Indian Journal of Ophthalmology

ORIGINAL ARTICLE
Year
: 2004  |  Volume : 52  |  Issue : 2  |  Page : 127--31

Sterile Endophthalmitis in Vitrectomised Eyes due to Suspected Heat Resistant Endotoxins in the Infusion Fluid


B Patnaik, C Biswas, Rajinder Kalsi Patnaik 
 Retina Associates Eye Institute, New Delhi, India

Correspondence Address:
B Patnaik
Retina Associates Eye Institute, New Delhi
India

Abstract

Purpose: To report on to the possibility of development of severe postoperative sterile endophthalmitis due to heat-resistant bacterial endotoxins in commercially available infusion fluids Methods: A case study of 4 eyes that had previously undergone vitreoretinal surgery, which developed clinical endophthalmitis within 18 hours of surgery and two eyes that had vitreous surgery with intraocular gas and did not develop clinical endophthalmitis following intraocular surgery on three consecutive operative days Results: The vitreous samples were sterile, both for bacteria and fungi. The only common supply in all cases was a new batch of Ringer�SQ�s lactate infusion fluid. Though the Ringer�SQ�s lactate solutions in the same batch were also sterile, the infusion fluids contained abnormally high levels of bacterial endotoxins detected by gel clot method. Conclusion: Commercially available infusion fluid may be sterile, yet contain endotoxins from killed bacteria. This could cause postoperative sterile endophthalmitis.



How to cite this article:
Patnaik B, Biswas C, Patnaik RK. Sterile Endophthalmitis in Vitrectomised Eyes due to Suspected Heat Resistant Endotoxins in the Infusion Fluid.Indian J Ophthalmol 2004;52:127-31


How to cite this URL:
Patnaik B, Biswas C, Patnaik RK. Sterile Endophthalmitis in Vitrectomised Eyes due to Suspected Heat Resistant Endotoxins in the Infusion Fluid. Indian J Ophthalmol [serial online] 2004 [cited 2024 Mar 29 ];52:127-31
Available from: https://journals.lww.com/ijo/pages/default.aspx/text.asp?2004/52/2/127/14610


Full Text

Infective endophthalmitis is the commonest cause of severe postoperative inflammation following intraocular surgery. However, not all cases of severe postoperative inflammation following intraocular surgery are infective. A condition of acute anterior segment inflammation following uneventful cataract surgery with intraocular lens (IOL) with no apparent cause (non infective) has been described as toxic anterior segment syndrome (TASS).[1],[2] The offending agents could be introduced through the IOL, the surgical instruments, the viscoelastic or the irrigation fluids. The chemicals used to polish and sterilise the IOLs, and even the intraocular lens design have been cited as sources of inflammation. pH and the ionic composition of the irrigating fluid may also be the cause of the inflammation. [2],[3],[4],[5] The endotoxins produced by the bacteria, which have been inactivated by the sterilisation process, are suspected to be the cause of the severe inflammation.[6],[7]

A series of four consecutive cases of sterile endophthalmitis during 3 successive days following intraocular procedures in previously vitrectomised eyes is reported here. It appears the infusion fluid, which filled the "empty" vitreous cavity, though sterile had contained heat-resistant endotoxins responsible for the severe postoperative inflammation. Two other cases operated for pars plana vitreous surgery during the same period using the same suspect infusion fluid, did not develop inflammation, probably because the fluid was replaced by gas at the end of the surgery.

 Materials and Methods



A series of six cases operated at Retina Associates Eye Institute, New Delhi over a period of 3 consecutive days is reported. Four consecutive cases developed acute endophthalmitis with markedly reduced vision, grade 3 to 4 media clarity, dense cells in the anterior chamber and in 3 out of 4 cases hypopyon following anterior segment intraocular surgery. All these cases had undergone vitreoretinal surgery in the past by one of the authors (B.P). All the anterior segment surgeries were done by one of the authors (RKP) during the same period (10-12 October, 2000) Ringer's lactate infusion fluid batch (OG. 5681, Mount Mettur Pharmaceuticals Limited, Bhopal) was used. Two other patients received trans pars plana vitreous surgery and gas tamponade in the same operating room using the same infusion fluid by the same vitreous surgeon (B.P). Both had an uneventful postoperative course.

