In Hereditary Heart Amyloidosis, Polyneuropathy Care Depends on Documentation Quality

When patients with hereditary transthyretin amyloidosis carry a diagnostic code for polyneuropathy, what actually backs that code varies enormously — and that variation appears to shape who receives powerful gene-silencing drugs. A retrospective chart review of 54 patients at MedStar Georgetown University Hospital, published in Cardiology and Therapy, found that only one third of the coded diagnoses were confirmed by objective electrodiagnostic testing, while one in five had no meaningful clinical documentation behind them at all. Gene silencer use tracked closely with the strength of that documentation rather than with the mere presence of a billing code, a finding the authors say exposes a widening gap between administrative coding and clinical medicine.

Hereditary transthyretin amyloidosis is an autosomal dominant disease caused by mutations in the TTR gene that force the transthyretin protein to misfold and deposit as amyloid fibrils in tissues. The heart and peripheral nervous system bear the brunt, but which organ dominates depends heavily on the specific variant, a person’s geographic ancestry and individual factors. The V142I variant is the most common pathogenic TTR mutation in the United States, carried by roughly 3 to 4 percent of African Americans — an estimated 1.2 to 1.5 million people. It classically produces a cardiac-predominant phenotype, with heart failure symptoms typically emerging after age 60, which is why it was historically characterized as a cardiomyopathy. Neurological involvement, however, appears more common than that older picture implied.

How common is genuinely contested, and the new study argues the answer depends less on the disease than on how hard clinicians look. A multinational biobank analysis found V142I carriers had significantly more polyneuropathy than non-carriers, though absolute rates drawn from diagnostic codes were low. US series that systematically assessed symptomatic or referred patients found polyneuropathy in nearly everyone evaluated, whereas a UK national cohort of the same variant, reviewed as patients presented, found it in roughly 40 percent and generally mild. Reported prevalence spans from approximately 2 to 9 percent in code-based studies to over 90 percent at specialized centers employing nerve conduction studies and skin biopsy. The study’s authors conclude that this variation reflects ascertainment intensity rather than true population differences: the more actively polyneuropathy is sought, the more it is found.

To assemble their cohort, the researchers queried institutional electronic medical records over approximately eight years, from 2017 to 2025, using co-occurring ICD-10 codes for transthyretin amyloidosis and polyneuropathy — E85.1, E85.82 and/or G63 — with confirmation of a pathogenic TTR variant through genetic testing. The wild-type code was deliberately included to cast a broad net, since hereditary cases are sometimes miscoded as wild-type before genetic testing settles the subtype. Individual chart review then verified clinical status and genotype. Patients with unknown genotype, asymptomatic gene carriers and pre-symptomatic carriers found through family screening were excluded. The authors stress that because cohort entry required a polyneuropathy code, the sample is enriched rather than representative, and the proportions reported are not population prevalence estimates.

The 54 patients were older — median age at last follow-up of 77 years, with an interquartile range of 69 to 81 — and overwhelmingly non-Hispanic Black (90.7 percent) and male (64.8 percent). V142I accounted for 87.0 percent of cases, with the remainder carrying V50M, A140S, T60A or P84L variants. Confirmed cardiac involvement was present in 51 patients, or 94.4 percent, consistent with the expected cardiac-predominant phenotype of V142I. Among patients with confirmed cardiomyopathy who were eligible for stabilizer therapy, 40 of 42 — 95.2 percent — were receiving treatment, reflecting established practice since pivotal trials demonstrated a mortality benefit for the stabilizer tafamidis in ATTR cardiomyopathy.

The polyneuropathy picture was far messier. Each patient was assigned to one of four mutually exclusive categories based on the highest level of documentation: polyneuropathy confirmed by abnormal electromyography or nerve conduction studies; provider-documented polyneuropathy without confirmatory testing; patient-reported neuropathic symptoms such as numbness, tingling or pain without examination findings or formal neurological evaluation; and unclear status with insufficient documentation to assess neurological function. Only 18 patients, 33.3 percent, had ancillary testing confirmation. Fourteen, or 25.9 percent, had provider documentation without testing, and 11, or 20.4 percent, had symptoms only. The remaining 11 — one in five — had insufficient documentation to support the code at all. Carpal tunnel syndrome and other mononeuropathies, though recognized early peripheral manifestations of the disease, were assessed separately and did not count toward the polyneuropathy classification.

