- Email: [email protected]
Sterile hypopyon secondary to ultrasonic cleaning solution
articles Sterile hypopyon secondary to ultrasonic cleaning solution Frederick A. Richburg, M.D. Fresno, California David J. Apple, M.D. James J. Reidy, M.D. Randall J. Olson, M.D. Salt Lake City, Utah ABSTRACT During a 20-week period, 21 cases of sterile hypopyon occurred 24 hours after uncomplicated extracapsular cataract extraction and inthe-bag posterior chamber intraocular len implantation at the Valley Eye Institute. While all ca es responded to teroid treatment and exhibited no residual ocular effect , the cause of the hypopyon remained a mystery de pite an on-going reevaluation of the surgical procedure and sterilization techniques. Several suspected cau es were inve tigated and eliminated. It wa clinically determined that urgical instruments placed in older ultrasonic cleaning olution cau ed the hypopyon . Further analysis at the Center for Intraocular Re earch at the University of Utah revealed that the in truments were contaminated by a heat-stable endotoxin that remained on the in truments through the autoclaving process. Surgeon currently u ing ultra onic in trument cleaner hould be aware that, with orne models, following the manufacturer's recommended fluid changing procedure could result in endotoxic contamination of surgical instruments which could, in turn, cau e evere inflammatory reactions in the immediate postoperative period. Key Word : autocla e, 'olution
ndotoxin , h pop on , ultra 'onic cl aning
Over a 20-week period, 21 cases of mild to severe sterile hypopyon presented 24 hours after uncomplicated extracapsular cataract extraction (ECCE) and inthe-bag posterior chamber intraocular lens (IOL) implantation at the Valley Eye Institute. A total of 266 procedures had been performed during the 20 weeks. All patients had surgery in an identical manner, using the same surgical instruments in the same operating room, attended by the same personnel, and using the same surgical techniques. REPORT OF CASES The first patient had surgery on June 7, 1984. Approximately 24 hours later, a 25% layered hypopyon
was seen in the anterior chamber (Figure 1). There was moderate to severe conjunctival inflammation. Visual acuity was reduced to hand motion and ocular pressure was 35 mm Hg. An anterior chamber tap was performed and a culture and smear were taken; both showed negative for bacteria. An initial protocol of subconjunctival steroids and antibiotics was prescribed. The antibiotics were discontinued after negative cultures were reported; topical and systemic steroid treatment was continued, and clinical improvement was observed. The protocol for the 20 subsequent cases prescribed only topical and periocular steroids, and occasionally
From the Valley Eye Institute, Fresno, California, and the Departments of Ophthalmology and Pathology, University of Utah Health Sciences Center, Salt Lake City, Utah. Reprint requests to Frederick A. Richburg, M.D., Valley Eye Institute, 1680 East Herndon Avenue, Fresno, California 93710. 248
J CATARACT REFRACT SURG-VOL
12, MAY 1986
Fig. 1.
(Richburg) Sterile hypopyon seen 24 hours after uncomplicated ECCE and in-the-bag posterior chamber IOL implantation.
