# Endocrine Debate: Thyroid Disease in Pregnancy: To Screen, or Not to Screen? **ENDO 2019 · Session E22 · Tuesday 26 March 2019, 11:45 AM–12:45 PM CT · New Orleans, LA** **Moderator:** David S. Cooper, MD (Johns Hopkins University School of Medicine) **For universal screening:** Alex Stagnaro-Green, MD, MHPE, MHA (University of Illinois College of Medicine Rockford) **Against universal screening:** Kristien Boelaert, MD, PhD (University of Birmingham, UK) --- ## About this transcript This is a verbatim machine transcript of the full 60-minute recording, rebuilt for this file. It differs from the earlier auto-transcript in four ways: (1) every paragraph carries a video timestamp in **HH:MM:SS** form, converted to session wall-clock time in the section headings; (2) speakers are identified by voice — ECAPA speaker embeddings clustered into the three voices and anchored on each speaker's solo stretches — not by guessing from content; (3) **every distinct screen state** (65 slides, 88 on-screen states) is reproduced in place, with chart/table descriptions where the slide was graphical; (4) known speech-recognition errors in names, acronyms, units and clinical terms were corrected against the slides, the speakers' own published papers and the guideline documents, and material that the recording does not contain was removed rather than guessed at. *Disfluencies (um, uh, false starts) are preserved because this is a verbatim record. Obvious recognition repetition artifacts were removed. Text marked **[V]** or † is flagged in Appendix D: [V] = wording is reconstructed/uncertain and must be checked against the audio before quotation; † = a short passage reconstructed from the earlier transcript where the audio is partly unintelligible.* ### Speakers and timing | Speaker | Role | Speaks | |---|---|---| | Dr. David S. Cooper | Moderator, Johns Hopkins | 00:00–05:08, 21:58–22:14, 35:33–35:55, 40:48–40:54, Q&A 44:53–54:20, 54:23–54:29, 56:12–56:28, 57:40–60:00 | | Dr. Alex Stagnaro-Green | FOR universal screening | 05:08–21:58 (presentation); 35:55–40:48 (rebuttal); 54:29–56:12 (final statement) | | Dr. Kristien Boelaert | AGAINST universal screening | 22:14–35:33 (presentation); 40:54–44:52 (rebuttal); 56:28–57:35 (final statement) | Recording starts 11:45 AM Central. Add the video timestamp to 11:45 AM for wall-clock time (e.g. 00:22:14 = 12:07 PM — which is why Dr. Boelaert says “Good afternoon”). The session was 60 minutes; the audio file is 60:12.7 and the final 12 seconds are a recording artifact. --- ## 00:00 — Opening and screening primer *11:45 AM CT* > **■ ON SCREEN — 00:00–00:38 · S01** > **Endocrine Debate: Thyroid Disease in Pregnancy: To Screen, or Not to Screen?** > > For - Alex Stagnaro-Green, MD, MHPE, MHA > Against - Kristien Boelaert, MD, PhD > Moderator - David S. Cooper, MD > > *Plain royal-blue title slide with a small green arrow pointer (presenter cursor); no charts.* **[00:00] Dr. David S. Cooper (Moderator):** Well good morning I think it's time to get started. My name is David Cooper from Johns Hopkins University School of Medicine and it's my pleasure and honor to welcome you to this endocrine debate. The topic is thyroid disease and pregnancy to screen or not to screen. It's sort of a Shakespearean title. And we have two expert debaters listed here on the slide. Dr. Stagnaro-Green, who will be arguing for this. And Dr. Kristien Boelaert, who will be arguing against. Um, I, uh, thought it would be good to just remind you about some of the basics of screening. > **■ ON SCREEN — 00:38–01:18 · S02** > **What is screening? (WHO definition)** > > Screening refers to the use of simple tests across a healthy population in order to identify individuals who have disease, but do not yet have symptoms. > Example: breast cancer screening using mammography. > For this morning: will screening (and treating) pregnant women for thyroid disease improve obstetrical and neurodevelopmental outcomes? > > *World Health Organization logo (top left); clip-art of a crowd under a magnifying glass with one figure highlighted in red (top right).* > > *Note: incremental build: second bullet added* **[00:44] Dr. David S. Cooper (Moderator):** Uh, screening refers to the use of a simple test in healthy people to identify people who have a disease but don't have any symptoms yet. An example, obviously, would be breast cancer screening using mammography. And for this morning, uh, the idea of screening and obviously treating, uh, if it's found a disease, uh, women who are pregnant, uh, will this improve obstetrical outcomes and neurodevelopmental outcomes in the child. > **■ ON SCREEN — 01:18–02:22 · S03** > **Screening Criteria** > > * Thyroid disease in pregnancy > ** Serum TSH; Serum TPO Ab > Adapted from: > Wilson JMG, Jungner G. Principles and Practice of Screening for Disease, World Health Organization, 1968 > > *Same slide with only the title visible (the 5 criteria have not yet appeared); WHO logo and crowd/magnifying-glass clip-art present.* > > *Note: incremental build: title-only state* **[01:18] Dr. David S. Cooper (Moderator):** Now, uh, to those of you who are into this concept of screening, you know that there are, uh, basic tenets or criteria for screening. And these were first put out, uh, uh, uh, by, uh, Wilson and Jungner in the world health organization bulletin 50 years ago now, but these are still criteria that are used. The first is that the condition is common. And I have an asterisk there at the condition we're talking about today is thyroid disease in pregnancy, that there's a screening tool for the condition that's available and reliable and cheap. And that's obviously TSH and TPO, which are in fact reliable and easily available. And then that there's an intervention that favorably influences the disease. It has limited adverse effects that the costs, uh, are cost effective. And then there's also an agreed upon policy on whom to treat. And I think that we'll agree that point one and point two, we all agree on. And it's really points three, four, and five that will be the heart of this, uh, discussion this morning. > **■ ON SCREEN — 02:22–02:48 · S04** > **The Debaters** > > For screening > Dr.. Alex Stagnaro-Green is the Regional Dean of the University of Illinois College of Medicine Rockford and is Professor of Medicine, Obstetrics and Gynecology, and Medical Education. > Against screening > Dr.. Kristien Boelaert is a Reader in Endocrinology at the University of Birmingham and a Consultant Endocrinologist at University Hospitals Birmingham, UK. > > *Two small portrait photos at left: Dr. Alex Stagnaro-Green (above 'For screening') and Dr. Kristien Boelaert (above 'Against screening'); green arrow pointer.* **[02:23] Dr. David S. Cooper (Moderator):** So just to introduce our two, uh, debaters, uh, Dr. Stagnaro-Green is I'm sure known to all of you. He's, uh, regional Dean at the University of Illinois College of Medicine in Rockford, professor of medicine, OBGYN and medical education. And from across the pond, uh, Dr. Kristien Boelaert is reader endocrinology at the University of Birmingham and consultant endocrinologist at university hospitals in Birmingham, UK. ## 02:47 — Format, rules, and the pre-debate poll *11:47 AM CT* **[02:47] Dr. David S. Cooper (Moderator):** So this is how we're going to do this. > **■ ON SCREEN — 02:48–03:38 · S05** > **Endocrine Debate: Thyroid Disease in Pregnancy: To Screen, or Not to Screen?** > > Audience Poll > Frame the Position/Share Evidence > Advocate For the Motion (Dr. Stagnaro-Green) > Advocate Against the Motion (Dr. Boeleart) > Rebuttals > Advocate For the Motion > Advocate Against the Motion > Questions from Audience > > *Agenda bullet list on solid blue background; green arrow pointer.* > > *Note: keyframe is before the 'Final Statements' sub-bullets were added (~t=200s, not captured in the 95-state list - see report)* **[02:49] Dr. David S. Cooper (Moderator):** First, we're going to ask you what you think of screening on your audience response system. So get out your cell phones. Then, uh, each of our two speakers is going to speak either for or against, and they will have about 15 minutes to do this. Then each one will have two minutes or five minutes, I think, sorry, to rebut the other person. Then we'll have some questions from you and you can ask questions and I'll have them up here on an iPad and I'll be able to transmit them to our, to our speakers. Then each speaker will have, I think, two minutes to advocate ultimately for their position. And then, uh, we're going to take another poll at the end to see if any of, if either Alex or Christian were able to persuade you one way or the other. > **■ ON SCREEN — 03:38–03:52 · S06** > **SESSION PARTICIPATION AND EVALUATIONS USING THE ENDO 2019 APP** > > STEP 1 Find the session > STEP 2 Tap the Participate Evaluate button > Complete session evaluations by clicking Evaluate This Session > > *Two smartphone screenshots of the ENDO 2019 app (session OR07 Growth, Puberty, and Insulin Action and Resistance; Room 356; Saturday, March 23 11:30 AM - 1:00 PM; CME/MOC 1.50), blue arrows with STEP 1 'Find the session' / STEP 2 'Tap the Participate Evaluate button'; app tags: Evaluate This Session, Polls, Question 1, Social Q&A, Ask, ADD TO MY SCHEDULE; ENDO 2019 logo.* **[03:39] Dr. David S. Cooper (Moderator):** Uh, we're almost ready to get started. This is just to remind you how to do this audience polling on your cell phone. You have to go to this particular event. And, uh, I think you've probably been doing this throughout the course of this meeting. And so here is the first question. > **■ ON SCREEN — 03:52–04:02 · S07** > **Question** > > 1. Are you in favor of screening all pregnant women for thyroid dysfunction? > – Yes > – No > > *Plain purple question slide; red banner at the bottom reads 'LIVE POLLING IS THE NEXT SLIDE'; green arrow pointer.* **[03:54] Dr. David S. Cooper (Moderator):** Are you in favor of screening all pregnant women for thyroid dysfunction? Yes or no? Okay. So just click in your answer, whether you think people should do this or not. > **■ ON SCREEN — 04:03–04:14 · S08** > **1. Are you in favor of screening all pregnant women for thyroid dysfunction?** > > Yes > No > 6 (yellow countdown box, bottom-right) > > *White live-poll screen of the ENDO 2019 app: question at top, 'Yes'/'No' options, large yellow countdown box bottom-right showing '6'.* > **■ ON SCREEN — 04:15–04:20 · S09** > **1. Are you in favor of screening all pregnant women for thyroid dysfunction?** > > Yes 71% > No 29% > Poll is Locked > > *White live-poll results screen of the ENDO 2019 app: question at top, Yes/No rows with percentages and horizontal bars (yellow for Yes, green for No), grey label 'Poll is Locked' bottom-right.* **[04:15] Dr. David S. Cooper (Moderator):** Okay. So lots of people think this is a good idea. And then here's the second question. > **■ ON SCREEN — 04:20–04:30 · S07** > **Question** > > 2. Are you in favor of screening all pregnant women for thyroid autoimmunity? > – Yes > – No > > *Plain purple question slide; red banner at the bottom reads 'LIVE POLLING IS THE NEXT SLIDE'; green arrow pointer.* **[04:22] Dr. David S. Cooper (Moderator):** Are you in favor of screening all pregnant women for thyroid autoimmunity? > **■ ON SCREEN — 04:32–04:42 · S08** > **2. Are you in favor of screening all pregnant women for thyroid autoimmunity?** > > Yes > No > 6 (yellow countdown box, bottom-right) > > *White live-poll screen of the ENDO 2019 app: question at top, 'Yes'/'No' options, large yellow countdown box bottom-right showing '6'.* **[04:40] Dr. David S. Cooper (Moderator):** Yes or no? > **■ ON SCREEN — 04:42–04:47 · S09** > **2. Are you in favor of screening all pregnant women for thyroid autoimmunity?** > > Yes 35% > No 65% > Poll is Locked > > *White live-poll results screen of the ENDO 2019 app: question at top, Yes/No rows with percentages and horizontal bars (yellow for Yes, green for No), grey label 'Poll is Locked' bottom-right.* **[04:42] Dr. David S. Cooper (Moderator):** Okay. That's an interesting outcome. All right. Well, I think we're ready to get going. And, uh, without further ado, I'm going to ask Dr. > **■ ON SCREEN — 04:50–05:01 · S11** > **Select your presentation** > > *Menu screen; the 'David Cooper 11:45 AM - 12:45 PM' card is highlighted/selected with 'Start' and 'My Files' buttons; cards also list Kristien Boelaert and Alex Stagnaro-Green sessions.* > > *Note: Operator screen: 'Select your presentation' list (the session is listed as 11:45 AM – 12:45 PM).* **[04:51] Dr. David S. Cooper (Moderator):** Stagnaro-Green to come up here. And, um, um, Alex, I'm just going to try to put your slides up for you. ## 04:55 — Handover and slide setup *11:49 AM CT* **[04:57] Dr. David S. Cooper (Moderator):** I think that's right here, right? Isn't that it? I think so. Okay. Great. Yep. Get you started. > **■ ON SCREEN — 05:06–05:33 · S12** > **UNIVERSAL SCREENING FOR THYROID DISEASE IN PREGNANCY SHOULD BE PERFORMED** > > Alex Stagnaro-Green, MD, MHPE, MHA > University of Illinois College of Medicine Rockford > > *Red title slide with a cropped Renaissance-style painting of a mother and child with an older woman (Madonna and Child); green arrow pointer.* **[05:06] Dr. David S. Cooper (Moderator):** Okay. Thanks. ## 05:08 — Stagnaro-Green: presentation for universal screening *11:50 AM CT* **[05:08] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Good morning, everybody. I want to thank the Endocrine Society and Dr. Cooper and Kristien for both arranging this debate and participating. Uh, my goal is at the end of this hour to get us all to agree that the data are compelling, that universal screening should be performed in all pregnant women. This, however, has been an area of hot debate over the last 25 years. And the various societies have come up with a very, what I would call, inadequate response, saying that we should go for targeted screening. > **■ ON SCREEN — 05:40–06:12 · S13** > **To Screen or Not to Screen: That is the Question** > > *Collage of 7-8 full-page journal-article screenshots (Clinical Thyroidology 'Comparison of universal...', Eur J Endocrinol 'Universal screening detects two times more thyroid disorders in early pregnancy...', BMJ 'Universal screening...', Lancet 'Universal screening in early pregnancy and treatment of overt and subclinical thyroid dysfunction...', NEJM 'Universal screening...', a Perspectives/Editorial, Frontiers in Endocrinology 'Thyroid Screening in Early Pregnancy: Pros and Cons'); green boxes/arrows and 'Free' tags highlight supportive sentences.* **[05:50] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** But targeted screening is very ineffective. It misses many of the individuals. And, in fact, when you target screen with all the various risk factors, you're screening almost 75% of the population. So, in point of fact, you are almost doing universal screening. Dr. Terry Davies, who was my mentor when I was an Endocrine Fellow and remains a good friend, wrote an editorial when he was the editor of Thyroid. > **■ ON SCREEN — 06:12–06:39 · S14** > **Terry Davies quote: “You can't sit on the fence forever” (2007)** > > "You can't sit on the fence forever. It is very uncomfortable for the backside." > -Terry Davies - 2007 > > *Red quote slide over a photo of a wrought-iron fence in front of a building, decorated with red ribbon graphics and three logos (stylised 'M', a blue square emblem, and the AACE seal).