Transcript The Latest Scientific Knowledge and WHO Guidance about IUDs

The Latest Scientific Knowledge and
WHO Guidance about IUDs
Roy Jacobstein, M.D., M.P.H.
Clinical Director, ACQUIRE
EngenderHealth
IUD Standardization Workshop
Accra, Ghana
June 2006
IUDs: Effectiveness and Safety

Highly effective, comparable to FS, V, Implants
(“Reversible sterilization”)
• 12-13 yrs with CU-T (FDA labels for 10)
• 3-8/1000 become pregnant in 1st yr of use
• WHO study:
– Average annual failure rate 0.4%
– Cumulative failure rate after 12 yrs 2.2%

Very safe for almost all women (e.g., PP, PA, interval; BF; HIV-infected;
young; nulliparous …)
Research (or K-to-P) Paradigm:
Evaluate Concerns via Medical Evidence
Medical
Evidence
Some
“Concern”
New
Understandings
New
 Perceptions
New
“Truths”
Key Provider Concerns Related to Whether or
not to Provide IUDs
Three main provider concerns:

Pelvic Inflammatory Disease (PID)

Infertility

HIV/AIDS
Evidence is reassuring …
Other Related Concerns

How to assess “High Individual Risk” for STIs in lowresource settings

Ruling out pregnancy before giving IUD

Prophylactic antibiotics at insertion — yes or no?

Relation of IUD to ectopic pregnancy

Relation of IUD to anemia
Concern: Does IUD Cause PID?



We know PID is an infectious disease, caused by
sexually-transmitted organisms (Chlamydia,
gonococcus)
But is the IUD an “accomplice”?
Two questions raised; “Is there …
– Q1: ↑Risk from IUD insertion process?”
– Q2: ↑Risk from post-insertion bacterial exposure? (i.e.,
“Does IUD facilitate later PID?”)
Medical Evidence: Low PID Rates among IUD
Users *

WHO study: 23,000 insertions, 51,000 wmn-yrs F/U

Overall rate of PID: 1.6 cases of PID/1000 women-years
(i.e., 998.4 /1000 do NOT get PID)

First 20 or so days: risk ↑ (though still very low:
7/1000 women-years) [This answers Q. 1]

Later periods: comparable risk as in women without
IUD; [This answers Q. 2; evidence-base for WHO SPR
rec. only 1 routine F/U visit needed
* Farley et al., Lancet 1992
Risk of PID: Very Low & Far Lower
Than Many Imagine or Believe
PID Incidence Rate by Time Since Insertion
PID Rate
(per 1000 woman years)
8
8
6
6
4
4
2
2
0
0
1
2
3 4 5 6 7
Months (first year)
8
9 10 11 12
2
3
4
5
Years
6
7
8+
Time Since Insertion
Source: Farley et al, 1992, in FHI 2004
But What About Risk of PID in HighPrevalent STI Settings?

Perhaps WHO data included only low risk women?
(Study done in Thailand and Latin America)

What about in low resource settings, where STI testing
is not feasible?

But no prospective studies exist,
(can only) estimate risks
thus we need to
Modeling the Attributable Risk *
High Risk Setting of 10% Cervical Infection
Simple Screening Questions
No Screening
Only 1 in 667 would
get PID from IUD
Only 1 in 333 would
get PID from IUD
(1.5 cases/1000)
(3 cases/1000)
0.15%
0.30%
* Shelton, Lancet 2001
PID Risk from IUD, in Perspective

PID Risk from IUD:
– 1.6 cases per 1000 women/year
– 1 in 667 insertions or 1 in 333 insertions in high-STI
settings

Other RH Risks in sub-Saharan Africa:
– Lifetime risk of maternal death: 1 in 16 *
– Induced abortion: 35/1000 women / year**
– Abortion death: nearly 7 per 1000 procedures**
*AbouZahr, Br Med Bull 2003
**Guttmacher Institute, 1999
Emerging View on PID
Insertion process, due to presence of sexually- transmitted
bacteria, ↑ short-term PID risk in some women (those at risk
of STIs). This risk is small
 IUD does not appear to facilitate development of PID in
post-insertion period
 Overall, risk of PID is very small
 Even in high-STI settings, risks appear small

(and much smaller than typically believed)
Risk must be considered in context of other risks
 These facts often not widely known -- our challenge

Concern: Does IUD Cause Infertility?
Concern understandable: We know
PID  infertility
 Single PID event  13% chance of tubal
occlusion* (Or, 87% will not get infertility from single

episode of PID)

More PID events  higher chance of infertility
* Westrom, Am J Obstet Gynecol 1975
Medical Evidence:
“IUD Not Associated with Tubal Infertility”*




