Authors: Haruka Kato, Takashi Yokoyama, Shun Manabe, Momoko Seki, Yusuke Ushio, Shiho Makabe, Shizuka Kobayashi, Hiroshi Kataoka, Kosaku Nitta, Junichi Hoshino
Categories: Case Report, Automated urine sediment analyzers, Clinical awareness, Fabry disease, Mulberry bodies, Mulberry cells, Urinary sediment, Podocyte
Source: BMC Nephrology
Authors: Haruka Kato, Takashi Yokoyama, Shun Manabe, Momoko Seki, Yusuke Ushio, Shiho Makabe, Shizuka Kobayashi, Hiroshi Kataoka, Kosaku Nitta, Junichi Hoshino
Fabry disease (FD) is an X-linked lysosomal storage disorder where early diagnosis is crucial to prevent irreversible organ damage. However, diagnosis is often delayed due to heterogeneous clinical presentations, particularly in heterozygous females and subjectively asymptomatic patients. Urinary mulberry cells and bodies are pathognomonic markers of FD, reflecting globotriaosylceramide (Gb3) accumulation in podocytes and consequent podocyte injury. Despite their diagnostic utility, these subtle morphological features are frequently missed by modern automated urine sediment analyzers and routine manual microscopy performed without specific clinical suspicion of FD. We emphasize the importance of “targeted” manual urine sediment examination triggered by clinical awareness.
We report a three-generation Japanese family with FD where targeted manual urine sediment examination served as the pivotal cue for diagnosis. The index case, a 61-year-old woman, was referred for incidental electrocardiographic abnormalities suspicious for left ventricular hypertrophy and trace proteinuria. Initial screening with an automated urine sediment analyzer and standard routine microscopy yielded non-specific findings. However, the revelation of a family history of FD during the clinical interview prompted a targeted manual re-evaluation of the urine sediment. This focused review successfully identified characteristic mulberry cells and bodies, triggering the genetic confirmation of a pathogenic GLA variant (c.761_763del). Subsequent screening of her subjectively asymptomatic 32-year-old daughter and 7-year-old grandson initially showed negative results on automated and routine non-targeted microscopy; however, targeted manual review revealed mulberry cells and bodies in the daughter and mulberry bodies in the grandson. These findings provided a compelling rationale for genetic testing in these subjectively asymptomatic relatives, confirming the diagnosis. Furthermore, urinary mulberry cell counts correlated with disease severity across the three family members and decreased following enzyme replacement therapy in the index case, paralleling the reduction in plasma lyso-Gb3.
This family illustrates that laboratory automation and routine non-targeted microscopy may have limitations in detecting Fabry nephropathy. Clinical awareness—the conscious suspicion of FD—serves as the decisive trigger for performing targeted manual urine sediment examination. Close communication between clinicians and laboratory technologists may be beneficial for identifying subtle diagnostic clues like mulberry cells and bodies that automated systems might overlook, ensuring early diagnosis and timely therapeutic intervention.
Fabry disease (FD) is an X-linked lysosomal storage disorder where α-galactosidase A (α-Gal A) deficiency leads to systemic globotriaosylceramide (Gb3) accumulation [1]. While enzyme replacement therapy (ERT) and chaperone therapy can prevent organ damage if initiated early, diagnosis is often delayed due to heterogeneous presentation, particularly in heterozygous females who may range from asymptomatic to severe phenotypes [2–3]. Consequently, non-invasive biomarkers capable of triggering genetic testing are clinically critical [4–5].
