The diagnosis of M35.4 does not come down to one “black-and-white” test. It is a puzzle-assembly process — clinical picture + abnormalities in blood tests + changes on imaging + histopathological picture from biopsy. The more pieces fit, the more certain the diagnosis.

Diagnostic criteria

In clinical practice, the most commonly used criteria are the Pinal-Fernandez criteria (2014, updated 2016) or the classic Lebeaux–Sène criteria (2012), based on Shulman’s 1974 work. The two sets are convergent; they differ in the details of severity classification. Below is the description according to Pinal-Fernandez:

Major criterion (required)

  • Symmetric, plate-like induration of the skin and subcutaneous tissue of the limbs (and/or trunk), sparing the fingers and toes.

Minor criteria (at least 2 required)

  • Peripheral eosinophilia > 0.5 × 10⁹/L
  • Hypergammaglobulinaemia > 1.5 g/L (especially of the IgG class)
  • Muscle weakness and/or elevated aldolase
  • Presence of the groove sign or peau d’orange
  • Characteristic MRI picture: thickening and hyperintensity of the fascia on T2 images and contrast enhancement on T1 images after gadolinium
  • Fascia biopsy with an inflammatory infiltrate (lymphocytic-eosinophilic) and thickening of the collagen fibres
/ note

In practice the diagnosis requires a clinical picture + biopsy in most cases, but when the clinical picture is classic and the minor criteria are met, some centres diagnose on the basis of criteria + MRI without biopsy. The centre’s experience and the risk of a false-negative biopsy are the deciding factors.

Blood tests — what to order

No laboratory parameter is specific for M35.4, but their combination is strongly suggestive:

  • Full blood count with differential — peripheral eosinophilia is present in 60–80% of patients in the active phase (norm < 0.5 × 10⁹/L; in EF often > 1.0 × 10⁹/L, sometimes much higher). In the remaining 20–40% it may be normal or mild — that does not exclude the diagnosis.
  • ESR — often significantly elevated (50–100 mm/h)
  • CRP — usually moderately elevated
  • Total protein and protein electrophoresis — polyclonal hypergammaglobulinaemia (mainly IgG), present in ~50–70% of patients
  • Aldolase and CK (creatine kinase) — aldolase may be elevated with muscle involvement; CK is most often normal or only minimally elevated (this distinguishes EF from inflammatory myopathies, where CK is often very high)
  • ANA (antinuclear antibodies) — most often negative or low-positive; a high titre suggests scleroderma instead
  • Anti-Scl-70, anti-centromere — negative (test that excludes scleroderma)
  • Renal, liver and thyroid function — initially normal, but monitored in view of subsequent treatment
/ note

First-eosinophilia trap: a blood count with peripheral eosinophilia often leads down the allergology and parasitology diagnostic path. That is understandable — these are the most common causes of eosinophilia. But if the patient simultaneously has symmetric induration of the limbs, a systemic connective-tissue disease must be considered from the outset.

MRI — what EF looks like on imaging

Magnetic resonance imaging is today the best non-invasive imaging method in M35.4. Characteristic findings on MRI of the affected limb:

  • Thickening of the superficial and deep fascia — visible on T1 sequences as a linear, thickened band around the muscles
  • Fascial hyperintensity on T2 images (especially with fat saturation, STIR) — reflects inflammatory tissue oedema
  • Fascial contrast enhancement on T1 after gadolinium — corresponds to active inflammation
  • Symmetry of changes — bilateral involvement of the same muscle groups (most often around the biceps femoris and calf muscles)
  • The underlying muscles are usually normal — in contrast to inflammatory myopathies, where the muscle signal is pathological

MRI does not replace biopsy (there is not yet a widely accepted picture-specificity sufficient for diagnosis without histology), but it has three important uses:

  • Reinforces the clinical diagnosis — when the MRI picture is classic, diagnostic certainty rises
  • Indicates the biopsy site — reduces the risk of a false-negative biopsy by guiding the operator to the area of clearest changes
  • Monitors treatment response — follow-up MRI after 3–6 months of therapy shows objective resolution of oedema and enhancement

