| | Structural integrity of the polyurethane female condom after multiple cycles of disinfection, washing, drying and relubricationReceived 6 June 2002; accepted 25 July 2002. Abstract The female condom provides an important alternative means of protection against HIV and other sexually transmitted infections for women, particularly in situations where partners are reluctant or refuse to use male condoms. The relatively high cost of the device, however, is a barrier to its use in resource-poor environments. This has led to some reuse of the product and two studies have demonstrated that female condoms can withstand a limited level of reuse without an excessive loss of structural integrity when washed with soap and water. A consultation on female condom reuse convened by the World Health Organization and the Joint United Nations Programme on AIDS in June 2000 recommended that all used female condoms should be disinfected immediately after use, before washing. The effect of such treatments on the female condom was not known. This study was undertaken to assess the effect of the disinfection, washing, drying and relubrication on the properties of the condom. Samples from three batches of female condoms were subjected to seven treatment cycles before being tested for structural integrity. In all cases the batches of condoms complied with the manufacturer’s release specification for the product after treatment. Some minor changes in properties were seen but these were not considered important. There was evidence of a small increase in the number of condoms with holes following repeated disinfection and washing cycles, suggesting that excessive or rough handling can damage the condom. Condoms should therefore be handled carefully and inspected thoroughly for signs of damage after washing and drying before being stored with the intention of subsequently reusing the device.
1. Introduction  The polyurethane female condom, sold and distributed under a number of names, including Femidom, Femy, Reality and Care, by the Female Health Company, provides an important additional means of protection against HIV and other sexually transmitted infections for women, particularly in situations where partners are reluctant or refuse to use male condoms [1]. The relatively high cost of the device limits its acceptability, and reuse of the device is perceived as one way of reducing the cost per protected act of intercourse. There have been some reports of reuse of the female condom [2], [3], and the structural integrity of the condoms has been assessed after multiple in vitro wash, dry, relubricate cycles [4], after a single in vivo use [5] and up to eight in vivo uses [6]. In response to requests for guidance, the World Health Organization (WHO) Department of Reproductive Health and Research and the Joint United Nations Programme on AIDS (UNAIDS) convened a consultation in June 2000 to consider the safety and feasibility of reuse of the female condom. The consultation included experts in microbiology, sexually transmitted infections, condom manufacture and quality assurance testing and programmatic issues. The objective was to evaluate data relevant to the safety of reusing the female condom, considering both the structural integrity (i.e., the ability of the condom to withstand washing and reuse without breaking or developing holes) and the potential risks of infection related to reuse. While the data suggested the condoms could withstand a limited level of reuse without loss of structural integrity when washed with soap and water between uses, the number of condoms tested was considered inadequate to provide guidance. In addition, there was insufficient evidence to determine whether sexually transmitted pathogens, including HIV, could be inactivated and safely removed from a used female condom with soap and water alone. The consultation considered that women could be at potential risk of infection when washing used condoms and that both the women and their partners could be at risk during subsequent use. The experts recommended that all used female condoms should be disinfected before washing with soap and water. Disinfection of the female condom had not been studied but it was agreed that standard disinfection procedures recommended for heavily contaminated materials—for example, soaking in 1% sodium hypochlorite solution (diluted household bleach) for at least 30 min [7]—followed by washing with soap and water and drying, would be likely to inactivate and remove sexually transmitted pathogens from the device. The impact of bleach on the structural integrity of the female condom had not been evaluated, although a preliminary study conducted by the Female Health Company had shown that seam strength and burst properties (peak pressure, burst pressure and burst volume) were essentially unaffected by five repeat immersions of 30 min duration in undiluted domestic bleach (5% sodium hypochlorite) and that seam strength was unaffected by soaking for 14 h in diluted bleach (0.5% sodium hypochlorite). The consultation concluded that further research was required to assess whether sexually transmitted pathogens could be safely removed by disinfection, washing and drying, and whether condoms could withstand repeat cycles of disinfection, washing, drying and relubrication [8]. This paper reports the results of an experiment undertaken at the premises of the Female Health Company under the supervision of an independent expert to assess the effect of the proposed treatment cycles on the key specification attributes that determine structural integrity of the female condom, namely, freedom from holes, air inflation properties and seam tensile strength. 2. Materials and methods 2.1. Materials Three batches of female condoms manufactured in June 1999, February 2000 and May 2000, respectively, were selected from retained samples. The condoms were packed and lubricated with silicone oil. They had been stored since manufacture under normal warehouse conditions on the premises of the Female Health Company in Park Royal, London. Domestos, a brand of bleach containing 5% sodium hypochlorite widely available in the United Kingdom (UK), was used for all the studies. Domestos also contains nonionic surfactants and soap. The bleach was diluted at the rate of one part by volume of bleach to four parts by volume of cold water. Washing was carried out in either Sunlight Dishwashing Liquid obtained from South Africa or Fairy Liquid Dishwashing Liquid obtained in the UK. The dishwashing liquid was diluted with tap water at the rate of 5 mL/L before use. Sunflower oil obtained from a UK supermarket, ASDA, was used to lubricate the condoms between washes. 2.2. Experimental procedure A total of 300 condoms from each batch were subjected to seven treatment cycles as described below. The condoms were processed in groups of 150 at a time in suitable containers capable of holding at least 10 L of liquid.