Cases of endophthalmitis detected within 18 hours of surgery were admitted to the hospital for close observation and prompt intervention. Clinical findings were recorded using a slitlamp and indirect ophthalmoscope on a daily basis, and sometimes more than once a day.

Oral prednisolone (40mg/day for patient 1 and 2, 60mg/day for patient 3,4) with systemic antibiotics (IV amikacin 500 mg twice daily, patient # 1 and oral cefuroxime 500 mg twice daily for the rest) and hourly corticosteriod and antibiotics drops were tried in each case of endophthalmitis for 12 to 24 hours before collecting a sample of fluid from the vitreous. The vitreous tap was done through a pars plana stab using a 26-gauge hypodermic needle and this was followed by injection of intravitreal antibiotics and dexamethasone, with or without vitreous lavage.

Since all the cases had undergone vitrectomy, it was easy to aspirate 0.5 ml of the fluid from the vitreous. The aspirate was immediately sent for microscopy and culture for both bacteria and fungi.

The sealed bottles of the Ringer's lactate infusion fluid from the same batch used in this series of cases were sent to the same laboratory for complete microbiological examination without resterilisation.

Biochemical analysis

Five sealed bottles of the Ringer's lactate solution of the same batch (OG. 5681, Mount Mettur Pharmaceuticals Limited, Bhopal) were sent to the Shriram Institute for Industrial Research, New Delhi for biochemical analysis. As a control, similar samples of Ringer Lactate infusion fluid from another manufacturer( Batch: 3.40.5346, Core Health Care Ltd., Sachana, Gujrat, India) was also tested at the same laboratory.

Vitreous lavage

This was carried out by classical 3 port pars plana approach in Case # 2 and # 4. The infusion fluid used was Ringer's lactate from a different manufacturer than the one used during the original surgery. Besides the vitreous lavage the anterior chamber also was flushed with the new infusion solution.

Routine resterilisation

The infusion fluids supplied by the manufacturers are routinely sterilised again by autoclaving immediately before use. The disposable tubings, needles and syringes are sterilised again by Ethylene oxide. The status of sterilisation is controlled by suitable indicator papers. Two representative cases are described briefly.

 Results



Patient 1. 22-year-old male. In a case of traumatic retinal detachment in the left eye was successfully treated by scleral buckling in 1994. Pars plana lensectomy with vitreoretinal surgery was done in April 2000 for redetachment with proliferative vitreoretinopathy in the same eye. The patient's best-corrected visual acuity was finger counting at half meter due to the presence of a scar over the macula. There were some residual perfluorocarbon liquid bubbles, frequently producing symptoms in the patient. The bubbles were removed using a 26 gauge hypodermic needle on a 2 ml glass syringe through limbal paracentesis in a sitting position on 10 October, 2000 and the intraocular volume was restored by injection of less than 0.75 ml of Ringer's lactate infusion fluid of the suspected batch. On the first postoperative day, visual acuity in the left eye was hand movement with accurate projection. Slitlamp biomicroscopy revealed 3+cell, 2+ flare in the anterior chamber, and fundus examination showed grade III media clarity.[8] He was put on oral prednisolone 40mg once daily, intravenous injection of amikacin 500mg twice daily and topical tobramycin 0.3% and dexamethazone 0.1% drops every one hour each. On the next day, hypopyon appeared and media clarity deteriorated. Intravitreal injection of vancomycin 1mg, amikacin 0.4mg and dexamethasone 0.2mg in 0.1ml each given on 12 October 2000, after taking vitreous sample for microbiological test through pars plana. At the same time oral prednisone was increased to 40mg twice daily. No bacteria or fungus was either identified or cultured in the vitreous specimen. The hypopyon disappeared on the next day. Systemic corticosteroid was continued for 3 weeks and topical steroid for 6 weeks. The inflammation resolved slowly over the next few weeks. His visual acuity improved to the preoperative level after one month.

