Journal Commentary – Dysautonomia and POSS

Twenty Years Later, POTS Is Still There, and the Patients Who Don’t Improve Are Telling Us Something

A new Journal of Internal Medicine study followed 44 adults with postural orthostatic tachycardia syndrome for a median of 23 years. The headline was that half improved. The finding that matters is which half didn’t, and why.

Andrea Vidali, MD – Endometriosis Surgical Specialists International – 3 August 2026

The Short Answer

POTS does not resolve. In a 20-year follow-up of 44 patients, only 2 percent had complete symptom resolution and 98 percent still had symptoms. The patients who worsened were separated from those who improved not by how sick they were at baseline, but by the presence of neuropathy, gastroparesis and overactive bladder, a distinct neuro-visceral phenotype. Outcome tracked mechanism, not severity.

Key Figures at a Glance

1. Median follow-up: 23 years from symptom onset, 17 years from diagnosis, an embedded diagnostic delay of roughly six years.
2. Complete symptom resolution: 2 percent. Ongoing symptoms: 98 percent.
3. Improved 46 percent, worsened 25 percent, unchanged 11 percent, variable course 16 percent. A slim majority did not improve.
4. Neuropathy, gastroparesis or overactive bladder present in all but one NOT IMPROVED patient, versus 20 percent of IMPROVED.
5. Orthostatic intolerance scores improved. Pupillomotor, gastrointestinal and bladder scores did not.
6. Baseline severity did not predict trajectory, and NOT IMPROVED patients recalled better initial health.

What the Study Actually Did

Bourne and colleagues contacted 120 people evaluated at the Vanderbilt autonomic centre between 1987 and 2014. Seventy-four responded, 46 completed a custom survey between March and September 2021, and 44 were analysed. Ninety-eight percent were female. Median age at survey was 48.

Participants were sorted into IMPROVED and NOT IMPROVED by reported symptom course, and completed COMPASS-31, the standard composite autonomic symptom instrument, both for their current state and, retrospectively, for how they remembered feeling at their first Vanderbilt visit.

This is a small, single-centre, survey-based, recall-dependent study. I will come back to that. But it is also nearly the only dataset of its kind, because almost nobody follows this population for two decades, and the questions it answers cannot be answered any other way.

Table 1: Study at a glance

Design	Single-timepoint custom survey, cross-sectional, retrospective recall of baseline
Setting	Vanderbilt University Medical Center Autonomic Dysfunction Center Research Unit
Recruitment	120 contacted, 74 responded, 46 completed, 44 analysed (62 percent response rate)
Original evaluation	1987 to 2014; survey fielded March to September 2021; published 2026
Population	98 percent female (43 of 44); median age 48; 60 percent premenopausal, 41 percent postmenopausal
Follow-up	Median 23 years from symptom onset; 17 years from diagnosis
Instrument	COMPASS-31, total and six subdomains
Principal limitation	Baseline COMPASS-31 reconstructed from memory across up to two decades

Four Findings That Matter

1. POTS is chronic. It does not burn out.

The persistent clinical folklore, that POTS is a disorder of adolescent girls that resolves in the twenties, does not survive this dataset. Neither does the menopause hypothesis: 41 percent of participants were postmenopausal at survey and the group as a whole was still symptomatic.

Artur Fedorowski, in the accompanying editorial, makes the point sharply: POTS is conceptualised as chronic from the moment of diagnosis, since the criteria require months of symptoms before the label can be applied at all. A condition that requires chronicity for entry cannot be casually described as self-limiting. He also draws the obvious extension: if long COVID-associated POTS shares mechanism with classical POTS, a very large post-2020 cohort is looking at decades, not years.

Figure 1: Self-reported symptom course since diagnosis (n = 44), 20-year follow-up

Outcome Category	Share of Patients
Resolved (fully recovered)	2 percent
Improved	46 percent
Variable course	16 percent
Unchanged	11 percent
Worsened	25 percent

Grouped differently: 48 percent resolved or improved, versus 52 percent not improved (variable, unchanged, or worsened). 98 percent of participants (43 of 44) reported ongoing POTS symptoms at the time of survey. Source: Bourne et al., J Intern Med 2026.

Figure 1 caption from the original: press coverage led with “almost half improved.” Read the other direction, a slim majority did not improve and essentially nobody recovered.