Mexico: case-control study of nulligravid infertile and
primigravid women
Similar patterns of previous Cu-IUD use
Blood tests for chlamydial antibodies:
infertile women: twice the % of antibodies
Thus, the real infertility “culprit” not IUD but
Chlamydia trachomatis (and gonococcus)
* Hubacher et al., N Engl J Med 2001
Infertility Risk from IUD, in Perspective:
“Harm of Doing” vs “Harm of Not Doing”


In hypothetical high-risk setting, if 25% of PID causes infertility,
without screening by history for every ~2,700 IUD insertions there
would be 1 case of infertility.
In same high STI-prevalence settings, the consequences of denying
IUDs to 2,700 women:
 2,160 pregnancies
 At least 400 serious obstetrical complications
 1-2 deaths from pregnancy and childbirth
 Unknown mortality & morbidity from unsafe abortion
* Shelton, Lancet 2001
Concern: Is IUD Use by HIV-Infected Women
Safe?”
Three possible questions to consider:
1. Does the IUD increase risk of HIV
acquisition?
2. Is IUD use safe for HIV-infected women, i.e.,
does the IUD increase her other health
risks?
3. Does an HIV+ IUD user  ↑risk to her seronegative male partner?
Medical Evidence, Question 1:
No ↑Risk of HIV Acquisition from IUD
Sinei 1996 Martin 1996
10
Harmful
Relative Risk
(Log Scale) 1
& 95% CI
Protective
0.1
Kapiga
1998
Musicco
1995
European
1989
Mati 1995
Carael
1988
Plourde
1992
Morrison
1997
Medical Evidence, Question 2:
IUD Use Is Safe for HIV-infected Women

Cohort studies in Kenya

Compared HIV-infected and HIV-non-infected
women using IUDs

Findings: Same low rates of overall (7-10%)
and infectious (0.2-2%) complications

Conclusion: HIV does not appear to increase risk
of IUD-related adverse events (inc. PID)
Sinei et al, Lancet, 1998
Morrison et al., Br J Obstet Gynaecol 2001
Medical Evidence, Q3: IUD Use by HIV+ Woman is
Safe for Sero-neg Partner *

Ancillary study to Kenyan cohort *

Asked if presence of IUD increases cervical
shedding of HIV? (Increased shedding a proxy for
increased risk of being infective.)
Found cervical shedding of HIV was not
increased with IUD use
 Inferential conclusion: IUD use by HIV+ women
appears safe for HIV-neg partner

* Richardson et al., AIDS 1999
Current WHO Guidance
About IUDs in Response to this
New Evidence
WHO’s Medical Eligibility Criteria for
Contraceptive Use (MEC, 2004)






Based on systematic reviews of latest
clinical & epid research
Covers 19 methods, 120 medical or
special conditions
~ 1700 recommendations on who can
use which methods
Gives guidance to programs &
providers
Informs natl. guidelines, policies &
standards
Helps reduce medical policy & practice
barriers, leading to improved quality &
use of FP methods and services
What Question Is Answered
by WHO’s MEC?
In the presence of a given condition or
classification, e.g., STIs or HIV/AIDS
can a particular FP method be used?
(And with what degree of caution or restriction,
as reflected in 4 categories or gradations
based on risks & benefits)?
WHO Medical Eligibility Criteria, Classification
Categories
Classification
Category
With Clinical
Judgment
With Limited
Clinical Judgment
1
No restriction: Use method in
any circumstances
Yes
Use the method
2
Generally use:
benefits generally outweigh
risks
Yes
Use the method
3
Generally do not use:
risks outweigh benefits
No
Do not use the method
4
Unacceptable health risk:
method not to be used
No
Do not use the method
WHO Medical Eligibility Criteria:
HIV/AIDS and Use of IUDs
HIV/AIDS
2nd Ed.
(1996)
Category
3rd Ed. (2004)
Category
I
C
High Risk of HIV
3
2
2
HIV-infected
3
2
2
AIDS
3
3
2
2
2
AIDS, clinically well on ARV therapy
WHO Medical Eligibility Criteria:
STIs and Use of IUDs
STIs
2nd Ed.
(1996)
Category
3rd Edition (2003)
Category
I
C
2
2
High individual risk of STIs
3
2
Other STIs (excluding HIV and hepatitis)
2
2
4
2
Increased (general) risk of STIs
Current purulent cervicitis or
chlamydial infection or gonorrhea
3
4
How to Determine High Individual Risk of STIs

It’s difficult

Providers need to assess risk for each individual
woman; not to base their decision to give IUD on
local STI levels. (Remember: even with levels of
STI as high as 10%, risk of clinical PID is low)