systemic steroids, depending upon the severity of the anterior chamber reaction and the effectiveness of treatment. Most of these cases were mild. Some even required gonioscopy to see the hypopyon clearly. Three cases were more severe and associated with a cloudy, inflammatory reaction in the vitreous which required a heavier oral steroid treatment. In one instance, it took 36 days for the vitreous to clear but the vision returned to 20/20. During this time, however, the patient developed transient moon facies that resolved three weeks following cessation of treatment. Follow-up studies on all patients have revealed no sequelae. Ninety percent have visual acuities of20/40 or better (Table 1), and the average corneal endothelial cell loss was 674 with no evidence of persistent corneal edema (Table 2). Table 1. Patient visual acuities three months postoperatively. Visual Acuity
Percentage
20/20 - 20/25
67
20/30 - 20/ 40
23
20/60
5
20/200 (macular degeneration)
5
Table 2. Endothelial cell loss in cases with sterile hypopyons secondary to endotoxic anticontamination. Condition
Cells/mm 2
Minimum cell loss seen
286
Maximum cell loss seen
1,433
Mean cell loss seen Lowest postoperative count Corneal edema seen
674 1,416
o
REEVALUATION METHODS Beginning with the second occurrence of sterile hypopyon, one author (FAR) and the Quality Assurance Committee of the institute instigated an on-going reevaluation of the entire surgical procedure, with particular emphasis on the sterilization techniques. A team, consisting of the surgeon, assistant surgeon, anesthesiologist, scrub nurse, and circulating nurse, identified and modified numerous suspected causes of the hypopyon, including: 1. Irrigating solutions - the brand was changed. 2. Technique for warming the balanced salt solution (BSS)-the preoperative technique of warming and keeping the BSS bottles warm while on the intravenous hanger was changed. 3. Additives to the irrigating solution-the epinephrine used and the mixing technique was suspected at one time. 4. Intravenous tubing from the BSS bottles to the silastic tubing-the brand of tubing was changed. 5. The silastic irrigation and aspiration tubing-first the technique of cleaning the tubing was changed, then the sterilization technique (from autoclave to gas) was changed. New tubing was installed before each surgery. 6. The scrub solution -a small, diluted amount was injected into the anterior chamber of a rabbit without reaction. 7. The implanted IOL-hypopyons were reported with three different lens styles. 8. The preoperative irrigation solution (used during part of the preparation procedures) -changed from artificial tears to BSS. 9. The preoperative antibiotics-brands changed. 10. Microscopic examination of the irrigation and aspiration handpiece revealed debris and grease around the O-rings-this item was carefully cleaned. 11. The phacoemulsification unit-another unit was brought in. 12. The autoclave-the generator and old water in the jacket were replaced; water filters were changed and a new steam filter added. None of these changes affected the hypopyon rate. In addition, discussions with leading ophthalmologists and pathologists across the country proved to be fruitless; however, one (RJO) recommended searching for a bacterial contaminant. After observing the entire surgical procedure, an operating room consultant reported that nothing was out of line and that the hypopyons must be related to instrumentation, set-up, or clean-up. Examination of the patients' charts revealed that 75% of the first 16 hypopyon cases occurred in the first surgical case of the day, and the other 25% occurred in the second case. No hypopyons occurred in the third, fourth, or fifth case of any surgical day (Table 3). This fact led to the initial supposition that the causative
J CATARACT REFRACT SURG-VOL
12, MAY 1986
249
Table 3. Incidence of hypopyon according to surgical day and order of implant cases.
Day and Order
Hypopyon Cases (percentage)
Day surgery performed Monday
48
Tuesday
10
Wednesday
18
Thursday
24
Order of cases 1st case of the day
76
2nd case of the day
24
3rd case of the day
0
