Renal hemodynamic alterations assessed by doppler ultrasonography in familial mediterranean fever patients without overt renal disease: a cross-sectional study

Abstract

Objective

Familial Mediterranean fever (FMF) is an autoinflammatory disease that may lead to renal parenchymal involvement and amyloidosis over time. Experimental and biopsy-based data from various renal diseases suggest that renal parenchymal injury may be accompanied by early alterations in renal hemodynamics. This study aimed to evaluate Doppler ultrasonography–derived renal hemodynamic parameters in adult FMF patients without overt renal involvement and to compare these findings with those of healthy controls.

Methods

This cross-sectional observational case-control study included 77 adult FMF patients without significant proteinuria (> 500 mg/g) and 75 age- and gender-matched healthy controls. Bilateral renal resistance index (RRI) and pulsation index (PI) measurements were performed in all participants using Doppler ultrasonography. In subgroup analysis, FMF patients were divided into two groups based on urinary protein excretion: those with normal or minimal proteinuria (< 150 mg/g) and those with low-grade proteinuria (150–500 mg/g).

Results

Both RRI and PI scores were significantly elevated in FMF patients than in controls (p < 0.05 for both). Subgroup analysis in FMF patients further revealed that RRI scores and PI scores were significantly higher in patients with low-grade proteinuria than in patients with normal or minimal proteinuria (p < 0.05 for both).

Conclusions

Renal hemodynamic changes detected by Doppler ultrasonography can be identified in FMF patients, even when there is no apparent renal involvement. The greater increase in RRI and PI values was found in patients with low-grade proteinuria, indicating that changes in renal hemodynamics may be present at an early stage of the disease. These changes detected by Doppler ultrasonography may be used as an adjunctive tool for early evaluation of renal involvement in FMF.

Introduction

Familial Mediterranean Fever (FMF) is a genetically inherited autoinflammatory disease characterized by recurrent attacks of fever and serositis [1]. Although these attacks are intermittent, subclinical inflammation can persist even during attack-free periods [1]. When recurrent attacks and chronic inflammation are not adequately controlled, amyloid deposition, most commonly affecting the kidneys and the most serious and life-threatening complication of FMF, can lead to irreversible kidney damage and ultimately end-stage renal failure [1–2]. Although specific MEFV gene mutations and persistently high serum amyloid A (SAA) levels have been associated with an increased risk of amyloidosis, no single clinical or laboratory marker has yet been shown to have sufficient reliability to identify renal involvement in the pre-proteinuria stage [3–4]. Proteinuria is usually the earliest clinical sign of renal amyloidosis; however, irreversible kidney damage may have already occurred by the time overt proteinuria (OP) is detected [5]. This highlights the critical need for early and non-invasive markers that can detect renal involvement before OP develops in FMF patients.

The renal resistance index (RRI), derived from Doppler waveforms of intrarenal arteries, is a measurement reflecting intrarenal vascular resistance and provides a non-invasive and cost-effective method for assessing renal hemodynamics [6]. An increased RRI not only indicates large vessel pathology but also reflects changes in intrarenal microcirculation related to tubulointerstitial edema, fibrosis, and decreased parenchymal compliance, even in the absence of renal artery stenosis [6]. It has been associated with renal parenchymal damage in various renal and systemic conditions, including chronic kidney disease, lupus nephritis, and diabetic nephropathy [6–9].

Pulsatility index (PI) is another Doppler-derived indicator (DDI) reflecting intrarenal vascular resistance and arterial compliance based on systolic, diastolic, and mean flow velocities [10]. Previous studies have shown that PI measured from interlobar arteries is particularly associated with tubulointerstitial damage and can be used as a non-invasive marker to monitor the progression of renal damage [11]. Given that renal involvement in FMF develops on a background of chronic inflammation and microvascular stress rather than a primary vascular disease, subtle parenchymal changes may lead to hemodynamic changes occurring before overt clinical renal involvement. However, data are limited in which both RRI and PI are simultaneously assessed as DDI of early renal involvement in FMF patients without OP.

Therefore, the aim of this study is to evaluate renal resistive and pulsatility indices, measured by Doppler ultrasonography, as non-invasive renal hemodynamic parameters in FMF patients without overt proteinuria, and to compare the findings with a healthy control group.