1.One hundred and fifty condoms (previously washed and dried) were placed into a container and 225 mL of sunflower oil added. The condoms were mixed with the oil by hand then removed and hung open-end down over pegs. Excess oil drained from the condoms. The condoms were left for 60 min from the time of first immersion in the oil. (This step was excluded from the first cycle).
2.The condoms were removed from the pegs and soaked in 10 L of diluted bleach (one part bleach to four parts water) for 30 min. (On the first cycle the condoms were removed from their individual packs and placed directly into the diluted bleach solution).
3.The condoms were removed to 10 L of fresh water and allowed to soak for 5 min.
4.The condoms were removed to 10 L of diluted dishwashing liquid (5 mL/L) and stirred gently by hand. After 5 min the condoms were removed to fresh rinsing water.
5.The condoms were removed one at a time from the rinsing water and washed by gently rubbing between fingers and palms in the diluted dishwashing liquid. The condoms were turned inside out during washing and finally rinsed in fresh water. The minimum washing time was 30 s, the maximum 45 s.
6.After washing, each condom was placed over a peg open-side down and left to dry for at least 1 h. On the last cycle, the condoms were thoroughly dried overnight before being stored in black plastic bags until tested. The procedure used to lubricate the condoms had the effect of introducing lubricant to the inside of the condom as well as lubricating the external surface. Condoms were also lubricated after the final disinfection—wash—dry cycle before being tested. This was done both to model the actual in-use situation more closely and because the normal quality control testing procedures were performed on lubricated condoms. To ensure even lubrication they were sprayed inside and out with a fine mist of sunflower oil. The oil was left in contact with the condom for at least 1 h and gently wiped off with a paper towel before testing for tensile strength and air burst properties. Water leak testing was carried out with the oil in situ. Final lubrication and testing was executed on small groups of condoms such that the total exposure to oil during this final stage was no longer than 2 h. For the first batch of condoms tested (batch B), all 300 were visually inspected after five cycles for signs of damage and tears, and 10 condoms were removed and tested for seam strength. For all batches condoms were visually inspected after seven cycles. Any condoms showing defects were removed. 2.3. Testing Condoms were selected at random from each set and tested for freedom from holes (n = 200), air burst pressure (n = 50) and seam tensile strength (n = 20) using the standard procedures employed by the Female Health Company. Untreated, control condoms from each batch were tested concurrently for air burst pressure (n = 50) and tensile seam tensile strength (n = 20). Freedom from holes was determined by mounting each condom on an acrylic plug and adding 300 mL of water containing a surfactant (0.02% Fluorad) to the condom. After hanging for 2 min, the outside of each condom was inspected for any signs of leaks. The acrylic plug was then sealed and each condom was rolled on absorbent paper to detect any small leaks that may have been missed during the hanging phase of the test. Air inflation properties were determined by inflating each condom with air at the rate of 0.5 L/s to bursting. The peak pressure, burst pressure and burst volume were recorded. Seam tensile strength was measured on a 20 mm wide ring cut across the condom from the mid-body region. Each ring was then mounted between the rollers of a tensile testing machine and stretched at the rate of 500 mm/min to failure. One of the rollers was rotated during the test to even out the stress distribution in the test sample. 2.4. Statistical methods Comparisons between treated and untreated condoms were made using two-sample t tests for continuous variables assuming equal variances. All confidence intervals are two-sided, 95% level and all p values are two-sided. Exact confidence intervals for binomial proportions were calculated from the F-distribution [9]. Fisher’s exact test was used to compare binomial proportions (STATA Statistical Software Release 6.0, Stata Corp, College Station, TX, USA). 