* **[06:22] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** And he was chiding the various endocrine and diabetes organizations for this position of targeted screening, saying you can't sit on the fence forever. It's very uncomfortable for the backside. And we'll come back to that. > **■ ON SCREEN — 06:39–06:56 · S15** > **Criteria for Screening** > > 1. The condition is common. > 2. Screening tool is readily available, reliable, inexpensive and acceptable. > 3. Intervention favorably influences the disease with limited adverse effects. > 4. Screening and intervention are cost-effective. > 5. Agreed-upon policy on whom to treat. > Wilson and Junger (1968) > > *Red slide: white box listing the 5 Wilson-Junger screening criteria at left; large blue UN/World Health Organization emblem at right.* **[06:40] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** As David laid out, we're going to look at five different components to see whether or not universal screening for thyroid disease meets all of the Wilson and Jungner criteria. And we'll start with the condition being common. > **■ ON SCREEN — 06:56–07:28 · S16** > **Prevalence in the Published Literature** > > Condition | Reported Prevalence Range > Overt Hypothyroidism - 0 - 13.1% > Subclinical Hypothyroidism - 1.5 - 42.9% > Isolated Hypothyroxinemia - 0.18 - 31% > Overt Hyperthyroidism - 0 - 2.5% > Subclinical Hyperthyroidism - 0.3 - 13% > > *Two-column table with dark-blue header (Condition | Reported Prevalence Range), 5 rows alternating white/light-blue.* **[06:58] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Although that's a relatively simple question, it could be quite confusing because we took a look at the world literature, and you see the range is quite dramatic. Most for subclinical hypothyroidism, anywhere between 1.5% and 42.9%. But it also ranges quite dramatically for subclinical hypothyroidism and all the other entities. > **■ ON SCREEN — 07:28–08:08 · S17** > **Differences in Diagnostic Criteria Mask the True Prevalence of Thyroid Disease in Pregnancy** > > Need consistent criteria to determine true prevalence > Meta analysis of all previously available data > Studies grouped based on criteria for diagnosis > For Hypothyroidism: > Gold-standard 97.5th percentile TSH cutoff is most accurate > For Hyperthyroidism: > Thyroid function testing done in 2nd trimester is most accurate > Dong & Stagnaro-Green, Thyroid 2019 > > *Text slide, no chart: 5 bullets (two with sub-bullets for Hypo- and Hyperthyroidism) and citation line.* **[07:28] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** So, a colleague and I did a study which was published in last month's Thyroid, where we took a look at what is the true prevalence of thyroid disease in pregnancy. Performed a meta-analysis and used the gold standard for hypothyroidism, 97.5th-percentile TSH cutoff. And for hyperthyroidism, a suppressed TSH in the second trimester, since, as we all know, a certain percentage of women will have a suppressed TSH in the first trimester, that's physiological and due to HCG. > **■ ON SCREEN — 08:08–08:45 · S18** > **Prevalence of Overt Hypothyroidism Based on TSH Cutoff Values** > > % Prevalence (y-axis 0.00% - 3.00%) > All N = 43 - 0.61% > 97.5th Percentile N = 19 - 0.50% > 95th Percentile N = 4 - 1.50% > ATA 2011^ N = 12 - 1.33% > TSH > 4* N = 9 - 0.80% > TSH Cutoff Criteria > Dong and Stagnaro-Green, Thyroid 2019 > > *Vertical bar chart, % prevalence y-axis 0-3%, red '% Prevalence' label; bars: All 0.61% (N=43), 97.5th Percentile 0.50% (N=19; highlighted with red border), 95th Percentile 1.50% (N=4), ATA 2011^ 1.33% (N=12), TSH >4* 0.80% (N=9); red banner 'TSH Cutoff Criteria'.* **[08:09] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** And I want to show you some of the data. So, here the prevalence of overt hypothyroidism is shown on the Y-axis. On the X-axis, if you look at all the studies, it's 0.61%. And if you look at the prior ATA guidelines, which used a cutoff of 2.5 mIU/L, it's 1.3%. But if you use the gold standard, it's half of 1%, or 1 of every 200 women who become pregnant will have overt hypothyroidism. > **■ ON SCREEN — 08:45–09:18 · S19** > **Prevalence of Subclinical Hypothyroidism Based on TSH Cutoff Values** > > % Prevalence (y-axis 0.00% - 25.00%) > All N = 56 - 5.10% > 97.5th Percentile N = 24 - 3.47% > 95th Percentile N = 4 - 3.89% > ATA 2011^ N = 25 - 14.39%* > TSH > 4^ N = 8 - 4.05% > TSH Cutoff Criteria > Dong and Stagnaro-Green, Thyroid 2019 > > *Vertical bar chart, % prevalence y-axis 0-25%; bars: All 5.10% (N=56), 97.5th Percentile 3.47% (N=24; highlighted with red border), 95th Percentile 3.89% (N=4), ATA 2011^ 14.39%* (N=25), TSH >4^ 4.05% (N=8); red banner 'TSH Cutoff Criteria'.* **[08:47] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** If you then look at subclinical hypothyroidism, where the data, as I showed you, is exceedingly broad, all of, if ps4. This is hyp Prideanza isn'th 2. And then finally, we're going to take a look at overt hypothyroidism, or overt hyperthyroidism. > **■ ON SCREEN — 09:18–09:38 · S20** > **Prevalence of Overt Hyperthyroidism Based on Timing of Screening** > > % Prevalence (y-axis 0.00% - 1.60%) > All* N = 21 - 0.64% > 1st Trimester^ N = 14 - 0.91% > 2nd Trimester^ N = 6 - 0.65% > TSH Cutoff Criteria > Dong and Stagnaro-Green, Thyroid 2019 > > *Vertical bar chart, % prevalence y-axis 0-1.6%; bars: All* 0.64% (N=21), 1st Trimester^ 0.91% (N=14), 2nd Trimester^ 0.65% (N=6; highlighted with red border); red banner 'TSH Cutoff Criteria'.* **[09:27] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** And overt hyperthyroidism, over all the studies, is 0.64. And when you focus in on the second trimester, it is 0.65. > **■ ON SCREEN — 09:38–10:30 · S21** > **Thyroid Disease Prevalence in Pregnancy** > > Prevalence Rates: > Overt Hypothyroidism 0.50% > Subclinical Hypothyroidism 3.47% > Isolated Hypothyroxinemia 2.05% > Overt Hyperthyroidism T1 0.91% > Overt Hyperthyroidism T2 0.65% > Subclinical Hyperthyroidism T1 2.18% > Subclinical Hyperthyroidism T2 0.98% > Thyroid Autoimmunity: TPOAb 5-14%; TgAb 3-18% (De Leo & Pearce, 2018) > Dong & Stagnaro-Green, Thyroid 2019 > > *Two boxes: left table 'Prevalence Rates:' (rows Overt Hypothyroidism 0.50%, Subclinical Hypothyroidism 3.47%, Isolated Hypothyroxinemia 2.05%, Overt Hyperthyroidism T1 0.91%, Overt Hyperthyroidism T2 0.65%, Subclinical Hyperthyroidism T1 2.18%, Subclinical Hyperthyroidism T2 0.98%; first row outlined red); right box 'Thyroid Autoimmunity' (TPOAb 5-14%, TgAb 3-18%, De Leo & Pearce, 2018).* **[09:39] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Putting all of these data together, you see that the prevalence for each individual component is quite high. Anywhere from 0.5 to 3.47%. And the point that I am going to drive home today is that based on the prevalence and the impact of overt thyroid disease in pregnancy, which is 1.15%, it is incumbent upon us to do universal screening, whether or not subclinical hypothyroidism or thyroid autoimmunity the identification and treatment of has any beneficial effect. And we'll come back to that. > **■ ON SCREEN — 10:30–11:02 · S22** > **Wilson–Jungner scorecard (progressive build)** > > (table) rows = screening criteria; columns = Overt Hypo, Subclinical Hypo, Isolated Hypothyroxinemia, Overt Hyper, Subclinical Hyper, Thyroid Auto-immunity > Row 1 'The condition is common.': + for all six columns (only row filled so far) > > *Checklist table build state: only criterion 1 row has green '+' cells across all six condition columns; rows 2-5 still empty.* > > *Note: build state 1 of 5* **[10:32] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Major League Baseball starts in two days. All teams are starting on the same day, so I thought I would do a scorecard for our Wilson and Jungner criteria, which you see on the left, and on the right, you see the various thyroid disorders that could occur during pregnancy. And a plus obviously means that it meets all the criteria. And as Dr. Cooper said, these first two are relatively straightforward. Is there a screening tool readily available, reliable, inexpensive, and acceptable? > **■ ON SCREEN — 11:02–11:13 · S15** > **Criteria for Screening** > > 1. The condition is common. > 2. Screening tool is readily available, reliable, inexpensive and acceptable. > 3. Intervention favorably influences the disease with limited adverse effects. > 4. Screening and intervention are cost-effective. > 5. Agreed-upon policy on whom to treat. > Wilson and Junger (1968) > > *Red slide: white box listing the 5 Wilson-Junger screening criteria at left; large blue UN/World Health Organization emblem at right.* > > *Note: near-duplicate of state 19 (same slide re-shown; only cursor position differs)* **[11:09] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Well, inexpensive, I'll talk with number four on cost effectiveness. > **■ ON SCREEN — 11:13–11:42 · S23** > **Test Acceptable to Population** > > Blood Draw at 1st Prenatal/Preconception Visit > Type and Screen > CBC > Rubella & Varicella IgG, RPR, HBsAg, HIV > Thyroid Screening > TSH > > *Red slide with photos of blood-collection tubes: 'Blood Draw at 1st Prenatal/Preconception Visit' showing Type and Screen (pink/red tube), CBC (lavender tube), and bottom row of yellow tubes for Rubella & Varicella IgG, RPR, HBsAg, HIV; 'Thyroid Screening' column shows a single TSH (yellow) tube.* **[11:13] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** But in terms of acceptable and accessible, that is a slam dunk. Blood drawn at the first prenatal or preconception visit includes type and screen, CBC, and all those other studies that you see on the, I guess it's your left hand of the slide. To add a thyroid screen is imperceptible to the patient, and therefore will, of course, be acceptable. And then when we go down to the second row of our Wilson and Jungner baseball scorecard, we see that it's all positives as well. > **■ ON SCREEN — 11:42–12:04 · S22** > **Wilson–Jungner scorecard (progressive build)** > > Row 1 'The condition is common.': + all columns > Row 2 'Screening tool-readily available, reliable, cheap and acceptable.': + all columns > > *Checklist table build state: rows 1 and 2 filled with green '+' across all six columns; rows 3-5 empty.* > > *Note: build state 2 of 5* **[11:53] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** The question is, will any of these six thyroid criteria continue to be positive when we look at the other three? > **■ ON SCREEN — 12:04–12:13 · S15** > **Criteria for Screening** > > 1. The condition is common. > 2. Screening tool is readily available, reliable, inexpensive and acceptable. > 3. Intervention favorably influences the disease with limited adverse effects. > 4. Screening and intervention are cost-effective. > 5. Agreed-upon policy on whom to treat. > Wilson and Junger (1968) > > *Red slide: white box listing the 5 Wilson-Junger screening criteria at left; large blue UN/World Health Organization emblem at right.* > > *Note: near-duplicate of state 19 (same slide re-shown)* **[12:04] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** The first one being an intervention favorably influences the disease with limited adverse effects. And for here, I brought it down to a look at the entire literature, breaking down the next six slides or so into correlation studies or treatment studies. > **■ ON SCREEN — 12:15–12:52 · S24** > **Overt Hypothyroidism Correlation Studies** > > GH/Preeclampsia: Correlation found - Davis et al. 1988, Leung et al. 1993, Buckshee et al. 1997, Sahu et al. 2010, Sharmeen et al. 2014 > Placental Abruption: Davis et al. 1988 > Low Birth Weight: Leung et al. 1993, Su et al. 2011 > Premature Birth: Buckshee et al. 1997, Yang et al. 2018 | No correlation: Abalovich et al. 2002, Nambiar et al. 2011 > Offspring Cognitive Delay/Deficits: Haddow et al. 1999, Klein et al. 2001 > Miscarriage: Abalovich et al. 2002, Hallengren et al. 2009, Nambiar et al. 2011, Aguayo et al. 2013, Taylor et al. 2014 > GDM: Sahu et al. 2010, Sharmeen et al. 2014 > > *Table: Outcome Studied | Correlation Found | No Correlation Found. Rows GH/Preeclampsia (Davis 1988, Leung 1993, Buckshee 1997, Sahu 2010, Sharmeen 2014), Placental Abruption (Davis 1988), Low Birth Weight (Leung 1993, Su 2011), Premature Birth (Buckshee 1997, Yang 2018 | Abalovich 2002, Nambiar 2011), Offspring Cognitive Delay/Deficits (Haddow 1999, Klein 2001), Miscarriage (Abalovich 2002, Hallengren 2009, Nambiar 2011, Aguayo 2013, Taylor 2014), GDM (Sahu 2010, Sharmeen 2014).* **[12:24] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** On the left hand side, the outcome here being gestational hypertension, placental abruption, low birth weight, et cetera. And then in the second column where a correlation is found and a third where a correlation is not found. Overt hypothyroidism, the correlation studies are overwhelmingly positive, showing an adverse impact from miscarriage to preeclampsia to low birth weight. And in the treatment studies that have been done, much less because it is now unethical to do a treatment study in this area, you see the data is also positive, especially for miscarriage. > **■ ON SCREEN — 12:52–13:06 · S25** > **Overt Hypothyroidism Treatment Studies** > > GH/Preeclampsia: Treatment benefit - Tan et al. 2006 | No benefit - Wolfberg et al. 2005 > Offspring Cognitive Delay/Deficits: Treatment benefit - Haddow et al. 1995, Downing et al. 2012 | No benefit - Liu et al. 1994, Lazarus et al. 2012, Hales et al. 2018 > Miscarriage: Treatment benefit - Abalovich et al. 2002, Hallengren et al. 2009, Hirsch et al. 2013, Taylor et al. 2014 > > *Table: Outcome Studied | Treatment Benefit | Treatment No Benefit. Rows GH/Preeclampsia (Tan 2006 | Wolfberg 2005); Offspring Cognitive Delay/Deficits (Haddow 1995, Downing 2012 | Liu 1994, Lazarus 2012, Hales 2018); Miscarriage (Abalovich 2002, Hallengren 2009, Hirsch 2013, Taylor 2014).* > **■ ON SCREEN — 13:06–13:28 · S26** > **Overt Hyperthyroidism Correlation Studies** > > Low Birth Weight: Correlation found - Davis et al. 1989, Momotami et al. 1991, Millar et al. 1994, Phoojaroenchanachai et al. 2001, Luewan et al. 2011, Andersen et al. 2013 > Miscarriage: Davis et al. 1989, Momotami et al. 1991, Andersen et al. 2014, Yang et al. 2018 > Congenital Abnormalities: Davis et al. 1989, Momotami et al. 1991 > Preterm Delivery: Davis et al. 1989, Luewan et al. 2011 > Preeclampsia: Millar et al. 1994 | No correlation: Luewan et al. 2011 > Heart Failure: Davis et al. 1989 > > *Table: Outcome Studied | Correlation Found | No Correlation Found. Rows Low Birth Weight (Davis 1989, Momotami 1991, Millar 1994, Phoojaroenchanachai 2001, Luewan 2011, Andersen 2013); Miscarriage (Davis 1989, Momotami 1991, Andersen 2014, Yang 2018); Congenital Abnormalities (Davis 1989, Momotami 1991); Preterm Delivery (Davis 1989, Luewan 2011); Preeclampsia (Millar 1994 | Luewan 2011); Heart Failure (Davis 1989).* **[13:08] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** The data are even stronger for overt hyperthyroidism. Looking at the left column, we look at low birth weight, miscarriage, congenital abnormalities, et cetera. The correlation found is very positive in every one of those areas, but only one study showed no correlation. > **■ ON SCREEN — 13:28–13:48 · S27** > **Overt Hyperthyroidism Treatment Studies** > > Low Birth Weight: Treatment benefit - Davis et al. 1989, Millar et al. 1994, Phoojaroenchanachai 2001, Aggarawal et al. 2014 > Miscarriage: Momotami et al. 1991 > Preterm Delivery: Davis et al. 1989, Momotami et al. 1991 > > *Table: Outcome Studied | Treatment Benefit | Treatment No Benefit. Rows Low Birth Weight (Davis 1989, Millar 1994, Phoojaroenchanachai 2001, Aggarawal 2014); Miscarriage (Momotami 1991); Preterm Delivery (Davis 1989, Momotami 1991).