Materials and methods

Study design and patients

This study was designed as a cross-sectional observational study. The study was conducted at Ataturk University Faculty of Medicine Hospital Rheumatology Outpatient Clinic, a tertiary referral center, between November 2021 and February 2022. Eligible participants were adults aged 18–65 years diagnosed with FMF according to the Tel Hashomer criteria, as revised by Livneh et al. [12].

Patients were excluded if they had other autoinflammatory or autoimmune diseases, active malignancy, ongoing infection, end-organ failure including hepatic failure, congestive heart failure, and renal failure, history of kidney transplantation, or cognitive impairment precluding adequate comprehension of the study procedures. Patients with conditions known to affect renal hemodynamics or protein excretion, including diabetes mellitus and hypertension, were also excluded. To minimize confounding related to structural renal abnormalities, participants with incidental renal or adrenal masses, loss of corticomedullary differentiation, or increased renal parenchymal echogenicity detected on baseline ultrasonography were excluded from the analysis. In addition, individuals with pre-existing CKD, OP, or an estimated glomerular filtration rate (eGFR) < 90 mL/min/1.73 m² were not included. After applying these criteria, 77 patients were included in the final analysis. Consecutive patients who met the inclusion criteria were screened for eligibility, and predefined exclusion criteria were applied prior to final enrollment.

A control group consisting of age-, sex-, and body mass index (BMI)–matched healthy individuals was recruited from hospital staff and community volunteers. Control participants were screened to exclude preexisting conditions such as FMF, other inflammatory rheumatic diseases, known kidney disease, hypertension, diabetes mellitus, cardiovascular disease, and medication use known to affect kidney function. Patients with abnormal renal ultrasound findings and proteinuria urinary protein-to-creatinine ratio (UPCR) > 150 mg/g) or eGFR < 90 mL/min/1.73 m² were also excluded from the study.

At the time of study evaluation, all FMF patients were receiving colchicine therapy, while a small subset of patients (n = 5) were additionally treated with IL-1 inhibitors.

The study was reported in accordance with the STROBE guidelines. The completed STROBE checklist is provided as Supplementary Table S1.

Measurements

After at least 8 h of overnight fasting, renal ultrasonographic examinations were performed in both the FMF and control groups using a Toshiba ultrasound system with a 2–5 MHz curvilinear transducer. All examinations were obtained by a radiologist blinded to the clinical grouping and were subsequently reviewed by a second radiologist. In cases of disagreement, measurements were re-evaluated jointly and a consensus value was recorded. The kidneys were evaluated in the supine position through a lateral lumbar oblique approach. B-mode imaging included longitudinal and transverse scans to assess renal size, parenchymal echogenicity, and corticomedullary differentiation, with inspiratory maneuvers used when necessary to improve visualization.

Renal doppler ultrasonography (RDUS) were performed at the level of interlobar arteries. Following the detection of vascular flow via color (CDI) and power Doppler imaging (PDI), spectral Doppler was employed at a 2–4 mm wide sample color box with an angle between 30 and 60 degrees. Measurements were recorded upon obtaining at least 3 uniform waveforms. At least three measurements were done in different interlobar arteries (superior, median, and inferior poles) in each kidney. Arithmetic average was recorded for each kidney. Peak systolic velocity (PSV), end-diastolic velocity (EDV), and PI values were obtained from the auto-measurements of the device. In order to minimize abnormal waveform appearance which could be attributable to true pathological process, real-time image acquiring parameters were adjusted by (1) increasing gain for amplification of the returning signals, (2) decreasing velocity scale and pulse repetition frequency (PRF) for decreasing aliasing artifact, (3) selecting low wall filter for incorporating low-amplitude velocities and normalizing waveform appearance, (4) decreasing angle of insonation for optimization of frequency shift that may occur in Doppler imaging. Resistive index was defined as (PSV-EDV/PSV) with < 0.7 considered as normal and PI defined as (PSV-EDV)/∑velocity with the range of 0.7–1.4 taken as normal value [13–14]. The PSV value of 180 cm/sec and above was considered as abnormal. Resistive index measurements were obtained from the interlobar arteries of both kidneys. To reduce intra-individual variability and because renal involvement in FMF is expected to be bilateral, the mean renal resistive index was calculated by averaging the right and left renal resistive index values and used for statistical analyses.