factor may have been washed away after the first operation of the day, and therefore must involve the irrigation/aspiration tubing or the surgical instruments. This conclusion proved to be wrong. The instrument cleaning protocol called for ultrasonic cleaning of the instruments at the end of each surgical day. Commercially prepared distilled water was used as a cleaning solution to fill the Mettler ultrasonic cleaner every Monday. Because the manufacturer's literature stated "the longer the cleaning solution was used the better it would function," it was not discarded until Thursday. The instruments were then wrapped and autoclaved for the next surgical session. Analysis revealed that 72% of the hypopyons occurred on Monday and Thursday of the surgery schedule, even though four or five cases were done on each of the first four days of the week. The hypopyons became more frequent each day that the ultrasonic cleaning solution was reused. Cultures of this solution were positive for Kelbsiella pneumoniae. This finding led us to suspect that, even though the instruments were autoclaved prior to usage, some toxic agents derived from these organisms in older, contaminated ultrasonic cleaning solution must have caused the hypopyons. It also appeared that the contaminant was washed away after one usage. LABORATORY RESULTS Samples of the ultrasonic cleaning solution were sent to the Utah Center for Intraocular Lens Research for pathologic analysis. Analysis of the solution by two of the authors (JJR, DJA) at the Utah Center revealed the presence of endotoxin in the solution. The quantity of endotoxin present was equivalent to 0.5 mg/ml of E. coli lipopolysaccharide B, or 2.5 enzyme units. Because the surgical instruments were autoclaved after being placed in the solution, one author (FAR) felt it was essential to determine the heat stability of the endotoxin present in the solution. To simulate the clinical situation, 15 cc of the contaminated solution 250
was autoclaved at 120°C for 15 minutes, the identical temperature and exposure time to which the surgical instruments were exposed. Three rabbits were then anesthetized, and 0.5 cc of aqueous fluid was withdrawn from their anterior chambers. In rabbit 1,0.5 cc of bacteriostatic 0.9% NaCI solution was injected into the anterior chamber. In rabbit 2, 0.5 cc of nonautoclaved contaminated solution was injected into the anterior chamber, and in rabbit 3,0.5 cc of the autoclaved contaminated solution was injected. Twenty-four hours later, there was no appreciable inflammatory reaction in rabbit 1, which had received the sterile bacteriostatic solution. Rabbit 2, which received the nonautoclaved solution, developed a late hypopyon. Rabbit 3, which received the autoclaved contaminated solution, developed severe ciliary infection and iritis accompanied by the formation of pupillary membranes (Figure 1). The endotoxins produced by the bacteria in the solution obviously remained stable even after exposure to autoclave temperatures. DISCUSSION Authors have attributed sterile hypopyons to numerous causes, including IOL lens polishing compounds and rough IOL edges,l reaction to lens protein,2,3,4 IOL sterilization techniques,S laser iridotomy, 6 leukemia, 7 iridocyclitis, Bechet's syndrome, and other types of uveitis. To our knowledge, this is the first presentation of hypopyon caused by endotoxin contamination. Our investigation proved that, when instruments are placed in a three-day old ultrasonic bath contaminated by Gram negative bacteria, endotoxins produced by these microorganisms could coat the instruments. The bacterial contamination is effectively eliminated when the instruments are autoclaved, but the heat-stable endotoxin remains on the instruments, with the potential of causing serious ocular inflammation. The manufacturer states in the instruction manual for this model of ultrasonic cleaner, "Normally, the longer the same liquid is used, the better it cleans." This may be true, since the longer the ultrasonic bath solution is used the more the liquid becomes degassed, which allows for maximum cavitation and better cleaning efficiency. In a private communication to one author (FAR), the manufacturer now recommends that the bath solution "be changed every 2-3 days or when needed. " The key phrase here is "when needed," since bacterial contamination in a highly used ultrasonic cleaner could occur in as little as 24 hours. The institute now uses the same ultrasonic cleaner with ordinary tap water that is changed daily. The unit is also wiped with an alcohol-soaked sponge every day. There have been no sterile hypopyons in the more than 744 surgical procedures performed since October 22, 1984.
J CATARACT REFRACT SURG-VOL 12, MAY 1986
Ophthalmic surgeons currently using ultrasonic instrument cleaners should be aware that, in some models, following the manufacturer's recommended fluid changing procedures could result in endotoxic contamination of surgical instruments. This contamination could cause patients to develop severe inflammatory reactions in the immediate postoperative period. The authors recommend that ultrasonic cleansing fluid be changed after each use to avoid such contamination. REFERENCES 1. Meltzer DW: Sterile hypopyon following intraocular lens sur-
gery. Arch Ophthalmol 98:100-104, 1980 2. Apple DJ, Mamalis N, Steinmetz RL, Loftfield K, et al: Phacoanaphylactic endophthalmitis associated with extra-
3.
4.
5. 6.