Fascia biopsy — gold standard

A full-thickness “skin-to-muscle” biopsy (full-thickness skin-to-muscle biopsy) remains the gold standard for diagnosing M35.4. It is a surgical, not a needle biopsy — a wedge of tissue is taken that includes all layers:

  • Epidermis and dermis
  • Subcutaneous tissue with vessels and septa
  • Superficial fascia
  • Deep fascia (key!) — adjacent to the muscle
  • Superficial layer of the muscle

A needle biopsy or punch biopsy alone is not enough, because it only reaches the dermis, while the most important changes are deeper. This is a common cause of first “negative” biopsies in patients who actually have EF.

Histopathological picture

The pathologist looks for specific features:

  • Marked fascial thickening (2–10× more than normal)
  • Inflammatory infiltrate in the fascia — lymphocytes, plasma cells, macrophages; eosinophils present in ~50–60% of biopsies, more often in the early phase
  • Thickening of collagen bundles, initially with active fibroblastic activity, later with mature, “hyaline” fibrosis
  • The inflammatory infiltrate may extend into adjacent layers: the subcutaneous tissue and the superficial layer of the muscle (perimysium)
/ note

The absence of eosinophils in the biopsy does not exclude EF. After prior steroid administration, eosinophils quickly disappear from the tissue even though the disease is ongoing. Also in the chronic, fibrotic phase, eosinophils may be few or absent. The diagnosis is made on the basis of the full clinical picture + histology, not the mere presence of eosinophils.

Ultrasound (US) — a supporting role

Ultrasound is a cheaper, faster alternative to MRI in several applications:

  • Initial assessment of fascial thickening (less sensitive than MRI, but available in a rheumatology office)
  • Monitoring progression/regression of changes over time
  • Assisting with biopsy (US-guided)
  • Excluding alternatives — e.g. deep vein thrombosis in a patient with a swollen limb

US is not sensitive enough to confirm a diagnosis of M35.4. In a systematic review of patients with EF, only ~25% had documented US changes despite an unambiguous clinical picture. This reflects the method’s limitations, not the absence of changes.

Sensitivity of each test for detecting M35.4

Proportion of M35.4 patients in whom a given investigation reveals typical changes. Full-thickness fascial biopsy is the gold standard — the other modalities support the clinical picture but do not replace it.

  • Fascial biopsy (full-thickness, skin-to-muscle)
    ≈ 95 % (gold standard)
  • Contrast MRI (T2 / STIR + gadolinium)
    ≈ 80–90 %
  • Peripheral blood eosinophilia
    60–80 %
  • Hypergammaglobulinaemia (electrophoresis)
    50–70 %
  • Musculoskeletal ultrasound
    ≈ 25 %

Full-thickness fascial biopsy stays the gold standard with about 95% sensitivity, because it reaches the layer where the disease actually lives. Contrast-enhanced MRI follows at 80–90% and is the best non-invasive tool — strong enough to guide the biopsy and monitor treatment, not strong enough to replace it. Peripheral eosinophilia and hypergammaglobulinaemia are supportive rather than confirmatory: present in roughly half to two thirds of patients, but normal in the rest. Musculoskeletal ultrasound rounds out the list at about 25% — useful for follow-up and as a referral aid, but inadequate as a stand-alone diagnostic step.