2. The outcome split is a domain split, not a severity split

This is the finding I want every pelvic pain clinician to read twice.

Figure 2: What separated the two outcome groups (presence of at least one of neuropathy, gastroparesis, or overactive bladder)

Outcome Group	Share With Neuropathy, Gastroparesis, or Overactive Bladder
IMPROVED	20 percent
NOT IMPROVED	All but one patient (about 96 percent)

Baseline COMPASS-31 severity as a predictor: no significant difference between groups at first evaluation. NOT IMPROVED patients recalled their overall health as better at baseline.

Figure 2 caption from the original: not a severity gradient, a phenotype. Neuropathy, gastroparesis and overactive bladder form a coherent neuro-visceral cluster, plausibly a small-fibre and autonomic neuropathy substrate, sitting inside a diagnosis that on paper is about heart rate on standing.

The patients who did well over twenty years were the ones whose problem really was orthostatic. The patients who did badly had a different disease wearing the same name.

3. Treatment worked on one domain and not the others

Total COMPASS-31 and the orthostatic intolerance subdomain improved across the follow-up interval. Pupillomotor scores went up (worse). Gastrointestinal and bladder scores also went up, though those differences did not survive Bonferroni correction.

Figure 3: Twenty years of treatment, by autonomic domain (direction of change in COMPASS-31 subdomains over follow-up)

Domain	Direction of Change	Statistical Note
Total score	Improved	Significant
Orthostatic intolerance	Improved	Significant
Gastrointestinal	Worsened	Not significant after correction
Bladder	Worsened	Not significant after correction
Pupillomotor	Worsened	Statistically worse

Figure 3 summary from the original: two decades of care moved the orthostatic needle and left the visceral and pupillomotor domains untouched or worse.

The authors are appropriately blunt that current treatment appears less effective for gastrointestinal symptoms, and name the difficulty of diagnosing and treating related conditions, mast cell activation syndrome among them, as part of the problem.

4. Baseline severity did not predict trajectory. In fact it inverted.

There were no significant baseline COMPASS-31 differences between groups. And NOT IMPROVED patients rated their overall health at the first Vanderbilt visit as better than IMPROVED patients did. The authors offer a mechanistic reading consistent with prior work: neuropathic POTS may deteriorate slowly from a higher starting point, while non-neuropathic POTS starts worse and recovers.

The patient who does not look that sick at intake may be the one on the worse twenty-year trajectory. Initial impression is not prognosis.

Reading This Through the POSS Lens

At ESSI we describe Pelvic Overlap Spectrum Syndrome (POSS) as a mechanism-based cluster rather than a diagnosis of exclusion: the patient with modest structural lesion burden and disproportionate multisystem symptom burden, the discordance phenotype. POTS is one of the ten canonical POSS domains, alongside mast cell, connective tissue, visceral hypersensitivity and central sensitisation.

The central claim of the framework is that in these patients, outcome tracks unaddressed domains, not lesion burden. Bourne et al. is the closest thing to a longitudinal test of that claim currently available in any adjacent literature, and it comes out the way the framework predicts. The prognostic split fell along mechanism lines. Single-domain therapy produced single-domain improvement. Initial severity was uninformative.

Table 2: What complete excision addresses, and what it does not

Domain	Addressed by Excision?	What to Say Before Surgery
Structural lesion (deep and superficial disease, adhesions, obliterated cul-de-sac)	Yes, directly	This is what the operation is for, and it is worth doing well.
Nerve entrapment (somatic nerve involvement, sciatic and obturator disease)	Yes, if identified	Requires it to be looked for preoperatively and mapped.
Autonomic / POTS (orthostatic intolerance, palpitations, exercise intolerance)	No	Twenty-year data say these symptoms persist. Promising resolution sets up a patient who will conclude surgery failed her.
Visceral hypersensitivity (bladder pain, bowel urgency, dysmotility)	Partially at best	Often the residual burden. Needs its own treatment plan, started before the operation.
Mast cell (flushing, food and drug reactivity, cyclic flares)	No	Perioperative stabilisation matters more than most teams appreciate.
Central sensitisation (widespread pain, allodynia, poor sleep)	Indirectly, over time	Removing nociceptive input helps, but does not reverse an established central state on its own.