* **[13:29] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** And in every one of the treatment studies that have been published to date on overt hyperthyroidism, they all show a beneficial effect. So it is clear that identifying, treating overt hypo, overt hyper, has a beneficial effect on both the mother and the fetus. > **■ ON SCREEN — 13:48–14:15 · S28** > **Thyroid Autoimmunity Correlation Studies** > > Miscarriage: Correlation found - Stagnaro-Green et al. 1990, Glinoer et al. 1991, Lejeune et al. 1993, Netto et al. 2004, Negro et al. 2006, Ghafoor et al. 2006, Nambiar et al. 2011, Lepoutre et al. 2012, Liu et al. 2014, Iqbal et al. 2016 | No correlation: Iijima et al. 1996, Plowden et al. 2016 > Preterm Delivery: Correlation found - Stagnaro-Green et al. 2005, Negro et al. 2006, Ghafoor et al. 2006, Karakosta et al. 2012, Kumru et al. 2015, Saki et al. 2015, Iqbal et al. 2016, Meena et al. 2016, Han et al. 2018 | No correlation: Iijima et al. 1996, Haddow et al. 2010, Nambiar et al. 2011 > Offspring Cognitive Delay: Pop et al. 1995, Wasserman et al. 2012 > > *Table: Outcome Studied | Correlation Found | No Correlation Found. Rows Miscarriage (Stagnaro-Green 1990, Glinoer 1991, Lejeune 1993, Netto 2004, Negro 2006, Ghafoor 2006, Nambiar 2011, Lepoutre 2012, Liu 2014, Iqbal 2016 | Iijima 1996, Plowden 2016); Preterm Delivery (Stagnaro-Green 2005, Negro 2006, Ghafoor 2006, Karakosta 2012, Kumru 2015, Saki 2015, Iqbal 2016, Meena 2016, Han 2018 | Iijima 1996, Haddow 2010, Nambiar 2011); Offspring Cognitive Delay (Pop 1995, Wasserman 2012).* **[13:54] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** The data on thyroid autoimmunity are fascinating and quickly changing as we had the Dhillon-Smith publication from the UK that just came out a couple of days ago. If you look at the correlation studies between miscarriage, preterm delivery, cognitive delay, very positive, vast majority of the studies. > **■ ON SCREEN — 14:15–14:30 · S29** > **Thyroid Autoimmunity Treatment Studies** > > Miscarriage: Treatment benefit - Negro et al. 2006, Negro et al. 2016, Lepoutre et al. 2012 | No benefit: Dhillin-Smith, March 23, 2019 > Preterm Delivery: Treatment benefit - Negro et al. 2006, Negro et al. 2016, Nazarpour et al. 2017 > > *Table: Outcome Studied | Treatment Benefit | Treatment No Benefit. Rows Miscarriage (Negro 2006, Negro 2016, Lepoutre 2012 | Dhillin-Smith, March 23, 2019); Preterm Delivery (Negro 2006, Negro 2016, Nazarpour 2017).* **[14:16] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Up until the Dhillon-Smith trial that came out just three days ago, the treatment benefit was positive both for miscarriage and preterm delivery. And that study, the Dhillon-Smith study, is just beginning to become analyzed. > **■ ON SCREEN — 14:30–14:58 · S22** > **Wilson–Jungner scorecard (progressive build)** > > Row 1: + all columns > Row 2: + all columns > Row 3 'Intervention favorably influences the disease': Overt Hypo +, Subclinical Hypo +/-, Isolated Hypothyroxinemia -, Overt Hyper +, Subclinical Hyper -, Thyroid Autoimmunity +/- > > *Checklist table build state: rows 1-3 filled; row 3 shows + for Overt Hypo and Overt Hyper, +/- for Subclinical Hypo and Thyroid Autoimmunity, - for Isolated Hypothyroxinemia and Subclinical Hyper; rows 4-5 empty.* > > *Note: build state 3 of 5* **[14:33] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** So our scorecard starts to change. Overt hypo, overt hyper, clearly indicated that intervention is favorable. I didn't go through the data for subclinical hyper, but there are no data that show intervention makes a difference. Subclinical hypo is mixed and now thyroid autoimmunity is mixed. What about screening and intervention being cost effective? This gets a little bit more complicated. > **■ ON SCREEN — 14:58–15:02 · S15** > **Criteria for Screening** > > 1. The condition is common. > 2. Screening tool is readily available, reliable, inexpensive and acceptable. > 3. Intervention favorably influences the disease with limited adverse effects. > 4. Screening and intervention are cost-effective. > 5. Agreed-upon policy on whom to treat. > Wilson and Junger (1968) > > *Red slide: white box listing the 5 Wilson-Junger screening criteria at left; large blue UN/World Health Organization emblem at right.* > > *Note: near-duplicate of state 19 (same slide re-shown)* **[15:02] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** And here in the United States, we used a cost effective analysis, an analytic tool comparing relative cost and benefit of at least two courses of action. > **■ ON SCREEN — 15:02–16:04 · S30** > **Cost Effectiveness Analysis** > > CEA-analytic tool comparing relative costs and benefits of at least two courses of action > COST (intervention A- B): Blood Test, Extra Visits, Treatment > BENEFIT (effectiveness A-B): Decrease miscarriage, Decrease preterm delivery, Improve IQ of child > > *Red slide: CEA definition; a balance-beam diagram: COST (intervention A- B) list = Blood Test, Extra Visits, Treatment; BENEFIT (effectiveness A-B) list = Decrease miscarriage, Decrease preterm delivery, Improve IQ of child; blue beam with white triangle fulcrum.* **[15:14] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** So under cost for screening, we could look at, and I'm not listing everything, just a couple of things, blood test, extra visits, treatment. And you compare intervention A to intervention B. Intervention A, for purposes of this talk, is universal screening. And intervention B could either be targeted screening or no screening. And then you take a look at benefit. And benefit, decreased miscarriage, decreased preterm delivery, improved IQ are all possibilities. And once again, you look at the effectiveness of universal screening as it pertains to those areas compared to either no screening or targeted screening. > **■ ON SCREEN — 16:04–17:26 · S31** > **Incremental Cost Effectiveness Ratio** > > (Cost of Intervention A-Cost of Intervention B) > / (Effectiveness of A-Effectiveness of B) > Effectiveness measured in Quality Adjusted Life Years (QALYs), > QALY calculated = (0 death, 1=optimal life) X (years) > ICER in US < $50,000/QALY deemed cost effective > > *Formula slide: ICER = (Cost of Intervention A - Cost of Intervention B) / (Effectiveness of A - Effectiveness of B); QALY definitions; bottom line 'ICER in US < $50,000/QALY deemed cost effective'.* **[16:05] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** You then create what's called an incremental cost effectiveness ratio. You take cost intervention A minus cost intervention B and divide it by effectiveness of A over effectiveness of B. But in the United States, to make it that we can use the ICER to compare hypertension screening to mammography to thyroid, for benefit, we translate into that something called a quality, a quality-adjusted life year. And that's calculated by zero being death, one being optimal life. And there are ways called the standard gamble, as well as others, a visual analog scale, where we put a number on that times the number of years. And at the end of the day, you find out the incremental cost effectiveness ratio in the United States is a dollar amount divided by quality. And what is accepted in the U.S. Is if you have a ICER that's under 50,000, that is thought to be cost effective and should be integrated into our screening modalities. To give you a sense, hypertension and mammography screening are both in the 20,000 to 30,000 range. > **■ ON SCREEN — 17:26–18:28 · S32** > **Cost-Effectiveness of Universal Screening** > > Dosiou et al. 2008 - USA - Universal TSH VS No Screening - Dominant > Thung et al. 2009 - USA - Universal TSH VS No Screening - Cost-Effective > Dosiou et al. 2012 - USA - Universal TSH and TPO VS Targeted - Cost-Effective > Donnay Candil et al. 2015 - Spain - Universal TSH and TPO VS Targeted - Dominant > Dosiou, 2012. Universal screening was cost-effective compared to both no screening and targeted screening even when eliminated any benefit of identifying and treating SCH > > *Table: Study | Location | Comparison | Universal Screening Was...; rows Dosiou et al. 2008 USA Universal TSH VS No Screening Dominant; Thung et al. 2009 USA Universal TSH VS No Screening Cost-Effective; Dosiou et al. 2012 USA Universal TSH and TPO VS Targeted Cost-Effective (highlighted row); Donnay Candil et al. 2015 Spain Universal TSH and TPO VS Targeted Dominant. Bottom red text: 'Dosiou, 2012. Universal screening was cost-effective compared to both no screening and targeted screening even when eliminated any benefit of identifying and treating SCH'.* **[17:27] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Four studies have looked at this at this point. Kristi Dosiou, and I was a co-author on this paper in the 2012 paper, looked at universal screening versus targeted screening, and found it to be cost effective, I.e., under 50,000. But in that, we also included a positive impact of subclinical hypothyroidism. We then did what's called a sensitivity analysis and said, let us make the conclusion that identifying and treating subclinical hypothyroidism has no beneficial effect. What is the impact on the ICER? And you can see that universal screening was cost effective compared to both no screening or targeted screening, even when eliminating any benefit and treating subclinical hypothyroidism and treating subclinical hypothyroidism. > **■ ON SCREEN — 18:28–18:58 · S33** > **Cost-Effectiveness - Overt Hyperthyroidism Logic Argument** > > Treatment of Overt Hyperthyroidism is beneficial > Screening method for Overt Hyperthyroidism is the same as for hypothyroidism > Screening for and treatment of Overt Hypothyroidism - shown to be cost-effective > Therefore Screening for Overt Hyperthyroidism can be assumed to be even more cost-effective > > *Bulleted logic argument (4 bullets) on red background; no chart.* **[18:31] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** And then there's a very easy logic argument here for overt hyperthyroidism. The treatment of overt hyperthyroidism is beneficial. It's also common, it's 0.65%. The screening is the same as for hypothyroidism. We've shown that the screening for overt hypo is cost effective. And therefore screening for overt hyperthyroidism can be assumed to be even more cost effective. > **■ ON SCREEN — 18:58–19:26 · S22** > **Wilson–Jungner scorecard (progressive build)** > > Row 1: + all columns > Row 2: + all columns > Row 3: Overt Hypo +, Subclinical Hypo +/-, Isolated Hypothyroxinemia -, Overt Hyper +, Subclinical Hyper -, Thyroid Autoimmunity +/- > Row 4 'Screening and intervention are cost-effective': Overt Hypo +, Subclinical Hypo +/-, Isolated Hypothyroxinemia -, Overt Hyper +, Subclinical Hyper -, Thyroid Autoimmunity +/- > > *Checklist table build state: rows 1-4 filled; row 5 still empty.* > > *Note: build state 4 of 5* **[18:59] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** And so as we continue our scorecard, we see overt hypo and overt hyper continue to meet all the Wilson and Jungner criteria. And then finally, what about an agreed on policy? And this in some ways is the most difficult because organizations not only in the OB world, but also in the endocrine world disagree here. > **■ ON SCREEN — 19:28–20:04 · S34** > **Agreement To Treat Meter** > > Never Treat: Isolated Hypothyroxinemia, Subclinical Hyperthyroidism > Unclear if Should Treat: Subclinical Hypothyroidism, Thyroid Autoimmunity > Always Treat: Overt Hypo, Overt Hyper > > *Horizontal blue meter with a blue dot: left 'Never Treat', centre 'Unclear if Should Treat', right 'Always Treat'; below: Never Treat = Isolated Hypothyroxinemia, Subclinical Hyperthyroidism; Unclear = Subclinical Hypothyroidism, Thyroid Autoimmunity; Always Treat = Overt Hypo, Overt Hyper.* **[19:28] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** So here's an agreement to treat meter. On the left, I think everyone agrees we would not treat isolated hypothyroxinemia or subclinical hypo. And on the right, we would all treat overt hypo or overt hyper. What about in the middle? Well, in the middle, it depends upon who you are, what your philosophy is. Is it a TSH of 2.5 [V5], is it a TSH of 4.0 with positive antibodies? So it's an area of uncertainty. It's part of the art of medicine. > **■ ON SCREEN — 20:04–20:17 · S22** > **Wilson–Jungner scorecard (progressive build)** > > Row 1: + all columns > Row 2: + all columns > Row 3: Overt Hypo +, Subclinical Hypo +/-, Isolated Hypothyroxinemia -, Overt Hyper +, Subclinical Hyper -, Thyroid Autoimmunity +/- > Row 4: Overt Hypo +, Subclinical Hypo +/-, Isolated Hypothyroxinemia -, Overt Hyper +, Subclinical Hyper -, Thyroid Autoimmunity +/- > Row 5 'Agreed-upon policy on whom to treat': Overt Hypo +, Subclinical Hypo -, Isolated Hypothyroxinemia -, Overt Hyper +, Subclinical Hyper -, Thyroid Autoimmunity +/- > > *Checklist table build state: all 5 rows filled (complete table).* > > *Note: build state 5 of 5 (final)* **[20:05] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** So we do have an agreed policy upon whom to treat. And in fact, if you look at overt hypo and overt hyper, they meet every one of the criteria of Wilson and Jungner. > **■ ON SCREEN — 20:17–20:46 · S35** > **Recommended Universal Screening** > > AACE (1999) - Pre-Pregnancy - TSH > AACE (2002) - Pre-Pregnancy or 1st Trimester - TSH > Endocrine Society (Minority) (2007) - 1st Visit - TSH > Italian Thyroid Association & Italian Association of Clinical Endocrinologists (2011) - 1st Visit - TSH, fT4 > Endocrine Society (Minority) (2012) - 1st Visit - TSH > Spanish Society of Endocrinology and Nutrition jointly with Spanish Society of Gynecology and Obstetrics (2012) - Pre-Pregnancy or 1st Trimester - TSH, fT4 Reflex > European Thyroid Association (Majority) (2014) - 1st Trimester - TSH, fT4 Reflex > > *Table: Professional Society | Timing | TSH | fT4 (check marks); rows AACE 1999 Pre-Pregnancy TSH ✓; AACE 2002 Pre-Pregnancy or 1st Trimester TSH ✓; Endocrine Society (Minority) 2007 1st Visit TSH ✓; Italian Thyroid Association & Italian Association of Clinical Endocrinologists 2011 1st Visit TSH ✓ fT4 ✓; Endocrine Society (Minority) 2012 1st Visit TSH ✓; Spanish Society of Endocrinology and Nutrition jointly with Spanish Society of Gynecology and Obstetrics 2012 Pre-Pregnancy or 1st Trimester TSH ✓ fT4 Reflex; European Thyroid Association (Majority) 2014 1st Trimester TSH ✓ fT4 Reflex.* **[20:19] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Now, it's not just Dr. Stagnaro-Green standing up here and saying we should do universal screening. AACE in 1999 and 2002 recommended universal screening. The Endocrine Society in 2007-2012, a minority group there. The Italian and the Spanish Society recommend universal screening. And our sister organization, the ETA, the majority recommend universal screening. > **■ ON SCREEN — 20:46–21:06 · S36** > **ATA 2017 Recommendation - Targeted Screening** > > History of or current signs of thyroid dysfunction > Known thyroid antibody positivity or presence of goiter > History of head/neck radiation or thyroid surgery > Age > 30 > Type 1 diabetes or other autoimmune disorders > History of pregnancy loss, preterm delivery or infertility > 2 or more prior pregnancies > Family history of autoimmune thyroid disease or thyroid dysfunction > Morbid obesity (BMI >= 40) > Use of amiodarone, lithium, or recent administration of iodinated contrast > Residing in an area of known moderate - severe iodine insufficiency > > *Bullet list of 10 risk factors for targeted screening; no chart.