Comparative efficacy of single and double povidone–iodine soaking protocols combined with draping sequence in donor cornea disinfection Authors: Guann-Jye Chiou, Hsin-Yu Liu, Hsiao-Sang Chu, Fung-Rong Hu Published: 2026 ABSTRACT Purpose:To evaluate the effectiveness of three povidone–iodine disinfection protocols on donor corneoscleral rim contamination and to identify factors influencing microbial contamination during corneal procurement.Design:Prospective interventional study.Methods:This single-center study at the National Eye Bank of Taiwan included 830 corneas from 440 donors. Three povidone–iodine protocols were compared: soaking after draping (single-soaking method 1 [SSM1]), soaking before draping (SSM2), and soaking both before and after draping (double-soaking method [DSM]). Corneoscleral rim cultures were obtained post-procurement, with conjunctival cultures performed before and after each disinfection step in DSM. Grafts with gentamicin-resistant isolates were re-cultured after hypothermic storage. Donor and procurement variables were analyzed for association with contamination. The primary outcome was rim contamination; secondary outcomes included conjunctival culture positivity, persistence of resistant isolates, and associated risk factors.Results:Of 830 corneas, 350 underwent SSM1, 364 SSM2, and 116 DSM. Rim contamination rates were similar (SSM1: 3.1%, SSM2: 3.0%, DSM: 2.6%; P > 0.05). In DSM, conjunctival positivity significantly decreased from 36.1% to 19.8% after double-soaking (P < 0.001). Staphylococcus epidermidis was the most frequent isolate. One gentamicin-resistant strain was identified but did not persist after storage. No donor or procedural factors correlated with rim contamination, and no post-transplant infections occurred.Conclusion:Double-soaking significantly reduced conjunctival microbial load and showed a favorable trend toward reducing rim contamination. These findings support its use as a simple, low-risk strategy to lower ocular surface bioburden and enhance graft safety, while confirming the effectiveness of current disinfection and storage practices. Corneal transplantation is vital for treating corneal diseases but carries a risk of donor-related infections, potentially compromising patient safety. The reported incidence of post-transplant infections, specifically endophthalmitis and keratitis, ranges from 0.08% to 0.77%[1] and 1.76% to 11.9%,[2] respectively, with both potentially leading to vision loss. To mitigate these risks, eye banks implement strict protocols to prevent microbial contamination of donor tissue. The selection of effective disinfection methods during procurement is critical for ensuring tissue safety.[34] Since its inception in 2014, the National Eye Bank of Taiwan (NEBT) has provided high-quality donor corneas and promoted cornea donations. An NEBT study found an 8.9% contamination rate using hypothermic preservation, showing no significant difference between bacteremic (7.1%) and nonbacteremic (9.1%) donors (P = 0.30).[5] NEBT adopted a “soaking after draping” method—applying 1% povidone–iodine soaking after sterile draping and before standard corneal procurement. In contrast, Gruenert et al.[6] reported a 7.8% contamination rate using organ culture and a “soaking before draping” technique with 1.25% povidone–iodine applied before field preparation. Although Gruenert et al.[6] reported a marginally lower contamination rate than NEBT protocols, variations in preservation methods and culture media limit the validity of direct comparisons. The effect of draping and soaking sequence on corneoscleral rim contamination remains uncertain, highlighting the need for further research. This study investigated corneoscleral rim contamination across three disinfection sequences and examined related factors, microbial isolates, and post-transplant infections. Methods This prospective interventional study was conducted at the NEBT. Institutional Review Board exemption was granted by the Research Ethics Committee of National Taiwan University Hospital (No. 202510019W) as the study utilized only deceased donor tissues and anonymized registry data. All procedures followed the