affect the inferences drawn.
Anatomic and visual results
Except one eye with endogenous endophthalmitis that needed evisceration, the rest could be sterilised. However, the anatomical results were satisfactory only in the cases with postoperative endophthalmitis. Seven of the 10 eyes in this group had a clear media and an attached retina; none of the eyes with posttraumatic and bleb related endophthalmitis had a satisfactory outcome; 2 of the 6 eyes with endogenous endophthalmitis had a clear media and an attached retina. Four of 10 eyes with postoperative infection regained 6/18 or better vision while 2 of 6 eyes with endogenous endophthalmitis recovered 6/36 vision. None of the eyes with posttraumatic and bleb related endophthalmitis recovered any useful vision.
Microbiological evaluation [Table - 3][Table - 4]
Culture. The criteria used to consider the isolated bacterium as the causative agent were its growth from vitreous humor (VH) or aqueous humor (AH) on two or more of the inoculated media or its growth on a single medium correlating with direct smear findings or repeated isolation of the same organism from two or more intraocular specimens of the patient. The culture proven cases included AH only from two eyes, VH only from 12 eyes, and both AH and VH from 6 eyes. In six eyes in which both the aqueous and vitreous were examined, the organism could be cultured from the vitreous in all the cases and from the aqueous in one case. The organism was also isolated from the lens capsule in one case and from the explanted IOL in another.
Smears. Smears were performed in all but two eyes. In these two eyes, the specimen was directly inoculated into liquid media in the operating room since the surgery was done late at night. Three of the 8 aqueous specimens and 13 of the 18 vitreous specimens on which smear examina-tions were performed were positive for organisms. The positive smears (n=16) consisted of gram-negative bacilli in 10 cases, gram-positive bacilli in 5 and gram-negative coccobacilli in one specimen. On comparison of the smears from AH and VH in the cases wherein both were examined on the eyes, it was found that 4 of 6 had vitreous smears positive and only 2 of 6 aqueous smears were positive. Where the vitreous smear was negative for organisms, the aqueous smear was also negative.
The smear positivity was also compared between the various clinical circumstances of endophthalmitis. Significantly, all the cases with endogenous endophthalmitis had positive smears, while 7 of the 10 postoperative cases and 2 of the 3 posttraumatic cases had smear positivity.
In 11 cases, the growth occurred in all four media and in three cases the growth occurred in two or three of the four media. Significantly, in four cases the growth occurred only in the liquid medium, although inoculation was done in all the media. Positive growth was seen in 1-4 days' time (average of 1.7 days).
The antibiotics susceptibility was tested against. gentamicin, amikacin, ciprofloxacin, ampicillin, cefotaxime, cefazolin, ceftazidime, and vancomycin. The susceptibility could not be tested for all the listed drugs for all the isolates due to logistic reasons. One isolate was resistant to all the drugs tested. Highest sensitivity was to ciprofloxacin (88.9% of the isolates tested), followed by amikacin (73.7% of the isolates tested). Resistance to both gentamicin and amikacin was seen in only four isolates. Of the three antibiotics in the cephalosporin category that were tested, the sensitivity was highest to cefotaxime (57.9%). All the cases with favourable outcome had isolates that were resistant to two or fewer antibiotics.
| Discussion|| |
Acinetobacter calcoaceticus belongs to the family Neisseriae. It is an ubiquitous saprophyte-gram-negative bacillus of low virulence. Up to 25% of healthy adults exhibit cutaneous colonisation. It was the commonest organism identified by the hospital personnel. The sub-species Iwoffi displays a predilection for the urinary tract. The organism is rod-shaped during rapid growth but is coccoid in the stationary phase. It has a tendency to retain crystal violet and be wrongly identified as a gram-positive organism.
The organism has no virulence factors and does not produce toxins. Hence, it is labelled as an opportunistic pathogen. However, it has been reported to cause purulent infection in otherwise healthy hosts. Acinetobacter has been known to cause infection in every organ of the body. Crawford et al reported this organism as a cause of recurrent posttraumatic endophthalmitis. Herbs described a case of corneal ulceration caused by contaminated soft contact lenses due to Herella species, which was the earlier name of Acinetobacter calcoaceticus . We have described a case of chronic endophthalmitis caused by sequestered Acinetobacter organism in the capsular bag, following cataract surgery with intraocular lens implantation. No large series of intraocular infection with this organism has been described in the literature.
The present series is a heterogeneous group of endophthalmitis cases due to exogenous and endogenous infection. In terms of severity of the disease, posttraumatic cases had the worst prognosis, while cases with endogenous endophthalmitis also had relatively poor prognosis with only 2 of 6 cases (33.3%) recovering useful vision. In contrast, cases with postoperative endophthalmitis did comparatively well, with 70% recovering moderate to good vision. The following factors could explain this behaviour. The organismal load is expected to be high with perforating injury especially with a retained intraocular foreign body. With the ubiquitous presence of the organism, heavy contamination of agents of injury (stone, metal chip and broomstick) is easily explained. Hence the severity of infection and the uniformly bad outcome in traumatic cases is partially explained by the organismal load.
