Introduction

Sarcoidosis is a systemic disorder of unknown cause with a wide variety of clinical and radiologic manifestations. The diagnosis is usually made on the basis of these manifestations supported by histologic findings. Systemic manifestations (eg, Löfgren syndrome, Heerfordt syndrome) are commonly seen at clinical examination. Bilateral hilar lymphadenopathy is the most common radiologic finding-frequently with associated pulmonary infiltrates-and typically has a characteristic perivascular distribution at high-resolution chest computed tomography. Radiologic findings in the short tubular bones of the hands and feet and magnetic resonance imaging findings of nodular involvement of muscle are often sufficient to raise suspicion for sarcoidosis. In the liver, spleen, kidneys, and scrotum, coalescing granulomas form nodules whose imaging features may occasionally be nonspecific, although familiarity with the relevant clinical settings will be helpful in recognizing the presence of sarcoidosis. Radiologic recognition of cardiac and central nervous system involvement is also important because patients may be only mildly symptomatic. The clinical course and prognosis of sarcoidosis are highly variable, often correlating with the mode of onset. Familiarity with the clinical and radiologic features of sarcoidosis in various anatomic locations plays a crucial role in diagnosis and management. Copyright RSNA, 2004

The resistive index (RI), calculated from the duplex Doppler waveform, was compared with clinical and laboratory findings and the results of renal biopsy in 41 patients with nonobstructive (medical) renal disease. Kidneys with active disease in the tubulointerstitial compartment had a mean RI of 0.75 +/- 0.07. This was statistically significantly different (p less than .01) from the RI in kidneys with disease limited to the glomeruli (mean RI of 0.58 +/- 0.05). Acute tubular necrosis resulted in an elevated RI (mean RI = 0.78 +/- 0.03) as did vasculitis/vasculopathy (mean RI = 0.82 +/- 0.05). Patients with hypertension, proteinuria, or hematuria did not have kidneys with a significantly higher RI than did patients without these clinical factors. Kidneys found to be abnormally echogenic did not have an RI significantly different from kidneys of normal echogenicity. There was a weak correlation between creatinine level and RI value, reflected by a linear correlation coefficient of 0.34. In patients with normal renal RIs, the mean creatinine level was 1.7 +/- 1.7, whereas in those with abnormal RI values (greater than or equal to 0.70), the mean creatinine level was 3.7 +/- 3.6. We conclude that some forms of nonobstructive renal disease can produce changes in the Doppler waveform detectable by RI measurement. The production of Doppler waveform changes is strongly influenced by the site of the main disease within the kidneys. Active disease within the tubulointerstitial compartment (acute tubular necrosis, interstitial nephritis) or vasculitis/vasculopathy generally resulted in an elevated RI, whereas disease limited to the glomeruli, no matter how severe, did not significantly elevate the RI. Degree of renal dysfunction as indicated by serum creatinine level probably affects the Doppler waveform to some degree, but the relationship is weak.

Ultrasonography (US) of the native kidneys is commonly requested for acute renal failure (ARF), although in most cases the examination results are negative. To detect changes in the Doppler waveform associated with ARF and determine whether Doppler US can provide significant diagnostic information not available with standard US, 91 patients with ARF were studied to determine a mean resistive index (RI) for each patient. Forty-six patients had acute tubular necrosis (ATN) with a mean RI +/- 1 standard deviation of .85 +/- .06, which was significantly higher than the mean RI of .67 +/- .09 in 30 patients with prerenal ARF (P less than .01). Fifteen patients had ARF due to non-ATN intrinsic renal disease (mean RI, .74 +/- .13). An elevated RI (greater than or equal to .75) occurred in 91% of patients with ATN versus only 20% of patients with prerenal azotemia. Patients with severe liver disease (hepatorenal syndrome) are a subset of those with prerenal ARF that accounted for most of the elevated RIs in this group. The study demonstrates that intrarenal Doppler US allows detection of changes associated with ARF far more often than standard US. More important, Doppler US may be helpful in distinguishing ATN from prerenal azotemia.