- Email: [email protected]
Optimizing the antisepsis protocol: Effectiveness of 3 povidone–iodine 1.0% applications versus a single application of povidone–iodine 5.0%
400
LABORATORY SCIENCE
Optimizing the antisepsis protocol: Effectiveness of 3 povidone–iodine 1.0% applications versus a single application of povidone–iodine 5.0% Megan R. Silas, MD, Richard M. Schroeder, MD, Richard B. Thomson, PhD, William G. Myers, MD
Purpose: To determine the minimum effective concentration of povidone–iodine that reduces the bacterial load by 3-log10, the U.S. Food and Drug Administration requirement for antiseptic agents, and to study alternative dosing schedules of povidone– iodine to optimize its bactericidal effect. Setting: Microbiology Laboratory, Evanston Hospital, Evanston, Illinois, USA. Design: Experimental study. Methods: A standard 0.5 McFarland solution of Staphylococcus epidermidis was applied to blood agar plates. The plates were treated with a single application of povidone–iodine solutions from 10.0% to 0.1% to define the range of interest. Another set of plates received 3 applications of various povidone–iodine solutions. Microbial growth was evaluated after 24 hours. Standard
A
lthough cataract surgery is typically a safe and effective operation, endophthalmitis remains a rare but serious postoperative complication. The incidence of endophthalmitis after cataract surgery was reported in a Swedish review of 2002 to 2004 data at 0.48 per 1000 rate of infection and a retrospective analysis of all 2004 United States Medicare claim data at 1.11 per 1000.1,2 Visual outcomes worse than 20/200 were seen in one third of endophthalmitis cases in another recent Swedish study.3 The organism most often responsible for postoperative endophthalmitis is coagulase-negative staphylococci (CoNS) (Staphylococcus epidermidis) followed by Staphylococcus aureus, and Streptococcus species.4,5 These organisms reflect the bacterial flora of the eyelids and
deviations with 99.0% and 99.9% confidence intervals for each concentration were estimated and used to estimate the minimum concentration that reduced the colony counts by at least 3-log10.
Results: Povidone–iodine at 2.5% and higher concentrations was effective in eliminating S epidermidis with a single application. Three 30-second applications of povidone–iodine at concentrations of 0.7% and higher resulted in at least a 3-log10 reduction of colonies. Conclusions: Povidone–iodine 5.0% has been the standard of care for preoperative ocular antisepsis for 3 decades. Povidone– iodine 0.7% was as effective as a bactericidal agent when applied multiple times. This suggests povidone–iodine 1.0%, applied in three 30-second applications for preoperative surface disinfection might be as effective for preoperative antisepsis. J Cataract Refract Surg 2017; 43:400–404 Q 2017 ASCRS and ESCRS
conjunctiva.6 These pathogens can enter the anterior chamber through clear corneal incisions, thus exposing the eye to potential endophthalmitis.7 Prophylaxis against endophthalmitis via antisepsis of the eyelids and conjunctiva has therefore been an important area of study. Perioperative povidone–iodine use is not without risk; povidone–iodine 10.0% and 5.0% solutions have been shown to be toxic to the corneal epithelium when placed topically, and povidone–iodine 1.0% placed intracamerally is toxic to the corneal endothelium.8,9 In addition, povidone–iodine 5.0% and 2.5% cause edema and irritation in rabbit corneas, while 1.0% and 0.5% concentrations do not.10 Because of such experimental results, an effort has been made to apply more dilute concentrations of povidone–iodine before or during surgery. Shimada et al.6
Submitted: September 7, 2016 | Final revision submitted: December 12, 2016 | Accepted: January 12, 2017 From the Department of Graduate Medical Education (Silas), MacNeal Hospital, Berwyn, the Department of Pathology (Thomson), NorthShore University Health System, Evanston, and the Feinberg School of Medicine (Myers), Northwestern University, Chicago, Illinois, and the Department of Ophthalmology and Visual Sciences (Schroeder), Washington University School of Medicine, St. Louis, Missouri, USA. We acknowledge the passing of Henry F. Edelhauser, PhD, who contributed to the design of the study. Dan Eisenberg, MD, performed the statistical analysis. Corresponding author: William G. Myers, MD, Jesse Brown Veterans Administration Hospital, Ogden Building Room 5388, 820 South Damen Avenue, Chicago, Illinois 60612, USA. E-mail: [email protected]. Q 2017 ASCRS and ESCRS Published by Elsevier Inc.
