A Phase I Randomized Postcoital Testing and Safety Study of the Caya
Diaphragm used with 3% Nonoxynol-9 gel, ContraGel, or No Gel

Christine K. Mauck, Vivian Brache, Thomas Kimble, Andrea Thurman,

Leila Cochon, Sarah Littlefield, Kim Linton, Gustavo F. Doncel, Jill L.
Schwartz

PII: S0010-7824(17)30163-4
DOI: doi: 10.1016/j.contraception.2017.05.016
Reference: CON 8918

To appear in: Contraception

Received date: 27 March 2017

Revised date: 30 May 2017
Accepted date: 31 May 2017

Please cite this article as: Mauck Christine K., Brache Vivian, Kimble Thomas, Thur-
man Andrea, Cochon Leila, Littlefield Sarah, Linton Kim, Doncel Gustavo F., Schwartz
Jill L., A Phase I Randomized Postcoital Testing and Safety Study of the Caya Di-
aphragm used with 3% Nonoxynol-9 gel, ContraGel, or No Gel, Contraception (2017),
doi: 10.1016/j.contraception.2017.05.016

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT

A Phase I Randomized Postcoital Testing and Safety Study of the Caya

Diaphragm used with 3% Nonoxynol-9 gel, ContraGel, or No Gel

T
Christine K. Maucka, Vivian Bracheb, Thomas Kimblec, Andrea Thurmanc, Leila Cochonb, Sarah

P
Littlefielda, Kim Lintona, Gustavo F. Donceld, Jill L. Schwartzd,e

RI
a
Formerly with CONRAD

SC
b
Profamilia, Nicolas de Ovando esq. Calle 16, Ensanche Luperon, Santo Domingo, Dominican

NU
Republic
c
CONRAD, Eastern Virginia Medical School, 601 Colley Ave, Norfolk, VA USA 23507
MA
d
CONRAD, Eastern Virginia Medical School, 1911 North Fort Myer Drive, Suite 900,
Arlington, Virginia, USA 22209
e
ED

Corresponding author: Jill L. Schwartz, MD, CONRAD, 1911 Fort Myer Drive, Suite 900,
Arlington, VA 22209; e-mail: jschwartz@conrad.org.
PT

Word count, abstract: 249

CE

Word count, manuscript text: 2742

AC

ClinicalTrials.gov identifier: NCT02309554

1
 ACCEPTED MANUSCRIPT

Abstract

Objectives: The Caya® Diaphragm is a newly approved single-size, non-latex diaphragm.

T
Contragel® is a personal lubricant containing lactic acid approved in Europe and other countries

P
for use with vaginal barrier devices. This study assessed the effectiveness in preventing sperm

RI
from penetrating midcycle cervical mucus of Caya with Contragel, Caya with 3% nonoxynol-9

SC
(N-9), and Caya alone.

Study Design: Phase I multi-center, single-blind, randomized, crossover, non-significant risk

NU
study at 2 sites: Eastern Virginia Medical School, Norfolk, VA, and Profamilia, Santo Domingo,
Dominican Republic. Healthy, sexually active women 18–45 years old, not at risk for pregnancy
MA
due to tubal occlusion were eligible. Each participant was seen in 9 visits, completing a baseline
cycle (without product use) followed by 3 test cycles (sequence determined by randomization),
each consisting of a cervical mucus check visit and a postcoital test visit. To proceed to test
ED

cycles, the baseline postcoital test had to show adequate cervical mucus and > 5 progressively
motile sperm per high power field (PMS/HPF).
PT

Results: All women had an average of < 5 PMS/HPF during the test cycle of each study arm,
CE

the primary endpoint. Caya with ContraGel and Caya with N-9 reduced the average number of
PMS/HPF from 22.5 to zero. Caya alone reduced the average number of PMS/HPF from 22.5 to
AC

0.4. There were 2 possibly product-related mild adverse events.

Conclusion: This study supports that Caya with ContraGel is safe and functions as well as Caya
with N-9 in preventing progressively motile sperm from reaching midcycle cervical mucus.