3. Results 3.1. Washing The washing scheme was highly effective in removing the vegetable oil from the condoms. In fact, most of the oil was removed within the first few seconds. Nevertheless, a minimum washing time of 30 sec was retained to ensure that the condoms were subjected to a similar degree of handling that could be expected in real use. In practice it would be necessary to remove semen and other body fluids as well as the lubricating oil. 3.2. Visual inspection All the condoms showed a degree of discoloration, turning slightly yellow/brown. This may have been associated with the treatment but could also have been caused by exposure to light during the treatment cycles. Two defective condoms were found and removed during the washing cycles, both from batch B. One was found on inspection after five cycles to have a small tear approximately 6–7 mm from the ring. The second was found after seven cycles with five small tears, all located within an area of approximately 10 mm diameter in the mid-body of the condom. In both cases the tears appeared to have been produced by physical damage during handling. 3.3. Freedom from holes The numbers of defective condoms (i.e., condoms with holes) in samples from each batch are summarized in Table 1 together with the original quality assurance (QA) data generated at the time of manufacture. | | |  | | Batch | |
|---|
| A | B | C | |
|---|
| Original QA data | | | | |
| Defectives/number tested | 0/315 | 0/200 | 0/315 | |
| 95% Confidence interval for defectives (%) | 0–1.2 | 0–1.8 | 0–1.2 | |
| | | | | |
| Defectives removed during treatment/number treated | 0/300 | 2/300 | 0/300 | |
| After seven treatment cycles | | | | |
| Defectives/number tested | 0/200 | 2/200 | 1/200 | |
| 95% Confidence interval for defectives (%) | 0.0–1.8 | 0.12–3.6 | 0.01–2.8 | | | | |
All three defective condoms were detected as soon as they were filled with water rather than when the condoms were rolled on absorbent paper, i.e., the holes were obvious and could be detected easily, although they had not been detected on visual inspection. All the defects were small tears and appeared to have been formed by physical damage rather than by chemical attack on the material. In the case of batch B, both the tears were in the seam. The single tear found in batch C was located near, but not in, the seam. Two of the tears, one each in batch B and C, were located towards the open (ring) end of the condom. The second tear in batch B was located in the distal, i.e., closed, end of the condom. All three batches would have been accepted at the 0.4% acceptable quality level for which the acceptance number is maximum two defectives per batch of size 200. Overall, 3 condoms out of the 600 subjected to the water test were found to have holes after the seven treatment cycles—percent defective 0.5% (95% confidence interval: 0.1%, 1.5%). This compares with a total of 0 holes found out of 830 condoms tested (0.0%, 0.4%) for the original QA data (Fisher’s exact test: p = 0.074). 3.4. Air inflation testing Air inflation results are summarized in Table 2 together with relevant QA data generated at the time of manufacture. Individual test results are plotted in Fig. 1. | | | | | Batch | |
|---|
| A | B | C | |
|---|
| Peak pressure (mbar) | | | | |
| Original QA data (SD)a | 82.3 (3.5) | 85.2 (3.7) | 81.1 (2.8) | |
| No treatment (SD)b | 88.8 (3.5) | 83.8 (4.4) | 82.2 (3.9) | |
| After seven cycles (SD)b | 82.4 (3.0) | 78.3 (4.5) | 77.6 (3.3) | |
| p-value for difference, no treatment vs. seven cycles | <0.001 | <0.001 | <0.001 | |
| Burst pressure (mbar) | | | | |
| Original QA data (SD)a | 50.6 (3.2) | 59.4 (2.5) | 50.4 (1.7) | |
| No treatment (SD)b | 59.1 (2.3) | 47.2 (1.9) | 45.7 (1.5) | |
| After seven cycles (SD)b | 50.2 (2.5) | 40.2 (2.2) | 42.3 (1.6) | |
| p-value for difference, no treatment vs. seven cycles | <0.001 | <0.001 | <0.001 | |
| Burst volume (L) | | | | |
| Original QA data (SD)a | 9.8 (1.4) | 10.5 (1.3) | 9.5 (1.3) | |
| No treatment (SD)b | 8.8 (1.4) | 10.0 (1.5) | 8.9 (1.1) | |
| After seven cycles (SD)b | 6.9 (2.1) | 9.3 (2.3) | 8.9 (1.2) | |