Urinary mulberry bodies and cells—comprising lamellated lipid inclusions within lysosomes shed from podocytes and exfoliated podocytes themselves—are highly specific markers for FD [6–8]. While lipiduria, such as oval fat bodies, is well-known in nephrotic syndrome [9], FD-associated particles exhibit distinct morphological differences, such as whorl-like structures [6–8]. Their detection offers a simple clue to Fabry disease even before overt organ dysfunction [10–11]. However, with the widespread adoption of automated urine sediment analyzers, morphological details required to identify these subtle structures are often lost. This limitation extends beyond FD; automated analyzers cannot visualize characteristic appearances, which inadvertently compromises the diagnostic value of urine sediment as a “liquid biopsy” [9, 12]. Therefore, despite technological advances, accurate detection of mulberry cells and bodies still relies heavily on visual assessment [6–8, 10–11]. Furthermore, given the low prevalence of FD and the frequent lack of recognition, maximizing the utility of this “liquid biopsy” requires a targeted manual review prompted by specific clinical suspicion.
We describe a three-generation family where mulberry cells and bodies were initially missed by automated urinalysis and routine manual microscopic examination but identified after targeted manual re-evaluation prompted by the clinical suspicion of FD. This report illustrates how clinical awareness and clinician-technologist communication can transform routine urinalysis into a decisive diagnostic tool triggering genetic confirmation.
Fig. 1Pedigree of the family. The proband (Index Case, III-3) is indicated by an arrow. Affected individuals are shown in black. (Case 2 is IV-3, Case 3 is V-1)
A 61-year-old Japanese woman was referred to the Department of Nephrology at Tokyo Rosai Hospital—a regional core community hospital with no prior experience in diagnosing FD—after an ECG suggested left ventricular hypertrophy (LVH), and trace hematuria and proteinuria were detected during a routine health checkup. Four years prior, she had undergone a detailed evaluation at a specialized cardiology clinic for similar findings, but these were judged to be “within normal limits”, and she was followed conservatively. Her son had been diagnosed with classic FD more than 30 years earlier. However, she had been informed at that time that she was “only a carrier,” leading to a misconception that she required no medical follow-up. Consequently, she did not disclose this critical family history during the evaluation at the specialized cardiology clinic.
At presentation, she was subjectively asymptomatic, with normal blood pressure and no history of neuropathic pain, hypohidrosis, or angiokeratomas. Her medical history included appendicitis, pyelonephritis, and uterine myoma. Laboratory findings showed serum creatinine 0.77 mg/dL, NT-proBNP 390 pg/mL, and urinary protein 0.32 g/gCr. Echocardiography revealed mild concentric LVH (LVDd 38 mm, LVDs 23 mm, IVSt 10 mm, LVPWt 12 mm, LV mass index 94.8 g/m²) (Fig. 2A). Brain MRI, Holter ECG, and ophthalmologic, otolaryngologic, and dermatologic evaluations were unremarkable.
Fig. 2Clinical and pathological findings of the Index Case. (A) Echocardiography showing mild concentric left ventricular hypertrophy. (B) Light microscopy of the urine sediment showing a characteristic mulberry cell. (C) Light microscopy showing mulberry bodies. (D) Electron microscopy of the kidney biopsy specimen showing laminated myeloid bodies (zebra bodies and Myeloid bodies) in the podocyte cytoplasm
Initial screening using an automated urine sediment analyzer (UF-1500; Sysmex Corporation, Kobe, Japan) was performed, followed by routine manual microscopic examination, which is standard protocol for all nephrology cases at our hospital. Both analyses yielded non-specific findings, and neither mulberry bodies nor mulberry cells were identified. However, while explaining the differential diagnosis of LVH, her family history of FD was revealed. We shared this clinical suspicion with the laboratory technologists, leading to a targeted manual re-examination of the urine sediment for mulberry bodies and cells. This focused review successfully revealed characteristic mulberry cells and bodies (Fig. 2B and C). Given the strong suspicion of FD based on the family history and these specific urinary findings, a kidney biopsy was performed despite the trace proteinuria. The biopsy demonstrated vacuolated podocytes on light microscopy and myelin/zebra bodies on electron microscopy (Fig. 2D). Genetic testing confirmed a known pathogenic GLA variant (c.761_763del), establishing the diagnosis of FD.
