OpenAI Reevaluates Tarlov Cyst Statistics

This is a transcript of a short chat that I had with OpenAI regarding Tarlov Cyst Statistics. It demonstrates how the under-reporting of Tarlov Cysts, beginning with radiologists refusing to include Tarlov Cysts in written reports, has affected patients and the industry with respect to TCD.

Scott Sullivan

6/5/20269 min read

Could you give me an overview of the current widely-accepted statistics on Tarlov Cyst Disease with a focus on The United States of America population?

8:26 PM

Worked for 52s

Here’s the current statistical picture, focused on the U.S.

Key Takeaway
Tarlov cysts themselves are not extremely rare on MRI, but symptomatic Tarlov Cyst Disease is much less common and is poorly counted because the U.S. does not appear to have a national registry or surveillance dataset for it.

Best Widely Cited Estimates

MeasureCurrent estimateU.S. population implicationAny Tarlov cyst in the U.S.about 3.82%about 13.1 million AmericansGlobal pooled prevalenceabout 4.18%broadly consistent with U.S. estimateSymptomatic among people with cysts5% to 8% in some clinical summaries; 15.59% in the 2021 meta-analysisroughly 650,000 to 2.0 million Americans, depending on definitionSex distributionMore common in womenmeta-analysis: 5.84% women vs. 3.03% men globallyTypical locationMostly sacral spineS2 was the most common sacral level, about 46.7% of sacral cystsPediatric prevalenceRaremeta-analysis: 0.53%

Using the U.S. Census Bureau’s July 1, 2025 population estimate of 341,784,857, the 3.82% U.S. cyst prevalence estimate implies about 13,056,000 people with a Tarlov cyst. If 15.59% are symptomatic, that implies about 2,035,000 symptomatic cases; if using the more conservative 5% to 8% symptomatic range cited by Cleveland Clinic, the estimate is about 653,000 to 1,044,000.

Important Caveats
“Tarlov cyst” and “Tarlov Cyst Disease” should not be treated as the same statistic. Many cysts are incidental and asymptomatic. The symptomatic disease count is the uncertain part because symptoms overlap with back pain, sciatica, pelvic pain, bladder/bowel disorders, and gynecologic or urologic conditions.

Also, older patient-facing sources sometimes cite 5% to 9% of the general population for small asymptomatic cysts, while the strongest pooled radiology meta-analysis gives 4.18% globally and 3.82% for North America/U.S.-relevant data. So the most defensible modern estimate is “around 4% have a cyst; a smaller subset are symptomatic.”

Sources: 2021 meta-analysis in Surgical and Radiologic Anatomy, StatPearls/NCBI, Cleveland Clinic, AANS, U.S. Census QuickFacts.

8:27 PM

Which study are you citing as "the strongest pooled radiology meta-analysis"

8:28 PM

I meant this study:

Klepinowski, Orbik & Sagan, “Global incidence of spinal perineural Tarlov’s cysts and their morphological characteristics: a meta-analysis of 13,266 subjects,” Surgical and Radiologic Anatomy, 2021

It pooled 22 radiologic studies involving 13,266 subjects and estimated:

Global pooled prevalence: 4.18%
North America prevalence: 3.82%
Symptomatic correspondence: 15.59%
Female predominance: 5.84% women vs. 3.03% men
Most common site: sacral spine, especially S2

I called it “strongest” because it is the most directly relevant pooled radiology meta-analysis I found for prevalence, not because it is flawless. The authors themselves note that the included studies were level 3 evidence and that stronger studies are still needed.

8:29 PM

This meta-analysis is familiar to me. They used the written MRI reports to identify Tarlov Cyst prevalence. But the study has a methodological flaw because it does not address the systematic underreporting of Tarlov Cysts by radiologists. Garreth Clarke did a study in england finding that only 20% of visible Tarlov Cysts were included in written MRI reports; and only 2.5% of patients were told. Please do an analysis of the Klepinowski et.Al. study to estimate how this may have affected their results. Also, not the work of Meike Hulens particularly "Tarlov Cysts are under-reported. Ten reasons why" to confirm that under-reporting is a serious flaw in the Klepinowski retrospective, analyzing each of the included studies to identify which ones were most likely affected by under-reporting, versus those less likely based upon their use of actual MRI imaging instead of written reports alone.