principles of the Declaration of Helsinki. Cornea procurement was performed in operating theaters or at hospital bedsides. Periocular skin was disinfected with three applications of 10% povidone–iodine. Sterile drapes were used, and eyelashes were covered using Opsite Flexigrid transparent dressing before eyelid speculum insertion. The conjunctival sacs were soaked with 5% povidone–iodine (3–5 mL) for 3 minutes, followed by irrigation with 0.9% saline until the fluid ran clear. Three different draping/soaking sequences were compared: (1) single-soaking method 1 (SSM1): soaking after draping (October 1, 2017–September 30, 2018); (2) SSM2: soaking before draping (December 1, 2018–November 30, 2019); (3) double-soaking method (DSM): soaking both before and after draping (January 11, 2022–December 19, 2022). The study flow diagram is demonstrated in Fig. 1. Study flow diagram: Donor inclusion, protocol allocation, and graft utilization Following soaking and draping, a 360-degree peritomy was performed. A No. 15 Bard-Parker surgical blade was used to create a sclerotomy 3–4 mm from the limbus. Corneoscleral discs were excised using Westcott scissors, avoiding contact with conjunctival sacs. Corneas from the same donor were procured using two separate instrument sets to avoid cross-contamination. Corneoscleral rim cultures were obtained by swabbing the outer and inner circumference. In the DSM group, conjunctival sac cultures were also collected after first and second soaking to evaluate the added benefit of the extra soaking. All corneas were hypothermically preserved in Optisol-GS (Bausch and Lomb, Garden City, New York, USA), containing gentamicin (100 mg/mL) and streptomycin (200 mg/mL) at 2–8°C. Swabs obtained from the corneoscleral rim and conjunctival sac were submitted for bacterial and fungal cultures to the National Taiwan University Hospital central laboratory. Drug susceptibility testing followed the Clinical and Laboratory Standards Institute guidelines.[7] Cultures were observed for 14 days for bacteria and 28 days for fungi. If the initial rim culture was positive, a repeat culture was performed. Persistently positive cases were discarded.[5] Donor data were obtained from the NEBT database and included age, time from admission to death, blood culture results, procurement locations (operating rooms or other locations), procuring personnel (ophthalmology residents or NEBT technicians), death-to-retrieval interval, transplantation outcomes, and microbial results with antibiotic susceptibility. Statistical analysis was performed with IBM® SPSS® Statistics 25. The Pearson’s Chi-square test was used to compare corneoscleral rim contamination rates among SSM1, SSM2, and DSM groups, as well as to assess differences between the first and second conjunctival sac cultures in the DSM group. Univariable logistic regression was conducted to evaluate factors associated with corneoscleral rim contamination. A P value of less than 0.05 was considered statistically significant. Results A total of 830 corneas from 440 donors were collected: 350 from 185 donors in SSM1, 364 from 190 in SSM2, and 116 from 65 in the DSM group. Donor demographics are summarized in Table 1. The mean donor age was 51.7 ± 15.5 years. The duration from hospital admission to death was 19.4 ± 24.5 days. The interval from death to cornea retrieval was 11.7 ± 7.3 hours; 45.9% of the corneas were retrieved within 12 hours. Final blood cultures were positive in 11.3% of cases. Most corneas (76.0%) were procured in operating rooms and 57.1% by ophthalmology residents. Table 1Summary of donor characteristics and the contamination rate among groups of different draping and disinfection sequencesSSM 1 (n=350)SSM 2 (n=364)SSM 1+2 (n=714)DSM (n=116)Age (years)52.6±15.049.5±15.851.0±15.555.6±15.2Admission-to-expiration time (days)21.2±27.918.3±22.419.7±25.317.7±18.8Death-to-retrieval time (hours)12.2±7.411.1±7.411.7±7.412.0±6.9Death-to-retrieval time<12 hours43.8%47.0%46.1%44.8%Positive blood culture rate11.1%12.1%11.6%9.5%Procuring in the operating room82.3%62.4%72.1%100.0%Procuring by doctors59.4%55.2%57.3%56.0%Positive 1st conjunctival sac culturesNANANA42 (36.2%)$Positive 2nd conjunctival sac culturesNANANA23 (19.8%)$Positive corneoscleral rim cultures11 (3.1%)*11 (3.0%)*22 (3.1%)#3 (2.6%)#Continuous variables are expressed in mean ± standard deviation. Categorical variables are expressed in percentage, *P value between SSM1 and SSM2 = 0.926, #P value between SSM 1+2 and DSM = 0.772, $P value between 1st and 2nd conjunctival sac cultures = <0.001, SSM=Single-soaking method, DSM=double-soaking method Summary of donor characteristics and the contamination rate among groups of different draping and disinfection sequences Continuous variables are expressed in mean ± standard deviation. Categorical variables are expressed in percentage, *P value between SSM1 and SSM2 = 0.926, #P value between SSM 1+2 and DSM = 0.772, $P value between 1st and 2nd conjunctival sac cultures = <0.001, SSM=Single-soaking method, DSM=double-soaking method Corneoscleral rim contamination rates were 11/350 (3.1%) in SSM1, 11/364 (3.0%) in SSM2, and 3/116 (2.6%) in DSM [Table 1]. Differences were not significant (SSM1 and SSM2, P = 0.926; SSM1 and DSM, P = 0.761; SSM2 and DSM, P = 0.808; SSM [SSM1 + 2] and DSM, P = 0.772). However, in DSM, conjunctival sac culture positivity decreased significantly from 36.2% (first culture) to 19.8% (second culture) (P < 0.001), and then further decreased to 2.6% in the final rim culture. Univariable logistic regression analysis showed that no variables significantly affected rim contamination [Table 2]. In the SSM (SSM1 + 2) group, Gram-positive bacteria predominated (83.3%), followed by Gram-negative bacteria (12.5%), and fungi (4.2%) [Table 3]. A similar pattern was observed in the DSM group. Staphylococcus epidermidis was the most commonly cultured pathogen across all groups [Table 4]. Table 2Factors associated with corneoscleral rim contamination rateSSM 1SSM 2SSM 1+2DSM OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P Age0.980 (0.944–1.017)0.2791.022 (0.980–1.067)0.3071.000 (0.973–1.028)0.9941.035 (0.956–1.120)0.393Admission-to-expiration time0.991 (0.961–1.022)0.5680.980 (0.939–1.023)0.3590.987 (0.962–1.012)0.3061.011 (0.958–1.067)0.691Death-to-retrieval time1.038 (0.953–1.129)0.3951.028 (0.947–1.116)0.5101.033 (0.974–1.095)0.2840.943 (0.791–1.125)0.515Death-to-retrieval time <12 hours1.457 (0.419–5.069)0.5541.065 (0.319–3.555)0.9181.243 (0.524–2.945)0.6220.397 (0.035–4.503)0.456Positive blood culture*0.9980.350 (0.089–1.374)0.1330.828 (0.240–2.860)0.7650.194 (0.016–2.334)0.196Procured in the operating room*0.9971.395 (0.418–4.661)0.5881.324 (0.482–3.640)0.586†Procured by doctors0.540 (0.141–2.070)0.3680.452 (0.118–1.733)0.2470.493 (0.191–1.275)0.1450.630 (0.056–7.149)0.709All P values and odds ratios were obtained using univariable logistic regression. DSM=double-soaking method, SSM=Single-soaking method. *The 95% confidence intervals for the odds ratios exceeded the representable range in SPSS and could not be expressed. †All of the corneoscleral rims were procured in the operating room and no odds ratio could be calculated Factors associated with corneoscleral rim contamination rate All P values and odds ratios were obtained using univariable logistic regression. DSM=double-soaking method, SSM=Single-soaking method. *The 95% confidence intervals for the odds ratios exceeded the representable range in SPSS and could not be expressed. †All of the corneoscleral rims were procured in the operating room and no odds ratio could be calculated Table 3Microbial isolates from single-soaking method (SSM)Corneoscleral rim cultures in SSMSSM 1SSM 2SSM 1+2Total (n, %)11 (100%)13 (100%)24 (100%)Gram-positive bacteria8 (72.7%)12 (92.3%)20 (83.3%) Staphylococcus epidermidis Staphylococcus aureus Other CoNS Streptococcus species Corynebacterium species Cutibacterium species Other Gram-positive bacteria 3 (27.3%)2 (18.2%)0 (0%)0 (0%)2 (18.2%)0 (0%)1 (9.1%)5 (38.5%)0 (0%)1 (7.7%)2 (15.4%)0 (0%)1 (7.7%)3 (23.1%)8 (33.3%)2 (8.3%)1 (4.2%)2 (8.3%)2 (8.3%)1 (4.2%)4 (16.7%)Gram-negative bacteria2 (18.2%)1 (7.7%)3 (12.5%) Pseudomonas aeruginosa Klebsiella species Acinetobacter baumannii Enterobacter cloacae complex Other Gram-negative bacteria 1 (9.1%)0 (0%)0 (0%)1 (9.1%)0 (0%)0 (0%)1 (7.7%)0 (0%)0 (0%)0 (0%)1 (4.2%)1 (4.2%)0 (0%)1 (4.2%)0 (0%)Fungi1 (9.1%)0 (0%)1 (4.2%) Candida albicans Candida species 1 (9.1%)0 (0%)0 (0%)0 (0%)1 (4.2%)0 (0%)SSM=Single-soaking method, CoNS=Coagulase-negative staphylococci Microbial isolates from single-soaking method (SSM) SSM=Single-soaking method, CoNS=Coagulase-negative staphylococci Table 4Microbial isolates from the double-soaking methods (DSM)Conjunctival sac and corneoscleral rim cultures of DSM1st conj. sac2nd conj. sacRim cultureTotal (n, %)52 (100%)27 (100%)3 (100%)Gram-positive bacteria45 (86.5%)21 (77.8%)3 (100%) Staphylococcus epidermidis Staphylococcus aureus Other CoNS Streptococcus species Corynebacterium species Cutibacterium species Other Gram-positive bacteria 24 (46.2%)1 (1.9%)2 (3.8%)0 (0%)7 (13.5%)9 (17.3%)2 (3.8%)16 (59.3%)1 (3.7%)1 (3.7%)0 (0%)1 (3.7%)2 (7.4%)0 (0%)2 (66.7%)0 (0%)0 (0%)0 (0%)1 (33.3%)0 (0%)0 (0%)Gram-negative bacteria5 (9.6%)6 (22.2%)0 (0%) Pseudomonas aeruginosa Klebsiella species Acinetobacter baumannii Enterobacter cloacae complex Other Gram-negative bacteria 0 (0%)0 (0%)3 (5.8%)0 (0%)2 (3.8%)0 (0%)1 (3.7%)1 (3.7%)0 (0%)4 (14.8%)0 (0%)0 (0%)0 (0%)0 (0%)0 (0%)Fungi2 (3.8%)0 (0%)0 (0%) Candida albicans Candida species 0 (0%)2 (3.8%)0 (0%)0 (0%)0 (0%)0 (0%)DSM=double-soaking method Microbial isolates from the double-soaking methods (DSM) DSM=double-soaking method Only one isolate (Staphylococcus lugdunensis) was gentamicin-resistant in SSM. In DSM, gentamicin-resistant isolates dropped from eight in the first conjunctival sac culture to three in the second; none were found in the rim culture. Repeat cultures after Optisol-GS storage were negative [Table 5]. Table 5Microbial isolates and selected drug susceptibility profiles in corneoscleral rim cultures across different disinfection protocolsNo.Microbial isolatesSusceptibilityRepeat cultureGMOXLVXSSM1 1 2 3 4 5 6 7 8 9 10 11Candida albicansEnterobacter cloacae complexPseudomonas aeruginosaStaphylococcus epidermidisLactobacillus plantarumStaphylococcus epidermidisCorynebacterium speciesStaphylococcus epidermidisCorynebacterium propinquumStaphylococcus aureusStaphylococcus aureusN/ASSIN/ASN/ASN/ASSN/AN/AN/ARN/ASN/ARN/ASSN/ASSRN/ASN/ASN/ASSAll negativeSSM2 1–1 1–2 1–3 2 3 4 5 6 7 8 9 10 11 Streptococcus mitis Streptococcus salivarius Rothia mucilaginosa Bacillus circulans Staphylococcus epidermidis Staphylococcus epidermidis Staphylococcus epidermidis Bacillus megaterium Staphylococcus epidermidis Staphylococcus epidermidis Klebsiella pneumoniae Cutibacterium acnes Staphylococcus lugdunensis N/AN/AN/AN/ASSSN/ASSSN/ARN/AN/AN/AN/ARRRN/ASRN/AN/ARRSN/AN/ASSSN/ASSSN/ASAll negativeDSM 1 2 3 Staphylococcus epidermidis Staphylococcus epidermidis Corynebacterium macginleyi SSN/ARRN/ASSN/AAll negativeDSM=double-soaking method, GM=gentamicin, OX=oxacillin, LVX=levofloxacin, S=susceptible, I=intermediate, R=resistant, N/A=not available, SSM=single-soaking method Microbial isolates and selected drug susceptibility profiles in corneoscleral rim cultures across different disinfection protocols DSM=double-soaking method, GM=gentamicin, OX=oxacillin, LVX=levofloxacin, S=susceptible, I=intermediate, R=resistant, N/A=not available, SSM=single-soaking method Among the 830 successfully collected corneas, three were discarded (two for positive donor Human T-lymphotropic virus (HTLV)-I/II antibody testing, one for stromal infiltration), with zero discards attributed to persistent contamination. The remaining 827 corneas were distributed: 557 (67.35%) for penetrating keratoplasty, 183 (22.13%) for endothelial keratoplasty, 44 (5.32%) for anterior lamellar keratoplasty, and 37 (4.47%) for therapeutic/tectonic keratoplasty. Six (0.73%) were not used due to preparation failure in endothelial keratoplasty. The infection prevalence remained 0% throughout all study periods, with no graft-related infections reported in the SSM1, SSM2, or DSM groups at 1 and 4 months post-surgery. This consistent outcome indicates that the baseline safety was maintained throughout the transition between different disinfection sequences. Discussion The overall corneoscleral rim contamination rate in our study was 3.1% in the SSM group and 2.6% in the DSM group. These rates are consistent with those reported in previous studies, which ranged from 0.53% in Hagenah’s study to 19.4% in Linke’s study.