0886-3350/$ - see frontmatter http://dx.doi.org/10.1016/j.jcrs.2017.01.007
EFFECTIVENESS OF DIFFERENT DOSAGES OF POVIDONE–IODINE
have shown that povidone–iodine 0.25%, when used as a surface irrigation agent throughout the procedure in addition to usual preoperative surface preparation, can reduce the number of bacteria found in aqueous sampling to zero at the completion of surgery. The importance of administering povidone–iodine at a concentration that is both safe and effective is critical.A In addition to the toxicities, higher concentrations of povidone–iodine cause burning and irritation while lower concentrations cause less discomfort.11,12 Most patients can tolerate low concentrations of povidone–iodine solution without prior use of anesthetic agents.12 Using lower concentrations of povidone–iodine would minimize the number of insults to the ocular surface and the potential for a barrier effect caused by some forms of topical anesthetic.13,14 MATERIALS AND METHODS Study Design A 0.5 McFarland suspension, approximately 1 108 colony forming units (CFU)/mL, of (CoNS) (S epidermidis RP62A) was prepared. A 0.25 mL of the 0.5 McFarland suspension was used to inoculate each 5.0% sheep blood agar plate by evenly flooding the surface. The fluid was allowed to soak into the agar for more than 2 minutes. This represents approximately 2.5 107 bacteria on the agar surface (0.25 mL of a 1 108 CFU/mL suspension), the typical bacterial load of infected conjunctiva. In the first exploratory experiment, 24 agar plates were covered with 0.25 mL of the 0.5 McFarland suspension as described above. As the bacterial solution was drying on the agar, povidone–iodine 10.0% (Betadine) was used to create the following diluted solutions by the following serial dilutions with Balanced Salt Solution (Alcon Laboratories, Inc.): 10.0%, 5.0%, 2.5%, 1.0%, 0.5%, 0.25%, and 0.1%. After all the solutions were made, 2 mL of each povidone–iodine solution was applied to the agar surfaces of 3 plates. After a 30-second exposure, excess povidone–iodine was poured off into a biohazard container. The 3 remaining plates were used as controls and were not treated with any solution. A second exploratory experiment was performed using 15 plates prepared as above. Three plates were covered with 2 mL of povidone–iodine 1.0% for 30 seconds. Another 3 plates received 6 mL of povidone–iodine 1.0% for 30 seconds. Three additional plates were treated with 2 mL of povidone–iodine 1.0% for 90 seconds. A final set of 3 plates was treated with 3 applications of 2 mL of povidone–iodine 1.0%, with each application lasting 30 seconds and with over 120 seconds between applications. The 3 remaining plates were used as controls. A final set of 55 plates inoculated with 0.25 mL of a 1:1000 dilution of the initial 0.5 McFarland suspension was applied to the surface of the agar. This dilution was used to better represent the bacterial load of a noninfected preoperative ocular surface. Given the starting bacterial suspension of 1 105 CFU/mL and inoculum volume of 0.25 mL, 2.5 104 CFU were inoculated to each plate prior to povidone–iodine application. Therefore, a 3-log10 reduction should lead to fewer than 25 colonies. The povidone–iodine 10.0% stock solution was serially diluted with a balanced salt solution to create a range of povidone–iodine solutions with concentrations of 1.0%, 0.9%, 0.8%, 0.7%, 0.6%, 0.5%, 0.4%, 0.3%, 0.2%, and 0.1%. After all the solutions were made, each plate was treated with 2 mL of povidone–iodine for 30 seconds and the process was repeated with each plate receiving a total of three 30-second applications of the specific povidone–iodine solution with at least 120 seconds between each application. The experiment was repeated 5 times using the same process to prepare the bacterial solution and povidone–iodine dilutions, thus each dilution was tested on 5 agar plates. The remaining 5 plates
401
were used as a control and were exposed to 2 mL of the balanced salt solution 3 times for 30 seconds. All plates were incubated at 36 C for 24 hours and then evaluated for growth. For the first 2 experiments, bacterial growth was categorized as heavy growth (confluent surface colonies), intermediate growth (O100 colonies), light growth (!100 colonies), or no growth. Because these experiments were performed to optimize the protocol for the final experiment, the exact number of bacterial colonies on each plate was not quantified. In the final experiment, a masked microbiology technician manually counted the colonies. A third experiment was used to further explore the effect of multiple application of povidone–iodine solution. For this study, a diluted McFarland inoculum was used to better replicate the bacterial load of a noninfected ocular surface. Statistical Analysis The basic mean and standard deviation was determined for each concentration using the 5 replicate samples. Confidence intervals of 99.0% and 99.9% were estimated, and the upper limit was used in the analysis for the concentration. An upper limit provides the worst-case result at the 0.01% and 0.001% level. The 0.01% cutoff exceeds the 0.05% t test values and the 0.001% level exceeds the 0.01% t test values, indicating statistical significance at the respective level. A regression line of upper limits was used to determine the intersection of the 3-log10 reduction in colony counts.
RESULTS Single Application of High Concentrations of Povidone– Iodine
The first experiment showed a dose-response relationship between the concentration of povidone–iodine solution used and the amount of bacterial growth (Figure 1). Single applications of highly concentrated povidone–iodine
Figure 1. Bacterial growth on plates inoculated with 2.5 107 CFU and then treated with a single application of various dilutions of povidone–iodine solution. A control plate is shown to show the effect of the diluent alone.
Volume 43 Issue 3 March 2017
402
EFFECTIVENESS OF DIFFERENT DOSAGES OF POVIDONE–IODINE
Alternative Dosing Schedules of Povidone–Iodine 1.0%
A second experiment was performed to further explore the bactericidal potential of povidone–iodine 1.0% (Figure 2). Tripling the volume of povidone–iodine by using 6 mL of solution did not lead to more effective bactericidal action, nor did tripling the application duration to 90 seconds. There was still light bacterial growth on all plates treated with these methods. However, when 3 separate 30-second applications of 2 mL of povidone–iodine 1.0% solution were used, there was greater bactericidal effectdthere was no growth on any plate. Determination of Effectiveness of Multiple Applications of Dilute Povidone–Iodine Solutions
The results of the third experiment showed no growth on all plates treated with 1.0%, 0.9%, and 0.7% solutions. Results for all concentrations of 0.7% and greater met the U.S. Food and Drug Administration (FDA) antiseptic efficacy requirement of at least a 3-log10 reduction in bacterial load because these plates all had fewer than 25 CFUs (Table 1 and Figures 3 and 4).A
Figure 2. Variations of povidone–iodine application on plates inoculated with 2.5 107 CFU. A: Two milliliters of povidone–iodine 1.0% solution applied for 30 seconds showing heavy growth. B: Six milliliters of povidone–iodine 1.0% solution applied for 30 seconds showing heavy growth. C: Two milliliters of povidone–iodine 1.0% solution applied for 90 seconds showing heavy growth. D: Two milliliters of povidone–iodine 1.0% solution applied 3 times for 30 seconds each, showing only light growth.
DISCUSSION Povidone–iodine has been used as a medical disinfectant for decades.15 It was introduced as a method of treating bacterial conjunctivitis by Isenberg et al. in 2002.12 Povidone–iodine, a hydrophobic iodophor, binds to the cell membrane of bacteria and releases molecular iodine into the cell, where it has a cytotoxic effect.10,16 Povidone (polyvinylpyrrolidone), a synthetic polymer, limits the amount of free iodine present in solution. Povidone–iodine 10.0% contains 10.0% povidone, not iodine, in aqueous solution with total titratable iodine content approximately 10.0%. An unusual chemical property of povidone–iodine iodophors is that the concentration of free iodine increases
solution (10.0%, 5.0%, and 2.5%) eradicated bacteria completely; all plates treated with these solutions had no growth. Single applications of lower concentrations (0.1%, 0.5%, and 0.25%) allowed intermediate or heavy growth. The povidone–iodine 1.0% solution application still allowed light bacterial growth on all 3 plates when applied 1 time for 30 seconds.