8:34 PM

Worked for 2m 45s

Yes. Your critique is valid, but I’d frame it precisely:

Klepinowski et al. did not directly audit routine MRI reports themselves; they pooled published radiologic studies. The vulnerability is that several source studies appear to depend on routine report/record ascertainment or non-dedicated incidental-finding review, while others used deliberate image review. Clarke’s audit shows why that distinction matters: in 1,064 reviewed lumbosacral MRIs, visible sacral Tarlov cysts were present in 14.85%, but only 21% of those cyst-positive scans mentioned them in the radiologist report, and only 2.5% of cyst-positive patients were told.

If Clarke’s ~20% reporting sensitivity were applied naively to Klepinowski’s pooled 4.18%, the implied imaging-visible prevalence would be about:

4.18% / 0.21 = 19.9%

That is probably too aggressive as a universal correction, because many included studies did re-review images. But it shows the possible direction and scale of bias: the pooled estimate could be materially low if report-dependent studies contributed meaningfully.

A more conservative sensitivity analysis:

AssumptionAdjusted prevalence estimateNo underreporting correction4.18%Only clearly vulnerable low-prevalence/report-like studies correctedroughly 6% to 8%All studies corrected using Clarke’s 21% report rateroughly 20%

My best estimate from this evidence is: Klepinowski’s 4.18% is better interpreted as a lower-bound pooled reported/detected prevalence, not a true anatomic prevalence. A plausible corrected range might be ~6% to 12%, with the upper tail potentially higher in low-back/sciatica MRI populations.

Study-Level Vulnerability

Klepinowski sourcePrevalenceUnderreporting riskBurdan8.91%Lower: detailed MRI review, high morphologic detailGleeson0.38%Moderate/high: very low rate; may reflect protocol/reporting limitsGopalakrishnan1.00%High: broad low-back MRI categorization, AQUA methodology high riskJeong10.73%Lower: anatomy-focused MRI reviewJoo1.72%Lower/moderate: MRI anatomy study, but not TC-focusedKomisaruk & Lee66.67%Lower for detection; special PGAD subgroup, not general prevalenceKuhn12.00%Lower: dedicated sacral MRI cohort with cyst morphologyLangdown1.53%High/moderate: large retrospective clinical MRI cohort; low rate suggests report/ascertainment lossLarsen17.65%Different issue: myelography, not routine MRI-report biasLim16.12%Lower: dedicated MRI review in pudendal neuralgia cohortOaklander50.00%Unclear; special PGAD subgroupPark2.13%Moderate: incidental findings study; may still miss low-priority cystsPaulsen4.60%Moderate/lower: older MRI prevalence study, but method details matterPetrasic1.62%Lower/moderate: MR neurography cohort, clinically selectedPink6.67%High/unclear: special subgroup, small sampleRamadorai1.49%Lower: MRI reread by radiologist and spine surgeon, pediatric cohortRamirez0.38%High: pediatric/back-pain evaluation; likely not TC-focusedSenoglu1.30%Lower/moderate: MRI anatomy study, not TC-focusedShi2.67%High/unclear: limited methodological visibilityTani9.80%Lower/moderate: sacral meningeal cyst prevalence focusZacharakis6.67%Moderate/unclear: small MRI cohortZeitoun0.46%High: very low rate; likely protocol/ascertainment sensitive

Hulens et al. strongly supports your broader concern. Their “Ten reasons why” review argues symptomatic Tarlov cysts remain underdiagnosed, often dismissed as incidental, and missed because clinicians do not ask about sacral, pelvic, bladder, bowel, and sexual symptoms. That does not prove every Klepinowski source is report-biased, but it makes routine-report-based prevalence a serious methodological weakness.

So I’d revise my earlier phrasing: Klepinowski is the strongest pooled prevalence meta-analysis available, but not necessarily the best estimate of true visible/anatomic prevalence if underreporting is systematic.

Sources: Klepinowski et al. 2021, Clarke et al. 2016/2017 audit, Hulens et al. “Ten reasons why” preprint.

8:37 PM

Now, can you give me a summary of the findings of Komisaruk and Goldstein and Kim et.Al. and a list of their published studies on PGAD and the link to Tarlov Cysts. In particular, I would like you to focus on how PGAD was always considered purely a mental illness until approximately 2013 when Dr. Komisaruk and Goldstein stumbled across this link between PGAD and Tarlov Cysts.