Key words: Caya; SILCS; diaphragm; postcoital test; ContraGel; nonoxynol-9

Implications: A single-size diaphragm used with a personal lubricant gel containing lactic acid
appears to be safe and to function as well as the same diaphragm used with N-9 in preventing
progressively motile sperm from reaching midcycle cervical mucus.

2
 ACCEPTED MANUSCRIPT

1. INTRODUCTION

Barrier contraceptives are useful for women who cannot or choose not to use hormonal

T
contraceptives or those who have sex infrequently and prefer a non-daily method. PATH, a

P
nonprofit global health organization, and CONRAD, a nonprofit entity with the goal of

RI
improving reproductive health, developed the single-size reusable non-latex diaphragm formerly
known as SILCS [1,2] and now branded the Caya® Contoured Diaphragm (Caya). It has a CE

SC
(Conformité Européenne) mark, mandatory for products in the European Economic Area and
certifying it meets medical device regulations, and 510(k) marketing clearance from the United

NU
States Food and Drug Administration (USFDA). As of 2016, the Caya diaphragm is marketed in
more than 25 countries. MA
The pivotal contraceptive study of Caya was a multi-center trial in 450 couples randomized to
Caya with BufferGel® (n=300) or 2% nonoxynol-9 (N-9) (n=150) [2]. The 6-month typical-use
ED

pregnancy probability was 9.6% for BufferGel, 12.5% for N-9, and 10.4% for both combined.
Estimated cumulative 12-month pregnancy probabilities were 17.8% for typical use of Caya and
13.7% for perfect use [2], comparable to the Ortho® diaphragm with BufferGel or N-9 [3].
PT

Typical-use 12-month contraceptive failure rates for diaphragms with N-9 are approximately
CE

16% [4].

ContraGel®, a personal lubricant containing lactic acid is CE mark approved for use with barrier
AC

devices in Europe and other countries where it has been marketed for nearly 40 years, but has not
been submitted for regulatory approval in the United States. Although ContraGel used alone
failed to show spermicidal activity and was not efficacious as a contraceptive in rabbits
(unpublished data, CONRAD), we hypothesize that, when used with Caya in humans, ContraGel
will potentiate the contraceptive effect of Caya, either via a detrimental effect on sperm due to
ContraGel’s low pH, or by enhancing the seal between Caya and the vagina, or both. In previous
studies, BufferGel, another gel with acid-buffering activity, also did not perform well when
tested alone in in vitro studies and yet performed well in the pivotal studies described above
[2,3]. However, BufferGel is not available on the market and N-9-based spermicide is not
recommended for use in developing countries where women are at high-risk of HIV [5,6]. Since

3
 ACCEPTED MANUSCRIPT

the 2% N-9 formulation used in the Caya pivotal study has been discontinued, Caya used with
ContraGel and with 3% N-9 were evaluated in this study.

Although the post coital test (PCT) was developed to diagnose infertility, it has been used to assess
contraceptive effectiveness of vaginal products such as gels and barriers prior to clinical contraceptive

T
efficacy trials [1,7,8,9]. The primary objective was to assess the effectiveness in preventing sperm from

P
penetrating midcycle cervical mucus of each of the following: Caya with ContraGel, Caya with 3% N-9,

RI
and Caya alone. The secondary objective was to assess safety among female participants and male

SC
partners. In addition, by studying these 3 products, the function of Caya alone and the contribution of
gel in preventing sperm from penetrating midcycle cervical mucus were assessed.

NU
2. MATERIALS AND METHODS MA
Study Design
This was a Phase I multi-center, single-blind, randomized, crossover, non-significant risk study
ED

of Caya with ContraGel, 3% N-9, and no gel carried out at 2 sites: Eastern Virginia Medical
School (EVMS), Norfolk, VA, and Profamilia, Santo Domingo, Dominican Republic. The study
PT

conformed to the Declaration of Helsinki and was approved by each organization’s institutional
review board. The participants took part voluntarily and signed informed consent forms prior to
CE

any study procedures.

Each participant was seen in 9 visits over 6-8 months, completing 4 study cycles with each cycle
AC

consisting of 2 visits: a cervical mucus check (CMC) visit and a postcoital test (PCT) visit. The
first cycle was a baseline cycle, without any product, to demonstrate the participant’s ability to
undergo normal ovulation and produce receptive midcycle cervical mucus, and the partner’s
ability to produce motile sperm capable of penetrating the mucus. Three test cycles followed,
with the sequence of study arms determined by randomization.