Table 1. Colony-forming units per plate of 5 plates after three 30-second applications of various concentrations of povidone–iodine. Povidone–Iodine (%) Parameter
0.00
0.1
0.2
0.3
0.4
0.5
0.6
Row 1
25000
3524
2143
1161
704
220
Row 2
25000
2799
3390
502
618
Row 3
25000
2829
2316
1471
409
Row 4
25000
3103
1674
957
973
172
96
0
4
0
0
Row 5
25000
3886
2088
1042
947
218
89
0
10
0
0
Mean
25000
3228.2
2322.2
1026.6
730.2
208.2
49.2
0.4
0
0
Max
25000
3886
3390
1471
973
243
96
2
0
0
SD 99% CI Max 99% Log (max99%) 3-log10?*
0.0
0.7
0.8
0.9
1.0
0
2
0
0
0
188
36
0
0
0
0
243
25
0
0
0
0
2.8 10
468.9
641.7
352.1
235.8
28.1
41.7
0.9
4.4
0.0
0.0
d
965.5
1321.4
724.9
485.5
57.9
85.9
1.8
9.0
0.0
0.0
25000.0
4193.7
3643.6
1751.5
1215.7
266.1
135.1
2.2
11.8
0.0
0.0
4.4
3.6
3.6
3.2
3.1
2.4
2.1
0.4
1.1
0.0
0.0
Yes
Yes
Yes
Yes
No
No
No
CI Z confidence interval *3-log intercept 99%: 0.65; 3-log intercept 99.9%: 0.68
Volume 43 Issue 3 March 2017
No
No
No
No
EFFECTIVENESS OF DIFFERENT DOSAGES OF POVIDONE–IODINE
Figure 3. Dilution trial to determine minimum effective concentration of povidone–iodine when applied 3 times to plates inoculated with 2.5 104 CFU.
with dilution of the complexed povidone–iodine, which helps explain the occasional bacterial contamination of povidone–iodine 10.0% bottles.16 In general, higher concentrations of free iodine are more reliable bactericidal agents than weaker concentrations.10,17 The known epithelial toxicity of povidone–iodine 5.0% solution to the cornea and conjunctiva has prompted an effort to achieve the same bactericidal effect with lower concentrations. This study found that povidone–iodine at any
403
given concentration was a more effective antiseptic when it was applied multiple times. Each application reduces further the bacterial load for the following application. This finding supports the practice of using a lower concentration of povidone–iodine flushed onto the ocular surface multiple times before the procedure and also applied throughout the case at an even greater dilution. A weakness of this study is that only 1 frequent pathogen was studied. Povidone–iodine is effective at higher concentrations against most pathogens, including many viruses, gram-positive and gram-negative bacteria, mycobacteria, and most fungi. However, lower concentrations of povidone–iodine have a greater free iodine concentration, the active killing agent.16 Therefore, it is reasonable to assume similar results with these other pathogens. Our study used agar plates as a surrogate for the mucosal membrane surface of the orbit. It is possible that the in vitro model does not account for all the biologic variability in the in vivo situation. Some studies have shown different effects in vivo than in vitro, where contaminants such as mucus and biofilms might be present; that is, higher concentrations of povidone–iodine were more effective.17 Further studies in a human surgical model will be required to determine whether pulsed flushing with dilute solutions can be more effective because flushing has been found to be superior to application of drops to the ocular surface.18 Even if the in vitro results were replicated in samples from human conjunctiva, one could still only conclusively infer a reduction in the incidence of endophthalmitis after analysis of a large-scale study of patients with this new iterative regimen. However, it should be noted that when a single 3-minute application of povidone–iodine 5.0% was recommended several decades ago, the process was based on laboratory studies and had not yet been shown to reduce endophthalmitis. Those studies did, however, lead to near universal acceptance of preoperative povidone–iodine antisepsis. This study was designed to further the understanding of the antibacterial effects of povidone–iodine at lower concentrations using an in vitro model. The application of
Figure 4. Colonies per plate remaining after 3 applications of various concentrations of povidone– iodine. This shows that 0.7%, 0.8%, 0.9%, and 1.0% povidone– iodine reduced the number of colonies to less than 25 (green zone), fulfilling the FDA requirement for sufficient antisepsis (3-log10 reduction in bacterial counts).
Volume 43 Issue 3 March 2017
404
EFFECTIVENESS OF DIFFERENT DOSAGES OF POVIDONE–IODINE
povidone–iodine is not limited only to the preoperative time period. Shimada et al.6,19 performed a series of trials using povidone–iodine diluted to a 0.25% concentration in a balanced salt solution used intraoperatively on the ocular surface and found a decrease in positive cultures from the anterior chamber. Our finding that multiple applications of povidone–iodine at low concentrations can limit bacterial survival is consistent with their results. In conclusion, in an in vitro environment, povidone– iodine at a 5.0% concentration was an effective bactericidal agent. Povidone–iodine 1.0% applied 3 times, spaced at least 2 minutes apart, was as effective in an in vitro model. This could be accomplished in an in vivo situation by dosing once in the preoperative area, once more on entering the procedure room, and finally a third time after the lid speculum is placed. Further research is needed to determine whether this method should be used before other intraocular procedures, such as intravitreal injections.
WHAT WAS KNOWN A single application of povidone–iodine 5.0% allowed to remain for 3 minutes is the current standard of care for preoperative antisepsis in ocular surgery and for intravitreal injections. Povidone–iodine 5.0% solution causes more irritation than povidone–iodine 1.0%.
WHAT THIS PAPER ADDS Three applications of povidone–iodine 0.7% and higher led to a 3-log10 units reduction in CFUs, the FDA requirement for surface antisepsis. Povidone–iodine 1.0% applied for 30 seconds 3 times, separated by 2 minutes, was as effective in reducing the bacterial count 3-log10 units in vitro as povidone–iodine 5.0% solution applied once.