* **[20:47] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** The ETA recommends targeted screening using these 10 or 11 criteria. It would take a lot of time for every OB to go through each of these. And we know that even if you do that, you miss a high percentage of women with overt or subclinical disease. > **■ ON SCREEN — 21:06–21:14 · S37** > **Women with overt and/or subclinical disease in pregnancy missed with targeted screening** > > y-axis 0%-100%; 12 vertical bars with no value labels > x-axis studies: Vaidya et al. 2007 [142]; Horacek et al. 2010 [143]; Jiskra et al. 2011 [144]; Chang et al. 2013 [145]; Akafajsi et al. 2013 [146]; Goel et al. 2012 [147]; Altomare et al. 2013 [148]; Kian et al. 2013 [149]; Granfors et al. 2014 [151]; Ahmed et al. 2014 [152]; Nazarpour et al. 2016 [153]; Akram et al. 2017 [154] > > *Vertical bar chart (y-axis 0-100%): 12 bars, no numbers printed, roughly 15-90%; x-axis lists 12 studies: Vaidya et al. 2007 [142], Horacek et al. 2010 [143], Jiskra et al. 2011 [144], Chang et al. 2013 [145], Akafajsi et al. 2013 [146], Goel et al. 2012 [147], Altomare et al. 2013 [148], Kian et al. 2013 [149], Granfors et al. 2014 [151], Ahmed et al. 2014 [152], Nazarpour et al. 2016 [153], Akram et al. 2017 [154] (some x labels partly unreadable - see report).* > > *Note: some x-axis study labels are small/blurred - see report for uncertainty* > **■ ON SCREEN — 21:14–21:29 · S38** > **Where We Are Today: Prevalence of Universal Screening Practices** > > y-axis 0%-100%, x-axis years > 2004 Maine Survey ~48% > 2009 ACOG Survey ~13% > 2010 European Survey ~42% > 2012 ATA Survey ~73% > 2013 Latin America, Asian and East Asian Surveys ~28% > 2014 Brazilian Survey ~83% > 2016 Israeli Survey ~62% > > *Vertical bar chart (y-axis 0-100%) by year: 2004 Maine Survey ~48%, 2009 ACOG Survey ~13%, 2010 European Survey ~42%, 2012 ATA Survey ~73%, 2013 Latin America, Asian and East Asian Surveys ~28%, 2014 Brazilian Survey ~83%, 2016 Israeli Survey ~62%.* > > *Note: percentages read approximately from bar heights - see report* **[21:16] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** And finally, we know that today, when you look at surveys, not only are the majority of physicians screening, but it's increasing from 2004 to 2016. So in summary, overt thyroid disease is important, simple and reliable screening, untreated overt disease causes harm, effective treatment available, targeted screening misses a significant number, > **■ ON SCREEN — 21:29–21:50 · S39** > **SUMMARY** > > 1. Overt thyroid disease is an important health problem in pregnancy > 2. Simple and reliable screening tests are available > 3. Untreated overt disease causes demonstratable harms > 4. Effective treatment is available for these conditions > 5. Targeted screening misses a significant number of women needing treatment > 7. Universal screening has been shown to be cost-effective > 8. A large proportion of providers already do universal screening > > *Bulleted summary, numbered 1-8 with number 6 missing on the slide (jump from 5 to 7); no chart.* > > *Note: slide numbering skips 6 (as printed; 'demonstratable' is the slide spelling)* **[21:46] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Universal screening, universal screening is cost effective, and many providers are doing it. > **■ ON SCREEN — 21:50–21:56 · S40** > **It's time to get off the fence (cartoon)** > > *Red slide with a photo of white gulls perched along a black metal fence; joke following the screening-criteria theme.* **[21:53] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** So going back to Dr. Davies, it's time to get off the fence. Thank you. ## 21:58 — Handover *12:06 PM CT* **[21:59] Dr. David S. Cooper (Moderator):** † I'm sorry — excuse me. **[22:00] Dr. David S. Cooper (Moderator):** † Dr. **[22:00] Dr. David S. Cooper (Moderator):** † Boelaert is going to speak next. > **■ ON SCREEN — 22:00–22:18 · S11** > **Select your presentation** > > *Menu screen; the 'Alex Stagnaro-Green 11:45 AM - 12:45 PM' card is highlighted/selected with Start and My Files buttons.* > > *Note: Operator screen: 'Select your presentation' list (the session is listed as 11:45 AM – 12:45 PM).* **[22:02] Dr. David S. Cooper (Moderator):** † Do you need to do it? **[22:14] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Right. ## 22:14 — Boelaert: presentation against universal screening *12:07 PM CT* **[22:15] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Good afternoon everyone. Thank you to the organizing committee for asking me to be part of this and thank you Alex for your speaking for the motion. > **■ ON SCREEN — 22:23–22:42 · S41** > **Thyroid Disease in Pregnancy: To Screen or Not to Screen - AGAINST the MOTION** > > Endocrine Debate > Thyroid Disease in Pregnancy: To Screen or Not to Screen > AGAINST the MOTION > Kristien Boelaert > Reader in Endocrinology > Consultant Endocrinologist > University of Birmingham > k.boelaert@bham.ac.uk > > *Boelaert title slide: ENDO 2019 banner top-left; University of Birmingham and IMSR logos top-right; three photos of Birmingham buildings along the bottom.* **[22:36] Dr. Kristien Boelaert (arguing AGAINST universal screening):** I'm going to try and convince you over the next 15 minutes now why I think we should not screen all pregnant women for thyroid dysfunction and thyroid autoimmunity. > **■ ON SCREEN — 22:42–22:46 · S42** > **Disclosure** > > Nothing to disclose > > *Mostly white slide with only the title and the sentence 'Nothing to disclose'; ENDO 2019 banner bottom-left; ENB/AACE-style flower graphics bottom-right.* **[22:45] Dr. Kristien Boelaert (arguing AGAINST universal screening):** I have no disclosures. > **■ ON SCREEN — 22:46–24:02 · S43** > **(1) Thyroid dysfunction in pregnancy** > > Increased maternal total T4 and total T3 > Declining free T4 and free T3 with advancing gestation > Overt Hypothyroidism: High TSH low FT4 0.2-0.6% > Overt Hyperthyroidism: Low TSH high FT4 0.2-0.4% > Subclinical Hypothyroidism: High TSH normal T4 2-18% > Isolated Hypothyroxinemia: Normal TSH low FT4 ~2% > Transient Gestational Thyrotoxicosis 1-3% > Chan and Boelaert Clin Endo 2015 82: 313-326 > Korevaar et al. Nature Reviews 2017 13: 610-622 > Taylor et al. Frontiers in Endocrinology 2018 9: 1-7 > > *Physiology diagram: central pregnant-woman silhouette with arrows labelled TSH, TBG, Estrogen, HCG, maternal thyroxine (T4); boxes 'Increased maternal total T4 and total T3' and 'Declining free T4 and free T3 with advancing gestation'; labels 'Placental development and D3 metabolism', 'Fetal development'; title above diagram 'Physiological changes during pregnancy'.* > > *Note: incremental build: right-hand prevalence column added* **[22:48] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Some of this will be repeated from what we've already heard, I think that's inevitable, but I think one of the first points to make is that in order to interpret thyroid function during pregnancy we need to remember the physiological changes that occur which cause changes in binding proteins and which make interpretation of thyroid function very difficult. So saying that this is a simple test that we can easily interpret I think is a gross overstatement. We've seen a significant number of slides from Alex on prevalence of these conditions and I'm not standing up here to state that overt thyroid dysfunction does not require treatment in pregnancy. I do believe that requires treatment and I do believe that we need to pick that up. However, what we do see when we screen a large number of people are a number of mild thyroid dysfunctions, subclinical thyroid disease, isolated hypothyroxinemia and in the case of hyperthyroidism transient gestational thyrotoxicosis with some prevalence rates given there. **[23:52] Dr. Kristien Boelaert (arguing AGAINST universal screening):** And I think it's an undeniable fact that we don't actually really know what to do with this, that evidence is controversial and that we do not have an agreed policy on this. > **■ ON SCREEN — 24:02–25:16 · S44** > **(1) Thyroid autoimmunity in pregnancy** > > TPO abs: 5-14% of unselected pregnancies (10-30% in subfertile/recurrent miscarriage population) > TABLET Trial: 9.5% TPO-positive (N = 19,237 screened) > TPO status affects treatment thresholds for starting levothyroxine > 4 line charts: FT4 in pmol/L and TSH in mIU/L vs hCG in IU/L, for TPOAb negative vs positive; p<0.0001 for FT4/TSH decline in negatives, p=0.79 / p=0.92 in positives. > Korevaar et al. JCEM 2017 102: 69-77 > > *Bullet slide (3 bullets); citation Korevaar et al. JCEM 2017; in the later state 4 small line charts appear (see per-state note).* > > *Note: incremental build: 4 charts added below bullets* **[24:03] Dr. Kristien Boelaert (arguing AGAINST universal screening):** So sticking with the first criterion that we're looking at that these are common conditions, yes indeed they are. TPO antibodies up until we had the TABLET trial we thought were present in about 5-14% of unselected pregnancies and much higher in the subfertile or recurrent miscarriage population. So when we conducted a TABLET trial we did not find this. We had set out to screen 5,000 women. We had to screen 20,000 in order to reach our target because we only found a TPO antibody positivity rate of 9.5%. A significant number of you would screen for thyroid function but not for TPO antibodies. Now all guidelines tell us that in order to make treatment decisions we need the TPO status. So this brings us to a further complication in interpreting thyroid function tests. These data from the Rotterdam group showing that in women who have positive TPO antibodies interpretation of thyroid function becomes even more difficult because the expected physiological changes such as a dip in serum TSH and a rise in free T4 do not occur. > **■ ON SCREEN — 25:16–26:24 · S45** > **(2) Screening tool: TFT and TPO testing** > > Interpretation of thyroid function tests not straightforward > Use of gestation-specific reference ranges recommended - not readily available > Continued debate and changes on "normal reference ranges" > Reference ranges derived from cross-sectional samples not considering longitudinal trajectories > FT4 concentration measurements often problematic > > *Red bullet slide (5 bullets); no chart.* **[25:17] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Coming on to the second criterion there, I've already referred to it, but we have an easily identifiable screening tool. Well I looked at, I can tell you, a lot, a lot of thyroid function tests in the TABLET trial and interpretation is not straightforward. Most guidelines tell us to use gestation specific reference ranges. That is okay for people like me who work in large centers, but for the majority of us in reality and in real life that is not possible. We are continuously debating both here and at the American Thyroid Association meeting, any endocrine meeting, about what normal reference ranges are. So again, there is no agreement on this. And often they don't, these are derived from cross-sectional samples, they don't take longitudinal changes such as, for example, what I showed on the previous slide, effects of TPO antibodies on interpretation of thyroid function tests. And then we all know that interpretation, for example, of a free T4 measurement can be very problematic and guidelines tell us that ideally we should be measuring total T4. **[26:18] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Again, that is not something that's available even in a large center where I work. So I would say that the second criterion is not fulfilled. > **■ ON SCREEN — 26:24–27:11 · S46** > **(3) Intervention favorably influences disease? - CATS TRIAL** > > CATS TRIAL > Treatment of sub. hypothyroidism or hypothyroxinemia with 150ug LT4 > Start treatment median week 13, IQ test at age 3 years > Mean offspring IQ: LT4 99.2 | Placebo 100.0 | Difference -0.8 | P 0.40 | (N/groups) (323/326) > No differences in obstetric outcomes > Lazarus et al. N Engl J Med. 2012 366: 493-5 > > *Table 'Mean offspring IQ': LT4 | Placebo | Difference | P | (N/groups); row Subcl. Hypo OR Hypothyroxinemia 99.2 | 100.0 | -0.8 | 0.40 | (323/326); note 'No differences in obstetric outcomes'.* **[26:27] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Do we know that the intervention favorably influences disease? Well, we have large trials now. So we have a CATS trial and then the Casey trial that was done in the US. These were powered on and set out to look at neurocognitive effects of treatment of subclinical hypothyroidism and hypothyroxinemia in the case of the CATS trial where a relatively large dose of levothyroxine, 150 micrograms, was given. This was started at a median time of 13 weeks and an IQ test was done at three years. And as you can see, this was a negative trial with no difference in offspring IQ. Secondary analysis looked at obstetric outcomes and there was no difference. So levothyroxine intervention did not favorably influence the disease. Similarly, the Casey trial, in this case it was women with subclinical hypothyroidism and hypothyroxinemia who got a slightly different dose of T4. > **■ ON SCREEN — 27:11–27:42 · S47** > **(3) Intervention favorably influences disease - NICHD TRIAL** > > NICHD TRIAL > Treatment of women with sub. hypothyroidism (100ug LT4) or hypothyroxinemia (50ug LT4) > Mean treatment start week 17, IQ test age 3/5 yrs > Powered to detect a 5 point IQ difference > Mean offspring IQ - Subclinical hypothyroidism: LT4 97 | Placebo 94 | Difference +3 | P 0.71 | (323/326) > Mean offspring IQ - Hypothyroxinemia: LT4 94 | Placebo 91 | Difference +3 | P 0.30 | (254/253) > No differences in obstetric outcomes > Casey et al. N Engl J Med 2017 376: 815-25 > > *Table 'Mean offspring IQ': LT4 | Placebo | Difference | P | (N/groups); rows Subclinical hypothyroidism 97 | 94 | +3 | 0.71 | (323/326); Hypothyroxinemia 94 | 91 | +3 | 0.30 | (254/253); note 'No differences in obstetric outcomes'.* **[27:22] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Mean start of treatment was at 17 weeks and IQ was measured at three to five years. This was powered to detect a five point IQ difference. And lo and behold, again, this was a negative trial with no effects on neurocognitive outcomes and no differences in obstetric outcomes. So again, the intervention does not favorably influence the disease. > **■ ON SCREEN — 27:42–28:19 · S48** > **(3) Intervention favorably influences disease - TABLET results** > > Primary outcome - Livebirth >=34 weeks: Levothyroxine 176/470 (37.4%) | Placebo 178/470 (37.9%) | Comparison 0.97 (0.83, 1.14) | p=0.74 > Secondary pregnancy outcomes (as a proportion of women who achieved pregnancy within 12 months; N=266 vs N=274): > Clinical pregnancy at 7 weeks: 237/266 (89.1%) vs 248/274 (90.5%); 0.98 (0.93, 1.04) > On-going pregnancy at 12 weeks: 194/266 (72.9%) vs 200/274 (73.0%); 1.00 (0.90, 1.11) > Miscarriage <24 weeks: 75/266 (28.2%) vs 81/274 (29.6%); 0.95 (0.73, 1.23) > Livebirth >=34 weeks: 176/266 (66.2%) vs 178/274 (65.0%); 1.02 (0.90, 1.15) > Dhillon-Smith et al. N Engl J Med 2019: epub > > *Table: Outcome | Levothyroxine | Placebo | Comparison; primary outcome Livebirth >=34 weeks 176/470 (37.4%) vs 178/470 (37.9%), 0.97 (0.83, 1.14), p=0.74 (red oval highlights this row); secondary pregnancy outcomes (N=266 vs N=274): clinical pregnancy at 7 weeks 237/266 (89.1%) vs 248/274 (90.5%) 0.98 (0.93, 1.04); on-going pregnancy at 12 weeks 194/266 (72.9%) vs 200/274 (73.0%) 1.00 (0.90, 1.11); miscarriage <24 weeks 75/266 (28.2%) vs 81/274 (29.6%) 0.95 (0.73, 1.23); livebirth >=34 weeks 176/266 (66.2%) vs 178/274 (65.0%) 1.02 (0.90, 1.15).