8:41 PM

Worked for 1m 31s

Short Version
The key turning point was 2012, not 2013: Komisaruk and Lee reported that 12 of 18 women with PGAD symptoms had sacral Tarlov cysts on MRI, a much higher rate than expected. Komisaruk later worked with Irwin Goldstein and others to frame PGAD/GPD as a condition with possible neurologic generators, especially sacral nerve-root irritation in the cauda equina.

I’d phrase the history this way: PGAD was not “always officially a mental illness,” but before the Tarlov cyst work it was often handled through sex therapy, psychiatry, pharmacology, pelvic floor care, or vague psychogenic explanations. The 2012 finding helped move the field toward a neurologic/sensory neuropathy model.

Main Findings
Komisaruk & Lee’s 2012 study began serendipitously: a colleague mentioned that his wife had PGAD and an incidental MRI finding of Tarlov cysts. They reasoned that S2/S3 dorsal roots carry genital sensory input, so cysts on those roots could generate unwanted genital sensations. They asked women in a PGAD support group to submit sacral MRIs. Tarlov cysts were found in 66.7%, all around S2/S3. They concluded that at least some PGAD may be a Tarlov-cyst-induced paresthesia, not primarily a psychiatric condition.

Komisaruk & Goldstein’s 2017 review broadened this into a model of PGAD as genital sensory neuropathy/radiculopathy. Their argument: if genital anesthesia or pudendal blocks fail, the source may be “upstream” at the sacral roots or cauda equina. They specifically named Tarlov cysts and herniated/annular disc pathology as possible drivers.

Goldstein, Komisaruk, Choll Kim, Noel Kim, and colleagues then helped formalize the modern PGAD/GPD framework in the 2021 ISSWSH consensus paper. That paper broadened the term to PGAD/GPD, or persistent genital arousal disorder/genito-pelvic dysesthesia, and proposed a five-region diagnostic model: end organ, pelvis/perineum, cauda equina, spinal cord, and brain. In that framework, Tarlov cysts are a Region 3 cauda equina/sacral nerve-root pathology.

Kim/Goldstein/Komisaruk’s later surgical work extended this model. A 2023 Journal of Sexual Medicine paper focused on PGAD/GPD from lumbosacral annular tear-induced sacral radiculopathy, not Tarlov cysts, and reported improvement after lumbar endoscopic spine surgery in selected patients. A 2026 NASSJ paper focused directly on Tarlov cyst surgery in PGAD/GPD: 19 selected patients underwent Tarlov cyst excision/imbrication; 14 of 19, or 73.7%, reported postoperative improvement, with no worsening symptoms reported in that cohort.

Key Publications

Komisaruk & Lee, 2012
“Prevalence of Sacral Spinal (Tarlov) Cysts in Persistent Genital Arousal Disorder.” Found Tarlov cysts in 12/18 PGAD MRIs.
Komisaruk & Goldstein, 2017
“Persistent Genital Arousal Disorder: Current Conceptualizations and Etiologic Mechanisms.” Framed PGAD as potentially caused by Tarlov cysts or disc-related cauda equina irritation.
Goldstein, Komisaruk, Pukall, Noel Kim, Choll Kim et al., 2021
ISSWSH consensus paper on PGAD/GPD. Created the modern diagnostic/process-of-care framework.
Kim, Goldstein, Komisaruk et al., 2023
Lumbar endoscopic spine surgery for PGAD/GPD caused by annular tear-induced sacral radiculopathy.
Goldstein, Choll Kim, Komisaruk et al., 2023 abstract
Step-care algorithm and surgical outcomes for PGAD/GPD patients with Tarlov cyst-induced sacral radiculopathy.
Coorapati, Irwin Goldstein, Sue Goldstein, Komisaruk & Choll Kim, 2026
“Tarlov Cysts and Sexual Dysfunction: A Multidisciplinary Approach to Evaluation and Surgical Treatment.” Reported 73.7% improvement after selected Tarlov cyst surgery.

A crucial related non-Goldstein/Kim paper is Feigenbaum & Boone, 2015, which reported improvement/elimination of PGAD symptoms after neurosurgical treatment of sacral spinal meningeal cysts.

8:43 PM

Now, could you give me just a reaction based upon what you found from the first Tarlov Cyst question to the PGAD link. Just...open forum...what does that make you think?