REFERENCES
€m M, Wejde G, Stenevi U, Thorburn W, Montan P. Endophthalmi1. Lundstro tis after cataract surgery; a nationwide prospective study evaluating incidence in relation to incision type and location. Ophthalmology 2007; 114:866–870 2. Keay L, Gower EW, Cassard SD, Tielsch JM, Schein OD. Postcataract surgery endophthalmitis in the United States; analysis of the complete 2003 to 2004 Medicare database of cataract surgeries. Ophthalmology 2012; 119:914–922. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3343208/pdf/nihms340631.pdf. Accessed January 24, 2017 €m M, Stenevi U, Montan P. Six-year incidence of endoph3. Friling E, Lundstro thalmitis after cataract surgery: Swedish national study. J Cataract Refract Surg 2013; 39:15–21 4. Vaziri K, Schwartz SG, Kishor K, Flynn HW Jr. Endophthalmitis: state of the art. Clin Ophthalmol 2015; 9:95–108. Available at: http://www.ncbi.nlm. nih.gov/pmc/articles/PMC4293922/pdf/opth-9-095.pdf. Accessed January 24, 2017 5. Nentwich MM, Ta CN, Kreutzer TC, Li B, Schwarzbach F, Yactayo~o de Kaspar H. Incidence of postoperative Miranda YM, Kampik A, Min endophthalmitis from 1990 to 2009 using povidone–iodine but no intracameral antibiotics at a single academic institution. J Cataract Refract Surg 2015; 41:58–66
Volume 43 Issue 3 March 2017
6. Shimada H, Arai S, Nakashizuka H, Hattori T, Yuzawa M. Reduction of anterior chamber contamination rate after cataract surgery by intraoperative surface irrigation with 0.25% povidone–iodine. Am J Ophthalmol 2011; 151:11–17 7. Panahibazaz M, Moosavian M, Khataminia G, Feghhi M, Yazdi F, Abbasi Montazeri E. Sub-conjunctival injection of antibiotics vs. povidone-iodine drop on bacterial colonies in phacoemulsification cataract surgery. Jundishapur J Microbiol 2014; 7 (9):e13108. Available at: https://www.ncbi. nlm.nih.gov/pmc/articles/PMC4255380/pdf/jjm-07-13108.pdf. Accessed January 24, 2017 8. Shibata Y, Tanaka Y, Tomita T, Taogoshi T, Kimura Y, Chikama T, Kihira K. Evaluation of corneal damage caused by iodine preparations using human corneal epithelial cells. Jpn J Ophthalmol 2014; 58:522–527 9. Naor J, Savion N, Blumenthal M, Assia EI. Corneal endothelial cytotoxicity of diluted povidone–iodine. J Cataract Refract Surg 2001; 27:941–947 10. Jiang J, Wu M, Shen T. The toxic effect of different concentrations of povidone iodine on the rabbit's cornea. Cutan Ocul Toxicol 2009; 28:119–124 11. Apt L, Isenberg S, Yoshimori R, Paez JH. Chemical preparation of the eye in ophthalmic surgery. III. Effect of povidone-iodine on the conjunctiva. Arch Ophthalmol 1984; 102:728–729 12. Isenberg SJ, Apt L, Valenton M, Del Signore M, Cubillan L, Labrador MA, Chan P, Berman NG. A controlled trial of povidone–iodine to treat infectious conjunctivitis in children. Am J Ophthalmol 2002; 134:681–688 13. Boden JH, Myers ML, Lee T, Bushley DM, Torres MF. Effect of lidocaine gel on povidone–iodine antisepsis and microbial survival. J Cataract Refract Surg 2008; 34:1773–1775 14. Schroeder RM, Silas MR, Thomson RM, Myers WG. The danger of viscous gel anesthetic use prior to povidone–iodine antisepsis: best practice recommendation. J Cataract Refract Surg 2015; 41:2333–2335 15. Maumenee AE, Michler RC. Sterility of the operative field after ocular surgery. Trans Pac Coast Oto-Ophthalmol Soc 1951; 32:172–183 16. Gottardi W. Iodine and iodine compounds. In: Block SS, ed, Disinfection, Sterilization, and Preservation, 4th ed. Philadelphia, PA, Lea & Febiger, 1991; 152–181 17. Ferguson AW, Scott JA, McGavigan J, Elton RA, McLean J, Schmidt U, Kelkar R, Dhillon B. Comparison of 5% povidone-iodine solution against 1% povidone-iodine solution in preoperative cataract surgery antisepsis: a prospective randomized double blind study. Br J Ophthalmol 2003; 87:163–167. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC1771501/pdf/bjo08700163.pdf. Accessed January 24, 2017 18. Safar A, Dellimore MC. The effect of povidone iodine flush versus drops on conjunctival colonization before intravitreal injections. Int Ophthalmol 2007; 27:307–312 19. Shimada H, Nakashizuka H, Hattori T, Kitagawa Y, Manabe A, Otani K, Yuzawa M. Reducing bacterial contamination inside fluid catch bag in 25gauge vitrectomy by use of 0.25 % povidone-iodine ocular surface irrigation. Int Ophthalmol 2013; 33:35–38 OTHER CITED MATERIAL A. U.S. Food and Drug Administration. Department of Health and Human Services. 21 CFR Part 310. Safety and Effectiveness of Health Care Antiseptics; Topical Antimicrobial Drug Products for Over-the-Counter Human Use; Proposed Amendment of the Tentative Final Monograph; Reopening of Administrative Record. Federal Register 2015; 80 (84). Available at: https://www. gpo.gov/fdsys/pkg/FR-2015-05-01/pdf/2015-10174.pdf. Accessed January 24, 2017
Disclosure: Dr. Myers is a consultant to Leiter's Enterprise, Inc. LLC. None of the other authors has a financial or proprietary interest in any material or method mentioned.