Her plasma lyso-Gb3 level was 12.4 nmol/L, and the urinary mulberry cell count and mulberry body count, quantified by our previously reported method [13], were 5.69 × 10 4 cells/gCr and 4.52 × 10 5 bodies/gCr, respectively. Enzyme replacement therapy (ERT) with agalsidase alfa was initiated. After six months, plasma lyso-Gb3 decreased to 6.2 nmol/L, and the urinary mulberry cell count declined to 2.32 × 10⁴ cells/gCr. Renal function and cardiac parameters remained stable for two years. Subsequently, she was switched to migalastat (chaperone therapy) due to the patient’s preference for oral medication and the verification that her specific variant is classified as amenable to migalastat based on in vitro assay data [14]. Following the switch, her α-Gal A activity increased from 9.4 to 12.4 nmol/h/mL to 23.8 nmol/h/mL. Her clinical course has remained stable, with plasma lyso-Gb3 at 7.9 ng/mL and urinary mulberry cell count at 1.72 × 10⁴ cells/gCr.
The 32-year-old daughter of the index case underwent evaluation following her mother’s diagnosis. Although she had undergone multiple urinalyses at our hospital over the preceding years, no abnormalities had been noted. At presentation, she was subjectively asymptomatic, with normal vital signs and no significant medical history. Laboratory findings revealed serum creatinine 0.60 mg/dL, NT-proBNP 22 pg/mL, and urinary protein 0.05 g/gCr. ECG, echocardiography (LV mass index 26.3 g/m²), brain MRI, and Holter ECG were normal. Even at this visit for detailed examination, the initial routine automated urinalysis and manual microscopy failed to detect abnormal cells. However, once the clinical suspicion of FD based on family history was shared with the laboratory technologists, the urine sediment was re-evaluated manually, revealing mulberry cells and bodies. Quantitative analysis showed 0.61 × 10⁴ cells/gCr, approximately one-tenth of her mother’s count (Table 1). Genetic analysis confirmed the same GLA c.761_763del variant, and plasma lyso-Gb3 was 3.8 nmol/L. At age 35, she began chaperone therapy with migalastat. She remains subjectively asymptomatic, with no evidence of renal or cardiac involvement during follow-up.
Table 1Clinical characteristics, laboratory findings, and urinary sediment analysis of the family membersAge at diagnosisIndex Case (III-3)Daughter (IV-3)Grandson (V-1)61327SexFemaleFemaleMaleNT-proBNP (pg/mL)3902221LV mass Index94.826.3NMeGFR (ml/min/1.73m^2^)58.692.7103.6UP (g/gCr)0.320.050.09GLA activity (nmol/mg P/h)10.7NM1.8Plasma lyso-Gb3 (nmol/L)12.43.833.8Mulberry Cell Count (x10^4^/gCr)5.690.610Mulberry BodiespresentpresentpresentTreatmentEnzyme replacement therapy → MigalastatMigalastatEnzyme replacement therapy
A 7-year-old boy, the grandson of the index case, was screened because of the family history. He was subjectively asymptomatic, with normal growth and no significant past medical history. Laboratory findings showed serum creatinine 0.43 mg/dL, NT-proBNP 21 pg/mL, and urinary protein 0.09 g/gCr. ECG, echocardiography, ophthalmologic, and dermatologic examinations were normal. Similar to the other cases, routine automated urinalysis and manual microscopy at the initial screening did not detect abnormal cells. However, upon communicating the specific suspicion of FD to the laboratory technologists, the urine sediment was re-evaluated manually, revealing mulberry bodies but no mulberry cells, consistent with the notion that mulberry bodies represent the degraded remnants of mulberry cells. His leukocyte α-Gal A activity was 1.8 nmol/mg protein/hour, consistent with classic FD, and plasma lyso-Gb3 was markedly elevated at 33.8 nmol/L (Table 1). ERT with agalsidase beta was initiated. At age 9, he experienced transient extremity pain during a febrile episode but has since remained symptom-free with preserved renal and cardiac function.