Healthy, sexually active women 18–45 years old, not pregnant and not at risk for pregnancy due
to tubal occlusion, at low risk for HIV or sexually transmitted infections (STIs), with menstrual
cycles of 24-35 days, were eligible. They were ineligible if Caya did not fit or they were unable
to insert, position, and remove it, or they were breastfeeding, within 2 months of last pregnancy
outcome, using hormonal contraceptives or an IUD, or had an unevaluated abnormal Pap test,

4
 ACCEPTED MANUSCRIPT

hysterectomy, drug/alcohol abuse, sensitivity to study products, genital findings suspicious for an
STI, a current vaginal infection, or urinary tract infection (UTI). Male partners had to be 18 years
old, at low risk for STIs, and without vasectomy or known fertility problems or sensitivity to
study products.

P T
Volunteers were consented at Visit 1, assessed for eligibility, and instructed to contact the site at the

RI
start of their next menses. The participant was instructed that, from the start of menses until 72 hours
before expected midcycle, she should use study-provided non-spermicidal condoms. Intercourse,

SC
intravaginal product use, and male partner ejaculation were prohibited for 72 hours before expected
midcycle. An ovulation predictor kit (OPK) (Clear Blue® Advanced Digital Ovulation Test, SPD Swiss

NU
Precision Diagnostics GmbH, Petit Lancy, Switzerland) or vaginal ultrasound to observe follicular
development could be used to determine if midcycle. Sites were allowed but not required to conduct
MA
vaginal ultrasounds to observe follicular development and help guide the appointment date for
midcycle. Participants using an OPK were instructed to begin daily urine testing on day 10 of their cycle
(adjusted as needed to normal cycle length) and to contact the site if positive before the scheduled
ED

appointment date. The CMC visit was then rescheduled for the day of the positive test or the next day
at the latest. However, if the participant had not had a positive OPK test result by the day of her
PT

scheduled appointment, she did not need to test further and came to the clinic as scheduled as
determined by the anticipated midcycle date by history. Vaginal pool and endocervical mucus aspirates
CE

(Aspirette®,Cooper Surgical, Shelton, CT) were obtained for evaluation of sperm and adequacy of
cervical mucus before and after intercourse at midcycle.
AC

Visits 2, 4, 6, and 8 were CMC visits, scheduled for the anticipated midcycle date. Samples of
cervical mucus were evaluated for midcycle characteristics, and were scored 0-3 points for each
of 5 categories [amount (none- to 0.3 ml), viscosity (thick to watery), spinnbarkeit (<1cm to
9cm), fern pattern (none to tertiary/quaternary ferning and cellularity (>20 cells per HPF to
none)] following the procedures previously described [10]. A score ≥10 is indicative of good
cervical mucus favoring sperm penetration and was a requirement for continuation. Vaginal
secretions were examined for the presence of sperm. If sperm were found, the cycle was
rescheduled. If the cervical mucus score was <10, the participant returned in 1-2 days or in the
next cycle and procedures were repeated. If no sperm were detected and the cervical mucus score
was ≥10, the participant was instructed to have intercourse without a condom 2-3 hours prior to

5
 ACCEPTED MANUSCRIPT

the PCT visit, scheduled for the same or following day. The participant was randomized to study
product sequence at Visit 4.

Visits 3, 5, 7, and 9 were PCT visits. The postcoital evaluation occurred within 27 hours of a
favorable cervical mucus check visit and within 2–3 h after intercourse. During test cycles, the

P T
participant received the assigned gel for that test cycle (or no gel) at the CMC visit, and was told

RI
not to remove the diaphragm after intercourse but to wear it to the PCT visit. Samples of cervical
and vaginal secretions were evaluated for midcycle characteristics and the presence of sperm

SC
with results managed as listed in Table 1.

The site contacted the participant 7 days after Visit 9 at which time she was exited from the study

NU
unless she had symptoms requiring follow-up.