Three consequences follow for the endometriosis clinic. Preoperative counselling has to be domain-specific. Failure to improve after complete excision is a differential, not a verdict, sometimes a lesion was missed and reoperation is right, but if the residual burden lives in the bladder, the bowel and the autonomic nervous system, that is a mechanistic pattern with a name, and it is not a psychiatric finding. And the neuropathy, gastroparesis, and overactive bladder triad deserves to be an explicit intake item rather than a footnote in the review of systems.

The Corollary: The Discipline of Continuing to Look

If the twenty-year outcome in a multisystem disorder is determined by which additional mechanisms are present, then looking for those mechanisms stops being a courtesy and becomes a prognostic act. Three things are systematically underlooked in this population.

Endometriosis

The ascertainment gap is visible in a single pair of numbers. In the long-term POTS outcomes survey of young women, roughly 72 percent reported that their POTS symptoms worsened before and during menses, a very loud cyclicity signal, while under 9 percent carried a diagnosis of endometriosis. Nobody is operating on the dysautonomia patient to find out. Meanwhile, on our side of the ledger, the endometriosis patient who mentions palpitations and standing fatigue is routinely told it is anxiety or deconditioning, and the autonomic question is never asked either.

Both specialties are producing the same artifact: a comorbidity that looks uncommon because nobody is measuring it. In POTS series where a gynaecologic specialist actually made the determination, POTS patients with endometriosis carried substantially higher rates of connective tissue and immune comorbidity than POTS patients without, exactly what you would expect if endometriosis is marking a shared terrain rather than sitting beside one.

Cyclic worsening of any multisystem symptom burden is a gynaecologic finding until proven otherwise. That is true of fatigue, orthostatic intolerance, migraine, bowel and bladder symptoms. It is not a soft sign.

Vascular Disease

Pelvic venous disorders are the second omission, and they sit precisely at the intersection of our two fields. Left common iliac vein compression (May-Thurner), left renal vein compression (nutcracker) and gonadal reflux with para-ovarian varices all produce pelvic pain, dyspareunia, deep ache worsened by standing and reduced venous return, and all are more prevalent in POTS cohorts than in controls. A systematic review published this year found compression anatomy detected frequently in established POTS, with symptom improvement reported after stenting, embolisation or bypass.

Necessary Caution

Anatomic compression is common in people who are perfectly well, roughly 40 percent of healthy controls. That means imaging-detected compression has real power to rule in a mechanism when the clinical picture already fits, and essentially none to rule out anything. It has no standing at all as a screening test in an asymptomatic patient. The treatment literature is retrospective, uncontrolled and unblinded in a population with high placebo responsiveness and a strong desire for an answer. Venous stents in young women are not a reversible decision.

So: image when the clinical syndrome points there, and image properly, since standard ultrasound frequently misses what dynamic contrast-enhanced MRA or venography will show. Then treat conservatively and stage interventions. Continuing to look is not the same as intervening on everything you find. The discipline is in holding both.

Autoimmunity

The third omission is immunologic, and it is closest to how we work at ESSI. The Bourne cohort itself listed autoimmune disease among its most frequently reported comorbidities, alongside IBS, ME/CFS and neuropathy. That is consistent with a substantial body of work: ganglionic acetylcholine receptor antibodies in roughly one in seven patients in the Mayo series; antinuclear antibodies in about a quarter and some marker of autoimmunity in nearly a third of 100 consecutive POTS patients; and adrenergic and muscarinic G-protein-coupled receptor autoantibodies reported at high frequency, with functional data showing patient IgG can activate alpha-1 and beta-1 adrenergic receptors and contract arterioles. There is even a report suggesting anti-muscarinic antibodies concentrate in the subgroup with persistent gastrointestinal symptoms, which, notably, is the subgroup Bourne et al. identified as doing worst over two decades.

The same caution applies, harder. GPCR autoantibody assays are not standardised, pathogenicity is inferred rather than established, and these are not yet clinical diagnostics. Ordering them outside a research context mostly generates numbers nobody can act on. But that is an argument against premature use of an unvalidated assay, not against immune evaluation. Thyroid autoimmunity, coeliac disease, Sjogren’s syndrome, antiphospholipid syndrome and inflammatory arthropathy are prevalent here, diagnosable with ordinary tests, and treatable. And the endometriosis-autoimmune association is established at meta-analytic level and reproduced in electronic health record cohorts numbering in the tens of thousands, where autoimmune clustering falls out of the data without being looked for.