Patient 6. 55-year-old male. In a Pars plana vitreoretinal surgery with intraocular gas tamponade was carried out successfully in 1999 for retinal detachment with proliferative vitreoretinopathy in his right eye. Subsequently he developed macular pucker in this eye. A repeat vitreoretinal surgery with removal of epiretinal membrane was successfully carried out in September 2000. The detachment recurred after 2 months, and vitreoretinal surgery with intraocular gas tamponade was carried out on 12 October 2000. Infusion fluid used in the procedure was the same Ringer's lactate solution used as in the other cases, fortified with adrenaline 0.3ml, dexamethasone 1mg and heparin 500 IU in 500ml. The epiretinal membrane was removed. Perfluorocarbon liquid was used to flatten the retina intraoperatively. Endolaser photocoagulation was applied. Perfluorocarbon liquid was exchanged with infusion fluid and then infusion fluid was exchanged with air. Air was replaced by SF6 gas at the end of surgery. The postoperative period was uneventful.

Observations

The relevant data on the 6 cases reported are summarised in [Table 1]. In Case: 1, the surgical intervention involved the entry of a sterile disposable hypodermic needle to the anterior chamber to remove by gentle suction the perfluorocarbon liquid from the anterior chamber in a sitting position followed by injection of about 0.75 ml of the Ringer's lactate solution out of a new batch from a reputed manufacturer. In Case: 3, a pars plana posterior capsulectomy was done using a sterile vitreous cutter. The infusion fluid used was the same Ringer's lactate solution. No other instruments or material entered the globe. The cases 2 and 4 were classical cases of small incision cataract surgery with IOL implantation. Several intraocular surgical instruments were used intraoperatively. A widely used viscous material was used and standard intraocular lenses were placed in the capsular bag after phacoemulsification. All of these cases showed severe intraocular inflammation, noticed at the first dressing, within 18 hours of the surgical intervention. The only supply common in all these 4 cases was the Ringer's lactate solution of newly acquired lot of a different batch to what was being used till the series of post operative inflammation were noticed. Interestingly all these eyes had been through a vitrectomy procedure on the previous occasions.

During the same period of 3 days, two cases (Cases: 5 and 6) were operated in the same operating room with the same staff, the same vitrectomy system, similar vitreous cutters and the same suspect Ringer's lactate solution as in the previous cases. Trans pars plana vitreous surgery was carried out with intraocular gas (SF6) injection. There were no postoperative inflammations.

All the samples, from the patients and the suspect bottles of Ringer's lactate were adequate in quantity for microbiological examination. There were no bacteria or fungi detected by microscopy. Culture was negative for all samples. The biochemical analysis of the Ringer's lactate solution of batch number OG 5681(Mount Mettur Pharmaceuticals, Bhopal) as reported by the Shriram Institute for Industrial Research is reproduced in [Table 2]. Most constituents were nearly similar as claimed. The pH was some what low at 6.21 However, what was significant was that the qualitative detection of endotoxins (Gel Clot method as per IP 1966) meaning the endotoxin level present was equal to or more than 0.5 EU/ml. These endotoxins were lipopolysaccharides (LPS) present in the outer cell walls of gram-negative bacteria.

All cases of endophthalmitis were treated with intravitreal injection of two antibiotics and dexamethasone. All the patients received oral corticosteroids and frequent topical corticosteroid seem to be ineffective during the first 24 hours. While continued oral and topical corticosteroid resolved the inflammation during the weeks to follow, the vitreous lavage using a different Ringer's lactate solution in two cases (Cases: 2 and 4) resulted in immediate and dramatic improvement.

While the Ringer lactate manufactured by the Mount Mettur Pharmaceutical was positive for endotoxin, the one manufactured by the core Health care was not.

 Discussion



The factor common to all 4 cases in this series of severe inflammation was the irrigation fluid. The fact that use of a new batch of Ringer's lactate had coincided with the spurt of these cases gave us added reason for suspicion. Even though sterilisation procedures for devices and supplies for surgery was adequate the chance of infection could not be entirely excluded. Since the consequences of infective endophthalmitis could be rapidly devastating, most surgeons would consider intravitreal antibiotics before microbiological confirmation. In this series all 4 eyes were treated with one time intravitreal injection of antibiotics and dexamethasone. But, the vitreous fluid was culture negative. Vitreous lavage was done in two of the four cases. The inflammation resolved in all eyes in two to three weeks with local and oral corticosteroids. The improvement occured in eyes where vitreous lavage was done using a different Ringer's lactate solution. It is tempting to conclude that the physical replacement of the suspect Ringer's lactate filling the 'empty' vitreous cavity (empty of vitreous) with fresh solution was immensely beneficial.