7.
capsular cataract extraction and posterior chamber intraocular lens. Arch Ophthalmol 102:1528-1532, 1984 Apple DJ, Mamalis N, Steinmetz RL, Loftfield K, et al: Phacoanaphylactic endophthalmitis following ECCE and IOL implantation. Am Intra-Ocular Implant Soc J 10:423-424, 1984 Abrahams IW: Diagnosis and surgical management of phacoanaphylactic uveitis following extracapsular cataract extraction with intraocular lens implantation. Am Intra-Ocular Implant Soc J 11:444-447, 1985 Worst JGF: A retrospective view on the sterilization of intraocular lenses and the incidence of sterile hypopyon. Am Intra-Ocular Implant Soc J 6:10-12, 1980 Cohen JS, Bibler L, Tucker D: Hypopyon following laser iridotomy. Ophthalmic Surg 15:604-606, 1984 Abramson DH, Wachtel A, Watson CW, Jereb B, et al: Leukemic hypopyon. J Pediatr Ophthalmol Strabismus 18(3):42-44, 1981
J CATARACT REFRACT SURG-VOL 12, MAY 1986
251
ndotoxin , h pop on , ultra 'onic cl aning
Over a 20-week period, 21 cases of mild to severe sterile hypopyon presented 24 hours after uncomplicated extracapsular cataract extraction (ECCE) and inthe-bag posterior chamber intraocular lens (IOL) implantation at the Valley Eye Institute. A total of 266 procedures had been performed during the 20 weeks. All patients had surgery in an identical manner, using the same surgical instruments in the same operating room, attended by the same personnel, and using the same surgical techniques. REPORT OF CASES The first patient had surgery on June 7, 1984. Approximately 24 hours later, a 25% layered hypopyon
was seen in the anterior chamber (Figure 1). There was moderate to severe conjunctival inflammation. Visual acuity was reduced to hand motion and ocular pressure was 35 mm Hg. An anterior chamber tap was performed and a culture and smear were taken; both showed negative for bacteria. An initial protocol of subconjunctival steroids and antibiotics was prescribed. The antibiotics were discontinued after negative cultures were reported; topical and systemic steroid treatment was continued, and clinical improvement was observed. The protocol for the 20 subsequent cases prescribed only topical and periocular steroids, and occasionally
From the Valley Eye Institute, Fresno, California, and the Departments of Ophthalmology and Pathology, University of Utah Health Sciences Center, Salt Lake City, Utah. Reprint requests to Frederick A. Richburg, M.D., Valley Eye Institute, 1680 East Herndon Avenue, Fresno, California 93710. 248
J CATARACT REFRACT SURG-VOL
12, MAY 1986
Fig. 1.
(Richburg) Sterile hypopyon seen 24 hours after uncomplicated ECCE and in-the-bag posterior chamber IOL implantation.
systemic steroids, depending upon the severity of the anterior chamber reaction and the effectiveness of treatment. Most of these cases were mild. Some even required gonioscopy to see the hypopyon clearly. Three cases were more severe and associated with a cloudy, inflammatory reaction in the vitreous which required a heavier oral steroid treatment. In one instance, it took 36 days for the vitreous to clear but the vision returned to 20/20. During this time, however, the patient developed transient moon facies that resolved three weeks following cessation of treatment. Follow-up studies on all patients have revealed no sequelae. Ninety percent have visual acuities of20/40 or better (Table 1), and the average corneal endothelial cell loss was 674 with no evidence of persistent corneal edema (Table 2). Table 1. Patient visual acuities three months postoperatively. Visual Acuity
Percentage
20/20 - 20/25
67
20/30 - 20/ 40
23
20/60
5
20/200 (macular degeneration)
5
Table 2. Endothelial cell loss in cases with sterile hypopyons secondary to endotoxic anticontamination. Condition
Cells/mm 2
Minimum cell loss seen
286
Maximum cell loss seen
1,433
Mean cell loss seen Lowest postoperative count Corneal edema seen
674 1,416
o