Endogenous Acinetobacter endophthalmitis presents several interesting features. It is obvious that these patients had transient Acinetobacter bacteraemia that caused the lodgement of the bacteria in the ocular tissue, leading to clinical intraocular infection. However, only one of these six cases had bilateral intraocular infection, and surprisingly none of the cases had historically any identifiable predisposition. Acinetobacter bacteraemia is most commonly reported with respiratory infections followed in frequency by urinary tract, wound and skin infections. Instances of patients who appeared healthy but had bacteraemia were recorded. This was generally noticed in patients with indwelling catheters. In these six cases of endogenous endophthalmitis, bacteraemia occurred asymptomatically, possibly due to the low virulence of the organism and the relatively good health of the patients. Once lodged in the eye, the vitreous served as a good culture medium and facilitated its growth. The growth of infection in the eye was also slow as is evident by the fact that most of these cases were misdiagnosed as idiopathic non-infective uveitis or neuro retinitis to start with. A high index of suspicion at the initial stage could have resulted in salvaging a larger number of eyes.The postoperative endophthalmitis cases presented with typical chronicity, with the referral doctor treating them as cases of sterile inflammation. This is evident from the fact that the mean interval between the occurrence of symptoms and presentation to us was 50.4 days (median, 55 days). Part of the delay could also be due to the socio-economic conditions prevailing in this country. Diabetes mellitus was present in 5 of 10 cases (50%) with postoperative Acinetobacter endophthalmitis. Acinetobacte r organisms are said to grow in all routine media. In 6 eyes where both the vitreous and aqueous were cultured, the aqueous yielded positive growth in only one eye, while the vitreous was culture positive in all. This culture pattern supports the general understanding that culturing the organism is easier in the vitreous than in the aqueous. The smears evaluation was also similar with 3 of 8 aqueous specimens and 13 of 18 vitreous being positive for the organisms. Smear positivity (for both vitreous and aqueous) was 100% in eyes with endogenous endophthalmitis, 70% in postoperative endophthal-mitis, and 66.6% in post traumatic cases.
In vitro ciprofloxacin proved to be the most effective drug, with 88.9% sensitivity and when the organism was resistant to ciprofloxacin, it was also resistant to all other tested antibiotics except amikacin in one case. Between the two tested, amikacin was marginally better compared to gentamicin. Cephalosporins were in general inferior to aminoglycosides and ciprofloxacin in their efficacy. Good outcome was associated with resistance of the organism to two or fewer antibiotics, which is a good correlation of the in-vitro results with the in vivo efficacy.
The literature suggests a seasonal variation in the incidence of Acinetobacter infection. In our series, the cases of postoperative endophthalmitis were distributed uniformly throughout the year, but all the endogenous endophthalmitis cases occurred in October and November. The explanation for this clustering of Acinetobacter endogenous postoperative endophthalmitis is not evident. There can obviously be no relevance of season for the traumatic infections.
| References|| |
Madhavan HN, Anand AR, Therese KL. Infectious endophthalmitis. Indian J Med Microbiol
Fallon RJ, Young H. Neisseria, Moraxella, Acinetobacter. In: Collee JG, Duguid JP, Fraser AG, Marmion BP, Sinamons A, editors. Mackic and Mc Cartney's Practical Medical Microbiology
. 13th edition. Churchill Livingstone, Edinburgh. 1996. pp 283-297.
Nauer AW, Kirby WMM, Sherris JC. Antibiotic susceptibility by a standardised single disc method. Am J Clin Pathol
Al-Khoja MS, Darrel JH. The skin as a source of Acinetobacter and Moraxella species occurring in blood cultures. J Clin Pathol
Larson EL. Persistent carriage of gram- negative bacteria on hands Am J Infect Control
Hoffmann S, Mabeck CE, Vejlsgaard R. Bacteriuria caused by Acinetobacter calcoaceticus Biovars in a normal population and in general practice. J Clin Microbiol
Allen DM, Harman BJ. Acinetobacter species. In: Mandell GL, Douglas RG, Bennet JE, editors. Principles and practice of infectious diseases. 3rd edition. New York: Churchill Livingstone, 1990. pp 1696-1699.
Crawford PM Jr, Carway PD, Payman GH. Trauma induced Acinetobacter Iwoffi endophthalmitis with multi organism recurrence. Strategies with intra vitreal treatment. Eye
Herbs RW. Herella corneal ulcer with use of soft contact lenses. Br J Ophthalmol
Gopal L, Madhavan HN, Ramaswamy AA, Saswade M, Battu RR. Postoperative endophthalmitis caused by sequestered Acinetobacter calcoaceticus. Am J Ophthalmol
Robinson RG, Garrison RG, Brown RW. Evaluation of the clinical significance of genus Herellea. Annals Intern Med
Retailliau HF, Hightower AW, Dixon RE, Allen JR. Acinetobacter calcoaceticus: A noscomial pathogen with an unusual seasonal pattern. J Infect Disease
[Table - 1], [Table - 2], [Table - 3], [Table - 4]