First author: Megan R. Silas, MD Department of Graduate Medical Education, MacNeal Hospital, Berwyn, Illinois, USA
LABORATORY SCIENCE
Optimizing the antisepsis protocol: Effectiveness of 3 povidone–iodine 1.0% applications versus a single application of povidone–iodine 5.0% Megan R. Silas, MD, Richard M. Schroeder, MD, Richard B. Thomson, PhD, William G. Myers, MD
Purpose: To determine the minimum effective concentration of povidone–iodine that reduces the bacterial load by 3-log10, the U.S. Food and Drug Administration requirement for antiseptic agents, and to study alternative dosing schedules of povidone– iodine to optimize its bactericidal effect. Setting: Microbiology Laboratory, Evanston Hospital, Evanston, Illinois, USA. Design: Experimental study. Methods: A standard 0.5 McFarland solution of Staphylococcus epidermidis was applied to blood agar plates. The plates were treated with a single application of povidone–iodine solutions from 10.0% to 0.1% to define the range of interest. Another set of plates received 3 applications of various povidone–iodine solutions. Microbial growth was evaluated after 24 hours. Standard
A
lthough cataract surgery is typically a safe and effective operation, endophthalmitis remains a rare but serious postoperative complication. The incidence of endophthalmitis after cataract surgery was reported in a Swedish review of 2002 to 2004 data at 0.48 per 1000 rate of infection and a retrospective analysis of all 2004 United States Medicare claim data at 1.11 per 1000.1,2 Visual outcomes worse than 20/200 were seen in one third of endophthalmitis cases in another recent Swedish study.3 The organism most often responsible for postoperative endophthalmitis is coagulase-negative staphylococci (CoNS) (Staphylococcus epidermidis) followed by Staphylococcus aureus, and Streptococcus species.4,5 These organisms reflect the bacterial flora of the eyelids and
deviations with 99.0% and 99.9% confidence intervals for each concentration were estimated and used to estimate the minimum concentration that reduced the colony counts by at least 3-log10.
Results: Povidone–iodine at 2.5% and higher concentrations was effective in eliminating S epidermidis with a single application. Three 30-second applications of povidone–iodine at concentrations of 0.7% and higher resulted in at least a 3-log10 reduction of colonies. Conclusions: Povidone–iodine 5.0% has been the standard of care for preoperative ocular antisepsis for 3 decades. Povidone– iodine 0.7% was as effective as a bactericidal agent when applied multiple times. This suggests povidone–iodine 1.0%, applied in three 30-second applications for preoperative surface disinfection might be as effective for preoperative antisepsis. J Cataract Refract Surg 2017; 43:400–404 Q 2017 ASCRS and ESCRS
conjunctiva.6 These pathogens can enter the anterior chamber through clear corneal incisions, thus exposing the eye to potential endophthalmitis.7 Prophylaxis against endophthalmitis via antisepsis of the eyelids and conjunctiva has therefore been an important area of study. Perioperative povidone–iodine use is not without risk; povidone–iodine 10.0% and 5.0% solutions have been shown to be toxic to the corneal epithelium when placed topically, and povidone–iodine 1.0% placed intracamerally is toxic to the corneal endothelium.8,9 In addition, povidone–iodine 5.0% and 2.5% cause edema and irritation in rabbit corneas, while 1.0% and 0.5% concentrations do not.10 Because of such experimental results, an effort has been made to apply more dilute concentrations of povidone–iodine before or during surgery. Shimada et al.6
Submitted: September 7, 2016 | Final revision submitted: December 12, 2016 | Accepted: January 12, 2017 From the Department of Graduate Medical Education (Silas), MacNeal Hospital, Berwyn, the Department of Pathology (Thomson), NorthShore University Health System, Evanston, and the Feinberg School of Medicine (Myers), Northwestern University, Chicago, Illinois, and the Department of Ophthalmology and Visual Sciences (Schroeder), Washington University School of Medicine, St. Louis, Missouri, USA. We acknowledge the passing of Henry F. Edelhauser, PhD, who contributed to the design of the study. Dan Eisenberg, MD, performed the statistical analysis. Corresponding author: William G. Myers, MD, Jesse Brown Veterans Administration Hospital, Ogden Building Room 5388, 820 South Damen Avenue, Chicago, Illinois 60612, USA. E-mail: [email protected]. Q 2017 ASCRS and ESCRS Published by Elsevier Inc.