Treatment patterns followed the documentation gradient with striking fidelity. Across the full cohort, 16 patients, or 29.6 percent, received a gene silencer — patisiran, vutrisiran or eplontersen — and 11 of them took one concurrently with a TTR stabilizer, with silencers layered onto continued stabilizer therapy rather than replacing it. Gene silencer use was highest among patients whose polyneuropathy was confirmed by ancillary testing, at 50.0 percent, versus 18.2 percent among those with symptoms alone. No patient in the cohort received a gene silencer under the recently approved ATTR cardiomyopathy indication, granted by the FDA in March 2025 on the basis of the HELIOS-B trial; all gene silencer access was predicated on neurological documentation. Neurology consultation and EMG completion also tracked with documentation level: consultation reached 97.7 percent of patients with confirmed or suspected polyneuropathy versus 36.4 percent of those with unclear status, and EMG completion 76.7 percent versus 9.1 percent, both differences highly significant.

Attribution complications loom over the entire enterprise. The majority of patients had at least one condition independently capable of causing polyneuropathy — diabetes mellitus, lumbar spinal stenosis or vitamin B12 deficiency — a burden the authors note exceeds the roughly one-half reported in a prior US series. Polyneuropathy in a given patient may therefore not be attributable to amyloid deposition alone, and both the method of assessment and the confounder burden must be weighed when interpreting prevalence figures. Three patients, carrying V142I, V50M and P84L respectively, lacked confirmed cardiac involvement; all three received gene silencers rather than stabilizers, consistent with a neurologic-predominant phenotype, and may represent early detection through cascade screening or genuinely neurologic-predominant disease. Among patients without confounders, 78.6 percent received stabilizers and 28.6 percent gene silencers; among those with one or more confounders, the figures were 72.5 percent and 30.0 percent.

The therapeutic stakes of this documentation gap are rising. TTR stabilizers — tafamidis, acoramidis and diflunisal — prevent tetramer dissociation, with evidence supporting neurological benefit in mixed-phenotype patients. Gene silencers reduce hepatic TTR synthesis through RNA interference or antisense mechanisms, with established efficacy for polyneuropathy and emerging cardiac data. With vutrisiran now approved for cardiomyopathy, cardiologists can prescribe gene silencers without any neurological documentation, and the authors warn that polyneuropathy may go uncharacterized precisely because documentation is no longer required for treatment access. Their data suggest systematic neurological assessment retains independent clinical value — establishing baseline severity, guiding monitoring and identifying patients who may benefit from multidisciplinary management. A recent network meta-analysis favoring gene silencers over stabilizers remains hypothesis-generating in the absence of head-to-head trials, and combination therapy is not proven superior.

The study’s limitations are considerable and honestly stated. As a retrospective, single-center chart review relying on administrative coding, it may have overcaptured or undercaptured patients; the V142I-predominant cohort limits generalizability to other variants; the three patients without cardiac involvement permit little inference; ancillary testing was limited to EMG and nerve conduction studies, so small-fiber polyneuropathy detectable by skin biopsy may be undercounted; and the design precluded determining symptom-onset sequence or the clinical reasoning behind individual treatment decisions. The authors call for standardized polyneuropathy assessment criteria in hereditary ATTR — including which diagnostic modalities to employ and how to account for competing etiologies — and point to emerging tools such as serum neurofilament light chain, nerve ultrasound and structured patient-reported outcome measures. They also flag a broader consequence: as machine learning tools increasingly draw on coded data, the gap between a diagnostic code and the documentation behind it becomes more consequential, and coding systems that better reflect clinical reality are needed.

Subject of Research: Documentation quality of polyneuropathy diagnoses and its influence on treatment selection in a predominantly V142I hereditary transthyretin amyloidosis cohort

Article Title: Polyneuropathy Assessment and Treatment Patterns in Patients with Hereditary Transthyretin Amyloidosis

Article References: Streicher, N., & Wubet, H. (2026). Polyneuropathy Assessment and Treatment Patterns in Patients with Hereditary Transthyretin Amyloidosis. Cardiology and Therapy, 15(3), 365-375. https://doi.org/10.1007/s40119-026-00472-6

Image Credits: AI Generated

DOI: 10.1007/s40119-026-00472-6

Keywords: hereditary transthyretin amyloidosis, polyneuropathy, V142I variant, ICD-10 coding, gene silencers, TTR stabilizers, electromyography, tafamidis, patisiran, vutrisiran, cardiomyopathy, retrospective chart review