| p-value for difference, no treatment vs. seven cycles | <0.001 | 0.075 | 0.96 | | | | |
After seven treatment cycles, peak pressures dropped by 6.4 (5.1, 7.7) mbar for batch A, 5.5 (3.7, 7.3) mbar for batch B and 4.6 (3.2, 6.0) mbar for batch C, representing percentage falls of 7.2%, 6.6%, and 5.6%, respectively. None of the tested condoms burst at a pressure below the 48 mbar minimum quality standard. Burst pressures dropped by 8.9 (7.9, 9.9) mbar for batch A, 7.0 (6.2, 7.8) mbar for batch B and 3.4 (2.7, 4.1) mbar for batch C, representing percentage falls of 15.1%, 14.8%, and 7.4%, respectively. Burst volumes showed much more variability, dropping by 1.9 (1.2, 2.6) L for batch A, 0.7 (−0.1, 1.5) L for batch B, and 0.0 (−0.5, 0.5) L for batch C, representing falls of 21.6%, 7.0%, and 0.0%, respectively. 4. Discussion A total of 900 polyurethane female condoms from three separate production batches were subjected to repeated in vitro chemical and physical challenges to determine whether they would withstand repeated disinfection, washing and relubrication during reuse. Some changes in the physical properties of the condoms were observed. While the appearance of the condoms changed, the degree of discoloration does not appear to be of any practical importance. There was no evidence of any significant change in surface texture or roughness. All three batches of condoms met the release criterion with regard to freedom from holes (AQL 0.4) after seven treatment cycles. The pooled data across all batches (three defectives in 600 samples) also complies with an AQL of 0.4 assuming an accept number (maximum number of defectives) of six for a sample size of 600. The ISO Standard for sampling by inspection (ISO 2859-1) does not include sampling plans for sample sizes of 600 but an acceptance number of six results in an operating characteristic curve comparable to those for the published sampling plans based on samples of sizes 500 and 800. Although the individual and cumulative levels of defectives comply with the 0.4 AQL, comparison between the original QA data from the time of manufacture and the number of defective condoms for the treated condoms suggests that some deterioration has occurred. The 95% confidence interval for the proportion defective from the cumulative QA data (0 to 0.4%) confirms that the level of defectives in freshly manufactured product is below 0.4%. After seven treatment cycles, 0.5% (0.1%, 1.5%) of condoms had holes, most likely due to the physical handling of the condoms rather than to deterioration of the product resulting from the chemical challenge. The results from the air inflation test indicate that there is a fall in peak inflation pressure of between 5.6% and 7.2% after seven treatment cycles. While the fall is statistically significant, it appears to be of little practical importance when compared with the minimum specified value and the normal variability seen between batches. All batches complied with the release specification for peak inflation pressure after seven treatment cycles and all condoms exceeded the minimum requirements of 48 mbar by a considerable margin. The fall in burst pressures after seven treatment cycles (between 7.4% and 15.1%) was generally larger than that seen for peak pressures (between 5.5% and 7.2%). None of the individual burst pressure results appeared to be particularly low when compared with the mean values and it is probable that the changes in burst pressure are of little practical significance. The differences in average burst volume between untreated condoms and those subjected to seven treatment cycles does not appear to be particularly important with statistical significance only being reached in the case of batch A. Inspection of the individual results, however, highlights the fact that in the case of two of the batches some condoms that were subjected to the seven treatment cycles had particularly low burst volumes of 2 and 3 L. This suggests the treatment did have some effect on the structural integrity of certain condoms. The seam strength appears to remain unaffected after seven treatment cycles. Only in the case of batch A was there a statistically significant difference between treated and untreated condoms and in this instance the seam strength