8:46 PM

Honestly? It makes me think this is a textbook case of a condition getting trapped between specialties.

The Tarlov cyst literature already has this odd pattern: cysts are visible, anatomically plausible, sometimes clearly symptomatic, but often dismissed because many are incidental. Then PGAD enters the picture, and the stakes get even stranger because the symptoms are intimate, embarrassing, hard to describe, and historically easy to psychologize. That is exactly the kind of clinical situation where “rare,” “incidental,” and “anxiety” can become a diagnostic dead end.

The Komisaruk/Lee finding feels important less because it proves every PGAD case is Tarlov-related, and more because it broke the frame. It said: wait, what if this “arousal” is not desire, not hypersexuality, not psychiatric excess, but abnormal sensory firing from sacral nerve roots? That is a profound reframing. It turns a morally and psychologically loaded symptom into a neuroanatomic problem.

The underreporting issue makes the whole thing feel even more consequential. If radiologists often omit Tarlov cysts because they assume they are incidental, then clinicians never connect them to pelvic, bladder, bowel, sexual, or genital sensory symptoms. And if clinicians do not ask those symptoms because they feel unrelated to back pain, the loop stays broken. The patient is left carrying the weirdest, most stigmatized part of the disease alone.

My impression is: the field is still messy, and causality is not settled. But the pattern is too coherent to dismiss. Sacral cysts, sacral radiculopathy, pelvic dysesthesia, bladder/bowel symptoms, sexual dysfunction, PGAD/GPD: these are not random neighbors. They sit on the same wiring map.

So what it makes me think is: Tarlov cyst disease may be much less “rare” than “rarely recognized,” and PGAD may be one of the clearest examples of why that recognition matters.

8:46 PM

How would you estimate the psychological toll upon patients, particularly those who are never told that they have Tarlov Cysts; and then those who do find out, but encounter a health care industry and insurance industry that are still in denial of the truth about this disease?

8:48 PM

Worked for 13s

I would estimate the psychological toll as severe, cumulative, and partly iatrogenic, meaning caused or worsened by the medical system itself.

There are really three injuries stacked together.

First is the primary neurologic burden: pain, sitting intolerance, pelvic symptoms, bladder/bowel issues, sexual dysfunction, PGAD/GPD, sleep disruption. Chronic pain alone is strongly associated with depression, anxiety, isolation, reduced work capacity, and suicidality.

Second is the “not knowing” burden. If a patient has visible Tarlov cysts but is never told, they may spend years trying to explain symptoms that feel bizarre, intimate, or frightening. In PGAD especially, the symptom itself can produce shame because it is mislabeled by others, and sometimes by the patient, as desire, hypersexuality, anxiety, trauma, or psychiatric instability. That is psychologically brutal: the body is sending abnormal nerve signals, but the person is left to interpret it morally or emotionally.

Third is the denial burden. Once a patient learns they have Tarlov cysts, they may reasonably feel relief: “There is something there.” But then they encounter clinicians, radiology reports, insurers, and sometimes family members saying, “That is incidental.” That creates a kind of epistemic injury. The patient is not only sick; they are forced to litigate whether their suffering is real.

My rough estimate:

Patients with symptomatic Tarlov cysts who are believed and treated appropriately: high distress, but with a coherent path forward.
Patients never told about visible cysts: higher risk of self-blame, psychiatric mislabeling, delayed diagnosis, and worsening disability.
Patients who know but are dismissed or denied coverage: highest risk group, because they carry both the symptoms and the burden of advocacy against institutional disbelief.

The PGAD link intensifies this dramatically. The literature describes PGAD/GPD as unwanted, intrusive, distressing, associated with catastrophization and sometimes suicidality. The 2006 descriptive study found distress predicted by intrusive unwanted arousal, continuous symptoms, shame, worry, and reduced sexual satisfaction. The ISSWSH consensus similarly emphasizes major psychosocial impact, including emotional lability, catastrophization, and suicidal ideation.

So my plain-language read is: the psychological toll is not secondary or incidental. It is part of the disease experience. And in many patients, the worst damage may come from the long delay between “there is an anatomic explanation” and “the system is willing to act like that explanation matters.”

Sources include the ISSWSH PGAD/GPD consensus, Komisaruk/Goldstein work, Oaklander’s PGAD neuropathy paper, and the PGAD descriptive literature.

8:48 PM