* **[27:43] Dr. Kristien Boelaert (arguing AGAINST universal screening):** And then data that have already been referred to, looking at thyroid autoimmunity, dr. Rima presented here on Saturday looking at giving levothyroxine or placebo to women who are euthyroid but have positive TPO antibodies found no difference in the primary outcome which was live birth at a minimum of 34 weeks between treatment and placebo. None of the secondary outcomes were significant either. So again, no evidence that giving levothyroxine has a positive influence here. > **■ ON SCREEN — 28:19–28:42 · S49** > **(3) Intervention favorably influences disease - TABLET subgroup forest plot** > > Subgroups: maternal age (<35, >=35 yr); No. of previous miscarriages (0, 1 or 2, >=3); Thyrotropin concentration at baseline (<=2.5, >2.5 mIU/liter); Infertility treatment at time of randomization (Yes/No); Thyroid peroxidase antibody concentration at baseline (Low/High); Race or ethnic group (White, South Asian, Chinese/black/other); BMI (<25, >=25) > All participants: Levothyroxine 176/470 vs Placebo 178/470; Relative Risk 0.97 (0.83-1.14); P for interaction 0.74 > Relative risk axis 0.50 / 0.75 / 1.00 / 1.50 / 2.50 with 'Placebo Better' and 'Levothyroxine Better' > Dhillon-Smith et al. N Engl J Med 2019: epub > > *Forest plot of relative risk (95% CI) by subgroup (maternal age, number of previous miscarriages, thyrotropin concentration at baseline <=2.5 vs >2.5 mIU/L [highlighted row], infertility treatment, TPO antibody, race/ethnic group, BMI) plus 'All participants' 176/470 vs 178/470, 0.97 (0.83-1.14), p interaction 0.74; axis 0.50-2.50 with 'Placebo Better' / 'Levothyroxine Better'.* > > *Note: dense table - subgroup rows condensed; individual cell values omitted* **[28:20] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Particularly what we looked at, particularly what we looked at, particularly what we looked at when we did a further subgroup analysis in the TABLET trial, looking at what's generally recommended in the current guidelines that we have, which is to treat women who are TPO antibody positive if they have a serum TSH over 2.5. I think that is what most people do. Again, we found no difference in our subgroups. > **■ ON SCREEN — 28:42–29:22 · S50** > **(3) Intervention favorably influences disease - TABLET trial (survival curve)** > > TABLET trial - line graph: Pregnancy (%) (0-100%) vs Weeks from Conception to Pregnancy End; two nearly identical curves (Levothyroxine, Placebo); number-at-risk table below (Levothyroxine 470, 324, 310, 301, 255, ...; Placebo 470, 324, 303, 291, 242, ...) > Dhillon-Smith et al. N Engl J Med 2019: epub > > *Line graph: pregnancy (%) y-axis 0-100% vs weeks from conception to pregnancy end; two nearly overlapping lines (Levothyroxine blue, Placebo pink); below the axis a 'number at risk' table (Levothyroxine/Placebo rows, ~470 starting) with counts at 0,3,6,9,12,15,20,25,30,35,40,45+ weeks.* > > *Note: number-at-risk values approximate - see report* **[28:45] Dr. Kristien Boelaert (arguing AGAINST universal screening):** And just to illustrate here how in the TABLET trial there was no difference at any point, which, OST? Of the percentage of women that were pregnant, and you can see that both in the treatment and in the placebo group, early and late miscarriages were very similar. Did women take their levothyroxine? Well, yes, they did. As you can see that in the women who were randomized to receive 50 micrograms of levothyroxine, serum TSH went down and serum 3-4 went up, all still within the normal range, but women were indeed taking their medication. Do we have clear evidence that this intervention is safe? > **■ ON SCREEN — 29:22–30:30 · S51** > **(3) Intervention may cause harm - optimal vs overtreatment** > > c Optimal treatment scenario > d Overtreatment scenario > y-axis: Mean cortex volume (cm3) 530-550 > x-axis: Maternal free T4 (pmol/l) - 7.4, 12.2, 20.1, 33.1, 54.6 > Shaded normal range band; blue fitted curve with pink observed points; overtreatment curve declines steeply at high maternal free T4 > Korevaar et al. Nature Reviews 2017 13: 610-622 > > *Two line graphs (c 'Optimal treatment scenario', d 'Overtreatment scenario'): y-axis Mean cortex volume (cm3) 530-550, x-axis Maternal free T4 (pmol/l) 7.4, 12.2, 20.1, 33.1, 54.6; shaded 'Normal range' band; blue fitted curve with pink observed points; the overtreatment curve declines at high fT4.* **[29:28] Dr. Kristien Boelaert (arguing AGAINST universal screening):** And I think for many years we all thought, well, we may not have good evidence that this is actually beneficial, but we're probably not doing much harm. I would like to contest that. So we're getting now increasing evidence that overtreating women during pregnancy does indeed cause harm. So these first data from Tim Korevaar, and these are derived from the Generation R study, where essentially what you can see is that there is an inverse U-shaped curve where both at the lower end and the higher end of free T4, the effects on neurocognitive outcome, or in this case, mean cortex volume, in the offspring brain is significantly affected by both low and high concentrations of free T4. And it appears that if you look here on this graph, if you go to an overtreatment scenario where you give too-high doses of levothyroxine, you will actually, you know, rather than get this optimal cortex volume, you will slide off the slope, indicating that overtreatment may cause harm. > **■ ON SCREEN — 30:30–31:46 · S52** > **(3) Intervention may cause harm - FT4 >97.5th percentile and verbal IQ** > > FT4 >97.5th percentile > INMA: High FT4 n=31, Reference n=628, beta -0.1 (-3.6, 3.7) > Generation R: n=78 vs 1,569, beta -2.4 (-5.8, 1.1) > ALSPAC: n=61 vs 1,162, beta -0.1 (-3.6, 3.3) > Overall: n=170 vs 3,359, beta -0.8 (-2.9, -1.2) > Forest plot: Lower verbal IQ <-> Higher verbal IQ > Right chart: Adjusted mean difference in verbal IQ (95% CI) vs Free T4 (pmol/L) categories <10.0, 10.0-10.99, 11.0-11.99, 12.0-12.99, 13.0-18.99, 19.0-19.99, >20.0 > Increased risk of ADHD in proportion of women overtreated in CATS (unpublished) > High fT4 associated with autism in offspring > Levie et al. JCEM 2018 103: 2967-2979 > Andersen et al. Thyroid 2018; 28: 537-546 > > *Left table: Cohort | High FT4 (n) | Reference (n) | beta (95% CI); INMA 31/628 -0.1 (-3.6, 3.7); Generation R 78/1,569 -2.4 (-5.8, 1.1); ALSPAC 61/1,162 -0.1 (-3.6, 3.3); Overall 170/3,359 -0.8 (-2.9, -1.2), with forest plot 'Lower verbal IQ / Higher verbal IQ'. Right: adjusted mean difference in verbal IQ (95% CI) by free T4 category (<10.0, 10.0-10.99, 11.0-11.99, 12.0-12.99, 13.0-18.99, 19.0-19.99, >20.0 pmol/l).* **[30:32] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Further studies, so the one on the left for you is from an IPD meta-analysis, looking at a number of studies where we look where the verbal IQ of children was studied in a number of trials or studies showing that those with a higher free T4 over the 97.5th centile had a lower verbal IQ, and then a similar study performed, so this was a case control study, the graph on the right from the Danish cohort, where, again, you can see this inverse shape, but confirming that if free T4 was at the higher end of normal, there is a significant reduction in verbal IQ. Importantly, what was also shown in both these studies was that higher free T4 was associated with autism in the offspring, and similar data, which are unpublished, but have been shown that there was a significant increased risk of ADHD in women who were given high doses of levothyroxine who had biochemical evidence of being overtreated in the CATS trials. So again confirming that giving this treatment and that certainly over-zealous levothyroxine replacement is associated with deleterious effects. > **■ ON SCREEN — 31:46–32:25 · S53** > **(3) Intervention may cause harm - Maraka adverse outcomes table** > > Pregnancy loss: 89 (10.6) vs 614 (13.5); Unadjusted 0.76 (0.60 to 0.96); Adjusted 0.62 (0.48 to 0.82); P 0.01 > Preterm delivery: 60 (7.1) vs 236 (5.2); 1.41 (1.05 to 1.88); 1.60 (1.14 to 2.24); 0.01 > Preterm labor: 111 (13.2) vs 569 (12.5); 1.08 (0.87 to 1.34); 1.14 (0.89 to 1.46); 0.29 > Premature rupture of membranes: 42 (5.0) vs 220 (4.8); 1.04 (0.74 to 1.45); 0.97 (0.66 to 1.42); 0.87 > Placental abruption: 7 (0.8) vs 36 (0.8); 1.05 (0.47 to 2.37); 1.60 (0.65 to 3.93); 0.30 > Gestational diabetes: 101 (12.0) vs 401 (8.8); 1.41 (1.12 to 1.78); 1.37 (1.05 to 1.79); 0.02 > Gestational hypertension: 49 (5.8) vs 221 (4.8); 1.24 (0.90 to 1.70); 1.27 (0.88 to 1.82); 0.21 > Pre-eclampsia: 46 (5.5) vs 177 (3.9); 1.43 (1.03 to 2.00); 1.61 (1.10 to 2.37); 0.01 > Poor fetal growth: 78 (9.3) vs 397 (8.7); 1.07 (0.83 to 1.38); 1.12 (0.84 to 1.50); 0.45 > Tachycardia: 18 (2.1) vs 90 (2.0); 1.08 (0.65 to 1.81); 1.77 (1.00 to 3.11); 0.05 > Maraka et al. BMJ 2017, 356: i6865 > > *Table: No. (%) events for thyroid hormone treatment (n=843) vs no thyroid hormone treatment (n=4562); Odds ratio (95% CI) unadjusted and adjusted; P value; outcomes pregnancy loss (89 (10.6) vs 614 (13.5); 0.76 (0.60 to 0.96); 0.62 (0.48 to 0.82); 0.01), preterm delivery (60 (7.1) vs 236 (5.2); 1.41 (1.05 to 1.88); 1.60 (1.14 to 2.24); 0.01), preterm labor, premature rupture of membranes, placental abruption, gestational diabetes (101 (12.0) vs 401 (8.8); 1.41 (1.12 to 1.78); 1.37 (1.05 to 1.79); 0.02), gestational hypertension, pre-eclampsia (46 (5.5) vs 177 (3.9); 1.43 (1.03 to 2.00); 1.61 (1.10 to 2.37); 0.01), poor fetal growth, tachycardia; several rows outlined in red.* **[31:47] Dr. Kristien Boelaert (arguing AGAINST universal screening):** And in this study, which is, and then this study which is a registry-based study looking at outcomes of women who were treated for subclinical hypothyroidism during pregnancy. You can see that what's highlighted in green that if you treat subclinical hypothyroidism in pregnancy, this was looking at about 5,000 women, there was indeed a beneficial effect on pregnancy loss. But other outcomes such as preterm delivery, gestational diabetes and preeclampsia were actually increased with treatment, again lending further support that this may not be altogether an innocent or an altogether safe treatment. > **■ ON SCREEN — 32:25–33:34 · S54** > **(4) Cost-effectiveness** > > Economic model demonstrating incremental cost-effectiveness ratio of 7258 USD per QUALY > Based on 2 small RCT in area of mild iodine deficiency > Overall costs of personnel training, public education, establishment of healthcare infrastructure > Burden of over-diagnosis and over-treatment > > *Bullet slide (4 bullets); later state adds a pasted image of Munch's 'The Scream' with the label 'Anxiety and inconvenience for patients'.* **[32:26] Dr. Kristien Boelaert (arguing AGAINST universal screening):** We've had a very elegant explanation on how we calculate qualities and what cost effectiveness is and I knew that Alex would refer to this study where indeed significant cost effectiveness was shown with 7,258 US dollars per QALY saved. However, this study was based on two small randomized controlled trials performed in Italy, which is an area of mild iodine deficiency and may therefore not be extrapolatable to everyone. And I don't think this took into account the overall cost of personnel training, public education and establishment of a healthcare infrastructure. But importantly, what has not been taken into account here is a burden of over-diagnosis and over-treatment, which has not been calculated in here. And subsequently, and I think it's very difficult to put a price on this, but the anxiety and inconvenience that we cause to patients by continuously screening them and then ending up with thyroid function tests where we don't actually really know what to do with, I think is something that we also need to take into account. > **■ ON SCREEN — 33:34–34:37 · S55** > **(5) No agreed policy on whom to treat - ATA 2017 Guidance** > > ATA 2017 Guidance: > Use trimester-specific reference ranges, similar assays or TSH > 4 mIU/L > Consider levothyroxine if TSH > 2.5 mIU/L in TPO+ or > 4.0 mIU/L in TPO- women > Don't routinely treat isolated hypothyroxinemia > ETA 2014 Guidance: > Ideally trimester and population-specific ranges, otherwise (2.5 mIU/L, 3.0 mIU/L and 3.5 mIU/L in different trimesters > Treat SCH before and during pregnancy > Consider treatment of isolated hypothyroxinemia in 1st trimester > Alexander et al. 2017 ATA guidelines, Thyroid 2017 27: 315-389 > Lazarus et al. 2014 ETA guidelines Eur Thyroid J 2014;3: 76-94 > > *Red slide with ATA 2017 Guidance bullets; later state adds a second green sub-heading 'ETA 2014 Guidance' with 3 more bullets.* > > *Note: incremental build: ETA 2014 Guidance section added* **[33:36] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Coming on to our fifth criteria, and I think this is probably where Alex and I will disagree the least, we do not have a policy on whom to treat. ATA guidance and ETA guidance is quite different. We are continuously still, at every conference we go to, debating what is an appropriate reference range. We are debating when should we start levothyroxine, should this be over 2.5, should this be over 4, should we take TPO antibodies into account. ATA guidance says do not treat isolated hypothyroxinemia. [V11] ETA guidance, and I defer to what you said, Alex, says consider isolated hypothyroxinemia treatment in the first trimester. There is no consensus here. And then we are all faced with women who attend these fertility clinics where they have been told, you actually need to keep your serum TSH around 1 mIU/L. There is no evidence for this, but those women come to us and they think we do not know what we are dealing with unless we do this. And again, do we have a policy? I do not think so. > **■ ON SCREEN — 34:37–35:48 · S56** > **Conclusions** > > Criteria > Condition is common - V > Screening tool available, reliable, inexpensive and acceptable - X > Intervention favourably influences disease and does not cause harm - X > Screening and intervention cost-effective - X > Agreed upon policy on whom to treat - X > PREGNANT WOMEN SHOULD NOT UNDERGO UNIVERSAL SCREENING FOR THYROID FUNCTION AND THYROID ANTIBODIES > > *Table 'Criteria' with right-hand check/cross marks: Condition is common ✓; Screening tool available, reliable, inexpensive and acceptable ✗; Intervention favourably influences disease and does not cause harm ✗; Screening and intervention cost-effective ✗; Agreed upon policy on whom to treat ✗. Large green text: 'PREGNANT WOMEN SHOULD NOT UNDERGO UNIVERSAL SCREENING FOR THYROID FUNCTION AND THYROID ANTIBODIES'.