0886-3350/$ - see frontmatter http://dx.doi.org/10.1016/j.jcrs.2017.01.007
EFFECTIVENESS OF DIFFERENT DOSAGES OF POVIDONE–IODINE
have shown that povidone–iodine 0.25%, when used as a surface irrigation agent throughout the procedure in addition to usual preoperative surface preparation, can reduce the number of bacteria found in aqueous sampling to zero at the completion of surgery. The importance of administering povidone–iodine at a concentration that is both safe and effective is critical.A In addition to the toxicities, higher concentrations of povidone–iodine cause burning and irritation while lower concentrations cause less discomfort.11,12 Most patients can tolerate low concentrations of povidone–iodine solution without prior use of anesthetic agents.12 Using lower concentrations of povidone–iodine would minimize the number of insults to the ocular surface and the potential for a barrier effect caused by some forms of topical anesthetic.13,14 MATERIALS AND METHODS Study Design A 0.5 McFarland suspension, approximately 1 108 colony forming units (CFU)/mL, of (CoNS) (S epidermidis RP62A) was prepared. A 0.25 mL of the 0.5 McFarland suspension was used to inoculate each 5.0% sheep blood agar plate by evenly flooding the surface. The fluid was allowed to soak into the agar for more than 2 minutes. This represents approximately 2.5 107 bacteria on the agar surface (0.25 mL of a 1 108 CFU/mL suspension), the typical bacterial load of infected conjunctiva. In the first exploratory experiment, 24 agar plates were covered with 0.25 mL of the 0.5 McFarland suspension as described above. As the bacterial solution was drying on the agar, povidone–iodine 10.0% (Betadine) was used to create the following diluted solutions by the following serial dilutions with Balanced Salt Solution (Alcon Laboratories, Inc.): 10.0%, 5.0%, 2.5%, 1.0%, 0.5%, 0.25%, and 0.1%. After all the solutions were made, 2 mL of each povidone–iodine solution was applied to the agar surfaces of 3 plates. After a 30-second exposure, excess povidone–iodine was poured off into a biohazard container. The 3 remaining plates were used as controls and were not treated with any solution. A second exploratory experiment was performed using 15 plates prepared as above. Three plates were covered with 2 mL of povidone–iodine 1.0% for 30 seconds. Another 3 plates received 6 mL of povidone–iodine 1.0% for 30 seconds. Three additional plates were treated with 2 mL of povidone–iodine 1.0% for 90 seconds. A final set of 3 plates was treated with 3 applications of 2 mL of povidone–iodine 1.0%, with each application lasting 30 seconds and with over 120 seconds between applications. The 3 remaining plates were used as controls. A final set of 55 plates inoculated with 0.25 mL of a 1:1000 dilution of the initial 0.5 McFarland suspension was applied to the surface of the agar. This dilution was used to better represent the bacterial load of a noninfected preoperative ocular surface. Given the starting bacterial suspension of 1 105 CFU/mL and inoculum volume of 0.25 mL, 2.5 104 CFU were inoculated to each plate prior to povidone–iodine application. Therefore, a 3-log10 reduction should lead to fewer than 25 colonies. The povidone–iodine 10.0% stock solution was serially diluted with a balanced salt solution to create a range of povidone–iodine solutions with concentrations of 1.0%, 0.9%, 0.8%, 0.7%, 0.6%, 0.5%, 0.4%, 0.3%, 0.2%, and 0.1%. After all the solutions were made, each plate was treated with 2 mL of povidone–iodine for 30 seconds and the process was repeated with each plate receiving a total of three 30-second applications of the specific povidone–iodine solution with at least 120 seconds between each application. The experiment was repeated 5 times using the same process to prepare the bacterial solution and povidone–iodine dilutions, thus each dilution was tested on 5 agar plates. The remaining 5 plates
401
were used as a control and were exposed to 2 mL of the balanced salt solution 3 times for 30 seconds. All plates were incubated at 36 C for 24 hours and then evaluated for growth. For the first 2 experiments, bacterial growth was categorized as heavy growth (confluent surface colonies), intermediate growth (O100 colonies), light growth (!100 colonies), or no growth. Because these experiments were performed to optimize the protocol for the final experiment, the exact number of bacterial colonies on each plate was not quantified. In the final experiment, a masked microbiology technician manually counted the colonies. A third experiment was used to further explore the effect of multiple application of povidone–iodine solution. For this study, a diluted McFarland inoculum was used to better replicate the bacterial load of a noninfected ocular surface. Statistical Analysis The basic mean and standard deviation was determined for each concentration using the 5 replicate samples. Confidence intervals of 99.0% and 99.9% were estimated, and the upper limit was used in the analysis for the concentration. An upper limit provides the worst-case result at the 0.01% and 0.001% level. The 0.01% cutoff exceeds the 0.05% t test values and the 0.001% level exceeds the 0.01% t test values, indicating statistical significance at the respective level. A regression line of upper limits was used to determine the intersection of the 3-log10 reduction in colony counts.
RESULTS Single Application of High Concentrations of Povidone– Iodine
The first experiment showed a dose-response relationship between the concentration of povidone–iodine solution used and the amount of bacterial growth (Figure 1). Single applications of highly concentrated povidone–iodine
Figure 1. Bacterial growth on plates inoculated with 2.5 107 CFU and then treated with a single application of various dilutions of povidone–iodine solution. A control plate is shown to show the effect of the diluent alone.
Volume 43 Issue 3 March 2017
402
EFFECTIVENESS OF DIFFERENT DOSAGES OF POVIDONE–IODINE
Alternative Dosing Schedules of Povidone–Iodine 1.0%
A second experiment was performed to further explore the bactericidal potential of povidone–iodine 1.0% (Figure 2). Tripling the volume of povidone–iodine by using 6 mL of solution did not lead to more effective bactericidal action, nor did tripling the application duration to 90 seconds. There was still light bacterial growth on all plates treated with these methods. However, when 3 separate 30-second applications of 2 mL of povidone–iodine 1.0% solution were used, there was greater bactericidal effectdthere was no growth on any plate. Determination of Effectiveness of Multiple Applications of Dilute Povidone–Iodine Solutions
The results of the third experiment showed no growth on all plates treated with 1.0%, 0.9%, and 0.7% solutions. Results for all concentrations of 0.7% and greater met the U.S. Food and Drug Administration (FDA) antiseptic efficacy requirement of at least a 3-log10 reduction in bacterial load because these plates all had fewer than 25 CFUs (Table 1 and Figures 3 and 4).A
Figure 2. Variations of povidone–iodine application on plates inoculated with 2.5 107 CFU. A: Two milliliters of povidone–iodine 1.0% solution applied for 30 seconds showing heavy growth. B: Six milliliters of povidone–iodine 1.0% solution applied for 30 seconds showing heavy growth. C: Two milliliters of povidone–iodine 1.0% solution applied for 90 seconds showing heavy growth. D: Two milliliters of povidone–iodine 1.0% solution applied 3 times for 30 seconds each, showing only light growth.
DISCUSSION Povidone–iodine has been used as a medical disinfectant for decades.15 It was introduced as a method of treating bacterial conjunctivitis by Isenberg et al. in 2002.12 Povidone–iodine, a hydrophobic iodophor, binds to the cell membrane of bacteria and releases molecular iodine into the cell, where it has a cytotoxic effect.10,16 Povidone (polyvinylpyrrolidone), a synthetic polymer, limits the amount of free iodine present in solution. Povidone–iodine 10.0% contains 10.0% povidone, not iodine, in aqueous solution with total titratable iodine content approximately 10.0%. An unusual chemical property of povidone–iodine iodophors is that the concentration of free iodine increases
solution (10.0%, 5.0%, and 2.5%) eradicated bacteria completely; all plates treated with these solutions had no growth. Single applications of lower concentrations (0.1%, 0.5%, and 0.25%) allowed intermediate or heavy growth. The povidone–iodine 1.0% solution application still allowed light bacterial growth on all 3 plates when applied 1 time for 30 seconds.