A urinary UPCR below 500 mg/g, corresponding approximately to a daily protein excretion below 500 mg/g, was used as an inclusion criterion, as this threshold is widely accepted in the literature to exclude patients with established renal amyloidosis or significant glomerular damage [15–16]. By applying this criterion, potential confounding effects related to advanced renal pathology were minimized.

To assess different stages of subclinical renal involvement, FMF patients without OP were stratified into two subgroups based on urinary protein excretion levels: Group 1 (normal or minimal proteinuria; UPCR < 150 mg/g) and Group 2 (low-grade proteinuria; UPCR 150–500 mg/g). According to commonly accepted nephrology guidelines, a UPCR value exceeding 150 mg/g represents the upper limit of normal urinary protein excretion and has been used as a marker of early renal involvement [17–18]. This stratification was applied to explore whether RDUS parameters, particularly the RRI demonstrate a graded association with increasing urinary protein excretion in the absence of overt proteinuria.

The primary outcome of the study was the RRI obtained by RDUS. Secondary variables included urinary protein excretion parameters, eGFR, and relevant clinical characteristics of the participants.Disease duration was defined as the time, in years, from the initial diagnosis of FMF to the date of study inclusion.

UPCR [(Spot urine protein(mg/dl)/ Spot urine creatinine(g/dl)] was measured using the first morning spot urine sample, and eGFR value developed by CKD epidemiology cooperation (CKD-EPI) was calculated from the urine and blood sample collected simultaneously from all patients [19–20].

Statistical methods

Continuous variables were assessed for normality using the Kolmogorov–Smirnov test. Comparisons between FMF patients and healthy controls were performed using the independent samples t-test for normally distributed variables and the Mann–Whitney U test for variables that did not meet normality assumptions. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate.

Following the comparison between FMF patients and controls, additional subgroup analyses were conducted within the FMF cohort. FMF patients were stratified into two subgroups according to UPCR (Group 1 < 150 mg/g and Group 2 150–500 mg/g). Demographic characteristics (age and sex), body mass index, disease duration, and relevant clinical and laboratory variables were compared between these two subgroups to assess differences associated with increasing levels of proteinuria.

Receiver operating characteristic curve analysis was used to assess the discriminatory performance of RDUS parameters for distinguishing FMF patients from healthy controls, and area under the curve values were calculated. A two-sided p value < 0.05 was considered statistically significant. All analyses were performed using SPSS software (version 23.0; IBM Corp., Armonk, NY, USA).

Sample size was calculated a priori based on the effect size reported in a previous study evaluating renal resistive index in FMF patients and healthy controls [21]. According to this analysis, a total of 128 participants (64 per group) were required to achieve 80% power at an alpha level of 0.05 for detecting a large effect size (Cohen’s d ≈ 0.8). To further increase statistical power, 77 FMF patients and 75 age- and sex-matched healthy controls were ultimately included in the study. Missing data were handled using complete-case analysis.

Results

A total of 128 patients were assessed for eligibility. After excluding 51 patients based on predefined exclusion criteria, the remaining 77 patients were included in the final analysis (Fig. 1).

Fig. 1.

Flowchart showing patient screening, exclusion criteria, and subgroup allocation according to UPCR levels. Abbreviations: FMF, Familial Mediterranean fever; UPCR, urine protein-to-creatinine ratio

Baseline demographic and clinical characteristics of the case and control groups are summarized in Table 1. The mean age was 32.9 ± 9.9 years in the FMF group and 31.8 ± 9.6 years in the control group, with no statistically significant difference between groups (p = 0.929). There were also no significant differences between the groups with respect to sex distribution, serum creatinine levels, or BMI (all p > 0.05).

Table 1.