REEVALUATION METHODS Beginning with the second occurrence of sterile hypopyon, one author (FAR) and the Quality Assurance Committee of the institute instigated an on-going reevaluation of the entire surgical procedure, with particular emphasis on the sterilization techniques. A team, consisting of the surgeon, assistant surgeon, anesthesiologist, scrub nurse, and circulating nurse, identified and modified numerous suspected causes of the hypopyon, including: 1. Irrigating solutions - the brand was changed. 2. Technique for warming the balanced salt solution (BSS)-the preoperative technique of warming and keeping the BSS bottles warm while on the intravenous hanger was changed. 3. Additives to the irrigating solution-the epinephrine used and the mixing technique was suspected at one time. 4. Intravenous tubing from the BSS bottles to the silastic tubing-the brand of tubing was changed. 5. The silastic irrigation and aspiration tubing-first the technique of cleaning the tubing was changed, then the sterilization technique (from autoclave to gas) was changed. New tubing was installed before each surgery. 6. The scrub solution -a small, diluted amount was injected into the anterior chamber of a rabbit without reaction. 7. The implanted IOL-hypopyons were reported with three different lens styles. 8. The preoperative irrigation solution (used during part of the preparation procedures) -changed from artificial tears to BSS. 9. The preoperative antibiotics-brands changed. 10. Microscopic examination of the irrigation and aspiration handpiece revealed debris and grease around the O-rings-this item was carefully cleaned. 11. The phacoemulsification unit-another unit was brought in. 12. The autoclave-the generator and old water in the jacket were replaced; water filters were changed and a new steam filter added. None of these changes affected the hypopyon rate. In addition, discussions with leading ophthalmologists and pathologists across the country proved to be fruitless; however, one (RJO) recommended searching for a bacterial contaminant. After observing the entire surgical procedure, an operating room consultant reported that nothing was out of line and that the hypopyons must be related to instrumentation, set-up, or clean-up. Examination of the patients' charts revealed that 75% of the first 16 hypopyon cases occurred in the first surgical case of the day, and the other 25% occurred in the second case. No hypopyons occurred in the third, fourth, or fifth case of any surgical day (Table 3). This fact led to the initial supposition that the causative
J CATARACT REFRACT SURG-VOL
12, MAY 1986
249
Table 3. Incidence of hypopyon according to surgical day and order of implant cases.
Day and Order
Hypopyon Cases (percentage)
Day surgery performed Monday
48
Tuesday
10
Wednesday
18
Thursday
24
Order of cases 1st case of the day
76
2nd case of the day
24
3rd case of the day
0
factor may have been washed away after the first operation of the day, and therefore must involve the irrigation/aspiration tubing or the surgical instruments. This conclusion proved to be wrong. The instrument cleaning protocol called for ultrasonic cleaning of the instruments at the end of each surgical day. Commercially prepared distilled water was used as a cleaning solution to fill the Mettler ultrasonic cleaner every Monday. Because the manufacturer's literature stated "the longer the cleaning solution was used the better it would function," it was not discarded until Thursday. The instruments were then wrapped and autoclaved for the next surgical session. Analysis revealed that 72% of the hypopyons occurred on Monday and Thursday of the surgery schedule, even though four or five cases were done on each of the first four days of the week. The hypopyons became more frequent each day that the ultrasonic cleaning solution was reused. Cultures of this solution were positive for Kelbsiella pneumoniae. This finding led us to suspect that, even though the instruments were autoclaved prior to usage, some toxic agents derived from these organisms in older, contaminated ultrasonic cleaning solution must have caused the hypopyons. It also appeared that the contaminant was washed away after one usage. LABORATORY RESULTS Samples of the ultrasonic cleaning solution were sent to the Utah Center for Intraocular Lens Research for pathologic analysis. Analysis of the solution by two of the authors (JJR, DJA) at the Utah Center revealed the presence of endotoxin in the solution. The quantity of endotoxin present was equivalent to 0.5 mg/ml of E. coli lipopolysaccharide B, or 2.5 enzyme units. Because the surgical instruments were autoclaved after being placed in the solution, one author (FAR) felt it was essential to determine the heat stability of the endotoxin present in the solution. To simulate the clinical situation, 15 cc of the contaminated solution 250