It was interesting to note that during the same 3-day period two cases were operated for trans pars plana vitreous surgery using the very same suspect Ringer's lactate solution and fluids to exchange. These eyes did not develop postoperative inflammation. Though same infusion fluid was used there was a distinct difference between the first 4 cases and later 2 cases The 'empty' vitreous cavity of the eyes of the former group were filled with or contained considerable amount of the suspect Ringer's lactate at the end of the surgery. On the other hand, the suspect Ringer's lactate solution in the 'empty' vitreous cavity of the eyes of the second group was replaced by air and then by a long acting gas (SF6) through a two stage exchange maneuver. Thus, the conclusion that the suspect Ringer's lactate, as it filled the vitreous cavity of eyes where the vitreous had been removed by vitreous surgery, was responsible for causing severe post operative inflammation, probably by some toxic effect, was inescapable. The biochemical examination of the Ringer's lactate solutions of the same batch confirmed our suspicion. It was found to contain an unacceptably high concentration of bacterial endotoxins. There seems to have been growth of bacteria in the Ringer's lactate solution during its manufacture or bottling. These had been killed by successful sterilisation by autoclaving, leaving behind heat-resistant endotoxins.

The endotoxins produced by the bacteria could be a distinct cause of the severe inflammation.[6],[7] The inflammation could manifest as mild anterior uveitis or could be severe enough to mimic endophthalmitis as in our series The difference could be endotoxin-dose dependent.

Thus, instruments, irrigation fluids or any other material introduced into the eye during any intraocular procedures must not only be sterile but free of toxic substances. The lack of confidence in the sterility of the commercially available infusion fluid is the reason why most surgeons in India play it safe by re-sterilising them before use. However, even then one may see epidemics of sterile endophthalmitis and one of the important reasons could be heat-resistant toxins. Considering the fact that all 9 bottles (4 used in 4 cases showing sterile endophthalmitis and 5 providing material for biochemical testing) seems to have contained bacterial endotoxin and was the part of the same batch, the bacterial contamination seems to have taken place somewhere in the manufacturing process. This worrisome possibility must be addressed by the manufacturers and the drug controllers. The surgeons should be aware of such a possibility.

References

1Monson MC, Mamalis N, Olson RJ. Toxic anterior segment inflammation following cataract surgery. J Cataract Refractive Surg 1992; 18: 184- 89.
2Faisal SJ, Mamalis N, Spencer TS, Kerstine RS, Olson RJ. Postoperative sterile endophthalmitis (TASS) associated with the Memory Lens. J Cataract Refractive Surg 2000; 26: 1773- 77.
3Apple DJ, Mamalis N, Loftfield K, Googe JM, Novak LC, Kavka-Van Norman D, Brady SE, Olsan RJ. Complication of intraocular lens: A historical and Histolopathological review. Surv Ophthalmol 1984 :29 ;1-54.
4Meltzer DW. Sterile hypopyon following intraocular lens surgery. Arch Ophthalmol 1980; 98 :100-4.
5Ratner BD. Analysis of surface contamination of intraocular lenses. Arch Ophthalmol 1983;101:1434- 38
6Kreisler KR, Martin, SS Young CW, Anderson CW & Mamalis, N. Postoperative inflammation following cataract extraction caused by bacterial contamination of the cleaning bath detergent. J Cataract Refractive Surg 1992; 18: 106-10.
7Richburg FA, Reidy J J, Apple DJ, Oslon RJ. Sterile hypopyon secondary to ultrasonic cleaning solution. J Cataract Refractive Surg 1986; 12: 248-51
8Nussenblatt RB. Palestine AG. Chan CC, Roberge F. Standardization of Vitreal Inflammatory activity in intermediate and posterior uveitis. Ophthalmology 1985;92:467-71.