[568910111213] A previous NEBT study by Chu et al.,[5] which used 1% povidone–iodine for soaking after draping, reported a contamination rate of 8.9%. In contrast, our study employed 5% povidone–iodine and demonstrated a lower contamination rate. While earlier studies suggested that both 1% and 5% povidone–iodine are equally effective for ocular surface decontamination,[14] a randomized study by Ferguson et al.[15] found 5% povidone–iodine to be more effective in eliminating conjunctival bacterial flora, although in the setting of cataract surgery. The Eye Bank Association of America Medical Standards also recommends the use of 5% povidone–iodine for disinfection,[4] which NEBT has adopted as a standard since 2017. In our analysis, the contamination rates were 3.1% for “soaking after draping” (SSM1) and 3.0% for “soaking before draping” (SSM2), indicating that the sequence of draping and soaking had no significant impact when a single soak was used (P = 0.926). Double soaks (DSM) resulted in a lower contamination rate of 2.6%, although this difference was not statistically significant compared with SSMs (P = 0.772). To determine whether the lack of statistical significance was due to limited sample size, we conducted a post hoc power analysis. Given the small differences in contamination rates (3.1% in SSM1, 3.0% in SSM2, and 2.6% in DSM), the estimated effect size was minimal (Cohen’s h ≈ 0.03). Based on standard assumptions for comparing two proportions (α = 0.05, power = 0.80), more than 17,000 samples per group would be required to detect such a small difference, indicating limited clinical relevance. Although our study may be underpowered for detecting minor differences, the observed trend suggests a potential practice benefit of double-soaking, particularly in reducing the microbial load within the conjunctival sac. While our findings on the effect of double-soaking on corneoscleral rim contamination are consistent with previous reports by Salisbury et al.[16] and Sawant et al.,[17] our study provides novel insights by directly comparing single versus double-soaking protocols across different draping sequences (before, after, or both) within a standardized hypothermic storage context. Salisbury et al.[16] reported reduced contamination with double-soaking using two separate 5-minute applications of povidone–iodine with a 5-minute interval; however, their study did not specify the order of draping and soaking. Sawant et al.[17] found only a nonsignificant reduction and similarly did not evaluate draping sequence, instead applying povidone–iodine for 5 minutes before and after conjunctival excision in paired donor eyes. Our study is the first to systematically assess both the sequence and repetition of povidone–iodine soaking during procurement and uniquely includes conjunctival sac culture data before and after soaking, an aspect not assessed in prior studies. The significant reduction in conjunctival sac culture positivity following double-soaking supports its decontamination effect. Although the reduction in rim contamination was not statistically significant, the trend is consistent with earlier findings and highlights the potential benefit of double-soaking. Notably, Sawant et al.[17] reported no adverse effects on tissue quality after double-soaking. An interesting finding in our study was the persistence of corneoscleral rim contamination despite double-soaking. This may be attributed to colonizing pathogens within the conjunctival sac, where achieving complete disinfection remains challenging. Builles et al.[10] similarly observed higher contamination when a single blepharostat was used for both eyes, suggesting possible transfer of microorganisms from the first eye’s conjunctival sac to the second. Microorganisms may persist in anatomical areas such as deep fornices or conjunctival folds, which can limit the uniform exposure of povidone–iodine. Given the conjunctival sac’s role as a potential reservoir for contamination, minimizing contact with this area during procurement is advisable. At NEBT, separate instrument sets were used for each eye to mitigate this risk. Surprisingly, our study found no significant association between corneoscleral rim contamination and donor-related factors such as age, hospital stay duration, ICU admission, or blood culture results. This aligns with findings from Linke et al.