Figure 4: Three lanes, and the rule-in / rule-out asymmetry (the discordance phenotype: modest lesion burden, disproportionate multisystem symptoms)

Lane	Look When	The Gap or Signal	Rule In / Rule Out Power
01. Endometriosis	Cyclic worsening of any multisystem symptom: fatigue, orthostasis, bowel, bladder, migraine	72 percent report perimenstrual worsening of POTS; fewer than 9 percent carry an endometriosis diagnosis	Strong to rule in; weak to rule out
02. Vascular	Left flank pain, haematuria, unilateral left leg swelling, post-coital ache, clear upright worsening	69 percent of POTS patients have greater than 50 percent left common iliac vein compression, versus 40 percent of healthy controls	Strong to rule in; weak to rule out
03. Autoimmunity	Sicca symptoms, arthralgia, thyroid symptoms, pregnancy loss, family autoimmune history, persistent GI symptoms	31 percent show any autoimmune marker; 25 percent ANA positive; 1 in 7 have ganglionic acetylcholine receptor (gAChR) antibody	Strong to rule in; weak to rule out

Figure 4 summary from the original: each lane is good enough to rule a mechanism in. None is good enough to rule one out. A positive finding in any of these lanes identifies a mechanism worth treating. A negative finding excludes almost nothing, which is precisely why the search has to continue past the first normal result.

Table 3: The three lanes, operationalised

Lane	Clinical Trigger	First-Line Evaluation	Rules In	Cannot Rule Out
Endometriosis	Cyclic worsening of any multisystem symptom; dysmenorrhoea; dyspareunia; cyclic bowel or bladder symptoms	Structured history, dynamic ultrasound with sliding sign, targeted MRI	Positive imaging or surgical confirmation establishes the lesion domain	Negative imaging excludes almost nothing; superficial and minimal disease is largely invisible
Pelvic venous	Left flank pain, haematuria or proteinuria, unilateral left leg swelling, post-coital ache, clear upright worsening	Dynamic contrast-enhanced MRA or venography, not standard ultrasound alone	Compression plus a matching clinical syndrome identifies a treatable mechanism	Compression is present in about 40 percent of healthy controls; absence of symptoms does not follow from absence of imaging findings, or vice versa
Autoimmune	Sicca, arthralgia, thyroid symptoms, recurrent pregnancy loss, family history, persistent GI symptoms	ANA, thyroid antibodies, coeliac serology, antiphospholipid panel where indicated	A positive, actionable diagnosis with established treatment	Seronegative autoimmune disease is real; a clean panel does not close the domain

These are three different failure modes of the same habit. The surgical lesion nobody looked for. The mechanical lesion nobody imaged. The immune process nobody serologised. In POSS terms, each is a separate domain, and Bourne et al. is telling us, with twenty years of follow-up, that domains are what determine outcome.

The patient does not need every test. She needs a clinician who has not stopped generating hypotheses at the first diagnosis that fit.

Where the Paper Is Weak, and It Should Be Said

Forty-four participants. A single national referral centre, which selects hard. A 62 percent response rate among those who could be reached at all, and survivorship in a survey means the people doing worst may simply not have answered.

The most serious limitation is the retrospective COMPASS-31. Participants were asked to recall an autonomic symptom state from as much as two decades earlier. That is not a measurement; it is a reconstruction, and it is systematically vulnerable to current state colouring the memory of the past. The finding that NOT IMPROVED patients recalled better baseline health is genuinely interesting, but it is also exactly what recall bias would manufacture if people who feel worse now remember the past as better.

The survey was fielded in 2021 and published in 2026, so the cohort’s experience predates the long COVID wave entirely. That is arguably a strength for mechanism, this is classical POTS, uncontaminated, and a limitation for generalising to the post-2020 population.

None of this makes the paper less useful. It makes it a hypothesis-generating dataset pointing at something the multisystem-pain literature has been circling for a decade. The right response is not to over-read 44 patients. It is to build the prospective cohort this paper argues for.

What I Take From It

The diagnostic delay embedded in the numbers is its own indictment: median 23 years from onset, 17 from diagnosis, roughly six years lost in this cohort, and the study’s own lead author has described a twelve-year personal delay. That interval is the same order of magnitude as the delay in endometriosis, in the same patients, driven by the same failure of pattern recognition across specialty boundaries.