Study products
MA
Caya (manufactured by Kessel medintim GmbH, Germany) is made of silicone rubber molded
over a circular nylon spring that allows Caya to fit women representing a range of diaphragm
ED

sizes without tight wedging (Figure 1). It has 2 cup-like structures; the larger fits over the cervix
and the smaller is for finger placement for removal.
PT

ContraGel (manufactured by DeltaMed GmbH, Germany) is designed for use with cervical
barriers. It contains water, lactic acid, sodium lactate, cellulose, and sorbic acid as active
CE

ingredients. Gynol II® Vaginal Contraceptive Gel (Extra Strength, 3% N-9), distributed by
Revive Personal Products Company, Madison, NJ, is a commercially available spermicide
AC

Participants were instructed to place about a teaspoon of gel into the cup and around the cervical
rim of Caya immediately prior to insertion. For test cycles without gel, participants were
instructed to insert Caya without water or other lubrication.

Randomization

Participants were randomized to treatment sequence in equal ratios, stratified by site.

Randomization was done by CONRAD using the random permuted blocks method to generate
sequences with block sizes chosen at random. There were 6 possible treatment sequences and 12
treatment codes (A-L): two codes per treatment sequence so that emergency or inadvertent
unblinding would not reveal assignments for all participants within a block.

6
 ACCEPTED MANUSCRIPT

Assignments were communicated to sites via sealed envelopes with center name and
randomization number on the outside and treatment code and sequence on the inside. After a
participant was cleared for randomization, the coordinator opened the next envelope in sequence
and provided the participant with appropriate products.

P T
ContraGel and Gynol II were not repackaged. The study coordinator and participant, by

RI
necessity, knew whether and which gel was used. The PI, examining clinician, and individual
evaluating cervical mucus for midcycle characteristics and sperm were blinded. The clinical

SC
study leader/analyst was also blinded until the data were locked.

NU
Statistical methods
Study endpoints MA
Endpoints for the primary objective (preliminary efficacy) were: proportion of women with an
average (across 9 high power fields) of fewer than 5 progressively motile sperm per high power
field (PMS/HPF), calculated separately for each test cycle; and mean, median, standard
ED

deviation, and interquartile range (IQR) for the average number (across 9 HPFs) of PMS/ HPF
for the baseline cycle and during each test PCT, across all women. Qualitative assessments of
PT

change from baseline were based on the median and IQR, because of expected non-normality of
test cycle data. Fewer than 5 PMS/HPF was considered indicative of acceptable barrier function.
CE

Endpoints for the secondary objective (safety) were treatment-emergent AEs, defined as
occurring on or after product use, that were urogenital, product-related, and/or serious, among
AC

female and male participants. These endpoints are described individually. There were no planned
statistical tests for this objective.

Determination of Sample Size

Up to 40 women could be consented per site, with the intent that at least 10 would complete the
study (both sites combined). Statistical considerations were not used to determine sample size;
instead it reflected the maximum number of subjects who could be enrolled and complete this
rigorous protocol in a reasonable timeframe.

7
 ACCEPTED MANUSCRIPT

3. RESULTS

Disposition of Patients
Twenty-seven women were enrolled (Figure 2). The first participant was enrolled on 1/5/15 and
the last participant visit was 1/29/16. Eleven participants completed a baseline cycle. Two of

T
these did not complete any test cycles. The remaining 9 participants completed all 3 test cycles.

P
RI
The most common reason for discontinuation (8 women) was a baseline PCT with sperm in the
vaginal pool, good cervical mucus, but < 5 PMS/HPF in the cervical mucus. Similar results in a

SC
test cycle would make it appear that the barrier had been successful.

NU
Demographic and Other Baseline Characteristics
On average, participants in the efficacy population were in their mid-thirties (range 28-44 years,
MA
Table 3). Two-thirds were Latina and just over half were of more than one race. There were, as
expected, differences between sites: all women at Profamilia vs. 25% of those at EVMS were
Latina, and all at Profamilia were of more than 1 race vs. none at EVMS. At EVMS, 75% were
ED

black or African-American and 25% Asian.