[12] and Patel et al.,[18] who also reported no impact of donor age on contamination risk or graft suitability. In contrast, Medina et al.[19] observed increased contamination with hospital stays exceeding 3 days and ICU stays over 4 days, though no additional risk was noted beyond 7 days. In our cohort, the mean hospital stay was 19.4 days, and not all donors had ICU admission. This variability in hospitalization settings may have diluted the potential association with contamination risk. Prolonged death-to-retrieval time has been identified as a risk factor for corneoscleral rim contamination in several studies. Gruenert et al.[6] and Linke et al.[12] reported median retrieval times of 24.7 and 40 hours, respectively, both linked to higher contamination rates. In contrast, our cohort had a shorter mean retrieval time of 11.7 hours (maximum 24), which may explain the absence of a significant association. While Gruenert et al.[6] also reported higher contamination rates outside operating rooms, we found no significant difference between retrievals conducted in operating rooms versus hospital bedsides. Similarly, contamination rates did not differ between procurements performed by physicians and trained NEBT technicians, highlighting the effectiveness of NEBT’s standardized training in maintaining low contamination risk across personnel and settings. Contamination during corneoscleral disc procurement remains a challenge due to the difficulty in disinfecting the conjunctival sac. In addition, cleansing of the periocular skin and adnexa with 10% povidone–iodine may partially reduce conjunctival microbial load before conjunctival soaking, which should be considered when interpreting the observed reductions. The incidence of post-keratoplasty infection varies by region and by definition of early versus late infection. A recent 10-year eye bank analysis from India[20] specifically evaluated infectious adverse events occurring within 6 weeks after elective optical keratoplasty, the time frame most consistent with donor-related transmission, and reported an early infection rate below 1%. In their cohort, Gram-negative organisms predominated, and most of these Gram-negative infections were multidrug-resistant. In contrast, Gram-positive bacteria—particularly Staphylococcus epidermidis—were the most common corneoscleral rim isolates in our present study. This predominance of Gram-positive bacteria is consistent with the microbiological profiles reported in a previous Taiwanese study of post-keratoplasty infections.[2] Our study also aligns with previous findings from NEBT,[5] which reported a similar distribution (79.1% Gram-positive, 19.4% Gram-negative, and 1.5% fungi; and Staphylococcus epidermidis being the most common). In contrast, Gruenert et al.[6] and Linke et al.[12] found higher rates of Gram-negative bacteria and fungal contamination, likely due to differences in preservation methods. NEBT used hypothermic storage, whereas both Gruenert and Linke employed organ culture systems with higher storage temperatures, which, as Fontana et al.[11] noted, may favor fungal over bacterial growth. Although some pathogens may evade standard disinfection, Optisol-GS proved effective in eliminating residual contamination. Antibiotic resistance among isolates was uncommon in our cohort. Only one Staphylococcus lugdunensis isolate demonstrated gentamicin resistance (among 27 isolates from 830 corneas), and its repeat culture after hypothermic storage in Optisol-GS was negative; therefore, no corneas were discarded due to persistent contamination. The high gentamicin concentration (100 μg/mL) in Optisol-GS, a concentration-dependent bactericidal agent, effectively suppressed most bacterial growth. This low prevalence of resistant organisms, particularly in contrast to the higher