actually increased after treatment. The results of this in vitro study are consistent with the experience obtained from a series of in vivo experiments on condom reuse by sex workers in Johannesburg [6]. In that study, groups of 50 women were asked to wash used condoms in liquid detergent, pat dry, store in a clean container and relubricate with vegetable oil immediately before reuse. Condoms were used up to a maximum of eight times and the returned condoms were assessed for water leakage, seam strength and burst pressure using identical testing procedures to those described here. Significant changes were observed in seam strength and burst pressure, but all condoms tested were within specification for fresh condoms and the changes were not considered to reflect important degradation of the condoms. However, a total of five holes was detected in 378 condoms—percent defective 1.3% (0.4%, 3.1%)—suggesting that some deterioration of the condoms had occurred. We cannot determine whether the holes were induced during intercourse or while handling during washing and drying. The condoms tested in vivo clearly had greater physical challenge than those subjected to the in vitro testing procedure. All five holes observed in the Johannesburg study were detected by the women themselves following a detailed inspection of the washed and dried condom. The treatment regimen used in this study was based on a draft disinfection/washing procedure outlined at the consultation meeting. The concentration of bleach (1% sodium hypochlorite), the exposure times (i.e., a total soaking time of 3.5 h in bleach and 6 h in sunflower oil) and the number of repeat cycles were deliberately selected to be severe in order to provide a margin of safety. The effect of any departure from the regimen tested in this study on the integrity of the female condom is not known. Critical features of the testing regimen include the following, which should not be exceeded in any reuse protocol without some additional formal assessment:
•Disinfection should be carried out as soon as possible after use and before the condom is washed. We have no information on whether a used condom with 3–5 mL semen will cause the condoms to deteriorate.
•The disinfection stage should be no longer than 30 min using household bleach (5% sodium hypochlorite) diluted at the rate of one part bleach to at least four parts of water. However, bleach is rapidly deactivated in the presence of proteinaceous material such as semen, so soaking used condoms containing semen in more concentrated bleach solution, or for longer periods, will probably not have a detrimental effect. Moreover, the limited experiments involving a 14-h soak in diluted bleach did not suggest any deterioration.
•Washing should be carried out using diluted dishwashing liquid or plain bar soap. In the latter case, the soap should be lathered first and not rubbed directly onto the condom. Only gentle manipulation is needed to clean the condom; it should not be scrubbed. The deterioration that was observed in this study was most likely due to the physical handling of the condoms.
•All the defects found in the treated condoms were fairly obvious, consisting of small tears. Those that were found during the water leakage test were identified when the condoms were filled with water rather than when they were rolled on absorbent paper. This suggests that whilst there does appear to be a risk that the condoms can be damaged during the disinfection and washing process, any defects that may be formed should be detectable by a careful inspection of the condom. It is important to include a detailed visual inspection stage in any protocol for condom reuse.
•The condoms should be air-dried or dried gently using, preferably, disposable paper towels or tissue.
•Once dry, the condom should be inspected carefully for any signs of damage or any tears, particularly in the seam region.
•The condom should be stored dry and not lubricated between uses, being relubricated with vegetable oil, preferably sunflower oil, or a water-based lubricant such as KY Jelly, immediately before use.
•Any condom showing signs of damage, leakage, tearing, excessive discoloration or significant changes in surface texture should be discarded.