Participants had, on average, about 11 years of education (range 7-16 years), with the mean
PT

higher at EVMS than Profamilia (13.0 vs. 10.2 years, respectively). All participants were living
with their partners, had been pregnant, and had experienced 1-5 vaginal deliveries, with a mean
CE

of 2.9. None had previously used a diaphragm. The average BMI was higher at EVMS than at
Profamilia: 35.2 (severely obese) vs 22.9 (normal), respectively.
AC

Primary objective – Preliminary Efficacy

The study’s endpoint for preliminary efficacy was the proportion of women who had < 5
PMS/HPF during the test cycle of each study product. Cervical mucus scores were good (≥10) on
the day of PCT in the baseline cycle and all 3 test cycles (Table 3), indicating adequate timing
for the PCT. All 3 study products showed a reduction in motile sperm in mid cycle cervical
mucus in 100% of women (Table 3). Both Caya with ContraGel and Caya with N-9 reduced the
average number of PMS/HPF from 22.5 to zero. Caya alone reduced the average number of
PMS/HPF from 22.5 to 0.4.

8
 ACCEPTED MANUSCRIPT

Secondary objective - safety

There were 16 AEs (EVMS: 3, Profamilia: 13, data not shown). All were mild except for a
moderate migraine headache (Caya with N-9) and a severe intermittent headache (Caya with

T
ContraGel). Two AEs were possibly product-related: bacterial vaginosis with onset 2 days after

P
use of Caya with N-9, and recurrent urinary tract infection (UTI) with onset 3 days after Caya

RI
with ContraGel use. There were no serious AEs and no male AEs reported.

SC
4. DISCUSSION

NU
All 3 study arms performed equally well, with all women meeting the study’s endpoint for
preliminary efficacy of having < 5 PMS/HPF during the test cycle. Both Caya with ContraGel
MA
and Caya with N-9 reduced the average number of PMS/HPF from 22.5 to zero. Caya alone
reduced the average number of PMS/HPF from 22.5 to 0.4, a value greater than zero but still
well below the threshold of 5.0.
ED

These results are very similar to a previous PCT study evaluating Caya with 2% N-9 and K-Y®
Jelly [1]. PCT studies measure progressively motile sperm in cervical mucus as a surrogate
PT

marker for pregnancy but there has been a correlation between results from PCTs and subsequent
performance in effectiveness trials. Three vaginal barriers used with N-9 have been evaluated in
CE

a PCT study and then in a contraceptive effectiveness study prior to FDA approval. No
PMS/HPF were seen among all 10 users of Lea’s Shield with N-9 or all 21 users of Caya with N-
AC

9; 6-month typical-use pregnancy rates of 8.7% and 12.5 were observed in subsequent
contraceptive efficacy studies, respectively [1,2,7,11]. Progressively motile sperm were seen in 1
of 7 users of FemCap with N-9 and a 6-month typical use pregnancy rate of 13.5% was observed
in the contraceptive study [8,12]. The results of this trial of Caya with ContraGel suggest that it
would perform similarly in a contraceptive effectiveness study.

Caya with ContraGel appeared to be safe in a single use. A UTI with onset 3 days after use of
Caya with ContraGel was observed in a woman with a UTI at enrollment. UTIs are common
problems in clinical practice and have been linked to use of diaphragms with spermicide and to
prior UTIs.

9
 ACCEPTED MANUSCRIPT

A limitation of this study is its small sample size, which is common in PCT trials that require
such extensive coordination among the site, participant, and partner.

The finding that Caya alone greatly reduced sperm penetration in cervical mucus suggests that
the barrier function of Caya, and probably diaphragms in general, may be largely responsible for

P T
its contraceptive effect. Nevertheless, gel probably adds additional protection. Three studies in

RI
the 1990s assessed the need for spermicide with a diaphragm. The study designs were
problematic, but the results suggested a trend toward decreasing effectiveness when the device

SC
was used without spermicide [13,14,15]. Additional research is needed before recommending use
of Caya without contraceptive gel. ContraGel, while not spermicidal, may be more effective

NU
than no gel at all due to its pH buffering activity, mechanical blockage of the cervical os, and
improvement of the seal between the diaphragm and the vaginal epithelium.
MA
ContraGel is marketed in more than 25 countries for use with the Caya diaphragm. Recent
studies support the feasibility of introducing Caya with ContraGel in South Africa, Uganda, and
ED

Zimbabwe [16,17,18].