Table 1. Colony-forming units per plate of 5 plates after three 30-second applications of various concentrations of povidone–iodine. Povidone–Iodine (%) Parameter
0.00
0.1
0.2
0.3
0.4
0.5
0.6
Row 1
25000
3524
2143
1161
704
220
Row 2
25000
2799
3390
502
618
Row 3
25000
2829
2316
1471
409
Row 4
25000
3103
1674
957
973
172
96
0
4
0
0
Row 5
25000
3886
2088
1042
947
218
89
0
10
0
0
Mean
25000
3228.2
2322.2
1026.6
730.2
208.2
49.2
0.4
0
0
Max
25000
3886
3390
1471
973
243
96
2
0
0
SD 99% CI Max 99% Log (max99%) 3-log10?*
0.0
0.7
0.8
0.9
1.0
0
2
0
0
0
188
36
0
0
0
0
243
25
0
0
0
0
2.8 10
468.9
641.7
352.1
235.8
28.1
41.7
0.9
4.4
0.0
0.0
d
965.5
1321.4
724.9
485.5
57.9
85.9
1.8
9.0
0.0
0.0
25000.0
4193.7
3643.6
1751.5
1215.7
266.1
135.1
2.2
11.8
0.0
0.0
4.4
3.6
3.6
3.2
3.1
2.4
2.1
0.4
1.1
0.0
0.0
Yes
Yes
Yes
Yes
No
No
No
CI Z confidence interval *3-log intercept 99%: 0.65; 3-log intercept 99.9%: 0.68
Volume 43 Issue 3 March 2017
No
No
No
No
EFFECTIVENESS OF DIFFERENT DOSAGES OF POVIDONE–IODINE
Figure 3. Dilution trial to determine minimum effective concentration of povidone–iodine when applied 3 times to plates inoculated with 2.5 104 CFU.
with dilution of the complexed povidone–iodine, which helps explain the occasional bacterial contamination of povidone–iodine 10.0% bottles.16 In general, higher concentrations of free iodine are more reliable bactericidal agents than weaker concentrations.10,17 The known epithelial toxicity of povidone–iodine 5.0% solution to the cornea and conjunctiva has prompted an effort to achieve the same bactericidal effect with lower concentrations. This study found that povidone–iodine at any
403
given concentration was a more effective antiseptic when it was applied multiple times. Each application reduces further the bacterial load for the following application. This finding supports the practice of using a lower concentration of povidone–iodine flushed onto the ocular surface multiple times before the procedure and also applied throughout the case at an even greater dilution. A weakness of this study is that only 1 frequent pathogen was studied. Povidone–iodine is effective at higher concentrations against most pathogens, including many viruses, gram-positive and gram-negative bacteria, mycobacteria, and most fungi. However, lower concentrations of povidone–iodine have a greater free iodine concentration, the active killing agent.16 Therefore, it is reasonable to assume similar results with these other pathogens. Our study used agar plates as a surrogate for the mucosal membrane surface of the orbit. It is possible that the in vitro model does not account for all the biologic variability in the in vivo situation. Some studies have shown different effects in vivo than in vitro, where contaminants such as mucus and biofilms might be present; that is, higher concentrations of povidone–iodine were more effective.17 Further studies in a human surgical model will be required to determine whether pulsed flushing with dilute solutions can be more effective because flushing has been found to be superior to application of drops to the ocular surface.18 Even if the in vitro results were replicated in samples from human conjunctiva, one could still only conclusively infer a reduction in the incidence of endophthalmitis after analysis of a large-scale study of patients with this new iterative regimen. However, it should be noted that when a single 3-minute application of povidone–iodine 5.0% was recommended several decades ago, the process was based on laboratory studies and had not yet been shown to reduce endophthalmitis. Those studies did, however, lead to near universal acceptance of preoperative povidone–iodine antisepsis. This study was designed to further the understanding of the antibacterial effects of povidone–iodine at lower concentrations using an in vitro model. The application of
Figure 4. Colonies per plate remaining after 3 applications of various concentrations of povidone– iodine. This shows that 0.7%, 0.8%, 0.9%, and 1.0% povidone– iodine reduced the number of colonies to less than 25 (green zone), fulfilling the FDA requirement for sufficient antisepsis (3-log10 reduction in bacterial counts).
Volume 43 Issue 3 March 2017
404
EFFECTIVENESS OF DIFFERENT DOSAGES OF POVIDONE–IODINE
povidone–iodine is not limited only to the preoperative time period. Shimada et al.6,19 performed a series of trials using povidone–iodine diluted to a 0.25% concentration in a balanced salt solution used intraoperatively on the ocular surface and found a decrease in positive cultures from the anterior chamber. Our finding that multiple applications of povidone–iodine at low concentrations can limit bacterial survival is consistent with their results. In conclusion, in an in vitro environment, povidone– iodine at a 5.0% concentration was an effective bactericidal agent. Povidone–iodine 1.0% applied 3 times, spaced at least 2 minutes apart, was as effective in an in vitro model. This could be accomplished in an in vivo situation by dosing once in the preoperative area, once more on entering the procedure room, and finally a third time after the lid speculum is placed. Further research is needed to determine whether this method should be used before other intraocular procedures, such as intravitreal injections.
WHAT WAS KNOWN A single application of povidone–iodine 5.0% allowed to remain for 3 minutes is the current standard of care for preoperative antisepsis in ocular surgery and for intravitreal injections. Povidone–iodine 5.0% solution causes more irritation than povidone–iodine 1.0%.
WHAT THIS PAPER ADDS Three applications of povidone–iodine 0.7% and higher led to a 3-log10 units reduction in CFUs, the FDA requirement for surface antisepsis. Povidone–iodine 1.0% applied for 30 seconds 3 times, separated by 2 minutes, was as effective in reducing the bacterial count 3-log10 units in vitro as povidone–iodine 5.0% solution applied once.