* **[34:38] Dr. Kristien Boelaert (arguing AGAINST universal screening):** So just to summarize all this for you, the only criterion I think I will agree with is present is that the condition is common. I don't think it is easy to interpret thyroid function tests during pregnancy and in pregnant women. It changes. I don't think we have an intervention that has been consistently shown to favorably influence disease. And I think increasing evidence mounting that this may cause harm. The screening and intervention, yes, there are some studies showing cost effectiveness, but I don't think all factors have been taken into account here. And I certainly do not believe that we have an agreed upon policy on whom to treat. So I would strongly urge that we vote against the motion and that not all pregnant women should undergo universal screening for thyroid function and thyroid antibodies. Thank you. **[35:40] Dr. David S. Cooper (Moderator):** Thank you very much, Dr. ## 35:40 — Handover *12:20 PM CT* **[35:41] Dr. David S. Cooper (Moderator):** Boelaert, for that passionate rebuttal. And now we're going to have Dr. Stagnaro-Green again come back up here and give his rebuttal to her. > **■ ON SCREEN — 35:48–35:55 · S11** > **Select your presentation** > > *Menu screen; the 'Kristien Boelaert 11:45 AM - 12:45 PM' card is highlighted/selected.* > > *Note: Operator screen: 'Select your presentation' list (the session is listed as 11:45 AM – 12:45 PM).* **[35:52] Dr. David S. Cooper (Moderator):** So I'll just get you going here, Alex. **[35:55] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** The discussion on whether or not to screen all women for thyroid disease in pregnancy has gotten lost and muddied ## 35:55 — Stagnaro-Green: rebuttal *12:20 PM CT* > **■ ON SCREEN — 36:02–38:22 · S57** > **Confusion on How to Monitor/Treat Women with SCH** > > Confusion on How to Monitor/Treat Women with SCH > AVOID UNCERTAINTY AT ALL COSTS > Targeted Screening Yields SCH > Ethical Concept of Autonomy > > *Red slide: 'AVOID UNCERTAINTY AT ALL COSTS' plus a photo of an ostrich with its head buried in the sand; bottom lines 'Targeted Screening Yields SCH' and 'Ethical Concept of Autonomy'.* **[36:12] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** In a discussion about subclinical hypothyroidism. But I am not making the argument that we should screen all women to diagnose and treat subclinical hypothyroidism. And I am not making the argument that we should screen all women to diagnose and treat women with autoimmune antithyroid disease. My argument is very simple. We have missed the forest of the trees. We have 1.2% of all women who have overt hyperthyroidism, overt hypothyroidism during pregnancy. These women are known to have increased miscarriage, preterm delivery, gestational hypertension. They need to be identified. They need to be treated. ACOG, the Society of Maternal Fetal Medicine, the American Thyroid Association, the Endocrine Society, and AACE all agree we need to identify and treat those patients. It is easy to identify with a blood test. We can then show it is cost effective. Intervention is helpful. So what we are really saying is we are trying to avoid uncertainty at all costs. We are putting our head in the sand. And we are saying we only want to diagnose that what we know how to handle. **[37:40] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** But that is not how medicine works. Medicine is science and medicine is art. And if we look at what is recommended right now by the ATA and the Endocrine Society, 65 to 70% of all women are considered high risk and will be screened and the vast majority of them will have subclinical hypothyroidism and we will need to have those discussions. And we will need to have those discussions because there's an ethical concept of autonomy. Our patients should know what risks they may or may not have and what we as professionals know and do not know. > **■ ON SCREEN — 38:22–39:15 · S58** > **Subclinical Hyperthyroidism Correlation Studies** > > Obstetrical Outcomes: No correlation - Casey et al. 2006, Rosario et al. 2016, Fraenkel et al. 2018 > Neonatal Outcomes: No correlation - Rosario et al. 2016, Fraenkel et al. 2018 > Decreased Offspring IQ/Cognitive Delay: Correlation found - Korevaar et al. 2015 | No correlation - Nelson et al. 2018, Pakkila et al. 2018 > Decreased risk of gestational hypertension: Correlation found - Casey et al. 2006 > > *Table: Outcome Studied | Correlation Found | No Correlation Found. Rows Obstetrical Outcomes (none | Casey 2006, Rosario 2016, Fraenkel 2018); Neonatal Outcomes (none | Rosario 2016, Fraenkel 2018); Decreased Offspring IQ/Cognitive Delay (Korevaar 2015 | Nelson 2018, Pakkila 2018); Decreased risk of gestational hypertension (Casey 2006).* **[38:24] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** In regards to subclinical hypothyroidism, here if we look at the correlation studies, the vast majority of correlation studies show no negative effect. The ATA guidelines do not recommend treating subclinical hypothyroidism. There are two studies that have found an impact in correlation, not in treatment. The Casey study from Parkland, Texas that followed 17,000 in women. In fact they showed that women who had subclinical hypothyroidism. In fact, they showed that women who had subclinical hyperthyroidism had a decrease in gestational hypertension. And the Korevaar study that came out in 2015 showed a decrease when you had that subclinical hyperthyroidism in terms of an IQ of 1 to 3 points. > **■ ON SCREEN — 39:15–39:56 · S59** > **Treatment of Graves' Disease - ATA Thyroid & Pregnancy Guidelines Recommendation 48** > > ATA Thyroid & Pregnancy Guidelines Recommendation 48 > "Antithyroid medication during pregnancy should be administered at the lowest effective dose of MMI or PTU, targeting maternal serum FT4/TT4 at or moderately above the reference range" > Strong Recommendation, High Quality Evidence > > *Quote slide: 'Antithyroid medication during pregnancy should be administered at the lowest effective dose of MMI or PTU, targeting maternal serum FT4/TT4 at or moderately above the reference range'; 'Strong Recommendation, High Quality Evidence'.* **[39:17] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** But that did not affect the ATA guidelines on the treatment of GRAVES, which is antithyroid medication during pregnancy, should be administered at the lowest effective dose, should be administered at the lowest effective dose, targeting maternal serum free T4 or total T4 at or moderately above the reference range, I.e. At subclinical hyperthyroidism. And if at some point we have confirmatory studies showing that subclinical hyperthyroidism has a deleterious impact on pregnancy, well then that's another reason why we need to screen and need to treat. And so the screening algorithm I put in front of you measures a TSH at preconception or at the first prenatal visit. > **■ ON SCREEN — 39:56–40:43 · S60** > **Screening Algorithm - Pre-conception or First Trimester** > > Screening Algorithm - Pre-conception or First Trimester > TSH: Low <0.1 mIU/L -> Reflex fT4 > TSH: Normal 0.1-4.0 -> No Further Workup > TSH: High >4.0 mIU/L -> Reflex fT4 + TPO > > *Flowchart: green TSH box; Low (<0.1 mIU/L) -> yellow 'Reflex fT4'; Normal (0.1-4.0) -> red 'No Further Workup'; High (>4.0 mIU/L) -> yellow 'Reflex fT4 + TPO'.* **[40:06] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** If it falls within the normal range of 0.1 to 4, no further workup. If it's elevated, as per the ATA, you get a reflex free T4 and TPO. And if it's decreased, you can get a reflex free T4 or a total T3. But the lack of our ability to know what to do exactly with SCH should not stop us for the 1% of women, that's 30,000 women per year in the United States, who will be found with overt hypo or overt hyper during pregnancy. Thank you. > **■ ON SCREEN — 40:43–40:58 · S11** > **Select your presentation** > > *Menu screen; the 'Alex Stagnaro-Green 11:45 AM - 12:45 PM' card is highlighted/selected and shows a 'Slides' button.* > > *Note: Operator screen: 'Select your presentation' list (the session is listed as 11:45 AM – 12:45 PM).* ## 40:48 — Handover *12:25 PM CT* **[40:48] Dr. David S. Cooper (Moderator):** Yeah, thank you, Alex. And Christian, you're next, please. Thank you. ## 40:54 — Boelaert: rebuttal *12:25 PM CT* > **■ ON SCREEN — 41:02–42:10 · S61** > **TABLET prevalence data** > > Total number of women who had thyroid function tested n = 19,213 > Euthyroid n = 18,293 (95.2%) > Isolated Hypothyroxinaemia n = 46 (0.3%) > Overt thyroid dysfunction n = 85 (0.4%) > SC thyroid dysfunction n = 694 (3.6%) > Other n = 95 (0.5%) > Euthyroid subgroup TSH<2.5 n = 14,919 (77.7%); TSH 2.5-4.5 n = 3,374 (17.6%) > Overt Hypothyroid n = 36 (0.2%); Overt Hyperthyroid n = 49 (0.3%) > SC Hypo TSH>4.5 n = 451 (2.4%); SC Hyper n = 243 (1.3%) > TPOAb positive: 9.5% > SC Hypo TSH>=2.5 n = 3,825 (19.9%); Moderate SCH TSH 4.5-10 n = 421 (2.2%); Severe SCH TSH>10 n = 30 (0.2%) > Dhillon-Smith et al. N Engl J Med 2019: epub > > *Flowchart of participants: total n=19,213 tested -> Euthyroid n=18,293 (95.2%), Isolated Hypothyroxinaemia n=46 (0.3%), Overt thyroid dysfunction n=85 (0.4%), SC thyroid dysfunction n=694 (3.6%), Other n=95 (0.5%); Euthyroid subgroup TSH<2.5 n=14,919 (77.7%) / TSH 2.5-4.5 n=3,374 (17.6%); Overt Hypothyroid n=36 (0.2%); Overt Hyperthyroid n=49 (0.3%); SC Hypo TSH>4.5 n=451 (2.4%); SC Hyper n=243 (1.3%); TPOAb positive 9.5%; SC Hypo TSH>=2.5 n=3,825 (19.9%), Moderate SCH 4.5-10 n=421 (2.2%), Severe SCH >10 n=30 (0.2%).* **[41:05] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Thank you. I'll show a few slides first and then I'll try and pick up on a few points that I disagree with, Alex. So, as I said, we have just screened over the last 10 years 20,000 women in the UK for thyroid function and TPO status. And so what we have is a very large cohort where we have prevalence data. Where we have prevalence data so as I said we screened nearly 20,000 women for thyroid function now these are women who had a history of at least one miscarriage or subfertility so they are already in a targeted treatment group and you can see whatever you know rather than focusing on this busy slide but what I would like to focus your attention to is that there was actually a very small minority that had overt thyroid dysfunction. This was 85 women out of 19,213 0.2% had overt hypothyroidism and 0.3% had overt hyperthyroidism so what we're talking about in a selected group of patients is already sort of quite a small minority. > **■ ON SCREEN — 42:10–42:49 · S62** > **CATS Results** > > Overall: N 390 vs 404; Mean IQ (SD) 99.2 (13.3) vs 100.0 (13.3); difference 0.8 (-1.1 to 2.6); p 0.40; IQ<85 12.1% vs 14.1% (2.1% (-2.6 to 6.7%)) > Positive on TSH1: N 209 vs 219; 100.4 (13.7) vs 100.7 (13.0); 0.3 (-2.2 to 2.8); p 0.82; IQ<85 11.0% vs 13.7% > Positive on FT4?: N 204 vs 207; 97.8 (12.7) vs 99.2 (13.3); 1.3 (-1.3 to 3.8); p 0.28; IQ<85 13.2% vs 14.5% > Positive on TSH and FT4: N 23 vs 22; 97.7 (11.6) vs 99.4 (11.4); 1.7 (-5.2 to 8.6); p 0.61; IQ<85 13.0% vs 13.6% > Lazarus et al. N Engl J Med. 2012 366: 493-5 > > *Table of mean offspring IQ and IQ<85 by subgroup (Overall N=390/404, mean IQ 99.2 (13.3) / 100.0 (13.3), difference 0.8 (-1.1 to 2.6), p=0.40; Positive on TSH1 N=209/219 100.4 / 100.7, 0.3 (-2.2 to 2.8), p=0.82; Positive on FT4? N=204/207 97.8 / 99.2, 1.3 (-1.3 to 3.8), p=0.28; Positive on TSH and FT4 N=23/22 97.7 / 99.4, 1.7 (-5.2 to 8.6), p=0.61), with IQ<85 percentages; last subgroup outlined in red.* **[42:10] Dr. Kristien Boelaert (arguing AGAINST universal screening):** What about treatment of overt hypothyroidism and you know as I've already said I'm not standing up here to say that you should not treat overt hypothyroidism but if you look at the supplementary data of the CATS trial and look at those women who actually had overt hypothyroidism so those who had a raised TSH and a low 34 you can see that when comparing the treatment and the placebo group that there was no difference. So I know it is small numbers of patients but even in these 50 women nearly 50 women that were treated there was no difference. So again I put it to you are we overegging things a little bit. > **■ ON SCREEN — 42:49–54:41 · S63** > **TABLET Results** > > Levothyroxine | Placebo > No. of women withdrawn from trial medication due to abnormal TFT: 46 (9.8%) | 45 (9.6%) > No. of SAEs: 28 (5.9%) | 18 (3.8%) > No. of women with pre-eclampsia: 9 (5%) | 5 (3%) > No. of women with gestational diabetes: 20 (11%) | 16 (9%) > Dhillon-Smith et al. N Engl J Med 2019: epub > > *Two-column table: Levothyroxine | Placebo; rows No. of women withdrawn from trial medication due to abnormal TFT 46 (9.8%) vs 45 (9.6%); No. of SAEs 28 (5.9%) vs 18 (3.8%); No. of women with pre-eclampsia 9 (5%) vs 5 (3%); No. of women with gestational diabetes 20 (11%) vs 16 (9%).* **[42:51] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Coming back to potential harm we again have data from the TABLET trial and these results are not significant but just coming back to what I was trying to state to you in my opening statement is that this may not be an entirely innocent treatment. Again similar to the study that was published in the BMJ where increased risk of gestational diabetes or preeclampsia was found we found that in these in those women who were treated with levothyroxine there was an increased risk of preeclampsia and an increased risk of gestational diabetes. So very similar to what was seen before. And at the top of this slide what I wanted to draw your attention to is the number of women that had to be withdrawn from the TABLET trial because their thyroid function became abnormal. So you might expect that in the placebo group, so these are women who are TPO antibody positive but who have thyroid function that's abnormal, if we're not treating them they're actually going to develop significantly abnormal thyroid function. **[43:51] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Well again there was no difference between those treated with levothyroxine and placebo. Dr. Stagnaro-Green says that we should screen all women in their preconception visit. I of course work in an area of the world where about 50% of pregnancies are unplanned so that would be quite difficult for me to do, Alex. And again, you know, I would just like to confirm that yes indeed we will pick up a number of patients with overt thyroid dysfunction. What we don't know is how many of these we would pick up because they actually present with clinical signs and symptoms and in screening we sometimes may forget that there is such a thing as a clinical evaluation of a patient but that we will be faced with a mass of thyroid function tests where we have no agreed policy on is it normal or is it abnormal and importantly how should we treat it. So again, you know, I stick with my motion that I don't think universal screening is appropriate or practical. Thank you. ## 44:53 — Audience questions *12:29 PM CT* **[44:53] Dr. David S. Cooper (Moderator):** Thank you both for those persuasive arguments. We have some questions from the audience that I have on my iPad here and I think you can just stay at your seats and answer these. So, and I think this one you both kind of alluded to but I'll ask it again and maybe you both can answer it again. The question is, does screening identify women early enough to allow treatment to make a difference and then should screening be done