This study suggests that Caya with ContraGel is safe and functions as well as Caya with N-9 in
PT

preventing progressively motile sperm from reaching midcycle cervical mucus.

CE
AC

10
 ACCEPTED MANUSCRIPT

TABLE 1: ACTIONS AT BASELINE CYCLE AND POSTCOITAL TEST

(from Schwartz et al, 2008 [1])

P T
RI
Vaginal pool - Mucus - Endocervix Action in
any sperm adequate Action in Test Cycle

SC
present
(including No Some sperm but >5 progressively Baseline Cycle
fragments)* sperm <5 progressively motile
motile sperm/HPF† sperm/HPF

NU
Yes Yes  Similar results in Barrier
test cycle would successful.
make it appear that
Yes Yes
MA Continued to
the product was next cycle.
successful.
Participant
No Yes 
discontinued
ED

Insufficient material Yes 

Yes Yes  Barrier failed.

PT

Good PCT. Continued to

Continued to next next cycle.
Yes No  cycle
CE

No Yes 
AC

No No 

Insufficient material Yes 

Insufficient material No 
Poor mucus may be Poor mucus
responsible for may be
Yes No 
scarcity or lack of responsible for
sperm in cervix. scarcity or lack
Yes No  Cycle repeated. of sperm in
cervix. Cycle
repeated.
No No 

Insufficient material No 
Ejaculation may not Ejaculation may
have taken place. not have taken
No Yes 

11
 ACCEPTED MANUSCRIPT

Cycle repeated. place. Cycle

repeated.
No No 

Insufficient material Yes 

T
Insufficient material No 

P
* Absence of sperm in vagina means either that ejaculation did not take place or the vaginal environment is unusually hostile to
sperm, possibly due to presence of spermicide in test cycle.

RI
† Confirms that ejaculation took place, but presumed to be insufficient for conception.

SC
NU
MA
ED
PT
CE
AC

12
 ACCEPTED MANUSCRIPT

TABLE 2: DEMOGRAPHIC CHARACTERISTICS AND OTHER BASELINE

CHARACTERISTICS

P T
EVMS Profamilia Both sites combined
N=4 N=5 N=9

RI
Age
Mean 35.3 36.8 36.1

SC
St dev 4.3 6.4 5.3

Median 35.5 36.0 36.0

NU
IQR 8.3 12.0 9.0

Min 30.0 28.0 28.0

MA
Max 40.0 44.0 44.0

Ethnicity
Hispanic/Latina 1 (25%) 5 (100%) 6 (67%)
ED

Not Hispanic/Latina 3 (75%) 0 (0%) 3 (33%)

Race
White 0 (0%) 0 (0%) 0 (0%)
PT

Black or African American 3 (75%) 0 (0%) 3 (33%)

Asian 1 (25%) 0 (0%) 1 (11%)

More than one race 0 (0%) 5 (100%) 5 (56%)
CE

Years of education
Mean 13.0 10.2 11.4

St dev 1.2 3.4 2.9

AC

Median 13.0 9.0 12.0

IQR 2.0 5.0 5.0

Min 12.0 7.0 7.0

Max 14.0 16.0 16.0

Study partner status

Living with 4 (100%) 5 (100%) 9 (100%)
Not living with 0 (0%) 0 (0%) 0 (0%)
Ever used a diaphragm?
Yes, less than 10 times 0 (0%) 0 (0%) 0 (0%)
No 4 (100%) 5 (100%) 9 (100%)
Ever pregnant?
Yes 4 (100%) 5 (100%) 9 (100%)
No 0 (0%) 0 (0%) 0 (0%)
# vaginal deliveries
Mean 2.8 3.0 2.9