Demographic and clinical characteristics of the control and case groups

Variable	All Groups	Control Group	FMF Group	p-value	Test
Age (years)	31.9 ± 9.9	31.8 ± 9.6	32.0 ± 9,9	0.929	m
Gender, n%					X²
- Male	57	30 (%40)	27 (%35.1)	0.530
- Female	95	45 (%60)	50 (%64.9)
BMI	23.4 ± 1.8	23.5 ± 1.8	23.4 ± 1.9	0.816	t
Creatinine	0.65 ± 0.14	0.62 ± 0.15	0.66 ± 0.13	0,264	m
eGFR	116.2 ± 22.7	111.2 ± 19.8	121.1 ± 24.4	0.002	m
Left RRI	0.64 ± 0.07	0.62 ± 0.06	0.66 ± 0.08	0.010	m
Right RRI	0.64 ± 0.07	0.62 ± 0.06	0.66 ± 0.08	0.006	m
RRI mean	0.64 ± 0.07	0.62 ± 0.06	0.66 ± 0.08	0.010	m
RRI mean ≥ 0,7, n%	33(%21.7)	4(%5.33)	29(%37.7)	0.000	X²
Left PI	1.14 ± 0.33	1.05 ± 0.26	1.24 ± 0.37	0.002	m
Right PI	1.13 ± 0.34	0.97 ± 0.17	1.29 ± 0.38	0.000	m
PI mean	1.14 ± 0.32	1.01 ± 0.19	1.27 ± 0.37	0.000	m
PI mean > 1.4	31(%20.4)	4(%5.3)	27(%35.1)	0.000	X²

Values are presented as mean ± standard deviation or median (interquartile range), as appropriate

Statistical tests: m, Mann–Whitney U test; χ², chi-square test; t, Student’s t-test

Abbreviations: FMF, familial Mediterranean fever; BMI, body mass index; RRI, renal resistive index; PI, pulsatility index; eGFR, estimated glomerular filtration rate

All DDI differed significantly between the FMF and control groups. The Left RRI, Right RRI, Left PI, Right PI, mean PI and Mean RRI values ​​were significantly higher in the Patient group compared to the Control group (all p < 0.05) (Table 1).

Receiver operating characteristic curve analysis demonstrated that LRRI, RRRI, LPI, and RPI values were significantly effective in distinguishing FMF patients from control subjects (Table 2; Fig. 2).

Table 2.

Receiver operating characteristic curve analysis of parameters for discriminating FMF patients from healthy controls

Variable	AUC	95% CI	p-value
Left RRI	0.621	0.532–0.710	0.010
Right RRI	0.629	0.540–0.718	0.006
RRI mean	0.620	0.530–0.710	0.011
Left PI	0.644	0.555–0.733	0.002
Right PI	0.770	0.694–0.847	0.000
PI mean	0.721	0.640–0.803	0.000

Abbreviations: FMF: Familial Mediterranean Fever, RRI: Renal Resistive Index, PI: Pulsatility Index

Fig. 2.

Receiver operating characteristic (ROC) curves showing the diagnostic performance of Doppler ultrasonographic indices for renal involvement. Footnote: Receiver operating characteristic (ROC) curves are presented for left and right renal resistive index (RRI), left and right pulsatility index (PI), and mean values of RRI and PI. The diagonal line represents the reference line indicating no discriminative ability

Among FMF patients without OP, 42 patients (54.5%) had normal or minimal proteinuria, whereas 35 patients (45.5%) exhibited low-grade proteinuria (Table 3). Mean RRI and PI values were significantly higher in FMF patients with low-grade proteinuria compared with those with normal or minimal proteinuria (RRI: 0.70 ± 0.07 vs. 0.63 ± 0.07; PI: 1.51 ± 0.33 vs. 1.07 ± 0.25; both p < 0.001).

Table 3.

Comparison of patients without overt proteinuria according to urinary protein excretion

Variable	Total Case	Group 1	Group 2	p-value	Test
Age (years)	32.01 ± 10.2	33.4 ± 11.0	30.8 ± 9.5	0.323	m
Gender, n%	77	35(45.5%)	42(54.5%)		X²
Male	27	11(31.4%)	16 (38%)	0.542
Female	50	24 (68.6%)	26 (62%)
BMI	23.5 ± 1,8	23.3 ± 1.5	23.6 ± 2	0.504	t
UPCR	160.4 ± 93.7	92.8 ± 23.2	241.5 ± 80.9	0.000	m
Colchicine mg/day	1.11 ± 0.32	1.21 ± 0.35	1.04 ± 0.28	0,021	t
Disease duration years	7(0–27)	5(0–20)	10(1–27)	0.068	m
Left RRI	0.66 ± 0.08	0.70 ± 0.07	0.64 ± 0.08	0.000	m
Right RRI	0.67 ± 0.08	0.70 ± 0.08	0.62 ± 0.07	0.000	t
RRI mean	0.66 ± 0.08	0.70 ± 0.07	0.63 ± 0.07	0.000	m
Left PI	1.23 ± 0.37	1.48 ± 0.33	1.03 ± 0.26	0.000	m
Right PI	1.31 ± 0.39	1.53 ± 0.35	1.09 ± 0.27	0.000	t
PI mean	1.27 ± 0.36	1.51 ± 0.33	1.07 ± 0.25	0.000	m