preconception? So I know you both touched on this. Why don't you just answer it again? Either one can go first, yeah. **[45:45] Dr. Kristien Boelaert (arguing AGAINST universal screening):** I agree. Screening should be done preconception. As I've alluded to, certainly in most areas in the United Kingdom, that is practically actually quite difficult. Getting patient thyroid function tests early on in pregnancy, again, is difficult. The intervention trials that have been done, you can see that treatment was actually started late and, you know, I would agree that both the CATS trial and the Casey study treatment was started too late but performing studies where we start this early is really difficult. You know, having been involved in the TABLET trial for 10 years, these studies are exceedingly difficult and it's probably a bit of wishful thinking that we will get decent data from intervention studies where this has actually started early enough. **[46:31] Dr. David S. Cooper (Moderator):** Alex? **[46:34] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Well, clearly preconception is the best as opposed to the first. Is that better? So preconception is obviously optimal versus the first prenatal visit and if it's the first prenatal visit, as early as possible and clearly not at the time that the CATS and the NICHD trials were done because most of us agree, although there were negative studies, they didn't disprove the point as to whether or not treating subclinical hypothyroidism or isolated hypothyroxinemia makes a difference. So the question is, can we get the patients early enough? In fact, the TABLET trial shows that you can. You can't do it in all women. So what we try and do is what's optimal if we find out and we can do a trial and we should do a trial, a TABLET equivalent for subclinical hypothyroidism to answer once and for all, does treating women with subclinical hypothyroidism pre-pregnancy if we find that's the case, then there's a public health message that has to go out there that we can get an ever-increasing number of women either preconception or early prenatal. **[47:56] Dr. David S. Cooper (Moderator):** Thank you. Okay, the questions are flooding in. I'm just, I have to, there's so many I have to choose here. So one of the issues about the CATS trial that you raised, I think, is that there may be over-treatment with its negative effects on neurocognitive development in the fetus. And the questioner is saying, well, yes, the T4 was high in those patients, but were there TSH levels in those patients that were low? Do you know the answer to that? **[48:32] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Yes, I think I do. So in a number of those patients, they are actually positively over-treated. So overt hyperthyroidism was induced [V10] and TSH was low in addition to raised free T4. It wasn't just a free T4, but it seems to be a free T4 being a sort of biological marker here of potential harm. **[48:51] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** But I want to add to that, David, I think we all recognize that the model of the CATS trial of giving 150 micrograms to all women without adjusting that, it is not a model that any of us would want to go forward with in treating subclinical hypothyroidism or hypothyroxinemia. [V11] So the data showing an overtreatment there is not surprising, but does not reflect what would happen in the real world. **[49:21] Dr. David S. Cooper (Moderator):** That's a good point. Thank you. Okay, this has to do with the TABLET results. The questioners are saying, okay, well, these patients all had positive antibodies, but normal thyroid function. Can you extrapolate your results to women who actually had subclinical or overt hypothyroidism? **[49:40] Dr. Kristien Boelaert (arguing AGAINST universal screening):** So in order to be eligible for the trial, thyroid function had to be within the normal reference range. So the only, I did show it on my slide, but so the only subanalysis there that we could do was those who had a TSH below 2.5 and above 2.5, but still within the normal reference range. It is difficult to extrapolate further because you weren't eligible if thyroid function tests were outside that reference range. You weren't included in the trial. **[50:14] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** I think there's two other points on the TABLET trial, which I honestly have not read very thoroughly, but in a cursory reading, we need to note that these were not random women who were selected, which is what's been found in the vast majority of the thyroid antibody studies. It's all coming women, not women who have miscarriage or infertility, which is what the precondition was, the inclusion criteria for the TABLET trial. And also there was a 30% pregnancy loss in the TABLET trial, which is quite high and I think needs to be looked at very carefully. **[50:52] Dr. Kristien Boelaert (arguing AGAINST universal screening):** If I can come back to that, you know, that's in that specific subpopulation of women, a 30% miscarriage rate is actually what we probably expect. **[51:02] Dr. David S. Cooper (Moderator):** Is actually what? **[51:03] Dr. Kristien Boelaert (arguing AGAINST universal screening):** What we expect. **[51:04] Dr. David S. Cooper (Moderator):** Okay. This is for you, Alex, but you both can answer this and several people have asked the same question. In women who have a low serum TSH, do you treat any of them? Or do you only treat women who have overt hyperthyroidism and particularly only those who have anti-TSH receptor antibodies, for example? Sure. **[51:29] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Sure. So an algorithm needs to be worked out. Do you just treat women who have suppressed TSH? And the answer is no, because they can be individuals who just have the HCG suppression, which is normal and physiologic. They can have twins, in which case that HCG suppression will be a longer time period. So the answer to number one is no. The answer to number two is if you get a free T4, and that free T4 is normal or right at the upper range of normal, you wouldn't treat those individuals because those individuals fall within the range that the ATA guidelines recommend that you want to achieve in women with Graves' disease. So you might have to do some further workup, but clearly the majority of the women who, or at least a portion of the women who have the suppressed TSH, will not require treatment. **[52:29] Dr. David S. Cooper (Moderator):** This is kind of a general question several people ask. Can the health care system cope with the burden of universal screening? **[52:40] Dr. Kristien Boelaert (arguing AGAINST universal screening):** No. **[52:42] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** Yes. In fact, if a health care system can't cope with the concept of universal screening, which, by the way, we do for gestational diabetes, we do for 17 other conditions during pregnancy, we screen all children for neonatal hypothyroidism. If our system can't cope with that, then we have some major issues with our system. **[53:10] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Can I just add to that, because I do not agree, is that for those other conditions, there are then clear, agreed policies on how to deal with it. We do not have that in this field. **[53:24] Dr. David S. Cooper (Moderator):** I think we have maybe just one more question. There are so many here, it is just hard to choose. How about this one? In women that you decide to treat for one reason or another, how do you initiate treatment? Do you start with the lowest dose of levothyroxine, 25 micrograms, or start with 50, or how would you go about treating such a woman? **[53:52] Dr. Kristien Boelaert (arguing AGAINST universal screening):** I often maybe practice too pragmatic medicine, but for me, it's weight-based. So if the woman is 75 kilograms or more, I start on 75 micrograms. If she's less than 75, I start on 50. **[54:11] Dr. David S. Cooper (Moderator):** Alex? **[54:11] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** So here we're pretty much in agreement. I might go 50 micrograms to begin with, but it also depends upon what the initial TSH is. **[54:19] Dr. David S. Cooper (Moderator):** Okay, thank you. Okay, I think we're almost at the end. ## 54:23 — Final statements *12:39 PM CT* **[54:23] Dr. David S. Cooper (Moderator):** We're going to have our two opponents come up and give us their summary for two minutes each. **[54:29] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** I have no slides for this component, and I want to conclude by, again, thanking the Endocrine Society, > **■ ON SCREEN — 54:41–56:24 · S11** > **Select your presentation** > > *Menu screen; the 'Alex Stagnaro-Green 11:45 AM - 12:45 PM' card is highlighted/selected (green arrow points at it).* > > *Note: Operator screen: 'Select your presentation' list (the session is listed as 11:45 AM – 12:45 PM).* **[54:53] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** David, and especially Kristien, for having this discussion. I think this is all really very straightforward, and for the reasons that I've articulated to date, so I'm not going to re-articulate them. However, I'm going to make one comment. In 2015, I published an article in JAMA, a viewpoint article, articulating this position regarding why universal screening should be done based on the impact of overt thyroid disease, hyper and hypo. JAMA, the editors had me rewrite it five times because they were concerned about the impact of the article. I think there must have been five obstetricians who gave me comments during that review process. I expected many, many rebuttals published in the literature to that concept. There was not one letter to the editor, nor have I seen to date any cogent argument why we should not screen all women for thyroid disease and pregnancy to identify and treat thyroid disease, overt hyper and overt hypo. **[56:11] Dr. David S. Cooper (Moderator):** Thank you. **[56:28] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Right. > **■ ON SCREEN — 56:28–57:10 · S64** > **Universal screening** > > It's cost-effective > Intervention does not cause harm > Screening tool is easy to use > There is an agreed treatment policy > > *Four portraits of UK political figures, each with an off-white speech bubble: 'It's cost-effective', 'Intervention does not cause harm', 'Screening tool is easy to use', 'There is an agreed treatment policy'; later state adds a photo of Theresa May with a thought bubble 'We must screen all pregnant women!'.* **[56:31] Dr. Kristien Boelaert (arguing AGAINST universal screening):** Sorry, I had to refer to it. I come from a part of the country where we're in deep trouble. I believe universal screening is based on lies, lies and more lies. We see a number of people there who have told a lot of people a lot of lies. It's not cost effective. There is no clear evidence that the intervention doesn't cause harm. The screening tool is not easy to use and there is no agreed treatment policy. But still, our leaders tell us that we should screen all pregnant women. What I would like, and I really struggled finding a trustworthy politician, > **■ ON SCREEN — 57:10–57:54 · S65** > **Targeted case-finding** > > Let's treat women who need Rx appropriately > Let's re-assess cost-effectiveness > Let's agree a policy > Let's conduct RCTs early in pregnancy > > *Photo of Michelle Obama at centre; four off-white thought bubbles around it: 'Let's treat women who need Rx appropriately', 'Let's re-assess cost-effectiveness', 'Let's agree a policy', 'Let's conduct RCTs early in pregnancy'.* **[57:11] Dr. Kristien Boelaert (arguing AGAINST universal screening):** So instead I've gone for one of my role models here to try and convince you that what we should do is targeted case finding. I think we need more trials. We need very clear evidence regarding cost effectiveness. I think we should try and agree on policies, and I think we should treat those women who require treatment appropriately. I put it to you, we should not screen all women during pregnancy. Thank you. **[57:40] Dr. David S. Cooper (Moderator):** Well, I want to thank both our debaters ## 57:40 — Re-poll, results, and close *12:42 PM CT* **[57:43] Dr. David S. Cooper (Moderator):** For their reasoned, logical, impassioned and intellectual arguments and now what we're going to do, if I can get the slides going, is to vote again. > **■ ON SCREEN — 57:54–58:02 · S11** > **Select your presentation** > > *Menu screen; the 'Kristien Boelaert 11:45 AM - 12:45 PM' card is highlighted/selected.* > > *Note: Operator screen: 'Select your presentation' list (the session is listed as 11:45 AM – 12:45 PM).* **[58:02] Dr. David S. Cooper (Moderator):** And we'll see whether there's been any change in anyone's mind about screening. > **■ ON SCREEN — 58:05–58:12 · S06** > **SESSION PARTICIPATION AND EVALUATIONS USING THE ENDO 2019 APP** > > STEP 1 Find the session > STEP 2 Tap the Participate Evaluate button > Complete session evaluations by clicking Evaluate This Session > > *Same ENDO 2019 app instructions slide, but the right-hand phone screenshot now shows the purple 'ENDO 2019' banner* > > *Note: near-duplicate of state 08 (same slide, slightly different app screenshot)* > **■ ON SCREEN — 58:12–58:26 · S07** > **Question** > > 1. Are you in favor of screening all pregnant women for thyroid dysfunction? > – Yes > – No > > *Plain purple question slide; red banner at the bottom reads 'LIVE POLLING IS THE NEXT SLIDE'; green arrow pointer.* **[58:13] Dr. David S. Cooper (Moderator):** So again, here is your question, going to come up in a minute. Are you in favor of screening all pregnant women for thyroid dysfunction per Dr. Stagnaro-Green's Entreaty? > **■ ON SCREEN — 58:28–58:39 · S08** > **1. Are you in favor of screening all pregnant women for thyroid dysfunction?** > > Yes > No > 6 (yellow countdown box, bottom-right) > > *White live-poll screen of the ENDO 2019 app: question at top, 'Yes'/'No' options, large yellow countdown box bottom-right showing '6'.* > **■ ON SCREEN — 58:40–58:45 · S09** > **1. Are you in favor of screening all pregnant women for thyroid dysfunction?** > > Yes 54% > No 46% > Poll is Locked > > *White live-poll results screen of the ENDO 2019 app: question at top, Yes/No rows with percentages and horizontal bars (yellow for Yes, green for No), grey label 'Poll is Locked' bottom-right.* **[58:43] Dr. David S. Cooper (Moderator):** Yes or no? > **■ ON SCREEN — 58:47–59:04 · S10** > **1. Are you in favor of screening all pregnant women for thyroid dysfunction?** > > Yes 71% (Pre-Test) / 55% (Post-Test) > No 29% (Pre-Test) / 45% (Post-Test) > Pre-Test | Post-Test toggle buttons > > *Live-poll comparison screen of the ENDO 2019 app on black background: Pre-Test / Post-Test toggle buttons, Yes/No rows with grey (pre-test) and blue (post-test) horizontal bars and percentages.* **[58:48] Dr. David S. Cooper (Moderator):** So — yeah, so here are the comparisons between the pre- and post-test. **[58:54] Dr. David S. Cooper (Moderator):** † So I think you've lost some ground here, Alex. **[59:00] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** † Have I? [V8] > **■ ON SCREEN — 59:04–59:12 · S07** > **Question** > > 2. Are you in favor of screening all pregnant women for thyroid autoimmunity? > – Yes > – No > > *Plain purple question slide; red banner at the bottom reads 'LIVE POLLING IS THE NEXT SLIDE'; green arrow pointer.