13
 ACCEPTED MANUSCRIPT

EVMS Profamilia Both sites combined

N=4 N=5 N=9
St dev 0.5 1.6 1.2

Median 3.0 3.0 3.0

T
IQR 0.8 3.0 1.5

P
Min 2.0 1.0 1.0

RI
Max 3.0 5.0 5.0

SC
BMI
Mean 35.2 22.9 28.4

St dev 9.0 1.7 8.6

NU
Median 36.3 23.5 24.2

IQR 17.0 2.6 13.2

MA
Min 24.4 20.0 20.0

Max 43.8 24.2 43.8

ED
PT
CE
AC

14
 ACCEPTED MANUSCRIPT

TABLE 3: PRELIMINARY EFFECTIVENESS - RESULTS OF PCTS

T
IP
EVMS Profamilia Both Sites Combined

CR
Baseline Caya Caya Caya Baseline Caya Caya Caya Baseline Caya Caya Caya
+N-9 +ContraGel + no gel +N-9 +ContraGel + no gel +N-9 +ContraGel + no gel

S
N=4 N=4 N=4 N=4 N=5 N=5 N=5 N=5 N=9 N=9 N=9 N=9

NU
Cervical Mucus Score Mean 11.5 12.3 11.8 12.8 13.0 11.8 12.6 12.8 12.3 12.0 12.2 12.8
SD 0.6 1.3 0.5 1.9 2.4 1.6 1.3 1.5 1.9 1.4 1.1 1.6
Percent of women with an average

MA
(across 9 HPFs) of fewer than 5
a
progressively motile sperm per HPF 0% 100% 100% 100% 0% 100% 100% 100% 0% 100% 100% 100%
Average number (across Mean 10.3 0 0 0.8 32.3 0 0 0.2 22.5 0 0 0.4
9 HPFs) of progressively Median 8.0 0 0 0 17.4 0 0 0 10 0 0 0

ED
motile sperm per HPF SD 6.8 0 0 1.5 43.9 0 0 0.4 33.4 0 0 1.0
IQR 12.3 0 0 2.3 58.9 0 0 0.5 14.5 0 0 0.5
a
Sperm counts at the EVMS site were performed by counting only one HPF felt to be representative of the whole slide and recording the number of sperm in that field.

PT
CE
AC

15
 ACCEPTED MANUSCRIPT

Figure 1. Caya® Diaphragm. (Image courtesy of Kessel medintim GmbH, Germany)

P T
RI
SC
NU
MA
ED
PT
CE
AC

16
 ACCEPTED MANUSCRIPT

Fig. 2 – Population tree

27 enrolled:

EVMS: 11

T
Profamilia: 16

P
16 did not complete a baseline cycle:

RI
SC
8 – Sperm in vagina, good cervical mucus score, < 5
PMS/HPF in baseline cycle. Similar result in test
cycle would give appearance of successful barrier.

NU
2 – Lost to follow-up before completing baseline cycle
MA 2 – Discontinued due to medical reasons before
completing baseline cycle

2 – Discontinued due to personal reasons before

completing baseline cycle
ED

1 – Discontinued due to inability to achieve good

cervical mucus in 2 attempts at the baseline cycle
PT

1 – Incorrectly discontinued in baseline cycle: no sperm

in vagina or mucus despite good mucus. Ejaculation
11 completed the baseline
CE

may not have taken place. Cycle should have been

cycle:
repeated.
EVMS: 4
AC

Profamilia: 7
2 completed no test cycles:

1 – Discontinued due to personal reasons before

completing a test cycle
9 completed all 3 test cycles:
1 – Discontinued due to inability to achieve good
EVMS: 4 cervical mucus in 2 attempts at the first test cycle

Profamilia: 5

17
 ACCEPTED MANUSCRIPT

Funding

Support for this project is made possible by the generous support of the American people through the United States
Agency for International Development (USAID) under the terms of the HealthTech V Cooperative Agreement
#AID-OAA-A-11-00051. The contents are the responsibility of PATH and CONRAD and do not necessarily reflect

T
the views of USAID or the US Government.

P
RI
SC
NU
MA
ED
PT
CE
AC

18
 ACCEPTED MANUSCRIPT

REFERENCES

T
[1] Schwartz JL, Ballagh SA, Creinin MD, Rountree RW, Kilbourne-Brook M, Mauck CK, Callahan MM.
SILCS diaphragm: postcoital testing of a new single-size contraceptive device. Contraception.

P
2008;78(3):237-44.