We have been telling these women, for a long time, that each of their problems belongs to a different doctor. Bourne et al. is a twenty-year receipt showing what that costs. The patients who did worst are the ones whose disease crossed the most boundaries, and the ones our system is structurally worst at treating.

Questions This Raises

Does POTS go away over time?

Rarely. In this 20-year follow-up only 2 percent had complete symptom resolution and 98 percent still had symptoms. Around 46 percent described improvement, but improvement is not remission. The idea that POTS is self-limiting is not supported by long-term data, and it does not reliably resolve at menopause either, 41 percent of this cohort was postmenopausal and still symptomatic.

Which POTS patients do worst over decades?

Those with neuropathy, gastroparesis and overactive bladder symptoms. All but one patient in the NOT IMPROVED group had at least one of these, compared with 20 percent of the IMPROVED group. This looks like a neuro-visceral phenotype, plausibly involving small-fibre and autonomic neuropathy, rather than a more severe version of the same illness.

If I have endometriosis and POTS, will excision surgery help my POTS?

It should not be expected to. Complete excision addresses the structural lesion domain and, where identified, nerve involvement. It does not treat autonomic dysfunction, mast cell activation or established central sensitisation. Surgery can meaningfully reduce pain while orthostatic symptoms continue unchanged, which is why counselling has to be domain-specific before the operation, not explained afterwards.

Should POTS patients be screened for pelvic venous compression?

Not as routine screening. Compression anatomy is found in roughly 40 percent of healthy people, so imaging can rule a mechanism in when the clinical picture fits but cannot rule anything out. Image when there is left flank pain, haematuria, unilateral left leg swelling, post-coital ache or clear upright worsening, and use dynamic contrast-enhanced MRA or venography, since standard ultrasound often misses it.

Is POTS an autoimmune disease?

In some patients, plausibly. Ganglionic acetylcholine receptor antibodies appear in roughly one in seven patients in the Mayo series, about a quarter of one 100-patient cohort were ANA positive, and adrenergic and muscarinic receptor autoantibodies have been reported at high frequency with supporting functional data. But these assays are not standardised and pathogenicity is inferred rather than proven, so they are not yet clinical diagnostics.

What is the discordance phenotype in POSS?

A patient with modest structural lesion burden and disproportionate multisystem symptom burden. Pelvic Overlap Spectrum Syndrome treats this as a mechanism-based cluster across ten domains rather than a diagnosis of exclusion, and predicts that outcome tracks which domains go unaddressed rather than how much disease is found at surgery.