Values are presented as mean ± standard deviation or median (interquartile range), as appropriate

Statistical tests: m, Mann–Whitney U test; χ², chi-square test; t, Student’s t-test

Abbreviations: IQR, interquartile range; BMI, body mass index; RRI, renal resistive index; PI, pulsatility index; UPCR, urinary protein-to-creatinine ratio

Patients with overt proteinuria (UPCR ≥ 500 mg/g) were excluded. Group 1 included patients with normal or minimal proteinuria (UPCR < 150 mg/g), and Group 2 included patients with low-grade proteinuria (UPCR 150–500 mg/g)

Discussion

FMF is characterized by marked heterogeneity in clinical presentation and disease activity and remains one of the most common underlying causes of secondary AA amyloidosis. Persistent subclinical inflammation, even during clinically silent periods, may predispose patients to progressive and irreversible renal damage. Although FMF is often considered clinically benign due to the availability of effective treatments such as colchicine, renal involvement may begin insidiously and progress from very early stages of the disease. In this setting, DDI obtained before overt clinical or biochemical deterioration may aid in the early identification of patients at increased risk of renal involvement. Accordingly, early detection of renal manifestations represents a key component of long-term management in FMF. The primary aim of this study was to compare RDUS findings between FMF patients and healthy controls and to evaluate whether hemodynamic parameters indicative of early renal parenchymal involvement can be detected before the development of overt proteinuria.

These findings are also consistent with a study by Sav et al. In that study, an increase in renal tissue stiffness was detected by shear wave elastography in children with FMF whose renal symptoms were in remission, and this was interpreted as subclinical involvement and possible amyloid deposition [22]. It is thought that the increase in parenchymal stiffness, when considered together with the high RRI and PI values ​​observed in the present study, may be related to the increase in renovascular resistance. These observations may point to the presence of early renal hemodynamic changes accompanying the parenchymal changes detected by RDUS.

Additionally, systemic vascular factors may also need to be considered when interpreting our results. As noted by O’Neill, RRI is influenced not only by intrarenal vascular properties, but also by systemic hemodynamic factors such as pulse pressure, arterial compliance, and vascular stiffness [13]. Increased vascular stiffness has been reported in FMF and has been linked to chronic inflammation and endothelial dysfunction [23]. In this context, elevated RRI may reflect early vascular changes in FMF rather than kidney injury alone. Such changes may occur before amyloidosis or clinically significant proteinuria becomes evident.

Over time, the relationship between DDI and renal pathology has gained increasing acceptance as a practical, non-invasive approach for monitoring CKD. In a cohort of patients with CKD, Chen et al. demonstrated a strong association between RRI and renal histopathological changes—particularly tubulointerstitial damage—in patients who underwent biopsy, underscoring the potential utility of this index in disease monitoring [11]. Similarly, Swelam et al. have also shown positive correlations of RRI, CKD stages, and histopathological findings in pediatric patients. These results suggest that RRI levels increased with disease progression [24]. These studies support that RRI may reflect renal structural or hemodynamic alterations in selected CKD populations, although such findings cannot be directly extrapolated to FMF. In this regard, one of the important findings of the current study was that RRI and PI levels were significantly higher in FMF patients compared to controls, even though there was no difference in age, gender distribution, serum creatinine levels, or BMI.