* **[59:05] Dr. David S. Cooper (Moderator):** † Okay. **[59:05] Dr. David S. Cooper (Moderator):** † And here's our next question. **[59:06] Dr. David S. Cooper (Moderator):** † Are you in favor of screening all pregnant women for thyroid autoimmunity? **[59:08] Dr. David S. Cooper (Moderator):** † Yes or no? > **■ ON SCREEN — 59:14–59:24 · S08** > **2. Are you in favor of screening all pregnant women for thyroid autoimmunity?** > > Yes > No > 6 (yellow countdown box, bottom-right) > > *White live-poll screen of the ENDO 2019 app: question at top, 'Yes'/'No' options, large yellow countdown box bottom-right showing '6'.* **[59:14] Dr. David S. Cooper (Moderator):** † Okay. **[59:14] Dr. David S. Cooper (Moderator):** † And I don't think Alex was really arguing for this at all. **[59:17] Dr. Alex Stagnaro-Green (arguing FOR universal screening):** † No. **[59:19] Dr. David S. Cooper (Moderator):** † No, you were not, right. > **■ ON SCREEN — 59:24–59:32 · S09** > **2. Are you in favor of screening all pregnant women for thyroid autoimmunity?** > > Yes 11% > No 89% > Poll is Locked > > *White live-poll results screen of the ENDO 2019 app: question at top, Yes/No rows with percentages and horizontal bars (yellow for Yes, green for No), grey label 'Poll is Locked' bottom-right.* > **■ ON SCREEN — 59:34–01:00:01 · S10** > **2. Are you in favor of screening all pregnant women for thyroid autoimmunity?** > > Yes 35% (Pre-Test) / 11% (Post-Test) > No 65% (Pre-Test) / 89% (Post-Test) > Pre-Test | Post-Test toggle buttons > > *Live-poll comparison screen of the ENDO 2019 app on black background: Pre-Test / Post-Test toggle buttons, Yes/No rows with grey (pre-test) and blue (post-test) horizontal bars and percentages.* **[59:34] Dr. David S. Cooper (Moderator):** Right and so um there's a little bit of a change here and we expected them I think so I'm not going to declare a winner or a loser here I think you know who you are but I want to thank both of you and also the endocrine society for allowing us to have this forum for this really lively discussion and hopefully we'll have some more answers in the years to come thank you to everybody --- ## Appendix A — Session timeline | Video time | Wall clock | Segment | |---|---|---| | 00:00 | 11:45 AM | Opening and screening primer | | 02:47 | 11:47 AM | Format, rules, and the pre-debate poll | | 04:55 | 11:49 AM | Handover and slide setup | | 05:08 | 11:50 AM | Stagnaro-Green: presentation for universal screening | | 21:58 | 12:06 PM | Handover | | 22:14 | 12:07 PM | Boelaert: presentation against universal screening | | 35:40 | 12:20 PM | Handover | | 35:55 | 12:20 PM | Stagnaro-Green: rebuttal | | 40:48 | 12:25 PM | Handover | | 40:54 | 12:25 PM | Boelaert: rebuttal | | 44:53 | 12:29 PM | Audience questions | | 54:23 | 12:39 PM | Final statements | | 57:40 | 12:42 PM | Re-poll, results, and close | ## Appendix B — Slide index (every distinct screen) | # | Video time | Slide | Screen states | |---|---|---|---| | 1 | 00:00–00:38 | Endocrine Debate: Thyroid Disease in Pregnancy: To Screen, or Not to Screen? | 1 | | 2 | 00:38–01:18 | What is screening? (WHO definition) | 2 | | 3 | 01:18–02:22 | Screening Criteria | 2 | | 4 | 02:22–02:48 | The Debaters | 1 | | 5 | 02:48–03:38 | Endocrine Debate: Thyroid Disease in Pregnancy: To Screen, or Not to Screen? | 1 | | 6 | 03:38–03:52 | SESSION PARTICIPATION AND EVALUATIONS USING THE ENDO 2019 APP | 1 | | 7 | 03:52–04:02 | Question | 1 | | 8 | 04:03–04:14 | 1. Are you in favor of screening all pregnant women for thyroid dysfunction? | 1 | | 9 | 04:15–04:20 | 1. Are you in favor of screening all pregnant women for thyroid dysfunction? | 1 | | 10 | 04:20–04:30 | Question | 1 | | 11 | 04:32–04:42 | 2. Are you in favor of screening all pregnant women for thyroid autoimmunity? | 1 | | 12 | 04:42–04:47 | 2. Are you in favor of screening all pregnant women for thyroid autoimmunity? | 1 | | 13 | 04:50–05:01 | Select your presentation | 1 | | 14 | 05:06–05:33 | UNIVERSAL SCREENING FOR THYROID DISEASE IN PREGNANCY SHOULD BE PERFORMED | 1 | | 15 | 05:40–06:12 | To Screen or Not to Screen: That is the Question | 1 | | 16 | 06:12–06:39 | Terry Davies quote: “You can't sit on the fence forever” (2007) | 1 | | 17 | 06:39–06:56 | Criteria for Screening | 1 | | 18 | 06:56–07:28 | Prevalence in the Published Literature | 1 | | 19 | 07:28–08:08 | Differences in Diagnostic Criteria Mask the True Prevalence of Thyroid Disease in Pregnancy | 1 | | 20 | 08:08–08:45 | Prevalence of Overt Hypothyroidism Based on TSH Cutoff Values | 1 | | 21 | 08:45–09:18 | Prevalence of Subclinical Hypothyroidism Based on TSH Cutoff Values | 1 | | 22 | 09:18–09:38 | Prevalence of Overt Hyperthyroidism Based on Timing of Screening | 1 | | 23 | 09:38–10:30 | Thyroid Disease Prevalence in Pregnancy | 1 | | 24 | 10:30–11:02 | Wilson–Jungner scorecard (progressive build) | 1 | | 25 | 11:02–11:13 | Criteria for Screening | 1 | | 26 | 11:13–11:42 | Test Acceptable to Population | 1 | | 27 | 11:42–12:04 | Wilson–Jungner scorecard (progressive build) | 1 | | 28 | 12:04–12:13 | Criteria for Screening | 1 | | 29 | 12:15–12:52 | Overt Hypothyroidism Correlation Studies | 1 | | 30 | 12:52–13:06 | Overt Hypothyroidism Treatment Studies | 1 | | 31 | 13:06–13:28 | Overt Hyperthyroidism Correlation Studies | 1 | | 32 | 13:28–13:48 | Overt Hyperthyroidism Treatment Studies | 1 | | 33 | 13:48–14:15 | Thyroid Autoimmunity Correlation Studies | 1 | | 34 | 14:15–14:30 | Thyroid Autoimmunity Treatment Studies | 1 | | 35 | 14:30–14:58 | Wilson–Jungner scorecard (progressive build) | 1 | | 36 | 14:58–15:02 | Criteria for Screening | 1 | | 37 | 15:02–16:04 | Cost Effectiveness Analysis | 1 | | 38 | 16:04–17:26 | Incremental Cost Effectiveness Ratio | 1 | | 39 | 17:26–18:28 | Cost-Effectiveness of Universal Screening | 1 | | 40 | 18:28–18:58 | Cost-Effectiveness - Overt Hyperthyroidism Logic Argument | 1 | | 41 | 18:58–19:26 | Wilson–Jungner scorecard (progressive build) | 1 | | 42 | 19:28–20:04 | Agreement To Treat Meter | 1 | | 43 | 20:04–20:17 | Wilson–Jungner scorecard (progressive build) | 1 | | 44 | 20:17–20:46 | Recommended Universal Screening | 1 | | 45 | 20:46–21:06 | ATA 2017 Recommendation - Targeted Screening | 1 | | 46 | 21:06–21:14 | Women with overt and/or subclinical disease in pregnancy missed with targeted screening | 1 | | 47 | 21:14–21:29 | Where We Are Today: Prevalence of Universal Screening Practices | 1 | | 48 | 21:29–21:50 | SUMMARY | 1 | | 49 | 21:50–21:56 | It's time to get off the fence (cartoon) | 1 | | 50 | 22:00–22:18 | Select your presentation | 1 | | 51 | 22:23–22:42 | Thyroid Disease in Pregnancy: To Screen or Not to Screen - AGAINST the MOTION | 1 | | 52 | 22:42–22:46 | Disclosure | 1 | | 53 | 22:46–24:02 | (1) Thyroid dysfunction in pregnancy | 2 | | 54 | 24:02–25:16 | (1) Thyroid autoimmunity in pregnancy | 2 | | 55 | 25:16–26:24 | (2) Screening tool: TFT and TPO testing | 1 | | 56 | 26:24–27:11 | (3) Intervention favorably influences disease? - CATS TRIAL | 1 | | 57 | 27:11–27:42 | (3) Intervention favorably influences disease - NICHD TRIAL | 1 | | 58 | 27:42–28:19 | (3) Intervention favorably influences disease - TABLET results | 1 | | 59 | 28:19–28:42 | (3) Intervention favorably influences disease - TABLET subgroup forest plot | 1 | | 60 | 28:42–29:22 | (3) Intervention favorably influences disease - TABLET trial (survival curve) | 1 | | 61 | 29:22–30:30 | (3) Intervention may cause harm - optimal vs overtreatment | 1 | | 62 | 30:30–31:46 | (3) Intervention may cause harm - FT4 >97.5th percentile and verbal IQ | 1 | | 63 | 31:46–32:25 | (3) Intervention may cause harm - Maraka adverse outcomes table | 1 | | 64 | 32:25–33:34 | (4) Cost-effectiveness | 2 | | 65 | 33:34–34:37 | (5) No agreed policy on whom to treat - ATA 2017 Guidance | 2 | | 66 | 34:37–35:48 | Conclusions | 1 | | 67 | 35:48–35:55 | Select your presentation | 1 | | 68 | 36:02–38:22 | Confusion on How to Monitor/Treat Women with SCH | 1 | | 69 | 38:22–39:15 | Subclinical Hyperthyroidism Correlation Studies | 1 | | 70 | 39:15–39:56 | Treatment of Graves' Disease - ATA Thyroid & Pregnancy Guidelines Recommendation 48 | 1 | | 71 | 39:56–40:43 | Screening Algorithm - Pre-conception or First Trimester | 1 | | 72 | 40:43–40:58 | Select your presentation | 1 | | 73 | 41:02–42:10 | TABLET prevalence data | 1 | | 74 | 42:10–42:49 | CATS Results | 1 | | 75 | 42:49–54:41 | TABLET Results | 1 | | 76 | 54:41–56:24 | Select your presentation | 1 | | 77 | 56:28–57:10 | Universal screening | 2 | | 78 | 57:10–57:54 | Targeted case-finding | 1 | | 79 | 57:54–58:02 | Select your presentation | 1 | | 80 | 58:05–58:12 | SESSION PARTICIPATION AND EVALUATIONS USING THE ENDO 2019 APP | 1 | | 81 | 58:12–58:26 | Question | 1 | | 82 | 58:28–58:39 | 1. Are you in favor of screening all pregnant women for thyroid dysfunction? | 1 | | 83 | 58:40–58:45 | 1. Are you in favor of screening all pregnant women for thyroid dysfunction? | 1 | | 84 | 58:47–59:04 | 1. Are you in favor of screening all pregnant women for thyroid dysfunction? | 1 | | 85 | 59:04–59:12 | Question | 1 | | 86 | 59:14–59:24 | 2. Are you in favor of screening all pregnant women for thyroid autoimmunity? | 1 | | 87 | 59:24–59:32 | 2. Are you in favor of screening all pregnant women for thyroid autoimmunity? | 1 | | 88 | 59:34–01:00:01 | 2. Are you in favor of screening all pregnant women for thyroid autoimmunity? | 1 | A complete slide-by-slide transcript, including each build state and every chart description, is in the companion file `ENDO2019_Slide_Appendix.md`. ## Appendix C — Audience poll results | Poll | Pre-debate | Post-debate | Change | |---|---|---|---| | “Are you in favor of screening all pregnant women for **thyroid dysfunction**?” | 71% yes / 29% no | 54% yes / 46% no on the live board; the comparison slide shows 55% / 45% | −17 points yes (71 → 54–55%) | | “Are you in favor of screening all pregnant women for **thyroid autoimmunity**?” | 35% yes / 65% no | 11% yes / 89% no | −24 points yes (35 → 11%) | The one-point difference in the first poll is a projection-slide inconsistency in the original deck (the live board reads 54%, the comparison slide 55%); both are reproduced on screen below. ## Appendix D — Verification notes and quotations requiring a check ### D.1 Passages marked [V] or † in the transcript These are the only places where the wording is not secured by the audio plus the slides. Quote them only after checking the recording. | # | Time | Passage | Issue | |---|---|---|---| | V1 | 56:28 | Boelaert: “I believe universal screening is based on lies, lies and more lies” | Deliberately inflammatory; exact wording (and whether “country” or “world”) must be checked before any quotation | | V2 | 56:49 | Boelaert: “There is no clear evidence that the intervention doesn't cause harm” | Double negative; the meaning flips if the negative was mis-heard | | V3 | 55:08–55:50 | Stagnaro-Green: JAMA 2015 viewpoint rewritten “five times”; “five obstetricians”; “not one letter to the editor” | Numbers and the “not one letter” claim are quotation-sensitive | | V4 | 36:41–37:04 | Stagnaro-Green: “We have 1.2% of all women who have overt hyperthyroidism, overt hypothyroidism during pregnancy” | As transcribed this asserts both conditions in the same women; verify the spoken wording | | V5 | 19:52 | Stagnaro-Green: “Is it a TSH of 2.5, is it a TSH of 4.0?” | Heavy recognition damage at this point; reconstructed from the agreement-to-treat slide and the ATA thresholds | | V6 | 46:05 | Boelaert: “both the CATS trial and the Casey study treatment was started too late” | Concession; verify wording | | V7 | 32:26 | Boelaert: “7,258 US dollars per QALY saved” | Slide itself says 7,258; both machine transcripts hear “7,558” — verify whether she misread her slide | | V8 | 59:00 | Stagnaro-Green: “Have I?” | Short exchange with Cooper; wording uncertain | | V9 | 39:03 | Stagnaro-Green: “showed a decrease… in terms of an IQ of one to three points” | Direction of the IQ effect (“increase” vs “decrease”) must be confirmed | | V10 | 48:36 | Boelaert: “so overt hyperthyroidism was induced and TSH was low…” | Biologically it must be hyperthyroidism; verify whether she misspoke on the day | | V11 | 49:04 | Stagnaro-Green: “…or hypothyroxinemia” | CATS enrolled subclinical hypothyroidism or hypothyroxinemia; the machine heard “overt hypothyroidism” | | V12 | 50:40 | Stagnaro-Green: “It's all women — women who have miscarriage or infertility” | Garbled in both transcripts; verify | ### D.2 Material removed because it is not in the recording * Recognition garbage at 10:01, 19:51–20:22, 29:49–30:13 and 59:11–60:12 (recogniser hallucination over silence/applause/slide changes) was deleted; the real speech was recovered by re-transcribing those windows continuously. * About 50 spurious one-word “No.” segments during the re-poll (audience noise) were deleted. * Prompts leaked into the recogniser output at the 49:40 chunk boundary (“Dr. David Cooper-…”); the false prefixes were stripped. ### D.3 Corrections applied (summary) * **Names:** Christian Bollard / Bullard / Borler → Kristien Boelaert; Stegner / Stegnero / Stagnero / Sagnawa → Stagnaro-Green; Christy Dissau → Kristi Dosiou; Dylan Smith → Dhillon-Smith; Tim Korova → Tim Korevaar; “core of our study” → “Korevaar study”; Gacy's → Casey; Wilson and younger → Wilson and Jungner; ACE → AACE (where the society is meant). * **Trials and tests:** tablet/TABLA/Tabla → TABLET; TPU/TPR/DPO → TPO; ICR → ICER; NIH trials → NICHD trials. * **Clinical terms:** 3T4 / 3D4 / 3-T4 → free T4; hyperthyroxenemia → hypothyroxinemia; thyroid toxicosis → thyrotoxicosis; growth hormone → gestational hypertension; main cortex volume → mean cortex volume; over thyroid → overt thyroid; paternal → maternal; function deaths → function tests; home to treat → whom to treat. * **Numbers/units:** 97.5th-percentile TSH cutoff; ATA-2011 cutoff 2.5 mIU/L; 1 million units per litre → 1 mIU/L; 7,558 → 7,258 USD/QALY (see V7); quality saved → QALY saved. ### D.4 Provenance of the session date The presentation screen in the recording reads “Saturday, March 23 11:30 AM – 1:00 PM”, but that is the stale advertised slot in the ENDO 2019 app. The Endocrine Society session archive for E22, the HCPLive report on the session, and the internal evidence in the recording (MLB Opening Day “starts in two days” = 26 March; the TABLET trial “presented here on Saturday” = 23 March; the NEJM paper “three days ago”) all date the session to **Tuesday 26 March 2019, 11:45 AM–12:45 PM CT**. Cite 26 March. ### D.5 What the earlier transcript got wrong or missed (for the record) * Its “Speaker 1 / 2 / 3” labels were mostly right but were never named, and they broke down in the Q&A and the final statements. * It contained no timestamps and no on-screen text. * Named errors: “Christian Ballard” (Boelaert), “Kristi Dzau” (Dosiou), “Dylan Smith” (Dhillon-Smith), “Wilson and younger” (Jungner), “KC trial” (NICHD/Casey), “tablet” (TABLET), “TPO” read as “TPU”. * It silently dropped the audience-poll percentages, the published cost-effectiveness numbers and the entire slide sequence — all of which are restored above. *Transcript produced from the ENDO 2019 recording (60:00.93, 640×360). Speech recognition: Whisper large-v3-turbo; speaker attribution: ECAPA-TDNN voice embeddings; screen capture: scene-change detection at 2 fps plus visual transcription of all 95 detected states.*