In all three family members, automated urine sediment analysis and routine microscopy initially failed to detect mulberry cells or bodies. However, once FD was clinically suspected, close communication between clinicians and laboratory technologists led to a targeted manual re-examination of the urine sediment, which successfully revealed Fabry-specific findings and enabled genetic confirmation. This family underscores that clinical awareness is the key trigger for performing a focused manual urine sediment examination, which remains indispensable for recognizing subtle but pathognomonic urinary features such as mulberry cells and bodies—findings that may otherwise be missed in routine or automated analyses.
Fabry disease is a multisystemic disorder whose renal manifestations may remain subclinical for decades, particularly in heterozygous females. Because specific therapies are most effective when initiated before irreversible organ damage develops [2, 3], early recognition of FD is essential but remains challenging. In this context, urinary mulberry cells and bodies serve as highly specific [6–8], non-invasive microscopic findings that reflect Gb3 accumulation within podocytes [15–17] and can even precede overt proteinuria.
Previous reports have established the diagnostic value of these structures. Selvarajah et al. [6] and Nakamura et al. [7] demonstrated that urine microscopy detects urinary mulberry bodies/cells with high sensitivity and specificity for FD when performed with a “targeted” approach. Furthermore, Fogazzi et al. [8] provided a detailed morphological classification of the fat particles in FD, highlighting significant differences from the oval fat bodies or lipid droplets typically seen in nephrotic syndrome. They reported that FD-associated particles exhibit diverse morphologies, such as “whorl-like” structures, and that isolated “bodies” are often more frequent than intact “cells.” These distinct morphological features underscore the utility of mulberry bodies/cells as a rationale for suspecting FD and proceeding to genetic testing. However, it must be emphasized that this high diagnostic yield is predicated on the examiner’s specific intent to identify them.
In daily clinical practice, commercial automated urine sediment analyzers are incapable of detecting these specific morphological features [18]. Furthermore, even with routine manual microscopy, identification without specific suspicion is notoriously difficult; successful detection in a general screening context is exceptionally rare and limited to a few case reports where they were fortuitously identified by chance [10–11]. Thus, a significant gap exists between the potential diagnostic utility of the test and its actual performance in daily clinical practice in a non-targeted routine setting.
Regarding their pathological significance, we and others previously provided immunocytochemical evidence that urinary mulberry cells originate from podocytes containing Gb3-enriched lysosomes [15–17]. Furthermore, it is reported that mulberry cell excretion correlates with the histological severity of podocyte vacuolation and decreases in response to ERT [13, 16]. These findings, consistent with reports on podocyturia [19, 20], indicate that urinary mulberry cells reflect ongoing podocyte injury mediated by Gb3 accumulation and may serve as biomarkers not only for early detection but also for monitoring therapeutic efficacy.
Our present family further supports this concept. Using the same GLA variant (c.761_763del) across three generations, we quantitatively measured urinary mulberry cells and found marked the grandmother exhibited 5.69 × 10^4^ cells/gCr, the daughter 0.61 × 10^4^ cells/gCr, and the grandson only detectable mulberry bodies. These generational differences likely represent different disease stages related to cumulative Gb3 accumulation. Notably, the decrease in urinary mulberry cell count following ERT in the index case paralleled the reduction in plasma lyso-Gb3, reinforcing its utility as a possible dynamic marker [13].
The key lesson from this family is the limitation of routine urinary screening. Although this report is limited to a single family and does not statistically evaluate how clinical information modifies diagnostic sensitivity and specificity, our experience compellingly illustrates that “clinical awareness”—the conscious suspicion of FD—remains the decisive trigger that bridges the gap between a negative routine urinalysis and a positive targeted examination. While laboratory automation enhances efficiency, it cannot replace morphological insight guided by clinical suspicion. Close collaboration between clinicians and technologists, along with targeted manual urine sediment examination, remains indispensable for identifying diagnostic clues that automated systems miss, ensuring early diagnosis of Fabry disease.