FHI’s STI (& R/O pregnancy) tools helpful
Key Screening Questions to Determine High
Individual Risk of STIs
1.
2.
3.
4.
5.
Within the last 3 months have you had more than one
sexual partner (outside of a polygamous union)?
Within the last 3 months, have you been told you have an
STI?
Within the last 3 months, has your partner been told he
has an STI?
Does your partner have any symptoms of STI (e.g., penile
discharge or sore, urinary pain or burning)?
Do you think your partner has had another sexual
partner within the last 3 months? (Outside of a regular
polygamous union, if one exists.)
Asking the Screening Questions: Some
Considerations
Need sensitive counseling in a private setting
 Simply asking questions directly may not yield
accurate assessment: sex = a sensitive topic
 Important to be neutral & non-judgmental
 Explain what places a woman at high individual risk
 If risk-associated behaviors / situations presented,
woman often best judge of her own risk and will usually
choose another method if risks explained
 Decide together, or ask her to assess her own risk

Other Provider Concerns about IUDs

Prophylactic antibiotics at insertion?

Ruling out pregnancy (FHI checklist)

Side effects management

Ectopic pregnancy

Anemia
Prophylactic Antibiotics?
“Data do not confirm
the utility
of prophylactic administration
of antibiotics …”*
* Cu-T 380A labeling
Prophylactic Antibiotics?*

Cochrane Review: Meta-analysis of RCTs (randomized,
controlled trials: ”gold standard”)

4 RCTs 1990-1998, compared oral doxycycline (or
azithromycin) vs. placebo or no rx; found:
– No significant ↓ in risk of PID (but ↓ 1/3 in Kenya)
– No ↓ in premature discontinuation of use
– Small ↓ (18%) in unscheduled return visits (33% in
Kenya)
* Grimes et al., Contraception 1999
Prophylactic Antibiotics:
WHO Recommendations*
Question 10. Should prophylactic antibiotics be
provided for Copper-bearing IUD
insertion?
– Prophylactic antibiotics generally not
recommended for Cu-bearing IUD insertion.
– In settings of high prevalence of cervical gonococcal and
chlamydial infections and limited STI screening, such
prophylaxis may be considered.
– Counsel the client to watch for symptoms of PID,
especially during the first month.
* WHO, Selected Practice Recommendations (SPR), 2004
Ruling Out Pregnancy*
Study in Kenya found:
 35% of clients FP were denied services
because they were non-menstruating, yet
 A checklist with six simple questions was able to
rule out pregnancy in 88% of women (who
could then get a FP method)
 had > 99% negative predictive value
(compared to pregnancy test)
Checklist for ruling out pregnancy among FP clients
in primary care”
* Stanback et al., Lancet 1999
Ectopic Pregnancy: Rates by Method
Typical ectopic pregnancy rate per 1000 woman years of method use
10
9
8
7
6.5
6
5
4
3
2.3
2
1.5
1.2
0.6
1
0.4
0.2
0.02
0
No
method
Diaphragm Condom
IUDs
Female
Sterilization
OCs
IUD
Male
(TCU 380A) Sterilization
Adapted from: Sivin 1991, in FHI 2004
Ectopic Pregnancy:
Risk With IUD Use Markedly Reduced




IUD highly protective against ectopic pregnancy — over
a thirty-fold reduction with Cu-T (Absolute risk low,
since IUD so effective a contraceptive)
Pregnancy with IUD in place 5-6x more likely to be
ectopic than in a non-user (relative risk higher)
The large majority (88-90%) of given pregnancies with
IUD in place will not be ectopic
Still must consider ectopics in women using IUDs who
have acute abdomen or sxs of pregnancy)
Anemia



The older, inert IUDs were sometimes associated with
significantly increased vaginal bleeding
With copper IUDs, vaginal blood loss largely unaffected
and significant reduction effect on hemoglobin
levels is uncommon
Thus, WHO classifies anemia as Category 2 for copper
IUDs
Progesterone-Releasing Intrauterine
Systems (IUS)






Mirena® — continuous release of a small amount (20
micrograms) of the progestin, levonorgestrel (same
hormone in Norplant and Jadelle)
Effective 5 years / Failure rate in 1 year: ~ 0.1-0.2%
Same benefits and side effects of progestins
Reduces menstrual cramps and flow
Reduced flow may reduce iron deficiency anemia
But too expensive for programs (30-80 times more costly: ~$40,
vs. $0.50 - $1.50 for Cu-T 380A)
ICA Foundation: Free & Subsidized
LNG-IUS

Partnership: Bayer Schering & Population Council

EngenderHealth on the Board of Directors

Gives combination of donations (free) & sales at public sector price of
US$40 per IUS
Projects in Africa: Ghana, Kenya, Nigeria, South Africa
Who can apply for a donation?: public health organizations (public &
private sector), NGOs
For more information:
– ICA Foundation, PO Box 581, FI-20101 Turku, Finland
– Website: http://www.ica-foundation.org