RI
SC
[2] Schwartz JL, Weiner DH, Lai JJ, Frezieres RG, Creinin MD, Archer DF, Bradley L, Barnhart KT,
Poindexter A, Kilbourne-Brook M, Callahan MM, Mauck CK. Contraceptive efficacy, safety, fit, and

NU
acceptability of a single-size diaphragm developed with end-user input.

Obstet Gynecol. 2015;125(4):895-903. MA

[3] Barnhart KT, Rosenberg MJ, MacKay HT et al. Contraceptive efficacy of a novel spermicidal
microbicide used with a diaphragm: a randomized controlled trial. Obstet Gynecol. 2007;110(3):577-86.
ED
PT

[4] Hatcher et al, editors. Percentage of women experiencing an unintended pregnancy during the first
year of typical use and the first year of perfect use. Table 31-1. Contraceptive Technology. New York, NY:
Ardent Media. 2011;792-93.
CE
AC

[5] Van Damme L, Ramjee G, Alary M, Vuylsteke B, Chandeying V, Rees H, et al. Effectiveness of COL-
1492, a nonoxynol-9 vaginal gel, on HIV-1 transmission in female sex workers: a randomized controlled
trial. Lancet. 2002;360(9338):971–77.

[6] WHO/CONRAD Technical Consultation on Nonoxynol-9. Geneva: World Health Organization;9–10

October 2001;Summary Report.

[7] Archer DF, Mauck CK, Viniegra-Sibal Al, Anderson FD. Lea's Shield: a phase I postcoital study
of a new contraceptive barrier device. Contraception. 1995;52(3):167-73.

19
 ACCEPTED MANUSCRIPT

[8] Mauck CK, Baker JM, Barr SP, Johanson W, Archer DF. A phase I study of Femcap used with
and without spermicide. Postcoital testing. Contraception. 1997;56(2):111-5.

P T
[9] Amaral E, Perdigão A, Souza MH, et al. Postcoital testing after the use of a bio-adhesive acid

RI
buffering gel (ACIDFORM) and a 2% nonoxynol-9 product. Contraception 2004;70:492–7.

SC
[10] WHO Laboratory Manual for the Examination and Processing of Human Semen. Geneva:

NU
World Health Organization;2010;Fifth Edition

MA
[11] Mauck CK, Glover LH, Miller E, et al. Lea's Shield: a study of the safety and efficacy of a new
vaginal barrier contraceptive used with and without spermicide. Contraception.
1996;53(6):329-35.
ED
PT

[12] Mauck C, Callahan M, Weiner DH, Dominik R. A comparative study of the safety and
efficacy of FemCap, a new vaginal barrier contraceptive, and the Ortho All-Flex diaphragm. The
FemCap Investigators' Group. Contraception. 1999;60(2):71-80.
CE
AC

[13] Bounds W, Guillebaud J, Dominik R, Dalberth BT. The diaphragm with and without spermicide. A
randomized, comparative efficacy trial. J. Reprod. Med. 1995;40(11):764–74.

[14] Ferreira AE, Araujo MJ, Regina CH, Diniz SG, Faundes A. Effectiveness of the diaphragm,
used continuously, without spermicide. Contraception. 1993;48(1):29–35.

[15] Smith C, Farr G, Feldblum PJ, Spence A. (1995) Effectiveness of the non-spermicidal fit-free
diaphragm. Contraception. 1995;51(5):289–91.

20
 ACCEPTED MANUSCRIPT

[16] Lépine A, Nundy N, Kilbourne-Brook M, Siapka M, Terris-Prestholt F. Cost-Effectiveness of

Introducing the SILCS Diaphragm in South Africa. PLoS One. 2015;10(8):e0134510.

P T
[17] Kyamwanga IT, Turyakira E, Kilbourne-Brook M, Coffey PS. Potential for revitalisation of the

RI
diaphragm for family planning in Uganda: a rapid assessment of the feasibility of introducing
the SILCS Diaphragm. Afr J Reprod Health. 2014;18(2):77-86.

SC
NU
[18] van der Straten A, Sahin-Hodoglugil N, Clouse K, Mtetwa S, Chirenje MZ. Feasibility and potential
acceptability of three cervical barriers among vulnerable young women in Zimbabwe. J Fam Plann
Reprod Health Care. 2010;36(1):13-9. MA
ED
PT
CE
AC

21