REFERENCES
€m M, Wejde G, Stenevi U, Thorburn W, Montan P. Endophthalmi1. Lundstro tis after cataract surgery; a nationwide prospective study evaluating incidence in relation to incision type and location. Ophthalmology 2007; 114:866–870 2. Keay L, Gower EW, Cassard SD, Tielsch JM, Schein OD. Postcataract surgery endophthalmitis in the United States; analysis of the complete 2003 to 2004 Medicare database of cataract surgeries. Ophthalmology 2012; 119:914–922. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3343208/pdf/nihms340631.pdf. Accessed January 24, 2017 €m M, Stenevi U, Montan P. Six-year incidence of endoph3. Friling E, Lundstro thalmitis after cataract surgery: Swedish national study. J Cataract Refract Surg 2013; 39:15–21 4. Vaziri K, Schwartz SG, Kishor K, Flynn HW Jr. Endophthalmitis: state of the art. Clin Ophthalmol 2015; 9:95–108. Available at: http://www.ncbi.nlm. nih.gov/pmc/articles/PMC4293922/pdf/opth-9-095.pdf. Accessed January 24, 2017 5. Nentwich MM, Ta CN, Kreutzer TC, Li B, Schwarzbach F, Yactayo~o de Kaspar H. Incidence of postoperative Miranda YM, Kampik A, Min endophthalmitis from 1990 to 2009 using povidone–iodine but no intracameral antibiotics at a single academic institution. J Cataract Refract Surg 2015; 41:58–66
Volume 43 Issue 3 March 2017
6. Shimada H, Arai S, Nakashizuka H, Hattori T, Yuzawa M. Reduction of anterior chamber contamination rate after cataract surgery by intraoperative surface irrigation with 0.25% povidone–iodine. Am J Ophthalmol 2011; 151:11–17 7. Panahibazaz M, Moosavian M, Khataminia G, Feghhi M, Yazdi F, Abbasi Montazeri E. Sub-conjunctival injection of antibiotics vs. povidone-iodine drop on bacterial colonies in phacoemulsification cataract surgery. Jundishapur J Microbiol 2014; 7 (9):e13108. Available at: https://www.ncbi. nlm.nih.gov/pmc/articles/PMC4255380/pdf/jjm-07-13108.pdf. Accessed January 24, 2017 8. Shibata Y, Tanaka Y, Tomita T, Taogoshi T, Kimura Y, Chikama T, Kihira K. Evaluation of corneal damage caused by iodine preparations using human corneal epithelial cells. Jpn J Ophthalmol 2014; 58:522–527 9. Naor J, Savion N, Blumenthal M, Assia EI. Corneal endothelial cytotoxicity of diluted povidone–iodine. J Cataract Refract Surg 2001; 27:941–947 10. Jiang J, Wu M, Shen T. The toxic effect of different concentrations of povidone iodine on the rabbit's cornea. Cutan Ocul Toxicol 2009; 28:119–124 11. Apt L, Isenberg S, Yoshimori R, Paez JH. Chemical preparation of the eye in ophthalmic surgery. III. Effect of povidone-iodine on the conjunctiva. Arch Ophthalmol 1984; 102:728–729 12. Isenberg SJ, Apt L, Valenton M, Del Signore M, Cubillan L, Labrador MA, Chan P, Berman NG. A controlled trial of povidone–iodine to treat infectious conjunctivitis in children. Am J Ophthalmol 2002; 134:681–688 13. Boden JH, Myers ML, Lee T, Bushley DM, Torres MF. Effect of lidocaine gel on povidone–iodine antisepsis and microbial survival. J Cataract Refract Surg 2008; 34:1773–1775 14. Schroeder RM, Silas MR, Thomson RM, Myers WG. The danger of viscous gel anesthetic use prior to povidone–iodine antisepsis: best practice recommendation. J Cataract Refract Surg 2015; 41:2333–2335 15. Maumenee AE, Michler RC. Sterility of the operative field after ocular surgery. Trans Pac Coast Oto-Ophthalmol Soc 1951; 32:172–183 16. Gottardi W. Iodine and iodine compounds. In: Block SS, ed, Disinfection, Sterilization, and Preservation, 4th ed. Philadelphia, PA, Lea & Febiger, 1991; 152–181 17. Ferguson AW, Scott JA, McGavigan J, Elton RA, McLean J, Schmidt U, Kelkar R, Dhillon B. Comparison of 5% povidone-iodine solution against 1% povidone-iodine solution in preoperative cataract surgery antisepsis: a prospective randomized double blind study. Br J Ophthalmol 2003; 87:163–167. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC1771501/pdf/bjo08700163.pdf. Accessed January 24, 2017 18. Safar A, Dellimore MC. The effect of povidone iodine flush versus drops on conjunctival colonization before intravitreal injections. Int Ophthalmol 2007; 27:307–312 19. Shimada H, Nakashizuka H, Hattori T, Kitagawa Y, Manabe A, Otani K, Yuzawa M. Reducing bacterial contamination inside fluid catch bag in 25gauge vitrectomy by use of 0.25 % povidone-iodine ocular surface irrigation. Int Ophthalmol 2013; 33:35–38 OTHER CITED MATERIAL A. U.S. Food and Drug Administration. Department of Health and Human Services. 21 CFR Part 310. Safety and Effectiveness of Health Care Antiseptics; Topical Antimicrobial Drug Products for Over-the-Counter Human Use; Proposed Amendment of the Tentative Final Monograph; Reopening of Administrative Record. Federal Register 2015; 80 (84). Available at: https://www. gpo.gov/fdsys/pkg/FR-2015-05-01/pdf/2015-10174.pdf. Accessed January 24, 2017
Disclosure: Dr. Myers is a consultant to Leiter's Enterprise, Inc. LLC. None of the other authors has a financial or proprietary interest in any material or method mentioned.
First author: Megan R. Silas, MD Department of Graduate Medical Education, MacNeal Hospital, Berwyn, Illinois, USA