was autoclaved at 120°C for 15 minutes, the identical temperature and exposure time to which the surgical instruments were exposed. Three rabbits were then anesthetized, and 0.5 cc of aqueous fluid was withdrawn from their anterior chambers. In rabbit 1,0.5 cc of bacteriostatic 0.9% NaCI solution was injected into the anterior chamber. In rabbit 2, 0.5 cc of nonautoclaved contaminated solution was injected into the anterior chamber, and in rabbit 3,0.5 cc of the autoclaved contaminated solution was injected. Twenty-four hours later, there was no appreciable inflammatory reaction in rabbit 1, which had received the sterile bacteriostatic solution. Rabbit 2, which received the nonautoclaved solution, developed a late hypopyon. Rabbit 3, which received the autoclaved contaminated solution, developed severe ciliary infection and iritis accompanied by the formation of pupillary membranes (Figure 1). The endotoxins produced by the bacteria in the solution obviously remained stable even after exposure to autoclave temperatures. DISCUSSION Authors have attributed sterile hypopyons to numerous causes, including IOL lens polishing compounds and rough IOL edges,l reaction to lens protein,2,3,4 IOL sterilization techniques,S laser iridotomy, 6 leukemia, 7 iridocyclitis, Bechet's syndrome, and other types of uveitis. To our knowledge, this is the first presentation of hypopyon caused by endotoxin contamination. Our investigation proved that, when instruments are placed in a three-day old ultrasonic bath contaminated by Gram negative bacteria, endotoxins produced by these microorganisms could coat the instruments. The bacterial contamination is effectively eliminated when the instruments are autoclaved, but the heat-stable endotoxin remains on the instruments, with the potential of causing serious ocular inflammation. The manufacturer states in the instruction manual for this model of ultrasonic cleaner, "Normally, the longer the same liquid is used, the better it cleans." This may be true, since the longer the ultrasonic bath solution is used the more the liquid becomes degassed, which allows for maximum cavitation and better cleaning efficiency. In a private communication to one author (FAR), the manufacturer now recommends that the bath solution "be changed every 2-3 days or when needed. " The key phrase here is "when needed," since bacterial contamination in a highly used ultrasonic cleaner could occur in as little as 24 hours. The institute now uses the same ultrasonic cleaner with ordinary tap water that is changed daily. The unit is also wiped with an alcohol-soaked sponge every day. There have been no sterile hypopyons in the more than 744 surgical procedures performed since October 22, 1984.
J CATARACT REFRACT SURG-VOL 12, MAY 1986
Ophthalmic surgeons currently using ultrasonic instrument cleaners should be aware that, in some models, following the manufacturer's recommended fluid changing procedures could result in endotoxic contamination of surgical instruments. This contamination could cause patients to develop severe inflammatory reactions in the immediate postoperative period. The authors recommend that ultrasonic cleansing fluid be changed after each use to avoid such contamination. REFERENCES 1. Meltzer DW: Sterile hypopyon following intraocular lens sur-
gery. Arch Ophthalmol 98:100-104, 1980 2. Apple DJ, Mamalis N, Steinmetz RL, Loftfield K, et al: Phacoanaphylactic endophthalmitis associated with extra-
3.
4.
5. 6.
7.
capsular cataract extraction and posterior chamber intraocular lens. Arch Ophthalmol 102:1528-1532, 1984 Apple DJ, Mamalis N, Steinmetz RL, Loftfield K, et al: Phacoanaphylactic endophthalmitis following ECCE and IOL implantation. Am Intra-Ocular Implant Soc J 10:423-424, 1984 Abrahams IW: Diagnosis and surgical management of phacoanaphylactic uveitis following extracapsular cataract extraction with intraocular lens implantation. Am Intra-Ocular Implant Soc J 11:444-447, 1985 Worst JGF: A retrospective view on the sterilization of intraocular lenses and the incidence of sterile hypopyon. Am Intra-Ocular Implant Soc J 6:10-12, 1980 Cohen JS, Bibler L, Tucker D: Hypopyon following laser iridotomy. Ophthalmic Surg 15:604-606, 1984 Abramson DH, Wachtel A, Watson CW, Jereb B, et al: Leukemic hypopyon. J Pediatr Ophthalmol Strabismus 18(3):42-44, 1981
J CATARACT REFRACT SURG-VOL 12, MAY 1986
251