References

1. Bourne KM, Gamboa A, Black B, Hall J, Biaggioni I, Shibao CA, et al. Long-term outcomes in patients with postural orthostatic tachycardia syndrome with an average follow-up of over 20 years. J Intern Med. 2026;1-13. doi:10.1111/joim.70104
2. Fedorowski A. The long and winding road of postural orthostatic tachycardia syndrome. J Intern Med. 2026. doi:10.1111/joim.70121
3. Vernino S, Bourne KM, Stiles LE, Grubb BP, Fedorowski A, Stewart JM, et al. POTS: state of the science and clinical care from a 2019 NIH expert consensus meeting, part 1. Auton Neurosci. 2021;235:102828.
4. Raj SR, Bourne KM, Stiles LE, Miglis MG, Cortez MM, Miller AJ, et al. POTS: priorities for POTS care and research from a 2019 NIH expert consensus meeting, part 2. Auton Neurosci. 2021.
5. Sletten DM, Suarez GA, Low PA, Mandrekar J, Singer W. COMPASS 31: a refined and abbreviated Composite Autonomic Symptom Score. Mayo Clin Proc. 2012;87(12):1196-1201.
6. Boris JR, Shadiack EC, McCormick EM, MacMullen L, George-Sankoh I, Falk MJ, et al. Long-term POTS outcomes survey: educational, economic, and social impact. J Am Heart Assoc. 2025;14(21):e042365.
7. Boris JR, et al. The Long-Term POTS Outcomes Survey, gynecologic findings: a cross-sectional survey in young women. Obstet Gynecol Int. 2025. doi:10.1155/ogi/8872884
8. Bourne KM, Sheldon RS, Hall J, Lloyd M, Kogut K, Sheikh N, et al. Compression garment reduces orthostatic tachycardia and symptoms in patients with POTS. J Am Coll Cardiol. 2021;77:285-296.
9. Taub PR, Zadourian A, Lo HC, Ormiston CK, Golshan S, Hsu JC. Randomized trial of ivabradine in patients with hyperadrenergic POTS. J Am Coll Cardiol. 2021;77:861-871.
10. Poor health-related quality of life in postural orthostatic tachycardia syndrome in comparison with a sex- and age-matched normative population. Clin Auton Res. 2023. doi:10.1007/s10286-023-00955-9
11. Fedorowski A, Sutton R. Autonomic dysfunction and postural orthostatic tachycardia syndrome in post-acute COVID-19 syndrome. Nat Rev Cardiol. 2023;20:281-282.
12. Thieben MJ, Sandroni P, Sletten DM, Benrud-Larson LM, Fealey RD, Vernino S, et al. Postural orthostatic tachycardia syndrome: the Mayo Clinic experience. Mayo Clin Proc. 2007;82(3):308-313.
13. Blitshteyn S. Autoimmune markers and autoimmune disorders in patients with postural tachycardia syndrome (POTS). Lupus. 2015;24(13):1364-1369.
14. Gunning WT, Kvale H, Kramer PM, Karabin BL, Grubb BP. Postural orthostatic tachycardia syndrome is associated with elevated G-protein coupled receptor autoantibodies. J Am Heart Assoc. 2019;8(18):e013602.
15. Fedorowski A, Li H, Yu X, Koelsch KA, Harris VM, Liles C, et al. Antiadrenergic autoimmunity in postural tachycardia syndrome. Europace. 2017;19(7):1211-1219.
16. Association of autoantibodies to muscarinic acetylcholine receptors with gastrointestinal symptoms and disease severity in patients with POTS. Clin Auton Res. 2022. PMID: 34993884.
17. Knuttinen MG, Zurcher KS, Khurana N, et al. Imaging findings of pelvic venous insufficiency in patients with postural orthostatic tachycardia syndrome. Phlebology. 2021.
18. Imami MR, Dos Santos Galdino V, Harvey P, et al. Pelvic venous disorders and orthostatic intolerance: a systematic review of diagnostic associations and treatment outcomes. Auton Neurosci. 2026.
19. Association and post-iliac vein stenting symptom improvement of POTS and orthostatic intolerance with pelvic venous disorders: two retrospective studies. eClinicalMedicine. 2026.
20. May-Thurner syndrome in patients with POTS and Ehlers-Danlos syndrome: a case series. Eur Heart J Case Rep. 2022;6(4):ytac161.
21. An overview of vascular compression syndromes and associations with autonomic dysfunction: a review. 2026. PMCID: PMC13023622.
22. Rahmioglu N, Mortlock S, Ghiasi M, et al. The genetic basis of endometriosis and comorbidity with other pain and inflammatory conditions. Nat Genet. 2023;55(3):423-436.
23. Shigesi N, Kvaskoff M, Kirtley S, et al. The association between endometriosis and autoimmune diseases: a systematic review and meta-analysis. Hum Reprod Update. 2019;25(4):486-503.
24. Comorbidity analysis and clustering of endometriosis patients using electronic health records. Cell Rep Med. 2025.
25. Pollack B, von Saltza E, McCorkell L, Santos L, Hultman A, Cohen AK, Soares L. Female reproductive health impacts of long COVID and associated illnesses including ME/CFS, POTS, and connective tissue disorders. Front Rehabil Sci. 2023;4:1122673.
26. Prevalence and predictors of gastrointestinal dysmotility in patients with hypermobile Ehlers-Danlos syndrome: a tertiary care center experience. 2020. PMCID: PMC7255528.
27. Vidali A. Pelvic Overlap Spectrum Complex. Endometriosis Surgical Specialists International. internationalendo.com/pelvic-overlap-spectrum-complex-pelvic-pain-essi/
28. Endometriosis, POTS, MCAS and long COVID: the shared genetic link. Endometriosis Surgical Specialists International. internationalendo.com/endometriosis-pots-mcas-long-covid-shared-genetic-link-essi/

Andrea Vidali, MD

Gynecologic surgeon, fellowship-trained reproductive endocrinologist and reproductive immunologist. Founder and Director, Endometriosis Surgical Specialists International.

This commentary is educational and does not constitute medical advice or establish a physician-patient relationship. Clinical decisions should be made with your own care team.