Pinal-Fernandez 2014 · Kissin 2015 · Daoussis 2022

Differential diagnosis

List of diseases that look similar to M35.4 and must be excluded:

  • Systemic sclerosis (SSc) — the most common “candidate”. Differs by acrosclerosis, Raynaud’s phenomenon, organ involvement, capillaroscopy changes, presence of anti-Scl-70 or anti-centromere antibodies.
  • Localised scleroderma (morphea, including pansclerotic) — may coexist with EF. Histologically, morphea begins in the dermis, EF in the fascia.
  • Nephrogenic systemic fibrosis (NSF) — associated with gadolinium in patients with advanced renal failure; the clinical picture is similar but the nephrological context is different.
  • Eosinophilia-myalgia syndrome (EMS) — historically linked to contaminated L-tryptophan (1989 epidemic); sporadic cases today.
  • Toxic Oil Syndrome — Spanish epidemic of 1981, now historical.
  • Inflammatory myopathies (dermatomyositis, polymyositis) — CK and aldolase markedly elevated, muscle biopsy with a different picture.
  • Chronic graft-versus-host disease (chronic GVHD) after bone-marrow transplantation — occasionally mimics EF; the clinical context settles it.
  • Scleromyxedema, scleredema diabeticorum, lipodermatosclerosis — similar skin changes, different location and context.
  • Primary lymphoedema, deep-vein thrombosis, circulatory failure — in the early phase of EF with swollen limbs, when induration is not yet present.

Why diagnosis takes so long

The median time from symptom onset to an M35.4 diagnosis in various series is 6–11 months. That is several times longer than for more common systemic diseases. Reasons:

  • Rarity of the disease — most primary-care physicians have never seen a patient with M35.4
  • Non-specific initial symptoms — swelling, pain and fatigue fit dozens of other conditions
  • First biopsies done as punch or needle — often negative because they do not reach the fascia
  • No rare-disease registry in many countries — limited access to referral centres
  • Seasonal symptom fluctuation — in the early phase, swelling may periodically subside, confusing the clinical picture
Median time from first symptom to diagnosis

In typical case series, M35.4 is diagnosed much later than more common autoimmune diseases. Every extra month of delay raises the risk of permanent contractures.

  • Median
  • M35.4 (eosinophilic fasciitis)
    6–11 mo.
  • Systemic sclerosis (SSc)
    ≈ 4 mo.
  • Systemic lupus erythematosus (SLE)
    ≈ 6 mo.
  • Rheumatoid arthritis (RA)
    ≈ 3 mo.

Median time to diagnosis in M35.4 is two to three times longer than in better-known autoimmune diseases. Rheumatoid arthritis is recognised in roughly three months, systemic sclerosis and lupus in four to six; M35.4 typically takes eight to eleven. The gap is driven by the rarity of the disease, the non-specific opening symptoms, and the high rate of first biopsies that don't reach the fascia. Every additional month of delay raises the risk of permanent joint contractures — which is why early referral to a rheumatologist matters more here than for the diseases sitting above on the chart.

Wright 2016 · Mango 2020 · NICE / EULAR comparators for RA and SLE

/ note

If you report to your doctor the combination of symptoms described here (symmetric swelling + gradual induration + restricted motion), ask for a full blood count with differential and — if eosinophilia is present — ESR, CRP, protein electrophoresis, ANA, anti-Scl-70. These tests are cheap, and their results within a day can steer diagnostics towards M35.4.

Sources

Full bibliography in the final version. Main sources:

  • Pinal-Fernandez I, et al. Diagnosis and classification of eosinophilic fasciitis. Autoimmun Rev 2014; 13: 379-382.
  • Lebeaux D, Sène D. Eosinophilic fasciitis (Shulman disease). Best Pract Res Clin Rheumatol 2012; 26: 449-458.
  • Kissin EY, et al. Diagnostic imaging in idiopathic inflammatory myopathy and eosinophilic fasciitis. J Rheumatol 2015.
  • Kransdorf MJ, Murphey MD. Eosinophilic fasciitis: spectrum of MRI findings. AJR Am J Roentgenol 2005; 184: 975-978.
  • Daoussis D, et al. The role of fascial thickness on MRI in the diagnosis of EF. Rheumatol Int 2022.
  • Lakhanpal S, et al. Eosinophilic fasciitis: clinical spectrum and therapeutic response in 52 cases. Semin Arthritis Rheum 1988; 17: 221-231.