•There should be a limit on the maximum number of uses. On the basis of this study up to five uses can be justified, and the experiments described here should be repeated for a larger number of cycles if it is desired to increase the maximum recommended number of uses. Parallel research is underway to establish whether the disinfect/wash/dry protocol drafted by the consultation is sufficient to remove all sexually transmissible pathogens, and determine whether any components of the protocol can be made less rigorous without compromising the user’s safety during washing and/or reuse. This study demonstrates that the physical integrity of the female condom is not seriously affected by seven treatment cycles involving disinfection in diluted sodium hypochlorite (one part of household bleach diluted with four parts of water) for 30 min, followed by washing in diluted dish washing liquid, drying and relubrication with sunflower oil. In all cases the batches of condoms complied with the release specification for the product after treatment. There was evidence of a slight increase in the number of condoms with holes following repeated disinfection and washing cycles, suggesting that excessive or rough handling can damage the condom. Condoms should therefore be handled carefully and inspected thoroughly for signs of damage after washing and drying before being stored with the intention of subsequently reusing the device. It should be noted that the results of the experiments described here only apply to the polyurethane female condom, the only female condom currently marketed. They cannot be extrapolated to other designs or materials for female condoms, nor to male condoms. In particular, the bleach disinfection and the relubrication with vegetable oil considered in these experiments are anticipated to result in rapid deterioration of latex. This study supports the opinion that the polyurethane female condom can be used up to a maximum of five times with an acceptable level of safety, providing that the disinfection and washing procedures described are used in conjunction with a detailed inspection of the condom. While used female condoms are likely to have more holes than fresh condoms, the proportion with holes remains very low. Whether condoms with this degree of holes are acceptable and safe depends on the alternatives available to women considering reuse. It is clear that a carefully disinfected, washed, dried and relubricated condom would provide better protection than no condom, but the results of this experiment and the in vivo studies from Johannesburg indicate that a reused condom will never be quite as good as a fresh condom. Acknowledgements The Female Health Company, Park Royal, London, provided the laboratory facilities and staff resources to conduct the experiments, including qualified Quality Assurance staff to carry out the tests according to Company Standard Operating Procedures. The experimental programme was overseen by Bill Potter. References [1].
[1]
World Health Organization
.
The female condom (a guide for planning and programming. Report No. WHO/RHR/00.8). Geneva: World Health Organization; 2000;. [2].
[2]
National AIDS Coordination Programme (NACP)
.
Piloting the female condom in Zimbabwe (a national survey).
Harare. 1998;. [3].
[3]
Smith JB, Nkhama G, Trottier DA.
Female condom reuse in Lusaka, Zambia (evidence from 12 cases).
J Urban Health. 2001;78:638–646. MEDLINE |
CrossRef
[4].
[4]
Pettifor AE, Rees HV, Beksinska ME, Kleinschmidt I, McIntyre J.
In vitro assessment of the structural integrity of the female condom after multiple wash, dry, and re-lubrication cycles.
Contraception. 2000;61:271–276. Abstract | Full Text |
Full-Text PDF (105 KB)
|
CrossRef
[5].
[5]
Joanis C, Latka M, Glover LH, Hamel S.
Structural integrity of the female condom after a single use, washing, and disinfection.
Contraception. 2000;62:63–72. Abstract | Full Text |
Full-Text PDF (770 KB)
|
CrossRef
[6].
[6]
Beksinska ME, Rees HV, Dickson-Tetteh KE, Mqoqi N, Kleinschmidt I, McIntyre JA.
Structural integrity of the female condom after multiple uses, washing, drying, and re-lubrication.
Contraception. 2001;63:33–36. Abstract | Full Text |
Full-Text PDF (49 KB)
|
CrossRef
[7].
[7]
World Health Organization
.
Biosafety guidelines for diagnostic and research laboratories working with HIV. Geneva: World Health Organization; 1992;. [8].
[8]
WHO/UNAIDS
.
Information update: consultation on re-use of the female condom. Geneva: World Health Organization; 2000;. [9].
[9]
Armitage P, Berry G.
Statistical methods in medical research. Oxford, UK: Blackwell Scientific Publishers; 1987;. a Stapleford Scientific Services, 3 Adcroft Place, Stapleford, Cambridge, CB2 SFD, UK b Enersol Pty Ltd, 127 Trafalgar Street, Annandale, NSW 2038, Australia c The Female Health Company, 1 Sovereign Park, Coronation Road, London NW10 7QP, UK d Department of Reproductive Health and Research, World Health Organization, 1211 Geneva 27, Switzerland  Corresponding author. Tel.: +41-22-791-33-10; fax: +41-22-791-41-71.
PII: S0010-7824(02)00418-3 © 2002 Elsevier Science Inc. All rights reserved. | |
|
FULL TEXT00418-3/journalimage?img=PIIS0010782402004183.gr1.lrg.gif&fig=FIG1&kwhquery=&issn=0010-7824&ishighres=false&allhighres=false&free=yes','journalimage');)
00418-3/journalimage?img=PIIS0010782402004183.gr2.lrg.gif&fig=FIG2&kwhquery=&issn=0010-7824&ishighres=false&allhighres=false&free=yes','journalimage');)