Studies have shown that renal involvement in FMF patients is not limited to amyloidosis; non-amyloid glomerulopathies and tubulointerstitial changes can also occur in the early stages of proteinuria. In particular, the study by Yazilitas et al. reported that in pediatric FMF patients, both amyloid and non-amyloid renal pathologies were detected in cases presenting with proteinuria, and the most common finding in the non-amyloid group was mesangioproliferative glomerulonephritis [25]. In this context, early detection of renal damage will also be important in terms of directing differential diagnoses.

Several biopsy studies in patients with inflammatory renal disease have established a strong correlation between high RRI values and the extent of histopathological lesions. In lupus nephritis, for example, a high level of RRI values was found to be associated with a more severe histological lesion, particularly class IV lupus nephritis disease [8]. In the context of FMF, Sezer et al. likewise reported increased RRI values in patients without OP; however, in contrast to our findings, no significant difference in PI was observed [22]. Some studies have also highlighted the potential prognostic value of renal PI in predicting renal function decline. In this context, Kosaki et al. demonstrated an independent association between PI and CKD progression and suggested that heterogeneity in renal PI may play a pathophysiological role in disease progression [26]. Consistent with the findings of Kosaki et al., our results support an increase in PI, while differing from the results reported by Sezer et al. The inter-study variability in PI findings may be due to the study’s design, the methods used to measure proteinuria, and the sensitivity of PI to various systemic hemodynamic influences, including the hardening of blood vessels that occurs with increasing age. Notably, our study was conducted in a relatively young population, and there were no differences between groups in terms of age, gender distribution, serum creatinine levels, or body mass index; however, both RRI and PI values ​​were found to be significantly higher in FMF patients compared to controls. The use of a standardized RDUS protocol and experienced specialists in the present study may have contributed to the low measurement variability.

Although Touitou et al. did not show a direct relationship between attack frequency and amyloidosis prevalence, Babaoğlu et al. identified persistent inflammation as an independent risk factor for proteinuria, amyloidosis, and renal failure [27–28]. In this context, the increases in RRI and PI observed in FMF on the background of chronic inflammation may be related to early and subclinical renal changes that may develop before amyloid deposition. Indeed, in this study, it was observed that renal hemodynamic changes could be detected in FMF patients even without significant proteinuria. While RRI values ​​generally remained within the normal range in patients with normal or minimal proteinuria, they approached the upper limit of normal in individuals with low-grade proteinuria; this suggested that RRI values ​​close to 0.70 may be significant in terms of early renal involvement. When evaluated together with Doppler findings, it is understood that even small amounts of proteinuria can have clinical significance. Furthermore, the limited number of studies evaluating renal hemodynamics with RDUS in FMF patients without significant proteinuria increases the importance of our findings in demonstrating that mild increases in intrarenal vascular resistance can be detected before significant biochemical abnormalities emerge.

However, there are some limitations to be considered. The cross-sectional nature of the study does not allow for the establishment of causality, and the sample size might affect the generalizability of the results. Moreover, the data on serum amyloid A, which might have added more information regarding the relationship between persistent inflammation and renal vascular abnormalities, were not available. Future prospective studies incorporating serial RRI measurements together with proteinuria assessment, inflammatory markers such as SAA, and attack-related disease activity parameters may help further clarify the diagnostic and prognostic role of renal hemodynamic abnormalities in FMF.

In summary, the observed increases in the DDI, especially the RRI values approaching 0.70, may represent early renal hemodynamic abnormalities preceding the onset of overt proteinuria in patients with FMF. In this context, RDUS should be considered a supplementary tool that may provide complementary information beyond conventional proteinuria assessment, rather than a substitute for established urinary markers.

Author contributions

M.A.M.: Conceptualization, Study design, Supervision, Critical revision of the manuscript. Z.T.: Conceptualization, Study design, Data collection, Data curation, Formal analysis, Writing – original draft. G.P.: Supervision, Critical revision of the manuscript. M.R.A.: Data collection, Formal analysis, Interpretation of data. E.B.D.: Data collection, Formal analysis, Interpretation of data. H.A.: Data collection, Interpretation of data, Writing – review & editing.

Funding

The authors did not receive support from any organization for the submitted work.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethical approval

This study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Clinical Research Ethics Committee of Ataturk University Faculty Of Medicine (approval date: December 04, 2021; protocol no: 07/43). Written informed consent was obtained from all participants prior to enrollment.

Competing interests

The authors declare no competing interests.