rate of multidrug-resistant Gram-negative infections reported in the Indian study,[20] may further contribute to the absence of donor-related infections observed. Importantly, no early postoperative infections, such as infectious keratitis or endophthalmitis, suggestive of donor-related transmission were identified during the study periods. This favorable outcome may be attributed to the standardized microbiological surveillance protocol implemented at NEBT. Following established protocols,[5] any rim with an initial positive culture undergoes repeat testing after storage, and persistently contaminated tissues are strictly discarded. This rigorous screening, combined with standardized povidone–iodine disinfection and Optisol-GS preservation, ensured a 0% donor-derived infection rate and underscores the efficacy of our protocols. This study has several limitations. First, conjunctival sac culture data were not available for the SSM group, limiting direct comparisons with DSM. Second, the relatively smaller sample size of the DSM group compared with the SSM groups may have reduced the statistical power to detect modest differences in corneoscleral rim contamination rates. Nevertheless, each study group was evaluated over a standardized 1-year period, which helped maintain consistency in procurement protocols, data collection procedures, and outcome assessment across groups. Third, only one gentamicin-resistant isolate was identified, and it did not survive storage; thus, our findings may not be generalizable to settings with a higher prevalence of resistant organisms. Fourth, because the three protocols were implemented sequentially rather than concurrently, temporal bias cannot be excluded. However, procurement environment, laboratory procedures, storage media, and personnel training protocols remained standardized throughout the study period. Additionally, conjunctival sac cultures were only available for the DSM group, limiting direct baseline comparisons with the SSM groups. Lastly, disinfection methods were not randomized. Although consistent protocols strengthen internal validity, future prospective randomized studies with larger samples are needed to confirm and expand these findings. Conclusion This is the first study to systematically evaluate the impact of different povidone–iodine soaking and draping sequences on donor corneal contamination under standardized hypothermic storage. Our results demonstrate that double-soaking significantly reduces the conjunctival microbial load and shows a favorable trend toward reducing corneoscleral rim contamination, though the sequence of soaking and draping did not independently yield significant differences. These findings support the clinical value of a double-soaking protocol in reducing ocular surface microbial load. No donor- or procurement-related factors were associated with contamination, highlighting the persistent challenge of fully disinfecting the conjunctival sac. Rigorous aseptic technique and minimal conjunctival contact remain essential. Additionally, gentamicin in Optisol-GS likely contributed to eliminating residual pathogens. Importantly, none of the 830 donor corneas from the NEBT resulted in post-transplant graft-related infections, affirming the safety and effectiveness of the disinfection and preservation protocols. These results offer valuable guidance for improving eye banking practices and support the broader adoption of enhanced antimicrobial strategies. Authors’ contributions Guann-Jye Chiou: Concept, design, definition of intellectual content, literature search, clinical studies, experimental studies, data acquisition, data analysis, statistical analysis, and manuscript preparation. Hsin-Yu Liu: Definition of intellectual content, literature search, clinical studies, data acquisition, manuscript editing, and manuscript review. Hsiao-Sang Chu: Literature search, data acquisition, data analysis, manuscript editing, and manuscript review. Fung-Rong Hu: Concept, design, definition of intellectual content, resources, supervision, statistical analysis, manuscript editing, and manuscript review. Conflicts of interest There are no conflicts of interest.